Determination of Mega‐Environments for Peanut Breeding Using the CSM‐CROPGRO‐Peanut Model

Abstract

Breeding for locally adapted cultivars requires a subdivision of the target region into mega‐environments. Crop models could assist in generating the required data for mega‐environment determination. The objective of this study was to determine whether subdividing the peanut (Arachis hypogaea L.) production areas in Thailand into mega‐environments using a crop simulation model would be justified. The Cropping System Model (CSM) CROPGRO‐Peanut was used to simulate pod yield of 17 diverse peanut lines for 130 locations covering all peanut production areas in Thailand. The data were statistically analyzed, and the genotype and genotype × environment (GGE) biplot method was used to subdivide the peanut production areas into subregions. The results reveal that the genotype × location interaction accounted for only a small proportion of total yield variation for all years. The analyses of yearly data by the GGE biplot shows inconsistent results across years for location grouping as well as for the winning genotypes of the individual location‐groups. The GGE biplot analysis of the mean data over 30 yr also indicates a similarity in genotype discrimination for all the locations. The results from this study show that the subdivision of peanut production areas into mega‐environments is not justified for Thailand. Therefore, for peanut breeding, Thailand should be considered as one mega‐environment.