Genetic Evaluation of Reproductive Traits of Ethiopian Sheep Breeds Under Community‐Based Breeding Programmes

ABSTRACT

The evaluation of breeding schemes against established objectives and selection traits is essential for assessing the performance, outputs, and overall impacts of breeding programmes. In Ethiopia, most Community‐Based Breeding Programmes (CBBPs) have prioritised growth traits, particularly live weight, as the main selection criteria. However, since productivity relies on both reproductive and growth traits, it is critical to evaluate how these traits are evolving to make necessary adjustments in management practices and breeding schemes. This study considered five indigenous sheep breeds (Menz, Semein, Horro, Bonga and Doyogena), managed under CBBPs since 2009. Fixed effects for reproductive traits were estimated using the GLM procedures of SAS 9.4. Genetic parameters were estimated for all traits using the restricted maximum likelihood (REML) method with WOMBAT software, employing a multivariate repeated model, except for age at first lambing (AFL), which was analysed using a non‐repeated multivariate model. Significant effects (p < 0.001) were observed for year of birth, breed of ewe, parity and birth season across all traits in the breeds studied. A general trend of improvement in litter size at birth (LSB), total litter weight at birth (TLWB), litter size at weaning (LSW), total litter weight at weaning (TLWW) and annual reproductive rate (ARR) was noted with increasing ewe parity until the seventh parity, followed by a decline thereafter. Direct heritability estimates for the traits according to the ewe breeds ranged from 0.03 to 0.25 for LSB, 0.02 to 0.16 for LSW, 0.08 to 0.21 for TLWB, 0.07 to 0.22 for TLWW, 0.03 to 0.19 for LI, 0.08 to 0.32 for ARR and 0.15 to 0.36 for AFL. Estimates of direct heritability and repeatability varied by breed and location, generally falling within small to medium ranges. Moderate to high genetic correlations were found between TLWW and other traits suggesting that selection for TLWW may significantly influence reproductive performances across most sheep breeds, with the exception of Menz sheep. The variations in genetic estimates across different breeds and locations indicate that genetic influences may vary depending on the specific context. Moderate to high genetic correlations between TLWW and other reproductive traits suggest that prioritising selection for TLWW could have a significant positive impact on reproductive performance across most sheep breeds, though the Menz breed may not exhibit the same expected benefits. These findings emphasise the need to integrate genetic selection with effective management practices tailored to each breed's specific needs, recommending the culling of unproductive ewes after the seventh parity to enhance the sustainability and productivity of CBBPs in Ethiopia.

1. Background

Ethiopia's sheep population was estimated at 42.91 million (CSA ²⁰²¹) and is distributed from the highlands' cool alpine climate to the lowlands' arid pastoral areas (Tibbo 2006). In Ethiopia, sheep are reared mainly by smallholder farmers and serve as a source of immediate cash needs, food, means of risk mitigation during crop failures, saving and investments in addition to other socio‐economic and cultural benefits (Legese and Fadiga 2014). Despite the large number of adapted sheep breeds and populations, the productivity of the stock remains low. Since the early 1960s, several efforts have been made to improve the genetic potential of the indigenous sheep population through importation and crossing with exotic genotypes (Tibbo 2006). However, this genetic improvement strategy yielded no significant effects on sheep productivity. The major shortcomings of these improvement programmes are that they do not fully address the farmers' preferences under low‐input systems and also fail to consider the indigenous practices (Gizaw et al. 2013; Haile et al. ²⁰¹⁸).

Opting towards community‐based breeding programme (CBBP) is being suggested as an attractive option for the genetic improvement of small ruminants in low‐input, smallholder production systems (Gizaw et al. 2013; Haile et al. ²⁰²⁰). In Ethiopia, Community‐Based Breeding Programmes (CBBPs) were initiated in 2009 by the Ethiopian National Agricultural Research System and its partners. The programme initially targeted four sheep breeds (Menz, Bonga, Horro and Afar) and later expanded to other sheep breeds (Doyogena, Semein, Washera, Dawuro) and many more breeds in Ethiopia and other African and Asian countries (Gizaw et al. 2013; Haile et al. ²⁰²⁰). Unlike the conventional centralised, top‐down breeding approaches, CBBP takes into account farmers' needs, views and active participation at every stage from planning to operation of the breeding programme (Gizaw et al. 2013; Mueller et al. ²⁰¹⁵; Haile et al. ²⁰²⁰). The major objectives of the breeding programmes were to maximise the rate of genetic progress for live weight traits. Promising achievements of CBBP in productivity improvement have been reported, including significant improvement of annual income of smallholder farmers from sale of sheep, and significant improvement of live body weight, body conformation and appearance of the animal (Gutu et al. 2015; Abebe et al. ²⁰²⁰; Haile et al. ²⁰²⁰; Areb et al. ²⁰²¹; Habtegiorgis et al. ²⁰²²). Conversely, genetic selection for increased body weight in sheep is inherently associated with higher nutritional demands, which may be unsustainable in an environment with limited feed resources. Thus, strategic improvement of local feed resources in conjunction with genetic improvement is crucial for the sustainability of the breeding programme in such anenvironment.

Reproductive efficiency is one of the most important factors affecting the profitability of small ruminants, and improvement of these traits leads to more efficient lamb production (Yavarifard et al. 2015). Several studies have reported that improvement of reproductive traits (fertility, lambing frequency, lambing interval, litter size) has more economic impact than improvement of growth rate and live weight traits in small ruminants (Yavarifard et al. 2015; Saghi and Shahdadi ²⁰¹⁷).

A robust rationale for the evaluation of breeding programmes lies in the critical need to assess both performance and impact against defined breeding objectives and selection traits. CBBPs have successfully prioritised growth traits, specifically live weight, resulting in notable improvements in live body weight, animal conformation and increased income for smallholder farmers. However, since productivity depends on the integration of both reproductive and growth traits, a thorough analysis of these combined traits is essential to guide modification in breeding and management practices. Despite improvements in growth traits, there is a significant gap in studies addressing the phenotypic and genetic trends for reproductive traits across diverse Ethiopian sheep breeds. This study aims to fill that gap by examining both reproductive and growth traits comprehensively, providing insights necessary to optimise selection strategies, accelerate genetic progress and develop efficient management systems for CBBPs. Such an evaluation supports the establishment of sustainable breeding programmes that address reproductive efficiency along with growth traits, both of which are key to increasing productivity and profitability in small ruminant production.

2. Materials and Methods

2.1. Description of the Sites, Breeds and Breeding Programme

Five indigenous sheep breeds (Menz, Semein, Horro, Bonga and Doyogena), managed under CBBPs since 2009, were considered for the study. According to Gizaw (2008) the Semein and Menz breeds belong to the short fat tail breed group category, while the other three are recognised as long fat‐tailed breeds. The Menz, Bonga and Horro CBBPs were the first CBBPs initiated in Ethiopia in 2009, while the Doyogena and Semein CBBPs were established in 2012 and 2016, respectively. The regions where Menz and Semein sheep are found are known for their subalpine sheep‐barley production system. On the other hand, Bonga and Doyogena breeds thrive in perennial crop‐livestock production systems, while the Horro sheep are reared in the mixed cereal‐livestock production system. The study areas demonstrated a significant variation in average annual rainfall, with Menz sheep‐rearing areas receiving the lowest amount at 980 mm, Bonga sheep‐keeping areas experiencing the highest amount at 1617 mm, and Semein, Horro and Doyogena sheep‐rearing areas receiving annual rainfall of 1550, 1244 and 1221 mm, respectively. Communal grazing has been practised in Menz, Horro and Semein areas, while tethering is common for Bonga and Doyogena sheep. Feed availability varies across CBBP villages within each respective breed. Additionally, reproductive management practices, including early culling of unselected rams and unproductive ewes, also differ among CBBP villages. In all CBBP sites, animals are grazed during the day and housed at night. During the dry season, they are supplemented with crop residues, particularly in feed‐scarce areas like Menz, Semein, and Horro. In contrast, Doyogena and Bonga CBBPs have relatively abundant feed resources, and animals in these areas are additionally supplemented with forages cultivated in backyard plots. Across all CBBP villages, animals are regularly treated for internal parasites and vaccinated against the common viral diseases prevalent in the respective areas. The average flock size per household is 25, 13, 15, 10 and 4 sheep for Menz, Semein, Horro, Bonga, and Doyogena sheep, respectively. The average size of HHs per CBBP villages is 50, 75, 120, 180 and 200 for Menz, Semein, Horro, Bonga, and Doyogena sheep, respectively. In general, controlled mating has been practised in all CBBP villages. To improve the efficiency of the breeding programmes, training was provided to farmers in areas including breeding programme implementation, reproductive management, health interventions, feed development, marketing and other critical animal husbandry practices.

Menz sheep are hardy and small in size and are adapted to the high‐altitude precipitous terrain characterised by scarcity of feed and low crop production due to extreme low temperatures and drought. Improving growth rate, lamb survival and lambing interval are the main farmers' objectives for Menz and Semein sheep (Surafel et al. 2012; Gizaw et al. ^{2013, 2018}; Haile et al. ²⁰²⁰). Bonga, Doyogena and Horro breeds are characterised by long fat tail, good growth rate, fattening ability, twinning rate and docile temperament. The main objectives of the farmers are to improve growth, twinning rate and body conformation (Edea et al. 2012; Gizaw et al. ²⁰¹³; Haile et al. ²⁰²⁰). Percentage of multiple births is 0.7%, 15%, 33%, 26% and 44% for Menz, Semein, Horro, Bonga, and Doyogena breeds, respectively.

For all the above‐described CBBPs, the selection of the replacement animals takes place at the age of 6 months, and selection was carried out on the breeding rams. Six‐month weight was the main selection criterion for Menz and Semein breeds. For Horro, Bonga and Doyogena breeds, both 6‐month weight and litter size at birth were considered as selection traits. In all CBBP sites, the selection of breeding rams was based on estimated breeding values (EBVs) using biological data recorded by enumerators; in each site, 2–3 selection rounds took place every year. In fast‐growing breeds like Bonga sheep, a two‐stage selection strategy was implemented. The first stage was conducted at 3 months of age, based on EBVs for 3‐month weight. This approach was designed to prevent potential candidate rams from being sold for slaughter at an early age. In all CBBP villages, the final ram selection was conducted at 6 months of age. Candidate rams were initially selected by researchers based on EBVs calculated using the WOMBAT software. However, the final approval was made by the farmers' selection committee, which, in addition to EBVs, also looked at morphological traits such as horn shape, tail type, coat colour, body conformation and other preferred traits by the farmers. One breeding ram was allocated to 25–30 breeding ewes by organising the farmers into ram groups based on their proximity. To avoid inbreeding, rams were allocated carefully to each ram group and exchanged among ram groups every year. Over the years of implementation of the CBBPs, a total of 645, 34, 209, 790 and 655 sires were selected and used for Menz, Semein, Horro, Bonga and Doyogena breeds, respectively.

2.2. Data Collection and Management

Data used for the study were collected between 2009 and 2021 and downloaded from the small ruminant digital database DTREO. The dataset contained pedigree information, date of birth, sex of lamb, type of birth, colour of lamb, records of body weight at different ages and owner of the animal. Traits considered in this study were litter size at birth (LSB), litter size at weaning (LSW), total litter weight at birth (TLWB), total litter weight at weaning (TLWW), age at first lambing (AFL), lambing interval (LI), annual reproductive rate (ARR). Annual reproductive rate refers to the number of young produced per breeding ewe per year and it was calculated as $\mathrm{ARR}=\left(365\mathrm{X}\mathrm{LSB}\right)/\mathrm{LI}$ (Asmare et al. 2021). Birth weight was measured within 24 h of the birth of the lamb. The weaning weight measurements were taken at approximately 90 days after birth, with a permissible range of plus or minus 5 days, and subsequently linearly adjusted to precisely 90 days to ensure precise comparison and reliable analysis. In order to enhance the robustness and reliability of the results, data cleaning and preprocessing were conducted prior to the main analysis. Thorough examination of the collected data was performed to identify any inconsistencies, missing values, or outliers that could potentially impact the analysis. Several functions in R such as hist(), qqnorm() Shapiro test() and merge() functions were used to detect outliers, test normality, and prepare data to the appropriate format for the analysis. In this regards, age at first lambing was considered in the analysis only for Menz and Bonga breeds, which is inconsistent for other sheep breeds. Weaning weight was adjusted (AdjWWT) using the following formula:

$$\text{AdjWWT}=\left(\frac{\text{Actual weaning weight}-\text{Birth weight}}{\text{Registered}\mathrm{age}\left(\text{Days}\right)\text{for weaning weight}}\right)\times 90+\text{Birth weight}$$

2.3. Data Analysis

Fixed and breed effects for reproductive traits were estimated using the GLM (General Linear Model) procedures of SAS (SAS ²⁰⁰⁴). Means were compared using Tukey‐Kramer's test. The model used for the analysis of fixed and breed effects was

$$Y_{\text{ijkl}}=\mu +Y_{\mathrm{i}}+B_{\mathrm{j}}+{\mathit{Bs}}_{\mathrm{k}}+P_l+e_{\text{ijkl}}$$

Where Y _ijkl is an observation for each dependent variable; μ is the overall mean; Y _i is the fixed effect of lambing/birth year (13 classes: 2009–2021); B _j is the fixed effect of ewe breed (five classes: Menz, Semein, Horro, Bonga and Doyogena); Bs_k is the fixed effect of lambing/birth season (three classes: main rainy, dry and short rainy season for all breeds except for Bonga (for Bonga two classes rainy and dry season)); P _l is the fixed effect of parity of dam (nine classes: parity 1–9), and e _ijkl is the residual error. Whereas the model used for the analysis of the village of CBBPs effect under the respective sheep breed breeding programmes was

$$Y_{\mathit{\text{ijkl}}}=\mu +Y_i+V_j+{\mathit{Bs}}_k+P_l+{\mathrm{e}}_{\mathit{\text{ijkl}}}$$

Where Y _ijkl is an observation for each dependent variable; μ is the overall mean; Y _i is the fixed effect of lambing/birth year (13 classes: 2009–2021); V _j is the fixed effect of the village of CBBPs (levels are varied for each sheep breeds); Bs_k is the fixed effect of lambing/birth season (three classes: main rainy, dry and short rainy season for all breeds except for Bonga (for Bonga two classes rainy and dry season)); P _l is the fixed effect of parity, and e _ijkl is the residual error.

(Co)variance components and genetic parameters were estimated using restricted maximum likelihood (REML) method fitting repeated animal model using WOMBAT software (Meyer 2012) for all traits except for age at first lambing, which was estimated fitting non‐repeated multivariate model. The pedigree structures of all studied breeds are presented in Table 1. The model used for the analysis of all traits except age at first lambing was as follows:

$$y=\mathrm{X\beta }+Z_{{\mathrm{a}}^{\mathrm{\alpha }}}+Z_{{\mathrm{pe}}^{\mathrm{pe}}}+e$$

Whereas the model used for the analysis of age at first lambing was.

TABLE 1.

Pedigree structure for random effects.

Parameters	Data set
	Menz	Semein	Horro	Bonga	Doyogena
Pedigree Structure for random effect ‘1’
Original no. of animals	9533	1063	2000	5464	2637
No. of animals after pruning	9060	1063	1449	4886	2508
Proportion (%) remaining	95.0	100	72.5	89.4	95.1
No. of levels w/out records	242	0	107	121	63
No. of levels with records	8818	1063	1342	4765	2445
No. of animals w/out offspring	6436	1063	1119	4016	2130
No. of animals with offspring	2624	0	330	870	378
No. of animals with offspring and records	2382	0	223	749	315
No. of sires with progeny in the data	263	0	98	121	63
No. of sires with records & progeny in data	21	0	18	0	1
No. of dams with progeny in the data	2361	0	232	749	315
No. of dams with records & progeny in data	2361	0	205	749	314
Pedigree Structure for random effect ‘2’
Original no. of animals	9533			5464
No. of animals after pruning	2660			769
Proportion (%) remaining	27.9			14.1
No. of levels w/out records	937			277
No. of levels with records	1723			492
No. of animals w/out offspring	1398			484
No. of animals with offspring	1262			285
No. of animals with offspring and records	325			8
No. of sires	213			59
No. of sires with progeny in the data	155			26
No. of sires with records & progeny in data	1			0
No. of dams	648			57
No. of dams with progeny in the data	513			32
No. of dams with records & progeny in data	185			5

Note: Pedigree Structure for random effect ‘1’ is for traits: litter size at birth (LSB), litter size at weaning (LSW), total litter weight at birth (TLWB), total litter weight at weaning (TLWW), lambing interval (LI) and annual reproductive rate (ARR); Pedigree Structure for random effect ‘2’ is for trait: age at first lambing (AFL).

$y=\mathrm{X\beta }+Z_{{\mathrm{a}}^{\mathrm{\alpha }}}+Z_{{\mathrm{m}}^{\mathrm{m}}}+e$ where y is a vector of observations on the considered traits; β is the vector of fixed effects (overall mean, site of CBBP, birth year, birth month and parity); α, m, pe and e are vectors of direct additive genetic effects, maternal genetic effects, permanent environment effects of the animal and the residual effects, respectively. Whereas X, Z _a, Z _m and Z _pe are corresponding incidence matrices relating the fixed effect, direct additive genetic effects, maternal additive genetic effects and permanent environmental effects of the animal. Repeatability (r) was estimated using: $r={\sigma }_{\mathrm{a}+}^2{\sigma }_{\mathrm{pe}}^2/{\sigma }_{\mathrm{p}}^2$ Where σ ² _a is additive variance, σ ² _pe is permanent environmental variance and σ ² _p is total phenotypic variance. Phenotypic and genetic trends were estimated by regressing yearly mean phenotype performance and estimating breeding value (EBV) on year of birth.

3. Results

3.1. Fixed Effects

Tables 2, 3 and 4 present the effects of non‐genetic factors on the reproductive performance of different sheep breeds. All traits under this study are significantly influenced by village of CBBPs under the respective sheep breed breeding programmes (p < 0.001). The effects of lambing year and season of lambing on all the studied traits were significant (p < 0.001). Ewes mated during the dry season (lambed during the main rainy season) had lower LSB, TLWB, LSW, TLWW and ARR than those mated in the main and short rainy seasons. Parity has a significant (p < 0.001) effect on all the studied traits for sheep breeds. There was a general tendency for improvement of LSB, TLWB, LSW, TLWW and ARR with the increase of ewe parity until the seventh parity, followed by a declining trend thereafter. Furthermore, the current analysis of the dataset revealed that the proportions of ewes after the seventh parity in the flocks were 7.3%, 6.8%, 0.42%, 1.88% and 4.7% for Menz, Semein, Horro, Bonga and Doyogena breeds, respectively.

TABLE 2.

Least squares means and standard errors of reproductive traits of some Ethiopian sheep breeds.

Parameters	N	LSB	TLWB (kg)	N	LSW	TLWW (kg)	N	LI (Days)	ARR
Overall mean	41,275	1.16 ± 0.01	3.52 ± 0.01	38,160	1.08 ± 0.01	13.44 ± 0.04	12,865	254.57 ± 0.46	1.74 ± 0.01
CV		30.68	32.88		38.94	44.62		19.89	37.39
Lambing year		***	***		***	***		***	***
Breed		***	***		***	***		***	***
Menz	22,736	1.05 ± 0.04d	2.87 ± 0.12d	21,739	0.89 ± 0.04d	8.31 ± 0.51e	7323	241.17 ± 5.25a	1.59 ± 0.07d
Semein	2006	1.14 ± 0.04c	3.16 ± 0.12c	2006	0.81 ± 0.04e	10.41 ± 0.53d	533	235.27 ± 5.64a	1.81 ± 0.07c
Horro	2211	1.29 ± 0.04b	4.01 ± 0.12b	2185	1.00 ± 0.04c	12.09 ± 0.53c	491	207.77 ± 5.82d	2.19 ± 0.07b
Bonga	10,574	1.31 ± 0.04b	4.74 ± 0.12a	8965	1.10 ± 0.04b	18.07 ± 0.51b	3744	228.51 ± 5.28b	2.10 ± 0.07b
Doyogena	3748	1.52 ± 0.04a	4.77 ± 0.12a	3265	1.22 ± 0.04a	19.33 ± 0.52a	744	221.64 ± 5.53c	2.53 ± 0.07a
Parity		***	***		***	***		***	***
1	12,049	1.19 ± 0.02d	3.49 ± 0.03d	10,930	0.96 ± 0.02d	12.75 ± 0.33d		NA	NA
2	7400	1.23 ± 0.02c	3.66 ± 0.03c	6832	1.01 ± 0.02c	13.61 ± 0.33c	3956	253.01 ± 2.53a	1.84 ± 0.03c
3	5260	1.25 ± 0.02b	3.75 ± 0.03b	4852	1.03 ± 0.02bc	13.89 ± 0.33bc	2935	242.25 ± 2.59b	1.92 ± 0.03b
4	3556	1.26 ± 0.02b	3.81 ± 0.03b	3313	1.05 ± 0.02ab	14.09 ± 0.34ab	2058	236.22 ± 2.65c	1.99 ± 0.03a
5	2161	1.30 ± 0.02a	3.93 ± 0.04a	2047	1.08 ± 0.02a	14.43 ± 0.35a	1266	233.33 ± 2.78cd	2.06 ± 0.04a
6	1298	1.30 ± 0.02a	3.94 ± 0.04a	1215	1.08 ± 0.03a	14.51 ± 0.37a	760	232.86 ± 3.02cd	2.04 ± 0.04a
7	707	1.30 ± 0.02a	3.95 ± 0.05a	675	1.09 ± 0.03a	14.51 ± 0.40a	426	231.76 ± 3.44cd	2.06 ± 0.04a
8	357	1.30 ± 0.03a	3.90 ± 0.07ab	333	1.06 ± 0.03a	14.14 ± 0.46a	240	229.04 ± 4.04cd	2.07 ± 0.05a
≥ 9	294	1.29 ± 0.03bc	3.80 ± 0.06b	277	1.04 ± 0.03bc	13.85 ± 0.49bc	211	222.22 ± 4.26d	2.13 ± 0.05a
Lambing season		***	***		***	***		***	***
Main rainy	14,892	1.25 ± 0.04b	3.86 ± 0.12b	13,505	0.99 ± 0.04b	13.41 ± 0.51b	4651	231.24 ± 5.31a	1.98 ± 0.07b
Short rainy	6947	1.26 ± 0.04b	3.89 ± 0.12ab	6507	1.02 ± 0.04a	13.94 ± 0.51a	2252	222.49 ± 5.36c	2.07 ± 0.07a
Dry	19,434	1.27 ± 0.04a	3.92 ± 0.12a	18,146	1.01 ± 0.04a	13.78 ± 0.51a	5962	226.88 ± 5.29b	2.08 ± 0.07a

Note: ^abcdeIn the same column, numbers bearing the same superscript are not statistically different at p = 0.05. Bold values indicate improves readability and helps draw attention as we have so many levels.

Abbreviations: ARR, annual reproductive rate; LI, lambing interval; LSB, litter size at birth; LSW, litter size at weaning; NA, not applicable; ns, not significant; TLWB, total litter weight at birth; TLWW, total litter weight at weaning.

^*** p < 0.001.

TABLE 3.

Effects of CBPs village on reproductive traits under community‐based breeding programmes.

CBBPs village	Traits									n	AFL (Days)
	n	LSB	TLWB (Kg)	n	LSW	TLWW (Kg)	n	LI (Days)	ARR
Menz		ns	***		***	***		***	***		***
Dargegn	5336	1.01 ± 0.01	2.83 ± 0.06b	5213	0.91 ± 0.02b	9.87 ± 0.30a	1554	241.91 ± 8.08a	1.60 ± 0.05c	271	570.68 ± 11.26a
Molale	5753	1.02 ± 0.01	3.02 ± 0.06a	5420	0.92 ± 0.02b	9.30 ± 0.30b	2245	228.18 ± 7.95b	1.69 ± 0.05b	275	509.58 ± 10.11b
Sinamba	8877	1.01 ± 0.01	2.48 ± 0.06d	8748	0.94 ± 0.02a	8.46 ± 0.30c	2741	240.37 ± 8.04a	1.61 ± 0.05c	830	560.52 ± 9.65a
Tsehaysina	1151	1.02 ± 0.01	2.70 ± 0.07c	850	0.84 ± 0.02c	9.92 ± 0.35a	246	216.58 ± 9.19bc	1.77 ± 0.06ab	17	509.46 ± 29.93b
Zeram	1619	1.01 ± 0.01	1.93 ± 0.06e	1508	0.93 ± 0.02ab	9.01 ± 0.34b	532	207.92 ± 8.88c	1.79 ± 0.06a	30	492.42 ± 23.83b
Semein		***	***		***	***		***	***
Afaf	850	1.06 ± 0.04	2.85 ± 0.11	850	0.78 ± 0.06	8.16 ± 0.79	180	249.58 ± 7.16	1.63 ± 0.09
Chamblege	1201	1.18 ± 0.04	2.98 ± 0.11	1156	0.88 ± 0.06	11.49 ± 0.78	350	233.60 ± 6.49	1.97 ± 0.08
Horro		***	***		***	ns		***	**
Gitlo	975	1.17 ± 0.09b	4.34 ± 0.34b	975	1.06 ± 1.11b	15.23 ± 1.29	237	220.95 ± 6.50a	2.21 ± 0.13b
Laku	639	1.43 ± 0.09a	4.60 ± 0.34a	639	1.28 ± 0.11a	14.57 ± 1.29	81	185.16 ± 7.07b	2.61 ± 0.14a
Meta‐welkite	597	1.20 ± 0.09b	3.89 ± 0.34c	594	1.12 ± 0.11b	15.31 ± 1.30	169	216.40 ± 6.29a	2.25 ± 0.13b
Bonga		***	***		***	***		***	***		ns
Boqa	5180	1.35 ± 0.05	4.87 ± 0.17	4636	1.36 ± 0.04	21.85 ± 0.64	1742	239.14 ± 6.17	2.10 ± 0.12	226	416.34 ± 7.20
Shuta	5394	1.28 ± 0.05	4.55 ± 0.17	4329	1.22 ± 0.04	17.76 ± 0.64	1992	245.31 ± 6.16	1.94 ± 0.12	266	431.29 ± 6.62
Doyogena		***	***		***	***		ns	ns
Ancha	883	1.55 ± 0.08a	4.47 ± 0.21b	746	1.26 ± 0.08ab	17.21 ± 1.25b	141	241.20 ± 5.67	2.47 ± 0.13
Begedamu	319	1.58 ± 0.08a	4.63 ± 0.22ab	300	1.31 ± 0.09a	19.59 ± 1.30a	51	251.62 ± 7.86	2.60 ± 0.19
Gomora	163	1.45 ± 0.09ab	4.27 ± 0.25bc	110	1.26 ± 0.10ab	20.18 ± 1.53a	35	240.82 ± 9.34	2.44 ± 0.22
Hawora	986	1.49 ± 0.08a	4.90 ± 0.21a	887	1.17 ± 0.08b	17.21 ± 1.25b	238	245.71 ± 5.11	2.47 ± 0.12
Lemi	198	1.38 ± 0.09b	3.77 ± 0.25c	141	1.14 ± 0.10b	19.64 ± 1.48a	21	244.99 ± 11.81	2.13 ± 0.28
Murasa	406	1.50 ± 0.08a	4.91 ± 0.22a	366	1.33 ± 0.09a	20.63 ± 1.31a	54	247.58 ± 7.62a	2.42 ± 0.18
Sarara	700	1.40 ± 0.08b	4.50 ± 0.22b	637	1.17 ± 0.08b	19.71 ± 1.26a	201	241.26 ± 5.40	2.27 ± 0.13
Women	93	1.25 ± 0.10b	3.82 ± 0.27c	78	1.06 ± 0.11b	17.41 ± 1.59b	29	236.21 ± 9.98	2.37 ± 0.24

Note: ^abcdIn the same column, numbers bearing the same superscript are not statistically different at p = 0.05.

Abbreviations: ARR, annual reproductive rate; LI, lambing interval; LSB, litter size at birth; LSW, litter size at weaning; ns, not significant; TLWB, total litter weight at birth; TLWW, total litter weight at weaning.

^*** p < 0.001.

^** p < 0.01.

TABLE 4.

Least squares means and standard errors of age at first lambing of some Ethiopian sheep breeds.

Parameters	N	AFL (Days)
Overall mean	2215	546.63 ± 2.81
CV		20.44
Birth year		***
Dam parity		***
Breed		***
Menz	1723	550.65 ± 8.82
Bonga	492	397.36 ± 9.86
Birth season		***
Main rainy	819	467.38 ± 9.34b
Short rainy	319	468.22 ± 10.50b
Dry	1077	486.41 ± 9.13a

Note: ^abIn the same column, numbers bearing the same superscript are not statistically different at p = 0.05. Bold values indicate improves readability and helps to draw attention.

Abbreviation: AFL, age at first lambing.

^*** p < 0.001.

3.2. Breed Effects

The least squares means and standard errors for reproductive traits across different ewe breeds are presented in Tables 2 and 4. All traits examined in this study were significantly influenced by the ewe breed (p < 0.001). The present results showed that the Doyogena breed had significantly higher (p < 0.001) LSB, LSW, TLWW, and ARR than the other four indigenous sheep breeds (1.52, 1.22, 19.33 kg and 2.53 respectively). Ranking second after Doyogena, the Bonga breed had significantly higher (p < 0.001) TLWB, LSW, and TLWW than the remaining three other indigenous sheep breeds (4.74 kg, 1.10 and 18.07 kg respectively). The Menz and Semein breeds had lower performance for all studied traits (p < 0.001). The current results indicate that the Bonga breed had shorter AFL (397 days) than the Menz breed (551). Least squares means and standard errors of LI for the studied sheep breeds were presented in Table 2. Shorter LI were found in the Horro, Doyogena, and Bonga breeds (208, 222 and 228 days) respectively; longer LI was computed for the Menz and Semein breeds (241 and 235 days respectively). AFL for the Menz breed showed a declining trend at a rate of 4.48 days over a period of 12 years (Figure 1). However, no improvement in AFL was recorded (not shown) in Bonga sheep.

FIGURE 1.

Phenotypic trend for AFL in Menz sheep.

3.3. Genetic Parameter Estimates

Estimates of direct heritability (h ² _a), ratio of variance due to permanent environmental effects to the total phenotypic variance (Pe²) and repeatability (r) for the studied traits are presented in Table 5. Current direct heritability estimates for the prolificacy traits were low to medium and ranged from 0.03 ± 0.01 to 0.25 ± 0.10 for LSB and 0.02 ± 0.02 to 0.16 ± 0.01 for LSW. The ratio of variance due to permanent environmental effects of the ewe was also low to medium and ranged from 0.04 ± 0.03 to 0.22 ± 0.10 for LSB and 0.04 ± 0.01 to 0.16 ± 0.04 for LSW. Estimates of direct heritability for TLWB, TLWW, ARR and LI were low to medium and ranged from 0.08 ± 0.03 to 0.21 ± 0.02, 0.07 ± 0.10 to 0.22 ± 0.02, 0.08 ± 0.04 to 0.32 ± 0.20 and 0.03 ± 0.09 to 0.19 ± 0.06, respectively. The estimated direct heritability for AFL was 0.36 ± 0.07 and 0.15 ± 0.18 for Menz and Bonga breed, respectively. The genetic trend of LI for Menz and Doyogena breed showed an improvement trend at a rate of 0.57 and 0.30 days over the years respectively (Figure 2). However, no improvement trend was recorded (not shown in the figure) for other sheep breeds.

TABLE 5.

Direct heritability and repeatability estimates for different reproduction traits.

Breed	Trait	LI	LSB	TLWB	LSW	TLWW	ARR	AFL
Menz	σ 2 a	303.99	0.0012	0.075	0.0035	2.447	0.020	5216.87
	h 2 a ± SE	0.09 ± 0.04	0.08 ± 0.02	0.21 ± 0.02	0.05 ± 0.01	0.22 ± 0.02	0.08 ± 0.04	0.36 ± 0.07
	pe2 ± SE	0.11 ± 0.04	0.08 ± 0.02	0.04 ± 0.02	0.04 ± 0.13	0.03 ± 0.02	0.15 ± 0.04
	r	0.21	0.10	0.25	0.09	0.25	0.23
Semein	σ 2 a	686.38	0.016	0.106	0.044	5.10	0.087
	h 2 a ± SE	0.19 ± 0.06	0.14 ± 0.03	0.14 ± 0.03	0.16 ± 0.01	0.12 ± 0.03	0.17 ± 0.05
	Pe2 ± SE	0.20 ± 0.01	0.15 ± 0.01	0.14 ± 0.01	0.16 ± 0.04	0.12 ± 0.01	0.17 ± 0.01
	r	0.39	0.29	0.27	0.32	0.24	0.33
Horro	σ 2 a	498.30	0.064	0.512	0.039	4.797	0.194
	h 2 a ± SE	0.07 ± 0.02	0.07 ± 0.03	0.11 ± 0.05	0.02 ± 0.02	0.07 ± 0.10	0.18 ± 0.25
	Pe2 ± SE	0.05 ± 0.01	0.04 ± 0.03	0.05 ± 0.05	0.07 ± 0.02	0.33 ± 0.11	0.17 ± 0.24
	r	0.13	0.10	0.17	0.08	0.41	0.35
Bonga	σ 2 a	49.268	0.022	0.293	0.018	4.065	0.082	1022.25
	h 2 a ± SE	0.03 ± 0.09	0.03 ± 0.01	0.08 ± 0.03	0.03 ± 0.01	0.08 ± 0.04	0.10 ± 0.09	0.15 ± 0.18
	Pe2 ± SE	0.20 ± 0.09	0.05 ± 0.01	0.08 ± 0.03	0.04 ± 0.01	0.12 ± 0.04	0.21 ± 0.09
	r	0.23	0.07	0.15	0.07	0.21	0.31
Doyogena	σ 2 a	197.435	0.107	0.708	0.046	9.783	0.434
	h 2 a ± SE	0.08 ± 0.31	0.25 ± 0.10	0.21 ± 0.09	0.09 ± 0.07	0.08 + 0.07	0.32 ± 0.20
	Pe2 ± SE	0.29 ± 0.32	0.22 ± 0.10	0.20 ± 0.10	0.09 ± 0.07	0.10 ± 0.08	0.25 ± 0.20
	r	0.36	0.48	0.41	0.18	0.19	0.58

Abbreviations: AFL, age at first lambing; ARR, annual reproductive rate; h ² _a, direct heritability; LI, lambing interval; LSB, litter size at birth; LSW, litter size at weaning; Pe², ratio of permanent environmental variance to phenotypic variance; r, repeatability; TLWB, total litter weight at birth; TLWW, total litter weight at weaning.

FIGURE 2.

Genetic trend by year of birth (LI).

3.4. Correlations Estimates

Estimates of direct genetic and phenotypic correlations among the studied traits are presented in Table 6. Direct genetic correlation estimates between LSB with other reproductive traits were positive and low to high in magnitude. LSB had a higher and positive direct genetic correlation with TLWB, LSW, TLWW and ARR for all studied breeds except Menz. Direct genetic correlations between TLWW with LSW, TLWB, ARR were positive and moderate to high in magnitude in all studied breeds except Menz.

TABLE 6.

Correlation estimates among reproduction traits.

Trait 1	Trait 2	Menz		Semein		Horro		Bonga		Doyogena
		r a12	r p12	r a12	r p12	r a12	r p12	r a12	r p12	r a12	r p12
LSB	TLWB	0.16 ± 0.10	0.21 ± 0.01	0.87 ± 0.05	0.84 ± 0.01	0.97 ± 0.03	0.92 ± 0.01	0.79 ± 0.11	0.27 ± 0.01	0.94 ± 0.17	0.61 ± 0.01
LSB	LSW	0.03 ± 0.17	0.17 ± 0.01	0.66 ± 0.08	0.51 ± 0.02	0.69 ± 0.31	0.71 ± 0.01	0.79 ± 0.16	0.17 ± 0.01	0.71 ± 0.32	0.46 ± 0.01
LSB	TLWW	0.01 ± 0.10	0.11 ± 0.01	0.75 ± 0.14	0.22 ± 0.02	0.50 ± 0.61	0.41 ± 0.02	0.78 ± 0.18	0.24 ± 0.01	0.60 ± 0.35	0.41 ± 0.02
LSB	ARR	0.11 ± 0.22	0.01 ± 0.01	0.84 ± 0.21	0.24 ± 0.04	0.56 ± 0.82	0.11 ± 0.05	0.93 ± 0.46	0.03 ± 0.01	0.80 ± 0.33	0.45 ± 0.03
LSB	AFL	0.05 ± 0.11	0.03 ± 0.02					−0.42 ± 0.42	0.01 ± 0.03
TLWB	LSW	−0.05 ± 0.12	0.08 ± 0.01	0.59 ± 0.16	0.46 ± 0.02	0.66 ± 0.32	0.67 ± 0.01	0.81 ± 0.15	0.22 ± 0.01	0.89 ± 0.43	0.33 ± 0.02
TLWB	TLWW	0.20 ± 0.07	0.13 ± 0.01	0.64 ± 0.13	0.25 ± 0.02	0.51 ± 0.61	0.39 ± 0.03	0.84 ± 0.13	0.35 ± 0.01	0.83 ± 0.43	0.31 ± 0.02
TLWB	ARR	−0.23 ± 0.15	0.04 ± 0.01	0.73 ± 0.22	0.15 ± 0.04	0.55 ± 0.82	0.10 ± 0.05	0.91 ± 0.42	0.08 ± 0.01	0.80 ± 0.34	0.43 ± 0.03
TLWB	AFL	0.05 ± 0.08	0.11 ± 0.02					−0.32 ± 0.36	0.03 ± 0.04
LSW	TLWW	0.32 ± 0.10	0.48 ± 0.01	0.98 ± 0.14	0.58 ± 0.02	0.94 ± 0.66	0.70 ± 0.02	0.81 ± 0.11	0.44 ± 0.01	0.96 ± 0.05	0.95 ± 0.01
LSW	ARR	0.35 ± 0.25	0.02 ± 0.01	0.55 ± 0.23	0.13 ± 0.04	0.65 ± 0.60	0.08 ± 0.05	0.89 ± 0.46	0.05 ± 0.01	0.60 ± 0.49	0.24 ± 0.03
LSW	AFL	0.17 ± 0.14	0.04 ± 0.02					−0.21 ± 0.40	0.03 ± 0.03
TLWW	ARR	0.22 ± 0.15	0.05 ± 0.01	0.66 ± 0.23	0.12 ± 0.04	0.48 ± 0.34	0.14 ± 0.05	0.80 ± 0.44	0.09 ± 0.02	0.60 ± 0.51	0.21 ± 0.03
TLWW	AFL	0.08 ± 0.09	0.05 ± 0.03					−0.13 ± 0.37	0.08 ± 0.05
ARR	AFL	0.52 ± 0.18	0.21 ± 0.03					−0.43 ± 0.53	−0.13 ± 0.07

Note:  r _p12 : phenotypic correlation between trait 1 and trait 2; r _a12 : direct genetic correlations between traits 1 and 2.

Abbreviations: AFL, age at first lambing; ARR, annual reproductive rate; LI, lambing interval; LSB, litter size at birth; LSW, litter size at weaning; TLWB, total litter weight at birth; TLWW, total litter weight at weaning.

4. Discussion

The significant effect of lambing year on reproductive traits has been reported by several authors (Yavarifard et al. 2015; Tesema et al. ²⁰²⁰; Tera et al. ²⁰²¹). The significant influence of lambing year on reproductive traits could be explained by variation in management, grazing pasture availability, climatic conditions, availability of breeding sire, and indirect effects of selective breeding for growth traits. Lambing interval and age at first lambing improved over years for Menz and Bonga breeds, which could be due to correlated response, as these traits are not directly targeted in the selection programme. The lower LSW, TLWW, ARR and LI performance observed in ewes lambing during the main rainy season could be due to environmental stress and low quality and quantity of feed resources for ewes that conceived during the dry season. Moreover, during the main rainy season, feed shortage is common in all CBBP sites because the arable land is covered by crop production, and grazing is restricted from grazing for hay production. As a result, animals are confined to marginal areas during this season. The variation in reproductive performance in favour of the dry season is attributed to feed abundance (following the main rainy season) and is reflected in better body condition at mating, positively influencing conception rate and ovulation rate (Yavarifard et al. 2015). We therefore suggest that when the mating season coincides with the dry season, farmers should consider having improved forages ready to supplement the flocks and hence reduce the negative impact of the season on the reproductive performance of sheep. As shown in Table 3, the village under the respective CBBPs had a significant effect on reproductive traits. The significant difference in reproductive traits among CBBP villages could be ascribed to differences in feed availability and reproductive management of the ewes.

Parity has a significant (p < 0.001) effect on all the studied traits for sheep breeds. Significant effects of parity on reproductive traits have been reported in literature (Yavarifard et al. 2015; Areb et al. ²⁰²¹; Tera et al. ²⁰²¹). Differences in maternal effects, capacity of milking, nursing, and maternal behaviour of ewes at different parities are probably the reason for the significant influence of ewe parity. The decline of the traits after parity seven is partly due to the biological reduction in the reproductive performance of the ewe and also its milk production and suckling potential (Yavarifard et al. 2015). Therefore, culling of unproductive ewes after the seventh parity can improve reproductive performance of the flock. The lower proportion of aged ewes in the breeding programmes could be due to the improvement of farmers' knowledge through frequent training given at CBBPs sites on sheep breeding and practice of culling older/unproductive ewes from their flock.

All traits under this study are significantly influenced by the ewe breed (p < 0.001). The significant difference in reproductive traits among indigenous breeds can be ascribed to the difference in natural habit, reproductive management, and genetic differences. Doyogena and Bonga breeds are reared in areas with more evenly distributed rainfall and better feed availability compared to the regions where Menz and Semein sheep are raised. LSB recorded in the current study for the Bonga breed is comparable to the findings reported for the same breed under similar breeding programmes (Areb et al. 2021; Tera et al. ²⁰²¹). However, ARR for the Bonga breed was lower than the values previously reported for the same breed (2.10 vs. 2.31) (Areb et al. 2021). Phenotypic performance of AFL for Menz breed showed a declined trend at a rate of 4.48 days over a period of 12 years (Figure 1). This indicates that selection for live weight at 6 months was associated with an improvement in the sexual maturity of the females leading to a shortening of the AFL. Selection for faster growth and higher live weight induces an earlier puberty in females and consequently the age at first conception and at first lambing (Nieto et al. 2013). Least squares means and standard errors of AFL for the Menz and Bonga breeds were presented in Table 4. AFL for the Bonga breed was in close agreement with the findings of previous reports (Edea et al. 2012), but lower than 375 days for the same breed (Areb et al. 2021). Moreover, AFL recorded in the current study for the Bonga breed was shorter than the values reported by Tera et al. (2021) for the same breed (453 days). In the current results, higher AFL was recorded for the Menz breed, which is 551 days. Such a wide variability in AFL between studied breeds could be attributed to differences in feed availability, growth rate, and rearing environment, in which Menz sheep are raised in highland and harsh environments which may negatively affect the onset of puberty. AFL is directly associated with the age at puberty, varying depending on feeding, reproductive management, live weight, and breed of the animal. Live weight of the ewe is the most important factor for the appearance of puberty; heavier maiden ewes have higher fertility and lamb at an earlier age than dams with lower body weight (Magaña‐Monforte et al. ²⁰¹³). LI recorded in Menz sheep is in agreement with the findings of (Getachew 2008) for the same breed (255 days). On the other hand, LI recorded in the current study for the Bonga breed was shorter than the previous reports by Areb et al. (2021); Tera et al. (2021) for the same breed. The LI for Horro sheep under the current study is longer than the values reported by Ayele and Urge (2019) for the same breed (267–300 days). Differences in reproductive performance recorded for the same breed across different studies under similar breeding programmes could be due to differences in data cleaning and the threshold used for the removal of outliers. In CBBPs, 64%, 48% and 54% of breeding rams were selected from 15%, 12% and 16% of participant farmers for Menz, Bonga and Doyogena respectively. This indicates that more than 50% of the breeding rams are contributed by about 15% of the participant farmers. Therefore, optimising the ongoing breeding programme with a specialising‐sire producer group could be an option to achieve better results.

Current direct heritability estimates for LSB were low to medium and ranged from 0.03 ± 0.09 for Bonga to 0.25 ± 0.10 for Doyogena breed. Direct heritability estimates for LSB in Doyogena and Bonga sheep show strong agreement with the results reported by Habtegiorgis et al. (2022) for Doyogena sheep and by Tera et al. (2021) for Bonga sheep. Estimates of direct heritability for LSB in our findings are in agreement with the estimates of (Mohammadi et al. 2015; Yavarifard et al. ²⁰¹⁵; Roudbar et al. ²⁰¹⁸) for different sheep breeds. The low direct heritability estimates of LSB in the current study, except for the Doyogena breed, indicate that a direct genetic selection for this trait does not considerably improve LSB in these breeds. Lower heritability estimates for LSW for Menz, Semein, Horro and Doyogena breeds, compared with the heritability estimates for LSB, suggested that the loss of lambs from birth to weaning is influenced mainly by environmental factors such as a particular disease and mortality of lambs. Lower heritability estimates for LSW are in close agreement with the findings of (Roudbar et al. 2018) for Lori‐Bakhtiari sheep breed. Low direct heritability estimates for TLWB in the current study fall in the range of 0.02–0.21 that has been published in the literature (Yavarifard et al. 2015; Roudbar et al. ²⁰¹⁸). Direct heritability estimates for TLWW in the current study were in the range of 0.02–0.26 as reported by several authors (Mohammadi et al. 2015; Yavarifard et al. ²⁰¹⁵; Roudbar et al. ²⁰¹⁸; Khattab et al. ²⁰²¹). TLWW is the most appropriate criterion for selecting ewes (Mohammadi et al. 2012). Thus, relatively higher direct heritability estimates of TLWW for Menz ewe (0.22) in our study indicate that considering this trait in the selection programme would yield considerable improvement in the reproductive performance of the breed.

Direct heritability estimates for AFL were 0.36 ± 0.07 and 0.15 ± 0.18 for Menz and Bonga breeds respectively. Relatively higher AFL observed in Menz sheep could be attributed to the presence of sizeable genetic diversity within the population. In agreement with these findings, (Rather et al. 2020) reported medium to high heritability estimates for AFL across different sheep breeds. Moreover, the relatively higher direct heritability estimates for AFL in Menz sheep indicate that selection based on AFL would be more effective for improving these traits. Contrary to the present finding, the lower direct heritability estimates for AFL were reported by Areb et al. (2021); Tera et al. (2021) for Bonga breed (0.07 and 0.015). The estimates of LI for Bonga ewes were congruent to estimates previously reported for the same breed (0.06) (Areb et al. 2021). However, lower direct heritability has been reported by Tera et al. (2021) for the same breed (0.009). Furthermore, Habtegiorgis et al. (2022) reported a higher heritability estimate for LI (0.20 ± 0.5), likely due to their use of a non‐repeated univariate model for repeated records. Lower direct heritability estimates for LI in the current study are in agreement with the finding of (Aguirre et al. 2017). ARR refers to the number of lambs produced per breeding ewe per year (Areb et al. 2021; Asmare et al. ²⁰²¹). Direct heritability estimates of ARR were 0.08, 0.17, 0.18, 0.10 and 0.32 for Menz, Semein, Horro, Bonga and Doyogena breeds respectively. These estimates are lower than direct heritability estimates reported by Areb et al. (2021) for Bonga breed (0.25). The low estimates of direct heritability for the most studied reproductive traits are probably due to the higher proportional influence of environmental effects and to their little genetic variability within the breed. This implies that much of the improvement in reproductive traits could be achieved by the improvement of the production environment, such as better nutritional management of ewes before mating and after pregnancy, using active breeding rams in the flock throughout the year and by improving the health of the flock, rather than by direct selection.

The ratio of permanent environmental variance to the phenotypic variance estimates for the studied traits ranged from 0.03 ± 0.02 for TLWW (Menz) to 0.33 ± 0.11 for TLWW (Horro breed) (Table 5). Estimates of the ratio of permanent environmental variance to the total phenotypic variance were higher than the estimates of direct heritability for Bonga breeds in all studied traits except TLWB, which is equal to direct additive heritability. This indicates that environmental factors have a highly significant effect on the expression of these traits. Furthermore, estimates of the ratio of permanent environmental variance to the total phenotypic variance were similar to the estimates of direct additive heritability for Semein breed. Estimates of the ratio of variance due to permanent environmental effects to the total phenotypic variance for the studied traits were in the range of reported estimates by previous authors (Boujenane et al. 2013; Mohammadi et al. ²⁰¹⁵; Baneh et al. ²⁰²⁰; Tera et al. ²⁰²¹). However, they are lower than those reported by Areb et al. (2021) for Bonga breed. Repeatability estimates for all studied traits were higher than the direct heritability estimates, suggesting that the studied traits are affected more by non‐additive genetic effects and permanent environmental effects. Therefore, the accuracy of selection for these traits in the first lambing should be medium as repeatability measures correlation between performance records in different lambings of the ewe. Repeatability estimates in our study ranged from 0.07 for LSB and LSW (Bonga) to 0.58 for ARR (Doyogena). Estimates of repeatability for the studied traits are in the range of the previously reported estimates (Amou Posht‐e‐ Masari et al. ²⁰¹³; Boujenane et al. ²⁰¹³; Mohammadi et al. ²⁰¹⁵; Yavarifard et al. ²⁰¹⁵; Tera et al. ²⁰²¹). However, higher estimates were reported by Areb et al. (2021) for ARR for Bonga breed (0.65).

LSB had a higher and positive direct genetic correlation with TLWB, LSW, TLWW and ARR for all studied breeds except Menz. These estimates show that ewes with a higher number of lambs born in each litter would have higher TLWB, LSW, TLWW and ARR. The higher direct genetic correlation estimates of LSB with TLWB, LSW and TLWW were in agreement with several reports for different sheep breeds (Amou Posht‐e‐ Masari et al. ²⁰¹³; Boujenane et al. ²⁰¹³; Mohammadi et al. ²⁰¹⁵; Yavarifard et al. ²⁰¹⁵). On the other hand, the direct genetic correlation between LSB with TLWB, LSW, TLWW, ARR and AFL was low and positive in the Menz breed. This implies that selection to improve genetic merit in LSB would have little influence on the genetic response in the other traits. Phenotypic correlations between LSB with TLWB, LSW and TLWW were generally moderate to high and positive. Selection for TLWW for those breeds may be desirable, even if the direct heritability is not high because of moderate to high direct genetic correlation between TLWW with LSB, LSW, TLWB and ARR except for Menz sheep. These estimates of direct genetic correlation were in consistency with those obtained by Amou Posht‐e‐ Masari et al. (2013); Mohammadabadi and Sattayimokhtari (2013). The weak direct genetic correlations between TLWB with LSW, TLWW, ARR for the Menz breed indicate that ewes with more lambs weight in each litter would have little effect on the genetic response in LSW, TLWW and ARR. On the other hand, estimates of direct genetic correlations between LSW with TLWW and ARR in the Menz breed were positive and weak in magnitude. This implies that ewes with a greater number of lambs at weaning in each lambing would have less effect on the genetic response of TLWW and ARR. Direct genetic correlations between TLWW and ARR were positive and medium to high in magnitude. Whereas the direct genetic correlation between TLWW with AFL and ARR with AFL were weak in magnitude. The weak correlation estimates showed that a fast‐growing lamb has no effect on achieving a higher TLWW and ARR.

5. Conclusion

The study highlights the significance of non‐genetic factors, particularly lambing year, season, parity, and breed, on the reproductive performance of indigenous sheep in Ethiopia. The findings underscore the need for tailored breeding strategies that consider these factors to enhance reproductive traits and overall flock productivity. While heritability estimates for these traits were generally acceptable, some were low, suggesting that genetic improvement is a viable approach within CBBPs. The variability in heritability estimates observed in this study suggests that enhancing reproductive performance necessitates a focus on both genetic and environmental factors, underscoring the importance of breed‐specific recommendations that align management practices and breeding strategies with each breed's unique characteristics. The presence of moderate to high genetic correlation between total litter weight at weaning (TLWW) and other traits suggests that incorporating TLWW into the selection index would significantly enhance ewe reproductive performance. By combining genetic selection with improved environmental conditions and effective culling strategies, especially removing unproductive ewes after their seventh parity, overall flock productivity can be increased. Additionally, improved management practices, such as supplemental feeding during critical periods, better nutritional management of ewes before mating and during pregnancy, maintaining healthy breeding rams year‐round, and ensuring overall flock health, can further enhance reproductive traits. The genetic parameter estimates provide valuable insights into heritability and trait correlations, aiding in the selection of superior breeding stock for future programmes.

Author Contributions

Conceptualization: Aynalem, Tesfaye, Mourad, Berhanu, Shanbel, Zelalem, and Barbara. Data compilation and follow up of the breeding programmes: Shanbel, Zelalem, Shenkute, Kebede, Temesgen, Ermiyas, Tusa, and Tesfaye. Writing – original draft preparation: Shanbel and Zelalem. Writing – review and editing: Shanbel, Zelalem, Aynalem, Tesfaye, Mourad, Berhanu, Barbara, Shenkute, Kebede, Temesgen, Ermiyas, and Tusa. Project administration: Aynalem, Mourad, Tesfaye, and Barbara. All authors have read and agreed to the submitted version of the manuscript.

Disclosure

Contribution to the filed: Reproductive efficiency is one of the most important factors affecting profitability in small ruminant breeding programmes. Improvement of reproductive traits has more economic impact and traits of interest than improvement of growth rate. However, no comprehensive study has been done yet to evaluate reproductive traits in different sheep breeds under CBBPs in Ethiopia. Thus, evaluating reproductive traits allows development of efficient animal management strategies, which would help to improve the efficiency of the ongoing CBBPs. The presence of positive and moderate to high genetic correlation between TLWW and other traits strongly suggests that incorporating TLWW into the selection index would result in significant enhancement in the reproductive performance of the ewes. By combining genetic selection with improvements in environmental conditions and implementing appropriate culling strategies, the reproductive performance of these sheep breeds can be enhanced effectively.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data will be made available on request.