Elevated atmospheric CO₂ enhances submergence tolerance in Sub1 introgressed rice varieties and their recurrent...

Rice, as a staple food for more than half of the global population, faces dual challenges from climate change and frequent submergence stress in lowland ecosystems. To elucidate the effect of elevated CO₂ on morpho-physiological responses in submergence tolerant and susceptible rice varieties, we conducted a field experiment featuring two levels of CO₂: ambient (400 ± 10 μmol mol^-1) and elevated (550 ± 20 μmol mol^-1). The experiment included four cultivars of Indica rice (Oryza sativa L.): Swarna-Sub1, IR64-Sub1 (which are tolerant to submergence), along with their recurrent parent varieties, Swarna and IR64 (which are susceptible to submergence). In this research, we employed path analysis utilizing the Partial Least Squares Path Modelling (PLS-PM) method to identify the direct and indirect factors affecting the grain yield of various varieties under the simultaneous effects of increased atmospheric CO₂ levels and submergence stress. The PLS-PM model was formed using 11 contributing factors, which were grouped into five conceptual (latent) variables: (1) antioxidant metabolites (e.g., catalase, peroxidase), (2) non-structural carbohydrates (e.g., starch and sugar), (3) chlorophyll (chlorophyll a & b), (4) yield attributes (fertile grain per spikelet, panicle length, panicle weight), and (5) plant survival parameters (shoot elongation and plant survival). The analysis identified antioxidant metabolites as the primary contributor to grain yield due to its highest regression (path) coefficient (β = 0.988). Non-structural carbohydrates (β = 0.231), yield attributes (β = 0.236), and chlorophyll (β = 0.445) contribute to grain yield. Furthermore, the PLS-PM model indicated a more pronounced impact under elevated CO₂ than ambient CO₂ levels with a direct effect of antioxidant metabolites and non-structural carbohydrates on grain yield. The PLS-PM model did not have multicollinearity issues, and the model fits (R²) were high (R² = 0.79 for ambient CO₂ and 0.76 for elevated CO₂), indicating that more than 75 % of the variance in grain yield was explained by the model constructs. These findings emphasize the potential for identifying and deploying CO₂-responsive genotypes and traits in breeding programs aimed at ensuring rice productivity under future climate scenarios.