Spectral vegetation indices as nondestructive tools for determining durum wheat yield

Remote sensing measurements may be a useful tool for quantifying crop development and yield. Our objective was to study the potential of using spectral reflectance indices to provide accurate and nondestructive estimates of physiological traits determining yield in durum wheat [Triticum turgidum L. subsp. durum (Desf.) Husn.]. Twenty-five genotypes were grown under rainfed and irrigated conditions in northeastern Spain. Reflectance from the vegetation at different growth stages was measured and the following spectral indices calculated: simple ratio (SR), normalized difference vegetation index (NDVI), and photochemical reflectance index (PRI). Crop dry mass (CDM), leaf area index (LAI), and green area index (GAI) were measured. All the indices and grain yield were greater under irrigated than rainfed conditions. LAI was the crop growth trait that most closely correlated with the spectral reflectance indices, with SR and PRI being the best and the worst indices, respectively, for the assessment of crop growth and yield. In rainfed conditions, the spectral reflectance indices measured at any crop stage were positively correlated (P < 0.05) with LAI and yield. Under irrigation, correlations were only significant during the second half of the grain filling. The integration of either NDVI, SR, or PRI from heading to maturity explained 52, 59, and 39% of the variability in yield within genotypes in rainfed conditions and 39, 28, and 26% under irrigation. Our results suggest that for durum wheat, the usefulness of the SR and NDVI for calculating green area and grain yield is limited to LAI values < 3.