Simulation of genetic and environmental interaction of three tropical hair sheep breeds for meat production

Abstract

Statistical and systems analysis techniques were used in a complementary way to achieve the objectives of obtaining estimates of genetic and environmental parameters for growth and reproduction characters of three tropical hair sheep breeds (Brazilian Somali, Santa Ines and Morada Nova) and examining genotype by environment interactions in order to optimally match these three genotypes with different environments. Initially, data from a total of 524 lambs sired by 21 rams and out of 190 ewes were included in the analysis. These data were collected from 1980 to 1984 at Sobral, Brazil from an experiment designed to compare the performance of the three breeds. Lamb characters analyzed were weight at birth, weaning, 8 mo, 10 mo and yearling, and mortality rate up to weaning. Ewe characters analyzed were total lamb weight born, total lamb weight weaned, weaning rate and prolificacy rate. In the analysis of binary characters such as mortality and prolificacy rates, two procedures, general linear model (GLM) and Linear Logistic Regression, were used and compared. The logistic analysis yielded similar results to that of GLM analysis only when all interactions were of small magnitude. Whenever some of these interactions were significant, the use of GLM yielded correct but misleading results. Factorial design experiments were simulated and analyzed. The first factorial was a 3 x 3 x 2 (3 breeds x 3 levels of nutritional supplementation x 2 types of supplement). In the second factorial (3 breeds x 2 quantities of supplement), one supplement (napier grass) with two different quantities (200g/d and 400g/d) was simulated. Analysis of the simulations indicated that digestible energy was the most critical factor limiting performance of the simulated flocks in the dry season. The analysis also indicated the need for supplementing lactating ewes at the beginning of lambing season. Increasing the amount of napier supplement increased productivity of the simulated Morada Nova and Santa Ines flocks. Analysis of the simulations indicated that in the northeast region of Brazil the Santa Ines has the highest level of productivity in areas with above average level of forage production. In harsher, more arid areas the Brazilian Somali has higher total flock productivity than the other two breeds. Finally, the Morada Nova is expected to perform well in most areas if moderately supplemented.