Stability and phenotypic performance of upland rice subjected to water deficit using a multi‐trait index

Abstract Drought stress is one of the major threats to food security and the sustainability of rice ( Oryza sativa L.) cultivation worldwide. Identifying drought‐tolerant genotypes requires multi‐environment evaluations that combine contrasting water regimes, locations, and crop years. In this context, the use of indices that integrate stability and performance across environments is essential. This study evaluated the stability of 25 upland rice lines in nine environments under both drought stress and favorable water conditions. The traits assessed were grain yield, days to flowering, spikelets per panicle, 1000‐grain weight, and spikelet sterility. Genotype selection was performed using the weighted average of absolute scores of BLUPs (WAASB) from the singular‐value decomposition of the best linear unbiased prediction interaction matrix and the multi‐trait stability index (MTSI). The results highlighted cultivar Douradão (G23) and lines G2, G21, and G11 as the most stable and high‐performing genotypes across environments, demonstrating their potential for recommendation under both water‐limited and favorable conditions. Relying solely on single‐trait evaluations, such as grain yield, may mislead the selection of upland rice genotypes for drought tolerance due to strong environmental influence and low heritability. In contrast, multi‐trait indices like WAASB and MTSI enable more accurate identification of genotypes that combine stability and performance across contrasting water regimes.