Breeding strategies for sustainable intensification of developing smallholder dairy cattle production systems

Abstract

Smallholder dairy cattle production systems in Africa are intensifying production through importation of germplasm from breeding programs conducted in temperate regions to improve commercial cow populations. Presence of genotype by environment interaction results in unfavorable correlated responses. The aim this thesis was to develop strategies for breeding programs in developing countries that can support sustainable intensification of these systems. Specific objectives were (a) to determine desired gains for breeding objective traits, (b) compare progeny testing (PT) and genomic selection (GS) selection strategies, (c) evaluate the economic performance of PT and GS selection strategies and (d) compare genetic gains for economic and non-economic breeding objectives; the Kenya dairy cattle sector was used as a working example. To account for the limited pedigree and performance recording, a five-trait breeding objective and small-sized breeding program were studied. Breeding objective traits, determined based on producer preferences, were milk yield (MY), production lifetime (PLT), calving interval (CI), fat yield (FY) and mature body weight (MBW). Producers were categorized into high intensive group, who placed highest preference on PLT and MY, and low intensity group that placed highest preferences on CI and PLT. MY and FY were the most important traits for processors. Consensus desired gains, based on weighted goal programming, were 2.51, 2.42, 0.22, 0.87 and 0.15% for PLT, MY, CI, FY and MBW, respectively. Comparison of breeding schemes shows that GS schemes had lower accuracies but gave higher responses per year due to shorter generation intervals. Besides genetic gains, economic performance underpins the adoption of selection strategies. GS schemes had between 3.2 and 5.2-fold higher cumulated genetic gain in the commercial cow population and higher gross margins compared to PT schemes. Semen storage made PT schemes more profitable but less so than GS schemes. Functional traits can increase the sustainability of resource poor smallholder systems under harsh environments. Economic breeding objectives yielded undesirable responses in functional traits. Breeding objectives based on desired gains or non-market objectives improved response in functional traits but at a monetary cost. It is concluded that sustainable productivity of smallholder systems can be improved by implementation of local breeding program based on GS, but this requires more emphasis on functional traits, which can be achieved by use of non-economic objectives.