Variation in Water‐Use Efficiency and Its Components in Wheat: II. Pot and Field Experiments

Abstract

Two models have been suggested to facilitate the study of genetic variation for seasonal water‐use efficiency (WUE) and its components in spring bread wheat (Triticum aestivum L.). These models were used to estimate the contribution of each component trait to variation in WUE. Eight diverse bread wheat cultivars were grown in replicated pot and field experiments under well‐watered and droughted conditions. Significant genotypic variation was observed for WUE (ratio of grain yield to water used), evapotranspiration efficiency (ETE, ratio of total dry matter to water used), and harvest index (HI, ratio of grain yield to total dry matter). Modern short cultivars, on average, had higher WUE than old tall cultivars in well‐watered (1.225 vs. 0.984 g kg−1) and droughted (1.198 vs. 0.860 g kg−1) pot conditions, due mainly to higher HI. Two old tall cultivars exhibited highest values for ETE. Harvest index and ETE accounted for 85 and 15% and for 81 and 19% of variation in WUE in well‐watered and droughted conditions, respectively. Harvest index had a higher direct effect and was more highly associated with WUE than ETE. The negative correlation between HI and ETE was not significant. Carbon isotope discrimination (Δ) showed negative correlation with ETE in well‐watered ( − 0.82) and droughted ( − 0.68) treatments. Grain yield and HI of modern short cultivars, on average, were higher than those of old tall cultivars under field conditions. Mean Δ was lower in droughted (17.9%o) than in well‐watered (18.9%o) field conditions. Plant growth rates estimated at boot stage and at anthesis were negatively associated with a drought susceptibility index based on grain yield. Drought susceptibility index was not correlated with Δ. Water‐use efficiency in wheat could be improved by improving either transpiration efficiency or ETE.