Dendrochronologia 93 (2025) 126378

Contents lists available at ScienceDirect

Dendrochronologia
journal homepage: [Link]/locate/dendro

Original article

Examining post-drought recovery in three sub-Mediterranean species: Some

trees not affected while some never recover
Luka Krajnc a ,∗, Polona Hafner a , Jernej Jevšenak a , Şule Ceyda Izmir c , Mitja Ferlan b ,
Jožica Gričar a
a
Gozdarski inštitut Slovenije, Ljubljana, Slovenia
b Slovenian Environment Agency, Ljubljana, Slovenia
c Istanbul University-Cerrahpaşa, Faculty of Forestry, Forest Botany Department, Istanbul, Turkey

ARTICLE INFO ABSTRACT

Keywords: The study aimed to evaluate and compare growth resilience to hotter droughts of three common sub-
Hotter droughts Mediterranean species: black pine (Pinus nigra Arnold), manna ash (Fraxinus ornus L.) and pubescent oak
Resilience (Quercus pubescens Willd). Over 200 trees were sampled across eight different sites in two countries, Slovenia
Recovery
and Italy. Our primary objective was to evaluate and compare diversity in response to drought within individual
Tree mortality
species. Increment cores were extracted from trees from each site, which were then processed and their tree-
Climate-growth relationships
ring width measured. Potential droughts were identified using a Summer-Heat moisture index in the following
years 1983, 1985, 1992, 2000 and 2003. Radial growth resilience was analyzed using two indices, post-drought
recovery period and average relative growth reduction. Climate-growth relationships were also examined. The
individual droughts affected the majority of studied trees. Of the three examined species, F. ornus appears to
be the least sensitive to drought stress. Across sites, species or individual droughts, a small proportion of trees
always remained unaffected by the droughts. Some of the affected trees never recovered their radial growth
within the studied period, indicating that their recovery period exceeded 20 years after individual drought
events. The proportion of both varied between species, site and drought. These facts would indicate that more
emphasis should be given in the future to studying ‘‘winner’’ trees, since they may hold the answer to species’
better adaptability to the occurrence of hotter droughts.

1. Introduction to cause mortality during prolonged droughts (Choat et al., 2018). It

was found recently that the frequency of mortality-triggering extreme
The sub-Mediterranean region in Europe is a gradual transition climate conditions increases non-linearly with warming (Hammond
from the Mediterranean to the Euro-Siberian phytogeographical zone et al., 2022). A tree’s sensitivity to environmental factors is therefore
(Sánchez et al., 2004). In this area, the climatic traits of the two most expressed in harsh conditions, for example in drought-prone
geographical areas are mixed and have intermediate characteristics of Mediterranean environments. Since in Mediterranean-type regions, hot-
both (Sánchez De Dios et al., 2009), with common dry and hot summers ter droughts and other extreme events (e.g., fires) are expected to
and fairly harsh winters (Gričar et al., 2019). The sub-Mediterranean cause significant challenges for the performance of trees and forest
region is highly sensitive to changes in climate (Lazoglou et al., 2024). ecosystems in the near future (Pachauri et al., 2014; Köhl et al., 2020),
This will also affect ecosystem functioning and result in a high turnover adaptation to these changes and the capacity of forests to mitigate
of species (Benito Garzón et al., 2008). climate change is being intensively studied.
Tree mortality is an important component of forest dynamics. Based Nevertheless, there is high diversity among tree species in responses
on climate model simulations, it is expected to increase in the future to drought stress (Grossiord et al., 2014). Their resistance, avoidance or
due to the occurrence of large-scale events and higher global tem- tolerance are based on various structural and/or physiological adjust-
peratures (Cailleret et al., 2019; Gazol and Camarero, 2022). Hotter ments (Bréda et al., 2006). These in turn contribute to the functional
droughts, defined as compound drought and heat-wave events, are im- balance among the tree organs and tissues responsible for water ac-
portant climate drivers of tree mortality (Allen et al., 2015). However, quisition, transport and transpiration, while preserving carbon gain,
it remains unclear how tree water relations and carbon balance interact

∗ Corresponding author.
E-mail address: [Link]@[Link] (L. Krajnc).

[Link]
Received 6 November 2024; Received in revised form 12 May 2025; Accepted 11 June 2025
Available online 1 July 2025
1125-7865/© 2025 The Authors. Published by Elsevier GmbH. This is an open access article under the CC BY license ([Link]
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

Fig. 1. Site location in North-East Adriatic region, country names in bold.

which is important for growth and supporting stress response mecha- of sub-Mediterranean climate from south to north and were selected
nisms (Sterck et al., 2008, e.g.). Apart from evolutionary adaptations for their species composition. First, we identified potentially suitable
to limited water availability and to other associated stresses, plants stands based on species’ distribution maps for F. ornus, P. nigra and Q.
have a certain phenotypic plasticity, adjusting structure and function in pubescens. The goal was to get as many different locations as possible
response to environmental conditions (Valladares et al., 2007). Short- with at least two of the three species co-occurring at the same location.
and long-term growth responses of trees to drought are often studied Unfortunately, due to the nature of the three species, this was not possi-
using tree-ring data, which allow a retrospective quantification of ble in all selected locations, so in two sites we only sampled one species.
drought effects at annual resolution for many individuals, sites and The sites can further be grouped by their climate into three clusters by
species (DeSoto et al., 2020). Several tree ring studies across various using mean annual precipitation and mean annual temperature from
tree species have shown drought-induced reductions in radial growth 1975–2023 as entry variables for clustering. The cluster analysis was
and resilience associated with higher mortality risk (Gazol et al., 2018; done by using K-means to group sites into three clusters ranging from
DeSoto et al., 2020). Moreover, forest resilience is related to drought drier to wetter micro-climate, i.e., sub-Mediterranean, transitional and
severity and forest composition (Gazol et al., 2018) and to taxonomic continental. The result of clustering can be seen in Fig. A.2.
groups of gymnosperms and angiosperms having different resilience
strategies (DeSoto et al., 2020). 2.2. Collection of tree-ring widths
Growth resilience can be defined as the capacity of a tree to reach
growth rates similar to those prior to drought. Defined in this way, At each site, at least 15 dominant living trees were cored using
resilience encompasses the capacity to reduce the impact of the dis- an increment borer in 2022, see Table 1. The sample size was deter-
turbance, i.e., resistance, and the ability to return to pre-disturbance mined following the guidelines proposed by Cherubini et al. (2020).
growth levels after drought, i.e., recovery (DeSoto et al., 2011, 2020). Additional data recorded for each tree was diameter at breast height
The current study aimed to evaluate and compare long-term recovery and height. No other stand characteristics were recorded, none of the
from hotter droughts of three common sub-Mediterranean species: stands were managed forests.
black pine (Pinus nigra Arnold), manna ash (Fraxinus ornus L.) and The extracted cores were then dried, glued into wooden holders
pubescent oak (Quercus pubescens Willd). This was achieved by ex- and scanned using the ATRICS system (Levanič, 2007), which uses
amining tree-ring widths of over 200 trees from eight different sites a stereo microscope with a digital camera to record the individual
in two countries, Slovenia and Italy. Both black pine and pubescent cores. The resulting images were measured using CooRecorder software
oak are relatively well studied species, while there has been relatively (Cybis Elektronik & Data AB) and analyzed using the R programming
little research done on manna ash. Our primary objectives were to (1) language (R. Core Team, 2021) using the package dplR (Bunn et al.,
assess and compare the drought response of studied species and, (2) 2023; Bunn, 2008). Based on Fig. A.1, it was decided to restrict the
to evaluate the climate-growth relationships among studied sites and starting year of the analysis to 1974, since this was roughly the year
species. in which most of the sampled trees were older than 20 years and
therefore considered to be mature with the exception of ash trees in
2. Material & methods the Kobjeglava site. The exact age at which a tree transitions between
juvenile and adult wood production cannot be determined in advance,
2.1. Sampling design as the change in properties is gradual with age. The transition age
from juvenile to mature wood depends on species, but varies from
Eight different sites were selected for analysis in southwest Slovenia 10–20 years (Abdel-Gadir and Krahmer, 1993; Gapare et al., 2006;
and northwest Italy (see Fig. 1). The sites were chosen along a gradient Adamopoulos et al., 2011; Giagli et al., 2017). Considering the species

2
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

Table 1 Table 2
Properties of sampled trees, sample size (N), mean diameter at breast height (DBH) in Median and maximum recovery period for tree-ring width in years, displaying median
centimeters (coefficient of variation in brackets), mean height (H) in meters (coefficient time to recovery and maximum recovery period in brackets, both in years. With the
of variation in brackets). exception of 1985, all differences in median recovery periods between species were
Site Species N DBH (cm) H (m) statistically significant (Kruskal–Wallis test, p < 0.05, confirmed with post-hoc analysis
by Dunn test with Bonferroni method for multiple comparisons).
F. ornus 15 13.1 (11) 10.5 (9)
Kobjeglava 1983 1985 1992 2000 2003
P. nigra 23 48.6 (9) 25.2 (11)
FO 0.6 (17.5) 2.3 (11.6) 1.6 (12.9) 0.7 (15.4) 0.6 (14.9)
F. ornus 15 12.9 (13) 11.9 (29) PN 1.6 (7.0) 3.7 (8.7) 4.2 (15.9) 3.8 (19.8) 3.9 (16.8)
Krempa QP 3.9 (14.7) 2.5 (5.7) 4.2 (17.9) 0.6 (14.0) 1.9 (7.8)
P. nigra 21 52.7 (12) 21.2 (13)
Osp Q. pubescens 12 35.5 (19)
F. ornus 15 14.1 (11) 8.8 (22)
Prosecco
Q. pubescens 15 20.5 (17) 7.9 (18) as the time needed to reach pre-disturbance growth levels, while the
F. ornus 15 14.6 (12) 11.5 (13) latter is the average growth reduction for the recovery period after
Rupinpiccolo
Q. pubescens 16 23.5 (16) 9.6 (13)
disturbances, shown as a percentage. It was calculated as the ratio of
total growth reduction divided by the length of the recovery period.
Sezana P. nigra 21 50.9 (11) 19.0 (19)
Both indices were calculated on the level of individual trees with the R
F. ornus 12 14.5 (24) 12.6 (37)
Trebusa
package pointRes (van der Maaten-Theunissen et al., 2015, 2021) using
P. nigra 21 39.1 (14) 20.8 (15) a maximum length of the recovery period of 20 years. The data used
Vremscica Q. pubescens 12 28.7 (19) were raw values of tree-ring width without any detrending to avoid
misinterpreting actual tree growth (Mazza and Sarris, 2021). All of the
trees included in this study were considered to be mature (over 20
years of age), and the effect of different tree age was disregarded in
analyzed in the cited studies, the transition age for species in the
the analysis. The statistical significance of potential differences between
current study can safely be estimated to between 10–15 years, enabling
species or sites was examined using the Kruskal–Wallis test (Kruskal
the inclusion of the ash trees from Kobjeglava in the analysis.
and Wallis, 1952), the post-hoc analysis was done with the Dunn
test (Dunn, 1964).
2.3. Climate data and drought determination
2.5. Climate-growth relationships
To evaluate the long-term climate conditions at the study sites,
we used the ClimateDT portal (Marchi et al., 2024), which provides
To study the relationships between climate and tree-growth, we
interpolated climate data with a high spatial and temporal resolution.
used the monthly_response() function from the dendroTools R pack-
Potential droughts were identified using the Summer-Heat moisture
age (Jevšenak and Levanič, 2018; Jevšenak, 2020), and calculated the
index (SHM), which is calculated by dividing Mean Warmest Month
Pearson correlation coefficients from both the previous and current
Temperature by Mean Summer (May–Sep) Precipitation (Marchi et al.,
years, employing window sizes ranging from 1 to 17 months, starting
2020). This index is considered to be representative of drought ex-
in June of the preceding year and ending in October of the current
posure, since higher values of SHM index have higher temperatures year. The time period used was all available tree-ring width data. To
and less precipitation. SHM index was used for drought evaluation control for age-related trends and other non-climatic factors affecting
over more frequently used SPEI (Vicente-Serrano et al., 2010) due to growth, we detrended the raw growth data with a cubic smoothing
sensitivity of SPEI calculations on choosing the estimation method for spline, using the dplR R package (Bunn, 2008). This spline applied a
potential evapotranspiration (Stagge et al., 2014), which can subse- 50% frequency cut-off at a 30-year wavelength (Klesse, 2021). We then
quently change the SPEI results. Based on the SHM index (Fig. 2), looked at autocorrelation and enhanced the high-frequency growth
we determined that the most extreme droughts occurred at individual signals by pre-weighting with a first-order autoregressive model (Cook
sites in the years 1983, 1985, 1992, 2000 and 2003. Later droughts and Kairiukstis, 1990). Finally, we used robust bi-weighted averages to
were not examined, since we were exclusively interested in droughts develop residual chronologies. To estimate the quality of the developed
that occurred at least 20 years before the year of sampling in order to chronologies, we calculated the coherence coefficient (Gleichläufigkeit
account for potentially longer recovery periods. Since the last major - GLK) (Buras and Wilmking, 2015) and the mean correlation between
drought occurred roughly 20 years before our sampling year (2022), series (rbar), which indicates the strength of the common signal (Cook
this period was deemed the most appropriate as it enabled us to and Kairiukstis, 1990; Wigley et al., 1984). To test the suitability
study all of the five droughts and their effect on our sample trees. of the created chronologies for climate-growth analysis, we applied
The drought severity of the five identified droughts roughly follows the sub-sample signal strength (SSS), first-order autocorrelation (AR1),
an inverse chronological order, with 2003 being the most extreme of and expressed population signal (EPS) (Table A.2). Climate-growth
the included five droughts. Due to their age, the data from ash trees correlations were calculated for precipitation, minimum and maximum
from site Kobjeglava and their resilience indicators are only comparable temperatures (all obtained from interpolated gridded data), considering
with other species/sites for the last two droughts, as they cannot be all years with subsample signal strength (SSS) above 0.90.
considered mature before 1994 using the criteria mentioned above.
Drought legacy effect was not considered to be an issue, as droughts 3. Results
were at least one growing season apart. Several recent studies have
reported that drought legacy effect in trees is relatively short-term and 3.1. Post-drought recovery period
rarely exceeds one year (Gazol et al., 2020; Kannenberg et al., 2019;
Miller et al., 2023). On average (Table 2), F. ornus recovered the fastest in the drought
of 2003 (median of 0.6 years), with P. nigra and Q. pubescens on average
2.4. The analysis of radial growth resilience taking considerably longer (3.9 and 1.9 years for the drought of 2003)
to recover to pre-drought growth rates. There are noticeable differences
As per the recommendations of Schwarz et al. (2020), the radial in recovery periods between droughts of different intensities.
growth resilience was analyzed using two indices, post-drought recov- When looking at maximum recovery periods of individual species
ery period and average relative growth reduction. The former is defined (Table 2), smaller differences can be observed between species and

3
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

Fig. 2. Summer-Heat moisture index (SHM) for sampled locations, identified droughts marked with a dashed vertical line (1983, 1985, 1992, 2000 and 2003).

Table 3 Table 4
Median recovery period for tree-ring width in years. The median time in years Median relative average growth reduction in proportion for analyzed species and years.
and number of trees are shown in brackets, grouping sites into clusters based on 1983 1985 1992 2000 2003
precipitation and temperature (see Appendix). Significant differences in recovery
periods between species are marked with * (Kruskal–Wallis test, p < 0.05, see Appendix FO 0.20 0.21 0.15 0.21 0.37
for a post-hoc analysis). PN 0.22 0.18 0.24 0.28 0.25
QP 0.18 0.13 0.18 0.12 0.15
Cluster Species 1983 1985 1992 2000 2003
FO 1.2 (14) 1.5 (8)* 1.7 (13) 0.7 (20)* 0.5 (25)*
sub-Mediterranean PN 6.1 (10) 4.1 (11)* 4.5 (7) 3.8 (11)* 0.5 (15)*
QP 2.5 (6) 1.9 (14)* 4.2 (17) 0.6 (15)* 2.0 (27)* 3.2. Post-drought relative radial increment reduction
FO 0.5 (12)* 2.8 (14) 1.3 (5) 0.6 (19)* 0.8 (25)*
transitional PN 1.6 (27)* 3.4 (7) 4.3 (6) 6.2 (26)* 4.4 (36)*
QP 6.0 (3)* 2.8 (10) 5.2 (6) 0.6 (10)* 1.4 (14)* Looking at the average reduction in tree-ring width of the drought-
FO 0.4 (6) 1.3 (4) 2.6 (5) 1.4 (7) 0.7 (7)
affected trees relative to the pre-drought growth (Fig. 3), the reduc-
continental tion almost never exceeded 50%. In the majority of droughts and
PN 1.4 (14) 4.0 (1) 4.1 (4) 1.8 (10) 3.6 (6)
species, the tree-ring width reduction rarely exceeded 30 %. The aver-
age relative reduction exhibited similar trends as the recovery period.
While higher drought severity could mostly be linked to higher relative
droughts than when looking at median recovery periods. Especially
reduction in tree-ring width, there were again some exceptions.
in relation to the drought of 2003, Q. pubescens had the shortest
The size of the growth reduction in the post-drought period was
maximum recovery period. Even though most of the population of F.
the lowest in Q. pubescens (12%–18%) across almost all droughts, while
ornus recovered comparatively quickly after individual droughts, some
the growth reduction of F. ornus was the highest, ranging from 15% in
individuals took significantly longer to recover.
1992 to 37% in the year 2003 (Table 4). No relationship was found
After grouping sites into clusters, significant differences in recovery
between the length of recovery period, relative growth reduction and
periods between species were found in sites with sub-Mediterranean
tree diameter or tree height.
and transitional climate in more recent droughts (Table 3). On drier,
sub-Mediterranean sites, F. ornus was the fastest to recover across
almost all droughts (median recovery periods ranging from 0.5 to 3.3. Trees with no recovery in 20 years after individual droughts
1.7 years). On sites with more annual precipitation and higher temper-
atures (transitional and continental), F. ornus also recovered the fastest The percentage of trees for each drought and by species with no
across all sites and droughts. While in some individual droughts and recovery within 20 years after individual droughts is shown in Table 5.
clusters the other two species had similar recovery periods to F. ornus In some trees, the recovery period exceeded 20 years. The drought of
(e.g., P. nigra in 2003 in the sub-Mediterranean cluster), they were not 2003 was not included in this analysis, since the trees were sampled in
as consistent in their growth response post-drought as F. ornus. The 2022 and the tree-ring data did not extend further than 20 years after
recovery periods were noticeably the shortest on the wettest site. the last drought (i.e. 2003).

4
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

Fig. 3. Histogram of average growth reduction.

Table 5 were calculated for all species in the transitional climate cluster. The
Proportion of trees across different species and droughts that did not recover within 20
Pinus nigra trees showed stronger dependency on temperature (Fig. 5).
years after individual droughts. The drought of 2003 is not shown, since the sampling
was done 19 years after this drought. Q. pubescens correlated positively with precipitation throughout the sea-
1983 1985 1992 2000
sons, including the previous growing season months. More significant
correlations were obtained for seasons that included more than a single
FO 0.07 0.04 0.01 0.07
PN 0.29 0.08 0.48 0.22 month. Fraxinus ornus showed more cluster-specific climate growth pat-
QP 0.19 0.13 0.38 0.00 terns than other two analyzed species. In the sub-Mediterranean cluster,
it generally correlated positively with higher summer precipitation and
less stronger with higher temperatures than in other two clusters.
Table 6
Proportion of drought-unaffected trees across different species and droughts, showing
the percentage of trees with no recovery period in the timespan of 20 years after 4. Discussion
individual droughts.
1983 1985 1992 2000 2003
The current study confirmed that the response to drought in three
FO 0.30 0.42 0.62 0.27 0.11 sub-Mediterranean species varied across species and drought intensity.
PN 0.12 0.70 0.33 0.14 0.20
QP 0.56 0.35 0.12 0.35 0.04
It also demonstrated that even within species at the same site or cluster,
noticeable differences in radial growth existed in drought response
between individual trees. Most trees were affected, while in some cases,
in the analysis of individual droughts and species showed that a higher
3.4. Drought-unaffected trees proportion of trees was unaffected by individual droughts than perma-
nently affected (i.e., no recovery to pre-drought growth in 20 years
There are trees in each drought whose tree-ring width was not after droughts). The proportion of both varied between species, site
affected by the individual drought either in the drought year or in years and drought. These facts would indicate that more emphasis should be
following the drought. The percentage of these drought-unaffected trees given in the future to studying ‘‘winner’’ trees, since they may hold
was examined first, Table 6 shows the overall percentage of unaffected the answer to species’ better adaptability to the occurrence of hotter
trees, grouped by species and droughts. While some differences can droughts.
be observed between species, more pronounced differences can be Of the three examined species, F. ornus appears to be the least
observed between droughts. For example, 2003 had the worst drought sensitive to drought stress and could be considered to be the most
according to climate data (Fig. 2) and this can also be confirmed by drought-adapted in terms of the response of radial growth. It had
the lowest percentages of unaffected trees across the three species. This the shortest recovery periods, with the minimum number of perma-
drought, which can be classified as the worst drought in the last seven nently affected trees. Although relatively small sizes of ash trees when
decades (Fig. 2), clearly affected the growth of the trees of all three compared to other two species in the current study could potentially
examined species, albeit to different degrees. explain the differences in drought tolerance, this effect appears to be
dependent on species and other factors (Pretzsch et al., 2018; McGregor
3.5. Climate-growth relationships et al., 2021; Deng et al., 2025). It has been shown that when soil
water availability is limited, F. ornus allocates more biomass to leaves
The results of climate-growth relationships by clusters are shown and phloem to compensate for the negative effects of drought (Kiora-
in Figs. 4 and 5. As the continental cluster only included one site, postolou and Petit, 2018). Q. pubescens in this study showed slightly
the results using that cluster should be interpreted with caution. The slower recovery after drought when compared to F. ornus. Although
correlations are shown for maximum temperature and for precipita- Q. pubescens is adapted to withstand moderate drought stress (Pasta
tion, while the correlations for minimum temperature can be found et al., 2016), expected severe drought events could cause hydraulic
in the Appendix. The most significant correlations with precipitation collapse and dieback due to its anisohydric strategy (Italiano et al.,

5
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

Fig. 4. Significant correlations (p < 0.05) between tree-ring widths of individual years and monthly precipitation of past (lowercase) and current (uppercase) years.

Fig. 5. Significant correlations (p < 0.05) between tree-ring widths of individual years and monthly maximum temperature of past (lowercase) and current (uppercase) years.

6
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

Fig. A.1. Individual tree-ring width measurements across the studied sites and species. FO – Fraxinus ornus, PN – Pinus nigra, QP – Quercus pubescens. Blue dashed line shows year
1974, starting year for analysis.

2023). Pines are more negatively affected by long-term drought in- the presence of intra-annual density fluctuations, IADFs (del Castillo
duced stress than oaks (Martín-Gómez et al., 2017), since pines rely et al., 2016; Krajnc et al., 2021; Gričar et al., 2020). Moreover, species-
on precipitation water from shallow soil layers while an oak’s roots specific inter-annual effects, i.e., the effects of climatic conditions
draw water from greater depths (del Castillo et al., 2016) and have and tree phenology in the previous growing season on the radial
access to more tightly bound water pools in the soil (Tang and Feng, growth in the current year, should be considered (Marchand et al.,
2001; Martín-Gómez et al., 2017). Furthermore, evidence shows that 2020; Gričar et al., 2022). In ring-porous species, such as F. ornus
bedrock can store considerable amounts of available water, which can and Q. pubescens, a large proportion of the earlywood is formed from
be used by trees to survive drought, especially in karstic areas in which carbohydrate reserves produced in previous years and stored in the
a large share of underground water sources is stored in rocks and parenchyma cells (Barbaroux and Bréda, 2002). Evergreen conifers may
not in soil (Nardini et al., 2024). When designing future studies, soil benefit from warm winter conditions in the Mediterranean region by
depth and type should be recorded on the level of individual trees enabling winter photosynthesis and carbohydrate accumulation, which
and taken into account when evaluating drought response. Some of would positively influence earlywood widths in the following growing
the recovery periods were on average longer in less intensive droughts, season (Martín-Benito et al., 2010).
which suggests that drought intensity is probably not directly linked to The two components of resilience, i.e., resistance and recovery, vary
recovery period. within taxonomic groups. DeSoto et al. (2020) found that Pinaceae
When considering the effects of drought on radial growth found in species tend to show stronger and longer legacies in tree-ring width
this study, several other environmental factors should be mentioned. after drought (slower recovery) than Fagaceae species. The use of
Although SHM index is considered to be a relatively good measure wood-anatomical variables (i.e., quantitative wood anatomy), such as
of drought exposure, its calculation is still based on data from an lumen area or vessel density, may allow the overcoming of some
individual year. It does not take account of the fact that the dynamics critical issues encountered when using tree-ring width to assess drought
of tree radial growth within individual seasons varies significantly impacts, such as different responses to spring or summer droughts
among tree species and years (Michelot et al., 2012; del Castillo et al., between co-occurring species due to different xylem phenology or
2016; Gričar et al., 2020, e.g.). Consequently, a drought in September dynamics of stored carbon use (Michelot et al., 2012).
will have a different effect on ring-porous species than on conifers, The effect of droughts could also potentially overlap. If a drought-
simply because conifers may continue growing longer in the season affected tree experiences another drought within the recovery period,
in the case of favorable growing conditions, as is often evident by the effect is likely to be compounded, making the tree more likely

7
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

Table A.1 cannot be directly observed in the current study (median relative
Mean annual temperature and mean annual precipitation for included sites.
growth reduction after 1985 drought is not higher than after drought
Site Mean temperature [℃] Mean precipitation [mm] of 1983, see Table 4 above), this could potentially affect the length
Kobjeglava 11.5 2362 of recovery period after the second drought. As there is no way of
Krempa 7.5 1641
evaluating the size of this effect of compounding drought effects on
Osp 12.7 1661
Prosecco 13.3 1718 recovery indices, the values on drought recovery from the current study
Rupinpiccolo 12.0 2056 are likely undervalued. They should be taken as the lowest possible
Sezana 11.1 2095 effect of droughts on tree growth and are quite likely more strongly
Trebusa 9.5 3145 pronounced in droughts that occurred close after each other.
Vremscica 9.1 1901
Without long-term monitoring of ‘‘winner’’ and ‘‘loser’’ trees, it is
not possible to determine which factors are crucial to drought resis-
Table A.2 tance. Although a large proportion of the non-affected trees remained
Quality assessment of the developed chronologies: Gleichläufigkeit (GLK), mean correla-
unaffected in terms of tree-ring width across all examined droughts,
tion between trees (rbar), expressed population signal (EPS), first order autocorrelation
(AR1) and sub-sample signal strength (SSS).
there was no clear indication of what makes a tree a ‘‘winner’’, in
Site GLK rbar EPS AR1 SSS above 0.90
terms of either being insensitive to an on-going drought or in terms of
a quick recovery post-drought. Individual tree vitality (combined with
FO-Kobjeglava 0.727 0.36 0.88 −0.194 1976
FO-Krempa 0.661 0.338 0.884 −0.053 1964 tree age, size and competition) could have a strong role in predicting
FO-Prosecco 0.670 0.506 0.968 0.018 1958 drought-response, but this would have to be evaluated yearly to make
FO-Rupinpiccolo 0.725 0.363 0.895 −0.198 1942 any meaningful observations. Tree vitality is the dynamic ability of
FO-Trebusa 0.578 0.083 0.522 −0.023 1940 a tree to grow under given conditions (Dobbertin, 2005). While the
PN-Kobjeglava 0.662 0.558 0.983 −0.081 1902
PN-Krempa 0.662 0.547 0.981 −0.014 1907
minimum vitality (i.e., tree death) can be identified, the hypothetical
PN-Sezana 0.660 0.563 0.981 −0.013 1899 optimum tree vitality cannot be directly determined. Among the com-
PN-Trebusa 0.605 0.64 0.987 0.127 1899 monly adopted indicators for assessing tree vitality, tree-ring widths are
QP-Kras 0.603 0.533 0.971 0.305 1948 of particular value (Beck, 2009), since decreasing growth curves are
QP-Prosecco 0.644 0.634 0.977 −0.047 1960
among the most obvious growth-related characteristics of dying trees
QP-Rupinpiccolo 0.671 0.343 0.879 0.226 1896
and are both species-specific and site-specific (Bigler and Bugmann,
2004). However, as evidenced in the current study, tree-ring widths
Table A.3 alone are not enough for evaluating drought resistance and many other
Post hoc analysis of differences in median recovery period between by clusters
factors on the level of individual trees as suggested above should be
and species, displaying only significant differences (Dunn test, Bonferroni method of
adjusting for multiple comparisons, p < 0.05). considered in future similar studies.
Drought year Cluster Group 1 Group 2 Z value p [Link]
2000 sub-Mediterranean FO PN 2.56 0.01 0.03 CRediT authorship contribution statement
2000 sub-Mediterranean PN QP −2.76 0.01 0.02
1983 transitional FO PN 3.22 0.00 0.00 Luka Krajnc: Writing – review & editing, Writing – original
1983 transitional FO QP 3.44 0.00 0.00
draft, Visualization, Validation, Methodology, Investigation, Funding
2000 transitional FO PN 3.79 0.00 0.00
2000 transitional PN QP −3.39 0.00 0.00 acquisition, Formal analysis, Data curation, Conceptualization. Polona
2003 transitional FO PN 3.91 0.00 0.00 Hafner: Writing – review & editing, Writing – original draft, Methodol-
2003 transitional PN QP −2.55 0.01 0.03 ogy, Investigation, Data curation, Conceptualization. Jernej Jevšenak:
Writing – review & editing, Writing – original draft, Visualization,
Methodology, Investigation, Formal analysis, Data curation, Concep-
tualization. Şule Ceyda Izmir: Writing – review & editing, Writing –
to suffer in the next drought. It could well be that permanently af-
original draft, Visualization, Validation, Methodology, Investigation,
fected trees from the current study will eventually die a premature
death. Cailleret et al. (2019) observed that there is a gradual increase in Data curation, Conceptualization. Mitja Ferlan: Writing – review
tree-ring width variability and a decrease in tree-ring width synchrony & editing, Writing – original draft, Project administration, Funding
in the last 20 years before the mortality of gymnosperms, irrespective acquisition, Conceptualization. Jožica Gričar: Writing – review &
of the cause of mortality. They speculated that these changes could editing, Writing – original draft, Supervision, Project administra-
be associated with drought-induced alterations in carbon economy and tion, Methodology, Investigation, Funding acquisition, Data curation,
allocation patterns. In contrast to the current study, Cailleret et al. Conceptualization.
(2019) did not find any consistent changes in any metric in the case
of angiosperms. They explained the lack of any signal by the rela- Declaration of competing interest
tively high capacity of angiosperms to recover after a stress-induced
growth decline. An increase in inter-annual variability in tree-ring The authors declare that they have no known competing finan-
widths and a decrease in tree-ring width synchrony may be powerful cial interests or personal relationships that could have appeared to
predictors of gymnosperm mortality risk, but not necessarily so for influence the work reported in this paper.
angiosperms (Cailleret et al., 2019). Several studies have, on the other
hand, reported, that drought legacy effect in trees is relatively short- Acknowledgments
term and rarely exceeds one year (Gazol et al., 2020; Kannenberg
et al., 2019; Miller et al., 2023). In the current study and contrary to This work was supported by the Slovenian Research Agency: re-
previously cited studies, trees from several sites exhibited longer recov- search core funding no. P4-0430 and P4-0107; projects J4-2540, J4-
ery periods than one year, indicating that the effect of droughts close 50130 and J4-4541. The work of LK was funded by the European
together could have a compound effect on post-disturbance growth as Union, contract no. 0005-404, Recovery and Resilience Facility (NOO)
suggested by Anderegg et al. (2015). When droughts follow one another and ‘‘NextGenerationEU’’. The authors kindly acknowledge the help of
in close succession, overlapping drought legacies can potentially obfus- Prof. Dr. Andrea Nardini with the selection of suitable sample sites
cate growth responses when evaluating the consequences of individual and Gregor Skoberne for his help with sampling and processing of
droughts. Although the effect of compounding on post-drought growth increment cores.

8
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

Fig. A.2. Site clustering based on average year precipitation and temperature.

Fig. A.3. Significant correlations (p < 0.05) between tree-ring widths of individual years and monthly minimum temperature of past (lowercase) and current (uppercase) years.

Appendix Data availability

Data will be made available on request.

See Figs. A.1–A.3 and Tables A.1–A.3.

9
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

References Dunn, O.J., 1964. Multiple comparisons using rank sums. Technometrics 6 (3),
241–252. [Link]
Abdel-Gadir, A.Y., Krahmer, R.L., 1993. Genetic variation in the age of demarcation Gapare, W.J., Wu, H.X., Abarquez, A., 2006. Genetic control of the time of transition
between juvenile and mature wood in douglas-fir. Wood Fiber Sci. 25 (4), 384–394. from juvenile to mature wood in PinusRadiata D. Don. Ann. For. Sci. 63 (8),
Adamopoulos, S., Karageorgos, A., Passialis, C., Chavenetidou, M., 2011. Mathematical 871–878. [Link]
approach for defining juvenile-mature wood transition zone in black locust and Gazol, A., Camarero, J.J., 2022. Compound climate events increase tree drought
chestnut. Wood Fiber Sci. 43 (3), 336–342. mortality across European forests. Sci. Total Environ. 816, 151604. [Link]
org/10.1016/[Link].2021.151604.
Allen, C.D., Breshears, D.D., McDowell, N.G., 2015. On underestimation of global
vulnerability to tree mortality and forest die-off from hotter drought in the Gazol, A., Camarero, J.J., Sánchez-Salguero, R., Vicente-Serrano, S.M., Serra-
anthropocene. Ecosphere 6 (8), 1–55. [Link] Maluquer, X., Gutiérrez, E., De Luis, M., Sangüesa-Barreda, G., Novak, K., Rozas, V.,
Tíscar, P.A., Linares, J.C., Martínez Del Castillo, E., Ribas, M., García-González, I.,
Anderegg, W.R.L., Schwalm, C., Biondi, F., Camarero, J.J., Koch, G., Litvak, M.,
Silla, F., Camisón, Á., Génova, M., Olano, J.M., Hereş, A.-M., Curiel Yuste, J.,
Ogle, K., Shaw, J.D., Shevliakova, E., Williams, A.P., Wolf, A., Ziaco, E., Pacala, S.,
Longares, L.A., Hevia, A., Tomas-Burguera, M., Galván, J.D., 2020. Drought legacies
2015. Pervasive drought legacies in forest ecosystems and their implications for
are short, prevail in dry conifer forests and depend on growth variability. In:
carbon cycle models. Science 349 (6247), 528–532. [Link]
Battipaglia, G. (Ed.), J. Ecol. 108 (6), 2473–2484. [Link]
science.aab1833.
2745.13435.
Barbaroux, C., Bréda, N., 2002. Contrasting distribution and seasonal dynamics of
Gazol, A., Camarero, J.J., Vicente-Serrano, S.M., Sánchez-Salguero, R., Gutiérrez, E.,
carbohydrate reserves in stem wood of adult ring-porous sessile oak and diffuse-
de Luis, M., Sangüesa-Barreda, G., Novak, K., Rozas, V., Tíscar, P.A., Linares, J.C.,
porous beech trees. Tree Physiol. 22 (17), 1201–1210. [Link]
Martín-Hernández, N., Martínez del Castillo, E., Ribas, M., García-González, I.,
treephys/22.17.1201.
Silla, F., Camisón, A., Génova, M., Olano, J.M., Longares, L.A., Hevia, A., Tomás-
Beck, W., 2009. Growth patterns of forest stands - the response towards pollutants
Burguera, M., Galván, J.D., 2018. Forest resilience to drought varies across biomes.
and climatic impact. IForest - Biogeosciences For. 2 (1), 4–6. [Link]
Glob. Chang. Biol. 24 (5), 2143–2158. [Link]
3832/ifor0472-002.
Giagli, K., Timko, L., Gryc, V., Fajstavr, M., Vavrčik, H., 2017. Tree-ring widths and
Benito Garzón, M., Sánchez De Dios, R., Sainz Ollero, H., 2008. Effects of climate
wood density variability of non- native species: A case study of douglas-fir growing
change on the distribution of Iberian tree species. Appl. Veg. Sci. 11 (2), 169–178.
in central Europe. In: International Conference on Information and Communication
[Link]
Technologies in Agriculture, Food and Environment. Chania, Crete, Greece, p. 10.
Bigler, C., Bugmann, H., 2004. Assessing the performance of theoretical and empirical
Gričar, J., Jevšenak, J., Hafner, P., Prislan, P., Ferlan, M., Lavrič, M., Vodnik, D.,
tree mortality models using tree-ring series of Norway spruce. Ecol. Model. 174
Eler, K., 2022. Climatic regulation of leaf and cambial phenology in Quercus
(3), 225–239. [Link]
pubescens: Their interlinkage and impact on xylem and phloem conduits. Sci. Total
Bréda, N., Huc, R., Granier, A., Dreyer, E., 2006. Temperate forest trees and stands
Environ. 802, 149968. [Link]
under severe drought: A review of ecophysiological responses, adaptation processes
Gričar, J., Vedenik, A., Skoberne, G., Hafner, P., Prislan, P., 2020. Timeline of leaf
and long-term consequences. Ann. For. Sci. 63 (6), 625–644. [Link]
and cambial phenology in relation to development of initial conduits in xylem and
1051/forest:2006042.
phloem in three coexisting sub-mediterranean deciduous tree species. Forests 11
Bunn, A.G., 2008. A dendrochronology program library in R (dplR). Dendrochronologia (10), 1104. [Link]
26 (2), 115–124. [Link] Gričar, J., Zavadlav, S., Jyske, T., Lavrič, M., Laakso, T., Hafner, P., Eler, K., Vodnik, D.,
Bunn, A., Korpela, M., Biondi, F., Campelo, F., Mérian, P., Qeadan, F., Zang, C., 2023. 2019. Effect of soil water availability on intra-annual xylem and phloem formation
dplR: Dendrochronology program library in R. and non-structural carbohydrate pools in stem of Quercus Pubescens. In: Epron, D.
Buras, A., Wilmking, M., 2015. Correcting the calculation of Gleichläufigkeit. (Ed.), Tree Physiol. 39 (2), 222–233. [Link]
Dendrochronologia 34, 29–30. [Link] Grossiord, C., Granier, A., Ratcliffe, S., Bouriaud, O., Bruelheide, H., Chećko, E.,
Cailleret, M., Dakos, V., Jansen, S., Robert, E.M.R., Aakala, T., Amoroso, M.M., Forrester, D.I., Dawud, S.M., Finér, L., Pollastrini, M., Scherer-Lorenzen, M.,
Antos, J.A., Bigler, C., Bugmann, H., Caccianaga, M., Camarero, J.-J., Cheru- Valladares, F., Bonal, D., Gessler, A., 2014. Tree diversity does not always improve
bini, P., Coyea, M.R., Čufar, K., Das, A.J., Davi, H., Gea-Izquierdo, G., Gillner, S., resistance of forest ecosystems to drought. Proc. Natl. Acad. Sci. 111 (41),
Haavik, L.J., Hartmann, H., Hereş, A.-M., Hultine, K.R., Janda, P., Kane, J.M., 14812–14815. [Link]
Kharuk, V.I., Kitzberger, T., Klein, T., Levanic, T., Linares, J.-C., Lombardi, F., Hammond, W.M., Williams, A.P., Abatzoglou, J.T., Adams, H.D., Klein, T., López, R.,
Mäkinen, H., Mészáros, I., Metsaranta, J.M., Oberhuber, W., Papadopoulos, A., Sáenz-Romero, C., Hartmann, H., Breshears, D.D., Allen, C.D., 2022. Global field
Petritan, A.M., Rohner, B., Sangüesa-Barreda, G., Smith, J.M., Stan, A.B., Sto- observations of tree die-off reveal hotter-drought fingerprint for Earth’s forests. Nat
janovic, D.B., Suarez, M.-L., Svoboda, M., Trotsiuk, V., Villalba, R., Westwood, A.R., Commun 13 (1), 1761. [Link]
Wyckoff, P.H., Martínez-Vilalta, J., 2019. Early-warning signals of individual tree Italiano, S.S., Julio Camarero, J., Borghetti, M., Colangelo, M., Pizarro, M., Ripul-
mortality based on annual radial growth. Front. Plant Sci. 9, [Link] lone, F., 2023. Radial growth, wood anatomical traits and remote sensing indexes
3389/fpls.2018.01964. reflect different impacts of drought on Mediterranean forests. For. Ecol. Manag.
Cherubini, P., Gargano, A., Grob, M., Nievergelt, D., Passardi, S., 2020. Dendrochronolo- 548, 121406. [Link]
gia’s tutoring recipes: How to take samples for small basic dendroecological studies. Jevšenak, J., 2020. New features in the dendroTools R package: Bootstrapped and par-
Dendrochronologia 64, 125774. [Link] tial correlation coefficients for monthly and daily climate data. Dendrochronologia
Choat, B., Brodribb, T.J., Brodersen, C.R., Duursma, R.A., López, R., Medlyn, B.E., 63, 125753. [Link]
2018. Triggers of tree mortality under drought. Nature 558 (7711), 531–539. Jevšenak, J., Levanič, T., 2018. dendroTools: R package for studying linear and nonlin-
[Link] ear responses between tree-rings and daily environmental data. Dendrochronologia
Cook, E., Kairiukstis, L. (Eds.), 1990. Methods of dendrochronology. Springer 48, 32–39. [Link]
Netherlands, Dordrecht, [Link] Kannenberg, S.A., Novick, K.A., Alexander, M.R., Maxwell, J.T., Moore, D.J.P.,
del Castillo, J., Comas, C., Voltas, J., Ferrio, J.P., 2016. Dynamics of competition over Phillips, R.P., Anderegg, W.R.L., 2019. Linking drought legacy effects across scales:
water in a mixed oak-pine Mediterranean forest: Spatio-temporal and physiological From leaves to tree rings to ecosystems. Global Change Biol. 25 (9), 2978–2992.
components. For. Ecol. Manag. 382, 214–224. [Link] [Link]
2016.10.025. Kiorapostolou, N., Petit, G., 2018. Similarities and differences in the balances between
Deng, R., Liao, J., Rademacher, T., Xu, Z., Du, M., Zheng, J., Fu, L., Zhang, X., 2025. leaf, xylem and phloem structures in Fraxinus ornus along an environmental
Species-specific influences of competition and tree size on drought sensitivity and gradient. In: Dannoura, M. (Ed.), Tree Physiol. 39 (2), 234–242. [Link]
resistance for three planted conifers in northern China. For. Ecosyst. 13, 100295. org/10.1093/treephys/tpy095.
[Link] Klesse, S., 2021. Critical note on the application of the ‘‘two-third’’ spline.
DeSoto, L., Cailleret, M., Sterck, F., Jansen, S., Kramer, K., Robert, E.M.R., Aakala, T., Dendrochronologia 65, 125786. [Link]
Amoroso, M.M., Bigler, C., Camarero, J.J., Čufar, K., Gea-Izquierdo, G., Gill- Köhl, M., Linser, S., Prins, K., 2020. State of Europe’s Forests 2020. Ministerial
ner, S., Haavik, L.J., Hereş, A.-M., Kane, J.M., Kharuk, V.I., Kitzberger, T., Conference on the Protection of Forests in Europe - FOREST EUROPE Liaison Unit
Klein, T., Levanič, T., Linares, J.C., Mäkinen, H., Oberhuber, W., Papadopoulos, A., Bratislava, Bratislava, [Link]
Rohner, B., Sangüesa-Barreda, G., Stojanovic, D.B., Suárez, M.L., Villalba, R., Krajnc, L., Hafner, P., Gričar, J., 2021. The effect of bedrock and species mixture on
Martínez-Vilalta, J., 2020. Low growth resilience to drought is related to future wood density and radial wood increment in pubescent oak and black pine. Forest
mortality risk in trees. Nat. Commun. 11 (1), 545. [Link] Ecol. Manag. 481, 118753. [Link]
s41467-020-14300-5. Kruskal, W.H., Wallis, W.A., 1952. Use of ranks in one-criterion variance analysis.
DeSoto, L., De la Cruz, M., Fonti, P., 2011. Intra-annual patterns of tracheid size in J. Amer. Statist. Assoc. 47 (260), 583–621. [Link]
the Mediterranean tree Juniperus thurifera as an indicator of seasonal water stress. 1952.10483441.
Can. J. For. Res. 41 (6), 1280–1294. [Link] Lazoglou, G., Papadopoulos-Zachos, A., Georgiades, P., Zittis, G., Velikou, K., Man-
Dobbertin, M., 2005. Tree growth as indicator of tree vitality and of tree reaction ios, E.M., Anagnostopoulou, C., 2024. Identification of climate change hotspots in
to environmental stress: A review. Eur. J. For. Res. 124 (4), 319–333. http: the Mediterranean. Sci Rep 14 (1), 29817. [Link]
//[Link]/10.1007/s10342-005-0085-3. 80139-1.

10
L. Krajnc et al. Dendrochronologia 93 (2025) 126378

Levanič, T., 2007. Atrics - a new system for image acquisition in dendrochronology. Pasta, S., de Rigo, D., Caudullo, G., 2016. Quercus pubescens in Europe: Distribution,
Tree- Ring Res. 63 (2), 117–122. [Link] habitat, usage and threats. In: Jesús San-Miguel-Ayanz, de Rigo, D., Giovanni
Marchand, L.J., Dox, I., Gričar, J., Prislan, P., Leys, S., Van den Bulcke, J., Fonti, P., Caudullo, Tracy Houston Durrant, Achille Mauri (Eds.), European Atlas of Forest
Lange, H., Matthysen, E., Peñuelas, J., Zuccarini, P., Campioli, M., 2020. Inter- Tree Species. Publication Office of the European Union, Luxembourg.
individual variability in spring phenology of temperate deciduous trees depends Pretzsch, H., Schütze, G., Biber, P., 2018. Drought can favour the growth of small in
on species, tree size and previous year autumn phenology. Agric. For. Meteorol. relation to tall trees in mature stands of Norway spruce and European beech. For.
290, 108031. [Link] Ecosyst. 5 (1), 20. [Link]
Marchi, M., Bucci, G., Iovieno, P., Ray, D., 2024. ClimateDT: A global scale-free R. Core Team, 2021. R: A language and environment for statistical computing. R
dynamic downscaling portal for historic and future climate data. Environments Foundation for Statistical Computing, Vienna, Austria.
11 (4), 82. [Link] Sánchez, E., Gallardo, C., Gaertner, M., Arribas, A., Castro, M., 2004. Future climate
Marchi, M., Castellanos-Acuña, D., Hamann, A., Wang, T., Ray, D., Menzel, A., 2020. extreme events in the Mediterranean simulated by a regional climate model: A first
ClimateEU, scale-free climate normals, historical time series, and future projections approach. Glob. Planet. Change 44 (1–4), 163–180. [Link]
for Europe. Sci Data 7 (1), 428. [Link] gloplacha.2004.06.010.
Martín-Benito, D., del Río, M., Cañellas, I., 2010. Black pine (Pinus Nigra Arn.) growth Sánchez De Dios, R., Benito-Garzón, M., Sainz-Ollero, H., 2009. Present and future
divergence along a latitudinal gradient in Western Mediterranean mountains. Ann. extension of the Iberian submediterranean territories as determined from the
For. Sci. 67 (4), 401. [Link] distribution of marcescent oaks. Plant Ecol 204 (2), 189–205. [Link]
Martín-Gómez, P., Aguilera, M., Pemán, J., Gil-Pelegrín, E., Ferrio, J.P., 2017. Contrast- 10.1007/s11258-009-9584-5.
ing ecophysiological strategies related to drought: The case of a mixed stand of Schwarz, J., Skiadaresis, G., Kohler, M., Kunz, J., Schnabel, F., Vitali, V., Bauhus, J.,
scots pine (Pinus Sylvestris) and a submediterranean oak (Quercus Subpyrenaica). 2020. Quantifying growth responses of trees to drought—a critique of commonly
Tree Physiol. 37 (11), 1478–1492. [Link] used resilience indices and recommendations for future studies. Curr. For. Rep. 6
Mazza, G., Sarris, D., 2021. Identifying the full spectrum of climatic signals controlling (3), 185–200. [Link]
a tree species’ growth and adaptation to climate change. Ecol. Indic. 130, 108109. Stagge, J.H., Tallaksen, L.M., Xu, C.Y., Van Lanen, H.A., 2014. Standardized
[Link] precipitation-evapotranspiration index (SPEI): Sensitivity to potential evapotranspi-
McGregor, I.R., Helcoski, R., Kunert, N., Tepley, A.J., Gonzalez-Akre, E.B., Her- ration model and parameters. In: Servat, E., Paturel, J.-E., Dezetter, A., Laaha, G.,
rmann, V., Zailaa, J., Stovall, A.E.L., Bourg, N.A., McShea, W.J., Pederson, N., Boyer, J.-F., Ruelland, D., Demuth, S., Mahe, G., Daniell, T.M., van Lanen, H.A.J.
Sack, L., Anderson-Teixeira, K.J., 2021. Tree height and leaf drought tolerance (Eds.), Hydrol. Chang. World. In: IAHS-AISH Proceedings and Reports, Copernicus
traits shape growth responses across droughts in a temperate broadleaf forest. New Publications, Germany, pp. 367–373.
Phytol. 231 (2), 601–616. [Link] Sterck, F.J., Zweifel, R., Sass-Klaassen, U., Chowdhury, Q., 2008. Persisting soil drought
Michelot, A., Simard, S., Rathgeber, C., Dufrene, E., Damesin, C., 2012. Comparing the reduces leaf specific conductivity in scots pine (Pinus Sylvestris) and pubescent oak
intra-annual wood formation of three European species (Fagus Sylvatica, Quercus (Quercus Pubescens). Tree Physiol. 28 (4), 529–536. [Link]
Petraea and Pinus Sylvestris) as related to leaf phenology and non-structural treephys/28.4.529.
carbohydrate dynamics. Tree Physiol. 32 (8), 1033–1045. [Link] Tang, K., Feng, X., 2001. The effect of soil hydrology on the oxygen and hydrogen
1093/treephys/tps052. isotopic compositions of plants’ source water. Earth Planet. Sci. Lett. 185 (3),
Miller, T.W., Stangler, D.F., Larysch, E., Honer, H., Puhlmann, H., Schindler, D., 355–367. [Link]
Jung, C., Seifert, T., Rigling, A., Kahle, H.-P., 2023. Later growth onsets or Valladares, F., Gianoli, E., Gómez, J.M., 2007. Ecological limits to plant phenotypic
reduced growth rates: What characterises legacy effects at the tree-ring level in plasticity. New Phytol. 176 (4), 749–763. [Link]
conifers after the severe 2018 drought? Sci. Total Environ. 854, 158703. http: 2007.02275.x.
//[Link]/10.1016/[Link].2022.158703. van der Maaten-Theunissen, M., Trouillier, M., Schwarz, J., Skiadaresis, G., Thurm, E.A.,
Nardini, A., Tomasella, M., Di Bert, S., 2024. Bedrock: The hidden water reservoir for van der Maaten, E., 2021. pointRes 2.0: New functions to describe tree resilience.
trees challenged by drought. Trees 38 (1), 1–11. [Link] Dendrochronologia 70, 125899. [Link]
023-02482-6. van der Maaten-Theunissen, M., van der Maaten, E., Bouriaud, O., 2015. pointRes: An R
Pachauri, R.K., Allen, M.R., Barros, V.R., Broome, J., Cramer, W., Christ, R., package to analyze pointer years and components of resilience. Dendrochronologia
Church, J.A., Clarke, L., Dahe, Q.D., Dasqupta, P., Dubash, N.K., Edenhofer, O., 35, 34–38. [Link]
Elgizouli, I., Field, C.B., Forster, P., Friedlingstein, P., Fuglestvedt, J., Gomez- Vicente-Serrano, S.M., Beguería, S., López-Moreno, J.I., 2010. A multiscalar drought in-
Echeverri, L., Hallegatte, S., Hegerl, G., Howden, M., Jiang, K., Jimenez Cis- dex sensitive to global warming: The standardized precipitation evapotranspiration
neros, B., Kattsov, V., Lee, H., Mach, K.J., Marotzke, J., Mastrandrea, M.D., index. J. Clim. 23 (7), 1696–1718. [Link]
Meyer, L., Minx, J., Mulugetta, Y., O’Brien, K., Oppenheimer, M., Pereira, J.J., Wigley, T.M.L., Briffa, K.R., Jones, P.D., 1984. On the average value of correlated time
Pichs-Madruga, R., Plattner, G.-K., Pörtner, H.-O., Power, S.B., Preston, B., Ravin- series, with applications in dendroclimatology and hydrometeorology. J. Clim. Appl.
dranath, N., Reisinger, A., Riahi, K., Rusticucci, M., Scholes, R., Seyboth, K., Meteor. 23 (2), 201–213. [Link]
Sokona, Y., Stavins, R., Stocker, T.F., Tschakert, P., van Vuuren, D., van Ypersele, J.- OTAVOC>[Link];2.
P., 2014. Climate Change 2014 Synthesis Report. Contribution of Working Groups
I, II, and III to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change. IPCC.

11