Genetic Study of Red Maasai Sheep and their Dorper Crossbred’s Performance in Sub-Saharan Africa Edwin Pancras Oyieng’ PhD DISSERTATION |JUSTUS - LIEBIG UNIVERSITY |2025 Institute of Animal Breeding and Genetics Justus-Liebig-University Gießen Genetic study of Red Maasai sheep and their Dorper crossbred’s performance in sub-Saharan Africa DISSERTATION For award of the doctoral degree (Dr. agr.) in the Faculty of Agricultural Science, Nutritional Science and Environmental Management at the Justus-Liebig University of Gießen Submitted by Edwin Pancras Oyieng’ From Nairobi, Kenya Gießen, 2025 Examiners ii With the approval of the Faculty of Agricultural Sciences, Nutritional Science and Environmental Management of the Justus-Liebig-University Gießen, Germany Dean: Prof. Dr. Klaus Eder Board of Examiners 1st reviewer: Prof. Dr. Sven König 2nd reviewer: Prof. Dr. Dr. Matthias Gauly Examiner: Prof. Dr. Wehrend Axel Examiner: Prof. Dr. Horst Brandt Chairperson: Prof. Dr. Lühken Gesine Date of Disputation: 8th July 2025 Supervisory team iii SUPERVISORY TEAM Prof. Dr. Sven König Professor of Animal Breeding and Genetics, Institute of Animal Breeding and Genetics, Justus-Liebig-University Gießen, Ludwigstraße 21 b, 35390 Gießen, Germany Prof. Dr. Dr. Matthias Gauly Professor of Animal and Veterinary Sciences, Faculty of Agricultural, Environmental and Food Sciences, Free University of Bozen – Bolzano, Universitatsplatz 5, 39100 Bolzano, Italy Prof. Dr. Raphael Mrode Professor of Animal Breeding and Genetics, Scotland’s Rural College, EH9 3JG Edinburgh, United Kingdom. Principal Scientist, Livestock Genetics Nutrition and Feed Resources Program, International Livestock Research Institute, P. O Box 30709-00100 Nairobi, Kenya Dr. Julie Ojango Senior Scientist, Livestock Genetics Nutrition and Feed Resources Program, International Livestock Research Institute, P. O Box 30709-00100 Nairobi, Kenya Dedication iv This thesis is dedicated to my sons (Edpancras and Pancras Oyieng’), Parents (Vincent and Consolata Ochieng’), Sisters (Nancy and Grace Ochieng’), Niece (Arozie) and nephew (Jabari). Thank you for your unconditional love, care and support during my PhD journey. Table of Contents v TABLE OF CONTENTS LIST OF TABLES ................................................................................................................... x LIST OF FIGURES ............................................................................................................... xii LIST OF ABBREVIATIONS .................................................................................................. xiv Summary .......................................................................................................................... xv Zusammenfassung ......................................................................................................... xviii CHAPTER 1: General Introduction ..................................................................................... 21 1.1 Livestock and livelihoods ................................................................................................... 21 1.2 Livestock and climate change ............................................................................................ 22 1.3 Livestock breeding for climate change .............................................................................. 22 1.4 Small ruminant genetic resources in sub-Saharan Africa .................................................. 23 1.5 Breeding strategies for sheep in sub-Saharan Africa ........................................................ 25 1.6 The Red Maasai sheep ....................................................................................................... 26 1.7 The Dorper sheep .............................................................................................................. 27 1.8 Objectives of this thesis ..................................................................................................... 28 References ............................................................................................................................... 29 CHAPTER 2: Evaluating reproduction traits in a crossbreeding program between indigenous and exotic sheep in semi-arid lands .................................................................................. 35 Abstract ................................................................................................................................... 36 2.1 Introduction ....................................................................................................................... 36 2.2 Material and Methods ....................................................................................................... 38 2.2.1 Study area .................................................................................................................. 38 2.2.2 Animal management and breeding ............................................................................ 38 2.2.3 Data structure and traits studied ............................................................................... 39 2.2.4 Data analysis .............................................................................................................. 41 2.2.4.1 Genetic parameter estimation ............................................................................ 41 2.2.4.2 Genetic gain ........................................................................................................ 43 Table of Contents vi 2.2.4.3 Rainfall index ...................................................................................................... 44 2.3 Results ............................................................................................................................... 44 2.3.1 Non-genetic factors .................................................................................................... 44 2.3.2 Genetic parameters ................................................................................................... 46 2.3.3 Genetic and phenotypic correlations ......................................................................... 48 2.3.4 Genetic trends and gain ............................................................................................. 49 2.3.5 Phenotypic trends and rainfall pattern ...................................................................... 51 2.4 Discussion .......................................................................................................................... 52 2.4.1 Non-genetic factors influencing reproduction of ewes ............................................. 52 2.4.2 Genetic influence on ewe fertility .............................................................................. 53 2.4.3 Phenotypic trends and genetic gain ........................................................................... 54 2.5 Conclusion ......................................................................................................................... 55 Ethics approval ........................................................................................................................ 56 Data and model availability statement ................................................................................... 56 Declaration of Generative AI and AI-assisted technologies in the writing process ................. 56 Authors ORCIDs ....................................................................................................................... 56 Declaration of interest ............................................................................................................ 56 Acknowledgement .................................................................................................................. 56 Financial Support ..................................................................................................................... 57 References ............................................................................................................................... 57 CHAPTER 3: Lamb survival and ewe longevity in a crossbreeding program between indigenous and exotic sheep in semi-arid lands ................................................................. 63 Abstract ................................................................................................................................... 64 3.1 Introduction ....................................................................................................................... 64 3.2 Materials and Methods ..................................................................................................... 65 3.2.1 Study area and flock management ............................................................................ 65 3.2.2 Data structure ............................................................................................................ 66 Table of Contents vii 3.2.3 Data analysis .............................................................................................................. 67 3.2.3.1 Survival analyses ................................................................................................. 67 3.2.3.2 Estimation of genetic parameters and genetic trends for lamb survival ............ 70 3.3 Results ............................................................................................................................... 70 3.3.1 Lamb survival ............................................................................................................. 70 3.3.2 Ewe longevity ............................................................................................................. 75 3.3.3 Genetic parameters and genetic trends for lamb survival ......................................... 77 3.4 Discussion .......................................................................................................................... 78 3.4.1 Factors affecting survival of lambs to yearling ........................................................... 78 3.4.2 Longevity of ewes ...................................................................................................... 81 3.5 Conclusion ......................................................................................................................... 82 Ethics approval ........................................................................................................................ 83 Declaration of generative AI and AI-assisted technologies in the writing process ................. 83 Authors ORCIDs ....................................................................................................................... 83 Declaration of interest ............................................................................................................ 83 Data availability statement ..................................................................................................... 83 Acknowledgement .................................................................................................................. 83 Author contributions ............................................................................................................... 84 Financial Support ..................................................................................................................... 84 References ............................................................................................................................... 84 CHAPTER 4: The impact of heat stress on growth and resilience phenotypes of sheep raised in a semi-arid environment of sub-Saharan Africa ............................................................ 89 Abstract ................................................................................................................................... 90 4.1 Introduction ....................................................................................................................... 90 4.2 Material and Methods ....................................................................................................... 92 4.2.1 Study area and animal management ......................................................................... 92 4.2.2 Animal and weather data ........................................................................................... 93 Table of Contents viii 4.2.3 Temperature-Humidity index (THI) ............................................................................ 94 4.2.4 Factors of variation on live weight gain ..................................................................... 95 4.2.5 Derivation of resilience phenotypes .......................................................................... 95 4.2.6 Fixed effects factors of variation on resilience phenotypes ...................................... 96 4.2.7 Genetic parameters of resilience indicators .............................................................. 97 4.3 Results ............................................................................................................................... 97 4.3.1 Fixed factors affecting growth ................................................................................... 97 4.3.2 Threshold for heat stress on growth .......................................................................... 98 4.3.3 Resilience phenotypes ............................................................................................. 100 4.3.4 Genetic parameters for resilience phenotypes ........................................................ 101 4.4 Discussion ........................................................................................................................ 102 4.5 Conclusion ....................................................................................................................... 106 Ethics approval ...................................................................................................................... 106 Declaration of generative AI and AI-assisted technologies in the writing process ............... 106 Authors ORCIDs ..................................................................................................................... 106 Declaration of interest .......................................................................................................... 107 Data availability statement ................................................................................................... 107 Acknowledgement ................................................................................................................ 107 Author contributions ............................................................................................................. 107 Financial Support ................................................................................................................... 107 References ............................................................................................................................. 108 CHAPTER 5: GENERAL DISCUSSION .................................................................................. 114 5.1 Reproduction efficiency .................................................................................................. 114 5.2 Pre- and post-weaning lamb survival .............................................................................. 115 5.3 Ewe’s length of productive life ........................................................................................ 116 5.4 Breeding for heat tolerance ............................................................................................ 116 5.5 Recommendations .......................................................................................................... 118 Table of Contents ix 5.6 Future considerations ...................................................................................................... 119 5.7 Conclusions ...................................................................................................................... 120 References ............................................................................................................................. 121 LIST OF PUBLICATIONS .................................................................................................... 129 ACKNOWLEDGEMENTS ................................................................................................... 130 FORMAL DECLARATION .................................................................................................. 131 List of Tables x LIST OF TABLES CHAPTER 2 Table 1: Descriptive statistics for the traits studied grouped by sheep breed ........................ 41 Table 2: Least square means (LSM) ± SE for age at first lambing (AFL), Lambing Interval (LI), ewes’ birth weight (EBWT), ewe’s weaning weight (EWWT), Litter birth weight (LBWT) and litter weaning weight (LWWT) for the sheep population ........................................................ 45 Table 3: Overall and within breed variance components and heritability estimates for age at first lambing (AFL), Lambing Interval (LI), ewes’ birth weight (EBWT), ewe’s weaning weight (EWWT), Litter birth weight (LBWT) and litter weaning weight (LWWT) grouped by sheep breed ....................................................................................................................................... 47 Table 4 Genetic and phenotypic ± SE correlations for age at first lambing (AFL), Lambing Interval (LI), ewes’ birth weight (EBWT), ewe’s weaning weight (EWWT), Litter birth weight (LBWT) and litter weaning weight (LWWT) grouped by sheep breed ..................................... 49 Table 5 Overall genetic gain ± SE for age at first lambing (AFL), Lambing Interval (LI), birth weight (EBWT), weaning weight (EWWT), Lambing weight at birth (LWB) and lambing weight at weaning (LWWT) from 2003 to 2020 grouped by sheep breed .......................................... 51 CHAPTER 3 Table 1 Summary statistics of lamb survival by breed ............................................................ 70 Table 2 Factors affecting lamb survival from birth to weaning and their risk ratio, mean age at failure (days) and Weibull parameter estimate (ρ)± Standard Error ...................................... 72 Table 3 Factors affecting lamb survival from weaning to yearling and their risk ratio, mean age at failure (days) and Weibull parameter estimate (ρ)± Standard Error .................................. 74 Table 4 Summary of statistics of ewe longevity from first lambing to culling at 2,190 days .. 75 Table 5 Factors affecting ewe longevity and their risk ratio, mean culling age (days) and Weibull parameter estimate (ρ)± Standard Error ................................................................... 77 Table 6 Genetic variance and heritability estimates ± standard error for lamb survival from birth to yearling by breed ........................................................................................................ 77 List of Tables xi CHAPTER 4 Table 1 Least square means (LSM) and SE (in parentheses) of fixed factors affecting growth ................................................................................................................................................. 98 Table 2 Least square means and SE (in parenthesis) of resilience phenotypes expressed as change in growth per unit increase in temperature-humidity index .................................... 100 Table 3 Variance components and heritability estimates (SE in parenthesis) of resilience phenotypes and liver weight gain ......................................................................................... 101 Table 4 Genetic correlations (above the diagonal) and phenotypic correlations (below the diagonal) of the traits ............................................................................................................ 102 List of Figures xii LIST OF FIGURES CHAPTER 1 Figure 1 Red Maasai ewe and its lamb………………………………………………………………..………………..26 Figure 2 Dorper ewe and its lamb…………………………………………………………………………………………27 CHAPTER 2 Figure 1 Genetic trends for age at first lambing by sheep breed from 2003 to 2020 (AFL = Age at first lambing, EBV=Estimated breeding values, DDDD=pure dorper, DDDR=75%Dorper- 25%Red Maasai, DDRR=50%Dorper-50%Red Maasai, RRRR=pure Red Maasai)………………50 …………………………………………………………………………………………………………………………..………….….50 Figure 2 Genetic trends for litter weight at weaning by sheep breed from 2003 to 2020 (LWWT = Litter weight at weaning, EBV=Estimated breeding values, DDDD=pure dorper, DDDR=75%Dorper-25%Red Maasai, DDRR=50%Dorper-50%Red Maasai, RRRR=pure Red Maasai)…………………………………………………………………………………………………………………..…………..50 Figure 3 Phenotypic trends for age at first lambing by sheep breed and rainfall patterns from 2003 to 2020 (DDDD=pure dorper, DDDR=75%Dorper-25%Red Maasai, DDRR=50%Dorper- 50%Red Maasai, RRRR=pure Red Maasai)…………………………………………………………………………...52 Figure 4 Phenotypic trends for litter weight at weaning by sheep breed and rainfall patterns from 2003 to 2020 (DDDD=pure dorper, DDDR=75%Dorper-25%Red Maasai, DDRR=50%Dorper-50%Red Maasai, RRRR=pure Red Maasai)……………………………………………52 ……………………………………………………………………………………………………………………………………………52 CHAPTER 3 Figure 1 Pre-weaning and post-weaning Kaplan – Maier survivor curves for lambs of different breeds (DDDD=pure dorper, DDDR=75%Dorper-25%Red Maasai, DDRR=50%Dorper-50%Red Maasai, RRRR=pure Red Maasai)…………………………………………………………………………………………75 ……………………………….……………………………………………………………………………………………………..…...75 Figure 2 Kaplan – Maier survivor curves of ewes from first lambing to culling for the different breeds (DDDD=pure dorper, DDDR=75%Dorper-25%Red Maasai, DDRR=50%Dorper-50%Red Maasai, RRRR=pure Red Maasai)…………………………………………………………………………………………76 ……………………………….………………………………………………………...76 List of Figures xiii Figure 3 Genetic trends for pre-weaning lamb survival from 2003 to 2022 of different breeds (DDDD=pure dorper, DDDR=75%Dorper-25%Red Maasai, DDRR=50%Dorper-50%Red Maasai, RRRR=pure Red Maasai)………………………………………………………………………………………………………78 ……………………………………….……….…………………………………………………….78 CHAPTER 4 Figure 1 The average monthly THI for each year in the study area……………………………………..99 Figure 2 Reaction norms for changes in weight in response to average temperature-humidity index for the population and each breed (DDDD=pure dorper, DDDR=75%Dorper-25%Red Maasai, DDRR=50%Dorper-50%Red Maasai, RRRR=pure Red Maasai)……………............……..100 List of Abbreviations xiv LIST OF ABBREVIATIONS AFL Age at first lambing ASAL Arid and Semi-Arid Lands CBBP Community based breeding program DDDD Pure Dorper breed DDDR 75% Dorper and 25% Red Maasai breed combination DDRR 50% Dorper and 50% Red Maasai breeding combination or F1 EBV Estimated breeding value EBWT Birth weight of the ewe EWWT Weaning weight of the ewe GHG Greenhouse gas LBWT Average birth weight in a litter for an ewe LI Lambing interval LWG1 Pre-weaning live weight gain LWG2 Post-weaning live weight gain LWWT Average weaning weight for the litter of an ewe RRRR Pure Red Maasai breed SRGR Small ruminant genetic resources SSA sub-Saharan Africa THI Temperature-Humidity Index Summary xv Summary The aim of this thesis was to study the performance of the Red Maasai sheep, the Dorper sheep and their crosses reared in a semi-arid environment. The study involved the analysis of reproduction parameters, survival of lambs and ewes length of productive life, the impact of heat stress on growth, and developed novel resilience phenotypes for the sheep population. The main breed groups in the breeding program comprised pure Dorper (DDDD), pure Red Maasai (RRRR), 75%Dorper-25%Red Maasai (DDDR) and 50%Dorper-50%Red Maasai – F1 (DDRR). R and ASReml-R software were used to evaluate reproduction parameters. Survival of lambs and the length of productive life of ewes were analysed using Cox and Weibull hazard models of the Survival Kit Version 6.12 software. Random regression models fitted with reaction norm functions were used to assess the impact of heat stress on growth and derive novel resilience phenotypes for growth in response to different levels of heat stress. Information generated through studying the reproductive performance of the flock is presented in Chapter 2. The pure Red Maasai sheep had significantly lower values for average age at first lambing (AFL), ewe birth weight (EBWT), ewe weaning weight (EWWT), litter birth weight (LBWT) and litter weaning weight (LWWT) compared to other breeds studied. The birth type (single or twins), sex of the lamb and parity in which the lambs were born significantly affected ewes’ birth and weaning weights. The overall heritability estimates of AFL (0.09±0.04) and LI (0.00±0.01) were not significant (P>0.05) while the heritability estimates for EBWT (0.38±0.04), EWWT (0.23±0.03), LBWT (0.19±0.03) and LWWT (0.09±0.02) were significant (P<0.05). The repeatability estimates were low for LBWT (0.25), LWWT (0.16) and that of lambing interval (LI) was near zero. Genetic and phenotypic correlations showed strong positive relationships between ewe and lamb weights. The Red Maasai had higher genetic and phenotypic correlations and genetic gains for the traits studied compared to the pure Dorper while the DDRR breed combination had a higher genetic gain among the crosses. LI had negative genetic correlations with LBWT and LWWT while AFL had positive genetic correlations with LBWT and LWWT. The phenotypic trends for AFL and LWWT showed a negative association with rainfall index over the years. Chapter 3 presents the results of pre- and post-weaning lamb survival to yearling, and ewes length of productive life. The pure Red Maasai lambs and ewes had better pre-weaning lamb survival rates and better productive life compared to the other breeds. Overall, 95% and 83% of lambs survived to weaning (90 days) and yearling (365 days), respectively. The Red Maasai lambs had the lowest mortality rates (2%) while Dorper lambs had the highest post- weaning mortality (24%) among the breeds. Lamb survival was significantly influenced by the Summary xvi season of birth, parity in which the lambs were born, birth type (single or twin) and birth weight. Single born lambs, those born during the wet season, offspring of multiparous ewes, and those having higher birth weights (>3kg) were associated with lower mortality risks. Ewe longevity was significantly affected by the breed, age at first lambing, parity and birth weight. Ewes of DDDR breed combination and those that had heavier birth weights (>3Kgs) had the highest risk of being culled, while ewes with a higher age at first lambing (>975 days old) and more than one lambing were less likely to be culled. Pre-weaning heritability estimates for survival (0.10–0.14) were higher than post-weaning estimates (0.01-0.05). The Red Maasai had the highest genetic gain for pre-weaning survival (-0.026) compared to DDDR (-0.018), F1 (-0.011), and Dorper (-0.012). General weakness, often due to poor nutrition, posed the highest risk for of lambs dying post-weaning (12.99 risk ratio), followed by diseases like enterotoxemia and sheep pox (6.006 risk ratio). The impact of heat stress on the growth of sheep and novel resilience phenotypes for growth are presented in Chapter 4. Heat stress, expressed as Temperature-Humidity Index (THI), significantly affected the growth of the sheep. The Red Maasai sheep had a higher tolerance for heat stress compared to the other breeds studied. The THI break points, when growth is affected by heat stress, were 78.75, 78.71, 78.42 and 77.93 for RRRR, DDDD, DDRR and DDDR respectively. At the THI break point, the growth rate declined at a rate of 0.06 Kgs, 0.09 Kgs, 0.05 Kgs and 0.15 in live weight gain per unit change in THI for RRRR, DDDD, DDRR and DDDR respectively. Random regression models fitted with reaction norm functions were used to develop two resilience phenotypes namely: Response and Stability. These resilience phenotypes were developed at THI 70 (representing low/no heat stress) and THI 85 (representing high heat stress). The breed, sex, type of birth, dams’ parity and season of birth significantly affected the stability of growth at low and high heat stress. Genetic correlations of resilience phenotypes at THI 85 with pre-weaning live weight gain (LWG1) were antagonistic and significant but not for post-weaning live weight gain (LWG2). Strong positive genetic and phenotypic correlations existed between response and its corresponding stability trait. The heritability estimates of resilience traits ranged from 0.12 for Response at THI 70 to 0.16 for Stability at THI 85. The better lamb survival and ewe longevity, and high tolerance to heat stress of the Red Maasai breed are an indication of their suitability for the harsh environment. Crossbreeding of the Red Maasai with Dorper has the potential to optimize growth and reproductive efficiency in the semi-arid environment. The moderate heritability estimates for resilience phenotypes in the population studied highlight opportunities for selective breeding Summary xvii to enhance resilience for growth under the changing climatic conditions. Context-specific improved animal management practices can increase the survival of lambs, improve their reproductive performance and reduce the impact of heat stress on growth. Summary xviii Zusammenfassung Das Ziel dieser Arbeit war die Untersuchung der Leistung von Roten Maasai-Schafen, Dorper-Schafen und ihren Kreuzungen, die in einer semi-ariden Umgebung gehalten werden. Die Studie umfasste die Analyse von Reproduktionsparametern, dem Überleben von Lämmern und Mutterschafen, der Länge des produktiven Lebens, dem Einfluss von Hitzestress auf das Wachstum und der Entwicklung neuartiger Resilienzphänotypen für die Schafpopulation. Die Hauptzuchtgruppen im Zuchtprogramm bestanden aus reinem Dorper (DDDD), reinem Roten Maasai (RRRR), 75% Dorper-25% Roten Maasai (DDDR) und 50% Dorper-50% Roten Maasai – F1 (DDRR). Die Software R und ASReml-R wurden zur Bewertung der Reproduktionsparameter verwendet. Das Überleben der Lämmer und die Länge des produktiven Lebens der Mutterschafe wurden mit Cox- und Weibull-Gefährdungsmodellen der Software Survival Kit Version 6.12 analysiert. Zufällige Regressionsmodelle, die mit Reaktionsnormfunktionen ausgestattet waren, wurden verwendet, um den Einfluss von Hitzestress auf das Wachstum zu bewerten und neue Resilienzphänotypen für das Wachstum als Reaktion auf unterschiedliche Hitzestressniveaus abzuleiten. Informationen, die durch die Untersuchung der Reproduktionsleistung der Herde gewonnen wurden, sind in Kapitel 2 dargestellt. Die reinen Roten Maasai-Schafe hatten signifikant niedrigere Werte für das durchschnittliche Alter beim ersten Lammen (AFL), das Geburtsgewicht der Mutterschafe (EBWT), das Absetzgewicht der Mutterschafe (EWWT), das Geburtsgewicht des Wurfs (LBWT) und das Absetzgewicht des Wurfs (LWWT) im Vergleich zu den anderen untersuchten Rassen. Der Geburtstyp (Einling oder Zwillinge), das Geschlecht des Lamms und die Parität, in der die Lämmer geboren wurden, beeinflussten die Geburts- und Absetzgewichte der Mutterschafe signifikant. Die Gesamtheritabilitätsschätzungen für AFL (0,09±0,04) und LI (0,00±0,01) waren nicht signifikant (P>0,05), während die Heritabilitätsschätzungen für EBWT (0,38±0,04), EWWT (0,23±0,03), LBWT (0,19±0,03) und LWWT (0,09±0,02) signifikant waren (P<0,05). Die Wiederholbarkeits-Schätzungen waren niedrig für LBWT (0,25), LWWT (0,16) und die des Lammintervalls (LI) war nahezu null. Genetische und phänotypische Korrelationen zeigten starke positive Beziehungen zwischen den Gewichten von Mutterschafen und Lämmern. Die Roten Maasai hatten höhere genetische und phänotypische Korrelationen und genetische Gewinne für die untersuchten Merkmale im Vergleich zu den reinen Dorper, während die DDRR-Rassenkombination den höchsten genetischen Gewinn unter den Kreuzungen aufwies. LI hatte negative genetische Korrelationen mit LBWT und LWWT, während AFL positive genetische Korrelationen mit LBWT Summary xix und LWWT aufwies. Die phänotypischen Trends für AFL und LWWT zeigten eine negative Assoziation mit dem Niederschlagsindex über die Jahre hinweg. Kapitel 3 präsentiert die Ergebnisse des Überlebens von Lämmern vor und nach dem Absetzen bis zum Jährling sowie der Länge des produktiven Lebens der Mutterschafe. Die reinen Roten Maasai-Lämmer und Mutterschafe hatten bessere Überlebensraten vor dem Absetzen und eine bessere produktive Lebensdauer im Vergleich zu den anderen Rassen. Insgesamt überlebten 95% und 83% der Lämmer das Absetzen (90 Tage) bzw. das Jährlingsalter (365 Tage). Die Roten Maasai-Lämmer hatten die niedrigsten Mortalitätsraten (2%), während Dorper-Lämmer die höchste Mortalität nach dem Absetzen (24%) unter den Rassen aufwiesen. Das Überleben der Lämmer wurde signifikant durch die Geburtssaison, die Parität, in der die Lämmer geboren wurden, den Geburtstyp (Einling oder Zwilling) und das Geburtsgewicht beeinflusst. Einzelgeborene Lämmer, solche, die während der Regenzeit geboren wurden, Nachkommen von mehrgebärenden Mutterschafen und solche mit höheren Geburtsgewichten (>3 kg) waren mit geringeren Mortalitätsrisiken verbunden. Die Langlebigkeit der Mutterschafe wurde signifikant durch die Rasse, das Alter beim ersten Lammen, die Parität und das Geburtsgewicht beeinflusst. Mutterschafe der DDDR- Rassenkombination und solche mit schwereren Geburtsgewichten (>3 kg) hatten das höchste Risiko, ausgesondert zu werden, während Mutterschafe mit einem höheren Alter beim ersten Lammen (>975 Tage) und mehr als einem Lammen weniger wahrscheinlich ausgesondert wurden. Die Heritabilitätsschätzungen für das Überleben vor dem Absetzen (0,10–0,14) waren höher als die Schätzungen nach dem Absetzen (0,01–0,05). Die Roten Maasai hatten den höchsten genetischen Gewinn für das Überleben vor dem Absetzen (-0,026) im Vergleich zu DDDR (-0,018), F1 (-0,011) und Dorper (-0,012). Allgemeine Schwäche, oft aufgrund schlechter Ernährung, stellte das höchste Risiko für das Sterben von Lämmern nach dem Absetzen dar (12,99 Risikoverhältnis), gefolgt von Krankheiten wie Enterotoxämie und Schafpocken (6,006 Risikoverhältnis). Der Einfluss von Hitzestress auf das Wachstum der Schafe und neue Resilienzphänotypen für das Wachstum sind in Kapitel 4 dargestellt. Hitzestress, ausgedrückt als Temperatur-Feuchtigkeits-Index (THI), beeinflusste das Wachstum der Schafe signifikant. Die Roten Maasai-Schafe hatten eine höhere Toleranz gegenüber Hitzestress im Vergleich zu den anderen untersuchten Rassen. Die THI-Breakpoints, bei denen das Wachstum durch Hitzestress beeinflusst wird, lagen bei 78,75, 78,71, 78,42 und 77,93 für RRRR, DDDD, DDRR bzw. DDDR. Am THI-Breakpoint nahm die Wachstumsrate um 0,06 kg, 0,09 kg, 0,05 kg und 0,15 kg an Lebendgewichtszunahme pro Einheit Änderung des THI für RRRR, DDDD, DDRR bzw. Summary xx DDDR ab. Zufällige Regressionsmodelle, die mit Reaktionsnormfunktionen ausgestattet waren, wurden verwendet, um zwei Resilienzphänotypen zu entwickeln, nämlich: Reaktion und Stabilität. Diese Resilienzphänotypen wurden bei THI 70 (geringer bis kein Hitzestress) und THI 85 (hoher Hitzestress) entwickelt. Die Rasse, das Geschlecht, der Geburtstyp, die Parität der Mutter und die Geburtssaison beeinflussten die Stabilität des Wachstums bei niedrigem und hohem Hitzestress signifikant. Genetische Korrelationen der Resilienzphänotypen bei THI 85 mit der Lebendgewichtszunahme vor dem Absetzen (LWG1) waren antagonistisch und signifikant, jedoch nicht für die Lebendgewichtszunahme nach dem Absetzen (LWG2). Starke positive genetische und phänotypische Korrelationen bestanden zwischen Reaktion und dem entsprechenden Stabilitätsmerkmal. Die Heritabilitätsschätzungen der Resilienzmerkmale reichten von 0,12 für Reaktion bei THI 70 bis 0,16 für Stabilität bei THI 85. Das bessere Überleben der Lämmer und die Langlebigkeit der Mutterschafe sowie die hohe Toleranz gegenüber Hitzestress der Roten Maasai-Rasse deuten auf ihre Eignung für die raue Umgebung hin. Die Kreuzung der Roten Maasai mit Dorper hat das Potenzial, das Wachstum und die reproduktive Effizienz in der semi-ariden Umgebung zu optimieren. Die moderaten Heritabilitätsschätzungen für Resilienzphänotypen in der untersuchten Population heben Möglichkeiten für selektive Zucht hervor, um die Resilienz für das Wachstum unter sich ändernden klimatischen Bedingungen zu verbessern. Kontextspezifische verbesserte Tierhaltungspraktiken können das Überleben der Lämmer erhöhen, ihre reproduktive Leistung verbessern und den Einfluss von Hitzestress auf das Wachstum reduzieren. Chapter One 21 CHAPTER 1: General Introduction 1.1 Livestock and livelihoods Livestock are a key global commodity for the provision of food, income, employment and risk insurance to mankind. Globally, livestock products contribute 18% to the kilocalorie consumption and 25% to the protein consumption (FAO, 2021). The livestock system is a significant global asset with an estimated value of at least USD.1.4 trillion and occupying about 45% of the earth’s surface area (Reid et al., 2008). The livestock systems in developing countries is rapidly changing as the demand for livestock products continues to increase due to increasing human population, rapid urbanization and increases in income (Thornton, 2010). In sub-Saharan Africa (SSA), livestock production contributes approximately 40% to the agricultural GDP (Enahoro et al., 2019; Sejian et al., 2015). Approximately 20% of the world’s cattle population, 25% of the world’s sheep and goat populations are found in SSA kept by 300 million livestock keepers, mostly concentrated in East and West Africa, with fewer in Southern and Central Africa (FAO, 2021). These livestock are reared in different production systems that vary from region to region. The systems can be broadly categorised as small-scale and large-scale production systems. The small-scale production systems include pastoralism, agropastoral and mixed smallholder farming. The large-scale systems include ranching, large-scale commercial farming, co-operative farming and state farming. The large-scale system still accounts for a relatively small proportion of agricultural output in SSA since the bulk of production occurs in the traditional small-scale system found in rural areas and Arid and Semi-Arid Lands (ASAL) (FAO, 2021). Majority of the rural households’ livelihoods are dependent on livestock farming while they own less than 2 Tropical Livestock Unit (TLU) and practice mixed crop-livestock farming (Otte et al., 2012). Manure and traction from livestock, are usually a non-monetized livestock inputs into household farming systems. Savings / asset accumulation and insurance represent another category of non-monetized services provided by livestock in traditional smallholder settings (Kayigema and Rugege, 2014; Lwelamira et al., 2010; Pender et al., 2004). In rural settings, livestock also serve as financial instruments because of the persistent absence of credit and financial markets in rural areas of developing countries (Pell et al.,2010). Livestock are usually sold to provide income for investment in other ventures such as land or small businesses and provide a source of income to meet planned and unplanned household needs such as school fees and hospital bills. Chapter One 22 1.2 Livestock and climate change Climate change is primarily caused by greenhouse gas (GHG) emissions that result in warming of the atmosphere with the livestock sector contributing 14.5% of global GHG emissions (Gerber et al., 2013; Solomon, 2007). As the global demand for livestock products increases, climate change threatens livestock production due to its impact on quality of feed crop and forage, water availability, animal and milk production, livestock diseases, animal reproduction, and biodiversity. Heat stress caused by climate change causes physiological and metabolic adjustments in animals resulting from thermoregulatory responses causing negative consequences in animal productivity and health (Renaudeau et al., 2012). Increased stress due to climate change weakens the immune systems of animals, making them more susceptible to infections (Nardone et al., 2010) hence increases the risk of livestock diseases outbreaks. Heat stress has also been shown to have negative effects on milky tied and growth in sheep (Tsartsianidou et al., 2021; Sánchez-Molano et al., 2020) and in cattle (Oloo et al., 2024). To cope with the challenge of climate change on livestock production, mitigation and adaptation strategies aimed at reducing the emission intensity of this sector are needed to meet the increasing demand for livestock products driven by population growth. One of the measures for mitigating GHG emissions is by improving feed and grazing management in ruminant animals since forage quality and digestibility affect enteric methane production. For example, harvesting forage (especially grass) for ensiling at an earlier stage to reduce lignification, chopping and grinding feeds (Hristov et al., 2013), addition of fats or fatty acids to the diets (Llonch et al., 2017a), adding feed additives such as electron receptors, ionophoric antibiotics and chemical inhibitors (Beauchemin et al., 2009), have shown a potential enteric methane reduction of about 5% per unit of fat protein corrected milk in dairy cattle (Knapp et al., 2014). Proper manure management can also help in mitigating the amount of GHG being emitted within the production system (Mohankumar Sajeev et al., 2018). The timing, quantity, and method of fertilizer applications are important factors which could help in optimizing biomass production and reduce soil greenhouse gas emissions (Grossi et al., 2019). 1.3 Livestock breeding for climate change Selecting and breeding animals that are resilient and adapted to the changing climate is one of the important ways of mitigating and adapting to the effects of climate change on livestock. Well adapted animals are able to maintain their productivity within their environment with Chapter One 23 little interventions. There is a direct link between greenhouse gas emission intensities and animal efficiency. The more productive the animal is, the lower the environmental impact will be per unit of product basis (Grossi et al., 2019). Both management quality and expression of full genetic potential are necessary to increase production efficiency. Genetic improvement is therefore an important tool to accumulate response to selection, and it can be used to reduce emissions (Cassandro, 2020). This can be done through breeding for improved efficiency which reduces the number of animals required to meet a given production level, improving functional traits such as fertility rates that can reduce wastage from the production system and selecting and breeding animals that are adaptable to the environment hence emitting less GHG. Breeding for more productive animals can result in lower nutrient requirements to achieve the same level of production. This reduces GHG emissions by diluting nutrient requirements for upkeep, allowing a given amount of production to be reached with fewer animals (Bell et al., 2011). Additionally, a more efficient animal will retain more dietary nitrogen protein and there will less nitrogen in faeces and urine (Gerber et al., 2013) . However, unless balanced by selection pressure on functional traits, selective breeding for higher productivity can harm animal health and welfare and lead to numerous unexpected consequences (Fraser et al., 2013; de Vries et al., 2011), as shown by the association between high milk production and an increased incidence of fertility problems and metabolic disorders such as ketosis in dairy cattle (Walsh et al., 2011) .Poor fertility requires more breeding animals in the herd to satisfy production requirements, as well as more replacements to keep the herd size stable, which increases GHG emissions. However, increasing reproductive pressure may increase the metabolic demands associated with pregnancy and breastfeeding, which could severely influence animal health and increase the risk of metabolic disorders, diminish immunological function, and thus reduce fertility (Llonch et al., 2017b) Therefore, developing selection indices that in-cooperates all traits of economic importance is necessary for breeding productive heat tolerant resilient livestock. 1.4 Small ruminant genetic resources in sub-Saharan Africa Small ruminant genetic resources (SRGR), sheep and goats are widely spread in sub-Saharan Africa (SSA) and are an important source of food, income, saving and socio-cultural values. SRGR make up 62% of domesticated ruminant livestock in SSA, with goats contributing 34% and sheep 28%. Approximately 90% of these SRGR are native breeds in pastoral, agro-pastoral, mixed crop-livestock systems. These production systems are characterized by different Chapter One 24 production goals, management strategies and constraints. The pastoral production system is mainly found in the ASALs where livestock is usually the main source of livelihood, breeds are mostly indigenous, and the animals are grazed. The agro-pastoral production system are found in semi-arid and in sub-humid areas. The production system is similar to the pastoral system however the pastoralists farm some crops to supplement their source of income and food. The mixed crop-livestock system is found in semi-arid, sub-humid and highlands areas. The systems has a densely populated a livestock population and are located near urban centres. The industrial system is a highly specialised system focusing on genetically improved and unform genotypes. The animals are kept under intensive management system (Kosgey, 2004). A larger proportion of goats (64%) and sheep (57%) are in the ASALs of SSA. Generally, goats out number sheep in most agro-ecological zones of SSA, except in the highlands where sheep are more prevalent. East and West Africa host the majority of SRGR, with East Africa leading in sheep numbers and West Africa in goats. Variations exist within regions; for instance, in southern Africa, South Africa has a higher sheep population than goats (FAO, 2021). SRGR in SSA are diverse consisting of about 61 sheep and 42 goat genotypes (Lebbie et al., 1996). It’s important to recognize that, aside from exotic and synthetic or composite breeds, the true genetic uniqueness of most major genotypes and their varieties or strains—especially indigenous ones—remains unclear. As observed, indigenous SRGR are often named after specific ethnic groups (e.g., Red Maasai sheep) or regions (e.g., West African Dwarf). Likewise, the categorization of these primary types relies predominantly on their physical or morphological traits. SRGR are particularly valued for their adaptability to harsh environments, rapid reproductive rates, and low maintenance requirements. For instance, indigenous breeds like the Red Maasai sheep and Small East African goats are well-suited to drought-prone areas due to their resistance to diseases and ability to thrive on low-quality forage (Mbuku et al., 2015; De Vries, 2008; Mugambi et al., 2005). The small ruminants are especially important to women who mostly manage them (Kariuki et al., 2022). Small ruminants also serve as a form of "insurance" for households, providing quick cash during emergencies or for social obligations such as weddings and school fees (Zonabend König et al., 2017). The small ruminants are known for their considerable ability to adapt and manage to survive and flourish in extremely hostile environments, including not only the arid zones but also the humid areas of West and central Africa. The weight and size of small ruminants in tropical Africa, are very much lower than in Chapter One 25 temperate countries, tend to reduce steadily from the traditional, arid production areas towards the humid zones in which dwarf breeds have developed. The domestic sheep (Ovis aries) is a very widespread small ruminant species raised primarily for its wool, meat, milk and hides. Sheep in Africa have been primarily a source of meat and much less important as milk and wool producers, than they are in Eurasia and Australasia (Da Silva et al., 2025). Africa hosts approximately a third of the world’s 1.3 billion heads of sheep majority of which are classified as indigenous (FAO, 2021). The indigenous African sheep genetic resources have been classified into two main groups, fat-tailed and thin-tailed sheep. The fat-tailed sheep are the most widely distributed, being found in mostly in the Arid and Semi-Arid Lands (ASAL) of North Africa, Eastern and Southern Africa (Muigai and Hanotte, 2013). 1.5 Breeding strategies for sheep in sub-Saharan Africa The main goal of animal breeding’s is to genetically improve livestock populations so that they produce more and improved quality animal products as a lower input cost. Animal breeding is a long-term activity that needs long term planning and commitment for it to achieve its goal. Through a structured breeding program, the breeding goal is defined and a consistent increment in the hereditary potential of livestock populations can be accomplished by continuously and selectively choosing the parents from one generation to the next (Falconer, 1996). In the midst of climate change, the breeding strategies for indigenous breeds should be aimed at not only improving the breed but also conserving it to maintain genetic diversity. In SSA, the most common and effective way of conserving indigenous breeds is in situ, where farmers and their community maintain their farm animal genetic resources as part of their livelihoods (FAO, 2007). Given the nature of the sheep production system in SSA, the breeding structures and strategies are very different compared to other species such as dairy cattle. Most flocks kept by farmers are either highly inbred or indiscriminately crossbred which leads to loss of important genetic material. This is mainly due to farmers lacking knowledge on sound animal breeding practices, lack of animal breeding policies and lack of record keeping infrastructures to aid in in selection and breeding. The breeding strategies that have been adopted for sheep in SSA are nucleus breeding programs, importation of exotic breeds and community-based breeding programs (CBBP). A nucleus breeding program is centralized and only few farmers participate in the program thereby requiring long-term commitment. Recording keeping in this system is easy and the Chapter One 26 genetic gain is higher (Kosgey and Okeyo, 2007). The recording keeping, genetic evaluation, delivery of genetic change breeding and selection happens at the centralised breeding. However, these breeding schemes has failed to sustainably provide the desired genetic improvement to small holders and engage the participation of the end users in the process (Haile et al., 2020). The importation of exotic or improved breeds is usually aimed at crossbreeding the indigenous breeds to upgrade them. However, this is mostly done without adequately assessing the viability and adaptability of these imported breeds and their consequent crosses to local production systems or conditions, and without a defined plan for determining the desired final genotype. Therefore, farmers end up practising indiscriminate crossbreeding with local populations resulting to genetic erosion of the local populations. Alternatively, in a CBBP, the selection and breeding is carried out by the farmers within the communities. The advantage of CBBP is that the farmer is actively involved throughout the process from inception through implementation. The success of CBBP is based upon proper consideration of farmers’ breeding objectives, available infrastructure, participation and ownership (Mueller et al., 2015; Wurzinger et al., 2011; Sölkner et al., 1998). CBBP have been shown to achieve genetic improvement for indigenous sheep breeds in ASAL regions(Haile et al., 2019, 2020). 1.6 The Red Maasai sheep The Red Maasai sheep breed is an indigenous breed native in East Africa (Figure 1). It is a fat- tailed transboundary indigenous breed mainly kept by Maasai pastoralists and neighbouring tribes in semi-arid and arid regions of Kenya and Tanzania. The breed is mainly used for meat, lard and other cultural practices for example, the fat of the Red Maasai is given after a woman has delivered a baby, after a cultural circumcision, and during sickness or injury. The breed is also used as a bride price where the Red Maasai is given to the mother-in-law (Liljestrand, 2012). The Red Maasai sheep is renowned for its adaptability to ASAL environments which are a low- input-output production system. The Red Maasai sheep is medium-sized, with a distinctive red-brown coat, polled head, and compact, resistance to gastrointestinal parasites, Figure 1 Red Maasai ewe and its lamb Chapter One 27 particularly Haemonchus contortus (Kwallah et al., 2008; Baker et al., 2004). The actual population size of the purebred Red Maasai is currently not available, and thus its risk status remains unknown. However, there is evidence to show that the Red Maasai is threatened due to the indiscriminate crossbreeding with the Dorper sheep that was imported from South Africa in the 1970s (Zonabend et al., 2014; Gibson and Pullin, 2005). Very few populations of pure Red Maasai can be found in research stations and with few farmers. In terms of production, the Red Maasai sheep has an average slaughter weight of 23.18 ± 0.20 kg at 9 months (Oyieng et al., 2022). A genetic analysis of reproduction and survival traits, and the effects of heat stress to the growth of the breed are yet to be studied and properly documented. Previous research on Red Maasai sheep has mainly been focusing on studying its resistance to Haemonchus contortus (Mugambi et al., 2005; Baker et al., 2004). 1.7 The Dorper sheep The Dorper is a synesthetic breed developed in South Africa (Figure 2). It was developed through the selective crossbreeding of the Dorset Horn rams and the Black Headed Persian ewes resulting into two variants of the breed one with a characteristic black head and white body and the other that is entirely white in colour. The Dorset Horn sheep is known for its quality meat production while the Black Headed Persian sheep is well adapted to ASAL environments (Cloete et al., 2000; Milne, 2000). The Dorper is popularly known for its high- quality carcass and relatively early maturing, and this made it a main breed of choice for meat in other African countries (de Waal and Combrinck, 2000). The breed is currently widespread across the globe, and research both on production and reproduction traits have been carried out (Cloete et al., 2000, 2021; Zishiri et al., 2013) When it was introduced to Kenya, no proper crossbreeding strategy was in place and farmers were not given instruction about how to maintain a continuous crossbreeding programme Figure 2 Dorper ewe and its lamb Chapter One 28 resulting into indiscriminate crossbreeding with indigenous breeds especially the Red Maasai. Within the country, it is nearly impossible to find purebred indigenous sheep in the field since most sheep have been crossed with the Dorper. The growth performance of the Dorper varies depending in the environment it is being reared with the highest birth weights of 4.19±0.09 kgs reported in South Africa (Cloete et al., 2021), low birth weight of 3.33±0.10kgs reported in Ethiopia (Goshme et al., 2021) and better birth weights of 3.8±0.02kgs in Kenya (Oyieng et al., 2022). The age of first lambing for the ewes ranges from 346 in South Africa (Cloete et al., 2000) to 786 days in Ethiopia (Goshme et al., 2021) while lambing interval ranges from 198 days in South Africa (Cloete et al., 2000) to 422 days in Kenya (Wanjala et al., 2023). A 32.1% mortality rate of the Dorper lambs before yearling have been reported in Ethiopia (Tesema et al., 2020). 1.8 Objectives of this thesis The overall objective of this thesis was to study the performance of the Red Maasai, Dorper and their crossbred combinations reared in an extensive production system in the semi-arid lands. The results of this study will not only help in informing the selection, breeding and conservation of the indigenous Red Maasai sheep but also in developing crossbreeding strategies between the Red Maasai and Dorper. More specifically, this thesis aims to: I. Evaluate the reproduction traits of the Red Maasai, Dorper and their crossbred combinations reared in an extensive production system in the semi-arid lands of Kenya. II. Assess the pre- and post-weaning to yearling survival of lambs and the length of productive life of ewes of the Red Maasai, Dorper and their crossbred combinations reared in an extensive production system in the semi-arid lands of Kenya III. Establish whether heat stress has an impact on the growth of Red Maasai, Dorper and their crossbred combinations reared in an extensive production system in the semi- arid lands of Kenya and develop novel resilience phenotypes for the sheep population. Chapter One 29 References Baker, R.L., Mugambi, J.M., J.O, A., Carles, A.B., Thorpe, W., 2004. Genotype by environment interactions for productivity and resistance to gastro-intestinal nematode parasites in Red Maasai and Dorper sheep. Animal Science 79, 343–353. Beauchemin, K.A., McAllister, T.A., McGinn, S.M., 2009. Dietary mitigation of enteric methane from cattle. CABI Reviews 1–18. doi:10.1079/PAVSNNR20094035 Bell, M.J., Wall, E., Simm, G., Russell, G., 2011. Effects of genetic line and feeding system on methane emissions from dairy systems. Animal Feed Science and Technology 166–167, 699–707. doi:https://doi.org/10.1016/j.anifeedsci.2011.04.049 Cassandro, M., 2020. Animal breeding and climate change, mitigation and adaptation. Journal of Animal Breeding and Genetics 137, 121–122. doi:https://doi.org/10.1111/jbg.12469 Cloete, S.W.P., Snyman, M.A., Herselman, M.J., 2000. Productive performance of Dorper sheep. Small Ruminant Research 36, 119–135. doi:https://doi.org/10.1016/S0921- 4488(99)00156-X Cloete, S.W.P., Thutwa, K., Scholtz, A.J., Cloete, J.J.E., Dzama, K., Gilmour, A.R., van Wyk, J.B., 2021. Breed effects and heterosis for weight traits and tick count in a cross between an indigenous fat-tailed and a commercial sheep breed. Tropical Animal Health and Production 53, 165. doi:10.1007/s11250-021-02612-7 Da Silva, A., Ahbara, A., Baazaoui, I., Jemaa, S. Ben, Cao, Y., Ciani, E., Dzomba, E.F., Evans, L., Gootwine, E., Hanotte, O., Harris, L., Li, M.-H., Mastrangelo, S., Missohou, A., Molotsi, A., Muchadeyi, F.C., Mwacharo, J.M., Tallet, G., Vernus, P., Hall, S.J.G., Lenstra, J.A., 2025. History and genetic diversity of African sheep: Contrasting phenotypic and genomic diversity. Animal Genetics 56, e13488. doi:https://doi.org/10.1111/age.13488 De Vries, J., 2008. Goats for the poor: Some keys to successful promotion of goat production among the poor. Small Ruminant Research 77, 221–224. doi:10.1016/j.smallrumres.2008.03.006 de Vries, M., Bokkers, E.A.M., Dijkstra, T., van Schaik, G., de Boer, I.J.M., 2011. Invited review: Associations between variables of routine herd data and dairy cattle welfare indicators. Journal of Dairy Science 94, 3213–3228. doi:https://doi.org/10.3168/jds.2011-4169 de Waal, H.O., Combrinck, W.J., 2000. The development of the Dorper, its nutrition and a perspective of the grazing ruminant on veld. Small Ruminant Research 36, 103–117. doi:https://doi.org/10.1016/S0921-4488(99)00155-8 Enahoro, D., Mason-D’Croz, D., Mul, M., Rich, K.M., Robinson, T.P., Thornton, P., Staal, S.S., 2019. Supporting sustainable expansion of livestock production in South Asia and Sub- Chapter One 30 Saharan Africa: Scenario analysis of investment options. Global Food Security 20, 114– 121. doi:https://doi.org/10.1016/j.gfs.2019.01.001 Falconer, D.S.., 1996. Introduction to quantitative genetics, 4th Edition. ed. Longman. FAO, 2021. Food and Agriculture Organization of the United Nations. FAOSTAT Statistical Database. FAO, 2007. The state of the world’s animal gentic resources for foos and agriculture. FAO, Rome. Fraser, D., Duncan, I.J.H., Edwards, S.A., Grandin, T., Gregory, N.G., Guyonnet, V., Hemsworth, P.H., Huertas, S.M., Huzzey, J.M., Mellor, D.J., Mench, J.A., Špinka, M., Whay, H.R., 2013. General Principles for the welfare of animals in production systems: The underlying science and its application. The Veterinary Journal 198, 19–27. doi:https://doi.org/10.1016/j.tvjl.2013.06.028 Gerber, Steinfeld, H., Henderson, B., Mottet, A., Opio, C., Dijkman, J., Falcucci, A., Tempio, G., 2013. Tackling climate change through livestock: a global assessment of emissions and mitigation opportunities. Gibson, J.P., Pullin, R.S. V, 2005. Conservation ofLivestock and Fish Genetic Resources: Joint Report of Two Studies Commissioned by the CGIAR Science Council. Rome, Italy. Goshme, S., Besufekad, S., Bisrat, A., Abebe, A., 2021. Reproductive and productive performance of Dorper sheep and their crossbreds with local highland sheep at Debre Birhan agricultural research center, Ethiopia. Livestock Research for Rural Development 33. Grossi, G., Goglio, P., Vitali, A., Williams, A.G., 2019. Livestock and climate change: impact of livestock on climate and mitigation strategies. Animal Frontiers 9, 69–76. doi:10.1093/af/vfy034 Haile, A., Getachew, T., Mirkena, T., Duguma, G., Gizaw, S., Wurzinger, M., Soelkner, J., Mwai, O., Dessie, T., Abebe, A., Abate, Z., Jembere, T., Rekik, M., Lobo, R.N.B., Mwacharo, J.M., Terfa, Z.G., Kassie, G.T., Mueller, J.P., Rischkowsky, B., 2020. Community-based sheep breeding programs generated substantial genetic gains and socioeconomic benefits. Animal 14, 1362–1370. doi:10.1017/S1751731120000269 Haile, A., Gizaw, S., Getachew, T., Mueller, J.P., Amer, P., Rekik, M., Rischkowsky, B., 2019. Community-based breeding programmes are a viable solution for Ethiopian small ruminant genetic improvement but require public and private investments. Journal of Animal Breeding and Genetics 136, 319–328. doi:https://doi.org/10.1111/jbg.12401 Chapter One 31 Hristov, A.N., Oh, J., Lee, C., Meinen, R., Montes, F., Ott, T., Firkins, J., Rotz, A., Dell, C., Adesogan, C., 2013. Mitigation of Greenhouse Gas Emissions in Livestock Production—A Review of Technical Options for Non-CO2 Emissions. Food and Agriculture Organization of the United Nations (FAO): Rome, Italy,. Kariuki, J., Galie, A., Birner, R., Oyieng, E., Chagunda, M.G.G., Jakinda, S., Milia, D., Ojango, J.M.K., 2022. Does the gender of farmers matter for improving small ruminant productivity? A Kenyan case study. Small Ruminant Research 206. doi:10.1016/j.smallrumres.2021.106574 Kayigema, V., Rugege, D., 2014. Women’s perceptions of the Girinka (one cow per poor family) programme, poverty alleviation and climate resilience in Rwanda. Knapp, J.R., Laur, G.L., Vadas, P.A., Weiss, W.P., Tricarico, J.M., 2014. Invited review: Enteric methane in dairy cattle production: Quantifying the opportunities and impact of reducing emissions. Journal of Dairy Science 97, 3231–3261. doi:https://doi.org/10.3168/jds.2013-7234 Kosgey, 2004. Breeding objectives and breeding strategies for small ruminants in the tropics. PhD Thesis, Wageningen University. Kosgey, I.S., Okeyo, A.M., 2007. Genetic improvement of small ruminants in low-input, smallholder production systems: Technical and infrastructural issues. Small Ruminant Research 70, 76–88. doi:10.1016/j.smallrumres.2007.01.007 Kwallah, A.B.O., Okeyo, A.M., Mburu, D., Hanotte, O., 2008. Genetic Diversity and Relationships Of Sheep Breeds Of Kenya : Preliminary Results And Evidence of Dilution Of the Indigenous Red Maasai 7–10. Lebbie, S.H.B., Yapi-Gnoare, C. V, Rege, J.E.O., Baker, R.L., 1996. Current developments in the management of small ruminant genetic resources: Sub-Saharan Africa. In Of IGA/FAO Round Table On The Global Management of Small Ruminant Genetic Resources. International Goat Association, Beijing, China. Liljestrand, J., 2012. Breeding practices of Red Maasai sheep in Maasai pastoralist communities. Master Thesis, Swedish University of Agricultural Sciences., Uppsala, Sweden. Llonch, P., Haskell, M.J., Dewhurst, R.J., Turner, S.P., 2017a. Current available strategies to mitigate greenhouse gas emissions in livestock systems: an animal welfare perspective. Animal 11, 274–284. doi:https://doi.org/10.1017/S1751731116001440 Chapter One 32 Llonch, P., Haskell, M.J., Dewhurst, R.J., Turner, S.P., 2017b. Current available strategies to mitigate greenhouse gas emissions in livestock systems: an animal welfare perspective. Animal 11, 274–284. doi:https://doi.org/10.1017/S1751731116001440 Lwelamira, J., Binamungu, H.K., Njau, F.B., 2010. Contribution of small scale dairy farming under zero-grazing in improving household welfare in Kayanga ward, Karagwe District, Tanzania. Livestock Research for Rural Development 22. Mbuku, S.M., Okeyo, A.M., Kosgey, I.S., Kahi, A.K., 2015. Optimum crossbreeding systems for goats in low-input livestock production system in Kenya. Small Ruminant Research 123, 55–61. doi:10.1016/j.smallrumres.2014.10.001 Milne, C., 2000. The history of the Dorper sheep. Small Ruminant Research 36, 99–102. doi:https://doi.org/10.1016/S0921-4488(99)00154-6 Mohankumar Sajeev, E.P., Winiwarter, W., Amon, B., 2018. Greenhouse Gas and Ammonia Emissions from Different Stages of Liquid Manure Management Chains: Abatement Options and Emission Interactions. Journal of Environmental Quality 47, 30–41. doi:https://doi.org/10.2134/jeq2017.05.0199 Mueller, J.P., Rischkowsky, B., Haile, A., Philipsson, J., Mwai, O., Besbes, B., Valle Zárate, A., Tibbo, M., Mirkena, T., Duguma, G., Sölkner, J., Wurzinger, M., 2015. Community-based livestock breeding programmes: essentials and examples. Journal of Animal Breeding and Genetics 132, 155–168. doi:https://doi.org/10.1111/jbg.12136 Mugambi, J.M., Audho, J.O., Baker, R.L., 2005. Evaluation of the phenotypic performance of a Red Maasai and Dorper double backcross resource population: natural pasture challenge with gastro-intestinal nematode parasites. Small Ruminant Research 56, 239–251. doi:https://doi.org/10.1016/j.smallrumres.2004.06.003 Muigai, A.W.T., Hanotte, O., 2013. The Origin of African Sheep: Archaeological and Genetic Perspectives. African Archaeological Review 30, 39–50. doi:10.1007/s10437-013-9129-0 Nardone, A., Ronchi, B., Lacetera, N., Ranieri, M.S., Bernabucci, U., 2010. Effects of climate changes on animal production and sustainability of livestock systems. Livestock Science 130, 57–69. doi:https://doi.org/10.1016/j.livsci.2010.02.011 Oloo, R.D., Ekine-Dzivenu, C.C., Mrode, R., Bennewitz, J., Ojango, J.M.K., Kipkosgei, G., Gebreyohanes, G., Okeyo, A.M., Chagunda, M.G.G., 2024. Genetic analysis of phenotypic indicators for heat tolerance in crossbred dairy cattle. Animal 18. doi:10.1016/j.animal.2024.101139 Otte, J., A. Costales, J. Dijkman, U. Pica-Ciamarra, T. Robinson, V. Ahuja, C. Ly, D. Roland-Holst, 2012. Livestock sector development for poverty reduction : an economic and policy Chapter One 33 perspective : livestock’s many virtues. Food and Agricultural Organization of the United Nations, Rome. Oyieng, E., Mrode, R., Ojango, J.M.K., Ekine-Dzivenu, C.C., Audho, J., Okeyo, A.M., 2022. Genetic parameters and genetic trends for growth traits of the Red Maasai sheep and its crosses to Dorper sheep under extensive production system in Kenya. Small Ruminant Research 206, 106588. doi:10.1016/j.smallrumres.2021.106588 Pender, J., Nkonya, E., Jagger, P., Sserunkuuma, D., Ssali, H., 2004. Strategies to increase agricultural productivity and reduce land degradation: evidence from Uganda. Agricultural Economics 31, 181–195. doi:https://doi.org/10.1016/j.agecon.2004.09.006 Reid, R.S., Galvin, K.A., Kruska, R.S., 2008. Global Significance of Extensive Grazing Lands and Pastoral Societies: An Introduction. In Fragmentation in Semi-Arid and Arid Landscapes: Consequences for Human and Natural Systems (eds. Galvin, K.A., Reid, R.S., Jr, R.H.B., Hobbs, N.T.). Springer Netherlands, Dordrecht, pp. 1–24. doi:10.1007/978-1-4020-4906- 4_1 Renaudeau, D., Collin, A., Yahav, S., De Basilio, V., Gourdine, J.L., Collier, R.J., 2012. Adaptation to hot climate and strategies to alleviate heat stress in livestock production. In Animal. pp. 707–728. doi:10.1017/S1751731111002448 Sánchez-Molano, E., Kapsona, V. V., Oikonomou, S., McLaren, A., Lambe, N., Conington, J., Banos, G., 2020. Breeding strategies for animal resilience to weather variation in meat sheep. BMC Genetics 21. doi:10.1186/s12863-020-00924-5 Sejian, V., Hyder, I., Ezeji, T., Lakritz, J., Bhatta, R., Ravindra, J.P., Prasad, C.S., Lal, R., 2015. Global warming: role of livestock. Climate change impact on livestock: Adaptation and mitigation 141–169. Sölkner, J., Nakimbugwe, H.N., Zárate, A.V., 1998. Analysis of determinants for success and failure of village breeding programmes. In In Proceedings of the 6th World Congress on Genetics Applied to Livestock Production, 12–16 January 1998, Armidale, Australia. Solomon, S., 2007. Climate change 2007-the physical science basis: Working group I contribution to the fourth assessment report of the IPCC. Cambridge university press. Tesema, Z., Deribe, B., Kefale, A., Lakew, M., Tilahun, M., Shibesh, M., Belayneh, N., Zegeye, A., Worku, G., Yizengaw, L., 2020. Survival analysis and reproductive performance of Dorper x Tumele sheep. Heliyon 6, e03840. doi:https://doi.org/10.1016/j.heliyon.2020.e03840 Thornton, P.K., 2010. Livestock production: Recent trends, future prospects. Philosophical Transactions of the Royal Society B: Biological Sciences. doi:10.1098/rstb.2010.0134 Chapter One 34 Tsartsianidou, V., Kapsona, V.V., Sánchez-Molano, E., Basdagianni, Z., Carabaño, M.J., Chatziplis, D., Arsenos, G., Triantafyllidis, A., Banos, G., 2021. Understanding the seasonality of performance resilience to climate volatility in Mediterranean dairy sheep. Scientific Reports 11. doi:10.1038/s41598-021-81461-8 Walsh, S.W., Williams, E.J., Evans, A.C.O., 2011. A review of the causes of poor fertility in high milk producing dairy cows. Animal Reproduction Science 123, 127–138. doi:https://doi.org/10.1016/j.anireprosci.2010.12.001 Wanjala, G., Kichamu, N., Cziszter, L.T., Astuti, P.K., Kusza, S., 2023. An On-Station Analysis of Factors Affecting Growth Traits of Pure Red Maasai and Dorper Sheep Breeds under an Extensive Production System. Animals 13, 300. doi:https://doi.org/10.3390/ani13020300 Wurzinger, M., Sölkner, J., Iñiguez, L., 2011. Important aspects and limitations in considering community-based breeding programs for low-input smallholder livestock systems. Small Ruminant Research 98, 170–175. doi:https://doi.org/10.1016/j.smallrumres.2011.03.035 Zishiri, O.T., Cloete, S.W.P., Olivier, J.J., Dzama, K., 2013. Genetic parameters for growth, reproduction and fitness traits in the South African Dorper sheep breed. Small Ruminant Research 112, 39–48. doi:10.1016/J.SMALLRUMRES.2013.01.004 Zonabend, E., Mirkena, T., Strandberg, E., Audho, J., Ojango, J., Malmfors, B., Okeyo, A.M., Philipsson, J., 2014. Breeding objectives for Red Maasai and Dorper sheep in Kenya – a participatory approach. In 10th World Congress on Genetics Applied to Livestock Production. pp. 1–3. Zonabend König, E., Strandberg, E., Ojango, J.M.K., Mirkena, T., Okeyo, A.M., Philipsson, J., 2017. Purebreeding of Red Maasai and crossbreeding with Dorper sheep in different environments in Kenya. Journal of Animal Breeding and Genetics 134, 531–544. doi:https://doi.org/10.1111/jbg.12260 Chapter Two 35 CHAPTER 2: Evaluating reproduction traits in a crossbreeding program between indigenous and exotic sheep in semi-arid lands E.Oyienga b, J.M.K Ojangob, M.Gaulyc, R.Mrodeb d, R.Dooso b e, A.M. Okeyo b, C.Kalinda f g , S.Königa a Institute of Animal Breeding and Genetics, Justus-Liebig-University Gießen, Ludwigstraße 21 b, 35390 Gießen, Germany b Livestock Genetics Nutrition and Feed Resources Program, International Livestock Research Institute, P. O Box 30709-00100 Nairobi, Kenya c Faculty of Agricultural, Environmental and Food Sciences, Free University of Bozen – Bolzano, Universitatsplatz 5, 39100 Bolzano, Italy d Animal and Veterinary Science, Scotland’s Rural College, EH9 3JG Edinburgh, United Kingdom e Animal Breeding and Husbandry in the Tropics and Subtropics, University of Hohenheim, Garbenstrasse 17, 70599 Stuttgart, Germany f Bill and Joyce Cummings Institute of Global Health, University of Global Health Equity, Kigali Heights, Plot 772 KG 7 Ave. P. O Box 6955, Kigali, Rwanda g School of Nursing and Public Health (SNPH), Discipline of Public Health Medicine, Howard College Campus, University of KwaZulu-Natal, Durban, South Africa Published in Animal, December 2024 https://doi.org/10.1016/j.animal.2024.101391 Chapter Two 36 Abstract Reproduction traits are important factors determining the efficiency of any sheep production system. This study evaluates the age at first lambing (AFL), lambing interval (LI), litter weight at birth (LBWT), litter weight at weaning (LWWT), birth weight of ewe (EBWT) and weaning weight of ewes (EWWT) in a crossbreeding program between the Red Maasai (RRRR) and Dorper sheep (DDDD) and their crosses, 75% Dorper (DDDR) and 50% Dorper (DDRR) breeds. All the traits significantly (P<0.05) differed across breeds and season of birth of the ewe. LBWT and LWWT were significantly affected by the sex of the lamb, type of birth of the lamb and parity in which the lambs were born in. AFL and LI had very high environment variances. Overall heritability estimates of AFL (0.09±0.04) and LI (0.00±0.01) were not significant from zero while the heritability estimates for EBWT (0.38±0.04), EWWT (0.23±0.03), LBWT (0.19±0.03) and LWWT (0.09±0.02) were significant (P<0.05). The RRRR had the highest genetic gain for all traits while the DDRR had a higher genetic gain among the crosses. LI had negative genetic correlations with LBWT (-0.53±0.08) and LWWT (-0.28±19.59) while AFL had positive genetic correlations with LBWT (0.27±0.46) and LWWT (0.31±0.34). The phenotypic trends for AFL and LWWT showed a negative and positive association, respectively, with the rainfall index over the years. With proper farm management, improved reproduction performance of ewes is possible by indirect selection using LBWT and LWWT for the Red Maasai, Dorper and their crosses within the semi-arid lands. Keywords: Genetic parameters, Lambing, Weights, Red Maasai, Dorper 2.1 Introduction Small ruminants can contribute to the efficient use of the semi-arid rangelands if harnessed for attributes that enable their optimal productivity and survival under the changing climates (Joy et al., 2020). Sheep and goats thrive through droughts more effectively than cattle, and flocks can be rebuilt more easily by the different communities following droughts (Haile et al., 2019; Muigai et al., 2017). The indigenous sheep and goat breeds raised by pastoral communities in the arid and semi-arid lands (ASAL) of East Africa are however reported to have low mature body weights and poor reproductive efficiency (Baker et al., 2004; Kosgey & Okeyo, 2007; Muigai et al., 2009; Wanjalaet al., 2023) limiting their potential to adequately contribute to improving livelihoods of pastoralist households. Despite indigenous sheep breeds being well adapted to their environment due to natural selection over time, targeted Chapter Two 37 structured breeding programs to improve their productivity are limited (Haile et al., 2020). Their adaptability can be transferred through crossbreeding with exotic breeds, however without a comprehensive crossbreeding strategy that includes selective improvement and conservation of the indigenous breeds, their genetic variation can be rapidly eroded. Conservation of indigenous breeds in situ is best achieved through profitable use of their inherent traits. When designing breeding programs for indigenous breeds, critical attention should be given to the reproductive performance of ewes as this influences the effectiveness and efficiency of sheep production (Tesema et al., 2020; Snyman et al., 1997). The ability of an ewe to regularly produce lambs over the years is considered a measure of the adaptability of the animal to the prevailing environmental conditions, and by extension, her long-term productivity (Kosgey et al., 2003). Since the 1970’s, in Eastern Africa there has been indiscriminate crossbreeding between exotic and indigenous breeds in attempts to increase the productivity of pastoral production systems and mutton market preferences (Ojango et al., 2023; Getachew et al., 2016; Zonabend König et al., 2016). However, these efforts and strategies have not been accompanied by concerted and organized selection programs, aimed at improving productivity in the existing indigenous sheep populations on a long-term basis (Deribe et al., 2021; Getachew et al., 2016; Kosgey et al., 2008). This, together with frequent and severe droughts followed by seasons of excessive rainfall (Calvin et al., 2023) have resulted in a rapid decline in the number of pure-bred indigenous sheep breeds, despite their renown tolerance to the changing climate and certain diseases such as parasitosis (Wanjalaet al., 2023). This underlines the need to selectively breed and conserve indigenous sheep populations while maintaining specific lines of their crossbreds Since 2003, the International Livestock Research Institute (ILRI) has implemented a selective crossbreeding program for purebred indigenous Red Maasai and their crosses with the Dorper, a composite breed introduced from South Africa (C. Milne, 2000), targeting improved growth performance while maintaining the inherent resistance to Haemonchus contortus of the Red Maasai. Positive genetic gains have been achieved in growth performance as reported by Oyieng et al., (2022). This paper presents reproductive parameters and the genetic progress in reproductive performance of ewes of the of Red-Maasai Sheep and their crosses with Dorper. The evaluation provides information on the fertility of the Red Maasai sheep and its crosses to guide crossbreeding strategies between indigenous and exotic breeds for optimum flock productivity in semi-arid areas. Chapter Two 38 2.2 Material and Methods 2.2.1 Study area The study is part of a selection breeding program for sheep in an arid environment run at the Kapiti Research Station and Wildlife conservancy (formerly known as Kapiti Plains Estate) in Machakos County, Kenya (https://www.ilri.org/research/facilities/kapiti-research-station- wildlife-conservancy). The conservancy, owned by the International Livestock Research Institute (ILRI), is situated between 1,650 and 1,900 meters above sea level and at -1.6 latitude and 37.1 longitude. The area receives an average annual rainfall of around 552 mm, and the temperature ranges between 22 — 27 °C. It has four distinct seasons: the long-wet season (March to May), the short-dry season (January and February), the long-dry season (June to September), and the short-wet season (October to December). The conservancy mostly covered in grasses and shrubs is home to both wild and domestic animals. Grazing pasture supply is often limited during the prolonged dry seasons. 2.2.2 Animal management and breeding The primary goals of the breeding program are to sustainably conserve the pure-bred Red Maasai sheep through within breed selection for growth, adaptability, and disease resistance and crossbreeding with the Dorper to increase the productivity and profitability of sheep raised in semi-arid and arid areas of Kenya. The Red Maasai sheep is characterized by thick red hair. Their growth characteristics have been described by (George Wanjala et al., 2023a; Oyieng et al., 2022). The crossbreeding program combines the faster growth rate and mutton- producing ability of the Dorper breed and the resilience of the Red Maasai breed to gastro- intestinal parasites. The main breed groups in the breeding program comprise pure Dorper (DDDD), pure Red Maasai (RRRR), 75%Dorper-25%Red Maasai (DDDR) and 50%Dorper- 50%Red Maasai (DDRR). The animals are reared in flocks comprising ewes that have lambed in the latest breeding cycle, and separate flocks of weaner and mature males, and weaner and mature females. The total flock size ranges from 1 300 to 1 500. This range is influenced by outflows due to culling and death, and inflows due to new births within the flock. Twice annually, in June and November, mating is conducted. The choice of the mating months is aligned to the rainfall seasons to enable lambing to take place at the best possible time of forage availability. When a lamb is three months old, it is weaned and separated from its mother. Every lamb born is weighed at birth, weaning (3 months), 6 months, 9 months, and one year of age. From nine months of age, when female lambs weigh at least 24 kgs, they are exposed to Rams for their first mating. Ewes are culled at the age of 6 years. Chapter Two 39 The selection of sires has a significant impact on mating and selection within the flock. Rams are chosen as sires based on three factors: their physical appearance, their dams' reproductive performance measured by the age at first lambing and intervals between lambings, and their genetic potential for growth to nine months using estimated breeding values. The number of Rams used is based on the number of female animals ready for mating and the desired number of offspring from each breed group required in the flock. The allocation of females to Rams in separate mating pens is managed to avoid inbreeding. Over a period of four weeks, each ram is given 20 to 30 ewes to mate. The performance of the Rams in the mating pens is monitored using coloured markers adhered to their underside that mark the rump of every ewe mounted. Inactive rams within a mating pen are replaced. Sires of the offsprings are recorded in line with the mating noted in each pen. To ensure genetic progress in the desired traits, 20% of the rams that were used in the previous mating season are replaced. The rams that have been culled but have good breeding values for growth and reproductive performance are sold to other sheep farmers as breeding animals while the rest are sold for mutton. Water is provided ad libitum for all the animals when in the pens during the night, and before setting out for grazing. When animals are grazing in the open fields, water is provided twice in the day. Scheduled vaccination against blanthrax, enterotoxaemia, and foot and mouth disease are provided annually, bi-annually and every 5 months, respectively. Anthelmintic treatment is given to animals before and after the rainy season based on the age and body condition of the animals. 2.2.3 Data structure and traits studied The pedigree used to construct the numerator relationship matrix of the animal models had 7 396 animals spanning 12 generations, including 206 sires and 2 591 dams. Lambing data of 2 056 ewes of the different breed types born between 2003 to 2020 was obtained for the analysis of reproduction of the sheep in an arid and semi-arid environment. The data was cleaned and boundaries determined based on the normal distribution for weight data. Weight records which were three standard deviations more or less from the mean were eliminated. Subsequently, only ewes with an age at first lambing between 330 and 1 080 days were included in the analyses. The lower limit was based on the possibility of including abortions that occurred in late pregnancy. The upper limit took care of the likelihood of a subsequent lambing event being misclassified as the first lambing due to an unrecorded first lambing. Lambing intervals below 240 days and above 600 days were also excluded. Following cleaning, records on 1 636 animals comprising 80% of the original data was used for the analyses. The traits evaluated were age at first lambing (AFL), lambing interval (LI), birth weight of the ewe Chapter Two 40 (EBWT), weaning weight of the ewe (EWWT), average birth weight in a litter for an ewe (LBWT) and average weaning weight for the litter of an ewe (LWWT). Descriptive statistics of the data used for the analysis of each trait is presented in Table 1. Chapter Two 41 Table 1: DescripZve staZsZcs for the traits studied grouped by sheep breed Breed group Trait DDDD DDDR DDRR RRRR Overall Age of first lambing (AFL) N 327 428 473 408 1636 Mean 739.66 756.85 727.29 702.65 731.35 SD 165.45 182.97 172.97 164.36 173.08 CV% 22 24 24 23 24 Min 411 398 422 363 363 Max 1082 1087 1087 1080 1087 Lambing interval (LI) N 236 433 592 630 1891 Mean 425.60 407.02 431.10 425.87 423.16 SD 91.80 88.14 90.94 91.89 91.12 CV% 22 22 21 22 22 Min 211 208 211 209 208 Max 600 599 600 599 600 Birth weight of the ewe (EBWT) N 327 428 473 408 1636 Mean 3.67 3.63 3.38 3.06 3.42 SD 0.65 0.61 0.59 0.50 0.63 CV% 18 17 17 16 18 Min 1.8 1.9 1.7 1.3 1.3 Max 5.8 5.1 5.1 5.1 5.8 Weaning weight of the ewe (EWWT) N 327 428 473 408 1636 Mean 17.23 16.79 16.38 14.26 16.12 SD 4.00 4.23 3.99 3.05 4.00 CV% 23 25 24 21 25 Min 7.1 4.8 6 5.7 4.8 Max 27 30 28 23.5 30 Litter birth weight (LBWT) N 550 836 1053 975 3414 Mean 3.51 3.71 3.63 3.20 3.50 SD 0.71 0.66 0.67 0.57 0.68 CV% 20 18 19 18 19 Min 1.5 2 1.5 1.6 1.5 Max 5.5 5.6 5.3 5.2 5.6 Litter weaning weight (LWWT) N 444 719 964 829 2956 Mean 17.19 17.21 16.76 15.33 16.53 SD 4.82 4.38 4.18 3.78 4.29 CV% 28 25 25 25 26 Min 5.2 6.34 5.76 4.34 4.34 Max 37.45 38.4 31.91 27.98 38.4 Breed code: DDDD= pure dorper, DDDR = 75%Dorper-25%Red Maasai, DDRR = 50%Dorper-50%Red Maasai, RRRR = pure Red Maasai 2.2.4 Data analysis 2.2.4.1 Genetic parameter estimation The Linear Model procedure of R (R Core Team, 2021) was used to evaluate the factors that influenced the AFL, LI, EBWT, EWWT, LBWT and LWWT. The fixed effects (that significantly influenced each trait (P<0.05) were included in the subsequent univariate and multivariate animal model analyses using ASReml-R 4.1 (Butler et al., 2018). Least-square means (LSM) of Chapter Two 42 different fixed effects groups for each model were calculated and contrasted using Tukey HSD Post Hoc test (Toothaker, 1993) at P<0.05. Different animal models were used to estimate the genetic parameters for the traits as the fixed factors affecting each trait differed. To analyse the AFL and LI, the model accounted for year-season of birth and breed effects. The breed, year of birth, season of birth and type of birth of the ewe were fitted to evaluate the EBWT and EWWT of the ewe. For LBWT and LWWT models, breed, sex of the lamb, type of birth of the lamb, year-season of lambing of the ewe, parity of the ewe, and age of the ewe nested within the parity were included as fixed effects. Additional effects included in the analyses of LBWT and LWWT were, type of birth, sex, and the breed of the lamb. Although 5 parities were represented in the data, parity effects were modelled with only four classes: parities 1, 2 and 3 separately, and parities 4 to 5 pooled into a fourth class. The effect of the Age of the ewe was fitted using a Legendre polynomial of order two. The univariate animal model 1 was used to estimate the heritability, additive, phenotypic and residual variances for AFL, EBWT and EWWT. 𝒚 = 𝑿𝜷 + 𝒁𝒂 + 𝒆 Model 1 Where 𝒚 is the vector of observations on the specific trait of the animal; 𝜷 and 𝒂 are vectors of fixed effects influencing the traits and direct additive genetic effects, respectively and e is the vector of residual errors. X is the incidence matrix relating observations to fixed effects; Z is the incidence matrix relating records to random animal effects. The vectors of random animal effects a and residual effects e were assumed to be normally distributed with a ~ N (0; 𝑨𝜎!" ) and e ~N (0; 𝐈𝜎#"), where A corresponds to the numerator relationship matrix, I correspond to the identity matrix, 𝜎!" is the additive genetic variance, and 𝜎#" is the residual variance. The repeatability animal Model 2 was fitted to estimate the heritability, additive, phenotypic and residual variances for LI, LBWT and LWWT. 𝒚 = 𝑿𝜷 + 𝒁𝒂 +𝑾𝒑𝒆 + 𝒆 Model 2 where 𝒚 is a measurement of individual trait, 𝜷 is the vector of the fixed effects in the model which included significant factors of variations for each in trait, 𝒂 is the solutions of random animal additive genetic effects, 𝒑𝒆 is the vector of random permanent environmental effects Chapter Two 43 and non-additive genetic effects and e is the vector of random residual effects. The vectors of random animal effects 𝒂, random permanent maternal environmental effects 𝒑𝒆, and residual effects e were assumed to be normally distributed with a ~ N (0; 𝑨𝜎!" ), pe ~ N (0; 𝑰𝜎$#" ) and e ~N (0; 𝐈𝜎#"), where A corresponds to the numerator relationship matrix, I correspond to the identity matrix, 𝜎!" is the additive genetic variance, 𝜎$#" is the permanent maternal environmental variance, and 𝜎#" is the residual variance. X is the incidence matrix relating observations to fixed effects; Z is the incidence matrix relating records to random animal effects. X, Z and W are incidence matrices relating observations to fixed, random animal and permanent environmental effects, respectively. Phenotypic and genetic correlations among traits were estimated using variances and covariances estimated from multivariate animal models. The following assumptions were made for the additive genetic effects in the multivariate models: / 𝒂𝟏 𝒂𝟐 𝒂𝒏 0 ~ 𝑵 45 𝟎 𝟎 𝟎 7 , 𝑨 ⊗: 𝜎𝟐𝒂𝟏 𝜎𝒂𝟏𝒂𝟐 𝜎𝒂𝟏𝒂𝒏 𝜎𝒂𝟏𝒂𝟐 𝜎𝟐𝒂𝟐 𝜎𝒂𝟐𝒂𝒏 𝜎𝒂𝟏𝒂𝒏 𝜎𝒂𝟐𝒂𝒏 𝜎𝟐𝒂𝒏 ;< where 𝑎) is the vector with additive genetic effects for trait i, 𝜎!$ " is the additive genetic variance of trait i, and 𝜎𝒂𝒊𝒂𝒋 is the genetic covariance between trait i and j. The residuals in the multivariate model were assumed to be: / 𝒆𝟏 𝒆𝟐 𝒆𝒏 0 ~ 𝑵 45 𝟎 𝟎 𝟎 7 , 𝑰 ⊗: 𝜎𝟐𝒆𝟏 𝜎𝒆𝟏𝒆𝟐 𝜎𝒆𝟏𝒆𝒏 𝜎𝒆𝟏𝒆𝟐 𝜎𝟐𝒆𝟐 𝜎𝒆𝟐𝒆𝒏 𝜎𝒆𝟏𝒆𝒏 𝜎𝒆𝟐𝒆𝒏 𝜎𝟐𝒆𝒏 ;< where 𝑒) is the vector with residuals for trait i, 𝜎#$ " is the residual variance of trait i, and 𝜎𝒆𝒊𝒆𝒋 is the residual covariance between trait i and j. The likelihood ratio test was used to test whether the heritability estimates differed significantly from zero by comparing the log likelihood of the tested model against a model without random animal genetic effects. 2.2.4.2 Genetic gain The overall genetic gain was the regression coefficient of estimated breeding values. The EBVs were averaged over each of the traits within year of birth, and resultant values regressed across the year of birth, model 3: 𝑦! = 𝑏+ + 𝑏,𝑥! Model 3 where 𝑦! is the average of EBV of ath year of birth; 𝑥! is the ath year of birth; 𝑏+ and 𝑏,, are the intercept and the linear regression coefficient, respectively. Chapter Two 44 2.2.4.3 Rainfall index An annual rainfall index was developed to model the variation in rainfall from 2003 to 2020 experienced on the ranch using the monthly rainfall data collected at the ranch weather station. Equation 1 was used to calculate the annual rainfall index. 𝑹𝑰𝒊 = 𝑿𝒊/ 𝝁 𝝈𝝁 Equation 1 Where 𝑹𝑰𝒊 is the raifall index for the ith year, 𝑿𝒊 is the average rainfall for the ith year, 𝝁 is the overall mean rainfall for the period under study (2003-2020) and 𝝈𝝁 is the standard deviation of the the overall mean rainfall. 2.3 Results 2.3.1 Non-genetic factors The least square means and their respective standard errors for significant (P<0.05) fixed effects on the traits studied are presented in Table 2. The type of birth of the ewe and the season of birth of the lamb did not significantly affect the LBWT and LWWT. The pure Red Maasai had a significantly (P<0.05) lower AFL, EBWT, EWWT, LBWT and LWWT compared to the other breeds. Among the crossbreds, the DDRR has significantly (P<0.05) lower AFL, EBWT, EWWT, LBWT and LWWT compared to the DDDR breed. The DDDR breed had the lowest lambing interval across all breeds. Single born ewes were significantly (P<0.05) heavier than ewes born as twins. Male and single birth lambs were also significantly (P<0.05) heavier than female and twin lambs for both LBWT and LWWT. The LBWT increased significantly (P<0.05) as the parity of the ewe increased. However, the difference in LWWT was only significant (P<0.05) between litters born in parity one and those born in parity four. Table 2: Least square means (LSM) ± SE for age at first lambing (AFL), Lambing Interval (LI), ewes’ birth weight (EBWT), ewe’s weaning weight (EWWT), Lider birth weight (LBWT) and lider weaning weight (LWWT) for the sheep populaZon Fixed effects AFL (days) LI (days) EBWT (kgs) EWWT (kgs) LBWT (kgs) LWWT (kgs) (N) LSM±SE (N) LSM±SE (N) LSM±SE (N) LSM±SE (N) LSM±SE (N) LSM±SE Breed *** ** *** *** *** *** DDDD (327 )806.66±7.56a (236) 424.60±5.95ab (317) 3.32±0.04a (265) 15.01±0.26a (838) 3.51±0.03a (681) 16.30±0.25a DDDR (428) 784.84±6.56a (433) 407.02±4.40a (428) 3.32±0.03a (413) 14.90±0.23a (1387) 3.56±0.03a (1138) 16.30±0.21a DDRR (473) 752.65±6.52b (592) 428.10±3.95b (467) 3.10±0.03b (456) 14.12±0.23b (1736) 3.44±0.03ab (1508) 15.80±0.20ab RRRR (408) 735.28±6.58b (630) 424.87±3.91b (411) 2.72±0.04c (378) 12.14±0.24c (1430) 3.03±0.03b (1232) 14.50±0.21b Type of birth (ewe) *** *** Single - - (3340) 3.45±0.02a (3268) 15.51±0.10a - - Twin - - (188) 2.78±0.06b (178) 12.50±0.35b - - Sex (lamb) *** *** Male - - - - (1644) 3.47±0.02a (1401) 16.30±0.18a Female - - - - (1770) 3.30±0.02b (1555) 15.20±0.18b Type of birth (lamb) *** *** Single - - - - (3179) 3.68±0.01a (2763) 17.00±0.11a Twin - - - - (235) 3.09±0.04b 1193) 4.50±0.30b Parity *** *** One - - - - (1602) 3.06±0.02a (1357) 14.50±0.19a Two - - - - (895) 3.39±0.03b (793) 16.20±0.20b Three - - - - (571) 3.49±0.03c (500) 16.50±0.23b Four - - - - (346) 3.60±0.04d (306) 15.90±0.28b Season of birth (ewe) *** *** *** *** * Long wet (446) 817.80±9.79a (633) 425.52±6.86a (498)13.42±0.35a (509) 3.38±0.04ab (470) 14.80±0.32a Short wet (640) 730.94±7.16b (731) 414.56±3.68a (585) 16.45±0.24b (667) 3.34±0.03a (606) 16.00±0.22b Long dry (350) 783.23±8.94a (353) 407.62±4.85bc (378) 12.41±0.27a (384) 3.39±0.03ab (361) 16.10±0.25b Short dry (200) 747.83±7.04b (174) 437.47±3.48b (233) 13.96±0.26a (229) 3.43±0.03b (167) 16.00±0.23b ***, ** , * means the fixed effect is significant at P<0.001, P<0.01 and P<0.05 respec^vely Breed code: DDDD= pure dorper, DDDR = 75%Dorper-25%Red Maasai, DDRR = 50%Dorper-50%Red Maasai, RRRR = pure Red Maasai Chapter Two 46 2.3.2 Genetic parameters The phenotypic, additive and residual variances, and heritability estimates for the traits by breed group and by population are presented in Table 3. The pure Dorper breed had higher heritability estimates for AFL, EBWT, and EWWT compared to the other breeds. Heritability estimates for LWWT were generally lower than that for LWB across all breeds. The heritability estimates for BWT, WWT and LBWT were all significant (P<0.05) while the heritability estimates for lambing interval were not significant (P>0.05). Overall, the heritability estimates for the ewe’s birth weights were higher compared to that of their weaning weights, the same pattern also being observed between their corresponding lambing birth and weaning weights. The heritability estimates of EBWT and EWWT across the breeds were higher (ranging between 0.28 and 0.42) than those LBWT and LWWT (ranging between 0.00 to 0.18). The maternal effects on LBWT and LWWT was 0.19 and 9.05 respectively which accounted for 76% of the total residual variance. The overall repeatability estimates for LBWT and LWWT were 0.25±0.03 and 0.16±0.03 respectively. The repeatability estimate for LI was almost zero. Chapter Two 47 Table 3: Overall and within breed variance components and heritability esZmates for age at first lambing (AFL), Lambing Interval (LI), ewes’ birth weight (EBWT), ewe’s weaning weight (EWWT), Lider birth weight (LBWT) and lider weaning weight (LWWT) grouped by sheep breed Breed 𝜎!" 𝜎#" 𝜎$" ℎ"± SE P Value DDDD AFL 15619.60 3377.745 12241.85 0.22±0.16 0.15 LI 6835.09 0.02 6835.07 0.00±0.09 0.09 EBWT 0.33 0.14 0.19 0.42±0.03 0.00 EWWT 9.39 3.44 6.49 0.33±0.02 0.01 LBWT 0.36 0.05 0.231 0.13±0.04 0.03 LWWT 16.59 0.00 16.59 0.00±0.00 0.34 DDDR AFL 14424.75 1701.17 12723.58 0.12±0.14 0.14 LI 5936.84