Drought is a major abiotic stressor significantly reducing global crop productivity by disrupting key morpho-physiological and biochemical processes, including stomatal conductance, photosynthetic efficiency, and water use dynamics. This study systematically assessed the effects of water scarcity on fifty genetically diverse cotton genotypes under three different conditions: normal field capacity (100% FC, monitored by soil moisture weight), moderate drought stress (60% FC), and severe drought stress (40% FC) during the flowering stage, a critical period for crop yield. Key morphological (shoot length, root length, biomass), physiological (relative water content, stomatal conductance), and biochemical traits (antioxidant enzyme activity, oxidative stress markers) were evaluated across all treatments. The results revealed a gradual decline in trait response as drought severity increased, with significant genotype-dependent variations. Histograms captured frequency distribution trends across different stress levels, while principal component analysis (PCA) differentiated trait expressions among genotypes. Pearson's correlation analysis highlighted strong positive and negative relationships among the traits based on their drought responses. The multi-trait genotype-ideotype distance index (MGIDI) proved to be a valuable tool, identifying G1 (FH-189), G2 (FH-453), G3 (N-5016), G7 (N-868), G26 (N-1011), G44 (FH-941), and G47 (135 − 44/BB) as the most drought-resistant genotypes. These genotypes showed superior performance and maintained higher water status, minimized oxidative damage, and activated effective defense mechanisms under stress. These findings provide valuable insights into the morpho-physiological and biochemical traits associated with drought resilience and lay the groundwork for targeted breeding programs to develop drought-tolerant cotton varieties. Such advancements are crucial for supporting cotton farmers in drought-prone areas, improving crop stability, and ensuring sustainable productivity in water-limited environments.
Shani et al. (2025) studied this question.