Comparative analysis of parametric and non-parametric statistics for grain yield stability in rice (Oryza sativa L.)

Basmati rice is an important cereal crop occupying a unique position in Indian agriculture. More than 90% of global rice is produced and consumed in Asia and plays a crucial role in the entry of mineral nutrients into the food chain. Identification of stable genotypes is of great significance because the environmental conditions vary from year to year and across the locations. In present study, thirty-six Basmati rice genotypes were evaluated in a randomized block design for two consecutive growing seasons (Kharif, 2016 and Kharif, 2017) under three crop establishment methods viz., direct seeded rice, conventional transplanted and system of rice intensification. Parametric and non-parametric stability methods were employed for the identification of high yielding and stable genotypes. Analysis of variance suggested that the main effect of environment contributed more than 55% of the total variation, followed by genotypic main effect (30.43%) and genotype × environment (G × E) interaction (12.95%). Best linear unbiased prediction (BLUP) indices identified Pusa 1734-8-3-85, Pusa Basmati 1 and HKR 08-425 while additive main effect and multiplicative interaction (AMMI) based indices identify HKR − 11–509, Improved Pusa Basmati 1 and Pusa 1656-10-705 as most stable genotypes. BLUP indices, Annicchiarico, EV of AMMI indices and NP_i⁽²⁾, NP_i⁽³⁾, NP_i⁽⁴⁾ of stability analysis showed a positive association with grain yield and could be used for the identification of high yielding and stable individuals.