[Link]-ethiopia.

org
[Link] Working Paper No. 25

Rangeland condition and feed

resources in Metema District,
North Gondar Zone, Amhara
Region, Ethiopia

Canadian International Agence canadienne de

Development Agency développement international



Federal Democratic Republic of Ethiopia
MINISTRY OF AGRICULTURE AND
RURAL DEVELOPMENT
 Rangeland condition and feed resources
in Metema District, North Gondar Zone,
Amhara Region, Ethiopia

Tesfaye Desalew, Azage Tegegne, Lisanework Nigatu and Worku Teka

Improving Productivity and Market Success of Ethiopian Farmers project (IPMS)–

International Livestock Research Institute (ILRI), Addis Ababa, Ethiopia

*Azage Tegegne is corresponding author: [Link]@[Link]

WorkingPaper_25.indd 1 1/10/2011 10:54:19 AM

 Authors’ affiliations
Tesfaye Desalew, Bureau of Agriculture and Rural Development, Amhara National Regional
State, Bahir Dar, Ethiopia
Azage Tegegne, Improving Productivity and Market Success (IPMS) of Ethiopian Farmers
Project, International Livestock Research Institute (ILRI), Addis Ababa, Ethiopia
Lisanework Nigatu, Haramaya University, Dire Dawa, Ethiopia
Worku Teka, IPMS, ILRI, Metema, Ethiopia

© 2010 International Livestock Research Institute (ILRI)

This publication is copyrighted by the International Livestock Research

Institute (ILRI). It is licensed for use under the Creative Commons Attribution-
Noncommercial-Share Alike 3.0 Unported License. To view this license, visit http://
[Link]/licenses/by-nc-sa/3.0/. Unless otherwise noted, you are free to copy,
duplicate, or reproduce, and distribute, display, or transmit any part of this publication
or portions thereof without permission, and to make translations, adaptations, or other
derivative works under the following conditions:
ATTRIBUTION. The work must be attributed, but not in any way that suggests
endorsement by ILRI or the author(s).
NON-COMMERCIAL. This work may not be used for commercial purposes.
SHARE ALIKE. If this work is altered, transformed, or built upon, the resulting work
must be distributed only under the same or similar license to this one.
NOTICE:
For any reuse or distribution, the license terms of this work must be made clear to others.
Any of the above conditions can be waived if permission is obtained from the copyright
holder.
Nothing in this license impairs or restricts the author’s moral rights.
Fair dealing and other rights are in no way affected by the above.
The parts used must not misrepresent the meaning of the publication. ILRI would appreciate
being sent a copy of any materials in which text, photos etc. have been used.

Editing, design and layout—ILRI Editorial and Publishing Services, Addis Ababa, Ethiopia.
Correct citation: Tesfaye Desalew, Azage Tegegne, Lisanework Nigatu, and Worku Teka.
2010. Rangeland condition and feed resources in Metema district, North Gondar Zone,
Amhara Region, Ethiopia. IPMS (Improving Productivity and Market Success) of Ethiopian
Farmers Project Working Paper 25. Nairobi, Kenya, ILRI.
International Livestock Research Institute
P O Box 30709, Nairobi 00100, Kenya P O Box 5689, Addis Ababa, Ethiopia
Phone + 254 20 422 3000 Phone + 251 11 617 2000
Email ILRI-Kenya@[Link] Email ILRI-Ethiopia@[Link]
[Link]

WorkingPaper_25.indd 2 1/10/2011 10:54:20 AM

 Table of Contents
List of Tables v
List of Figures vii
Abbreviations viii
Acknowledgements ix
Abstract x
1 Introduction 1
2 Background 3
2.1 Livestock resources in Ethiopia 3
2.2 Rangelands of Ethiopia 3
2.3 Major livestock feed resources in Ethiopia 4
2.4 Causes of rangeland degradation 6
2.5 Factors affecting rangeland vegetation 8
2.6 Factors influencing vegetation composition of the rangeland 8
2.7 Range condition assessment 10
2.8 Biomass estimation methods 13
3 Materials and methods 14
3.1 Description of the study area 14
3.2. Selection of kebeles and farmers 16
3.3 Rangeland condition assessment 16
3.4 Evaluating the quality of major feed resources 20
3.5 Statistical analysis 20
4 Results and discussion 22
4.1 Socio-economic conditions 22
4.2 Land holding and land use pattern 24
4.3 Livestock holdings 25
4.4 Livestock herd/flock structure 26
4.5 Major feed resources 27
4.6 Feed management and utilization practices 28

iii

WorkingPaper_25.indd 3 1/10/2011 10:54:20 AM

 4.7 Availability of natural pasture 29
4.8 Floristic composition 38
4.9 Range condition assessment 47
4.10 Biomass production 52
4.11 Correlation among variables studied for range condition assessment 55
5 Conclusion and recommendations 57
5.1 Conclusion 57
5.2 Recommendations 59
References 61
Appendices 65
Appendix Table 1. Criteria for the scoring of the different factors determining
range conditions (Herbaceous vegetation and soil condition) 65
Appendix Table 2. Criteria for the scoring of the different factors determining
range conditions 66
Appendix Table 3. Common and/or dominant woody species and their percentage
composition in the farming system of different grazing areas 67
Appendix Table 4. Herbaceous species identified in different farming system and
grazing types of the district 68
Appendix Table 5. Correlation matrix among variables studied in the communal
grazing areas of CLFS 69
Appendix Table 6. Correlation matrixes among variables studied in the road side
grazing areas of CLFS 70
Appendix Table 7. Correlation matrixes among variables studied in the enclosure
areas of CLFS 71
Appendix Table 8. Correlation matrix among variables studied in the grazing
areas of CLFS 72
Appendix Table 9. Correlation matrixes among variables studied in the grazing
areas of SLFS 73
Appendix Table 10. Correlation matrix among variables studied and their
interactions in range condition assessment in Metema 74

iv

WorkingPaper_25.indd 4 1/10/2011 10:54:20 AM

 List of Tables
Table 1. Livestock population and regional distribution in Ethiopia 3
Table 2. Nutritional quality of major feed resources in Metema district
as influenced by season 5
Table 3. Experimental units used for analysis in each composite sample site 21
Table 4. Sex, ethnic group, education, and major occupation of respondents
in the cotton–livestock and sesame–livestock farming systems in Metema
district 22
Table 5. Family size, age of respondents and family members in Metema district 23
Table 6. Mean ± SE land holding and land use pattern in Metema district 25
Table 7. Livestock holdings in TLU of the sampled households in Metema 26
Table 8. Livestock herd structures in surveyed households by farming system
in Metema 27
Table 9. Major livestock feed resources and feeding systems in Metema district 28
Table 10. Percentage response of respondents on availability of feed resources
in Metema 30
Table 11. Estimates of grain and crop residues production for major crops grown
in Metema 31
Table 12. Percentage of respondents using stubble grazing by farming system
in Metema 32
Table 13. Percentages of respondents showing the types and amount
of supplementation of feeds in the two farming systems in Metema 33
Table 14. Availability of different feed resources by month in the district 34
Table 15. Chemical composition (Mean ± SE) of different feeds in the study districts 34
Table 16. Effect of season on chemical composition (Mean ± SE) of composite
samples of dominant feed species in Metema 35
Table 17. Estimates of feed balance in Metema district 36
Table 18. Sources of water and watering frequency by farming system for cattle
in Metema (percentage of respondents) 37
Table 19. Percentage of respondents showing major problems to livestock
production in Metema 37

WorkingPaper_25.indd 5 1/10/2011 10:54:20 AM

 Table 20. Common and dominant grass species identified by farming systems
and different grazing types in Metema 39
Table 21. Common and dominant woody species and their percentage
composition by farming system in different grazing areas in Metema 40
Table 22. Percentage of height class (metres) distribution of trees and shrubs
in Metema 41
Table 23. Herbaceous species composition (% DM biomass) and their desirability
by grazing type in cotton–livestock farming system in Metema 42
Table 24. Woody species and percentage composition in different grazing system
in cotton–livestock farming system in Metema 44
Table 25. Herbaceous species composition (% DM biomass) and their desirability
by grazing type in sesame–livestock farming system in Metema 45
Table 26. Woody species in the sesame based farming system by grazing type
in sesame–livestock farming system (SLFS) 47
Table 27. Range condition score (Mean ± SE) of communal grazing areas
by farming system in Metema 48
Table 28. Range condition score (Mean ± SE) for road side grazing areas found
in different farming systems of the study district 50
Table 29. Range condition score (LSM ± SE) in enclosure grazing areas
by farming system in Metema 51
Table 30. Dry matter biomass (kg/ha) of the communal grazing areas by farming
system in Metema district 52
Table 31. Dry matter biomass (kg/ha) of the road side grazing areas by farming
system in Metema 53
Table 32. Dry biomass (kg/ha) of enclosure grazing areas by farming system
in Metema 54

vi

WorkingPaper_25.indd 6 1/10/2011 10:54:20 AM

 List of Figures
Figure 1. Location of Metema woreda in Ethiopia 14
Figure 2. Regression graph of dry matter biomass and range condition 55
Appendix Figure 1. Metema district farming systems 75
Appendix Figure 2. Land use and land cover of Metema district 75
Appendix Figure 3. The view of communal grazing lands at the dry season prior
to flash burning 76
Appendix Figure 4. The view of communal grazing lands at peak dry season 76
Appendix Figure 5. Partial view of communal grazing lands in the sesame-based
farming system 77
Appendix Figure 6. Hay storage in open field 77
Appendix Figure 7. P. spheslatum grass dominated communal grazing land
at SLFS 78
Appendix Figure 8. Part of an enclosure site in the sesame–livestock farming
system (Agam wuha) 78
Appendix Figure 9. Communal grazing land after flash burning 79
Appendix Figure 10. H. rufa grass dominated wood land (Guange river side areas) 79

vii

WorkingPaper_25.indd 7 1/10/2011 10:54:20 AM

 Abbreviations
ADF Acid Detergent Fibre
ADL Acid Detergent Lignin
AEZ Agro-ecological Zone
ANOVA Analysis of Variance
ANRS Amhara National Regional State
AOAC Association of Official Analytical Chemist
BOA Bureau of Agriculture
CACC Central Agricultural Census Commission
CBFS Cotton Based Farming System
CRs Crop Residues
CSA Central Statistic Authority
DCP Digestible Crude Protein
DM Dry Matter
FAO Food and Agriculture Organization of the United Nations
GLM General Linear Model
ha hectare
HHC Household Count
HH Household
ILCA International Livestock Center for Africa
ILRI International Livestock Research Institute
IVDMD Invitro Dry Matter Digestibility
LSM Least square mean
masl Metre above sea level
ME Metabolizable Energy
MOA Ministry of Agriculture
NDF Neutral Detergent Fibre
OoARD Office of Agriculture and Rural Development
OM Organic Matter
PAs Peasant Associations
SE Standard error
SBFS Sesame Based Farming System
SPSS Statistical Package for Social Science
SSA Sub-Saharan Africa
SRM Society of Range Management
t tonne
TLU Tropical Livestock Unit

viii

WorkingPaper_25.indd 8 1/10/2011 10:54:20 AM

 Acknowledgements
The authors would like to extend their appreciation to the contributions of Solomon
Legesse, Research and Development Assistant (RDA) and Teklehymanot Seyoum (driver)
in IPMS Metema PLW, Ato Dawit Negassa and Ato Mulugeta Habte-Michael at the
ILRI nutrition laboratory and Ato Yasin Getahun, IPMS GIS technician. The assistance
of Ato Melaku Wondafrash at the National Herbarium, Addia Ababa University, in the
identification of vegetation specimens is duely acknowledged. Ato Meseret Adane,
Metema district computer technician, for his kind cooperation to use the internet, is
highly acknowledged. Deep appreciation and thanks goes to all the staff members of
Metema Office of Agriculture and Rural Development, for their kind cooperation and
provision of all the relevant information. Sincere appreciation and thanks also goes to
Aderajew Molla, Bekele Kerebih, Tesfaye Mengistie and Solomon Nega for their technical
support and encouragement during the study. Last but not least, we would also like to
thank the farmers in Metema for their full cooperation and contribution to the study.

ix

WorkingPaper_25.indd 9 1/10/2011 10:54:20 AM

 Abstract
The study was conducted in 2006/07 in Metema district, North Gondar Zone of Amhara
region, Ethiopia, with the objectives to characterize the existing rangeland and to
determine the feed resources utilization practices, to assess the natural grazing land
condition based on herbaceous, woody and soil condition and to evaluate the chemical
composition of major livestock feed resources of the area. A single-visit formal survey,
group discussions and visual observations are used to collect the primary information
and secondary sources are also used in livestock feed resources assessment of the
district. A total of 140 respondents from 7 kebeles were selected for interviewing by
stratified random sampling techniques. To assess the range conditions, the samples
were collected by classifying the district into cotton–livestock and sesame–livestock
farming systems. Within a farming system, grazing lands were further stratified into three
sampling areas: communal, road side and enclosure grazing areas. In each of the range
sites a sampling block of 4 km × 1 km was demarcated and this was further stratified
into four sampling plots of equal size. In each of the plot, a belt transects of 50 m × 4 m
was laid out randomly. Then, the parameter used for (herbaceous, soil and woody) grass
species composition, basal cover, litter cover, soil erosion, soil compaction, seedling
count, age distribution and woody density enumeration, canopy cover and hedging
were determined. For the height classes <0–1 m, >1–3 m, >3–4.5 m and >4.5 m was
used. Feed samples were stratified by season and types and subjected to chemical
analysis for determination of DM, ash, CP, ADF, ADL and IVDMD. About 83% of the
inhabitants in the district practice mixed crop–livestock farming system. The mean family
size is 5.31 ± 0.20 persons per household, while the average land holding is 6.78 ±
1.33 ha/household. The mean livestock holding per household is 12.52 ± 6.23 TLU,
and is composed of cattle, goats, sheep, donkeys and camels. Natural pasture (55.7%),
crop residues (20.7%), stubble (14.3%) and hay (9.3%) are the major feed resources
for dry season whereas in the wet season only natural pasture serves as feed resource.
The total estimated DM yield of grazing land and stubble is 780,750 and 51,954 t DM
per annum, respectively. The total estimated available feed supply is 833,531.2 t DM
per annum. Of the identified 33 herbaceous species, 14 and 19 are different grasses
and non-grass species. From the non-grass species 6 legumes and 13 sedges and other
species are recorded. Of the grasses, 23.07%, 38.46% and 30.77% are highly desirable,
desirable and less desirable, respectively. Of the identified 20 woody species, 15%,
35%, and 50% are highly desirable, desirable and less desirable, respectively. The largest
proportion of woody vegetation is contributed by different species of acacia (20%)
and commbretum (10%). Trees and shrubs grouped within the height class of >1–3 m
constituted 41.2% in communal grazing areas, 38.5% in road side grazing and 33.3%

WorkingPaper_25.indd 10 1/10/2011 10:54:20 AM

 in enclosures. Range condition assessment factors such as basal cover, litter cover,
grass species composition, woody vegetation density, canopy cover, hedging effect, age
distribution and total condition score are significant (P<0.05) in communal grazing areas
of the sesame–livestock than in the cotton–livestock farming system. The communal
grazing areas have significantly (P< 0.05) higher values of grass species composition,
basal cover, litter cover, age distribution, and woody species density score, than the road
side grazing areas and lower (P<0.05) than the enclosure grazing areas. The dry matter
biomass of grass, highly desirable, desirable species of grasses and legumes and others
obtained in the same farming system were significantly (P<0.05) higher than the road
side grazing types and lower (P<0.05) in total grass biomass, highly desirable grass, and
total biomass than enclosure areas. The total dry matter biomass, dry matter biomass of
grass and highly desirable grasses, and legumes are significantly (P<0.05) higher in the
enclosure followed by communal grazing and the road side grazing areas. In general,
there is low feed resources conservation and utilization and very poor traditional grazing
land management system in Metema. The abundant feed resources in the wet season
are wasted. In the dry season, grasses are turned to ash by wild and man-made fire in
the process of forest honey harvesting and crop land cleaning. As a result, the livestock
populations seriously suffer from the critical feed shortage during the long dry season.
The rangeland, species composition and biomass production are also affected by human,
livestock and natural factors (biotic and abiotic factors). The human population of the
district has increased due to settlement programs, investment induced settlers, expanding
crop cultivation and have increased the pressure on the rangelands and natural grazing
areas. Bush encroachment and overgrazing are also serious problems. Shifting cultivation
practice is also contributing to the increased bush encroachment. The seasonal
movement and transhumant livestock production by highlanders in adjacent districts
also increases the grazing intensity. Absence of adequate baseline information about the
rangeland resources, unsynchronized seasonal availability of feed resources and cropland
encroachment to the rangeland are some of the main constraints of the district, and
studies on rangeland management systems and improved livestock production should be
initiated.

xi

WorkingPaper_25.indd 11 1/10/2011 10:54:20 AM

 1 Introduction
Ethiopia, with its diverse climate and topography, is a country with wide resources and
traditional skill and experience in livestock rearing (Ayana 1999). About 62% of the total
land surface in the country is suitable for grazing (Mengistu 1998). In most developing
countries, rangelands have contributed to the major portion of feed consumed by
ruminants. In Ethiopia more than 90% of the ruminant livestock feed on natural pastures,
which vary in composition depending on the agro-ecology (Alemayehu 2005). Rangeland
is defined as land producing natural forage for animal consumption (Coppock 1994).
Most rangelands are at best only marginally suitable for arable cropping, and in Ethiopia
there are extensive areas where livestock raising on the natural vegetation is the only
possible types of land use.

The lowlands of the country are found below 1500 metre above sea level (masl) and
are estimated to cover about 78 million hectares, which is about 61–65% of the total
land area of the country (Friedel et al. 2000). They are home for about 12% of the human
and 26% of the livestock population (Beruk and Tafesse 2000). Pastoral communities
dominate the lowland areas of the country. Low human population density and highly
variable and uncertain rainfall characterize the lowland areas. In the pastoral community,
grazing biomass is entirely determined by the amount, pattern and timing of rainfall.

The rangelands are presently undergoing extensive deterioration both in quantity

and quality (Belaynesh 2006). The rangelands have limited capabilities in vegetative
production and in providing reasonable animal sustenance and production due primarily
to adverse environments including low and seasonal rainfall; moisture gathering winds;
varying degrees of poor soil; soil erosion; lack of or inadequate forage and grazing
management; and overstocking rates (Alemayehu 2005). Intensity of grazing and
browsing and restriction of livestock mobility have more serious consequences on the
rangelands than the number of animals owned by the pastoralists. Community structure is
vastly altered when improper grazing continues for long periods (Holcheck et al. 1998).
In seasonally dry environments, the main limitations to animal production are the lack
of green feed for at least half of the year coupled with the low nutritive quality of forages
during most of the active pasture growth period (Alemayehu 2005). The low nutritive
quality of the forage during the growth period is mainly due to environmental stresses
such as high temperatures and infertile soils (van Soest 1988). The major determinant of
livestock productivity is dry-matter intake which, in turn, is influenced by the palatability,
chemical composition and physical attributes of the diet, provided that the livestock are
disease-free. The chemical composition of the fibre is important in providing indications
of fermentation rates. However, interactions between the environment and plant

WorkingPaper_25.indd 1 1/10/2011 10:54:20 AM

 physiology and growth are sufficient to render associations between fibre components
and nutritive value unreliable (van Soest 1988).

The Amhara Regional State is located in the northwestern part of the Ethiopia. The region
covers about one-eighth of the total area of the country and is home to about 27% of the
total human and 35% of the total livestock population (BoA 1999; Befekadu and Berhanu
2000). The study area (Metema) district is one of the 104 districts of the Regional State,
which is found in North Gondar Administrative Zone bordering the Sudan. The district is
broadly characterized as lowland agro-ecology, and livestock production is an integral
part of the land use system. Cattle and goat rearing is a common practice. The area has
relatively high potential natural pasture during the wet season (IPMS 2005; Elias et al.
2007) According to Sisay (2006), the major livestock feeds available in the study areas are
natural pasture, crop residues, crop aftermath and hay. Based on his estimation, the mean
utilizable DM biomass of crop residues in the study areas was 3.68 t per household.
Furthermore, the total ME (Metabolizable Energy), and DCP (Digestible Crude Protein)
produced in the district were 76,636.81 MJ and 736.75 kg, respectively which could
satisfy the nutrient requirements of the livestock owned per household. These results may
hold true for the wet season because of the availability of extensive grazing land in the
area as compared to the district’s livestock population. Feed shortage in the dry season is
becoming a serious problem and the situation of the area is changing very fast because of
the current resettlement program and the large number of seasonal transhumant livestock
movement from the neighbouring three highland districts (Chilga, Dembia and Gondar
Zuria) into the district. The population pressure is undoubtedly increasing resulting in
deforestation, encroachment of farmlands into forest areas, and diminishing grazing
lands. The poor feed conservation practice of the community has also resulted in decline
of the potential of the rangelands and has exacerbated the feed shortage, particularly
during the dry season. According to Sisay (2006), the number of livestock and the
available feed resources is not proportional to be beneficial for livestock rearing.

Absence of adequate baseline information about the rangeland resources is considered

as one of the bottlenecks for development of rangelands in Ethiopia (Amsalu 2000). Thus,
to ensure sustainable use and development, characterization of the rangeland condition,
assessing and identifying the major feed resources of the area is imperative. This study
was therefore undertaken to generate information on the rangeland resources, to
characterize the existing rangeland, to assess the natural grazing land condition based on
herbaceous, woody and soil condition, to determine the major feed resources, to evaluate
their nutritive value and their utilization practices, and to forward suggestions for future
sustainable management of the rangelands. A number of districts in Tigray and Amhara
Regional States have similar agro-ecology to Metema, and the results of this study will
have broader relevance to the large area of the lowlands of northwestern Ethiopia.
2

WorkingPaper_25.indd 2 1/10/2011 10:54:21 AM

 2 Background
2.1 Livestock resources in Ethiopia
Although statistical data on livestock in Ethiopia have never been consistent, the latest
estimates indicated that Ethiopia has the largest livestock population and the highest
draught animal population in the African continent. Data on Ethiopian livestock resources
are presented in Table 1. Despite its huge numbers, the livestock subsector in Ethiopia
is not as productive as compared to its great potential and the direct contribution to the
national economy is very limited. The poor genetic potential for productive traits, in
combination with the suboptional general management situation that the animals are
exposed to are the main contributors to the low productivity.

Table 1. Livestock population and regional distribution in Ethiopia

Thousand heads
Region
Cattle Sheep Goats Equines Camels
Tigray 3103 1376 3107 476 32
Afar 473 403 801 26 171
Amhara 12,748 8987 6022 2438 50
Oromia 2245 9098 7439 3738 255
Somali 620 1162 283 96 24
Benishangul Gumuz 411 84 321 49 –
SNNPR 9263 3838 2626 732 –
Gambela 130 17 31 – –
Harari 44 4 36 8 –
Dire Dawa 48 43 122 13 5
Ethiopia—Total 49,297 25,017 21,884 7209 759
Source: CSA survey (2008/09).

2.2 Rangelands of Ethiopia

Natural vegetation integration reflects the whole of the natural environment. If
topography, geology and soil are not altered markedly, the change in vegetation
usually reflects a change in rainfall (Alemayehu 2005; Abule et al. 2007a). Basically,
the vegetation of an area is a product of the material available and the environmental
conditions prevailing. The latter includes the environment, landform, soil, climate and
factors such as fire and grazing and modification, circulation of minerals and plant
decay. Furthermore, for a complete understanding of vegetation, it is necessary to
consider the past as well as the present, for each sample of vegetation have a history
and a background of plant colonization and succession. Often, of course, the present
vegetation represents a stage of regression from a more highly developed or vigorous
community that has been brought under stress, perhaps through overgrazing.
3

WorkingPaper_25.indd 3 1/10/2011 10:54:21 AM

 The pastoral rangelands of Ethiopia are located around the peripheral or the outer edge
of the country, almost surrounding the central highland mass (Alemayehu 2004). The
areas are classified as marginal arable and non-arable land and comprise about 62%
(767,600 km2) of the country’s land area. Most of these areas are below 1500 metres
above sea level (masl) with the southwest and the southeastern areas having an altitude
of around 1000 masl and the southeastern and southwestern rangelands rising up to
1700 masl (Kidane 1993). Climate in the lowlands includes arid (64%), semi-arid (21%)
and subhumid (15%) zones largely defined by four rainfalls and temperature regimes.
These zones vary markedly in terms of number of plant growing days per year, forage
production, common plant associations, livestock and human carrying capacities and
incidence of important livestock diseases. Ethiopia has over 70 million heads of livestock,
and the lowlands are home to 12% of the human and 26% of the livestock population
(CSA 2009). Various forms of pastoralism and agropastoralism dominate. Livestock
depend upon rangelands consisting of native vegetation, with crop residues increasing in
importance as livestock feed as annual rainfall increases. According to Coppock (1994),
calculated for the lowlands overall, roughly six people/km² are dependent on 11 Tropical
Livestock Units (TLUs), which are composed of cattle (49%), goats (16%), equines (16%),
camels (12%) and sheep (7%). In contrast, the highlands support 72 people/km² and
dependent on 44 TLUs/km² which are dominated by cattle (76%), equines (14%), sheep
(8%) and goats (2%). Thus, although the lowlands comprise over 50% more land area
than the highlands, the lowlands have only 40% as many TLUs at one-quarter the density.

2.3 Major livestock feed resources in Ethiopia

The major livestock feed resources in Ethiopia are natural grazing and browse, crop
residues, improved pasture, and agro-industrial by-products (Alemayehu 2004). The
feeding systems include communal or private natural grazing and browsing, provision
of crop residues and cut-and-carry feeding. At present, stock are fed almost entirely on
natural pasture and crop residues. Livestock are grazed on permanent pastures, fallow
land and cropland aftermath (Alemayehu 2004; Abule et al. 2007a).

2.3.1 Feed availability and nutritive value of range forage

Feed availability

Natural pasture comprises the largest feed resource and estimates of its contribution to
overall feed resource vary greatly. This is because the quantity and quality of native pasture
varies with altitude, rainfall, soil and cropping intensity. Seasonal fluctuations of feed
resources in the tropics also follow the pattern of vegetation growth which is modified by

WorkingPaper_25.indd 4 1/10/2011 10:54:21 AM

 availability of rainfall. Alemayehu (1998) estimated that about 80–85% of the ruminant feed
in Ethiopia comes from natural pasture. The total area of grazing and browsing is estimated
at 62,280 million hectares, of which 12% is in the crop farming areas (with more than 600
mm rainfall) and the rest around the pastoral areas (Alemayehu 2005). Communal grazing
is managed as a common property resource. Estimating carrying capacity, if calculated
on plant availability, should allow a plant use of 30–50% (de Leeuw and Toothill 1993).
Important principles of rangeland management that affect feed availability include stocking
rate, rest, frequency of grazing, top hamper and litter cover (Alemayehu 2005; Abule et al.
2007a). Trees are an important component of the rangelands and serve for environmental
conservation; provide fuel wood and building materials. They are also important source of
feed for browsers and their value of tree litter as feed and shade to livestock should not be
underestimated (Alemayehu 2005; Abule et al. 2007a).

Feed quality

Chemical analysis of range forage plants serves as a comparative measure of differences

between species and changes with season. Chemical analyses are also useful for
measuring differences on effect of stage of growth, and effect of site quality on chemical
constituents. Simbaya (1998) reported that the quality of natural pastures is influenced
by the absence of legume species in communal grasslands. This limits the nutritional
quality of available fodder and animals are thus unable to meet their protein, energy
and mineral requirements. As suggested by Osuji et al. (1993), poor nutrition is one of
the major constraints to livestock productivity in sub-Saharan Africa. This is because
animals thrive predominantly on high-fibre feeds (straws, stover and native pasture hay)
which are deficient in nutrients (nitrogen, sulphur, minerals, phosphorus etc.) essential for
microbial fermentation. Consequently, the digestibility and intake of digestible nutrients
are unavoidably low. Data on chemical composition of major feed resources as affected
by season in Metema district are presented in Table 2.

Table 2. Nutritional quality of major feed resources in Metema district as influenced by season

% chemical composition
Feed types Season
DM ASH CP ADF ADL NDF IVDMD
Natural pasture August 93.4 9.4 6.4 45.8 4.7 78.3 57.8
October 94.1 10.3 5.7 47.8 5.5 78.5 56.9
Fodder August 92.3 15.4 14.4 37.0 13.3 54.9 51.6
October 92.9 8.6 13.2 40.6 13.2 54.5 47.9
Hay 94.5 10.5 7.2 41.2 4.8 76.8 54.3
Sorghum stover 92.4 7.9 3.1 44.9 6.3 75.5 46.9
Teff straw 93.2 8.9 4.3 46.9 8.0 76.0 41.0
Millet stover 92.4 9.8 4.2 44.2 5.9 72.3 52.8
Source: Sisay (2006).

WorkingPaper_25.indd 5 1/10/2011 10:54:21 AM

 Range forage varies in quality from time to time and from place to place. Plants are most
nutritious during the growth stages. Once mature, plants are subject to leaching and
dilution of nutrients and reduction in nutritive value. Declines in nutrient composition
and leaching are especially serious in the case of herbaceous plants (Alemayehu
2006). As plants mature, contents of crude protein, readily digested carbohydrates and
phosphorus decrease, while contents of fibre, lignin and cellulose increase (Osuji et
al. 1993). Most grasses and tree leaves in arid environments are low in nutritive value
because of high contents of lignin and relatively indigestible (compared with starch)
cellulose and hemicellulose. The plants require such substances as an adaptation
mechanism to protect themselves from high temperatures and evapotranspiration.
Unfortunately, these substances reduce their nutritional content and digestibility (Mathur
et al. 1991). The stage of growth, maturity of grasses and taste influence their nutritional
value. Mathur et al. (1991) reported that cattle kept solely on grazing mature grass
exhibited loss in body weight during the dry season. The use of some roughages is also
limited by their low contents of protein and digestible energy (Ncube and Mpofu 1994),
especially as the season progresses. Further problems are caused by secondary factors
such as antinutritive or toxic constituents in the plants, which restrict acceptability of the
diet (Osuji et al. 1993; Kumar and D’Mello 1995).

Crop residues are roughages that become available as livestock feeds after crops have
been harvested. Residues can usually be grouped by crop type such as cereals, grain
legumes, roots and tubers. Apart from being a source of animal feed, residues are also
used as building, roofing and fencing materials, as fuel, and as fertilizer or surface mulch
in cropland. Their value when used as feed depends on the demand from livestock
owners, which varies with the overall supply and demand situation for feeds. This in turn
depends on the livestock density and the supply of other feed resources, in particular
forages and browse from natural vegetation (de Leeuw and Rey 1995). The supply of crop
residues is a function of the proportion of land used for cropping and of the edible feed
yields per unit of land, where consumable feed from the crop residues exceeds grazing
from natural pastures (expressed in tonne of dry matter per hectare, t DM/ha). Depending
on the production system, expansion of cropland may have a positive effect on overall
feed supply

2.4 Causes of rangeland degradation

Rangeland degradation may be defined as the loss of utility or potential utility or
the reduction, loss or change of features of rangeland ecosystem, which cannot be
replaced (Chrisholm and Dumsday 1987). In general, rangeland degradation implies a
reduction in rank or status, which includes a loss of topsoil, a change to a simple floral/

WorkingPaper_25.indd 6 1/10/2011 10:54:21 AM

 fauna composition or a transition from one organic form to a lower organic form, and
continuous reduction of productivity/biomass of the ecosystem. Generally, a lower
biological diversity is supposed to occur in a degraded rangeland.

2.4.1 Drought and shortage of rain

Prolonged drought including shortage and erratic rainfall can cause serious range
degradation. Rainfall during drought is hardly adequate to allow grasses to grow and
unable to fill the surface water ponds (Cossins and Upton 1988).

2.4.2 Bush encroachment

Overall, woody vegetation reduces grass cover through increasing competition for
available water and nutrients and reducing light reaching the grass layer. In addition to
competing with grasses, these noxious woody plants are commonly thorny and thicket
forming and reduce the grazing capacity of the rangeland (Alemayehu 2004). From
rangeland management perspective, understanding the factors that contribute to invasion
process of undesirable woody vegetation is important. Many factors may be involved
in bush encroachment. Overgrazing, including high stocking rates, is claimed to be the
major problem and a high concentration of woody plants are found around water points
where stocking densities and grazing intensities are relatively high (Cossins and Upton
1988). As reported by Archer (2003), the characteristics common to many woody species
that increase in grazed environments include high seed production, seeds that persist in
soil for many years, ability to disperse over long distances, ability to sprout following top
removal, tolerance to low levels of water and nutrients and low palatability. A report by
Alemayehu (2004) indicates that the ecological succession in the Borana rangelands of
Ethiopia indicates that the potential of grassland is threatened by bush encroachment in
many areas.

2.4.3 Over population and overstocking

Increase in human population necessitates the increase in livestock population in
rangelands in order to maintain survival. In pastoral areas of Ethiopia, the animal and
human populations are growing at an increasing rate, while the pasture resource on
which they depend is limited or diminishing both in terms of grazing area and range
productivity (Coppock 1994). These increases in population and over stocking are
increasing the imbalances in the Borana range system and have already resulted in over
grazing and range degradation (Alemayehu 2004). Gamedo (2004) also reported that
overgrazing has been one of the major factors for rangeland degradation in Borana and

WorkingPaper_25.indd 7 1/10/2011 10:54:21 AM

 the relatively good rangeland condition in ranches and kalos may show the fact that
overgrazing has contributed to rangeland degradation.

2.5 Factors affecting rangeland vegetation

2.5.1 Climate and soil
Climate plays a primary role in determining the main types and growth responses of
vegetation used for grazing (Alemayehu 2004; 2005). The amount and distribution
of rainfall determine the form and productivity of the natural vegetation. Edaphic
characteristics may substantially modify climatic factors in various ways. High natural soil
fertility levels increase the vegetation’s response to moisture while soil volume and water
holding characteristics condition the soil water storage capacity. Surface characteristics
determining run-off and infiltration of water, and subsurface characteristics determined
the level of drainage. Over grazing can lead to bareness and loss of topsoil by erosion
to such an extent that the vegetation assumes a drier appearance than the rainfall data
suggests.

2.5.2 Animal management

Livestock grazing can have profound impact on vegetation. The general pattern of
grazing-induced vegetation change is well documented (Alemyehu 2004; 2005). It is
known that less palatable plants increase at the expense of more palatable species.
Community structure is vastly altered when improper grazing continues for long period.
The poor animal production experienced in rangelands has long been attributed to the
poor quality of forage. This is generally confirmed by chemical analysis of the range
plants. However, sampling of such material often overlooked selective grazing by
animals between species and deferent parts of species and selectivity is of considerable
importance (Holechek et al. 1994).

2.6 Factors influencing vegetation composition

of the rangeland
The species composition of rangeland varies depending on topography, climate and
soil types. Different grasslands contain various types of grasses, legumes, and other
herbaceous species. The botanical composition of a plant community can change
because of many factors, including altitude, grazing practices, burning, drought, and
temperature effects, pest, and erosion. Depending on the nature of this compositional
change, the productivity of an area (in terms of its capacity to support livestock) may

WorkingPaper_25.indd 8 1/10/2011 10:54:21 AM

 change. A change in plant composition results because of the relative adaptability of
different species to these influences (Alemyehu 2004; 2005).

2.6.1 Effects of grazing on vegetation composition

Natural pasture communities are very complex consisting of a large range of grasses,
shrubs and herbaceous species among which only few species are palatable. Livestock
are able to selectively graze a small proportion of the palatable herbage available and
ignore the undesirable ones. The most palatable species are selected first and closely
defoliated. If the grazing pressure is high, then a decline in the quality and productivity
of rangeland occurs (Cossins and Upton 1985). This causes reduced vigour, less seed
production and eventually plant death. Overgrazing can also lead to extensive sheet and
gully erosion (Alemyehu 2005). Since the whole of plant organism, leaves, stems, and
roots react to the degree of trampling or grazing which it receives, a weakening of the top
growth results in a lighter short root system that dies back from the bottom. Grass roots
continue normal growth when not more than about 40–50% of the vegetative parts are
removed (grazed) during the period of active growth. Therefore, the effects of overgrazing
can be overcome if rangelands are properly managed. According to Laze and Swain
(1969), grasses naturally need certain rest periods to develop, to grow, to seed and to be
able to build reserves for the next growing season.

2.6.2 Response of plants to grazing

The response to grazing by various plants is dependent on the reduced vigour of plants
grazed repeatedly without the opportunity to replenish food reserves. It is also based
on the high degree of selectivity exhibited by grazing livestock. Cattle prefer grasses,
sheep prefer forbs, deer prefer browse. Within each category, there are ice-cream plants
(decreasers), which the grazing animal will seek out. Repeated grazing will have the same
effect as repeated clipping. A third factor independent of the plant–animal interaction is
the presence or absence in the climax plant community. The classification of decreaser
and increaser is based largely on the preference exhibited by the livestock. This is
dependent on two major criteria: (1) the range site and (2) the livestock. The range site
is important because it limits the selectivity available. This explains why some plants
are decreasers on one site and increasers on another. In the first case they are the most
palatable species, while in the second there are other plants that are more palatable. The
kind of livestock will determine general categories of preference.̕

WorkingPaper_25.indd 9 1/10/2011 10:54:21 AM

 2.7 Range condition assessment
Range condition is the present status of vegetation of a range site in relation to the climax
(natural potential) plant community for that site. It is an expression of the relative degree
to which the kinds, proportions, and amounts of plants in a plant community resemble
that of the climax plant community for the site (SRM 1999).

According to Pratt and Gwynne (1977), range condition is the state and health of the
range, which can be assessed on the basis of an area’s vegetation composition, plant
vigour, ground cover and soil status. The concept of ‘condition’ implies that an optimal or
desired vegetation cover (in terms of quantity and composition) exists for each particular
land system. However, since it will often be uncertain what the desired or ‘optimum’
condition is (particularly in areas which have undergone misuse for a considerable period
of time), and since optimum range condition will differ according to the manner in which
the range is used (e.g. cattle, sheep, wildlife), the comparison used should be clearly
stated as well as whether this is based on actual measurements or whether it is assumed.
As cited by Amaha (2006), rangeland condition is a concept encompassing the levels
of specific indicators such as plant species composition, vegetation cover (basal cover),
forage production (productivity), land condition (soil erosion and compaction) and
management at a particular location(s) aimed at sustained livestock production (Friedel et
al. 2000).

According to Mannetje et al. (1976), the determination of the botanical composition

of rangeland is important in understanding the fodder value of individual species and
their reaction to biotic and edaphic factors, which may be explained in terms of type of
species, yield, and frequency of occurrence, density and basal cover. Plant dry matter
yield is often directly related to animal production, while the other parameters are useful
to describe and quantify the plant population and successional trends of the rangeland
vegetation (DuToit and Aucamp 1985) and to assess the rangeland condition (van der
Westhuizen et al. 1999; 2001). Methods of classifying range condition have emphasized
species composition, growth form composition, forage productivity, a combination of
a number of different vegetation and soil attributes, such as cover, composition, vigour,
palatability, litter and soil stability.

If overuse is excessive or continued over long periods of time, invaders or undesirable

plants are found. The invader plants were thought to be absent in the original vegetation
but through grazing pressure have replaced the decreaser and increaser plants. In
favourable years invaders can provide considerable forage for a short period of time
but sound range management cannot be based on this uncertain forage production.

10

WorkingPaper_25.indd 10 1/10/2011 10:54:22 AM

 The four classes of range condition are based on production percentages of decreaser
and increaser plants present on the site as compared to the original vegetation. A site
composed of decreasers and increasers indicates a high condition range. As decreasers
on the site are replaced with increaser and invader plants, the site shows need of
improvement. The four range condition classes are as follows:

Excellent condition: 76–100% of allowable vegetation is mixture of original highly

palatable, desirable perennial decreasers and increasers. Legumes and desirable forbs
may be present.

Good condition: 51–75% of vegetation is mixture of original highly palatable, desirable

perennial decreasers and increasers. Some legumes and forbs may be present.

Fair condition: 26–50% of allowable vegetation is mixture of original highly palatable,

desirable perennial decreasers and increasers. Some legumes may occur, but most forbs
are increasers and invaders. Overall vegetation appearance is shorter and amount of bare
ground is increasing.

Poor condition: less than 25% of all vegetation is composed of highly palatable, desirable
perennial decreasers and increasers. Invader plants and unallowable increasers comprise
majority of vegetation.

2.7.1 Estimating range condition

Range condition refers to the present ecological status in terms of productivity of a
vegetation community relative to its natural potential for particular range site and types
of land use (SRM 1999). In other words the concept ‘condition’ implies that an optimal
or desired vegetation cover (in terms of quantity and composition) exists for each
particular land system. Range condition is based on the species composition of the plant
community as estimated by the percentage of the total annual air-dry weight of each
species. Species must be classified as decreasers, increasers, or invaders. Each species has
an allowable percentage that occurred in climax.

2.7.2 Range sites

Before range condition can be assessed the range sites must be located. Range sites
are the basic unit of land for practical use. Ideally, each range site should respond in
the same manner to climatic variation, have uniform topography and productivity and
respond uniformly to experimental treatment (Alemayehu 2006).

11

WorkingPaper_25.indd 11 1/10/2011 10:54:22 AM

 2.7.3 Range trends
The direction of change in ecological status or resource value rating observed over time
(SRM 1999). It describes the current health of the range, and then range trend indicates
whether it is getting better, worse or staying the same. Consequently, range trend is the
best single indicator of the success or failure of current management practices. It can be
determined by two range condition determinations. However, changes in range condition
can be fairly slow and the ability to properly estimate range condition has sufficient error
to require that meaningful changes must be based on readings at least five years apart.
This is too long; management must be evaluated over shorter periods of time. Indicators
of trend are very useful tools.

2.7.4 Range condition classification

Range condition classification is often included in a range inventory. Changes in range
condition scores overtime are usually the basis for monitoring management effectiveness.
Range condition classification provides an induction of management inputs necessary.
If ranges are in good or excellent condition, maintaining them in a stable condition
may be the best management strategy. However, if they are in poor or fair condition,
management, that is aimed at ‘improvement’ may be indicated. Generally, four or five
condition classes are recognized: excellent, good, fair and poor. Sometimes a fifth
category is added. Many approaches have been used to determine range condition
on different range sites or habitat types. Dyksterhuis (1949 1958) developed the most
familiar method. This approach is ecological, in that range condition is measured in
degree of departure from climax. The approach assumes that climax can be determined
for each range sites. Excellent class would represent climax, i.e. excellent (76–100);
good (50–75); fair (26–50); and poor (0–25). Originally, species occurring on each
site were classified, by their reaction to grazing, as decreasers, increasers, or invaders.
Dcreasers are highly palatable plants that decline in abundance with grazing pressure.
Plants classified as increaser I types are moderately palatable and ‘serve as secondary
forage plants’. They may increase slightly, or remain stable under moderate grazing
condition and reaches fair condition, they also decline. Other plant species present in
the climax vegetation but that are unpalatable may increase under grazing pressure or as
site deterioration occurs. These species are classified as increaser II plants. Invaders are
species that encroach on to the sites from adjacent sites in a later stage of deterioration.
Type I invaders may eventually decrease if forced utilization occurs at later stages of
deterioration; while type II invaders are generally unpalatable and increase even at the
final stages of deterioration.

12

WorkingPaper_25.indd 12 1/10/2011 10:54:22 AM

 2.8 Biomass estimation methods
Biomass or standing crop usually refers to the weight of organisms present at one time
(SRM 1999). Increases in biomass through growth process of photosynthesis over time are
generally considered productivity estimates that include a time dimension. In the English
system biomass is generally expressed in pound per acre, and in the metric system, kg per
hectare. Productivity estimates would include a unit of time (e.g. annual per day, week,
month etc.). Most estimates of plant biomass or standing crop include only that above
the soil surface. This material is commonly available to larger herbivores. Below ground
biomass is very important for plant functions, but is difficult to measure and generally not
included in inventory or monitoring procedures. Direct harvesting is considered the most
reliable method of determining above ground biomass. However, this method is too time-
consuming to be of practical value for inventory or monitoring of extensive range areas.
Several weight estimate techniques have been developed for rapid and fairly reliable
determination of herbage weight. These procedures involve estimating herbage weight by
species from small quadrates in the field. Training of observers in the field is necessary.
This can be done easily by checking the estimates with clipped quadrate (Alemayehu
2005). The method is considered reliable enough to be used on detailed research studies.
Weight estimates can be adjusted by clipping a portion of the quadrates that have been
estimated.

13

WorkingPaper_25.indd 13 1/10/2011 10:54:22 AM

 3 Materials and methods
3.1 Description of the study area
3.1.1 Location, climate and topography
The study was conducted in Metema district, North Gondar Zone of Amhara Regional
State (Figure 1). Metema is located between 12º40’00” N and 36º8’00” E. It is 925
km northwest of Addis Ababa and 180 km west of Gondar town. The district has an
international boundary of more than 60 km long distance between Ethiopia and Sudan.
It has 18 rural and 2 urban kebeles. According to IPMS (2005), out of the total 18 rural
kebeles in the district, 16 are under moist kola and the remaining two kebeles are dry
kola. Mean annual temperature ranges from 22ºC to 28ºC and daily temperature reaches
as high as 43ºC during the months of March to May. Mean annual rainfall ranges from
850 mm to 1100 mm (IPMS 2005) and has unimodal pattern. The rainy months extend
from June to end of September. Altitudes range from 550 to 1068 masl.

Metema

Figure 1. Location of Metema woreda in Ethiopia.

3.1.2 Human population and livestock resources

According to CSA (2005), the human population in the woreda is estimated at 76,084
rural (54.2% male) and 18,467 urban (50.0% male). The majority of the population (80%)
lives in rural areas with farming as a major occupation. The original residents of the area

14

WorkingPaper_25.indd 14 1/10/2011 10:54:22 AM

 are the Gumuz people. Until recently, they practised slash and burn and hunting wild
animals. The area has been gradually populated by settlers from the highlands. According
to OoARD (2007), through a new resettlement program, the government has settled a
total of 33,159 people between 2003 and 2005.

Livestock production was an integral part of the land use system. Production of cattle
for draught power, milk, and meat; small ruminants for income and meat and donkeys
and camels for transport and poultry for income is a common practice. The livestock
population is composed of 136,910 cattle, 32,024 goats, 1686 sheep, 7164 donkeys,
7127 poultry, 400 camels and 23,789 beehives (OoARD 2007). Transhumance
production system was commonly practised by the highlanders. According to Azage et
al. (2008), about 80% of the inhabitants of the highland Chilga, Dembia and Gondar
Zuria districts trek their livestock to Metema in search of feed during the main rainy
season from May to October. The major cattle breed of the study area is Fogera crossbred
with other highland zebu cattle. Ruthana cattle originally from Sudan and Felata cattle
from Niger and Nigeria also constituted smaller proportion of the cattle population. The
main small ruminant resource is goat production. There is a small proportion of sheep
population locally known as the ‘Gumuz sheep’. Important livestock diseases include
infectious diseases, internal and external parasites.

3.1.3 Land use and farming systems

According to OoARD (2007), land use pattern of Metema district is comprised of forest
and rangelands 72.0% (312,300 ha), cultivated land 23.6%, (103,908 ha of which
71,324 ha, 13,908 ha and 18,676 ha are smallholder farms, commercial farms, and
potential cultivable land, respectively), and uncultivable land 5% (23,877 ha). About
60% of the district is plain, and 20, 15, and 5% are sloppy, undulating and valley
bottoms, respectively.

Cotton–livestock farming system

According to IPMS (2005), 4 out of the 18 peasant associations (PAs) belong to this
farming system. They are Maka, Awlala, Genda Wuha and Kemechela. They are found
in the northeastern part of the woreda. These PAs predominantly grow cotton followed
by sorghum and sesame in few areas. The PAs in this farming system have some different
features in terms of suitability for crop production and amount of rainfall received. These
PAs are relatively cooler in temperature and have higher altitude and rainfall. Most of the
area is flat with black Vertisol and have water logging problem. As a result, the area is
suitable for growing cotton and rice. Farmers in these PAs practice slightly early planting

15

WorkingPaper_25.indd 15 1/10/2011 10:54:22 AM

 of crops. Cotton is grown in bigger plots, while sorghum and sesame are planted in very
smaller areas. The primary and secondary sources of income in this farming system were
crops (97.7%) and livestock (97.6%). Off-farm activity was the third important source of
income for 82% of the respondents.

Sesame–livestock farming system

Fourteen PAs belong to this farming system, and sesame, cotton and sorghum are the
major crops produced (in order of importance). Altitude (550 to 700 masl) and rainfall
(700 to 900 mm) in this farming system is lower than the cotton–livestock farming system.
The underground water table is high, and in some places water could be obtained at
less than 10 m depth. In addition, there are three rivers in this farming system that could
be used for irrigated crop production and livestock rearing. Extensive grazing areas and
natural gum and incense trees are found in this farming system. Sorghum is an important
crop and it is the main food crop for all smallholder households and labourers in
commercial sesame and cotton farms. The major sources of income in this farming system
were crops (95.2%), livestock (84.8%) and off-farm (57.1%).

3.2 Selection of kebeles and farmers

A single-visit formal survey method (ILCA 1990) was used to collect information on
livestock feed resources and rangeland utilization practices in the district. A total of
7 kebeles were selected randomly out of the total 18 kebeles. Then, a total of 140
respondents were selected using purposive sampling technique and interviewed for
collection of primary data. A structured questionnaire was prepared as per the objective
of the study and was pre-tested. In addition, observations, group discussions with elders,
development agents and key informants were undertaken. Secondary data were reviewed
and browsed to consolidate the information.

3.3 Rangeland condition assessment

3.3.1 Range sites selection and sampling procedures
Discussions were made with the community members, elders and agricultural experts
about the locations and conditions of major grazing areas in the two farming systems.
Short visits were made to these sites for assessment. The number of sites was then decided
based on proportions of the available grazing land within farming system. Categories of
the grazing lands were further stratified into three sampling areas: communal grazing,
road side and enclosure using the systematically stratified random sampling technique.

16

WorkingPaper_25.indd 16 1/10/2011 10:54:23 AM

 Enclosures used as benchmark was selected from protected areas in schools and
government protected acacia wooded grass land for sesame–livestock farming system.
The road sides were selected 200 metres away form the main road to avoid any edge
effect. Road side grazing lands were selected because they are exposed to continuous
grazing and trampling by transhumants and inhabitants.

In each site, a sampling block of 4 km × 1 km, considered homogenous and

representative of the vegetation cover under investigation, was demarcated. This was
further stratified into four sampling plots of equal size (1 km × 1 km each) in order to
encompass vegetations of herbaceous and woody layers. In each plot, a belt transects
of 50 m × 4 m was laid out randomly. A total of 68 composite samples (24 from
communal and 24 from road side) for herbaceous and 21 composite samples for woody
vegetation were collected from both farming systems. In enclosure areas, 8 and 12
sampling units were used in cotton–livestock and sesame–livestock farming systems,
respectively. The numbers of quadrates laid in the composite sampling units differed
according to the proportion of the grazing areas and the factors assessed. For grass
species composition, three and four quadrates per one composite unit were used in
the cotton–livestock and sesame–livestock farming systems, respectively. For woody
vegetation assessment three belt transects per one composite sampling unit was used
in both farming systems. For biomass estimation four quadrates per one composite
sampling unit was employed in both farming systems. Compass and GPS were used to
measure the transect locations and coordinates. Sampling was done from 10 August to
10 September 2007 when almost all the pasture plants were fully-grown with over 50%
at flowering stage.

3.3.2 Species composition assessment

At each sample site, the composition of herbaceous species was assessed by harvesting
three or four quadrates of 1 m × 1 m area (depending on the homogeneity and
heterogeneity of the range sites) randomly by throwing a quadrate each time towards the
back and the herbaceous species within the quadrates and this was cut to ground level.
The sample was weighed immediately using a spring balance and was transferred into
properly labelled paper bag and fasten at the top. Samples were kept under a shade until
sampling for the day was completed. Each of the sample in the paper bag was then hand
separated into different species and separately weighed again. Subsequently, samples
were sun dried. Dry matter (DM) of each species was determined in an oven (60ºC for 72
hours) at the ILRI laboratory Addis Ababa and the percent composition of each species
were determined on DM weight basis (ILCA 1990).

17

WorkingPaper_25.indd 17 1/10/2011 10:54:23 AM

 3.3.3 Identification of plant species
For identification of species, representative plants with flowering head and other
vegetative parts from each species were collected and dried in presses. After drying,
the specimens were mounted. Very few common species were identified right in the
field (Azene et al. 1993) and specimens were coded and transported to the National
Herbarium of Addis Ababa University for final identification.

3.3.4 Range condition assessment

The rangeland condition was assessed considering the three layers (grass, woody, and
soil factors) based on the suggestion made by Friedle (1991). For grass and soil layer
factors, the criteria developed in South Africa (Tainton 1981) and adapted to fit semi-
arid and subhumid environments by Baars et al. (1997) was used. Woody vegetation
layer was assessed using density enumeration based on the suggestion of Pratt and
Gwynne (1977) for east African rangelands. The assessment factors were based on the
composition of the grass layer, basal cover, litter cover, soil erosion, soil compaction,
and woody vegetation density (Appendix Table 1) summing up to a total of 50 points,
the overall rating of vegetation was interpreted as excellent (65–58.5 points), good
(47.5–41 points), fair (40–33.5 points), poor (32.5–26 points), very poor (≤ 25 points)
(Baars et al. 1997).

3.3.5 Herbaceous species composition

For herbaceous (mainly grasses) species composition, 1 to 10 points were considered.
Grass species were divided into desirable species likely to decline with heavy grazing
pressure (deceasers), intermediate species likely to increase with heavy grazing pressure
(increasers), and undesirable species likely to invade with heavy grazing pressure
(invaders), according to the succession theory (Abule et al. 2007). Classification of grass
species into these three categories was done by conducting detailed interview with
members of the local community about the palatability and distribution of each identified
grass species in relation to the intensity of grazing, and cross checking with the list of
grasses from the literature.

3.3.6 Basal and litter cover

For basal and litter cover a score rating of 0 to 10 points were considered as indicated
in Appendix Table 1. A representative sample area of 1 m2 was selected for detail
assessments of basal and litter cover. For the surface of basal cover of tufted grasses,

18

WorkingPaper_25.indd 18 1/10/2011 10:54:23 AM

 the distribution was assessed as follows: The 1 m2 was divided into halves, one of
which was again divided into eighths. All plants in the selected 1 m2 were removed
and transferred to the eighth to facilitate visual estimation. Only basal cover of living
parts was considered. The rating of basal cover for tufted species was considered
excellent if the eighth was completely filling (12.5) or very poor if the cover becomes
less than 3%.

3.3.7 Number of seedlings and age distribution

The number of seedlings was counted in three randomly selected areas, each the size
of an A4 paper (30 cm × 21 cm) (Appendix Table 1). An A4 paper was dropped from an
approximate height of 2 metre above the ground and the number of seedlings within the
paper was counted. The category: ‘no seedlings’, was given 0 point, and more than 4
seedlings were given the maximum score of 5 points. Similarly, if all age categories, i.e.
young, medium and old plants of the dominant species are present, a maximum score
of 5 points was given. If there were only young plants, then the minimum score of 1 was
given (Appendix Table 1).

3.3.8 Woody vegetation layer

In woody vegetation assessment, species composition, density, canopy cover, plant height
and hedge effect were considered. All species in the belt transect of 50 m × 4 m were
recorded and identified. The desirability and palatability of each species was recorded
based on group discussion with livestock owners with respect to the woody plants
sensitivity to grazing, abundance and preference by livestock as feed. This discussion was
supported by literature (Azene et al. 1993). The criteria developed by Kuchar (1995) were
used for scoring the percent of canopy cover of woody species. Height of a particular
species in the belt was measured using calibrated poles of appropriate sizes for different
woody species and five height classes (>0.5–1 m, >1–2 m, >2–3 m, 3–4 m and >4–5 m)
were employed (Amaha 2006). The density of woody species were enumerated from each
belt transect of 200 m2 area. Only live woody plants regardless of being single or multiple
stemmed were counted and recorded to estimate the woody vegetation density.

The number of trees/shrubs per plot area was determined and the lowest score (0 point)
was given for the highest density: i.e. 0 = > 5000/ha, 1 = 4001–5000/ha, 2 = 3001–4000/
ha, 3 = 2001–3000/ha, 4 = 1001–2000/ha, and 5 = 0–1000/ha (Abule 2003). The percent
canopy cover of the species in the belt transect was measured using a tape meter. The
ratio was computed as the measured canopy area by the remaining tape length and the
result percentage canopy cover was rated by the criteria developed by Kuchar (1995).

19

WorkingPaper_25.indd 19 1/10/2011 10:54:23 AM

 The rates given for cover in the range of 1–5 points, for woody covers of >45%, 36–45%,
26–35%, 15–25% and <15% points of 5, 4, 3, 2, and 1 were given, respectively. Hedging
was estimated visually based on the state of palatability of the species. The rating of
hedging effect for woody species (1–5 points) was based on criteria developed by Kuchar
(1995).

3.3.9 Soil erosion and compaction

Soil erosion was evaluated based on the amount of pedestals (higher parts of soils kept
together by plant roots, with eroded soil around the tuft) and in severe cases the presence
of pavements (terraces flat soil, normally without basal cover, with a line of tufts between
pavements). Soil compaction was assessed based on the amount of capping (crust
formation) scoring of 0–5 points (Appendix Table 1).

3.4 Evaluating the quality of major feed resources

3.4.1 Sample preparation
Representative samples of the major feed resources were collected. The samples were
stratified based on type and season. For each season, samples of the same feed type
were bulked and then thoroughly mixed and subsampled. Finally the subsamples were
oven dried at 65ºC for 72 h, and ground in Willey mill to pass through 1 mm sieve and
were kept in air tight containers pending analysis for chemical composition (van Soest
1988).

3.4.2 Laboratory analysis

Feed samples were analysed for DM and ash using the method of AOAC (1990).
Nitrogen (N2) is determined by micro-Kejeldhal method, and then CP (crude protein)
was calculated as N × 6.25. Neutral detergent fibre (NDF) and acid detergent lignin
(ADL) were analysed by the method of van Soest et al. (1991). The method of Tilley
and Terry as modified by van Soest and Robertson (1985) was used to determine
IVDMD.

3.5 Statistical analysis

Data on various aspects of livestock feed resources were analysed using Statistical
Packages for Social Sciences (SPSS 2003). Descriptive statistics (mean, percentage, range,
standard error of means) were calculated. For range condition assessments, 68 composite

20

WorkingPaper_25.indd 20 1/10/2011 10:54:23 AM

 samples for herbaceous species and 21 composite sample units respectively, and all in all
89 composite sample units (Table 3) were subjected to analysis using the General Linear
Model (GLM) procedure of SPSS version 12. One-way analysis of variance (ANOVA)
was used for variables of range condition. Correlation analysis was used to test if there
were relationships between rangeland degradation, biomass production and vegetation
variables.

Table 3. Experimental units used for analysis in each composite sample site
Grazing type
Farming system
Communal Road side Enclosure Subtotal
Herbaceous layer CLFS 12 12 8 32
SLFS 12 12 12 36
Total 24 24 20 68
Woody layer CLFS 4 4 2 10
SLFS 4 4 3 11
Total 8 8 5 21
Composite samples = 89; SLFS = sesame–livestock farming system, CLFS = cotton–livestock farming system.

21

WorkingPaper_25.indd 21 1/10/2011 10:54:23 AM

 4 Results and discussion
4.1 Socio-economic conditions
4.1.1 Household characteristics
Sex, ethnic group, education, and major occupation of respondents under the two
farming systems in Metema district are presented in Table 4. Overall, 87.1% of the
respondents are male, 10% spouses, 1.4% sons and 1.4% are daughters. From overall
respondents, 88.6 and 11.4% are male and female, respectively. There were five ethnic
groups in the districts. Of these, the Amhara dominate in both farming systems, being
76.3% in the sesame–livestock and 60% in cotton–livestock system. Overall, 86.4% are
Amhara, 10% Tigre 5% Gumuz, 3% Agew and 1% are others.

Table 4. Sex, ethnic group, education, and major occupation of respondents in the cotton–livestock
and sesame–livestock farming systems in Metema district
CLFS SLFS Overall
Variables
HHC % HHC % HHC %
Respondent status N = 80 N = 60 N = 140
Husband 67 83.8 55 91.7 122 87.1
Spouse 10 12.5 4 6.7 14 10.0
Son 1 1.3 1 1.7 2 1.4
Daughter 2 2.5 2 1.4
Sex of household head
Male 68 85.0 56 93.3 124 88.6
Female 12 15.0 4 6.7 16 11.4
Ethnic group of household
Amhara 61 76.3 60 100 121 86.4
Agew 3 3.8 3 2.1
Tigre 10 12.5 10 7.1
Gumuz 5 6.3 5 3.6
Other 1 1.3 1 0.7
Education status of household
Illiterate 33 41.3 35 58.3 68 48.6
Read and write only 29 36.3 18 30.0 47 33.6
Grade 1–3 7 8.8 3 5.0 9 6.4
Grade 4–6 9 11.3 2 3.3 9 6.4
Grade 7–9 2 2.5 2 3.3 7 5.0
Major occupation
Trade 4 5.0 4 2.9
Livestock rearing only 3 3.8 1 1.7 4 2.9
Crop production only 10 12.5 2 3.3 12 8.6
Mixed farming 61 76.3 55 91.7 116 82.9
Others 2 2.5 2 3.3 3 2.1
SLFS = sesame–livestock farming system, CLFS = cotton–livestock farming system, HHC = household count.

22

WorkingPaper_25.indd 22 1/10/2011 10:54:24 AM

 The overall educational status of the respondents indicate that 48.6% are illiterate; being
higher in sesame–livestock (58.3%) than in the cotton–livestock (41.3%) farming system.
Similar findings were also reported for pastoralist communities in South Omo (Admasu
2006) and Bale Zones (Teshome 2006) of Ethiopia. The findings may suggest that this
situation could impede technology uptake. About 33.5% of the respondents could read
and write. About 5% of the respondents had grade 7 to 9 education; which is the highest
among the total respondents.

The majority of the households (82.9%) are farmers and practised crop and livestock
mixed farming system. This could be due to the fact that most of the settlers came to the
district from highland and medium altitude areas of Gondar, Wollo, North Shoa and
Gojjam Zones of the Region, which practised predominantly mixed farming system. The
native Gumuz people still practice hunting and wild plant gathering with some farming
(IPMS 2005).

4.1.2 Age and family size

The overall mean age of respondents is 40.16 ± 0.84 (Table 5). This result was lower than
the mean average age of 44.3 years reported by (Teshome 2006) in Fogera district of the
Amhara region, whereas it was in agreement with the mean average age of 41.2 ± 0.65
years reported by (Tesfaye 2008) for Metema district.

Table 5. Family size, age of respondents and family members in Metema district

CLFS SLFS Overall

Variables Mean ± SE Mean ± SE Mean ± SE
(N = 60) (N = 80) (N = 140)
Age of respondents, years 41.05 ± 1.43 39.49 ± 1.01 40.16 ± 0.84
Age of family members
> 60 years 3.00 ± 1.00 2.00 ± 0.00 2.44 ± 0.44
16 to 60 years 2.74 ± 0.17 3.13 ± 0.19 2.97 ± 0.13
6 to 15 years 3.29 ± 0.26 2.89 ± 0.24 3.09 ± 0.18
< 6 years 2.46 ± 0.22 2.28 ± 0.14 2.35 ± 0.13
Family size
Male 5.47 ± 0.34 5.32 ± 0.32 5.41 ± 0.24
Female 6.02 ± 1.32 4.98 ± 0.25 5.44 ± 0.65
Total 5.37 ± 0.28a 5.26 ± 0.27a 5.31 ± 0.20a
SLFS = sesame–livestock farming system, CLFS = cotton–livestock farming system.
Means with same superscript within the same row do not differ significantly (P<0.05).
SE = standard error.

The mean family size is 5.37 ± 0.28 and 5.26 ± 0.27 persons per household in the
cotton–livestock and sesame–livestock farming systems, respectively; the overall being

23

WorkingPaper_25.indd 23 1/10/2011 10:54:24 AM

 5.31 ± 0.20 persons per household. This mean family size was similar to the national
average of 5.2 (CACC 2002), but lower than the values of 6.6 ± 0.31 reported by Sisay,
(2006) for the same district and 7.3 for Hamer and Bena-Tsemay by Admasu (2006).
The age structure showed that 50% are within the productive workforce age of 16 to 60
years. Only 2.5% are above 60 years of age. The remaining 30.1% are between 6–15
years of age, while 17.4% are children of less than 6 years of age. In rural Ethiopia, all
people above 10 years of age are involved in agricultural and related activities (CSA
2003).

4.2 Land holding and land use pattern

Data on land holding and land use pattern of the respondents are presented in Table
6. The overall average land holding per household is 6.78 ± 1.33 ha; being slightly
higher (P>0.05) in the cotton–livestock (7.53 ± 1.53 ha) than in the sesame–livestock
(6.21 ± 1.17 ha) farming system. The possible reason may be associated with the time
of settlement in both areas. According to the group discussion with key informants,
the settled inhabitants in the cotton–livestock farming system arrived earlier (by
their own and government), benefited from less controlled open access land and are
therefore well established. The majority of the inhabitants in the sesame–livestock
system are relatively new settlers and only 1–2 ha of farm land is allocated per settler
(OoARD, Settlement Desk, personal communication). According to Sisay (2006), this
is greater than the average landholdings of the highland districts of Debark (1.66 ha)
and Layarmachiho (2.03 ha). As reported by IPMS (2005), previously settled and the
indigenous people officially own about 5 ha of land, but many farmers cultivate more
than this.

According to the respondents, about 91.4% of the land is used for annual crop
production; while 0.8% is covered by perennial crops, 3.1% is left for private grazing and
4.7% fallowed (Table 6). The trend was similar to the result reported by Sisay 2006. This
fallow land was associated with the widely practised shifting cultivation in the area as a
means of soil fertility improvement measure. Based on the group discussion held with
elders, some 10 years back, fallowing of farmland was practised for about 5–8 years and
was a common practice in the district. The trend has been decreasing, and currently, land
is left for fallow for about 2 to 4 years due to the increase in human population and the
use of extensive areas for private investment.

24

WorkingPaper_25.indd 24 1/10/2011 10:54:24 AM

 Table 6. Mean ± SE land holding and land use pattern in Metema district

CLFS SLFS Overall

Variables
HHC % HHC % HHC %
Land holding
1–5 ha 21 42.9 38 52.8 64 50.4
5.1–10 ha 15 30.6 23 31.9 38 29.9
10.1–15 ha 10 20.4 11 15.3 22 17.3
>15 ha 3 6.1 0 0.0 3 2.4
Mean (SE) 55 7.53 ± 1.53a 72 6.21 ± 1.17b 127 6.78 ± 1.33b
Land use
Annual crops 51 89.5 66 93 117 91.4
Perennial crops – – 1 1.4 1 0.8
Private grazing – – 4 5.6 4 3.1
Fallow 6 10.5 – – 6 4.7
HHC = household count, CLFS = cotton–livestock farming system, SLFS = sesame–livestock farming system.
Means with different superscripts differ significantly (P<0.05).

4.3 Livestock holdings

Data on livestock holdings of the respondents by farming system are presented in Table
7. The average cattle holding per household that was 6.34 ± 0.59 TLU per household in
the sesame based farming system was significantly (P<0.05) different than mean cattle
holding in the cotton based farming system cattle holding that was 6.24 ± 0.45 TLU per
HH. Goat, sheep, donkey, and camel were 0.43 ± 0.6, 0.11 ± 0.03, 0.31 ± 0.04 and
0.06 ± 0.04, respectively, are greater than the cotton based farming system of 0.36 ±
0.04, 0.01 ± 0.1, 0.34 ± 0.03, and 0.00 for goat, sheep, donkey and camel respectively.
The difference between the two may be because of the availability of more free grazing
area in sesame based farming favoured to have large number of livestock than the cotton
based farming. The over all mean livestock holding per household in the district was
12.52 ± 6.23, 0.80 ± 0.40, 0.13 ± 0.07, 0.65 ± 0.32 and 0.07 ± 0.04, cattle, goat, sheep,
donkey and camel respectively (Table 5). The total livestock holding per household was
27.58 ± (13.72) TLU (Table 5). This finding was comparable in number with the reports
of IPMS (2005) and Sisay (2006) 9.41 ± 0.03, 9.4 ± 0.33, 0.3 ± 0.33, and 0.9 ± 0.08 for
cattle, goat, sheep and donkey for the same area. With regard to livestock composition of
the area, cattle were dominant followed by goat, donkey, sheep and camel in that order.
The result is also similar with the findings reported by the above mentioned authors.

25

WorkingPaper_25.indd 25 1/10/2011 10:54:24 AM

 Table 7. Livestock holdings in TLU of the sampled households in Metema

CLFS SLFS
Species
Mean ± SE Mean ± SE
Cattle 6.24 ± 0.45 6.34 ± 0.59
Goats 0.36 ± 0.04 0.43 ± 0.60
Sheep 0.01 ± 0.1a 0.11 ± 0.03b
Donkey 0.34 ± 0.03 0.31 ± 0.04
Camel 0.00a 0.06 ± 04b
Poultry (No.) 15.7a 17.2b

CLFS = cotton–livestock farming system; SLFS = sesame–livestock farming system; TLU = Tropical Livestock
Unit.
Means followed by different superscripts differ significantly (P<0.05).

4.4 Livestock herd/flock structure

The livestock herd structure by farming system is presented in Table 8. Mean cattle
holding per household is higher in the sesame–livestock (16.34 ± (0.59) TLU/hh) than
in cotton–livestock (6.25 ± (0.45 TLU/hh) farming system. This is due to the better
access to extensive gazing area in sesame livestock farming system. In both farming
systems, cows dominate followed by calves, oxen, heifers, and steers, respectively.
Female animals are highly valued than males and are a manifestation of wealth and
prestige. Oxen are purchased for farming activities and sold immediately after the
ploughing season. Young bulls (locally called ‘shelba’) are highly demanded in the
Sudan market.

With regard to goat flocks, the structure that was also in a similar trend, was revealed
as that of she-goats population followed by kids and he-goats respectively (Table 8).
The number of he-goats is smaller than she-goats because he-goats are either sold
at relatively younger age for cash or used for household consumption, while she-
goats are retained as breeding stock. Ownership of camels and sheep were small in
number as compared to other species of livestock. Male camels are purchased from
Afar Region and Sudan and are mainly used for sesame oil extraction (locally called
‘ansara’). Male donkeys are the third largest populated stock in Metama because they
are in high demand for the heavy burden of donkey cart (‘caroo’) and transportation
activities.

26

WorkingPaper_25.indd 26 1/10/2011 10:54:24 AM

 Table 8. Livestock herd structures in surveyed households by farming system in Metema
Mean ± SE Mean ± SE
Variables
CLFS SLFS
Cattle
Oxen 1.19 ± 0.09 1.06 ± 0.15
Cow 2.04 ± 0.19 2.08 ± 0.26
Heifer 1.02 ± 0.13 1.17 ± 0.15
Steer 0.71 ± 0.13 0.73 ± 0.11
Calves 1.30 ± 0.13 1.30 ± 0.18
Total 6.24 ± 0.45 6.34 ± 0.59
Goats
He-goat 0.03 ± 0.01 0.01 ± 0.01
She-goat 0.23 ± 0.03 0.22 ± 0.03
Kids 0.11 ± 0.02 0.20 ± 0.04
Total 0.36 ± 0.04 0.43 ± 0.60
Sheep
Male 0.01 ± 0.01 0.06 ± 0.02
Female 0.0 0.01 ± 0.01
Lambs 0.01 ± 0.0 0.04 ± 0.01
Total 0.01 ± 0.1 0.11 ± 0.03
Donkeys 0.34 ± 0.03 0.31 ± 0.04
Camels 0.0 0.06 ± 0.04
Total 13.60 ± 0.9a 14.11 ± 1.24b
Poultry (No.) 15.7a 17.2b
CLFS = cotton–livestock farming system; SLFS = sesame—livestock system; SE = standard error.
Means followed by different superscripts differ significantly (P<0.05).

4.5 Major feed resources

The major feed resources for different species of livestock are presented in Table 9 by
farming system and season of the year. In the wet season natural pasture is the sole
sources of livestock feed, while in the dry season, natural pasture (55.7%), crop residues
(20.7%), stubble grazing (14.3%) and grass hay (9.3%) are the major feed resources
for cattle, 73.6%, 12.1%, and 14.3% respectively for sheep, 72.9% and 27.1% natural
pasture and crop residues for goats, and crop residues (65.7%), natural pasture (17.9%)
and crop residues (16.4%) respectively for donkeys. Among the major feed resources
described above, natural pasture and crop residues rank as the first and second largest
source of livestock feed. Natural pastures support animal productivity in the rainy season,
while in the dry season these pastures can hardly maintain the animals as most of the
feed resources are less available and of poor nutritional quality. This could be due to the
poor practices of feed conservation and flash burning of the feed resources during the
dry season. Other studies have reported similar results (Simbaya 1998; Alemayehu 2006;
Sisay 2006; Elias et al. 2007).

27

WorkingPaper_25.indd 27 1/10/2011 10:54:24 AM

 Table 9. Major livestock feed resources and feeding systems in Metema district

CLFS SLFS Overall

Feeding system and
Wet Dry Wet Dry Wet Dry
major feed sources
HHC % HHC % HHC % HHC % HHC % HHC %
Major feeds for
Cattle: 60
– Natural pasture 100.0 30 50.0 80 100.0 48 60.0 140 100.0 78 55.7
– Crop residues 14 23.3 15 18.8 29 20.7
– Aftermath 9 15.0 11 13.8 20 14.3
– Hay 7 11.7 6 7.5 13 9.3
Goat 60
– Trees and shrubs 100.0 45 75.0 80 100.0 57 71.3 140 100.0 102 72.9
– Crop aftermath 15 25.0 23 28.8 38 27.1
Sheep 60
– Natural pasture 100.0 41 68.4 80 100.0 62 77.5 140 100.0 103 73.6
– Crop residues 8 13.3 9 11.3 17 12.1
– Crop aftermath 11 18.3 9 11.3 20 14.3
Donkey 60
– Natural pasture 100.0 13 21.7 80 100.0 12 15.0 140 100.0 25 17.9
– Crop residues 35 58.3 57 71.3 92 65.7
– Crop aftermath – 12 20.0 11 13.8 23 16.4
Feeding system
Cattle 60
– Tethering 6 10.0 2 3.3 4 5.0 4 5.0 10 7.1 6 4.3
– Free grazing 54 90.0 58 96.7 68 85.0 73 91.3 122 87.1 131 93.6
– Cut-and-carry – – – – 8 10.0 3 4.8 8 5.7 3 2.1
Goat 60
– Tethering 13 21.7 – – 11 13.8 – – 24 17.1
– Free browsing 47 78.3 60 100.0 69 86.3 80 100.0 116 82.9 140 100
Sheep 60
– Free grazing 60 100.0 60 100.0 80 100.0 80 100.0 140 100.0 140 100
Donkey 60
– Tethering 14 23.3 39 65.0 15 18.8 49 61.3 29 20.7 88 62.9
– Free grazing 46 81..7 21 35.0 65 81.3 26 32.5 111 79.3 47 33.6
– Cut-and-carry 5 6.3 5 3.5
HHC = household count, CLFS = cotton–livestock farming system, SLFS = sesame–livestock farming system.

4.6 Feed management and utilization practices

Three types of feeding systems were practised in the district (Table 9). In the cotton–
livestock farming system, 90 and 96.7% of the respondents practised free grazing for
their cattle during the wet and the dry season, respectively; while the respective values
were 85 and 91.3% in the sesame–livestock farming system. Overall, 87.1, 5.7 and 7.1%
of the respondents use free grazing, cut-and-carry and tethering during the wet season,

28

WorkingPaper_25.indd 28 1/10/2011 10:54:25 AM

 respectively, while 93.6, 4.3 2.1% respectively practice these feeding management
systems during the dry season. Variations in labour shortage, number of animals
kept by herders and area covered by crops contribute to seasonal differences in herd
management.

Overall, about 82.9% of the respondents allowed their goats to browse while 17.1%
tethered them. Sheep production is based on free grazing. There were some variations
in feeding management between the two farming system. Donkeys are important
animals in the district, and particularly male donkeys are bought from the neighbouring
highland districts for the purpose of transportation services. Most of the respondents
(62.9%) tethered their donkeys while some (33.6%) practised cut-and-carry system. This
is because donkeys work hard during the day and are left to graze at night. Unlike the
highlands, hyenas are not found in the district, making night time grazing possible.

During the wet season, natural pasture is the sole feed resource for all species of livestock
(Table 9). In the cotton–livestock farming system, the major feed resources for cattle
during the dry season are natural pasture (50%), crop residues (23.3%), crop aftermath
(15%) and grass hay (11.7%). Whereas, in the sesame–livestock farming system, the
use of natural pasture is higher (60%; P<0.05), while the contributions of crop residues
(18.8%), crop aftermath (13.8%) and grass hay (7.5%) are lower (P<0.05) than the
cotton–livestock farming system. This is attributed to the fact that most of the communal
grazing areas are found in the sesame–livestock farming system. Feed conservation
practices and crop residue utilization are better in the cotton–livestock than in the
sesame–livestock farming system.

4.7 Availability of natural pasture

Data on availability of natural pasture during the wet and dry seasons are presented in
Table 10. About 62 and 65% of the respondents in the cotton–livestock and sesame–
livestock farming system, respectively, indicated that availability of natural pasture for
cattle is inadequate during the dry season. Contrary to this, availability of feed for goats
was rated as adequate by 75 and 70% of the respondents in the cotton–livestock and
sesame–livestock farming systems, respectively.

During the wet season, availability of livestock feed in both farming systems is rated as
adequate and abundant by 50 and 42.9% of the respondents, respectively. According
to IPMS (2005), out of the total land area of the district, estimated at 440,085 ha, about
71% (312,300 ha) is natural pasture and forest land. In addition, fallow lands and private
grazing lands (not clearly known) are major sources of natural pastures in the district, as
confirmed by earlier reports (Alemayehu 2006; Sisay 2006; Elias et al. 2007).

29

WorkingPaper_25.indd 29 1/10/2011 10:54:25 AM

 Table 10. Percentage response of respondents on availability of feed resources in Metema

CLFS SLFS Overall

Variables
HHC % HHC % HHC %
Availability of feeds Dry season
Cattle:
Adequate 23 38.3 28 35.0 51 36.4
Inadequate 37 61.7 52 65.0 89 63.6
Goat
Adequate 45 75.0 56 70.0 115 82.1
Inadequate 15 25.0 24 30.0 39 27.9
Sheep
Adequate 33 55.0 45 56.3 78 55.7
Inadequate 27 45.0 35 44.8 62 44.3
Donkey
Adequate 51 85.0 62 77.5 113 80.7
Inadequate 9 15.0 18 22.1 27 19.3
Availability of feeds Wet season
Adequate 28 48.3 42 52.5 70 50.0
Inadequate 10 16.7 10 7.1
Abundant 22 36.7 38 47.5 60 42.9
Quality of the feeds
Very good 48 80.0 70 87.5 118 84.3
Average 12 20.0 10 12.5 22 15.7
N = 60 for CLFS; N = 80 for SLFS, CLFS = cotton–livestock farming system, SLFS = sesame–livestock farming
system, HHC = household count.

4.7.1 Hay making

According to the information gathered from group discussion with elders, hay making
started only recently. About 10 years ago, there was no feed shortage during the dry
season because bamboo trees were abundant and they shed leaves during the dry
season and these leafy foliages were the principal sources of livestock feed. After the
disappearance of the bamboo forest about 10 years ago, communities were forced to
collect and conserve grass hay due to increased human and livestock population and
drought. Hay making is practised from end of October up to November and curing takes
place for a period of two week to one month. The quality of hay is often poor and is
bleached by the strong sunshine in the area. Farmers have not received any training on
good practices of hay making by the extension service. This poor hay making practice
was also reported by Sisay (2006) who estimated hay production per household was only
0.77 t DM/ha. Tesfaye (2008) also reported the percentage of households that practice
hay making is lower in the sesame–livestock (70.9%) than in the cotton–livestock (86.0%)
farming system.

30

WorkingPaper_25.indd 30 1/10/2011 10:54:25 AM

 4.7.2 Crop residues production and utilization
In the cotton–livestock farming system, the main crops grown are cotton, rice, sorghum,
maize, sesame, finger millet and fruits, while in the sesame livestock farming system,
sesame, cotton, sorghum, maize, soya bean, teff, chickpea, groundnut and fruits are
grown. Sorghum is the most important cereal crop grown in the district and constitutes
about 90% of the total crop residues produced in the area. Crop residues from maize
(8.4%), teff (1.3%), finger millet (0.14%), rice (0.25%), soya bean and chickpea make
up for the difference (Table 11). Next to natural pastures, crop residues are other main
sources of livestock feed during the dry season. In the cotton–livestock farming system,
23.3, 13.3 and 58.3% of the crop residues are used for cattle, sheep and donkeys,
respectively, while in the sesame–livestock farming system the respective values are
18.8, 11.3 and 71.3%. Overall, out of the total amount of crop residues produced in the
district, 20.0, 12.1 and 65.7% are used for cattle sheep and donkeys, respectively. The
value of crop residues produced in a particular area depends on the amount and type of
crops grown in the area.

Table 11. Estimates of grain and crop residues production for major crops grown in Metema

Cultivated Grain produced Crop residue G:CR %

Types of crops Sources
(ha) (Qt) (CR) (t) ratio Share
Teff 1005 8033 12.1 1.5 MOA (1984) 1.3
Finger millet 365 2555 1.3 2.04 Awassa, Bako 0.14
Sorghum 15,820 329,056 822.6 2.5 Jhanke (1982) 89.5
Maize 1809 40,143 80.3 2.0
Chickpea 86 344 0.41 1.2
Rice 117 2,340 2.34 1 Devendra (1997) 0.25
Soya bean 15.5 93 0.09 1 Devendra (1997)
Total 919.1
Sources: OoARD (2006).
Means with different superscripts differ significantly (P<0.05).

4.7.3 Stubble grazing

The proportion of respondents using stubble grazing in the two farming systems is
presented in Table 12. Stubble grazing is an important feed resource and is practised
soon after the crops have been harvested from October to early December. Livestock
are allowed to graze stubbles of sorghum, maize, teff, peanut and millet fields. Stubble
grazing contributes to about 14.3% of the basal diet of cattle; being slightly higher in
the cotton–livestock (15%) than in the sesame–livestock farming system (13.8%). The
proportion of stubble as feed for goats (25%) and sheep (18.3%) is slightly lower in
the cotton–livestock than in the sesame–livestock (28.8%) farming system, while the

31

WorkingPaper_25.indd 31 1/10/2011 10:54:25 AM

 reverse was true for sheep (18.3% vs. 11.3%), respectively. The differences in stubble
grazing usage between the two farming systems could be associated with the types and
size of cereal crop production. In the cotton–livestock system cereals like sorghum are
extensively produced, while in the sesame–livestock system sesame is the dominant crop
and herbicides are extensively used to control important weeds and forbs minimizing
stubble grazing. This finding agrees with the reports of Sisay (2006).

Table 12. Percentage of respondents using stubble grazing by farming system in Metema

CLFS SLFS Overall

Species
HHC % HHC % HHC %
Cattle 9 15.0 11 13.8 20 14.3
Goat 15 25.0 23 28.8 38 27.1
Sheep 11 18.3 9 11.3 20 14.3
Donkey 12 20.0 11 13.8 23 16.4
CLFS = cotton–livestock farming system; SLFS = sesame–livestock farming system, HHC = household count.

4.7.4 Supplementary feeding

The types and the amount of supplementary feeding for livestock are summarized in
Table 13. Crop residues, sesame oil by-products (‘embaze’), noug cake, hay, wheat bran
and local brewery by-product (‘atela’) are the major supplementary feed used in the dry
season. More number of respondents (28.3%) in the cotton–livestock farming system use
crop residues as supplementary feed than those in the sesame–livestock farming system
(20.0%). This could be due to the more availability of residues and experiences of the
farmers in collecting and utilization of residues in the cotton–livestock farming system. As
expected, the use of embaze as a supplementary feed is higher in the sesame–livestock
(26.3%) than the cotton–livestock farming system (6.7%), due to availability and easy
access in the former system.

Amount of supplementation

Data on the amounts of supplementary feed offered to livestock are presented in Table 13.
Most of the respondents (53%) provide 1 kg of supplementary feed per cattle per day, and
this is slightly higher in the cotton–livestock (55%) than in the sesame–livestock (51.25%)
farming system. About 22 and 19% of the respondents also indicated that they provide
1.5 and 0.5 kg per cattle per day, respectively. About 66 and 31% of the respondents also
indicated that they provide 1 kg and 1.5 kg of supplementary feed per donkey per day,
respectively. This indicates that donkeys are important assets to the household and the
community at large in Metema.

32

WorkingPaper_25.indd 32 1/10/2011 10:54:25 AM

 Table 13. Percentages of respondents showing the types and amount of supplementation of feeds in
the two farming systems in Metema

CLFS (N = 60) SLFS (N = 80) Overall (N = 140)

Variables
HH % HH % HH %
Types of supplementation
Cattle
Crop residues only 17 28.3 16 20.0 31 12.9
Crop residues + embaze 10 16.7 12 15.0 22 15.7
Embaze (sesame cake) 4 6.7 21 26.3 25 17.7
Noug cake 9 15.0 7 8.8 16 11.4
Grass hay 8 13.3 10 12.5 18 12.9
Wheat bran 5 8.3 9 11.3 14 10.0
Atela 3 5.0 – 5 3.6
Atela + hay 4 6.7 5 6.3 9 6.4
Donkey
Crop residues 20 33.3 29 36.3 49 35.0
Hay only 4 6.7 7 8.8 11 7.9
Grains 36 60.0 44 55.0 80 57.1
Amount of supplementation, kg/animal per day
Cattle
2 4 6.7 9 11.3 13 5.4
1.5 15 25.0 6 7.5 31 22.1
1 33 55.0 41 51.3 74 52.9
0.5 6 10.0 20 25.0 26 18.6
0.75 2 3.3 4 5.0 6 4.2
Donkey
2 1 1.7 4 5.0 5 3.6
1.5 19 31.7 24 30.0 43 30.7
1 40 66.7 52 65.0 92 65.7
Crop residues only = 1 kg/animal per day; Crop residues + Embaze = 1.5 kg/animal per day; embaze (sesame
cake) = 0.5 kg/animal per day; noug cake = 0.75 kg/animal per day; Grass hay = 2 kg/animal per day.

The amount of wheat bran, atela and atela plus hay offered are not estimated due to small
quantity offered to animals. During the field work it was also noted that the amount of
supplementary feeds offered, especially sesame cake, noug cake and wheat bran was
very small, and the major purpose of supplementation is not to increase productivity, but
is focused to those animals that are susceptible to diseases, weak animals, suckling cows
and calves.

4.7.5 Feeding calendar and seasonal availability of feed resources

Seasonal availability of feed resources in the study area, as obtained from group
discussion and observation during the field work is presented in Table 14. The pattern
of feed resources availability in the district is influenced by season as in the highland

33

WorkingPaper_25.indd 33 1/10/2011 10:54:26 AM

 areas (Ahmed 2006) and the trend followed similar pattern with the findings of Sisay
(2006).

Table 14. Availability of different feed resources by month in the district

Month
Type of feed
Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May
Natural pasture * * * * * * ----- ---- ----- ---- ----- -----
Crop residues * * * *
Stubble grazing ---- * * *
Hay * * * *
Supplementation --- * *
Fodder trees * * * *
----- = Availability of few green browse species and dry grasses; * = Abundant

4.7.6 Chemical composition of the major feed resources

The effect of species on chemical composition

Chemical composition of different species of feed resources is given in Table 15.

Comparison of the chemical composition of the three major grass species (Pennisetum
spheslatum, Cenchrus ciliaris and Pterocarpus lucens) indicated that there is a species
difference in nutrient contents.

Table 15. Chemical composition (Mean ± SE) of different feeds in the study districts

Species
Chemical composition,%
P. spheslatum C. ciliaris P. lucens
DM 90.2 ± 0.14a 89.9 ± 0.14ab 89.0 ± 0.14b
Ash 20.1 ± 2.54a 15.3 ± 2.54ab 7.7 ± 2.54b
OM 79.9 ± 2.54ab 84.7 ± 2.54ab 92.3 ± 2.54a
CP 6.3 ± 0.43ab 6.4 ± 0.43ab 15.8 ± 0.43a
NDF 69.3 ± 0.64b 74.6 ± 0.64a 58.2 ± 0.64c
ADF 45.8 ± 0.55a 45.9 ± 0.55a 43.1 ± 0.55b
ADL 8.1 ± 0.74ab 5.9 ± 0.74ab 21.3 ± 0.74a
IVDMD 45.9 ± 1.18b 39.3 ± 1.18c 47.8 ± 1.18a
Means with different superscripts in a row differ significantly (P<0.05).

The mean DM and ash contents in P. spheslatum were significantly (P < 0.05) higher than
in P. lucens and C. ciliaris and also mean ADF and ADL contents in P. spheslatum were
significantly (P < 0.05) higher than in P. lucens. The DM, ash, NAF and ADF contents
of P. lucens were significantly (P<0.05) lower than the mean DM, ASH, NAF and ADF
contents of C. ciliaris. The mean OM and CP contents in C. ciliaris were higher than in
P. spheslatum. Contents of NDF and ADF in C. ciliaris were significantly (P<0.05) higher

34

WorkingPaper_25.indd 34 1/10/2011 10:54:26 AM

 than in P. spheslatum and C. ciliaris. Mean OM, CP and IVDMD contents in P. lucens
were significantly (P < 0.05) higher than the other two species. The differences in nutrient
content between species could be associated with their inherent nature, i.e. fodder trees
contain more protein than grass species. As reported by Mathur et al. (1991) most grasses
and tree leaves in arid environments are low in nutritive value because of high contents
of lignin and relatively indigestible (compared with starch) cellulose and hemicelluloses.

The effects of season on chemical composition

Data on the effect of season on chemical composition of dominantly growing two

grass and one fodder species are given in Table 16. The DM, OM, NDF, ADF and ADL
contents of feeds were significantly (P < 0.05) higher in October than in August. Whereas,
the percentages of ash, CP, IVDMD of the feeds were significantly (P < 0.05) lower
in October than in August. This may be associated with the temperature, the stages of
growth and the plant species that influenced the quality of feeds. This finding agrees with
earlier reports (Alemayehu 2006; Sisay 2006), that the structural constituents of plant
materials (lignin, cellulose and hemicelluloses) increase with maturity. During the growth
stage, plants are most nutritious. Once mature, plants are subject to leaching and dilution
of nutrients and reduction in nutritive value. Declines in nutrient composition and
leaching are especially serious in the case of herbaceous plants. As the plant matures,
the contents of crude protein, the more readily digestible carbohydrates and phosphorus
decrease and that of fibre, lignin and cellulose increase. The decline in nutritive value
of these feeds in October indicates that feed conservation practised by the communities
between mid October and November is not appropriate time and should be done earlier
than in mid October.

Table 16. Effect of season on chemical composition (Mean ± SE) of composite samples of domi-
nant feed species in Metema

Season
Chemical composition, %
Wet (August) Dry (October)
DM 88.9 ± 0.11b 90.5 ± 0.11a
ASH 18.2 ± 2.07a 10.5 ± 2.07b
OM 81.8 ± 2.07b 89.5 ± 2.07a
CP 10.8 ± 0.35a 8.2 ± 0.35b
NDF 60.8 ± 0.52b 73.9 ± 0.52a
ADF 40.3 ± 0. 45a 49.6 ± 0.45a
ADL 10.1 ± 0.60b 13.5 ± 0.60a
IVDMD 50.7 ± 0.96a 37.9 ± 0.96b
Means with different superscripts in a row differ significantly (P<0.05).

35

WorkingPaper_25.indd 35 1/10/2011 10:54:26 AM

 4.7.7 Feed balance
The estimate of the total available feeds in terms of dry matter (calculated from grazing
land, cultivated land (crop residues and stubble), and forest land) is presented in Table 17.
The total estimated annual DM from crop residues, grazing land and stubble is 827.16,
780,750.0 and 51,954.0 t; totalling 833,531.16 t. The total livestock population in TLU
(tropical livestock unit) is estimated at 103,190. Daily DM requirement for maintenance
of one TLU is estimated as 2.5% of the body weight (ILCA 1990) that is, 250 × 2.5% =
6.25 kg DM per day and (6.25 × 365 = 2280 kg) per annum. Therefore, the total DM
requirement for maintenance per annum is estimated at 235,273.2 t. Based on this
calculation, the estimated feed balance in the district is 598,258 t DM. This finding is
in agreement with the estimates of Sisay (2006) who reported that the nutrient balance
in Metema is sufficient to support the livestock holdings per household. However, the
quality of the feed is poor.

Table 17. Estimates of feed balance in Metema district

A
Livestock species B A×B C C – (A × B) D
(TLU)
Cattle 95,837 2.28 218,508.4
Goat 3202 2.28 7300.6
Sheep 169 2.28 385.3
Donkey 3582 2.28 8167.0
Camel 400 2.28 912.0
Total 103,190 235,273.3 833,531.2 598,257.9 0.00
A = Livestock number in the district (TLU), B = Feed requirement of per animal (t Dm/year), A × B = Total feed
requirement of all animals (t DM/year), C = Estimate of available feed resources (t DM/year), C–(A × B) = Esti-
mated feed balance of the district, D = Deficit (if any).

4.7.8 Water resources

Availability of water in grazing areas is a good opportunity for livestock and human
beings living in a lowland agro-ecology like Metema district. Based on respondents and
observations made during the study, the major sources of water in the district (Table 18)
are rivers (64.3%), springs (21.4%) and ponds (10.7%). Rivers are equally important
sources of water in both farming systems. There are three large continuous rivers passing
through the district and used for livestock and irrigation activities. Ponds are more
important sources of water in the cotton–livestock (18.3%) than in the sesame–livestock
(5%) farming system, while springs are more important in the sesame–livestock (28%)
than in the cotton–livestock (13.3%) farming system. IPMS (2005) also reported that
farmers and agriculturists believe that the underground water table is high and in some
places sufficient amount of water could be obtained at less than 10 m depth. The majority

36

WorkingPaper_25.indd 36 1/10/2011 10:54:26 AM

 of the respondents (69.3%) water their animals once a day, while about 31% water twice
a day (Table 18). This result is in agreement with Teshome (2006).

Table 18. Sources of water and watering frequency by farming system for cattle in Metema (per-
centage of respondents)

CLFS SLFS Overall

Parameter
N % N % N %
Sources of water
River 39 65.1 51 63.7 90 64.3
Pond 11 18.3 4 5.0 15 10.7
Spring 8 13.3 22 27.5 30 21.4
Wells 2 3.3 3 3.8 5 3.6
Watering frequency
Once a day 37 61.7 60 75.0 97 69.3
Twice a day 23 38.3 20 25.0 43 30.7
CLFS = cotton–livestock farming system; SLFS = sesame–livestock farming system.

4.7.9 Major constraints to livestock production

The major problems associated with livestock production as perceived by the respondents
are presented in Table 19. About a quarter of the respondents indicated livestock theft as
their major problem in both farming systems. During group discussions, elders reported
that at times up to 90 heads of cattle are stolen. The trend appears to be escalating from
time to time. The location of the district encourages theft as there is an extensive area
bordering the Sudan making it easier for cattle rustlers to trek the animals across the
boarder.

Table 19. Percentage of respondents showing major problems to livestock production in Metema

CLFS (N = 60) SLFS (N = 80) Overall

Variables
HH % HH % HH %
Feed shortage 4 15.0 4 5.0 8 5.7
Biting insects 8 13.3 10 12.5 18 12.9
Livestock diseases 10 16.7 15 18.8 25 17.9
Labour shortage 4 6.7 5 6.3 9 6.4
Drought 5 8.3 7 8.8 12 8.6
Livestock theft 14 23.3 19 23.8 33 23.6
Conflict with highlanders 5 8.3 8 10.0 13 9.3
Crop land encroachment 7 11.7 8 10.0 15 10.7
Shortage of crop residues 3 5.0 4 5.0 7 5.0
HH = Households, CLFS = cotton–livestock farming system, SLFS = sesame–livestock farming system.

Livestock diseases (16.7%) and biting insects (13.3%) were the secondly and thirdly
prioritized constraints in both CLFS and in the SLFS, respectively. About 11% of the
respondents indicated that crop land encroachment is affecting livestock production. The

37

WorkingPaper_25.indd 37 1/10/2011 10:54:27 AM

 possible causes of the crop land encroachment might be due to the human population
pressure through continuous settlement program by government, increasing number of
voluntary settlers and settler’s relatives from the neighbouring highland districts and the
natural increase in population.

Conflict between transhumants and the local inhabitants for livestock feed resources
was not a major problem and was reported by about 9% of the respondents. Feed
shortage, drought and shortage of crop residues were minor concerns and were reported
by only 5.7, 8.6 and 5% of the respondents, respectively. During the group discussion,
it was noted that there are two major varieties of sorghum grown by farmers. These
are ‘wodiaker’ and ‘zole’, and the former is highly preferred by livestock. The residue
from the later variety is left as mulch or burned in the field. Farmers also reported that
the productivity of wodiaker variety is declining at an alarming rate due to loss of soil
fertility, while the zole variety grows well on less fertile soil and is more productive than
the wodiaker variety. This could contribute to feed shortage during the long dry season.
An alternative dual purpose sorghum variety suitable for the agro-ecology should be
developed in the future.

4.8 Floristic composition

4.8.1 Herbaceous species composition
A total of 33 herbaceous species were recorded, and out of these, 14 (42.4%) were
grasses, 6 (18.2%) were legumes, and 13 (39.4%) were sedges and others (Table 20;
Appendix Table 6). Of the grass species identified, 23.1% were categorized as highly
desirable, 38.5% desirable and 30.8% less desirable. This might be due to the gradual
disappearance of highly desirable species through over use and disturbance by livestock
and human beings. From the discussion held with the farmers, it was understood that the
major factors that cause the decline in the abundance of highly desirable species were
drought followed by overgrazing due to the increase in grazing pressure, invader plants
have replaced decreaser and increaser plants. Furthermore, overgrazing reduces ground
cover, plant height, forage quality and productivity, changes are induced in the dominant
growth forms of herbaceous plants; tall perennial bunch grass species give way to shorter
rhizomatous and sotoloniferous perennial grasses which are replaced by annual grass and
forbs species (Herlocker 1999; Abule et al. 2007a). Besides, overgrazing tends to reduce
perennial grassland vegetation types and allows invasion by annual forbs and grasses.
Drought and overgrazing could be the causal factors for the decline in plant species
composition and diversity of plants over time (Admasu 2006; Alemayehu 2006; Amaha
2006; Abule et al. 2007a).

38

WorkingPaper_25.indd 38 1/10/2011 10:54:27 AM

 Table 20. Common and dominant grass species identified by farming systems and different grazing
types in Metema

CLFS SLFS
Grasses Category
CG RS EN CG RS EN
C. ciliars DS D C C C C
P. sphacelatum DS C C D C C
Setaria pumila DS C C C C C
Brachiaria lata DS C C
Urochloa fatamensis LD C C
Rhamphicarpa fistulosa LD C C C
Temeda triandria HD D
Cynodon dyctlon HD C
Cyprus spp. DS C C
Eleusine floccifolia LD D D C
Hyparrhenia rufa HD D D
Panicum coloratum HD D
Sporobolus pyramidialis DS C
HD = highly desirable; DS = desirable; LD = less desirable; CG = communal grazing; RS = road side; EN =
enclosure; P = present (<5% of DM); C = common (>5% and <20% of DM), D = dominant (>20% of DM).

4.8.2 Woody species composition

A total of 20 woody species were identified in the study district (Table 21; Appendix Table
5), and 15, 35, and 50% are highly desirable, desirable and less desirable, respectively.
The highest proportion of the woody vegetation is composed of different species of acacia
(20%) and commbretum (10%). Species such as Anogeissus leiocarpus, P. lucens, and
Ziziphus spina-christi are highly desirable and they are dominant in enclosure areas.
Besides different species of acacia, Balanites aegyptiaca, Boswelia papyrifera, Combretum
collinum, C. mole, Dichrostachys cinerea, Ficus sycomorus, Fluegea virosa, Gardenia
ternifolia, Grewia villosa, Stereospermum kunthianum, Terminalia laxiflora and Ximenia
Americana are commonly found woody species (Table 21). In sesame based framing
system, Acacia polyacantha, A. seyal, Balanites aegyptiaca and B. papyrifera were the
dominant species in communal grazing areas and A. tortilis, A. leiocarpus, A. senegal, C.
collinum, C. mole, D. cinerea, F. sycomorus, F. virosa, P. lucens, and Z. spina-christi, were
common species. In road side grazing areas, A. polyacantha and D. cinerea dominated
and A. seyal, B. aegyptiaca, B. papyrifera, T. laxiflora, C. collinum, C. mole and X.
Americana were common species found. From this result, the grazing land in the district
could be characterized as acacia dominated wood land and the woody vegetation were
the important sources of feed for ruminant animals in the area.

39

WorkingPaper_25.indd 39 1/10/2011 10:54:27 AM

 Table 21. Common and dominant woody species and their percentage composition by farming
system in different grazing areas in Metema

CLFS SLFS
Woody species Category
CG RS EN CG RS EN
G. ternifolia LD C
A. polyacantha DS D D C D D
T. laxiflora C C C
A. tortilis DS C D C C C
A. seyal DS D C D D C C
D. cinerea DS D C D
A. leiocarpus DS C C D C D
P. lucens HD C D C D
F. sycomorus DS C C
C. collinum LD C C C
X. Americana LD C
S. kunthianum DS C
B. aegyptiaca DS C C D C C
G. villosa LD C
Z. spina-christi HD C C D C D
Piliostigma toningii DS
F. virosa LD C C
B. papyrifera UD C C C D C D
A. senegal DS D C C C
C. mole LD C C
CLFS = cotton–livestock farming system; SLFS = sesame–livestock farming system; CG: communal grazing; RS:
road side; EN: enclosure; C = common (>10% and <20% density); D = dominant (>20% density).

According to the result obtained from group discussions, P. lucens is one of the most
important fodder trees in the district. After the long dry season during the on set of the
main rain season, the community practised to search P. lucens to heal their emaciated
animals. The communities take care of this tree species by their own against destruction
unlike other species. Thus, woody plants contribute significantly to the sources of
livestock feed in the district. The finding is agreed with the report suggested that woody
species are important source of food, fodder, fuel wood, medicine, fibre and gums
(Herlocker 1999; Alemayehu 2006; Abule et al. 2007b).

Height classes of woody vegetation

The height class distribution of trees and shrubs in the farming systems are presented in
(Table 22). There is no difference in height class category of < 0–1 m between grazing
types found in both farming systems. The highest percentage of distribution of trees and
shrubs is found in height class category of > 1–3 m in both farming systems and are also
comparable. Generally, most height classes of the vegetation could be considered at

40

WorkingPaper_25.indd 40 1/10/2011 10:54:27 AM

 browsing height of the animals. According to Tainton (1999) height of 1.5 m represents
the mean browsing height for goats.

Table 22. Percentage of height class (metres) distribution of trees and shrubs in Metema

CLFS SLFS
Height class
CG RS EN CG RS EN
< 0–1 m 22.6 25.0 21.1 20.6 23.1 20.8
> 1–3 m 45.2 33.3 42.1 41.2 38.5 33.3
>3–4.5 m 19.4 16.7 15.8 26.5 23.1 12.5
> 4.5 m 12.9 25.0 21.1 11.8 15.4 33.3
CLFS = cotton–livestock farming system; SLFS = sesame–livestock farming system; CG = communal grazing; RS
= road side grazing, EN = enclosure.

Hence, these reachable heights of different woody browse species make the area
favourable to raise browsing animal species such as goats and at the same time maintain
the balance between the woody and herbaceous species. Studies suggested that
integration of grazers and browsers having different feeding habitats makes more efficient
use of natural vegetations. According to Taylor (1985), when cattle were partly replaced
by goats and/or sheep, individual cattle performance increased because forage demand
for the grass component was reduced. Likewise, production of ewes increased when
some sheep were replaced by cattle and goats because grazing pressure on the forbs
component declined. Herd diversification through increasing the number of browsers
such as camels and goats would also contribute towards efficient resource utilization and
decreases woody plant encroachments in rangelands (Gemedo 2004).

4.8.3 Vegetation in the cotton–livestock farming system

Herbaceous species composition

From a total of 24 herbaceous species recorded in the cotton–livestock farming system,

50, 20.8 and 29.2% are different species of grasses, legumes and sedges (Table 23).
Among the grass species identified, 25% are highly desirable, 33.3% are desirable
and 41.7% are less desirable, respectively. The ratios of the herbaceous species
(grasses:legumes:sedges) is 12:5:7. Bracheria lata, P. spheslatum, R. fistulosa and Seteria
pumila are the common grass species. In the communal grazing areas, C. ciliaris is
the dominant grass species, while Cyprus spp, Eurochloa fatamensis fistulosa, R. and
S. pumila, are the common grass species. In the road side grazing areas, E. flocifolia is
dominant. The enclosure areas have a relatively higher percentage of highly desirable
grass species than the communal and road side grazing areas. In the enclosure areas, C.
ciliaris, Cyprus spp, E. fatamensis, P. spheslatum and S. pumila, are the common species,
while Themeda teriandra and H. rufa are dominant. This could be attributed to a better

41

WorkingPaper_25.indd 41 1/10/2011 10:54:27 AM

 management practice and a lower livestock impact in the enclosure areas. This result is in
agreement with earlier reports (Admasu 2006; Amaha 2006; Teshome 2006).

Table 23. Herbaceous species composition (% DM biomass) and their desirability by grazing type
in cotton–livestock farming system in Metema

Grazing types Herbaceous species Category % composition Remark

Communal C. ciliaris DS 21.2 Grasses
P. sphacelatum DS 5.8
S. pumila DS 10.4
B. lata DS 6.2
U. fatamensis LD 1.4
R. fistulosa LD 16.66
Urochloa cf. brchyura LD 1.2
T. triandria HD 2.6
C. dyctlon HD 1.4
Ahiya abish (local name) UD Legumes
Alysicarpus quartinianus LD
Vigna membranacea LD
Chamaecrista mimosoides (L) UD
Euphhorbia indica UD Sedges
Spermacoce sphaerostima UD 1.3
Hygrophilla schulli UD 10.4
Kedrostis foetidissima LD 0.63
Commelina subula UD 0.5
Road side C. dyctlon HD 3.4 Grasses
Cypres spp. LD 6.7
C ciliaris DS 14.9
S. pumila DS 17.2
U. fatamensis LD 5.3
R. fistulosa LD 5.2
E. floccifolia LD 23.6
Ahiya abish (local name) UD 13.2 Legumes
A. quartinianus LD 8.0
V. membranacea DS
Road side Cyanotis barbata UD Sedges
H. schulli UD
S. sphaerostima UD
Bidens setigera UD
K. foetidissima LD
Zennia elegans Jaquin. UD

42

WorkingPaper_25.indd 42 1/10/2011 10:54:27 AM

 Grazing types Herbaceous species Category % composition Remark
Enclosure P. sphacelatum DS 6.1 Grasses
C. ciliaris DS 6.5
S. pumila DS 5.2
B. lata DS 1.9
U. fatamensis LD 5.2
T. triandra HD 22.6
H. rufa HD 24.4
C. dyctlon HD 5.8
Indigofera spicata DS Legumes
A. quartinianus LD
Dismodium dichotomum (Klein) DS
Hibiscus articulatus HD Sedges
C. barbata UD
C. subula UD
H. vitifolius L. DS
HD = highly desirable, DS = desirable, LD = less desirable.

Woody vegetation

A total of 17 woody species were recorded (Table 24), and 17.7, 41.2, 35.3 and 5.9%
are highly desirable, desirable, less desirable and undesirable species, respectively. A.
polyacantha, A. seyal, A. senegal and B. papirefera are the dominant species whereas, A.
tortolis, A. leiocarpus, B. aegyptiaca, F. sycomorus, G. ternifolia, P. lucens, T. laxiflora and
Z. spina-christi are commonly found in communal grazing areas. In road side grazing
areas, A. polyacantha, A. tortilis and D. cinerea are the dominant species, while A. seyal,
A. leiocarpus, A. senegal, B. papyrifera, Z. spina-christi are common. In the enclosure
grazing areas, A. leiocarpus, P. lucens, A. seyal and Z. spina-christi are dominant woody
plant species and A. polyacantha, B. aegyptiaca, B. papyrifera, A. senegal, and T. laxiflora
are common. The proportion of desirable species in this farming system is considerably
higher than the highly desirable and less desirable species. This could favour raising
browsing livestock species like goats and camels.

4.8.4 Vegetation in the sesame–livestock farming system

Herbaceous species

Herbaceous species obtained in sesame based farming system and their percentage
composition is presented in (Table 25).

43

WorkingPaper_25.indd 43 1/10/2011 10:54:27 AM

 Table 24. Woody species and percentage composition in different grazing system in cotton–live-
stock farming system in Metema

Grazing types Woody species Category % composition

Communal T. laxiflora LD 0.9
G. ternifolia LD 24.8
A. polyacantha DS 1.5
A. tortilis DS 12.1
A. seyal DS 8.5
D. cinerea LD 5.5
A. liocarpus HD 15.2
P. lucens HD 7.0
F. sycomorus DS 2.4
C. collinum LD 1.2
S. kunthianum DS 0.6
X. americana LD 0.9
Blanites aegyptyca DS 5.5
Z. spina-christi HD 1.8
P. toningii LD 5.2
B. paperiferra UD 0.6
A. senegal DS 3.9
Road side G. ternifolia LD 21.0
A. polyacantha DS 1.1
A. tortilis DS 14.8
A. seyal DS 2.3
D. cinerea LD 15.3
A. liocarpus HD 2.8
P. lucens HD 5.7
F. sycomorus DS 14.8
C. collinum LD 1.1
S. kunthianum DS 1.7
X. americana LD 9.1
B. aegyptyca DS 7.5
Z. spina-christi HD 1.1
P. toningii LD
Enclosure T. laxiflora LD 31.0
A. liocarpus HD 3.4
A. polyacantha DS 10.3
A. seyal DS 20.7
P. lucens HD 6.9
B. aegyptyca DS 13.8
Z. spina-christi HD 5.7
B. paperiferra UD 3.4
A. senegal DS 2.8
D = desirable, HD = highly desirable; UD = undesirable; LD = less desirable.

44

WorkingPaper_25.indd 44 1/10/2011 10:54:28 AM

 Table 25. Herbaceous species composition (% DM biomass) and their desirability by grazing type
in sesame–livestock farming system in Metema

Grazing types Herbaceous species Category % composition Species types

Communal U. fatamensis LD 2.6 Grasses
P. sphacelatum DS 22.0
U. bracyra LD 3.2
T. triandra HD 2.9
C. ciliaris DS 16.5
E. floccifolia LD 18.3
Ahiya abish UD Legumes
A. quartinianus LD
I. spicata DS
C. barbata UD Sedges
E. indica UD
S. sphaerostima UD
C. subul UD
H. schulli UD
Road side P. sphacelatum DS 8.5 Grasses
S. pumila DS 17.4
C. dactylon HD 3.6
R. fistulosa LD 9.2
E. floscifolia LD 23.1
Ahiya abish UD Legumes
C. mimosoides (L) UD
E. indica UD Sedges
H. schulli UD
S. sphaerostima UD
B. setigera UD
Z. elegans Jaquin. UD
C. subula UD
Enclosure P. sphacelatum DS 6.4
T. triandra HD 2.6
S. pumila DS 5.2
B. lata DS 8.3
P. coloratum HD 20.5
H. rufa HD 25.1
S. pyramidialis DS 5.1
E. flocifolia LD 5.6
I. spicata DS Legumes
A. quartinianus LD
H. articulatus HD
V. membranacea DS
Corchorus trilocularis L DS Sedges
B. setigera UD
C. barbata UD
D = Desirable, HD = highly desirable; UD = undesirable; LD = Less desirable.

45

WorkingPaper_25.indd 45 1/10/2011 10:54:28 AM

 A total of 25 herbaceous species are recorded in this farming system, and the
majority (56%) are grass species, followed by sedges (24%) and legumes (20%). In
the communal grazing areas, 13 herbaceous species are identified, out of which 6
(46.2%) are grass species, and 3 (23.1%) and 4 (30.8%) are legumes and sedges/others
species, respectively. From the grass species recorded, 16.7, 33.3 and 50.0% are highly
desirable, desirable, and less desirable, respectively. In road side grazing area, out of
the recorded 5 species of grasses, 20, 40 and 40% are highly desirable, desirable, and
less desirable, respectively. in enclosure grazing area, 8 species of grasses, 4 species of
legumes and 2 species of sedges are recorded. From the grass species identified, 37.5%
are highly desirable, 50% desirable and 12.5% less desirable. In road side grazing
areas, the less desirable species of Eleunine flocifolia is dominant (23.1%), followed by
the desirable species of S. pumila (17.4%) and P. spheslatem (9.2%). Highly desirable
species accounted for 14.28, 42.86, and 42.86% for the desirable and less desirable R.
fistulosa species of grasses. In enclosure areas, the highly desirable grass species of H.
rufa (25.1%) and P. coloratum (20.5%) are the dominant specie. There is a relatively high
percentage of highly desirable (decreasers) grass species in the enclosure followed by
communal and the road side grazing areas. This might suggest that the highly desirable
species are replaced by less desirable and unpalatable species as the grazing pressure
increased.

Woody vegetation in sesame based farming system

A total of 18 woody species are identified in the sesame–livestock farming system

(Table 26). These are composed of highly desirable (16.7%), desirable (38.9%), and less
desirable (44.4%) species. In the communal grazing areas, A. tortolis, A. leiocarpus, B.
egyptica, P. lucens, T. laxiflora, X. Americana, and Z. spina-christi are the common species,
while A. polyacantha, A. seyal, A. senegale and B. papirefera are the dominant species.
In the road side grazing areas, the frequency of A. polyacantha, A. tortolis, A. seyal and
D. cinerea was dominant, and A. senegal, A. leiocarpus, B. papyrifera, Z. spina-christi are
common species. In the enclosure areas, highly desirable species like, A. leiocarpus, P.
lucens, Z. spina-christi and the desirable species of A. seyal are dominant. A. Balanites
egyptica, B. papirefera, A. senegal, A. polyacantha, and T. laxiflora are common species.

The proportion of desirable tree species in the communal and road side is lower than in
the enclosure areas, probably due to the influence of people and animals. For instance,
group discussions revealed (confirmed during the field work) that about 10 years ago,
almost all of the district was covered by bamboo tree species, and the remains of bamboo
trees were observed on steep slope areas and undisturbed remote kebeles of Shimele
Gara, Kemechela, Zebach Bahir and Lemlem Terara. A. polyacantha was found in very

46

WorkingPaper_25.indd 46 1/10/2011 10:54:28 AM

 low density a decade ago, but currently it is the dominant species. This result agrees with
the suggestion of Alemayehu (2006).

Table 26. Woody species in the sesame based farming system by grazing type in sesame–livestock
farming system (SLFS)

SLFS
Woody species Category
CG RS EN
A. polyacantha DS C D
T. laxiflora LD P C
A. tortilis DS C C C
A. seyal DS D C C
D. cinerea DS C D P
A. leiocarpus DS C P D
P. lucens HD C P D
F. sycomorus DS P P
C. collinum LD C P C
X. Americana LD P – –
B. aegyptiaca DS C C C
G. villosa LD P – C
Z. spina-christi HD C C D
P. toningii DS P P
F. virosa LD C
B. papyrifera UD C C D
A. senegal DS C P C
C. mole LD C C
CG: communal grazing; RS: road side; EN: enclosure; P: Present (<10% density); C: common (>10% and <20%
density); D: dominant (>20% density).

4.9 Range condition assessment

4.9.1 Effect of farming systems on rangeland condition
Communal grazing areas

In the communal grazing areas, basal cover, litter cover and grass species composition
were significantly (P<0.05) higher in sesame–livestock than in the cotton–livestock
farming system (Table 27). This may be associated with grazing intensity and disturbances
of the grazing land in the cotton–livestock farming system to be relatively higher than the
sesame–livestock farming system. Different literature also suggested that the frequency
and intensity of grazing influences the rate of live biomass accumulation on a site thereby
affecting the rate of competitive displacement in a multi species community. Additionally,
grazing affects the amount of plant litter at the soil surface with important indirect effects
on patterns of germination and seedling establishment (Belaynesh 2006; Teshome 2006;
Lishan 2007). Productivity of most rangelands has been reduced by human and livestock

47

WorkingPaper_25.indd 47 1/10/2011 10:54:28 AM

 pressure and natural hazards. Because vegetation integrates all environmental factors
acting on a site, knowledge on its type may be used to make inference about prevailing
environmental patterns (Herlocker 1999). Major causes of changes in rangelands are
excessive grazing by domestic and/or wildlife animals, cultivating for cropping and
harvesting resources like firewood, foods and building materials (Teshome 2006; Lishan
2007).

Table 27. Range condition score (Mean ± SE) of communal grazing areas by farming system in
Metema

Parameter CLFS SLFS

Grass species composition score 4.64 ± 0.31b 6.62 ± 0.31a
Basal cover 4.80 ± 0.25a 5.20 ± 0.25a
Litter cover 4.62 ± 0.37a 5.15 ± 0.37a
Soil erosion 3.11 ± 0.30a 2.77 ± 0.30a
Soil compaction 2.79 ± 0.28a 3.33 ± 0.28a
Age distribution of grasses 2.36 ± 0.27a 3.25 ± 0.27a
Seedling distribution 2.91 ± 0.34a 3.22 ± 0.34a
Woody density score 2.25 ± 0.27a 1.50 ± 0.27a
Canopy cover score 8.92 ± 0.77a 6.97 ± 0.77a
Hedging 1.90 ± 0.25a 2.15 ± 0.25a
Total range condition score 38.32 ± 1.21a 40.22 ± 1.21a
Range condition Fair Fair
Woody density 3354.17 ± 292.28a 4137.52 ± 292.28a
Canopy cover 146.55 ± 15.98a 179.1 ± 15.98a
CLFS: cotton–livestock farming system; SLFS: sesame–livestock farming system.
Means with different superscripts in a row differ significantly (P<0.05).

Soil erosion and compaction did not differ between farming system. Soil erosion and
compaction depend on a number of man made and natural factors including vegetation
cover, soil type, intensity and pattern of rainfall, degree of wind erosion, high percentage
value of bare ground and grazing management systems in an area.

The woody vegetation density, canopy cover, hedging effect, age distribution and total
condition score in the sesame–livestock farming system was significantly (P<0.05)
higher than the cotton–livestock farming system (Table 27). The possible reason for
the differences between the two could be the disturbance of grazing areas in cotton–
livestock farming system by humans and livestock. Based on density value (plants/ha),
the most common and/or dominant woody species in the communal grazing areas of
cotton–livestock farming are A. polyacantha (3050), A. tortolis (1450), A. seyal (1250),
A. senegal (800), A. liocarpus (1800), B. aegyptyca (1050), B. paperiferra (650), P. lucens
(1150), F. sycomorus (400) S. kunthianum (150), X. Americana (300) and Z. spina-christi
(1550). Studies in Borana area have shown that when the woody plant density is greater
than 2400 plants per hectare, the area is moving towards bush encroachment (Gemedo

48

WorkingPaper_25.indd 48 1/10/2011 10:54:28 AM

 2004). Based on the results of this study, the communal grazing areas are affected by
bush encroachment.

Woody vegetation reduces grass cover through increasing the competition for available
water and nutrients and reduces the light reaching the grass layer. In addition to
competing with grasses, these noxious woody plants are commonly thorny and thicket
forming and the grazing capacity of the rangeland may be extremely reduced (Alemayehu
2004). Factors regulating the balance between graminoid and woody plant life-forms
include climate, soils, disturbance (e.g. grazing, fire), and their interactions. Changes
in one or more of these factors may enable woody plants to increase in abundance.
Characteristics common to many woody species that increase in grazed environments
include high seed production, seeds that persist in soil for many years, ability to disperse
over long distances, ability to sprout following top removal, tolerance to low levels of
water and nutrients and low palatability (Archer 2003). These authors also suggested
that as climate changes occur over time, undesirable woody species vigorously spread
out on communal grazing lands. In semi-arid ecosystems, drought, absence of fire
and overgrazing are some of the major factors that cause conversion of grasslands to
woodlands (Coppock 1994; Archer 2003). The increase in woody plant encroachment,
loss of palatable grass cover and increase of unpalatable forbs are the main threats to the
communal grazing areas in Metema. The result of this study revealed that the condition of
the communal grazing lands is in fair condition, and this implies that there is need to take
measures to improve the rangelands. Herd diversification through increasing the number
of browsers such as camels and goats will contribute towards efficient resource utilization
and decrease encroachment of woody plants.

Road side grazing areas

The mean range condition scores for herbaceous species variables considered are
presented in (Table 28). Grass species composition, basal cover and litter cover were
significantly (P<0.05) higher in the sesame–livestock than in the cotton–livestock farming
system. This could possibly be associated with the grazing pressure exerted by livestock
and climate changes that favour the replacement of most palatable tall and erect species
such as (H. rufa, T. triandra and P. coloratum) by creeping, spreading and grazing resistant
species such as C. dactylon and less palatable E. floccifolia grass species which cover the
soil. Similar studies made in Borana (Ayana 1999), Middle Awash (Amsalu 2000; Abule
et al. 2007a), Bena-Tsemay (Admasu 2006), Somali Region (Lishan 2007) and Bale Zone
(Teshome 2006) in Ethiopia revealed that the percent of cover decreased as the condition
of the range decline, due to the replacement of tall and erect species with low growing
and spreading species.

49

WorkingPaper_25.indd 49 1/10/2011 10:54:28 AM

 Table 28. Range condition score (Mean ± SE) for road side grazing areas found in different farming
systems of the study district

Parameters CLFS SLFS

Grass species composition 4.43 ± 0.22b 5.62 ± 0.22a
Basal cover 4 ± 0.52a 4.25 ± 0.52a
Litter cover 3.51 ± 0.39a 4.87 ± 0.39a
Soil erosion 2.40 ± 0.26a 2.0 ± 0.26a
Soil compaction 2.32 ± 0.27a 2.49 ± 0.27a
Age distribution of grasses 2.25 ± 0.32a 2.71 ± 0.32a
Seedling distribution 1.96 ± 0.38a 2.18 ± 0.38a
Woody density score 2.75 ± 0.27a 2.5 ± 0.27a
Canopy cover score 3.02 ± 0.79a 1.9 ± 0.79a
Hedging 1.25 ± 0.29a 2.10 ± 0.29a
Total range condition score 27.92 ± 1.42a 30.65 ± 1.42a
Range condition Poor Poor
Woody density 2678.33 ± 196.34a 2795.85 ± 196.34a
Canopy cover 68.38 ± 13.04a 75.40 ± 13.04a
CLFS: cotton–livestock farming system; SLFS: sesame–livestock farming system.
Means with different superscripts in a row differ significantly (P<0.05).

The mean density of woody plants along the road side grazing lands is 2678.3 ± 13.04/
ha in the cotton–livestock farming system and 2795.9 ± 13.04/ha in the sesame–livestock
farming system. Based on density value (plants/ha), the woody species in the road
side grazing areas of the cotton–livestock farming system identified as common and/or
dominant are A. tortilis, (1600), A. seyal, (350); B. egyptica, (1300), D. cinerea (1655),
G. ternifolia, (1150), P. toninngii (300), while A. syal (2200), B. egyptica, (1350), B.
periphera, C. collinum, (300), T. laxiflora (750), (1660), and Z. spina-christi (1850) are
identified as common and/or dominant in the sesame–livestock farming system. Similar
to the communal grazing areas, the road side grazing areas are also encroached by
bush. The density of woody plants as well as the total range condition scores indicated
that the riverside grazing areas are in poor condition, and it is imperative to improve the
condition of the rangeland.

Enclosure

In the enclosure areas, the grass species composition, basal cover, litter cover, woody
density and canopy cover scores are significantly (P < 0.05) higher in the sesame–
livestock than in the cotton–livestock farming system. This could be attributed to the
variation in land use pattern of the sites and the response of the species to protection
from grazing. Studies indicated that knowledge of the land use history of a site is
imperative (Fleischner 1994). In addition to this spatial scale, e.g. plant distribution
patterns in an enclosures plot may be caused by different processes than patterns found

50

WorkingPaper_25.indd 50 1/10/2011 10:54:28 AM

 at the landscape scale; short temporal scales (the initial response of a community to
protection or release from grazing may not represent the long-term response) and the
pool of species that are present or able to disperse to the protected area. There was no
significant (P < 0.05) difference in the soil erosion, compaction, age distribution, seedling
count, and hedge effects between the two farming systems (Table 29). Generally, these
parameters suggested that the condition of the enclosure sites is in a good condition. The
number of livestock that graze in the enclosure areas limits the grazing pressure and had
a positive effect on rangeland condition. This is considered as a means to protect and
conserve local resource under threat from increasing human and livestock population
pressure and interventions.

Table 29. Range condition score (LSM ± SE) in enclosure grazing areas by farming system in
Metema

Parameter CBFS SBFS

Grass species composition score 7.80 ± 0.61a 8.11 ± 0.50a
Basal cover 6.91 ± 0.46a 7.03 ± 0.37a
Litter cover 6.0 ± 0.33a 7.05 ± 0.27a
Soil erosion 3.88 ± 0.16b 4.89 ± 0.20a
Soil compaction 4.93 ± 0.38a 3.64 ± 0.31a
Age distribution of grasses 3.70 ± 0.44a 3.67 ± 0.35a
Seedling distribution 4.0 ± 0.21a 4.64 ± 0.17a
Woody density score 3.5 ± 0.28a 3.0 ± 0.23a
Canopy cover score 2.75 ± 0.61a 3.46 ± 0.50a
Hedging 2.20 ± 0.41a 2.43 ± 0.33a
Total range condition score 46.65 ± 0.78a 46.93 ± 0.63a
Range condition Good Good
Woody density 2316.7 ± 244.68a 2340.0 ± 199.78a
Canopy cover 64.05 ± 7.27a 71.93 ± 3.44a
CLFS: cotton–livestock farming system; SLFS: sesame–livestock farming system.
Means with different superscripts in a row differ significantly (P<0.05).

The woody vegetation density in the enclosure areas did not differ between the two
farming systems (Table 29), and the overall mean density of woody vegetation is 1029.9
plants/ha. Based on density value (plants/ha) the common/dominant species in the
cotton–livestock farming system are A. leiocarpus (300), A. senegal (540), B. aegyptiaca
(500), B. papyrifera (500), P. lucens (400), and Z. spina-christi (1540). In the sesame–
livestock farming system, A. polyacantha (650), A. tortilis (500), A. seyal (2250), A.
leiocarpus (600), A. senegal (1550) B. aegyptiaca (1480), B. papyrifera (3200), G. villosa
(450), C. mole (100), P. lucens (2200), S. kunthianum (300) and Z. spina-christi (1540) are
identified as the common and/or dominant species.

The enclosure areas in the two farming systems have different range condition class. In
general, the condition of the rangeland in the enclosure areas is good, and implies that

51

WorkingPaper_25.indd 51 1/10/2011 10:54:29 AM

 establishing enclosures is an alternative method of improving the rangelands. A similar
observation was made by Ayana (1999); Amsalu (2000); Admasu (2006); Amaha (2006);
Teshome (2006); Lishan (2007) who reported that good range condition class in enclosure
areas in Bale, Somali, Borana, Rift valley, Somali and Hamer and Bena-Tesmay areas of
Ethiopia, respectively.

4.10 Biomass production

4.10.1 Dry matter biomass production in different grazing types
by farming system
Communal grazing areas

Data on dry matter biomass production of highly desirable, intermediate and less
desirable grasses in the communal grazing areas are presented by farming system in Table
30.

Table 30. Dry matter biomass (kg/ha) of the communal grazing areas by farming system in Metema
district

CLFS SLFS
Parameters
Mean ± SE Mean ± SE
Total grasses 2396.6 ± 370.17a 3282.2 ± 598.77b
Highly desirable 143.3 ± 22.14a 145.1 ± 27.46a
Intermediate 1562.8 ± 242.38a 1929.0 ± 352.6b
Less desirable 690.2 ± 106.61a 1206.5 ± 221.29b
Legumes 699.2 ± 56.69a 923.3 ± 213.74b
Others 488.7 ± 128.09a 802.1 ± 163.14b
Total biomass 3584.4 ± 403.34a 5007.6 ± 664.25b

CLFS = cotton–livestock farming system, SLFS = sesame–livestock farming system.

Means with different superscripts in a row differ significantly (P<0.05).

The dry matter biomass from grasses and legumes and the total dry matter biomass
production are higher in the sesame–livestock than in the cotton–livestock farming
system, and these differences may be attributed to the higher grazing intensity and
anthropogenic disturbance of the cotton based farming system. Studies on soil erosion
and soil compaction generally found out that exposure to livestock grazing compacts soil
and this again increases with grazing intensity (Fleischner 1994). Therefore, compaction
is directly related to soil productivity, because it reduces water and air movement into
and through the soil and reduces water and air availability to the root system of plants.

52

WorkingPaper_25.indd 52 1/10/2011 10:54:29 AM

 Soil compaction also directly restricts root growth because as soils become compacted
only fewer large pores are present and so there is little space for roots to enter (Amaha
2006). In semi-arid rangelands, under similar rainfall conditions, soil type will have an
effect on plant biomass production. Plant biomass and standing crop are affected by
species composition and density (Alemayehu 2005).

Road side grazing areas

In the road side grazing areas, dry matter biomass production of total grass, legumes,
others and total biomass production are significantly (P<0.05) higher in the sesame–
livestock than in the cotton–livestock farming system (see Table 31). The difference
could be associated with the increasing grazing intensity and anthropogenic
disturbances applied in the cotton–livestock farming system where heavy and
continuous grazing pressure resulted in decreased biomass production. There was,
however, no significant (P<0.05) difference in the dry matter biomass of legumes
between the two farming systems. Grasslands are able to tolerate a moderate degree of
grazing intensity before changing in composition, diversity, or productivity. However, as
grazing intensity increases or becomes continuous, tall and medium grasses eventually
give way to short-stature perennial grasses, which, in turn, gives way to annuals and
unpalatable perennials with a concomitant loss of primary and secondary productivity,
diversity, cover, and soil. Different reports on Ethiopian rangelands indicated that
decline in perennial grass and increase in unpalatable forbs and annual grass cover
influenced types of grazing management, rainfall and livestock population pressure
where overgrazing coupled by a high population of livestock and prolonged drought
may lead to reduction of dry matter biomass production and aggravate rangeland
deterioration (Ayana 1999; Amsalu 2000; Gemedo 2004; Amaha 2006; Abule et al.
2007a, b, c).

Table 31. Dry matter biomass (kg/ha) of the road side grazing areas by farming system in Metema

CLFS SLFS
Parameters
Mean ± SE Mean ± SE
Grasses 1631.6 ± 370.17a 1963.7 ± 598.77b
Highly desirable 79.3 ± 18.99a 114.8 ± 35.85b
Intermediate 751.9 ± 170.60a 822.8 ± 250.88b
Less desirable 798.7 ± 181.19a 1026.2 ± 312.92b
Legumes 422.1 ± 106.45a 497.7 ± 72.21a
Others 288.7 ± 52.98a 712.9 ± 225.98b
Total biomass 2342.5 ± 403.34a 3174.2 ± 664.25b
CLFS = cotton–livestock farming system, SLFS = sesame–livestock farming system.
Means with different superscripts in a row differ significantly (P<0.05).

53

WorkingPaper_25.indd 53 1/10/2011 10:54:29 AM

 Enclosure

In the enclosure areas, total dry matter biomass, dry matter biomass of grasses and
legumes are significantly higher (P < 0.05) in the sesame–livestock than in the cotton–
livestock farming system (Table 32). This may be associated with the presence of more
dominant herbaceous species in the sesame–livestock system that may have contributed
to the higher biomass production. Some literature suggested that the most dominant
species contributed the highest amount of biomass (Kamau 2004). The total dry matter
biomass value obtained for the enclosure areas in this study is much higher than those
reported by Amsalu (2000), Amaha (2006), and Lishan, (2007) for the arid and semi-arid
rangelands of Middle Awash Somali Region of Ethiopia. This implies that the productivity
of enclosure areas in Metema is better than other rangelands in Ethiopia and can support
livestock population provided that good management practice is applied.

Table 32. Dry biomass (kg/ha) of enclosure grazing areas by farming system in Metema

CLFS SLFS
Parameters
Mean ± SE Mean ± SE
Grasses 4814.9 ± 453.36a 8438.6 ± 598.77b
Highly desirable 3343.0 ± 259a 5161.0 ± 367.21b
Desirable 1165.2 ± 15.55a 2676.7 ± 198.9b
Less desirable 305.7 ± 23.51a 599.1 ± 42.51b
Legumes 697.1 ± 164.73a 1071.3 ± 116.27b
Others 375.6 ± 116.27a 1186.0 ± 203.98b
Total biomass 5887.6 ± 493.9a 10695.8 ± 664.2b
CLFS = cotton–livestock farming system, SLFS = sesame–livestock farming system.
Means with different superscripts in a row differ significantly (P<0.05).

4.11 Correlation among variables studied for range

condition assessment
Among the different variables correlated, only the most important ones were presented
(Appendix Tables 5–10). Correlation analysis showed that grass species composition
was positively correlated with basal cover, total dry matter biomass and total condition
score. Similarly, grass species composition was negatively correlated with soil erosion,
soil compaction, woody density and canopy cover. Basal cover was highly and positively
correlated (P ≤ 0.05) with grass species composition and total dry matter biomass and
also positively correlated with total condition scores. These finding were in agreement
with the report of Gemedo (2004) from the Borana rangelands.

54

WorkingPaper_25.indd 54 1/10/2011 10:54:29 AM

 Dry matter biomass
18000

16000

14000

12000

10000
y = 282.13x - 4066.4
8000 R2= 0.4509
6000

4000

2000

0
0 10 20 30 40 50 60
Range condition score

Figure 2. Regression graph of dry matter biomass and range condition.

There were positive and significant correlations between basal cover and grass species
composition (P < 0.01); between grass species composition and age distribution (P <
0.01) whereas strong negative correlations were noted among basal cover, soil erosion,
soil compaction, woody density and canopy cover. Total dry matter biomass was also
negatively correlated with soil erosion, soil compaction, woody density and canopy cover
in the communal grazing areas found in the two farming systems. Similarly, the total dry
matter biomass was positively correlated with age distribution, basal cover and grass
species composition in the road side grazing areas (Appendix Tables 5–10).

In the enclosure grazing areas, correlation analysis showed that woody density and
canopy cover were negatively correlated (P ≤ 0.05) with basal cover, total dry matter
biomass and positively correlated (P ≤ 0.05) with soil erosion and soil compaction. On
the other hand, density of woody vegetation and canopy cover were positively correlated
(P ≤ 0.05) with each other (Appendix Tables 5–10). Similarly, total dry matter biomass
production was negatively correlated (P ≤ 0.05) with soil erosion, soil compaction,
woody vegetation density and canopy cover. Besides, the total dry matter biomass was
positively correlated (P ≤ 0.05) with basal cover and grass species composition.

55

WorkingPaper_25.indd 55 1/10/2011 10:54:29 AM

 5 Conclusion and recommendations
5.1 Conclusion
This study was conducted in Metema district, North Gondar Zone, Amhara National
Regional State of northwestern Ethiopia. The aims of the study were to characterize the
existing rangeland, to determine the available feed resources and utilization practices,
to assess the natural grazing land based on herbaceous and woody biomass and soil
condition and to evaluate the chemical composition of major livestock feed resources
of the area. The feed resources utilization practice of the district was assessed through
interviewing 140 households using a semi-structured questionnaire, group discussions
and personal observations. For the study of rangeland condition, the district was classified
into cotton–livestock and sesame–livestock farming systems and three grazing types
(communal grazing, enclosure and road side grazing areas). Data on grass species
composition, basal and litter covers, age distribution, soil erosion, soil compaction and
woody species density, tree height class, canopy cover and hedging effect were collected.

In Metema, the mean family size is 5.31 ± 0.20 persons and the level of education is
low. Smallholder crop–livestock mixed farming is dominant and is the main occupation
of the people. The mean land holding is 6.78 ± 1.33 ha per household, and the land use
pattern includes annual cropping, perennial cropping, communal grazing, fallowing and
private grazing. Cattle are the dominant livestock followed by small ruminant (goats and
sheep), donkeys and camel. The mean total livestock holding per household is 27.58 ±
13.72 TLU, and is composed of 12.52 ± 6.23 TLU cattle, 0.80 ± 0.40 TLU goats, 0.13 ±
0.07 TLU sheep, 0.65 ± 0.32 TLU donkeys and 0.07 ± 0.04 TLU camels. Female stocks
of cattle and goats are dominant in both farming systems because male stocks are sold
at earlier ages and females are retained as replacement breeding stock. In addition,
ownership of large number of cows is considered as prestige and wealth ranking by the
community.

The major livestock feed resources in Metema are natural pasture, crop residues, crop
aftermath and fallow land. Natural pasture is the major sources of livestock feed,
and contributes to 59.3, 72.9, 72.1 and 42.1% of the diets of cattle, goat, sheep and
donkeys, respectively. Although there is surplus feed during the rainy season, these
pastures can hardly maintain animals in the dry season due to limited availability and
very low nutritive value. Free grazing, tethering and cut-and-carry feeding systems are
the commonly practised feeding systems in both farming systems. Animals are allowed
to graze in communal grazing land, forest land and fallow lands (privately owned land)
during the wet season and from October onwards. From the total area of the district,

56

WorkingPaper_25.indd 56 1/10/2011 10:54:30 AM

 i.e. 312,300 ha, about 71% is natural pasture and forest land indicating the existence of
extensive grazing land resources. The abundantly available feeds in the wet season are
mostly wasted because of poor feed conservation practices. Farmers make hay late in the
season between mid-October and end of November, when the herbaceous species are
lignified and low in nutrient quality.

Crop residues are the second major feed resources. Sorghum is the most important
cereal crop grown in Metema next to sesame and cotton, and accounts for about 90%
of the total crop residues produced in the district. Crop residues of maize, teff, finger
millet and rice make up for the difference. As observed during the field work, there is no
proper collection, handling, storage and utilization of crop residues. The total mean crop
residues utilization as livestock feed is low, and most of it is either burned in the field or
left as mulch. Generally, the feed balance estimate of the district on DM basis is sufficient
to support the maintenance requirement of livestock per household, but the quality of the
feeds is very poor.

In the rangelands, a total of 32 herbaceous and 20 woody species were recorded.

Among the herbaceous species, 41.9% are different grasses and 58.1% are non-grass
species. The non-grass species are comprised of five species of legumes, seven species
of sedges and six species of forbs. Of the grass species, 23.1% are highly desirable,
38.5% desirable and 30.8% are less desirable. P. spheslatum (Jingra) and C. ciliaris
(Zemen) are the dominant and desirable grass species found in the communal grazing
areas of the sesame–livestock and cotton–livestock farming system, respectively. Of
the identified woody species, 15, 35, and 50% are highly desirable, desirable and less
desirable, respectively. The largest proportion of the woody vegetation is made up of
different species of acacia (20%) and commbretum (10%), hence the rangeland can be
categorized as an acacia dominated grazing land. P. lucens (Charia) is a very important
fodder tree used as a main livestock feed during the onset of the main rainy season. The
woody vegetation falls within the height class of >1–3 m in about 41% of the communal
grazing areas, 39% of the road side grazing land and 33% of the enclosure areas,
indicating that most of the vegetation is at the browsing height of the animals.

In cotton–livestock farming system, a total of 24 herbaceous and 17 woody species are

recorded. Among the herbaceous species, 50% are different grass species, 20.8% legumes
and 29.2% sedge species. Among the woody species, 17.7, 41.2, and 35.3% are highly
desirable, desirable and less desirable, respectively, while 5.9% are undesirable. In the
sesame–livestock farming system, 25 herbaceous and 18 woody species were identified.
Among the herbaceous species, 56% are grasses, 20% legumes and 24% sedges species.
About 17% of the grass species are highly desirable, while 38.9 and 44.4% are desirable

57

WorkingPaper_25.indd 57 1/10/2011 10:54:30 AM

 and less desirable, respectively. In the communal grazing areas, basal cover, litter cover and
grass species composition are significantly (P<0.05) higher in the sesame–livestock than in
the cotton–livestock farming system. Higher soil erosion and compaction is observed in the
cotton–livestock than in the sesame–livestock farming system. The woody vegetation density
is higher (P<0.05) in the sesame–livestock than in the cotton–livestock farming system.
Grass species composition, basal cover and litter cover of road side grazing is significantly
(P<0.05) higher in sesame–livestock than in the cotton–livestock farming system. Total dry
matter biomass production is significantly (P < 0.05) higher in enclosure areas followed by
the communal and road side grazing areas.

In general, there is low feed resources conservation and utilization and very poor
traditional grazing land management system in Metema. The abundant feed resources
in the wet season are wasted. In the dry season, grasses are turned to ash by wild and
man-made fire (forest honey harvesters and crop land cleaning). As a result, the livestock
population seriously suffer from critical feed shortage during the long dry season. The
rangeland, species composition and biomass production are also affected by human,
livestock and natural factors (biotic and abiotic factors). The human population of the
district has increased due to settlement programs, investment induced settlers, expanding
crop cultivation and have increased the pressure on the rangelands and natural grazing
areas. Bush encroachment and overgrazing are also serious problems. Shifting cultivation
practice is also contributing to the increased bush encroachment. The seasonal movement
and transhumant livestock production by highlanders in adjacent districts also increases
the grazing intensity.

5.2 Recommendations
• The rangelands in the lowlands of northwestern Ethiopia are important resources
for the country and beyond. They are rich in biodiversity, have huge economic
importance and play an important ecological buffering role between the Sahel and the
highlands of Ethiopia and should be managed properly.
• The rich biodiversity of incense and gum trees is critically important. The fast
disappearing species such as B. paperiferra and Acacia spp. should be restored
through fast propagation and multiplication and aggressive planting campaigns. The
excessive and indiscriminate burning of important species of trees and bushes for fire
wood should be regulated.
• The livestock resources are enormous and should be properly utilized.
• Conservation of grass hay, which otherwise is burned down during the dry season,
should be averted and could be developed as a marketable commodity to the
adjacent highland woredas, where feed shortage is crucial.

58

WorkingPaper_25.indd 58 1/10/2011 10:54:30 AM

 • Transhumance production system is increasingly leading to poor rangeland
management and social conflicts over feed resources between the highlanders and
lowlanders. Balance has to be established between the two systems.
• Settlers and investors or any new comers to the region should be educated ahead of
time about the ecological and economic importance of the natural resources base and
should be regulated in terms of utilization. There should be regular campaigns of tree
planting.
• Absence of adequate baseline information about the rangeland resources,
unsynchronized seasonal availability of feed resources and cropland encroachment
to the rangeland are some of the main constraints of the district. Studies on rangeland
management systems and improved livestock production should be initiated.
• The deteriorating condition of the rangelands in Metema should be reversed through
rangeland rehabilitation, proper management and clearly demarcated land use system
of the natural grazing lands.
• Provision of integrated extension service is required on range management, feed
resources development and management and training on feed collection, storage, and
proper feeding systems.
• Balancing grazer and browser livestock species is essential in order to keep ecological
balance as well as to increase livestock productivity.
• Introduction of dual purpose food–feed crops such as sorghum species should be
considered.
• The influence of different sub-habitats of woody plants (under canopy, between
canopies) on herbaceous species composition and diversity, biomass production and
their influence on livestock productivity should be studied.

59

WorkingPaper_25.indd 59 1/10/2011 10:54:30 AM

 References
Abule, E., Snyman, H.A., Smit, G.N. 2007a. Rangeland evaluation in the middle Awash valley of
Ethiopia. I. Herbaceous vegetation cover. Journal of Environments 70:253–271.
Abule, E., Snyman, H.A., Smit, G.N. 2007b. Rangeland evaluation in the middle Awash valley of
Ethiopia. II. Woody vegetation. Journal of Environments 70:272–292.
Abule, E., Snyman, H.A., Smit, G.N. 2007c. Rangeland evaluation in the middle Awash valley of
Ethiopia. I. Herbaceous vegetation cover. III. Relationship among soil and vegetation variables.
Journal of Environments 70:293–303.
Admasu Terefe, 2006. Pastoralists perceptions on range–livestock management practices and
rangeland assessment in Hamer and Benna-Tsemay districts of South Omo Zone. MSc Thesis.
School of Graduate Studies, Haramaya University, Ethiopia. 159p.
Ahmed Hassen, 2006. Assessment and evaluation of utilization practices of natural pastures and
other forages in Basona Worana wereda of North Shoa Zone. MSc Thesis. School of Graduate
Studies, Haramaya University, Ethiopia. 174p.
Alemayehu, M. 1998. Natural pasture improvement study around smallholder dairy Areas. MoA,
Small Dairy Development Project (SDDP), Addis Ababa, Ethiopia
Alemayehu, M. 2004. Rangelands Biodiversity: Concepts, Approaches, and the Way Forward. Addis
Ababa, Ethiopia.
Alemayehu Mengistu, 2005. Rangelands: Biodiversity Conservation and Management and Inventory
and Monitoring. Addis Ababa University, Faculty of Science, Addis Ababa, Ethiopia. 103 pp
Alemayehu Mengistu, 2006. Range Management for East Africa: Concepts and Practices, Sponsored
by RPSUD and Printed by A.A.U Printed Press. Addis Ababa, Ethiopia
Association of Official Analytical Chemists (AOAC), 1990. Official methods of analysis. 12th ed.
Washington, DC
Amaha Kassahun, 2006. Characterization of rangeland resources and dynamics of the pastoral
production system in the Somali region of Eastern Ethiopia. PhD Thesis. University of the Free
State, Bloemfontein, South Africa. 232p.
Amsalu Sisay, 2000. Herbaceous species composition, dry matter and condition of the major
grazing areas in the middle rift valley, Ethiopia. MSc. Thesis, School of Graduate Studies,
Alemaya University, Dire Dewa, Ethiopia
Archer, S. 2003. Proliferation of woody plants in grasslands and savannas. Retrieved on 15/3/2008
from the internet: [Link] Tamu. Edu/rlem/ faculy/ archer/ [Link]/.
(AOAC), 1980. Association of Official Analytical Chemists methods of analysis. 12th ed.
Washington, DC
Ayana Angasa, 1999. Range condition traditional grazing management system in Borena. MSc
Thesis. School of Graduate Studies, Alemaya University, Dire Dawa, Ethiopia .55p
Azene Bekele, A. Anne Birnie and B. Tengnas, 1993. Useful trees and shrubs for Ethiopia,
Identification, Propagation and Management for agricultural and pastoral communities. English
press, Nairobi, Kenya. 473p.
Baars R. M. T., E.C Chileshe, and D .R. Kalokoni, 1997. Technical notes: Range condition in high
cattle density areas in the Western province of Zambia. Tropical grasslands, 31:569–573.
Befekadu Degefe and Berhanu Nega. 2000. Annual report on the Ethiopian economy: Volume I,
1999/2000. The Ethiopian Economic Association, Addis Ababa, Ethiopia. 429 pp.
Belaynesh Debalkie, 2006. Floristic composition and diversity of the vegetation, soil seed bank
flora and condition of the rangelands of the Jijiga zone, Somali Regional State, Ethiopia. MSc
Thesis. School of Graduate Studies, Alemaya University. 144p.

60

WorkingPaper_25.indd 60 1/10/2011 10:54:30 AM

 Beruk Yemane and Tafesse Mesifin, 2000. Pastoralism and agropastoralism: Past and present. In:
Proceeding of 8th annual conference of the Ethiopia Society of Animal Production (ESAP). Addis
Ababa, Ethiopia, 24–26 August 2000.
BoA (Bureau of Agriculture).1999. General basic agricultural data: Woreda level. BOA, Bahir Dar,
Ethiopia. 81 pp.
CACC (Central Agricultural Census Commission). 2002. Ethiopian agricultural sample enumeration
report, held 2001/02 (1994 EC). Addis Ababa, Ethiopia.
Chrisholm. A. and R. Dumsday, 1987, Land Degradation. Cambridge University Press, Cambridge.
Coppock, D.L., 1994. The Borena Plateau of Southern Ethiopia: Synthesis of Pastoral Research,
Development and Changes, 1980–1990. International Livestock Center for Africa (ILCA), Addis
Ababa, Ethiopia .72p.
Cossins, N. and M. Upton, 1985. The Productivity and Potential of the Southern Rangelands of
Ethiopia. JESS (Joint Ethiopian Pastoralists system study).Draft summary report, International
Livestock Centre for Africa, Addis Ababa, Ethiopia.
Cossins, N.J. M. Upton, 1988. Options for Improvement of the Borana Pastoral System’ Agricultural
system 27 pp 251–278.
CSA. 2003. Ethiopian Agricultural Sample Enumeration . 2001/02 results for Amhara region,
statistical report on livestock and farm implements, part 4 , Addis Ababa, Ethiopia.
CSA 2005. Statistical Abstracts, Addis Ababa, Ethiopia.
CSA 2009. Statistical Abstracts, Addis Ababa, Ethiopia.
de Leeuw, P.N. and B. Rey, 1995. An analysis of the current trends in the distribution of ruminant
livestock in sub-Saharan Africa. World Animal Review 83: 43–59.
de Leeuw P.N. and Toothill J.C. 1993. The concept of rangeland carrying capacity in sub-Saharan
Africa—Myth or reality. In: Behnke R.H. Jr., Schoones I. and Kerven C. (eds), Range ecology at
disequilibrium: New models of natural variability and pastoral adaptation in African savannas.
ODI (Overseas Development Institute), London, UK. pp. 77–88.
Elias Mulugeta, Berhanu Gebremendihin, Hoekstra, D and Jabbar M. 2007. Analysis of Ethio–Sudan
cross-border cattle trade: The case of Amhara Regional State. IPMS (Improving Productivity and
Market Success) of Ethiopian Farmers Project Working Paper 4. ILRI (International Livestock
Research Institute), Nairobi, Kenya. 41 pp.
Fleischner, T.L.1994. Ecological costs of livestock grazing in Western North America. Conservation
Biology 8(3):629–44
Friedel MH, Laycock WA and GN. Bastin, 2000. Assessing Rangeland Condition and Trend. In:
Mannetje, L.’t., and Jones RM (Eds.). Field and Laboratory Methods for Grassland and Animal
Production Research, pp 305–360. CABI International, Wallingford, UK.
Gemedo Dalle, 2004. Vegetation Ecology, Rangeland Condition and Forage Resources Evaluation
in the Borana Lowlands, Southern Oromia, Ethiopia. PhD Thesis. University of the Gottingen,
Germany. 253 pp.
Herlocker, D.,1999. Rangeland Ecology and Resource Development in East Africa .GTZ. Nairobi,
Kenya. 393 pp.
Holechek, J. L., R.D. Pieper, and C.H., Herbel, 1994. Range Management Principles and Practices.
2nd Edition. Prenticehall. Englewood Cliffs, NJ.
ILCA (International Livestock Center for Africa), 1990. Livestock Systems Research Manual. Vol. 1.
ILCA Working Paper 1. ILCA, Addis Ababa, Ethiopia.
IPMS. 2005. Metema pilot learning site diagnosis and program design. ILRI, Addis Ababa, Ethiopia.

61

WorkingPaper_25.indd 61 1/10/2011 10:54:30 AM

 Kidane, G. 1993. Effects of Cutting date on Botanical Composition and Nutritive of value of Native
pasture in the Central Ethiopian high lands. MSc. Thesis. Alemaya University of Agriculture.
Alemaya, Ethiopia.
Kuchar, P., 1995. Range monitoring, evaluation and range survey methods, Southeast Range Project
Technical Report. Addis Ababa, Ethiopia. 92p.
Lishan Tsehay, 2007. Woody and herbaceous species composition and the condition of the
rangelands in Shinile zone of condition of the rangelands in Shinile zone of Somali regional
state, Ethiopia. MSc Thesis. School of Graduate Studies, Haramaya University, Ethiopia.136p.
Mannetje L’t, Jones RJ and Stobbs TH 1976. Pasture Evaluation by Grazing Experiments. In:
Shaw NH and Bryan WW (eds), Tropical pasture research Principles and Methods. Bull. 51.
Commonwealth Agricultural Bureaux. Farnham Royal. Bucks. England. pp. 222–234.
MoA (Ministry of Agriculture). 1984. General Agricultural Survey. Preliminary Report. 1983/84. Vol.
II MoA, Addis Ababa, Ethiopia.
Osuji P O, Nsahlai I V and Khalili H.. 1993. Feed Evaluation. ILCA Manual 5. ILCA (International
Livestock Centre for Africa), Addis Ababa, Ethiopia. 40 pp.
Pratt, D.J and M. D. Gwynne, 1977. Range Management and Ecology in East Africa, Frieger
publishing company, Huntington, New York. 391 pp.
Simbaya, J.,1998. Development and field evaluation of dry-season feed supplementation packages
for smallholder farms, Livestock and Pest Research Centre (LPRC), National Institute for
Scientific and Industrial Research (NISIR) Annual report, Chilanga, Zambia.
Sisay Amare, 2006. Qualitative and quantitative aspects of animal feed in different agro-ecological
areas of North Gonder. MSc. Thesis. Alemaya University, Ethiopia.
van Soest, P. J., 1988. Effect of environment and quality of fiber on the nutritive value of crop
residues. In: Reed J D, Capper B S and Neate P J H (eds), Plant breeding and the nutritive
value of crop residues. Proceedings of a workshop held at ILCA, Addis Ababa, Ethiopia, 7–10
December 1987. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. pp.
71–96.
van Soest .P.J. and J. B. Robesrtson, 1985. Analysis of forage and fibrous feeds. A laboratory manual
for animal science, 613. Department of animal sciences, Cornell university, Ithaca, NY, USA.
202 pp.
van Soest, P.J., J. B. Robertson and B. A. Lewis, 1991. Methods for dietary fibers, neutral detergent
fibers and non-starch polysaccharides in relation to animal nutrition. J. Dairy Science. 74.
35883–3597.
SPSS (Statistical Procedure for Social Science). 2003. (Version 12). INC. Chicago, USA.
SRM (Society for Range Management). 1999. A glossary of terms used in range management.
Society for Range Management. Denver, Colorado. 20 pp.
Tainton, N. M., 1981. The Assessment of Veld Condition. In: N.M. Tainton (ed),Veld and pasture
management in South Africa. Shutter and shooter Ltd: Pietermaritzburg, South Africa. pp.
46–55.
Tainton, N.M., 1999. Veld management in South Africa. In: N.M. Tainton (ed). University of Natal
Press, Pietermarizburg, South Africa. 532p.
Tesfaye Mengistie. 2008. Characterization of cattle milk and meat production, processing and
marketing system in Metema District. MSc Thesis. School of Graduate Studies, Hawasa
University, Ethiopia, 200 pp.
Teshome Abate, 2006. Traditional utilization practices and condition assessment of the rangelands
in Rayitu district of Bale zone, Ethiopia. MSc Thesis. School of Graduate Studies, Haramaya
University, Ethiopia. 129 pp.

62

WorkingPaper_25.indd 62 1/10/2011 10:54:30 AM

 van der Westhuizen, HC van Rensburg WLJ, Snyman HA 1999. The quantification of rangeland
condition in a semi-arid grassland of Southern Africa. African Journal of Range and Forage
Science 16:9–21.
van der Westhuizen HC, Snyman HA, van Rensburg WLJ and Potgieter JHJ 2001. The quantification
of grazing capacity from grazing and production values for species in the semi-arid grassland
biome of Southern Africa. African Journal of Range and Forage science, 18:43–52.

63

WorkingPaper_25.indd 63 1/10/2011 10:54:30 AM

 64
Appendices

WorkingPaper_25.indd 64
Appendix Table 1. Criteria for the scoring of the different factors determining range conditions (Herbaceous
vegetation and soil condition)
Score Grasses spp composition Basal cover Number of seedlings Litter cover Soil erosion Soil compaction Age distribution
> 12%, no bare > 12%, no bare
10 91–100% decreaser
areas areas
9 81–90% decreaser – –
>9% evenly >9% evenly
8 71–80% decreaser
distributed distributed
>9% occasional >9% occasional
7 61–70% decreaser
bare spots bare spots
>6% evenly >6% evenly
6 51–60% decreaser
distributed distributed
>4 seedlings on No sign of soil
5 41–50% decreaser >6%, bare spots >6%, bare spots No compaction Young, medium ,old
A4 paper erosion
10–40% decreaser and >3%, mainly 4 seedlings on >3%, mainly Slight sand
4 Isolated capping Two size categories
>30% increaser perennials A4 paper perennials mulch
10-40% decreaser and >3%, mainly 3 seedlings on >3%, mainly an- Weak-side
3 >50% capping Only old
<30% increaser annuals A4 paper nuals pedestals
< 10% decreaser and > 2 seedlings on Steep-sided
2 1–3% 1–3% >75% capping Only medium
50% increaser A4 paper pedestals
<10% decreaser and 1 seedling on Almost 100%
1 <1% <1% Pavement Only young
<50% increaser. A4 paper capping
0 0% No seedlings 0% Gullies
Source: Baars et al. (1997).

1/10/2011 10:54:30 AM
 Appendix Table 2. Criteria for the scoring of the different factors determining range conditions

WorkingPaper_25.indd 65
Parameters Value Total point Descriptions
Hedging 3 5 Highly palatable and palatable shrubs share dominance
Most hedge able plants are lightly to moderately hedged
Few or no decadent plants
2 3 Palatable plants dominant. Hedgeable plants moderately to
Heavily hedged
Some shrubs decadent due to hedging
1 2 Palatable and less palatable plants dominant
Hedge able plants heavily to very heavily hedged
Considerable numbers of decadents’ shrubs present
Some may be dead due to hedging
0 1 Less palatable and unpalatable shrubs dominant
Some normally unhedgeable shrubs are hedged
Hedge able shrubs very heavily hedged the crowns often reduced to
Canopy cover 3 5 >45% cover
2 4 36–45% cover
1.5 3 26–35%
1 2 15–25%
0 1 <15% cover
Density 5 0–1000/ha
4 >1000–2000/ha
3 >2000–3000/ha
2 >3000–4000/ha,
1 >4000/ha
Source: Kuchar (1995).

65

1/10/2011 10:54:30 AM
 66
Appendix Table 3. Common and/or dominant woody species and their percentage composition in the farming
system of different grazing areas

WorkingPaper_25.indd 66
CLFS SLFS
Woody species Category
CG RS EN CG RS EN
G. ternifolia LD C – P – – –
A. polyacantha DS D D C C D
T. laxiflora C P C P C
A. tortilis DS C D – C C C
A. seyal DS D C D D C C
D. cinerea DS D – C D P
A. leiocarpus DS C C D C P D
P. lucens HD C P D C P D
F. sycomorus DS C P – P P
C. collinum LD P – – C P C
X. Americana LD C – P – –
S. kunthianum DS C P – – P
B. aegyptiaca DS C P C C C C
G. villosa LD – – – – C
Z. spina-christi HD C C D C C D
P. toningii DS P P P
F. virosa LD C – – C
B. papyrifera UD C P C C C D
A. senegal DS D P C – P C
C. mole LD – C C
CLFS: cotton livestock farming system; SLFS: sesame livestock farming system; CG: communal grazing; RS: road side; EN: enclosure; P: Present (<10% density); C:
common (>10% and <20% density); D: dominant (>20% density).

1/10/2011 10:54:31 AM
 Appendix Table 4. Herbaceous species identified in different farming system and grazing types of the district

WorkingPaper_25.indd 67
Botanical name Local name Category CLFS SLFS
Grasses: CG RS EN CG RS EN
Eriochloa fatamensis (Hochst. and Steud.) Metruqe LD P C C P
P. sphacelatum (Nees) Jingra DS C C C D C C
Urochloa cf. brachyura (Hack.) Stapf Metruq LD P P
T. triandra Chechewa HD P D P P
C. ciliaris Zemen DS D C C C
E. floccifolia Ahya dagusa LD D C D C
S. pumila (Poir.) R. and S. Dimemo DS C C C C
C. dyctlon Serdo HD P P C P
[Link] (Hochst.) Benth. Ejargew LD C C C
B. lata (Schumach.) C. E. Hub Mashilo DS C P C
P. coloratum L. kok sar HD D D
H. rufa Topas HD D
S. pyramidialis NA DS C
Cyprrus Spp. Gicha LD C
Legumes:
I. spicata Forsk. Kola maget DS C C C
A. quartinianus A. Rich Adoye DS C C C D
C. mimosoides (L) Greene) Koina UD P C p
------------------- Yeahiya abish UD C D D D
V. membranacea Yeayit guaya DS C P C C
D. dichotomum (Klein ex W) Aborida P D
Sedges and others:
C. barbata Mech mesel UD P P C
E. indica Na UD C C C D
S. sphaerostima wuha ankur UD C P C
H. schulli Amakela LD P

67

1/10/2011 10:54:31 AM
 68
Botanical name Local name Category CLFS SLFS
K. foetidissima Hareg ressa UD P P

WorkingPaper_25.indd 68
C. subula Yekola mech UD D D C C P
H. articulatus Hoccht. Abo hareg HD P D C
B. setigera (Sch. Bip.) Sherff Yebereha adey UD C C C C
Phyllanthus rotundifolius Willd. UD p p
C. trilocularis L Amirra DS p
H. vitifolius L. Gimel waika DS p P p
Z. elegans Jaquin. Adey abeba UD C C C C

1/10/2011 10:54:31 AM
 Appendix Table 5. Correlation matrix among variables studied in the communal grazing areas of CLFS

WorkingPaper_25.indd 69
fs spc bc lc se sc ag sed wds ccs heg tor wd cc
Fs 1
Spc 0.767** 1
Bc 0.758** 0.540** 1
C 0.526** 0.383 0.548** 1
Se 0.571** 0.315 0.744** 0.576** 1
Sc 0.388 0.387 0.335 0.126 0.135 1
Ag 0.462* 0.575** 0.243 0.017 –0.14 0.132 1
Sed 0.234 –0.023 0.179 0.338 0.118 0.228 –0.087 1
Wds –0.203 –0.323 –0.056 –0.314 0.491 –0.691 –0.240 –0.612 1
Ccs 0.664 0.728* 0.312 –0.103 0.121 0.188 0.383 –0.046 0.172 1
Heg 0.887** 0.737** 0.856** 0.383 0.353 0.555 0.716* 0.118 –0.058 0.616 1
Tor 0.666** 0.593** 590** 0.468* 0.497 0.383 0.196 0.345 –0.169 0.649 0.902** 1
Wd 0.612 0.859** 0.490 0.511 0.033 0.807* 0.515 0.219 –0.611 0.410 0.489 0.731* 1
Cc 0.507 0.351 0.303 0.254 0.374 –0.180 0.088 –0.196 0.237 0.576 0.304 0.417 0.170 1
** Correlation is significant at 0.01 levels;* Correlation is significant at 0.05 levels.
Fs: farming system; Spc: species composition; Bc: basal cover; Lc: litter cover; Se: soil erosion; Sc: soil compaction; Ag: age distribution; Sed: seedling count; Wds:
woody density score; Ccs: canopy cover score; Heg: hedging effect; Tor: total range condition score; Wd: woody density; Cc: canopy cover.

69

1/10/2011 10:54:31 AM
 70
Appendix Table 6. Correlation matrixes among variables studied in the road side grazing areas of CLFS

WorkingPaper_25.indd 70
fs spc bc lc se sc ag sed wds ccs heg tor wd cc
Fs 1
Spc 0.707** 1
Bc 0.497* 0.226 1
Lc 0.406* 0.288 0.040 1
Se –0.073 –0.058 –0.057 0.321 1
Sc 0.080 0.108 –0.240 0.021 0.014 1
Ag 0.132 0.138 –0.070 0.507** 0.128 0.197 1
Sed 0.330 0.060 0.426* 0.137 –0.070 –0.154 –0.248 1
Wds 0.460 0.246 0.127 –0.038 0.060 0.185 –0.183 0.395 1
Ccs –0.220 –0.196 0.488 0.025 0.017 0.334 –0.201 0.141 –0.003 1
Heg 0.482 0.306 –0.083 –0.073 0.119 0.030 –0.200 0.198 0.792* –0.275 1
Tor 0.462* 0.523** 0.222 0.418* 0.167 0.271 0.434* 0.235 0.380 0.457 0.199 1
wd 0.142 0.036 0.156 0.150 0.453 –0.295 0.149 0.240 0.392 –0.446 0.374 0.138 1
Cc –0.071 –0.331 0.472 0.445 0.331 0.111 0.061 0.753* –0.002 0.298 –0.310 0.469 –.203 1
**= correlation is significant at 0.01 levels,*= correlation is significant at 0.05 levels.
Fs: farming system; Spc: species composition; Bc: basal cover; Lc: litter cover; Se: soil erosion; Sc: soil compaction; Ag: age distribution; Sed: seedling count; Wds:
woody density score; Ccs: canopy cover score; Heg: hedging effect; Tor: total range condition score; Wd: woody density; Cc: canopy cover.

1/10/2011 10:54:32 AM
 Appendix Table 7. Correlation matrixes among variables studied in the enclosure areas of CLFS

WorkingPaper_25.indd 71
Fs spc bc lc se sc ag sed wds ccs heg tor wd
Fs 1
Spc 0.339 1
Bc 0.439 0.595** 1
Lc 0.302 –0.229 0.027 1
Se 0.202 0.342 0.089 –0.294 1
Sc 0.013 0.411 –0.050 –0.034 0.197 1
Ag –0.258 –0.494* –0.323 0.235 –0.548* –0.213 1
Sed –0.146 –0.347 –0.066 0.467* –0.115 –0.261 0.214 1
Wds –0.968** –0.289 0.009 –0.639 –0.045 –0.069 –0.163 –0.362 1
Ccs –449 –219 0.747 –0.575 –0.349 –0.170 –0.265 0.180 0.509 1
Heg –425 0.509 –0.071 –0.702 0.730 0.697 0.041 –0.886* 0.360 0.304 1
Tor –0.241 0.198 –0.022 –0.621 0.250 0.155 –0.193 0.105 0.896* 0.360 0.304 1
wd 0.043 –0.405 –0.754 0.631 –0.399 –0.354 0.885* 0.459 –0. 247 –0.603 –0.185 –0.107 1
**= correlation is significant at 0.01 levels;*= correlation is significant at 0.05 levels.
Fs: farming system; Spc: species composition; Bc: basal cover; Lc: litter cover; Se: soil erosion; Sc: soil compaction; Ag: age distribution; Sed: seedling count; Wds:
woody density score; Ccs: canopy cover score; Heg: hedging effect; Tor: total range condition score; Wd: woody density; Cc: canopy cover.

71

1/10/2011 10:54:32 AM
 72
Appendix Table 8. Correlation matrix among variables studied in the grazing areas of CLFS

WorkingPaper_25.indd 72
spc bc lc se sc ag sed wds ccs heg tor wd
Grt 1
Spc 0.665** 1
Bc 0.611** 0.671** 1
Lc 0.583** 0.697** 0.714** 1
Se 0.642** 0.644** 0.582** 0.586** 1
Sc 0.592** 0.600** 0.626** 0.386** 0.521** 1
Ag 0.579** 0.689** 0.802** 0.654** 0.620** 0.648** 1
Sed 0.790** 0.720** 0.477 0.715* 0.820** 0.701* 0.669* 1
Wds 0.124 0.048 –0.087 –0.235 –0.056 0.092 –0.139 0.147 1
Ccs 769** 0.693** 0.426 0.788** 0.866** 0.503 0.669* 0.902** –0.159 1
Heg 0.757** 0.790** 0.747** 0.764** 0.723** 0.660** 0.801** 0.878** 0.043 0.826** 1
Tor –0.281 –0.140 0.112 –0.068 –0.406 –0.096 –0.056 –0.540 0.151 –0.309 –0.233 1
wd 0.214 –0.080 0.137 –0.078 –0.268 –0.216 0.043 –0.490 0.522 –0.299 –0.137 0.657*
**= correlation is significant at 0.01 levels; *= correlation is significant at 0.05 levels.
Fs: farming system; Spc: species composition; Bc: basal cover; Lc: litter cover; Se: soil erosion; Sc: soil compaction; Ag: age distribution; Sed: seedling count; Wds:
woody density score; Ccs: canopy cover score; Heg: hedging effect; Tor: total range condition score; Wd: woody density; Cc: canopy cover.

1/10/2011 10:54:32 AM
 Appendix Table 9. Correlation matrixes among variables studied in the grazing areas of SLFS

WorkingPaper_25.indd 73
Spc bc lc se sc ag sed wds ccs heg tor wd
Spc 1
Bc 0.631** 1
Lc 0.718** 0.718** 1
Se 0.652** 0.615** 0.728** 1
Sc 0.833** 0.602** 0.807** 0.641** 1
Ag 0.689** 0.0.470** 0.734** 0.617** 0.732** 1
Sed 0.568** 0.784** 0.714** 0.458** 0.699** 0.437** 1
Wds 0.247 0.484 0.516 0.545 0.291 0.052 0.395 1
Ccs 0.143 0.219 –0.019 0.182 0.049 0.125 0.023 –0.624* 1
Heg 0.633* 0.574 0.563 0.718* 0.572 0.401 0.582 0.556 –0.240 1
Tor 0.846** 0.634** 0.811** 0.672** 0.819** 0.685** 0.663** 0.307 0.336 0.591 1
wd 0.162 –0.114 –0.147 –0.054 0.038 0.124 –0.168 –0.835** 0.828** –0.271 0.117 1
**= correlation is significant at 0.01 levels,*= correlation is significant at 0.05 levels.
Fs: farming system; Spc: species composition; Bc: basal cover; Lc: litter cover; Se: soil erosion; Sc: soil compaction; Ag: age distribution; Sed: seedling count; Wds:
woody density score; Ccs: canopy cover score; Heg: hedging effect; Tor: total range condition score; Wd: woody density; Cc: canopy cover.

73

1/10/2011 10:54:32 AM
 74
Appendix Table 10. Correlation matrix among variables studied and their interactions in range condition
assessment in Metema

WorkingPaper_25.indd 74
Fs Grt spc bc lc se sc ag sed wds ccs heg tor
FS 1
Grt 0.078 1
Spc 0.509** 0.345** 1
Bc 0.443** 0.473** 0.726** 1
Lc 0.306** 0.451** 0.690** 0.727** 1
Se 0.247* 0.323** 0.630** 0.677** 0.739** 1
Sc 0.192 0.454** 0.704** 0.634** 0.703** 0.603** 1
Ag 0.157 0.328** 0.615** 0.549** 0.682** 0.513** 0.628** 1
Sed 0.207 0.576** 0.588** 0.733** 0.772** 0.577** 0.669** 0.571** 1
Wds –0.086 0.799** 0.418 0.556** 0.406 0.590** 0.520* 0.411 0.523* 1
Ccs 0.117 –0.678** 0.173 0.169 0.005 0.041 0.040 0.060 –0.024 –0.359 1
Heg 0.285 0.536* 0.719** 0.677** 0.529* 0.702** 0.734** 0.515* 0.633** 0.737** –0.142 1
Tor 0.282* 0.440** 0.797** 0.745** 0.732** 0.773** 0.679** 0.756** 0.634** 0.205 0.770** 1
**= correlation is significant at 0.01 levels’;*= correlation is significant at 0.05 levels.
Fs: farming system; Spc: species composition; Bc: basal cover; Lc: litter cover; Se: soil erosion; Sc: soil compaction; Ag: age distribution; Sed: seedling count; Wds:
woody density score; Ccs: canopy cover score; Heg: hedging effect; Tor: total range condition score; Wd: woody density; Cc: canopy cover.

1/10/2011 10:54:33 AM
 Source: IPMS (2005).
Appendix Figure 1. Metema district farming systems.

Source: IPMS (2005).

Appendix Figure 2. Land use and land cover of Metema district.

75

WorkingPaper_25.indd 75 1/10/2011 10:54:35 AM

 Appendix Figure 3. The view of communal grazing lands at the dry season prior to flash burning.

Appendix Figure 4. The view of communal grazing lands at peak dry season.

76

WorkingPaper_25.indd 76 1/10/2011 10:54:38 AM

 Appendix Figure 5. Partial view of communal grazing lands in the sesame-based farming system.

Appendix Figure 6. Hay storage in open field.

77

WorkingPaper_25.indd 77 1/10/2011 10:54:40 AM

 Appendix Figure 7. P. spheslatum grass dominated communal grazing land at SLFS.

Appendix Figure 8. Part of an enclosure site in the sesame–livestock farming system (Agam wuha).

78

WorkingPaper_25.indd 78 1/10/2011 10:54:43 AM

 Appendix Figure 9. Communal grazing land after flash burning.

Appendix Figure 10. H. rufa grass dominated wood land (Guange river side areas).

79

WorkingPaper_25.indd 79 1/10/2011 10:54:47 AM

 [Link]
[Link] Working Paper No. 25

Rangeland condition and feed

resources in Metema District,
North Gondar Zone, Amhara
Region, Ethiopia

Canadian International Agence canadienne de

Development Agency développement international



Federal Democratic Republic of Ethiopia
MINISTRY OF AGRICULTURE AND
RURAL DEVELOPMENT