Long-term spatiotemporal surface water dynamics using Google Earth Engine in southeastern Morocco

Abstract

Monitoring changes in water areas within arid regions is vital for directing water resource development and ensuring efficient use, and addressing the impacts of drought and climate change. Thus, in this research, we examined the annual changes in surface water at the El Mansour Eddahbi (EME) dam (southeastern Morocco) from 1984 to 2023. To achieve this, we exploited the Google Earth Engine’s potential and applied four water indices – MNDWI, EWI, NWI, and AWEI – to Landsat satellite images. Subsequently, for each year, a supervised classification utilizing random forest was implemented to accurately extract and identify water areas. The findings revealed that the EME dam’s water surface area fluctuates substantially on an inter-annual basis. Besides, Pearson’s analysis demonstrates that the EME dam’s water surface area has a strong positive correlation with drought indices such as SPI-12 and SPEI-12. This highlights the link between changing surface water, drought, and the need for adaptive water management under climate change.