Use of remote sensing (RS) in pistachio yield estimation

In this study, TERRA-ASTER images were used. At the first, pixel(s), with sufficient density were identified by means of colour composition, multi-spectral classification, PC, NDVI and other image processing methods. Study area was Kabootarkhan, a town near Rafsanjan in Kerman Province. Good crop performance "Yield" was expressed with field measurement of an index, named "Index of Leaf Area (ILA)" and as "Crop Variable" of selected pixels. "Yield" and "ILA" index were considered as the dependent variables for multivariate analysis with spectral reflectance and soil properties as independent variables. A data set with sufficient variation in the dependent and independent variables was compiled. It was expected that the variation could be explained by the spectral and soil variables. Furthermore investigated, how well "ILA" index variable is related to crop productivity. Can a "Yield" index be defined from "ILA" index? The results showed that the variability in the pistachio yield in the study area, accepted to be for 94.3 percent (adjusted R2 = 94.3) dependent on three main independent variable attributes: "Crop", "Soil", and "Other" parameters related to crop yield and production.