Thèse de doctorat de

L‘UNIVERSITE DE RENNES

Ecole Doctorale n° 600

Ecologie, Géosciences, Agronomie, Alimentation
Spécialité : Ecologie et évolution

Par
Valentin Cabon
Effets de la variabilité microclimatique sur la faune en milieu urbain dans
un contexte de changement climatique

Thèse présentée et soutenue à Rennes, le 07/10/2024

Par
Valentin Cabon
Unité de recherche : UMR 6553 ECOBIO

Rapporteurs avant soutenance :

Effets de la variabilité microclimatique sur la faune en milieu urbain dans
Sophie Beltran-Bech Maîtresse de conférences HDR – Université de Poitiers
un contexte
Yves Richard de changement climatique
Professeur HDR – Université de Bourgogne

Thèse présentée et soutenue à Rennes, le 07/10/2024

Composition du Jury :
Unité de recherche : UMR 6553 ECOBIO
Président
Thèse N° : : Sylvain Pincebourde Directeur de recherche HDR – CNRS UMR 7261 IRBI

Rapporteurs : Sophie Beltran-Bech Maîtresse de conférences HDR – Université de Poitiers

Yves Richard Professeur HDR – Université de Bourgogne

Examinateurs : Sandra Lavorel Directrice de recherche HDR – CNRS UMR 5553 LECA
Sylvain Pincebourde Directeur de recherche HDR – CNRS UMR 7261 IRBI

Dir. de thèse : Benjamin Bergerot Maître de conférences HDR – Université de Rennes

Hervé Quénol Directeur de recherche HDR – CNRS UMR 6554 LETG
Avant-propos

Cette thèse a été réalisée à Rennes, d'octobre 2021 à septembre 2024 au sein de l'UMR ECOBIO, en
partenariat avec le laboratoire LETG. Elle a été cofinancée par la Région Bretagne au travers du
dispositif ARED et par l'Université de Rennes. Un soutien financier a également été apporté par le
CNRS dans le cadre du programme MITI (Mission pour les Initiatives Transverses et
Interdisciplinaires) BiodivR. Ce projet de thèse a également reçu le soutien financier et logistique de
la zone Atelier Armorique.
Remerciements
Pour commencer, je remercie chaleureusement mes encadrants Benjamin et Hervé. Aussi bien
professionnellement qu’humainement, cela aura été un vrai plaisir de travailler avec vous deux
durant ces trois années. Merci tout particulièrement pour votre implication dans ce projet de thèse,
pour votre dynamisme et pour votre disponibilité de tous les instants !
Je remercie également les membres du jury, Sophie Beltran-Bech, Yves Richard, Sandra Lavorel et
Sylvain Pincebourde d’avoir accepté l’invitation à évaluer ce travail.
Merci également aux membres de mon comité de suivi individuel Magali Deschamps-Cottin et
Vincent Dubreuil pour leurs conseils avisés et bienveillants.
Vincent, merci également pour ta disponibilité dans les échanges, des retours toujours très
pertinents, et le partage des données d’ICU, sans lesquelles ce travail n’aurait pas été possible. Merci
également à Charlotte Brabant de m’avoir aidé à exploiter les données issues de ce fabuleux réseau !
Je remercie aussi Jonathan Lenoir, Eva Gril et Sylvain Pincebourde de m’avoir guidé dans les analyses
microclimatiques. Hâte de poursuivre !
Merci à mes proches collaborateurs du SpiderLab, en particulier à Dylan Amiar et Aurélien Ridel pour
leur aide précieuse sur l’identification des araignées, je garderai un très bon souvenir des heures
passées sur les loupes binoculaires en votre compagnie. Merci d’avoir su me transmettre les bons
tuyaux pour progresser rapidement, et ce toujours dans la bonne humeur. Merci également à Julien
Pétillon de m’avoir encouragé à travailler sur les araignées en début de thèse et merci à Julien et
Kaïna Privet de m’avoir invité à participer aux réunions du SpiderLab, c’est une vraie chance d’avoir
pu participer à ces temps d’échanges réguliers !
Un grand merci à Romain Georges pour son aide précieuse sur le terrain et sur l’identification des
carabes. L’été 2022 a été particulièrement marqué par les vagues chaleur (on s’en souvient !), idéal
comme contexte d’étude, moins quand il faut crapahuter de longues journées sur le terrain ;-)
Merci à Yann Laurent de m’avoir proposé son aide sur l’identification des staphylins et de m’avoir
intégré aux travaux réalisés sur ce taxon par les étudiants de l’Institut Agro.
Le travail n’aurait pas été si stimulant sans les nombreux collaborateurs qui m’ont aidé à récolter les
données en participant activement aux relevés. En particulier, Merci à Baptiste Bongibault, Maxence
Rocoplo, Nathalie Boulier-Monthean et Didier Le Cœur de l’Institut Agro de m’avoir ouvert les portes
du domaine de l’institut pour réaliser les piégeages, et d’avoir mobilisé des étudiants motivés pour
les séances de tri. Longue vie au projet TAXON !
Je remercie chaleureusement Matthieu Beaufils pour son implication dans cette thèse, merci
Matthieu de m’avoir fait parfois décoller les yeux du sol et de ses arthropodes pour observer et
écouter les oiseaux ! Merci également à toi et Sylvie pour vos invitations à découvrir de tout autres
écosystèmes sur la côte normande, de vraies bouffées d’air frais durant la thèse !
Certains travaux n’auraient pas été possibles sans le soutien de stagiaires : Maud Launoy, Nytia
Triveillot, Sarah Carpentier, Mathis Ellien, Brendan Blanchard, Alice Furet, Louis Copin et Benoît
Deletre. Un grand merci à vous toutes et tous d’avoir été aussi investis et rigoureux durant vos stages.
J’espère vous recroiser bientôt !
Merci à la DJB et Rennes Métropole, en particulier Charlotte Vincent et Bertrand Martin, ainsi qu’à
tous les agents de la ville de Rennes que j’ai pu croiser sur le terrain pour votre collaboration dans la
mise en place des campagnes de relevés. Merci également aux services des jardins des communes
de Cesson-Sévigné, Saint-Jacques-de-la-Lande et Chantepie.
Je remercie également les associations qui m’ont ouvert les portes de leurs jardins, et qui ont ainsi
largement contribué au bon déroulement des sessions de terrain. Merci notamment au GAEC du
canal Saint-Martin, au collectif 35 Volts de la ferme de Quincé, à Vert le Jardin et aux Cols Verts !
Étudier l’écologie est l’affaire de tous, y compris (et surtout !) des plus jeunes. C’est pourquoi je
remercie chaleureusement Delphine Ducourtioux du lycée Emile Zola de m’avoir non seulement
permis d’échantillonner en plein cœur de l’îlot de chaleur rennais, mais aussi de m’avoir invité à
communiquer sur mon projet de thèse auprès de ses élèves en classes de SVT.
Et parce que tout gradient a deux extrémités, je remercie également la DIR Ouest, et
particulièrement Astrid Thomas-Bourgneuf et Guénaël Kernen, de m’avoir permis l’accès aux
dépendances vertes plus « fraîches » que le centre-ville, aux abords de Rennes.
Également un merci tout spécial à Julio qui a donné beaucoup de son temps à me transmettre son
savoir geek, qui in fine m’a permis d’économiser un temps précieux. On partait de loin, mais grâce à
toi manier le cluster est devenu un jeu d’enfant ;-)
Mon gout pour l’écologie urbaine ne s’étant originellement pas dévoilé à Rennes mais à Berlin, je
profite de ces remerciements pour saluer mes anciens collègues de l’institut d’écologie de
l’Université Technique de Berlin, Sascha Buchholz, Moritz von der Lippe, Birgit Seitz, Ingo Kowarik,
Tanja Straka et tous ceux que j’ai pu croiser à la Rothenburgstraße 12. Merci à vous pour le prêt des
capteurs qui ont été essentiels pour mener à bien mes recherches sur Rennes.
Enfin, je n’oublie pas mes parents, ma sœur Juliette, et Cédric. Merci à vous 4 pour votre soutien
sans faille ! Merci aussi à tous/tes les copain/ines qui se reconnaîtront, pour m’avoir aidé à sortir la
tête du guidon parfois ;-)
Et aussi et surtout à Anna, qui m’a soutenu et encouragé pendant ce doctorat, un énorme merci à
toi d’avoir été là.
 Table des matières
Table des matières ............................................................................................................................... I
Liste des figures ................................................................................................................................. V
Liste des tableaux ............................................................................................................................. VII
Liste des annexes ............................................................................................................................ VIII

1 Partie I : Introduction générale ..................................................................................................... 1

1.1 Réchauffement climatique et activités humaines 1
1.1.1 Urbanisation ...................................................................................................................................... 1
1.1.2 Effets de l’urbanisation sur le climat ..................................................................................................... 2
1.1.3 Genèse de l’îlot de chaleur urbain ........................................................................................................ 4
1.2 Réponse des organismes aux réchauffements 7
1.2.1 Impact sur les organismes endothermes............................................................................................... 7
1.2.2 Les ectothermes : des organismes particulièrement vulnérables ............................................................. 8
1.2.3 Décalages spatiaux .......................................................................................................................... 10
1.2.4 Décalages temporels ........................................................................................................................ 11
1.2.5 Adaptations locales et plasticité phénotypique..................................................................................... 12
1.2.6 Altération des interactions biotiques et des communautés .................................................................... 12
1.3 Ecosystèmes urbains 13
1.3.1 La ville : un modèle idéal pour étudier les réponses des organismes au changement climatique .............. 13
1.3.2 Interactions du climat avec les autres composantes urbaines ............................................................... 15
1.3.3 Pertes d’habitat et de connectivité associées à l’urbanisation ............................................................... 16
1.3.4 Potentiels refuges pour la biodiversité ................................................................................................ 16
1.3.5 Services écosystémiques .................................................................................................................. 16
1.4 L’ICU à Rennes 17
1.4.1 Contexte géographique et climatique ................................................................................................. 17
1.4.2 Historique de la caractérisation de l’ICU rennais .................................................................................. 19
1.4.3 Intensité de l’ICU durant la période d’étude ......................................................................................... 23
1.4.4 Climat urbain et faune à Rennes : un contexte idéal ............................................................................ 25
1.5 Terrains d’étude et modèles biologiques 25
1.5.1 Prairies et pelouses urbaines à Rennes.............................................................................................. 25
1.5.2 Les arthropodes comme indicateurs écologiques ................................................................................ 26
1.5.3 Acquisition des données ................................................................................................................... 27
1.6 Problématique et questions de recherche 29
1.7 Organisation de la thèse 31

I
2 Partie II : Effet de la température sur la diversité des arthropodes.................................................35
2.1 Chapitre 1 : Importance de l’échelle pour dissocier les effets de l'habitat et de la température sur les
communautés d'araignées urbaines 36
2.1.1 Introduction...................................................................................................................................... 38
2.1.2 Materials and methods...................................................................................................................... 40
[Link] Study area and sampling design 40
[Link] Community data collection 41
[Link] Landscape predictors 42
[Link] Local-scale predictors 42
[Link] Statistical analyses 43
2.1.3 Results ............................................................................................................................................ 44
[Link] Sampling results 44
[Link] Effects of habitat and temperature on spider communities 44
2.1.4 Discussion ....................................................................................................................................... 49
[Link] Effects of temperature at the landscape and local scales 49
[Link] Effect of habitat at the local scale 51
[Link] Implications for urban planning 52
2.1.5 Appendices...................................................................................................................................... 54
2.2 Chapitre 2 : Effets des facteurs liés à l’habitat et à la température sur les communautés de
staphylins et de carabes en ville 64
2.2.1 Introduction...................................................................................................................................... 66
2.2.2 Material and methods ....................................................................................................................... 67
[Link] Sampling sites 67
[Link] Arthropod sampling 68
[Link] Community indices 68
[Link] Environmental variables 68
[Link] Data analysis 69
2.2.3 Results ............................................................................................................................................ 70
[Link] Community composition 70
[Link] Density and diversity indices 70
2.2.4 Discussion ....................................................................................................................................... 72
[Link] Effect of habitat 72
[Link] Absence of temperature effect 73
2.2.5 Appendices...................................................................................................................................... 75

3 Partie III : Mécanismes expliquant les liens entre diversité des arthropodes et température ...........85
3.1 Chapitre 3 : Filtres environnementaux et composition fonctionnelle des communautés 86
3.1.1 Introduction...................................................................................................................................... 88
3.1.2 Materials and Methods...................................................................................................................... 90
[Link] Study area and sampling sites 90
[Link] Spider sampling 91
[Link] Functional trait selection 92
[Link] Environmental variables 92
[Link] Data analyses 94
II
3.1.3 Results ............................................................................................................................................ 95
[Link] Sampling results and urbanization gradient 95
[Link] Spider community composition and indicator species along an urbanization gradient 96
[Link] Relationships between functional traits and environmental conditions 100
[Link] Characterization of changes in functional trait community indices 101
3.1.4 Discussion ..................................................................................................................................... 103
[Link] Spider community composition and indicator species along an urbanization gradient 103
[Link] Relationships between functional traits and environmental conditions 104
[Link] Changes in functional trait community indices 106
3.1.5 Appendices.................................................................................................................................... 109
3.2 Chapitre 4 : Les traits d’histoire de vie modulent la contrainte exercée par le climat urbain sur la
taille corporelle 114
3.2.1 Introduction.................................................................................................................................... 116
3.2.2 Materials & Methods ....................................................................................................................... 117
[Link] Study area and sampling design 117
[Link] Species selection 119
[Link] Morphological measurements 121
[Link] Explanatory variables 122
[Link] Data analysis 123
3.2.3 Results .......................................................................................................................................... 124
[Link] Sampling results 124
[Link] Predictors of body size 124
3.2.4 Discussion ..................................................................................................................................... 126
3.2.5 Appendices.................................................................................................................................... 129

4 Partie IV : Importance des conditions microclimatiques à l’échelle des arthropodes ....................137

4.1 Chapitre 5 : Rôle de la thermorégulation comportementale en contexte d’îlots de chaleur urbains138
4.1.1 Introduction.................................................................................................................................... 140
4.1.2 Material and methods ..................................................................................................................... 143
[Link] Model species and sampling sites 143
[Link] Urban heat island (UHI) 144
[Link] Thermal preference 146
[Link] Data analysis 147
4.1.3 Results .......................................................................................................................................... 148
[Link] Sampled individuals 148
[Link] Period selection for Tpref parameters 149
[Link] Influence of endogenous factors and UHI on Tpref 149
4.1.4 Discussion ..................................................................................................................................... 152
[Link] Role of endogenous factors 152
[Link] Role of the urban heat island 153
4.1.5 Appendices.................................................................................................................................... 156

III
4.2 Chapitre 6 : Potentiel des prairies urbaines pour atténuer le réchauffement urbain à l’échelle des
arthropodes 162
4.2.1 Introduction.................................................................................................................................... 164
4.2.2 Material and methods ..................................................................................................................... 166
[Link] Study area and study period 166
[Link] Near-ground temperature data recording 168
[Link] Rural and urban reference temperature recording 168
[Link] Descriptors of the environment structure 169
[Link] Daily UHI characterization 169
[Link] Data analysis 169
4.2.3 Results .......................................................................................................................................... 171
[Link] Comparisons of daily minimal temperatures 171
[Link] Effect of the environment structure on temperature offset 171
4.2.4 Discussion ..................................................................................................................................... 176

5 Partie V : Discussion générale ..................................................................................................179

5.1 Expliquer les patrons observés à l’échelle des communautés 182
5.1.1 L’ICU atmosphérique filtre les espèces selon leurs traits fonctionnels .................................................. 182
5.1.2 La thermorégulation comportementale permet l’adaptation aux variations de température dans les micro-
habitats ......................................................................................................................................... 185
5.1.3 Une gestion extensive des prairies urbaines favorise le maintien de la diversité ................................... 186
5.1.4 Des micro-habitats complexes favorisent le maintien de la diversité fonctionnelle ................................. 190
5.2 Perspectives de recherche 192
5.2.1 Approfondir les connaissances climatologiques à l’échelle des micro-habitats...................................... 192
5.2.2 Valider l’hypothèse d’inertie comportementale................................................................................... 193
5.2.3 Dissocier plasticité phénotypique et adaptations évolutives ................................................................ 193
5.2.4 Evaluer l’impact du réchauffement urbain sur les interactions biotiques ............................................... 195

6 Références ..............................................................................................................................199

IV
 Liste des figures
Figure 1.1 - Croissance de la population mondiale vivant en milieu rural et urbain. .......................................... 2
Figure 1.2 - Illustration des différences de température formant deux types d’ICU : l’ICU atmosphérique et l’ICU de
surface. ............................................................................................................................................. 5
Figure 1.3 - Carte schématique illustrant les caractéristiques typiques des isothermes d'un ICU atmosphérique
nocturne, et leur correspondance avec le degré de développement urbain. .............................................. 6
Figure 1.4 - Courbe exposition-réponse de la mortalité liée à la température. ................................................... 8
Figure 1.5 - Impact du réchauffement nocturne sur l'activité des organismes ectothermes. ................................ 9
Figure 1.6 - La répartition des espèces est généralement déterminée par le climat. ........................................ 10
Figure 1.7 - Comparaison des caractéristiques de trois gradients thermiques spatiaux.. .................................. 14
Figure 1.8 - Composants biophysiques compris dans l’écosystème urbain.. ................................................... 15
Figure 1.9 - Réseau hydrographique rennais et emplacement de la zone humide des Prairies-Saint-Martin proche
du centre-ville. .................................................................................................................................. 18
Figure 1.10 - ICU mesuré lors de la canicule de 2003 (du 7 au 11 août) à Rennes.. ........................................ 20
Figure 1.11 - Exemples d’une station météorologique et d’un capteur connecté à Rennes. .............................. 21
Figure 1.12 - Carte de la localisation des capteurs connectés à Rennes en fonction de la classification des LCZ.
........................................................................................................................................................ 22
Figure 1.13 - Cartes mensuelles moyennes de l’intensité de l’ICU en 2022.. .................................................. 23
Figure 1.14 - Températures moyennes annuelles (en °C) à Rennes.. ............................................................ 24
Figure 1.15 - Organisation des parties et chapitres de la thèse ..................................................................... 33
Figure 1.16 - Coupe schématique d’un piège Barber.................................................................................... 28
Figure 1.17 - Piège Barber enterré et muni d’un abri transparent. Une sonde de température est disposée à
proximité directe du piège, afin de mesurer la température du micro-habitat, 5 cm au-dessus de la surface du
sol.................................................................................................................................................... 28
Figure 2.1 - Map of the study area.. ............................................................................................................ 41
Figure 2.2 - Distribution of landscape and local predictors among four groups: landscape habitat, local habitat,
atmospheric UHI, and near-ground temperature. .................................................................................. 43
Figure 2.3 - Plots showing significant relationships between the number of trapped individuals, species diversity or
functional diversity of spiders and predictors related to near-ground temperature, atmospheric UHI and local
habitat. ............................................................................................................................................. 47
Figure 2.4 - Relative variance explained independently by landscape- and local-scale predictors according to GLMM
models for spiders’ number of trapped individuals, taxonomic diversity and functional diversity indices. ..... 48
Figure 2.5 - Results of NMDS ordination (Bray–Curtis distance) based on staphylinid and carabid datasets.. .... 70
Figure 2.6 - Venn diagrams of the activity density, the taxonomic diversity, and the functional diversity of staphylinids
and carabids.. ................................................................................................................................... 71
Figure 2.7 - Significant relationship between the functional diversity of staphylinids and the vegetation height.. . 72
Figure 3.1 - Location of the 36 sampling sites within and around Rennes.. ..................................................... 91
Figure 3.2 - Mean scaled values and standard deviations of environmental variables after classification of sampling
sites into three clusters by hierarchical clustering. ................................................................................ 96

V
Figure 3.3 - Results of NMDS ordination based on 85 species (Bray-Curtis distance) and the four significant
environmental variables. .................................................................................................................... 97
Figure 3.4 - Community-averaged body size in sites belonging to three categories identified by hierarchical
classification.. ................................................................................................................................. 100
Figure 3.5 - Community-averaged thermal affinity based on mean, minimum and maximum temperature data, in
sites belonging to three clusters identified by the hierarchical classification.. ......................................... 101
Figure 3.6 - Absolute and relative species richness. Upper charts display the total number of species, the number
of species belonging to the three body size classes and the number of species belonging to three classes of
thermal affinity. Below charts display the relative species richness belonging to the three body classes and
relative species richness belonging to three classes of thermal affinity.. ................................................ 102
Figure 3.7 - Map of Rennes and surrounding area.. ................................................................................... 106
Figure 3.8 - Map of the study area.. .......................................................................................................... 118
Figure 3.9 - Map of the study area.. .......................................................................................................... 121
Figure 3.10 - Relationships between P. prativaga and P. pullata body size and temperature variables.. .......... 126
Figure 4.1 - Conceptual hypothesis framework displaying predicted responses on thermal preferences according
to the urban heat island intensity (UHI).. ............................................................................................ 143
Figure 4.2 - Location of the 10 sampling sites within the city borders of Rennes and surroundings according to the
spatialized UHI mean intensity obtained from the sensor network during A) the experiment period, B) the entire
period P. saltans individuals have spent as subadults or adults before and during the experimentation, C) the
winter period preceding the experiment. ............................................................................................ 145
Figure 4.3 - A) Variation of the mean Tpref of 522 individuals of Pardosa saltans over 90 minutes in an experimental
thermal gradient, B) Histogram displaying the global distribution of Tpref values within the 31-90 min time
interval considered in further analysis... ............................................................................................. 149
Figure 4.4 - Graphs representing significant relationships between Tpref parameters (mean Tpref and Tpref range)
and the tested endogenous and environmental factors.. ...................................................................... 151
Figure 4.5 - Map of the study area.. .......................................................................................................... 167
Figure 4.6 - Histograms displaying averaged temperature offset values of daily minimum temperatures obtained on
39 ‘urban reference – rural reference’, ‘near-ground – rural reference’ and ‘near-ground – urban reference’ site
comparisons. .................................................................................................................................. 171
Figure 4.7 - Relationships between temperature offset and mean tree height calculated within 10-m and 20-m radius
buffers. ........................................................................................................................................... 173
Figure 4.8 - Relationships between temperature offset and mean building height calculated within 10-m, 20-m and
50-m radius buffers.......................................................................................................................... 174
Figure 4.9 - Relationships between temperature offset and mean herbaceous vegetation height (left plots) and
relationships between temperature offset and cover of herbaceous vegetation (right plots)..................... 175
Figure 5.1 - Synthèse des principaux résultats issus des différents chapitres de la thèse .............................. 181
Figure 5.2 – Mécanismes expliquant la relation négative entre le nombre d’individus piégés et le réchauffement du
micro-habitat. .................................................................................................................................. 186

VI
Liste des tableaux
Table 1.1 - Liste des causes potentielles des îlots de chaleur dans la couche canopée urbaine…………………… 3
Table 1.2 - Les différentes échelles du climat, telles que définies par plusieurs auteurs en fonction du contexte
d’étude……………………………………………………………………………………………………………………4
Table 2.1 - Relationships estimated by GLMMs (estimate ± standard error) between the number of trapped
individuals, species diversity orders or functional diversity orders of spiders, and predictors related to near-
ground temperature, landscape habitat measured within 100-m-radius buffers, landscape temperature, local
habitat or local temperature…………………………………………………………………………………………..46
Table 3.1 - Environmental variables used to characterize sampling sites……………………………………………………..93
Table 3.2 - List of the 19 most abundant species occurring in each cluster…………………………………………..98
Table 3.3 - Responses of the 29 most abundant species to environmental variables, tested by GLM…….………99
Table 3.4 - Significant responses of absolute species richness to environmental variables within classes of species
sorted by body size and thermal affinity……………………………………………………………………………103
Table 3.5 - Summary of the characteristics of the 11 selected species………………………………….…………..120
Table 3.6 - Effects of temperature variables (‘UHI’ and ‘near-ground temperature’), intraspecific concurrence, ‘sex’
(two levels: female/male) and ‘generation’ (two levels: G1/G2) and their interactions, on body size estimated
by GLMMs…………………………………………………………………………………………………………….125
Table 4.1 - Summary of individual parameters for two development stages (adults and juveniles) in males and
females of Pardosa saltans………………………………………………………………………………….………148
Table 4.2 - Effects of endogenous (development stage, sex, prosoma length), environmental factors (UHI intensity)
and their interactions on mean Tpref and Tpref range estimated by generalized linear mixed models
(GLMM)………………………………….…………………………………………………………………………….150
Table 4.3 - Effects of environmental variables measured within 2-m-radius buffers (herbaceous canopy density and
height) and within 10, 20 and 50-m-radius buffers (height of trees and buildings) as well as the effect of daily
UHI intensity and its interaction with each previous variable on the daily offsets measured between daily
minimal near-ground temperature and daily minimal atmospheric temperatures, estimated by
GLMMs………………………………………………………………………………………………………………..172

VII
 Liste des annexes
Appendix 2.1 - Sampling site selection procedure ........................................................................................ 54
Appendix 2.2 - Species list of spiders and related traits based on literature data. ............................................ 56
Appendix 2.3 - Matrix of correlations among predictors. ............................................................................... 61
Appendix 2.4 - Relationships estimated by GLMMs between number of trapped individuals, species diversity
orders or functional diversity orders of spiders, and predictors related to near-ground temperature, landscape
habitat measured within 200-m-radius buffers, landscape temperature, local habitat or local temperature. . 62
Appendix 2.5 - Relationships estimated by GLMMs between number of trapped individuals, species diversity
orders or functional diversity orders of spiders, and predictors related to near-ground temperature, landscape
habitat measured within 500-m-radius buffers, landscape temperature, local habitat or local temperature. . 63
Appendix 2.6 - Mean scaled values and standard deviations of environmental variables after classification of
sampling sites into three clusters by hierarchical clustering. Clustering was performed based on the two and
three variables measured at the landscape and local scale, respectively. ............................................... 75
Appendix 2.7 - List of staphylinid species and corresponding trait information relative to diet and body size (given
in millimetres). The number of caught individuals is indicated in the column ‘N’. ...................................... 76
Appendix 2.8 - List of carabid species and corresponding trait information relative to diet and body size (given in
millimetres). The number of caught individuals is indicated in the column ‘N’. .......................................... 80
Appendix 2.9 - Matrix of correlations among predictors. Variables describing the proportion of impervious surface
are indicated by ‘imp_’, followed by the radius of the buffer considered (in meters). ................................. 82
Appendix 2.10 - Venn diagrams of the activity density, the taxonomic diversity, and (c) the functional diversity of
staphylinids and carabids. .................................................................................................................. 83
Appendix 3.1 - Species list of spiders and related traits based on literature data. .......................................... 109
Appendix 3.2 - Distribution of the number of females and males collected of the eleven selected species
according to the sampling month....................................................................................................... 129
Appendix 3.3 - Results of correlation tests between prosoma width and three other measured morphological
attributes (prosoma length, femur length and tibia length).................................................................... 130
Appendix 3.4 - Summary of the individual parameters of the 11 selected species after excluding outliers, arranged
by sex and generation/cohort.. .......................................................................................................... 132
Appendix 3.5 - Differences in body size between male and female individuals as well as differences between
body size of individuals from generations (G) or cohorts (C). ............................................................... 134
Appendix 4.1 - Laboratory gradient and corresponding thermal image. ........................................................ 156
Appendix 4.2 - Calibration curves related to gradients A) and B), obtained by temperatures measured every
centimetre with thermocouples as a function of the distance to the warmer point. .................................. 157
Appendix 4.3 - Significant negative relationship between Tpref range of juveniles P. saltans and three UHI
variables measured on distinct time intervals. .................................................................................... 157
Appendix 4.4 - Effects of time interval on mean Tpref values estimated by generalized linear mixed models
(GLMM). Results of four models are shown, differing in range of time intervals included as fixed effects.
Significant results (P<0.05) are displayed in bold. ............................................................................... 158
Appendix 4.5 - Results of GLMM post hoc testing all possible combinations of interacting factors ‘sex’ and
‘developmental stage’. Post hoc tests were performed separately on models related to mean Tpref and Tpref
range. Significant results (P<0.05) are in bold. ................................................................................... 159
Appendix 4.6 - Effects of UHI intensity on Tpref range estimated by linear models. Results of two models are
shown, computed on two separated datasets including respectively only adult individuals or juveniles.
Significant results (P<0.05) are displayed in bold. ............................................................................... 159
Appendix 4.7 - Effects of ontogenetic (dev. stage, sex, body size and leg length) and environmental factors (UHI
intensity) on juveniles’ Tpref range estimated by generalized linear mixed models (GLMM).. .................. 160

VIII
IX
X
 Introduction générale

1 Partie I : Introduction générale

1.1 Réchauffement climatique et activités humaines

L'augmentation de la température due au changement climatique d’origine anthropique est
largement reconnu comme affectant les écosystèmes (Lenoir and Svenning 2015, Scheffers et al.
2016, IPCC 2021). Parmi ceux-ci, les écosystèmes urbains sont particulièrement exposés au
réchauffement en raison du phénomène d'îlot de chaleur urbain (ICU), l'un des exemples les plus
évidents de modification locale du climat induite par l'homme (Oke et al. 2017). Si les effets néfastes
de ces réchauffements locaux sur la santé humaine ont été mis en évidence par de nombreuses
études (Piracha and Chaudhary 2022), les ICU sont également susceptibles d’impacter la biodiversité
au sein des villes (Youngsteadt et al. 2015, Kaiser et al. 2016, Diamond and Martin 2021). Dans un
contexte de réchauffement global amplifié par le réchauffement urbain, les études menées au cours
de cette thèse visent à déterminer dans quelle mesure la faune urbaine est affectée par les variations
spatiales de température à l’échelle de la ville de Rennes.

1.1.1 Urbanisation
Aujourd’hui plus de 50% de la population humaine vit et travaille dans un environnement urbain
(ONU 2018). Le terme « urbanisation » désigne à la fois le phénomène d’artificialisation des terres
agricoles et naturelles, mais aussi le processus socio-économique conduisant à la concentration des
populations humaines dans les zones urbanisées. Les activités professionnelles y sont
majoritairement composées d’industries et de services, ce qui contraste avec les zones rurales où le
secteur de l’agriculture concentre la majorité des emplois. Cette profonde transition dans les modes
de vie des humains s'est produite très rapidement et de manière exponentielle au cours du siècle
dernier (Figure 1.1). En 1900, 16 villes majeures comptant plus d’un million d’habitants étaient
recensées autours du globe (principalement situées en Europe et dans l'est de l'Amérique du Nord)
alors qu’en 2000, ce chiffre s’élevait à près de 300, dont certaines mégalopoles comptaient déjà plus
de 10 millions d'habitants (Modelski 2003). A l’heure actuelle, la majorité des citadins vivent dans
des agglomérations plus modestes de moins d'un demi-million d'habitants (Oke et al. 2017).

1
PARTIE I

Figure 1.1 - Croissance de la population mondiale vivant en milieu rural et urbain. (a) Population mondiale
totale. (b) Pourcentage de la population mondiale. Adapté de Oke et al. (2017).

Malgré ce développement récent, la couverture urbaine mondiale représenterait entre 2 et 3 % de

la surface continentale de la planète, selon certaines estimations (Schneider et al. 2009). Malgré
cette proportion relativement peu élevée en termes de surface, les zones urbanisées sont associées
à une utilisation intensive des ressources et génèrent plus de 50 % des émissions contribuant au
changement climatique mondial d'origine anthropique (Oke et al. 2017). Plusieurs décennies de
recherches en climatologie ont montré que le développement des villes (associé à une destruction
des habitats naturels et au développement d’infrastructures) a entraîné des changements marqués
dans les paysages thermiques (Mills 2014). L’urbanisation (au sens artificialisation des terres) et la
dé-végétalisation ont localement généré des augmentations de températures au sein des habitats
terrestres et aquatiques (Nelson and Palmer 2007, Grimm et al. 2008). Ce phénomène, connu sous
le nom d'îlot de chaleur urbain (ICU), est l'un des exemples les plus évidents de modification
involontaire du climat par l'homme (Oke et al. 2017).

1.1.2 Effets de l’urbanisation sur le climat

La variabilité spatio-temporelle du climat urbain est le résultat d'une très forte hétérogénéité de
l'espace urbanisé qui contient des surfaces horizontales et verticales qui modifient les
caractéristiques physiques de la couche limite atmosphérique (la couche proche de la surface). Ces
effets climatiques sont principalement thermiques, bien que des éléments climatiques tels que le
vent, les précipitations et l'humidité atmosphérique soient également affectés (Sturman and Quénol
2023). Deux caractéristiques des surfaces horizontales et verticales urbaines modifient en particulier
les conditions atmosphériques : les propriétés de surface (nature et rugosité des matériaux) et de
géométrie (forme des surfaces) (Table 1.1). Plus particulièrement, la couche limite atmosphérique
urbaine peut-être affectée par :
1) Les matériaux de construction qui composent les éléments urbains, tels que les bâtiments,
les routes ou encore la végétation déterminent les propriétés radiatives, thermiques et
hygrométriques d'une surface et donc sa capacité à absorber, réfléchir et émettre des radiations, et
ainsi transfèrent ou retiennent la chaleur et l'eau.

2
 Introduction générale

2) Les proportions de recouvrement de surface occupées par différents matériaux (par

exemple les surfaces bâties, pavées, végétalisées ou aquatiques) sont particulièrement importantes
pour la répartition spatiale de la chaleur.
3) La configuration tridimensionnelle des éléments urbains joue également un rôle
important. Par exemple, les dimensions des bâtiments et des espaces qui les séparent (largeur et
espacement des rues) contribuent à déterminer la rugosité aérodynamique de l’air à l’échelle d’une
ville complète, et contrôlent les échanges radiatifs et de circulation de l'air à l’échelle d’une rue (Oke
et al. 2017).

De plus, les activités humaines émettent de la chaleur, ce qui peut impacter le climat de manière
directe. On compte par exemple parmi ces impacts le réchauffement de l'air à proximité des maisons
climatisées. Les activités humaines rejettent également des polluants atmosphériques sous forme
gazeuse ou particulaire. Les impacts indirects comprennent les processus par lesquels les polluants
atmosphériques interfèrent avec le transfert radiatif dans l'atmosphère, ou encore les gaz à effet de
serre qui modifient le bilan radiatif de la planète (Oke et al. 2017).

Table 1.1 - Liste des causes potentielles des îlots de chaleur dans la couche canopée urbaine. Adapté de Oke et
al. (2017).

Cause Description de la cause

Géométrie des surfaces (a) Augmentation de la surface.

(b) Bâtiments peu espacés :
- Réflexion multiple et absorption plus importante des ondes courtes.
- Le faible facteur de visibilité du ciel (Sky View Factor) réduit la perte
nette d'ondes longues, en particulier la nuit.
- La réduction du vent réduit les pertes de chaleur par convection et
advection.

Les matériaux de construction ont souvent une plus grande capacité à

Propriétés thermiques
stocker et à restituer ultérieurement la chaleur sensible.
(a) L'imperméabilisation des bâtiments et des pavés réduit l'humidité
Etat de surface
du sol et des surfaces.
(b) La convection favorise la densité du flux de chaleur sensible par
rapport à la densité du flux de chaleur latente.

Le dégagement de chaleur anthropique dû à la combustion de

Chaleur anthropique combustibles et à la consommation d'électricité est beaucoup plus
important dans les villes.

L'atmosphère urbaine plus chaude, polluée et souvent plus humide

Effet de serre urbain émet davantage de rayonnement de grande longueur d'onde vers la
couche de canopée urbaine.

3
PARTIE I

1.1.3 Genèse de l’îlot de chaleur urbain

L'expression la plus concrète de la modification des échanges d'énergie en surface est la création de
l'îlot de chaleur urbain. Les températures en ville sont souvent plus élevées de plusieurs degrés que
dans les zones rurales environnantes. L’ICU est communément mesuré par l’écart entre la
température en ville et la température en zone rurale avoisinante (Oke 1995). L’étude de différentes
composantes du milieu urbain permet d’étudier les variations spatiales de l’ICU à différentes échelles
spatiales imbriquées (Oke 1995), notamment l’échelle microclimatique et l’échelle locale, telles que
classiquement définies en climatologies (Table 1.2).

Table 1.2 - Les différentes échelles du climat, telles que définies par plusieurs auteurs en fonction du contexte
d’étude. Adapté de Bramer et al. (2018).
Geiger et al. Barry and Orlanski (1975),
Littmann (2008)
(2009) Blanken (2016) WMO (2010)
Microclimat :
< 50 m (défini par
Échelle horizontale 0.001 à 100 m la hauteur de la 10 à 100 m² <100 m
canopée)
< quelques
Échelle verticale -10 à +10 m centaines de
mètres
Échelle temporelle <10s <Minutes
Climat local :
Échelle horizontale 100 m à 10 km 100 m à 10 km 100 m à 2 km 100 m à 3 km
Échelle verticale 5 m à 1 km 500 m à 1,5 km
Échelle temporelle 10 s à 1 h Minutes - heures
Mesoclimat :
Échelle horizontale 1 km à 200 km 10 km à 50 km 3 à 100 km
Échelle verticale 500 m à 4 km
Échelle temporelle Heures - jours Heures - jours
Marcoclimat :
Échelle horizontale >200 km >50 km 100 à 3000 km
Échelle verticale 1 à 10 km
Échelle temporelle Jours - semaines > heures

A l’échelle microclimatique, l’ICU de surface désigne les différences entre (1) la température à
l'interface entre la couche limite atmosphérique et les matériaux solides urbains et (2) la
température à l'interface air-sol équivalente en milieu rural (Figure 1.2). Cette mesure, informant sur
le différentiel des températures de surface, est principalement réalisée à l’aide de données issues
d’imagerie thermique. L’utilisation de capteurs de température au sol, mesurant la température de

4
 Introduction générale

l’air proche de la surface, permet quant à elle la mesure de l’ICU proche du sol. La résolution spatiale
de ce type d’ICU est de l’ordre du centimètre au mètre.

A l’échelle du climat local l’ICU de la couche limite de canopée urbaine désigne la différence entre la
température de l'air contenue dans la couche de canopée urbaine (c’est-à-dire la couche entre la
surface du sol et le niveau des toits, et la hauteur équivalente dans la couche atmosphérique en
milieu rural (Figure 1.2) (N.B. : pour plus de clarté, nous désignerons « l’ICU de la couche de canopée »
par « ICU atmosphérique » ou « atmospheric UHI » tout au long de ce manuscrit). La résolution
spatiale de ce type d’ICU est de l’ordre de la centaine de mètres.

Figure 1.2 - Illustration des différences de température formant deux types d’ICU : l’ICU atmosphérique et l’ICU
de surface. Adapté de (Oke et al. 2017).

L'ICU atmosphérique est le phénomène de réchauffement urbain le plus couramment étudié.

L'impact de l'urbanisation sur la formation de l'îlot de chaleur urbain a donné lieu à de nombreuses
études dans de grandes agglomérations comme Athènes (Katsoulis and Theoharatos 1985), Delhi
(Mohan et al. 2009), Melbourne (Morris and Simmonds 2000), New York (Gaffin et al. 2008), et Tokyo
(Hung et al. 2006). Dans les plus grandes villes, l'îlot de chaleur urbain peut s'étendre au-delà de la
périphérie sur plusieurs dizaines de kilomètres. Sur le plan temporel, l'heure du jour et les conditions
météorologiques (vent, nuages) influencent fortement l'ICU atmosphérique. A l’échelle d’une
journée, la différence de température entre zone urbaine et zone rurale est plus importante dès le
coucher de soleil. Ces écarts de température particulièrement remarquables en situations nocturnes
correspondent à un rafraîchissement plus rapide et plus important de l’air en campagne qu’en ville.

5
PARTIE I

Le terme "îlot de chaleur" a émergé en raison de la similarité entre les courbes isothermes de
l’intensité de l’ICU à l’échelle d’une ville, et les contours de hauteur d'une île océanique (Oke et al.
2017). Dans le cas des ICU atmosphériques, cette configuration « en îlot » ne décrit cependant que
les situations nocturnes (Figure 1.3). L’analogie entre phénomène climatique du réchauffement
urbain et topographie s’est également étendue aux « falaises » désignant les changements brusques
d’intensité d’ICU entre zone rurale et urbaine. De plus, un « pic urbain » désigne souvent un centre-
ville densifié où le différentiel de température est le plus élevé. Les zones de « collines » sont
assimilées à des zones particulièrement densément urbanisées disséminées dans différents
quartiers, où l’ICU atteint des intensités intermédiaires. Enfin les « vallées » ou « îlots de fraicheur »
désignent les zones faiblement densifiées telles que les espaces verts. De fait, la distribution spatiale
de l’intensité de l’ICU atmosphérique et les contours de ses isothermes sont intrinsèquement liés à
l’agencement de la matrice d’occupation du sol ainsi que d’intensité de l’activité humaine.

Figure 1.3 - Carte schématique illustrant les caractéristiques typiques des isothermes d'un ICU atmosphérique
nocturne, et leur correspondance avec le degré de développement urbain. La situation représente un ICU par
temps calme et clair, dans une ville située sur un terrain relativement plat. Adapté de (Oke et al. 2017).

Les contrastes d’ICU atmosphériques entre zones urbanisées et zones rurales sont particulièrement
remarquables en situation de nuits claires et par absence de vent, c’est-à-dire lors des situations
permettant un refroidissement important de la couche de canopée urbaine par rayonnement, ainsi
qu’un faible mélange turbulent. Lorsque ces critères sont remplis, la distribution spatiale de l’ICU
atmosphérique répond principalement à des variables locales, et son intensité est accrue (Foissard
2015).

Plus généralement, l'îlot de chaleur urbain se caractérise non seulement par un gradient de
température entre le centre-ville et la périphérie, mais aussi par des variations souvent très
complexes du climat à l'intérieur des zones urbaines, créées par des différences dans l'utilisation des
sols (par exemple, bâtiments, routes, espaces verts, points d’eau, etc.). La prise en compte des

6
 Introduction générale

facteurs d’occupation du sol aux échelles micro-locales et locales (comme la présence de bâtiments,
la hauteur du bâti, la proximité d’un parc etc.), et à l’échelle macro (comme la proximité de l’océan
ou la typologie de la campagne avoisinante) détermine l’intensité de l’ICU atmosphérique dans le
contexte particulier d’une ville. De la même manière, les variations temporelles de l’intensité de l’ICU
sont directement liées aux facteurs climatiques agissant à court ou moyen terme (Cantat 2004).
Parmi ceux-ci, une nébulosité importante aura pour effet de réduire le flux d’énergie stockée en ville
durant la journée. La pluie modifie également le bilan énergétique en ville et réduit l’écart de
température entre la ville et la campagne. Par ailleurs, le vent équilibre les flux de chaleur entre ville
et campagne, ce qui a pour effet de diminuer les écarts de température caractérisant l’ICU. Une prise
en considération méticuleuse de l’ensemble de ces facteurs spatio-temporels permet l’estimation à
une résolution fine de l’ICU grâce à des modèles prédictifs.

1.2 Réponse des organismes aux réchauffements

Le scénario le plus optimiste défini par le GIEC1, reflétant une réduction maximale des émissions de
gaz à effet de serre (SSP1-1.9), prévoit une augmentation de 1,5°C à l’horizon fin 2100 (IPCC 2013).
Le scénario le plus pessimiste (SSP5-8.5), quant à lui, prédit une augmentation de plus de 5°C (IPCC
2013, O’Neill et al. 2016). En ville, le phénomène d’ICU se combine au réchauffement global pour in
fine amplifier les impacts potentiels sur les organismes (Angilletta 2009). Parmi les risques dus au
changement climatique identifiés par le GIEC (IPCC 2013), le risque de mortalité et de morbidité
humaine pendant les périodes de chaleur extrême est mentionné comme étant particulièrement
élevé pour les populations urbaines (Schär and Jendritzky 2004).

1.2.1 Impact sur les organismes endothermes

Parmi les organismes endothermes, c’est-à-dire les organismes dotés de la capacité de réguler leur
température corporelle de manière interne (Angilletta et al. 2010), les effets du réchauffement sur
les humains est au cœur d'une grande partie de la recherche sur le climat urbain (Masson et al. 2014,
Lemonsu et al. 2015, Salmond et al. 2018), dont une grande partie concerne l'application des
connaissances scientifiques à la conception des villes (de Munck et al. 2018, Broadbent et al. 2018).
L'humain possède une température corporelle interne quasi constante de 37°C, maintenue par le
système de thermorégulation du corps, qui assure l'homéostasie dans des conditions normales. Ce
système de thermorégulation interne, propre aux organismes endothermes, comprend un ensemble
de processus physiologiques qui gèrent les échanges de chaleur entre le corps et le milieu ambiant,
de manière à ce que le gain ou la perte nette d'énergie soit proche de zéro. Il s'agit d'un processus
complexe qui régule l'élimination de la chaleur générée par les processus métaboliques internes en
réponse aux conditions environnementales externes, y compris le rayonnement et les échanges de
chaleur sensible et latente (Oke et al. 2017).
En réponse à un stress thermique chaud, en tant qu’organisme endotherme, le corps humain doit
évacuer la chaleur pour que la production de chaleur interne soit maintenue à un niveau de base. Le
mécanisme le plus efficace pour faire face à des conditions chaudes est la transpiration et
l'évaporation régulatrices. Malgré ces mécanismes, des conditions de chaleur trop extrêmes

1 Groupe d'experts intergouvernemental sur l'évolution du climat

7
PARTIE I

provoquent une augmentation de la température interne, ce qui accroît le taux métabolique et la

chaleur interne générée. Lors d’une exposition trop longue à ces conditions, le système de
thermorégulation interne échoue, ce qui provoque une situation d'hyperthermie (Oke et al. 2017).
Les conditions d’apparition d'hyperthermie varient d'un individu à l'autre, notamment selon leur âge
et leur état de santé. Cependant, à l’échelle d’une population il existe une relation claire entre la
mortalité et la température de l'air (Figure 1.4). En dehors de certaines limites de température, la
mortalité est corrélée à l'ampleur de l'écart à ces limites (Li et al. 2013b). Cependant, ces limites
varient selon l’origine des populations étudiées. Par exemple, les populations européennes se sont
adaptées avec succès à des températures estivales moyennes de 16°C au nord de la Finlande et 24°C
en Grèce (Keatinge et al. 2000).

Figure 1.4 - Courbe exposition-réponse de la mortalité liée à la température. La ligne continue montre les
estimations centrales. La ligne en pointillé montre les intervalles de confiance à 95 %. Adapté de Li et al. (2013b).

1.2.2 Les ectothermes : des organismes particulièrement vulnérables

Bien que les humains, tout comme l’ensemble des organismes endothermes, soient impactés par les
variations de température, d’autres organismes sont largement considérés comme étant, plus
encore, vulnérables face aux variations de température associées au réchauffement global : les
ectothermes. Chez les ectothermes, la température corporelle n’est pas régulée de manière interne
comme chez les endothermes, et reste de fait proche de la température de leur environnement
immédiat. Les ectothermes constituent la grande majorité de la biomasse des organismes et environ
99 % de toutes les espèces dans le monde (Atkinson and Sibly 1997). Il est donc particulièrement
important de comprendre comment les ectothermes réagissent aux changements climatiques. Chez
les ectothermes, un réchauffement de l’environnement est généralement associé à des taux
métaboliques plus élevés, qui peuvent conduire à une augmentation des coûts physiologiques (Irwin
& Lee, 2003 ; Williams et al., 2012). Jusqu'à une certaine température, dite optimale, le
réchauffement stimule la croissance, le développement et la reproduction des ectothermes. Au-delà,
les coûts métaboliques augmentent très rapidement et des lésions thermiques se créent, ce qui
entraîne un ralentissement du développement, une fécondité réduite, une capacité de dispersion

8
 Introduction générale

amoindrie et, en pour finir, une diminution de la condition physique et une augmentation de la
mortalité (Gilbert and Raworth 1996).
Le réchauffement nocturne causé par les ICU atmosphériques classiquement mesurés en ville est
caractérisé par une élévation des températures minimum journalières. A défaut d’élever les
températures environnementales maximales, ce type de réchauffement peut induire différentes
réponses des organismes ectothermes. Par exemple, les températures corporelles peuvent être
maintenues dans une gamme thermique nocturne suffisamment élevée pour améliorer les
performances (en élevant les taux métaboliques et de développement), et pour étendre la fenêtre
temporelle d'activité quotidienne de recherche de nourriture (Rutschmann et al. 2024). En revanche,
le réchauffement nocturne est également susceptible d’affecter la gamme thermique de
récupération physiologique, notamment chez les organismes diurnes. Des températures minimales
plus chaudes peuvent déclencher une augmentation des taux physiologiques (par exemple les taux
métaboliques de repos ou encore le taux de réparation des dommages cellulaires) et une réduction
du temps alloué au repos (Figure 1.5), pouvant résulter en une augmentation de la consommation
d'énergie (Speights et al. 2017, Ma et al. 2020) et affecter in fine négativement les performances.

Figure 1.5 - Impact du réchauffement nocturne sur l'activité des organismes ectothermes. Avec des
températures nocturnes plus élevées, les températures opératives sont susceptibles de dépasser le seuil
minimum de température nécessaire à l’activité (VTmin ; ligne horizontale en pointillés) plus tard dans la nuit
et plus tôt dans la matinée (flèches noires). Ce décalage temporel offre davantage de possibilités de recherche
de nourriture ou d'activité physiologique de nuit (surface jaune). Cette extension de l'activité se fait néanmoins
au détriment du temps de repos (surface bleue), affectant la qualité de la récupération physiologique. Adapté
de (Rutschmann et al. 2024).

Ces impacts physiologiques du réchauffement de l’environnement peuvent dès lors mener à des
adaptations aux nouvelles conditions locales, mais aussi à des changement biogéographiques
comme le décalage des aires de répartition, et des abondances au sein de ces aires, ou encore dans
les cas les plus extrêmes, à des extinctions locales, voir globales (Hof et al. 2011, Abram et al. 2017,
Harvey et al. 2022) (Figure 1.6).

9
PARTIE I

Figure 1.6 - La répartition des espèces est généralement déterminée par le climat (A). Elles suivent le
changement climatique (flèches rouges) si les populations peuvent se disperser et s'établir dans de nouveaux
habitats appropriés, et disparaissent là où le climat est devenu inadapté (B). Les espèces peuvent être menacées
d'extinction si la taille de leur habitat diminue (par exemple, aux pôles ou au sommet des montagnes) (C), ou si
des barrières migratoires (D) ou une capacité de dispersion limitée (E) les empêchent d'atteindre un nouvel
habitat approprié. La capacité des espèces à s'adapter (ou à modifier leur comportement), les interactions entre
les espèces et d'autres facteurs de stress liés au changement climatique représentent des incertitudes majeures
(F) qui affectent notre capacité à prédire la perte de biodiversité avec le changement climatique. Extrait de
Lambers (2015).

1.2.3 Décalages spatiaux

En réponse au changement climatique, l'aire de répartition géographique de nombreuses espèces
s'est déplacée soit vers les pôles, soit vers des altitudes plus élevées. Au cours des dernières
décennies, sur 35 espèces de papillons, 63 % ont déplacé l'ensemble de leur aire de répartition vers
le nord, alors que 3 % seulement se sont déplacés vers le sud (Parmesan et al. 1999). De plus,
Parmesan et Yohe (2003) recensent 80% des 434 espèces examinées à s’être historiquement
déplacées conformément aux prévisions basées sur le changement climatique. Une méta-analyse
mondiale a estimé que la vitesse des changements de distribution des espèces terrestres dans le
cadre du changement climatique s'est produite à un rythme médian de 1,69 km par an vers des
latitudes plus élevées dans l'hémisphère nord et à un rythme médian de 1,1 m par an vers des
altitudes plus élevées dans les écosystèmes montagneux (Chen et al. 2011). La contraction de l'aire
de répartition ou l'extinction globale est une issue probable pour de nombreuses espèces, en
particulier lorsque la taille du corps, le cycle de vie ou une interaction écologique limite la dispersion

10
 Introduction générale

(Figure 1.6c). Certaines espèces ont par exemple vu la partie sud de leur aire de répartition diminuer
(Franco et al. 2006, Angilletta 2009).
Par ailleurs, le changement climatique opère généralement plus rapidement que la possibilité des
espèces à adapter leur aire de répartition (Lenoir and Svenning 2015). Dans ce contexte, l’accès aux
micro-habitats (les habitats immédiats des individus) et aux microclimats associés joue un rôle
essentiel dans l’atténuation de l'exposition aux conditions macroclimatiques et aux épisodes de
chaleur extrême (Scheffers et al. 2014), ce qui peut ralentir les changements de répartition de
certaines espèces. Les changements de répartition sont également souvent hétérogènes sur les
gradients géographiques lorsque des facteurs autres que la température entraînent une
redistribution des espèces. Par exemple, les changements de précipitations ou les interactions
interspécifiques peuvent entraîner des déplacements vers des altitudes moins élevées, à mesure que
le climat se réchauffe (Lenoir et al. 2010).
En général, les espèces sont plus abondantes au centre de leur aire de répartition et déclinent
progressivement vers les marges de l'aire de répartition (Sagarin et al. 2006). De ce fait, les
changements de répartition en réponse au réchauffement climatique devraient se traduire par des
changements d'abondance au sein d'une localité. Plus précisément, les espèces adaptées aux
conditions plus froides devraient décliner, tandis que les espèces adaptées aux conditions chaudes
devraient augmenter (Barry et al. 1995). Sur ce principe, de nombreuses études mentionnent des
changements d'abondance, bien qu'aucun changement d’étendue globale de l'aire de répartition ne
soit observable (Lenoir and Svenning 2015). Les changements spatiaux d'abondance ne sont pas
strictement considérés comme des décalages de l’aire de répartition. Cependant, ils peuvent être
considérés comme des états intermédiaires dans un processus de déplacement en cours ou comme
des signes précurseurs d’adaptation d'aire de répartition des espèces (Maggini et al. 2011, Lenoir
and Svenning 2015). La détection des changements de densités d’individus à l'intérieur des aires de
répartition est donc un paramètre important à prendre en compte pour estimer l’impact du
réchauffement sur la distribution spatiale des espèces.

1.2.4 Décalages temporels

Aux changements spatiaux en réponse à l’élévation des températures, s’ajoutent les décalages
d’ordre temporel. Les décalages phénologiques ont été parmi les signes les plus évidents du
réchauffement climatique ont ainsi fait l’objet de nombreuses recherches. La hausse des
températures s'est accompagnée d'une apparition plus précoce de nombreux ectothermes (Beebee
1995, Forrest 2016, Gutiérrez and Wilson 2021). Des relevés à long terme ont notamment montré
que les premiers vols de papillons survenaient un mois plus tôt que durant les trois décennies
précédentes (Roy and Sparks 2000, Forister and Shapiro 2003). De nombreuses études font
également état d’un décalage des phénologies printanières vers des dates plus précoces, en lien avec
l'augmentation des températures (Diamond et al. 2011, Brooks et al. 2014, Ge et al. 2015). Dans une
méta-analyse mondiale de 143 études portant sur les amphibiens, les oiseaux, les invertébrés et les
plantes, Root et al. (2003) ont mis en évidence une avancée significative de l'activité vernale et font
état d’un taux moyen d'avancement phénologique des espèces de 5,1 jours par décennie. Dans une
analyse similaire portant sur 677 espèces, 62% des espèces ont montré une activité plus précoce au
printemps, comme la reproduction, la nidification, la floraison, la migration ou le bourgeonnement
(Parmesan and Yohe 2003). Dans cette dernière étude, le taux moyen d'avancement calculé est de

11
PARTIE I

2,3 jours par décennie. Le décalage est plus important chez les espèces des hautes latitudes que chez
les espèces des basses latitudes (Parmesan 2007). De même, la phénologie est davantage impactée
dans les environnements urbains que dans les environnements ruraux (Roetzer et al. 2000, White et
al. 2002), ce qui révèle les effets additifs du réchauffement global et urbain.

1.2.5 Adaptations locales et plasticité phénotypique

En outre, l'adaptation aux conditions environnementales locales joue un rôle fondamental dans le
maintien des espèces face aux changements environnementaux (Blanquart et al. 2013), en
influençant notamment la contraction ou l'expansion des aires de répartition géographiques
(Kirkpatrick and Barton 1997) ou leur phénologie (Parmesan and Yohe 2003). Les adaptations locales
résultent des interactions entre les génotypes et l'environnement. En l'absence d'autres forces et
contraintes que celles exercées par l’environnement, chaque population est susceptible de
développer des traits qui lui confèrent un avantage dans ses conditions environnementales propres.
Ainsi, la constitution génétique des populations locales doit procurer aux individus une meilleure
valeur adaptative (fitness) dans leur environnement propre, qu’aux individus venant d’autres
environnements (Kawecki and Ebert 2004). Par exemple, les marqueurs génétiques associés à la
tolérance à la chaleur chez Drosophila subobscura ont suivi le réchauffement climatique sur plusieurs
continents (Balanyá et al. 2006). Grâce à une expérience de transplantation réciproque visant à
comparer les réponses de fourmis en milieu rural et urbain, Martin et al. (2021) ont constaté que les
individus issus de populations rurales dans l'environnement urbain étaient plus vulnérables en été,
probablement en raison d'une faible tolérance à la chaleur. A l’inverse, les fourmis urbaines dans
l'environnement rural étaient plus vulnérables en hiver, probablement en raison d'une perte
évolutive de la tolérance au froid. Les auteurs ont également obtenu des résultats cohérents avec la
théorie de l'adaptation locale en étudiant la fécondité.
Les changements phénotypiques observés sur la base des traits peuvent effectivement résulter de
changements génétiques, mais aussi de plasticité phénotypique. Cette dernière se produit lorsque
les individus d'un génotype donné adaptent leur phénotype en fonction des conditions qu'ils
rencontrent. De nombreuses études ont mis en évidence le rôle de la plasticité dans l'élaboration
des réponses phénotypiques au changement climatique contemporain (Chown et al. 2007, Chevin
et al. 2013, Oostra et al. 2018). La plasticité permet aux organismes de s'adapter aux environnements
variables et hétérogènes (Stearns 1989). La plasticité comportementale est notamment largement
reconnue comme jouant un rôle important dans les adaptations thermiques (Abram et al. 2017).
Cette aptitude est particulièrement déterminante dans le contexte du changement climatique, dès
lors que certaines espèces peuvent diminuer leur risque d'extinction en ajustant leur phénotype de
manière plastique aux variations de température (Parmesan 2006).

1.2.6 Altération des interactions biotiques et des communautés

Les espèces n’étant pas impactées de manière homogène par les changements climatiques, les
décalages spatio-temporels (mentionnés dans les parties précédentes) peuvent conduire à de
nouvelles communautés biotiques (Gilman et al. 2010). Ces bouleversements peuvent avoir des
répercussions importantes sur les interactions entre espèces, notamment au sein des réseaux
trophiques. Sur 11 espèces étudiées, Visser and Both (2005) ont recensé une incompatibilité

12
 Introduction générale

phénologique (« phenological mismatch ») avec leur source de nourriture chez sept d’entre elles.
De plus, certaines espèces voyant leur aire de répartition s’étendre en réponse au changement
climatique, peuvent être attaquées par moins d'ennemis naturels et subir des niveaux de mortalité
réduits dans les zones nouvellement colonisées, ce qui a pour conséquence de favoriser leur
expansion (Menéndez et al. 2008). À leur tour, les interactions biotiques modifiées peuvent entraver
ou faciliter de nouveaux déplacements d’aires de répartition, pouvant entraîner un effet cascade
impliquant les dynamiques de prédation, l'herbivorie, les associations hôte-plante, les compétitions
et les mutualismes (Sorte et al. 2010, Cahill et al. 2013, Vergés et al. 2014). L'impact sur les
communautés d'un déplacement d'espèce dans le temps ou l’espace peut être comparable aux
effets produits par l'introduction d'espèces non indigènes (Sorte et al. 2010), ce phénomène étant
lui-même reconnu comme l'un des principaux moteurs de la perte de biodiversité (Sala et al. 2000).
Les conséquences pour la dynamique des communautés dépendent donc du fait que les espèces
envahissantes perturbent les interactions entre les espèces indigènes ou comblent le vide créé par
l'extinction.
La redistribution des espèces en réponse au changement climatique peut donc modifier la
composition des communautés dans l'espace et dans le temps (diversité bêta), le nombre d'espèces
cooccurrentes à un endroit donné (diversité alpha) et/ou le nombre d'espèces présentes dans une
région plus vaste (diversité gamma) (Pecl et al. 2017). La diversité et la composition des traits
fonctionnels au sein des communautés peuvent également être influencées (Buisson et al. 2013),
sans que toutefois des changements soient observés dans la richesse des espèces (diversité alpha)
(Pecl et al. 2017). Alors qu’il est largement accepté qu’un niveau élevé de diversité, tant sur le plan
taxonomique que fonctionnel, indique une stabilité du fonctionnement des écosystèmes aux
échelles locales et régionales (Oliver et al. 2015). Des modifications dans les communautés suite à
des redistributions des espèces ont également des conséquences sur l'état des écosystèmes (Pecl et
al. 2017).

1.3 Ecosystèmes urbains

1.3.1 La ville : un modèle idéal pour étudier les réponses des organismes au
changement climatique
Deux approches sont couramment utilisées pour prédire les effets du réchauffement sur les
organismes. D’une part, un réchauffement climatique expérimental peut être appliqué
instantanément en conditions contrôlées et sur des échelles de temps brèves, limitant le réalisme
biologique de ces simulations (Wolkovich et al. 2012). D’autre part, les substitutions du temps par
l’espace servent de substituts au réchauffement climatique futur (Hargrove and Pickering 1992).
L’espace est dès lors défini sur un territoire, souvent étendu, de manière à inclure un gradient de
température allant de conditions climatiques naturellement plus chaudes, c’est à dire dans des zones
de plus basse latitude ou de plus basse altitude, et inversement pour l’extrémité froide du gradient.
Ces systèmes peuvent cependant contenir des variables confondantes empêchant l’identification de
relations causales claires entre les variations climatiques et la physiologie des individus, la
persistance des populations ou encore la structure des communautés (Damgaard 2019).

13
PARTIE I

Les villes (et leurs îlots de chaleur urbains) sont de plus en plus utilisées comme modèles pour la
mise au point et la conduite d’expérimentations visant à étudier les réponses des organismes
ectothermes au réchauffement climatique (Youngsteadt et al. 2015, Diamond and Martin 2021). Les
villes imposent des réchauffements locaux aux écosystèmes (1.1.2 Effets de l’urbanisation sur le
climat). Il est possible de mesurer les températures environnementales le long de gradients urbains
pour tenir compte de l'hétérogénéité thermique dans les différentes parties des villes (Zhao et al.
2014b). Si au préalable des sites d’étude sont identifiés de manière à isoler au maximum l’effet du
climat parmi les variables confondantes (Szulkin et al. 2020) ou de les prendre en compte dans les
analyses, les écosystèmes urbains peuvent être utilisés comme des modèles efficaces pour
comprendre les relations entre température, physiologie et persistance/adaptation des populations
(Diamond and Martin 2020, 2021).
Les réchauffements locaux dus au phénomène d’îlots de chaleur urbains mesurés dans les villes sont
du même ordre de grandeur que les prédictions publiées par le GIEC d’ici 2100 à l’échelle globale
(Oke et al. 2017, IPCC 2021), faisant des villes des laboratoires à ciel ouvert pour étudier les effets
du changement climatique sur les organismes. De plus, les distances permettant de couvrir les
gradients de températures couvrant l’ensemble des scénarios du GIEC sont relativement faibles, en
comparaison avec les gradients de latitudes et altitudes classiques (Figure 1.7) (Verheyen et al. 2019).

Figure 1.7 - Comparaison des caractéristiques de trois gradients thermiques spatiaux. Les caractéristiques sont
décrites dans le sens de la flèche verticale lorsque l'on va de la région la plus froide vers la région la plus chaude.
Extrait de Verheyen et al. (2019).

14
 Introduction générale

1.3.2 Interactions du climat avec les autres composantes urbaines

Les écosystèmes urbains sont formés par l’ensemble de la biosphère établie en ville incluant la
végétation, les animaux ou les humains ainsi que leurs habitats. Les composants abiotiques des villes
quant à eux comptent aussi bien l'atmosphère, l'hydrosphère, la pédosphère (sols), la lithosphère
(géologie), ou encore le système construit (bâtiments, routes, mobilier urbain etc.) ( Figure 1.8). Ce
système construit est caractéristique de l’occupation des sols en milieu urbains et a de nombreux
impacts sur la biosphère. Parmi les impacts les plus visibles, on compte la destruction des habitats
pour la faune et la flore ou encore le morcellement de ces habitats (Liu et al. 2016). L’intensité de la
gestion dans les écosystèmes urbains entretenus (par exemple parcs, jardins, cimetières, golfs) et la
structure des habitats associés ont également un impact sur la biosphère que ces zones abritent
(Fontaine et al. 2016). Chaque écosystème urbain présente donc un ensemble d’habitats pour la
faune et la flore, intégré dans un agencement unique de variables abiotiques (Oke et al. 2017). Pour
étudier les effets du réchauffement urbain sur les organismes, il est donc primordial de quantifier
ces variables abiotiques urbaines, qui si elles ne sont pas considérées, peuvent agir comme des
variables confondantes. De plus, le développement du système construit provoque des « climate
feedbacks » qui se traduisent par les ICU (1.1.3 Genèse de l’îlot de chaleur urbain). Il est donc
essentiel de minimiser les covariations entre les variables abiotiques décrivant l’habitat et le
réchauffement par un choix minutieux des sites d’étude.

Figure 1.8 - Composants biophysiques compris dans l’écosystème urbain. Cela inclut tous les éléments de
l’environnement naturel préurbain, modifiés par l’introduction de l’infrastructure construite. Extrait de Oke et
al. (2017).

15
PARTIE I

1.3.3 Pertes d’habitat et de connectivité associées à l’urbanisation

La perte d'habitat fait généralement référence à la diminution de l'étendue spatiale de l'habitat
naturel, y compris les forêts, les prairies ou encore les zones humides (Fahrig 1997) sur un territoire
défini. La fragmentation quant à elle désigne une mesure de la configuration de l’habitat et est
souvent estimée grâce au nombre de parcelles d’habitat, à la taille de ces parcelles, ou encore à
l’isolement de ces parcelles sur un territoire défini (Fahrig 2003). La perte et la fragmentation de
l'habitat sont généralement concomitantes et interdépendantes, et influencent toutes deux
négativement la biodiversité et les processus écologiques (Fahrig 1997, Haddad et al. 2015, Beninde
et al. 2015, Fenoglio et al. 2020). L’urbanisation croissante (1.1.1 Urbanisation) est une cause
importante de perte et de fragmentation de l'habitat à l’échelle du globe (Swenson and Franklin
2000, Syphard et al. 2011). Au cours de l'urbanisation, les pertes d'habitat sont principalement
causées par la conversion de vastes zones naturelles en surfaces construites imperméables et
artificielles (Mcdonald et al. 2008, Seto et al. 2012). Dans le même temps, le développement des
infrastructures linéaires telles que les routes ou les voies ferrées accumulées aux surfaces
imperméables entraînent une fragmentation de l'habitat (Swenson and Franklin 2000, Scolozzi and
Geneletti 2012). Pour évaluer les impacts relatifs des différentes composantes abiotiques des villes
sur les organismes, y compris l’effet du réchauffement, il est essentiel de s’intéresser aux variables
décrivant l’habitat à l’échelle du paysage urbain, en estimant leur perte et leur fragmentation, donc
en mesurant le système construit (Figure 1.8).

1.3.4 Potentiels refuges pour la biodiversité

Malgré les effets négatifs de la perte d’habitat et de la fragmentation ayant inspiré l’image de
« jungles de béton » abritant peu d'espèces (McKinney 2002), les environnements urbains ont un
potentiel pour le maintien d’espèces à fort enjeux de conservation (Ives et al. 2016). Contrairement
aux paysages d’agriculture intensive, les villes sont composées d'une grande diversité d'habitats,
eux-mêmes de qualité, de taille et de connectivité diverses (Baldock 2020). Dans le cas de villes
situées dans des paysages agricoles intensifs, cette hétérogénéité de la matrice urbaine peut
favoriser des niveaux de biodiversité plus élevés que dans les zones adjacentes (Spotswood et al.
2021). Par exemple, dans une étude comparant les communautés de pollinisateurs dans plusieurs
type d’habitats urbains, les jardins domestiques, familiaux et communautaires de quatre villes du
Royaume-Uni (Bristol, Reading, Leeds et Edinburgh) ont été identifiés comme des « hotposts » de
diversité et d’abondance (Baldock et al. 2019). De plus en plus d’études mettent en évidence la
biodiversité que ces espaces recensent (Schwartz et al. 2002, Aronson et al. 2014, Ives et al. 2016,
Soanes and Lentini 2019, 2019, Knapp et al. 2021), ou encore l'importance des nombreux types
d’espaces verts jouant le rôle d’habitats (Aronson et al. 2017, Lepczyk et al. 2017, 2023). De plus,
des études ont montré que même des espaces verts relativement nouvellement créés, tels que des
délaissés urbains, peuvent abriter des espèces à fort enjeu de conservation (Planchuelo et al. 2020).

1.3.5 Services écosystémiques

Les services écosystémiques sont les avantages que l’être humain tire, directement ou
indirectement, des fonctions des écosystèmes. Dans une revue récente de la littérature, Schwarz et

16
 Introduction générale

al. (2017) identifient de nombreuses relations entre la biodiversité et les services écosystémique
qu’elle fournit en ville. Pami ceux-ci, les auteurs listent un certain nombre de services d’ordre culturel
comme la plus-value esthétique, le potentiel pour l’éducation, ou encore pour les loisirs ou le bien-
être et la santé, physique comme mentale. L’enjeu important de l’accès aux espaces verts en ville
pour le bien-être et la santé des citadins a notamment été révélé par la crise du covid 19 et les
mesures de confinement prises au début de la pandémie. Bien qu’ayant permis de sauver des
millions de vies (Flaxman et al. 2020), ces mesures ont suscité des inquiétudes quant à leur impact
sur la santé mentale des personnes en raison du stress, de l'anxiété et de la solitude (Karatzias et al.
2020, Holttum 2020, Galea et al. 2020). De nombreuses études ont montré que l'accès aux espaces
verts joue un rôle positif pour la santé et le bien-être des personnes (Houlden et al. 2018, Twohig-
Bennett and Jones 2018) et qu'il peut agir comme un tampon contre les expériences de vie
stressantes (van den Berg et al. 2010), tout particulièrement en période de confinement (Poortinga
et al. 2021). La biodiversité urbaine est également associée à l’approvisionnement en nourriture par
le biais de structures d’agriculture urbaine. Enfin, de nombreux services sont associés à différents
rôles de régulation joués par les organismes des villes. Parmi eux, on compte la séquestration du
carbone, le maintien de la fertilité, la réduction de l’érosion des sols ou encore la pollinisation.
Cependant, le service de régulation tout particulièrement mis en avant par l’étude de Schwarz et al.
(2017) est la régulation du climat local et notamment l’atténuation du climat local par la biodiversité
végétale. Dès lors, les plans d'action contre le changement climatique en ville par l’implémentation
d’infrastructures vertes à vocation de régulation du climat peuvent devenir une opportunité pour
l’intégration de mesures de conservation de la biodiversité et ainsi générer un co-bénéfice climat-
biodiversité (Butt et al. 2018).

1.4 L’ICU à Rennes

Le suivi des conditions climatiques est devenu un enjeu contemporain majeur, notamment en ville
où l’effet du changement climatique global est amplifié localement par l’ICU (Dubreuil et al., 2022).
Dans ce contexte, une connaissance fine de la distribution spatiale et temporelle de l’ICU peut aider
les collectivités locales à développer des mesures d’atténuation. En France, plusieurs villes moyennes
comme Rennes (Dubreuil et al. 2020), mais aussi Toulouse (Dumas et al. 2021), Dijon (Richard et al.
2021) ou encore Strasbourg ont été équipées de réseaux de mesures climatiques avec l’objectif
d’étudier l’ICU à fine échelle.

1.4.1 Contexte géographique et climatique

Cette thèse a été réalisée sur la ville de Rennes, ainsi que les municipalités voisines de Thorigné-
Fouillard, Cesson-Sévigné, Chantepie, Saint-Jacques-de-la-Lande, Vezin le coquet et Pacé. Avant de
se concentrer sur la caractérisation de l’ICU rennais, il est important de détailler le contexte de la
zone d’étude en termes de topographie, dynamique démographique et climat car l’ensemble de ces
variables influencent l’ICU. Rennes est la 11ème ville de France la plus peuplée avec une population
municipale légale en vigueur au 1er janvier 2024 de 225 081 habitants selon l’INSEE. La ville de
Rennes et les communes des alentours sont situées dans une vaste cuvette nommée le bassin
rennais. Cette cuvette est relativement plane et son altitude moyenne est d’environ 40 m. Le réseau
hydrographique rennais sur la zone géographique concernée par cette thèse est composé d’un cours

17
PARTIE I

d’eau principal : la Vilaine, dans lequel se jettent l’Ille, dans le centre-ville de Rennes. Au sud-ouest
de Rennes, un ensemble d’étangs borde la Vilaine pour former une vaste zone humide appelée « les
étangs d’Apigné ». Proche du centre-ville de Rennes au nord, on retrouve une autre zone humide
correspondant au parc des Prairies Saint Martin, qui borde l’Ille (Figure 1.9). En dehors de la forêt de
Rennes (2 915 ha) située au nord-est de la zone d’étude, on retrouve quelques formations boisées
éparpillées et de petite taille, ainsi qu’une trame bocagère plus ou moins dégradée.

Figure 1.9 - Réseau hydrographique rennais et emplacement de la zone humide des Prairies-Saint-Martin proche
du centre-ville. (Données : OCSGE 201 ; [Link]).

En termes d’urbanisme, le quartier historique densément urbanisé du centre-ville est entouré d’une
couronne pavillonnaire formée essentiellement de maisons individuelles mitoyennes avec jardins
particuliers en arrière-cour. Une deuxième couronne est composée d’une part de grands ensembles
de barres et de tours d’immeubles et d’autre part de quartiers pavillonnaires. Aux abords de cette
deuxième couronne, on retrouve les zones industrielles (Foissard 2015). La région Rennaise subit
une forte croissance démographique. Selon l’AUDIAR (l’agence d’urbanisme de Rennes), le pays de
Rennes accueillerait 130 000 habitants supplémentaires à l’horizon 2040. Pour répondre aux besoins
croissants de logements, le PLH (programme local de l’habitat) fixe l’objectif progressif de 3 800 à 4
600 logements par an sur l’ensemble de la métropole rennaise. L’étalement du tissu urbain couplé
à la densification des zones urbaines existantes a généralement pour conséquence une
intensification progressive de l’ICU (Oke et al. 2017). Les aménagements futurs de la ville de Rennes
et de ses abords devraient donc avoir des conséquences directes sur la modification du climat urbain
local (Foissard 2015).

18
 Introduction générale

Rennes et ses environs sont caractérisés par un climat océanique tempéré (Joly et al. 2010) et des
conditions thermiques relativement douces (Tmoy=12,4°C ; Tmin=8,0°C ; Tmax=16,8°C ; mesurées
sur la période 1991-2020 par la station Rennes-Saint Jacques). L’été est relativement doux avec une
température maximale moyenne de 24,8°C au mois de juillet. En moyenne la température maximale
est supérieure à 25°C durant 43,5 jours et supérieure à 30°C durant 8,9 jours. Bien que ces conditions
thermiques tempérées indiquent une prédominance du climat océanique à Rennes, une légère
dégradation de ce type de climat y est observable, comparativement à la partie Ouest de la Bretagne
(Foissard 2015). Les précipitations sont réparties sur l’ensemble de l’année et leur cumul est de 691
mm. Les précipitations sont moins importantes dans le bassin rennais que sur le reste de la région
Bretagne du fait de sa localisation à l’intérieur des terres. Le vent moyen observé est réparti
équitablement sur l’ensemble de l’année. Les journées avec un vent soutenu (> 57 km/h) concernent
en moyenne 41,7 jour par an. En revanche, les jours de tempêtes sont rares puisqu’ils concernent
en moyenne 0,6 jours par an. Ces épisodes sont observés principalement d’octobre à janvier. La forte
variabilité des types de temps observés à Rennes provoque une hétérogénéité temporelle forte de
l’ICU (Foissard 2015). La formation de l’ICU est favorisée par une combinaison entre ciel clair et vent
faible, correspondant généralement à des conditions anticycloniques, ce qui équivaut à environ 10%
des situations (Foissard et al. 2012).

1.4.2 Historique de la caractérisation de l’ICU rennais

Le phénomène d’ICU atmosphérique (correspondant à la couche de canopée urbaine ; voir 1.1.3
Genèse de l’îlot de chaleur urbain) est suivi depuis une vingtaine d’années sur l’agglomération
rennaise par les chercheurs de l’unité LTEG-Rennes. L’étude de la variabilité spatio-temporelle de
l’ICU à l’échelle de Rennes a nécessité la mise en place d’un réseau de capteurs conséquent qui a été
étoffé au fur et à mesure des programmes de recherche. Un premier réseau de mesure a été installé
dans le cadre du programme ECORURB (écologie du rural vers l’urbain). Il s’agit d’un programme de
recherche transdisciplinaire qui s’est déroulé de 2003 à 2012 et qui portait sur les « Dynamiques des
espèces et des espaces à l’interface ville-campagne » (Clergeau 2011). Dès 2003, des premiers
transects mobiles ont permis de localiser les secteurs de la ville présentant des températures
particulièrement élevées (Dubreuil et al. 2022). Durant la canicule d’août 2023, à laquelle est
attribuée le décès de 15 000 personnes en France et 70 000 en Europe (Besancenot 2005), les
relevés effectués à Rennes montrent comment l’ICU a pu accentuer le phénomène avec deux nuit
consécutives durant lesquelles la température en ville dépasse de plus de 5°C la température en
campagne, et reste bien au-dessus des seuils de récupération physiologique (Dubreuil et al. 2022).

19
PARTIE I

Figure 1.10 - ICU mesuré lors de la canicule de 2003 (du 7 au 11 août) à Rennes. Température horaire indiquée
en degrés Celsius). Extrait de Dubreuil et al. 2022.

Toujours dans le cadre du projet ECORURB, 16 stations météorologiques (Weather Monitor II puis
Vantage Pro 2) ont été installées en 2004. Les stations ont alors été réparties de manière à couvrir
trois classes de sites : urbain, suburbain et périurbain (Quénol et al. 2010). Les sites urbains sont
définis comme étant fortement artificialisés, avec une fraction de bâti variant de 20 à 40 %. A
l‘inverse, les sites périurbains sont des espaces essentiellement agricoles, caractérisés par un
recouvrement de la végétation qui représente 50% à 98% de l’occupation du sol avoisinant la station,
et une fraction du bâti inférieure à 5%, correspondant à des maisons individuelles plus ou moins
clairsemées. Les sites suburbains sont des espaces intermédiaires entre les sites urbains et
périurbains. Cinq stations supplémentaires viendront compléter ce réseau en 2010.
En 2013, dans le cadre de la thèse de Xavier Foissard, un second réseau de mesure composé de 20
capteurs de température a été déployé dans Rennes afin d’étudier et modéliser la variabilité spatiale
de l’ICU (Foissard 2015). En 2019, le réseau d’observation climatologique permanent de l’ICU à
Rennes était constitué de 25 stations automatiques, ainsi que deux stations supplémentaires gérées
respectivement par Météo France et l’INRAE (Dubreuil et al. 2020).
L’avènement des solutions connectées et la baisse du coût des capteurs ont ensuite permis une
refonte du système d’observation rennais, qui s’appuie depuis 2020 sur la technologie LoraWan.
LoRaWAN est un protocole de télécommunication permettant la transmission de données à bas
débit. Cette technologie, utilisée dans le cadre du suivi de l’ICU, a l’avantage de permettre la
multiplication des points de mesure, étant donné qu’il n’est plus indispensable de relever chaque
capteur manuellement. Elle permet également de suivre en temps réel le phénomène et de détecter
immédiatement toute défaillance sur le réseau, ce qui réduit considérablement la perte de données.

20
 Introduction générale

En 2021, année à laquelle cette thèse a débuté, le réseau baptisé RUN (Rennes Urban Network) est
constitué de 30 stations DAVIS Vantage-Pro-2, et d’un total de 93 capteurs connectés RisingHF
mesurant la température atmosphérique toutes les 15 minutes sous un abri anti-radiation normé.
Afin de standardiser au mieux les mesures, les capteurs sont disposés de manière à limiter l’influence
de facteurs pouvant affecter la température de l’air. Les capteurs sont également exposés au sud
pendant les heures chaudes de la journée (de 11 à 15 heures environ), de manière à éviter tout effet
d’ombre portée et ainsi mesurer correctement les températures maximales. De plus, les capteurs
sont installés à 3 mètres de hauteur (Figure 1.11), de manière à limiter le vol ou la dégradation du
matériel (Dubreuil et al. 2022).

Figure 1.11 - Exemples d’une station météorologique (à gauche) et d’un capteur connecté (à droite) à Rennes.
Photos : Alice Furet (2023).

Le choix des points de mesure s’est fait selon plusieurs critères. Les études antérieures menées
depuis 2003 ont permis d’évaluer la variabilité spatiale des températures suivant les caractéristiques
de la ville. De plus, les résultats des modélisations de l’ICU réalisées lors de la thèse de X. Foissard
ont permis d’identifier le volume du bâti, de la végétation, ou encore les surfaces imperméables
comme étant des facteurs jouant un rôle important dans l’intensité de l’ICU. Pour finir, la méthode
internationale de classification des formes urbaines nommée LCZ (Local Climate Zones) proposée
par Stewart et Oke (2012) a permis aux climatologues de répartir les capteurs de manière à obtenir
une couverture représentative de l’ensemble des climats locaux sur la zone d’étude (Dubreuil et al.
2022). La Figure 1.12 détaille les différents types de LCZ sur la zone d’étude et indique la répartition
des capteurs sur celles-ci.

21
PARTIE I

Figure 1.12 - Carte de la localisation des capteurs connectés à Rennes en fonction de la classification des LCZ
(extrait de Dubreuil et al. 2022).

22
 Introduction générale

1.4.3 Intensité de l’ICU durant la période d’étude

La densité du réseau de capteurs a permis de spatialiser l’ICU sur l’ensemble de la zone d’étude en
interpolant les valeurs entre les points au plus proche voisin sur une maille de 100 mètres. Par
exemple, les cartes de la Figure 1.13 sont construites à partir des 93 capteurs RisingHF opérationnels
dès 2021. Ces cartes permettent de visualiser l’intensité de l’ICU atmosphérique moyen au cours de
chaque mois de l’année 2022. A l’échelle d’une année complète, ces cartes mensuelles montrent
que l’ICU est particulièrement intense à la saison chaude, c’est-à-dire au printemps et l’été
(données : Vincent Dubreuil et Charlotte Brabant, LETG).

Figure 1.13 - Cartes mensuelles moyennes de l’intensité de l’ICU en 2022. Différence avec la température
minimale moyenne de la station la plus froide. Toutes les journées sont prises en compte pour le calcul des ICU
mensuels. Les valeurs affichées dans la légende sont exprimées en °C. (Données : Vincent Dubreuil et Charlotte
Brabant, LETG).

En effet, mars à septembre correspond à la période durant laquelle les « enveloppes chaudes » (en
rouge sur les cartes ; Figure 1.13) sont les mieux développées, avec un différentiel de température
entre le centre et la périphérie atteignant les 3°C et un ICU supérieur à 1,5°C dans toute la partie
intra-rocade de la ville. En automne et en hiver, l’intensité maximale de l’ICU ne dépasse pas 2°C
dans le centre-ville, et on observe un faible différentiel entre les températures des zones rurales et
les quartiers périphériques de Rennes.

23
PARTIE I

L’année 2022 est une année remarquable par ses conditions climatiques en France, comme sur
l’ensemble du globe. A Rennes, cette année a été marquée par un record de chaleur historique avec
plus de 40,5°C enregistrés le 18 juillet 2022 (station Météo-France Rennes-Saint Jacques). La
moyenne annuelle a également été dépassée de plus de 4 dixièmes de degrés par rapport au
précédent record enregistré en 2020 (Figure 1.14) (HCBC 2023). En 2022, presque tous les mois ont
connu des anomalies positives de températures. Trois vagues de chaleur se sont succédé entre juin
et juillet. Le nombre de jours chauds (plus de 25°C) ou très chauds (plus de 30°C) a également
dépassé les valeurs jusqu’alors mesurées (HCBC 2023).

Figure 1.14 - Températures moyennes annuelles (en °C) à Rennes. Les observations de 1950 à 2022 sont
représentées en vert. Les projections climatiques sont en orange pour le scénario intermédiaire (RCP4.5) et en
rouge pour le scénario à fortes émissions de gaz à effet de serre (RCP8.5). Les valeurs moyennes trentenaires
sont indiquées pour quelques périodes représentatives ; les valeurs pour Nice, Avignon, Toulouse et Angoulême
pour la période de référence 1951-1980 correspondent à la position du nom de ces villes sur l’échelle des
températures. Source des données : Météo France et portail Drias, réalisation HCBC. Extrait du rapport 2023
du HCBC (HCBC 2023).
En ce qui concerne l’ICU, Les cartes de la Figure 1.13 montrent sa distribution spatiale à l’échelle de
la zone d’étude. L’ICU est particulièrement intense dans le centre-ville mais aussi dans les quartiers
du centre-sud. Le secteur nord conserve en revanche des intensités plus basses, notamment à
proximité du centre-ville où le rôle d’îlot de fraîcheur des Prairies-Saint-Martin et de la coulée verte
sont particulièrement visibles. Les mesures réalisées à Rennes depuis 20 ans montrent que les ICU
atmosphériques intenses (plus de 4 °C d’écart entre ville et campagne), sont observés dans 16,5%
des nuits et surtout l’été (Dubreuil et al. 2020). Le 18 juillet 2022, l’ICU à Rennes a battu un nouveau
record absolu avec une différence de 9°C mesurée entre centre-ville et campagne. Cette situation
intense est par ailleurs bien visible sur la carte de l’intensité moyenne de l’ICU de juillet ( Figure 1.13).
L’année 2022 est dès lors particulièrement représentatrice du réchauffement climatique global, mais
également de l’accentuation des réchauffements locaux provoqués par l’urbanisation (HCBC 2023).
Le « Bulletin de santé publique canicule en Bretagne » indique dans un bilan relatif à l’été 2022 que
durant les trois vagues de chaleur parvenues respectivement en juin, juillet et août, les seuils d’alerte
biométéorologique ont été dépassés, entraînant un emballement de la mortalité humaine avec un
excès de décès de 168 personnes, ce qui représente une surmortalité relative de +20%.

24
 Introduction générale

1.4.4 Climat urbain et faune à Rennes : un contexte idéal

Si le lien entre vagues de chaleur et santé humaine est formellement établi (Piracha and Chaudhary
2022), les conséquences pour les écosystèmes locaux restent moins connues. Une première étude
mettant en relation des données biologiques avec l’intensité de l’ICU à Rennes a permis de mettre
en évidence un effet du réchauffement urbain sur la phénologie des arbres (Quénol et al. 2010). Les
auteurs ont constaté un déclenchement plus précoce de la floraison en ville, qui concorde avec des
écarts thermiques particulièrement élevés entre les stations des secteurs urbains, suburbains et
périurbains au moment du débourrement. En revanche, les impacts du climat urbain à Rennes n’ont
jusqu’alors pas été explorés au regard de la faune, bien que de nombreuses études d’écologie
urbaine se soient penchées sur les effets de l’occupation du sol et de l’habitat (Croci et al. 2008,
Varet et al. 2011a, 2013, 2014a). La mise en place du réseau de mesure à fine échelle sur Rennes en
2021, ainsi que les records d’intensité d’ICU enregistrés en 2022 forment un contexte idéal pour
étudier les liens encore inexplorés entre le climat de la ville de Rennes et la faune des écosystèmes
urbains.

1.5 Terrains d’étude et modèles biologiques

1.5.1 Prairies et pelouses urbaines à Rennes
A Rennes, les espaces verts publics couvrent 16% de la surface de la ville, et plus de la moitié de ces
espaces (56%) sont représentés par des surfaces herbacées (soit 445 ha) (Chollet et al. 2018). Depuis
un trentaine d’années, la municipalité de Rennes gère ses espaces verts selon une approche
différenciée, c’est-à-dire que l’intensité de la gestion des différents espaces est adaptée selon
l’intensité d’utilisation et selon le type d’usage associé aux espaces (Aggéri 2011). Selon ce principe,
les surfaces herbacées rennaises comprennent une large gamme de structures allant des pelouses
de haute intensité de gestion, avec une fréquence de tonte pouvant monter jusqu’à 20 passages par
an, aux prairies peu gérées fauchées une seule fois dans l’année. Une majorité d’espaces enherbés
sont quant à eux gérés de manière intermédiaire, avec 7 à 9 passages par an.
La disponibilité en prairies et pelouses, ainsi que le gradient d’intensité de gestion, font de ces
espaces des modèles idéaux pour la conduite d’études d’écologie expérimentale. Des études
récentes portant sur les effets de l’intensité de la gestion des prairies et pelouses rennaises sur les
communautés végétales ont d’ailleurs montré un fort impact des pratiques sur diverses
composantes de la diversité (taxonomique, fonctionnelle et phylogénétique) (Chollet et al. 2018),
ainsi que sur les assemblages d’espèces (Gros et al. 2023). La gestion semble en effet agir comme un
filtre environnemental sur les espèces, en excluant les espèces compétitrices et en favorisant les
espèces tolérantes au stress (Gros et al. 2023). La faune quant a elle n’a jusqu’alors pas fait l’objet
d’études sur les prairies et pelouses urbaines à Rennes. Cependant, de nombreuses études menées
dans d’autres villes indiquent que les communautés animales, notamment chez les arthropodes,
sont également touchées par les stratégies de gestion des surfaces herbacées urbaines (Buchholz et
al. 2018, Proske et al. 2022, Bauer et al. 2024).

25
PARTIE I

1.5.2 Les arthropodes comme indicateurs écologiques

Les arthropodes vivant au sol constituent un groupe important de la communauté animale des
systèmes prairiaux et jouent un rôle clé dans ces écosystèmes, tout en étant sensibles aux
changements environnementaux causés par les perturbations anthropiques (Solascasas et al. 2022),
faisant d’eux des indicateurs écologiques performants. En effet, leur taille réduite les rend sensibles
aux conditions locales, tandis que leur mobilité, bien que limitée, leur permet de se déplacer en
réponse à des conditions locales changeantes (Gerlach et al. 2013). De plus, des temps de génération
souvent courts entraînent des réponses rapides, et la variabilité de leurs exigences écologiques
fournit un large éventail d’espèces aux réponses environnementales spécifiques (Gerlach et al.
2013). Aux vues de ces caractéristiques, ces arthropodes sont largement utilisés comme modèles
biologiques dans les études d’écologie, y compris en milieu urbain (Eckert et al. 2017, Buchholz et al.
2018, Buchholz and Egerer 2020, Bauer et al. 2024).
Parmi ces arthropodes, les araignées constituent un taxon particulièrement adapté pour étudier la
réponse des organismes aux changements environnementaux. Ce groupe est extrêmement diversifié
avec plus de 50 000 espèces connues dans le monde (Nentwig et al. 2023), dont plus de 1600 en
France métropolitaine (UICN et al. 2023). Leur écologie, très variée, leur a permis de coloniser la
plupart des niches écologiques terrestres (Foelix 2011). Par exemple, les diverses méthodes de
prédation que l’on retrouve au sein de ce taxon, nommées « guildes de chasse » (Cardoso et al.
2011), sont en grande partie responsable du micro-habitat fréquenté par les espèces (Uetz 1979,
1991), par ailleurs elles sont toutes considérées comme « prédateurs généralistes » (Wise 1993).
Parmi les huit guildes communément considérées (Cardoso et al. 2011), les tisseuses de toiles dites
« orbitales » chassent de manière passive à l’aide de toiles géométriques leur permettant de
capturer principalement des insectes volants. Ces espèces peu mobiles sont tributaires de la
présence d’éléments structurels du micro-habitat leur permettant d’accrocher leurs toiles (Uetz
1991). Au contraire, les chasseuses au sol, très mobiles, ne tissent pas de toile dans un but de
prédation, mais chassent leurs proies à vue et peuvent donc aussi bien se trouver dans une
végétation dense, que sur un sol nu ou couvert de litière (Uetz et al. 1999). D’autres espèces se
placent « en embuscade » sur des fleurs en attentes de proies, ou encore adoptent une méthode de
chasse intermédiaire comme les « chasseuses errantes » (Cardoso et al. 2011).
Cette diversité de guildes s’accompagne d’une large gamme de tailles corporelles et de capacités de
dispersion contrastées. En effet, certaines espèces sont capables de se déplacer passivement par les
airs sur de longues distances. C’est ce qu’on appelle le phénomène de « ballooning » (Bell et al.
2005). En condition de vent et de température particulières, ces espèces peuvent tisser une toile
qui, une fois portée par la force du vent, leur permet de se soulever et d’être emmenées à leur tour.
En plus de cette diversité fonctionnelle, leur abondance, leur diversité en espèces et leur facilité
d'échantillonnage permet l’établissement de jeux de données scientifiquement robustes (Hacala et
al. 2024). En milieu urbain, le filtre environnemental sur les espèces est tel qu’il est particulièrement
important de choisir un modèle biologique suffisamment représenté (en termes de densité
d’individus et de diversité d’espèces) depuis l’extrémité rurale du gradient d’urbanisation, jusqu’à
son extrémité urbaine. Les araignées se sont montrées jusqu’alors particulièrement propices à la
conduite d’études d’écologie urbaines, en permettant l’obtention de jeux de données robustes en
ville (Alaruikka et al. 2002, Varet et al. 2013, Buchholz et al. 2018, de la Flor et al. 2020).

26
 Introduction générale

Certains groupes de coléoptères, en particulier les carabes, sont également des indicateurs
écologiques très fréquemment étudiés en ville (Niemelä and Kotze 2009, Magura and Lövei 2021).
Les carabes sont un groupe diversifié en termes de nombre d’espèces et abondant en ville comme
en campagne. Leurs caractéristiques écologiques sont souvent étudiées au regard de traits
fonctionnels similaires à ceux des araignées : les tailles corporelles varient beaucoup au sein de ce
groupe, ainsi que leurs capacités de dispersion qui sont définies par la taille des ailes, permettant le
vol ou non. Dans le cas des carabes, les habitudes alimentaires ne sont pas définies par des guildes
de chasse, mais par leur régime alimentaire qui peut être herbivore, granivore, prédateur, ou
omnivore.
Enfin, toujours au sein de l’ordre des coléoptères, les staphylins sont quant à eux beaucoup moins
fréquemment étudiés que les araignées ou les carabes, mais présentent des caractéristiques
écologiques toutefois tout aussi intéressantes que les deux précédents groupes (Bohac 1999). Les
staphylins sont décrits comme un groupe « hyper divers » dans la littérature (Betz et al. 2018), et
présentent des régimes alimentaires ainsi que des tailles corporelles très variées (Clough et al. 2007).
Ces caractéristiques représentent une opportunité pour l’étude de la diversité et de la composition
fonctionnelle au sein des communautés. Les staphylins remplissent dès lors les critères d’un
indicateur écologique fiable. Leur faible occurrence dans la littérature provient probablement de la
complexité à l’identification et du peu de littérature dédiée à leur détermination.

1.5.3 Acquisition des données

Au cours de cette thèse, trois groupes taxonomiques d’arthropodes ont été étudiés comme modèles
biologiques. Le groupe principal, au cœur de cette thèse, est celui des araignées (vivant au sol). Deux
groupes secondaires sont les carabes et les staphylins. L’échantillonnage de ces trois groupes a été
réalisé sur des prairies et pelouses urbaines, distribuées le long d’un gradient d’ICU à Rennes. Au
cours de cette thèse, nous avons choisi d’étudier un type d’habitat urbain en particulier : les prairies
et pelouses urbaines. Pour étudier la faune arthropode du sol dans les prairies et pelouses urbaines
(habitats échantillonnés dans cette thèse), le piégeage à interception de type « Barber » (Barber
1931) est un dispositif largement utilisé (Spence and Niemelä 1994). Ce type de piège consiste en un
récipient enterré jusqu’à la surface du sol, dans lequel les arthropodes tombent au cours de leurs
déplacements au sein du micro-habitat échantillonné (Figure 1.15). Il existe différentes variantes de
ce genre de dispositif : munis d’appât ou non, remplis de liquide de piégeage ou non. Il présente
aussi différentes déclinaisons de tailles ou de couleurs (Brown and Matthews 2016).

27
PARTIE I

Figure 1.15 - Coupe schématique d’un piège Barber

Les dispositifs utilisés durant cette thèse ont été conçus de manière à introduire le moins de biais
d’échantillonnage possibles dans la représentativité des communautés. Pour cela, des gobelets en
plastique transparents (85 mm de diamètre, 115 mm de hauteur) ont été utilisés, et remplis de 150
ml de solution saline à 100 g/L pour empêcher la décomposition des arthropodes, et d'une goutte
de savon inodore pour éviter leur flottaison. Afin d’éviter le débordement des pièges en cas
d’épisodes pluvieux, et par la même occasion la perte d’individus piégés, un couvercle de plastique
transparent a systématiquement été disposé 50 mm au-dessus du bord supérieur de chaque piège
(Figure 1.16).

Figure 1.16 - Piège Barber enterré et muni d’un abri transparent. Une sonde de température est disposée à
proximité directe du piège, afin de mesurer la température du micro-habitat, 5 cm au-dessus de la surface du
sol.

28
 Introduction générale

Au total, 36 sites ont été identifiés, tous situés sur des prairies ou pelouses (la procédure de sélection
des sites est détaillée dans le Chapitre 1). Les sites d'échantillonnage sont situés dans des parcs
publics, des jardins communautaires, des espaces verts adjacents à des bâtiments scolaires ou
encore des jardins privés. Nous avons échantillonné les arthropodes de mars à septembre 2022, en
utilisant trois pièges Barber par site d'échantillonnage, placés à cinq mètres les uns des autres pour
former un triangle et favoriser l'efficacité du piégeage (Ward et al. 2001). Les trois pièges ont été
orientés de manière standard vers le nord, le sud-ouest et le sud-est. Le piégeage a été réalisé durant
les deux premières semaines de chaque période de sept mois d'échantillonnage, pour un total de
756 unités d'échantillonnage (c.-à-d. trois pièges sur chacun des 36 sites sur sept sessions). A
proximité directe des pièges, des sondes de température (Lascar EL-USB-2+ ; Tinytag Talk 2 TK-4023)
ont été placées à cinq centimètres au-dessus de la surface du sol afin d’enregistrer les températures
proches du sol toutes les 15 minutes pendant chaque session (Figure 1.16). Ces sondes ont été posées
sans abris afin de mesurer les indices actinothermiques, correspondant aux températures ressenties
par les arthropodes étudiés, dans leur micro-habitat.
Après chaque session d’échantillonnage, tous les individus d'araignées, de carabes et de staphylins
ont été conservés séparément par site et par session d'échantillonnage, dans l'éthanol à 70 %. Les
araignées matures ont été identifiées au niveau de l'espèce à l'aide des clés d'identification de
Roberts (1985, 1987) (nomenclature conforme au World Spider Catalog version 22 ;
[Link] Les staphylins ont été identifiés au niveau de l'espèce en utilisant Lott (2009)
et Lott and Anderson (2011). Les carabes ont été identifiés à l’espèce à l'aide de Jeannel (1941). Sur
la base de ces données de communautés, nous avons analysé la réponse des arthropodes au
conditions de température et d’habitat en considérant séparément les araignées (Chapitre 1), les
staphylins et les carabes (Chapitre 2). Etant donné que le nombre d’individus piégés s’est révélé être
moindre chez les staphylins et les carabes, par rapport aux araignées, nous nous sommes limités aux
analyses de communautés chez ces deux groupes. En revanche, le nombre élevé d’individus
d’araignées piégées a permis de mener des études supplémentaires, notamment morphologiques
sur les populations de plusieurs espèces (Chapitre 4). Par ailleurs, la capture d’individus d’araignées
vivants (capturés à vue) et en nombre suffisant, a également permis de mener une étude
comportementale détaillée dans le Chapitre 5.

1.6 Problématique et questions de recherche

Dans le contexte du réchauffement global et plus particulièrement de l’amplification de ce
phénomène par l’ICU en ville, les objectifs de cette thèse sont (1) de déterminer si la faune
ectotherme, étudiée par le prisme des arthropodes vivant au sol, est affectée par les variations
spatiales de température et (2) d’identifier les échelles du climat déterminantes pour ces
organismes. L’hypothèse générale est que les arthropodes terrestres urbains, de par leur
vulnérabilité aux températures et leurs capacités de dispersions limitées, sont contraints par les
variations thermiques à l’échelle d’une ville. Nous supposons que l’ICU agit comme un filtre
environnemental, en excluant certaines espèces particulièrement vulnérables de par leurs
caractéristiques écologiques, impactant indirectement la diversité et la composition des
communautés.

29
PARTIE I

Sur la base de cette hypothèse générale, nous dérivons les questions de recherche suivantes :

o Peut-on dissocier l’effet des variables liées à la modification de l’habitat en milieu urbain et les
variations de température ?

Le réchauffement de l’air en ville est intrinsèquement lié au développement urbain et notamment

aux surfaces couvertes par les matériaux de construction et à la forme des constructions (1.1.2 Effets
de l’urbanisation sur le climat), Dès lors, étudier l’effet du climat en ville sur les organismes nécessite
de pouvoir décorréler les variables décrivant la structure de l’habitat d’une part, et les températures
d’autre part. Nous pensons que cet objectif peut être atteint par une procédure de sélection
spécifique de sites d’étude.

o L’échelle à laquelle le réchauffement urbain est communément caractérisé (l’ICU atmosphérique ;

1.1.3 Genèse de l’îlot de chaleur urbain) est-elle pertinente pour étudier les effets des variations de
températures sur les arthropodes ?

Du point de vue de la santé humaine, les conditions au sein de la couche de canopée urbaine sont
les plus pertinentes, car elles correspondent aux conditions ressenties par les personnes. Les
températures peuvent varier à l’échelle de quelques centaines de mètres seulement, ce qui
correspond à de courtes distances pour les humains, ainsi que pour les organismes dont le domaine
vital est étendu. Les organismes dont le domaine vital ne s’étend pas au-delà de cette distance
peuvent donc être particulièrement impactés par ces conditions locales, ce qui est le cas pour de
nombreux arthropodes. Cependant, les conditions au sein de leur habitat au niveau de la surface du
sol, dont l’échelle correspond à leur taille très réduite (de l’ordre d’un mètre), peuvent différer des
conditions atmosphériques communément considérées pour caractériser le réchauffement urbain.
Il nous paraît donc essentiel d’estimer l’importance relative des variations de température 1) à
l’échelle du paysage urbain et 2) à l’échelle du micro-habitat proche du sol.

o Si la hausse des températures en ville est effectivement reliée à une modification de diversité ou
d’assemblage d’espèces au sein des communautés, quelles peuvent-être les caractéristiques
fonctionnelles déterminantes du maintien ou de l’exclusion des espèces ?

Au sein d’un même taxon, les espèces peuvent présenter des caractéristiques écologiques très
différentes, leur permettant de se maintenir ou non dans un environnement donné. Nous pensons
qu’étudier la composition fonctionnelle des communautés sur la base des traits fonctionnels des
espèces permettra d’améliorer notre compréhension des mécanismes responsables des éventuels
patrons de biodiversité observés.

o Si certaines espèces sont capables de se maintenir dans des conditions thermiques contrastées, la
température induit-elle des variations morphologiques intraspécifiques ?

Outre l’extinction locale, diverses réponses écophysiologiques peuvent permettre aux individus
d’une même espèce de se maintenir sous de nouvelles conditions environnementales. Ces réponses
des individus à des contraintes nouvelles peuvent être dues à des adaptations locales de nature
évolutives, ou le résultat de la plasticité phénotypique. Les relations entre la température et taille

30
 Introduction générale

corporelle sont particulièrement étudiées dans le contexte du changement climatique. Nous

pensons qu’identifier de telles relations en milieu urbain et réaliser une comparaison interspécifique
permettra d’améliorer notre compréhension des mécanismes responsables de la vulnérabilité plus
ou moins accrue chez les espèces, y compris dont des populations persistent en conditions urbaines
chaudes.

o Quel rôle le comportement joue-t-il dans la capacité des espèces à se maintenir dans un
environnement thermique contraignant ?

La thermorégulation comportementale est très courante chez les ectothermes. Cette capacité à tirer
profit des variations thermiques au sein du micro-habitat peut permettre aux organismes de
maintenir leur température corporelle proche de l’optimum, y compris lorsque l’environnement
thermique global leur est défavorable. Nous pensons que l’étude de ce comportement chez des
populations issues d’environnements plus ou moins « chauds » permettra d’améliorer notre
compréhension de l’échelle du climat écologiquement pertinente pour étudier les impacts sur les
arthropodes.

o Y a-t-il une interdépendance des températures aux échelles du micro-habitat et de la ville, et quels
sont les leviers d’action pour atténuer le réchauffement sur les arthropodes en ville ?

Les températures ressenties par les arthropodes vivant au sol sont susceptibles de différer de
manière importante des températures atmosphériques communément étudiées pour caractériser
le climat urbain. Pour la conception de mesures efficaces d’atténuation du réchauffement en faveur
de la biodiversité en ville, il est impératif que ces relations entre échelles du climat soient étudiées.
Il nous apparaît notamment essentiel d’identifier les éléments structurels des espaces verts ouverts
qui doivent être considérés en priorité dans la conception d’infrastructures vertes génératrices de
microclimats propices à la biodiversité.

1.7 Organisation de la thèse

Dans la Partie I, nous nous intéressons aux effets des variations spatiales de température sur les
communautés d’arthropodes le long d’un gradient d’urbanisation. Plus précisément :

o Le Chapitre 1 explore l’importance de considérer les échelles paysagères et locales pour

déterminer l’influence relative de l’habitat et de la température sur l’abondance, la diversité
taxonomique et la diversité fonctionnelle au sein des communautés d’araignées.

o Dans le Chapitre 2, nous étudions l’effet des mêmes facteurs environnementaux que dans le
Chapitre 1, mais cette fois à l’aide d’une approche multi-taxonomique sur les communautés de
staphylins et de carabes.

Dans la Partie II, nous nous focalisons sur le modèle « araignées » afin d’explorer les mécanismes
fonctionnels expliquant les liens entre patrons de diversité (observés dans le Chapitre 1) et variables
climatiques urbaines mesurées à plusieurs échelles :

31
PARTIE I

o Dans le Chapitre 3 (Cabon et al. 2024b), nous nous intéressons aux relations entre traits
fonctionnels et environnement à l’échelle des communautés, dans le but d’identifier les
caractéristiques des espèces qui déterminent leur établissement ou non en milieu urbain.

o Dans le Chapitre 4 (Cabon et al. 2024a), nous analysons les relations intraspécifiques entre
taille corporelle et gradients de température chez 11 espèces d’araignées présentant des
traits d’histoire de vie contrastés. La comparaison interspécifique de ces relations permet
de formuler des hypothèses sur les traits d’histoire de vie jouant un rôle dans la sensibilité
des espèces aux variations climatiques urbaines.

Dans la Partie III, nous nous intéressons tout spécifiquement au rôle de la température considérée à
l’échelle du micro-habitat des arthropodes étudiés jusqu’alors. Pour cela :

o Dans le Chapitre 5 (Cabon et al. 2023) nous étudions s’il existe un lien entre (1) les
températures sélectionnées par les individus d’une espèce d’araignée (Pardosa saltans)
dans un environnement thermique contrôlé (on parle communément de préférence
thermique), et l’intensité de l’îlot de chaleur urbain associée au milieu d’origine des
populations prélevées.

o Dans le Chapitre 6, nous analysons les relations entre échelles du climat mesurées au cours
des études présentées dans les précédentes parties. Pour cela, nous étudions le potentiel
d’atténuation de la végétation herbacée, par rapport aux températures atmosphériques,
communément considérées pour caractériser l’ICU.

Enfin, une discussion générale permettra de discuter les résultats obtenus dans les chapitres
nommés ci-dessus et nous permettra de suggérer des perspectives des recherches portant sur
l’écologie des organismes ectothermes en milieu urbain. L’organisation des chapitres de cette thèse
est résumée en Figure 1.17.

32
 Introduction générale

Figure 1.17 - Organisation des parties et chapitres de la thèse.

33
PARTIE I

34
 CHAPITRE 1

2 Partie II : Effet de la température sur la

diversité des arthropodes

▪ Chapitre 1 : Importance de l’échelle pour dissocier les effets de l'habitat et de la

température sur les communautés d'araignées urbaines

▪ Chapitre 2 : : Effets des facteurs liés à l’habitat et à la température sur les

communautés de staphylins et de carabes en ville

En bref :
▪ Le design expérimental mis en place pour l’échantillonnage commun à ces deux chapitres
permet de simultanément dissocier (1) l’effet de l’habitat et de l’ICU et (2) l’effet de
l’échelle (paysage versus micro-habitat) sur les communautés d’arthropodes.

▪ Les groupes d’arthropodes répondent différemment aux variables environnementales

testées : les communautés d’araignées sont influencées par les températures, alors
qu’aucun signal n’est observé chez les staphylins et les carabes.

▪ Les variations de température affectent différemment les communautés d’araignées selon

l’échelle considérée : L’ICU atmosphérique (échelle du paysage) est négativement relié à
la diversité taxonomique. En revanche, le réchauffement proche du sol (échelle du micro-
habitat) est négativement relié au nombre d’individus piégés.

▪ La structure du micro-habitat, et notamment la hauteur de végétation, est positivement

reliée à la diversité fonctionnelle chez les araignées, mais aussi chez les staphylins. Cela
indique un fort potentiel des micro-habitats pour atténuer l’effet négatif des
réchauffements sur les arthropodes en ville.

▪ L’absence de réponse aux variables testées chez les carabes indique que d’autres facteurs
environnementaux sont susceptibles d’impacter les communautés. Le faible nombre
d’individus piégés pourrait également expliquer ce résultat.

35
PARTIE II

2.1 Chapitre 1 : Importance de l’échelle pour dissocier les effets de

l'habitat et de la température sur les communautés d'araignées
urbaines

Résumé
L'urbanisation est associée à de multiples changements environnementaux qui affectent les
écosystèmes du monde entier. La température et l'habitat sont les principaux moteurs des
communautés animales dans les villes, mais leur importance relative reste peu étudiée, car les îlots
de chaleur urbains (ICU) sont intrinsèquement liés à la perte d’habitat due à l’artificialisation
croissante des surfaces. Ce premier chapitre vise à dissocier les effets et l'échelle pertinente de la
température et de l'habitat sur les communautés d'araignées vivant au sol. Sur la base d'un plan
d'échantillonnage original, nous avons collecté 20 761 individus d'araignées appartenant à 137
espèces sur 36 sites d'échantillonnage de la ville de Rennes (nord-ouest de la France). Nous avons
caractérisé les communautés en évaluant le nombre d'individus piégés sur chaque site, ainsi qu'en
calculant plusieurs métriques permettant d’estimer les diversités taxonomiques et fonctionnelles.
Les mesures de température ont été obtenues à partir de deux réseaux de capteurs mesurant
indépendamment l'ICU (résolution de 100 m) et la température proche du sol (résolution de 1 m).
L’occupation du sol et l'isolement sont utilisés pour décrire l’habitat à l’échelle du paysage, et la
structure de la végétation pour décrire les habitats locaux. Nous avons utilisé des modèles linéaires
mixtes généralisés pour dissocier les effets de la température de ceux de l'habitat à l'échelle du
paysage et à l'échelle locale, et nous avons identifié les relations entre les variables décrivant les
communautés (activité-densité, diversité taxonomique et fonctionnelle), et les prédicteurs. Nous
montrons que les paramètres liés à la température sont des prédicteurs importants des
communautés d'araignées, et que les échelles locale et paysagère ont des effets indépendants. La
température près du sol explique à elle seule 24 % du nombre d'individus piégés, tandis que l'ICU
explique 20 % de la diversité taxonomique. Nous montrons que la hauteur et la couverture de la
végétation locale sont des prédicteurs significatifs de la diversité fonctionnelle et expliquent
respectivement 22 % et 25 % de la variance. Nous concluons que des mesures d'aménagement
appliquées localement pourraient atténuer la perte de diversité taxonomique induite par l'ICU
atmosphérique, et favoriser l'établissement de communautés plus diversifiées.

Article en révisions mineures chez Urban Ecosystems.

36
 CHAPITRE 1

When scale matters to disentangle the effect of habitat and temperature on

ground-dwelling spider communities in urban environments

Valentin Cabon1,2, Dylan Amiar1, Romain Georges1,2, Vincent Dubreuil2,3, Julien Pétillon1,4, Hervé
Quénol2,3, Benjamin Bergerot1,2

1. University of Rennes, CNRS, UMR 6553 ECOBIO, Rennes, France

2. LTSER ZA Armorique, F-35000 Rennes, France
3. University of Rennes 2, CNRS, UMR 6554 LETG, Rennes, France
4. Institute for Coastal and Marine Research – Nelson Mandela University – Port Elizabeth – South
Africa

Abstract
Urbanisation is an ongoing process associated with multiple environmental changes affecting
ecosystems worldwide. Temperature and habitat are main drivers of animal communities within
cities, but quantifying their relative weights remains a challenge, as urban heat islands (UHI) often
co-vary with land-cover. This study aims to disentangle the effects and relevant scale of temperature
and habitat on ground-dwelling spider communities. Based on an original sampling design, we
collected 20,761 spiders individuals belonging to 137 species at 36 sampling sites in the city of
Rennes (northwest France). We characterised communities by assessing the number of trapped
individuals at each site, as well as calculating several metrics to estimate taxonomic and functional
diversities. Temperature metrics were obtained from two sensor networks monitoring UHI (100-m
resolution) and near-ground temperature (1-meter resolution) independently. Land-cover and
isolation were used to describe landscapes, and vegetation structure to describe local habitats. We
used generalized linear mixed models to disentangle the effects of temperature from those of
habitat at the landscape and local scales, and identified relationships between community
descriptors and predictors. We show that temperature-related metrics are important predictors of
spider communities, and that the landscape and local scales have independent effects. Near-ground
temperature alone explained 24% of the number of trapped individuals, whereas UHI explained 20%
of taxonomic diversity. Local vegetation height and cover were significant predictors of functional
diversity, and explained 22% and 25% of variance, respectively. We conclude that locally applied
planning measures could mitigate the loss of taxonomic diversity induced by the atmospheric UHI
and promote the establishment of more diverse communities.

Key words
Araneae, climate warming, functional diversity, taxonomic diversity, urban heat island

37
PARTIE II

2.1.1 Introduction
With more than 50% of the world’s population living in urban areas (United Nations 2019), cities are
rapidly expanding worldwide. The associated transformation of wildlife habitats is considered a
leading cause of biodiversity loss (Fahrig 2003), but awareness of the need for urban wildlife
protection has increased in the last decades (Lin and Chen 2022). Strategies for sustaining
biodiversity and related ecosystem services have been rising in cities through the integration of
green infrastructures (Aznarez et al. 2022), also mitigating urban warming (Bartesaghi-Koc et al.
2020).
The urban physical environment host a myriad of environmental conditions that represent a multi-
dimensional ecological filter often associated with a loss of taxonomic and functional diversities
among animal and plant communities (McKinney 2006, Aronson et al. 2014). Biotic and abiotic
factors occurring across the urban matrix determine the presence and the abundance of species,
based on their ability to cope with urban conditions (MacGregor-Fors et al. 2022, Delgado de la flor
et al. 2024). Accordingly, species can be classified as urban dwellers, utilisers or avoiders (Fischer et
al. 2015). It depends on whether they can pass through the urban filter, but also establish over a
longer term. The resulting community distribution of species across space leads to the emergence
of diversity patterns along urbanisation gradients (Aronson et al. 2014, Hahs et al. 2023). To explain
these patterns, recent urban ecology studies have aimed to disentangle the roles of specific urban
factors (Beninde et al. 2015) and investigated landscape (e.g., connectivity, heterogeneity or the
proportion of impervious surface) and local (e.g., vegetation structure, microtopography or
management practices) predictors distinctly. Landscape predictors are generally considered as
upper-level filters that determine the permeability of the surrounding urban matrix for the
immigration of species from rural adjacent areas into cities (Faeth et al. 2011, Beninde et al. 2015).
Local predictors are intended to determine habitat suitability and are mainly related to species
survival, reproduction success and abundance (Beninde et al. 2015, Proske et al. 2022). Although
ecological filters operate at both spatial scales (Christie et al. 2010, Norton et al. 2014, Philpott et al.
2014, Otoshi et al. 2015, Kyrö et al. 2018, Piano et al. 2020d), the relative weights of landscape and
local-scale predictors in shaping diversity patterns differs across taxa, and is strongly linked to the
functional traits of species (Croci et al. 2008).
Urban temperature-related predictors are increasingly taken into account, in link with habitat-
related factors. The urban heat island (UHI) is characterised by an increase of the mean temperature
of urban cores relatively to adjacent rural areas (Oke et al. 2017), mainly resulting from warmer night
temperatures (Oke et al. 2017, Foissard et al. 2019a, Dubreuil et al. 2020). Atmospheric UHIs depict
air warming above the surface (e.g., at a 2-m height, Foissard et al. 2019), and are characterised as
large areas of heat at the landscape scale (Oke et al. 2017). While many ecological studies have dealt
with the relationship between atmospheric UHIs and vegetation, including in our study area (Mimet
et al. 2009), the relationship with arthropod fauna is much less documented. Arthropods are
ectothermic species whose metabolism depends on outside temperature. They are particularly
interesting ecological targets in cities because of their sensitivity to temperature and consequently
to warming (Menke et al. 2011, Hamblin et al. 2018, McGlynn et al. 2019). First results showed that
the urban atmospheric UHI measured at the canopy level was associated with shifts in arthropod
phenology (Meineke et al. 2014) and community composition (McGlynn et al. 2019). Other studies
measuring local temperature within a few meters reported associations between near-ground
temperature and changes in arthropod abundance (Shochat et al. 2004, Hamblin et al. 2018, Piano

38
 CHAPITRE 1

et al. 2020a). Although most studies emphasize the physiological implication of near-ground
temperature for single taxa (Angilletta et al. 2007, Diamond et al. 2018b, Yilmaz et al. 2022),
information about community-wide responses to near-ground temperature still remains scarce (but
see McGlynn et al. 2019, Piano et al. 2020).
We expect different ecological responses in cities depending on the scale considered to characterise
temperature, as in the case of land cover and habitat. The landscape-scale atmospheric
measurements strongly differs from local-scale near-ground temperatures (Lembrechts et al. 2020).
The standardised atmospheric UHI is a reliable metric commonly used to characterise warming
patterns at the landscape scale, whereas near-ground temperatures are spatially heterogeneous and
commonly referred to as thermal mosaics (Ziter et al. 2019, McGlynn et al. 2019) due to the fine-
scale complexity of vegetated and built surfaces within cities. Temperature measured at the ground
level matches the scale at which arthropods experience their thermal environment (Pincebourde
and Woods 2020). As a result, local thermal heterogeneity provides opportunities for small
organisms to buffer adverse thermal conditions through behavioural thermoregulation (Woods et
al. 2015). As with habitat-related predictors, it is necessary to consider multiple – local and landscape
– scales to study the response of arthropod communities to temperature. Near-ground (local) and
atmospheric (landscape) measurements of temperature have rarely been taken into account
simultaneously in urban ecology studies because of the lack of sensor networks for monitoring
temperature at different spatial scales.
Spiders are ideal biological models for urban ecology studies because they are convenient to collect
and can be found abundantly along urbanisation gradients (Varet et al. 2011a, Vergnes et al. 2014,
Piano et al. 2020b). Furthermore, they are highly diverse groups that respond rapidly to urban
environmental stressors, and substantial literature is available about landscape and local habitat
predictors driving their taxonomic and functional diversities (Sattler et al. 2010, Varet et al. 2011a,
Vergnes et al. 2014, Buchholz et al. 2018, Piano et al. 2020b). Moreover, spiders are prey and
predators, making them important links in trophic chains and the basis of many ecosystem services.
They are also ectotherms, and this makes them particularly sensitive to urban warming (Meineke et
al. 2017a, Johnson et al. 2020b).
In this study, we aimed to disentangle landscape factors (land cover and the atmospheric UHI) from
local factors (habitat and near-ground temperature) to explain diversity patterns among spider
communities. The main novelty of our study consists in analysing the role of temperature at the
landscape (atmospheric UHI) and local (near-ground temperature) scales, thanks to two
independent networks of temperature sensors. We expected taxa to display different responses to
temperature at the landscape and local scales. Indeed, in a recent study examining the decoupled
effect of urban warming on fly communities, the atmospheric UHI overweighted land-cover factors
as a driver of the community composition (McGlynn et al. 2019). In addition, urban near-ground
temperature was found closely related to the abundance of Coleoptera and Hemiptera (Piano et al.
2020). We therefore expected near-ground temperature to modulate the number of spider
individuals trapped by affecting habitat thermal suitability, as proposed by Shochat et al. (2004),
whereas the atmospheric UHI should be more related to taxonomic and functional diversity patterns
(McGlynn et al. 2019).

39
PARTIE II

2.1.2 Materials and methods

[Link] Study area and sampling design
We conducted our study in Rennes, a city of 223,000 inhabitants in north-western France. City
planning is characteristic of European medium-sized cities, with a historical city centre composed of
densely built residential and commercial neighbourhoods (> 40% built-up area) surrounded by
suburban areas characterised by a lower building density (10 - 30%) and more vegetation cover (15
- 50%). The city is surrounded by a peri-urban conurbation where land cover is mainly composed of
fields and individual houses. Rennes is located 70 km from the sea and is under the influence of
temperate oceanic climate. In 2022, the mean annual temperature was 13.7 °C, the mean winter
temperature 7.7 °C and the mean summer temperature 19.6 °C (Saint-Jacques airport station).
Despite its mild climate, the city faces regular UHI events. Between 2004 and 2019, strong UHI
situations (above 4 °C) represented more than 17% of the total amount of nights, and this proportion
reached 30% in specific months (e.g., September) (Dubreuil et al. 2020).
To identify suitable sampling areas where land cover and the atmospheric UHI were not correlated,
we first performed a spatial correlation analysis, as described in Appendix 2.1 at the end of this
chapter. This sampling site selection procedure is an important prerequisite because it determines
the opportunity to discriminate individual effects of predictors related to landscape habitat and
atmospheric UHI on spider communities.
To characterize the atmospheric UHI, we used the long-term Rennes Urban Network
([Link] established in 2004. Since 2020, 30 DAVIS Vantage-Pro-2 automatic weather
stations and 93 connected temperature sensors (Rising-HF- RHF1S001) have been deployed across
the city centre and the peri-urban and rural areas of Rennes, and collect temperature data at 60-
and 15-minute frequencies, respectively. Using this raw temperature data, we measured the daily
UHI intensity by calculating the difference between the minimum daily temperature data at a
reference cold station located outside the urban area located about 11 km in the north of Rennes
(Melesse, 48°12′18.1″N 1°40′52.8″W) and the minimum daily temperature data recorded by all other
sensors. In a second step, we interpolated these discrete measures of daily UHI intensity by inverse
distance weighting to obtain maps of estimated values covering the entire city extent at 100-m
resolution (Dubreuil et al. 2020). Since the UHI intensity varies over time, we averaged the daily
intensity values to the whole study period (from 1st March to 30th September 2022) to obtain a single
raster map of mean UHI intensity (Figure 2.1).
We selected 36 sampling sites (Figure 2.1) located on the identified suitable sampling areas, all
located on grasslands. The sampling sites were located in public parks, community gardens, green
spaces adjacent to educational buildings, or private gardens.

40
 CHAPITRE 1

Figure 2.1 - Map of the study area. The mean atmospheric UHI (from 1st March to 30th September 2022) is
illustrated by a coloured gradient ranging from blue (low intensity; minimum = 0 °C) to red (high intensity;
maximum = 3 °C). Grey areas display impervious surface.

[Link] Community data collection

We sampled arthropods from March to September 2022, using three pitfall traps per sampling site
placed five meters from each other to form a triangle and favour trapping efficiency (Ward et al.
2001). The three traps standardly headed north, south-west and south-east. We used plastic cups of
85 mm diameter and 115 mm height buried to the ground level and filled with 150 mL of saline
solution at 100 g/L to prevent decomposition, and a drop of neutral soap to avoid floating. The traps
were opened during the first two weeks of each seven months of sampling and emptied directly after
each trapping session, resulting in 756 sampling units (i.e., three traps on each of the 36 sites and
seven sessions). We sorted all spider individuals per site and sampling session and kept them in
separate vials containing 70% ethanol. Mature spider individuals were identified to the species level
using Roberts' (1985, 1987) identification keys (nomenclature following the World Spider Catalog
version 22; [Link] and the number of trapped individuals per species , site and
session was assessed. Three functional traits were retrieved from literature data: body size (mm),
dispersal (ballooning capacity) and hunting guild (Appendix 2.2). These traits are regularly considered
to cover a broad range of functional attributes and to respond to environmental changes along
urbanisation gradients (Buchholz et al. 2018). More precisely, body size affects species physiology
and fitness, and determines the ecological niches and spatial distribution of species (Merckx et al.
2018c, Magura and Lövei 2021). Considering the sexual dimorphism in most spiders, we considered
the body size of females as a proxy for a given species (Merckx et al. 2018c). Hunting guild informs
on the foraging behaviour, and provide relevant information on the species’ rank in the food web
(Cardoso et al. 2011). The dispersal capacity informs on the species’ ability to escape their home
range so as to avoid competition or adverse environmental conditions and favour gene flow across
populations (Bonte and Dahirel 2017).

41
PARTIE II

[Link] Landscape predictors

We calculated habitat and temperature predictors at the landscape scale based on land-cover and
atmospheric UHI data, respectively. We used circular buffers of 100-m radius around the sampling
sites to calculate the landscape predictors because this extent is (1) an appropriate scale to
investigate landscape-scale drivers of arthropod communities (Lafage et al. 2015, McCary et al. 2018,
Chatelain et al. 2023), (2) the finest spatial resolution of atmospheric UHI data available in our study
area, and (3) the scale at which the correlation between urban land-cover (i.e., the built-up area)
and the atmospheric UHI in Rennes remains low (Foissard et al. 2019a). As a complement, we
calculated habitat and temperature predictors within 200-m and 500-m radius buffers, to compare
landscape predictors across multiple scales.
We derived three landscape-habitat predictors based on land-cover data: (1) the proportion of
impervious surface (buildings, roads and parking sites), (2) the proportion of grassland (lawns and
meadows), and (3) patch isolation, known to be a relevant landscape-scale predictor of spider and
carabid communities in urban areas (Buchholz et al. 2018) (Figure 2.2). We estimated patch isolation
as the mean nearest-neighbour distance of each grassland patch from edge to edge using the
‘lsm_l_enn_cv’ function of the ‘landscapemetrics’ R-package (version 1.5.5 Hesselbarth et al., 2019)
in R software. Land-cover data was retrieved at five-meter resolution from the French portal for
geographic data (dataset OCSGE 2017; [Link]; accessed on 1 December 2021).
We derived one temperature predictor at the landscape scale – the mean atmospheric UHI. To assign
UHI intensities to our 36 sampling sites, we extracted the mean pixel (100 x 100 m) value within
circular buffers (i.e., 100, 200 and 500-m radius) around the sampling sites from the UHI-raster map,
using the ‘exact_extract’ function of the ‘exactextractr’ R-package (version 0.9.1; Baston et al., 2023)
in R software.

[Link] Local-scale predictors

For local-scale habitat predictors, we measured the percent cover of plant species and the height of
the vegetation within the triangle delineated by the three pitfall traps during each sampling session
(Figure 2.2). We averaged the values to obtain one final value per site and per predictor. These two
predictors are regularly used to characterise arthropod habitats (de la Flor et al. 2017). We also
calculated an index of management pressure based on the mowing regime, obtained by summing
the total number of cuts observed in the field during the sampling season (ranging from zero to six).
Management intensity is known to have a negative effect on the taxonomic diversity of spiders
(Buchholz et al. 2018).
Near-ground temperatures were recorded locally every 15 minutes during each trapping session
with a specific network of temperature loggers (Lascar EL-USB-2+; Tinytag Talk 2 TK-4023) placed
five centimeters above the ground surface, in the direct vicinity of the north-headed pitfall trap, at
each sampling site. In the same way as for the atmospheric UHI, we calculated the differences
between the daily minimum temperatures obtained from the coldest station of the near-ground
sensor network, used as a reference, and that of each of the other sites. We obtained a single mean
value of near-ground temperature by averaging daily differences throughout the study period.

42
 CHAPITRE 1

Figure 2.2 - Distribution of landscape and local predictors among four groups: landscape habitat, local habitat,
atmospheric UHI, and near-ground temperature.

[Link] Statistical analyses

[Link].1 Community indices
Pitfall traps overestimate active species. Therefore, we considered the sum of individual counts at
each site as a measure of species’ activity density (Topping and Sunderland 1992), but refer to
‘number of trapped individuals’ for the sake of clarity. Taxonomic and functional diversity were
quantified using the Hill numbers of orders Q = 0; 1; 2 with the ‘iNEXT.3D’ R-package (version 1.0.1;
Chao et al. 2021). This method allows to account for the possible influence of dominance patterns,
with indices of order Q0 being the most sensible to rare species (i.e., the actual species richness) and
Q2 the less sensible (Hill 1973). For functional indices, a species’ functional distance matrix was
calculated on multiple traits using the Gower method (Gower 1971) with the ‘daisy’ function of the
‘cluster’ R-package (version 2.1.4; Maechler et al., 2023). Dispersal and hunting guild were treated
as nominal whereas body size was a continuous numerical variable. Based on these functional
distances among species, the computed functional diversity indices quantify the effective number of
equally distinct virtual functional groups, which can be thus be considered as ‘functional species’
(Chao et al. 2021). Prior to analysis, we checked the absence of any significant spatial autocorrelation
for all indices using Moran’s test in the ‘spdep’ R-package (version 1.2.8; Bivand et al., 2023),
considering the nearest-neighbour distance (N = 3) as the metric and performing 599 permutations.
To ensure that diversity indices could reliably be compared between sites, we also checked sample
coverages and found more than 90% completeness at all sites. Therefore, we considered the
observed in the subsequent analysis and have not rarefied or interpolated individuals counts.

43
PARTIE II

[Link].2 Spider response to landscape and local predictors

To disentangle which of the habitat or temperature and spatial scale (i.e., landscape- or local-scale)
predictors mostly explained spider community indicators, we computed generalized linear mixed
models (GLMMs) using the ‘lme4’ R-package (Bates et al. 2015). Prior to the analysis, low correlations
among predictors were checked (r < 0.7; Appendix 2.3). We built distinct models, structured with a
single index of number of trapped individuals, species diversity or functional diversity as the response
variable. The taxonomic and functional indices calculated at Q0, Q1 and Q2 orders were tested
individually, resulting in a total of seven distinct models. We fitted each model with all predictors
listed in the Figure 2.2 and selected the best model among all possible candidate models using the
‘dredge’ function in ‘MuMin’ R-package (Bartoń 2016). Since habitat size, has previously been
stressed to modulate the establishment of habitat-specific assemblages in the urban matrix (Beninde
et al. 2015), we accounted for its possible effect in each GLMMs, by adding patch area as a random
categorical variable with levels ‘small’, ‘medium’ and ‘large’ defined so that the three groups are
composed of a balanced number of sites. Secondly, we estimated the contribution of each individual
predictor to the models, using the ‘[Link]’ R-package (Lai et al. 2022). All statistical analyses were
performed in R version 4.0.3.

2.1.3 Results
The total area suitable for sampling after spatial correlation analysis covered 6000 ha, i.e., 58% of
the entire extent of the study area. Moreover, 1685 ha of urban grasslands were located in suitable
areas, i.e., 82% of the total urban grasslands of the study area. The 36 selected grasslands ranged
from 0.1 to 4.7 ha in size (mean = 0.8 ± 1.0 ha) and the distance between them ranged from 181 m
to 11.36 km (mean = 4458 ± 2173 m). Grasslands were characterised by diverse vegetation
structures, with an average vegetation height ranging from 5 to 44cm (mean = 20 ± 11.5 cm) and a
vegetation cover ranging from 60 to 100% (mean = 90.5 ± 9.4 %). The mowing frequency ranged
from zero to six times during the sampling period.

[Link] Sampling results

In total, data from four sampling units out of the 756 (i.e., 0.5%) was not available due to damaged
traps and was estimated by the averaged data from the two other traps forming the triplet. The data
set includes 20,761 mature spider individuals assigned to 21 families and 137 species (Appendix 2.2).
Among them, the most represented families in terms of number of trapped individuals were
Linyphiidae (N = 9045; 44%), Lycosidae (N = 5462; 26%) and Tetragnathidae (N = 4937; 24%). The
species with the highest number of trapped individuals were Pachygnatha degeeri (N = 4913; 24%)
and Pardosa cf. tenuipes (N = 2603; 13%). We did not detect any spatial autocorrelation related to
spiders’ number of trapped individuals or to spiders’ taxonomic and functional diversity.

[Link] Effects of habitat and temperature on spider communities

Our models showed that the number of trapped individuals was significantly negatively related to
near-ground temperature (Table 2.1, Figure 2.3a). Species diversity at Q0 order (Table 2.1, Figure 2.3b)
was negatively related to the atmospheric UHI and to the mowing frequency ( Table 2.1, Figure 2.3c).

44
 CHAPITRE 1

Functional diversity at Q0 order was significantly positively related to local vegetation height ( Table
2.1, Figure 2.3d) and negatively related to local vegetation cover (Table 2.1, Figure 2.3d).

The hierarchical partitioning analysis showed that for the number of trapped individuals, near-
ground temperature alone significantly explained a quarter (24%) of total variation ( Figure 2.4a).
Regarding species diversity at a Q0 order, atmospheric UHI and the local habitat alone explained 18%
and 5% of the variance, respectively and these values decreased with increasing orders (Figure 2.4b).
For functional diversity, no predictors exceeded 5% of variance explained, even when cumulated by
variable type (e.g., landscape habitat or local habitat factors; Figure 2.4c). The same analysis
performed with landscape habitat and atmospheric UHI predictors calculated at larger scales (i.e.,
200-m and 500-m radius buffers) showed similar results than those obtained at the 100-m scale
(Appendix 2.4, Appendix 2.5). Therefore, we only consider results related to landscape predictors
calculated at a 100-m scale in the rest of the paper.

45
 PARTIE II

46
Table 2.1 - Relationships estimated by GLMMs (estimate ± standard error) between the number of trapped individuals, species diversity orders or functional diversity orders of
spiders, and predictors related to near-ground temperature, landscape habitat measured within 100-m-radius buffers, landscape temperature, local habitat or local temperature.
Only parameters related to predictors retained during model selection are displayed. Significant results are in bold (***: P-value < 0.001).

Near-
Impervious Open Atmospheric Vegetation Vegetation Mowing Marginal Conditional
Isolation ground
surface vegetation UHI cover height regime R² R²
temperature

-44.89 ± 19.81 ± -88.93 ± 62.98 ± 13.62 ± -106.04 ± -70.47 ± -220.98 ±

Number of trapped individuals 0.38 0.38
64.67 59.68 47.76 67.76 64.27 89.72 75.10 60.7 (***)

Species diversity
-1.73 ± -5.06 ± 1.04 -2.93 ±
Q0 - - - - - 0.71 0.71
1.08 (***) 0.83 (**)
Q1 - - - -1.09 ± 0.63 - - - - 0.08 0.08
-0.47 ±
Q2 - - - - - - - 0.03 0.07
0.67

Functional diversity
-0.58 ± 0.55 ±
Q0 - - - - 0.16 0.16 - - 0.35 0.35
(***) (***)
-0.25 ±
Q1 - - - - - - - 0.07 0.07
0.15
0.20 ±
Q2 - - - - - - - 0.06 0.06
0.13
 Chapitre 1

Figure 2.3 - Plots showing significant relationships between the number of trapped individuals, species diversity
or functional diversity of spiders, and predictors related to near-ground temperature, atmospheric UHI and local
habitat (i.e., mowing regime, vegetation height, vegetation cover). Confidence intervals at 95% are shown.

47
PARTIE II

Figure 2.4 - Relative variance explained independently by landscape- and local-scale predictors according to
GLMM models for spiders’ (a) number of trapped individuals, (b) taxonomic diversity (orders Q = 0, 1 and 2) and
(c) functional diversity indices (orders Q = 0, 1 and 2).

48
 Chapitre 1

2.1.4 Discussion
Understanding the factors that shape biodiversity patterns in environments facing global changes is
of prime importance for predicting and mitigating the adverse effects of global warming. Cities and
their related UHIs provide an excellent framework to conduct climate-related ecological studies.
However, atmospheric UHIs and land-cover artificialisation tend to co-vary, so that disentangling the
predictors from each other is a challenge. The present study proposes a reproductible approach that
partially de-correlates urbanisation from the UHI (Cabon et al. 2024b). Near-ground temperature
and the atmospheric UHI outperformed landscape and local habitat predictors in explaining
variations in arthropod communities. In addition, we went beyond this result by comparing for the
first time the relative weights of two temperature predictors (near-ground temperature and the
atmospheric UHI) that vary differently across the urbanisation gradient at two independent spatial
scales (100 meters vs. metric).
Based on the ecological characteristics of ground-dwelling spiders, our results show that 100 meters
is a valuable landscape scale to decouple the effect of land cover from that of the atmospheric UHI.
This result is partially supported by recent ecological studies. McGlynn et al. (2019) found
proportions of impervious surface (within 50-m buffers) decoupled from temperature across the Los
Angeles Basin. Similar results were obtained by Hamblin et al. (2018) with 200-m buffers. However,
these two studies only considered near-ground temperature when measuring correlations with land
cover. Our study finally provides first evidence that atmospheric UHI patterns can also be partially
decoupled from land cover data.

[Link] Effects of temperature at the landscape and local scales

An increasing number of studies at the community level not only focus on habitat-related predictors
to identify urban drivers, but also consider temperature (Meineke et al. 2017a, Hamblin et al. 2018,
McGlynn et al. 2019, Piano et al. 2020d). Yet, unlike land cover predictors that are almost
systematically assessed at the landscape and local scales, temperature remains mainly considered at
single spatial scales. However, near-ground temperature deviates from the atmospheric UHI (Oke et
al. 2017), as confirmed by the low correlation observed in our study (r = 0.46). The main difficulty in
studying temperature at different scales lies in the complexity of setting up devices at the scale of
the entire city and at each sampling site. Two sensor networks need to be deployed – one spread
evenly across the city to capture the atmospheric UHI, the other consisting of one (or more) logger
per sampling site to capture near-ground temperature. Last but not least, all this represents a large
number of devices and measurements associated with a cumbersome data analysis process. We
approximated near-ground temperature with a single sensor per site that provided a significant
predictor. However, an interesting approach to go further would be to upgrade the local sensor
network by multiplying the number of measurement points at each sampling site to obtain a measure
of local temperature heterogeneity.
To the best of our knowledge, this study is the first that attempts to specify the effects of
temperature simultaneously measured at the landscape and local scales. At the landscape scale, our
results are in accordance with our first expectation: variations in spider taxonomic diversity were
associated to the atmospheric UHI, explaining 18% of the variance alone. The observed shift in
taxonomic diversity reflects changes in community compositions suggesting the existence of a
filtering process in response to the atmospheric UHI. Species may be filtered by large areas of heat
acting as thermal barriers at the city’s edge, where only species physiologically able to cope with

49
PARTIE II

urban warming persist (e.g., urban-adapted heat-tolerant species; see Hamblin et al. 2017, Piano et
al. 2017), whereas others are extirpated. The analysis confirmed that atmospheric UHI is an
important driver constraining the species richness of spiders. Different mechanisms can lead to the
observed decrease in species richness with increased temperature. Heat-sensitive native species
could be replaced by a lower number of non-native heat-tolerant ones, suggesting a decrease in
species richness along with a high turnover rate (McKinney 2008). Alternatively, the reduced number
of species may result from the persistence of only a subset of the species pool able to cope with
warming conditions (Piano et al. 2017). For example, ant communities are better represented by
heat- and dry-tolerant species in urbanised habitats than in natural adjacent areas (Menke et al.
2011). Along the same urbanisation gradient as in the present study, Cabon et al. (2024) examined
changes in spider community functional composition and found a community-wide shift toward
thermophilic species, explained by the exclusion of heat-sensitive species in urban areas. Here
however, the absence of relationship between functional indices and the atmospheric UHI could be
explained by the lack of thermal attributes in the set of traits selected to compute the functional
diversity indices.
At the local scale, our results also confirm our theoretical framework. Near-ground temperature had
no effect on taxonomic or functional diversity, but was negatively associated with the number of
trapped individuals (24%). This result is in line with several recent studies that identified urban near-
ground temperature as a prevailing driver of arthropod abundance (Meineke et al. 2013, Hamblin et
al. 2018, McGlynn et al. 2019). Yet, unlike several studies reporting an increased abundance of
phytophagous arthropods with temperature elevation (Meineke et al. 2013, Dale and Frank 2014,
2018), our results show a negative relationship between the number of trapped individuals and
near-ground temperature. Although phytophagous arthropods and predator spiders display
opposite patterns to urban thermal conditions, our result is supported by recent studies on flying
arthropods (Hamblin et al. 2018, McGlynn et al. 2019), showing that temperature is associated with
decreased abundance. Since near-ground temperature was not associated with changes in
taxonomic or functional diversities, it might not be directly involved in a species filtering process
according to their ability to persist in urban environments. Instead, near-ground temperature may
determine the abundance patterns of the locally present species. Near-ground temperatures are
heterogeneous, and this variability matches the local perception range of most arthropods, i.e.,
about 20 cm (Pincebourde and Salle 2020). For example, the measured mean atmospheric UHI
spanned 3 °C over a distance of 5 km, whereas our temperature loggers recorded near-ground
temperatures spanning up to 5 °C. Since these temperature-related predictors are based on daily
minimum temperature records, they mostly depict nocturnal thermal situations. However, a daytime
experiment conducted by Pincebourde & Salle (2020) showed that temperatures can range up to 16
°C within less than 1 meter distance on leaf litter. Therefore, mobile ground-dwelling arthropods
might have plenty of opportunities to behaviourally thermoregulate over a local spatial scale in
grassland vegetation during daytime. Many ground-dwelling spiders are known to behaviourally
thermoregulate to avoid deleterious thermal conditions (Humphreys 1974, Cabon et al. 2023).
Therefore, species-specific thermoregulatory behaviour can lead to community-wide avoidance of
hot deleterious conditions. This idea is supported by previous results obtained by Shochat et al.
(2004), who hypothesized that local deficits of spider abundance resulted from behavioural reactions
(i.e., local emigration) in response to warmer microenvironments.

50
 Chapitre 1

[Link] Effect of habitat at the local scale

As already found in previous ecological studies on urban environments (Alaruikka et al. 2002,
Magura et al. 2010a, Buchholz et al. 2018), our results show that landscape habitat predictors did
not affect spider taxonomic and functional diversities. This suggests that habitats located in highly
urbanised areas can host diversified communities. Yet, it is worth pointing out that the scale selected
here to describe the landscape (100-m radius buffers) corresponds to the optimal scale allowing a
maximized local decorrelation of UHI and land cover, and can be interpreted as matching the home
range of mobile ground-dwelling spiders. Although this scale is the most suitable to answer our
research questions, relationships between spider community indices and landscape habitat
predictors may also be observed at larger scales, yet without opportunity to dissociate the effects of
temperature from those of habitat. Local habitat predictors outperformed landscape habitat
predictors in explaining spider taxonomic diversity. If we refer to the usual dichotomous framework
opposing the local and landscape scales (Beninde et al. 2015, Egerer et al. 2017, Buchholz et al.
2020), the urban landscape should filter species according to their capacity to disperse into
urbanised areas. Therefore, landscape descriptors are mainly referred to as drivers of the taxonomic
and functional diversities of urban communities. By contrast, local habitats should determine the
availability of resources needed for survival and reproduction, and control the establishment of
viable populations. Our results confirm the importance of local habitat for spiders, but suggest that
this predictor rather drive taxonomic diversity. Other studies have reported similar results, e.g., in
semi-natural habitats (Sarthou et al. 2014, Lafage et al. 2015), where local habitat features have been
found to be more significant drivers of spider species richness than landscape predictors. Similarly,
spider species richness was mainly related to local habitat characteristics such as the bare-ground
cover or the vegetation structure, in both non-urban (Pétillon et al. 2008) and urban (Argañaraz et
al. 2018) habitats.
Only predictors describing the local habitat explained the variation of spider functional diversity. This
could be explained by the fact that two out of the three traits used for calculating functional diversity
indices (body size and ballooning capacity) are related to dispersal capacities (Jenkins et al. 2007,
Entling et al. 2011). Species with a high dispersal capacity are generally more abundant in urbanised
areas (Piano et al. 2017, Merckx et al. 2018a). As discussed by Varet et al. (2011) and Argañaraz et
al. (2018), the inclusion of highly dispersive spider species in urban communities might explain the
absence of a decrease in taxonomic diversity with urbanisation. In addition, such a replacement of
species from the original species pool by highly dispersive ones at a high turnover rate might
maintain functional diversity in urban areas and lead to the functional uniformity of urban-adapted
communities. Based on this hypothesis, local habitat might become the major driver of spider
functional diversity, independently from the landscape context.
We observed a positive effect of vegetation height on spider functional richness. This result was quite
expected because the occurrence of a species within a specific microhabitat is intimately linked to
the hunting method practiced by the foraging guild it belongs to (Gibson et al. 1992, Barriga et al.
2010, Diehl et al. 2013, Rao 2017). As a result, different hunting guilds need different microhabitat
structures (Uetz 1979, 1991, Morris 2000, Bell et al. 2001). The vertical structure of herbaceous
vegetation can vary drastically among grasslands, depending in particular on the management
regime (Bell et al. 2001, Chollet et al. 2018). A high herbaceous vegetation provides more
opportunities than a low vegetation for the establishment of weaver species, because elements for
fixing webs (i.e., anchorage points) are more available (Uetz 1991, Mcnett and Rypstra 2000,
Langellotto and Denno 2004, Diehl et al. 2013). Nevertheless, the presence of high herbaceous
vegetation does not rule out the maintenance of a surface structure suitable for the establishment
51
PARTIE II

of wandering species (Maelfait and De Keer 1990, Uetz et al. 1999). In addition, high vegetation
should provide increased potential for niche differentiation through the diversification of
phytophagous prey (Morris 2000, Woodcock et al. 2009) and indirectly favour the co-occurrence of
diverse hunting guilds (Lyons et al. 2018), which could explain the observed enhanced functional
richness.
The local proportion of vegetation cover was significantly negatively related to spider functional
richness and evenness. In other words, the sites with a heterogeneous vegetation cover supported
more functionally diverse spider communities. Similar outcomes have been found in irregularly
managed habitats, which hosted functionally diverse communities composed of more web-building
and ambush-hunting species than intensively managed ones in urban (Buchholz et al. 2018) and
agricultural landscapes (Diehl et al. 2013). As for vertical structures, horizontal heterogeneity (e.g.,
vegetation cover, bare soil or litter) tends to support functionally more diverse communities
compared to homogeneous environments, in particular when habitat conditions provide
opportunities for multiple foraging guilds to co-occur (Uetz et al. 1999).
Although spider communities can be structured according to inter-annual climatic fluctuations, in
terms of activity (Bowden et al. 2018, Muster et al. 2020) and composition (Relys et al. 2002), the
results discussed here were obtained after a single year of sampling. Longer-term studies are
therefore needed to determine whether the effects we observed fluctuate over time or not.
However, in view of the particularly high temperatures for the region measured in 2022 and the
particularly intense UHI conditions for Rennes (HCBC 2023), we believe that these results should be
representative of the years to come, during which this type of heatwave is likely to be repeated and
intensified (HCBC 2023).

[Link] Implications for urban planning

The factors promoting urban biodiversity and the efficient scales of management action need to be
clearly identified in order to guide urban planning. Ecological studies of urban green spaces are
needed to identify the landscape- and local-scale factors that drive floristic and faunistic diversity
and to guide urban planning in the development of efficient conservation or restoration measures
(Lepczyk et al. 2017). In light of conservation measures, our results show that arthropod
communities can be enhanced by developing strategies at both local and landscape scales. At the
landscape scale, the atmospheric UHI was a prevailing filter of spider assemblages that sorted species
according to their ability to adapt and persist in warm urban habitats (Merckx and Van Dyck 2019,
Piano et al. 2020d). Therefore, increasing biodiversity requires that additional species can pass
through the urban thermal filter. Implementing green infrastructures is a widely used strategy that
efficiently generates cool islands (Norton et al. 2014, Manso et al. 2021). However, present
knowledge highlights the importance of sustaining connectivity between urban vegetated areas such
as grasslands, and thermal connectivity within cities to facilitate immigration and the establishment
and maintenance of species coming from rural surroundings. By highlighting the climate-biodiversity
co-benefits of green spaces, our results can be used to design planning solutions that meet the
ambitions set out in the roadmaps developed by local authorities (Rennes Métropole 2019). Our
results also show that local characteristics (temperature and habitat) are particularly promising
levers to enhance arthropod conservation in urban grasslands because they can promote diversity
locally (e.g., through the diversification of vegetation structures). In particular, the present study
shows that decreasing near-ground temperature and favouring high herbaceous vegetation coupled
with heterogeneous ground coverage maintained at a low frequency would increase the number of
52
 Chapitre 1

trapped individuals and diversity of spiders in urban grasslands. As previously demonstrated for
grassland vegetation in our study area (Chollet et al. 2018), developing measures locally to guide
grassland management practices appears as a first important step for arthropod conservation. Local
measures at the grassland scale are often easier to achieve than landscape-scale ones (e.g., related
to land-cover changes at a neighborhood scale). Furthermore, our results show that prioritising
grassland management measures could not only mitigate the loss of taxonomic diversity induced by
the intensification of the atmospheric UHI in cities, but could also promote the establishment of
more functionally diverse species pools.

53
PARTIE II

2.1.5 Appendices
Appendix 2.1 - Sampling site selection procedure

Prior to data collection, we performed a spatial analysis in order to identify suitable sampling areas
where land cover and atmospheric UHI were not significantly correlated. We based this sampling site
selection procedure on two maps: a raster map showing the proportions of built-up areas and a
raster map showing the intensity of the atmospheric UHI in 2020 (both at 100 m resolution). As the
proportion of built-up areas were identified as a major factor in atmospheric UHI intensity in Rennes
(Foissard et al. 2019a), we identified the areas where the two maps were least spatially correlated.
To do this, we proceeded as follows:
We performed a Spearman spatial correlation analysis to identify all pixels showing a non-significant
covariation between atmospheric UHI intensity and the proportion of built-up areas. This analysis
was performed using the ‘rasterCorrelation’ function from the ‘spatialEco’ package (version 1.3.7).
Correlations were calculated within a 500 m square sliding window (i.e., 25 pixels), centered on each
map pixel. This window size corresponds to a trade-off between a number of pixels larger enough to
calculate a reliable correlation (N = 25) and an area small enough (25 ha) to identify uncorrelated
areas. We extracted all pixels corresponding to non-significant correlations (p-value > 0.05), resulting
in a raster map indicating suitable (i.e., uncorrelated) areas for sampling. We selected all patches of
herbaceous vegetation (excluding agricultural fields) from the Rennes land cover map that
overlapped the suitable sampling area (see map Appendix 2.1 below). We identified 39 accessible
sampling sites encompassing a broad gradient of built-up cover proportions and UHI intensities.
Once fieldwork was completed in 2022, we repeated the first four steps with updated atmospheric
UHI data from 2022 corresponding to the arthropod sampling period, to check that all sampling sites
were still located in suitable (i.e., uncorrelated) areas. We removed three sites located in unsuitable
areas, resulting in 36 remaining sampling sites. This sampling site selection procedure was an
important prerequisite, as it determined the subsequent discrimination of individual effects of
landscape habitat and atmospheric UHI related to arthropod communities.

54
 Appendix 2.1 - Map of Rennes (black line) and its surroundings. Areas identified as suitable for sampling (uncorrelated) after spatial correlation

55
analysis are in purple (p-value > 0.05) and associated intersecting grasslands are in green. Impervious surface is displayed in grey. Black crosses
display sampling sites.
Chapitre 1
PARTIE II

Appendix 2.2 - Species list of spiders and related traits based on literature data (1/5).

Body length of
females (mm)
Ballooning
Hunting guild
(Roberts 1985,
Species (Blandenier and
1987, Macías- (Uetz et al. 1999, Cardoso
Fürst 1998, Bell
Hernández et al. et al. 2011)
et al. 2005)
2020, Nentwig
et al. 2023)
Agroeca brunnea 8.0 Ground hunter no
Agroeca inopina 7.5 Ground hunter no
Agyneta affinis 2.0 Wandering web builder no
Agyneta decora 2.5 Sheet web builder no
Agyneta mollis 2.0 Wandering web builder yes
Agyneta rurestris 2.4 Wandering web builder yes
Alopecosa cuneata 8.0 Ground hunter no
Alopecosa pulverulenta 10.0 Ground hunter yes
Antistea elegans 3.0 Sheet web builder no
Argenna subnigra 2.5 Ground hunter no
Asagena phalerata 5.0 Space web builder no
Atypus affinis 15.0 Ambusher yes
Aulonia albimana 4.5 Ground hunter yes
Bathyphantes gracilis 2.5 Sheet web builder yes
Centromerita bicolor 3.5 Wandering web builder yes
Centromerus sylvaticus 4.0 Sheet web builder yes
Ceratinella brevis 1.5 Wandering web builder no
Cicurina cicur 7.0 Ground hunter no
Civizelotes civicus 4.3 Ground hunter no
Clubiona diversa 5.0 Foliage runner no
Clubiona neglecta 8.0 Foliage runner no
Clubiona terrestris 7.0 Foliage runner no
Cnephalocotes obscurus 1.9 Wandering web builder yes
Collinsia inerrans 2.5 Wandering web builder yes
Cryptachaea blattea 3.2 Space web builder yes
Dicymbium nigrum 2.5 Wandering web builder yes
Dicymbium tibiale 2.6 Wandering web builder no
Diplocephalus latifrons 2.0 Wandering web builder no
Diplocephalus picinus 1.0 Wandering web builder no

56
 Chapitre 1

Appendix 2.2- Species list of spiders and related traits based on literature data (2/5).
Body length of
females (mm)
Ballooning
Hunting guild
(Roberts 1985,
Species (Blandenier and
1987, Macías- (Uetz et al. 1999, Cardoso
Fürst 1998, Bell
Hernández et al. et al. 2011)
et al. 2005)
2020, Nentwig
et al. 2023)
Diplostyla concolor 3.0 Sheet web builder yes
Drassodes cf. lapidosus 18.0 Ground hunter no
Drassodes pubescens 9.0 Ground hunter no
Drassyllus lutetianus 7.5 Ground hunter yes
Drassyllus praeficus 6.0 Ground hunter yes
Drassyllus pusillus 5.0 Ground hunter yes
Drassyllus villicus 6.9 Ground hunter yes
Dysdera crocata 15.0 Ground hunter no
Dysdera erythrina 10.0 Ground hunter no
Enoplognatha latimana 6.0 Space web builder no
Enoplognatha mordax 8.5 Space web builder yes
Enoplognatha testacea 3.2 Space web builder no
Enoplognatha thoracica 4.0 Space web builder no
Episinus truncatus 4.0 Space web builder no
Eratigena picta 7.0 Sheet web builder no
Erigone atra 2.8 Wandering web builder yes
Erigone dentipalpis 2.6 Wandering web builder yes
Ero furcata 3.2 Stalker yes
Ero tuberculata 4.0 Stalker no
Euophrys frontalis 5.0 Stalker no
Gnathonarium dentatum 3.0 Wandering web builder yes
Gongylidiellum vivum 1.9 Wandering web builder yes
Hahnia nava 2.0 Sheet web builder no
Haplodrassus dalmatensis 6.5 Ground hunter no
Haplodrassus signifer 9.0 Ground hunter no
Hypomma cornutum 2.9 Wandering web builder yes
Lasaeola prona 2.5 Space web builder no

57
PARTIE II

Appendix 2.2- Species list of spiders and related traits based on literature data (3/5).
Body length of
females (mm)
Ballooning
Hunting guild
(Roberts 1985,
Species (Blandenier and
1987, Macías- (Uetz et al. 1999, Cardoso
Fürst 1998, Bell
Hernández et al. et al. 2011)
et al. 2005)
2020, Nentwig
et al. 2023)
Lathys humilis 2.5 Ground hunter yes
Maso sundevalli 1.8 Wandering web builder no
Mermessus trilobatus 2.1 Wandering web builder no
Micaria albovittata 8.0 Ground hunter no
Micaria micans 4.1 Ground hunter yes
Micrargus herbigradus 1.9 Wandering web builder yes
Micrargus subaequalis 2.0 Wandering web builder yes
Microneta viaria 3.0 Sheet web builder yes
Monocephalus fuscipes 2.2 Wandering web builder yes
Neottiura bimaculata 3.3 Space web builder yes
Neottiura suaveolens 2.3 Space web builder no
Neriene clathrata 5.0 Sheet web builder no
Oedothorax apicatus 3.3 Wandering web builder yes
Oedothorax fuscus 2.9 Wandering web builder yes
Oedothorax retusus 3.0 Wandering web builder yes
Ostearius melanopygius 2.6 Wandering web builder no
Ozyptila brevipes 4.0 Ambusher no
Ozyptila praticola 4.0 Ambusher no
Ozyptila sanctuaria 4.0 Ambusher yes
Ozyptila simplex 5.0 Ambusher yes
Pachygnatha clercki 7.0 Orb weaver yes
Pachygnatha degeeri 3.7 Orb weaver yes
Palliduphantes ericaeus 1.9 Sheet web builder yes
Palliduphantes insignis 2.5 Sheet web builder no
Palliduphantes pallidus 2.3 Sheet web builder no
Panamomops sulcifrons 1.5 Wandering web builder no
Pardosa amentata 8.0 Ground hunter yes
Pardosa cf. tenuipes 6.5 Ground hunter yes

58
 Chapitre 1

Appendix 2.2- Species list of spiders and related traits based on literature data (4/5).
Body length of
females (mm)
Ballooning
Hunting guild
(Roberts 1985,
Species (Blandenier and
1987, Macías- (Uetz et al. 1999, Cardoso
Fürst 1998, Bell
Hernández et al. et al. 2011)
et al. 2005)
2020, Nentwig
et al. 2023)
Pardosa hortensis 5.5 Ground hunter no
Pardosa nigriceps 7.0 Ground hunter yes
Pardosa palustris 6.0 Ground hunter yes
Pardosa prativaga 6.0 Ground hunter yes
Pardosa pullata 6.0 Ground hunter yes
Pardosa saltans 5.7 Ground hunter no
Pardosa vittata 6.3 Ground hunter no
Pelecopsis parallela 2.0 Wandering web builder yes
Philodromus praedatus 6.0 Ambusher yes
Phlegra fasciata 7.0 Stalker yes
Phrurolithus festivus 3.0 Ground hunter no
Phrurolithus minimus 3.5 Ground hunter no
Phycosoma inornatum 3.0 Space web builder no
Piratula latitans 5.0 Ground hunter no
Pisaura mirabilis 15.0 Ambusher yes
Pocadicnemis juncea 2.2 Wandering web builder no
Porrhomma pygmaeum 2.4 Sheet web builder yes
Prinerigone vagans 3.2 Wandering web builder yes
Psilochorus simoni 2.5 Space web builder no
Robertus arundineti 2.5 Space web builder yes
Robertus lividus 4.0 Space web builder no
Saaristoa abnormis 4.0 Sheet web builder no
Scotina palliardii 3.5 Ground hunter no
Stemonyphantes lineatus 6.4 Wandering web builder no
Styloctetor compar 2.3 Sheet web builder no
Synageles venator 4.0 Stalker no
Tallusia experta 3.6 Sheet web builder yes
Tapinocyba praecox 1.8 Wandering web builder yes

59
PARTIE II

Appendix 2.2- Species list of spiders and related traits based on literature data (5/5).
Tenuiphantes flavipes 2.5 Sheet web builder yes
Tenuiphantes tenuis 3.2 Sheet web builder yes
Tibellus oblongus 10.0 Ambusher yes
Tiso vagans 2.2 Wandering web builder yes
Trachyzelotes pedestris 8.0 Ground hunter no
Trichoncus saxicola 2.1 Wandering web builder no
Trochosa ruricola 14.0 Ground hunter no
Trochosa terricola 14.0 Ground hunter yes
Walckenaeria acuminata 4.0 Wandering web builder yes
Walckenaeria antica 2.6 Wandering web builder yes
Walckenaeria atrotibialis 2.8 Wandering web builder no
Walckenaeria cucullata 2.7 Wandering web builder no
Walckenaeria incisa 3.2 Wandering web builder no
Xysticus acerbus 8.0 Ambusher no
Xysticus cristatus 8.0 Ambusher yes
Xysticus erraticus 8.0 Ambusher no
Xysticus kochi 8.0 Ambusher yes
Zelotes apricorum 9.0 Ground hunter no
Zelotes electus 5.5 Ground hunter no
Zelotes latreillei 8.0 Ground hunter yes
Zelotes petrensis 7.0 Ground hunter no
Zelotes tenuis 9.6 Ground hunter no
Zodarion italicum 3.0 Ambusher no
Zodarion rubidum 4.4 Ambusher no
Zora spinimana 6.5 Ground hunter yes

60
 Chapitre 1

Appendix 2.3 - Matrix of correlations among predictors.

Veg. Manag. Veg. Patch Open Imp. Atm. Near-gr.

Cov. Freq. height Isol. Veg. Surf. UHI Temp.
Landscape predictors at 100m scale
Vegetation cover 1.00 0.14 0.30 0.04 -0.06 0.22 0.22 -0.25
Management freq. 0.14 1.00 -0.66 -0.17 -0.22 0.27 0.10 0.26
Vegetation height 0.30 -0.66 1.00 0.08 0.45 -0.33 -0.32 -0.54
Patch isolation (100m) 0.04 -0.17 0.08 1.00 -0.19 -0.02 0.05 -0.20
Open vegetation (100m) -0.06 -0.22 0.45 -0.19 1.00 -0.53 -0.38 -0.40
Impervious surface (100m) 0.22 0.27 -0.33 -0.02 -0.53 1.00 0.64 0.40
Atmospheric UHI (100m) 0.22 0.10 -0.32 0.05 -0.38 0.64 1.00 0.46
Near-ground temperature -0.25 0.26 -0.54 -0.20 -0.40 0.40 0.46 1.00
Landscape predictors at 200m scale
Vegetation cover 1.00 0.14 0.30 0.15 -0.03 0.20 0.22 -0.25
Management freq. 0.14 1.00 -0.66 -0.17 -0.22 0.19 0.10 0.26
Vegetation height 0.30 -0.66 1.00 0.12 0.40 -0.28 -0.33 -0.54
Patch isolation (200m) 0.15 -0.17 0.12 1.00 -0.42 -0.14 -0.09 -0.11
Open vegetation (200m) -0.03 -0.22 0.40 -0.42 1.00 -0.48 -0.35 -0.33
Impervious surface (200m) 0.20 0.19 -0.28 -0.14 -0.48 1.00 0.69 0.34
Atmospheric UHI (200m) 0.22 0.10 -0.33 -0.09 -0.35 0.69 1.00 0.45
Near-ground temperature -0.25 0.26 -0.54 -0.11 -0.33 0.34 0.45 1.00
Landscape predictors at 500m scale
Vegetation cover 1.00 0.14 0.30 0.04 -0.03 0.34 0.22 -0.25
Management freq. 0.14 1.00 -0.66 0.22 -0.06 0.08 0.08 0.26
Vegetation height 0.30 -0.66 1.00 -0.19 0.21 -0.19 -0.33 -0.54
Patch isolation (500m) 0.04 0.22 -0.19 1.00 -0.53 -0.37 -0.27 -0.17
Open vegetation (500m) -0.03 -0.06 0.21 -0.53 1.00 -0.16 -0.31 -0.29
Impervious surface (500m) 0.34 0.08 -0.19 -0.37 -0.16 1.00 0.83 0.36
Atmospheric UHI (500m) 0.22 0.08 -0.33 -0.27 -0.31 0.83 1.00 0.45
Near-ground temperature -0.25 0.26 -0.54 -0.17 -0.29 0.36 0.45 1.00

61
 PARTIE II

62
Appendix 2.4 - Relationships estimated by GLMMs (estimate ± standard error) between number of trapped individuals, species diversity orders or functional diversity orders of
spiders, and predictors related to near-ground temperature, landscape habitat measured within 200-m-radius buffers, landscape temperature, local habitat or local temperature.
Only parameters related to predictors retained during model selection are displayed. Significant results are in bold (***: P-value<0.001; **: 0.001<P-value<0.01; **: 0.01<P-
value<0.05).

Impervious Open Near-

Isolation Atmospheric Vegetation Vegetation Mowing Marginal Conditional
surface vegetation ground
(200m) UHI (200m) cover height regime R² R²
(200m) (200m) temperature

Number of trapped -24.99 ± -24.51 ± -108.53 ± 16.50 ± 32.88 ± -96.90 ± -86.83 ± -205.75 ±
0.37 0.37
individuals 74.94 68.10 59.26 73.11 65.35 88.02 76.02 58.32 (***)

Species diversity
-2.40 ± 1.17 -2.82 ± 0.97 -5.23 ± 1.04 -3.72 ± 0.80
Q0 -1.85 ± 1.12 - - - 0.63 0.63
(*) (**) (***) (***)
Q1 - - - -1.08 ± 0.62 - - - - 0.08 0.08
Q2 - - 0.49 ± 0.43 - - - - - 0.04 0.04

Functional diversity
-0.58 ± 0.55 ±
Q0 - - - - - - 0.35 0.35
0.16 (***) 0.16 (***)
Q1 - - - - - - -0.25 ± 0.15 - 0.07 0.07
0.20 ±
Q2 - - - - - - - 0.06 0.06
0.13
 Appendix 2.5 - Relationships estimated by GLMMs (estimate ± standard error) between number of trapped individuals, species diversity orders or functional diversity orders of
spiders, and predictors related to near-ground temperature, landscape habitat measured within 500-m-radius buffers, landscape temperature, local habitat or local temperature.
Only parameters related to predictors retained during model selection are displayed. Significant results are in bold (***: P-value<0.001; **: 0.001<P-value<0.01; **: 0.01<P-
value<0.05).

Open
Impervious Isolation Atmospheric Vegetation Vegetation Mowing Near-ground Marginal Conditional
vegetation
surface (500 m) (500 m) UHI (500 m) cover height regime temperature R² R²
(500 m)

Number of trapped 133.66 ± 116 ± 105.60 ± -36.48 ± -11.11 ± -37.40 ± -165.21 ±

15.09 ± 88.95 0.36 0.36
individuals 70.39 78.58 94.83 70.16 95.83 75.37 63.36 (***)

Species diversity
-6.49 ± 0.88 -1.37 ± -4.25 ±
Q0 - - - - - 0.61 0.61
(***) 1.16 1.10 (***)
-1.56 ± -2.09 ± 0.75
Q1 - -1.96 ± 0.84 - - - - 0.20 0.20
0.83 (*) (**)
Q2 - -0.61 ± 0.43 - - - - - - 0.05 0.05

Functional diversity
-0.50 ± 0.15 -0.41 ±
Q0 - - - - - - 0.36 0.36
(***) 0.15 (**)
-0.25 ±
Q1 - - - - - - - 0.07 0.07
0.15
Q2 - - - - - 0.20 ± 0.13 - - 0.06 0.06

63
Chapitre 1
PARTIE II

2.2 Chapitre 2 : Effets des facteurs liés à l’habitat et à la

température sur les communautés de staphylins et de carabes
en ville

Résumé
Ce deuxième chapitre examine l'influence des variables environnementales urbaines sur les
communautés de staphylins et de carabes, complétant ainsi les résultats sur les araignées détaillés
dans le premier chapitre. Contrairement aux araignées, les communautés de staphylins et de carabes
ne sont pas principalement affectées par les variables liées à la température. La composition des
espèces de staphylins, groupe peu examiné en écologie urbaine, est influencée par l'habitat et la
température, tandis que leur diversité fonctionnelle est positivement corrélée à la hauteur de la
végétation herbacée uniquement. En revanche, les communautés de carabes n'ont montré aucun
lien significatif avec les variables environnementales testées. Cette étude souligne les différences de
réponses des communautés d'ectothermes aux environnements thermiques entre groupes
taxonomiques. L'absence de réponse observée chez les carabes suggère que d'autres paramètres
environnementaux, non étudiés ici, pourraient influencer les communautés de ce taxon. Le nombre
relativement faible d'individus capturés peut également expliquer l'absence de réponses. Chez les
staphylins, les recommandations visant à promouvoir la diversité fonctionnelle au sein des
assemblages comprennent la réduction de la fréquence de tonte et/ou l'augmentation de la hauteur
de coupe. Cependant, des recherches plus approfondies sur les caractéristiques spécifiques aux
espèces, et en particulier sur les capacités de dispersion des staphylins seraient nécessaires pour
mieux comprendre l'impact de l'urbanisation sur les communautés d'arthropodes.

Article en préparation pour Diversity

64
 Chapitre 2

Effects of habitat and temperature-related factors on urban staphylinid and

carabid communities

Valentin Cabon1,2, Yann Laurent3, Romain Georges1,2, Vincent Dubreuil2,4, Hervé Quénol2,4, Benjamin
Bergerot1,2

1. University of Rennes, CNRS, UMR 6553 ECOBIO, Rennes, France

2. LTSER ZA Armorique, F-35000 Rennes, France
3. Institut Agro, Rennes, France
4. University of Rennes 2, CNRS, UMR 6554 LETG, Rennes, France

Abstract
This study examines the influence of urban environmental variables on rove and ground beetle
communities, complementing previous research on spiders in the same area. Unlike spiders,
staphylinid and carabid communities are not primarily affected by temperature-related variables.
Staphylinids, often overlooked in urban ecology, showed species composition variation to be
influenced by habitat and temperature, whereas their functional diversity was positively correlated
with herbaceous vegetation height only. In contrast, carabid communities exhibited no significant
relationship with the tested environmental variables. This study underscores the taxon-dependent
nature of ectotherm community responses to thermal environments. A lack of responses observed
in carabids suggests that other environmental parameters, non-investigated here, might drive
communities in this taxon, whereby the relatively low number of individuals caught might also
explain the lack of pattern. In staphylinids, recommendations to promote functional diversity within
assemblages include reducing the mowing frequency and/or increasing the cutting height. However,
further investigation into species-specific traits, and particularly dispersal capacities in staphylinids,
is needed to advance our understanding of urbanization's impact on arthropod communities.

Key words
activity density, Staphylinidae, Carabidae, functional diversity, taxonomic diversity, urban heat island

65
PARTIE II

2.2.1 Introduction
Urbanisation strongly influences the environmental conditions experienced by city-dwelling
organisms at landscape and local scales (Norton et al. 2014, Philpott et al. 2014, Otoshi et al. 2015,
Kyrö et al. 2018). In particular, climatic conditions are altered through the proliferation of heat-
absorbing materials and the canyon-like morphology of streets, leading to an increase of mean
temperatures within city centres, as commonly referred to as the urban heat islands (UHI)
phenomenon (Oke et al. 2017, Foissard et al. 2019a). In the previous chapter of this manuscript
(chapter 1), we aimed to disentangle the effect of habitat- and temperature-related variables along
a rural-urban gradient on spider communities. In particular, we quantified habitat and temperature
variables at the landscape scale (i.e., considering 100-m to 500-m-radius buffers) and at the local
scale (i.e., 1-m). This study highlighted that the number of caught spider individuals and the
taxonomic diversity were related to near-ground thermal conditions and atmospheric warming,
respectively. This suggests that climatic spatial variability, more than to land cover, impacts spider
communities. Furthermore, intense UHI conditions were found to favour communities composed of
in average smaller and more thermophilic species than those of cooler rural areas, as described in
the chapter 3 (Cabon et al. 2024b).
Yet, investigations on other urban-dwelling arthropods are essential to draw a representative
assessment of the footprint left by urbanisation on biodiversity (Piano et al. 2020a), notably by the
UHI. Among beetle dwellers of soil surface, staphylinids and carabids are the most abundant groups
in European ecosystems and are recognised as valuable biological indicators of anthropic pressures
(Bohac 1999, Rainio and Niemelä 2003, Michaels 2007). Along urbanization gradients, both groups
are highly diverse and respond rapidly to urban environmental stressors, such as habitat loss or
fragmentation (Niemelä et al. 2002, Deichsel 2006, Vergnes et al. 2012, Magura et al. 2013). Further,
as ectothermic animals, beetles are likely to be constrained by temperature variations induced by
UHI, with (Piano et al. 2017, Lövei and Magura 2022a).
Although staphylinids can reach abundance levels similar to those of spiders (Irmler and Lipkow
2018), the deficit of community ecology studies focusing on this taxon may be explained by the
complex identification of individuals at the species level. Moreover, the knowledge about species’
ecology and life history remains much more limited than for carabids (Deichsel 2006), making
functional approaches less straightforward. If carabids have been extensively investigated in the
urban context (Magura and Lövei 2021), staphylinids were mainly studied in agricultural landscapes
due to their potential for crop protection, and only few studies aimed to test how this taxon is
constrained by the urban environment (but see Deichsel 2006, Vergnes et al. 2012, 2014, Magura et
al. 2013).
These few previous researches, conducted on urban woodlots, have shown that the community
composition of both, staphylinid and carabid beetles, can be affected by landscape fragmentation
within the urban matrix (Deichsel 2006, Croci et al. 2008, Vergnes et al. 2012). At the local scale (i.e.
the woodlot patch scale), the increase in disturbance frequency (e.g. through management
measures), was hypothesized to homogenise urban forests patches, removing microhabitats needed
by some specialist carabid species (Niemelä et al. 2002). Regarding staphylinids, the intensification
of management practices in urban green spaces was associated with a decrease in species richness,
doubtlessly due to a reduction of plant debris, preventing the establishment of phytodetriticol
species and leading to biotic homogenization (Magura et al. 2013). Responses to microhabitat

66
 Chapitre 2

structure are particularly expected in staphylinids because they harbour hyper-divers feeding habits
and many species are microhabitat specialists depending of particular structures such as dead wood
or carcass for decomposers, or even the presence of fungi in the case of fungivorous species (Irmler
and Lipkow 2018).
To date, climatic variables were little considered to predict patterns of abundance, diversity or
composition within staphylinid and carabid urban communities (Piano et al. 2020a). In this study, we
first compared the community compositions in both taxa across 36 urban grasslands clustered in
three categories of urbanization intensity, according to the landscape and local environmental
characteristics. In a second step, we disentangled the effect of the landscape factors from those of
the local ones, to explain activity-density and diversity (i.e., taxonomic and functional) patterns
among staphylinid an carabid communities. Under “landscape factors”, we dissociated the role of
atmospheric temperature (i.e., UHI-induced warming) from the role of habitat loss (i.e., cover of
impervious surface). Under “local factors”, we distinguished microclimatic temperature (i.e., near-
ground warming) from micro-habitat structure (i.e., herbaceous vegetation height and density).

2.2.2 Material and methods

[Link] Sampling sites
We conducted this study in Rennes, a city of 223,000 inhabitants in northwestern France. Located
70 km from the sea, Rennes experiences a temperate oceanic climate. In 2022, the mean annual
temperature was 13.7 °C, with an average winter temperature of 7.7 °C and an average summer
temperature of 19.6 °C (Saint-Jacques airport station). Despite its mild climate, Rennes frequently
experiences UHI events. From 2004 to 2019, intense UHI conditions (with temperatures exceeding
4 °C above the rural surrounding) accounted for over 17% of the total nights. In certain months (i.e.,
September) this proportion increased to 30% (Dubreuil et al. 2020). We identified appropriate
sampling areas where land cover and the atmospheric UHI were uncorrelated, by initially performing
a spatial correlation analysis, following the methodology outlined in the Appendix 2.1 of the Chapter
1. This site selection process allowed us to distinguish the separate impacts of landscape habitat and
atmospheric UHI predictors on spider communities.
We used sensors from Rennes Urban Network ([Link] to characterize the
atmospheric UHI. Since 2020, 30 DAVIS Vantage-Pro-2 automatic weather stations and 93 connected
temperature sensors (Rising-HF- RHF1S001) collect temperature data at 60- and 15-minute
frequencies, respectively, across the city centre and the peri-urban and rural areas of Rennes. We
calculated the daily UHI intensity by the difference between the minimum daily temperature data at
a reference cold station located outside the urban area (Melesse - about 11 km in the north of
Rennes, 48°12′18.1″N 1°40′52.8″W) and the minimum daily temperature data recorded by all other
sensors. These discrete measures of daily UHI intensity were then interpolated by inverse distance
weighting, to obtain maps of estimated values covering the entire city extent at 100-m resolution
(Dubreuil et al. 2020). We averaged the daily UHI intensity values to the whole study period (from
1st March to 30th September 2022) to obtain a single raster map of mean UHI intensity ( Figure 2.1;
Chapter 1). Following our site selection procedure (Appendix 2.1 ; Chapter 1), we selected 36 sampling
sites (Figure 2.1; Chapter 1) located on grasslands belonging to public parks, community gardens,
green spaces adjacent to educational buildings, or private gardens.

67
PARTIE II

[Link] Arthropod sampling

Staphylinids and carabids were sampled using three pitfall traps per sampling sites, during a two-
week period each month, from March to September 2022. The traps were strategically positioned 5
meters apart in a triangular formation to optimize capture rates (Ward et al. 2001) and were oriented
towards the north, southeast, and southwest directions. Traps were made of plastic cups measuring
85 mm in diameter and 115 mm in height. Each trap was filled with 150 mL of saline solution (at a
concentration of 100 g·L-1) to enhance insect preservation, along with a drop of neutral soap to
prevent floating. Following each sampling session, staphylinid and carabid individuals were sorted
and placed into separate vials containing 70% ethanol for storage. Staphylinids were identified to the
species level using Lott (2009) and Lott and Anderson (2011), except for 12 species that have been
considered as morphospecies (Appendix 2.7). Carabid individuals were identified using Jeannel (1941).
To investigate the functional responses of staphylinid and carabid communities, we retrieved trait
information related to feeding group and body size for each species from the literature ( Appendix 2.7,
Appendix 2.8). If feeding information was unavailable for a particular species, we inferred based on
the ecological characteristics of other species within the same genus, where available. Otherwise,
the feeding group was categorized as unknown (Clough et al. 2007).

[Link] Community indices

To compute community indices, the raw arthropod data was pooled by site to obtain a total
abundance value per species and site. Independently for staphylinids and carabids, we computed (1)
the total number of collected individuals per site (i.e., the activity density), (2) the taxonomic diversity
using the species richness and Shannon diversity, and (3) the functional diversity using equivalent
indices computed based on the two functional traits considered here (i.e., feeding group and body
size). To correct for bias due to potential difference of sampling coverages among sampling sites, we
computed the taxonomic and functional diversity indices using the function ‘estimate3D’ from the
R-package ‘iNEXT.3D’ (Chao et al. 2021). This method enables to set a reference sampling coverage
(we used the default value ‘0.95’) and to calculate rarefaction- or extrapolation-based diversity
indices, which can reliably be compared.

[Link] Environmental variables

To characterise the habitat and temperature at each sampling sites, two variables were measured at
the landscape scale: the proportion of impervious surface as a proxy for urbanisation and the mean
UHI intensity (° C) calculated over the entire study period (as a proxy for urban warming). We
computed these landscape variables independently by choosing circular buffers of respectively 100
m, 200 m, 500 m and 1000 m, in order to compare their effects among multiple scales (Philpott et
al. 2014). Three variables were considered at the local scale: the vegetation cover, the vegetation
height (cm) and a measure of the air temperature (° C) as experienced by arthropods just above the
ground surface (i.e., 5 cm).

68
 Chapitre 2

[Link] Data analysis

[Link].1 Classification of study sites
To establish a typology of sites according to their contrasts in environmental characteristics, we
performed a hierarchical clustering analysis (HCA) based on all scaled environmental variables. We
used the ‘HCPC’ function and the ‘Ward’ method to construct the tree in the ‘FactoMineR’ package
(version 2.7) (Lê et al. 2008). The optimal number of clusters was determined according to the
partition with the higher relative loss of inertia (Lê et al. 2008). Once the sites were grouped into
clusters, we performed a silhouette analysis to check the agreement of individual sites with their
own cluster, using the ‘silhouette’ function from the ‘cluster’ package (version 2.1.4) (Maechler et
al. 2013). The site classification resulted in 3 clusters of sites defined as ‘high vegetated rural’, ‘low
vegetated rural’ and ‘low vegetated urban’. The mean scaled values and standard deviations of
environmental variables into three clusters after classification of sampling sites are reported in
Appendix 2.6.
[Link].2 Community composition analysis
The difference in community composition among site clusters was analysed using non-metric
multidimensional scaling (NMDS, Bray–Curtis dissimilarity, ‘vegan’ package) and pairwise
permutational multivariate analysis of variance (9999 permutations) using the
‘[Link]’ function from the ‘RVAideMemoire’ package (version 0.9-83-3). Finally, to
test for a potential environmental effect on communities’ functional composition, we computed
community weighted means of numerical trait (i.e., body size) and tested whether significant
differences in body size CWM existed between clusters by performing a one-way ANOVA using the
‘aov’ function, followed by a post-hoc Tukey test using the ‘Tukey-HSD’ function.
[Link].3 Community diversity analysis
To determine the main predictors of staphylinid and carabid community indices, we conducted a
variation partitioning analysis to separate the effects of habitat and temperature at different spatial
scales (i.e., landscape- or local-scale predictors) with the ‘vegan’ package (version 2.6.4; Oksanen et
al. 2022). We separately sorted the response indices of staphylinids and carabids among three
independently analysed response datasets: activity density, taxonomic diversity, and functional
diversity. Predictors were sorted among four datasets: local habitat (composed of two datasets:
vegetation cover and vegetation height), near-ground temperature, landscape habitat, and
atmospheric UHI. Since one dataset contained more than a single predictors, we reported the
adjusted R² as an estimator of the explained variation to prevent inflated R² values (Peres-Neto et al.
2006). We identified significant explanatory datasets by performing permutation tests for
redundancy analysis (RDA), using the ‘[Link]’ function (Legendre et al. 2011). In case a
community dataset was significantly linked to an explanatory dataset, we additionally ran a GLM
fitted with the involved variables, to determine how single community indices were linked to
predictors.
All statistical analyses were performed in R version 4.0.3.

69
PARTIE II

2.2.3 Results
In total, 3317 staphylinid individuals of 121 species and 505 carabid individuals of 60 species were
included in the analysis. The correlation matrix between predictors is provided in the at the end of
this Chapter (Appendix 2.9).

[Link] Community composition

The results of NMDS showed that the community composition of staphylinids and carabids differed
among clusters (staphylinids: stress = 0.215, p = 0.005; carabids: stress = 0.210, p = 0.003). The
communities from the two rural clusters were not significantly different from each other, but
significantly differed from the ‘short vegetated urban’ cluster. In staphylinids, the community
composition was significantly structured by all the tested variables (Figure 2.5). In carabids instead,
none of the environmental variables explained the community structure.

Figure 2.5 - Results of NMDS ordination (Bray–Curtis distance) based on staphylinid (A) and carabid (B) datasets.
Only significant environmental variables are displayed (p < 0.05).

[Link] Density and diversity indices

Variation partitioning showed that only a small percentage of functional diversity was constrained
by the environment, and habitat structure was the only significant predictor in staphylinids. This
result was found with a 100-m scale considered to calculate landscape and UHI data sets included in
the analyze (Figure 2.6). The results remained similar when these variables were considered at larger
scales to perform variation partitioning (i.e., 200 m, 500 m and 1000 m; Appendix 2.10). In contrast,
none of the predictor datasets significantly explained the variance of carabids, neither in activity
density, nor in taxonomic or in functional diversity with any of the scale considered to calculate
landscape and UHI variables (Figure 2.6, Appendix 2.10)

70
 Chapitre 2

Figure 2.6 - Venn diagrams of (a) the activity density, (b) the taxonomic diversity, and (c) the functional diversity
of staphylinids (left column) and carabids (right column). The predictor datasets are represented by coloured
ellipses where the landscape habitat within a 100-m radius buffer is dark green (upper left), local habitat is light
green (upper right), the atmospheric UHI within a 100-m radius buffer is dark blue (lower left) and near-ground
temperature is light blue (lower right). Variation proportion lower than 1% are not displayed. Significant results
are underlined.

The GLMs indicated that only the functional diversity index was significantly linked to the local
habitat, and that vegetation height, not vegetation cover, positively affected the functional diversity
index (coeff. = 0.038, p < 0.001, adjusted R² = 0.38; Figure 2.7). Finally, the community functional
composition assessed by the CWM index of body size did not relate to any of the tested
environmental variable, neither in staphylinids, nor in carabids.

71
PARTIE II

Figure 2.7 - Significant relationship between the functional diversity (q1) of staphylinids and the vegetation
height. Confidence intervals at 95% are shown.

2.2.4 Discussion
This study explores which urban environmental variables affect rove and ground beetle communities
and these outcomes directly complement previous results obtained on spiders in the same study
area (see Cabon et al. 2024). We show that unlike spiders, staphylinid and carabid communities are
not primarily affected by temperature-related variables. Staphylinid remains to date a rarely
considered in urban ecology studies, although our sampling results indicate that they are a diverse
and abundant arthropod group of urban grasslands. If the variation of the staphylinid species
composition was already found to be impacted by urbanization (Deichsel 2006), our study provides
new outcomes by investigating the relative importance of underlying variables related to habitat and
temperature, at multiple scales.

[Link] Effect of habitat

Landscape variables (i.e., 72rbanization and UHI) had no effects on staphylinids’ activity density and
diversity. Instead, we found the functional diversity of staphylinids to be positively linked to the
height of the herbaceous vegetation layer, locally. This pattern was already identified in other
arthropods (e.g., spiders; see results of chapter 1 and Alaruikka et al. 2002, Magura et al. 2010,
Buchholz et al. 2018), where local habitat predictors performed better to predict functional diversity
than landscape predictors. The fact that the functional diversity of staphylinids is favoured by a high
vegetation layer may be linked to the increased diversity of vertical structures provided by grown
vegetation. In particular, complex vertical structures within the herbaceous vegetation may sustain
more diverse feeding groups within staphylinid communities than shortly-mown homogenous lawns.
For instance, a high vegetation may offer more opportunities for phytophagous species to feed
directly on plants. Indirectly, higher vegetation may maintain a high moisture amount at the ground

72
 Chapitre 2

level and favour the fungal development that is needed by fungivorous species to feed. Further, high
plants may host more preys (e.g., aphids), which might be beneficial for predatory staphylinids.
In carabids, the community composition analysis showed that assemblages are different from a site
cluster to another, whereby none of the tested environmental variables were significant predictors.
This result, combined with the absence of relationships between diversity indices and any of the
predictor data sets, suggest that carabid communities are shaped by the environment, but the
environmental variables that we investigate here are not the determinant ones. A previous study
conducted in the same study area by Croci et al. 2008 evidenced that carabid communities can be
primarily impacted by changes in the urban matrix, at the landscape scale (i.e., 600 m). Because of
their limited dispersal abilities, carabids might be strongly subject to large-scale urban filtering and
become globally less abundant and diverse in urban areas, but adapt to urban environmental
constraints (Croci et al. 2008). However, we should be cautious when comparing these results since
in previous studies led in Rennes (also in Varet et al. 2013), carabids were sampled in urban
woodlands, and patterns may diverge from grassland communities. In addition, it is important to
emphasize that the lack of responses observed in carabids in our study may also be influenced by
the relatively low number of individuals caught. Future works will be needed to fully understand our
non-significant results. For instance, other habitat characteristics (e.g., vegetation composition, soil
characteristics, biotic interactions) not included here might be determinant habitat predictors
(Harvey et al. 2008).

[Link] Absence of temperature effect

In the chapter 1, we have shown that spiders were primarily constrained by urban temperature
variation, with the local temperature constraining spiders’ activity density, whereas warming at the
landscape scale determined their taxonomic diversity. Like spiders, staphylinids and carabids are
ground-dwelling predators occurring in urban grasslands. However, unlike spiders, they did not
respond to temperature. Differences in behaviors, interactions, and life histories can modulate the
response of terrestrial ectotherms to temperature increase (Tylianakis et al. 2008, Diamond et al.
2011, Sunday et al. 2014). However, contrasting physiological heat tolerance among taxonomical
groups is expected to be the main factor explaining inconsistent responses of arthropods to
temperature increase (Youngsteadt et al. 2017). For example, warmer thermal conditions tend to
support the fitness of taxa tolerating a wide temperature range, whereas those with narrow thermal
tolerance breadth are more negatively affected (Piano et al. 2020a). A strong physiological variability
among taxa can therefore be expected to explain differences in terms of response strength. Rove
and ground beetle species from assemblages investigated here may therefore tolerate higher
temperature or have a wider thermal tolerance breadth than spiders.
In addition to the effect of physiological thermal tolerances, the responses to temperature increase
may be modulated by arthropods’ life-history traits. For example, dispersal capacity has been
demonstrated as a determinant factor affecting the direction of the relationship between body size
variation and the UHI intensity among arthropod groups (Merckx et al. 2018c). Recent results on
spiders captured in Rennes have shown that the strength of the relationship between body size and
the environmental temperature varied according to species’ capacity to disperse on long distance or
not (Cabon et al. 2024a) (chapter 4). More precisely, this latest study revealed that small species also
able to aerially disperse were less affected by warming than larger and less mobile wandering

73
PARTIE II

species. When brought in relation with the present results, these outcomes suggest that the pattern
variability observed among species belonging to a particular taxon may explain that community-wide
signals of response to warming can be blurred by interspecific variations, making effect-prediction
related to urban warming challenging (Piano et al. 2020a). In the light of our results, life-histories of
staphylinid and carabid species belonging to sampled communities may be diverse, which may
explain the lack of signal at the community scale. Therefore, gathering information relative to
dispersal (e.g., flying capacity, a poorly documented trait in staphylinids) may help to determine
whether patterns occur within high and low dispersers, independently.
In conclusion, our study brings new evidences that the responses of ectotherm communities in terms
of composition and diversity to the thermal environment are taxon-dependent. Therefore, general
patterns cannot be drawn regarding the effect of UHI or near-ground temperature on the arthropod
fauna. Our results suggest that reducing mowing frequency or increasing cutting height should
promote the functional diversity within staphylinid assemblages. Yet, the consideration of additional
life-history traits (e.g., dispersal capacities) as potential predictors of the community-wide thermal
response may enable to advance our mechanistic understanding. However, the difficulty of a multi-
taxonomic approach currently lies in the lack of information available on the functional traits of the
species making up arthropod communities. In particular, future research effort should be made to
better document life history traits of staphylinids, a reliable indicator of environmental changes
related to urbanization.

74
 Chapitre 2

2.2.5 Appendices

Appendix 2.6 - Mean scaled values and standard deviations (black segments) of environmental variables
after classification of sampling sites into three clusters by hierarchical clustering. Clustering was
performed based on the two and three variables measured at the landscape and local scale, respectively.

75
PARTIE II

Appendix 2.7 - List of staphylinid species and corresponding trait information relative to diet and body size (given
in millimetres). The number of caught individuals is indicated in the column ‘N’ (1/4).
Body length
Diet
Staphylinid species N (Lott 2009,
(Clough et al. 2007)
Lott and Anderson 2011)
Aleochara curtula 47 Parasitic 6
Aleochara gr. Bipustulata 19 Parasitic 4
Aleochara sp3 2 Parasitic 5
Alevonota gracilenta 2 Parasitic 2.6
Aloconota gregaria 3 Predatory 3.25
Amarochara forticornis 1 Predatory 3.75
Amischa gr. Decipiens 39 Predatory 2.1
Anotylus inustus 1 Detritivorous 3.5
Anotylus nitidulus 13 Detritivorous 2.5
Anotylus sculpturatus 19 Detritivorous 3.9
Anotylus tetracarinatus 47 Detritivorous 1.95
Astenus immaculatus 14 Unknown 3.75
Astenus lyonessius 14 Unknown 3.75
Astenus sp3 2 Unknown 3.75
Atheta gr. Aeneicollis 15 Fungivorous 3.32
Atheta gr. Crassicornis 18 Fungivorous 3.1
Atheta triangulum 1 Fungivorous 3.55
Athetini sp4 3 Fungivorous 3.32
Athetini sp5 1 Fungivorous 3.32
Athetini sp6 2 Fungivorous 3.32
Athetini sp7 1 Fungivorous 3.32
Athetini sp8 1 Fungivorous 3.32
Autalia gr. Longicornis 1 Predatory 2.9
Bisnius fimetarius 2 Predatory 6.25
Bisnius gr sordidus 4 Predatory 6
Bolitobius castaneus 24 Predatory 6.5
Brachida exigua 1 Fungivorous 2.25
Callicerus obscurus 10 Predatory 2.75
Callicerus rigidicornis 3 Predatory 4.25
Cordalia obscura 3 Predatory 2.4

76
 Chapitre 2

Appendix 2.7 - List of staphylinid species and corresponding trait information relative to diet and body size (given
in millimetres). The number of caught individuals is indicated in the column ‘N’ (2/4).
Body length
Diet
Staphylinid species N (Lott 2009,
(Clough et al. 2007)
Lott and Anderson 2011)
Cypha longicornis 15 Fungivorous 1.2
Dinaraea angustula 6 Predatory 3.35
Drusilla canaliculata 462 Myrmecophilous 4.9
Encephalus complicans 4 Unknown 2
Falagrioma thoracica 4 Predatory 2.7
Gabrius breviventer 4 Predatory 4.25
Gabrius nigritulus 7 Predatory 4.25
Gauropterus fulgidus 1 Predatory 10
Geostiba circellaris 1 Predatory 2.67
Hypomedon debilicornis 1 Detritivorous 2.5
Ilyobates propinquus 2 Predatory 4.5
Ischnosoma splendidum 51 Fungivorous 3.25
Leptacinus batychrus 2 Predatory 5
Liogluta longuscula 3 Predatory 3.7
Lomechusa paradoxa 1 Myrmecophilous 4
Mocyta gr. Fungi 285 Fungivorous 2.6
Mycetoporus angularis 1 Fungivorous 3.5
Mycetoporus longulus 6 Fungivorous 4.5
Mycetoporus sp3 1 Fungivorous 4
Nehemitropia lividipennis 2 Predatory 3.25
Neohilara subterranea 7 Fungivorous 2.65
Ocypus aeneocephalus 2 Predatory 13
Ocypus aethiops 27 Predatory 20.25
Ocypus olens 91 Predatory 27.5
Oligota pusillima 7 Predatory 1.15
Omalium italicum 1 Detritivorous 3
Othius laeviusculus 2 Predatory 5.5
Othius punctulatus 10 Predatory 11
Othius subuliformis 1 Predatory 5
Oxypoda brevicornis 8 Predatory 2.85

77
PARTIE II

Appendix 2.7 - List of staphylinid species and corresponding trait information relative to diet and body size (given
in millimetres). The number of caught individuals is indicated in the column ‘N’ (2/4).
Body length
Diet
Staphylinid species N (Lott 2009,
(Clough et al. 2007)
Lott and Anderson 2011)
Oxypoda induta 1 Predatory 3
Oxypoda lurida 1 Detritivorous 4.4
Oxypodini sp4 2 Predatory 2.76
Oxypodini sp5 1 Predatory 2.76
Oxytelus laqueatus 1 Predatory 2.76
Paedurus littoralis 67 Predatory 8.5
Pella limbata 65 Myrmecophilous 5
Philonthus carbonarius 112 Predatory 7.5
Philonthus cognatus 217 Predatory 9.5
Philonthus corruscus 5 Predatory 8
Philonthus gr. Varians 6 Predatory 7
Philonthus laminatus 2 Predatory 11
Philonthus sanguinolentus 1 Predatory 7
Philonthus sp10 1 Predatory 9.36
Philonthus splendens 1 Predatory 12.75
Philonthus succicola 4 Predatory 11.5
Philonthus tenuicornis 3 Predatory 10
Plataraea brunnea 43 Predatory 3.75
Platydracus chalcocephalus 86 Myrmecophilous 16
Platydracus stercorarius 7 Myrmecophilous 14.5
Platystethus cornutus 1 Herbivorous 3.25
Platystethus sp2 1 Herbivorous 3.25
Pycnota paradoxa 2 Predatory 2.25
Quedius fumatus 2 Predatory 8.5
Quedius gr. Curtipennis 49 Predatory 12.5
Quedius gr. Simplicifrons 1 Predatory 11.25
Quedius levicollis 52 Predatory 8.77
Quedius nigriceps 3 Predatory 8.5
Quedius picipes 1 Predatory 10
Quedius schatzmayri 29 Predatory 6

78
 Chapitre 2

Appendix 2.7 - List of staphylinid species and corresponding trait information relative to diet and body size (given
in millimetres). The number of caught individuals is indicated in the column ‘N’ (3/4).
Body length
Diet
Staphylinid species N (Lott 2009,
(Clough et al. 2007)
Lott and Anderson 2011)
Quedius semiaeneus 2 Predatory 7.5
Quedius semiobscurus 27 Predatory 8.5
Rugilus erichsonii 6 Predatory 4.25
Rugilus geniculatus 1 Predatory 5.5
Rugilus orbiculatus 14 Predatory 4.5
Rugilus rufipes 3 Predatory 5.75
Rugilus similis 18 Predatory 5.75
Rugilus subtilis 1 Predatory 5.5
Scopaeus gracilis 1 Predatory 3.25
Sepedophilus gr. Testaceus 2 Fungivorous 4
Sepedophilus nigripennis 40 Fungivorous 2.5
Stenus aceris 1 Predatory 4.5
Stenus clavicornis 28 Predatory 5.75
Stenus fulvicornis 2 Predatory 3.5
Stenus gr. Brunnipes 22 Predatory 3.5
Stenus ossium 25 Predatory 4.5
Stenus providus 1 Predatory 5.75
Stenus sp7 2 Predatory 4.89
Sunius propinquus 33 Predatory 3.75
Tachinus marginellus 43 Predatory 5.75
Tachinus rufipes 59 Predatory 3.75
Tachinus subterraneus 1 Predatory 3.75
Tachyporus dispar 171 Predatory 3.5
Tachyporus hypnorum 556 Predatory 2.5
Tachyporus nitidulus 68 Predatory 3.45
Tachyporus sp4 27 Predatory 5.75
Tasgius ater 3 Detritivorous 16.5
Tasgius morsitans 7 Detritivorous 14
Xantholinus elegans 9 Predatory 10.5
Xantholinus linearis 10 Predatory 7.5
Xantholinus longiventris 39 Predatory 7.75

79
PARTIE II

Appendix 2.8 - List of carabid species and corresponding trait information relative to diet and body size (given in
millimetres). The number of caught individuals is indicated in the column ‘N’ (1/2).
Body length
Diet
Carabid species N (Jeannel 1941,
(Larochelle 1990)
Schirmel et al. 2012)
Abax parallelepipedus 3 Omnivorous 18.59
Agonum afrum 3 Predatory 8.25
Amara (Zezea) fulvipes 1 Herbivorous 10.25
Amara (Zezea) strenua 18 Herbivorous 8.83
Amara aenea 32 Omnivorous 7.06
Amara communis 55 Omnivorous 6.29
Amara convexior 53 Omnivorous 7.6
Amara curta 1 Omnivorous 6.45
Amara eurynota 1 Omnivorous 10.12
Amara familiaris 3 Omnivorous 6.1
Amara lunicollis 15 Omnivorous 7.5
Amara ovata 2 Omnivorous 9
Amara proxima 1 Omnivorous 7.75
Anchomenus dorsalis 5 Omnivorous 7.1
Asaphidion curtum 1 Omnivorous 4.13
Asaphidion flavipes 2 Omnivorous 4.18
Asaphidion stierlini 3 Herbivorous 7.5
Badister bullatus 5 Predatory 5.38
Bembidion biguttatum 3 Herbivorous 3.75
Bembidion lampros 5 Omnivorous 3.2
Bembidion obtusum 9 Omnivorous 2.95
Bembidion properans 46 Predatory 2.95
Bembidion quadrimaculatum 17 Omnivorous 3.04
Brachinus sclopeta 1 Unknown 5.75
Bradycellus harpalinus 3 Omnivorous 3.89
Bradycellus verbasci 2 Omnivorous 2.5
Callistus lunatus 1 Unknown 6.53
Diachromus germanus 1 Herbivorous 8.96
Harpalus affinis 10 Omnivorous 10.56
Harpalus atratus 1 Herbivorous 11.25

80
 Chapitre 2

Appendix 2.8 - List of carabid species and corresponding trait information relative to diet and body size (given in
millimetres). The number of caught individuals is indicated in the column ‘N’ (2/2).
Body length
Diet
Carabid species N (Jeannel 1941,
(Larochelle 1990)
Schirmel et al. 2012)
Harpalus attenuatus 5 Herbivorous 8
Harpalus cupreus 1 Predatory 13
Harpalus distinguendus 1 Omnivorous 9.85
Harpalus litigiosus 1 Omnivorous 9.5
Harpalus melancholicus 1 Omnivorous 9
Harpalus rubripes 15 Omnivorous 9.5
Harpalus smaragdinus 1 Omnivorous 9.42
Leistus fulvibarbis 3 Unknown 7.38
Loricera pilicornis 6 Predatory 7.29
Microlestes maurus 0 Predatory 2.5
Microlestes minutulus 1 Predatory 2.8
Nebria brevicollis 16 Predatory 11.7
Notiophilus biguttatus 7 Predatory 5.13
Notiophilus quadripunctatus 4 Predatory 5.25
Notiophilus rufipes 1 Predatory 5.66
Ocys harpaloides 1 Unknown 5.25
Ophonus puncticeps 32 Herbivorous 7.5
Paratachys bistriatus 1 Unknown 2.18
Parophonus maculicornis 5 Unknown 6.5
Philochthus lunulatus 1 Predatory 3.5
Poecilus cupreus 32 Omnivorous 11.43
Poecilus versicolor 2 Omnivorous 10.24
Pseudoophonus rufipes 10 Omnivorous 13.71
Pterostichus madidus 12 Omnivorous 15.75
Pterostichus vernalis 2 Predatory 6.95
Semiophonus signaticornis 2 Omnivorous 6.5
Stenolophus teutonus 3 Herbivorous 5.75
Stomis pumicatus 2 Predatory 7.13
Syntomus obscuroguttatus 7 Predatory 3.25
Trechus quadristriatus 27 Predatory 3.69

81
 PARTIE II

Appendix 2.9 - Matrix of correlations among predictors. Variables describing the proportion of impervious surface are indicated by ‘imp_’, followed by the radius of the buffer
considered (in meters). Variables describing the UHI intensity are indicated by ‘UHI_’, followed by the radius of the buffer considered (in meters). The cover of local herbaceous
vegetation is indicated by ‘veg_cover’. The cover of local herbaceous vegetation is indicated by ‘veg_cover’. The height of local herbaceous vegetation is indicated by ‘veg_height’.
Near-groud-
imp_100 imp_200 imp_500 imp_1000 UHI_100 UHI_200 UHI_500 UHI_1000 veg_cover veg_height
temperature
imp_100 1.00 0.91 0.74 0.69 0.64 0.65 0.66 0.67 0.22 -0.33 0.40
imp _200 0.91 1.00 0.86 0.81 0.68 0.69 0.71 0.72 0.20 -0.28 0.34
imp _500 0.74 0.86 1.00 0.96 0.80 0.80 0.83 0.85 0.34 -0.19 0.36
imp _1000 0.69 0.81 0.96 1.00 0.84 0.84 0.88 0.92 0.29 -0.27 0.38
UHI_100 0.64 0.68 0.80 0.84 1.00 1.00 0.99 0.95 0.22 -0.32 0.46
UHI_200 0.65 0.69 0.80 0.84 1.00 1.00 0.99 0.96 0.22 -0.33 0.45
UHI_500 0.66 0.71 0.83 0.88 0.99 0.99 1.00 0.98 0.22 -0.33 0.45
UHI_1000 0.67 0.72 0.85 0.92 0.95 0.96 0.98 1.00 0.17 -0.37 0.46
veg_cover 0.22 0.20 0.34 0.29 0.22 0.22 0.22 0.17 1.00 0.30 -0.25
veg_height -0.33 -0.28 -0.19 -0.27 -0.32 -0.33 -0.33 -0.37 0.30 1.00 -0.54
Near-ground-
0.40 0.34 0.36 0.38 0.46 0.45 0.45 0.46 -0.25 -0.54 1.00
temperature

82
 Appendix 2.10 - Venn diagrams of (a) the activity density, (b) the taxonomic diversity, and (c) the functional diversity of staphylinids (left columns) and carabids (right
columns). The predictor datasets are represented by coloured ellipses where the landscape habitat is dark green (upper left), local habitat is light green (upper right), the
atmospheric UHI within a 100-m radius buffer is dark blue (lower left) and near-ground temperature is light blue (lower right). Results are given separately with landscape
habitat and UHI predictors calculated within 200-m, 500-m and 1000-m radius buffers. Variation proportion lower than 1% are not displayed. Significant results are
underlined.

83
Chapitre 2
PARTIE I

84
 Chapitre 3

3 Partie III : Mécanismes expliquant les liens

entre diversité des arthropodes et
température

▪ Chapitre 3 : Filtres environnementaux et composition fonctionnelle des

communautés

▪ Chapitre 4 : Les traits d’histoire de vie modulent la contrainte exercée par

le climat urbain sur la taille corporelle

En bref :
▪ Dans cette partie, nous montrons que la baisse de diversité taxonomique en
situation d’ICU atmosphérique intense est liée à un phénomène de filtrage
environnemental des espèces selon leurs caractéristiques écologiques. En
particulier, les espèces les plus thermosensibles et de taille relativement
importante sont exclues des zones dans lesquelles le réchauffement s’intensifie.

▪ Les espèces de Lycosidae semblent particulièrement négativement impactées

par le réchauffement, au regard de leurs activités-densités négativement reliées
à l’ICU atmosphérique.

▪ Certaines Lycosidae restent cependant présentes sur l’ensemble du gradient

d’ICU. En comparant les réponses intraspécifiques des tailles corporelles aux
variations de température, nous nous apercevons que les Lycosidae sont plus
contraintes par les réchauffements que les autres espèces testées.

▪ Cet effet accru du réchauffement sur ces espèces pourrait s’expliquer par leur
cycle de vie (univoltin) ou encore par leurs faibles capacités de dispersion.

85
PARTIE III

3.1 Chapitre 3 : Filtres environnementaux et composition

fonctionnelle des communautés

Résumé
Ce chapitre étudie les liens entre les compositions taxonomiques et fonctionnelles des
communautés d’arthropodes et des gradients d’habitat et de température. Pour cela, nous avons
comparé les effets de l’échelle du paysage (avec l’artificialisation des sols, l’îlot de chaleur urbain
(ICU)) et de l’échelle locale (avec la hauteur et la couverture de la végétation, la température près
du sol). Nous avons échantillonné 20 499 araignées correspondant à 137 espèces, sur 36 prairies et
pelouses à Rennes (nord-ouest de la France). Contrairement aux zones rurales, les résultats
montrent que les sites urbains (caractérisés par une végétation courte et un fort réchauffement de
l’air), abritent des communautés pauvres en espèces, composées d’espèces thermophiles de petite
taille. L’intensification de l’ICU et l’homogénéisation de la structure du micro-habitat (végétation
courte et dense) sont associées au déclin des espèces de grande taille, et sensibles à la chaleur. Ces
résultats mettent en évidence le rôle prépondérant du réchauffement urbain, plutôt que de la
modification de l’occupation des sols, en tant que filtre urbain. En outre, nous montrons que l’ICU à
l’échelle du paysage, et non la température locale, filtre les espèces en fonction de leurs
caractéristiques fonctionnelles. L’ICU atmosphérique peut donc être considéré comme une barrière
thermique, filtrant les espèces en fonction de leur capacité physiologique à se maintenir en
conditions thermiques urbaines. Enfin, pour contrebalancer l’homogénéisation biotique, nous
insistons sur l’importance de mettre en place des structures d’habitat complexes à l’échelle locale
au sein des infrastructures vertes urbaines.

Article publié chez Land :

Cabon, V., H. Quénol, V. Dubreuil, A. Ridel, and B. Bergerot. 2024. Urban heat island and reduced
habitat complexity explain spider community composition by excluding large and heat-sensitive
species. Land 13:83.

86
 Chapitre 3

Urban heat island and reduced habitat complexity explain spider community
composition by excluding large and heat-sensitive species

Valentin Cabon1,2, Hervé Quénol2,3, Vincent Dubreuil2,3, Aurélien Ridel1, Benjamin Bergerot1,2

1. University of Rennes, CNRS, ECOBIO – UMR 6553, Rennes, France

2. LTSER ZA Armorique, F-35000 Rennes, France
3. University of Rennes 2, CNRS, LETG – UMR 6554, Rennes, France

Abstract
Along with the worldwide urbanization, upheavals in habitat and temperature are major threats for
biodiversity. However, due to their interdependence, their relative role as drivers of animal
community composition remains entangled. Here, we investigated how taxonomic and functional
compositions of arthropod communities were related to uncorrelated habitat and temperature
gradients, and compared landscape (i.e., urbanization, Urban Heat Island (UHI)) to local variables
(i.e., vegetation height and cover, near-ground temperature). We sampled 20,499 spiders (137
species) on 36 study sites in Rennes (northwestern France). Unlike rural areas, urban sites were
characterized by short vegetation and intense UHI, hosted species-poor communities, and were
composed of small thermophilic species. UHI intensification and local habitat homogenization (short
and dense vegetation) were associated with declining large and heat-sensitive species. These results
highlight the prevalent role of urban warming, rather than land cover change, as an urban filter.
Further, we show that landscape-scale UHI, not local temperature, filters species according to their
functional attributes. UHI can therefore be considered as a thermal barrier, filtering species
according to their physiological capacity to cope with urban thermal conditions. Finally, to
counterbalance biotic homogenization, we argue for the importance to implement complex habitat
structures at the local scale within urban green infrastructure.

Keywords
Araneae; arthropod; climate warming, community composition; environmental filter; functional
composition; functional traits; urbanization

87
PARTIE III

3.1.1 Introduction
The rapid expansion of cities is widely considered to be a major contributor to global biodiversity loss
(Elmqvist et al. 2013). However, despite an acknowledged global negative impact, the potential of
urban areas to conserve biodiversity is often highlighted (Shwartz et al. 2014, Beninde et al. 2015).
For instance; urban ecosystems have been shown to support different animal taxa (MacGregor-Fors
et al. 2016), including ground-dwelling arthropods such as carabid beetles and spiders (Buchholz et
al. 2018). Whether urbanization supports or disadvantages communities depends on intraspecific
and interspecific responses to the new conditions induced by the urban environment (McIntyre
2014). Indeed, urban environments can affect functional trait frequencies and composition, and
therefore act as an environmental filter shaping urban community composition according to the
ability of species to overcome urban conditions and establish themselves (Piano et al. 2017, Buchholz
et al. 2020). Distinct variables of the urban environment filter have been identified as important
drivers of arthropod community composition at local and landscape scales (Sattler et al. 2010,
Philpott et al. 2014, Beninde et al. 2015). These local and landscape variables described in the
literature can be grouped in two main categories, habitat and temperature related variables (Piano
et al. 2020a).
About habitat, changes in land cover at the landscape scale were repeatedly identified as important
drivers of the loss of arthropod diversity along urbanization gradients (Fenoglio et al. 2020, Maher
et al. 2022). For example, parasitoid diversity was significantly negatively affected by the
replacement of vegetated areas with impervious surface (Bennett and Gratton 2012), ground-
dwelling beetles with building coverage (Delgado de la Flor et al. 2017), while ant, beetles and spiders
were related to landscape connectivity variables (Peng et al. 2020). In addition to landscape specific
effects, studies also provide evidences of the common role of habitat variables measured at
landscape and local scales in explaining arthropod communities structure (Philpott et al. 2014,
McCary et al. 2018). Variables such as the mowing regime or the habitat structural complexity were
found to influence arthropod species composition at local scale (Sattler et al. 2010, Buchholz et al.
2018).
In addition to habitat conditions, temperature is another widely studied variable. Urban warming has
been identified as an important driver of arthropod community composition (Meineke et al. 2017a)
because arthropods are ectotherms and their metabolism is directly dependent on external
temperature (Angilletta 2009, Huey et al. 2012). At landscape scale, temperatures are strongly
influenced by the urban heat island (UHI) phenomenon. The term UHI refers to an increase in the
average atmospheric temperature within a city, compared to adjacent rural areas (Oke et al. 2017).
This phenomenon mainly leads to an increase in night temperatures within urbanized areas (Oke et
al. 2017, Foissard et al. 2019a, Dubreuil et al. 2020). At local scale, near-ground temperatures also
influence arthropod communities (McGlynn et al. 2019, Piano et al. 2020a). However, near-ground
temperatures, as experienced by arthropods, can be very different from atmospheric temperatures
(Daly et al. 2010, Pincebourde and Salle 2020, Lembrechts et al. 2020). In fact, near-ground
temperatures show greater spatial variation than atmospheric UHI and are therefore often referred
to as thermal mosaics (Ziter et al. 2019, McGlynn et al. 2019). In urban areas, this is due to the
complex interweaving of vegetated and built surfaces (Buyantuyev and Wu 2010, Li et al. 2011,
2013a). Considering the temperature recorded at near-ground scale in addition to the UHI in studies
of arthropod communities is necessary, as this corresponds to the scale at which arthropods perceive
their environment (Pincebourde and Woods 2020). Consequently, the presence of near-ground

88
 Chapitre 3

thermal variations provides opportunity to mitigate unfavorable thermal conditions for ground-
dwelling arthropods (i.e., behavioral thermoregulation) (Woods et al. 2015).
Although a few studies have shown direct relationships between atmospheric UHI and arthropod
community composition (Meineke et al. 2017a, Merckx and Van Dyck 2019, McGlynn et al. 2019), to
our knowledge, no study has aimed to link functional trait composition (i.e., the distribution of
individual functional traits within a community (Díaz et al. 2007)) to thermal metrics (at both
landscape and local scales). Characterizing the links between species’ functional traits and
environmental variables would enable us to better understand how species maintain themselves in
cities (Buchholz et al. 2020), identify species from regional pools that are able to cope with urban
environmental conditions, and the relevant environmental variables that shape species composition
(Moretti et al. 2017). For example, a landscape habitat variable (i.e., urbanization) was found to shift
the functional composition of communities towards more dispersive and thermophilic carabid
species (Piano et al. 2017). If this study provides fundamental insights into the temperature induced
functional response of arthropods to urban conditions, the actual temperature measurements were
not related to functional shifts. Ideally, both near-ground (local) and atmospheric UHI (landscape)
temperatures should be recorded. However, this may rarely be possible due to the technical
difficulties of implementing comprehensive sensor networks capable of monitoring temperature at
different spatial scales.
In ground-dwelling arthropods, several functional traits have been associated with local and
landscape environmental conditions. Among them, body size is a key functional trait related to many
physiological traits (Kingsolver and Huey 2008) and constrained by environmental conditions. For
example, leaf litter depth is positively associated with carabid body size (Koivula et al. 1999, Small et
al. 2002, Tyler 2008, Philpott et al. 2019). Reduced grassland management favors both large-bodied
carabids and spiders occurrence (Buchholz et al. 2018, Delgado de la flor et al. 2020). At the
landscape scale, urbanization increases the dispersal capacity of spiders (Piano et al. 2020b), while
patch isolation favors small, highly mobile carabids (Buchholz et al. 2018) and spiders (Piano et al.
2020b). Finally, body size changes in several arthropods are related to urbanization rate and are
thought to be due to temperature increases caused by UHI (Merckx et al. 2018a, 2018c). Hence, UHI-
induced warming is generally predicted to drive shifts in ectotherm communities toward smaller
species, according to the Atkinson’ temperature-size rule (Atkinson 1994). The thermal affinity of
species is also expected to be an important functional trait driving species community composition
at local and landscape scales (Piano et al. 2017). If many studies have examined intraspecific changes
in critical thermal limits (Angilletta et al. 2007, Brans et al. 2017, Diamond et al. 2018c, 2018a, Sato
and Takahashi 2022) and thermal preferences (Cabon et al. 2023) along UHI gradients, almost no
studies have considered interspecific thermal traits to test whether changes also occur at the
community level. Although a few studies have reported changes in thermal affinities of urban versus
rural communities in ants and carabids (Menke et al. 2011, Piano et al. 2017), the identification of
explanatory variables remains to be investigated. Indeed, our knowledge of functional trait-
environment relationships remains scarce with respect to the respective roles of habitat and
temperature in explaining arthropod communities. To date, no studies have gone beyond a rural-
urban comparison with the aim of disentangling the role of different urbanization-related variables
and testing temperature metrics as predictors.
In this study, we used ground-dwelling spiders as a model to assess functional trait-environment
relationships along an urbanization gradient. Spiders are ideal biological models because they are
abundant, highly diverse and respond to urban environmental stressors, including landscape and
89
PARTIE III

local predictors (Varet et al. 2011a, Vergnes et al. 2014, Buchholz et al. 2018, Piano et al. 2020b).
Moreover, spiders effectively maintain many ecosystem services by playing important regulatory
roles as abundant prey and predators in the food chain (Michalko et al. 2019), notably because of
their diversity in terms of functional characteristics, their behavioral or physiological adaptations but
also their evolutionary strategies (Foelix 2011). Finally, spiders are ectotherms, which makes them
particularly sensitive to temperature changes (Meineke et al. 2017a). For example, individuals from
warmer locations exhibit higher thermal tolerance (Malmos et al. 2021a). Firstly, we analyzed spider
community composition along an urbanization gradient, identified indicator species, and determined
which predictor variables drive the abundance of the mainly represented species. Secondly, we
characterized the functional traits of spider communities and their relationships with the
environment. More precisely, we quantified changes in community-averaged body size, community
dispersal capacity and community thermal affinity in response to urbanization gradient and thermal
conditions. Thirdly, we tested whether the observed changes in community functional composition
were the result of a functional replacement (i.e., species with particular functional trait scores are
discarded while others are favored) or a filtering process (i.e., species with particular trait scores are
discarded with little or no replacement).
In order to disentangle the relative effects of habitat and temperature at landscape scale on the
functional composition of communities, we selected study sites using a specific procedure that
minimizes the covariation between both habitat and temperature variables (for details see Appendix
2.1, Chapter 1). To our knowledge, our study is the first to analyze functional trait-environment
relationships by simultaneously comparing (1) uncorrelated habitat and temperature variables at
landscape scale, (2) local habitat and temperature variables, and (3) temperature variables at
landscape and local scales. In line with previous findings on ground-dwelling spiders (Merckx et al.
2018c), we expect Atkinson’s rule to prevail (Atkinson 1994). As a result, the size of spiders should
decrease in urbanized areas, while dispersal capacity should increase in relation to the rate of
urbanization, as densely urbanized landscapes are also more fragmented (Concepción et al. 2015,
Cheptou et al. 2017). We can assume that urban communities will be composed of more
thermophilic species than rural ones in response to the warmer conditions generated by increased
UHI intensity (Menke et al. 2011, Piano et al. 2017).

3.1.2 Materials and Methods

[Link] Study area and sampling sites
This study was conducted in and around Rennes city in the north-west of France (48°06′N–1°40′W).
Rennes is a city of 227,000 inhabitants. The proximity of the sea (70 km) maintains a temperate
oceanic climate. Despite its mild climate, Rennes city faces regular and strong UHI events (Dubreuil
et al. 2020). We chose grasslands as a model ecosystem because they represent 445 ha (i.e., 56%)
of urban green spaces in Rennes (Chollet et al. 2018). They provide good experimental conditions,
as they can support a wide range of vegetation structures, from short and densely seeded lawns in
highly managed parks or courtyards, to higher and heterogenous vegetation where management is
less frequent. Finally, urban grasslands support abundant ground-dwelling arthropod communities
(Proske et al. 2022), ensuring suitable sampling size.

90
 Chapitre 3

To conduct our experiment, 36 sampling sites were selected on grasslands that had been previously
identified as suitable (Figure 3.1). To select suitable grasslands, we determined areas where
urbanization rate and UHI intensity was not highly correlated by performing a spatial correlation
analysis based on land cover and UHI maps (for details see Appendix 2.1, Chapter 1). The output map
allowed us to identify the main decorrelated areas and then select the 36 study sites. The sites were
located in public green spaces (N = 10), roadside green spaces (N = 6), communal gardens (N = 12),
educational institutions (N = 5) and private gardens (N = 3). Grassland structures ranged from poorly
managed meadows to manicured lawns.

Figure 3.1 - Location of the 36 sampling sites (black crosses) within and around Rennes (black line). Built surfaces
are in grey, grasslands in green and waterbodies in blue.

[Link] Spider sampling

Spiders were sampled each month from March to September 2022 using three pitfall traps per
sampling site. The three pitfall traps were placed 5 m apart, forming a triangle to ensure capture
efficiency (Ward et al. 2001). The traps were oriented north, southeast and southwest. The traps
were made of plastic cups (85 mm in diameter and 115 mm in height). We filled them with 15 cl of
saline solution at a concentration of 100 g.L-1 to improve spider preservation and a drop of neutral
soap to prevent floating. The traps were opened for two weeks each month. After each sampling
session, spider individuals were sorted and stored in separate vials containing 70% ethanol. All
mature spiders were identified to species level using identification keys (Roberts 1985, 1987).
Nomenclature followed the World Spider Catalogue version 22 ([Link] The dataset
used for analysis was obtained by pooling all three traps per sampling site and all seven sampling
sessions together.

91
PARTIE III

[Link] Functional trait selection

To investigate the functional response of spider communities to their environment, we retrieved
information related to body size (Roberts 1985, 1987, Macías-Hernández et al. 2020, Nentwig et al.
2023) and dispersal capacity (i.e., ballooning (Blandenier and Fürst 1998, Bell et al. 2005)) for each
species from literature data (Appendix 3.1). These traits are known to respond to environmental
change along urbanization gradients (Buchholz et al. 2018). Body size influences species physiology,
ecological niches and species spatial distribution (Merckx et al. 2018c, Magura and Lövei 2021).
While sexual dimorphism is common in spiders, we retrieved only the body size of females (Merckx
et al. 2018c). Dispersal capacity informs on species ability to escape their home range to avoid
competition or adverse environmental conditions (Bonte and Dahirel 2017).
We also estimated the thermal niche of species to test whether urban environmental conditions
filter species according to their thermal affinities. We followed a method described in the literature
(Bowler et al. 2015, Piano et al. 2017) to assign temperature attributes to species according to their
spatial distribution in Europe and North Africa. Accordingly, we extracted species distribution
information for all recorded species from the online database Spiders of Europe (Nentwig et al. 2023)
and we retrieved gridded European mean, maximum and minimum daily temperatures between
2011 and 2022 (Cornes et al. 2018). From temperature data, we extracted the averaged mean,
maximum and minimum temperature per country. The average temperatures of countries where a
species is present were then used to calculate the thermal niches of the species, resulting in three
thermal affinity values per species: mean, maximum and minimum thermal affinity.

[Link] Environmental variables

To characterize habitat and temperature variables, we collected two and three environmental
variables at landscape and local scales respectively (see
Table 3.1 for details). Variables measured at landscape scale were the proportion of impervious
surface (a commonly used proxy for urbanization, (Moll et al. 2019) and the mean UHI intensity
calculated over the entire study period as a proxy for urban atmospheric warming (Dubreuil et al.
2020, Cabon et al. 2023). To calculate these two landscape variables, we chose 100 m as the radius
size of the circular buffers around the sampling sites. This buffer size has been found to cover an
adequate area to provide landscape scale information related to urban matrix that drive arthropod
communities (McCary et al. 2018, Chatelain et al. 2023). Furthermore, it corresponds to the finest
spatial resolution of atmospheric UHI data available in our study area and it matches the scale at
which the correlation between urban land cover (i.e., built-up area) and atmospheric UHI remains
lowest in Rennes (Foissard et al. 2019a). We did not include any additional landscape variables
related to vegetation cover, since urbanization and vegetation (including wooded areas) were
strongly negatively correlated (r = -0.8).
At the local (i.e., site) scale, we characterized vegetation structure by measuring the percentage of
vegetation cover and the height of the vegetation (
Table 3.1). These predictors are regularly used to characterize the habitat of arthropod species
(Delgado de la Flor et al. 2017). In addition, we monitored the air temperature just above the surface
(i.e., 5 cm) during the study period to obtain a record of the near-ground temperature experienced
by the spiders.

92
 Chapitre 3

Table 3.1 - Environmental variables used to characterize sampling sites.

Variable Mean + SD Unit Details

Landscape scale

Percentage of impervious surface within a 100 m radius buffer

36.80 ± around sampling sites. Land cover data were obtained at a
Urbanization %
21.94 resolution of 5 m from the French portal for geographical data
(dataset OCSGE 2017; [Link]).

Average difference between the minimum daily temperature at a

reference cold station located outside the urban area and the
minimum daily temperature at each other sensor, over the sampling
period. Atmospheric temperature data used to calculate the UHI
were retrieved from the long-term Rennes Urban Network
Urban heat ([Link] Based on 30 DAVIS Vantage-Pro-2
1.71 ± 0.56 °C
island automated weather stations and 93 connected temperature
sensors (Rising HF RHF1SOO1), UHI was interpolated by inverse
distance weighting to spatialize the estimated UHI values over the
city extent at a resolution of 100 m (Foissard et al. 2019a). From this
map, we finally extracted the mean pixel values within circular
buffers of 100m around the sampling sites.
Local scale
Mean vegetation height measured within the triangle materialized
Vegetation 19.73 ± by the three pitfall traps during each of the seven sampling sessions.
cm
height 11.54 All seven measurements were averaged to obtain one final value per
site.

Percentage of herbaceous vegetation cover estimated in 10%

increments within the triangle materialized by the three pitfall traps
Vegetation 90.50 ±
% during each of the seven sampling sessions. All seven measurements
cover 9.38
were averaged to obtain one final value per site. A low vegetation
cover typically indicates the presence of litter or bare soil.

Average difference between the minimum daily temperature at a

reference cold station and the minimum daily temperature at each
other sensor, over the sampling period. Temperatures were
Near-ground recorded every 15 minutes during each trapping session using a
2.58 ± 1.39 °C
temperature dedicated network of temperature loggers (Lascar EL-USB-2+;
Tinytag Talk 2 TK-4023). The loggers were placed 5 cm above the
ground surface at each sampling site, in close proximity to the north
trap.

93
PARTIE III

[Link] Data analyses

[Link].1 Characterization of the urbanization gradient
To establish a typology of sites according to their contrasts in environmental characteristics, we
performed a hierarchical clustering analysis (HCA) based on all scaled environmental variables. We
used the ‘HCPC’ function and the ‘Ward’ method to construct the tree in the ‘FactoMineR’ package
(Lê et al. 2008). The optimal number of clusters was determined according to the partition with the
higher relative loss of inertia (Lê et al. 2008). Once the sites were grouped into clusters, we
performed a silhouette analysis to check the agreement of individual sites with their own cluster,
using the ‘silhouette’ function from the ‘cluster’ package (Maechler et al. 2013).

[Link].2 Spider community composition along an urbanization gradient

The composition of spider communities across environmental clusters obtained by HCA was analyzed
using non-metric multidimensional scaling (NMDS, Bray-Curtis dissimilarity) with the ‘vegan’ package
(Oksanen et al. 2019). For ordination, species abundances were square root transformed. We then
tested whether communities differed significantly between clusters by performing a pairwise
permutational multivariate analyze of variance (9999 permutations) using the
‘[Link]’ function from the ‘RVAideMemoire’ package (Herve 2023). Finally, all
environmental variables were fitted in ordination space and their significance was tested using a
Monte Carlo randomization procedure (9999 permutations).
Indicator species were assessed using the fidelity and exclusivity of species to the resulting clusters
of sites using the IndVal (indicator value) procedure in the ‘labdsv’ package (Roberts 2016). Indicator
values are calculated from the species abundance matrix, taking into account both relative
abundances and frequencies of occurrence within each individual cluster. Significance of indicator
values was tested at a P-value < 0.05 using a Monte Carlo randomization procedure (9999
permutations). Species with both an indicator value > 0.5 and a significant P-value were considered
to be accurate indicators.
To test the response of the 30 most represented species (including indicator species) to
environmental variables, we fitted Poisson generalized linear models (GLMs). Since overdispersion
was detected, we corrected standard errors by using quasi-Poisson GLM (Zuur et al. 2009). We
assessed the goodness of fit by calculating the adjusted R², with the ‘rsq’ function from the ‘rsq’
package (Zhang 2017). To reduce statistical noise, all observations comprising fewer than three
individuals per site and per sampling period were removed from the dataset used to perform
statistical analysis on community composition and indicator species.

[Link].3 Relationships between functional traits and environmental conditions

For each trait (i.e., body size, ballooning and thermal affinity), we obtained a trait community index
by calculating community-averaged trait scores. As a binary trait, ballooning was considered as
numerical and a community-averaged value between 0 and 1 was calculated. In order to account for
the functional attributes of rare species in the calculation of community-averaged indexes, we
considered all species and did not weight the community-averaged traits by species abundance. To
test whether significant differences in trait community indices existed between clusters, we
performed a one-way ANOVA using the ‘aov’ function, followed by a post-hoc Tuckey test using the

94
 Chapitre 3

‘Tuckey-HSD’ function. To identify the direction of variations of each trait, we built a GLM with each
community-averaged trait score as the response variable and all five environmental variables as
predictors. For all models, we assumed a normal distribution and assessed goodness of fit by the
adjusted R² calculated using the ‘rsq’ package (Zhang 2017).

[Link].4 Characterization of changes in functional trait community indices

Finally, we analyzed whether changes in functional trait community indices were the result of
replacement by species with particular trait scores within communities, or whether they were due
to a decline in certain species (i.e., species filtering). To achieve this, we categorized species
according to their continuous functional trait scores (Piano et al. 2017) (i.e., body size and thermal
affinity). To define species categories, cutoff functional trait scores were chosen to yield three groups
with balanced numbers. Species are thus divided into small, medium and large species with respect
to body size, or into low, medium or high according to their affinity to temperatures with respect to
thermal niche. We used one-way ANOVA followed by a post-hoc Tuckey test to determine whether
species with particular functional trait categories were over- or under-represented in the cluster
communities in terms of absolute and relative species richness. We also tested whether specific
environmental variables were associated with increases or decreases in the richness of species with
specific functional trait using the same GLM as described above, but with absolute species richness
as the response variable.
All statistical analyses were performed in R version 4.2.2.

3.1.3 Results
[Link] Sampling results and urbanization gradient
A total of 20,499 adult spiders were captured, belonging to 21 families and 137 species (Appendix
3.1). The most represented families were Linyphiidae (N = 8,807; 43%), Lycosidae (N = 4,951; 24.2%)
and Tetragnathidae (N = 5,023; 24.5%). The most abundant species were Pachygnatha degeeri (N =
4,913; 24%) and Pardosa cf. tenuipes (N = 2,490; 12%).
The hierarchical classification of sampling sites led to the identification of three distinct clusters
characterized by contrasted environmental conditions (Figure 3.2). The first cluster (‘high vegetated
rural’) includes sites with both low UHI intensity and urbanization at the landscape scale, while at
the local scale the herbaceous vegetation was high, dense with low near-ground temperatures. The
second cluster (‘short vegetated rural’) includes sites with both low UHI intensity and urbanization,
while at the local scale, the vegetation was short, sparse with intermediate near-ground
temperatures. The third cluster (‘short vegetated urban’) includes sites with both high UHI intensity
and urbanization, while at the local scale, the vegetation was short, sparse with high near-ground
temperatures. The spatial distribution of clusters is illustrated in Figure 3.7.

95
PARTIE III

Figure 3.2 - Mean scaled values and standard deviations (black segments) of environmental variables after
classification of sampling sites into three clusters by hierarchical clustering. Clustering was performed based on
the two and three variables measured at the landscape and local scale, respectively.

[Link] Spider community composition and indicator species along an urbanization gradient
Excluding species with fewer than three occurrences per site and per sampling session resulted in
20,049 adult spiders (i.e., 98% from total abundance) belonging to 85 species (62% from total species
richness) subjected to NMDS ordination (Figure 3.3). The final stress value of 0.17 indicates a good
representation of the community distribution by the scaling. Spider communities from both rural
clusters (i.e., ‘high vegetated rural’ and ‘short vegetated rural’) did not significantly differ among
each other (P = 0.81), but were significantly different from communities belonging to the ‘short
vegetated urban’ cluster (P = 0.002, respectively). Examination of individual environmental variables
revealed that spider communities were significantly structured by UHI intensity (P < 0.001),
urbanization (P = 0.004), near-ground temperature (P < 0.001) and vegetation height (P = 0.002).

96
 Chapitre 3

Figure 3.3 - Results of NMDS ordination based on 85 species (Bray-Curtis distance) and the four significant
environmental variables.

Although the composition of communities differed significantly between clusters, several dominant
species were shared by the three clusters. For example, Pachygnatha degeeri, Pardosa cf. tenuipes
or Tenuiphantes tenuis were abundant in the three clusters (Table 3.2). We identified six indicator
species in total, three for ‘high vegetated rural’ sites and three for ‘short vegetated urban’ sites.

97
 PARTIE III

98
Table 3.2 - List of the 19 most abundant species occurring in each cluster. Indicator species are in bold. For each cluster, the number of individuals per species are indicated in the column
‘N’ and the related percentage in the column ‘%’.

High vegetated rural N % Short vegetated rural N % Short vegetated urban N %

Total 7535 100 Total 6529 100 Total 5983 100
Pachygnatha degeeri 2159 28.6 Pachygnatha degeeri 1650 25.3 Pachygnatha degeeri 1128 18.9
Pardosa cf. tenuipes 691 9.2 Pardosa cf. tenuipes 1464 22.4 Erigone dentipalpis 1080 18.1
Oedothorax fuscus 657 8.7 Tenuiphantes tenuis 319 4.9 Tenuiphantes tenuis 522 8.7
Oedothorax retusus 551 7.3 Alopecosa pulverulenta 227 3.5 Pardosa cf. tenuipes 442 7.4
Tenuiphantes tenuis 368 4.9 Oedothorax fuscus 195 3.0 Erigone atra 393 6.6
Pardosa pullata 357 4.7 Pardosa pullata 195 3.0 Bathyphantes gracilis 327 5.5
Agyneta affinis 240 3.2 Diplostyla concolor 189 2.9 Oedothorax fuscus 222 3.7
Bathyphantes gracilis 214 2.8 Hahnia nava 188 2.9 Hahnia nava 213 3.6
Aulonia albimana 207 2.7 Aulonia albimana 182 2.8 Oedothorax retusus 169 2.8
Diplostyla concolor 194 2.6 Bathyphantes gracilis 169 2.6 Tiso vagans 161 2.7
Pardosa prativaga 177 2.3 Oedothorax retusus 153 2.3 Diplostyla concolor 115 1.9
Hahnia nava 174 2.3 Tiso vagans 121 1.9 Ozyptila simplex 115 1.9
Alopecosa pulverulenta 151 2.0 Agyneta affinis 115 1.8 Agyneta mollis 104 1.7
Agyneta mollis 144 1.9 Erigone dentipalpis 107 1.6 Agyneta affinis 92 1.5
Panamomops sulcifrons 138 1.8 Erigone atra 104 1.6 Pardosa pullata 88 1.5
Erigone dentipalpis 112 1.5 Pardosa prativaga 103 1.6 Aulonia albimana 82 1.4
Trochosa terricola 91 1.2 Trochosa terricola 101 1.5 Cryptachaea blattea 82 1.4
Tiso vagans 86 1.1 Phrurolithus festivus 78 1.2 Panamomops sulcifrons 80 1.3
Erigone atra 61 0.8 Agyneta mollis 76 1.2 Phrurolithus festivus 72 1.2
 Chapitre 3

The GLM analysis of the abundances of the 29 most abundant species revealed that UHI and near-
ground temperature were each associated with 7 species (Table 3.3); representing 24.21% and
41.09% of the individual count; respectively. Near-ground temperature was negatively related to
abundances; whereas UHI was associated to contrasted responses. For instance; Erigone dentipalpis
indicated ‘short vegetated urban’ sites and was positively related to UHI. In contrast; Pardosa pullata
indicated ‘high vegetated rural’ sites (Table 3.2) and was less abundant when UHI and near-ground
temperature increased (Table 3.3). Similarly; indicators of ‘short vegetated urban’ sites became rarer
with increasing vegetation height (Table 3.3); whereas P. pullata was supported

Table 3.3 - Responses of the 29 most abundant species to environmental variables, tested by GLM. Positive and negative
relationships are indicated by ↑ and ↓ respectively. P-values are indicated between brackets. Indicator species are in bold. ‘N’
is the number of individuals per species and ‘%’ is the percentage of individual count per species related to the total number of
individuals. ‘Urb’ displays urbanization, ‘Near-ground temp’ the near-ground temperature, ‘Veg height’ the vegetation height and
‘Veg cover’ the vegetation cover. Adjusted R² is given in the ‘adj-R²’ column.

Environmental variables

Near-
adj-R²
Species N % Urb UHI ground Veg height Veg cover
temp

Pachygnatha degeeri 4937 24.62 ↓ (0.020) 0.27

Pardosa cf. tenuipes 2597 12.95 0.00
Erigone dentipalpis 1299 6.48 ↑ (0.002) ↓ (0.023) ↓ (0.033) 0.51
Tenuiphantes tenuis 1209 6.03 0.02
Oedothorax fuscus 1074 5.36 ↑ (< 0.001) ↓ (0.002) 0.47
Oedothorax retusus 873 4.35 ↑ (0.003) ↓ (0.020) 0.33
Bathyphantes gracilis 710 3.54 0.02
Pardosa pullata 640 3.19 ↑ (0.010) ↓ (0.020) ↓ (< 0.001) ↑ (0.010) 0.55
Hahnia nava 575 2.87 0.06
Erigone atra 558 2.78 ↓ (0.030) ↑ (0.010) 0.34
Diplostyla concolor 498 2.48 0.06
Aulonia albimana 471 2.35 ↓ (0.040) ↑ (0.030) 0.11
Agyneta affinis 447 2.23 0.06
Alopecosa pulverulenta 403 2.01 ↓ (0.005) ↓ (0.030) 0.14
Tiso vagans 368 1.84 0.00
Agyneta mollis 324 1.62 ↑ (0.010) 0.16
Pardosa prativaga 304 1.52 0.10

99
PARTIE III

Table 3.3 – (cont.)

Panamomops sulcifrons 288 1.44 0.13

Trochosa terricola 213 1.06 ↓ (0.020) 0.26
Phrurolithus festivus 207 1.03 0.00
Ozyptila simplex 203 1.01 0.01
Mermessus trilobatus 142 0.71 0.00
Pachygnatha clercki 100 0.50 ↓ (0.050) 0.06
Pelecopsis parallela 99 0.49 0.00
Palliduphantes pallidus 94 0.47 ↑ (0.050) 0.15
Dicymbium nigrum 84 0.42 0.07
Monocephalus fuscipes 84 0.42 0.00
Cryptachaea blattea 82 0.41 ↑ (0.001) ↓ (0.002) 0.85
Pardosa palustris 76 0.38 0.58

[Link] Relationships between functional traits and environmental conditions

Mean community body size varied significantly between clusters ( Figure 3.4, P = 0.04). The mean
community size of spiders was significantly lower (4.00 mm) in high vegetated urban sites, compared
to short and high vegetated rural sites (4.34 mm and 4.58 mm, respectively). GLM showed that body
size was positively related to vegetation height (coeff = 0.28, P = 0.02, adjusted R² = 0.21).
Community ballooning capacity did not differ significantly between clusters (P = 0.50). GLM showed
a significant negative relationship between community ballooning capacity and vegetation height
(coeff = - 0.03, P = 0.02, adjusted R² = 0.16).

Figure 3.4 - Community-averaged body size in sites belonging to three categories identified by hierarchical
classification. Differences between clusters are givent by non-matching lowercase letters (Tuckey tests; P ≤
0.05).

100
 Chapitre 3

We found significant differences in community thermal affinities between ‘high vegetated rural’ and
urban sites (Figure 3.5). Urban sites systematically had higher community thermal affinities with mean
affinities of 10.30 °C (mean thermal affinity), 15.10 °C (maximum thermal affinity) and 5.80 °C
(minimum thermal affinity), whereas high vegetated rural sites had lower values with 10.10 °C (mean
thermal affinity), 14.90 °C (maximum thermal affinity) and 5.60 °C (minimum thermal affinity). No
relationship with environmental variables were indentified by GLM analysis regarding the thermal
affinity. As the changes in the community-averaged thermal affinity were similar whether the index
was calculated from mean, maximum or minimum temperatures, only the mean community thermal
affinity is considered in the following results and discussion. The spatial distribution of mean
community body size and thermal affinity values are illustrated in Figure 3.7.

Figure 3.5 - Community-averaged thermal affinity based on mean (A), minimum (B) and maximum (C)
temperature data, in sites belonging to three clusters identified by the hierarchical classification. Differences
between clusters are givent by non-matching lowercase letters (Tuckey tests; P ≤ 0.05).

[Link] Characterization of changes in functional trait community indices

The total number of species was significantly lower in the ‘short vegetated urban’ cluster, compared
to ‘high vegetated rural’ and ‘short vegetated rural’ (Figure 3.6A). Among species classes sorted by
body size, the number of small and large species was significantly lower in the urban cluster
compared to both rural ones, whereas the number of medium species differed only between the
urban cluster and the ‘high vegetated rural’ (Figure 3.6B). Proportionally, small species increased
whereas large species decreased in the urban cluster, compared to both rural ones ( Figure 3.6D). We
observed a decrease in the absolute number of species associated to low and intermediate
temperatures in the urban cluster, compared to both rural ones (Figure 3.6C). In the case of species
related to high temperatures, the number of species did not differ between clusters. The decreasing
absolute species richness in species associated to low and intermediate resulted in an increased
proportion of species associated to high temperatures in the urban cluster, compared to the rural
ones (Figure 3.6E).

101
PARTIE III

Figure 3.6 - Absolute and relative species richness. Upper charts display the total number of species (A), the
number of species belonging to the three body size classes (B) and the number of species belonging to three
classes of thermal affinity (C). Below charts display the relative species richness belonging to the three body
classes (D) and relative species richness belonging to three classes of thermal affinity (E). Differences between
clusters are given by non-matching lowercase letters (Tuckey tests, P ≤ 0.05) within each single body size or
thermal affinity class.

Analysis of the effect of environmental variables on variation in absolute species richness showed
that the variables studied had a strong influence on large species (Table 3.4). In particular, the UHI
intensity is negatively related to the number of large species, whereas the vegetation height is
positively related to the number of large species. In addition, the number of species related to low
and intermediate temperatures were negatively related to UHI intensity, whereas the number of
species related to intermediate temperatures was also explained by vegetation attributes ( Table 3.4).

102
 Chapitre 3

Table 3.4 - Significant responses of absolute species richness to environmental variables within classes of species
sorted by body size and thermal affinity. Significant positive and negative relationships are indicated by ↑ and
↓ respectively. P-values are indicated between brackets. Near-ground temperature is not shown as no changes
in absolute species richness were associated with this variable.

Environmental variables
Adj-R²
Urbanization UHI Vegetation height Vegetation cover
Body size classes
Small ↓ (0.04) - - - 0.36
Medium - ↓ (0.02) - - 0.29
Large - ↓ (< 0.001) ↑ (< 0.001) ↓ (0.01) 0.60
Thermal affinity classes
Low temperatures - ↓ (0.03) - - 0.45
Intermediate temperatures - ↓ (0.002) ↑ (0.003) ↓ (< 0.001) 0.75
High temperatures - - - - 0.10

3.1.4 Discussion
[Link] Spider community composition and indicator species along an urbanization gradient
We showed that spider community composition varies between urban grasslands, depending on
their environmental characteristics. Urbanization rate, UHI intensity, near-ground temperature and
vegetation height were identified as factors dissociating the three clusters of sites with contrasting
environmental conditions. While some species are shared with similar abundance across the three
clusters (e.g., Pachygnatha degeeri, Pardosa cf. tenuipes, Tenuiphantes tenuis), we found that
species composition contrasted strongly in terms of occurrence and abundance between clusters
(Table 3.2). In particular, we found that ‘high vegetated rural’ and ‘short vegetated urban’ clusters
present indicator species.
In high vegetated rural sites, the two wolf spider species identified as indicators are closely related
species (i.e., Pardosa pullata and Pardosa prativaga). They are known to commonly co-occur in a
widely overlapping range of vegetation structures, although some differences in habitat preference
have been observed, particularly in case of interspecific competition (Hollander and Lof 1972). Both
species are typical of herbaceous vegetation (i.e., agricultural meadows), although P. pullata prefers
open microhabitats with low vegetation, whereas P. prativaga prefers denser and higher vegetation
(Hollander and Lof 1972). Our results show that near-ground temperature is the variable mostly
related to the abundance of P. pullata across the 36 study sites. Therefore, this result suggests that
the stronger association of P. pullata with high-vegetated grasslands (i.e., meadows) than with short-
vegetated grasslands (i.e., lawns) may not be determined by a particular habitat structure, but could
rather result mainly from locally adapted temperature conditions. We identified two common and
widespread linyphiid spiders, Erigone atra and Erigone dentipalpis, as indicator species associated
with urban sites characterized by intense UHI conditions, short and dense vegetation and high near-

103
PARTIE III

ground temperatures. These two species are phylogenetically close and co-occur frequently. They
share identical life cycles, niches and habits (de Keer and Maelfait 1988, Downie et al. 2000). Both
species have been well studied in agricultural landscapes and previous studies have reported high
dispersal capacities (i.e., by ballooning) as well as a particular ability to colonize new habitats in a
short time after human-induced disturbance. In the light of our results, these conclusions drawn
from the agricultural context can easily be transposed to the urban environment. If agricultural fields
are regularly disturbed by multiple operations throughout the year, urban grasslands may also be
subject to multiple mowing operations. The fact that E. atra and E. dentipalpis are indicators of short
vegetated grasslands found in urban areas is therefore not surprising, since these are also the most
frequently mown. In addition, Cryptachea blattea was indicator for urban areas. This species was
first described in New Zealand and is spreading rapidly throughout Europe, as evidenced by the first
records made in the last decade in several European countries (Germany in 2008 (Sührig 2010),
Britain in 2011 (Marriott 2012), Switzerland in 2013-2014 (Hänggi and Straub 2015), France in 2014
(Le Divelec et al. 2018), the Netherlands in 2014 (Bink 2014) and Ireland in 2019 (Nolan 2020)). The
majority of these new records were made in anthropogenic environments, such as ornamental
gardens or nurseries (Nolan 2020). The strong positive relationship observed between C. blattea
abundance and the urbanization rate is therefore in line with previous studies. It indicates that the
presence of C. blattea is probably related to the intensive use of horticultural plants, which are widely
planted in cities’ urban parks and green infrastructures. The absence of indicator species associated
with the ‘short vegetated rural’ cluster may be explained by heterogeneous environmental
characteristics of sites (Hacala et al. 2020).
[Link] Relationships between functional traits and environmental conditions
As a first step, to determine whether the spider communities belonging to the three identified
clusters were functionally distinct, we tested whether the community-averaged body size differed
between clusters. We observed a significant decrease in body size in urban spider communities
associated with the highest levels of urbanization, UHI intensity and near-ground temperature (Figure
3.7). This result is consistent with previous research showing that community-averaged body size of
terrestrial arthropods generally tends to decrease with increasing urbanization (Magura et al. 2004,
Gaublomme et al. 2005, Sadler et al. 2006). Furthermore, this relationship has recently been linked
to temperature increases in urban areas caused by the UHI effect (Merckx et al. 2018c). However,
our examination of the links between body size and environmental variables (taken individually) calls
this latter hypothesis into question. It revealed that vegetation structure (more that the UHI) was
related to changes in community-averaged body size. Interestingly, community-averaged dispersal
capacity was also related to local vegetation attributes, although no differences among clusters were
observed. This is consistent with previous studies, showing a lack of relationship between landscape
scale urbanization rate and community-averaged dispersal capacity of spiders (Buchholz et al. 2018).
It is therefore not surprising that the community-averaged functional traits (i.e., body size and
dispersal capacity) showed significant relationships with the vegetation height. The fact that short-
vegetated habitats, whether in rural or urban areas, host smaller, more dispersive species than high-
vegetated habitats is supported by previous findings concerning spiders. Indeed, community-
averaged body size is driven by local vegetation height (Delgado de la flor et al. 2020). In addition, a
decrease of vegetation structure complexity may prevent colonization and establishment of large
preys, which in turn may limit feeding opportunities for large spiders. Although we focus on the
community-averaged body size response, intraspecific changes in functional trait can occur along

104
 Chapitre 3

urban environmental gradients, resulting either from phenotypic plasticity or genetic adaptation
(Alberti et al. 2017). For example, moths showed that intraspecific body size changes toward smaller
individuals with increasing temperature. It is consistent with the observed global trend observed at
the community level, characterized by a negative relationship (Merckx et al. 2018a). In spiders,
however, only a few species have been the subject of such studies (e.g., (Johnson et al. 2019b)).
Future urban studies on spiders should compare the intraspecific variation in body size across
species. This approach would enable to determine whether community-averaged and intraspecific
changes are consistent, or whether the direction of relationships with environmental drivers is
species-dependent (Merckx et al. 2018a).
In a second step, we tested whether changes in community-averaged thermal affinity occurred
between clusters. We observed a significant increase in thermal affinity in the urban cluster,
compared to communities in the ‘high vegetated rural’ cluster. This suggests that thermophilic
species are favored by urban environments (Figure 3.7). Similar community-averaged functional
changes along an urbanization gradient have been observed in ants (Menke et al. 2011) and
grassland carabid communities (Piano et al. 2017). In these studies, changes were interpreted as a
result of species being selected to succeed in intense UHI conditions thanks to an affinity for high
temperatures (which diverges from our UHI-index based on daily minimum temperatures). Species
tolerating a wider range of temperatures were hypothesized to be favored in urban areas (Piano et
al. 2017). These authors found only the thermal affinity index based on maximum temperatures to
increase with urbanization, while the index based on minimum temperatures did not respond. This
latter assumption is contradicted by our results, since we observed a constant change, whether the
mean, minimum, or maximum temperature index was considered. If the range of temperature
tolerance does not change in our study system, interspecific variation in critical thermal limits (i.e.,
CTmax) may be of prime importance. For example, in a study exploring the relationship between
thermal tolerance of bee species and their ability to persist under intense UHI conditions, thermal
tolerance was shown to be a critical thermal trait for predicting species-specific responses to
warming (Hamblin et al. 2017). This study highlights the importance of species' thermal physiology
in shaping community compositions through functional filtering processes. This importance is also
supported by the large number of studies focusing on intraspecific variation in thermal tolerance
(Angilletta et al. 2007, Brans et al. 2017, Diamond et al. 2018c, 2018a, Sato and Takahashi 2022). An
alternative explanation for the observed change in community-averaged thermal affinities could be
an indirect effect of body size, through the covariation of species' thermal affinity and body size.
Intraspecific studies on spiders have already shown that larger individuals tend to prefer higher
temperatures (Cabon et al. 2023), and the general positive association between body size and
thermal tolerance is striking in arthropods (Blanckenhorn 2000, Nielsen and Papaj 2015, Klockmann
et al. 2016). However, as we found that species' trait for body size and mean thermal affinity are
weakly correlated (r = 0.22), we argue that the predominant patterns from intraspecific studies
cannot be extended to the community level in our system. Furthermore, communities sampled in
warm urban sites, and therefore belonging to the urban cluster, were composed of smaller species
than rural communities, which in itself invalidates this alternative interpretation. Nevertheless, it
cannot be excluded that traits other than thermal affinity are associated with biogeographical
distribution and directly or indirectly influence the ability of species to succeed in urban areas.

105
PARTIE III

Figure 3.7 - Map of Rennes (black line) and surrounding area. Symbols represent sampling sites. The shape of
the symbols indicates the cluster to which a sampling site belongs (circles = ‘high vegetated rural’, triangles =
‘short vegetated rural’, squares = ‘short vegetated urban’). Symbol size indicates the mean community body
size (from 3.30 mm to 5.73 mm). Symbol colour indicates the mean community thermal affinity, with white
symbols indicating low values (between 9.89 °C to 10,16 °C), light red symbols indicating intermediate values
(between 10.17 °C to 10,43 °C) and dark red symbols indicating high values (between 10.44 °C to 10,70 °C). The
atmospheric UHI (March 1 to September 30, 2022) is illustrated by a colour gradient from blue (low intensity;
minimum = 0 °C) to red (high intensity; maximum = 3 °C). Impervious surface is shown in grey.

[Link] Changes in functional trait community indices

Finally, we aimed to investigate the mechanisms associated with the observed changes in
community-averaged functional traits, by assessing the role of functional replacement versus
filtering between environmental clusters. We found that sites belonging to the urban cluster were
less rich in species than sites belonging to the two rural clusters. This confirms previous results
obtained on carabids, supporting the filtering process theory (Piano et al. 2017). Under this
condition, community-averaged functional changes result, at least partially, from a subset of species
with particular traits values, enabling them to establish in urban environmental conditions (Niemelä
and Kotze 2009, Piano et al. 2017). We further aimed to understand the mechanisms lying under the
observed community-averaged functional changes, by studying the variations in species richness
partitioned into classes of species characterized by particular functional trait values.
We found that urban sites were linked to a decreased in species richness in all body size classes (e.g.,
small, medium, large). This decrease in species number was more important in large species than in
small and medium species. Therefore, the decrease in community-averaged body size in urban sites
results from a size-related loss of species, with the urban filter being more effective for larger

106
 Chapitre 3

species. Furthermore, this acute loss of large species within urban spider communities could be the
direct result of a negative effect of temperature increase. Indeed, the number of large species was
strongly negatively related to the UHI intensity. This result is in line with the global prediction of a
warming-induced decrease in body-size within communities of ectothermic taxa (Scheffers et al.
2016). At the scale of a city, UHI could limit the body size of spider community by increased metabolic
costs in warmer conditions than in adjacent rural areas, allowing only small and medium sized species
to maintain themselves. Such filtering of large species reflects common temperature size rules
observed at the intraspecific level. For example, Bergmann’s rule states that warmer conditions (e.g.,
low latitudes or altitudes) are associated with intraspecific changes in body size in favor of smaller
individuals (Blanckenhorn and Demont 2004). Furthermore, according to Aktinson’s rule,
ectothermic individuals that have developed in warm conditions generally become smaller in the
adult stage than individuals from colder environments (Atkinson 1994). Furthermore, our analysis of
species richness confirms the importance of local vegetation structure as a driver of the community-
averaged body size. It also indicates that these variables are particularly determinant of the presence
of large species. In the light of our results, the loss of availability of vertical and horizontal structural
complexity of the vegetation through decreasing height and increasing density limits the
establishment of large species. This suggests that vegetation homogenization, e.g., by increasing
mowing frequency, limits the establishment of large spiders, possibly by reducing hiding
opportunities or the availability of large preys (Delgado de la flor et al. 2020).
The hypothesis of a functional filtering process is also supported by an examination of species
richness distribution within thermal affinity classes. Indeed, we found that the increased in
community-averaged thermal affinity in the urban cluster resulted from an increased proportion of
thermophilic species, combined with a reduced proportion of species associated with intermediate
temperatures. Closer examination of the absolute number of species reveals that these changes in
proportions are due to a relatively greater deficit of species with low and intermediate temperatures
affinities in the urban, compared to the rural clusters. As with body size, the difference in community-
averaged thermal affinity observed between urban and rural clusters is therefore the result of
decreasing species richness when comparing the rural clusters to the urban one. It indicates a
functional filtering process. Indeed, the absence of an absolute increase in thermophilic species in
sites belonging to the urban cluster shows that the observed functional trait change toward higher
temperature values is due to a filtering process against heat-sensitive species in urbanized areas.
Detailed analysis of individual variables showed that this functional filtering is linked to the
intensification of UHI conditions. This result provides a better understanding of the mechanisms.
Indeed, previous studies have linked changes in community-averaged temperature affinities to
global urbanization but have not formally identified the underlying driving variables. Our study
provides first evidence that temperature is the determining variable and goes further by showing
that landscape-scale UHI, rather than local temperature increase, filters species according to their
thermal affinities. Thus, the large heat zones generated by the UHI effect can be considered as
thermal barriers at the edge of the city, where only species physiologically able to cope with urban
warming persist. On the contrary, near-ground temperature in city has been shown to be a dominant
factor associated with the abundance of locally present species, as supported by previous studies on
different arthropod groups (Meineke et al. 2013, Hamblin et al. 2018, McGlynn et al. 2019).

107
PARTIE III

Conclusions
Our study shows that changes in the functional composition of spider community occurs at both
local and landscape scales and those functional traits are filtered at both scales. Landscape-scale UHI
negatively affects species richness by filtering species according to their thermal affinity, leading to
biotic homogenization at the community level. Furthermore, the local habitat structure is also
related to the functional composition of spider communities. Indeed, the reduction of vegetation
height led to a reduction in the number of species, with larger species being extirpated. As the
vegetation in urban sites was on average shorter than in rural sites in our study system, urban
communities changed towards low-diversity communities composed of small species. As predatory
arthropods, spiders play an important role in the trophic chain, controlling a wide range of
invertebrates. Consequently, lower abundance of specific functional traits due to reduced species
richness may lead to imbalance in urban ecosystems caused by a lack of biocontrol at lower trophic
levels. Here, we provide further evidence that complexifying grassland vegetation, e.g., by increasing
mowing height or decreasing management frequency, enables the establishment of more
functionally diverse communities. Improving the structure of green infrastructures could therefore
not only compensate for the loss of species induced by urban warming at taxonomic and functional
levels, but also help to locally mitigate the effects of the UHI (Francoeur et al. 2021), thus offering
valuable co-benefits to urban climate and biodiversity.

108
 Chapitre 3

3.1.5 Appendices
Appendix 3.1 - Species list of spiders and related traits based on literature data (1/5).
Body length of
females (mm) Ballooning
Mean Maximal Minimal
(Roberts 1985, (Blandenier thermal thermal thermal
Species 1987, Macías- and Fürst affinity affinity affinity
Hernández et al. 1998, Bell et (°C) (°C) (°C)
2020, Nentwig al. 2005)
et al. 2023)
Agroeca brunnea 8.0 no 9.1 13.8 4.8
Agroeca inopina 7.5 no 13.1 18.2 8.2
Agyneta affinis 2.0 no 8.9 13.4 4.6
Agyneta decora 2.5 no 7.8 11.9 3.7
Agyneta mollis 2.0 yes 10.5 15.4 5.9
Agyneta rurestris 2.4 yes 9.7 14.5 5.3
Alopecosa cuneata 8.0 no 9.8 14.5 5.3
Alopecosa pulverulenta 10.0 yes 10.3 15.1 5.7
Antistea elegans 3.0 no 9.6 14.3 5.1
Argenna subnigra 2.5 no 9.2 13.8 4.8
Asagena phalerata 5.0 no 10.3 15.1 5.7
Atypus affinis 15.0 yes 10.8 15.5 6.4
Aulonia albimana 4.5 yes 10.4 15.4 5.7
Bathyphantes gracilis 2.5 yes 10.1 14.9 5.6
Centromerita bicolor 3.5 yes 9.1 13.8 4.7
Centromerus sylvaticus 4.0 yes 9.3 13.9 4.9
Ceratinella brevis 1.5 no 9.7 14.5 5.3
Cicurina cicur 7.0 no 9.5 14.2 5.1
Civizelotes civicus 4.3 no 12.4 17.0 8.1
Clubiona diversa 5.0 no 8.9 13.4 4.6
Clubiona neglecta 8.0 no 10.0 14.8 5.3
Clubiona terrestris 7.0 no 11.3 16.5 6.4
Cnephalocotes obscurus 1.9 yes 8.8 13.3 4.5
Collinsia inerrans 2.5 yes 9.1 13.6 4.8
Cryptachaea blattea 3.2 yes 11.0 15.5 6.7
Dicymbium nigrum 2.5 yes 9.2 13.8 4.9
Dicymbium tibiale 2.6 no 8.7 13.2 4.4

109
PARTIE III

Appendix 3.1 - Species list of spiders and related traits based on literature data (2/5).
Body length of
females (mm) Ballooning
Mean Maximal Minimal
(Roberts 1985, (Blandenier thermal thermal thermal
Species 1987, Macías- and Fürst affinity affinity affinity
Hernández et al. 1998, Bell et (°C) (°C) (°C)
2020, Nentwig al. 2005)
et al. 2023)
Diplocephalus latifrons 2.0 no 9.0 13.5 4.7
Diplocephalus picinus 1.0 no 9.3 13.9 5.0
Diplostyla concolor 3.0 yes 10.0 14.8 5.4
Drassodes cf. lapidosus 18.0 no 10.3 14.9 6.0
Drassodes pubescens 9.0 no 9.6 14.4 5.1
Drassyllus lutetianus 7.5 yes 9.7 14.4 5.2
Drassyllus praeficus 6.0 yes 10.1 15.1 5.5
Drassyllus pusillus 5.0 yes 9.3 13.9 4.9
Drassyllus villicus 6.9 yes 11.0 16.1 6.3
Dysdera crocata 15.0 no 12.4 17.8 7.3
Dysdera erythrina 10.0 no 10.6 15.2 6.1
Enoplognatha latimana 6.0 no 10.7 15.6 6.2
Enoplognatha mordax 8.5 yes 10.1 14.8 5.8
Enoplognatha testacea 3.2 no 18.5 25.8 11.6
Enoplognatha thoracica 4.0 no 10.5 15.4 5.8
Episinus truncatus 4.0 no 11.2 16.4 6.4
Eratigena picta 7.0 no 11.2 16.1 6.7
Erigone atra 2.8 yes 9.2 13.7 4.9
Erigone dentipalpis 2.6 yes 10.7 15.6 6.1
Ero furcata 3.2 yes 9.4 14.1 4.9
Ero tuberculata 4.0 no 9.6 14.4 5.2
Euophrys frontalis 5.0 no 11.7 17.0 6.6
Gnathonarium dentatum 3.0 yes 10.1 15.0 5.5
Gongylidiellum vivum 1.9 yes 10.7 15.5 6.1
Hahnia nava 2.0 no 10.7 15.4 6.2
Haplodrassus dalmatensis 6.5 no 9.5 14.1 5.1
Haplodrassus signifer 9.0 no 11.9 17.1 7.1
Hypomma cornutum 2.9 yes 10.7 15.6 6.1

110
 Chapitre 3

Appendix 3.1 - Species list of spiders and related traits based on literature data (3/5).
Body length of
females (mm) Ballooning
Mean Maximal Minimal
(Roberts 1985, (Blandenier thermal thermal thermal
Species 1987, Macías- and Fürst affinity affinity affinity
Hernández et al. 1998, Bell et (°C) (°C) (°C)
2020, Nentwig al. 2005)
et al. 2023)
Lasaeola prona 2.5 no 9.4 14.0 5.0
Lathys humilis 2.5 yes 9.1 13.6 4.7
Maso sundevalli 1.8 no 10.5 15.4 5.9
Mermessus trilobatus 2.1 no 9.7 14.4 5.2
Micaria albovittata 8.0 no 10.0 14.7 5.6
Micaria micans 4.1 yes 12.3 17.6 7.4
Micrargus herbigradus 1.9 yes 8.5 12.9 4.4
Micrargus subaequalis 2.0 yes 9.6 14.2 5.3
Microneta viaria 3.0 yes 9.8 14.6 5.2
Monocephalus fuscipes 2.2 yes 9.9 14.6 5.5
Neottiura bimaculata 3.3 yes 10.9 15.2 6.8
Neottiura suaveolens 2.3 no 10.2 15.0 5.6
Neriene clathrata 5.0 no 10.2 15.3 5.6
Oedothorax apicatus 3.3 yes 10.6 15.6 5.9
Oedothorax fuscus 2.9 yes 9.3 13.9 4.9
Oedothorax retusus 3.0 yes 10.0 14.6 5.7
Ostearius melanopygius 2.6 no 8.8 13.3 4.5
Ozyptila brevipes 4.0 no 9.0 13.6 4.8
Ozyptila praticola 4.0 no 10.7 15.4 6.3
Ozyptila sanctuaria 4.0 yes 9.7 14.4 5.2
Ozyptila simplex 5.0 yes 9.6 14.3 5.1
Pachygnatha clercki 7.0 yes 13.0 18.3 7.9
Pachygnatha degeeri 3.7 yes 11.3 16.5 6.5
Palliduphantes ericaeus 1.9 yes 9.9 14.8 5.3
Palliduphantes insignis 2.5 no 10.7 15.6 6.1
Palliduphantes pallidus 2.3 no 8.9 13.5 4.6
Panamomops sulcifrons 1.5 no 9.6 14.3 5.2
Pardosa amentata 8.0 yes 9.7 14.4 5.2

111
PARTIE III

Appendix 3.1 - Species list of spiders and related traits based on literature data (4/5).
Body length of
females (mm) Ballooning
Mean Maximal Minimal
(Roberts 1985, (Blandenier thermal thermal thermal
Species 1987, Macías- and Fürst affinity affinity affinity
Hernández et al. 1998, Bell et (°C) (°C) (°C)
2020, Nentwig al. 2005)
et al. 2023)
Pardosa cf. tenuipes 6.5 yes 11.2 16.0 6.8
Pardosa hortensis 5.5 no 11.1 16.0 6.5
Pardosa nigriceps 7.0 yes 9.7 14.4 5.2
Pardosa palustris 6.0 yes 9.2 13.8 4.9
Pardosa prativaga 6.0 yes 9.9 14.8 5.4
Pardosa pullata 6.0 yes 9.8 14.5 5.3
Pardosa saltans 5.7 no 10.5 15.3 5.8
Pardosa vittata 6.3 no 11.9 17.4 6.8
Pelecopsis parallela 2.0 yes 9.6 14.3 5.1
Philodromus praedatus 6.0 yes 11.6 16.7 6.8
Phlegra fasciata 7.0 yes 10.0 14.8 5.4
Phrurolithus festivus 3.0 no 10.0 14.6 5.6
Phrurolithus minimus 3.5 no 9.6 14.3 5.2
Phycosoma inornatum 3.0 no 9.9 14.7 5.5
Piratula latitans 5.0 no 11.3 16.5 6.5
Pisaura mirabilis 15.0 yes 10.7 15.6 6.1
Pocadicnemis juncea 2.2 no 10.8 15.7 6.1
Porrhomma pygmaeum 2.4 yes 9.1 13.7 4.8
Prinerigone vagans 3.2 yes 9.5 14.3 5.1
Psilochorus simoni 2.5 no 12.6 17.9 7.7
Robertus arundineti 2.5 yes 10.3 15.1 5.7
Robertus lividus 4.0 no 10.3 15.1 5.7
Saaristoa abnormis 4.0 no 9.0 13.6 4.7
Scotina palliardii 3.5 no 8.7 13.1 4.4
Stemonyphantes lineatus 6.4 no 8.8 13.4 4.5
Styloctetor compar 2.3 no 9.5 14.2 5.0
Synageles venator 4.0 no 9.2 13.8 4.7
Tallusia experta 3.6 yes 9.7 14.4 5.3

112
 Chapitre 3

Appendix 3.1 - Species list of spiders and related traits based on literature (5/5).
Body length of
females (mm) Ballooning
Mean Maximal Minimal
(Roberts 1985, (Blandenier thermal thermal thermal
Species and Fürst
1987, Macías- affinity affinity affinity
Hernández et al. 1998, Bell et (°C) (°C) (°C)
2020, Nentwig al. 2005)
et al. 2023)
Tapinocyba praecox 1.8 yes 8.9 13.5 4.6
Tenuiphantes flavipes 2.5 yes 9.3 13.6 5.0
Tenuiphantes tenuis 3.2 yes 10.6 15.5 6.1
Tibellus oblongus 10.0 yes 10.2 15.1 5.7
Tiso vagans 2.2 yes 11.2 16.2 6.5
Trachyzelotes pedestris 8.0 no 10.2 15.1 5.7
Trichoncus saxicola 2.1 no 10.0 15.0 5.4
Trochosa ruricola 14.0 no 10.5 15.5 5.9
Trochosa terricola 14.0 yes 10.0 14.7 5.5
Walckenaeria acuminata 4.0 yes 10.1 15.0 5.5
Walckenaeria antica 2.6 yes 9.8 14.6 5.3
Walckenaeria atrotibialis 2.8 no 8.9 13.4 4.7
Walckenaeria cucullata 2.7 no 8.9 13.4 4.6
Walckenaeria incisa 3.2 no 10.0 14.6 5.6
Xysticus acerbus 8.0 no 11.4 16.7 6.4
Xysticus cristatus 8.0 yes 9.8 14.6 5.3
Xysticus erraticus 8.0 no 9.6 14.3 5.1
Xysticus kochi 8.0 yes 10.6 15.7 5.9
Zelotes apricorum 9.0 no 10.0 14.7 5.6
Zelotes electus 5.5 no 9.4 14.0 5.0
Zelotes latreillei 8.0 yes 9.3 14.0 5.0
Zelotes petrensis 7.0 no 8.9 13.5 4.5
Zelotes tenuis 9.6 no 14.2 20.0 8.9
Zodarion italicum 3.0 no 12.0 17.3 6.9
Zodarion rubidum 4.4 no 10.3 15.0 5.8
Zora spinimana 6.5 yes 10.1 14.9 5.5

113
PARTIE III

3.2 Chapitre 4 : Les traits d’histoire de vie modulent la contrainte

exercée par le climat urbain sur la taille corporelle

Résumé
Dans ce chapitre, nous comparons les relations intraspécifiques entre la température et la taille
corporelle de 2283 individus d’araignées représentant 11 espèces appartenant à diverses familles
(Hahniidae, Linyphiidae, Lycosidae et Tetragnathidae). Les araignées ont été collectées sur 36
prairies et pelouses, réparties le long d'un gradient d'ICU.
Des changements intraspécifiques de taille corporelle sont reliés à la température, mais uniquement
chez les Lycosidae Pardosa prativaga et Pardosa pullata, alors qu'aucun effet clair n’est observé pour
les autres espèces examinées. Chez ces deux espèces, le sexe est un facteur important qui module
les réponses individuelles : les femelles sont plus affectées que les mâles par la température.
Les amplitudes contrastées des réponses observées entre les Lycosidae et les autres espèces
indiquent que les espèces de grande taille, peu dispersives et univoltines sont plus fortement
contraintes par la température que les petites espèces bivoltines qui peuvent disperser par voie
aérienne (ballooning). Ces résultats sur les araignées soulignent que la prise en compte des traits
d'histoire de vie spécifiques aux espèces peut aider à comprendre des réponses parfois contrastées
au sein d’un même groupe taxonomique. Cette approche peut être essentielle pour dégager des
patrons généraux de changements phénotypiques chez les arthropodes, sous l'effet du
réchauffement anthropique.

Article publié chez Functional Ecology :

Cabon, V., H. Quénol, B. Deletre, L. Copin, V. Dubreuil, and B. Bergerot. 2024. Body size responses to
urban temperature variations are driven by life history traits in spiders. Functional Ecology 38:1578–
1589.

114
 Chapitre 4

Body size responses to urban temperature variations are driven by life history
traits in spiders

Valentin Cabon1,2, Hervé Quénol2,3, Benoît Deletre1, Louis Copin1, Vincent Dubreuil2,3, Benjamin
Bergerot1,2

1. University of Rennes, CNRS, UMR 6553 ECOBIO, Rennes, France

2. LTSER ZA Armorique, F-35000 Rennes, France
3. University of Rennes 2, CNRS, UMR 6554 LETG, Rennes, France

Abstract

1. Urban ecosystems exhibit altered environmental conditions compared to the rural surroundings,
including higher temperatures, the so-called urban heat island (UHI) effect. Along urbanisation
gradients, temperature variations occur at various scales ranging from the landscape to the
microhabitat, with possible consequences on living organisms. As ectothermic animals, arthropods
are particularly affected by UHIs because warming may directly induce physiological effects on their
metabolism.
2. Body size is a commonly used metric to investigate arthropod responses to thermal stress, but the
scale-dependent role of temperature variation as predictor of body size remains unclear. To assess
temperature-size relationships across spatial scales and test if a common pattern to several species
emerges, we performed morphological measurements on 2283 individuals representing 11 spider
species belonging to Hahniidae, Linyphiidae, Lycosidae and Tetragnathidae. Spiders were collected
on 36 grasslands distributed along a broad UHI gradient.
3. Intraspecific shifts in body size were related to temperature in the Lycosids Pardosa prativaga and
Pardosa pullata at the landscape and local scales, respectively, whereas no clear effects were
observed regarding the other examined species. In both species, sex was an important endogenous
factor that modulated individual responses: females were more affected than males by temperature.
4. The contrasting amplitudes of the responses observed among Lycosids and the other species indicate
that large, low dispersive and univoltine species are under a stronger influence of temperature than
small bivotline species that can disperse aerially. Our results on spiders highlight that considering
species-specific life history traits helps to understand contrasting responses among species, and this
may be essential for drawing general patterns of phenotypic changes among arthropods under
anthropogenic warming.

Key words: Araneae, climate warming, intraspecific trait variation, morphological traits, urban
ecology, urban heat island

115
PARTIE III

3.2.1 Introduction

In cities, temperature increase not only contributes to global warming but is also exacerbated by it
and is commonly referred to as the urban heat island (UHI) phenomenon (Oke et al. 2017). UHIs are
characterised by an elevation of mean temperatures in urban centres relative to nearby rural areas
and regional average temperatures (Foissard et al. 2019b). Assessing changes in body size is a
recognised efficient method to investigate adaptations of individuals under environmental thermal
stress, notably in UHI contexts (Johnson et al. 2019a, 2020a). In a recent meta-analysis, Lövei &
Magura (2022) summarised body size as having crucial ecological implications from the individual to
the ecosystem level. Body size is closely related to several functional and life history traits such as
morphology, physiology, survival or fecundity, with consequences on the fitness, ecological niches
and spatial distribution of individuals (Wanninger 2015). In addition, body size influences biotic
interactions, and in turn community dynamics, food webs, and ecological networks (Lövei and
Magura 2022b). Among other organisms, arthropods are critical players in urban ecosystems
(Johnson et al. 2019a), by providing a wide range of ecosystem services (e.g., pollination,
decomposition, predation). As ectotherms, urban arthropods are particularly sensitive to warming
because increased temperature has a physiological effect on their metabolism (Angilletta et al. 2007,
Brans et al. 2017, Diamond et al. 2018c, 2018b, Sato and Takahashi 2022). Therefore, investigating
changes in the body size of arthropods may provide a better understanding of urbanisation-induced
stress and instability, including the consequences of UHIs on arthropod species (Merckx et al. 2018b).
Ecological and evolutionary body size responses to temperature have been observed across
geographical thermal gradients (latitudinal and altitudinal ones; Ameline et al. 2018; Beckers et al.
2020; Yang et al. 2018) for many arthropods – e.g., spiders –, and more recently across thermal
gradients in urban areas (Johnson et al. 2019, Malmos et al. 2021). Thermal variations can be
measured at various scales ranging from the broad landscape to the microhabitat along urbanisation
gradients (Cabon et al. 2024a). Air temperature 3 m above the surface is commonly recorded to
monitor UHIs (Foissard et al. 2019b), i.e., large areas of heat at the landscape scale (Oke et al. 2017,
Foissard et al. 2019b). Yet, arthropods also experience their thermal micro-environment near to the
ground level, where fine-scale temperature variations can be highly heterogenous due to habitat
microstructures (Harvey et al. 2020, 2022, Pincebourde and Woods 2020, Ma et al. 2021). However,
these two scales are rarely considered simultaneously in studies on body size because sensor
networks dedicated to the monitoring of temperature at the landscape and local scales are rarely
available.
The body size responses of arthropods to thermal gradients also vary according to the life history
strategies and ecological characteristics of the species (Forster et al. 2012, Horne et al. 2015, Fenberg
et al. 2016, Wilson-Brodie et al. 2017). For instance, the dispersal strategy determines how body
size is linked to UHI in several arthropod groups (Merckx et al. 2018b). Since escaping from adverse
conditions within a fragmented environment requires high dispersal capacities, increased mobility
should enhance the chances of survival of a focus taxon (Merckx et al. 2018a). Among arthropod
groups, positive and negative size-dispersal links can be found and this parameter has been
evidenced to play a central role in the way community-averaged body size respond to UHI (Merckx

116
 Chapitre 4

et al. 2018b). Accordingly, taxa showing negative size-dispersal links should become smaller within
warm fragmented urban environments. In addition, life history strategies can influence how the
environment constrains the development of individuals. For example, the developmental time may
determine whether growing larger should be prioritised at the expense of early maturity under harsh
thermal conditions (Wilson et al. 2019), or if a converse trade-off occurs. Voltinism is expected to be
an important explanatory factor of interspecific patterns of temperature-size relationships (Horne et
al. 2015). Terrestrial arthropods with a univoltine life cycle (i.e., one generation per year) and long
developmental times are expected to respond differently to warming than bivoltine and multivoltine
species (i.e., two or more generations per year, respectively) (Blanckenhorn and Demont 2004, Opell
2010). At the intraspecific level, body size changes along temperature clines may also vary seasonally
because individuals from distinct generations may grow at different rates and experience contrasting
thermal conditions (Wilson et al. 2019). For instance, spiders that emerge early in the season can be
larger than those that emerge later (Masumoto 1994). In addition, the body size response of spiders
to temperature has been repeatedly found to be sex-biased (Høye and Hammel 2010, Bowden et al.
2013), possibly due to contrasting activity patterns (Harvey and Dong 2023).
The present study aims to (1) explore which temperature-size patterns occur in common spider
species along urban thermal clines, (2) evaluate the respective roles of temperature (at the
landscape and microhabitat scales), the biotic environment and endogenous factors in shaping
intraspecific variations in body size, and (3) determine if trait syndromes explain interspecific
differences in the temperature-size rule followed by different species. To reach these objectives, we
conducted morphological measurements on 11 spider species collected along urban temperature
gradients. Spiders are a particularly well suited model because communities are highly abundant and
diverse in cities (Varet et al. 2014b, Meineke et al. 2017b, Piano et al. 2020c). Moreover, urban
populations exhibit diversified behavioural and/or physiological responses to UHIs (Johnson et al.
2019a, 2020a). According to recent results on Lycosid spiders obtained in natural areas (Ameline et
al. 2018), and to the general trend observed in univoltine arthropods, we expected intraspecific
relationships to vary according to species-specific trait syndromes. For instance, we expect positive
relationships between body size and urban temperature within large, univoltine and low dispersive
species. Although knowledge is still scarce regarding spiders from other families, we hypothesized,
based on the reported importance of life history traits for temperature-size relationships (Horne et
al. 2015), that negative size-temperature relationships will prevail in small, highly dispersive and
multivoltine species (Blanckenhorn and Demont 2004) such as Linyphiids.

3.2.2 Materials & Methods

[Link] Study area and sampling design
We conducted this study in Rennes (northwestern France, 48°06′N-1°40′W) and its surroundings.
Rennes is a city of 227,000 inhabitants under a strong oceanic climate influence. According to a long-
term urban climate monitoring program, Rennes faces regular and strong UHI events (Dubreuil et al.
2020) that provide ideal experimental conditions to investigate if morphological changes occur along
a thermal gradient. We chose grasslands as a model ecosystem because they represent a major part
of urban green spaces in Rennes (445 ha, 56%) (Chollet et al. 2018) and are available from the centre
to the outskirts of the city. Furthermore, the diversified management strategies applied across
Rennes result in grassland structures ranging from short and dense to high and heterogenous

117
PARTIE III

vegetation. These conditions provide a broad range of microclimatic conditions, a determinant

ecological driver of arthropod fauna (Harvey et al. 2020, 2022, Pincebourde and Woods 2020, Ma et
al. 2021). Finally, grasslands support abundant spider communities along a broad UHI gradient in
Rennes (Cabon et al. 2024a) ensuring a suitable sampling size to perform morphological
comparisons. Our site selection procedure aimed to maintain the cofounding effect of urbanisation
at a low level, in order to focus on the effects of temperature. To do so, we first performed a spatial
correlation procedure allowing for the identification of areas where urbanisation (i.e., the proportion
of built-up areas) and UHIs are weakly correlated (P < 0.05), as thoroughly described in Cabon et al.
(2024a). Then, we identified all grassland patches (dataset ocsge_rm_2017; [Link];
accessed on 1 December 2021) overlapping the areas previously identified as ‘weakly correlated’
and selected 36 accessible sites encompassing a broad gradient of UHI (Figure 3.8). Three pitfall traps
per sampling site were installed and opened 15 days per month from March to September 2022. We
used plastic cups (85 mm in diameter and 115 mm in height), filled with 150 mL of saline solution at
a concentration of 100 g.L-1 to improve spider preservation and a drop of neutral soap to prevent
floating. The three traps were placed 5 m apart and oriented north, southeast and southwest,
forming a triangle to ensure capture efficiency (Ward et al. 2001). Each month, spiders were sorted
and stored in individual vials containing 70% ethanol. All mature spiders were identified to the
species level using identification keys (Roberts 1985, 1987). Nomenclature followed the World Spider
Catalogue version 22 ([Link] accessed on 1 December 2022). Prior to the start of the
fieldwork, sampling authorisations were granted by each city council and land owner concerned. Our
study did not require ethical approval.

Figure 3.8 - Map of the study area. The atmospheric UHI (from 1st March to 30th September 2022) is illustrated
by a coloured gradient ranging from blue (low intensity; minimum = 0 °C) to red (high intensity; maximum = 3
°C). Small black dots, temperature sensors at 3 m height used to record the atmospheric UHI. Large coloured

118
 Chapitre 4

dots, sampling sites with near-ground temperature measured at 5 cm height, ranging from blue (low intensity;
minimum = 0 °C) to red (high intensity; maximum = 5 °C). Grey, impervious surfaces.

[Link] Species selection

We analysed the distribution of the spider species and selected the most abundant species evenly
distributed along the UHI gradient. To do so, we sorted the sites according to their UHI intensity in
three classes labeled "high," "medium," and "low", using quantiles as cut-off values. Then, we
selected the species with at least 20 individuals per UHI class. The selection resulted in six Linyphiid
species, three Lycosids, one Tetragnathid and one Hahniid ( Table 3.5). The species’ life history traits
related to dispersal (i.e., ballooning capacity) and voltinism were retrieved from the literature ( Table
3.5). Species selection resulted therefore to two main trait syndromes: (1) small, highly dispersive,
bivoltine species and (2) large, low dispersive, univoltine species. In addition, one species
(Pachygnatha degeeri) shared characteristics with both groups (Table 3.5).

119
PARTIE III

Table 3.5 - Summary of the characteristics of the 11 selected species. ‘Ballooning capacity’ and ‘Voltinism’
columns, possible interspecific comparisons. ‘Sex’ and ‘Generation/Cohort’ columns, possible intraspecific
comparisons. The generation of each multivoltine species is indicated by the number following the letter G. The
cohort of each univoltine species showing cohort splitting is indicated by the number following the letter C.
“Intraspecific competition” column, mean number (and standard deviation) of co-occurring individuals from a
same species at the sampling sites.
Voltinism
(Duffey 1956,
Hallander 1970a,
Ballooning Hollander 1971, De Intraspecific
capacity Keer and Maelfait Generation competition (total
1987a, 1987b, (G) or number of
Family Species (Simonnea Sex
Alderweireldt and De individuals
u et al. Cohort (C) trapped per
Keer 1988, De Keer
2016) sampling site)
and Maelfait 1988,
de Keer and Maelfait
1988, Alderweireldt
and De Keer 1990)
Bathyphantes
Linyphiidae High Bivoltine m /f G2 19.97 ± 17.92
gracilis
Diplostyla
Linyphiidae High Bivoltine m/f G2 14.03 ± 18.42
concolor
Erigone
Linyphiidae High Bivoltine m G2 36.19 ± 57.76
dentipalpis
Hahniidae Hahnia nava High Bivoltine m G1 16.25 ± 17.87
Oedothorax
Linyphiidae High Bivoltine m/f G1/G2 29.94 ± 89.68
fuscus
Oedothorax
Linyphiidae High Bivoltine m G2 24.53 ± 87.43
retusus
Pachygnatha
Tetragnathidae High Univoltine m C1/C2 137.14 ± 120.72
degeeri
Pardosa cf.
Lycosidae Low Univoltine m/f C1/C2 72.31 ± 110.21
tenuipes
Pardosa
Lycosidae Low Univoltine m/f - 8.75 ± 14.72
prativaga
Pardosa
Lycosidae Low Univoltine m/f - 18.06 ± 28.73
pullata
Tenuiphantes
Linyphiidae High Bivoltine m/f G2 33.58 ± 24.74
tenuis

120
 Chapitre 4

[Link] Morphological measurements

We conducted morphological measurements in the laboratory with a stereomicroscope (Leica M205
C) equipped with a binocular camera (Leica DMC4500). For each individual of each selected species,
we took pictures of the prosoma and right leg IV (Pétillon et al. 2018, Cabon et al. 2023) using
software LEICA LAS X. The length and width of the prosoma and the length of the leg segments were
measured to the nearest 0.01 mm with software ImageJ (Rasband 2007) (Figure 3.9).

Figure 3.9 - Laboratory morphological measurements of the 11 selected spider species. Blue segments, prosoma
measurements (light blue, width; dark blue, length). Pink segments, measurements of right legs IV.

121
PARTIE III

[Link] Explanatory variables

[Link].1 Atmospheric urban heat island (UHI)

Thanks to a long-term climate observation network established in Rennes in 2004 (Rennes Urban
Network; [Link] the UHI can be accurately measured across space and time
(Foissard et al. 2019b, Dubreuil et al. 2020). Since 2020, temperatures have been monitored by 30
DAVIS Vantage-Pro-2 automatic weather stations (Davis Instruments Corporation, Hayward, CA
94545, USA) and 93 RisingHF RHF1SOO1 connected temperature sensors (LoRaWAN Solutions
RisingHF, Shenzhen 518057, China) distributed across rural, suburban and city centre areas and
collecting data at a standardised 3 m height and at 60 (Vantage-Pro-2) and 15 min (RisingHF)
frequencies. We calculated the UHI from the 2022 dataset as the mean difference between the
minimum daily temperatures measured by each sensor and the minimum daily temperatures
measured at a rural reference station. The rural reference station is located about 11 km in the north
of Rennes (Melesse, 48°12'18.1"N 1°40'52.8"W). It is characterised by agricultural land with hedges
and the local climate is considered as not influenced by urbanisation (Foissard et al. 2019b).
Therefore, the resulting metric corresponds to the number of °C above the temperature measured
at the rural reference station located outside the city. Using these differences, we spatialised the
UHI by inverse distance weighting interpolation to the entire study area at 100 m resolution. As
seasonal variations in UHI are known to occur, we averaged monthly raster maps to generate a single
UHI map corresponding to the exact period of the Tpref measurements (i.e., March to September
2022). Finally, we extracted the UHI values corresponding to the raster cells on which the 36
sampling sites were located.

[Link].2 Near-ground temperature

At the microclimate scale, we monitored near-ground temperature 5 cm above the ground using a
specific network of temperature loggers composed of 21 Lascar EL-USB-2+ (Lascar Electronics Ltd.,
Whiteparish, Wiltshire SP5 2SJ, UK) and 15 Tinytag Talk 2 TK-4023 (Gemini Data Loggers Ltd.,
Chichester, West Sussex PO198UJ, UK). At each sampling site, the logger was placed in the direct
vicinity of the north-headed pitfall trap and recorded temperature every 15 minutes. As for the
atmospheric UHI, we calculated the differences between the daily minimum temperatures obtained
from the station showing the lowest averaged near-ground temperature (Cesson-Sévigné,
48°06'36.6"N-1°36'29.9"W) and that of each of the other sites. We obtained a single near-ground
temperature value by averaging daily differences throughout the study period.

[Link].3 Intraspecific competition

To account for the possible effects of intraspecific competition and test if local abundance affected
body size, we estimated the population density at each sampling site. To do so, we calculated the
total number of trapped individuals of a given species per sampling site throughout the study period
(Beckers et al. 2020).

122
 Chapitre 4

[Link].4 Endogenous factors

Considering the sex-biased response of body size to temperature reported in previous studies (Høye
et al. 2009, Høye and Hammel 2010, Bowden et al. 2013), we included male and female individuals
whenever both sexes were sufficiently represented. Out of eleven selected species, only males were
considered in Hania nava, Erigone dentipalpis, and Oedothorax retusus because females were
underrepresented. Pachygnatha degeeri females were not considered because their internal
epigyne did not allow for reliably checking maturity. In addition, we analysed the phenological
variations of species abundance to control for possible seasonal changes in body size throughout the
year. All individuals from a monthly peak and those from the previous and next months were selected
for measurement, except for Pardosa prativaga, whose peak was not clear so that we had to select
five months instead of three to consider enough individuals ( Appendix 3.2). By doing so, we also
characterised individuals according to their generation in the case of bivoltine species, or their
cohort when univoltine species exhibited cohort splitting (Samu et al. 1998).

[Link] Data analysis

[Link].1 Correlations among traits
All statistical analyses were performed in R version 4.2.2 (R Core Team 2022). We first performed an
analysis of correlations among the measured morphological attributes to determine if prosoma
width could be considered as a reliable proxy of the body size of each selected species. The prosoma
is widely considered as a reliable proxy of spiders’ body size because this sclerotised body part is less
prone to deformation and damage than their unsclerotised soft abdomen (Beckers et al. 2020).
Prosoma width was significantly positively correlated to all other attributes in every species (except
to femur length in O. retusus) (Appendix 3.3). Therefore, we considered prosoma width as an overall
proxy of body size and referred to it as body size (Høye et al. 2009, Ameline et al. 2018, Hein et al.
2019, Beckers et al. 2020).

[Link].2 Effects of environmental and endogenous factors on body size

First, we pooled all individuals per species and sex, and removed outliers with median body size
values outside the 95% confidence interval to reduce statistical noise in the dataset (Cabon et al.
2023). Then, we assessed the effect of environmental variables and endogenous factors on body size
by computing one generalised linear mixed model (GLMM) per species for a total 11 models, using
the ‘lmer’ function in ‘lme4’ package. ‘Intraspecific competition’, ‘near-ground temperature’ and
‘UHI’ were included as fixed effects, and so were ‘sex’ and/or ‘generation’ whenever available.
‘Sampling site’ was added as a random effect to correct for a potential site-specific bias. As the body
size response to environmental changes is likely to differ strongly among sexes and generations of
spiders (Høye et al. 2009, Bowden et al. 2013, Yang et al. 2018), we added an interaction term
between ‘sex’, ‘generation’ and every other explanatory variable included as a fixed effect. In the
case of Oedothorax fuscus, generation G1 was only represented by females, whereas G2 had
individuals from both sexes available. Therefore, we split the data related to this species by (1)
excluding G1 individuals to include ‘sex’, but not keeping ‘generation’ as an explanatory variable in a
first sub-model (O. fuscus (A), Table 3.6), and (2) excluding G2 male individuals to include ‘generation’,
but not ‘sex’ in a second sub-model (O. fuscus (B), Table 3.6). We selected the best models among all

123
PARTIE III

possible candidate models using the ‘dredge’ function and averaged the top-ranking models showing
a delta AIC < 2, using the ‘[Link]’ function in ‘MuMin’ package (Bartoń 2016). When the final
models included interactions involving ‘sex’ or ‘generation’, we computed linear models to identify
the underlying significant relationships. We tested the significance of the explanatory variables by
performing type III ANOVA tests using the ‘Anova’ function in ‘car’ package (Fox et al. 2007). Finally,
we calculated the coefficients of determination to assess the proportion of variation explained by
the selected models, shown in Table 3.6.

3.2.3 Results
[Link] Sampling results
The morphological attributes of 2283 adult spiders were measured on the 36 sampling sites, and
data from 2174 individuals were kept in the dataset for analysis after removing 109 outliers. The
largest species in averaged size were P. prativaga (2.19 ± 0.13 mm), P. pullata (2.03 ± 0.09 mm) and
P. cf. tenuipes (1.98 ± 0.28 mm), whereas the smallest ones were Bathyphantes gracilis (0.66 ± 0.05
mm) and Hahnia nava (0.66 ± 0.05 mm) (Appendix 3.4). The UHI at the sampling sites ranged from 0.8
to 2.9 °C, whereas near-ground temperature ranged from 0 to 5.1 °C.

[Link] Predictors of body size

[Link].1 Effect of the thermal environment
Thermal conditions were found related to the body size of specific species (Table 3.6). The body size
of B. gracilis and P. prativaga was significantly related to the ‘UHI: sex’ interaction. In accordance
with the procedure described in the 'data analysis' section, we computed linear models to identify
the underlying significant relationships, since these final models included interactions involving ‘sex’.
The linear models computed separately on both sexes of B. gracilis revealed that neither males (α =
-0.006; P > 0.05), nor females (α = 0.002; P > 0.05) were significantly related to the UHI. In P.
prativaga, only females were positively related to the UHI (α = 0.06; P < 0.001, Figure 3.10). The body
size of P. pullata was significantly related to the ‘near-ground temperature: sex’ interaction, with
only females negatively related to near-ground temperature (α = -0.02; P = 0.05, Figure 3.10).

[Link].2 Effect of endogenous factors

‘Sex’ had a predominant independent effect on body size in all species with individuals from both
sexes available, except P. prativaga for which ‘sex’ was only significant in interaction with the UHI
(Table 3.6). The males of all species were smaller than the females, except Diplostyla concolor, in
which females were smaller (Appendix 3.5). ‘Generation’ was another determinant driver of body size
in all three species for which a comparison among generations was possible (i.e., O. fuscus, P.
degeeri, P. cf. tenuipes; Table 3.6), and the second-generation individuals were consistently the
smallest (Appendix 3.5).

124
 Table 3.6 - Effects of temperature variables (‘UHI’ and ‘near-ground temperature’), intraspecific concurrence, ‘sex’ (two levels: female/male) and ‘generation’ (two levels: G1/G2)
and their interactions, on body size estimated by GLMMs. Estimates are indicated, followed by significance levels between brackets (* 0.01 < P < 0.05; ** 0.001 < P < 0.01; *** P
< 0.001). Significant results (P < 0.05) are in bold. Results of averaged models after variable selection. ‘-’, non-selected variables; ‘na’, variables not included in the full models.

UHI Near-ground Intraspecific Sex Genera UHI: Near-ground Intraspecific Generation: Intraspecific Conditional Marginal
temperature competition tion/Co Sex temperature competition: Sex competition: R² R²
hort : Sex Sex Generation
Bathyphantes -0.041 -0.020
0.006 0.007 0.004 na -0.009 - na na 0.39 0.35
gracilis (*) (*)
0.031
Diplostyla concolor 0.007 0.005 - na - - - na na 0.24 0.15
(***)
Erigone dentipalpis -0.014 -0.007 -0.007 na na na na na na na 0.15 0.03
Hahnia nava 0.024 -0.01 0.004 na na na na na na na 0.19 0.08
Oedothorax fuscus -0.134
0.006 0.004 0.004 na - - 0.007 na na 0.84 0.84
(A) (***)
Oedothorax fuscus -0.018
- - - na na na na na - 0.07 0.07
(B) (***)
Oedothorax retusus - 0.004 -0.006 na na na na na na na 0.13 0.06
Pachygnatha -0.026
-0.009 -0.004 0.007 na na na na na - 0.2 0.11
degeeri (***)
-0.190 -0.405
Pardosa cf. tenuipes - 0.023 -0.035 - - 0.026 -0.036 -0.018 0.74 0.69
(***) (***)
0.125 -0.132
Pardosa prativaga 0.01 -0.025 0.116 na - - na na 0.3 0.27
(**) (**)
-0.030
Pardosa pullata 0.032 -0.017 0.012 na - 0.034 (*) - na na 0.21 0.09
(*)
-0.074
Tenuiphantes tenuis -0.009 -0.004 - na - - - na na 0.41 0.39
(***)

125
Chapitre 4
PARTIE III

Figure 3.10 - Relationships between P. prativaga and P. pullata body size and temperature variables.
Relationships were only significant in females (white dots, dashed regression lines). Black dots, males (solid
regression lines).

3.2.4 Discussion
Urbanisation is hypothesized to induce changes in body size within arthropod populations through
the UHI effect (Merckx et al. 2018a, 2018b). In a previous study conducted in the same area, the
community-averaged body size of spiders was found to be reduced in urban areas compared to
adjacent rural areas due to the exclusion of individuals from large-bodied species in response to
intense UHI conditions (Cabon et al. 2024a). While communities consisting of smaller species are
thought to result from increased metabolic rates and associated costs for a given body size (Brown
et al. 2004, Scheffers et al. 2016), intra-specific shifts in body size within urban populations match
the community-level pattern in flying arthropod species (Merckx et al. 2018a). The present study
shows that general conclusions on patterns of temperature-related shifts in body size cannot reliably
be drawn for spiders, as already observed in non-urban areas (Ameline et al. 2018, Hein et al. 2019).
Indeed, we found body size response to temperature gradients to be mediated by species’ trait
syndromes, with only large, low dispersive, univoltine species showing response patterns.
The thermal response of body size has been repeatedly identified as size-dependent (Blanckenhorn
and Demont 2004, Klok and Harrison 2013). That only the largest species tested here (P. prativaga
and P. pullata) are involved in temperature-size relationships contradicts the general pattern of
terrestrial arthropods according to which larger species are less sensitive to warming than smaller
ones (Forster et al. 2012, Klok and Harrison 2013). Among spiders, the general pattern might not be
applicable, since size is negatively related to the species’ capacity to use large-scale aerial dispersal.
Small spider species weighing less than 2.0 mg (i.e., eight out of the eleven species tested in the
present study) can be lifted easily by the wind due to biomechanical constraints, and this mass
threshold is not reached by adult linyphiids (Foelix 2011, Simonneau et al. 2016). Accordingly,
Simonneau et al. (2016) noticed that several Linyphiids and Tetragnathid species also tested in our
study were dominant ballooners (e.g., B. gracilis, E. dentipalpis, O. fuscus, O. retusus, Tenuiphantes
tenuis, P. degeeri). In addition, aerial dispersal almost systematically occurred during “hot” days (i.e.,
with a mean daily atmospheric temperature above 19 °C) and reduced wind speed conditions (below

126
 Chapitre 4

∼ 4 m.s−1) (Simonneau et al. 2016). These parameters also match the conditions observed during
intense UHI events in our study area (Dubreuil et al. 2020). Therefore, small Linyphiid, Tetragnathid
and Hahniid species should be able to escape unfavourable thermal habitats caused by UHI events
by using large-scale aerial dispersal (Samu et al. 1999). This capacity to escape adverse temperatures
might explain why UHI exerts a low selection pressure on the body size of these species, compared
to adult Lycosids that can only cover distances of up to 100 m (Richter 1970, Morse 1997, 2002). The
smallest species investigated here, which were also the bivoltine ones ( Table 3.5) were less
constrained by temperature than the univoltine species, as shown in previous studies on arthropods
(Horne et al. 2015).
Thanks to a low time pressure during their developmental cycle, univoltine species are expected to
exploit their developmental time to grow larger over the season and improve their fecundity (Horne
et al. 2015). Therefore, we expected positive size-temperature relationships for univoltine species.
We observed this pattern in P. prativaga, possibly resulting from enhanced development rates under
intense UHI conditions, yet only in females. Sex-biased responses have also been observed along
altitudinal clines (Bowden et al. 2013, Høye & Hammel 2010). These authors found that the males of
several wolf spider species remained unchanged in body size in response to elevation, whereas
females grew smaller. Our results bring additional support to the rule stating that the body size of
female terrestrial arthropods is more plastic than that of males in relation to constraining
environments (Stillwell et al. 2010). Intensified UHI conditions in urban areas may extend the growing
seasons compared to rural adjacent areas owing to elevated daily minimum temperatures in early
spring and late autumn (Foissard et al. 2019b, Dubreuil et al. 2020) and a reduced number of frost
days per year. Consequently, the sex-biased body size response to UHI variations that we report in
P. prativaga may result from sex-dependent growth strategies (Uhl et al. 2004, Høye and Hammel
2010). Urban females may increase their number of moults over the season (Esperk et al. 2007) or
increase their growth increments between moults (Blanckenhorn et al. 2007) compared to their rural
counterparts. In Lycosids, fecundity is known to be limited by female body size, and larger females
produce more and/or larger eggs (Simpson 1993). Therefore, intensified UHI conditions appear to
allow females to become more fecund by growing larger. In males, body size may not be constrained
by the environment because no fecundity-size relationship is involved and the growth strategy might
consist in becoming mature earlier instead of reaching a larger size (Uhl et al. 2004).
Contrary to our expectations, P. pullata females grew smaller with increasing near-ground
temperature. In this species, the constraints of large-scale atmospheric temperatures, as estimated
by the UHI, were overridden by the microclimatic cline. Variations of the body size response to
environmental factors across closely related Lycosid species has been reported (Bowden et al. 2013,
Ameline et al. 2018). However, to our knowledge, this is the first time that diverging body size
responses to temperature are highlighted to be scale-dependent among Lycosids. This might be due
to the contrasting habitat preferences of closely related species. Indeed, although P. prativaga is
mainly found in buffered microhabitats, P. pullata prefers poorly grown herbaceous vegetation with
low-growth plants, where temperature at the ground level is directly influenced by solar radiation
and wind velocity (Hollander & Lof, 1972). Given that Lycosid spiders are known to locally use
locomotion to select a specific temperature range across their thermal microhabitat (Humphreys
1974, Cabon et al. 2023), the body size of P. pullata may primarily be constrained by near-ground
temperatures: this species might be a more active and precise thermoregulator than P. prativaga
due to its complex thermal micro-environment.

127
PARTIE III

Conclusion
Our results support that interspecific differences in spider life-history traits (e.g. spider mobility or
voltinism) determine whether intraspecific body size variations are related to urban temperature
clines or not, as previously suggested along altitudinal and latitudinal gradients (Bowden et al. 2013,
Hein et al. 2019, Beckers et al. 2020). The body size of large and low-dispersive univoltine species
(Lycosids) was particularly sensitive to temperature changes. For Pardosa species, both positive (in
P. prativaga) and negative (in P. pullata) responses occurred, with body size variations associated to
temperature measured at the landscape (‘UHI’) and microhabitat (‘near-ground temperature’)
scales, respectively. In small and highly dispersive bivoltine species (Linyphiids, Hahniids,
Tetragnathids), body size variations were mainly associated with endogenous factors (‘sex’,
‘generation/cohort’), but were not constrained by the environment. These results based on a large
number of spider individuals belonging to 11 species with contrasting trait syndromes provide a new
understanding on why a general pattern of the relationship between body size and urbanisation-
induced warming cannot be drawn. We conclude that predictions on species responses to thermal
stress generated by cities should consistently be made with regard to species life history
characteristics.

128
 3.2.5 Appendices

Appendix 3.2 - Distribution of the number of females (upper line) and males (lower line) collected of the eleven selected species according to the sampling month
(from March to September). The dashed lines represent the trend lines of the number of individuals, obtained using the ‘geom_smooth’ function from the
‘ggplot2’ package in R. Colored bars indicate individuals selected for measurement (light red = no seasonal distinction, orange = first generation/cohort, dark
red = second generation/cohort).

129
Chapitre 4
PARTIE III

Appendix 3.3 - Results of correlation tests between prosoma width and three other measured morphological
attributes (prosoma length, femur length and tibia length). For each of the eleven selected species, ‘t’ gives the
t statistics, ‘df’ the degrees of freedom, ‘p-val’ the p-values and ‘cor’ the correlation estimates. (1/2)
Prosoma length Femur length Tibia length
t 12.766 13.172 6.512
df 217 212 175
Bathyphantes gracilis
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.655 0.671 0.442
t 11.049 10.615 7.324
df 174 174 113
Diplostyla concolor
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.642 0.627 0.567
t 13.445 12.222 9.733
df 102 102 92
Erigone dentipalpis
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.800 0.771 0.712
T 20.591 15.640 11.079
df 127 127 127
Hahnia nava
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.877 0.811 0.701
t 27.653 22.045 13.300
df 241 241 235
Oedothorax fuscus
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.872 0.818 0.655
t 3.474 1.879 3.246
df 95 94 89
Oedothorax retusus
p-val 0.001 0.063 0.002
cor 0.336 0.190 0.325
t 6.824 8.979 7.849
df 238 238 238
Pachygnatha degeeri
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.405 0.503 0.453
t 65.044 32.956 32.602
df 389 388 387
Pardosa cf. tenuipes
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.957 0.858 0.856

130
 Chapitre 4

Appendix 3.3 - Results of correlation tests between prosoma width and three other measured morphological
attributes (prosoma length, femur length and tibia length). For each of the eleven selected species, ‘t’ gives
the t statistics, ‘df’ the degrees of freedom, ‘p-val’ the p-values and ‘cor’ the correlation estimates (2/2).
Prosoma length Femur length Tibia length
t 12.976 11.697 10.362
df 169 168 165
Pardosa prativaga
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.706 0.670 0.628
t 10.612 9.468 9.109
df 154 154 154
Pardosa pullata
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.650 0.607 0.592
t 20.467 14.783 6.577
df 246 246 183
Tenuiphantes tenuis
p-val < 0.0001 < 0.0001 < 0.0001
cor 0.794 0.686 0.437

131
PARTIE III

Appendix 3.4 - Summary of the individual parameters of the 11 selected species after excluding outliers,
arranged by sex and generation/cohort. The mean prosoma width and standard deviation per species are
indicated in millimeters. The number of individuals and the corresponding percentage within each species are
indicated in the “Count” lines (1/2).
G1/C1 G2/C2 Total
Bathyphantes gracilis Males Count - - 116 (53%)
Prosoma width - - 0.64 ± 0.03
Females Count - - 103 (47%)
Prosoma width - - 0.69 ± 0.04
Total Count - - 219 (100%)
Prosoma width - - 0.66 ± 0.05
Diplostyla concolor Males Count - 99 (56%) 99 (56%)
Prosoma width - 0.83 ± 0.04 0.83 ± 0.04
Females Count - 77 (44%) 77 (44%)
Prosoma width - 0.80 ± 0.04 0.80 ± 0.04
Total Count - 176 (100%) 176 (100%)
Prosoma width - 0.82 ± 0.04 0.82 ± 0.04
Erigone dentipalpis Males Count - 104 (100%) 104 (100%)
Prosoma width - 0.88 ± 0.05 0.88 ± 0.05
Hahnia nava Males Count - - 129 (100%)
Prosoma width - - 0.66 ± 0.05
Oedothorax fuscus Males Count - 83 (34.5%) 83 (34.2%)
Prosoma width - 0.72 ± 0.03 0.72 ± 0.03
Females Count 71 (29.2%) 89 (36.6%) 160 (65.8)
Prosoma width 0.87 ± 0.04 0.85 ± 0.03 0.86 ± 0.04
Total Count 71 (29.2%) 172 (70.8%) 243 (100%)
Prosoma width 0.87 ± 0.04 0.79 ± 0.07 0.81 ± 0.08
Oedothorax retusus Males Count - 97 (100%) 97 (100%)
Prosoma width - 0.74 ± 0.03 0.74 ± 0.03
Pachygnatha degeeri Males Count 114 (47.5%) 126 (52.5%) 240 (100%)
Prosoma width 1.15 ± 0.05 1.13 ± 0.04 1.14 ± 0.05
Pardosa cf. tenuipes Males Count 97 (24.8%) 103 (26.3%) 200 (51.2%)
Prosoma width 2.11 ± 0.12 1.67 ± 0.11 1.89 ± 0.25
Females Count 89 (22.8%) 102 (26.1%) 191 (48.8%)
Prosoma width 2.29 ± 0.15 1.89 ± 0.21 2.08 ± 0.27
Total Count 186 (47.6%) 205 (52.4%) 391 (100%)
Prosoma width 2.20 ± 0.16 1.78 ± 0.20 1.98 ± 0.28

132
 Chapitre 4

Appendix 3.4 - Summary of the individual parameters of the 11 selected species after excluding outliers,
arranged by sex and generation/cohort. The mean prosoma width and standard deviation per species are
indicated in millimeters. The number of individuals and the corresponding percentage within each species are
indicated in the “Count” lines (2/2).
G1/C1 G2/C2 Total
Pardosa prativaga Males Count - - 95 (55.5%)
Prosoma width - - 2.15 ± 0.08
Females Count - - 76 (44.5%)
Prosoma width - - 2.23 ± 0.16
Total Count - - 171 (100%)
Prosoma width - - 2.19 ± 0.13
Pardosa pullata Males Count - - 88 (56.5%)
Prosoma width - - 2.01 ± 0.09
Females Count - - 68 (43.5%)
Prosoma width - - 2.04 ± 0.09
Total Count - - 156 (100%)
Prosoma width - - 2.03 ± 0.09
Tenuiphantes tenuis Males Count 124 (50%) - 124 (50%)
Prosoma width 0.78 ± 0.04 - 0.78 ± 0.04
Females Count 124 (50%) - 124 (50%)
Prosoma width 0.85 ± 0.05 - 0.85 ± 0.05
Total Count 248 (100%) - 248 (100%)
Prosoma width 0.81 ± 0.06 - 0.81 ± 0.06

133
 PARTIE III

134
Appendix 3.5 - Differences (ANOVA, *: p-value <0.05) in body size (in mm) between male and female individuals (white dots represent females and black dots
are males) as well as differences (ANOVA, *: p-value <0.05) between body size of individuals from generations (G) or cohorts (C).
 Chapitre 4

135
PARTIE III

136
 Chapitre 5

4 Partie IV : Importance des conditions

microclimatiques à l’échelle des arthropodes

▪ Chapitre 5 : Rôle de la thermorégulation comportementale en contexte

d’îlots de chaleur urbains

▪ Chapitre 6 : Potentiel de la végétation herbacée pour atténuer le

réchauffement urbain à l’échelle des micro-habitats

En bref :
▪ Dans la partie précédente, nous avons montré que certaines espèces, bien que
particulièrement sensibles au réchauffement atmosphérique urbain (en particulier les
Lycosidae), peuvent maintenir des populations en ville, notamment jusqu’à l’extrémité
chaude du gradient d’ICU. Dans cette partie nous explorons le rôle que peuvent jouer les
micro-habitats dans le maintien de ces populations en situation d’ICU intense.

▪ Les températures préférées mesurées en laboratoire ne sont pas reliées à l’intensité d’ICU
d’origine chez les individus adultes de Pardosa saltans (Lycosidae) issus de populations
prélevées sur l’ensemble du gradient d’ICU. Les individus thermorégulent par ailleurs de
manière précise en conditions contrôlées.

▪ Le découplage des conditions thermiques au sein des micro-habitats par rapport aux
conditions atmosphérique pourrait dès lors atténuer l’exposition des individus au
réchauffement et expliquer l’absence de réponse du trait thermique mesuré.

▪ Les micro-habitats prairiaux peuvent atténuer, voire compenser le réchauffement

atmosphérique urbain, favorisant ainsi le maintien des espèces les plus sensibles au
réchauffement en ville. La capacité d’atténuation du micro-habitat est cependant modulée
par la structure de l’environnement proche, et de la végétation herbacée. Ces paramètres
peuvent être optimisés par une gestion appropriée qui peut in fine favoriser le maintien de la
diversité des arthropodes en ville.

▪ 137
PARTIE IV

4.1 Chapitre 5 : Rôle de la thermorégulation comportementale en

contexte d’îlots de chaleur urbains

Résumé

La plupart des ectothermes s'appuient sur la thermorégulation comportementale pour maintenir

leur température corporelle proche de leur optimum physiologique. Les ectothermes peuvent donc
limiter considérablement leur exposition aux extrêmes thermiques en choisissant une gamme de
températures étroite, incluant leurs températures préférées (Tpref). Malgré les preuves que la
thermorégulation comportementale peut être ajustée par la plasticité phénotypique ou limitée par
la sélection naturelle, les variations intraspécifiques de la Tpref à travers les gradients
environnementaux restent peu étudiées par rapport à d'autres traits thermiques comme la tolérance
thermique. Dans ce chapitre, nous analysons les variations de Tpref des populations d'araignées
échantillonnées le long du gradient d'ICU rennais. Nous avons mesuré deux composantes de la
préférence thermique en laboratoire, à savoir la Tpref moyenne et l’amplitude de Tpref (c'est-à-dire
l'écart-type) chez 557 individus d'une Lycosidae commune (Pardosa saltans), collectés sur le terrain.
Nous avons testé si les valeurs de Tpref diffèrent entre dix populations issues de zones thermiques
contrastées. Nous montrons que des facteurs endogènes tels que la taille corporelle ou le sexe
déterminent principalement la Tpref moyenne et l'amplitude de la Tpref. L'amplitude de la Tpref est
également liée à l'intensité de l'ICU, mais dans une moindre mesure et seulement chez les juvéniles.
L'absence de relation entre les mesures de Tpref et l'ICU chez les araignées adultes suggère un effet
Bogert, selon lequel la capacité des individus à détecter et à exploiter des microclimats optimaux
affaiblit la pression de sélection des températures (ici induite par l'ICU) sur leur physiologie
thermique. Par ailleurs, cette absence de relation pourrait indiquer que les conditions de
température à l'échelle du micro-habitat diffèrent des conditions mesurées à l’échelle de l'ICU. Cette
étude montre l'importance de prendre en compte les variations inter-individuelles et inter-
populations de l'amplitude de la Tpref, et confirme que l'amplitude de la Tpref est une mesure
pertinente pour informer sur la précision (et donc l’efficacité) de la thermorégulation
comportementale dans une population donnée.

Article publié dans Journal of Thermal Biology :

Cabon, V., S. Pincebourde, H. Colinet, V. Dubreuil, R. Georges, M. Launoy, J. Pétillon, H. Quénol, and
B. Bergerot. 2023. Preferred temperature in the warmth of cities: Body size, sex and development
stage matter more than urban climate in a ground-dwelling spider. Journal of Thermal Biology
117:103706.

138
 Chapitre 5

Preferred temperature in the warmth of cities: body size, sex and

development stage matter more than urban climate in a ground-dwelling
spider

Valentin Cabon1,2, Sylvain Pincebourde3, Hervé Colinet1, Vincent Dubreuil4, Romain Georges1,2, Maud
Launoy1, Julien Pétillon1,5, Hervé Quénol4, Benjamin Bergerot1,2

1. University of Rennes, CNRS, ECOBIO [(Ecosystèmes, biodiversité, évolution)] - UMR 6553, Rennes,
France
2. LTSER ZA Armorique, F-35000 Rennes, France
3. Institut de Recherche sur la Biologie de l'Insecte, UMR 7261, CNRS – Université de Tours, Tours,
France
4. University of Rennes 2, CNRS, LETG - UMR 6554, Rennes, France
5. Institute for Coastal and Marine Research – Nelson Mandela University – Port Elizabeth – South
Africa

Abstract:
Most ectotherms rely on behavioural thermoregulation to maintain body temperatures close to their
physiological optimum. Hence, ectotherms can drastically limit their exposure to thermal extremes
by selecting a narrower range of temperatures, which includes their preferred temperature (Tpref).
Despite evidence that behavioural thermoregulation can be adjusted by phenotypic plasticity or
constrained by natural selection, intraspecific Tpref variations across environmental gradients
remain overlooked as compared to other thermal traits like thermal tolerance. Here, we analysed
Tpref variation of spider populations found along a gradient of urban heat island (UHI) which displays
large thermal variations over small distances. We measured two components of the thermal
preference, namely the mean Tpref and the Tpref range (i.e., standard deviation) in 557 field-
collected individuals of a common ground-dwelling spider (Pardosa saltans, Lycosidae) using a
laboratory thermal gradient. We determined if Tpref values differed among ten populations from
contrasting thermal zones. We showed that endogenous factors such as body size or sex primarily
determine both mean Tpref and Tpref range. The Tpref range was also linked to the UHI intensity to
a lesser extent, yet only in juveniles. The absence of relationship between Tpref metrics and UHI in
adult spiders suggests a Bogert effect according to which the ability of individuals to detect and
exploit optimal microclimates weakens the selection pressure of temperatures (here driven by UHI)
on their thermal physiology. Alternatively, this lack of relationship could indicate that temperature
patterns occurring at the scale of the spiders' micro-habitat differ from the measured UHI. This study
shows the importance of considering both inter-individual and inter-population variations of the
Tpref range when conducting Tpref experiments, and supports Tpref range as being a relevant
measure to inform on the strength of behavioural thermoregulation in a given population.

Key words: Behavioural thermoregulation, Bogert effect, Climate change, Ectotherm, Pardosa
saltans, urbanization, Selected temperature

139
PARTIE IV

4.1.1 Introduction
Ectothermic animals depend on external heat to regulate their body temperature, which directly
drives their physiological performance, from metabolism, growth, reproduction or movement, to
interaction strength and distribution (Angilletta 2009, Huey et al. 2012). Ectotherms are thus widely
affected by climate change (Deutsch et al. 2008, Sinclair et al. 2016). To maintain their body
temperature close to their physiological optimum and maximize their performance and fitness,
ectotherms’ ability to behaviourally thermoregulate is of critical importance (Martin and Huey 2008).
This mechanism consists in moving within a specific range of temperatures that bring the organism
near its optimum for a given performance metric across heterogeneous thermal environments. Thus
ectotherms evolved particular adaptations to select temperatures within this range, hence the term
of preferred temperature (Tpref), which limits the exposure to deleterious thermal extremes and
maximize energetic and physiological benefits (Martin and Huey 2008, Angilletta 2009).
Among the variety of thermal traits, Tpref and related metrics are considered as the most relevant
trait to assess behavioural thermoregulation (Porter et al. 1973, Stevenson 1985, Hertz et al. 1993,
Kearney et al. 2009, Gvoždík 2012). Tpref reflects the selected temperature within a controlled
thermal environment, in absence of confounding factors interacting with the thermoregulatory
behaviour (e.g., foraging opportunities or predator avoidance) (Huey and Slatkin 1976). Traditionally,
the mean of the Tpref was mostly used (Labra et al. 2009) but the range of Tpref is also an informative
component to characterize thermoregulation capacities. The inter-individual variability of Tpref
often displays contrasting levels between species. For example, all the tested individuals selected
temperatures within a narrow range of only 2-3 °C in the common brown shrimp (Reiser et al. 2014)
and in the León rock lizard (Ortega et al. 2016). By contrast, the Tpref distribution encompassed a
range of >10°C in several drosophilids (MacLean et al. 2019). Several studies showed that
intraspecific Tpref variations are sex- and size-dependant (Forsman 2000, Dillon et al. 2009, Berec et
al. 2014), as well as being driven by developmental or reproductive stages (Dillon et al. 2009,
MacLean et al. 2019). Although the drivers of such differences, both at the inter and intra specific
levels, are not well understood, the range of Tpref can be used as an indicator of a population’s
ability to buffer against temperature extremes like heat waves events (Brandt et al. 2020). In general,
thermal ecology theory predicts that a narrow Tpref range improves the ability to buffer
environmental extremes as long as the targeted temperature remains in some microclimates
(Woods et al. 2015). However, a narrow Tpref range can reduce selection pressure on thermal limits,
thereby making the species particularly vulnerable when the adequate temperature is not present
anymore in the environment (Muñoz et al. 2016). Thus, the Tpref range and mean are often expected
to vary along environmental gradients such as altitude but available studies suggest that this relation
is more complex and species specific than anticipated (Gvoždík 2012, Trochet et al. 2018).
Most of the complex and highly variable responses of Tpref to temperature variations along
geographical clines were derived from studies on vertebrate ectotherms (Llewelyn et al. 2016,
Trochet et al. 2018, Muñoz 2022). Three types of ectotherms’ responses can be drawn from these
studies that mostly focused on the median or mean of Tpref ( Figure 4.1). Firstly, the adaptation of
local optima hypothesis, according to which individuals from colder environments select lower Tpref
relative to conspecifics from warmer environments, following a co-gradient response (Brattstrom
1965, Levinton 1983, Mitchell and Lampert 2000). Secondly, the counter-gradient relationship
hypothesis predicts that individuals experiencing extreme temperatures should be more sensitive to
adverse thermal conditions (Llewelyn et al. 2016), and therefore they select lower temperatures (i.e.,

140
 Chapitre 5

lower Tpref) when they are given the opportunity. Hence, individuals from cold environments select
higher Tpref to compensate for lower opportunities to find favourable temperatures within the
thermal environment (Conover and Schultz 1995, Fangue et al. 2009, McElroy 2014, Llewelyn et al.
2016, Hodgson and Schwanz 2019). Thirdly, the Bogert effect (Bogert 1949) suggests that
behavioural thermoregulation inhibits the plastic response of Tpref (Huey et al. 2003, Marais and
Chown 2008). This behavioural inertia is suspected to buffer selection pressure and thus to reduce
plastic or evolutive adaptation to changing thermal environments. Consequently, this process
involves the absence of correlation between geographical clines and Tpref. Alternatively, an absence
of correlation could also indicate that patterns of temperature measured along the geographical
cline do not occur at the scale of the microhabitat occupied by the investigated taxa. Different
patterns can apply in closely related species; for instance, the variation of Tpref in a community of
Drosophila species along geographical clines illustrated a co-gradient pattern (Nevo et al. 1998), a
counter-gradient pattern (Yamamoto and Ohba 1984) as well as an absence of Tpref response to
elevation (Krstevska and Hoffmann 1994) depending on the species. Here, we applied this framework
to test if one of these mechanisms prevail in the variation of Tpref (both mean and range) of a spider
species along a broad thermal cline generated by the urban heat island.
Urbanisation gradients, alongside with their strong deleterious impacts on local biodiversity
(McKinney 2008, Groffman et al. 2014, Fenoglio et al. 2020), have been recently recognised as
promising open-sky laboratories to surrogate climate change impacts on arthropods (Youngsteadt
et al. 2015, Lahr et al. 2018). During the night, the mean air temperature in cities is higher than in
the countryside due to the combination of heat-absorbing impervious materials, reduction of
vegetation cover and associated water vapor flux, canyon-like urban morphology, and accumulations
of anthropogenic heat sources (such as traffic or heating) (Heisler and Brazel 2010). This
phenomenon, commonly referred to as the atmospheric urban heat island (UHI) (Oke et al. 2017),
provides multiple advantages for conducting thermal ecology studies, particularly on ectothermic
species such as arthropods (Verheyen et al. 2019). A first interesting component of the UHI lies in
the steep slope of the thermal gradient (Oke et al. 2017). Differences in average temperatures can
reach 4°C between locations separated by less than 10 km, providing a broad thermal gradient within
a narrow geographic area (Verheyen et al. 2019). The second component of interest lies in the
magnitude of night/day time temperature fluctuations which is modulated by the UHI’s intensity.
Much of the increased average temperatures in urban areas is the result of warmer nocturnal
temperatures relative to adjacent rural areas (Oke et al. 2017, Foissard et al. 2019a, Dubreuil et al.
2020). Therefore, the magnitude of temperature variations between day- and night-time is lower in
urban areas than in rural surroundings (Oke et al. 2017). An urban dwelling animal living under
intense UHI conditions thus experience a narrower range of night/day time temperature fluctuations
than its rural conspecific. If multiple studies on arthropods demonstrated the existence of plastic or
evolutive responses of thermal tolerance to the UHI (Angilletta et al. 2007, Brans et al. 2017,
Diamond et al. 2018d, 2018b, Sato and Takahashi 2022), almost no studies aimed at linking
arthropods’ thermal preferences across a UHI gradient to infer the impact of temperature variations
on their thermoregulatory behaviour (but see Youngsteadt et al., 2023). Although fine-scale surface
temperatures are likely to be an important driver of the Tpref response to the thermal environment,
repeating investigations into the potential effects of large-scale warming through atmospheric UHI
on rural-urban thermal gradients represents a first step in providing crucial information on how
climate change relates to the physiology and behaviour of ectotherms. If large-scale temperature
patterns across geographical clines (i.e., atmospheric UHI) cannot be linked to Tpref variations,

141
PARTIE IV

surface temperature may better match the scale at which arthropods experience their thermal
environment and therefore be the determining metric. In addition, we lack information on the
variation of the range of Tpref along these environmental gradients. By applying the same reasoning
than for the mean, however, we expect a positive relationship between the Tpref range and the
magnitude of the night/day time temperature fluctuations in the local optima adaptation hypothesis,
and therefore a negative relationship between the mean and range of Tpref (Figure 4.1). By contrast,
we expect the range of Tpref to decrease with the magnitude of the night/day time temperature
fluctuations under the counter-gradient hypothesis, while no change is anticipated in the Bogert
effect hypothesis (Figure 4.1).
In this study, we determined the variation of thermal preference in a free-moving wolf-spider along
an UHI gradient to infer which mechanism (Figure 4.1) is at play. Lycosids are relevant models for such
a study because (i) they are dominant ground-dwelling predators in many habitats, including along
urbanization gradients (Varet et al. 2011b), and (ii) previous studies showed the relationship
between local abiotic factors and micro-habitat partitioning (e.g., for temperature and humidity, see
Frick et al., 2007 and DeVito et al., 2004, respectively), suggesting they adjust their local distribution
according to their Tpref, including in the field (Humphreys 1974). We collected individuals along an
UHI gradient to measure their Tpref in the lab under standardized conditions. We conducted
individual-based and repeated measurements of Tpref over time, which permitted to calculate both
mean Tpref and Tpref range for every individual. We then tested if mean Tpref and Tpref range co-
vary with the UHI intensity. We further tested whether the response to UHI dominates the influence
of three endogenous factors on mean Tpref and Tpref range, namely body size, sex and life stage
(juveniles versus adults). This comparison can be useful to infer the relative biological importance of
any effect of the UHI on Tpref.

142
 Chapitre 5

Figure 4.1 - Conceptual hypothesis framework displaying predicted responses on thermal preferences (mean
Tpref: top line and Tpref range: bottom line) according to the urban heat island intensity (UHI). The UHI intensity
axis is characterized by an increase in mean temperature and a concomitant decrease of magnitude in
night/day time temperature fluctuations. The three columns describe the local optima adaptation, the counter-
gradient hypothesis and the behavioural inertia hypothesis (Bogert effect) respectively.

4.1.2 Material and methods

[Link] Model species and sampling sites
The wolf spider Pardosa saltans (Töpfer-Hofmann et al. 2000) (Araneae: Lycosidae) is widely
distributed across Europe. This spider completes its life cycle over two years. Spiderlings emerge
during summer in year N, to spend winter exclusively in the leaf litter. Individuals reach the subadult
stage in September of N+1 and overwinter a second time as subadults. Finally, subadults reach the
adult stage between early March and the end of April of the year N+2. Since mating occurs as soon
as the adult stage is reached, most females are carrying egg-sacs by the end of May and spiderlings
start to emerge in July (Edgar 1971). If adults are active on the leaf-litter during daytime in woody
habitats (Edgar 1971, Hendrickx et al. 2001, Pétillon et al. 2011), males commonly explore a wider
range of habitats than females and can also be found in clearings, hedgerows, or thickets (Ruhland
et al. 2016, Trabalon 2022). In urbanised areas, P. saltans can be found in woody patches such as
parks, wastelands, and natural remnants. These spiders use the litter layer to behaviourally
thermoregulate and protect themselves from high temperature variations (Trabalon 2022). Due to
its abundance in urban habitats and its thermoregulatory capacity, P. saltans represents an excellent
model species to study if temperature changes induced by the UHI modulate the ability of
arthropods to behaviourally thermoregulate.

143
PARTIE IV

We collected adults and juveniles (subadults) from both sexes by visual inspection of the litter of ten
deciduous woodlots located across a gradient of UHI in the area of Rennes, France (coordinates:
48°06'41.6"N 1°40'46.5"W), using 200 ml plastic tubes. Spiders were all collected in similar woody
habitats regardless of the corresponding UHI intensity. For females, only individuals without egg-sac
where collected. To account for spiders’ acclimation to field temperature variation over the season,
we conducted two sampling sessions, respectively in March and April 2022, between 10 am and 3
pm, during sunny conditions. To ensure that individuals remained acclimated to field temperature
conditions, spiders were brought to the laboratory and experiments began within two hours
following field collection (Reynolds and Casterlin 1979).

[Link] Urban heat island (UHI)

In Rennes, a long-term climate observation network was established in 2004 with the aim to
characterize the UHI intensity across space and time (Foissard et al. 2019a, Dubreuil et al. 2020).
Since 2004, the Rennes Urban Network ([Link] has grown from 12 to 30 DAVIS
Vantage-Pro-2 automatic weather stations, to which 93 connected temperature sensors were added
in 2020 (Dubreuil et al. 2022). The weather stations and temperature sensors are spread over rural,
suburban and city-centre areas, collecting data at a 60- and 15-minutes frequency, respectively. We
used this database to calculate the UHI intensity for each month between June 2021 and June 2022,
as the mean difference between minimal daily temperatures measured at the cold reference station
and the minimal daily temperatures measured by each other sensor. Therefore, this UHI intensity
corresponds to a temperature anomaly of a particular site, i.e., the number of degrees C above the
cold reference that is necessarily outside the urban landscape. These temperature difference values
were used to spatialize the UHI intensity by inverse distance weighting interpolation to the entire
city area at a 100m resolution. The UHI intensity is known to vary over the year, so we averaged the
monthly raster maps to calculate UHI value corresponding to three periods that are ecologically
relevant for our study species (Figure 4.2): (1) the exact period of the Tpref measurements (i.e., March
and April 2022), (2) the period corresponding to the entire time spent by the collected P. saltans
individuals since they reached the subadult stage (i.e., September 2021 to April 2022), and (3) the
winter period before the experiment (i.e., December 2021 to February 2022), during which
individuals overwintered as subadults. At last, we extracted the UHI intensity values corresponding
to the raster cells on which the ten sampling sites were located for those 3 periods defined above.

144
 Chapitre 5

Figure 4.2 - Location of the 10 sampling sites (black dots) within the city borders of Rennes and surroundings
according to the spatialized UHI mean intensity obtained from the sensor network (red dots) during A) the
experiment period (March and April 2022), B) the entire period P. saltans individuals have spent as subadults
or adults before and during the experimentation (September 2021 to April 2022), C) the winter period preceding
the experiment (December 2021 to February 2022). The UHI is depicted by a colour gradient ranging from blue
(low intensity) to red (high intensity). Transparent grey patches represent build-up areas.

145
PARTIE IV

[Link] Thermal preference

We determined the Tpref of each individual collected at the ten sites using two home-made thermal
gradient systems in the laboratory. Each unit was composed of an aluminium plate (dimensions L x l
x h: 50 x 18 x 1 cm), heated at one side using a thermal resistance and cooled at the other side with
a cryostat bath (Polystat CC3, Huber Kältemaschinenbau AG, Germany) filled with heat transfer fluid
(Thermofluid SilOil, Huber, Germany), creating a linear temperature gradient orientated along the
plates’ length (Appendix 4.1). A system of 14 ramps was created on top of the plate (along the length)
by applying 15 aluminium bars of 1 cm height, 35 cm length and separated by 1 cm from each other.
Each spider was placed individually in one of the ramps. The grid of ramps was covered with a
transparent acrylic plate to avoid the animals to escape. The two units therefore allowed us to
determine the Tpref of 28 individuals at a time.
We conducted preliminary tests to set the temperature range of the gradient at 14-39°C, in
accordance to the range of preferred temperatures of spiders, thereby avoiding an edge effect with
individuals stuck at one end of the gradient. We then measured the temperature distribution over
the two gradients using thermocouples (Type-K) connected to Testo 175T3 temperature data loggers
(Testo SE& Co. Germany) that were placed every centimetre along the plates (the edge of gradients
was graduated at a 5 mm resolution), and checked for temperature stability over time. The
temperatures measured as a function of the distance to the warmer end allowed us to generate
calibration curves that were used for post-conversions of distance (location of a spider in a ramp) to
temperature. Conjointly, air humidity was measured along the gradient using five iButton
Hygrochron Temperature/Humidity Logger (DS1923, Dallas Semiconductor/Maxim, Dallas, USA).
First trials revealed humidity proportions ranging from 20% at the warm side, to 100% at the cold
extremity. To avoid a temperature-humidity co-gradient effect and enable the decoupling of
preferred temperatures from humidity preference, we placed a fragment of wet sponge at each
ramp’s warm extremity. This method maintained the relative humidity above 70% along the whole
gradients.
At the time of an experimental run, the set-up was turned on one hour prior to experiment’s start
to ensure the thermal gradient is stable. Then, the collected spiders were individually sexed,
weighted (to the nearest 0.01 mg using a Sartorius A200S Analytical Balance) and randomly placed
individually in the ramps in the centre of the gradient (where surface temperature was about 27°C).
Stabilization times exceeding one hour have already been observed in arthropods (Refinetti 1984).
Therefore, we monitored spiders’ position every minute during 90 minutes with cameras (AKASO
Brave 7) suspended above the set-up. Light was maintained in the lab at the minimum level so that
the position of spiders in the images could be clearly observed. At the end of the experiment, each
spider was numbered and kept in a separate Eppendorf tube, in 70° alcohol, for morphological
measurements. All spiders were individually checked under binocular magnifier to confirm the prior
identification of species, sex and development stage (juvenile or adult). In addition, we took pictures
of the cephalothorax and the right leg IV (Pétillon et al. 2018) using the software Leica LAS X, and
used ImageJ to measure the length of cephalothorax and leg as proxies of body size.

146
 Chapitre 5

[Link] Data analysis

We extracted the position in the ramps of each individual obtained each minute during the 90 min
experimental period, and converted them in temperature values using the gradient calibration
curves (Appendix 4.2). We tested if the Tpref varied over the 90 minutes by dividing the temperature
measurements into 10 min intervals. We then calculated a mean Tpref per spider and per 10 min
interval. We performed generalized mixed models (GLMM) to test the effect of time (fixed effect),
while accounting for spider individuals as random effect. We computed multiple models considering
at first the whole experiment time (i.e., 90 min), and successively removed 10 min time intervals to
allow for the determination of a breakpoint at which spiders reached their (stable) final Tpref. To
estimate the spiders’ stabilized Tpref considered in further analyses, we determined the mean and
range values based on all temperature measurements available after this breakpoint.
To determine mean Tpref and Tpref range, we fitted a Gaussian distribution curve to each
individual’s temperature distribution by maximum likelihood (mle). Based on the obtained estimates,
we attributed a mean and a range value calculated as the standard deviation value of the mean of
the fitted Gaussian distribution. A low Tpref range indicates that a spider explored a narrow range of
the thermal gradient, whereas a high value corresponds to an individual exploiting a large portion of
the ramp. To reduce the risk of including individuals for which the movement was inhibited by low
temperatures at the cold end of the gradient (see Dillon et al. 2009), we removed outliers with
median Tpref values outside the 95% dispersion interval, after pooling all individuals.
Statistically, we tested separately the response of mean Tpref and Tpref range to endogenous
variables and UHI intensity. We computed generalized linear mixed models (GLMM) and included
developmental stage (juvenile versus adult), sex, two morphological measurements (length of
prosoma and leg) and a measure of UHI intensity (i.e., March to April 2022) as predictors. As habitat
and behaviour are likely to differ strongly in juveniles and adults of P. saltans, we added an
interaction term between developmental stage and each other predictor. We selected the best
models using the dredging procedure from the ‘MuMin’ package (Bartoń 2016). In cases where
interactions were retained in the final models, we computed post hoc tests to identify the underlying
significant relationships. We performed Tukey tests using the ‘emmeans’ R-package (Lenth 2023) in
the case of categorical variables and linear models for continuous data. Additional similar models
were independently tested to verify that the predictive power of the two other UHI metrics (i.e.,
September 2021 to April 2022 and December 2021 to February 2022) remained unchanged. In each
model, we defined all the endogenous and environmental variables as fixed effects and sampling
session (March or April) as a random effect. We also calculated the coefficient of determination
(marginal r-squared) to assess the proportion of variation explained by the fixed effects.
All statistical analyses were performed in R version 4.0.3, using the package ‘fitdistrplus’ for gaussian
distribution fitting (Delignette-Muller and Dutang 2015), ‘lme4’ for GLMMs computing (Bates et al.
2015) and ‘performance’ (Lüdecke et al. 2021) for computing marginal r-squared values from mixed
effects models.

147
PARTIE IV

4.1.3 Results
[Link] Sampled individuals
The number of collected individuals at each sampling site ranged from 27 to 71 for a total number
of 557 individuals that were assayed. We obtained the Tpref for every minute during 90 minutes on
522 individuals of P. saltans, including 259 males (102 juveniles and 157 adults) and 263 females
(117 juveniles and 146 adults) after removing only 35 outliers in the complete dataset. Globally,
spiders had a mean Tpref of 33.4 ± 2.0°C and a Tpref range of 1.8 ± 1.7°C (Table 4.1). The average UHI
intensity of the study sites ranged from 0.85°C to 2.5°C when calculated over the exact sampling
period (March 2022 – April 2022), from 0.83°C to 2.1°C when considering the preceding winter
period (Dec. 2021 – March 2022) and from 0.93°C to 2.8°C when calculated over the entire previous
year (July 2021 – June 2022).

Table 4.1 - Summary of individual parameters for two development stages (adults and juveniles) in males and
females of Pardosa saltans. Mean and standard deviation values are indicated for each variable, except in the
case “count”, where the number of individuals and the corresponding percentage are indicated. Values related
to Tpref (mean and range) in this table were calculated on the experimental time ranging from the 31 st to 90th
minute.
Males Females Total
Adults Count 157 (30%) 146 (28%) 303 (58%)
Prosoma length
2.3 ± 0.3 3.6 ± 0.7 2.9 ± 0.8
(mm)
Leg length (mm) 10.2 ± 0.9 12.4 ± 1.2 11.3 ± 1.5
Mean Tpref (°C) 32.8 ± 2.0 34.3 ± 2.3 33.5 ± 2.3
Tpref range (°C) 2.6 ± 1.6 2.3 ± 1.5 2.1 ± 1.6
Juveniles Count 102 (19.5%) 117 (22.5%) 219 (42%)
Prosoma length
2.7 ± 0.3 3.0 ± 0.4 2.9 ±0.4
(mm)
Leg length (mm) 8.6 ± 0.9 9.4 ± 1.3 9.0 ± 1.2
Mean Tpref (°C) 33.4 ± 1.7 33.0 ± 1.5 33.2 ± 1.6
Tpref range (°C) 1.2 ± 1.5 1.4 ± 1.8 1.3 ± 1.6
Total Count 259 (49.6%) 263 (50.4%) 522 (100%)
Prosoma length
2.5 ± 0.3 3.4 ± 0.6 2.9 ± 0.7
(mm)
Leg length (mm) 9.6 ± 1.2 11.0 ± 2.0 10.3 ± 1.8
Mean Tpref (°C) 33.0 ± 1.9 33.7 ± 2.1 33.4 ± 2.0
Tpref range (°C) 2.0 ± 1.7 1.5 ± 1.6 1.8 ± 1.7

148
 Chapitre 5

[Link] Period selection for Tpref parameters

Spiders’ Tpref was influenced by experimental time (GLMM, p < 0.001, Figure 4.3A, Appendix 4.4).
Mean Tpref values were lower during the first three time-intervals considered (0-10 min: 32.9 ±
3.0°C, 11-20 min: 33.2C ± 2.7°C, 21-30 min: 33.3 ± 2.5°C), but stabilized after 30 minutes and
remained unchanged until the end of the experiment (31-90 min: 33.4 ± 2.0°C, Figure 4.3B).
Consequently, we excluded Tpref data collected during the exploration phase, and considered only
stabilized Tpref values measured after the 30th minute of the experiment.

Figure 4.3 - A) Variation of the mean Tpref of 522 individuals of Pardosa saltans over 90 minutes in an
experimental thermal gradient. Tpref data were divided and averaged into 10 minutes intervals. Semi-
transparent points represent individual values of averaged Tpref. The central line of each box indicates the
median of the averaged Tpref and the black line connecting these values represents its variation through time.
The dashed line indicates the 10 min interval from which Tpref values remained stable in time and were
considered in analyses. B) Histogram displaying the global distribution of Tpref values within the 31-90 min time
interval considered in further analysis.

[Link] Influence of endogenous factors and UHI on Tpref

We found that females had a warmer mean Tpref (Table 4.2, Figure 4.4A) and a narrower Tpref range
(Table 4.2, Figure 4.4B) than males. Post hoc analyses revealed that these two relationships were
present in adults, but not in juveniles (Table 4.2, Figure 4.4C and D). In addition, larger individuals (i.e.,
adult females) had warmer mean Tpref (Table 4.2, Figure 4.4E). We detected an interaction between
development stage and UHI intensity on Tpref range (Table 4.2). Post hoc tests on interactions
showed that Tpref range became more precise, as UHI intensity increased, but only in juveniles
(Appendix 4.5, Figure 4.4F). Models fitted with UHI intensities measured over longer time periods
showed similar relationships between the interaction of UHI intensity and developmental stage, and
the Tpref range (Appendix 4.7, Appendix 4.3).

149
PARTIE IV

Table 4.2 - Effects of endogenous (development stage, sex, prosoma length), environmental factors (UHI
intensity) and their interactions on mean Tpref and Tpref range estimated by generalized linear mixed models
(GLMM). Significant results (P<0.05) are in bold. Results of models after variable selection are displayed.

Est. SE P-val R²
Mean Tpref 0.105
Development stage -1.005 0.260 <0.001
Sex -0.876 0.327 0.007
Prosoma length 0.491 0.185 0.008
Development stage : Sex 1.401 0.384 <0.001
Tpref range 0.140
Development stage 0.834 0.438 0.057
Sex 1.053 0.180 <0.001
UHI intensity -0.128 0.148 0.387
Development stage : Sex -1.420 0.279 <0.001
Development stage : UHI intensity -0.580 0.238 0.015

150
 Chapitre 5

Figure 4.4 - Graphs representing significant relationships between Tpref parameters (mean Tpref and Tpref
range) and the tested endogenous and environmental factors. White dots represent females and black dots are
males. A and B show the differences in Tpref among sexes. C and D display relationships observed between
Tpref parameters and developmental stage, was well as significant interactions (*: p-value < 0.05) between
developmental stage and sex. E represents the positive relationship between mean Tpref and the prosoma
length. F represents the relationship between Tpref range, the significant interaction of the UHI intensity (from
March to April 2022) and developmental stage (solid line represents adults, dashed line represents juveniles).

151
PARTIE IV

4.1.4 Discussion
While the plasticity of thermal tolerance limits across urban gradients has received attention (Brans
et al. 2017, Diamond et al. 2017), little is known on the plasticity of Tpref in ectotherms. In this study,
we showed that Tpref varies only marginally (and only in juveniles) along an UHI gradient in a
common spider species. Pardosa saltans showed a preferred temperature of 33.40 °C (± 2.0°C),
which is in line with previous results obtained on lycosids spiders (Nørgaard 1951, Sevacherian and
Lowrie 1972, Humphreys 1978). Tpref is more strongly modulated by endogenous factors such as
sex, life stage and body size than the UHI gradient. This result implies that traits may be more
important than the environment in defining the parameters of behavioural thermoregulation. Our
results further imply that the Bogert effect may prevail in this spider system, although we cannot
exclude that the thermal optima hypothesis rules the response in juveniles.

[Link] Role of endogenous factors

Sex and body size are generally identified as an important determinant of several thermal traits
(Angilletta 2009). Our study confirms the importance of sex and body size on thermal responses. Sex
was the only factor, among the predictors we tested, to explain part of the variance for both the
mean Tpref and the Tpref range. The effect, however, was opposite for the two parameters. If mean
Tpref was warmer in females, their Tpref range was narrower. The difference of Tpref range among
sexes suggests that females, which tend to be larger than males, choose more accurately the
temperatures to which they expose themselves. Hence, this observation coupled with the warmer
mean Tpref observed in females suggests that they tend to be 'specialist' thermoregulators, placing
themselves accurately in the mosaic of thermal environments where temperatures are rather high.
Contrastingly, smaller male individuals may be more 'generalists', having lower preferred
temperature but exploring a wider range of thermal conditions. These patterns of variation in Tpref
mean and Tpref range in adult spiders may be linked to the framework 'suboptimal is optimal'
(Martin and Huey 2008). This theory relates the asymmetric shape of temperature-performance
curves to the expected Tpref mean, which should shift below optimal temperature to avoid exposure
to challenging temperatures. This should be especially true for thermal specialists, like Lyscosid
females, for which the temperature range between optimum and critical maximal temperature is
particularly narrow. Nevertheless, we lack data on this spider thermal biology (e.g., optimum) to test
this theory any further. Similar complexity was already observed in the relationship between body
size (or mass) and thermal tolerance traits in arthropods, yet with a general pattern suggesting a
positive association between body size and heat tolerance (Blanckenhorn 2000, Nielsen and Papaj
2015, Klockmann et al. 2016). Given that the CTmax, the optimal temperature and Tpref (mean) are
often related to some extent (Angilletta 2009), our result on a similar positive relationship between
Tpref and body size is not surprising. Yet, contrasting patterns of body size effect on other
thermoregulation traits such as the onset temperature for refuge seeking were also observed
(Nielsen and Papaj 2015). More experimental studies are needed to identify a general pattern
between body size and Tpref metrics and the functional underpinnings of this relation.
Further, the observed divergence between the mean Tpref of males and females in P. saltans
contrasts with previous observations in other Lycosids (Nørgaard 1951, Frick et al. 2007). However,
the mean Tpref of Lycosids is well known to strikingly vary in adult females carrying an egg-sac
(Sevacherian and Lowrie 1972), as they select warmer microhabitats that are more suitable for their

152
 Chapitre 5

cocoon. Our results suggest that females P. saltans are already more thermophilic than males prior
to egg laying. In addition, the difference of Tpref range between sexes has never been revealed
before in Lycosids. We propose that the higher variability of Tpref in males is related to their
exploratory behaviour. In Pardosa spiders, males and females display contrasting movement and
exploratory behaviours. Heydemann (1960) reported that populations with high abundance are
structured with females mainly located within optimal biotopes whereas males move across adjacent
marginal biotopes where density is lower. Further, P. saltans males can be twice as active as females
and they can explore larger surfaces (Trabalon 2022). Indeed, adult males start to actively search for
females during the reproductive period (Vlijm and Richter 1966, Hallander 1967, Cady 1983), which
can in turn be accompanied by an increase of their Tpref range. More experimental works are
needed to generalize (or not) this effect, especially by comparing species with various activity levels
and contrasting exploratory behaviours.
Our study further brings evidence that Tpref range might increase, especially in males, once
individuals reach sexual maturity in response to a change in their ecology relative to juveniles.
Previous studies on species from the genus Pardosa have pointed out that habitats used by
conspecifics are stage-dependant (Hallander 1967, 1970b, Edgar 1971). For example, juveniles of P.
pullata were shown to hide in grass tussocks, whereas adults stayed at the surface of the meadow
vegetation (Hallander 1970b). Sub-adults of P. lugubris (former name of P. saltans) were reported to
overwinter until March in shady parts of forest leaf litter, whereas adults explore a broader habitat
range (Edgar 1971). These relationships between habitat preferences across spiders’ life stages are
likely to be accompanied by shifts in experienced thermal environments and could explain why Tpref
range and development stage are closely related.

[Link] Role of the urban heat island

[Link].1 Mean Tpref suggests Bogert hypothesis
We found no relationship between the mean Tpref values and the thermal environment measured
as UHI intensities. This result gives therefore support to the Bogert effect hypothesis which states
that adjustments of traits to local climate conditions can be inhibited by thermoregulatory
behaviour. P. saltans relies on behavioural thermoregulation in its usual habitat that displays high
thermal heterogeneity during spring and summer. Its thermophilic nature makes it particularly active
and abundant in sunny patches such as wood clearings and hedges (Trabalon 2022). Therefore, this
thermoregulatory ability could constrain the evolution of Tpref mean as long as favourable thermal
microenvironments are available in its habitats across the urban landscape.
P. saltans is active on the leaf-litter during daytime in the growth season. Despite soil surfaces are
often considered as relatively homogenous and buffered thermal environments (Lembrechts et al.
2020), they can display higher thermal heterogeneity at small spatial scales. Indeed, local (i.e., spider-
scale) thermal heterogeneity may be of primary importance because the UHI gradient studied results
in an increase in mean air temperatures of up to 2.5°C (no sampling sites were available in the city
centre, where the UHI intensity reaches 3°C). The amplitude of the UHI gradient (0 - 2.5°C) may be
too weak to cause adaptation or acclimation leading to a change in behaviour that can be detected
in the Tpref assay. In contrast, Pincebourde & Salle (2020) measured soil surface temperature over
the horizontal ground layer of a temperate forest using thermal cameras and loggers and reported
temperature ranges of up to 16°C within less than 1 meter. This small scale heterogeneity results

153
PARTIE IV

from the interaction between multiple factors, including sun exposure, leaf litter microtopography
and shading effects by above-ground vegetative elements (Pincebourde and Salle 2020). The scale
at which this variability is perceptible in forest litter surface matches the scale corresponding to the
size of many arthropods, i.e., an area of about 20 cm in length (Pincebourde and Woods 2020).
Therefore, there is plenty of thermal opportunities for P. saltans to thermoregulate behaviourally
over fine spatial scales on the leaf litter in semi-open areas that are typical in wood patches across
urban landscapes. Future studies need to integrate more deeply temperatures at micro-scale that
are pertinent for those arthropods at the soil surface, to infer the role of this microclimatic mosaic
on the thermoregulation behaviour and ability to escape extremes (Woods et al. 2015). The fact that
P. saltans has a predominantly diurnal activity may explain its particular ability to take advantage of
a thermally heterogeneous diurnal environment to actively thermoregulate. However, the nocturnal
nature of air warming in an urban environment challenges the theory that daytime temperatures
release evolutionary constraints on Tpref. Although active during daytime, P. saltans individuals
experience the nocturnal temperature increase in a thermal microhabitat that is much more
homogeneous than during the day, because sun exposure and associated shading effects disappear.
However, apart from a Bogert effect or a low explanatory power of the UHI metric considered,
several other factors may explain Tpref lack of response to urban warming. Among these, Tpref may
exhibit a lack of plasticity or heritable variation in P. saltans (but see Paranjpe et al., 2013). In
addition, urban conditions may rarely allow spiders to experience their preferred temperatures as
predator, prey or even humidity distribution may force individuals to thermoconform, as observed
by Stork et al. (2021) in a Lycosid spider.

[Link].2 Tpref range suggests hypothesis of local optima adaptation for juveniles only
Although results for adults and the Tpref mean suggest the Bogert hypothesis, another scenario is
observed for the Tpref range of juveniles. We observed that the Tpref range decreased significantly
with increasing UHI intensity, corresponding to the local optima adaptation (Figure 4.1, left-side,
bottom line). This indicates that, while the mean of Tpref remains similar across urban populations,
the spiders are more selective relative to temperature (or more precise) under more intense UHI
conditions compared to rural areas. This result can be explained by the microclimate conditions that
juveniles experience during the winter. As previously mentioned, P. saltans individuals overwinter as
juveniles in forests’ dry-leaves litter. Indeed, in their natural environment, low-mobile juveniles
spend most of their lifetime under relatively shady and thermally homogeneous conditions. These
microenvironments offer limited opportunities to behaviourally thermoregulate and favour
therefore selection pressure on thermoregulation traits. Hence, the overwintering period might be
responsible for shifts in Tpref range among rural and urban populations since also winters are
marked by UHI events in Rennes (Dubreuil et al. 2020). These winter UHI events could have driven
beneficial local acclimation in juveniles’ Tpref ranges by imposing novel selective pressure. The
observed inter-population variations in Tpref range might disappear as soon as sub-adults moult to
reach their final adult stage, and thereby start exploring broader thermal conditions. Endogenous
variations in thermal traits are likely to occur in arthropods in which a single stage overwinters, or in
which development stages differ in habitat preference and mobility (Marais and Chown 2008).
However, to date only few studies explicitly considered life-stages when testing for intra-specific
differences in arthropod’s Tpref response across geographic clines (but see MacLean et al. 2019). To
infer the factors affecting behavioural thermoregulation under natural conditions, we tested

154
 Chapitre 5

individuals collected in the field, i.e., non-acclimated to controlled temperatures. Experiments

combining field and laboratory approach are ideal for conducting behavioural experiments, although
multiple factors may still be confounding (Calisi and Bentley 2009), as the limited part of variation
explained in our study (14%) suggests. To further our interpretations, conducting experiments in
common gardens and comparing thermoregulatory patterns along a UHI gradient could provide
additional information to elucidate how phenotypic plasticity and adaptive changes interact in
thermal preference under rapid urban development for different life stages.

Conclusion

We showed that contrasting thermoregulatory behaviour patterns along an UHI gradient were
modulated by the developmental stage in populations of P. saltans. We propose the hypothesis of
Bogert effect to be relevant for adult individuals and the adaptations of local optima in juveniles. Our
proposed conceptual framework further brings evidences that the mean Tpref and the Tpref range
can respond differently. Considering both parameters in future studies could help to better predict
complex and variable responses of ectotherms’ Tpref along geographic clines. If we further brought
evidence that urban thermal macro-conditions can trigger changes in thermoregulatory behaviour,
our interpretations also suggest microclimates, and particularly their local heterogeneity, to be a
major driver of arthropods’ thermal preferences. Including fine-scale local temperature data (e.g.,
from thermal imagery) to compare the implication of the thermal environment for arthropods’
thermoregulatory behaviour at multiple scales would therefore be a promising approach.

155
 PARTIE IV

156
4.1.5 Appendices

Appendix 4.1 - Laboratory gradient and corresponding thermal image.

 Chapitre 5

Appendix 4.2 - Calibration curves related to gradients A) and B), obtained by temperatures measured every
centimetre with thermocouples as a function of the distance to the warmer point.

Juveniles

2.0
Tpref range (°C)

1.5
Legend
UHI (December 2021 - February 2022)
UHI (March 2022 - April 2022)
UHI (September 2021 - April 2022)
1.0

0.5

1.0 1.5 2.0 2.5

UHI intensity (°C)

Appendix 4.3 - Significant negative relationship between Tpref range of juveniles P. saltans and three UHI
variables measured on distinct time intervals.

157
PARTIE IV

Appendix 4.4 - Effects of time interval on mean Tpref values estimated by generalized linear mixed models
(GLMM). Results of four models are shown, differing in range of time intervals included as fixed effects.
Significant results (P<0.05) are displayed in bold.
Random effects (Id individual): Fixed effects (time interval): ANOVA:
Groups [Link]. Time interval Estimate Chisq Df P-value
1. Id_individual 1.861 Intercept 32.358 134.74 7 2.2E-16
Residual 1.907 10-20min 0.6098
20-30min 0.8662
30-40min 1.1071
40-50min 1.0198
50-60min 0.978
60-70min 1.0206
70-80min 1.1094
2. Id_individual 1.871 Intercept 32.9675 31.245 6 2.3E-05
Residual 1.717 20-30min 0.2542
30-40min 0.4957
40-50min 0.4092
50-60min 0.3698
60-70min 0.4065
70-80min 0.4984
3. Id_individual 1.918 Intercept 33.2193 8.5458 5 0.1286
Residual 1.562 30-40min 0.2431
40-50min 0.1589
50-60min 0.1211
60-70min 0.1507
70-80min 0.2442
4. Id_individual 1.973 Intercept 33.46064 3.0011 4 0.5576
Residual 1.45 40-50min -0.082864
50-60min -0.119363
60-70min -0.095265
70-80min 0.001278

158
 Chapitre 5

Appendix 4.5 - Results of GLMM post hoc testing all possible combinations of interacting factors ‘sex’ and
‘developmental stage’. Post hoc tests were performed separately on models related to mean Tpref and Tpref
range. Significant results (P<0.05) are in bold.
Estimate Std. error P-value

Mean Tpref
Adult F - Juvenile F 1.005 0.260 <0.001
Adult F - Adult M 0.876 0.327 0.037
Adult F - Juvenile M 0.480 0.300 0.377
Juvenile F - Adult M -0.129 0.269 0.963
Juvenile F - Juvenile M -0.525 0.265 0.197
Adult M - Juvenile M -0.396 0.253 0.398

Tpref range
Adult F - Juvenile F 0.085 0.194 0.972
Adult F - Adult M -1.053 0.180 <0.001
Adult F - Juvenile M 0.452 0.202 0.114
Juvenile F - Adult M -1.137 0.191 <0.001
Juvenile F - Juvenile M 0.368 0.214 0.312
Adult M - Juvenile M 1.505 0.199 <0.001

Appendix 4.6 - Effects of UHI intensity on Tpref range estimated by linear models. Results of two models are
shown, computed on two separated datasets including respectively only adult individuals or juveniles.
Significant results (P<0.05) are displayed in bold.
Estimate Std. error P-value R²

Adults
UHI intensity (March to
-0.202 0.154 0.191 0.006
April 2022)

Juveniles
UHI intensity (March to
-0.658 0.189 <0.001 0.005
April 2022)

159
PARTIE IV

Appendix 4.7 - Effects of ontogenetic (dev. stage, sex, body size and leg length) and environmental factors (UHI
intensity) on juveniles’ Tpref range estimated by generalized linear mixed models (GLMM). Results of three
models are shown, differing in the UHI variable used, calculated on distinct time intervals. Significant results
(P<0.05) are displayed in bold.
Estimate Std. error P-value R²

UHI: March to April 2022 0.140

Development stage 0.834 0.438 0.057
Sex 1.053 0.180 <0.001
UHI intensity -0.128 0.148 0.387
Development stage : Sex -1.420 0.279 <0.001
Development stage : UHI
-0.580 0.238 0.015
intensity

UHI: September 2021 to April 2022 0.131

Development stage 0.771 0.455 0.090

Sex 1.058 0181 <0.001

UHI intensity -0.080 0.161 0.622

Development stage : Sex -1.400 0.280 <0.001
Development stage : UHI
-0.575 0.265 0.030
intensity

UHI: December 2021 to February 2022 0.127

Development stage 0.899 0.518 0.082

Sex 1.064 0.181 <0.001

UHI intensity -0.025 0.196 0.897

Development stage : Sex -1.374 0.280 <0.001

Development stage : UHI

-0.697 0.330 0.034
intensity

160
 Chapitre 5

161
PARTIE IV

4.2 Chapitre 6 : Potentiel des prairies urbaines pour atténuer le

réchauffement urbain à l’échelle des arthropodes

Résumé

Dans les villes, les écosystèmes sont particulièrement vulnérables au réchauffement nocturne dû à
la formation d'îlots de chaleur urbains (ICU). Si le réchauffement urbain nocturne est bien
documenté comme ayant des effets délétères sur l'homme, ses effets sur les organismes vivant en
ville sont beaucoup moins documentés. Les ICU sont souvent cartographiés à une résolution trop
grossière pour rendre compte de l'environnement thermique réel auquel sont soumis les organismes
de taille réduite tels que les plantes herbacées ou les animaux ectothermes. Dans cette étude, nous
comparons la température minimale journalière mesurée à une hauteur de 3 m pour le suivi de l’ICU
atmosphérique, avec les températures minimales journalières à une hauteur de 5 cm dans les
prairies adjacentes. Nous utilisons des températures atmosphériques de référence en milieu rural et
en milieu urbain pour calculer les décalages de température ou « offsets ». Nous montrons qu'à une
échelle écologiquement pertinente, les prairies ont le potentiel de limiter l’exposition des
organismes au réchauffement nocturne induit par les ICU. Nous constatons également que la
structure des prairies et de leur environnement direct détermine le potentiel d'atténuation de l'ICU.
Par exemple, la proximité de couverts arborés et bâtis limite les effets de refroidissement nocturne.
À l'inverse, la hauteur et la densité de la canopée herbacée sont positivement liées au
refroidissement. Ces résultats peuvent aider les gestionnaires de parcs et les urbanistes à concevoir
et à mettre en œuvre des infrastructures vertes dont les caractéristiques climatiques soutiennent
efficacement la biodiversité urbaine.

Article en préparation

162
 Chapitre 6

Urban grasslands can efficiently buffer organisms from nocturnal warming

induced by the urban heat island

Valentin Cabon1,2, Benjamin Bergerot1,2, Charlotte Brabant2,3, Vincent Dubreuil2,3, Eva Gril4, Jonathan
Lenoir4, Sylvain Pincebourde5, Hervé Quénol2,3

1. Université de Rennes, CNRS, UMR 6553 ECOBIO, Rennes, France

2. LTSER ZA Armorique, F-35000 Rennes, France
3. Université de Rennes 2, CNRS, UMR 6554 LETG, Rennes, France
4. Université de Picardie Jules Verne, CNRS, UMR 7058 EDYSAN, Amiens, France
5. Université de Tours, CNRS, UMR 7261 IRBI, Tours, France

Abstract
In cities, ecosystems are vulnerable to night-time warming through the formation of urban heat
islands (UHI), particularly in the context of climate change. If urban nocturnal warming is well
documented as having deleterious effects on humans, few less is known about its consequences on
city-dwelling organisms. UHI is often mapped at a resolution which is too coarse to capture the actual
thermal environment experienced by small organisms such as herbaceous plants or ectothermic
animals. In this study, we compare daily minimal temperature typically measured at 3-m height to
monitor UHIs, with daily minimal temperatures at a 5-cm height in adjacent grasslands. We used
reference rural, as well as reference urban atmospheric temperature to calculate temperature
offsets. We show that at an ecologically relevant scale, grasslands have the potential to buffer
organisms from nocturnal warming induced by UHI. We found that the structure of grasslands
determined the potential of UHI mitigation. For example, the proximity of tree- and built-canopies
limited nocturnal cooling effects. Conversely, the height and the density of the herbaceous canopy
were positively related to cooling. These finding may help parks managers and urban planners to
conceive and implement climate-smart green infrastructures efficiently supporting urban
biodiversity.

Key words
climate change, green infrastructure, microclimate, temperature buffering, urban biodiversity, urban
planning

163
PARTIE IV

4.2.1 Introduction

Temperature increase due to anthropogenic climate change is widely acknowledged to affect

ecosystems (Lenoir and Svenning 2015, Scheffers et al. 2016, IPCC 2021). Urban ecosystems are
particularly exposed to warming through the urban heat island (UHI) phenomenon, one of the most
obvious examples of human-induced local modification of the climate (Oke et al. 2017). This
phenomenon is mainly attributable to the transformation of natural habitats by constructed
landscapes, leading to new topography and new thermal properties of surfaces, which in turn
increase the temperature (Stewart and Oke 2012). As is the case for global warming (Speights et al.
2018), a disproportionate amount of urban warming can be attributed to “less cool” nocturnal
conditions rather than warmer diurnal conditions. A daily minimal temperature in a city can be up to
12°C degrees higher than in a surrounding rural areas (Oke 1982, 1995). This nocturnal warming is
typically measured at a spatial scale matching the limit between local climate and microclimate (i.e.
100-m; Littmann 2008, Geiger et al. 2009, Barry and Blanken 2016), and is increasingly considered
as an important predictor of biotic changes in urban ecology studies (Singh et al. 2020). UHI-induced
nocturnal warming is not expected to constrain organisms by extreme temperatures since mostly
nocturnal air-cooling decreases. However, it may prolongs organisms’ exposure to abnormally warm
conditions at night, with possible short-term deleterious consequences on populations (Zhao et al.
2014a, Tougeron and Abram 2017). For example, ectothermic animals rely on the thermal
environment to regulate their body temperature (Huey and Kingsolver 1989). If night cooling is
determinant for physiological recovery in many ectotherms, nocturnal warming can have adverse
effects on their physiology, behavior, and population size (Rutschmann et al. 2024). Previous urban
studies have also shown that nocturnal warming can reduce ectotherm species richness, by
excluding warming-sensitive species from communities (Piano et al. 2017, Cabon et al. 2024b).
In cities, green infrastructures are acknowledged to be efficient tools to mitigate nocturnal warming
induced by atmospheric UHI (Zölch et al. 2016). Urban green infrastructures include a wide range of
habitats such as woodlands, parks, trees alignments or green roofs and facades (Gaffin et al. 2012).
Multiple studies have demonstrated the cooling potential of urban trees via shading and
evapotranspiration (Hall et al. 2012, Skelhorn et al. 2014, Moser et al. 2015), while others reported
cooling due to green facades (Jänicke et al. 2015) and green roofs (Mutani and Todeschi 2020, Wang
et al. 2022). However, the vast majority of these studies are conducted in order to assess the human
thermal comfort, and the reported atmospheric temperatures do not reflect the thermal conditions
experienced by many smaller organisms, such as plants or ectothermic animal (Potter et al. 2013,
Lenoir et al. 2017, Bramer et al. 2018, Pincebourde and Woods 2020, Harvey et al. 2022, Haesen et
al. 2023). Among urban green infrastructures, grasslands have a great potential for biodiversity
conservation, as they have the capacity to host species-rich plant and animal communities (Klaus
2013), including species of conservation concern (Buchholz et al. 2018, Chollet et al. 2018). Due to
their manyfold management regimes, the structure of urban grasslands often ranges from
intensively managed lawns (with short and dense vegetation), to extensive meadows (with higher
and more heterogenous herbaceous canopy) within a same city (Chollet et al. 2018, Norton et al.
2019).

164
 Chapitre 6

Such variation in grassland structures leads to highly heterogenous microclimatic conditions at the
micro-habitat scale, which may in turn have important implications for organisms living under the
herbaceous canopy (Geiger et al. 2009). In addition, open grassland ecosystems experience stronger
spatial and temporal microclimatic variation than wooded habitats (Gril et al. 2023) and climate
variability is generally greater near the surface because of reduced wind speed, heat stored in the
ground and less efficient heat-transfer than in upper air strata (Geiger et al. 2009, Bramer et al.
2018). Despite the climatic vulnerability of grassland-dwelling organisms, few microclimatic studies
have been to date conducted in these urban microhabitats.
Calculating temperature offsets is an efficient method to assess to what extent a microclimate
deviates from adjacent atmospheric conditions (Zellweger et al. 2019, Haesen et al. 2023). Offsets,
obtained by subtracting atmospheric from microclimate temperatures, enable to determine
whether the micro-habitat provides warmer (positive values) or cooler (negative values) conditions,
compared to the adjacent atmospheric environment. Temperature offsets can be affected by
multiple environmental predictors that can operate from the landscape to the local scale
surrounding a micro-habitat (Zellweger et al. 2019). For example, climatic conditions may constrain
offsets at the landscape scale, whereas local vegetation characteristics can determine the warming
rate attributable to anthropogenic climate change that organisms experience (Zellweger et al. 2019).
For example, wooded edges were found to trap grasslands’ outgoing thermal emittance at night,
favoring microclimatic nocturnal warming (Bernath-Plaisted et al. 2023). In addition, the grassland’s
herbaceous canopy structure might also be a determinant driver of local temperature and humidity
conditions and modulate the UHI-induced warming experienced by grassland-dwelling organisms.
Therefore, identifying the structural features mitigating UHI-induced nocturnal warming within
grasslands could help city councils and urban planners to develop microclimate-oriented
management strategies and green infrastructure design that not only improve human thermal
comfort, but also better support urban biodiversity.
In this study, we aim to determine to which extent thermal micro-habitats provided by grasslands
within a city enable to buffer organisms against atmospheric UHI-induced warming. Since this
phenomenon occurs at night, we investigate daily minimum temperatures (Oke 1995). To evaluate
the cooling potential of micro-habitats, we quantified the daily temperature offsets between air
temperature measured five centimeters above the ground within urban grasslands and the
atmospheric air temperature at a 3-m height, measured by urban and rural reference stations. In a
second step, we tested which structural features surrounding micro-habitats better explain the
offset variability observed among sites. To account for the potential effect of local climatic conditions
on grassland’s offsetting capacity, we quantified the UHI intensity throughout the entire study
period, and investigated whether temperature offsets varied under strong versus weak UHI
conditions.

165
PARTIE IV

4.2.2 Material and methods

[Link] Study area and study period

We conducted this study in Rennes city (northwestern France, 48°06′ N-1°40′ W) and its
surroundings (Figure 4.5). Rennes is a city of 227,000 inhabitants under a strong oceanic climate
influence, that regularly faces strong UHI events (Dubreuil et al., 2020). In this study, all temperature
measurements were performed between June and September 2022. This period corresponds to a
particularly warm summer in France with the three consecutive heatwave events in mid-June, mid-
July and early August. For the city centre of Rennes, records of UHI intensity, estimated as the
difference between the daily minimal temperature measured by the urban and the rural reference
weather stations, reached up to 8°C during that period. This study period therefore provides ideal
experimental conditions to investigate to which extent temperatures in the herbaceous vegetation
layer deviate from extreme atmospheric measurements.

166
 Chapitre 6

Figure 4.5 - Map of the study area. Green dots are sensors at 5 cm height used to record the near-ground
temperature within grassland vegetation. Red crosses are sensors at 3 m height used to record reference urban
atmospheric temperature. The yellow triangle is the rural reference station. Black thin lines display the 39
couples of near-ground sensors and urban reference stations used to calculate temperature offsets. Grey areas
are impervious surfaces, blue water streams and ponds and black thick line the administrative borders of Rennes
city.

167
PARTIE IV

[Link] Near-ground temperature data recording

In Rennes, grasslands represent a major part of the urban green infrastructure with a total of 445
ha, corresponding to 56% of the total surface of green infrastructure (Chollet et al., 2018). Moreover,
the various management strategies implemented throughout Rennes lead to grassland canopy
structures that vary from short to high and from dense to sparse. These contrasting herbaceous
canopy structures may lead to diverse microclimatic conditions, a determinant ecological driver for
the flora and fauna living inside or under the grassland canopy (Lenoir et al. 2017, Harvey et al. 2020,
Pincebourde and Woods 2020, Haesen et al. 2023, Harvey and Dong 2023). A total of 39 grasslands
were selected (Figure 4.5) such that they encompass a broad UHI gradient and a broad range of
canopy structures. At each grassland site, we monitored near-ground temperature conditions at 5
cm above the ground (Bernath-Plaisted et al. 2023) using a specific network of temperature sensors
composed of 21 Lascar EL-USB-2+ (Lascar Electronics Ltd., Whiteparish, Wiltshire SP5 2SJ, UK) and
15 Tinytag Talk 2 TK-4023 (Gemini Data Loggers Ltd., Chichester, West Sussex PO198UJ, UK). To
measure the actinothermal indices corresponding to the temperatures experienced by organisms
near the ground, we did not use shields to cover the loggers. Near-ground temperatures were
monitored during 15 contiguous days per month from June to September 2022 resulting in a total of
60 (4×15) days of temperature monitoring.

[Link] Rural and urban reference temperature recording

To study how near-ground temperatures in grasslands deviate from standardized atmospheric
records, we gathered simultaneous records of temperature obtained by (1) a network of urban
reference sensors scattered across the city of Rennes and affected by the UHI effect as well as by (2)
a single rural reference weather station located outside the city of Rennes and thus unaffected by
the UHI effect. Urban reference sensors belong to a long-term climate observation network
established in Rennes in 2004 with the aim to characterize the UHI intensity across space and time
(Dubreuil et al., 2020; Foissard et al., 2019). Since 2004, the Rennes Urban Network
([Link] has grown from 12 to 30 DAVIS Vantage-Pro-2 automatic weather stations
(Davis Instruments Corporation, Hayward, CA 94545, USA), to which 93 RisingHF RHF1SOO
connected temperature sensors (LoRaWAN Solutions RisingHF, Shenzhen 518057, China) were
added in 2020 (Dubreuil et al., 2022). These automatic weather stations and connected sensors are
spread over suburban and urban areas and record temperature at 3 m height, following the
standards of the World Meteorological Organisation (WMO 2023), every 60 min and 15 min,
respectively. To pair each near-ground temperature sensor installed in a focal grassland with its local
urban reference, we selected the geographically closest urban sensor available from the Rennes
Urban Network. This selection resulted in 39 pairs of sensors (near-ground vs. urban) from 32 distinct
urban reference sensors: one urban sensor can be paired with several neighbouring near-ground
temperature sensors (Figure 4.5). The geographical distance between a near-ground sensor and its
paired urban sensor was 583 ± 338 m on average. The rural reference weather station is located
about 11 km north of Rennes (Melesse, 48°12′18.1″N 1°40′52.8″W). It also belongs to Rennes Urban
Network but the landscape surrounding this weather station located outside the city of Rennes is
characterised by agricultural fields and hedgerows and thus is unaffected by the UHI effect (Foissard
et al., 2019).

168
 Chapitre 6

[Link] Descriptors of the environment structure

To quantify the environmental structures in the surrounding of each study site, which are likely to
influence near-ground temperatures, we quantified building and tree height. To do so, we subtracted
Digital Terrain Models (DTM) raster data from Digital Elevation Models (DEM) raster data, derived
from LiDAR and made available in open access by Rennes Metropole. We performed this raster
subtraction twice, using (1) the DEM data related to trees only
([Link] dataset/mnt_2021_arbres/information/, accessed on 1st
May 2024) and (2) the DEM model related to buildings only ([Link]
explore/dataset/mnt_2021_batiments/ information/, accessed on 1st May 2024). By doing so, we
converted elevation (above-sea level) to height data (in meters). The obtained two datasets present
an altimetric accuracy of 0.2 m and a vertical resolution of 0.5 m. To explore the effect of building
and tree height at multiple spatial scales, we cropped each two rasters (i.e., tree and building height)
in circular buffers of 10, 20 and 50-m-radius around the GPS coordinates of each near-ground and
urban reference sensor. We then estimated the mean height of buildings and trees (in meters) within
each buffer by computing the mean pixel value, and considered ‘NA’ pixels as zero values in order to
account for the proportions of cover in the final variable. The processing of variable computation
from DTM and DEM was conducted in R (R Core Team 2022) using the ‘sf’ (Pebesma et al. 2024) and
‘raster’ (Hijmans et al. 2023) packages. In addition, to investigate the influence of the herbaceous
canopy structure of the studied grasslands in the direct vicinity of near-ground sensors, we measured
the mean height and the mean cover of the herbaceous vegetation layer within a 2-m buffer around
each near-ground sensor and repeated this measure every month during the four months of the
study period.

[Link] Daily UHI characterization

Because the UHI is expected to influence atmospheric temperatures and therefore the way near-
ground measurements relate to it, we estimated the temporal dynamic, at a daily resolution, of the
UHI intensity throughout the entire study period. To estimate the UHI intensity on a daily basis, we
calculated the difference between daily minimum temperature as measured by one of the urban
reference stations located right in the city center (Champs-Libres, 48°06'17.5"N 1°40'26.2"W) and
the daily minimum temperature as measured by the rural reference station located outside the city
of Rennes.

[Link] Data analysis

To explore how near-ground temperature measured within the herbaceous vegetation layer
deviates from atmospheric conditions and thereby determine whether urban grasslands can mitigate
the UHI effect at a microhabitat scale, we focused our analysis on daily minimum temperature data
solely. Indeed, the UHI effect occurs at night essentially and considering daily minimum
temperatures only ensure to capture the temporal thermal variability due to this phenomenon. We
thus computed the difference or “offset” in daily minimum temperature between a given grassland
temperature sensor and its paired urban reference sensor, resulting in a time series of 1,961 offset
values for each of the 39 pairs of grassland-urban sensor. We did the same with the rural reference
weather station to obtain 39 time series of offset values between near-ground air temperature

169
PARTIE IV

conditions at 5 cm height inside grasslands and ambient air temperature conditions at 3-m height in
a rural landscape outside the influence of the UHI effect emanating from Rennes. In addition, we
also calculated time series of daily offset values in minimum temperature between the rural
reference weather station and each of the 32 urban reference sensors. This allows to compare how
urban atmospheric and near-ground temperatures deviate from standardized reference
temperature measured at a single location.
To explain the observed variation in the thermal offsetting capacity of urban grasslands, we
computed generalized mixed-effects models (GLMM) using the ‘lmer’ function from the ‘lme4’
package (Bates et al. 2015) in R (R Core Team 2022). We modelled separately, as two distinct
response variables: (R1) the offset in daily minimum temperatures between near-ground conditions
in urban grassland and the rural reference conditions and (R2) the offset in daily minimum
temperatures between near-ground conditions in urban grasslands and its neighboring urban
reference conditions at 3 m height. We first included, as fixed effects variables in the model formula,
all descriptors of the environment surrounding the sensors (i.e., mean height of trees, mean height
of buildings, monthly herbaceous vegetation cover, monthly herbaceous vegetation height) as well
as daily UHI. For the second response variable (R2), the two predictor variables of mean height of
trees and mean height of building surrounding the sensors are available for both sensors involved in
a given pair. Hence, we computed the difference in mean height of trees and buildings and used it
as a predictor for the second response variable (R2) while we kept the original values of mean height
of trees and building around the grassland sites for the first response variable (R1). Note also that
we fitted three different GLMMs for each of these two response variables (R1 and R2) to test the
three different buffer sizes (10-m, 20-m and 50-m radius) that are available for the two predictor
variables of mean height of trees and mean height of building surrounding the sensors. As the effect
of each of the predictor variables on the daily offset in minimum temperature between near-ground
and atmospheric temperature (cf. R1 and R2) is likely to differ depending on the UHI intensity, we
added an interaction term between ‘UHI intensity’ and every other explanatory variable included as
fixed effect. To account for methodological effects, we added the type of sensor and the type of land
cover under the control sensors (vegetated or sealed surface) to all 6 GLMMs as fixed effects. Finally,
we added the identity of the urban reference sensor as a random intercept term in GLMMs testing
R2 as response variable, since one urban reference sensor could have been paired with multiple
near-ground sensors for offset calculations (Figure 4.5). Prior to model computation, we performed
an analysis of correlations among the measured predictor variables to check for independence. We
then selected the best models among all possible candidate models using the ‘dredge’ function and
averaged the top- ranking models showing a delta AIC <2, using the ‘[Link]’ function in ‘MuMin’
package (Bartoń, 2016). Finally, we tested the significance of the explanatory variables by performing
type III ANOVA tests using the ‘Anova’ function in ‘car’ package (Fox et al., 2007) and calculated the
Conditional and marginal R² to assess the proportion of variation explained by the selected models
(Table 4.3).

170
 Chapitre 6

4.2.3 Results
[Link] Comparisons of daily minimal temperatures
Our results confirm that an urban heat island effect occur at the atmospheric level in Rennes
(Dubreuil et al. 2020), since all the control sensors distributed throughout the city recorded warmer
daily minimal temperatures than the reference rural station. In average, the daily minimal urban
reference temperatures were 1.89 ± 0.75 °C warmer than the daily minimal temperatures measured
at the reference rural station (Figure 4.6a). In addition, we observe an offsetting capacity of grassland
vegetation toward cooler near-ground micro-habitats compared to atmospheric conditions.
Grassland cooling at the near-ground level is partially occurring when daily minimal temperatures
are compared to the rural reference station (R1), with one third of studied grassland (13 from 39)
showing negative offsets (Figure 4.6b). However, the average temperature offset of 0.56 ± 1.65 °C
indicates that warming remains the dominant situation. Finally, when near-ground daily minimal
temperatures of grasslands are compared to the temperature records from urban reference sensors
(R2), cooling becomes the dominant situation, with 80% of studied grassland (31 from 39) showing
negative offsets and daily minimal temperatures being in average 1.33 ± 1.62 °C cooler than at the
control stations (Figure 4.6c).

Figure 4.6 - Histograms displaying averaged temperature offset values of daily minimum temperatures
obtained on 39 ‘urban reference – rural reference’, ‘near-ground – rural reference’ and ‘near-ground – urban
reference’ site comparisons.

[Link] Effect of the environment structure on temperature offset

The daily UHI alone was positively related to the offset between near-ground and rural reference
temperatures (Table 4.3). warming is therefore exacerbated under strong UHI conditions at the near-
ground level. Contrastingly, when near-ground and urban reference temperatures are compared,
we observe an opposite link, with cooling capacity prevailing under strong UHI, compared to weaker
conditions (Table 4.3).
Further, all tested variables describing the environment structure were found to relate to the
offsetting capacity of urban grasslands. The interaction between these variables and the daily UHI
was also significant for most variables. The median daily UHI value during the whole study period
was 3°C. We therefore considered this value as a threshold to categorize weak (< 3°C) and strong (>
3°C) UHI conditions.

171
PARTIE IV

Table 4.3 - Effects of environmental variables measured within 2-m-radius buffers (herbaceous canopy density
and height) and within 10, 20 and 50-m-radius buffers (height of trees and buildings) as well as the effect of
daily UHI intensity and its interaction with each previous variable on the daily offsets measured between daily
minimal near-ground temperature and daily minimal atmospheric temperatures, estimated by GLMMs. The
tested offset values were obtained by (A) the difference between daily minimal near-ground temperature and
daily minimal rural reference temperature and (B) the difference between daily minimal near-ground
temperature and daily minimal urban reference temperature. Empty estimate and P-value cells refer to
variables not included in the models. Bold values refer to significant results.
10-m-radius buffer 20-m-radius buffer 50-m-radius buffer
Estimate P-value Estimate P-value Estimate P-value
(A) Difference between near-ground and rural reference temperature

Fixed effects
Sensor type 0.222 0.691 0.289 0.595 0.112 0.830
Mean height of trees 0.218 0.215 0.174 0.180 0.077 0.538
Mean height of buildings 0.791 0.133 0.497 0.044 0.412 0.021
Herbaceous vegetation cover -0.002 0.420 -0.002 0.398 -0.002 0.307
Herbaceous vegetation height 0.055 < 0.001 0.054 < 0.001 0.054 < 0.001
Daily UHI 0.403 < 0.001 0.386 < 0.001 0.402 < 0.001
Daily UHI : Sensor type -0.095 0.002 -0.077 0.011 -0.110 < 0.001
Daily UHI : Mean height of trees 0.038 < 0.001 0.017 0.020 - -
Daily UHI : Mean height of buildings 0.122 < 0.001 0.071 < 0.001 0.069 < 0.001
Daily UHI : Herbaceous vegetation cover - - - - - -
Daily UHI : Herbaceous vegetation height -0.008 < 0.001 -0.008 < 0.001 -0.008 < 0.001
Conditional R² 0.767 0.760 0.761
Marginal R² 0.152 0.160 0.186

(B) Difference between near-ground and urban reference temperature

Fixed effects
Sensor type 2.177 < 0.001 2.162 < 0.001 -3.096 < 0.001
Δ Mean height of trees 1.036 < 0.001 0.993 < 0.001 -0.010 0.741
Δ Mean height of buildings 1.273 0.002 1.757 0.005 1.639 < 0.001
Herbaceous vegetation cover 0.005 0.270 0.009 0.035 0.008 0.038
Herbaceous vegetation height 0.005 0.350 0.011 0.065 0.014 0.028
Land cover atmospheric sensor 0.767 0.366 1.150 0.377 -1.983 0.067
Daily UHI -0.178 0.002 -0.086 0.278 -0.012 0.889
Daily UHI : Sensor type -0.216 < 0.001 -0.146 < 0.001 -0.108 0.002
Daily UHI : Δ Mean height of trees 0.025 0.003 -0.007 < 0.001 - -
Daily UHI : Δ Mean height of buildings 0.024 0.150 0.045 < 0.001 0.109 < 0.001
Daily UHI : Herbaceous vegetation cover -0.001 0.097 -0.002 0.047 -0.002 0.018
Daily UHI : Herbaceous vegetation height -0.008 < 0.001 - - -0.008 < 0.001
Daily UHI : Land cover atmospheric sensor - - - - -0.060 0.066
Conditional R² 0.755 0.755 0.758
Marginal R² 0.194 0.183 0.193

172
 Chapitre 6

The offsetting capacity of grasslands was positively related to the mean height of trees. However,
this relationship solely occurred in a close vicinity (i.e., 10-m, 20-m, but not 50-m radius buffers) and
was more pronounced under strong UHI conditions than under weak UHI conditions ( Figure 4.7).

Figure 4.7 - Relationships between temperature offset and mean tree height calculated within 10-m and 20-m
radius buffers. Top plots display the relationships involving offset values calculated as the difference between
daily minimal near-ground and rural reference temperature (A). Bottom plots display the relationships involving
the temperature offset values calculated as the difference between daily minimal near-ground and urban
reference temperature (B). Relationships under strong UHI conditions are depicted in red (daily UHI ≥ 3°C),
whereas relationships under weak UHI conditions are in blue (daily UHI < 3°C). Significant interaction between
UHI conditions and the vegetation structure are displayed by ‘*’ (0.01 < P-value < 0.05), ‘**’ (0.001 < P-value <
0.01) or ‘***’ (P-value < 0.001).

173
PARTIE IV

The offsets between near-ground and both atmospheric measurements (i.e., rural and urban
reference) were also positively related to the mean height of buildings. The relationships were
significant at all tested spatial scales, except at the 10-m scale regarding offsets between near-
ground and urban reference temperature. All significant relationships observed along with building
height were most pronounced under strong than under weak UHI conditions (Figure 4.8).

Figure 4.8 - Relationships between temperature offset and mean building height calculated within 10-m, 20-m
and 50-m radius buffers. Top plots display the relationships involving offset values calculated as the difference
between daily minimal near-ground and rural reference temperature (A). Bottom plots display the relationships
involving the temperature offset values calculated as the difference between daily minimal near-ground and
urban reference temperature (B). Relationships under strong UHI conditions are depicted in red (daily UHI ≥
3°C), whereas relationships under weak UHI conditions are in blue (daily UHI < 3°C). ‘***’ indicates a significant
interaction between UHI conditions and the building height (P-value < 0.001), whereas ‘NS’ a non-significant
interaction.

174
 Chapitre 6

The offsets between near-ground and both atmospheric measurements (i.e., rural and urban
reference) were negatively related to the variables describing the herbaceous vegetation structure,
namely height and cover. These negative relationships were the most pronounced under strong UHI
conditions (Figure 4.9)

Figure 4.9 - Relationships between temperature offset and mean herbaceous vegetation height (left plots) and
relationships between temperature offset and cover of herbaceous vegetation (right plots). Top plots display
offset values calculated as the difference between daily minimal near-ground and rural reference temperature
(A). Bottom plots display offset values calculated as the difference between daily minimal near-ground and
urban reference temperature (B). Temperature offset values were calculated as the difference between daily
minimal near-ground temperatures and daily minimal reference temperatures. Relationships under strong UHI
conditions are depicted in red (daily UHI ≥ 3°C), whereas relationships under weak UHI conditions are in blue
(daily UHI < 3°C). Significant interaction between UHI conditions and the vegetation structure are displayed by
‘*’ (0.01 < P-value < 0.05) or ‘***’ (P-value < 0.001), whereas ‘NS’ a non-significant relationship.

175
PARTIE IV

4.2.4 Discussion
Daily minimum temperatures reported by urban reference sensors decreased much less than daily
minimum temperatures which were in average 1.89 °C cooler at the rural reference site, confirming
the occurrence of an atmospheric UHI in Rennes during the study period. Urban grasslands had the
potential to offset this UHI-induced nocturnal warming in micro-habitats, with a daily minimal
temperature in average 1.33 °C cooler than the urban reference temperature. Some urban
grasslands even provided cooler conditions than the rural reference site, although daily minimum
temperature remained in averaged slightly warmer. Grassland capacity to offset atmospheric
warming became more pronounced under intense UHI conditions, which suggests that the
herbaceous micro-habitats could play an important ecological role as refugia for organisms during
extreme nocturnal warming.
Providing cool nocturnal thermal refugia within cities might offer opportunities for species to seek
resting temperatures and avoid long-term fitness costs induced by nocturnal warming (Rutschmann
et al. 2024). The consideration of the UHI intensity in our analysis enabled us to compare the cooling
potential of grasslands under contrasting weather situations. Indeed, intense UHI mostly occur
during clear-sky nights with low wind velocity (Morris et al. 2001). We observed that cooling
amplitude within grasslands’ herbaceous canopy is particularly high under strong UHI conditions.
This might be due to the absence of clouds that could absorb long-wave radiations from the ground
and return it to the surface, or to the absence of wind which prevent air mixing and favor the thermal
vertical stratification of air masses (D’Odorico et al. 2013). Even when temperatures were compared
to the rural reference station, urban grasslands had the potential to entirely offset UHI. However,
such extreme offsetting values were only reached in grasslands with particular structural properties.
Green infrastructures are key urban features to maintain biodiversity within cities (Aronson et al.
2017). Yet, our results show that their structural organization can be determinant for the quality of
the thermal micro-habitat they provide to the organisms they host. In particular, we found UHI
mitigation by near-ground nocturnal cooling in grasslands to be favored by the openness of the tree
and built canopy surrounding micro-habitats. These results are in accordance with previous
outcomes from forest microclimates, showing that tree canopy cover increases daily minimum
temperature and thus favors nighttime warming (Li et al. 2015, Zellweger et al. 2019). Forested areas
generally have a lower albedo than grasslands, increasing daytime heat storage by solar radiation
(Peng et al. 2014). Further, during the night tree canopies tend to inhibit surface radiative cooling
from the ground because they absorb a part of the long-wave radiation and return it back to the
ground, favoring nocturnal warming near the surface (Chen et al. 1993, Grimmond et al. 2000,
D’Odorico et al. 2013). If nocturnal warming of the near-ground air under tree canopies is a well-
documented phenomenon in rural surrounding (Bader et al. 2007, Renaud and Rebetez 2009, Maher
et al. 2022), our study brings evidences that this phenomenon also occurs in urban green
infrastructures. As for trees, we show that the mean building height in the surrounding of the studied
micro-habitats favor nocturnal warming near the ground. This was expected, since the radiative and
thermal properties of the construction materials that make up buildings provide particularly heigh
capacities to absorb diurnal solar radiations, retain the produced heat, and release it at night (Oke
et al. 2017). Besides, the 3D geometrical configuration of urban buildings is widely acknowledged as
playing an important role in regulating heat loss through longwave radiation (Unger 2004). The close
vicinity of high buildings to micro-habitats may restrain the share of open-sky seen from the ground,

176
 Chapitre 6

creating a “canyon” effect (Nunez and Oke 1977) which restrict longwave radiation and resulting
cooling effects (Unger 2004).
At the micro-habitat scale, we found that the structure of the herbaceous canopy in the direct vicinity
of the micro-habitats affected temperature offsets in the opposite way as did tree and built canopies.
This may be due to diurnal shading effects, limiting the amount of solar radiations received by the
ground beneath highly and densely grown herbaceous canopy, compared to surfaces covered by
shorter and sparser vegetation (D’Odorico et al. 2013). More intense ground-heating may be
followed by higher nocturnal upwelling long-wave radiation in shortly or sparsely grown vegetation,
which favors nocturnal warming (He et al. 2010).
Grasslands represent an important part of urban green infrastructures and can be subject to a wide
range of management regimes, whereby the height and the frequency of mowing can broadly vary
across a single city. Understanding how management operations affect grassland microclimates can
enable city councils to adapt practices in order to provide high-quality thermal micro-habitats to city-
dwelling organisms, by improving climate-adapted green spaces (Gardiner and Hassall 2009, Thomas
et al. 2009). Based on our results, we believe that preserving patches of open grasslands, i.e., non-
forested areas, within urban green infrastructures networks can help to provide nighttime cooling
refugia. This recommendation should not be understood as undermining the importance of urban
trees, which play a key role in reducing solar radiation on organisms through shading during the day
(Li et al. 2015) and play a crucial role in mitigating the atmospheric UHI, and therefore improving
human well-being (Armson et al. 2012). We argue that, besides wooded areas, open grasslands are
efficient tools for mitigating the UHI-effect at an ecologically relevant scale. We provide further
evidence that this potential can be enhanced by maintaining high and dense herbaceous canopies,
which can be easily achieved with minimal mowing intensity, e.g., once a year to prevent the growth
of woody vegetation (Chollet et al. 2018). If measures aiming to cool urban conditions are likely to
be beneficial for most diurnal organisms requiring tempered nocturnal resting conditions
(Rutschmann et al. 2024), they mays also restrain counterparts that benefit from increased minimal
temperature. For example, nocturnal warming may increase foraging activity and physiological rates,
which has been documented to have positive effects on multiple insect species (Ma et al. 2020).
Such positive responses to warming are however likely to have possible subsequent alterations in
biotic interactions, and imbalance in ecosystems. Increased activity due to nocturnal warming has
notably been reported to favor dispersal outbreaks in the pine processionary moth (Thaumetopoea
pityocampa) (Battisti et al., 2005), which might be particularly important to prevent in urban areas.
This study shows that a common type of urban green infrastructure has the potential to mitigate
warming induced by the UHI phenomenon, at the scale of the organisms it hosts. In line with previous
microclimatic research, we confirm that considering the ecological relevant scale is crucial to predict
organism responses to climatic stimuli. The analyze of grasslands’ structural parameters driving
temperature offsets near to the ground provide new mechanistic understandings of urban
microclimates, with implications for urban biodiversity conservation through management and
design of green infrastructures. Near-ground microclimates should be considered by local authorities
and urban planners to design and implement climate-smart green infrastructures within cities, that
could help to counteract the global negative impact of urbanization on biodiversity. Future
examinations could compare whether the differences reported in our study are generally applicable
to cities, or whether they depend on the background climate. This approach could open up new
perspectives in terms of urban biodiversity conservation.

177
PARTIE IV

178
 Discussion générale

5 Partie V : Discussion générale

L’objectif général de cette thèse est de déterminer si la faune ectotherme, étudiée par le prisme des
arthropodes du sol, est affectée par les variations spatiales de température dans le contexte du
réchauffement urbain mesuré sur la ville de Rennes et ses alentours. La zone atelier Armorique, dont
la ville de Rennes fait partie, s’est avérée être particulièrement adaptée pour la conduite d’une étude
pluridisciplinaire, à l’interface entre climatologie et écologie urbaine. Sur la base de données de
températures obtenues à l’aide d’un double réseau de capteurs dense et d’un échantillonnage
robuste des arthropodes du sol, nous avons montré que le réchauffement atmosphérique urbain est
négativement relié à la diversité des espèces au sein de communautés (Chapitre 1).
Face à ce constat, nous avons testé l’hypothèse selon laquelle le patron de diversité observé à
l’échelle des communautés serait dû à un filtre environnemental exercé par l’ICU, excluant certaines
espèces particulièrement vulnérables de par leurs caractéristiques écologiques. L’étude de la
distribution de trois traits fonctionnels au sein des communautés d’araignées sur 36 prairies et
pelouses montre que l’affinité thermique moyenne est plus élevée au sein de communautés
échantillonnées sur des sites urbains, que sur des communautés d’origine rurale. La taille corporelle
moyenne des espèces quant à elle diminue en ville, par rapport aux communautés rurales. L'ICU à
l'échelle du paysage, et non la température des micro-habitats, filtre les espèces en fonction de leurs
traits fonctionnels. L'ICU peut donc être considéré comme une barrière thermique, filtrant les
espèces en fonction de leur capacité physiologique à faire face aux conditions thermiques urbaines
(Chapitre 3).
Des résultats complémentaires sur les tailles des araignées indiquent que la prise en compte des
traits d'histoire de vie spécifiques aux espèces aide à comprendre les réponses contrastées entre
espèces. La prise en compte de ces traits peut aider à identifier des patrons généraux de diversité.
En effet, chez les espèces présentes en ville, nous montrons des amplitudes de réponses contrastées
entre les Lycosidae et les autres espèces. Ces résultats indiquent que des espèces de relativement
grande taille, peu dispersives et univoltines sont susceptibles d’être plus fortement contraintes par
la température que les petites espèces bivoltines, qui peuvent disperser par voie aérienne
(ballooning) (Chapitre 4).
A la lumière de ces résultats, une étape supplémentaire a été d’examiner par quelle stratégie des
espèces, a priori particulièrement sensibles aux variations de températures, peuvent se maintenir
en conditions urbaines. Une stratégie particulièrement courante chez les ectothermes est la
thermorégulation comportementale. Cela consiste à maintenir sa température corporelle proche de
son optimum physiologique en se déplaçant au sein d’une gamme particulière de températures
disponibles dans le micro-habitat, incluant les températures préférées (Tpref). Nous avons montré
qu’une Lycosidae commune (Pardosa saltans), prélevée sur un gradient d’ICU, thermorégule très
précisément par le comportement, et pourrait dès lors utiliser cette stratégie pour atténuer la
contrainte environnementale exercée par le réchauffement (Chapitre 5).

179
Discussion générale

Dans le Chapitre 1, nous avons également identifié un lien négatif entre les températures mesurées
à l’échelle du microclimat (donc ressenties par les arthropodes) et le nombre d’individus piégés. La
thermorégulation comportementale utilisée par certaines espèces (Chapitre 5), au sein des
communautés, pourrait expliquer que les micro-habitats soumis au réchauffement soient « évités »,
donc moins fréquentés par les individus d’araignées, que les micro-habitats plus « frais ».
Les conditions climatiques au sein des micro-habitats semblent donc décisives pour atténuer les
effets du réchauffement dus à l’ICU et favoriser le maintien de certaines populations d’arthropodes
en ville. La capacité du micro-habitat à atténuer les effets du réchauffement urbain peut être
modulée par la structure de la végétation herbacée, ou encore la structure des infrastructures
situées aux alentours des micro-habitats (arbres, bâtiments) (Chapitre 6). Dès lors, intégrer la qualité
thermique des micro-habitats dans les stratégies de gestion et la conception des espaces verts
constituerait une opportunité nouvelle pour la conservation de la biodiversité urbaine.
Une synthèse de ces principaux résultats est représentée graphiquement par la Figure 5.1. Au cours
de cette discussion, je confronte dans un premier temps les résultats obtenus dans les six chapitres
constituant cette thèse. Plus précisément, je discute (1) les mécanismes expliquant le lien négatif
observé entre ICU et diversité spécifique, (2) le lien possible entre le comportement de
thermorégulation et le nombre réduit d’individus relevés dans les micro-habitats « chauds », et (3)
l’importance de la qualité thermique des micro-habitats pour le maintien de la diversité. Au cours de
cette partie, je mentionne également différents leviers possibles pour atténuer l’impact du
réchauffement sur la biodiversité des arthropodes en ville. Dans un second temps, j’expose diverses
perspectives de recherche issues de ce travail.

180
 Figure 5.1 - Synthèse des principaux résultats issus des différents chapitres de la thèse.

181
Discussion générale
Discussion générale

5.1 Expliquer les patrons observés à l’échelle des communautés

5.1.1 L’ICU atmosphérique filtre les espèces selon leurs traits fonctionnels
Un des principaux apports de cette thèse est d’avoir pu relier la variabilité spatiale de l'ICU
atmosphérique avec le niveau de diversité taxonomique au sein des communautés d’araignées,
comme détaillé dans le Chapitre 1. De plus, les résultats du Chapitre 4 ont montré que différentes
espèces issues de ces communautés sont plus ou moins contraintes par l’intensité du réchauffement
lié à ICU, et que le niveau de la contrainte exercé par cette variable peut s’expliquer par les traits
fonctionnels qui caractérisent ces espèces. Face à ces constats, nous avons fait le choix d’étudier la
composition des communautés, en particulier sur la base de ces traits. En effet, les baisses de
diversité taxonomique en réponse à des intensités d'ICU atmosphérique élevées suggèrent
l'existence d'un processus de filtrage de certaines espèces face à des conditions environnementales
particulières. Des études récentes ont montré que diverses espèces d’arthropodes sont filtrées par
le réchauffement urbain, qui peut donc agir comme une « barrière thermique » à la périphérie de la
ville, sous l’effet de laquelle seules les espèces physiologiquement adaptées aux conditions
thermiques chaudes persistent (Hamblin et al. 2017, Piano et al. 2017).
Pour étudier les capacités physiologiques des espèces, nous avons estimé les affinités thermiques
de l’ensemble des espèces qui composent les communautés échantillonnées, sur la base de leur
distribution géographique à l’échelle de l’Europe et de l’Afrique du nord. Plus précisément, nous
avons estimé l’affinité thermique des espèces en calculant la moyenne des températures (minimale,
maximale et moyenne) de l’ensemble des états dans lesquels elles sont recensées. Cette méthode,
largement utilisée dans les études s’intéressant à la « thermophilisation » de la faune ou de la flore
(Khaliq et al. 2024), permet d’obtenir un proxy de trait thermique sans nécessiter la réalisation de
mesures phénotypiques. Pour cela, la méthode se base sur l’hypothèse selon laquelle les espèces
distribuées sur des zones géographiques « chaudes » sont plus thermophiles que les espèces
distribuées sur des zones plus « froides ». Nos résultats ont montré que l'augmentation de l'affinité
thermique moyenne de la communauté dans le groupe urbain résultait d'une proportion accrue
d'espèces thermophiles, combinée à une proportion réduite d'espèces associées à des températures
intermédiaires. Nous avons par ailleurs constaté une absence d'augmentation absolue des espèces
thermophiles dans les sites appartenant au groupe urbain, ce qui montre que le changement de trait
fonctionnel observé vers des valeurs de températures plus élevées est dû à un processus de filtrage
excluant les espèces sensibles à la chaleur dans les zones urbanisées. L'analyse détaillée de l’effet
des variables environnementales a montré que ce filtrage fonctionnel est lié à l'intensification des
conditions d’ICU atmosphérique.
Le Chapitre 3 nous permet donc de conclure qu’une différence d'affinité thermique moyenne entre
les communautés urbaines et rurales existe, et résulte d’un processus de filtrage fonctionnel des
espèces en réponse à l’intensité d’ICU. Ce filtrage peut expliquer la diminution de la diversité
taxonomique observée dans le Chapitre 1. Cette méthode s’est avérée efficace pour identifier des
patrons biogéographiques de composition fonctionnelle à l’échelle de la ville et les relier aux patrons
globaux de diversité. La distribution géographique des araignées européennes étant renseignée de
manière complète dans la littérature, l’approche utilisée nous a permis de calculer des indices de
composition fonctionnelle des communautés, en incluant l’ensemble des espèces du jeu de
données. Une perspective intéressante serait la prise en considération de traits thermiques issus de

182
 Discussion générale

mesures directes. Une telle approche pourrait s’avérer plus exacte pour définir le degré de
« thermophilie » d’une espèce. Cependant, des données de traits thermiques (par exemple la
CTmax2 ou la Tpref3) ne sont disponibles que pour une partie infime des espèces qui composent les
communautés d’araignées échantillonnées dans notre étude. Pourtant, une analyse fine de ces traits
biologiques pourrait permettre d’identifier des seuils de tolérance thermique permettant
l’implantation des espèces en ville (Hamblin et al. 2017).
Si la diminution de la diversité en conditions intenses d’ICU s’explique par l’exclusion d’espèces
présentant des traits fonctionnels particuliers, il semblerait logique d’observer une relation négative
entre diversité fonctionnelle au sein des communautés et intensité de l’ICU, en raison d’une
homogénéisation. Or, les résultats du Chapitre 1 indiquent que la diversité fonctionnelle est
uniquement reliée aux caractéristiques du micro-habitat. Cela pourrait s'expliquer par l'absence de
traits fonctionnels relatifs à la thermobiologie des espèces dans l'ensemble des traits sélectionnés
pour le calcul des indices de diversité fonctionnelle. L’ajout de traits thermiques décrivant la
tolérance (par exemple CTmax) ou la préférence (Tpref) dans le jeu de données pourrait permettre
le calcul de métriques de diversité fonctionnelles sensibles à l’intensité de l’ICU.
Par ailleurs, la capacité de dispersion aérienne (ballooning) ne ressort pas comme étant une
caractéristique écologique liée au filtrage des espèces dans le Chapitre 3. Ce résultat peut s’expliquer
par le fait que ce trait soit codé selon la capacité « potentielle » à ballooner dans les bases de
données. Dès lors, bon nombre d’espèces se sont vu décrites comme « capables » de ballooner, alors
qu’elles n’utilisent dans les faits cette méthode de dispersion que très rarement. Par exemple,
certaines Lycosidae, dont Pardosa pullata et Pardosa prativaga étudiées dans le Chapitre 4, sont
capables de ballonner au stade juvénile, mais deviennent trop lourdes au stade adulte (Simonneau
et al. 2016). Effectuer des tests en conditions contrôlées (par exemple Weyman 1995) afin d’obtenir
des mesures de probabilité d’observation du comportement de ballooning chez les espèces pourrait
permettre le calcul de métriques fonctionnelles plus sensibles aux conditions environnementales
que la variable booléenne utilisée dans cette thèse.
Pour finir, il est intéressant de mentionner le rôle important que semblent jouer les espèces peu
représentées en termes de nombre d’individus piégés pour expliquer les patrons de diversité
observés sur les communautés d’araignées. Dans le Chapitre 1, nous avons pu constater que seuls
les indices de diversité d’ordre Q0 répondent aux changements de conditions environnementales.
Ces indices correspondent au nombre d’espèces (ou richesse spécifique) dans le cas de la diversité
taxonomique, ou encore au nombre de « groupes fonctionnels » dans le cas de la diversité
fonctionnelle (il s’agit alors du nombre de combinaisons de traits observés). Ces résultats indiquent
que les changements de diversité observés sont dus aux espèces dites « rares », car étant
représentées par un faible nombre d’individus piégés. Cette observation explique par ailleurs que
seules les métriques de composition fonctionnelle non pondérés par le nombre d’individus piégés
répondent aux variables environnementales dans le Chapitre 3. Ces précisions permettent de
rappeler que les patrons de diversité et de composition discutés dans cette partie résultent d’une
faible proportion de la somme totale des individus piégés.

2
Critical Thermal Maximum : Température critique maximale
3 Preferred Temperature : Température préférée
183
Discussion générale

Encadré 1 - Penser l’atténuation du réchauffement atmosphérique à l’échelle de la ville.

Les résultats du Chapitre 1 indiquent qu’atténuer l’intensité de l’ICU atmosphérique pourrait

favoriser la diversité spécifique d’arthropodes dans les pelouses et prairies urbaines. A l’échelle
d’une ville, augmenter la densité d'espaces verts représente l'un des moyens les plus efficaces
pour atténuer le réchauffement (Emmanuel and Loconsole 2015, Lai et al. 2019). Cette stratégie
est cependant largement limitée par l’espace disponible en ville, souvent déjà fortement densifié.
Face à ce constat, l’optimisation du réseau d’infrastructures vertes incluant tous les espaces verts
naturels, semi-naturels et artificiels ayant des fonctions écologiques au sein et aux alentours de
la ville (Norton et al. 2015) est un levier possible pour favoriser l’atténuation de l’ICU. Pour cela,
certaines études de modélisation examinent par exemple la possibilité d’identifier des corridors
de refroidissement à l'échelle d’agglomérations entières, basées sur des données d’occupations
du sol et de géométrie urbaine (Wu et al. 2020). Des corridors de refroidissement constitués
d’infrastructures vertes ont la capacité d'abaisser leur propre température et celle des zones
environnantes, mais aussi de ventiler l'îlot de chaleur (Wu et al. 2020). Il est dès lors envisageable
que de tels corridors remplissent un rôle écologique permettant l’entrée de certaines espèces et
le maintien de leurs populations dans les espaces urbains, desquels elles étaient jusqu’alors
exclues, en raison de leur tolérance limitée aux températures élevées.

Les exemples de leviers décrits ici pour favoriser la biodiversité animale urbaine à l’échelle de
l’habitat, de l’espace vert ou de la ville entière s’inscrivent tous dans une démarche plus globale
d’adaptation aux changements climatiques par le biais de mesures basées sur les écosystèmes
(Jones et al. 2012). Le concept de ces mesures est d’utiliser la biodiversité et les services
écosystémiques qu’elle génère pour s’adapter aux effets néfastes du changement climatique. En
milieu urbain, où le réchauffement est particulièrement intense, ces mesures englobent aussi
bien le développement d’infrastructures vertes dédiées à l’agriculture urbaine, aux nouveaux
espaces verts comme les toitures et façades végétalisées, aux bassins d’orage, ou encore à la
valorisation des espaces verts existants (McVittie et al. 2018). Nous venons de discuter des effets
positifs que la mise en œuvre de mesures basées sur les écosystèmes pourrait avoir sur les
arthropodes. En plus des effets positifs sur la biodiversité animale étudiée dans le cadre de cette
thèse, de nombreux autres groupes taxonomiques pourraient être soutenus par les mesures
d’adaptations basées sur les écosystèmes. La création de nouveaux habitats dont la conception
favorise les conditions microclimatiques, mais encore considère le cycle de vies d’espèces cibles
(concept ‘Animal-Aided-Design’; Weisser and Hauck 2017) peut aussi bien favoriser l’implantation
des invertébrés, que des reptiles, amphibiens, mammifères ou oiseaux.

184
 Discussion générale

5.1.2 La thermorégulation comportementale (i.e., la mobilité) permet

l’adaptation aux variations de température dans les micro-habitats
Comme discuté dans la première partie de cette discussion, certaines espèces présentes à
l’extrémité « froide » du gradient d’ICU sont exclues des communautés échantillonnées dans les
milieux les plus chauds, dès lors que les conditions de réchauffement s’intensifient. L’étude des
variations intraspécifiques de la taille corporelle décrite dans le Chapitre 4 indique que certaines
espèces, bien que représentées sur l’ensemble du gradient d’ICU (en l’occurrence les Lycosidae
Pardosa pullata et Pardosa prativaga), sont également contraintes par les températures. De plus,
lors de notre étude sur les températures préférées chez une autre Lycosidae (Pardosa saltans)
(Chapitre 5), nous n’avons pas observé de réponse du trait thermique étudié à l’ICU atmosphérique
chez les adultes. Nous interprétons ce résultat comme étant dû à un comportement de
thermorégulation prononcé chez cette espèce. Ce résultat suggère que les conditions thermiques à
l’échelle du micro-habitat sont déterminantes pour le maintien de ces espèces. Comme mentionné
dans le Chapitre 4, nous rappelons ici que les Lycosidae ne balloonent que peu ou pas au stade adulte
(Simonneau et al. 2016). Dès lors, il est probable que ces espèces tirent profit de la variabilité
thermique de leur micro-habitat pour compenser leur faible capacité à se déplacer sur de longues
distances.
Ces résultats issus des Chapitres 4 et 5 nous permettent de proposer une explication à la relation
négative observée à l’échelle des communautés dans le Chapitre 1 entre le nombre d’individus
piégés, et la température proche du sol. En effet, les micro-habitats « chauds » peuvent être
« évités » par les espèces capables de thermoréguler par le comportement. Il est important de
rappeler que la variabilité thermique des micro-habitats mesurée dans le Chapitre 1 décrit un
réchauffement nocturne (basé sur les températures minimum journalières). Cela suggère que
« l’évitement » d’environnements thermiques « chauds » peut être déterminé par l’environnement
thermique nocturne, s’il présente par exemple des températures inadaptées à la récupération
physiologique pour les espèces diurnes (Rutschmann et al. 2024). Presque un quart (24,6 %) des
individus piégés et analysés dans le Chapitre 1 sont des Lycosidae. Dès lors, un évitement, même
partiel, des micro-habitats les plus chauds par les individus de cette famille pourrait expliquer le
déficit en nombre d’individus piégés. Cette hypothèse est notamment renforcée par le résultat des
analyses d’activité-densités réalisées individuellement pour chacune des 29 espèces les plus
présentes dans le Chapitre 3. Ces résultats montrent que la température proche du sol est
négativement reliée au nombre d’individus de Lycosidae comme Pardosa pullata, Alopecosa
pulverulenta ou Trochosa terricola, mais aussi à d’autres espèces, y compris certaines capables de
ballooner (Oedothorax fuscus, Oedothorax retusus, Pachygnatha clercki, Pachygnatha degeeri). Nous
montrons dans le Chapitre 5 que le comportement de thermorégulation est caractérisable chez une
Lycosidae, mais il est probable que d’autres espèces d’araignées appartenant à d’autres familles,
observent un comportement similaire (par exemple chez les Salticidae ; Brandt et al. 2020), ce qui
renforce l’idée que la capacité de thermoréguler par le comportement puisse être à l’origine de la
baisse du nombre d’individus piégés, observée dans le Chapitre 1. Les relations logiques telles que
décrites entre les différents Chapitres de la thèse sont représentées graphiquement dans la Figure
5.2.

185
Discussion générale

Figure 5.2 – Mécanismes expliquant la relation négative entre le nombre d’individus piégés et le réchauffement
du micro-habitat
.

5.1.3 Une gestion extensive des prairies urbaines favorise le maintien de la

diversité
Les deux premières parties de cette discussion se sont intéressées aux relations entre les
communautés d’araignées, et les variations de températures mesurées à deux échelles spatiales
distinctes. D’une part, nous avons pu voir que l’ICU atmosphérique intervient à l’échelle du paysage
urbain et pourrait agir négativement sur la diversité taxonomique en réduisant le nombre d’espèces
dans les communautés par filtrage selon leurs traits fonctionnels. D’autre part, le réchauffement à
l’échelle des micro-habitats est associé à une réduction du nombre d’individus piégés, et ce,
possiblement en raison d’un évitement de ces zones par les espèces capables de thermoréguler par
le comportement. Dès lors, les résultats jusqu’alors abordés suggèrent que les réchauffements, à
l’échelle du paysage, comme à l’échelle du micro-habitat, agissent négativement sur les
communautés.

186
 Discussion générale

Dans ce contexte, la structure des micro-habitats semble pouvoir compenser, au moins

partiellement, les baisses de diversité liées au réchauffement des environnements urbains. Dans le
Chapitre 1, nous avons observé qu’une fréquence réduite de tonte ou de fauche était associée
positivement à la richesse spécifique. De nombreuses études réalisées durant les dernières
décennies ont montré des associations entre la diversité taxonomique des araignées et la gestion
des prairies en contexte agricole. Notre étude révèle que ces résultats en milieux agricoles sont
transposables au contexte urbain. La réduction de la tonte ou de la fauche favorise une végétation
plus diversifiée sur le plan taxonomique, comme sur le plan architectural, ce qui favorise la diversité
des araignées (Bell et al. 2001). Cela peut s’expliquer par le fait que certaines espèces, comme les
Clubionidae, n'utilisent que des plantes vivantes pour tisser leurs toiles, alors que certaines
Araneidae n'utilisent que du matériel végétal sec comme support. De plus, une fréquence de tonte
importante, associée à une évacuation des résidus végétaux entraîne une diminution de la litière, et
par la même occasion une diminution des opportunités pour certaines espèces de s’établir (Maelfait
and De Keer 1990, Baines et al. 1998). Comme montré dans le Chapitre 3, les habitats gérés
intensivement (associées aux sites urbains) favorisent les espèces pionnières, notamment les
Linyphiidae appartenant au genre Erigone. Ces mêmes résultats ont été observés en contexte
agricole par Nentwig (1988) sur des prairies fréquemment fauchées. Cet auteur montre qu’au cours
des années suivant la réduction de la fréquence de fauche, la dominance en espèces passe d’une
quasi exclusivité de Lyniphiidae, pour progressivement inclure les Lycosidae. De plus, une relation
linéaire positive a été observée entre le temps écoulé suite à la réduction de la fréquence de fauche
et la diversité taxonomique (Hassall et al. 1992). Par ailleurs, la disponibilité en proies, comme les
collemboles, peut être réduite après une coupe (Purvis and Curry 1980) ce qui peut entraîner une
hausse de la compétition et le déclin de certaines espèces en réponse aux interventions de gestion.
Bien que les relations entre régime d’entretien et diversité soient complexes sur les prairies agricoles
(Bell et al. 2001), la structure de la végétation est un facteur important (Uetz 1991, Gibson et al.
1992, Morris 2000, Diehl et al. 2013, Nardi and Marini 2021), comme le suggèrent également nos
résultats sur la diversité fonctionnelle et taxonomique en contexte urbain, présentés dans les
Chapitres 1 et 2. La structure des surfaces herbacées sur lesquelles les communautés ont été
échantillonnées durant cette thèse varie par leur hauteur et leur densité. Le gradient de hauteur
s’étend d’une extrémité courte avec des pelouses gérées intensivement à une extrémité haute
comprenant des prairies à faibles fréquences de fauche. Notre design expérimental présente
l’avantage de décorréler partiellement fréquence d’entretien et hauteur de végétation du fait des
hauteurs de coupe, qui peuvent différer d’une quinzaine de centimètres entre le passage d’une
tondeuse et d’un gyrobroyeur agricole.
Une étude de De Keer et al. (1989) a montré que les microclimats diffèrent selon les types de
végétation prairiale, haute ou courte. De plus, ces auteurs montrent que certaines espèces
d’araignées se déplacent en alternant ces deux types de micro-habitats sur une période de 24
heures. Dès lors, les changements microclimatiques induits par une intervention de gestion (tonte
ou fauche) peuvent affecter le comportement de thermorégulation, et induire des changements de
densité d’individus chez les espèces sensibles selon le processus décrit dans la partie précédente. En
effet, une canopée dense de végétation permet d’abriter les organismes des conditions climatiques
défavorables. Dès lors, un raccourcissement immédiat de celle-ci induit une exposition accrue à la
pluie ou au vent, pouvant endommager les toiles ou encore augmenter les risques de dessiccation
(Purvis and Curry 1980, Uetz 1991, Wise 1993). De plus, en conditions diurnes, la part du
rayonnement solaire direct atteignant le substrat augmente considérablement suite à une réduction
de la hauteur de végétation (Purvis and Curry 1980, Bell et al. 2001). Ces résultats relatifs à la
187
Discussion générale

structure de la végétation herbacée, couplés à notre étude sur la thermorégulation

comportementale (Chapitre 5) mettent en évidence le rôle central de la complexité du micro-habitat
sur les assemblages d’araignées. La simplification des micro-habitats dans les prairies et pelouses
urbaines est susceptible d’induire une réduction de l’hétérogénéité thermique des habitats. Cette
baisse de « qualité thermique » peut avoir pour conséquence la nécessité d’intensifier le
comportement de thermorégulation chez les ectothermes (Díaz 1997). Par conséquent, la
répartition spatiale des microclimats au sein des micro-habitats peut influer sur l'énergie nécessaire
à la thermorégulation. L'énergie dépensée durant les déplacements étant un coût inhérent à la
thermorégulation comportementale (Huey and Slatkin 1976, Sears and Angilletta 2015), une baisse
de l’hétérogénéité thermique spatiale peut avoir des répercussions importantes sur les organismes.
Par exemple, elle peut augmenter le risque de prédation et la concurrence pour les ressources
thermiques (Downes and Shine 1998). De plus, le surplus du cout énergétique investit dans les
déplacements sur des distances plus étendues peut induire à son tour une réduction de la viabilité
des populations (Díaz 1997). La structure du micro-habitat détermine donc la vulnérabilité au
changement climatique des organismes vivant au sol, par le biais de ses effets sur le comportement
thermorégulateur. Le lien entre gestion des espaces et réduction de la qualité thermique des micro-
habitats a notamment déjà été identifié comme ayant des implications fortes pour la conservation
d’ectothermes vertébrés (Neel and McBrayer 2018).

188
 Discussion générale

Encadré 2 - Quelles mesures de gestion pour favoriser un climat adapté à la faune des prairies et
pelouses urbaines ?

Les pelouses, c'est-à-dire les prairies monospécifiques, sont des milieux particulièrement pauvres
en biodiversité (Chollet et al. 2018, Norton et al. 2019, Sehrt et al. 2020, Proske et al. 2022, Bauer
et al. 2024). Des études récentes indiquent que les mesures de gestion intensives, telles que les
coupes rases et fréquentes, impactent les arthropodes (Buchholz et al. 2018, Norton et al. 2019,
Proske et al. 2022, Bauer et al. 2024).
Les changements brusques dans la structure de la végétation modifient notamment les
microclimats et réduisent l’accès aux ressources alimentaires et aux opportunités structurelles
pour la prédation et la nidification. Les facteurs clés ayant été identifiés pour promouvoir
l’abondance et la diversité des arthropodes sont l’augmentation de l’hétérogénéité des structures
verticales et horizontales des prairies. Pour cela, divers auteurs ont préconisé des mesures
comme (1) la réduction de la fréquence de tonte à un ou deux passages par an, (2) l’utilisation
des barres de coupe plutôt que des gyrobroyeurs ou tondeuses rotatives et (3) l’exportation les
résidus de fauche pour permettre aux communautés végétales en place de se redévelopper et de
se diversifier après intervention (Wastian et al. 2016, Lerman et al. 2018, Proske et al. 2022).
Vingt-cinq années après la mise en place de mesures de gestion différenciées (Aggéri 2011) à
Rennes, Chollet et al. (2018) ont constaté que la fréquence de fauche influence positivement les
diversités taxonomiques, fonctionnelles et phylogénétiques au sein des communautés végétales.
Les résultats du Chapitre 1 de cette thèse permettent d’étendre ces résultats aux arthropodes
sur la zone d’étude rennaise. En effet, nous montrons une hausse de la diversité en espèces
associées à une réduction de la fréquence de gestion chez les araignées, ainsi qu’une hausse de
la diversité fonctionnelle avec la complexification de la strate herbacée chez les araignées et les
staphylins.
De plus, les résultats du Chapitre 6 montrent qu’une végétation herbacée haute a le potentiel
d’atténuer, voire de compenser le réchauffement dû à l’ICU atmosphérique, à l’échelle du
microclimat ressenti par les arthropodes. La mise en relation des résultats décrits dans les
Chapitres 1 et 6 nous permet de supposer qu’un lien indirect existe entre le mode de gestion et
la diversité des arthropodes, induit par les conditions microclimatiques.
Les recommandations concernant les types de gestions favorisant la faune arthropode rejoignent
celles déjà formulées par Chollet et al. (2018), à savoir, une forte réduction de la fréquence de
tonte appliquée de manière systématique dans les zones utilisées de manière non-intensive par
les citadins. Une date de fauche permettant la repousse de la végétation, avant la fin de la saison
végétative (par exemple au mois d’août) permettrait également de favoriser la disponibilité en
abris hivernaux, nécessaires aux cycles reproductifs de nombreuses espèces (Proske et al. 2022).
Pour finir, la fauche régulière de certains espaces, au sein de larges espaces non fauchés (par
exemple des cheminements) permettrait d’améliorer la qualité thermique du micro-habitat, en
favorisant l’hétérogénéité des températures nécessaire aux ectothermes qui régulent leur
température corporelle par le comportement.

189
Discussion générale

5.1.4 Des micro-habitats complexes favorisent le maintien de la diversité

fonctionnelle
Nos résultats montrent une relation positive entre la complexité de la végétation et la diversité
fonctionnelle chez les araignées (Chapitre 1). Ce résultat est en parfaite adéquation avec les patrons
observés jusqu’alors en contexte rural (Uetz 1991, Gibson et al. 1992, Morris 2000, Entling et al.
2007, Diehl et al. 2013, Nardi and Marini 2021). En effet, les changements de composition et de
diversité au sein des communautés d’araignées sont davantage associés à des changements
structurels de la végétation d’ordre architectural, que de sa composition en espèces végétales (Bell
et al. 2001). Les araignées tisseuses sont notamment particulièrement contraintes par la présence
des structures physiques qui permettent l’implantation de leurs toiles (Uetz 1991, Rypstra et al.
1999). Leur diversité augmente à mesure que la diversité de structures adaptées à la construction
de différents types de toiles augmente également (Bell et al. 2001, Woodcock et al. 2009). Les
chasseuses au sol dépendent quant à elles d’une hétérogénéité de refuges et d’espaces ouverts dans
la végétation herbacée, tels qu’une succession d’agrégats herbeux formés par certaines graminées
et de zones de sol nu (Maelfait and De Keer 1990, Woodcock et al. 2009). Tout comme dans une
étude récente menée sur des habitats prairiaux en contexte rural (Lyons et al. 2018), les chasseuses
au sol dans les prairies et pelouses urbaines que nous avons étudiées sont dominées par des espèces
appartenant au genre Pardosa. Ces espèces utilisent différentes composantes de la structure de la
végétation pour la capture des proies, l'hivernage et le refuge contre les prédateurs (Maelfait and
De Keer 1990, Lyons et al. 2018).
Concernant les staphylins, peu d’études ont jusqu’alors mis en relation la structure de la végétation
herbacée avec la diversité fonctionnelle. Ce taxon regroupe pourtant des espèces aux habitudes
alimentaires hautement diversifiées, dont certaines sont des spécialistes de micro-habitats aux
structures particulières. Par exemple, certains staphylins dépendent de la présence de bois mort ou
de carcasses (les décomposeurs), ou encore de la présence de champignons dans le cas d'espèces
fongivores (Irmler and Lipkow 2018). Tout comme chez les araignées, nos résultats révèlent que leur
diversité en termes de nombre d’espèces est favorisée par la complexité de la végétation (Chapitre
2). Les traits fonctionnels considérés dans notre étude ne semblent cependant pas être en lien avec
la structure de la végétation car aucune relation entre diversité fonctionnelle et hauteur ou densité
de végétation n’a été détectée. La non prise en considération de leurs capacités de dispersion en
raison du manque d’informations dans les bases de données fonctionnelles pourrait expliquer cette
absence de relations. La compilation de données fonctionnelles relatives aux capacités de dispersion,
ainsi qu’au affinités thermiques, chez les espèces de staphylins rapportées dans notre étude,
pourrait permettre une caractérisation plus fine du mécanisme de filtrage environnemental lié à la
structure de micro-environnement, observé dans le Chapitre 2.
Les résultats des chapitres que nous confrontons dans la première partie de cette discussion mettent
tout d’abord en évidence le fait qu’il est possible de décorréler partiellement les variations de
températures dues au réchauffement urbain et les variables liées à la modification de l’habitat. En
effet, l’analyse des corrélations spatiales entre l’ICU atmosphérique et la proportion de surface
construite conduite à l’aide de fenêtres glissantes de 25 ha (ou 25 pixels de 100 x 100 m) nous a
permis d’identifier à une résolution de 100 m des zones de « décorrélation locale » entre les deux
variables. La sélection de nos 36 sites d’étude sur ces espaces préalablement identifiés a abouti à
une décorrélation partielle à l’échelle de la zone d’étude entière. Cette décorrélation, bien que
partielle, nous a permis de tester l’effet des variables liées aux températures et liées à l’habitat dans

190
 Discussion générale

les mêmes modèles, tout en garantissant des coefficients de corrélation entre les prédicteurs qui
soient sous le seuil communément considéré de 0.7 (Dormann et al. 2013). De la même manière,
nous avons observé des corrélations inférieures au même seuil entre les variables décrivant la
structure de la strate herbacée et la température à l’échelle du micro-habitat. Nous interprétons
cette décorrélation partielle comme résultant de la configuration des infrastructures (bâtiments,
arbres) à proximité des sites (Chapitre 6).
L’échelle macro à laquelle le climat est communément mesuré (échelle kilométrique, température à
2 ou 3 m du sol) peut donc être considérée comme inadaptée pour l’étude d’organismes
ectothermes de taille réduite (1 cm ou moins), qui ressentent les conditions climatiques à l'échelle
du centimètre ou du mètre (Potter et al. 2013, Pincebourde and Woods 2020). L’échelle de 100 m
considérée tout au long de cette thèse pour caractériser le gradient d’ICU atmosphérique à l’échelle
de la ville de Rennes et de ses alentours apparaît comme plus pertinente pour expliquer les patrons
de diversité taxonomique chez les araignées du sol. Cette échelle, bien que représentative des
conditions atmosphériques et non du micro habitat (capteurs à 3 m au-dessus du sol) permet de
cartographier les variations spatiales à une résolution suffisamment fine pour identifier le
mécanisme de filtrage des espèces par les conditions thermiques. En effet, nous avons pu observer
dans le Chapitre 3 que les caractéristiques fonctionnelles déterminantes du maintien des espèces en
conditions d’ICU atmosphérique intenses sont 1) une taille corporelle réduite et 2) une affinité aux
températures élevées. En relation avec ces résultats, nous avons observé que certaines araignées
considérées comme « de grande taille » dans notre étude, bien que moins représentées en nombre
d’individus piégés, persistent en milieu urbain à fort ICU. Notre étude portant sur les variations
intraspécifiques de la taille corporelle chez diverses espèces (Chapitre 4) montre que ces « grandes »
espèces sont cependant plus contraintes par des conditions thermiques défavorables que leurs
homologues de petite taille, et donc particulièrement vulnérables aux réchauffements. Enfin, nous
montrons que le comportement de thermorégulation est susceptible d’aider les populations
d’espèces contraintes par la température à se maintenir, en tirant profit des variations thermiques
à l’échelle du micro-habitat. Nous mettons par ailleurs cette stratégie en relation avec les résultats
du Chapitre 1, montrant que la température du micro-habitat est négativement liée au nombre
d’individus, ce qui peut indiquer un évitement local des températures élevées par l’usage du
déplacement.
Au vu du rôle important de la température à l’échelle du micro-habitat démontré par nos études, il
nous a semblé cohérent d’étudier les liens entre températures atmosphériques et températures
proches du sol. L’étude présentée dans le Chapitre 6 met en évidence la capacité du micro-habitat à
atténuer localement, voir à fortement limiter le réchauffement atmosphérique induit par le
phénomène d’ICU. Nous montrons dans ce chapitre que cette capacité d’atténuation est liée à la
configuration du site d’étude. La hauteur moyenne des infrastructures telles que les arbres ou les
bâtiments aux alentours des prairies ou pelouses diminue par exemple la capacité d’atténuation de
l’ICU par le micro-habitat. La hauteur de la végétation herbacée ainsi que sa densité sont quant à eux
des facteurs favorisant l’atténuation du réchauffement proche du sol. Ces premiers résultats sur les
températures à l’échelle des micro-habitats prairiaux fournissent des pistes intéressantes pour
l’ajustement des mesures de gestion et d’aménagement des espaces verts en ville. Cependant,
caractériser la variabilité spatiale à l’échelle fine des micro-habitats et conduire des
expérimentations in-situ de thermorégulation comportementale constituent une prochaine étape
logique, qui devrait permettre de vérifier les interprétations émises sur la base des résultats obtenus
au cours de cette thèse.

191
Discussion générale

5.2 Perspectives de recherche

Les résultats des chapitres que nous confrontons dans la première partie de cette discussion mettent
tout d’abord en évidence le fait que l’ICU atmosphérique, bien que non-représentatif des conditions
thermiques à l’échelle des micro-habitats, décrit les variations spatiales du réchauffement à une
résolution suffisamment fine pour identifier le mécanisme de filtrage des espèces par les conditions
thermiques. Cependant, le rôle important du micro-habitat est également démontré pour atténuer
localement, voir à fortement limiter le réchauffement atmosphérique induit par le phénomène
d’ICU, et permettre aux organismes de thermoréguler. Dans cette seconde partie de la discussion,
je propose diverses perspectives de recherche découlant des résultats de la thèse.

5.2.1 Approfondir les connaissances climatologiques à l’échelle des micro-

habitats
Les températures de surface d'une zone de quelques mètres carrés à un temps T peuvent être aussi
variables que les températures atmosphériques mesurées par les stations météorologiques à
l’échelle du macroclimat (par exemple d'un continent) (Pincebourde and Woods 2020). Si la mesure
ponctuelle des températures proches du sol est une première étape importante pour la
caractérisation de l’environnement thermique ressenti par les organismes, plusieurs études ont
définit l’intégration de la distribution spatiale des températures comme essentielle pour caractériser
la qualité thermique d’un micro-habitat (Sears and Angilletta 2015, Sears et al. 2016). Mesurer les
variations spatiales de température et plus largement du microclimat à l’échelle des micro-habitats
herbacés urbains constituerait dès lors une étape importante pour compléter les résultats obtenus
pendant cette thèse. Pour cela, des perspectives intéressantes seraient de reproduire les conditions
thermiques représentatives des micro-habitats en conditions contrôlées, ou de conduire des
expérimentations directement sur le terrain. Dans les deux cas, les mesures de variation de
températures intra-site pourraient être réalisées à l’aide d’imagerie thermique pour mesurer les
températures de surface, et d’un réseau de thermocouples disposés très proches les uns des autres
(quelques centimètres) pour mesurer la variabilité thermique proche du sol. Nous avons montré
dans le Chapitre 5 que les Lycosidae thermorégulent en se positionnant de manière précise le long
d’un gradient de température linéaire, et donc nécessairement simplifié par rapport aux conditions
réelles. Mesurer les Tpref dans une mosaïque thermique complexe permettrait de valider
l’interprétation selon laquelle la température du micro-habitat joue un rôle important dans les
patrons de densité d’individus, comme observé dans le Chapitre 1. Par ailleurs, il serait également
important de prendre les conditions diurnes en considération. Pour cela, il est primordial de s’assurer
que les caractéristiques de la sonde utilisée (notamment sa taille, sa forme, sa matière ou encore sa
couleur) permettent de mesurer les températures opératives4 représentatives du modèle biologique
étudié (Dzialowski 2005).

4
La température d'un objet inanimé de capacité thermique nulle ayant la même taille, la même forme et les mêmes propriétés
radiatives que l'animal étudié, lorsque exposé au même environnement.
192
 Discussion générale

5.2.2 Valider l’hypothèse d’inertie comportementale

Les gammes de température préférées très étroites mesurées sur notre gradient thermique en
laboratoire (Chapitre 5) suggèrent une thermorégulation comportementale active chez Pardosa
saltans. Une perspective intéressante serait de conduire une expérimentation comportementale en
mésocosme, permettant de reproduire un micro-habitat au sein duquel la distribution des
températures opératives puisse être contrôlée. La gamme des températures préférées de P. saltans
est déjà connue (Chapitre 5). Il serait dès lors possible de démontrer que les individus
thermorégulent, si les températures corporelles mesurées se situent plus souvent dans la gamme
des températures préférées que par chance, au regard de l’ensemble des conditions thermiques
disponibles (Hertz et al. 1993). Dans un second temps, il serait possible de vérifier si la
thermorégulation comportementale est effectivement associée à un changement physiologique plus
limité que celui auquel on s'attendrait en l'absence de comportement régulateur. Les performances
des individus dépendent de la température corporelle chez les ectothermes (Angilletta 2009). Dès
lors, comparer les vitesses de déplacement ou encore la survie ou le succès reproducteur selon
plusieurs conditions thermiques (et donc opportunités de thermorégulation) permettrait de valider
ou d’invalider qu’un effet Bogert contribue à l’absence d’adaptations locales de la température
préférée aux environnements urbains plus chauds (Huey et al. 2003, Logan et al. 2014, Gilbert and
Miles 2017).

5.2.3 Dissocier plasticité phénotypique et adaptations évolutives

Outre les résultats sur la température préférée du Chapitre 5, les travaux menés durant cette thèse
ont permis de mettre en lien des changements phénotypiques avec l’élévation des températures en
ville chez plusieurs espèces d’araignées. Notamment, dans le Chapitre 4 nous avons identifié des
réponses de la taille corporelle aux conditions thermiques chez deux Lycosidae : Pardosa prativaga
et Pardosa pullata. Si cette étude permet d’identifier des relations entre conditions
environnementales et phénotypes des individus par des approches corrélatives, elle ne permet pas
d’établir de lien de causalité entre ces observations. Dès lors, les mécanismes responsables des
réponses phénotypiques observées restent incertains. En écologie évolutive, deux mécanismes
pouvant mener à des changements phénotypiques en réponse aux conditions
environnementales sont communément dissociés : la plasticité phénotypique et l’adaptation
évolutive (Diamond and Martin 2016). Dès lors, dans notre contexte urbain, le fait qu'un phénotype
varie le long d’un gradient environnemental n'indique pas nécessairement que la variation du trait
résulte d’une adaptation évolutive (Donihue and Lambert 2015). En effet, la plasticité phénotypique
peut également intervenir pour permettre aux organismes de maximiser leur valeur sélective 5 (ou
« fitness ») dans un environnement urbain, sans qu’aucun changement génétique ne se produise
(Sommer 2020). Par exemple, la diminution de la taille corporelle des ectothermes dans des
conditions de réchauffement de l’air est un modèle bien établi de plasticité phénotypique (Atkinson
1994).
Pour dissocier lequel de ces mécanismes, plasticité ou évolution, est responsable des changements
phénotypiques observés dans cette thèse, différentes méthodes utilisées de manière standard en
biologie évolutive pourraient être appliquées lors de futures expérimentations (Merilä and Hendry
2014, Donihue and Lambert 2015). Par exemple, les expérimentations de « jardin commun » sont

5 Capacité d’un individu à diffuser ses gènes par la reproduction.

193
Discussion générale

une approche intéressante permettant d’atténuer les effets à court terme de l’acclimatation
plastique des individus à leurs conditions environnementales d’origine. L’approche consiste à
maintenir des individus issus de conditions environnementales distinctes (par exemple prélevés le
long d’un gradient d’ICU) sous les mêmes conditions durant au moins une génération avant de
mesurer leurs phénotypes. Si des différences persistent malgré le traitement « commun », il est alors
probable qu’elles soient d’origine génétique, ce qui indique alors une divergence évolutive entre les
populations (Diamond and Martin 2020). Une seconde expérience envisageable, de terrain cette fois,
serait la transplantation réciproque. Appliquée à un gradient d’ICU, cette approche consiste à mettre
en situations d’ICU intense des individus issus de conditions d’ICU faibles, et vice-versa. La présence
d’une adaptation locale est alors confirmée lorsque l’on observe une valeur sélective plus élevées
chez individus lorsqu’ils sont testés dans leur environnement d’origine, que dans tout autre
environnement (Martin et al. 2021). Ces méthodes permettant de distinguer les réponses plastiques
des réponses évolutives nécessitent la mise en œuvre d’expérimentations plus coûteuses en temps
qu’une mesure des phénotypes le long d'un gradient environnemental. Elle nécessite par ailleurs de
recueillir des individus vivants, ce qui limite le recours au piégeage. Ces contraintes supplémentaires
peuvent expliquer le fait que peu d'études aient jusqu’alors distingué les réponses plastiques des
réponses évolutives face au changements environnementaux d’origine anthropiques (Diamond and
Martin 2020). Pourtant, ce type d’expérimentation serait particulièrement intéressant dans notre
contexte car l’héritabilité des traits morphologiques et physiologiques est en général assez forte,
alors que l’héritabilité de nombreux traits d'histoire de vie est relativement faible (Mousseau and
Roff 1987). Cette observation indique une plus grande probabilité de trouver une adaptation
évolutive en réponse au réchauffement urbain pour les traits morphologiques (par exemple la taille
corporelle) ou physiologiques (par exemple la température préférée) par rapport à d'autres traits.
En contexte urbain, San Martin y Gomez and Van Dyck (2012) rapportent par exemple à l’aide d’une
expérimentation en jardin commun que les femelles du criquet Chorthippus brunneus issues de
conditions urbaines à ICU intense sont plus lourdes que leurs congénères rurales (ce que suggèrent
nos résultats sur Pardosa prativaga dans le Chapitre 4). Le fait que ce constat soit réalisé sur des
individus descendant d’une première génération élevée en conditions communes confirme qu’il
s’agit d’effet héréditaires et donc probablement d’adaptations génétique au réchauffement urbain
- bien qu’il puisse également s’agir d’autre effets héréditaires, comme les effets maternels (San
Martin y Gomez and Van Dyck 2012). De plus, ces résultats obtenus en laboratoire sont complétés
par une étude récente de terrain, montrant des adaptations locales aux villes chez les insectes, dues
à une évolution récente de leurs tolérances thermiques (Martin et al. 2021).
La base génétique de ces adaptations locales peut dès lors être étudiée par le biais d’approches de
génomique des populations, dans le but d’identifier les signatures génétiques de l'adaptation à
l'urbanisation (Savolainen et al. 2013). En utilisant une approche RAD-seq6, Theodorou et al. (2018)
ont par exemple mis en évidence une différenciation génétique entre les populations rurales et
urbaines d’une espèce commune de bourdon. Ces auteurs rapportent qu’une partie des loci7
identifiés présentent une proportion élevée de différenciation génétique et que ces changements
sont fortement associés l’urbanisation, indiquant une divergence adaptative entre les populations.
Ce type d’étude, encore rares sur les ectothermes en milieu urbain, pourraient permettre d’identifier

6La technique de « Restriction site Associated DNA (RAD) tag Sequencing » permet d’identifier et de génotyper des polymorphismes de
séquence d’ADN dans de nombreux échantillons, sans avoir besoin d’un génome de référence.

7 Position fixe d'un gène ou d'un marqueur génétique sur un chromosome.

194
 Discussion générale

les gènes impliqués dans les adaptations locales, notamment dans le contexte du réchauffement
urbain.

5.2.4 Evaluer l’impact du réchauffement urbain sur les interactions biotiques

Bien que de très nombreuses études se soient intéressées jusqu’alors aux effets de l’urbanisation
sur des populations d’arthropodes particulières, les effets de l’urbanisation sur les interactions
trophiques dans lesquelles ces arthropodes sont impliqués restent relativement peu explorées
(Theodorou 2022). Pourtant, les interactions entre espèces peuvent être considérées comme
l'architecture des écosystèmes (McCann 2007). Les espèces d’arthropodes étudiées dans le cadre de
cette thèse sont impliquées essentiellement dans la prédation, ou l'herbivorie dans le cas de certains
staphylins et carabes. Ces interactions jouent un rôle central dans le fonctionnement des
écosystèmes urbains et présentent dès lors un intérêt fort pour la conservation de la biodiversité ou
encore pour l’apport de services écosystémiques (Theodorou 2022). Si une espèce s'adapte, migre
ou s'éteint, les effets en cascade sur les écosystèmes sont dus à ses interactions biotiques.
Les résultats de cette thèse mettent en évidence qu’en réponse au climat urbain, les communautés
d’arthropodes prédateurs en ville, notamment chez les araignées, se distinguent des communautés
rurales en termes de composition et de diversité. Le filtre environnemental induit par les
températures sur ces communautés d’arthropodes prédateurs peut directement ou indirectement
altérer les niveaux trophiques inférieurs et supérieurs dans les écosystèmes urbains. Nous montrons
par exemple qu’une baisse du nombre d’individus prédateurs (araignées) piégés est associée au
réchauffement du micro-habitat. A l’inverse, des études ont montré que les proies herbivores sont
particulièrement abondantes en conditions de réchauffement urbain (Meineke et al. 2013, Dale and
Frank 2014). Selon ces auteurs, ce résultat s’explique par le fait que des conditions de températures
élevées en ville favorisent la survie, la fécondité et l’accroissement de ces populations. Cependant,
une recrudescence de proies herbivores en ville pourrait également être la conséquence d’un
« relâchement » du contrôle biologique, dû (1) à la baisse d’abondance de leurs prédateurs, (2) à
l’exclusions de certaines espèces jouant un rôle de prédation important dans les zones à fort ICU, ou
encore (3) à un décalage phénologiques de l’activité entre les proies et leurs ennemis naturels. Par
exemple, Meineke et al. (2014) ont montré que la température provoque un avancement
phénologique chez des insectes herbivores en conditions urbaines, alors qu’aucun changement n’a
été observé au sein des communautés de parasitoïdes. Les auteurs prédisent que dans le contexte
de leur étude, ce type de décalage phénologique peut favoriser la prolifération d’insectes ravageurs
des arbres en ville. Dans le cadre de cette étude, la prolifération des herbivores a été reliée
directement à la hausse des températures ; et indirectement à l’effet des niveaux trophiques
supérieurs (ou effet « top-down ») que provoque le relâchement du contrôle biologique sur une
fenêtre temporelle donnée.
En plus de ces effets, le réchauffement en ville peut également induire de nombreux effets
provenant de niveaux trophiques inférieurs (ou effets « bottom-up »). En effet, des changements sur
la végétation peuvent induire des réactions en chaîne sur plusieurs niveaux trophiques supérieurs
(Burghardt et al. 2009). La présence ou l’absence d’une plante hôte en réponse au réchauffement
influence logiquement les interactions entre végétation et herbivores, qui eux-mêmes déterminent
les communautés de prédateurs (Dale and Frank 2018). Même à l’échelle d’une simple feuille, le
réchauffement peut provoquer la fermeture des stomates, modifiant ainsi le microclimat pour in fine
altérer le taux de visite des arthropodes, et ainsi les interactions biotiques sur cette surface
195
Discussion générale

(Pincebourde et al. 2016). Les ICU peuvent décaler la floraison des plantes dans la saison, perturbant
ainsi l'utilisation des ressources florales par les pollinisateurs, et participant au déclin des
pollinisateurs (Neil and Wu 2006, Goulson et al. 2015). Ces effets peuvent se répercuter sur les
insectes qui remplissent des fonctions doubles en tant que pollinisateurs et prédateurs (Luder et al.
2018). Les études menées jusqu’alors sur les interactions biotiques en ville considèrent souvent
l’urbanisation comme une variable englobant une gamme de divers changements abiotiques (perte
et fragmentation des habitats, pollution, température), mais ne dissocient que peu ces différentes
composantes (Dale and Frank 2018). Le design expérimental utilisé au cours de cette thèse
constituerait un cadre idéal pour conduire des expérimentations visant à déterminer si le
réchauffement influence les interactions trophiques. Sur la base des résultats obtenus dans le
Chapitre 1 sur les communautés d’araignées, une thématique de recherche intéressante serait de
déterminer les effets indépendants du réchauffement à l’échelle de l’ICU et du micro-habitat sur les
cortèges de proies en étudiant, par exemple, la prédation par le biais des contenus stomacaux des
arthropodes prédateurs le long de gradients de température.
Etudier l’effet du réchauffement sur les interactions biotiques impliquant les arthropodes prédateurs
en tant que proies serait également envisageable. Dès lors, des taxons plus mobiles, comme les
oiseaux, devraient être considérés. Des études ont suggéré que le taux de prédation des arthropodes
par les oiseaux devrait diminuer avec l'urbanisation (Isaksson 2018). Cependant, certaines espèces
ont été relevées comme étant particulièrement abondantes autour des sites d’étude « urbains »
échantillonnés en 2022 dans le cadre cette thèse (Données : Matthieu Beaufils, Bretagne Vivante).
Notamment, le Martinet noir et le Moineau domestique semblent prospérer en ville puisqu’ils
représentent respectivement 25% et 15% de l’abondance totale des espèces affiliées aux sites
« urbains ». Il est cependant techniquement compliqué d’étudier le lien entre urbanisation et
prédation des arthropodes par les oiseaux. Plusieurs études utilisant des chenilles d'insectes
artificielles ont montré à la fois des taux de prédation plus élevés (Faeth et al. 2005, Kozlov et al.
2017) et plus faibles par les oiseaux en ville (Eötvös et al. 2020, Kurucz et al. 2021). Les arthropodes
étudiés au cours de cette thèse sont principalement sédentaires (à l’exception des araignées
capables de ballooner), et n'ont aucun moyen de disperser d’une zone à forte intensité d’ICU
atmosphérique à une zone à intensité plus réduite. Quant à eux, les oiseaux peuvent se déplacer à
des échelles spatiales beaucoup plus étendues et ainsi traverser des zones d’ICU plus ou moins
intenses de la ville. Ces prédateurs d’arthropodes peuvent donc aisément thermoréguler en se
déplaçant vers des parties plus fraîches à l’échelle de la ville de Rennes. Dans notre système, il est
donc peu probable d’observer une réaction de ce taxon à l’ICU. Un éventuel décalage des
interactions dû aux variations thermiques, qu’il se produise dans le temps (phénologique) ou dans
l’espace, résulterait dès lors plus probablement d’un effet sur les proies arthropodes, que sur les
oiseaux
Ce travail a montré que les conditions climatiques urbaines à Rennes influencent les communautés
d’arthropodes. De plus, différentes échelles du climat sont liées à des réponses distinctes au sein de
ces communautés (Chapitres 1 et 2). D’une part, le réchauffement atmosphérique lié au phénomène
d’ICU est associé à une diminution de la diversité en espèces. Nous expliquons ce résultat par l’action
d’un filtre environnemental sur les communautés, excluant les espèces les plus sensibles à la chaleur
et de taille plus importante (Chapitre 3). Une comparaison interspécifique des réponses aux
variations de température chez plusieurs espèces d’araignée présentant des caractéristiques
écologiques contrastées (Chapitre 4) a permis de confirmer ce constat. Nous pensons que la capacité
limitée à disperser sur de longues distances par les airs chez les espèces de taille plus importante

196
 Discussion générale

pourrait être un facteur empêchant la fuite en cas de conditions intenses de réchauffement urbain,
expliquant la sensibilité élevée aux températures observée chez les Lycosidae testées. D’autre part,
nous avons montré qu’un réchauffement à l’échelle du micro-habitat (températures proches du sol,
dans la végétation herbacée) est lié à une diminution du nombre d’individus piégés (Chapitre 1). Une
étude en conditions microclimatiques contrôlées nous a permis d’observer qu’une espèce de
Lycosidae thermorégule de manière précise en adaptant sa position sur un gradient de température
grâce au comportement (Chapitre 5). Ce mécanisme peut expliquer, au moins partiellement,
l’association négative entre l’abondance totale observée à l’échelle des communautés d’araignées
et le réchauffement du micro-habitat. La capacité du micro-habitat à atténuer le réchauffement
urbain peut être influencée par la structure de la végétation, et la géométrie urbaine avoisinant
l’espace vert (Chapitre 6). Intégrer la composante climatique dans les mesures de gestion, et la
conception des infrastructures vertes constituerait donc un levier efficace pour améliorer l’efficacité
des actions de conservation de la biodiversité en ville.

197
Discussion générale

198
 Références

6 Références
Abram, P. K., G. Boivin, J. Moiroux, and J. Brodeur. 2017. Behavioural effects of temperature on ectothermic animals:
unifying thermal physiology and behavioural plasticity. Biological Reviews 92:1859–1876.
Aggéri, G. 2011. Inventer les villes-natures de demain... Gestion différenciée, gestion durable des espaces verts. Educagri
éditions. Dijon.
Alaruikka, D., D. J. Kotze, K. Matveinen, and J. Niemelä. 2002. Carabid Beetle and Spider Assemblages along a Forested
Urban–Rural Gradient in Southern Finland. Journal of Insect Conservation 6:195–206.
Alberti, M., C. Correa, J. M. Marzluff, A. P. Hendry, E. P. Palkovacs, K. M. Gotanda, V. M. Hunt, T. M. Apgar, and Y. Zhou.
2017. Global urban signatures of phenotypic change in animal and plant populations. Proceedings of the National
Academy of Sciences 114:8951–8956.
Alderweireldt, M., and R. De Keer. 1988. Comparison of the life cycle history of three Oedothorax-species (Araneae,
Linyphiidae) in relation to laboratory observations. XI. Europäisches Arachnologisches Colloquium, Tech. Univ.
Berlin, Dokumentation Kongresse u. Tagungen:169–177.
Alderweireldt, M., and R. De Keer. 1990. Field and laboratory observations on the life cycle of Pachygnatha degeeri
Sundevall, 1830 and Pachygnatha clercki Sundevall, 1823 (Araneae, Tetragnathidae). Acta Zoologica Fennica
190:35–39.
Ameline, C., T. T. Hoye, J. J. Bowden, R. R. Hansen, O. L. P. Hansen, C. Puzin, P. Vernon, and J. Petillon. 2018. Elevational
variation of body size and reproductive traits in high-latitude wolf spiders (Araneae: Lycosidae). Polar Biology
41:2561–2574.
Angilletta, M. J. 2009. Thermal adaptation: A theoretical and empirical synthesis. Oxford University Press.
Angilletta, M. J., B. S. Cooper, M. S. Schuler, and J. G. Boyles. 2010. The evolution of thermal physiology in endotherms.
Frontiers in Bioscience E 2:861–881.
Angilletta, M. J., R. S. Wilson, A. C. Niehaus, M. W. Sears, C. A. Navas, and P. L. Ribeiro. 2007. Urban physiology: city ants
possess high heat tolerance. PLoS ONE 2:e258.
Argañaraz, C. I., G. D. Rubio, and R. M. Gleiser. 2018. Spider communities in urban green patches and their relation to
local and landscape traits. Biodiversity and Conservation 27:981–1009.
Armson, D., P. Stringer, and A. R. Ennos. 2012. The effect of tree shade and grass on surface and globe temperatures in
an urban area. Urban Forestry & Urban Greening 11:245–255.
Aronson, M. F. J., F. A. La Sorte, C. H. Nilon, M. Katti, M. A. Goddard, C. A. Lepczyk, P. S. Warren, N. S. G. Williams, S.
Cilliers, B. Clarkson, C. Dobbs, R. Dolan, M. Hedblom, S. Klotz, J. L. Kooijmans, I. Kuhn, I. MacGregor-Fors, M.
McDonnell, U. Mortberg, P. Pysek, S. Siebert, J. Sushinsky, P. Werner, and M. Winter. 2014. A global analysis of
the impacts of urbanization on bird and plant diversity reveals key anthropogenic drivers. Proceedings of the Royal
Society B-Biological Sciences 281.
Aronson, M. F., C. A. Lepczyk, K. L. Evans, M. A. Goddard, S. B. Lerman, J. S. MacIvor, C. H. Nilon, and T. Vargo. 2017.
Biodiversity in the city: key challenges for urban green space management. Frontiers in Ecology and the
Environment 15:189–196.
Atkinson, D. 1994. Temperature and organism size: a biological law for ectotherms? Advances in Ecological Research
25:1–58.
Atkinson, D., and R. M. Sibly. 1997. Why are organisms usually bigger in colder environments? Making sense of a life
history puzzle. Trends in ecology & evolution 12:235–239.
Aznarez, C., J.-C. Svenning, G. Taveira, F. Baró, and U. Pascual. 2022. Wildness and habitat quality drive spatial patterns
of urban biodiversity. Landscape and Urban Planning 228:104570.
Bader, M. Y., M. Rietkerk, and A. K. Bregt. 2007. Vegetation Structure and Temperature Regimes of Tropical Alpine
Treelines. Arctic, Antarctic, and Alpine Research 39:353–364.
Baines, M., C. Hambler, P. J. Johnson, D. W. Macdonald, and H. Smith. 1998. The effects of arable field margin
management on the abundance and species richness of Araneae (spiders). Ecography 21:74–86.

199
Références
Balanyá, J., J. M. Oller, R. B. Huey, G. W. Gilchrist, and L. Serra. 2006. Global Genetic Change Tracks Global Climate
Warming in Drosophila subobscura. Science 313:1773–1775.
Baldock, K. C. R. 2020. Opportunities and threats for pollinator conservation in global towns and cities. Current Opinion
in Insect Science 38:63–71.
Baldock, K. C. R., M. A. Goddard, D. M. Hicks, W. E. Kunin, N. Mitschunas, H. Morse, L. M. Osgathorpe, S. G. Potts, K. M.
Robertson, A. V. Scott, P. P. A. Staniczenko, G. N. Stone, I. P. Vaughan, and J. Memmott. 2019. A systems approach
reveals urban pollinator hotspots and conservation opportunities. Nature Ecology & Evolution 3:363–373.
Barber, H. S. 1931. Traps for Cave-Inhabiting Insects. Journal of the Elisha Mitchell Scientific Society 46:259–266.
Barriga, J. C., L. Lassaletta, and A. G. Moreno. 2010. Ground-living spider assemblages from Mediterranean habitats under
different management conditions. The Journal of Arachnology 38:258–269.
Barry, J. P., C. H. Baxter, R. D. Sagarin, and S. E. Gilman. 1995. Climate-Related, Long-Term Faunal Changes in a California
Rocky Intertidal Community. Science 267:672–675.
Barry, R. G., and P. D. Blanken. 2016. Microclimate and Local Climate. Cambridge University Press, Cambridge.
Bartesaghi-Koc, C., P. Osmond, and A. Peters. 2020. Quantifying the seasonal cooling capacity of ‘green infrastructure
types’ (GITs): An approach to assess and mitigate surface urban heat island in Sydney, Australia. Landscape and
Urban Planning 203:103893.
Bartoń, K. 2016. MuMIn: Multi-Model Inference. R package version 1.47.5.
Baston, D. 2023, September 20. exactextractr: Fast extraction from raster datasets using polygons.
Bates, D., M. Machler, B. M. Bolker, and S. C. Walker. 2015. Fitting Linear Mixed-Effects Models Using lme4. Journal of
Statistical Software 67:1–48.
Bauer, T., H. Höfer, and J. Schirmel. 2024. Dry grasslands in urban areas can harbour arthropod species of local
conservation concern and should be prioritised for biodiversity-friendly mowing regimes. Insect Conservation and
Diversity n/a.
Beckers, N., N. Hein, A. Anneser, K. A. Vanselow, and J. Loeffler. 2020. Differences in mobility and dispersal capacity
determine body size clines in two common alpine-tundra arthropods. Insects 11:74.
Beebee, T. J. C. 1995. Amphibian breeding and climate. Nature 374:219–220.
Bell, J., D. Bohan, E. Shaw, and G. Weyman. 2005. Ballooning dispersal using silk: world fauna, phylogenies, genetics and
models. Bulletin of entomological research 95:69–114.
Bell, J. R., C. P. Wheater, and W. R. Cullen. 2001. The implications of grassland and heathland management for the
conservation of spider communities: a review. Journal of Zoology 255:377–387.
Beninde, J., M. Veith, and A. Hochkirch. 2015. Biodiversity in cities needs space: a meta-analysis of factors determining
intra-urban biodiversity variation. Ecology Letters 18:581–592.
Bennett, A. B., and C. Gratton. 2012. Local and landscape scale variables impact parasitoid assemblages across an
urbanization gradient. Landscape and Urban Planning 104:26–33.
Berec, M., Z. Stará, and S. Poláková. 2014. Relation between body-size and thermoregulation behavior: postprandial
thermophily in spiny-tailed Agama, Uromastyx acanthinura Bell. Polish Journal of Ecology 62:139–145.
van den Berg, A. E., J. Maas, R. A. Verheij, and P. P. Groenewegen. 2010. Green space as a buffer between stressful life
events and health. Social Science & Medicine 70:1203–1210.
Bernath-Plaisted, J. S., C. A. Ribic, W. B. Hills, P. A. Townsend, and B. Zuckerberg. 2023. Microclimate complexity in
temperate grasslands: implications for conservation and management under climate change. Environmental
Research Letters 18:064023.
Besancenot, J.-P. 2005. Climat et santé. A propos de la vague de chaleur de l’été 2003 en France. Climat et santé. A propos
de la vague de chaleur de l’été 2003 en France:215.
Betz, O., U. Irmler, and J. Klimaszewski. 2018. Biology of Rove Beetles (Staphylinidae). Springer.
Bink, J. 2014. Cryptachaea blattea (Urquhart, 1886)(Araneae, Theridiidae) voor het eerst waargenomen in Nederland.
Nieuwsbrief SPINED 34:11–11.
Bivand, R., M. Altman, L. Anselin, R. Assunção, O. Berke, F. G. Blanchet, M. Carvalho, B. Christensen, Y. Chun, C. Dormann,
S. Dray, D. Dunnington, V. Gómez-Rubio, E. Krainski, P. Legendre, N. Lewin-Koh, A. Li, G. Millo, W. Mueller, H. Ono,
200
 Références
J. Parry, P. Peres-Neto, G. Piras, M. Reder, J. Sauer, M. Tiefelsdorf, R. Westerholt, L. Wolf, and D. Yu. 2023,
November 23. spdep: Spatial Dependence: Weighting Schemes, Statistics.
Blanckenhorn, W. U. 2000. The evolution of body size: what keeps organisms small? The Quarterly Review of Biology
75:385–407.
Blanckenhorn, W. U., and M. Demont. 2004. Bergmann and Converse Bergmann Latitudinal Clines in Arthropods: Two
Ends of a Continuum?1. Integrative and Comparative Biology 44:413–424.
Blanckenhorn, W. U., A. F. Dixon, D. J. Fairbairn, M. W. Foellmer, P. Gibert, K. van der Linde, R. Meier, S. Nylin, S. Pitnick,
and C. Schoff. 2007. Proximate causes of Rensch’s rule: does sexual size dimorphism in arthropods result from sex
differences in development time? The American Naturalist 169:245–257.
Blandenier, G., and P. Fürst. 1998. Ballooning spiders caught by a suction trap in an agricultural landscape in Switzerland.
Pages 178–186 Proceedings of the 17th European colloquium of arachnology, Edinburgh. British Arachnological
Society Buckinghamshore.
Blanquart, F., O. Kaltz, S. L. Nuismer, and S. Gandon. 2013. A practical guide to measuring local adaptation. Ecology Letters
16:1195–1205.
Bogert, C. M. 1949. Thermoregulation in reptiles, a factor in evolution. Evolution 3:195–211.
Bohac, J. 1999. Staphylinid beetles as bioindicators. Agriculture, Ecosystems & Environment 74:357–372.
Bonte, D., and M. Dahirel. 2017. Dispersal: a central and independent trait in life history. Oikos 126:472–479.
Bowden, J. J., O. L. P. Hansen, K. Olsen, N. M. Schmidt, and T. T. Høye. 2018. Drivers of inter-annual variation and long-
term change in High-Arctic spider species abundances. Polar Biology 41:1635–1649.
Bowden, J. J., T. T. Høye, and C. M. Buddle. 2013. Fecundity and sexual size dimorphism of wolf spiders (Araneae:
Lycosidae) along an elevational gradient in the Arctic. Polar Biology 36:831–836.
Bowler, D. E., P. Haase, I. Kröncke, O. Tackenberg, H. G. Bauer, C. Brendel, R. W. Brooker, M. Gerisch, K. Henle, T. Hickler,
C. Hof, S. Klotz, I. Kühn, S. Matesanz, R. O‘Hara, D. Russell, O. Schweiger, F. Valladares, E. Welk, M. Wiemers, and
K. Böhning-Gaese. 2015. A cross-taxon analysis of the impact of climate change on abundance trends in central
Europe. Biological Conservation 187:41–50.
Bramer, I., B. J. Anderson, J. Bennie, A. J. Bladon, P. De Frenne, D. Hemming, R. A. Hill, M. R. Kearney, C. Körner, A. H.
Korstjens, J. Lenoir, I. M. D. Maclean, C. D. Marsh, M. D. Morecroft, R. Ohlemüller, H. D. Slater, A. J. Suggitt, F.
Zellweger, and P. K. Gillingham. 2018. Chapter Three - Advances in Monitoring and Modelling Climate at
Ecologically Relevant Scales. Pages 101–161 in D. A. Bohan, A. J. Dumbrell, G. Woodward, and M. Jackson, editors.
Advances in Ecological Research. Academic Press.
Brandt, E. E., K. T. Roberts, C. M. Williams, and D. O. Elias. 2020. Low temperatures impact species distributions of jumping
spiders across a desert elevational cline. JOURNAL OF INSECT PHYSIOLOGY 122.
Brans, K. I., M. Jansen, J. Vanoverbeke, N. Tüzün, R. Stoks, and L. De Meester. 2017. The heat is on: Genetic adaptation to
urbanization mediated by thermal tolerance and body size. Global Change Biology 23:5218–5227.
Brattstrom, B. H. 1965. Body temperatures of reptiles. The American Midland Naturalist 73:376–422.
Broadbent, A. M., A. M. Coutts, N. J. Tapper, M. Demuzere, and J. Beringer. 2018. The microscale cooling effects of water
sensitive urban design and irrigation in a suburban environment. Theoretical and Applied Climatology 134:1–23.
Brooks, S. J., A. Self, F. Toloni, and T. Sparks. 2014. Natural history museum collections provide information on
phenological change in British butterflies since the late-nineteenth century. International Journal of
Biometeorology 58:1749–1758.
Brown, G. R., and I. M. Matthews. 2016. A review of extensive variation in the design of pitfall traps and a proposal for a
standard pitfall trap design for monitoring ground-active arthropod biodiversity. Ecology and Evolution 6:3953–
3964.
Brown, J. H., J. F. Gillooly, A. P. Allen, V. M. Savage, and G. B. West. 2004. Toward a metabolic theory of ecology. Ecology
85:1771–1789.
Buchholz, S., and M. H. Egerer. 2020. Functional ecology of wild bees in cities: towards a better understanding of trait-
urbanization relationships. Biodiversity and Conservation 29:2779–2801.
Buchholz, S., A. K. Gathof, A. J. Grossmann, I. Kowarik, and L. K. Fischer. 2020. Wild bees in urban grasslands: Urbanisation,
functional diversity and species traits. Landscape and Urban Planning 196:103731.

201
Références
Buchholz, S., K. Hannig, M. Moller, and J. Schirmel. 2018. Reducing management intensity and isolation as promising tools
to enhance ground-dwelling arthropod diversity in urban grasslands. Urban Ecosystems 21:1139–1149.
Buisson, L., G. Grenouillet, S. Villéger, J. Canal, and P. Laffaille. 2013. Toward a loss of functional diversity in stream fish
assemblages under climate change. Global Change Biology 19:387–400.
Burghardt, K. T., D. W. Tallamy, and W. Gregory Shriver. 2009. Impact of Native Plants on Bird and Butterfly Biodiversity
in Suburban Landscapes. Conservation Biology 23:219–224.
Butt, N., D. F. Shanahan, N. Shumway, S. A. Bekessy, R. A. Fuller, J. E. M. Watson, R. Maggini, and D. G. Hole. 2018.
Opportunities for biodiversity conservation as cities adapt to climate change. Geo: Geography and Environment
5:e00052.
Buyantuyev, A., and J. Wu. 2010. Urban heat islands and landscape heterogeneity: linking spatiotemporal variations in
surface temperatures to land-cover and socioeconomic patterns. Landscape Ecology 25:17–33.
Cabon, V., S. Pincebourde, H. Colinet, V. Dubreuil, R. Georges, M. Launoy, J. Pétillon, H. Quénol, and B. Bergerot. 2023.
Preferred temperature in the warmth of cities: Body size, sex and development stage matter more than urban
climate in a ground-dwelling spider. Journal of Thermal Biology:103706.
Cabon, V., H. Quénol, B. Deletre, L. Copin, V. Dubreuil, and B. Bergerot. 2024a. Body size responses to urban temperature
variations are driven by life history traits in spiders. Functional Ecology 38:1578–1589.
Cabon, V., H. Quénol, V. Dubreuil, A. Ridel, and B. Bergerot. 2024b. Urban heat island and reduced habitat complexity
explain spider community composition by excluding large and heat-sensitive species. Land 13:83.
Cady, A. B. 1983. Microhabitat selection and locomotor activity of Schizocosa ocreata (Walckenaer) (Araneae: Lycosidae).
The Journal of Arachnology 11:297–307.
Cahill, A. E., M. E. Aiello-Lammens, M. C. Fisher-Reid, X. Hua, C. J. Karanewsky, H. Yeong Ryu, G. C. Sbeglia, F. Spagnolo, J.
B. Waldron, O. Warsi, and J. J. Wiens. 2013. How does climate change cause extinction? Proceedings of the Royal
Society B: Biological Sciences 280:20121890.
Calisi, R. M., and G. E. Bentley. 2009. Lab and field experiments: Are they the same animal? Hormones and Behavior 56:1–
10.
Cantat, O. 2004. L’îlot de chaleur urbain parisien selon les types de temps. Norois. Environnement, aménagement,
société:75–102.
Cardoso, P., S. Pekár, R. Jocqué, and J. A. Coddington. 2011. Global Patterns of Guild Composition and Functional Diversity
of Spiders. PLOS ONE 6:e21710.
Chao, A., P. A. Henderson, C.-H. Chiu, F. Moyes, K.-H. Hu, M. Dornelas, and A. E. Magurran. 2021. Measuring temporal
change in alpha diversity: A framework integrating taxonomic, phylogenetic and functional diversity and the
iNEXT.3D standardization. Methods in Ecology and Evolution 12:1926–1940.
Chatelain, M., J. Rüdisser, and M. Traugott. 2023. Urban-driven decrease in arthropod richness and diversity associated
with group-specific changes in arthropod abundance. Frontiers in Ecology and Evolution 11.
Chen, I.-C., J. K. Hill, R. Ohlemüller, D. B. Roy, and C. D. Thomas. 2011. Rapid Range Shifts of Species Associated with High
Levels of Climate Warming. Science 333:1024–1026.
Chen, J., J. F. Franklin, and T. A. Spies. 1993. Contrasting microclimates among clearcut, edge, and interior of old-growth
Douglas-fir forest. Agricultural and Forest Meteorology 63:219–237.
Cheptou, P.-O., A. L. Hargreaves, D. Bonte, and H. Jacquemyn. 2017. Adaptation to fragmentation: evolutionary dynamics
driven by human influences. Philosophical Transactions of the Royal Society B: Biological Sciences 372:20160037.
Chevin, L.-M., S. Collins, and F. Lefèvre. 2013. Phenotypic plasticity and evolutionary demographic responses to climate
change: taking theory out to the field. Functional Ecology 27:967–979.
Chollet, S., C. Brabant, S. Tessier, and V. Jung. 2018. From urban lawns to urban meadows: Reduction of mowing
frequency increases plant taxonomic, functional and phylogenetic diversity. Landscape and Urban Planning
180:121–124.
Chown, S. L., S. Slabber, M. A. McGeoch, C. Janion, and H. P. Leinaas. 2007. Phenotypic plasticity mediates climate change
responses among invasive and indigenous arthropods. Proceedings of the Royal Society B: Biological Sciences
274:2531–2537.

202
 Références
Christie, F. J., G. Cassis, and D. F. Hochuli. 2010. Urbanization affects the trophic structure of arboreal arthropod
communities. Urban Ecosystems 13:169–180.
Clergeau, P. 2011. Ville et biodiversité. Les enseignements d’une recherche pluridisciplinaire. Presses Universitaires de
Rennes.
Clough, Y., A. Kruess, and T. Tscharntke. 2007. Organic versus conventional arable farming systems: Functional grouping
helps understand staphylinid response. Agriculture, Ecosystems & Environment 118:285–290.
Concepción, E. D., M. Moretti, F. Altermatt, M. P. Nobis, and M. K. Obrist. 2015. Impacts of urbanisation on biodiversity:
the role of species mobility, degree of specialisation and spatial scale. Oikos 124:1571–1582.
Conover, D. O., and E. T. Schultz. 1995. Phenotypic similarity and the evolutionary significance of countergradient
variation. Trends in Ecology & Evolution 10:248–252.
Cornes, R. C., G. van der Schrier, E. J. M. van den Besselaar, and P. D. Jones. 2018. An Ensemble Version of the E-OBS
Temperature and Precipitation Data Sets. Journal of Geophysical Research: Atmospheres 123:9391–9409.
Croci, S., A. Butet, A. Georges, R. Aguejdad, and P. Clergeau. 2008. Small urban woodlands as biodiversity conservation
hot-spot: a multi-taxon approach. Landscape Ecology 23:1171–1186.
Dale, A. G., and S. D. Frank. 2014. Urban warming trumps natural enemy regulation of herbivorous pests. Ecological
Applications 24:1596–1607.
Dale, A. G., and S. D. Frank. 2018. Urban plants and climate drive unique arthropod interactions with unpredictable
consequences. Current Opinion in Insect Science 29:27–33.
Daly, C., D. R. Conklin, and M. H. Unsworth. 2010. Local atmospheric decoupling in complex topography alters climate
change impacts. International Journal of Climatology 30:1857–1864.
Damgaard, C. 2019. A Critique of the Space-for-Time Substitution Practice in Community Ecology. Trends in Ecology &
Evolution 34:416–421.
De Keer, R., and J.-P. Maelfait. 1987a. Laboratory observations on the development and reproduction of Oedothorax
fuscus (Blackwall, 1834)(Araneida, Linyphiidae) under different conditions of temperature and food supply. Revue
d’Ecologie et de Biologie du Sol 24:63–73.
De Keer, R., and J.-P. Maelfait. 1987b. Life history of Oedothorax fuscus (Blackwall, 1834)(Aranae, Linyphiidae) in a heavily
grazed pasture. Revue d’Ecologie et de Biologie du Sol 24:171–185.
De Keer, R., and J.-P. Maelfait. 1988. Laboratory observations on the development and reproduction of Erigone atra
Blackwall, 1833 (Araneae, Linyphiidae). Bulletin of the British Arachnological Society 7:237–242.
Deichsel, R. 2006. Species change in an urban setting—ground and rove beetles (Coleoptera: Carabidae and
Staphylinidae) in Berlin. Urban Ecosystems 9:161–178.
Delgado de la Flor, Y. A., C. E. Burkman, T. K. Eldredge, and M. M. Gardiner. 2017. Patch and landscape-scale variables
influence the taxonomic and functional composition of beetles in urban greenspaces. Ecosphere 8:e02007.
Delgado de la flor, Y. A., K. I. Perry, L. M. Collis, P. L. Phelan, and M. M. Gardiner. 2024. Biotic and abiotic factors drive
multi-trophic interactions among spiders at different spatial scales in urban greenspaces. Journal of Urban Ecology
10:juae008.
Delgado de la flor, Y. A., K. I. Perry, K. J. Turo, D. M. Parker, J. L. Thompson, and M. M. Gardiner. 2020. Local and landscape-
scale environmental filters drive the functional diversity and taxonomic composition of spiders across urban
greenspaces. Journal of Applied Ecology 57:1570–1580.
Delignette-Muller, M. L., and C. Dutang. 2015. fitdistrplus: An R package for fitting distributions. Journal of Statistical
Software 64:1–34.
Deutsch, C. A., J. J. Tewksbury, R. B. Huey, K. S. Sheldon, C. K. Ghalambor, D. C. Haak, and P. R. Martin. 2008. Impacts of
climate warming on terrestrial ectotherms across latitude. Proceedings of the National Academy of Sciences
105:6668–6672.
DeVito, J., J. M. Meik, M. M. Gerson, and D. R. Formanowicz, Jr. 2004. Physiological tolerances of three sympatric riparian
wolf spiders (Araneae: Lycosidae) correspond with microhabitat distributions. Canadian Journal of Zoology
82:1119–1125.

203
Références
Diamond, S. E., L. D. Chick, A. Perez, S. A. Strickler, and R. A. Martin. 2018a. Evolution of thermal tolerance and its fitness
consequences: parallel and non-parallel responses to urban heat islands across three cities. Proceedings of the
Royal Society B: Biological Sciences 285:20180036.
Diamond, S. E., L. D. Chick, A. Perez, S. A. Strickler, and R. A. Martin. 2018b. Evolution of thermal tolerance and its fitness
consequences: parallel and non-parallel responses to urban heat islands across three cities. Proceedings of the
Royal Society B: Biological Sciences 285:20180036.
Diamond, S. E., L. D. Chick, A. Perez, S. A. Strickler, and C. Zhao. 2018c. Evolution of plasticity in the city: urban acorn ants
can better tolerate more rapid increases in environmental temperature. Conservation Physiology 6:coy030.
Diamond, S. E., L. D. Chick, A. Perez, S. A. Strickler, and C. Zhao. 2018d. Evolution of plasticity in the city: urban acorn ants
can better tolerate more rapid increases in environmental temperature. CONSERVATION PHYSIOLOGY 6.
Diamond, S. E., L. Chick, A. Perez, S. A. Strickler, and R. A. Martin. 2017. Rapid evolution of ant thermal tolerance across
an urban-rural temperature cline. Biological Journal of the Linnean Society 121:248–257.
Diamond, S. E., A. M. Frame, R. A. Martin, and L. B. Buckley. 2011. Species’ traits predict phenological responses to climate
change in butterflies. Ecology 92:1005–1012.
Diamond, S. E., and R. A. Martin. 2016. The interplay between plasticity and evolution in response to human-induced
environmental change. F1000Research 5:2835.
Diamond, S. E., and R. A. Martin. 2020. Evolutionary consequences of the urban heat island. Pages 91–110 in M. Szulkin,
J. Munshi-South, and A. Charmantier, editors. Urban Evolutionary Biology. Oxford University Press.
Diamond, S. E., and R. A. Martin. 2021. Physiological adaptation to cities as a proxy to forecast global-scale responses to
climate change. Journal of Experimental Biology 224:jeb229336.
Díaz, J. A. 1997. Ecological correlates of the thermal quality of an ectotherm’s habitat: a comparison between two
temperate lizard populations. Functional Ecology 11:79–89.
Díaz, S., S. Lavorel, F. S. Chapin, P. A. Tecco, D. E. Gurvich, and K. Grigulis. 2007. Functional Diversity — at the Crossroads
between Ecosystem Functioning and Environmental Filters. Pages 81–91 in J. G. Canadell, D. E. Pataki, and L. F.
Pitelka, editors. Terrestrial Ecosystems in a Changing World. Springer, Berlin, Heidelberg.
Diehl, E., V. L. Mader, V. Wolters, and K. Birkhofer. 2013. Management intensity and vegetation complexity affect web-
building spiders and their prey. Oecologia 173:579–589.
Dillon, M. E., G. Wang, P. A. Garrity, and R. B. Huey. 2009. Review: Thermal preference in Drosophila. Journal of Thermal
Biology 34:109–119.
D’Odorico, P., Y. He, S. Collins, S. F. J. De Wekker, V. Engel, and J. D. Fuentes. 2013. Vegetation–microclimate feedbacks
in woodland–grassland ecotones. Global Ecology and Biogeography 22:364–379.
Donihue, C. M., and M. R. Lambert. 2015. Adaptive evolution in urban ecosystems. AMBIO 44:194–203.
Dormann, C. F., J. Elith, S. Bacher, C. Buchmann, G. Carl, G. Carré, J. R. G. Marquéz, B. Gruber, B. Lafourcade, P. J. Leitão,
T. Münkemüller, C. McClean, P. E. Osborne, B. Reineking, B. Schröder, A. K. Skidmore, D. Zurell, and S. Lautenbach.
2013. Collinearity: a review of methods to deal with it and a simulation study evaluating their performance.
Ecography 36:27–46.
Downes, S., and R. Shine. 1998. Heat, safety or solitude? Using habitat selection experiments to identify a lizard’s
priorities. Animal Behaviour 55:1387–1396.
Downie, I. S., I. Ribera, D. I. McCracken, W. L. Wilson, G. N. Foster, A. Waterhouse, V. J. Abernethy, and K. J. Murphy. 2000.
Modelling populations of Erigone atra and E. dentipalpis (Araneae: Linyphiidae) across an agricultural gradient in
Scotland. Agriculture, Ecosystems & Environment 80:15–28.
Dubreuil, V., C. Brabant, G. Delaunay, J. Nabucet, H. Quenol, F. Clain, F. Leprince, J. Dreano, and L. Georget. 2022. Rennes,
une ville climato-intelligente ? - L’IoT au service du suivi des îlots de chaleur. Les technologies numériques au
service de la ville et de la personne.
Dubreuil, V., X. Foissard, J. Nabucet, A. Thomas, and H. Quénol. 2020. Fréquence et intensité des îlots de chaleur à rennes :
bilan de 16 années d’observations (2004-2019). Climatologie 17:6.
Duffey, E. 1956. Aerial dispersal in a known spider population. The Journal of Animal Ecology:85–111.
Dumas, G., V. Masson, J. Hidalgo, V. Edouart, A. Hanna, and G. Poujol. 2021. Co-construction of climate services based on
a weather stations network: Application in Toulouse agglomeration local authority. Climate Services 24:100274.

204
 Références
Dzialowski, E. M. 2005. Use of operative temperature and standard operative temperature models in thermal biology.
Journal of Thermal Biology 30:317–334.
Eckert, S., M. Moller, and S. Buchholz. 2017. Grasshopper diversity of urban wastelands is primarily boosted by habitat
factors. Insect Conservation and Diversity 10:248–257.
Edgar, W. D. 1971. The life-cycle, abundance and seasonal movement of the wolf spider, Lycosa (Pardosa) lugubris, in
central Scotland. Journal of Animal Ecology 40:303–322.
Egerer, M. H., C. Arel, M. D. Otoshi, R. D. Quistberg, P. Bichier, and S. M. Philpott. 2017. Urban arthropods respond variably
to changes in landscape context and spatial scale. Journal of Urban Ecology 3:jux001.
Elmqvist, T., M. Fragkias, J. Goodness, B. Güneralp, P. J. Marcotullio, R. I. McDonald, S. Parnell, M. Schewenius, M.
Sendstad, and K. C. Seto. 2013. Urbanization, biodiversity and ecosystem services: challenges and opportunities:
a global assessment. Springer Nature.
Emmanuel, R., and A. Loconsole. 2015. Green infrastructure as an adaptation approach to tackling urban overheating in
the Glasgow Clyde Valley Region, UK. Landscape and Urban Planning 138:71–86.
Entling, M. H., K. Stämpfli, and O. Ovaskainen. 2011. Increased propensity for aerial dispersal in disturbed habitats due to
intraspecific variation and species turnover. Oikos 120:1099–1109.
Entling, W., M. H. Schmidt, S. Bacher, R. Brandl, and W. Nentwig. 2007. Niche properties of Central European spiders:
shading, moisture and the evolution of the habitat niche. Global Ecology and Biogeography 16:440–448.
Eötvös, C. B., G. L. Lövei, and T. Magura. 2020. Predation Pressure on Sentinel Insect Prey along a Riverside Urbanization
Gradient in Hungary. Insects 11:97.
Esperk, T., T. Tammaru, S. Nylin, and T. Teder. 2007. Achieving high sexual size dimorphism in insects: females add instars.
Ecological entomology 32:243–256.
Faeth, S. H., C. Bang, and S. Saari. 2011. Urban biodiversity: patterns and mechanisms. Annals of the New York Academy
of Sciences 1223:69–81.
Faeth, S. H., P. S. Warren, E. Shochat, and W. A. Marussich. 2005. Trophic Dynamics in Urban Communities. BioScience
55:399–407.
Fahrig, L. 1997. Relative Effects of Habitat Loss and Fragmentation on Population Extinction. The Journal of Wildlife
Management 61:603–610.
Fahrig, L. 2003. Effects of Habitat Fragmentation on Biodiversity. Annual Review of Ecology, Evolution, and Systematics
34:487–515.
Fangue, N. A., J. E. Podrabsky, L. I. Crawshaw, and P. M. Schulte. 2009. Countergradient variation in temperature
preference in populations of killifish Fundulus heteroclitus. Physiological and Biochemical Zoology 82:776–786.
Fenberg, P. B., A. Self, J. R. Stewart, R. J. Wilson, and S. J. Brooks. 2016. Exploring the universal ecological responses to
climate change in a univoltine butterfly. Journal of Animal Ecology 85:739–748.
Fenoglio, M. S., M. R. Rossetti, and M. Videla. 2020. Negative effects of urbanization on terrestrial arthropod
communities: A meta-analysis. Global Ecology and Biogeography 29:1412–1429.
Fischer, J. D., S. C. Schneider, A. A. Ahlers, and J. R. Miller. 2015. Categorizing wildlife responses to urbanization and
conservation implications of terminology. Conservation Biology 29:1246–1248.
Flaxman, S., S. Mishra, A. Gandy, H. J. T. Unwin, T. A. Mellan, H. Coupland, C. Whittaker, H. Zhu, T. Berah, J. W. Eaton, M.
Monod, A. C. Ghani, C. A. Donnelly, S. Riley, M. A. C. Vollmer, N. M. Ferguson, L. C. Okell, and S. Bhatt. 2020.
Estimating the effects of non-pharmaceutical interventions on COVID-19 in Europe. Nature 584:257–261.
de la Flor, Y., C. Burkman, T. Eldredge, and M. Gardiner. 2017. Patch and landscape-scale variables influence the
taxonomic and functional composition of beetles in urban greenspaces. Ecosphere 8.
de la Flor, Y., K. Perry, K. Turo, D. Parker, J. Thompson, and M. Gardiner. 2020. Local and landscape-scale environmental
filters drive the functional diversity and taxonomic composition of spiders across urban greenspaces. JOURNAL OF
APPLIED ECOLOGY 57:1570–1580.
Foelix, R. 2011. Biology of spiders. OUP USA.
Foissard, X. 2015, October 14. L’îlot de chaleur urbain et le changement climatique : application à l’agglomération
rennaise. phdthesis, Université Rennes 2.

205
Références
Foissard, X., O. Cantat, V. Dubreuil, E. Savouret, H. Quénol, A. Bensaïd, and O. Planchon. 2012. Climat urbain, changement
climatique et types de temps dans le Grand Ouest de la France. Pages 73–92 Changement climatique dans l’Ouest.
Presses Universitaires de Rennes.
Foissard, X., V. Dubreuil, and H. Quenol. 2019a. Defining scales of the land use effect to map the urban heat island in a
mid-size European city: Rennes (France). Urban Climate 29.
Foissard, X., V. Dubreuil, and H. Quénol. 2019b. Defining scales of the land use effect to map the urban heat island in a
mid-size European city: Rennes (France). Urban Climate 29:100490.
Fontaine, B., B. Bergerot, I. Le Viol, and R. Julliard. 2016. Impact of urbanization and gardening practices on common
butterfly communities in France. Ecology and Evolution 6:8174–8180.
Forister, M. L., and A. M. Shapiro. 2003. Climatic trends and advancing spring flight of butterflies in lowland California.
Global Change Biology 9:1130–1135.
Forrest, J. R. 2016. Complex responses of insect phenology to climate change. Current Opinion in Insect Science 17:49–
54.
Forsman, A. 2000. Some like it hot: Intra-population variation in behavioral thermoregulation in color-polymorphic pygmy
grasshoppers. Evolutionary Ecology 14:25–38.
Forster, J., A. G. Hirst, and D. Atkinson. 2012. Warming-induced reductions in body size are greater in aquatic than
terrestrial species. Proceedings of the National Academy of Sciences 109:19310–19314.
Fox, J., D. Bates, D. Firth, M. Friendly, G. Gorjanc, S. Graves, R. Heiberger, G. Monette, H. Nilsson, B. Ripley, S. Weisberg,
and A. Zeleis. 2007. The car package.
Franco, A. M. A., J. K. Hill, C. Kitschke, Y. C. Collingham, D. B. Roy, R. Fox, B. Huntley, and C. D. Thomas. 2006. Impacts of
climate warming and habitat loss on extinctions at species’ low-latitude range boundaries. Global Change Biology
12:1545–1553.
Francoeur, X. W., D. Dagenais, A. Paquette, J. Dupras, and C. Messier. 2021. Complexifying the urban lawn improves heat
mitigation and arthropod biodiversity. Urban Forestry & Urban Greening 60:127007.
Frick, H., C. Kropf, and W. Nentwig. 2007. Laboratory temperature preferences of the wolf spider Pardosa riparia
(Araneae: Lycosidae). Arachnology 14:45–48.
Gaffin, S. R., C. Rosenzweig, R. Khanbilvardi, L. Parshall, S. Mahani, H. Glickman, R. Goldberg, R. Blake, R. B. Slosberg, and
D. Hillel. 2008. Variations in New York city’s urban heat island strength over time and space. Theoretical and
Applied Climatology 94:1–11.
Gaffin, S. R., C. Rosenzweig, and A. Y. Y. Kong. 2012. Adapting to climate change through urban green infrastructure.
Nature Climate Change 2:704–704.
Galea, S., R. M. Merchant, and N. Lurie. 2020. The Mental Health Consequences of COVID-19 and Physical Distancing: The
Need for Prevention and Early Intervention. JAMA Internal Medicine 180:817–818.
Gardiner, T., and M. Hassall. 2009. Does microclimate affect grasshopper populations after cutting of hay in improved
grassland? Journal of Insect Conservation 13:97–102.
Gaublomme, E., H. Dhuyvetter, P. Verdyck, and K. Desender. 2005. Effects of urbanisation on carabid beetles in old beech
forests. DIAS report 114:111–123.
Ge, Q., H. Wang, T. Rutishauser, and J. Dai. 2015. Phenological response to climate change in China: a meta-analysis.
Global Change Biology 21:265–274.
Geiger, R., R. H. Aron, and P. Todhunter. 2009. The Climate Near the Ground. Rowman & Littlefield.
Gerlach, J., M. Samways, and J. Pryke. 2013. Terrestrial invertebrates as bioindicators: an overview of available taxonomic
groups. Journal of Insect Conservation 17:831–850.
Gibson, C. W. D., C. Hambler, and V. K. Brown. 1992. Changes in Spider (Araneae) Assemblages in Relation to Succession
and Grazing Management. Journal of Applied Ecology 29:132–142.
Gilbert, A. L., and D. B. Miles. 2017. Natural selection on thermal preference, critical thermal maxima and locomotor
performance. Proceedings of the Royal Society B: Biological Sciences 284:20170536.
Gilbert, N., and D. A. Raworth. 1996. Insects and temperature — a general theory. The Canadian Entomologist 128:1–13.

206
 Références
Gilman, S. E., M. C. Urban, J. Tewksbury, G. W. Gilchrist, and R. D. Holt. 2010. A framework for community interactions
under climate change. Trends in Ecology & Evolution 25:325–331.
Goulson, D., E. Nicholls, C. Botías, and E. L. Rotheray. 2015. Bee declines driven by combined stress from parasites,
pesticides, and lack of flowers. Science 347:1255957.
Gower, J. C. 1971. A general coefficient of similarity and some of its properties. Biometrics 27:857–871.
Gril, E., F. Spicher, C. Greiser, M. B. Ashcroft, S. Pincebourde, S. Durrieu, M. Nicolas, B. Richard, G. Decocq, R. Marrec, and
J. Lenoir. 2023. Slope and equilibrium: A parsimonious and flexible approach to model microclimate. Methods in
Ecology and Evolution 14:885–897.
Grimm, N. B., S. H. Faeth, N. E. Golubiewski, C. L. Redman, J. G. Wu, X. M. Bai, and J. M. Briggs. 2008. Global change and
the ecology of cities. Science 319:756–760.
Grimmond, C. S. B., S. M. Robeson, and J. T. Schoof. 2000. Spatial variability of micro-climatic conditions within a mid-
latitude deciduous forest. Climate Research 15:137–149.
Groffman, P. M., J. Cavender-Bares, N. D. Bettez, J. M. Grove, S. J. Hall, J. B. Heffernan, S. E. Hobbie, K. L. Larson, J. L.
Morse, C. Neill, K. Nelson, J. O’Neil-Dunne, L. Ogden, D. E. Pataki, C. Polsky, R. R. Chowdhury, and M. K. Steele.
2014. Ecological homogenization of urban USA. Frontiers in Ecology and the Environment 12:74–81.
Gros, C., A. Bulot, S. Aviron, V. Beaujouan, and H. Daniel. 2023. Both management practices and landscape influence plant
communities in urban grasslands. Frontiers in Ecology and Evolution 11.
Gutiérrez, D., and R. J. Wilson. 2021. Intra- and interspecific variation in the responses of insect phenology to climate.
Journal of Animal Ecology 90:248–259.
Gvoždík, L. 2012. Plasticity of preferred body temperatures as means of coping with climate change? Biology Letters
8:262–265.
Habel, J. C., J. Dengler, M. Janišová, P. Török, C. Wellstein, and M. Wiezik. 2013. European grassland ecosystems:
threatened hotspots of biodiversity. Biodiversity and Conservation 22:2131–2138.
Hacala, A., M. Le Roy, J. Sawtschuk, and J. Pétillon. 2020. Comparative responses of spiders and plants to maritime
heathland restoration. Biodiversity and Conservation 29:229–249.
Hacala, A., M. Salgueiro-Simon, A. Bain, D. Marguerie, and J. Pétillon. 2024. Spider and vascular plant assemblages in
subarctic peat bogs are complementary ecological indicators of variation in local and landscape factors. Ecological
Indicators 158:111389.
Haddad, N. M., L. A. Brudvig, J. Clobert, K. F. Davies, A. Gonzalez, R. D. Holt, T. E. Lovejoy, J. O. Sexton, M. P. Austin, C. D.
Collins, W. M. Cook, E. I. Damschen, R. M. Ewers, B. L. Foster, C. N. Jenkins, A. J. King, W. F. Laurance, D. J. Levey,
C. R. Margules, B. A. Melbourne, A. O. Nicholls, J. L. Orrock, D.-X. Song, and J. R. Townshend. 2015. Habitat
fragmentation and its lasting impact on Earth’s ecosystems. Science Advances 1:e1500052.
Haesen, S., J. Lenoir, E. Gril, P. De Frenne, J. J. Lembrechts, M. Kopecký, M. Macek, M. Man, J. Wild, and K. Van Meerbeek.
2023. Microclimate reveals the true thermal niche of forest plant species. Ecology Letters 26:2043–2055.
Hahs, A. K., B. Fournier, M. F. J. Aronson, C. H. Nilon, A. Herrera-Montes, A. B. Salisbury, C. G. Threlfall, C. C. Rega-Brodsky,
C. A. Lepczyk, F. A. La Sorte, I. MacGregor-Fors, J. Scott MacIvor, K. Jung, M. R. Piana, N. S. G. Williams, S. Knapp,
A. Vergnes, A. A. Acevedo, A. M. Gainsbury, A. Rainho, A. J. Hamer, A. Shwartz, C. C. Voigt, D. Lewanzik, D. M.
Lowenstein, D. O’Brien, D. Tommasi, E. Pineda, E. S. Carpenter, E. Belskaya, G. L. Lövei, J. C. Makinson, J. L.
Coleman, J. P. Sadler, J. Shroyer, J. T. Shapiro, K. C. R. Baldock, K. Ksiazek-Mikenas, K. C. Matteson, K. Barrett, L.
Siles, L. F. Aguirre, L. O. Armesto, M. Zalewski, M. I. Herrera-Montes, M. K. Obrist, R. K. Tonietto, S. A. Gagné, S. J.
Hinners, T. Latty, T. D. Surasinghe, T. Sattler, T. Magura, W. Ulrich, Z. Elek, J. Castañeda-Oviedo, R. Torrado, D. J.
Kotze, and M. Moretti. 2023. Urbanisation generates multiple trait syndromes for terrestrial animal taxa
worldwide. Nature Communications 14:4751.
Hall, J. M., J. F. Handley, and A. R. Ennos. 2012. The potential of tree planting to climate-proof high density residential
areas in Manchester, UK. Landscape and Urban Planning 104:410–417.
Hallander, H. 1967. Range and movements of the wolf spiders Pardosa chelata (O. F. Müller) and P. pullata (Clerck). Oikos
18:360–364.
Hallander, H. 1970a. Environments of the wolfspiders Pardosa chelata (OF MULLER) and Pardosa pullata (Clerck). Ekol Pol
Ser A.

207
Références
Hallander, H. 1970b. Prey, cannibalism and microhabitat selection in the wolf spiders Pardosa chelata O. F. Müller and P.
pullata Clerck. Oikos 21:337–340.
Hamblin, A. L., E. Youngsteadt, and S. D. Frank. 2018. Wild bee abundance declines with urban warming, regardless of
floral density. Urban Ecosystems 21:419–428.
Hamblin, A. L., E. Youngsteadt, M. M. Lopez-Uribe, and S. D. Frank. 2017. Physiological thermal limits predict differential
responses of bees to urban heat-island effects. Biology Letters 13.
Hänggi, A., and S. Straub. 2015. Storage buildings and greenhouses as stepping stones for non-native potentially invasive
spiders (Araneae)–a baseline study in Basel, Switzerland. Arachnologische Mitteilungen.
Hargrove, W. W., and J. Pickering. 1992. Pseudoreplication: a sine qua non for regional ecology. Landscape Ecology 6:251–
258.
Harvey, J. A., and Y. Dong. 2023. Climate Change, Extreme Temperatures and Sex-Related Responses in Spiders. Biology
12:615.
Harvey, J. A., R. Heinen, R. Gols, and M. P. Thakur. 2020. Climate change-mediated temperature extremes and insects:
From outbreaks to breakdowns. Global Change Biology 26:6685–6701.
Harvey, J. A., W. H. van der Putten, H. Turin, R. Wagenaar, and T. M. Bezemer. 2008. Effects of changes in plant species
richness and community traits on carabid assemblages and feeding guilds. Agriculture, Ecosystems & Environment
127:100–106.
Harvey, J. A., K. Tougeron, R. Gols, R. Heinen, M. Abarca, P. K. Abram, Y. Basset, M. Berg, C. Boggs, J. Brodeur, P. Cardoso,
J. G. de Boer, G. R. De Snoo, C. Deacon, J. E. Dell, N. Desneux, M. E. Dillon, G. A. Duffy, L. A. Dyer, J. Ellers, A.
Espíndola, J. Fordyce, M. L. Forister, C. Fukushima, M. J. G. Gage, C. García-Robledo, C. Gely, M. Gobbi, C. Hallmann,
T. Hance, J. Harte, A. Hochkirch, C. Hof, A. A. Hoffmann, J. G. Kingsolver, G. P. A. Lamarre, W. F. Laurance, B.
Lavandero, S. R. Leather, P. Lehmann, C. Le Lann, M. M. López-Uribe, C.-S. Ma, G. Ma, J. Moiroux, L. Monticelli, C.
Nice, P. J. Ode, S. Pincebourde, W. J. Ripple, M. Rowe, M. J. Samways, A. Sentis, A. A. Shah, N. Stork, J. S. Terblanche,
M. P. Thakur, M. B. Thomas, J. M. Tylianakis, J. Van Baaren, M. Van de Pol, W. H. Van der Putten, H. Van Dyck, W.
C. E. P. Verberk, D. L. Wagner, W. W. Weisser, W. C. Wetzel, H. A. Woods, K. A. G. Wyckhuys, and S. L. Chown.
2022. Scientists’ warning on climate change and insects. Ecological Monographs 93:e1553.
Hassall, M., A. Hawthorne, M. Maudsley, P. White, and C. Cardwell. 1992. Effects of headland management on
invertebrate communities in cereal fields. Agriculture, Ecosystems & Environment 40:155–178.
HCBC. 2023. Bulletin 2023 du HCBC : le changement climatique en Bretagne.
He, Y., P. D’Odorico, S. F. J. De Wekker, J. D. Fuentes, and M. Litvak. 2010. On the impact of shrub encroachment on
microclimate conditions in the northern Chihuahuan desert. Journal of Geophysical Research: Atmospheres 115.
Hein, N., J. Pétillon, R. Pape, H. Feilhauer, K. A. Vanselow, and J. Löffler. 2019. Broad-scale rather than fine-scale
environmental variation drives body size in a wandering predator (Araneae, Lycosidae). Arctic, Antarctic, and
Alpine Research 51:315–326.
Heisler, G. M., and A. J. Brazel. 2010. The urban physical environment: Temperature and urban heat islands. Pages 29–56
Urban Ecosystem Ecology. John Wiley & Sons, Ltd.
Hendrickx, F., K. De Cock, D. De Bakker, and J.-P. Maelfait. 2001. Differences in geographic distribution and habitat of
some cryptic species in the Pardosa lugubris group (Lycosidae, Araneae) in Belgium. Belgian journal of zoology
131:79–84.
Hertz, P. E., R. B. Huey, and R. D. Stevenson. 1993. Evaluating temperature regulation by field-active ectotherms: the
fallacy of the inappropriate question. The American Naturalist 142:796–818.
Herve, M. 2023, September 26. RVAideMemoire: Testing and Plotting Procedures for Biostatistics.
Hesselbarth, M. H. K., M. Sciaini, K. A. With, K. Wiegand, and J. Nowosad. 2019. landscapemetrics: an open-source R tool
to calculate landscape metrics. Ecography 42:1648–1657.
Heydemann, B. 1960. Verlauf und Abhängigkeit von Spinnensukzessionen im Neuland der Nordseeküste. Verhandlungen
der deutschen Zoologischen Gesellschaft 54:431–457.
Hijmans, R. J., J. van Etten, M. Sumner, J. Cheng, D. Baston, A. Bevan, R. Bivand, L. Busetto, M. Canty, B. Fasoli, D. Forrest,
A. Ghosh, D. Golicher, J. Gray, J. A. Greenberg, P. Hiemstra, K. Hingee, A. Ilich, I. for M. A. Geosciences, C. Karney,
M. Mattiuzzi, S. Mosher, B. Naimi, J. Nowosad, E. Pebesma, O. P. Lamigueiro, E. B. Racine, B. Rowlingson, A.
Shortridge, B. Venables, and R. Wueest. 2023, October 14. raster: Geographic Data Analysis and Modeling.

208
 Références
Hill, M. O. 1973. Diversity and Evenness: A Unifying Notation and Its Consequences. Ecology 54:427–432.
Hodgson, M. J., and L. E. Schwanz. 2019. Drop it like it’s hot: Interpopulation variation in thermal phenotypes shows
counter-gradient pattern. Journal of Thermal Biology 83:178–186.
Hof, C., I. Levinsky, M. B. Araújo, and C. Rahbek. 2011. Rethinking species’ ability to cope with rapid climate change. Global
Change Biology 17:2987–2990.
Hollander, J. D., and H. Lof. 1972. Differential use of the habitat by Pardosa pullata (Clerck) and Pardosa prativaga (L.
Koch) in a mixed population (Araneae, Lycosidae). Tijdschrift voor entomologie 115:205--215.
Hollander, J. den. 1971. Life histories of species in the Pardosa pullata group, a study of ten populations in the Netherlands
(Araneae, Lycosidae). Tijdschrift voor Entomologie.
Holttum, S. 2020. Research watch: Coronavirus (COVID-19), mental health and social inclusion in the UK and Ireland.
Mental Health and Social Inclusion 24:117–123.
Horne, C. R., Andrew. G. Hirst, and D. Atkinson. 2015. Temperature-size responses match latitudinal-size clines in
arthropods, revealing critical differences between aquatic and terrestrial species. Ecology Letters 18:327–335.
Houlden, V., S. Weich, J. P. de Albuquerque, S. Jarvis, and K. Rees. 2018. The relationship between greenspace and the
mental wellbeing of adults: A systematic review. PLOS ONE 13:e0203000.
Høye, T. T., and J. U. Hammel. 2010. Climate change and altitudinal variation in sexual size dimorphism of arctic wolf
spiders. Climate Research 41:259–265.
Høye, T. T., J. U. Hammel, T. Fuchs, and S. Toft. 2009. Climate change and sexual size dimorphism in an Arctic spider.
Biology Letters 5:542–544.
Huey, R. B., P. E. Hertz, and B. Sinervo. 2003. Behavioral drive versus behavioral inertia in evolution: a null model
approach. The American Naturalist 161:357–366.
Huey, R. B., M. R. Kearney, A. Krockenberger, J. A. M. Holtum, M. Jess, and S. E. Williams. 2012. Predicting organismal
vulnerability to climate warming: roles of behaviour, physiology and adaptation. Philosophical Transactions of the
Royal Society B: Biological Sciences 367:1665–1679.
Huey, R. B., and J. G. Kingsolver. 1989. Evolution of thermal sensitivity of ectotherm performance. Trends Ecol Evol 4:131–
5.
Huey, R. B., and M. Slatkin. 1976. Cost and benefits of lizard thermoregulation. The Quarterly Review of Biology.
Humphreys, W. F. 1974. Behavioural thermoregulation in a wolf spider. Nature 251:502–503.
Humphreys, W. F. 1978. The thermal biology of Geolycosa godeffroyi and other burrow inhabiting Lycosidae (Araneae) in
Australia. Oecologia 31:319–347.
Hung, T., U. Daisuke, O. Shiro, and Y. Yoshifumi. 2006. Assessment with satellite data of the urban heat island effects in
Asian mega cities. International Journal of Applied Earth Observation and Geoinformation 8:34–48.
IPCC. 2013. Climate Change 2013 the Physical Science Basis: Final Draft Underlying Scientific-Technical Assessment :
Working Group Contribution to the IPCC 5th Assessment Report [Stocker, T.F., D., Qin, G.-K., Plattner, M., Tignor,
S.K., Allen, J., Boschung, A., Nauels, Y., Xia, V., Bex & P.M., Midgley (eds.)]. Cambridge University Press, Cambridge,
UK and New York, USA.
IPCC. 2021. Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment
Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte, V., P. Zhai, A. Pirani, S. L. Connors,
C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M. I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.
K. Maycock, T. Waterfield, O. Yelekçi, R. Yu and B. Zhou (eds.)]. Cambridge University Press, Cambridge, UK and
New York, USA.
Irmler, U., and E. Lipkow. 2018. Effect of Environmental Conditions on Distribution Patterns of Rove Beetles. Pages 117–
144 in O. Betz, U. Irmler, and J. Klimaszewski, editors. Biology of Rove Beetles (Staphylinidae): Life History,
Evolution, Ecology and Distribution. Springer International Publishing, Cham.
Isaksson, C. 2018. Impact of Urbanization on Birds. Pages 235–257 in D. T. Tietze, editor. Bird Species: How They Arise,
Modify and Vanish. Springer International Publishing, Cham.
Ives, C. D., P. E. Lentini, C. G. Threlfall, K. Ikin, D. F. Shanahan, G. E. Garrard, S. A. Bekessy, R. A. Fuller, L. Mumaw, L. Rayner,
R. Rowe, L. E. Valentine, and D. Kendal. 2016. Cities are hotspots for threatened species. Global Ecology and
Biogeography 25:117–126.

209
Références
Jänicke, B., F. Meier, M.-T. Hoelscher, and D. Scherer. 2015. Evaluating the Effects of Façade Greening on Human
Bioclimate in a Complex Urban Environment. Advances in Meteorology 2015:747259.
Jeannel, R. 1941. Faune de France: Coléoptères carabiques. Office central de faunistique.
Jenkins, D. G., C. R. Brescacin, C. V. Duxbury, J. A. Elliott, J. A. Evans, K. R. Grablow, M. Hillegass, B. N. Lyon, G. A. Metzger,
M. L. Olandese, D. Pepe, G. A. Silvers, H. N. Suresch, T. N. Thompson, C. M. Trexler, G. E. Williams, N. C. Williams,
and S. E. Williams. 2007. Does size matter for dispersal distance? Global Ecology and Biogeography 16:415–425.
Johnson, J. C., E. Garver, and T. Martin. 2020a. Black widows on an urban heat island: extreme heat affects spider
development and behaviour from egg to adulthood. Animal Behaviour 167:77–84.
Johnson, J. C., E. Garver, and T. Martin. 2020b. Black widows on an urban heat island: extreme heat affects spider
development and behaviour from egg to adulthood. Animal Behaviour 167:77–84.
Johnson, J. C., J. Urcuyo, C. Moen, and D. R. S. Ii. 2019a. Urban heat island conditions experienced by the Western black
widow spider (Latrodectus hesperus): Extreme heat slows development but results in behavioral
accommodations. PLOS ONE 14:e0220153.
Johnson, J. C., J. Urcuyo, C. Moen, and D. R. II. Stevens. 2019b. Urban heat island conditions experienced by the Western
black widow spider (Latrodectus hesperus): Extreme heat slows development but results in behavioral
accommodations. PLoS One 14:e0220153.
Joly, D., T. Brossard, H. Cardot, J. Cavailhes, M. Hilal, and P. Wavresky. 2010. Les types de climats en France, une
construction spatiale. Cybergeo: European Journal of Geography.
Jones, H. P., D. G. Hole, and E. S. Zavaleta. 2012. Harnessing nature to help people adapt to climate change. Nature
Climate Change 2:504–509.
Kaiser, A., T. Merckx, and H. Van Dyck. 2016. The Urban Heat Island and its spatial scale dependent impact on survival
and development in butterflies of different thermal sensitivity. Ecology and Evolution 6:4129–4140.
Karatzias, T., M. Shevlin, J. Murphy, O. McBride, M. Ben-Ezra, R. P. Bentall, F. Vallières, and P. Hyland. 2020. Posttraumatic
Stress Symptoms and Associated Comorbidity During the COVID-19 Pandemic in Ireland: A Population-Based
Study. Journal of Traumatic Stress 33:365–370.
Katsoulis, B. D., and G. A. Theoharatos. 1985. Indications of the Urban Heat Island in Athens, Greece. Journal of Applied
Meteorology and Climatology 24:1296–1302.
Kawecki, T. J., and D. Ebert. 2004. Conceptual issues in local adaptation. Ecology Letters 7:1225–1241.
Kearney, M., R. Shine, and W. P. Porter. 2009. The potential for behavioral thermoregulation to buffer “cold-blooded”
animals against climate warming. Proceedings of the National Academy of Sciences 106:3835–3840.
Keatinge, W. R., G. C. Donaldson, E. Cordioli, M. Martinelli, A. E. Kunst, J. P. Mackenbach, S. Nayha, and I. Vuori. 2000.
Heat related mortality in warm and cold regions of Europe: observational study. BMJ 321:670–673.
de Keer, R., M. Alderweireldt, K. Decleer, H. Segers, K. Desender, and J.-P. Maelfait. 1989. Horizontal distribution of the
spider fauna of intensively grazed pastures under the influence of diurnal activity and grass height. Journal of
Applied Entomology 107:455–473.
de Keer, R., and J.-P. Maelfait. 1988. Observations on the life cycle of Erigone atra (Araneae, Erigoninae) in a heavily
grazed pasture. Pedobiologia 32:201–212.
Khaliq, I., C. Rixen, F. Zellweger, C. H. Graham, M. M. Gossner, I. R. McFadden, L. Antão, J. Brodersen, S. Ghosh, F. Pomati,
O. Seehausen, T. Roth, T. Sattler, S. R. Supp, M. Riaz, N. E. Zimmermann, B. Matthews, and A. Narwani. 2024.
Warming underpins community turnover in temperate freshwater and terrestrial communities. Nature
Communications 15:1921.
Kingsolver, J., and R. Huey. 2008. Size, temperature, and fitness: three rules. Evolutionary Ecology Research 10:251–268.
Kirkpatrick, M., and N. H. Barton. 1997. Evolution of a Species’ Range. The American Naturalist 150:1–23.
Klaus, V. H. 2013. Urban Grassland Restoration: A Neglected Opportunity for Biodiversity Conservation. Restoration
Ecology 21:665–669.
Klockmann, M., F. Günter, and K. Fischer. 2016. Heat resistance throughout ontogeny: body size constrains thermal
tolerance. Global Change Biology 23:686–696.
Klok, C. J., and J. F. Harrison. 2013. The temperature size rule in arthropods: independent of macro-environmental
variables but size dependent. Integrative and comparative biology 53:557–570.
210
 Références
Knapp, S., M. F. J. Aronson, E. Carpenter, A. Herrera-Montes, K. Jung, D. J. Kotze, F. A. La Sorte, C. A. Lepczyk, I. MacGregor-
Fors, J. S. MacIvor, M. Moretti, C. H. Nilon, M. R. Piana, C. C. Rega-Brodsky, A. Salisbury, C. G. Threlfall, C. Trisos,
N. S. G. Williams, and A. K. Hahs. 2021. A Research Agenda for Urban Biodiversity in the Global Extinction Crisis.
BioScience 71:268–279.
Koivula, M., P. Punttila, Y. Haila, and J. Niemelä. 1999. Leaf Litter and the Small-Scale Distribution of Carabid Beetles
(Coleoptera, Carabidae) in the Boreal Forest. Ecography 22:424–435.
Kozlov, M. V., V. Lanta, V. Zverev, K. Rainio, M. A. Kunavin, and E. L. Zvereva. 2017. Decreased losses of woody plant
foliage to insects in large urban areas are explained by bird predation. Global Change Biology 23:4354–4364.
Krstevska, B., and A. A. Hoffmann. 1994. The effects of acclimation and rearing conditions on the response of tropical and
temperate populations of Drosophila melanogaster and D. simulans to a temperature gradient (Diptera:
Drosophilidae). Journal of Insect Behavior 7:279–288.
Kurucz, K., J. J. Purger, and P. Batáry. 2021. Urbanization shapes bird communities and nest survival, but not their food
quantity. Global Ecology and Conservation 26:e01475.
Kyrö, K., S. Brenneisen, D. J. Kotze, A. Szallies, M. Gerner, and S. Lehvävirta. 2018. Local habitat characteristics have a
stronger effect than the surrounding urban landscape on beetle communities on green roofs. Urban Forestry &
Urban Greening 29:122–130.
Labra, A., J. Pienaar, and T. F. Hansen. 2009. Evolution of Thermal Physiology in Liolaemus Lizards: Adaptation,
Phylogenetic Inertia, and Niche Tracking. The American Naturalist 174:204–220.
Lafage, D., S. Maugenest, J.-B. Bouzillé, and J. Pétillon. 2015. Disentangling the influence of local and landscape factors
on alpha and beta diversities: opposite response of plants and ground-dwelling arthropods in wet meadows.
Ecological Research 30:1025–1035.
Lahr, E. C., R. R. Dunn, and S. D. Frank. 2018. Getting ahead of the curve: cities as surrogates for global change.
Proceedings of the Royal Society B: Biological Sciences 285:20180643.
Lai, D., W. Liu, T. Gan, K. Liu, and Q. Chen. 2019. A review of mitigating strategies to improve the thermal environment
and thermal comfort in urban outdoor spaces. Science of The Total Environment 661:337–353.
Lai, J., Y. Zou, S. Zhang, X. Zhang, and L. Mao. 2022. [Link]: an R package for computing individual effect of predictors
in generalized linear mixed models. Journal of Plant Ecology 15:1302–1307.
Lambers, J. H. R. 2015. Extinction risks from climate change. Science 348:501–502.
Langellotto, G. A., and R. F. Denno. 2004. Responses of invertebrate natural enemies to complex-structured habitats: a
meta-analytical synthesis. Oecologia 139:1–10.
Larochelle, A. 1990. The Food of Carabid Beetles: (coleoptera: Carabidae, Including Cicindelinae). Association des
entomologistes amateurs du Québec.
Le Divelec, R., C. Courtial, and C. Rollard. 2018. Cryptachaea blattea (Urquart, 1886) et Pseudomaro aenigmaticus (Denis,
1966): deux araignées nouvelles pour la France (Araneae: Theridiidae, Linyphiidae). Revue Arachnologique 2:2–6.
Lê, S., J. Josse, and F. Husson. 2008. FactoMineR: an R package for multivariate analysis. Journal of statistical software
25:1–18.
Legendre, P., J. Oksanen, and C. J. F. ter Braak. 2011. Testing the significance of canonical axes in redundancy analysis.
Methods in Ecology and Evolution 2:269–277.
Lembrechts, J. J., J. Aalto, M. B. Ashcroft, P. De Frenne, M. Kopecký, J. Lenoir, M. Luoto, I. M. D. Maclean, O. Roupsard, E.
Fuentes-Lillo, R. A. García, L. Pellissier, C. Pitteloud, J. M. Alatalo, S. W. Smith, R. G. Björk, L. Muffler, A. Ratier
Backes, S. Cesarz, F. Gottschall, J. Okello, J. Urban, R. Plichta, M. Svátek, S. S. Phartyal, S. Wipf, N. Eisenhauer, M.
Pușcaș, P. D. Turtureanu, A. Varlagin, R. D. Dimarco, A. S. Jump, K. Randall, E. Dorrepaal, K. Larson, J. Walz, L. Vitale,
M. Svoboda, R. Finger Higgens, A. H. Halbritter, S. R. Curasi, I. Klupar, A. Koontz, W. D. Pearse, E. Simpson, M.
Stemkovski, B. Jessen Graae, M. Vedel Sørensen, T. T. Høye, M. R. Fernández Calzado, J. Lorite, M. Carbognani, M.
Tomaselli, T. G. W. Forte, A. Petraglia, S. Haesen, B. Somers, K. Van Meerbeek, M. P. Björkman, K. Hylander, S.
Merinero, M. Gharun, N. Buchmann, J. Dolezal, R. Matula, A. D. Thomas, J. J. Bailey, D. Ghosn, G. Kazakis, M. A. de
Pablo, J. Kemppinen, P. Niittynen, L. Rew, T. Seipel, C. Larson, J. D. M. Speed, J. Ardö, N. Cannone, M. Guglielmin,
F. Malfasi, M. Y. Bader, R. Canessa, A. Stanisci, J. Kreyling, J. Schmeddes, L. Teuber, V. Aschero, M. Čiliak, F. Máliš,
P. De Smedt, S. Govaert, C. Meeussen, P. Vangansbeke, K. Gigauri, A. Lamprecht, H. Pauli, K. Steinbauer, M.
Winkler, M. Ueyama, M. A. Nuñez, T.-M. Ursu, S. Haider, R. E. M. Wedegärtner, M. Smiljanic, M. Trouillier, M.
Wilmking, J. Altman, J. Brůna, L. Hederová, M. Macek, M. Man, J. Wild, P. Vittoz, M. Pärtel, P. Barančok, R. Kanka,

211
Références
J. Kollár, A. Palaj, A. Barros, A. C. Mazzolari, M. Bauters, P. Boeckx, J.-L. Benito Alonso, S. Zong, V. Di Cecco, Z.
Sitková, K. Tielbörger, L. van den Brink, R. Weigel, J. Homeier, C. J. Dahlberg, S. Medinets, V. Medinets, H. J. De
Boeck, M. Portillo-Estrada, L. T. Verryckt, A. Milbau, G. N. Daskalova, H. J. D. Thomas, I. H. Myers-Smith, B. Blonder,
J. G. Stephan, P. Descombes, F. Zellweger, E. R. Frei, B. Heinesch, C. Andrews, J. Dick, L. Siebicke, A. Rocha, R. A.
Senior, C. Rixen, J. J. Jimenez, J. Boike, A. Pauchard, T. Scholten, B. Scheffers, D. Klinges, E. W. Basham, J. Zhang, Z.
Zhang, C. Géron, F. Fazlioglu, O. Candan, J. Sallo Bravo, F. Hrbacek, K. Laska, E. Cremonese, P. Haase, F. E. Moyano,
C. Rossi, and I. Nijs. 2020. SoilTemp: A global database of near-surface temperature. Global Change Biology
26:6616–6629.
Lemonsu, A., V. Viguié, M. Daniel, and V. Masson. 2015. Vulnerability to heat waves: Impact of urban expansion scenarios
on urban heat island and heat stress in Paris (France). Urban Climate 14:586–605.
Lenoir, J., J.-C. Gégout, A. Guisan, P. Vittoz, T. Wohlgemuth, N. E. Zimmermann, S. Dullinger, H. Pauli, W. Willner, and J.-
C. Svenning. 2010. Going against the flow: potential mechanisms for unexpected downslope range shifts in a
warming climate. Ecography 33:295–303.
Lenoir, J., T. Hattab, and G. Pierre. 2017. Climatic microrefugia under anthropogenic climate change: implications for
species redistribution. Ecography 40:253–266.
Lenoir, J., and J.-C. Svenning. 2015. Climate-related range shifts – a global multidimensional synthesis and new research
directions. Ecography 38:15–28.
Lenth, R. V. 2023. emmeans: Estimated marginal means. R package version 1.8.6.
Lepczyk, C. A., M. F. J. Aronson, K. L. Evans, M. A. Goddard, S. B. Lerman, and J. S. MacIvor. 2017. Biodiversity in the City:
Fundamental Questions for Understanding the Ecology of Urban Green Spaces for Biodiversity Conservation.
BioScience 67:799–807.
Lepczyk, C. A., M. F. Aronson, and F. A. La Sorte. 2023. Cities as sanctuaries. Frontiers in Ecology and the Environment
21:251–259.
Lerman, S. B., A. R. Contosta, J. Milam, and C. Bang. 2018. To mow or to mow less: Lawn mowing frequency affects bee
abundance and diversity in suburban yards. Biological Conservation 221:160–174.
Levinton, J. S. 1983. The latitudinal compensation hypothesis: growth data and a model of latitudinal growth
differentiation based upon energy budgets. I. Interspecific comparison of Ophryotrocha (Polychaeta: Dorvilleidae).
Biological Bulletin 165:686–698.
Li, J., C. Li, F. Zhu, C. Song, and J. Wu. 2013a. Spatiotemporal pattern of urbanization in Shanghai, China between 1989
and 2005. Landscape Ecology 28:1545–1565.
Li, J., C. Song, L. Cao, F. Zhu, X. Meng, and J. Wu. 2011. Impacts of landscape structure on surface urban heat islands: A
case study of Shanghai, China. Remote Sensing of Environment 115:3249–3263.
Li, T., R. M. Horton, and P. L. Kinney. 2013b. Projections of seasonal patterns in temperature- related deaths for
Manhattan, New York. Nature Climate Change 3:717–721.
Li, Y., M. Zhao, S. Motesharrei, Q. Mu, E. Kalnay, and S. Li. 2015. Local cooling and warming effects of forests based on
satellite observations. Nature Communications 6:6603.
Lin, B.-S., and T.-W. Chen. 2022. The plant and faunal species composition and diversity on rooftop farms: Seasonal
variation and the effects of site and surrounding characteristics. Landscape and Urban Planning 226:104483.
Littmann, T. 2008. Topoclimate and Microclimate. Pages 175–182 in S.-W. Breckle, A. Yair, and M. Veste, editors. Arid
Dune Ecosystems. Springer, Berlin, Heidelberg.
Liu, Z., C. He, and J. Wu. 2016. The Relationship between Habitat Loss and Fragmentation during Urbanization: An
Empirical Evaluation from 16 World Cities. PLOS ONE 11:e0154613.
Llewelyn, J., S. Macdonald, A. Hatcher, C. Moritz, and B. L. Phillips. 2016. Thermoregulatory behaviour explains
countergradient variation in the upper thermal limit of a rainforest skink. Oikos 126:748–757.
Logan, M. L., R. M. Cox, and R. Calsbeek. 2014. Natural selection on thermal performance in a novel thermal environment.
Proceedings of the National Academy of Sciences 111:14165–14169.
Lott, D. A. 2009. The Staphylinidae (rove beetles) of Britain and Ireland. Part 5: Scaphidiinae, Piestinae, Oxytelinae.
Lott, D. A., and R. Anderson. 2011. The Staphylinidae (rove beetles) of Britain and Ireland. Parts 7 & 8: Oxyporinae,
Steninae, Euaesthetinae, Pseudopsinae, Paederinae, Staphylininae.

212
 Références
Lövei, G. L., and T. Magura. 2022a. Body size and the urban heat island effect modulate the temperature–size relationship
in ground beetles. Journal of Biogeography 49:1618–1628.
Lövei, G. L., and T. Magura. 2022b. Body size and the urban heat island effect modulate the temperature–size relationship
in ground beetles. Journal of Biogeography 49:1618–1628.
Lüdecke, D., M. S. Ben-Shachar, I. Patil, P. Waggoner, and D. Makowski. 2021. performance: An R Package for Assessment,
Comparison and Testing of Statistical Models. Journal of Open Source Software 6:3139.
Luder, K., E. Knop, and M. H. M. Menz. 2018. Contrasting responses in community structure and phenology of migratory
and non-migratory pollinators to urbanization. Diversity and Distributions 24:919–927.
Lyons, A., P. A. Ashton, I. Powell, and A. Oxbrough. 2018. Habitat associations of epigeal spiders in upland calcareous
grassland landscapes: the importance for conservation. Biodiversity and conservation 27:1201–1219.
Ma, C. S., G. Ma, and S. Pincebourde. 2021. Survive a Warming Climate: Insect Responses to Extreme High Temperatures.
Annual review of entomology:163–184.
Ma, G., C. Le Lann, J. van Baaren, and C.-S. Ma. 2020. NightWarming Affecting Interspecific Interactions: Implications for
Biological Control. Pages 39–53 in Y. Gao, H. M. T. Hokkanen, and I. Menzler-Hokkanen, editors. Integrative
Biological Control: Ecostacking for Enhanced Ecosystem Services. Springer International Publishing, Cham.
MacGregor-Fors, I., F. Escobar, R. Rueda-Hernández, S. Avendaño-Reyes, M. L. Baena, V. M. Bandala, S. Chacón-Zapata,
A. Guillén-Servent, F. González-García, F. Lorea-Hernández, E. Montes de Oca, L. Montoya, E. Pineda, L. Ramírez-
Restrepo, E. Rivera-García, and E. Utrera-Barrillas. 2016. City “Green” Contributions: The Role of Urban
Greenspaces as Reservoirs for Biodiversity. Forests 7:146.
MacGregor-Fors, I., M. García-Arroyo, and J. Quesada. 2022. Keys to the city: an integrative conceptual framework on
avian urban filtering. Journal of Urban Ecology 8:juac026.
Macías-Hernández, N., C. Ramos, M. Domènech, S. Febles, I. Santos, M. Arnedo, P. Borges, B. Emerson, and P. Cardoso.
2020. A database of functional traits for spiders from native forests of the Iberian Peninsula and Macaronesia.
Biodiversity Data Journal 8:e49159.
MacLean, H. J., J. Overgaard, T. N. Kristensen, C. Lyster, L. Hessner, E. Olsvig, and J. G. Sørensen. 2019. Temperature
preference across life stages and acclimation temperatures investigated in four species of Drosophila. Journal of
Thermal Biology 86:102428.
Maechler, M., P. Rousseeuw, A. Struyf, M. Hubert, K. Hornik, P. Roudier, J. Gonzalez, K. Kozlowski, E. Schubert, and K.
Murphy. 2023, December 1. cluster: “Finding Groups in Data”: Cluster Analysis Extended Rousseeuw et al.
Maechler, M., P. Rousseeuw, A. Struyf, M. Hubert, K. Hornik, M. Studer, P. Roudier, and J. Gonzalez. 2013. Package
‘cluster.’
Maelfait, J.-P., and R. De Keer. 1990. The border zone of an intensively grazed pasture as a corridor for spiders araneae.
Biological Conservation 54:223–238.
Maggini, R., A. Lehmann, M. Kéry, H. Schmid, M. Beniston, L. Jenni, and N. Zbinden. 2011. Are Swiss birds tracking climate
change?: Detecting elevational shifts using response curve shapes. Ecological Modelling 222:21–32.
Magura, T., R. Horvath, and B. Tothmeresz. 2010a. Effects of urbanization on ground-dwelling spiders in forest patches,
in Hungary. Landscape Ecology 25:621–629.
Magura, T., R. Horváth, and B. Tóthmérész. 2010b. Effects of urbanization on ground-dwelling spiders in forest patches,
in Hungary. Landscape Ecology 25:621–629.
Magura, T., and G. L. Lövei. 2021. Consequences of Urban Living: Urbanization and Ground Beetles. Current Landscape
Ecology Reports 6:9–21.
Magura, T., D. Nagy, and B. Tóthmérész. 2013. Rove beetles respond heterogeneously to urbanization. Journal of Insect
Conservation 17:715–724.
Magura, T., B. Tóthmérész, and T. Molnár. 2004. Changes in carabid beetle assemblages along an urbanisation gradient
in the city of Debrecen, Hungary. Landscape Ecology 19:747–759.
Maher, G. M., G. A. Johnson, and J. D. Burdine. 2022. Impervious surface and local abiotic conditions influence arthropod
communities within urban greenspaces. PeerJ 10:e12818.
Malmos, K. G., A. H. Lüdeking, T. Vosegaard, A. Aagaard, J. Bechsgaard, J. G. Sørensen, and T. Bilde. 2021a. Behavioural
and physiological responses to thermal stress in a social spider. Functional Ecology 35:2728–2742.

213
Références
Malmos, K. G., A. H. Lüdeking, T. Vosegaard, A. Aagaard, J. Bechsgaard, J. G. Sørensen, and T. Bilde. 2021b. Behavioural
and physiological responses to thermal stress in a social spider. Functional Ecology 35:2728–2742.
Manso, M., I. Teotónio, C. M. Silva, and C. O. Cruz. 2021. Green roof and green wall benefits and costs: A review of the
quantitative evidence. Renewable and Sustainable Energy Reviews 135:110111.
Marais, E., and S. L. Chown. 2008. Beneficial acclimation and the Bogert effect. Ecology Letters 11:1027–1036.
Marriott, D. 2012. Cryptachaea blattea (Urquhart, 1886) a theridiid new to Great Britain. Newsletter of the British
arachnological Society 123:9–10.
Martin, R. A., L. D. Chick, M. L. Garvin, and S. E. Diamond. 2021. In a nutshell, a reciprocal transplant experiment reveals
local adaptation and fitness trade-offs in response to urban evolution in an acorn-dwelling ant. Evolution 75:876–
887.
Martin, T. L., and R. B. Huey. 2008. Why “suboptimal” is optimal: Jensen’s inequality and ectotherm thermal preferences.
The American Naturalist 171:E102-118.
Masson, V., C. Marchadier, L. Adolphe, R. Aguejdad, P. Avner, M. Bonhomme, G. Bretagne, X. Briottet, B. Bueno, C. de
Munck, O. Doukari, S. Hallegatte, J. Hidalgo, T. Houet, J. Le Bras, A. Lemonsu, N. Long, M.-P. Moine, T. Morel, L.
Nolorgues, G. Pigeon, J.-L. Salagnac, V. Viguié, and K. Zibouche. 2014. Adapting cities to climate change: A systemic
modelling approach. Urban Climate 10:407–429.
Masumoto, T. 1994. Male emergence timing and mating success in the funnel-web spider,Agelena limbata (Araneae:
Agelenidae). Ecological Research 9:159–165.
McCann, K. 2007. Protecting biostructure. Nature 446:29–29.
McCary, M. A., E. Minor, and D. H. Wise. 2018. Covariation between local and landscape factors influences the structure
of ground-active arthropod communities in fragmented metropolitan woodlands. Landscape Ecology 33:225–239.
Mcdonald, R. I., P. Kareiva, and R. T. T. Forman. 2008. The implications of current and future urbanization for global
protected areas and biodiversity conservation. Biological Conservation 141:1695–1703.
McElroy, M. T. 2014. Countergradient variation in locomotor performance of two sympatric Polynesian skinks (Emoia
impar, Emoia cyanura). Physiological and biochemical zoology: PBZ 87:222–230.
McGlynn, T. P., E. K. Meineke, C. A. Bahlai, E. J. Li, E. A. Hartop, B. J. Adams, and B. V. Brown. 2019. Temperature accounts
for the biodiversity of a hyperdiverse group of insects in urban Los Angeles. Proceedings of the Royal Society B:
Biological Sciences 286.
McIntyre, N. E. 2014. Wildlife Responses to Urbanization: Patterns of Diversity and Community Structure in Built
Environments. Pages 103–115 in R. A. McCleery, C. E. Moorman, and M. N. Peterson, editors. Urban Wildlife
conservation: Theory and Practice. Springer US, Boston, MA.
McKinney, M. L. 2002. Urbanization, biodiversity, and conservation. Bioscience 52:883–890.
McKinney, M. L. 2006. Urbanization as a major cause of biotic homogenization. Biological Conservation 127:247–260.
McKinney, M. L. 2008. Effects of urbanization on species richness: A review of plants and animals. Urban Ecosystems
11:161–176.
Mcnett, B. J., and A. L. Rypstra. 2000. Habitat selection in a large orb-weaving spider: vegetational complexity determines
site selection and distribution. Ecological Entomology 25:423–432.
McVittie, A., L. Cole, A. Wreford, A. Sgobbi, and B. Yordi. 2018. Ecosystem-based solutions for disaster risk reduction:
Lessons from European applications of ecosystem-based adaptation measures. International Journal of Disaster
Risk Reduction 32:42–54.
Meineke, E. K., R. R. Dunn, and S. D. Frank. 2014. Early pest development and loss of biological control are associated
with urban warming. Biology Letters 10:20140586.
Meineke, E. K., R. R. Dunn, J. O. Sexton, and S. D. Frank. 2013. Urban Warming Drives Insect Pest Abundance on Street
Trees. PLOS ONE 8.
Meineke, E. K., A. J. Holmquist, G. M. Wimp, and S. D. Frank. 2017a. Changes in spider community composition are
associated with urban temperature, not herbivore abundance. Journal of Urban Ecology 3:juw010.
Meineke, E. K., A. J. Holmquist, G. M. Wimp, and S. D. Frank. 2017b. Changes in spider community composition are
associated with urban temperature, not herbivore abundance. Journal of Urban Ecology 3:juw010.

214
 Références
Menéndez, R., A. González-Megías, O. T. Lewis, M. R. Shaw, and C. D. Thomas. 2008. Escape from natural enemies during
climate-driven range expansion: a case study. Ecological Entomology 33:413–421.
Menke, S. B., B. Guenard, J. O. Sexton, M. D. Weiser, R. R. Dunn, and J. Silverman. 2011. Urban areas may serve as habitat
and corridors for dry-adapted, heat tolerant species; an example from ants. Urban Ecosystems 14:135–163.
Merckx, T., A. Kaiser, and H. Van Dyck. 2018a. Increased body size along urbanization gradients at both community and
intraspecific level in macro-moths. Global Change Biology 24:3837–3848.
Merckx, T., C. Souffreau, A. Kaiser, L. F. Baardsen, T. Backeljau, D. Bonte, K. Brans, M. Cours, M. Dahirel, N. Debortoli, K.
De Wolf, J. M. T. Engelen, D. Fontaneto, A. T. Gianuca, L. Govaert, F. Hendrickx, J. Higuti, L. Lens, K. Martens, H.
Matheve, E. Matthysen, E. Piano, R. Sablon, I. Schon, K. Van Doninck, L. De Meester, and H. Van Dyck. 2018b. Body-
size shifts in aquatic and terrestrial urban communities. Nature 558:113–117.
Merckx, T., C. Souffreau, A. Kaiser, L. F. Baardsen, T. Backeljau, D. Bonte, K. I. Brans, M. Cours, M. Dahirel, N. Debortoli, K.
De Wolf, J. M. T. Engelen, D. Fontaneto, A. T. Gianuca, L. Govaert, F. Hendrickx, J. Higuti, L. Lens, K. Martens, H.
Matheve, E. Matthysen, E. Piano, R. Sablon, I. Schön, K. Van Doninck, L. De Meester, and H. Van Dyck. 2018c. Body-
size shifts in aquatic and terrestrial urban communities. Nature 558:113–116.
Merckx, T., and H. Van Dyck. 2019. Urbanization-driven homogenization is more pronounced and happens at wider spatial
scales in nocturnal and mobile flying insects. Global Ecology and Biogeography 28:1440–1455.
Merilä, J., and A. P. Hendry. 2014. Climate change, adaptation, and phenotypic plasticity: the problem and the evidence.
Evolutionary Applications 7:1–14.
Michaels, K. F. 2007. Using staphylinid and tenebrionid beetles as indicators of sustainable landscape management in
Australia: a review. Australian Journal of Experimental Agriculture 47:435–449.
Michalko, R., S. Pekár, and M. H. Entling. 2019. An updated perspective on spiders as generalist predators in biological
control. Oecologia 189:21–36.
Mills, G. 2014. Urban climatology: History, status and prospects. Urban Climate 10:479–489.
Mimet, A., V. Pellissier, H. Quénol, R. Aguejdad, V. Dubreuil, and F. Rozé. 2009. Urbanisation induces early flowering:
evidence from Platanus acerifolia and Prunus cerasus. International Journal of Biometeorology 53:287–298.
Mitchell, S. E., and W. Lampert. 2000. Temperature adaptation in a geographically widespread zooplankter, Daphnia
magna. Journal of Evolutionary Biology 13:371–382.
Modelski, G. 2003. World Cities: -3000 to 2000. FAROS 2000, Washington, DC.
Mohan, M., Y. Kikegawa, B. Gurjar, S. Bhati, A. Kandya, and K. Ogawa. 2009. Assessment of urban heat island intensities
over Delhi.
Moll, R. J., J. D. Cepek, P. D. Lorch, P. M. Dennis, E. Tans, T. Robison, J. J. Millspaugh, and R. A. Montgomery. 2019. What
does urbanization actually mean? A framework for urban metrics in wildlife research. Journal of Applied Ecology
56:1289–1300.
Moretti, M., A. T. C. Dias, F. de Bello, F. Altermatt, S. L. Chown, F. M. Azcárate, J. R. Bell, B. Fournier, M. Hedde, J. Hortal,
S. Ibanez, E. Öckinger, J. P. Sousa, J. Ellers, and M. P. Berg. 2017. Handbook of protocols for standardized
measurement of terrestrial invertebrate functional traits. Functional Ecology 31:558–567.
Morris, C. j. g., and I. Simmonds. 2000. Associations between varying magnitudes of the urban heat island and the synoptic
climatology in Melbourne, Australia. International Journal of Climatology 20:1931–1954.
Morris, C. J. G., I. Simmonds, and N. Plummer. 2001. Quantification of the Influences of Wind and Cloud on the Nocturnal
Urban Heat Island of a Large City. Journal of Applied Meteorology and Climatology 40:169–182.
Morris, M. G. 2000. The effects of structure and its dynamics on the ecology and conservation of arthropods in British
grasslands. Biological Conservation 95:129–142.
Morse, D. H. 1997. Distribution, movement, and activity patterns of an intertidal wolf spider Pardosa lapidicina population
(Araneae, Lycosidae). Journal of Arachnology:1–10.
Morse, D. H. 2002. Orientation and movement of wolf spiders Pardosa lapidicina (Araneae, Lycosidae) in the intertidal
zone. The Journal of Arachnology 30:601–609.
Moser, A., T. Rötzer, S. Pauleit, and H. Pretzsch. 2015. Structure and ecosystem services of small-leaved lime (Tilia cordata
Mill.) and black locust (Robinia pseudoacacia L.) in urban environments. Urban Forestry & Urban Greening
14:1110–1121.

215
Références
Mousseau, T. A., and D. A. Roff. 1987. Natural selection and the heritability of fitness components. Heredity 59:181–197.
de Munck, C., A. Lemonsu, V. Masson, J. Le Bras, and M. Bonhomme. 2018. Evaluating the impacts of greening scenarios
on thermal comfort and energy and water consumptions for adapting Paris city to climate change. Urban Climate
23:260–286.
Muñoz, M. M. 2022. The Bogert effect, a factor in evolution. Evolution 76:49–66.
Muñoz, M. M., G. M. Langham, M. C. Brandley, D. F. Rosauer, S. E. Williams, and C. Moritz. 2016. Basking behavior predicts
the evolution of heat tolerance in Australian rainforest lizards. Evolution 70:2537–2549.
Muster, C., M. Krebs, and H. Joosten. 2020. Seven years of spider community succession in a Sphagnum farm. The Journal
of Arachnology 48:119–131.
Mutani, G., and V. Todeschi. 2020. The Effects of Green Roofs on Outdoor Thermal Comfort, Urban Heat Island Mitigation
and Energy Savings. Atmosphere 11:123.
Nardi, D., and L. Marini. 2021. Role of abandoned grasslands in the conservation of spider communities across
heterogeneous mountain landscapes. Agriculture, Ecosystems & Environment 319:107526.
Neel, L. K., and L. D. McBrayer. 2018. Habitat management alters thermal opportunity. Functional Ecology 32:2029–2039.
Neil, K., and J. Wu. 2006. Effects of urbanization on plant flowering phenology: A review. Urban Ecosystems 9:243–257.
Nelson, K. C., and M. A. Palmer. 2007. Stream Temperature Surges Under Urbanization and Climate Change: Data, Models,
and Responses1. JAWRA Journal of the American Water Resources Association 43:440–452.
Nentwig, W. 1988. Augmentation of beneficial arthropods by strip-management. Oecologia 76:597–606.
Nentwig, W., T. Blick, R. Bosmans, D. Gloor, A. Hänggi, and K. Christian. 2023. Spiders of Europe - Version 09.2023.
[Link]
Nevo, E., E. Rashkovetsky, T. Pavlicek, and A. Korol. 1998. A complex adaptive syndrome in Drosophila caused by
microclimatic contrasts. Heredity 80:9–16.
Nielsen, M. E., and D. R. Papaj. 2015. Effects of developmental change in body size on ectotherm body temperature and
behavioral thermoregulation: caterpillars in a heat-stressed environment. Oecologia 177:171–179.
Niemelä, J., and D. J. Kotze. 2009. Carabid beetle assemblages along urban to rural gradients: A review. Landscape and
Urban Planning 92:65–71.
Niemelä, J., D. J. Kotze, S. Venn, L. Penev, I. Stoyanov, J. Spence, D. Hartley, and E. M. de Oca. 2002. Carabid beetle
assemblages (Coleoptera, Carabidae) across urban-rural gradients: an international comparison. Landscape
Ecology 17:387–401.
Nolan, M. 2020. Notes on ‘irish’ spiders (Arachnida): Atypus affinis EICHWALD (Atypidae) and Entelecara errata O.P.-
CAMBRIDGE (Linyphiidae) removed from the Irish list; Cryptachaea blattea (Urquhart)(Theridiidae) new to Ireland.
Bulletin of the Irish Biogeographical Society:189–204.
Nørgaard, E. 1951. On the Ecology of Two Lycosid Spiders (Pirata piraticus and Lycosa pullata) from a Danish Sphagnum
bog. Oikos 3:1–21.
Norton, B. A., G. D. Bending, R. Clark, R. Corstanje, N. Dunnett, K. L. Evans, D. R. Grafius, E. Gravestock, S. M. Grice, J. A.
Harris, S. Hilton, H. Hoyle, E. Lim, T. G. Mercer, M. Pawlett, O. L. Pescott, J. P. Richards, G. E. Southon, and P. H.
Warren. 2019. Urban meadows as an alternative to short mown grassland: effects of composition and height on
biodiversity. Ecological Applications 29:e01946.
Norton, B. A., A. M. Coutts, S. J. Livesley, R. J. Harris, A. M. Hunter, and N. S. G. Williams. 2015. Planning for cooler cities:
A framework to prioritise green infrastructure to mitigate high temperatures in urban landscapes. Landscape and
Urban Planning 134:127–138.
Norton, B. A., L. J. Thomson, N. S. G. Williams, and M. J. McDonnell. 2014. The effect of urban ground covers on
arthropods: An experiment. Urban Ecosystems 17:77–99.
Nunez, M., and T. R. Oke. 1977. The Energy Balance of an Urban Canyon. Journal of Applied Meteorology and Climatology
16:11–19.
Oke, T. R. 1982. The energetic basis of the urban heat island. Quarterly Journal of the Royal Meteorological Society 108:1–
24.

216
 Références
Oke, T. R. 1995. The Heat Island of the Urban Boundary Layer: Characteristics, Causes and Effects. Pages 81–107 in J. E.
Cermak, A. G. Davenport, E. J. Plate, and D. X. Viegas, editors. Wind Climate in Cities. Springer Netherlands,
Dordrecht.
Oke, T. R., G. Mills, A. Christen, and J. A. Voogt. 2017. Urban Climates. Cambridge University Press, Cambridge.
Oksanen, J., F. G. Blanchet, M. Friendly, R. Kindt, P. Legendre, D. McGlinn, P. R. Minchin, R. B. O’Hara, G. L. Simpson, P.
Solymos, M. H. H. Stevens, E. Szoecs, and H. Wagner. 2019. vegan: Community Ecology Package. R package version
2.5-6. [Link] (accessed 01.12.2020).
Oksanen, J., G. L. Simpson, F. G. Blanchet, R. Kindt, P. Legendre, P. R. Minchin, R. B. O’Hara, P. Solymos, M. H. H. Stevens,
E. Szoecs, H. Wagner, M. Barbour, M. Bedward, B. Bolker, D. Borcard, G. Carvalho, M. Chirico, M. D. Caceres, S.
Durand, H. B. A. Evangelista, R. FitzJohn, M. Friendly, B. Furneaux, G. Hannigan, M. O. Hill, L. Lahti, D. McGlinn, M.-
H. Ouellette, E. R. Cunha, T. Smith, A. Stier, C. J. F. T. Braak, and J. Weedon. 2022, October 11. vegan: Community
Ecology Package. R package version 2.6-4. [Link] (accessed 01.12.2020).
Oliver, T. H., M. S. Heard, N. J. B. Isaac, D. B. Roy, D. Procter, F. Eigenbrod, R. Freckleton, A. Hector, C. D. L. Orme, O. L.
Petchey, V. Proença, D. Raffaelli, K. B. Suttle, G. M. Mace, B. Martín-López, B. A. Woodcock, and J. M. Bullock.
2015. Biodiversity and Resilience of Ecosystem Functions. Trends in Ecology & Evolution 30:673–684.
O’Neill, B. C., C. Tebaldi, D. P. van Vuuren, V. Eyring, P. Friedlingstein, G. Hurtt, R. Knutti, E. Kriegler, J.-F. Lamarque, J.
Lowe, G. A. Meehl, R. Moss, K. Riahi, and B. M. Sanderson. 2016. The Scenario Model Intercomparison Project
(ScenarioMIP) for CMIP6. Geoscientific Model Development 9:3461–3482.
ONU. 2018. World Urbanization Prospects: The 2018 Revision. United Nations, Department of Economic and social affairs,
Population Division, United Nations, New York.
Oostra, V., M. Saastamoinen, B. J. Zwaan, and C. W. Wheat. 2018. Strong phenotypic plasticity limits potential for
evolutionary responses to climate change. Nature Communications 9:1005.
Opell, B. D. 2010. Bergmanns’s size cline in New Zealand marine spray zone spiders (Araneae: Anyphaenidae:
Amaurobioides). Biological Journal of the Linnean Society 101:78–92.
Orlanski, I. 1975. A rational subdivision of scales for atmospheric processes. Bulletin of the American Meteorological
Society:527–530.
Ortega, Z., A. Mencía, and V. Pérez-Mellado. 2016. Adaptive seasonal shifts in the thermal preferences of the lizard
Iberolacerta galani (Squamata, Lacertidae). Journal of Thermal Biology 62:1–6.
Otoshi, M. D., P. Bichier, and S. M. Philpott. 2015. Local and Landscape Correlates of Spider Activity Density and Species
Richness in Urban Gardens. Environmental Entomology 44:1043–1051.
Paranjpe, D. A., E. Bastiaans, A. Patten, R. D. Cooper, and B. Sinervo. 2013. Evidence of maternal effects on temperature
preference in side-blotched lizards: implications for evolutionary response to climate change. Ecology and
Evolution 3:1977–1991.
Parmesan, C. 2006. Ecological and Evolutionary Responses to Recent Climate Change. Annual Review of Ecology,
Evolution, and Systematics 37:637–669.
Parmesan, C. 2007. Influences of species, latitudes and methodologies on estimates of phenological response to global
warming. Global Change Biology 13:1860–1872.
Parmesan, C., N. Ryrholm, C. Stefanescu, J. K. Hill, C. D. Thomas, H. Descimon, B. Huntley, L. Kaila, J. Kullberg, T. Tammaru,
W. J. Tennent, J. A. Thomas, and M. Warren. 1999. Poleward shifts in geographical ranges of butterfly species
associated with regional warming. Nature 399:579–583.
Parmesan, C., and G. Yohe. 2003. A globally coherent fingerprint of climate change impacts across natural systems. Nature
421:37–42.
Pebesma, E., R. Bivand, E. Racine, M. Sumner, I. Cook, T. Keitt, R. Lovelace, H. Wickham, J. Ooms, K. Müller, T. L. Pedersen,
D. Baston, and D. Dunnington. 2024, March 24. sf: Simple Features for R.
Pecl, G. T., M. B. Araújo, J. D. Bell, J. Blanchard, T. C. Bonebrake, I.-C. Chen, T. D. Clark, R. K. Colwell, F. Danielsen, B.
Evengård, L. Falconi, S. Ferrier, S. Frusher, R. A. Garcia, R. B. Griffis, A. J. Hobday, C. Janion-Scheepers, M. A. Jarzyna,
S. Jennings, J. Lenoir, H. I. Linnetved, V. Y. Martin, P. C. McCormack, J. McDonald, N. J. Mitchell, T. Mustonen, J. M.
Pandolfi, N. Pettorelli, E. Popova, S. A. Robinson, B. R. Scheffers, J. D. Shaw, C. J. B. Sorte, J. M. Strugnell, J. M.
Sunday, M.-N. Tuanmu, A. Vergés, C. Villanueva, T. Wernberg, E. Wapstra, and S. E. Williams. 2017. Biodiversity
redistribution under climate change: Impacts on ecosystems and human well-being. Science 355:eaai9214.

217
Références
Peng, M.-H., Y.-C. Hung, K.-L. Liu, and K.-B. Neoh. 2020. Landscape configuration and habitat complexity shape arthropod
assemblage in urban parks. Scientific Reports 10:16043.
Peng, S.-S., S. Piao, Z. Zeng, P. Ciais, L. Zhou, L. Z. X. Li, R. B. Myneni, Y. Yin, and H. Zeng. 2014. Afforestation in China cools
local land surface temperature. Proceedings of the National Academy of Sciences 111:2915–2919.
Peres-Neto, P. R., P. Legendre, S. Dray, and D. Borcard. 2006. Variation Partitioning of Species Data Matrices: Estimation
and Comparison of Fractions. Ecology 87:2614–2625.
Pétillon, J., A. Georges, A. Canard, J.-C. Lefeuvre, J. P. Bakker, and F. Ysnel. 2008. Influence of abiotic factors on spider and
ground beetle communities in different salt-marsh systems. Basic and Applied Ecology 9:743–751.
Pétillon, J., K. Lambeets, B. Ract-Madoux, P. Vernon, and D. Renault. 2011. Saline stress tolerance partly matches with
habitat preference in ground-living wolf spiders. Physiological Entomology 36:165–172.
Pétillon, J., B. Leroy, E. A. Djoudi, and V. Vedel. 2018. Small and large spatial scale coexistence of ctenid spiders in a
neotropical forest (French Guiana). Tropical Zoology 31:85–98.
Philpott, S., J. Cotton, P. Bichier, R. Friedrich, L. Moorhead, S. Uno, and M. Valdez. 2014. Local and landscape drivers of
arthropod abundance, richness, and trophic composition in urban habitats. Urban Ecosystems 17:513–532.
Philpott, S. M., S. Albuquerque, P. Bichier, H. Cohen, M. H. Egerer, C. Kirk, and K. W. Will. 2019. Local and Landscape
Drivers of Carabid Activity, Species Richness, and Traits in Urban Gardens in Coastal California. Insects 10:112.
Piano, E., F. Bona, and M. Isaia. 2020a. Urbanization drivers differentially affect ground arthropod assemblages in the city
of Turin (NW-Italy). Urban Ecosystems 23:617–629.
Piano, E., K. De Wolf, F. Bona, D. Bonte, D. E. Bowler, M. Isaia, L. Lens, T. Merckx, D. Mertens, M. Van Kerckvoorde, L. De
Meester, and F. Hendrickx. 2017. Urbanization drives community shifts towards thermophilic and dispersive
species at local and landscape scales. Global Change Biology 23:2554–2564.
Piano, E., D. Giuliano, and M. Isaia. 2020b. Islands in cities: Urbanization and fragmentation drive taxonomic and
functional variation in ground arthropods. Basic and Applied Ecology 43:86–98.
Piano, E., D. Giuliano, and M. Isaia. 2020c. Islands in cities: Urbanization and fragmentation drive taxonomic and functional
variation in ground arthropods. Basic and Applied Ecology 43:86–98.
Piano, E., C. Souffreau, T. Merckx, L. F. Baardsen, T. Backeljau, D. Bonte, K. I. Brans, M. Cours, M. Dahirel, N. Debortoli, E.
Decaestecker, K. De Wolf, J. M. T. Engelen, D. Fontaneto, A. T. Gianuca, L. Govaert, F. T. T. Hanashiro, J. Higuti, L.
Lens, K. Martens, H. Matheve, E. Matthysen, E. Pinseel, R. Sablon, I. Schön, R. Stoks, K. Van Doninck, H. Van Dyck,
P. Vanormelingen, J. Van Wichelen, W. Vyverman, L. De Meester, and F. Hendrickx. 2020d. Urbanization drives
cross-taxon declines in abundance and diversity at multiple spatial scales. Global Change Biology 26:1196–1211.
Pincebourde, S., C. C. Murdock, M. Vickers, and M. W. Sears. 2016. Fine-Scale Microclimatic Variation Can Shape the
Responses of Organisms to Global Change in Both Natural and Urban Environments. Integrative and Comparative
Biology 56:45–61.
Pincebourde, S., and A. Salle. 2020. On the importance of getting fine-scale temperature records near any surface. Global
Change Biology 26:6025–6027.
Pincebourde, S., and H. A. Woods. 2020. There is plenty of room at the bottom: microclimates drive insect vulnerability
to climate change. Current Opinion in Insect Science 41:63–70.
Piracha, A., and M. T. Chaudhary. 2022. Urban Air Pollution, Urban Heat Island and Human Health: A Review of the
Literature. Sustainability 14:9234.
Planchuelo, G., I. Kowarik, and M. von der Lippe. 2020. Endangered Plants in Novel Urban Ecosystems Are Filtered by
Strategy Type and Dispersal Syndrome, Not by Spatial Dependence on Natural Remnants. Frontiers in Ecology and
Evolution 8.
Poortinga, W., N. Bird, B. Hallingberg, R. Phillips, and D. Williams. 2021. The role of perceived public and private green
space in subjective health and wellbeing during and after the first peak of the COVID-19 outbreak. Landscape and
Urban Planning 211:104092.
Porter, W. P., J. W. Mitchell, W. A. Beckman, and C. B. DeWitt. 1973. Behavioral implications of mechanistic ecology :
Thermal and behavioral modeling of desert ectotherms and their microenvironment. Oecologia 13:1–54.
Potter, K. A., H. Arthur Woods, and S. Pincebourde. 2013. Microclimatic challenges in global change biology. Global
Change Biology 19:2932–2939.

218
 Références
Proske, A., S. Lokatis, and J. Rolff. 2022. Impact of mowing frequency on arthropod abundance and diversity in urban
habitats: A meta-analysis. Urban forestry & urban greening 76:127714-.
Purvis, G., and J. P. Curry. 1980. Successional Changes in the Arthropod Fauna of a New Ley Pasture Established on
Previously Cultivated Arable Land. Journal of Applied Ecology 17:309–321.
Quénol, H., V. Dubreuil, A. Mimet, V. Pellissier, R. Aguejdad, P. Clergeau, and S. Bridier. 2010. Climat urbain et impact sur
la phénologie végétale printanière. La Météorologie 2010:50–57.
R Core Team. 2022. R: A language and environment for statistical computing. R Foundation for Statistical Computing,
Vienna, Austria.
Rainio, J., and J. Niemelä. 2003. Ground beetles (Coleoptera: Carabidae) as bioindicators. Biodiversity & Conservation
12:487–506.
Rao, D. 2017. Habitat Selection and Dispersal. Pages 85–108 in C. Viera and M. O. Gonzaga, editors. Behaviour and Ecology
of Spiders: Contributions from the Neotropical Region. Springer International Publishing, Cham.
Rasband, W. 2007. ImageJ. U.S. National Institutes of Health, Bethesda, Maryland, USA. [Link] ij/,1997–
2012.
Refinetti, R. 1984. Behavioral Temperature Regulation in the Pill Bug, Armadillidium vulgare (Isopoda). Crustaceana
47:29–43.
Reiser, S., J.-P. Herrmann, and A. Temming. 2014. Thermal preference of the common brown shrimp (Crangon crangon,
L.) determined by the acute and gravitational method. Journal of Experimental Marine Biology and Ecology
461:250–256.
Relys, V., S. Koponen, and D. Dapkus. 2002. Annual differences and species turnover in peat bog spider communities. The
Journal of Arachnology 30:416–424.
Renaud, V., and M. Rebetez. 2009. Comparison between open-site and below-canopy climatic conditions in Switzerland
during the exceptionally hot summer of 2003. Agricultural and Forest Meteorology 149:873–880.
Rennes Métropole. 2019. Plan Climat- Air-Énergie Territorial 2019-2024. Service Transition Energétique et Ecologique,
Direction Economie Emploi Innovation.
Reynolds, W. W., and M. E. Casterlin. 1979. Behavioral Thermoregulation and the “Final Preferendum” Paradigm.
American Zoologist 19:211–224.
Richard, Y., B. Pohl, M. Rega, J. Pergaud, T. Thevenin, J. Emery, J. Dudek, T. Vairet, S. Zito, and C. Chateau-Smith. 2021. Is
Urban Heat Island intensity higher during hot spells and heat waves (Dijon, France, 2014–2019)? Urban Climate
35:100747.
Richter, C. J. 1970. Aerial dispersal in relation to habitat in eight wolf spider species (Pardosa, Araneae, Lycosidae).
Oecologia 5:200–214.
Roberts, D. W. 2016. labdsv: Ordination and multivariate analysis for ecology.
Roberts, M. J. 1985. The spiders of Great Britain and Ireland, Volume 1: Atypidae to Theridiosomatidae. Harley Books,
Colchester, UK.
Roberts, M. J. 1987. The spiders of Great Britain and Ireland, Volume 2: Linyphiidae and check list. Harley Books,
Colchester, UK.
Roetzer, T., M. Wittenzeller, H. Haeckel, and J. Nekovar. 2000. Phenology in central Europe – differences and trends of
spring phenophases in urban and rural areas. International Journal of Biometeorology 44:60–66.
Root, T. L., J. T. Price, K. R. Hall, S. H. Schneider, C. Rosenzweig, and J. A. Pounds. 2003. Fingerprints of global warming on
wild animals and plants. Nature 421:57–60.
Roy, D. B., and T. H. Sparks. 2000. Phenology of British butterflies and climate change. Global Change Biology 6:407–416.
Ruhland, F., J. Pétillon, and M. Trabalon. 2016. Physiological costs during the first maternal care in the wolf spider Pardosa
saltans (Araneae, Lycosidae). Journal of Insect Physiology 95:42–50.
Rutschmann, A., C. Perry, J.-F. Le Galliard, A. Dupoué, O. Lourdais, M. Guillon, G. Brusch IV, J. Cote, M. Richard, J. Clobert,
and D. B. Miles. 2024. Ecological responses of squamate reptiles to nocturnal warming. Biological Reviews 99:598–
621.

219
Références
Rypstra, A. L., P. E. Carter, R. A. Balfour, and S. D. Marshall. 1999. Architectural Features of Agricultural Habitats and Their
Impact on the Spider Inhabitants. The Journal of Arachnology 27:371–377.
Sadler, J., E. Small, H. Fiszpan, M. Telfer, and J. Niemelä. 2006. Investigating environmental variation and landscape
characteristics of an urban–rural gradient using woodland carabid assemblages. Journal of biogeography 33:1126–
1138.
Sagarin, R. D., S. D. Gaines, and B. Gaylord. 2006. Moving beyond assumptions to understand abundance distributions
across the ranges of species. Trends in Ecology & Evolution 21:524–530.
Sala, O. E., F. Stuart Chapin, III, J. J. Armesto, E. Berlow, J. Bloomfield, R. Dirzo, E. Huber-Sanwald, L. F. Huenneke, R. B.
Jackson, A. Kinzig, R. Leemans, D. M. Lodge, H. A. Mooney, M. Oesterheld, N. L. Poff, M. T. Sykes, B. H. Walker, M.
Walker, and D. H. Wall. 2000. Global Biodiversity Scenarios for the Year 2100. Science 287:1770–1774.
Salmond, J., C. E. Sabel, and S. Vardoulakis. 2018. Towards the Integrated Study of Urban Climate, Air Pollution, and Public
Health. Climate 6:14.
Samu, F., J. Németh, F. Tóth, É. Szita, B. Kiss, and C. Szinetár. 1998. Are two cohorts responsible for bimodal life history
pattern in the wolf spider Pardosa agrestis in Hungary. Pages 215–221.
Samu, F., K. D. Sunderland, and C. Szinetar. 1999. Scale-dependent dispersal and distribution patterns of spiders in
agricultural systems: a review. Journal of Arachnology:325–332.
San Martin y Gomez, G., and H. Van Dyck. 2012. Ecotypic differentiation between urban and rural populations of the
grasshopper Chorthippus brunneus relative to climate and habitat fragmentation. Oecologia 169:125–133.
Sarthou, J.-P., A. Badoz, B. Vaissière, A. Chevallier, and A. Rusch. 2014. Local more than landscape parameters structure
natural enemy communities during their overwintering in semi-natural habitats. Agriculture, Ecosystems &
Environment 194:17–28.
Sato, A., and Y. Takahashi. 2022. Responses in thermal tolerance and daily activity rhythm to urban stress in Drosophila
suzukii. Ecology and Evolution 12:e9616.
Sattler, T., D. Borcard, R. Arlettaz, F. Bontadina, P. Legendre, M. Obrist, and M. Moretti. 2010. Spider, bee, and bird
communities in cities are shaped by environmental control and high stochasticity. Ecology 91:3343–3353.
Savolainen, O., M. Lascoux, and J. Merilä. 2013. Ecological genomics of local adaptation. Nature Reviews Genetics 14:807–
820.
Schär, C., and G. Jendritzky. 2004. Hot news from summer 2003. Nature 432:559–560.
Scheffers, B. R., L. De Meester, T. C. L. Bridge, A. A. Hoffmann, J. M. Pandolfi, R. T. Corlett, S. H. M. Butchart, P. Pearce-
Kelly, K. M. Kovacs, D. Dudgeon, M. Pacifici, C. Rondinini, W. B. Foden, T. G. Martin, C. Mora, D. Bickford, and J. E.
M. Watson. 2016. The broad footprint of climate change from genes to biomes to people. Science 354:aaf7671.
Scheffers, B. R., T. A. Evans, S. E. Williams, and D. P. Edwards. 2014. Microhabitats in the tropics buffer temperature in a
globally coherent manner. Biology Letters 10:20140819.
Schirmel, J., I. Blindow, and S. Buchholz. 2012. Life-history trait and functional diversity patterns of ground beetles and
spiders along a coastal heathland successional gradient. Basic and Applied Ecology 13:606–614.
Schneider, A., M. A. Friedl, and D. Potere. 2009. A new map of global urban extent from MODIS satellite data.
Environmental Research Letters 4:044003.
Schwartz, M. W., N. L. Jurjavcic, and J. M. O’Brien. 2002. Conservation’s Disenfranchised Urban Poor. BioScience 52:601–
606.
Schwarz, N., M. Moretti, M. N. Bugalho, Z. G. Davies, D. Haase, J. Hack, A. Hof, Y. Melero, T. J. Pett, and S. Knapp. 2017.
Understanding biodiversity-ecosystem service relationships in urban areas: A comprehensive literature review.
Ecosystem Services 27:161–171.
Scolozzi, R., and D. Geneletti. 2012. A multi-scale qualitative approach to assess the impact of urbanization on natural
habitats and their connectivity. Environmental Impact Assessment Review 36:9–22.
Sears, M. W., and M. J. Angilletta. 2015. Costs and Benefits of Thermoregulation Revisited: Both the Heterogeneity and
Spatial Structure of Temperature Drive Energetic Costs. The American Naturalist 185:E94–E102.
Sears, M. W., M. J. Angilletta, M. S. Schuler, J. Borchert, K. F. Dilliplane, M. Stegman, T. W. Rusch, and W. A. Mitchell.
2016. Configuration of the thermal landscape determines thermoregulatory performance of ectotherms.
Proceedings of the National Academy of Sciences of the United States of America 113:10595–10600.

220
 Références
Sehrt, M., O. Bossdorf, M. Freitag, and A. Bucharova. 2020. Less is more! Rapid increase in plant species richness after
reduced mowing in urban grasslands. Basic and Applied Ecology 42:47–53.
Seto, K. C., B. Güneralp, and L. R. Hutyra. 2012. Global forecasts of urban expansion to 2030 and direct impacts on
biodiversity and carbon pools. Proceedings of the National Academy of Sciences 109:16083–16088.
Sevacherian, V., and D. C. Lowrie. 1972. Preferred temperatures of two species of Lycosid spiders, Pardosa sierra and P.
ramulosa. Annals of the Entomological Society of America 65:111–114.
Shochat, E., W. L. Stefanov, M. E. A. Whitehouse, and S. H. Faeth. 2004. Urbanization and Spider Diversity: Influences of
Human Modification of Habitat Structure and Productivity. Ecological Applications 14:268–280.
Shwartz, A., A. Turbé, R. Julliard, L. Simon, and A.-C. Prévot. 2014. Outstanding challenges for urban conservation research
and action. Global Environmental Change 28:39–49.
Simonneau, M., C. Courtial, and J. Pétillon. 2016. Phenological and meteorological determinants of spider ballooning in
an agricultural landscape. Comptes Rendus Biologies 339:408–416.
Simpson, M. R. 1993. Reproduction in two species of arctic arachnids, Pardosa glacialis and Alopecosa hirtipes. Canadian
journal of zoology 71:451–457.
Sinclair, B. J., K. E. Marshall, M. A. Sewell, D. L. Levesque, C. S. Willett, S. Slotsbo, Y. Dong, C. D. G. Harley, D. J. Marshall,
B. S. Helmuth, and R. B. Huey. 2016. Can we predict ectotherm responses to climate change using thermal
performance curves and body temperatures? Ecology Letters 19:1372–1385.
Singh, N., S. Singh, and R. K. Mall. 2020. Chapter 17 - Urban ecology and human health: implications of urban heat island,
air pollution and climate change nexus. Pages 317–334 in P. Verma, P. Singh, R. Singh, and A. S. Raghubanshi,
editors. Urban Ecology. Elsevier.
Skelhorn, C., S. Lindley, and G. Levermore. 2014. The impact of vegetation types on air and surface temperatures in a
temperate city: A fine scale assessment in Manchester, UK. Landscape and Urban Planning 121:129–140.
Small, E. C., J. P. Sadler, and M. G. Telfer. 2002. Carabid beetle assemblages on urban derelict sites in Birmingham, UK.
Journal of Insect Conservation 6:233–246.
Soanes, K., and P. E. Lentini. 2019. When cities are the last chance for saving species. Frontiers in Ecology and the
Environment 17:225–231.
Solascasas, P., F. M. Azcárate, and V. Hevia. 2022. Edaphic arthropods as indicators of the ecological condition of
temperate grassland ecosystems: A systematic review. Ecological Indicators 142:109277.
Sommer, R. J. 2020. Phenotypic Plasticity: From Theory and Genetics to Current and Future Challenges. Genetics 215:1–
13.
Sorte, C. J. B., S. L. Williams, and J. T. Carlton. 2010. Marine range shifts and species introductions: comparative spread
rates and community impacts. Global Ecology and Biogeography 19:303–316.
Speights, C. J., J. P. Harmon, and B. T. Barton. 2017. Contrasting the potential effects of daytime versus nighttime warming
on insects. Current Opinion in Insect Science 23:1–6.
Speights, C. J., C. L. Wolff, M. E. Barton, and B. T. Barton. 2018. Why and How to Create Nighttime Warming Treatments
for Ecological Field Experiments. The Yale Journal of Biology and Medicine 91:471–480.
Spence, J. R., and J. K. Niemelä. 1994. Sampling carabid assemblages with pitfall traps: the madness and the method. The
Canadian Entomologist 126:881–894.
Spotswood, E. N., E. E. Beller, R. Grossinger, J. L. Grenier, N. E. Heller, and M. F. J. Aronson. 2021. The Biological Deserts
Fallacy: Cities in Their Landscapes Contribute More than We Think to Regional Biodiversity. BioScience 71:148–
160.
Stearns, S. C. 1989. The Evolutionary Significance of Phenotypic Plasticity: Phenotypic sources of variation among
organisms can be described by developmental switches and reaction norms. BioScience 39:436–445.
Stevenson, R. D. 1985. The relative importance of behavioral and physiological adjustments controlling body temperature
in terrestrial ectotherms. The American Naturalist 126:362–386.
Stewart, I. D., and T. R. Oke. 2012. Local Climate Zones for Urban Temperature Studies. Bulletin of the American
Meteorological Society 93:1879–1900.

221
Références
Stillwell, R. C., W. U. Blanckenhorn, T. Teder, G. Davidowitz, and C. W. Fox. 2010. Sex differences in phenotypic plasticity
affect variation in sexual size dimorphism in insects: from physiology to evolution. Annual review of entomology
55:227–245.
Stork, R., P. Smith, C. Aaen, and S. Cooper. 2021. Field observations of body temperature for the wolf spider, Rabidosa
rabida (Araneae: Lycosidae), differ from reported laboratory temperature preference suggesting
thermoconforming behavior. Environmental Entomology 50:405–409.
Sturman, A., and H. Quénol. 2023. Climate Change - Impacts and Adaptation at Regional and Local Scales. Oxford
University Press.
Sührig, A. 2010. Cryptachaea blattea , eine weitere nach Deutschland eingeschleppte Spinnenart (Araneae: Theridiidae).
Arachnologische Mitteilungen 39:1–4.
Sunday, J. M., A. E. Bates, M. R. Kearney, R. K. Colwell, N. K. Dulvy, J. T. Longino, and R. B. Huey. 2014. Thermal-safety
margins and the necessity of thermoregulatory behavior across latitude and elevation. Proceedings of the National
Academy of Sciences 111:5610–5615.
Swenson, J. J., and J. Franklin. 2000. The effects of future urban development on habitat fragmentation in the Santa
Monica Mountains. Landscape Ecology 15:713–730.
Syphard, A. D., K. C. Clarke, J. Franklin, H. M. Regan, and M. Mcginnis. 2011. Forecasts of habitat loss and fragmentation
due to urban growth are sensitive to source of input data. Journal of Environmental Management 92:1882–1893.
Szulkin, M., C. J. Garroway, M. Corsini, A. Z. Kotarba, and D. Dominoni. 2020. How to Quantify Urbanization When Testing
for Urban Evolution? Page 0 in M. Szulkin, J. Munshi-South, and A. Charmantier, editors. Urban Evolutionary
Biology. Oxford University Press.
Theodorou, P. 2022. The effects of urbanisation on ecological interactions. Current Opinion in Insect Science 52:100922.
Theodorou, P., R. Radzevičiūtė, B. Kahnt, A. Soro, I. Grosse, and R. J. Paxton. 2018. Genome-wide single nucleotide
polymorphism scan suggests adaptation to urbanization in an important pollinator, the red-tailed bumblebee
(Bombus lapidarius L.). Proceedings of the Royal Society B: Biological Sciences 285:20172806.
Thomas, J. A., D. J. Simcox, and R. T. Clarke. 2009. Successful Conservation of a Threatened Maculinea Butterfly. Science
325:80–83.
Töpfer-Hofmann, G., D. Cordes, and O. v Helversen. 2000. Cryptic species and behavioural isolation in the Pardosa lugubris
group (Araneae, Lycosidae), with description of two new species. Bulletin of the British Arachnological Society
11:257–274.
Topping, C. J., and K. D. Sunderland. 1992. Limitations to the Use of Pitfall Traps in Ecological Studies Exemplified by a
Study of Spiders in a Field of Winter Wheat. Journal of Applied Ecology 29:485–491.
Tougeron, K., and P. K. Abram. 2017. An Ecological Perspective on Sleep Disruption. The American Naturalist.
Trabalon, M. 2022. Effects of wolf spiders’ captive environment on their locomotor and exploratory behaviours. Insects
13:135.
Trochet, A., A. Dupoué, J. Souchet, R. Bertrand, M. Deluen, S. Murarasu, O. Calvez, A. Martinez-Silvestre, I. Verdaguer-
Foz, E. Darnet, H. L. Chevalier, M. Mossoll-Torres, O. Guillaume, and F. Aubret. 2018. Variation of preferred body
temperatures along an altitudinal gradient: A multi-species study. Journal of Thermal Biology 77:38–44.
Twohig-Bennett, C., and A. Jones. 2018. The health benefits of the great outdoors: A systematic review and meta-analysis
of greenspace exposure and health outcomes. Environmental Research 166:628–637.
Tyler, G. 2008. Differences in abundance, species richness, and body size of ground beetles (Coleoptera: Carabidae)
between beech (Fagus sylvatica L.) forests on Podzol and Cambisol. Forest Ecology and Management 256:2154–
2159.
Tylianakis, J. M., R. K. Didham, J. Bascompte, and D. A. Wardle. 2008. Global change and species interactions in terrestrial
ecosystems. Ecology Letters 11:1351–1363.
Uetz, G. W. 1979. The influence of variation in litter habitats on spider communities. Oecologia 40:29–42.
Uetz, G. W. 1991. Habitat structure and spider foraging. Pages 325–348 in S. S. Bell, E. D. McCoy, and H. R. Mushinsky,
editors. Habitat Structure: The physical arrangement of objects in space. Springer Netherlands, Dordrecht.
Uetz, G. W., J. Halaj, and A. B. Cady. 1999. Guild Structure of Spiders in Major Crops. The Journal of Arachnology 27:270–
280.

222
 Références
Uhl, G., S. Schmitt, M. A. Schäfer, and W. Blanckenhorn. 2004. Food and sex-specific growth strategies in a spider.
Evolutionary Ecology Research 6:523–540.
UICN, OFB, MNHN, and AsFrA. 2023. La liste rouge des espèces menacées en France - Chapitre Araignées de France
métropolitaine. Paris, France.
Unger, J. 2004. Intra-urban relationship between surface geometry and urban heat island: review and new approach.
Climate Research 27:253–264.
United Nations. 2019. Department of Economic and Social Affairs, Population Division. World Urbanization Prospects:
The 2018 Revision (ST/ESA/SER.A/420). New York: United Nations.
Varet, M., F. Burel, D. Lafage, and J. Pétillon. 2013. Age-dependent colonization of urban habitats: a diachronic approach
using carabid beetles and spiders. Animal Biology 63:257–269.
Varet, M., F. Burel, and J. Pétillon. 2014a. Can urban consolidation limit local biodiversity erosion? Responses from carabid
beetle and spider assemblages in Western France. URBAN ECOSYSTEMS 17:123–137.
Varet, M., F. Burel, and J. Petillon. 2014b. Can urban consolidation limit local biodiversity erosion? Responses from carabid
beetle and spider assemblages in Western France. Urban Ecosystems 17:123–137.
Varet, M., J. Pétillon, and F. Burel. 2011a. Comparative responses of spider and carabid beetle assemblages along an
urban-rural boundary gradient. The Journal of Arachnology 39:236–243.
Varet, M., J. Pétillon, and F. Burel. 2011b. Comparative responses of spider and carabid beetle assemblages along an
urban-rural boundary gradient. The Journal of Arachnology 39:236–243.
Vergés, A., P. D. Steinberg, M. E. Hay, A. G. B. Poore, A. H. Campbell, E. Ballesteros, K. L. Heck, D. J. Booth, M. A. Coleman,
D. A. Feary, W. Figueira, T. Langlois, E. M. Marzinelli, T. Mizerek, P. J. Mumby, Y. Nakamura, M. Roughan, E. van
Sebille, A. S. Gupta, D. A. Smale, F. Tomas, T. Wernberg, and S. K. Wilson. 2014. The tropicalization of temperate
marine ecosystems: climate-mediated changes in herbivory and community phase shifts. Proceedings of the Royal
Society B: Biological Sciences 281:20140846.
Vergnes, A., I. Le Viol, and P. Clergeau. 2012. Green corridors in urban landscapes affect the arthropod communities of
domestic gardens. BIOLOGICAL CONSERVATION 145:171–178.
Vergnes, A., V. Pellissier, G. Lemperiere, C. Rollard, and P. Clergeau. 2014. Urban densification causes the decline of
ground-dwelling arthropods. Biodiversity and Conservation 23:1859–1877.
Verheyen, J., N. Tüzün, and R. Stoks. 2019. Using natural laboratories to study evolution to global warming: contrasting
altitudinal, latitudinal, and urbanization gradients. Current Opinion in Insect Science 35:10–19.
Visser, M. E., and C. Both. 2005. Shifts in phenology due to global climate change: the need for a yardstick. Proceedings
of the Royal Society B: Biological Sciences 272:2561–2569.
Vlijm, L., and C. Richter. 1966. Activity fluctuations of Pardosa lugubris (Walckenaer), Araneae Lycosidae, during the
breeding-season. Entomologische Berichten 26:222–230.
Wang, X., H. Li, and S. Sodoudi. 2022. The effectiveness of cool and green roofs in mitigating urban heat island and
improving human thermal comfort. Building and Environment 217:109082.
Wanninger, A. 2015. Morphology is dead – long live morphology! Integrating MorphoEvoDevo into molecular EvoDevo
and phylogenomics. Frontiers in Ecology and Evolution 3.
Ward, D. F., T. R. New, and A. L. Yen. 2001. Effects of Pitfall Trap Spacing on the Abundance, Richness and Composition
of Invertebrate Catches. Journal of Insect Conservation 5:47–53.
Wastian, L., P. A. Unterweger, and O. Betz. 2016. Influence of the reduction of urban lawn mowing on wild bee diversity
(Hymenoptera, Apoidea). Journal of Hymenoptera Research 49:51–63.
Weisser, W. W., and T. E. Hauck. 2017, June 15. ANIMAL-AIDED DESIGN – using a species’ life-cycle to improve open space
planning and conservation in cities and elsewhere. bioRxiv.
Weyman, G. S. 1995. Laboratory Studies of the Factors Stimulating Ballooning Behavior by Linyphiid Spiders (Araneae,
Linyphiidae). The Journal of Arachnology 23:75–84.
White, M. A., R. R. Nemani, P. E. Thornton, and S. W. Running. 2002. Satellite Evidence of Phenological Differences
Between Urbanized and Rural Areas of the Eastern United States Deciduous Broadleaf Forest. Ecosystems 5:260–
273.

223
Références
Wilson, R. J., S. J. Brooks, and P. B. Fenberg. 2019. The influence of ecological and life history factors on ectothermic
temperature–size responses: Analysis of three Lycaenidae butterflies (Lepidoptera). Ecology and Evolution
9:10305–10316.
Wilson-Brodie, R. J., M. A. MacLean, and P. B. Fenberg. 2017. Historical shell size reduction of the dogwhelk (Nucella
lapillus) across the southern UK. Marine Biology 164:190.
Wise, D. 1993. Spiders in ecological webs Cambridge University Press. New York ISBN-13:978–0521.
WMO. 2010. Manual on the Global Observing System. World Meteorological Organization, Geneva.
Wolkovich, E. M., B. I. Cook, J. M. Allen, T. M. Crimmins, J. L. Betancourt, S. E. Travers, S. Pau, J. Regetz, T. J. Davies, N. J.
B. Kraft, T. R. Ault, K. Bolmgren, S. J. Mazer, G. J. McCabe, B. J. McGill, C. Parmesan, N. Salamin, M. D. Schwartz,
and E. E. Cleland. 2012. Warming experiments underpredict plant phenological responses to climate change.
Nature 485:494–497.
Woodcock, B. A., S. G. Potts, T. Tscheulin, E. Pilgrim, A. J. Ramsey, J. Harrison-Cripps, V. K. Brown, and J. R. Tallowin. 2009.
Responses of invertebrate trophic level, feeding guild and body size to the management of improved grassland
field margins. Journal of Applied Ecology 46:920–929.
Woods, H. A., M. E. Dillon, and S. Pincebourde. 2015. The roles of microclimatic diversity and of behavior in mediating
the responses of ectotherms to climate change. Journal of Thermal Biology 54:86–97.
Wu, J., S. Li, N. Shen, Y. Zhao, and H. Cui. 2020. Construction of Cooling Corridors with Multiscenarios on Urban Scale: A
Case Study of Shenzhen. Sustainability 12:5903.
Yamamoto, A., and S. Ohba. 1984. Temperature preferences of eleven Drosophila species from Japan: The relationship
between preferred temperature and some ecological characteristics in their natural habitats. Zoological Science
1:631–640.
Yang, J., Q. Wu, R. Xiao, J. Zhao, J. Chen, and X. Jiao. 2018. Seasonal variations in body melanism and size of the wolf
spider Pardosa astrigera (Araneae: Lycosidae). Ecology and Evolution 8:4352–4359.
Yilmaz, A. R., A. Yoder, S. E. Diamond, and R. A. Martin. 2022. Adaptation to Urban Heat Islands Enhances Thermal
Performance Following Development under Chronic Thermal Stress but Not Benign Conditions in the Terrestrial
Isopod Oniscus asellus. Physiological and Biochemical Zoology 95:302–316.
Youngsteadt, E., A. G. Dale, A. J. Terando, R. R. Dunn, and S. D. Frank. 2015. Do cities simulate climate change? A
comparison of herbivore response to urban and global warming. Global Change Biology 21:97–105.
Youngsteadt, E., A. F. Ernst, R. R. Dunn, and S. D. Frank. 2017. Responses of arthropod populations to warming depend
on latitude: evidence from urban heat islands. Global Change Biology 23:1436–1447.
Youngsteadt, E., S. G. Prado, K. J. Keleher, and M. Kirchner. 2023. Can behaviour and physiology mitigate effects of
warming on ectotherms? A test in urban ants. Journal of Animal Ecology 92:568–579.
Zellweger, F., D. Coomes, J. Lenoir, L. Depauw, S. L. Maes, M. Wulf, K. J. Kirby, J. Brunet, M. Kopecký, F. Máliš, W. Schmidt,
S. Heinrichs, J. den Ouden, B. Jaroszewicz, G. Buyse, F. Spicher, K. Verheyen, and P. De Frenne. 2019. Seasonal
drivers of understorey temperature buffering in temperate deciduous forests across Europe. Global Ecology and
Biogeography 28:1774–1786.
Zhang, D. B. 2017. A Coefficient of Determination for Generalized Linear Models. American Statistician 71:310–316.
Zhao, F., W. Zhang, A. A. Hoffmann, and C.-S. Ma. 2014a. Night warming on hot days produces novel impacts on
development, survival and reproduction in a small arthropod. Journal of Animal Ecology 83:769–778.
Zhao, L., X. Lee, R. B. Smith, and K. Oleson. 2014b. Strong contributions of local background climate to urban heat islands.
Nature 511:216–219.
Ziter, C. D., E. J. Pedersen, C. J. Kucharik, and M. G. Turner. 2019. Scale-dependent interactions between tree canopy
cover and impervious surfaces reduce daytime urban heat during summer. Proceedings of the National Academy
of Sciences 116:7575–7580.
Zölch, T., J. Maderspacher, C. Wamsler, and S. Pauleit. 2016. Using green infrastructure for urban climate-proofing: An
evaluation of heat mitigation measures at the micro-scale. Urban Forestry & Urban Greening 20:305–316.
Zuur, A., E. N. leno, N. Walker, A. A. Saveliev, and G. M. Smith. 2009. Mixed Effects Models and Extensions in Ecology with
R. Springer-Verlag, New York.

224
 Références

225
Références

Effets de la variabilité microclimatique sur la faune en milieu urbain dans un contexte de changement climatique
Mots-clés : Arthropodes, Araneae, Carabidae, Staphylinidae, Diversité fonctionnelle, Diversité taxonomique,
Infrastructures vertes, Îlots de chaleur urbains, Urbanisation
Le réchauffement climatique affecte les écosystèmes. Les relativement importante sont effectivement les plus
écosystèmes urbains sont particulièrement vulnérables au sensibles aux réchauffements, possiblement en raison de
réchauffement, de par leur exposition aux îlots de chaleur leur cycle de vie plus long, ou encore de leur capacité
urbains (ICU). Cette thèse aborde les relations entre limitée à disperser par voie aérienne (ballooning).
différents groupes d’arthropodes (principalement Cependant, la qualité structurelle et thermique du micro-
araignées) et les variations spatiales de température habitat peut atténuer ces effets négatifs en favorisant la
mesurées sur la ville de Rennes et ses alentours. En diversité fonctionnelle et l’abondance. Les micro-habitats
particulier, l’effet du réchauffement sur les arthropodes est prairiaux, bien qu’exposés à des conditions thermiques
comparé entre diverses échelles spatiales, s’étendant du diurnes extrêmes en ville, peuvent atténuer, voire
micro-habitat au paysage. La diversité taxonomique est compenser le réchauffement nocturne induit par l’ICU
négativement reliée à l’intensité de l’ICU. Cette réduction atmosphérique à l’échelle des arthropodes. Bien que le
du nombre d’espèces en ville correspond notamment à réchauffement atmosphérique à l’échelle du paysage
une exclusion des espèces les plus sensibles à la chaleur. En urbain constitue un filtre environnemental fort sur les
plus de leurs affinités thermiques, les espèces sont filtrées espèces, des mesures de gestion locales visant à
selon leurs caractéristiques morphologiques. La taille optimiser la qualité thermique des micro-habitats
corporelle moyenne au sein des communautés diminue peuvent constituer un levier efficace pour la conservation
avec une intensité d’ICU croissante. Les espèces de taille de la biodiversité urbaine.

Effects of microclimatic variability on the urban fauna in a context of climate change

Keywords: Arthropods, Araneae, Carabidae, Staphylinidae Functional diversity, Green infrastructure, Taxonomic diversity,
Urbanisation, Urban heat islands

Global warming affects ecosystems. Urban ecosystems are intensity. Relatively large species are indeed the most
particularly vulnerable to warming, due to their exposure sensitive to warming, possibly because of their longer life
to urban heat islands (UHI). This thesis addresses the cycle, or their limited ability to aerially disperse
relationships between different groups of arthropods (ballooning). However, the structural and thermal quality
(mainly spiders) and spatial variations in temperature of the micro-habitat can mitigate these negative effects
measured in and around the city of Rennes. In particular, by promoting functional diversity and abundance.
the effect of warming on arthropods is compared between Grassland micro-habitats, although exposed to extreme
various spatial scales, ranging from the micro-habitat to the daytime thermal conditions in the city, can attenuate or
landscape. Taxonomic diversity is negatively related to the even compensate for the nocturnal warming induced by
UHI intensity. This reduction in the total number of species atmospheric UHI at the arthropod scale. Although
in the city corresponds in particular to the exclusion of atmospheric warming at the scale of the urban landscape
heat-sensitive species. In addition to their thermal constitutes a strong environmental filter on species, local
affinities, species are filtered according to their management measures designed to optimizing the
morphological characteristics. The average body size thermal quality of micro-habitats can efficiently support
within communities decreases with increasing UHI urban biodiversity conservation.

226