Evaluation Of Sen2agri System Over Semi-Arid Conditions: A Case Study Of The Haouz Plain In Central Morocco

For semi-arid ecosystems, the remote sensing based crop type classification is often hampered by heterogeneous agricultural landscapes with mixed crops and small scale land use. Under these complicated conditions, the use of high spatio-temporal satellite time series data can improve classification results. In this context, Sentinel-2 for Agriculture (Sen2Agri) provides an open source processing system to exploit Sentinel-2 imagery in an operational way for crop monitoring. This research was conducted to explore and evaluate the potentiality of Sen2Agri system for crop type mapping in a highly fragmented and heterogeneous agricultural landscape in Haouz plain, Morocco. Based on the phenological information from field campaign, Sentinel-2 time series data for 2018 agricultural season, were processed to classify the main crop types dominated in the study area. After classification, accuracy assessment was undertaken based on generated crop map and resulted in Overall Accuracy (OA) of 85.6 and Kappa coefficients of 0.81.