Association between canopy reflectance indices and yield and physiological traits in bread wheat under drought and well-irrigated conditions

Spectral reflectance (SR) indices [NDVI (R900 – R680/R900 + R680); GNDVI (R780 – R550/R780 + R550); and water index, WI (R900/R970)]; and 6 chlorophyll indices (R740/R720, NDI = R750 – R705/R750 + R705, R780 – R710/R780 – R680, R850 – R710/R850 – R680, mND = R750 – R705/R750 + R705 – 2R445, and mSR = R750 – R445/R705 – R445) were measured with a FieldSpec spectroradiometer (Analytical Spectral Devices, Boulder, CO) on bread wheat (Triticum aestivum L.) genotypes adapted to irrigated and drought conditions to establish their relationship with yield in field-grown plots. Bread wheat genotypes from the International Maize and Wheat Improvement Center (CIMMYT) were used for this study in 3 experiments: 8 genotypes in a trial representing historical progress in yield potential, and 3 pairs of near-isolines for Lr19, both of which were grown under well-watered conditions; and the third experiment included 20 drought tolerant advanced genotypes grown under moisture stress. These were grown during the 2000 and 2001 spring cycles in a temperate, high radiation environment in Obregón, NW México. The 9 SR indices were determined during grain filling along with canopy temperature depression (CTD), flag leaf photosynthetic rate, and chlorophyll estimates using a SPAD meter. The relationship of SR indices with grain yield and biomass fitted best with a linear model. NDVI and GNDVI showed positive relationships with grain yield and biomass under well-irrigated conditions (r = 0.35–0.92), whereas NDVI showed a stronger association with yield under drought conditions (r = 0.54). The 6 chlorophyll indices showed significant association with yield and biomass of wheat genotypes grown under well-irrigated conditions (r = 0.39–0.90). The association between chlorophyll indices and chlorophyll estimates was correlated (r = 0.38–0.92), as was the case for photosynthetic rate (r = 0.36–0.75). WI showed a significant relationship with grain yield in wheat genotypes grown under drought stress conditions (r = 0.60) as well as with grain yield and biomass under well-irrigated conditions (r = 0.52–0.91). The relationship between WI and CTD was significant (P = 0.05) in both environments (r = 0.44–0.84). In conclusion, the SR showed potential for identifying higher-yielding genotypes in a breeding program under dry or irrigated conditions, as well as for estimating some physiological parameters.