Screening drought tolerance in bread wheat genotypes (Triticum aestivum L.) using drought indices and multivariate

Abstract

Drought is a wide-spread problem seriously influencing wheat (Triticum aestivum L.) production, but development of tolerant genotypes is hampered by the lack of effective selection criteria. The objective of research study was to evaluate the efficiency of several selection indices to identify drought tolerant genotypes under drought stress conditions. Twenty seven bread wheat genotypes differing in yield performance were grown under drought stress, and normal irrigation during 2014-2015 growing season, were evaluated in split plot design with three replications Significant and high positive correlation was found between grain yield in the stress condition (Ys) and (Yp) with indices STI, GMP, YI, and Significant negative correlation was found between Ys with SSI, SI, SSPI, TOL indices. Principal component analysis (PCA) based on the Spearman's correlation matrix, indicated that first PCA (80.6%) and second PCA (18.1%) accounted for 98.7% of variations among the indices. The results of PCA revealed that the screening methods were significantly inter-correlated with each other indicating that several of the statistics probably measure similar aspects of drought tolerance. Cluster analysis classified the cultivars into four groups according to drought tolerance. The results showed that MP, GMP and STI were more effective in identifying high yielding genotypes in both drought-stressed and irrigated conditions, identifying G1 and G10 as more tolerant and G25 and G26 as more sensitive genotype to drought stress.