Characterization of Sheep Production, Feeding and Marketing Systems in Arba Minch Zuria Woreda of Gamo Gofa Zone, Southern Ethiopia

Home , Arba Minch, Arba Minch Zuria, Chencha, Chencha (woreda), Dita (woreda)

CHARACTERIZATION OF SHEEP PRODUCTION, FEEDING AND MARKETING SYSTEMS IN ARBA MINCH ZURIA WOREDA OF GAMO GOFA ZONE, SOUTHERN ETHIOPIA

MSc. THESIS

ESATU BEKELE BAYDE

NOVEMBER, 2017 ARBA MINCH, ETHIOPIA

CHARACTERIZATION OF SHEEP PRODUCTION, FEEDING AND MARKETING SYSTEMS IN ARBA MINCH ZURIA WOREDA OF GAMO GOFA ZONE, SOUTHERN ETHIOPIA

A THESIS SUBMITTED TO THE DEPARTMENT OF ANIMAL AND RANGE SCINECES, COLLEGE OF AGRICULTURAL SCIENCES, SCHOOL OF GRADUATE STUDIES, ARBA MINCH UNIVERSITY IN PARTIALFULFILLMENTOF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN ANIMAL PRODUCTION

BY ESATU BEKELE BAYDE

Major advisor: Yisehak Kechero (PhD, Associate Professor) Co-advisor: Chencha Chebo (MSc, Assistant Professor)

NOVEMBER, 2017 ARBA MINCH, ETHIOPIA

DEDICATION

This thesis is dedicated to my beloved father Bekele Bayde who passed away and I deeply wish that God might give him peaceful rest forever.

III

DECLARATION

I, the undersigned, hereby declare that the thesis entitled "Characterization of sheep production, feeding and marketing systems in Arba Minch Zuria woreda of Gamo Gofa zone, Southern Ethiopia" is the outcome of my own work and all sources of materials used for this thesis have been accordingly acknowledged. This thesis has been submitted in partial fulfillments of the requirements for the degree of Master of Science in Animal Production at Arba Minch University and is deposited at the University Library to be available to borrowers under rules of the library. I truly declare that this thesis is not submitted to any other institution anywhere for the award of any academic degree, diploma or certificate. I concede copy right of the thesis in favor of the Arba Minch University College of Agriculture as well as Animal Sciences department.

Name: Esatu Bekele Bayde Signature:

Place: Arba Minch University, College of Agriculture. Date of Submission: October 2017

BIOGRAPHICAL SKETCH The author Esatu Bekele Bayde was born on October 5, 1986 G.C. in Dita woreda, Done kebele Gamo Gofa Zone, Southern Nation Nationalities people Region. He attended his elementary education at Done Awassa in 1996 -2001 G.C.; he started his junior secondary education at Zada Junior secondary school (Grade 7) in 2002 G.C.; at Arba Minch Chamo junior secondary School (Grade 8 -9) in 2003-2004 G.C. and at Arba Minch compressive high school (Grade 10 - 12) in 2005 – 2007 G.C. Then he joined Arba Minch University in 2008 and graduated with BSC in Animal science on 10th July, 2010 G.C. After graduation he was employed by the Dita Woreda of marketing and cooperative office in 13 July 2010 – June 30 2011 G.C., at Arba Minch Zuria woreda Agriculture office Animal Production and forage development expert 1st July 2011 – 30 March 2013 G.C. and South West Zone –Ethiopian Kale Heywet Church (SWZ-EKHC) Office community livelihood and children based project at Arba Minch Zuria Woreda April 1st 2013 - 2016 G.C. until he joined Arba Minch University, school of graduate studies for the degree master of science in Animal production 2012016 G.C.

SCHOOL OF GRADUATE STUDIES ARBA MINCH UNIVERSITY ADVISORS’ THESIS SUBMISSION APPROVAL SHEET

This is to certify that the thesis entitled “Characterization of sheep production, feeding and marketing systems in Arba Minch Zuria woreda of Gamo Gofa zone, Southern Ethiopia” has submitted in partial fulfillment of the requirements for the degree of Master of Science in Animal Production, the Graduate Program of the Department of Animal Sciences, and has been carried out by Mr. Esatu Bekele Bayde ID.No RMSc/223/08 under my/our supervision. Therefore I/we recommend that the student has fulfilled the requirements and hence hereby can submit the thesis to the department for defense.

Yisehak Kechero (PhD, Associate professor) ______Name of Principal advisor Signature Date

Mr. Chencha Chebo (MSc, Assistant professor) ______Name of Co-advisor Signature Date

VII

EXAMINERS’ APPROVAL PAGE

Approved by: We the examiners’ board approve that this thesis has passed through the defense and review process

Chairperson Signature Date

External Examiner Signature Date

Internal Examiner Signature Date

Yisehak Kechero (PhD, Associate professor) Principal advisor Signature Date

(DGC/SGC) Signature Date

Signature and Stamp of the Department/School

ACKNOWLEDGMENT

First and for most, I thank my almighty God for his will to help, protect, provide and support me to reach at this stage; without His assistance nothing would have happened.

My special thanks and heartfelt gratitude extends to my lovely wife Loza Awoke for her patience, support and encouragement, her special prayer helped me to have strength and endurance in my work. You are a source of my pleasure and well-being.

I would like to express my deepest and sincere appreciation to my respected major advisor, Yisehak Kechero (PhD, Associate Professor), for his unreserved advice, guidance and his excellent teaching and constructive criticism and comments from the initial conception to the end of this work are highly appreciated. I am also thankful to my co-advisor Mr. Chencha Chebo (MSc, Assistant professor), who helped me from the time of proposal development to the time of finalizing my thesis work.

I wish to express my deepest love and gratitude to my brother Mr. Belay Bekele his follow up, sustained support and constructive comments and also Haile Petros, Sintayehu Tomas, Elias Zale and Frehiwot Tarekegn for their sustained support, collected data and encouragement.

My appreciations and thanks also go to my mother organization “Ethiopian kale Hiywot Church South West Zone” for giving me this MSc. study fellowship and providing me all the necessary supports including salary and financial support.

I also wish to express my thanks to Arba Minch University Animal and Range science department thought me and all developmental agents (DA), kebeles administrations and also farmers who responded to my numerous questions with patience.

VIII

LIST OF ABBREVIATIONS

ADG Average Daily Gain AEZ Agro Ecological Zones ANOVA Analysis of Variance AFL Age at First Lambing CSA Central Statistical Authority DA Development Agent DM Dry matter EKHC-SWZ Ethiopian Kale Hiywet Church – South West Zone EPA Ethiopia Privatization Agency EIAR Ethiopian Institute of Agriculture Research FAO Food and Agriculture Organization of the United Nations FGD Focus Group Discussion GO Governmental Organization HSD Honestly Significant Difference HH Household HL Highland ILCA International Livestock Center for Africa LI Lambing Interval LL Lowland LS Litter Size ML Midland MOA Ministry of Agriculture MT Metric Tone NGO Non-Governmental Organization TLU Tropical livestock unit PI Parturition Interval PRA Proportionality of Respondents Across

VIII

LIST OF ABBREVIATIONS (Continued)

RDC Respiratory Disease Complex SE Standard Error SD Standard Deviation SNNPR Southern Nation Nationality People Region SPSS Statistical Procedures for Social Science

TABLE OF CONTENTS CONTENTS PAGE DEDICATION ...... III DECLARATION ...... IV BIOGRAPHICAL SKETCH ...... VI ACKNOWLEDGMENT ...... VIII LIST OF ABBREVIATIONS ...... VIII LIST OF ABBREVIATIONS (Continued) ...... IX TABLE OF CONTENTS ...... VI LIST OF TABLE ...... XV LIST OF FIGURE ...... XVI ABSTRACT ...... XVII 1. INTRODUCTION ...... 1 1.1 General Objective ...... 4 1.1.1 Specific objective ...... 4 2. LITERATUREREVIEW ...... 5 2.1. Major Sheep Production Systems in Ethiopia ...... 5 2.2. Socio-Economic Importance of Sheep ...... 5 2.3. Indigenous Knowledge and Community/Village-based Breeding Strategies ...... 6 2.4. Flock structure ...... 7 2.5. Off take rate ...... 7 2.6. Reproductive Performances ...... 7 2.6.1. Age at sexual maturity ...... 8 2.6.2. Age at first lambing ...... 8 2.6.3. Lambing interval ...... 9 2.6.4. Lambing pattern (occurrence of lambing)...... 10 2.6.5. Litter size (LS) ...... 11 2.6.6. Seasonality in breeding ...... 12 2.7. Feed resource and feeding system ...... 12 2.7.1 Feed resource ...... 12 2.7.2. Feed and supplementation ...... 13 2.8. Sheep Marketing ...... 14 2.9. Sheep Production Constraints and Opportunities ...... 15

Table of content (continued) 2.9.1 Production constraints of sheep ...... 15 2.9.1.1 Feed shortage ...... 15 2.9.1.2 Disease and Mortality of Sheep ...... 15 2.9.1.3 Market access and information ...... 16 2.10. Sheep production opportunities...... 16 3. MATERIALS AND METHODS ...... 16 3.1 Description of the Study Area ...... 16 3.2. Sampling technique and sample size ...... 17 3.3. Data sources and Data Collection Tools ...... 18 3.4.1. Dry matter yield of pasture ...... 18 3.4.2. Dry matter yield of crop residue, fallow land ad after math grazing ...... 19 3.4.3. Quantity of trees and shrubs ...... 19 3.4.4. Estimating available concentrates ...... 19 3.4.5. Estimation of balance between feed supply and feed requirement ...... 19 3.5. Statistical Analysis ...... 20 4. RESULTS AND DISCUSSION ...... 21 4.1. Socio-economic characteristics of the respondents ...... 21 4.1.1. Household characteristics ...... 21 4.1.2. Landholding and land use pattern of the households ...... 22 4.1.3. Farming activities in the study area ...... 24 4.1.4. Livestock holding and composition ...... 24 4.1.5. Spearman’s correlation coefficient between altitudes, farm land size, family size and performance parameters of the sheep ...... 27 4.1.7. Purposes of keeping sheep...... 30 4.1.8. Labor division in sheep management ...... 31 4.2. Sheep breeding and reproductive performance ...... 32 4.2.1. Sheep breeding management ...... 32 4.2.2. Breed types and preference to sheep ...... 33 4.2.3. Reproductive performances of sheep ...... 34 4.2.3.1. Age at puberty ...... 34 4.2.3.2. Age at first parturition ...... 35 4.2.3.3. Parturition interval...... 35 4.2.3.4. Litter size ...... 36

VII

Table of content (continued) 4.2.3.5. Slaughter age ...... 36 4.2.3.6. Average life span and average lamb crop of the sheep ...... 36 4.2.4. Management practices ...... 37 4.2.4.1. Housing ...... 37 4.2.4.2. Lamb rearing, Selection of castration and fatting ...... 39 4.2.4.3. Purchase, disposal and culling practices of sheep ...... 42 4.2.4.4. Consumption of sheep ...... 44 4.2.4.5. Sheep disease and health managements ...... 44 4.2.5. Feed resources and feeding system of sheep ...... 47 4.2.5.1 Feed resource availability ...... 47 4.2.5.2. Common supplement offered to sheep ...... 48 4.2.5.3. Feed seasonal availability ...... 49 4.2.6. Feeding system (grazing system)...... 49 4.2.6.1. Reason of tethering ...... 50 4.2.7. Estimation of Feed Supply ...... 51 4.2.7.1. Dry matter yield of pasture...... 51 4.2.7.2. Crop residues production ...... 52 4.2.7.3. Crop aftermath ...... 53 4.2.7.4. Forest land dry matter production ...... 54 4.2.7.5. Indigenous fodder trees and shrubs in the study area ...... 54 4.2.7.6. Non – conventional feeds ...... 55 4.2.8. Dry matter requirement of livestock in the study wereda ...... 55 4.2.9. Estimated annual feed balance ...... 56 4.2.10. Water resource and watering practices ...... 57 4.2.11. Sheep marketing system ...... 58 4.2.11.1. Reasons of selling, buying and market places ...... 58 4.2.11.2. Sold classes of sheep for urgent income need ...... 59 4.2.11.3. Season of marketing sheep ...... 60 4.2.11.4. Mode of price setting...... 61 4.2.11.5. Marketing channels ...... 62 4.3. Sheep production constraints and opportunities ...... 62 4.3.1. Sheep production constraints ...... 62

VIII

Table of content (continued) 4.3.2. Feed and grazing land shortage; ...... 64 4.3.3. Opportunities of sheep production ...... 64 5. CONCLUSION AND RECOMMENDATION ...... 67 5.1. Conclusion...... 67 5.2. Recommendation ...... 68 6. REFERENCES ...... 70 7. ANNEX ...... 80 7.1.Questionnaires ...... 80

LIST OF TABLE Table 1; Age at first lambing of Ethiopian indigenous sheep breeds/types ...... 9 Table 2; Lambing Interval of Ethiopian indigenous sheep breeds/types ...... 10 Table 3; Litter size of Ethiopian indigenous sheep breeds/types ...... 11 Table 4; The brief description of the study area ...... 17 Table 5; Household characteristics of the respondents in the study wereda ...... 22 Table 6; Mean land holding and use patterns of the sampled households in study area ...... 23 Table 7; Livestock holding and composition ...... 25 Table 8; Trends of the sheep population of the respondents in the study area ...... 25 Table 9; Livestock population (in numbers and TLU) of Arba Minch zuria wereda ...... 26 Table 10; Correlation matrix between variables studied (N=138)...... 28 Table11; Purpose of keeping sheep ...... 31 Table 12; Household preferences to sheep ...... 34 Table 13; Reproductive parameters of sheep flock ...... 37 Table 14; Sheltering sheep during night, housing and management practices ...... 39 Table 15; Management practices of young, breeding and fattening flocks ...... 42 Table 16; Sheep flock dynamics ...... 43 Table17; Major sheep disease in the study area ...... 45 Table 18; Commonly available of feed resources ...... 48 Table 19; Common supplements offered to sheep ...... 49 Table 20; Grazing management practices of sheep in the study area ...... 50 Table 21; The possible reasons of sheep tethering reported by the respondents ...... 51 Table 22; Estimated quantity of feed DM obtained from different land use types of the respondents in the study area ...... 52 Table 23; Crop and straw yield production in the study area ...... 53 Table 24; Crop and straw yield production in Arba Minch zuria wereda ...... 53 Table 25; The estimated biomass yield of indigenous fodder trees and shrubs identified in the study area ...... 55 Table 26; Feed resources category and their DM supply according to regional differences ...... 57

List Table (continued) Table 27; Yearly differences in the balance between feed resource availability and dietary requirement ...... 57 Table 29; The reasons for sheep sale of the sampled respondents ...... 59 Table 30; Major constraints of sheep production in the study area ...... 63 Table 31; Reasons of feed shortage in the study area ...... 64 Table 32; Opportunities of sheep production in the study area ...... 66

XVI

LIST OF FIGURE

Figure 1; Local sheep, Dawuro/Bonga sheep and Woliata/Doyogena sheep (left to right) ...... 29 Figure 2; Flock structures of sheep populations in the study area ...... 30 Figure 3; Division of family labor for sheep management ...... 32 Figure 4; sheep culling reasons ...... 44 Figure 5; The photo image of gaso ...... 46 Figure 6; Classes of sheep sold for urgent income needs...... 60 Figure 7; Sheep market routes in Arba Minch Zuria woreda ...... 62 Figure 8; Sheep market channels in Arba Minch Zuria...... 62

XVI

CHARACTERIZATION OF SHEEP PRODUCTION, FEEDING AND MARKETING SYSTEMS IN ARBA MINCH ZURIA WOREDA OF GAMO GOFA ZONE, SOUTHERN ETHIOPIA By Esatu Bekele (Bsc), Yisehak Kechero (PhD, Associate professor), Chencha Chebo (MSc Assistant professor) ABSTRACT The study was carried to characterize sheep production, feeding and marketing systems as well as to identify constraints and opportunities of sheep production in Arba Minch Zuria Woreda of Gamo Gofa zone, southern Ethiopia. The woreda was selected based on its potential for sheep production. It was stratified in to three distinctive agro-ecologies, namely highland (HL), midland (ML) and lowland (LL). A total of 138 households were interviewed using structured questionnaire. Probability proportional to size sampling technique was followed to select respondents. A Focus group discussion composed of 12 households was conducted in each agro-ecology. Data were analyzed using statistical procedures for social science (SPSS) version 20. Statistical variations for categorical data were tested by means of cross tabs, with significant differences at P<0.05 while the descriptive statistics for the numerical data was subjected to one way analysis of variance (one-way ANOVA) using the general linear model procedure of SPSS. The survey revealed that the overall average, total land, crop land, and grazing land per households were 1.34, 0.903 and 0.228 hectares, respectively (P<0.05). The average sheep flock size holding per household was 10.03, 5.07 and 4.73 in HL, ML and LL, respectively. The overall purpose of sheep rearing across all AEZs was source of income, saving, meat production, sharing the risk and manure production which were ranked first to fifth in the given order. The major available feed was natural pasture and crop residues in wet and dry seasons. Feeding systems widely practiced in the all AEZs were tethering, stall feeding and free grazing system in dry and wet seasons. About 51.4 % of the respondents were grazing sheep with other livestock. About 62.8% and 100% of respondents had their own grazing land in HL and ML AEZs, respectively. The overall age at first lambing was 12.81±0.14 months lambing interval was 7.52±.08 months and litter size was 1.50±.08 lamb per sheep (P<0.05). A high lambing month was extended from April to May. Majority of respondents’ across the agro-ecologies said that the practice of selling sheep was at age of 6 months until1 year. Feed and grazing land shortage, diseases, and labor shortage were major constraints ranked first, second, third and fourth with an overall index value of 0.3, 0.26, 0.14 and 0.14 of sheep production, respectively. Opportunities of sheep production were immediate return 60.9%, 57.2% had low startup cost, and 55.8% got GO and NGO interventions. About 51.8% of respondents got a suitable market conditions that motivate them for sheep production in the area. Sheep in Arba Minch Zuria have

XVII appreciable reproductive performances and are coming important integral parts among livestock in this area. Introduction and use of best parties and alternative feed production technologies are, highly recommended. Further research is also recommended to identify and select the superior and economically important traits in sheep production within this eco types and quantitative aspects of marketing require providing complete marketing information. Keys words: Arba Minch Zuria, marketing, performance, production, sheep feeding

XVIII

1. INTRODUCTION

Ethiopia is believed to have the largest sheep population in Africa, which is estimated to be 30,697,942 million heads (CSA (2016/17). Sheep are highly adaptable to a broad range of environments owned by smallholder farmers as an integral part of the livestock sub-sector Zelalem (2016). Approximately 75% of the sheep are kept in smallholder mixed farms in the highland areas, which cover areas of over 1500 altitude and receive over 700 mm of annual rainfall, while the remaining 25% are found in the lowlands. Sheep are traditionally kept in smallholdings and are associated with the small-scale resource poor livestock keepers Zewdu (2008).

Indigenous sheep in Ethiopia have a multipurpose role for smallholder farmers as sources of income, meat, milk, skin, manure and coarse wool or long hairy fleece. They are also a means of risk avoidance during crop failure. Thus, increasing the current level of productivity of sheep is essential to meet the demands of the ever-increasing human population Mengesha and Tsega (2012). Sheep serve as living bank for their owners and serve as source of immediate cash need and insurance against crop failure especially where land productivity is low and unreliable due to erratic rainfall, severe erosion, frost, and water logging problems. According to Zewdu (2008), sheep also play an important role in cultural, social livelihoods and religious values for large and diverse human population.

Getahun (2008) reported four sheep production systems in Ethiopia; in annual crop-based systems (Northern, North-Western and central Ethiopia), in perennial crop-based systems (mainly southern and south-western highlands), in cattle-based systems (agro-pastoral and arid areas), and small ruminants dominated systems (pastoral and arid eastern and Northeastern areas).

Reproductive performances are important early indicators of adaptability and management adequacy Abegaz et al. (2002); Getahun (2008). The most important components of reproductive performance in sheep are age at first service (puberty), age at first lambing, lambing interval, litter size, slaughter age, ewe productive life and ewe life time lamb crop. Puberty in the ewe lamb is the point in which she first exhibits estrus. From the point of farmers view, puberty is the age at first service. Studies revealed that age at first mating for both sexes is not fixed and sheep are left to nature to reproduce Zewdu (2008).

The major feed resources for sheep in Ethiopia include grazing on communal natural pasture, private pastures, crop stubbles, fallow grazing, road side grazing, river side grazing, crop residues, browses, grains, improved forages, and non-conventional feeds including household food leftovers, weeds from crop fields, tillers from dense crop fields, fillers (crops intentionally planted on part of crop lands or around homestead to be used as feed) and traditional brewers grains (locally known as atella). Importance of the different feed resources varies depending on the production system, farmers’ livestock management practice and the production environment (e.g. availability of grazing land, climate) Solomon et al. (2010).

According to Solomon et al. (2010), in most production systems, agro-ecologies and geographic regions, extensive free grazing in communal grazing lands and stubble grazing are the most common practices of feeding sheep. It is estimated that natural pasture provides from 80–90%, and crop residues 10–15% of the total livestock feed intake in Ethiopia Alemayehu (2003). A relatively more intensive controlled feeding is practiced in the perennial crop–livestock systems and sub moist crop–livestock system. The form of controlled feeding includes tethered grazing on private lands and marginal lands, cut-and-carry feeding of grass, tillers, fillers and weeds Getahun (2008).

Improving marketing success of sheep or livestock producers provides incentives to adopt technological interventions that improve sheep productivity, which in turn improves marketing success. Access to local market is the most important economic determinant to adopt technologies Zelalem (2007) and choice of production enterprises. Market locations in primary and secondary markets are usually not fenced; there are no permanent animal routes and no feed and watering infrastructures. Yet, buyers and sellers are subjected to various service charges by the local authorities as well as other bodies Ayele et al. (2003). Nearly in all parts of the country, there is no regular market information on prices and supplies, nor formalized grades and standards of sheep and other livestock Kebede and Ray (1992); Ayele et al. (2003). As a result, there is excess supply of animals beyond demands in some seasons. The more mobile trader is better informed on market prices which combined with excess supply places the trader in a better position during price negotiation Tesfaye (2009).

Based on research finding of Kassahun (2000), locally available breeds of livestock are important economic resources since they are adapted to the existing production constraints such as feed shortages, prevalent diseases, lack of strong market options and limited extension service.

The productivity of indigenous breeds is low compared to temperate breeds, but their ability to survive and produce in the harsh and mostly unpredicted tropical environment is remarkable.

Currently contribution of the livestock subsector in Ethiopia is far below from its potential. This is because of socio economic and technical limitations like inadequate feed quality and quantity, diseases, poor breeding and genetic potentials, inadequate livestock policies, veterinary service, livestock extension services, marketing and infrastructure that affect the livestock potentials Addis (2015).

Characterizing the current production systems, feeding, breeding and marketing practices, and productivity level of the breed/type in their habitat are prerequisites to set up an improved management program at the smallholder and pastoral levels Kosgey et al. (2004). Arba Minch Zuria woreda, having distinctive agro-ecologies, is geographically located in Gamo Gofa Zone, Southern Nations, Nationalities and Peoples Regional State (SNNPRS). Although sheep husbandry is back bone of the small holder farmers in the study area, the production system, marketing, feed resources and feeding system is not yet characterized in this administrative zone. Besides these, the potentials and constraints of sheep production is not well studied in the area. Since, availing basic information is important to improve the small ruminant sector in general and sheep production system in particular with the focuses of improved management practices, it is important to carry out research in this regard to filling the information gap in the study area. Therefore, the present study is planned with the following general and specific objectives:

1.1 General Objective

This study is initiated to characterize production, feeding and marketing systems as well as to identify the constraints and opportunities of sheep production in Arba Minch zuria woreda, Gamo Gofa zone, southwestern Ethiopia.

1.1.1 Specific objective

1. To assess the socio-economic characteristics of the survey population and link some of these variables with performance parameters of sheep; 2. To assess the existing production, feeding and marketing systems of sheep breeds under prevailing management conditions; 3. To identify the major constraints and opportunities of sheep production in the study area.

2. LITERATUREREVIEW

2.1. Major Sheep Production Systems in Ethiopia

Based on input-output flow, Tibbo (2006) reported sheep production system of Ethiopia into two major categories and three different production systems. The first and the most common system is the traditional smallholder management system. Sheep is kept as an adjunct to other agricultural activities along with other livestock species. When closely examined, these two broad categories could be further classified as three major different production systems; highland mixed crop-livestock and pastoral and agro-pastoral production systems Tibbo (2006); Solomon et al. (2008) characterized by different production goals and priorities, management strategies and practices, and constraints. The traditional subsistence smallholder management system is the most common one in the country. But the parastatal production system represents a minor portion of the sheep production system in the country. The sheep production in the country is based mainly on indigenous breeds Markos (2006).

Based on the previous agricultural activity, Getahun (2008) reported four production system categories; small ruminant in annual crop-based systems (Northern, North-Western and central Ethiopia), small ruminant in perennial crop-based systems (mainly southern and south-western highlands), small ruminant in cattle-based systems (agro-pastoral and arid areas), and small ruminant dominated systems (pastoral and arid eastern and Northeastern areas).

According to the report of Fsahatsion et al. (2013), indicate the primary reason of sheep keeping by the farmers is for source of income generations through the sale of live animals with and the cash obtained might be used to buy clothing and food items, pay taxes, additional fertilizers to manures and household supplies (children schools). The second main reason of sheep keeping is for meat production with and the keeping of sheep production for manure and social and cultural function.

2.2. Socio-Economic Importance of Sheep

Ethiopian producers are careful about wasting their livestock investment just to meet a seasonal market demand providing high prices. On the other hand, farmers are forced to sale their sheep whenever crop failures; abnormal rainfall patterns and food gap happened in the year. At such situations, most farmers become victims of nature and loose or highly reduce their flocks. As a

5 result, the price of local sheep fluctuates over the year according to the timing of feasts and crop failures. As a result, price of sheep deteriorates radically since the supply highly exceeds the demand Seare (2007).

Sheep are of great importance as major sources of livelihood Kosgey et al. (2004) and contribute to the sustenance of landless, smallholder and marginal farmers especially to the poor in the rural areas throughout the developing countries. The ownership of small ruminants is regarded as a safe investment for the family as well as to gain social prestige within the community. They are sold to meet compelling family financial obligations or slaughtered for consumption at home or festivals. The small sizes of the animal, their high reproductive efficiency and low initial investment make them suitable for rearing in the small holder farmer Ademosum (1994). In Ethiopia annual national mutton production is 78 thousand MT, largely because of the high average off take rates estimated at about 35% for sheep Workneh (2006). Sheep contribute some 20.9% of the total ruminant livestock meat output or about 13.9% of the total domestic meat production, with a live animal and chilled meat export surpluses. Per capita consumption of sheep (kg/person per year) in Ethiopia is 2.1 kg Zewdu (2008).

2.3. Indigenous Knowledge and Community/Village-based Breeding Strategies

Sheep are kept for multiple objectives with source of income being rated highest. Breeding ram was generally uncontrolled. Size, color and long and fatty tail type were the most frequently reported traits in selecting breeding rams; whereas size, color, tail size and twining rate were mentioned as traits given due emphasis in choosing future breeding ewes as a example Bonga breeding ewes. Farmers consider size, color, pedigree, age at first lambing, lambing interval, mothering ability and twining rate were selecting for breeding female.

Adaptive traits such as tolerance to diseases and feed shortage were given low emphasis in selecting replacement stocks in haighland sheep breeds. Disease, feed and grazing land shortage identified several constraints such as early disposal of breeding stocks, small flock sizes with only a few breeding males, uncontrolled mating, communal grazing in wet season and free roaming during dry season that make controlled breeding or mating very difficult. Traditional breeding practices such as sharing of breeding rams should be further strengthened. This in turn requires the full participation and long-term commitment of sheep keepers and other livestock development actors. To realize full benefits of a breeding strategy; approaches should be holistic

6 with concurrent improvement in the non-genetic factors (disease resistance and feed efficiency) as well Edea (2012).

2.4. Flock structure

Flock structure is defined as the proportion (in terms of head) of the flock of sheep, which is formed by different age and sex classes of animals. Flock composition in terms of age and sex classes has been taken as an indicator of the management objectives for the owner and the production (reproductive, mortality and off -take rate) of the flock Ayalew et al. (2002). There were higher numbers of ewes (88.7%) than rams (11.3%). The flock structure of lamb and ewe was higher in the dega than weyna dega. This might be attributed to the prevalent practice of keeping ewe for breeding purpose which accounted the greater portion of the newly born animals but rams are either castrated or sold when they reach market age Mengistie et al. (2010) and Fsahastion et al. (2013).

2.5. Off take rate

The indigenous sheep are year round breeders and mating is not controlled. Annual lambing rates are only 1.2. Annual off-take rate for sheep is estimated at 33% EPA (2002) with an average carcass weight of about 10 kg, which is the second lowest amongst sub-Saharan African countries FAO (2004). Most of the off take is due to commercial purposes (sales for generation of income) and for household needs Getahun (2008). Similarly, Tsedeke (2007) reported the need of sheep to generate an immediate cash income. Belete (2009) also reported that sales accounted for 27.5% of the off-take for sheep. The same author indicated that slaughtering during festivals and other reasons accounted for 26.6% and 15.8% off-take for sheep and goats, respectively.

2.6. Reproductive Performances

Good reproductive performance is a prerequisite for any successful genetic improvement and it determines production efficiency. Study suggests that differences exist in reproductive performance between indigenous sheep breeds and their variation allow for the selection of suitable breeds for a given environment. The most important components of reproductive performance in sheep are age at first lambing, lambing interval, and litter size and ewe productive life. Puberty in the ewe lamb is the point in which she first exhibits estrus. From the

7 farmers’ point of view, puberty is the age at first service. Results revealed that age at first mating for both sexes is not fixed and sheep are left to nature to reproduce Zewdu (2008).

2.6.1. Age at sexual maturity

According to Zewdu (2008), ages at first service for Bonga and Horro breeds were 8.5 ± 2.5 and 7.2 ± 2.4 months for females and males, respectively. Horro female lambs reach age at first mating earlier (P<0.01) than Bonga ewe lambs. According to Amelmal (2011) age at sexual maturity (puberty) was 11.05±1.6, 10.88±1.7 and 9.5±1.4 months for males and 11.13±2.7, 10.8±1.9 and 9.5±1.4 months for females in Tocha, Mareka and Konta respectively. The sexual maturity (puberty) in local sheep in Illu Abba Bora and Gumuz female sheep was reported to be 5-8 and 7.21 •± 1.75 months respectively Dhaba (2013) and Solomon (2007). The findings by Tsedeke (2007) for age at puberty of local Alaba sheep were 6.7 and 6.9 months for male and female respectively. The sexual maturity of sheep in Ethiopia Showed variation among breeds / types, locations and differences or variation allowing for the selection of suitable breeds for a given environment.

2.6.2. Age at first lambing

Total life time production (life time lamb crop) can be increased by encouraging first lambing at an early age Amelmal (2011). Age at first lambing is based on breed, husbandry and management practices and has wide variation among African sheep. In most traditional systems, first lambing occurs at 450-540 days when ewe weights are 80-85 percent of mature size (Wilson (1986) and Poor nutrition, disease or parasitic burdens and genotype limit early growth and it can put obstacle for early maturity for giving first birth. Year and season of birth in which the ewe lamb born influence age at first lambing through their effect on feed supply and quality during different season Mukasa and Lahlou (1995).

Age at first parturition of 12.97 months for sheep reported in western Ethiopia Belete (2009) is slightly higher than the age at first lambing 12.43 months reported by Derib (2009). Samuel (2005) reported age at first parturition of 17.01months for sheep. In the Table 1 more of the Ethiopian indigenous sheep breeds or types of age at first lambing under different management conditions are differed. When breeding males are available in the flocks, age at first parturition is a good indicator of early sexual maturity in ewes Zelalem (2016).

Table 1; Age at first lambing of Ethiopian indigenous sheep breeds/types Breed/Type AFL(months) Source Breed/Type Breed/Type AFL(months) AFL(months) Source Source Gumuz 13.67 Solomon (2007) Menz 17.06 Niftalem (1990) Thin-tailed sheep 13.7 Mukasa-Mugerwa et al. (1986) Washera 15.46 Mengiste (2008) Blackhead Ogaden 23.56±3.63 Fikrte (2008) Bonga 14.9±3.1 Zewdu (2008) Horro 13.3±1.7 Zewdu (2008) Arsi-bale 12.7 Tsedeke( 2007) Local sheep in Alaba 12.7 Tsedeke (2007) Doyogena 14.6 Getahun (2008) Local sheep in AdaaLiban 17.07 Samuel (2005) Local sheep in Tocha 12.88±1.7 Amelmal (2011) Local sheep in Mareka 14.75±1.8 Amelmal (2011) Amelmal (2011) Local sheep in Konta 14.77±1.8 Local sheep in Illu Abba Bora 10 – 13 Dhaba (2013) Local sheep in Gamogofa Zone 12.4±0.28 Fsahatsion et al. (2013) Local sheep in Ada Barga and Ejere 14.29±0.08 Yadeta (2016)

2.6.3. Lambing interval

Parturition interval (Lambing interval) refers to the number of days between successive parturitions. Reproductive efficiency is related to the length of parturition interval; i.e. ewe with long lambing interval has lower reproductive efficiency. At least three times lambing is expected per two years under normal circumstances Girma (2008). To attain this, lambing interval should not exceed 8 months (245 days). There are reports on the possibility of attaining three parturitions from indigenous small ruminants in two years Getahun (2008) though parturition interval of the traditionally managed sheep was influenced by various factors including previous litter type, parity and lambing season Dibissa (2000).

Table 2; Lambing Interval of Ethiopian indigenous sheep breeds/types Breed/Type LI(months) Source Menz 8.5 Tesfaye (2008) Afar sheep 9 Tesfaye (2008) Local sheep around Dire Dawa 11.2-11.3 Aden (2003) Washera 9.16 Mengiste (2008) Blackhead Ogaden 10.46 Fikrte (2008) Bonga 8 Belete (2009) Local sheep in Gomma woredat 7.87-8.04 Belete (2009) Bonga 8.9±2.1 Zewdu (2008) Arsi-bale 12.7 Tsedeke (2007) Local sheep in Gamo Gofa 7.34±0.13 Fsahatsion et al. (2013) Local sheep in Alaba 9.19±0.08 Deribe (200 Local sheep in Tocha 11.62±3.8 Amelmal (2011) Local sheep in Mareka 10.33±4 Amelmal (2011) Local sheep in Konta 11.02±3.8 Amelmal (2011) Local sheep in Illu Abba Bora 9-12 Dhaba (2013) Local sheep in Ada Barga and 8.83±0.44 Yadeta (2016) Ejere 2.6.4. Lambing pattern (occurrence of lambing)

Long reproductive life span in tropical (unfavorable) condition is one of the adaptation traits of tropical livestock. According to Zewdu (2008) the average reproductive life span of Horro and Bonga ewes were 7.9 ± 3.1 years and 7.4 ±2.7 years, respectively. Long term reproductive performance (long living, high fertility, ability to produce more offspring) of dams should be given more importance in selection programs (Zewdu (2008).

According to Solomon (2007) in a circumstance that there is lack of comparative figures for Ethiopian breeds, quite long reproductive life span of Gumuz breed (8.5 years for ewes and 3.67 years for rams) was reported. The average reproductive life span of Tocha, Mareka and Konta local ewes were 9.17±1.70, 9.82±1.51and 9.28±1.62 years, respectively. The life time lamb crop is very important trait to improve sheep productivity and profitability. According to Zewdu (2008) on an average a Bonga and Horro ewe delivers 12.2 ± 1.80 and 15.3 ±4.3 lambs in her life time. Also, similar result was reported for Gumuz sheep (13.5 ± 1.76 lambs) in Metema areas Solomon (2007).

2.6.5. Litter size (LS)

Litter size is a combination of ovulation rate and embryo survival, number of lambs born per parturition. There is a positive relationship between litter size and age and litter size and parity Getahun (2008); Girma (2008). An increase in ewe weight (prior to mating) by 1 kg over the average of the population results is in an increase of about 3.8% in litter size. Litter size varies between 1.08 and 1.75 with average of 1.38 for tropical breeds Girma (2008). Twin born ewes tended to produce more and heavier lambs than did those born single. Litter size is significantly affected by year of lambing, parity and weight of ewes at mating Abegaz et al. (2002); Gemeda et al. (2002a); Berhanu and Aynalem (2009).

Level of nutrition has effect on litter size in that, poor nutrition during service period lead to reduced ovulation rates and increase embryonic mortality and consequently decrease litter size Zelalem (2016). According to Zewdu (2008) a twining rate of 39.9 % or litter size of 1.40 and 36 % or litter size of 1.36 were obtained for Bonga and Horro sheep breeds, respectively Litter size is influenced by genotype, parity, season, and ewe body weight at mating Zelalem (2016) and management system is also a major source of variation in litter size. Table 5 represents litter size of Ethiopian indigenous sheep types.

Table 3; Litter size of Ethiopian indigenous sheep breeds/types Breed/Type Litter size Source Gumuz 1.17 Solomon (2007) Menz 1.13 Mukasa-Mugerwa et al. (2002) Afar sheep 1.03 Wilson (1982) Washera 1.11 Mengiste (2008) Bonga 1.40 Zewdu (2008) Horro 1.36 Doyogena sheep 1.42 Getahun (2008) Local sheep in Gamo Gofa 1.3±0.04 Fsahatsion (2013) Local sheep in Alaba 1.51±0.04 Deribe (2009) Local sheep in Ada Barga and Ejere 1.19±0.42 Yadeta (2016) Twining rate (percent) Local sheep in Tocha 24.75±7.9 Amelmal (2011) Local sheep in Mareka 37.8±12.9 Amelmal (2011) Local sheep in Konta 39.06±17.9 Amelmal (2011)

2.6.6. Seasonality in breeding

Tsedeke (2007) identified two possible breeding seasons, between November to January (major breeding season), and between April to June (minor breeding season), when relatively adequate nutrition is available for reproduction and access to breeding males, as rams and bucks roam freely. Lambing occurred year-round, with the majority of lambing in September to October and late April to June in the Alaba especial woreda southern Ethiopia. This period of high lambing percentage is characterized by small rainy seasons with adequate feed. Lambing was lowest in January to early April, which characterize dry period under the Ethiopian conditions. The high lambing rate recorded during September to October and April to June might be attributed to availability of adequate feed. The high percent of lambing is during the rainy season in Ethiopian high lands. The higher rate of lambing during wet season is attributed to high fertility resulting from availability of adequate and high quality feeds from natural pasture Dhaba (2013).

2.7. Feed resource and feeding system

2.7.1 Feed resource

The major livestock feed resources in Ethiopia are grazing and browsing on natural pastures; crop residues and agro-industrial by-products; and cultivated pasture and forage–crop species. Availability and quality of native pastures vary with altitude, rainfall, soil type and cropping intensity. The total area of grazing and browsing in Ethiopia is estimated at 62,280 million hectares, of which 12% is in the farming areas (more than 600 mm rainfall) and the rest around the pastoral areas Alemayehu (2005).

Natural pastures which provide more than 90% of the livestock feed are very poorly managed in both ecological zones in Ethiopia. In the mixed farming mid-altitude areas, better soils are used for cropping and the main permanent natural pasture lands are found on the highland slopes of hills and seasonally waterlogged areas. In the lowlands where pastoralism is practiced most of the land except for rivers, swamps, lakes and deserts contains natural pasture which may be associated with woodland in the wetter areas. Considering the country as a whole, grazing lands contribute 53% of the total land area. Even though the amount of grazing area seems to be large, the yield and quality of the pasture is very low. Due to poor management and overstocking, natural pastures in highland and midland are highly overgrazed resulting in serious land

12 degradation, loss of valuable species and dominance by unpalatable species Alemayehu (1985). In sub humid mid-altitude areas, natural pastures are dominated mainly by Hyparrhenia species, which tend to grow fast and become stemmy and fibrous within short period of time thus losing their palatability and feed value. In these areas, the overgrazed pastures are dominated by unpalatable Sporobolus and Pennisetum species. Herbage growth is dense during the wet season and this gives large bulk of herbage during the dry season, which is burnt to encourage re-growth in subsequent rains. In semi-arid mid-altitude zones, rainfall is the major factor influencing primary productivity Alemayehu (1985). Because of diversity of climate, a number of forage species, mainly grasses are found in both ecological zones. Natural grasslands of the highland areas are rich in legume species, while grasslands of the mid-altitude and lowland zones have lower proportion of legume. The proportion tends to decrease with decrease in altitude. The less abundant native legumes of the lower altitude have sprawling growth Endale (2015).

Cereal straw from teff, barley and wheat is the largest component of livestock diet in the intermediate and highland areas of Ethiopia. Straw is stacked after threshing and fed to animals during the dry season, as are pulse–crop residues (e.g. horse beans, chickpeas, haricot beans, field peas and lentils). At lower altitudes in the highland areas maize, sorghum and millet stovers occur to a greater extent than at higher altitudes. Tef is grown at intermediate altitudes and barley replaces wheat at the higher altitudes, where pulses are also grown to a great extent. The nutritive value of tef straw is equivalent to medium-quality hay and the residue of other cereal crops is only of poor to fair quality. On the other hand, pulse haulms are high-quality roughage with 5- 8% protein content Alemayehu (2003).

2.7.2. Feed and supplementation

According to Workneh (2004), irrespective of the AEZs, production systems or livestock densities herded grazing is the most common practice, followed by un herded (particularly in the dry season) and tethered (particularly in the wet season). Animals were rarely kept in stall/yards or paddocks. Herded grazing is more common in the pastoral and crop–livestock systems than the agro-pastoral system. Tethering, stall/yard and paddock feeding was not practiced in the pastoral system. Sheep are supplemented with minerals and vitamins, roughage/crop residues and concentrates in that order. In general, roughage/crop residue supplementation is higher during the dry than during the wet season, especially in the lowland AEZ. In contrast, mineral/vitamins supplementation is more frequent during the wet season than during the dry

13 season. Higher proportions of the households in the low and medium livestock density categories have supplemented minerals/vitamins during wet season than those in the high and very high livestock densities. Roughage/residues or concentrates supplementation is not practiced in the pastoral production system. Adult male and female sheep tend to be more frequently supplemented than young sheep across AEZs, livestock densities and production systems. Except in the pastoral production system, where adult male sheep are ranked first for receiving feed supplements, adult female sheep generally receive priority for feed supplementation Workneh (2004).

2.8. Sheep Marketing

In terms of marketing channels, there were primary markets and secondary markets. The primary markets were mainly at the grass roots while the secondary markets were mainly in the major urban centers. It was observed that there were no clearly designated primary market yards where farmers could sell their livestock (Yadeta, 2016). Berhanu et al. (2015) reported most of sheep sold at district market. The traders would buy the animals directly from the farmers. Tsedeke (2007) in southern Ethiopia reported that most of producers market their animals based on eye- ball estimation.

According to Yenesew (2010) in Burie woreda, west Gojjam reported that one household sold on average 1.1 heads of sheep per year. The buyers on the market places are mainly farmers (56%), hotel, civil servants (13%), sheep traders (3%) and others (7%). Berhanu et al. (2015) in highland of Ethiopia reported 62% of farmers buy sheep for reproduction purpose, 14.4% for household consumption and 13.8% for trading.

Farmers usually sell their sheep during Easter, New Year and Christmas. During this period the demand for sheep increases thereby resulting in steep rise in market prices of sheep. The male animals are either sold or slaughtered at home during festivals. The study of Yenesew et al. (2013) in Burie district, north western Ethiopia indicated that, both young ram and ewe lambs aged 6 month to 1 year mostly sold. Rams not required for breeding would be sold or castrated before puberty Taye et al. (2009).

The studies have shown that sheep was the only readily available assets to be sold to meet immediate house expenditure by marginal and landless farmers. Farmers sold their animals when they needed cash for home consumption, to pay loan/children's school fees, at time of crop

14 failure or drought and because of difficulty in management Belete (2009) and Ramesh et al. (2012).

2.9. Sheep Production Constraints and Opportunities

2.9.1 Production constraints of sheep

There are several sheep production constraints in the country. These include feed scarcity, transport and infrastructural problems, lack of market information and lack of trained personnel and absence of recording Markos (2006). In mixed crop-livestock systems, relatively high inbreeding coefficient because of uncontrolled mating and absence of sharing communal land for communal herding might potentially increase the risk unless appropriate measure is taken Zewudu et al. (2012). The study of Deribe and Taye (2014) in southern Ethiopia indicated that, non-genetic factors influence reproductive traits and pre-weaning mortality of lambs. Flock management in groups due to resource endowment, parity, litter size, and season (due to seasonal fluctuations in both quantity and quality of feed) were important factors that need to be considered in the improvement plan of sheep.

2.9.1.1 Feed shortage

Many authors described the seasonal feed shortages, both in quality and quantity, and the associated reduction in livestock productivity in different parts of the country Tibbo (2006); Tsedeke (2007); Getahun (2008). Feed shortage problem is similar throughout the country, being serious in high human population areas where land size is diminishing due to intensive crop cultivation and soil degradation. The better use of available feeds and the use of non- conventional feeds for supplementation are growing Belete (2009) to alleviate the problem.

2.9.1.2 Disease and Mortality of Sheep

Diseases and parasites are also contributing for higher production losses, particularly in young stocks. Respiratory Disease Complex (RDC) Tibbo (2006) is among the most important diseases and associated complexes in small ruminants’ husbandry and management. Poor Management is creating a favorable environment for disease incidences. Early mortalities (as high as 50% in lambs) are among the most important losses associated to managements like cold stress, starvation, miss-mothering Tibbo (2006). Predators such as foxes and hyenas are also contributing for the losses of lambs Tsedeke (2007) and Belete (2009).

2.9.1.3 Market access and information

Currently, the contribution of the livestock subsector in Ethiopia is below from its potential. This is because of socio economic and technical limitations like inadequate feed quality and quantity, diseases, poor breed potentials and inadequate livestock policies of credit, veterinary service, extension, marketing and infrastructure that affect the livestock potentials Addis (2015).The major problems in traditional management system are that the system is not market oriented, under developed marketing and infrastructure system, and poor financial facility Azage et al. (2006). Long market chain is an important barrier for producers and inhibits them from direct benefiting through sell of their animals without involvement of brokers. Poor marketing information and problems of credit facilities reduced the benefit gained by the smallholders. Inadequate infrastructure like road accessibility and marketing facilities are also contributing for the reduced benefit made from the sale of animals by the producers Tibbo (2006).

2.10. Sheep production opportunities

High demand of the small ruminants in the local market as a result of population increase, urbanization and also all household member involvement in their management can be considered as an opportunity for the small ruminant production Tsedeke (2009). The study of Okpebholo (2007) showed that low start-up cost as an important factor in providing opportunity for the development of a small ruminant production system by a small-scale farmer with limited resources. Similarly, incensement of mutton demand, as found in present study, was in agreement with finding reported by Solomon et al. (2010) indicating that sheep breeds in the lowlands of the country were in good demand in the middle east markets. Tsedeke (2007) and Zawudu et al. (2012) in western and south-western reported, gender participation is another sheep and goat production opportunities.

3. MATERIALS AND METHODS

3.1 Description of the Study Area

The study was conducted in Arba Minch zuria woreda of Gamo Gofa zone, Ethiopia. The woreda is selected based on its high sheep production potential, large sheep population, pioneer in introducing Bonga, Dawuro and Doyogena sheep breeds and have suitable agro-ecology. Arba Minch zuria woreda is characterized here under in the Table 4.

Table 4; The brief description of the study area Climatic conditions Arba Minch Zuria woreda, Gamo Gofa Zone Major Agro-ecological zones based Below 1500 (27.6%) (Low land), 1500- 2300(41.4%) on elevation of meter above sea (midland) & 2300 – 3300 (31%) (highland) level Mean Annual T0 (0C) 16.3-30.6 Mean annual R.F (mm) 800 – 1600 Sheep Farming system Mixed with crop production

Source: Arba Minch Zuria wereda livestock and fishery office (2016)

There are 29 rural kebeles in Arba Minch Zuria woreda with an estimated area of 168,172 hectare. This woreda has an estimated total population of 211,437 of whom 105,304 were male and 106,133 female

3.2. Sampling technique and sample size

Multi-stage sampling procedures were followed at three different stages. In the first stage, three distinctive agro-ecologies (altitude based regions) that are potential for sheep farming were purposive selected. This three strata are low lands (<1500 masl), midlands (1500-2300masl) and high lands (>2300masl) (MoA (2000).In the second stage, based on reconnaissance survey and participatory rural appraisal (PRA) information one representative farmers administrations (kebeles) from each stratum, three from the woreda, were selected following stratified random sampling technique. In the third stage, individual household heads having sheep of any breed and size and/or adopted improved sheep farming practices were identified and selected using systematic random sampling technique. A list of households in each survey site was identified with the help of the chief of kebeles and agricultural development agents. The identified household heads were questioned using a pretested structured questionnaire. The total sample size for household interview was determined using probability proportionate to size-sampling technique Cochran (1977).

2 *( p)(q) no  z d 2 Where:

17 no = desired sample size according to Cochran (1977), when population (household, HH) greater than 10,000 Z = standard normal deviation (1.96 for 95% confidence level) P = 0.1 (proportion of population to be included in sample i.e. 10%) q = is 1-P i.e. (0.9) d =is degree of accuracy desired (0.05) or 5% error term The sample size was 138; which included 43 HHs from highland, 57 midland and low land 38 HHs were selected.

3.3. Data sources and Data Collection Tools

A focus group discussion consisting of 12 famers were conducted in each agro-ecology. The focus group included model and well experienced sheep farmers, representatives of poor famers, women representatives, development agents and kebele governor at each agro-ecology. discussion was mainly focused on demographic and farming system characteristics, objective of sheep keeping, sheep herd size and composition, herd structure by sex, reproductive performances like age at first lambing and lambing interval, life span lambing, litter size, parturition length and puberty age, general management and breeding practices, productivity and adaptation ability. In addition, the researcher's own observation to production system, breeding practices and physical observation of sheep, besides group discussion and communication with concerning livestock experts at different levels were conducted, farmers trait/breed preferences and sheep resources and utilization systems, marketing systems, constraints and opportunities of sheep production in the studied areas were collected using a pretested structured questionnaire.

3.4. Estimation of Feed Supply

3.4.1. Dry matter yield of pasture

The total amount of dry matter (DM) available in pasture was determined by multiplying the average value of grazing land holding within the per hectare DM yield of the natural pastures with conversion factor of 2tDM/ha/year FAO (1987, 1995). Amount of DM obtained from communal grazing land was factored in to total communal grazing areas for each total households and their associate TLU suitable to graze on this land unit.

3.4.2. Dry matter yield of crop residue, fallow land and after math grazing

The quantity of available crop residues (DM basis) was estimated from the total crop yields of the households, which was obtained from questionnaire survey, according to FAO (1987) conversion factor. The conversion factors are 1.5 for barley, wheat, tef (Eragrostistef); 2 for maize, 1.2 for pulse and oil crop straws and 2.5 for sorghum. The quantities of available DM in fallow land and aftermath grazing are determined by multiplying the available land by the conversion factors of 1.8 for fallow land and 0.5 for grazing aftermath. Quantification of DM yield from road sides, bush/wood lands of common and private sources and household compounds were done following direct assessment Yisehak and Geert (2014). DM yield from crop thinning, and crop weeds were assessed through questionnaire and direct measurement.

3.4.3. Quantity of trees and shrubs

The potential fodder yield of shrubs and trees were estimated by measuring stem diameter using measuring tape and applying the equation of Petmak (1983). Accordingly, leaf DM yield of fodder trees was predicted using the algometric equation of log W = 2.24logDT-1.50, where W = leaf yield in kilograms of dry weight and DT is trunk diameter (cm) at 130 cm height. Similarly, trunk diameter (DT) can be obtained by DT = 0.636C; where C = circumference in centimeter (cm). For the leaf DM yield of a shrub this algometric equation was used log W = 2.62logDS - 2.46, where DS is the stem diameter in cm at 30 cm height. In quantifying tree feed resources from common property resources (e.g., open forest areas) similar approaches as communal grazing area mentioned earlier, was used.

3.4.4. Estimating available concentrates

The quantity (DM basis) of conventional and non-conventional concentrates (supplements) is available for each household was obtained by interviewing the farmers during the cross-sectional questionnaires survey.

3.4.5. Estimation of balance between feed supply and feed requirement

Total available DM in the main dry season from natural pasture, crop residues, crop aftermath, tree legumes and concentrates were compared to the annual DM requirements of the livestock population in the sampled households. Data of livestock population in the sample households

19 were obtained from the interview of household heads during the survey. To compare, the number of livestock population was converted into tropical livestock units (TLU) using the conversion factors of Varvikko et al. (1993). The DM requirements of the livestock population were calculated according to the daily DM requirements for maintenance of 1 TLU (250kg livestock consumes 2.5% of its body weight (6.25 kg DM/d) Kearl (1982).

3.5. Statistical Analysis

Data (both qualitative and quantitative) was cleaned and entered into Microsoft office Excel every day after administering questionnaire to prevent loss of data. All the surveyed data was analyzed using statistical procedures for social science (SPSS) version 20 SPSS (2009). Statistical variations for categorical data was tested by means of cross tabs, with significant differences at P< 0.05; while the descriptive statistics for the numerical data was subjected to one way analysis of variance (one-way ANOVA) using the general linear model procedure of SPSS. Mean comparisons were carried out using Turkey and significant differences were declared when P<0.05. The appropriate statistical model used was:

Yij = μ + i +ij th Where,  is the overall mean;i,, the effect i location( agro-ecology, i= 1, 2, 3) ij, random error For parameters required ranking, indices were calculated to provide ranking of major constraints for the sheep production in the study area and purpose of sheep keeping in the different agro- ecologies. The indices were calculated as follows: Index = Sum of (3 × number of household ranked first + 2 × number of household ranked second + 1 × number of household ranked third) given for an individual reason, criteria or preference divided by the sum of (3 × number of household ranked first + 2 × number of household ranked second + 1 × number of household ranked third) for overall reasons, criteria or preferences. Analyzed data was presented using tables, figures, percentages, means, and standard error in the process of examining and describing sheep production, feeding, marketing functions, facilities, services, intermediaries, and market and animal characteristics.

4. RESULTS AND DISCUSSION

4.1. Socio-economic characteristics of the respondents

4.1.1. Household characteristics

The household characteristics of the respondents are presented in Table 5. The respondents’ age was ranged from 20 to 65 years. This showed similar mean and productive age range in all AEZs. In the lowland, the average family size was relatively looking at the first than highland and mid altitude and this might be due to difference in farming systems, migration of children from high land to low land for job searching and practice of family planning program among farmers according to the focused group discussion during the study.

Overall average family size in the study woreda was 7.1 individuals. Generally, in the study woreda, the average family size of the respondents was higher than national average family size of rural areas (4.9) per household CSA (2011), but comparable finding of this study on highland and mid altitude average family size (6.59 heads per household) was reported by Fsahatsion et al. (2013). The result of present study was also similar to the value reported (6.7 heads per household) for Halaba special woreda of southern Ethiopia by (Tsedeke, 2007). In the result it was also stated that mainly labor (livestock herding and tethering) demanding agricultural activities in the area contributed for such higher family sizes.

The large percentage value of illiterate (81.4%) in the highland and mid altitude (64.9%) compared to the low land can hinder in the adoption of agricultural technologies and farming sheep and other animals during young age or children in the study area (Table 5). In lower altitude, students are forced to drop from education at elementary school mostly because of socio-economic factors like wealth difference of the farmers so that the wealthy family send their children to higher grades as to the group discussion finding during the study. This is higher than the findings of Yilikal (2015) who reported (27.33%) in his research work at Chencha and Mirab Abaya woredas in Gamo Gofa Zone, Southern Ethiopia.

Table 5; Household characteristics of the respondents in the study wereda Agro-ecology, % Characteristics HL ML LL Overall p-value Male 74.4 89.5 73.7 80.4 NS Female 25.6 10.5 26.3 19.6 NS Average age 42.12 42.18 38.97 41.28 NS Average family size 6.8 6.6 7.8 7 NS religion orthodox 65.1 59.6 13.2 48.6 *** protestant 27.9 40.4 86.8 49.3 *** Indigenous religion 7a - - 2.2 *** Educational status Illiterate 81.4 64.9 23.7 58.7 *** Can read and write 7 10.5 26.3 13.8 *** Elementary 9.3 17.5 31.6 18.8 *** Above secondary school 2.3 7 18.4 8.7 *** NS=not significant; *** = p < .001

4.1.2. Landholding and land use pattern of the households

The average land holding of the respondents was found to be the lowest in low land (0.79 ha/HH) as compared to 1.52 ha/HH in the highland and 1.56 ha/HH in mid land (Table 6).This is mostly due to small amount of total land and large human population in low land. In the study woreda, the average total land owned by the households was 1.34 hectares. The overall average landholding of the respondents in the study woreda was higher than the average national landholding size (0.96 ha/HH) CSA (2011). In the low land average land holding was lower than the national land holding. The average land size allocated for crop production was 0.903 ha/HH while that of grazing land was 0.228 ha/HH, fallow land 0.16 ha/HH and tree covered 0.045 ha/HH Table 8.

Private vegetation cover was 0.06, 0.07, and 0 ha of land in the highland, midland and lowland respectively (P<0.01). From the household land holding in the study woreda allocated for crop production was about 0.92, 0.97 and 0.79 ha for highland, mid altitude, and low land, respectively was not significant in the AEZs. Land allocated for grazing of livestock in highland was 0.31, mid altitude 0.38 and 0 low land ha respectively. Average land allocation of grazing land, fallow land, vegetation covered land and total average were highly significant in the AEZs

(p<0.001). The land allocated for crop production and animal grazing in the low land was small as compared to high and mid altitudes and this was due to the availability of smaller size of land per household in the low altitude as it was mentioned above. As the result, mostly animal grazing was restricted to communal grazing land and zero grazing. Yilikal (2015) reported that the amount of land size allocated for crop production was 1.47 ha and grazing land was 0.232 ha in Chencha and Mirab Abaya, Gamo Gofa zone. The land allocation for crop production was lower than his finding but grazing land was similar to these neighboring woredas of the study area. The current results were comparable with earlier reports of Fsahatsion et al. (2013) for Gamo Gofa Zone (SNNP regional state).

The possible reasons could be human population growth rate, land degradation and soil erosion resulting in declining of landholding per household across the three agro-ecologies of the woreda. The increasing trend of land holding of the respondents in the highland, midland and low altitude was reported to 54.5%, 36.4% and 9.1%, respectively. Majority of the respondents disclosed the decreasing trends of land holdings was recorded in highland (77.8%), and midland (22.2%) regions. According to 25.9%, 42.2% and 31.9% of the respondents, the current status of land holding was stable in the three respective altitude regions. The land holding in the highland of the study area was both increased and decreased because land was divided or shared to family members and rented to near farmers contributed for the decreased land holding whereas land rented from near farmers contributed for increased land holding. Table 6; Mean land holding and use patterns of the sampled households in study area Agro-ecology, mean Land types HL ML LL Overall p-value Crop land 0.92 0.97 0.79 0.91 NS Private grazing land 0.31 0.38 - 0.23 *** Fallow land 0.24 0.21 - 0.16 *** Private vegetation cover 0.06 0.07 - 0.05 ** Total landholding 1.52 1.56 0.79 1.34 *** Land holding trends (%) increasing 54.5 36.4 9.1 34.8 *** decreasing 77.8 22.2 - 34.1 *** stable 25.9 42.2 31.9 34.1 *** NS=not significant; **=P<0.01; ***=p<0.001

4.1.3. Farming activities in the study area

In the study area, farmers practice both crop and livestock production. Farmers grow barley, wheat, tef, sorghum, maize, haricot bean, bean, pea, potato, sweet potato, banana and enset. The income status of the respondents by the ecology was high, medium and low of percentage by 4.3%, 39.9% and 55.8%, respectively. The overall, low income status was found to be high percentage value (55.8%) based on the wealth status. This might be associated with less land holding and practices of unimproved farming systems where farmers are mostly dependent on rain fed agriculture. The estimated overall income percentage in the area of general livestock production, sheep production, crop production and trade were the share of 18.97%, 26.83%, 54.03%, and 0.25%, respectively (P<0.001).

Across the different agro-ecology (100%) sheep was the next to the cash crop for income source is responded by all households. This is possibly due to the fact that sheep having important productive traits like prolific, short gestation interval, easy to produce on small plot of land, easy to manage and easy to be sold, “sheep is the pocket money” as also heard from the focused group discussion.

4.1.4. Livestock holding and composition

The livestock composition and holding differed at different agro-ecologies (Table 7). The overall average number of livestock per household for cattle, sheep, goats, equines and chickens was 3.27, 6.5, 2, 0.73, 0.31 and 0.78, respectively. The major livestock species kept were cattle, sheep and goat with the average holding size of 3.21, 10.03 and 0 in the highland, 3.03, 5.07 and 1.25in mid altitude and 3.73, 4.73 and 0.91 low lands which was highly significant in the AEZs (P<0.001). This result was lower than the findings of Tsedeke (2007) who reported the overall average number of livestock per household as 8.5 cattle, 5.0 sheep, 6.5 goats, 1.5 equines and 3.8 chickens in Halaba special woreda, Southern Ethiopia but sheep holding was higher than his finding. The mean sheep holdings in this study lower with 18.7 reported by Fsahatsion et al. (2013) in Gamo Gofa zone. In the three respective AEZs, the mean value for sheep owners was 10.03, 5.07 and 4.73.The possible reasons for this effect were that more productive (more prolific, short gestation interval), easily produced on a small plot of land, contribute to more flexible short-term form of investment, easily marketable related to cattle’s and also respondents

24 traditionally believe that “sheep is a pocket money” if they need money, they immediately go for sales of sheep as it is explained during focused group discussion. Table 7; Livestock holding and composition Livestock species Agro-ecology (average value) overall p-value highland midland lowland Cattle 3.21 3.03 3.73 3.27 NS Sheep 10.03 5.07 4.73 6.52 *** Goat - 1.25 0.91 0.73 *** Equine 0.51 0.1 0.43 0.31 *** Chicken 0.16 0.6 1.7 0.78 *** NS=not significant; ***=p<0.001 The comparison of ecologies for their major livestock species in the study area as it is presented in the Table 7 indicated that sheep population in the highland was higher than the mid land and the low land where as the populations of cattle and goats were high for the low land. For the highland farmers, sheep is the main income source and comfortable in the highland AEZ for production hence, they rearing high sheep population than other livestock. The sheep population in the different ecology has 100% increasing trend in low land, 90.7% increase in the high land and 87.7% increase in the mid land (Table 8). The main reason is that sheep handling is easier than other livestock due to increased population and thus high computation for farm land resulted in low grazing land left for the livestock. Besides, the sheep easily multiply and provide immediate income. This could possibly attract the farmers’ interest to go for sheep keeping. The trend of the sheep population is presented in Table 8. Table 8; Trends of the sheep population of the respondents in the study area Trends Agro-ecology (%) overall p-value highland midland lowland Increase 90.7 87.7 100 92 NS Decrease 8.7 8.8 - 5.1 NS Stable 4.7 3.5 - 2.9 NS

In the current study, the major reasons responsible for declining sheep number were shortage of grazing land, population growth, expansion of crop land, shortage of feeds and animal diseases. These finding is in agreement with the finding of Samson and Frehiwot (2014) who reported that crop production is dominant in the highland production system of Ethiopia where the farmer requires cattle mainly for tillage and herd size might have to be reduced due to grazing land or

25 feed shortage. The total populations of livestock in the woreda were estimated to be 131,884.6 TLU. As indicated in Table 9, cattle comprised 87.2%, sheep 2.22% of the total TLU of the livestock population in the woreda. The overall average TLU of livestock per household in the study woreda is 4.22, 0.11, 0.13, 0.3 and 0.1 for cattle, sheep, goats, equine, and poultry, respectively. The large number of Cattle (4.220) and other small ruminants like of sheep (0.11 TLU) and goats (0.13TLU) owned per HH might indicate the fitness of these animals in that production system as the area was suitable for sheep production. The present findings with respect to cattle are higher than the results reported by Tsedeke (2011), which is 3.78 in the highland of Wolaita and Dawuro zones, Southern Ethiopia and 3.01 reported by Assen and Aklilu (2012) in high land of Dgua Tembien, Laelay Mychew and Mereb Lehe woredas of Tigray Regional State. Table 9; Livestock population (in numbers and TLU) of Arba Minch zuria wereda Livestock species Population TLU factors TLU population Cattle 143432 115017.9 Cow 46432 0.7 32502.4 Oxen 36869 1 36869 Heifer 29255 0.5 14627.5 Bull 31019 1 31019 Sheep 29587 0.1 2925.5 Goat 36632 0.1 3663.2 Donkey 6091 0.5 3045.5 Horse 4344 0.8 3475 Mule 2300 0.7 1610 Poultry 214838 0.01 2148.4 Total 580799 131884.6 Sources: TLU conversion Jahnke (1982); total number of livestock (2016/17) Annual report of Arba Minch zuria woreda Livestock and fish resource development office.

Across the three agro-ecologies of the study woreda, the household owned mixed species composition as mechanism for coping the situation of scarce feed availability by decreasing the competition for feed resources (different species tend to make use of different components), and thus reduced risk by lessening the dependency on one species for animal product for basic household consumption to improve the livelihood. According to all respondents majority of livestock were privately owned. Although lower number of livestock shared holding were found, discussion with key informants revealed that it is an important means of building livestock assets

26 in the area. The majority of the inventory stocks were built through birth. Purchase, gifts from different sources (different NGOs) and inheritances from family were the important ways of building livestock holdings.

4.1.5. Spearman’s correlation coefficient between altitudes, farm land size, family size and performance parameters of the sheep

The statistical correlation between altitude and farm land size was significant (P <0.001) and negatively correlated (correlation coefficient (r) =0.292) but other socio-economic parameters, age and family size of respondents not significantly varied with altitude variation (Table10). Moreover, altitude variation exhibited a highly significant correlation for sheep performance parameters like PAm, LI, SAm, SAf and LT (P <0.001). Liter size and life time lamb crop were positively correlated with altitude (r 0.413 and 0.568, P <0.001). Puberty age of female sheep and lambing interval significantly affected and negatively correlated with altitude (P<0.05). Ages of the respondents’ were positively correlated with sheep performance parameters except life time lamb crop of the sheep. Farm land size positively correlated with sheep performance parameters (p<0.001) and family size values were negatively correlated with the majority of sheep performance parameters (P <0.001). In general, the results of correlation matrix analyses revealed that sheep performance significantly affected by farm land size and family size with the extent varied among different agro-ecology (Table10).

According to the majority of the respondents’ farm land size positively correlated with sheep performance parameters which was linked grazing land obtain ability for sheep for the production. Household size or family size negatively correlated with the sheep performance parameters thus, family size increase, performance decreases. Family size affecting performance by the fulfillment of family needs early sold and slaughter of rams and ewes. Another factors were rearing individuals in the household and competing with other livestock of feed that creates feed shortage in the household and grazing land shortage occurred the expansion of the crop cultivation for the met of feeding large size family.

Age of respondents positive relation with sheep performance parameters resulted from young to adult rearing sheep for meeting family need, sheep easy to control, past replacement and which needs to rear small capital compared to other livestock.

Table 10; Correlation matrix between variables studied (N=138) AL AR FL FS PAm PAf AFL LI LS SAm SAf LT LLC AL 1 -.116 -.292 .162 -.408 -.169 -.170 -.233 .413 -.318 -.405 -.282 .568 AR NS 1 .375 .096 .048 .032 .025 .135 .005 .197 .154 .081 -.027 FL *** *** 1 -.018 .235 .057 .045 -.046 -.215 .308 .259 .222 -.342 FS NS NS NS 1 -.277 -.089 -.100 .032 .418 -.283 -.307 -.050 .262 Pam *** NS *** *** 1 .588 .589 .113 -.499 .688 .670 .220 -.376 PAf * NS NS NS *** 1 .994 .178 -.170 .366 .349 .087 -.137 AFL * NS NS NS *** *** 1 .165 -.164 .360 .345 .100 -.121 LI *** NS NS NS NS * NS 1 -.096 .153 .211 -.019 -.216 LS *** NS * *** *** * NS NS 1 -.511 -.538 -.186 .482 Sam *** * *** *** *** *** *** NS *** 1 .940 .191 -.414 SAf *** NS *** *** *** *** *** * *** *** 1 .243 -.402 LT *** NS *** NS *** NS NS NS * * NS 1 .190 LLC *** NS *** *** *** NS NS * *** *** *** * 1 *=p<.05; ***=p<.001; NS=Not Significant; AL=Altitude; AR=Age of Respondents; FL=Farm Land size; FS=Family Size; PAm=Puberty Age of male; PAf=Puberty Age of female; AFL=Age at First Lambing; LI=Lambing Interval; LS=Liter Size; SAm=Slaughter Age of male; SAf=Slaughter Age of female; Life Time of sheep; LLC=Life time Lamb Crop

4.1.6. Flock structure of the sheep in the study area

The flock structure of the study area is found to be Local sheep breed/type, Dawro/Bonga breed and Wolyta/Doyogena breed. Total number of flock composition of the sheep with sex and age group owned by the respondents were 103 male lambs< six months, 81 Male six months to one year, 56 Male > one year (Intact), 21 Castrated male, 139 female lambs < six months, 169 Female six months to one year and 344 Female > one year. The overall male to female ratio is 1:6 in the study area. This finding is lower than Getachew et al. (2010) finding of Menz sheep breeding ram to breeding ewe (1:8.3).

Flock structure of sheep for the three agro-ecologies is indicated in Figure 7. The sheep flocks were composed of less than six month or sucking male lambs 9.68%, 11.26% and 15.05 with overall 11.28 % in the highland, midland and lowland respectively, which was not significantly different. Weaned male lambs of or six month to one year age or puberty age accounted 11.75%, 5.12% and 8.06% with overall 8.87 % in the highland, midland and lowland respectively (P<0.001), intact/entire males of above one year 6.22%, 6.48% and 5.36% with overall 6.13 % in the highland, midland and lowland respectively (P<0.05) and castrates and fattening males 2.53%, 2.05% and 2.15% with overall 2.30 % in the highland, midland and lowland respectively

28 were not significantly different with in AEZs.

Breeding female (37.68% ewes), puberty aged sheep 27.38% and pre-weaned young (26.51% lambs) constitute the larger proportion of sheep flocks. This finding was in agreement with those reports of Zewdu et al. (2008), Fsahatsion et al. (2013) and Mengistie et al. (2010). The sex- proportion of the flocks represents 28.59% male and 71.41% female in sheep. Breeds of the sheep in the study area were local sheep type (47.6%) and Dawuro/Bonga (1%) in the highland, local sheep type (25.9%) and Dawuro/Bonga (6.95%) in the midland and 100% Wolyta/Doyogena sheep type in the low land of the study area. In the highland and mid altitude large figure of local sheep type were kept. The Dawuro/Bonga sheep breed which was introduced from Dawuro zone by NGOs and GO comprises of small number as compared to the local sheep type. In the low land sheep breed was originated from Wolyita by the traders as it is explained during focused group discussion. Since it has long and fat tail this sheep looks like the Doyogena breed as it is indicated in the following Figure.

Figure 1; Local sheep, Dawuro/Bonga sheep and Woliata/Doyogena sheep (left to right)

Figure 2; Flock structures of sheep populations in the study area

4.1.7. Purposes of keeping sheep

In the study area, sheep were kept for sale (income generation),meat, manure, saving, share risks with other animals and social functions (sacrifices or ritual) in highland with the index value of 0.31, 0.16 ,0.20 , 0.20, 0.12 and 0.01,respectively and in the mid altitude 0.30, 0.19, 0.16, 0.12, 0.15 and 0.01, respectively. In the low land the index value of 0.40, 0.13, 0.32 and 0.16 for sale (income generation), meat, saving, and share risks with other animals respectively (Tables 11). The overall purposes of sheep production were ranked 0.32, 0.23, 0.17, 0.14, 0.13, and 0.01 for sale (income generation), saving, meat, share risks with other animals, manure and social functions (sacrifices or ritual), respectively. The primary reason of sheep keeping by the farmers was for source of income generations through the sale of live animals with an index value of 0.31, 0.30, and 0.4 in highland, midland and lowland AEZs, respectively. During the focus group discussion it was described that from the sales of live animals they generate cash that might be used to buy clothing and food items, pay taxes, to purchase additional fertilizers and household supplies (children schools). These findings were in agreement with reports of earlier research works of Solomon et al. (2010), Mekuriaw et al. (2012), Arse et al. (2013), Yadeta (2016), Hundie and Geleta (2015) and Yadeta (2016).

The second main reason of sheep keeping was for saving sheep as a pocket money being a live bank for the farmers that they use it at any time where the need for cash arises according to the focused group discussion. The rank of the index value was 0.20, 0.21 and 3.2 in the highland, midland and lowland respectively. Thirdly, meat production for the sheep keepers during festivals, family birth and wedding and personal consumptions was another purpose with index value of 0.17 overall AEZs which indicated in Table 10. The fourth purpose was to distribute risk and benefit which was insurance for the sheep keepers during unexpected problems. When they face crop loss, they go for sales of the sheep for the large animals were not easily sold and their long replacement period does not help the farmer to solve his immediate problems so that they prefer to save those large animals. Its index value of this purpose was 0.12, 0.15 and 0.14 in the highland, midland and lowland respectively. Lastly, the manure production was the final purpose of keeping sheep mostly in the high land and midland farmers; manure is used as an organic fertilizer for increasing their soil fertility. Table11; Purpose of keeping sheep Purpose of Agro-Ecological Zones sheep prodn. HL ML LL Overall I Rank

R1 R2 R3 I R1 R2 R3 I R1 R2 R3 I Sale 22 18 3 0.31 29 21 7 0.3 18 13 7 0.4 0.32 1 Meat 16 4 2 0.16 20 16 - 0.19 4 5 4 0.13 0.17 3 Manure 19 6 2 0.20 23 3 4 0.16 - - - - 0.13 5 Saving 11 16 - 0.2 25 15 - 0.21 20 10 1 0.32 0.23 2 Share risks 6 11 - 0.12 21 6 2 0.15 11 4 - 0.16 0.14 4 Scarification 1 - - 0.01 - 1 - 0.01 - - - - 0.01 6

R1= first rank, R2 = second rank, R3 = third rank, I = index 4.1.8. Labor division in sheep management

The division of activities by different members of households in the area is shown in Figure 8 Purchasing and selling of sheep was the responsibility of men (mostly household heads). However, women and children were responsible for other activities. About 60.90% traditional and modern veterinary services for sick flock were taken by men. Large proportion of the flocks was owned by the husband (49.3%) and women possess 23.20% while boys own some 25.4% of the flocks. About 68.1% of sheep were sold by husbands. According to Endashew (2007) women and children may have property right over the flocks, but are not decision makers when it comes to selling of animals. Husbands possess more power in deciding the use of incomes generated

31 from sale of animals. Most of the females are trader in the study area. The present results were in agreement with Tsedeke (2007) and Zawudu et al. (2012) who reported similar labor division in relation to sheep production and management in western and south-western Ethiopia.

Figure 3; Division of family labor for sheep management

4.2. Sheep breeding and reproductive performance

4.2.1. Sheep breeding management

There was often no selective mating policy in the study area. Ram run with flock throughout the year, some select the breeding ram from the flock and neighbor’s with 29.7% and 70.3% of overall altitudes. Only 29.7% of overall respondents that owned sheep had breeding ram. More households retain young male for sales, castration and fattening. This implies that retention of young male for breeding is rarely practiced by owners. The obtained result is in agreement with that of Tesfaye et al. (2011) and Fsahation et al. (2013).

The majority of breeding rams originated from the same or other villages. In dry season, immediately after crop harvest, ram from different flocks while roam freely mates females within the same village or from other villages and some were using market rams or use neighbors’ ram. Two possible breeding seasons of the flocks were identified. Sheep owners (78.26%) indicated that the major breeding time of the flocks was between Octobers to January. Until crops were

32 harvested, flocks usually tethered and maintained under nutritional stress. Between October and January, immediately after crops harvested the flocks freely graze on crop stubbles and aftermaths. Thus, adequate nutrition for reproductive process and access to breeding males creates favorable situation to the flock breeding during this period. About 65.22% of the sheep keepers indicated that the second minor breeding season was between April to June in the different agro-ecologies. As in main season, this was also attributed to improved feed supply from grasses and browses grown immediately after the belg shower rains. Lambing of the flocks mated in major breeding season occur in an entry to wet season from April to May whereas births of the minor breeding season occurs in dry season from October to December. This finding was in agreement with Dhaba et al. (2013) who reported high lambing/kidding rate during April to June and Yadeta (2016). It was observed that there was no deliberate practice of making selective breeding to avoid risks of the inbreeding depression in the flocks. This has of more concern that almost all breeding rams originated from their respective flocks that might imply that the relationship of animals within a flock and even within a village was narrow and inbreeding was widespread and increasing.

4.2.2. Breed types and preference to sheep

Types of sheep observed in the study AEZs were of short and thin-tailed, fat and long tailed and fat-tailed coarse hair type. Local sheep types, Dawro/Bonga breed and lowland sheep type Wolyta/Doyogena breed are found in the study area. According to Solomon (2008) characterized phenotypically Bonga sheep breed as fatty and long tail and Doyogena also fatty and long tail with course hair. In the highland and mid altitude local (in area usually found) sheep type was dominant. In the last few years GO and NGOs introduced sheep from Dawro into the highland and midland. In the low land Doyogena sheep breed was dominant or only found. About 60.9% 54.7% 51.8%, 51.4%, 29.7% and 21.9% prefer for sheep keepers for immediate return, easy to manage keeping/raring, high market demand, distribute loss and benefits, market incentive price and appropriate for slaughter and home consumption, respectively. Sheep keeping was easy to manage 66.7%, 42.1% and 60.5 % in the highland, midland and lowland respectively (p<0.05) and distribute loss and benefits 69.8%, 35.1% and 55.3% in the highland, midland and lowland respectively (p<0.05). Compared to other large animals all agro-ecologies were indicated in Table 12. Immediate return due to high reproduction efficiency was comparably appreciated by large animals like cattle.

Table 12; Household preferences to sheep Agro-Ecology % overall p-value Characteristics HL ML LL High market demand 64.3 36.8 60.5 51.8 NS Incentive market price 27.9 28.1 34.2 29.7 NS Easy to manage 66.7 42.1 60.5 54.7 * Immediate return 65.1 57.9 60.5 60.9 NS Distribute loss and benefits 69.8 35.1 55.3 51.4 * Appropriate for slaughter and home consumption 18.6 17.9 31.6 21.9 NS *=p<.05; NS=not significant

4.2.3. Reproductive performances of sheep

The results on reproductive performance of all agro-ecological zones AFL, slaughter age, age at sexual maturity, LI, LS, reproductive life span of the sheep had significant difference between AEZs as it is presented in Table 13.

4.2.3.1. Age at puberty

Age at puberty for highland local sheep male and female were 7.34±0.123 months and 7.95±0.136 months respectively, Dawuro/Bonga sheep breed/type 7.00±0.555 months and 7.50±0.614 months for male and female respectively which was not significant differ. The sexual maturity of the breed types closely related but there was significantly different with in the agro- ecological zone (p<0.05). In the highland puberty age of male and female local sheep were 7.34±0.123 months 7.95±0.136 months and for midland local it was 7.74±0.120 months and 8.02±.132 months respectively. The case of Dawuro/bonga sheep breed/type in the highland male and female was 7.00±0.555 months and 7.50±0.614 months and in the mid altitude male and female was 7.79±0.120 months and 8.14±0.232 months, respectively. In the highland Dawuro/Bonga Breeds/types and local sheep types had earlier puberty age than the midland sheep types/breeds in the study area. During the focused group discussion it was explained that due to feed shortage and the grazing land shortage, sheep keepers in the mid land mostly keep their sheep tethered. And in the low land it was found that the sheep breed was Woliata/Doyogena, average age at first mate male and female was 6.32±0.127 months and

7.42±0.141 months respectively. In both sex, it was earlier than that of the highland and midland sheep breeds/types. The reproductive performances of sheep are presented in Table 13. Overall 7.23±0.14 and 7.24±0.11 months puberty age for male and female was found in all sheep types. This finding was longer than that of Yilikal (2015), 6.9 months in Chencha and 6.4 months in Mirab Abaya, but closer with Helen et al. (2015) that was reported, 7.63±0.14 for male and 7.24±.14 for female in indigenous sheep production system in eastern Ethiopia in her study on the implications for genetic improvement and sustainable use

4.2.3.2. Age at first parturition

Mean age at first parturition was 12.95±0.138 months for local sheep type and 12.50±0.626 months for Dawuro sheep type/breed in the highland and 13.02±0.135 months for local sheep type/breed and 13.14±0.237 months Dawuro sheep type/breed in the mid altitude. In the highland and midland for local sheep 12.95±0.138 and 13.02±0.135 months and 12.50±0.626 13.14±0.237 months in the highland and midland months Dawuro sheep types/breed, respectively represents the age at first parturition. Woliata/Doyogena sheep found only in the lowland and their age at first parturition was earlier than that of high and midland and the same is true for local and Dawuro sheep types. Average age at first parturition observed in this study was 12.808±.142 months for parturition over the three agro-ecologies. In the mid altitude average age at first parturition was 13.02±0.135 month which was higher than the high land and low land. This finding is in agreement with Deribe (2009), Fsahatsion et al. (2013) in their study on farm performance evaluation of indigenous sheep (12.40) in Alaba, traditional sheep production and breeding practice (12.4±0.28) in Gamo Gofa Zone.

4.2.3.3. Parturition interval

Highland local sheep breed/type parturition interval was 7.76±0.75 and for midland it was 7.60±0.75 months but Dawuro breed/type has 7.50±0.339 and 7.29±0.128 months in the highland and mid land respectively. Doyogena/wolyta sheep type parturition interval was closer to the Dawuro sheep type as indicated in Table 13. Parturition interval was significantly different between AEZs (p<0.05) with 7.519±0.077 months overall agro-ecologies of the system. Lambing interval was closer with Fsahatsion et al. (2013) traditional sheep production and breeding practice (7.34±0.28) in Gamo Gofa zone and Yilikal (2015) small ruminant production and marketing: constraints and opportunities in Chencha and Mirab Abaya woredas, Southern Ethiopia.

4.2.3.4. Litter size

Litter size of the local sheep type in the highland and mid land was lower than that of the Dawuro sheep in the highland and midland as well as Doyogena sheep closer to the Dawuro sheep type as presented in the Table 13. The average litter size or prolificacy as obtained in the present study was 1.495±0.082 lambs per head in the overall agro-ecologies significant (p<0.001) of the study area. The average litter size was in agreement with Deribe (2009) on farm performance evaluation of indigenous sheep (1.51) in Alaba. The current results were within the range (1.08 - 1.75) as reported by Girma (2008) for tropical breeds. But the current liter size were higher than those reported by Yadeta (2016), 1.21, 1.18 and 1.16 in high, mid and lowland, respectively, for Ada Barga and Ejere woreda sheep of West Shoa Zone, Tadele (2010) for Menz and Afar sheep breeds (close to one lamb per lambing), Bonga sheep (1.13) and Washera sheep (1.11) reported by Solomon et al. (2010).

4.2.3.5. Slaughter age

Average slaughter age of sheep was 7.78±0.226 and 7.945±0.224 months for male and female in the overall AEs. Female slaughter age was higher than that of the male in the study area and highly significantly different in the AEZs (p<0.001). In Arba Minch zuria wereda there was consumption tradition of young lambs that might lead to culling of future breeding stock. That there was a belief on this practice to contribute to building the body and health of sick family better specially for mothers and this could attribute to the observed early slaughter ages of the sheep. The current finding is lower than the results reported by Assen and Aklilu (2012) who reported ASMM of 8.42 and 8.8 months in HL and ML of Tigray region respectively. This finding was higher than that of Yadeta (2016) finding but it was in agreement for the low land sheep.

4.2.3.6. Average life span and average lamb crop of the sheep

The variation in the reproductive life span of female sheep was significant due to AEZs (p<0.05) (Table 13). The reproductive life spans of sheep were 10.5, 9.5 and 9.3 years in the highland, midland and lowland respectively. This finding was closer to Yaeta (2016) reported. The mean lamb crops of sheep on its life span were 10.25, 9.7 and 15 lambs in the highland, midland and lowland respectively of agro-ecological zones (p<0.001).

Table 13; Reproductive parameters of sheep flock Performance measurements Agro-Ecology mean ± SE months Overall p-value Local sheep HL ML LL Age at first mate male 7.34±0.12a 7.74±0.12a - 7.24±0.13 *** Age at first mate female 7.95±0.14a 8.02±0.13a - 7.81±0.14 * Age at first parturition 12.95±0.14a 13.02±0.13b - 12.81±0.14 * Parturition interval 7.76±0.08a 7.60±0.08a - 7.52±0.o8 * Litter size 1.29±0.08a 1.21±0.08b - 1.50±0.08 *** Slaughtering age of male 7.73±0.22b 9.09±0.22a - 7.80±0.20 *** Slaughtering age of female 8.00±0.22b 9.16±0.21a - 7.95±0.22 *** Dawro/Bonga sheep Age at first mate male 7.00±0.56b 7.79±0.210a - 7.24±0.13 *** Age at first mate female 7.50±0.61b 8.14±0.232a - 7.81±0.14 * Age at first parturition 12.50±0.6b 13.14±0.24a - 12.81±0.14 * Parturition interval 7.50±0.34a 7.29±0.128b - 7.52±0.o8 * Litter size 1.50±0.36a 1.50±0.136a - 1.5±0.08 *** Slaughtering age of male 7.00±1.00b 8.64±0.38a - 7.78±0.23 *** Slaughtering age of female 7.50±0.38b 8.64±0.37a - 7.95±0.22 *** Wolyta/Doyogena seep Age at first mate male - - 6.32±0.13a 7.24±0.13 *** Age at first mate female - - 7.42±0.14a 7.81±0.14 * Age at first parturition - - 12.4±0.14a 12.81±0.14 * Parturition interval - - 7.45±0.08a 7.52±0.08 * Litter size - - 1.97±0.08a 1.5±0.08 *** Slaughtering age of male - - 6.42±0.23a 7.78±0.23 *** Slaughtering age of female - - 6.42±0.23a 6.42±0.64 *** Average life span in year 10.5±0 .44a 9.5±0.19b 9.3 ±0.09b 9.8 ±0.16 * Average lamb crop of life 10.25±0.4b 9.7±.36c 15±0.37a 11.7±0.30 *** time of sheep in number Mean ± SE (standard error); ***=p<0.001; *=p<0.05

4.2.4. Management practices

4.2.4.1. Housing

About 48.8% of the respondents in high land respondents accommodate their flocks in the main houses together with the family members. According to 51.2% respondents in highland, sheep flocks were confined in separately constructed sheep house due to the large the flock size and farmers confining their sheep in the separately constructed known by local language /Gamogna/ “Dharxxe” meaning night house. This is witnessed by key informant farmers during group

37 discussion indicated. Small sheep flock size was confined in main house during night. In the midland 87.7% of respondents were confining their sheep in the main house and 12.3% were kept in separately constructed house during night time highly significant difference in the AEZs (P<0.001). The small flock size of sheep found in the midland as compared to the highland dwellers forced the respondents to accommodate their sheep in the main house. In lowland, 100% of the respondents accommodate in the separately constructed house for their animals, the reason was that the hot weather in the lowland forced them to confine their sheep and other animals in the separately constructed house. Generally in all AEZs of the study area, 51.4% and 48.6% farmers accommodate their sheep in main house and separately constructed sheep house, respectively (P<0.001).

The major reason for housing flocks at night with the family is to minimize attack by predators and to avoid theft. Predators rarely destroy separate houses and also main houses and causes whole loss of flocks in high land and mid altitude. Fox and hyenas were the major predators. Some sheep groups separately accommodated at night from the flock like fattened, rams and young lambs 16.3%, 33.3% and 2.6% in highland, midland and lowland, respectively (P<0.001). Sheep not separately accommodated at night from the flock group were 83.7%, 66.7% and 97.4% in highland, midland and lowland areas, respectively (P<0.001). This might create unwanted mating, and physical stress to lambs and abortion of pregnant ewes. In the all agro- ecological zones young lambs were kept around the homestead for gazing to avoid energy loss while walking long distances in search of feed and water and to minimize exposure to predators. This result was consistent with Tsedeke (2007) who reported similar accommodation of sheep housing in Southern Ethiopia.

The roofs and floor of the sheep house was 100% grass and deep floor in highland and midlands. However, in the lowland about 63.2% and 36.8% of respondents use corrugated iron sheet and grass houses. Overall 61.2% floor was earth and 38.8% wooden paved/stone during dry season and72.5% earth and 27.5% wooden paved/stone during wet season. In the all altitudes Table 14 was explains about all the sheep housing management. Overall, 61.6% of respondents clean sheep house during dry season was ones a day. About 58.7% of respondents clean sheep stables ones per week during wet season. In the all altitudes cleaning frequencies are presented in Table 14.

Table 14; Sheltering sheep during night, housing and management practices Agro-Ecology % overall p- Sheltering sheep during night HL ML LL value

In main house 48.8 87.7 - 51.4 *** Separately built house 51.2 12.3 100 48.6 *** Separated sheep groups from the flock 16.3 33.3 2.6 19.6 *** during night (fattened, rams and lambs) Non separated sheep groups from the flock 83.7 66.7 97.4 80.4 *** Roof type Grass 100 100 36.8 82.6 *** Corrugated iron sheet - - 63.2 17.4 *** Floor type dry season Earth 72.1 27.9 100 61.2 *** Wooden/stone paved 57.9 42.1 - 38.8 *** Floor type wet season Earth 67.4 57.9 100 72.5 *** Wooden/stone paved 32.6 42.1 - 27.5 *** Cleaning frequencies of week dry season Daily 18.6 36.6 - 21.7 *** One time 39.5 52.6 100 61.6 *** Two times 4.7 - - 1.4 *** Three times 37 8.8 - 15.2 *** Cleaning frequencies of week wet season Daily 44.2 - - 13.8 *** One time 44.2 100 13.2 58.7 *** Two times 11.6 - 86.8 27.5 ***

***=P<0.001

4.2.4.2. Lamb rearing, Selection of castration and fatting

About14%, 12.3% and 28.9% of lambs were offered with green leaves and leftover food until weaning in highland, midland and low land areas, respectively. This was not significant difference in the AEZs. In cases of loss of their dams, young animals were offered with cow’s milk and household leftover foods until they start grazing and browsing. About 13.95% of the highland respondents practiced early weaning (preventing suckling by herding separately or tethering, until 4 months) of young animals to maintain body condition of mothers for the subsequent reproduction, that was earlier than Tsedeke (2007) reported 4.6 months and Adugna (1998) 6.6 months. These kinds of practices were not recorded in agro-ecologies. Dams were

39 self-weaned their lamb (dams mated and prevent suckling, lamb grow and stop suckling).37.2% in the highland, 24.6% in the midland and 27% in the lowland of respondents were castrating their male animals, which was not significantly different in the AEZs.

In the highland, midland and lowland 34.9%, 33.3% and 26.8% of the households respectively used traditional methods of castration (use wood or stone to crash vas deference of the testes), it was highly significant difference in the AEZs (P<0.001), and2.3%, 3.5% and 21.1% households in the highland, mid altitude and lowland respectively use modern methods (using Burdizzo) at the kebeles DAs office, which also was highly significant difference in the AEZs (P<.001). According to 95% of the total respondents the major reason of castration, was to fatten the sheep and sale them, The difference in this practice experienced in the agro ecology was closely related i.e. 93.8%, 100% and 90% in the high land, mid altitude and low land respectively. This was in agreement with Yilikal (2015) and Sisay and Kefyalew (2015) reported closer castration practices of sheep in Chencha and Mirab and Eastern Ethiopia.

About 2.5% of the total flock owners in all AEs castrate males with undesirable physical characteristics like black coat color and small body size at early age to avoid breeding. Several intact males in a household make noises and become restless and difficult to handle thus 2.5% in all AEZs of the respondent castrate their animals to tame, in high land it was 6.2% but in the midland and lowland there was no more rams tamed in the flock. The average age of castration was one year for sheep. This is attributed to the households’ interest to castrate and fatten sheep as early as possible to take advantage of the higher demand and prices at the present market. No one of the sheep owned respondents practice docking the tail of female sheep to facilitate mating and sanitary purpose.

Sheep for sale at targeted festival/holyday markets either home born or purchased were selectively castrated and receive fattening managements of supplementation with locally available feeds and health management (deworming and treatment). Households select fattening sheep with desirable physical characteristics. Sheep in the study area is preferred for their quick response to fattening management, better adaptation to the local environments and possess market desirable physical traits (tail, body length coat colors and horns). Above 27.5% of respondents of sheep owners select animals for fattening based on body conformation (length

40 and body height). Particularly, 48.6% of sheep owning respondents consider the physical characteristics of the animals.

31.2% of respondents of overall altitudes sheep owners indicated that coat color was some of the good traits. White, light brown and mixture coat colors have a preference in that order in sheep, 78.9% of the respondents prefer these most important traits in the low land. The presence of medium size and up-ward orientation of horns add high aesthetic value as it was elaborated during the focused group discussion. The width, length and fatness of the tail were vitally important in sheep in the overall altitudes so that 2.9% of the respondents look intently for these traits. 23.9% of Respondents in the overall altitudes who were sheep owners mentioned that age of the animals was important factor at the market. Young fattened animals fetch higher prices as to the respondents.

This study identified that body conformation (height and length) and physical characteristics (coat color, horn, tail) were the major criteria that household consider selecting sheep for castration and fattening. Coat color (like red, red mixed with black) and horn (medium and twisted) and fat and long tail in sheep were also very important traits farmers preferences. This clearly depicts body conformation and certain physical traits (tail, coat color, horn) were foremost criteria that producers, traders and consumers critically consider and accordingly breeding efforts needs to assimilate the stakeholder preferences.

Table 15; Management practices of young, breeding and fattening flocks Agro-Ecology % overall p- Particulars HL ML LL value Provide lambs additional feed 14 12.3 28.9 17.4 NS Castration of sheep 37.2 24.6 27 29.2 NS Castrating by local methods 34.9 33.3 7.9 26.8 *** Castrating by modern (burdizo) 2.3 3.5 21.1 8 *** methods Average castration age in month 6.79±1.56 5.68±1.4 3.17±.87 5.37±.8 NS Castration to fatten and sale 93.8 100 90 95 NS Castration to control unwanted breed - - 10 2.55 NS Castration to tame 6.2 - - 2.5 NS Selection for fattening body 23.3 33.3 23.7 27.5 *** conformation Selection for fattening physical 48.8 29.8 76.3 48.6 *** characteristics Selection for fattening age of the sheep 27.9 36.8 - 23.9 *** Selection of physical characteristics 14 12.3 78.9 31.2 *** color Selection of physical characteristics - 7 - 2.9 * horn Selection of physical characteristics tail 34.9 43.9 78.9 50.7 *** Selection of physical characteristics 23.3 36.8 78.9 44.2 *** body length NS=not significant; *=p<0.05; ***=P<0.001 4.2.4.3. Purchase, disposal and culling practices of sheep

The major factors that account to increase and reduce flock size were shown in Table 16. Mortality represented 10.8% of overall loss in sheep and it was found to be the medium cause for flock size decline in the studied areas. The sheep sales average value of 1.12±0.112 was one reason for the reduction of flock size per household over years in the three AEZs which account 17.1% of the total disposal of sheep and highly significantly different in the AEZs (P<0.001). Intact male, castrate and weaned young animals were usually sold in need of immediate cash. Pre-weaned young and breeding females are disposed as the last option. Fattened flocks were sold at targeted seasonal festival markets. Slaughter represents a total disposal overall altitudes of 7.7% in sheep. Birth at home was a major source of building household flock and constituted about 93.9% of overall altitudes in sheep flocks. Purchase and gifts from family and relatives and NGOs was 1.7% and 6.1% respectively contributed the overall altitudes of sheep acquisitions.

Table 16; Sheep flock dynamics Agro-Ecology % overall p-value Routes HL ML LL Exit through (subtraction) Sale 12.5 23.5 17.8 17.1 *** Slaughter 7.7 8.7 2.2 7.7 ** Died 10.4 5.9 19.4 10.8 *** Enter through (addition) Home born 96 86.9 100 93.9 ** Purchase 2.1 2.1 - 1.7 NS Gift 3.9 13.1 - 6.1 *** ***=p<0.001; **=p<0.01; NS=not significant

The flocks’ dynamics through major exit routes for overall agro-ecologies was 39.7% of the respondents that practiced culling out of unproductive or poor performing flocks. From these 36.2% and 35.5% respondents practiced cull flocks with reproductive problems (repeat abortion, poor mothering of its lamb, repeatedly lamb death) and old age respectively in overall altitudes of study area. Aged breeding females are also culled and replaced by young flocks. 29.7% of respondents culled their animals due to health problems like physical damage and mechanical or pathogenic factors in the overall altitudes. Similarly, 24.6% of the respondents were culled their animals with some undesirable phenotypic characteristics, for example abnormal leg, black and ash coat color. The following figure shows sheep culling reasons. Similar findings were reported by Yadeta (2016) in West Shao and Sissay and Kefyalew (2015) Eastern Ethiopia.

Figure 4; sheep culling reasons

4.2.4.4. Consumption of sheep

Sheep was slaughtered for household meat consumption. Major slaughter was made during festivals and various family and cultural events representing total slaughter of 7.7% contribute to considerable flock off-take that indicated in table16. Large flock was slaughtered in major festival (meskel and Easter) periods. In occasions when the family birth, circumcision, wedding etc. sheep was slaughtered. During focused group discussion elaborated that sheep owners were commonly slaughter male sheep. Female were primarily kept for breeding and rarely slaughtered. Some of the sheep owners were non-selectively slaughter the sheep that was available at home, either male or female animal during festivals and events.

4.2.4.5. Sheep disease and health managements

Sheep owners rated diseases and parasites, mechanical and undetermined causes as the main cause of sheep mortality, 53.5%, 14% and 50%in the highland, mid altitude and lowland respectively with overall 36.2%. The major diseases in the study area were Pastureloesis, pneumonia, blackleg, external and internal parasites, liver fluke, inter dermal parasite and diarrhea. These diseases were reported as the most prevalent flock health fears across all AEZs

44 as indicated in Table 17. This result was in agreement with the Fsahatsion et al (2013). Sometimes attack by predators (monkey attack young lambs, foxes and some large birds attack lambs like Eagle) was causes of animal mortality as it was elaborated during focused group discussion. This was particularly apparent in the bush and forest abundant widespread grazing areas where wild animals are concentrated. Table17; Major sheep disease in the study area Agro-Ecology % Overall p-value Types of disease HL ML LL Labe/black leg 62.8 85.96 34.42 76.09 *** External and internal parasites 48.8 36.84 5.26 31.88 *** (including liver fluke) Diarrhea 13.95 12.28 12.32 16.67 NS Chepe/ovine pastureloesis 41.86 21.05 - 21.74 ** Oddo/ inter dermal parasite 4.65 5.26 18.42 9.42 NS Samba/pneumonia 27.9 21.05 76.3 38.41 *** ***=p<0.001; **=p<0.05; NS=not significant

About 71.7% of respondents were reported flock sick/death due to cause of poisonous local botanicals in overall AEs. Sheep sick/death due to poison plants was 58.1%, 91.2% and 57.9% in the high land, mid altitude and low land of the sheep respectively, which was significant difference in the AEZs (P<0.001). Key informants listed sheep poisoning local grass and browse varieties which can cause problem more apparent during dry season when flock was less fortune to obtain adequate feed. The known poison plant found in the high land and mid altitude was local name “gaso” shrub and low land was cassava young leaves in the area. The “gaso” is highly green during the dry season which was browsed by animals can cause the sickness like vomiting, froth making and make animals not to browse and gaze anymore and may lead to death.

Figure 5; The photo image of gaso

Households apply various methods of treatments for their flocks with health problems. About 44.9 percent of the respondents in overall AEs applied the various methods for treating their flock in the study area. 60.5 and 63.2 percent respondents in the high land and mid altitude respectively applied traditional treatments on their sheep using different parts of various plants (leaves, stems, roots), water, kerosene, soil, ash and local extractions mixed in varying proportions or alone for different diseases and parasites. Elders were often skilled and experienced in providing the treatments. In the low land, farmers did not use traditional treatments because they were nearer to town for easy access to veterinary service.

Burning of sheep with hot metal was one of the traditional methods practiced to treat sick animals constituting 48.8% and 70.2% of the respondents in the high land and mid altitude, respectively but in the low land no one use the hot iron treatment for their sheep. About 51.2%, 29.8% and 100% in the high land, mid land and low land, respectively with 55.8% of overall altitude, of the considered flock owners do not practice cutting/burning of animals because it causes serious infection of the cut/burned body parts which often causes mortality or further health complications. During the focused group discussion with the key informants extension worker expressed that the cut or burn of hot iron has damaged the skin quality and reduces the market value. Therefore, whenever their animals get sick, they bring them to the nearer veterinary center or DA’s office.

In general, at all kebeles of Arba Minch Zuria woreda animal health experts were placed, but the distance from the center was differ in altitude. On average, the veterinary center was located at 2.23 km, 2.18 km and 0.21 km in high land, mid altitude and low land areas respectively. The distance was closely related for the high land and mid land from considered households. High land and mid altitude average distance was longer than low land therefore low land farmers get veterinary service easily. Less adequately, 16.3%, 31.6% and 92.1% in the high land, mid altitude and low land of the respondents’ flocks received vaccination with overall 43.5%. Some 4.7% respondents in the highland and 5.3% in the mid altitude indicated that their sick animals were immediately slaughtered and 4.7% of the highland and 1.8% of the midland respondents immediately sale their sick animal. About 20.9%, 17.5% and 5.3% in the highland, midland and low land respectively of the flock owners treat their flocks with veterinary supplies purchased from open (illegal) markets, because of the center far from their residence no awareness created.

4.2.5. Feed resources and feeding system of sheep

4.2.5.1 Feed resource availability

The common forms of grazing and feed resource are presented in Table 18. From the interviewed household, 50.7%, 71.7%, 71.7%, 98.7%, 99.3% and 80.4% of utilize communal grazing, private pasture, riverside grazing, grazing aftermath, crop residues and road side grazing, overall AEs respectively. Generally, common sheep feed resources were natural pasture and crop residues in the study area. The main natural pasture was communal grazing, grazing riverside, road side grazing, browses and private pasture grazing.

And crop residues was comprised of different crop residues, enset and banana leaves as sheep feed significant difference with AEZs (P<0.05), root crops and grazing aftermath in the three agro-ecologies. Communal land grazing was 62.8, 8.8 and 100 percent in the highland, midland and lowland respectively, highly significant with AEZs (P<0.001) and road side grazing also 48.8, 91.2 and 100 percent in the highland, midland and lowland respectively, highly significant with AEZs (P<0.001). Similarly, many researchers Tesfaye (2009); Funte et al. (2010); Hassen et al. (2010); Fsahatsion et al. (2013) indicated that natural pasture were the main source of feed for livestock species in Ethiopia.

Table 18; Commonly available of feed resources Available of feed Agro-Ecology % Overall p-value resources HL ML LL Communal grazing land 62.8 8.8 100 50.7 *** Road side grazing 48.8 91.2 100 80.4 *** River side grazing 97.7 100 - 71.7 NS Grazing aftermath 95.3 100 100 98.6 NS Private pasture grazing 97.7 100 - 71.7 NS Crop residues 97.7 100 100 99.3 * ***=p<0.001; *=p<0.05; NS=not significant 4.2.5.2. Common supplement offered to sheep

Supplementary feeding of sheep was common practices during the dry season when feed shortage occurred in the study area. The flock holders during the dry season provide feed supplements in the afternoon when the animal return to home. The common supplements were indicated in the Table 19. According to the focus group discussion results during dry season (from January to April) feed shortage occurred. This does not mean that there was no feed shortage in other months of the year. From the respondents results it was 100% for the period from end of December to April 100% for highland and midland, thus flock holders feed enset leaves and in the low land banana leave. In the overall altitudes 92.8% of the respondents feed their sheep (young lambs, ewes given birth and fattening sheep) food leftover. Different grains fed for fatten and a ewe with lamb in the AEZs with significant difference (P<0.001).

In the low land and mid altitude respondents fed their flock sweet potato roots for fatten sheep and ewes to pass feed shortage periods. Also in the high land 62.8% fed potato and enset roots for their sheep. Concentrate feeding was not common in the highland and midland but in the low land 68.4% fed wheat brain purchased from market; also atella was not in highland and mid land but in the low land about 47.5% feeding atella during dry season. They do not feed the sheep atella during rainy season for fear of edema on sheep. Most of the highland respondents feed their flock mineral soil/bole purchased from market. About 58.1% for Highland and 28.07% for midland have this experience but in low land respondents did not feed mineral soil/salt. The frequency of supplementation of all supplements in all altitudes was carried out when the supplement feeds are available, except enset and banana leaves which they feed one to two times a day during dry season.

Table 19; Common supplements offered to sheep Agro-Ecology % overall p-value Supplements HL ML LL Green fodders (enset & banana 100 100 100 100 NS leaves) Food leftover 88.4 91.2 100 92.8 * Grains 42.2 36.8 34.2 38.4 *** Roots and tubers 62.8 100 97.4 87.7 *** Concentrates(agro-industry by - - 68.4 18.8 *** products, wheat brain) Atella - - 47.5 13 *** Bole (mineral soil) 58.1 28.07 - 29.7 *** ***=p<0.001; *=p<0.05; NS=not significant 4.2.5.3. Feed seasonal availability

Availability of feeds depends on the season of the year when lands were covered with either Meher or Belg season crops. Major Belg season crops including maize, sorghum, haricot bean, barley, tef and potato and these were available on lands from around March/April to harvest time in August/December. Tef, wheat, barley, maize and haricot bean were major Meher crops planted in July and harvested around November/December. The duration extending from planting of major Belg and Meher crops until their harvest makes major challenge to the availability of sheep feeds. After harvest of Belg crops (December/January) sheep graze the crop stubbles for few months and then the Meher crops take over from July to part of December. During these extended times when land was covered with crops, grazing on communal and private Pasture and roadside grazing constitute major sheep feed. Cut and- carry or grazing of local grasses was also major feed source for sheep from crops planting to harvest. Tillers and weeds from crop fields make minor part of sheep feeding. After harvest, crop residues (straws and fresh tops and thinning of Stover) and crop stubble were the major feeds for sheep. 96.4% of the total households responded that during both the Meher and Belg cropping seasons, sheep graze tethered or herded.

4.2.6. Feeding system (grazing system)

Sheep on average grazes for about nine hours a day either roam freely, tethered or herded. Sheep were grazed alone or together with other livestock species (Table 20). According to the respondents’ results about 48.8%, 29.8% and 76.3% in the high land, mid altitude and low land respectively, grazed sheep alone and 51.2 %, 70.2% and 23.7% for high land, mid altitude and

49 low land respectively (P<0.001) with overall 48.6%. Graze together with other animals 51.2%, 70.2% and 23.7% highland, midland and lowland respectively (P<0.001) with overall 51.4% in the AEZs. In wet season, when the major feed resource was communal or private grazing (road side grazing) about 88.4%, 100% and 76.3% in the highland, midland and lowland respectively significant difference (P<.001) in the AEZs of the respondents use fully kept/tethered grazing system with overall 89.9% in the study area. This means sheep do not go in to the crop fields. In the dry season crop residues, aftermath and indigenous browses and leaves were available. Majority of the respondents practiced partly kept/tethered grazing (71%) their animals overall in the study area. while in low land 32% of the respondents used tether and cut and carry method of feeding systems in the study area. Herding, paddocking and zero grazing in wet season and dry feeding season were presented in the Table 20 in the all AEZs of the study area. This result was agreement with Fsehatsion et al. (2013). Table 20; Grazing management practices of sheep in the study area Grazing management practices Agro-Ecology % Overall p-value HL ML LL Sheep alone 48.8 29.8 76.3 48.6 *** Together with other animals 51.2 70.2 23.7 51.4 *** Grazing in dry season Fully kept/tethered 16.3 24.6 7.9 17.4 *** Partly kept/tethered 69.8 71.9 71.1 71 NS Herding 65.1 96.5 38 70.3 *** Free grazing 14 3.5 21.1 11.6 * Zero grazing - - 32 8.7 *** Paddock 2.3 - - 0.7 NS Use of supplementary feeds 92.9 100 100 97.8 * Wet season Fully tethering 88.4 100 76.3 89.9 *** Partly tethering 11.6 - 13.2 7.2 * Free grazing - - 10.5 2.9 NS Herding 83.7 31.6 68.4 58 *** Paddock 11.6 - - 3.6 NS *=p<0.05; ***=p<0.001; NS=not significant 4.2.6.1. Reason of tethering

The result from Table 21 indicated that about 96.4% of the respondents tether their animals in the study area. However, majority of respondents tether their animals in the highland and mid altitude when compared to those in the low land. The main reasons for tethering (as presented in

Table 21) were to prevent crop damage or disturbance, followed by optimal usage of labor, protect from predators and to prevent unwanted breeding. The result shows that there are significant (P<0.05) variation across the three agro-ecologies for the tethering management practice. To prevent or avoid crop damage has no significant variation across the three altitudes. However, there was no significant variation between three agro-ecologies for same reasons. The explanations as discussed above are in agreement with that of Tsedeke (2007), Belete (2009), Funte et al. (2010) and Fsehation et al. (2013). Table 21; The possible reasons of sheep tethering reported by the respondents

Possible reasons Agro-ecology % overall p-value

highland midland lowland

To avoid crop and vegetable damage 97.7 100 89.5 96.4 *

Save labor 55.8 31.6 89.5 55.1 ***

Protect from predators and theft 4.7 1.8 73.7 22.5 ***

Utilize marginal land and hillsides 32.6 22.8 - 19.6 ***

4.2.7. Estimation of Feed Supply

4.2.7.1. Dry matter yield of pasture

About 15342 ha of grazing land were available in the study area. Therefore, the total dry matter production from pasture equals to 30684 tons per year. The amount of pasture produced by the respondents was estimated from the pasture land holding and communal grazing land of the respondents. The average pasture land holding of the total respondents in highland, midland and lower altitudes was 120.75, 65.88 and 15 ha, respectively. Therefore, dry matter production from pasture in the highland, midland and lower altitudes was 241.5, 131.75 and 30 tons/year, respectively which sums up to a total of 403.25 tons dry matter per year. Trees and shrub land covered in the highland, midland and lowland was 3.69 ha, 5 ha and 0 ha, respectively. Hence, the dry matter production in highland and midland were 2.58 and 3.5 tons respectively (Table 22).

Table 22; Estimated quantity of feed DM obtained from different land use types of the respondents in the study area Land use types Agro-Ecology, Area (Total Conv TDM production total ha) ersion (tones) HL ML LL factor HL ML LL Grazing land 23.8 29.9 0 53.6 2 47.5 59.8 0 107.3 Aftermath grazing 12 28.9 11.7 52.6 0.5 6 14.5 5.8 26.3 Forest land 3.7 5 0 8.7 0.7 2.6 3.5 0 6. Communal grazing 97 36 15 148 2 194 72 30 296 land Sub total 136.5 99.8 26.7 263 250.1 149.7 35.8 435.6 4.2.7.2. Crop residues production

A crop residue was one of the dominant feed sources in most parts of Ethiopia especially during the dry season of the year. A total of 73,043 tons of crop residues were produced from different crop types in the study area (Table 23). According to Tolera (1990), 10% of the crop residue loss was expected due to several factors. Therefore, 7304.3 tons of dry matter of crop residue could be lost from the total crops produced, whereas 65738.7 tons of DM obtained from crop residues in the study area. Based on the woreda Agricultural and natural resource office annual report, the proportion of crop residues as animal feed (56.96%) was higher as compared to other feed types in the wereda.

Table 23; Crop and straw yield production in the study area Crop type Total land Grain yield in Conversion Crop residues plowed(ha) tones factors yield in tones Barley 2000 5600 1.5 8400 Wheat 1200 3000 1.5 4500 Teff 2957 3548.4 1.5 5322.6 Sorghum 400 720 2.5 1800 Maize 6562 24279.4 2 48558.8 Bean 200 40 1.2 48 Pea 200 360 1.2 432 Haricot bean 2765 3318 1.2 3981.6 Total 16284 40865.8 73043

According to the respondents the total crop residues produced per year in the high, mid and low altitudes was 52.82, 84.27 and 70.68 tons, respectively. A total of 207.76 tons of crop residues were produced from different crop types in the sampled households (Table 24). Therefore, 20.78 tons of dry matter of crop residue could be lost from the total crops produced whereas 186.98 tons of DM obtained from crop residues of respondents in overall altitudes in the study area. The total crop residues of DM produced per year in the high, mid and low altitudes was 47.54, 75.84 and 63.6 tons, respectively. Based on sampled results, the proportion of crop residues as animal feed (35.38%) was covered as compared to other feed types in the study household of the study area. Table 24; Crop and straw yield production in Arba Minch zuria wereda Crop type Total land plowed(ha) Grain yield in tones Conversi Crop residues HL ML LL HL ML LL on factors yield in tones Barley 8.67 4.08 - 24.28 11.42 - 1.5 53.56 Wheat 3 6.5 - 10.5 22.75 - 1.5 49.88 Tef - 1.5 0.713 - 1.8 0.86 1.5 3.98 Sorghum - 2.35 - - 4.22 - 2.5 10.56 Maize - 5.8 9 - 2.14 33.3 2 70.88 P. Bean 0.08 0.075 - 0.15 0.15 - 1.2 0.36 Pea 0.22 0.12 - 0.39 0.21 - 1.2 0.72 H.bean - 8.5 1.94 - 12.53 2.33 1.2 17.82 total 11.97 28.93 11.65 35.32 55.23 36.48 207.76 4.2.7.3. Crop aftermath

The contribution of crop aftermath in livestock feeding was important especially in dry season when feed availability was limited to crop residue and aftermath grazing. In the study area the

53 total of 52.55 ha of land were covered by different crop types (according to the respondents’ responses). The conversion factor of stubble gazing into total dry matter yields is 0.5 (FAO, 1987). Therefore, 26.275 tons of DM feed was obtained per year from crop aftermath in the study area. 5.99, 14.47 and 5.83 tons of DM crop aftermaths were produced in the high land, midland and lowland respectively. As indicated in table 24, these crop aftermath were majorly obtained from wheat, barley, tef and maize Stover in the highland and mid altitudes whereas from tef, wheat, maize and sorghum Stover in mid and lower altitude. In the woreda, total of 16284 hectare covered by different crop types which was indicated in the table 22. Hence, 8142 tons of feed obtained per year from crop aftermath in the study woreda.

4.2.7.4. Forest land dry matter production

In the wereda, the total area of land covered by forest was 15501 ha Annual report of Agricultural office (2016/17). The conversion factor used to get total dry matter production from forest land is 0.7 (FAO, 1987). Therefore, a total of 10850.7 tons of feed dry matter was produced in the wereda. According to the respondents total dry matter production from forest land was 2.58, 3.5 and0 tons in the high land, mid and lower altitude. This indicated that forest land availability which could be the source of livestock feed in the mid altitude was higher as compared to the high land and lowland.

4.2.7.5. Indigenous fodder trees and shrubs in the study area

In the three altitudes (highland, midland and lowland) of the study area fodder trees and shrubs (1.1 tons) supplemented sheep feeds. The available fodder trees and shrubs that were known by the farmers are indicated in the Table 25.

Table 25; The estimated biomass yield of indigenous fodder trees and shrubs identified in the study area Local name Scientific name Brose type Average C Average W Buna Coffee Arabica Shrub 32.3 35.34 Woybeta Terminalia brownii Fresen Tree 75.1 41.28 Shola Ficus sur Tree 291 159.95 Wanza Cordiya Africana Tree 82 45.38 Bisana Croton macrostachyus Tree 76 43.17 Grawa Vernoniya amygadilina Shrub 89 97.4 Girar Acacia nilotica Tree 51.2 28.34 Zage/birbra Millettia ferruginea Shrub 25.7 22.25 Karkaha/Bamboo Tree 32 17.59 Korch Erythrina brucei Tree 317 174.24 Kosso Hagenia abyssinica Tree 310.5 170.67 Shomboko shrub 10.3 11.27 Badena Balanites aegyptiaca (L.) Del. tree 99 54.42 ‘Lolash’ shrub 97.7 106.89 ‘Shinka’ shrub 65.7 71.88 total 1079.97 4.2.7.6. Non – conventional feeds

In the study woreda, utilization of non-conventional feeds other than local alcohol waste (Atella) and food leftover was very low. In the highland, midland and lowland dry matter production was 1, 1, and 3 tons per year, respectively.

4.2.8. Dry matter requirement of livestock in the study wereda

The DM requirement was calculated based on the daily DM requirement of 250 kg dual purpose tropical cattle (an equivalent of one TLU) for maintenance requirement that needs 6.25kg/day/animal or 2281 kg/year/animal Jahnke (1982). Therefore, the total dry matter requirement of 129,736.2 TLU is 295,928.27 tons per year (poultry was excluded because of mono gastric nature of the animal).

Based on the respondents, TLU value were 157.9, 158.9 and 129.1 in the high land, mid altitude and low altitudes, respectively, without poultry. The TLU annual dry matter requirement was 360.17, 362.51 and 294.5 tons in the highland, midland and lower altitudes, respectively. From this result, the total dry mater requirement in the highland and mid altitudes was higher than

55 lower altitude. This was due to relatively large number of livestock in highland and mid altitudes.

4.2.9. Estimated annual feed balance

The current DM production of feed from pasture grazing, crop residues, crop aftermath grazing, foliage’s of fodder trees and shrubs of the study area was 115,416.5 tons per year. The total dry matter requirement for 129,736.2 TLU (poultry was excluded because it is mono gastric) is 295,928.27 tons per year. The total dry matter produced per year in the woreda, can only supply the animals for approximately 6 months only. In the rest of the year, animals suffer from feed shortage.

According to respondents in highland, mid and lower altitudes, the total dry matter of feed obtained per year was 299.01, 227.01, and 102.76 tons, respectively whereas the total DM requirement for TLU 157.9, 158.9 and 129.1 is 360.17, 362.51 and 294.5 tons in highland, mid and lower altitudes, respectively. Therefore, the total dry matter produced in these areas can supply the animals for 8months, 6.22 months and 4.05 months in the year in highland, mid and lower altitudes, respectively. In the overall AEZs was only 6 months satisfied. According to respondents, in the highland annual feed balance was higher than midland and lowland, highland farmers have large communal grazing land. Highland, midland and lowland because 4, 5.78 and 7.95 months of feed shortage, respectively in this areas. In Arba Minch Zuria wereda, the existing feed supply on a year round basis satisfies only 61.2% of the maintenance. DM requirement of livestock, lower than the findings of Tessema et al (2003) reported 72.7% maintenance DM requirement of livestock in the Metema district and lower than Endale (2015) reported 7.29 and 5.55 midland and lowland, respectively, assessment of feed resources and determination of mineral status of livestock feed in Meta Robi district. But in the highland finding was higher than his finding.

Table 26; Feed resources category and their DM supply according to regional differences Nutrient Feed supply Agro-ecology, yield total Highland Midland Lowland DM Grazing feed 247.5 146.3 35.8 429.6 Forest land grazing 2.6 3.5 - 6.1 Crop residues 47.54 75.84 63.6 186.98 Indigenous fodders and shrubs .37 .37 .36 1.1 No-conventional feed sources 1 1 1.5 3.5 Concentrate(wheat brain) - - 1.5 1.5 total 299.01 227.01 102.76 628.78 Table 27; Yearly differences in the balance between feed resource availability and dietary requirement Nutrient Feed supply Agro-ecology, yield total HL ML LL DM Available tone 299.01 227.01 102.76 628.78 Requirement tone 360.17 362.5 294.5 1017.17 Balance tone -61.16 -135.49 -191. 74 -388.39 4.2.10. Water resource and watering practices

The watering systems in the highland, mid and lower altitudes were similar with 100% respondents in all the AEZs practicing group watering system. Sheep get water from river (100 %). In the present study, livestock get water on average distance of less than 1 km. getting water sources at the nearest distance can save their energy that was otherwise wasted in searching water. In the study wereda, 22.5% of the respondents water their animals twice a day, 5.8% once a day and 71.7% ad libitum overall AEZs highly significant (P<0.001). In the dry season and rainy season farmer Provide water to their sheep differently, during the rainy season they provide once a day in the lowland but in the mid and highland they do not give water to their sheep, as expressed during group discussion. All over the study area sheep owners did not faced water related problems, because of Arba Minch zuria woreda rich with a lot of rivers, ponds and spring accesses.

Table 28; Watering systems and frequency of watering in the study wereda Watering Agro-Ecology % Overall p-value frequency HL ML LL Any time needed 81.4 96.5 23.7 71.7 *** Once a day 14 3.5 - 5.8 *** Twice a day 4.7 - 76.3 22.5 *** ***=p<0.001

4.2.11. Sheep marketing system

4.2.11.1. Reasons of selling, buying and market places

Sheep rearing was one of the main cash income sources for the farmers in the study area. There were six village sheep markets and one zone market places in the wereda. Selling and purchasing of all species of animals in the village markets and Zonal town markets was observed throughout the year, though peak marketing was observed when high cash demand for agricultural inputs was needed. The farmers generate income for farm inputs (fertilizer, seed, others) from the sales of sheep. 72.1%, 70.2%and 47.3% was the percentage of income generation in the high land, mid altitude and low land respectively (P<0.05) with64.5% for overall altitudes.

Most of the high land and mid altitude farmers purchasing fertilizer and seed during planting season by selling their sheep based on the above survey result which was significant difference with in the AEZs (P<0.05). The income generation for children schooling/purchase cloth was 48.8%, 43.9% and 76.3% in the high land, mid altitude and low land respectivelywith54.3% of overall (P<0.05). Based on the survey result low land farmers sold sheep for the children school fee/purchasing cloth. The income generation for family health and animal health treatments was 55.8%, 50.9% and 26.3% in the high land, mid altitude and low land respectively with 45.7%of overall altitude (P<0.05). Traditionally ‘sheep is a pocket money’ in the high land and mid altitude as elaborated during group discussion for the immediate cash need come from the sheep.

For Shortage of feed and grazing land the response result was 55.8%, 50.9% and 81.6% for the high land, mid altitude and low land, respectively (P<0.05) with 60.9% of overall. Population growth was the factor for declined grazing land for it has created competition of land for crop cultivation and settlement of residents, which was mostly observed in the low land as it was specified during focus group discussion. Sheep owners generate income from sales of sheep to purchase food when the crop production fails. In this finding overall percentage of the

58 respondents for purchase of food was 46.5%, 50.9% and 23.7% in the high land, mid altitude and low land respectively (P<.05) with 42% of overall. Farmers who were paying credit/tax from their sheep was 58.1%, 61.4% and 26.3% in the high land mid altitude and low land, respectively (P<.05) with 50% of overall average value. This result was closer trend with Berhanu et al (2015) and Yadeta (2016) in the highlands of Ethiopia and West Shewa zone.

The respondents’ market place for sale and purchase of sheep in the highland, mid altitude and lowland was mainly Gina, Gatse, Batale (Wusamo), Garbo, Gaga and Chano market places. Reasons for purchase of sheep were 2.9% for slaughter during festivals and 7.25% for breeding in the overall altitudes of the study area. 16.3% and 5.3% in the high land and mid altitude and 4.7% and 3.5% of respondents in the high land and mid altitude were used for slaughter during festivals and breeding, respectively. Table 29; The reasons for sheep sale of the sampled respondents Reasons of sheep sale Agro-Ecology, % Overall p-value HL ML LL Purchase farm inputs 72.1 70.2 47.3 64.5 * For children school fee/cloth purchase 48.8 43.9 76.3 54.3 * For family and animal health treatments 55.8 50.9 26.3 45.7 * Grazing land and feed shortage 55.4 50.9 81.6 60.9 * Purchase for family food 46.5 50.9 23.7 42 * Pay for credit/tax 58.1 61.4 26.3 50 * *=p<.05 4.2.11.2. Sold classes of sheep for urgent income need

Generally, male sheep at young age were sold on market ranking of sale. On this study the 69.8%, 75% and 81.6% in the highland, midland and lowland respectively with overall 75.4% of respondents sold male lamb less than 6 month age for urgent income needs, which was not significant in the AEZs and 79.1%, 93% and 94.7% in the highland, midland and lowland respectively, with overall 89.1% of respondents sold ram lamb between six month and one year age. It was significant in the AEZs (P<0.05). Remained class of sheep sold for urgent income needs female lambs less than six months age, breeding rams, castrated, old ewes and old rams overall percent value of 19.6, 10.9, 40.6, 60.1 and 60.1 respectively in the AEZs. Which were highly significant in the AEZs (P<0.001) in the study area. This shows that farmers in the three altitude sold sheep classes mostly on young male lambs and old sheep (both sex) as presented on the figure 12.

Figure 6; Classes of sheep sold for urgent income needs.

4.2.11.3. Season of marketing sheep

One household in the study area sold on average overall 1.12±0.112 heads of sheep, minimum sold 0 and maximum 8 sheep per year. In the sampled respondents 1.26±0.205, 1.19±0.207 and 0.84±0.112 sold on average in the high land, mid altitude and low land respectively and the minimum sale in all altitudes were zero but the maximum was 5, 8 and 2 sheep in the high land, mid altitude and low land respectively. Farmers usually sold sheep during Easter, New Year, planting season and Christmas. During this period there were high demand on market and market price for sheep was higher. During festivals the demand was very high and the animals sold fetch better prices. On average, 903.57 Birr per head was a market price of overall altitudes during the study period and during planting season (average price 479) it was very cheap because the farmers sold sheep for the input purchasing. Average sheep sold and purchase in the year for overall altitude was 17.1% and 1.7%. Purchasing and sales seasons were planting season’s belg and meher and festivals like Easter, meskel, Christ mass and epiphany as elaborated during focus group discussions in the all altitudes. In the Table 16 sheep sold and purchased in the last 12 months of the three AEZs is presented. Which is agreement with Tsedeke (2007) in Alaba

60 southern Ethiopia; Belete (2009) in western Ethiopia who reported that marketing and consumption of sheep and goats was targeted to holidays of the year rounds

4.2.11.4. Mode of price setting

Producers and purchasers choose 100% prefer “eye-ball” marketing mode. Animals were marketed on individual basis and agreement to prices reached after a long one-to-one bargaining between buyers and sellers and sometimes via brokers. Local and terminal traders were better informed of the demands and prices of animals and decisive to fix prices. Producers usually sale with the trader prices for their immediate income needs. The market price information obtained overall was 14.5% and not obtained was 85.5% in the study area. Market information obtained was 11.6%, 12.3%and 21.1% in the high land, mid land lowland, respectively, not significant difference in the agro-ecology. The source of information 4.65%, 12% and 15.8% in the highland, midland and lowland was development agents, respectively with 10.9% overall altitudes so that the study area has insignificant (P<0.05) access for market information.

The market challenge faced was 7%, 1.8% and 7.9% for highland, midland and lowland respectively. Types of market challenges were seasonality of market demand, tax and brokers interference 9.3%, 1.75% and 8.57% for highland, midland and lowland respectively, which was not significantly different in the AEZs and 88.4%, 87.7% and 78.9% in the highland, midland and lowland respectively for lack of market price information.

Wusamo Gaga Gina Garbo market market marke market t Chano market Gatse marke t

Arba Minch town Collected in bere and transported to Addis Ababa

Figure 7; Sheep market routes in Arba Minch Zuria woreda

4.2.11.5. Marketing channels

The major marketing channels in the study area linking farmers with end users were present in the figure 15. These different channels represented the full range of available outlets through which sheep move from the different collection markets in production area to local markets to meet end-users need. Farmers’ consumers Farmers Hotels consumers Farmers’ small/medium trader hotels consumers Farmers’ small/medium traders’ large traders’ hotels consumers

Figure 8; Sheep market channels in Arba Minch Zuria

4.3. Sheep production constraints and opportunities

4.3.1. Sheep production constraints

The main constraints that had limited sheep production and productivity in the study area were ranked and presented in Table 30. Feed and grazing land shortage was the first rank and which was major constraint in the three altitudes in the study area with index value of 0.3. The present findings were in agreement with Yenesew et al. (2013) who reported that in Bure district of North western Ethiopia feed shortage was very severe especially in the highland kebeles and

Mesay et al. (2013) in Lemu-Bilibilo district in Arsi zone also reported that, shortage of feed at the end of dry season when all crop residues have been consumed and pasture growth is poor, was the major constraint for livestock production in the area. The second constraint was disease and parasite occurrences with the index value of overall altitude 0.26, but in the lowland first rank accounted by its index value was 0.32 and the third and fourth rank were labor shortage and genotype with index value of 0.14 and 0.14, respectively, and other remained were minor constraints of sheep production. Table 30; Major constraints of sheep production in the study area Agro-ecology Constraints Rank Index Highland 1 2 3 4 5 6 7 Overall Feed shortage 11 26 - - - - - 0.26 0.30 Disease and parasite 29 ------0.22 0.26 Labor shortage - 5 25 - - - - 0.17 0.14 Poor Marketing - - 1 5 - - - 0.03 0.01 Inadequate extension - - 2 14 8 1 - 0.1 0.09 and support, Low productivity of 1 7 3 5 11 10 4 0.15 0.14 local breeds Midland Feed shortage 23 33 - - - - - 0.30 0.30 Disease and parasite 34 ------0.20 0.26 Labor shortage - 12 26 1 - - - 0.17 0.14 Poor marketing - - - 1 - 1 - 0.005 0.01 practices Inadequate extension - - 8 10 10 - - 0.09 0.09 and support, Low productivity of - 11 7 16 10 10 1 0.18 0.14 local breeds Low land Feed shortage 2 36 - - - - - 0.3 0.30 Disease and parasite 36 ------0.32 0.26 Labor shortage, - 2 33 - - - - 0.23 0.14 Inadequate extension - - - - 7 10 - 0.05 0.09 and support Low productivity of - - - 3 2 - 7 0.03 0.14 local breeds Index= ((7 for rank 1) + (6 for rank 2) + (5 for rank 3) + (4 for rank 4) + (3 for rank 5) + (2 for rank 6) + (1 for rank 7) divided by the sum of all weighed mentioned by the respondents

4.3.2. Feed and grazing land shortage;

Feed and grazing land shortage was very severe in the three agro-ecological zones presented in table 31. The primary problem was caused by cultivation, settlement and protection on grazing lands for crop production as the human population was increasing significant in the AEZs (P<0.05). That was accounted the percent value of 90.7, 98.2 and 44.7 in the highland, midland and lowland respectively in the study area. The next problem was shrinking and decline in productivity of grazing land was 76.7%, 75.9% and 89.5% in the highland, midland and lowland respectively, which was not significant in the AEZs in the study area. the third factor was increase in human population about 79.1%, 86% and 13.2% in the highland, midland and lowland respectively contributed grazing land shortage by expanding crop land and settlement place, which was highly significant difference in the AEZs (P<0.001). The remained factors were increasing of animal population 34.8% and drought 19.6% of overall altitudes those were resulted in overgrazing and poor productivity of grazing lands and which leads to drought. Table 31; Reasons of feed shortage in the study area Agro-Ecology % Over p-value Particular HL ML LL all Shrinking and decline in productivity of grazing land 76.7 75.9 89.5 79.7 NS Increase of animal population 9.3 10.5 44.7 34.8 *** Cultivation, settlement and protection on grazing 90.7 98.2 44.7 81.2 * land Drought - - 44.7 19.6 *** Increase in human population 79.1 86 13.2 63.8 ***

*=p<0.05; ***=p<0.001; NS=not significant 4.3.3. Opportunities of sheep production

Even though there were many constraints that affect sheep production in the study area, there were also a couple of opportunities to produce and improve sheep production such as presence of adaptive breed, low startup cost, GO and NGOs intervention (breed improvement, disease and parasite control, forage development skill introduce, knowledge transfer), high market demand, easy to manage, immediate return and increased demand of sheep meat. From the study, as indicated in Table 32, immediate return was the primary opportunity like fast reproductive rate(prolific, short lambing interval, early lambing, short time maturation) with percent value of 65.1, 57.9 and 60.5 in the highland, midland and lowland with overall 60.9, respectively, not significant difference with AEZs. The second opportunity of sheep production was low startup

64 cost with percent value of 51.2, 54.4 and 68.4 in the highland, midland and lowland with overall 57.2, respectively (P<0.05). Government non-government organization intervention, easy to manage, market demands and access, adaptive breed type and increased demand of sheep meat were ranked as third, fourth, fifth sixth and seventh opportunities with percent values of overall 55.8, 54.7, 51.8, 37 and 25.8 respectively. Easy to be managed by children and women are advantages to be integrated with crop production. And also, high demand of the sheep in both the domestic and export markets as a result of population increase, urbanization can be considered as an opportunity for the sheep producers.

The need for young males is also opportunity for the producers. Nowadays, many abattoirs flourish in the country; so agents and assemblers purchase sheep. Fattened flocks have high demand and incentive price during seasonal Holiday markets with peak demands in New Year, Meskel and Easter holidays which was significant difference with AEZs (P<0.05) in the study area.

Adaptive sheep type, interventions of government and non-government organizations and increased demand of sheep meat in the study area were highly significant difference (P<0.001). Organizations disseminating improved sheep breeds and providing management trainings and health care are already in place. The local feeding management system entirely depended on natural pastures of spicy herbs drawn the preference of urban consumers like the current “organic agricultural products”.

Moreover, the area (Arba Minch Zuria) receives enough amount of rainfall that can be used to develop various types of grasses, legumes and browses through different production strategies. Landless youth and farmers, retired people and other members of society can be engaged in fattening activities that make them benefited as result of high market demand and higher prices. The local NGO (EKHC – SWZ and Catholic) church based and the working on the sheep breed improvement for small holder farmers, its target kebeles. This was the one of the opportunities of the sheep production in the study area these results are consistent with those with Fsahatsion et al. (2013) and yilikal (2015).

Table 32; Opportunities of sheep production in the study area Agro-Ecology % Overall p-value Particular HL ML LL High market demand 64.3 36.8 60.5 51.8 * Easy to manage 66.7 42.1 60.5 54.7 * Immediate return 65.1 57.9 60.5 60.9 NS Low startup cost 51.2 54.4 68.4 57.2 * Adaptive breed/sheep types 44.2 36.8 28.9 37 *** Interventions from GO and NGO 58.1 64.9 39.5 55.8 *** Increased demand of sheep meat 44.8 21 10.1 25.8 *** *p<0.05; ***p<0.001; NS=not significant

5. CONCLUSION AND RECOMMENDATION

5.1. Conclusion

The major purpose of this study was to generate baseline information on sheep production, marketing, feeding system and identifying production constraints and opportunities. Information was obtained from 138 sample sheep holders. Household characteristics and socio economic features, production systems and purposes of keeping, health and diseases of sheep, reproduction performances, resources and systems of feed utilizations, and other routine management aspects related to sheep production systems information were gathered through interview and focus group discussion with key informants.

Livestock holdings indicated that larger population of cattle, sheep and goat in Arba Minch Zuria woreda. Sheep were kept by every household (N= 138, 100%) and it is potential district for sheep production. According to the respondents, availability of suitable environmental conditions, seasonal abundances of crop residues, weeds and crop leftovers and residual parts of perennial crops (Enset and banana) and annual crops together with introduction of best practices would help to enhance sheep productivity in the area.

Major sheep feeds were obtained from grazing on crop stubbles, private pasture, road side, communal pasture, local browses and cut-and-carry of local browses and grasses. Feed availability largely depends on the season of the year when lands are covered with either autumn/Meher or summer/Belg season crops. Quantity of the seasonal available feeds is usually inadequate for existing sheep populations. In the area, sheep keepers were practicing traditional way of sheep production, feeding and marketing system. Poor housing, feeding and health care were the prevailing conditions, which lead to low productivity of animals.

Sheep in Arba Minch Zuria woreda have appreciable reproductive performances such as: early maturity (P<0.001), short lambing interval (P<0.01) as compared to published reports of many other sheep breeds in Ethiopia. Sheep are coming important and integral parts among livestock in this area. Some of the respondent farmers were keeping sheep with no heads of cattle and other livestock species. All of respondents were said that sheep were kept for source of income, meat, saving, share risk/benefits with other animals, organic fertilizer/manure etc. Thus, they are a good substitute for the resource poor smallholder who cannot afford money to buy cattle.

Interdependence between sheep and crop production is supportive to overcome the problem of land shortage, and it encourages the involvement of all the family members in providing feed for sheep and develops the division of labor among family members in the different sectors of household farm activities. Information on marketing systems, market class, marketing channels, seasonality of prices, mode of marketing, price setting in relation to supply and demand of sheep were collected considerable differences was seen among the study area. Fattened flocks are largely marketed during festival markets and highly demanded consumers. New Year, Easter, and Christmas festivities are peak seasonal holidays in major markets.

The area in general, has high potential for sheep production due to the presence of favorable agro-ecology to grow different feed resources, high market demand of live animal and meat, availability of universities, research centers and different organizations working on improvement of sheep productivity. The major constraints of sheep production in the area are shortage of feed and grazing land, disease, low performing, genotypes and labor shortage during ploughing season.

5.2. Recommendation

The sheep population of the Arba Minch Zuria woreda needs a total of 295,928.27 tons of dry matter feed per year for maintenance requirement alone but the current production can only support for 6 months which is very low. Therefore, alternative feed production technologies such as development of improved forages, efficient feed utilization technologies (e.g. provision of chopper, storage/preservation system of crop residues) and natural pasture land improvement measures should be undertaken.

In the current study Dawuro sheep and local sheep reproductive performances were closer except litter size. Therefore, GOs and NGO should introduce well known breeds like Bonga, Horro in Ethiopia for their better reproduction and production performances. However, the introductions should also consider Similarities of agro- ecological conditions of receiving area to improve local sheep, thus, superior and economically important traits.

Livestock sector and other related organization should provide training on sheep production improvement technologies and flock management before introduce improved sheep breed for the farmers who were engaged on sheep rearing.

Further researches are important in order to identify and select the superior and economically important traits with in this ecotype. Thus the researchers and organization should intervene improving this ecotype in order to utilize local /indigenous potential sheep breed/type.

Quantitative aspects of marketing (supply, demand, prices, producer, alternative marketing mode and consumer behavior) are required for further investigation in order to collect complete marketing information in relation to marketing of sheep.

6. REFERENCES Abate, A.L., M. Meyer (1997). Prediction Gilgel of the useful energy in tropical feeds from proximate composition and in vivo derived energetic contents. 1. Metabolisable energy. Small Rum.Res., Abegaz, S., E. Negussie, G. Duguma, J.E.O. Rege (2002). Genetic parameter estimates for growth traits in Horro sheep. J.Anim.breed.Genet, 119:35-45. Addis, G. (2015). Review on Challenges and Opportunities Sheep Production: Ethiopia Department of Animal Production and Extension, Faculty of Veterinary Medicine, University of Gondar, Ethiopia Ademosum, A. (1994). Constraints and prospects for small ruminant research development in Africa. In: S.H.B. Lebbie, B. Rey, and E.K Irungu (eds). Small ruminant research and development in Africa. pp. 1-5.In: Proceedings of the 2nd biennial conference of the African Small Ruminant Research Network. AICC, Arusha, Tanzania, 7-11 December, 1992. ILCA, Addis Ababa, Ethiopia. Aden, T. (2003). Evaluations of local sheep breed under traditional management around rural Areas of Dire Dawa. An M.Sc Thesis presented to the School of Graduate Studies of Alemaya University of Agriculture, Dire Dawa, Ethiopia.128p. Alemayehu, M. (1985). Feed resource in Ethiopia. In: Proceedings of the second PANESA workshop on animal feed resources for small-scale livestock producers, Nairobi, Kenya,11–15 November 1985. pp. 35. Alemayehu, M. (2005). Feed resources base of Ethiopia: Status and opportunities for integrated development. Pp. 377 – 386. Proceedings of the 12th Annual Conference of the Ethiopian Society of Animal Production (ESAP). Addis Ababa, Ethiopia, August 12 – 14,2004, ESAP (Ethiopian Society of Animal Production). Alemayehu, M. (2003). Country Pasture /Forage Resource Profile, Ethiopia. http://www.fao.org/ag/AGP/AGPC/doc/counprof/Ethiopia/Ethiopia.Htm (Accessed on Feb. 2009). Amelmal, A. (2011). Phenotypic Characterization of Indigenous Sheep Types of Dawuro Zone and Konta special Woreda of SNNPR, Ethiopia. MSc thesis submitted to the department of animal science, school of graduate studies, Haramaya University, Ethiopia.

Arba Minch Zuria Wereda livestock and fishery office (2016) updated record. Arba Minch, Zuria woreda, Arba Minch, Ethiopia Arba Minch Zuria Wereda agriculture and natural resource office (2016) updated record. Arba Minch, Zuria woreda, Arba Minch, Ethiopia Arse, Gebeyehu, Feyisa, Hundessa, Gurmessa, Umeta, Merga, Muleta and Girma, Debele, (2013). Assessment on challenges and opportunities of goat farming system in Adami Tulu, Arsi Negelle and Fantale woreda of Oromia Regional State, Ethiopia. Afr. J. Agric. Res. 8(1):26-31 Assen, Ebrahim and Aklilu, Hailemichael (2012). sheep and goat production and utilization in different agro-ecological zones in Tigray, Ethiopia. Livestock Research For Rural Development 24 (1) 2012. Ashenafi, A. (2010). Performance evaluation and economic analysis of local sheep under farmers` condition supplemented with grass pea residues, wheat bran and noug Seedcake College of Agriculture and Environmental Sciences, Department of Animal Production and Technology, School of Graduate Studies Bahir Dar University Ayalew, W., EwBurns, J.M King and Rischkowsky, B. (2002). Practicalities of sustaining a goat crossbreeding programme in eastern Ethiopia. Ethiopia. J. Anim. Prod., 2: 71-85. Azage, T., Berhanu, G., Dirk H. (2006). Institutional arrangements and challenges in market oriented livestock agriculture in Ethiopia. ESAP (Ethiopian Society of Animal Production). In: Proceedings of the 14th Annual Conference of the Ethiopian Society of Animal Production (ESAP) held in Addis Ababa, Ethiopia, September 5–7, 2006. ESAP, Addis Ababa. pp 1-20. Belete, S. (2009). Production and marketing systems of small ruminants in Goma woreda of jimma zone, western Ethiopia. Msc. Thesis.Hawassa University, April, 2009, Awassa, Ethiopia.Pp 38-54 Berhan, A., J. and Van Arendonk (2006). Reproductive performance and mortality rate in MenzandHorro sheep following controlled breeding inEthiopia. Small Ruminant Research 63: 297–303. Berhanu, B., Aynalem, H. (2009). Factors affecting growth performance of sheep under village management conditions in the south western part of Ethiopia. Livestock Research for Rural Development, 21 (11); pp1-11.

Berhanu, Gebremedhin, Dirk, Hoekstra, Azage, Tegegne, Kaleb, Shiferaw and Aklilu, Bogale (2015). Factors determining household market participation in small ruminant production in the highlands of Ethiopia. Lives working paper, 31 (2) Cochran, W.G. (1977). Sampling techniques (3rd Ed.) New York: John Wiley & Sons. CSA (Central statistical Authority) (2003). Ethiopian Agricultural Sample Enumeration, 2001/2002. Southern Nation Nationalities and Peoples Region. Statistical report on livestock and farm implements. Addis Ababa, Ethiopia, 35- 43 CSA (Central Statistical Agency) (2011). Ethiopia Demographic and health Survey.55p CSA (Federal Democratic Republic of Ethiopia Central Statistical Agency) (2016/17) Agricultural Sample Survey Volume II Report on Livestock and Livestock Characteristics (Private Peasant Holdings) Deribe, G. T. (2009). On-farm performance evaluation of indigenous sheep and goats in Alaba, Southern Ethiopia m. Sc. Thesis Hawassa University, Ethiopia. Dibissa, N (2000). Sheep production on smallholder farmers in the Ethiopian Highlands-a farming system Approach. Ph.D dissertation. Humboldt University, Berlin, Germany. Dhaba, U., Belay, D., Solomon, D., Taye, T. (2013). Breeding Practices and Reproductive Performance of Traditionally Managed Indigenous Sheep and Goat in Ilu Abba Bora Zone of Oromia Regional State, Ethiopia Illu Abba Bora Zone Office of Agriculture and Rural Development, Mettu, Ethiopia Department of Animal Sciences, College of Agriculture, Jimma University, P.O. Box 307, Jimma, Ethiopia Endashew, Assefa (2007) Assessment on production system and marketing of goats at Dale woreda (Sidama Zone). MSc Thesis. University of Hawassa, Awassa, Ethiopia. EPA (Ethiopian Privatization Agency). (2002). (web accessed on March 29, 2009: (http://www.telecom.net.et/~epa/sectors/leather.html). Ermias, B. (2014) On- Farm Performance Evaluation Of Dorper Sheep Breed Crosses In WolaitaAndSiltie Zones, Southern Ethiopia Msc Thesis Msc Program In Tropical Animal Production And HealthAddis Ababa University Debrezeit, Ethiopia FAO (Food and Agricultural Organization of the United Nations) (1987). Master Land Use Plan, Ethiopian Range Livestock Consultancy Report Prepared for the Government of the Peoples Republic Development, Ethiopia. Technical Report.AG/ETH/82/020/FAO, Rome, 94pp. Morrilton, Arkansas. FAO (1995). FAOSTAT.PC Data Base.Food and agriculture Organization, Rome, Italy.

FAO (2004). Livestock sector brief: Ethiopia. Food and Agricultural Organization of the United Nations.Livestock Information, sector Analysis and Policy Branch. FAO, Rome. Fsahatsion, H., Aberra, M., Sandip, B. (2013).Traditional sheep production and breeding practice in Gamogofa Zone, Southern Ethiopia Department of Animal Production and Technology, Aksum University, Tigray, Ethiopia Hawassa University, PO Box 05, Hawassa, Ethiopia. International Journal of Livestock Production Research, 1(3):26 - 43. Fikrte, F. (2008). On-farm characterization of Blackhead Somali sheep breed and its production system in Shinile and Erer woreda of Shinile zone. M.Sc. Thesis presented to the school of Graduate Studies of Hramaya University of Agriculture, Dire Dawa, and Ethiopia.134p. Gemeda, D., S.J.S choeman, S.W.P. Cloete, G.F Jordan (2002) a. the influence of non-genetic factors on early growth traits in the Tygerhoek Merino lambs. Eth.J.Anim.Prod.2 (1)127-141. Getachew1, A. Haile2, M. Tibbo3, A. K. Sharma4, J. Sölkner5 and M. Wurzinger 5, 6 (2010). herd management and breeding practices of sheep owners in a mixed crop-livestock and a pastoral system of Ethiopia 1Debre Berhan Agricultural Research Center, P. O. Box 112, Debre Berhan, Ethiopia. 2International Livestock Research Institute (ILRI), Animal Genetic Resources, P. O. Box 5689, Addis Ababa, Ethiopia. 3International Center for Agricultural Research in the Dry Areas (ICARDA), P. O. Box 5466, Tel Hadya Aleppo, Syria. 4Indian Veterinary Research Institute, Mukteswar, Uttrakhand, India. 5BOKU - University of Natural Resources and Applied Life Sciences, Department of Sustainable Agricultural Systems, Vienna, Austria. 6International Livestock Research Institute (ILRI), P. O. Box 30709, Nairobi, Kenya. Getahun, L. (2008). Productive and Economic performance of Small Ruminant production in production system of the Highlands of Ethiopia.Ph.D dissertation. University of Hohenheim, Stuttgart-Hoheinheim, Germany

Girma, A. (2008). Reproduction in sheep and goats. AlemuYami and R.C. MERKEL (eds.). IN: Sheep and goat Production Hand Book for Ethiopia. Ethiopia Sheep and Goats Productivity Improvement Program (Esgpip), Addis Ababa, Ethiopia. pp. 57-72. Helen Nigussiea, Yoseph Mekashab, Solomon Abegaz c,Kefelegn Kebeded, Sanjoy Kumar, Pale (2015). Indigenous sheep production system in Eastern Ethiopia: implication for genetic improvement and sustainable use a Ambo University, Department of Animal Science, P.O.Box 19, Ambo, Ethiopia b,d,e Haramaya University, School of Animal and Range Sciences, P.O.Box 138, Dire Dawa, Ethiopia bInternational Livestock Research Institute,

Ethiopia cEthiopian Institutes of Biodiversity, P.O.Box 30726, Addis Ababa, Ethiopia Hundie, Demissu and Geleta, Gobena (2015). Assessment on production situation and breeding practices of ‘Horro sheep’ under traditional management in Horro Guduru and East Wollega Zones, West Ethiopia. Global Journal of Animal Breeding and Genetics, SSN: 2408-5502 Vol. 3(3), pp. 146-152.

ILCA (International Livestock Centre for Africa) (1987) Annual Report, ILCA, Addis Ababa, Ethiopia.pp 17-21

Jahnke, H. E. (1982). Livestock Production Systems and Livestock Development in Tropical Africa. Kieler Wissenschaftsverlag Vauk. Kiel, Germany.

Kassahun, A. (2000) Comparative performance evaluation of Horro and Menz sheep of Ethiopia under grazing and intensive feeding conditions eingereicht and derLandwirtschaftlichGärtnerischenFakultät der Humboldt-UniversitätzuBerlin (M.Sc.Animal Science, University of Wales, UK) Kearl, L.C. (1982). Nutrient Requirements of Ruminants in Developing Countries. Utah Agricultural Experimental Station, Utah State University, International Food Stuff Institute, Logan, USA.

Kosgey, I.S. (2004). Breeding objectives and breeding strategies for small ruminants in the Tropics. Ph.D. Thesis, Wageningen University, the Netherlands. (ISBW: 90-5808-990-8) Germany.271p.

Markos, T. (2006). Productivity and health of indigenous sheep breeds and crossbreds in the central Ethiopian Highlands. PhD thesis, Swedish University of Agriculture, Uppsala. McDonald, P., J.F.D. Greenhalgh, C.A. Morgan, R. Edwards, Liam Sincliar and Robert Wilkinson (2010). Animal nutrition. 7 ed. Prentice Hall, NY, USA Mekuriaw,, S, Mekuriaw Z, Taye, M, Yitayew, A, Assefa, H, Haile, A. (2012). Traditional management system and farmers’ perception on local sheep breeds (Washera and Farta) and their crosses in Amhara Region, Ethiopia. Livestock Res. Rural Develop. p.24, Article #4. Retrieved November 9, 2012, from http://www.lrrd.org/lrrd24/1/meku24004.htm Mekuriaw, S., Haile, A. (2014). Genetic parameter estimates for growth and reproductive traits of Sheep for genetic improvement and designing breeding program in Ethiopia: A Review. Open Access Library Journal, 1:e589. http://dx.doi.org/10.4236/oalib.1100589.

Mengesha, M., W., Tsega (2012). Indigenous Sheep Production in Ethiopia: Ethiopian Institute of Agricultural Research, DebreZeit Agricultural Research Center, Addis Ababa, Ethiopia Addis Ababa University, Faculty of Veterinary Medicine, Addis Ababa, Ethiopia Mengitie, T. (2008). On-farm performance of Washera sheep at Yilmanadensa and QuaritWoredas of the Amhara National Regional state. MSc Thesis submitted to Hawassa University, Ethiopia, Mesay, Yami, Bedada, Begna and Teklemedihin,Teklewold (2013). Enhancing the productivity of livestock production in highland of Ethiopia: Implication for improved on‐farm feeding strategies and utilization, Ethiopian Institute of Agricultural Research (EIAR), Assela, Ethiopia. MOA (Ministry of Agriculture) (2000). Agro ecological Zonation of Ethiopia. Addis Ababa, Ethiopia. Mukasa, M., E., D. Anindo, S. Sovani, A. Lahlou-Kassi, S. Tembely, J. E. O. Rege and R.L. Baker. (2002). Reproductive performance and productivity of Menz and Horro sheep lambing in the wet and dry seasons in the highlands of Ethiopia. Small Ruminant Research 45:261-271. Mukasa, M., E., and A., Lahlou-K. (1995). Reproductive performance and productivity of Menz sheep in the Ethiopian highlands.Small Rumin.Res.17: 167-177. Mukasa, M., E., Ephreim ,B., and Taddesse, T. (1986). Productivity of Indigenous sheep and goats in the Adaa woredat of Ethiopian highlands. Pp.81-87In: K. O. Adeniji and J. A .Kategile (eds). Proceedings of the workshop on the improvement of Small Ruminants in Eastern and Southern Africa, 18-22 August, 1986. Nairobi, Kenya. Niftalem, D. (1990). On farm study of reproductive and growth performance of Menzsheep around DebreBrahan area. An M.Sc Thesis presented to the School of Graduate Studies of Alemaya University of Agriculture, Dire Dawa, Ethiopia.103p. Okpebholo, Fidelis and Tyrell, Kahan (2007). Opportunities and Challenges for Developing Small Ruminant Systems. Cooperative Extension Program, College of Engineering Sciences, Technology and Agriculture, Florida A&M University, Tallahassee, Florida the Caprine Chronicle, Volume 22, Issue 4. Petmak, M. (1983). Primary productivity, nutrient cycling and OM turnover of tree plantation after agricultural intercropping practices in north east Thailand. PhD thesis, University of Philadelphia, Los Banos Philadelphia, pp: 228.

Ramesh., D, H R, Meena and K L, Meena (2012). Analysis of Small ruminant market system in different agro-climatic zones of Southern India. Vet. World, 2012, Vol.5 (5): 288-293 R.L (2002). Reproductive performance and productivity of Menz and Horro sheep lambing in the wet and dry seasons in the highlands of Ethiopia. Small Ruminant Research 45: 261–271. Rey, B., Lebbi S. H. B., Reynolds (1992). Small ruminant research and Development in Africa. Proceedings of the 1stBinnial Conference of the African Small ruminant Research ILCA (international Livestock Centre for Africa), Nairobi, Kenya, pp 425-437. Samson, L. and Frehiwot, M. (2014). Spatial analysis of cattle and shoat population in Ethiopia: growth trend, distribution and market access.http://www.springerplus.com/content/3/1/310. Samuel, M. (2005). Characterization of livestock production system; A case study of Yerer water shed, Adaa Liben woreda of east Showa, Ethiopia. An M.Sc Thesis presented to the School of Graduate Studies of Alemaya University of Agriculture, Dire Dawa, Ethiopia.184p.

Seare, T. (2007). Study on physical characteristics, management practices and performance of Abergelle and Degua sheep breeds feed on urea treated wheat straw with cats. Msc thesis Mekelle University the School of Graduate Studies Mekelle, Ethioia Sisay, Fikru and Kefyalew, Gebeyew* (2015). Sheep and Goat Production Systems in Degehabur Zone, Eastern Ethiopia:Challenge and Opportunities Jijiga University, Collage of Dry Land Agriculture, PO Box 1020, Jijiga, Ethiopia Solomon, G., AzageT., Berhanu G., Dirk H. (2010). Sheep and goat production and marketing systems in Ethiopia: Characteristics and strategies for improvement Improving Productivity and Market Success (IPMS) of Ethiopian Farmer Project, International Livestock Research Institute (ILRI), Addis Ababa, Ethiopia Solomon, A., Girma A., Kassahun A. (2008). Sheep and goat production system in the tropics. AlemuYami and R.C. Merkel (eds.). IN: Sheep and goat Production Hand Bookfor Ethiopia. Ethiopia Sheep and Goats Productivity Improvement Program (ESGPIP), Addis Ababa, Ethiopia. pp. 57-72. Solomon, A. (2007). In situ characterization of Gumuz sheep under farmers’ management in north western lowland of Amhara region. An. M.Sc. Thesis presented to the School of Graduate Studies of Alemaya University, Dire Dawa, and Ethiopia. 32p. Solomon, G. (2008). Sheep resource of Ethiopia, Genetic diversity and breeding strategy.PhD thesis,Wageningen University, The Netherlands With Summary in Dutch and English

SPSS (Statistical Procedures for Social Sciences) (2009). SPSS (Version 20). Statistical Procedures for Social Sciences (SPSS) INC. Chicago, Illinois, USA. Tadele, Mirkena. (2010). Identifying breeding objectives of smallholders/pastoralists and optimizing community-based breeding programs, for adapted sheep breeds in Ethiopia. Accept for award of Doctoral Thesis, University of Natural Resources and Live Sciences, Vienna, Boku.2010 pp 125. Taye, M., G., Abebe., S., Gizaw, S., Lemma, A. Mekoya and M., Tibbo. (2009). Growth performances of Washara sheep under smallholder management systems in Yilmanadensa and Quarit woredats, Hawassa, Ethiopia. pp 1-11. Tembely, S Lahlou-Kassi, A, Sovani, S, Rege J E O, Baker RL, Mukassa-Mugerwa, E. (1994). Small ruminant management practices and control of helminthosis under traditional production systems in cool Ethiopian highlands. Proceedings of Small Ruminant Research Networks, UICC, Kampala, Uganda, 5-9 December 1994. ILRI, Nairobi, Kenya. Tesfaye, G. (2008). Characterization of Menz and Afar Indigenous Sheep Breeds of Smallholders and Pastoralist for Designing Community Based Breeding Strategies in Ethiopia. An MSc Thesis presented to the School of Graduate Studies of Haramaya University, Dire Dawa, and Ethiopia. Tesfaye, Getachew, Solomon, Gizaw, Sisay, Lemma and Mengistie, Taye (2011) Breeding practices, growth, and carcass potential of fat-tailed Washera sheep breed in Ethiopia. Tropical Animal Health Production 43:1443–1448. Tessema, Z., Aklilu, A. and Ameha, S. (2003). Assessment of the Livestock Production System, Available Feed Resources and Marketing Situation in Belesa Woreda: A Case Study in Drought Prone Areas of Amhara Region. In: Proceedings of the 10th annual conference of the Ethiopian Society of Animal Production (ESAP) held in Addis Ababa, Ethiopia, August 21-23, 2003. Tibbo, M. (2006). Productivity and health of indigenous sheep breeds and crossbreds in the Central Ethiopian highlands.Faculty of Medicine and Animal Science Department of Animal Breeding and Genetics. A Ph.D Dissertation. Swedish University of Agricultural Sciences, Uppsala, Sweden, pp 1163. Tolera, A. (1990). Animal production and feed resource constraints in Welayita Sodo and the supplementary value of Desmodium intortum, Stylosanthes guianensis and Macratyloma

axillare when fed to growing sheep on a basal diet of Maize Stover. M.Sc. Thesis. Agricultural University of Norway, 133p. Tsedeke, K. (2007). Production and marketing of sheep and goats in Alaba, SNNPR.Msc thesis, Hawassa University. Hawassa, Ethiopia. Tsedeke, Kocho and Endrias, Geta (2011). Agro-ecologic mapping of livestock system in smallholder crop-livestock mixed farming of Wolaita and Dawuro woreda, Southern Ethiopia. Livestock Research for Rural Development 23 (3) 2011. Varvikko, T., G.M. Kidane, G. Geda (1993). Importance of early hay making in improving the standard of dairy cow feeding on small holder farms in the Ethiopian highlands. Proceedings: VIIth world congress on animal production, Edmonton, Canada, Wilsion, R.T. (1986c). Strategies to increase sheep production in east Africa. pp. 118-123. In: Small Ruminants Production in the Developing Countries. FAO Animal Production and Health Paper 58, Rome, Italy. Wilson, R.T. (1982). Husbandry, nutrition, and productivity of goat and sheep in tropical Africa. pp. 67-75. In: R.M. Gatenby and J.C.M.Trail (eds.).Small Ruminants Breed Productivity in Africa, ILCA (International Livestock Center for Africa), Addis Ababa, Ethiopia. Winrock International Institute for Agricultural Development (1992).Assessment of Animal Agriculture in Sub-Saharan Africa. Winrock: Morrilton, Arkansas. Workneh, A. (2006).Getting the Incentives Right: Concerns Associated with Expansion of Cattle Export Markets in Ethiopia. Eth. J. Anim. Prod. 6: 99-103. Workneh, A., J.Rowlands (2004), Design, execution and analysis of the livestock breed survey in Oromiya regional State, Ethiopia. Yadeta, N. (2016). Production and Reproductive Performances, Procedures Traits Preferences and Marketing Systems of Small Ruminants in Ada Barga and Ejere Woredats of West Shoa zone. MSc thesis submitted to the department of animal science, school of graduate studies, Jimma University, Ethiopia. Yenesew, Abebe, Solomon, Melaku and Azage, Tegegne, (2013). Assessment of sheep marketing system in Burie district, North Western Ethiopia.Wudpecker Journal of Agricultural Research 2(3): 97 - 102 Yilikal, T. (2015). Small ruminant production and marketing: constraints and opportunities in Chencha and Mirab Abaya woreda, Southern Ethiopia. Department of animal science, Arba Minch University, Ethiopia.

Yilmaz, O., H. Denk and D.Bayram. (2007). Effects of lambing season, sex and birth type on growth performance in Norduzlambs. Small Ruminant Research, 68 (3):336-339. Yisehak and Geert (2014). The Impacts of Imbalances of Feed Supply and Requirement on Productivity of Free-Ranging Tropical Livestock Units: Links of Multiple Factors 1Department of Animal Sciences, Jimma University, P.O. Box 307 Jimma, Ethiopia 2Laboratory of Animal Nutrition, Faculty of Veterinary Medicine, Ghent University, Heidestraat 19, 9820, Merelbeke, Belgium Zelalem, A. (2016). Review of the Reproductive Performances of Indigenous Sheep in Ethiopia Bonga Agricultural Research Center, P. O .Box 101, Bonga, Ethiopia Journal of Biology, Agriculture and Healthcare, Vol.6, No.9, Zewdu, E., Aynalem, H., Markos, T., A K Sharma, J Sölkner, M Wurzinger (2012). Sheep production systems and breeding practices of smallholders in western and south-western Ethiopia: Implications for designing community-based breeding strategies. Yabello Pastoral and Dry land Agriculture Research Centre, P.O.Box, 85, Yabello, Ethiopia Zewdu, E. (2008). Characterization Of Bonga and Horro Indigenous Sheep Breeds of Smallholders for Designing Community Based Breeding Strategies in Ethiopia. Master of Science Degree in Agriculture (Animal Genetics and Breeding) Haramaya University, Dire Dawa, Ethiopia.

7. ANNEX

7.1.Questionnaires

7.1.1 General information and socio economic aspects Date of interview ------Name of respondent ------2. Sex 1. Male 2. Female 3. Age ------4. Educational status 1. illiterate 2.Can read &write 3. elementary 4.junior secondary 5.Secondary 6.above secondary 5. Religion 1. Orthodox Christian 2. Muslim 3. Protestant 4. other (specify) -- 6. respondant’s Position in household 1. Household head 2. household member 3. employee 7. Number of people living in the house by age and sex 1. Children ------2. Adult------8. Land holding (in ha) 1. Crops land __ timad2. Fallow land__timad3.Grazingland____ timad 9. Trend in land holding 1. Decreasing 2. Increasing 3. Stable 10. Reason: _____ 11. Type of grazing land and ownership rank 1 Open grassland Own Rent Communal 2 Tree covered grassland 3 Bush/shrub grassland 4 Stone covered grassland 5 Swampy grassland 12. Income status 1. Low 2. Medium 3. High 13. What is your major farming activity? 1. Livestock production 2. Crop production 3. Mixed. On which do you depend more for? 1. Food ______2. Income source ______

14. Numbers of livestock kept livestock cattle sheep goats chickens donkeys mules horses local improved number Own Share origin Family Home born Purchase gift 15. Population trend in major livestock species increasing decreasing stable reason Sheep Cattle goats 16. Sheep Number by age group and breed. Number------breed------a. Male 6 months to 1 year ______------b. Female 6 months to 1 year ______------c. Male > 1 year (Intact) ______------d. Female > 1 year ______------e. < 6 months male lambs ______------f. < 6 months female lambs ______------g. Castrated male ______------

17. Major crop grown s.n crop season Main Short rain In quintal In quintal 1. Barley 2. Wheat 3. Teff 4. Sorghum 5. Maize 6. Bean 7. Pea 8. haricot bean 9. enset 10. banana 11. potato 18. List three most important crops During long rain (‘Meher’) During short rain (‘Belg’) 1. ______2. ______3. ______19. What are your major sources of income and estimate % contribution? Sources of income % contribution Rank in terms of priority

Livestock production sheep production Crop production Trade 20. Purpose of keeping sheep in the households1. Why you keep sheep? (Rank) 1=Sale (income source) 2=Meat 3=Manure 4=Sacrifices/rituals 5=Social and cultural functions 6=saving 7=Distribute benefits/risks with other animals 8= Wool/hair 11= Skin 10= Dowry

Who do the different tasks and decides on benefits obtained from sheep? activities men women boy girl Herding and/or tether of sheep Feeding and watering of sheep House cleaning Health care of the sheep Owner of the sheep Sell and purchase sheep 8.2. Sheep breeding and reproductive managements 1. Do you select your male and female animals for breeding purpose? 1=Yes 2=No 2. Do you have your own breeding male animals (ram)? 1=Yes 2=No 3. What are the common sources of breeding males for your flocks? 1. Own 2.neihbors 3. Others, specify

4. Mating and parturition seasons Sheep breed Mating and parturition seasons Oc Nov Dec Jan Fab Mar Apr Ma Jun July Aug Sep t y e Bong Service a time/concepti sheep on time Parturition Time Local Service sheep time/concepti on time Parturition Time Any other

6. Sheep reproductive performance particular In month/number male female breed Age at first mate Age at first parturition Parturition interval Average litter sizes (single, twin, triplets) in number Slaughter age Number of females mated in the past 12 months Number of females gave offspring in the past 12 months Total number of offspring born from mated females in the past 12 months Total number of offspring weaned out of total born in the past12 months 7. What are the reasons you justify that hinders fertility and reproduction of sheep? 1=Inadequate feed and water supply 2=Inconvenient climatic conditions 3=Disease and parasite burdens 4=Lack/shortage of breeding male 5=Drought in the area 6=others, specify 8. Do you fix age at first mating for the females? A. Yes b. No. Age ------If yes, which breed? 9. Do you fix age at first mating for the males? A. Yesb. No. age ---- If yes which breed 1.Average reproductive lifetime of ewe (in years) _2. Specify the breed3. Average number of lambing per ewes’ life time _____4. Specify the breed 10. Lambing pattern, occurrence of most births (Rank top three)1. January July 2. February August 3. March September 4. April October 5. May November 6. June December 8.2.1 Lamb rearing, castration and culling 1. Do you provide lambs additional feed to their mother’s milk until they begin grazing? 1=Yes 2=No 2. If yes, what types of feed resources and feeding? Feed types ______how feed ------3. Do you practice weaning of lambs? 1=Yes 2=No 4. If yes, when? Lambs______months

5. Do you practice castration of sheep? 1=Yes 2=No 6. If yes, why? 1=to fatten and sale 2=to control unwanted breeding 3=to tame them 4=others 7. At what age you usually castrate? __years (months) 8. How you select sheep for fattening? (Rank) 1=Conformation (height, length and appearance) 2=Breed (known local ecotypes) 3=Physical characteristics (color, horn, tail length and width, etc.) 4=Age 5=others, specify 9. If you practice select with physical characteristics, (Rank) 1=Color 2=Horn 3=Tail 4=Body length and height 5=others, specify 10. Do offer specific feeding and other management practices for castrated sheep? 1=Yes 2=No 11. If yes, what feeds and for how long? Feed types Duration ______12. What is the common method you castrate your sheep? 1=Local methods (using stone, stick, metal) 2=Burdizzo (OoFLR) 3=others, specify 13. Do you practice fattening of sheep for targeted market seasons and market places? 1=Yes 2=No 14. If yes, which season/months (Rank)? 1=New Year festival 2=Easter 3=Christmas 4=Meskel 5=Ed Al Fetir 6=Arefa 7=others, specify 15. is there and emerging opportunity of increased demand and incentive price for fattened sheep? 1=Yes 2=No 16. Do you practice culling of sheep from flock? 1=Yes 2=No 17. If yes, reasons for culling (rank)? 1=Old 2=Sick 3=Reproductive problem 4=Physical defect 5=Unwanted physical characteristics (black color) 7=others, specify 18. How sheep left from your flock over the last 12 months? 1=Sale 2=Death 3=Slaughter for home consumption 4=Theft 5=Predator 6=Gift 7=Share arrangements 8=others, specify 19. How you replace/own sheep left the household flock in various ways? 1=Home born 2=Share arrangements 3=Gift 4=Purchase 5=Not replace 6=others, specify 20. If you sale sheep for urgent income needs, which you prefer to sale? 1=Lambs 2=Rams 3=Ewes 4=Castrates 5=Others 21. How you sale young male sheep? 1=Sale all when reach to marketing age 2=Sale holding some for breeding 3=Sale holding some to castrate and fattening 4=others, specify