In this study, we developed a prediction model for NDVI time series data obtained from the MODIS sensor, covering the period from 2002 to 2022, for rainfed crops near four meteorological stations in the province of Soria. The analysis was conducted at the pixel level, using the Box-Jenkins methodology, and the accuracy of the models was evaluated using the U-Theil index. We found that the models predicted with good accuracy for 30% of the analyzed pixels, with annual seasonality identified as a key factor. The results were published in Sáenz et al., 2023.
