Statistical analysis of multi-environmental rice yield trial in Bastar district of Chhattisgarh

Abstract

The main purpose of adaptability of crop varieties in multi-location trials (MVAT) is a problem in the presence of genotype by environment (G×E) interaction. Various statistical techniques are available to analyse the data in MVAT. But, the information on these methods and their relative performance on evaluation of adaptability of rice varieties are limited. Therefore, a study was conducted to statistically analyze of multi-environmental rice yield data to compare the statistical methods available for analysis of MVAT through application to MVAT data of rice. MVATs data of rice were collected from the farmer’s field at five diverse location of Bastar district. It consisted of data on performance of varieties at 5 locations over 3 replicated fields. The statistical techniques i.e. stability parameters, ranking and multivariate techniques were tested with yield data. Different stability method was considered for various aspect of variability across environment and hence produces different results for same data. Therefore, interpretations of stability of varieties vary according to different parameters considered. Multivariate methods describe G×E interaction effectively with plots that are easy to understand. Therefore, considering result from different stability and multivariate method altogether recorded for evaluation of adaptability of rice varieties.