Monitoring drought-induced degradation of olive and citrus tree crops in the Tadla plain (Morocco) with multi-sensor remote sensing

¹ Geomatics, Georesources and Environment Laboratory, Faculty of Sciences and Techniques, Sultan Moulay Slimane University, Beni Mellal, 23000, Morocco
² Center for Remote Sensing Applications (CRSA), Mohammed VI Polytechnic University (UM6P), Ben Guerir 43150, Morocco
³ Laboratory of Data Science for Sustainable Earth, Sultan Moulay Slimane University, Beni Mellal, 23000, Morocco

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

The Tadla plain, located in central Morocco, has experienced ongoing drought episodes in recent years, causing significant degradation of citrus orchards and olive groves. This study aims to evaluate the changes in these crops from 2018 to 2024 using multi-sensor satellite data (Sentinel-1 and Sentinel-2) and a supervised classification algorithm, the Support Vector Machine (SVM). To characterize the vegetation and water conditions of the crops, several biophysical indices were extracted, including the Normalized Difference Vegetation Index (NDVI), the Modified Soil-Adjusted Vegetation Index (MSAVI), the Normalized Difference Water Index (NDWI), and the Normalized Difference Moisture Index (NDMI). These indices help assess vegetation vigor, biomass, and water stress. Combining optical and radar data improved the detection of degraded areas, especially in sectors exposed to high water stress. Applying the SVM classifier to the combined Sentinel-1 and Sentinel-2 data achieved a high overall accuracy (OA) of 0.927, confirming the reliability of the mapping for monitoring the ongoing degradation of orchards. The diachronic analysis of cultivated areas shows a significant decline: citrus orchards lost about 38% of their area from 2019 to 2024, while olive groves decreased by 32%. At the same time, the reduction in vegetation and water indices indicates a decline in biomass, photosynthetic activity, and leaf water content, highlighting the combined effects of drought and groundwater overexploitation. These findings demonstrate the effectiveness of integrating remote sensing with machine learning techniques for environmental monitoring and agricultural planning. They offer a strategic tool to predict the impact of water stress on perennial crops and support the development of sustainable water management practices in vulnerable regions.