Investigation of farmer-led breeding goals and strategies in smallholder dairy farming systems to cope with variation in feed sources and quality
Chawala, Aluna Raphael
Breeding goals and long term breeding strategies for smallholder dairy farmers in Sub-Saharan Africa (SSA) have not been defined clearly. There is a need for a clear breeding goal that reflects priorities, need and expectations of smallholder farmers, who represent the main farming system. Crossbreeding of Bos taurus and Bos indicus has been advocated as a way of improving biological efficiency of dairy cows in SSA. However, there is little realised genetic gain among the crossbred dairy animals mainly due to lack of clear breeding goals and long-term breeding strategy beyond the production of the first filial generations (F1s). The overarching aim of this PhD thesis was to investigate breeding goals and strategies of dairy cattle evaluation that can be used as a guide for future dairy cattle breeding programmes in low and medium production environments in Sub-Saharan Africa, using Tanzania as a case study. The key hypothesis of this PhD study was that there is a mismatch between the priorities and preferences of smallholder farmers with regards to animal traits and priorities and policies set centrally by government. In order to address this hypothesis, the following specific objectives were addressed: i) to analyse previous scientific literature on the improvement of animal traits in Sub-Saharan Africa and assess the impact of production systems and breed types on dairy cattle performance; ii) to examine the level of genetic progress that has been achieved in the centrally-run breeding programmes of Mpwapwa cattle in Tanzania; iii) to identify farmer-preferred traits and assess their relative importance in smallholder dairy production systems; and iv) to develop and assess different breeding strategies incorporating smallholder farmer preferred traits.To address the first objective, a systematic literature review of 64 research studies published between 1980 and 2018 on farmers' and institutional dairy trait preferences was made. Meta-analysis was used to examine the production and reproduction performance of dairy cattle in smallholder and large-scale dairy production systems. Overall, qualitative analysis and meta-analysis of published literature showed a main research focus towards fertility traits (age at first calving, calving interval, days open and number of services per conceptions), milk yield and disease resistance or survival traits. Butterfat, protein and workability traits such as temperament were rarely considered. The performance of dairy cattle was higher in large-scale farms and centrally-based breeding stations compared to smallholder dairy farmers. Low performance of dairy cattle in smallholder dairy farms was associated with the use of inferior bulls and poor animal husbandry practices. This study suggested that production systems and genetics play a large role in increased dairy productivity in sub-Saharan Africa. Future dairy research should place a greater emphasis on the interactions between improved feeding, disease control and genetics at a production system level to inform profitable combinations of animal traits.The second objective was to assess the effectiveness of an actual centrally-based breeding programme in Tanzania, which was set up 40 years ago as a possible nucleus to provide genetic improvement to the smallholder dairy farmers. A total of 1,003 total lactation milk records from 385 cows and 78 sires collected from 1967 to 2012 from Tanzania Livestock Research Institute – Mpwapwa were used to estimate genetic parameters and genetic progress for production and fertility traits in a synthetic dual-purpose breed of cattle. Genetic parameters were estimated using mixed animal models of statistical analyses. Heritability estimates for 305-day milk yield (h2 = 0.44 ± 0.04) and calving interval (0.10 ± 0.05) were statistically greater than zero (p<0.05) suggesting that these traits could be genetically improved with genetic selection. However, overall negative phenotypic and genetic progress was observed over the past four decades. The key recommendation was that a further understanding of the breeding goal and performance of the breed in a smallholder dairy production system is required for an effective and sustainable genetic improvement programme. Because of the absence of animal performance data in smallholder farms and lack of clear breeding goals, the next step was to identify breeding goals for smallholder dairy farmers in a participatory manner. Objective 3 was addressed by conducting interviews with stakeholders, including senior research scientist from International Livestock Research Institute (ILRI), livestock and extensions officers from local government authorities, and a choice experiment survey with smallholder dairy farmers in Tanzania to identify farmer-preferred traits and their relative importance. Participation of smallholder dairy farmers was considered as critical in the establishment of breeding goals. The study was conducted through visits to 555 randomly selected dairy farms in the sub-humid Eastern coast and temperate Southern highlands of Tanzania. The choice experiment data were analysed using a conditional logit model. Results showed that from a farmer’s viewpoint, the most important dairy traits included high milk yield (emphasis coefficient = 1.43±0.059), good fertility (0.85±0.050), easy temperament (0.76±0.066), low feed requirements (-0.56±0.092) and disease resistance (0.48±0.048). Farmers’ trait preferences differed between agro-ecological zones and production systems due to variation of climatic conditions, feed resources and local infrastructure. Adaptability to the local environment was considered as a fundamental trait for selecting dairy cattle across production systems and agro-ecological zones. Farmers were willing to invest in improved dairy cattle showing desired traits at an affordable price (coefficient = -0.001±0.0003). Results from addressing this objective provide evidence for designers of breeding programmes to take consideration of specific farmers’ preferred traits. A smallholder farmer index comprising all these traits can be developed based on the relative emphasis placed by the farmers on each trait.A semi-stochastic simulation was then used to investigate dairy cattle breeding scheme designs for different breeding strategies and to determine genetic progress for the farmer preferred traits in smallholder dairy production systems (Objective 4). Identified animal traits in smallholder dairy production systems from objective 3 were simulated. Three key breeding strategies were investigated. The first breeding strategy assumed genetic improvement through continuous importation of genetically superior exotic dairy bulls to SSA. Importation strategies were considered based on either the smallholder farmer index from objective 3 or the index in the country of export, which is most often the case. The second strategy assumed that semen from elite exotic bulls were imported and used only once to produce F1 animals. Thereafter, elite animals were selected from within the F1 population based on the smallholder farmer selection index, to establish a synthetic breed with continuous crossing. Finally, the third strategy was to improve the indigenous population by genetically selecting the best bulls based on the smallholder farmer selection index from within the population. Results from the three breeding strategies showed a positive genetic progress of all breeding goal traits over generations. Increase in proportion of cows bred by improved sires through artificial insemination (AI) resulted in a greater genetic response for all selection scenarios. Three mating schemes of 100%, 50% and 30% AI uptake were investigated to quantify the genetic gain of animal traits and selection indexes. All scenarios showed that the importation breeding strategy led to an overall higher genetic progress compared to synthetic breed and within indigenous breed selection strategies, probably due to the high relative emphasis on milk yield by the smallholder farmers. For example, after 15 generations of selection, the genetic response of the importation strategy regarding the farmer index exceeded the corresponding genetic response of the synthetic breed strategy by 20.26%, 51.41% and 65.55% for the three AI schemes, receptively. The former also exceeded the genetic response of the indigenous breed strategy by 43.06%, 64.89% and 75.02% for the three AI schemes, receptively. Our simulation studies have demonstrated the value of including farmer preferred traits in genetic selection of imported genetic material compared to selection based on the circumstances of the exporting country. After 15 generations of selection, the genetic gains from continuous importation of animals based on farmer preferred traits were 8.31% - 9.83% higher compared to importing based on circumstances of the exporting country. Potentially there is an opportunity for breeders to choose an appropriate breeding strategy that fits a particular production environment.In summary, participatory breeding goals for smallholder dairy farmers in Tanzania were identified and documented. The amount of emphasis for identified breeding goal traits was quantified. Furthermore, appropriate breeding strategies were developed and the genetic progress for each individual trait in a particular breeding strategy was quantified. These findings contribute to the body of knowledge on breeding and genetics in smallholder dairy production systems in SSA. Results from this thesis can be used by farmers, government and other development partners in planning sustainable breeding programmes in Tanzania and Sub-Saharan Africa at large.