Evaluation of agricultural systems using remote sensing time series and dynamic prediction models

Currently, in the context of climate change, the need to monitor crops and predict their evolution and production has become an urgent need. The current remote sensing systems allow this monitoring to be carried out synoptically in large regions, at different scales and with a high temporal frequency. Specifically, the satellites of the European Union’s COPERNICUS Program provide information in the optical and Radar domains with a high spatial and temporal resolution that will allow a highly detailed monitoring, and which should be both operational and reliable. For this, it is necessary to develop analysis and classification models for large amounts of information and high computing capacities. In the academic environment of agriculture, there are capabilities to develop monitoring models, but automatic artificial intelligence algorithms and computing capabilities are not available yet. In contrast, in the technology industry, computing capabilities are available but not the knowledge of crops functioning and their relationship to remote sensing information.