Methodology for the Definition of Durum Wheat Yield Homogeneous Zones by Using Satellite Spectral Indices

One of the main questions facing precision agriculture is evaluation different algorithms for delineation homogeneous management zones. In present study, a new approach based on use time series satellite imagery, collected during two consecutive growing seasons, was proposed. Texture analysis performed using Gray-Level Co-Occurrence Matrix (GLCM) used to integrate and correct sum vegetation indices maps (NDVI MCARI2) define homogenous productivity zones ten durum wheat fields in southern Italy. The identified through method that integrates GLCM with spectral studied showed greater accuracy (0.18–0.22 Mg ha−1 ∑NDVIs + 0.05–0.49 ∑MCARI2s GLCM) respect methods considered only indices. Best results were also obtained derived by yield previous year or acquired single day. Therefore, survey data over entire study period provided best terms estimated yield; addition clustering allowed further improve estimate better fields.