Modeling changes in climatic variables in Morocco using remote sensing data and statistical techniques

Abstract

In the arid Rhamna region of Morocco, climate change has taken a significant toll on the vegetation in the area; severe drought since 2018 has heavily impacted the agricultural sector of the nation, which makes up 39% of the labor workforce. Existing major crops, such as olive trees, require more water than can be sustainably provided, and will deplete the groundwater of the region in the coming decades if they continue to be grown. In order to accelerate the transition towards more sustainable cropping systems, it is crucial to understand the changes in key climatic variables in the region, including spatial patterns and temporal trends. In order to understand these changes, statistical analyses of time-series data for soil moisture, precipitation, land surface temperature, and evapotranspiration were conducted, modeling the changes over time from the year 2012 to 2022. The data used is available to the public in Google Earth Engine, and processing was done locally using Python libraries. Preliminary analyses on 8-day aggregates of the data yield an R² value of 0.6, suggesting that seasonal trends in the Rhamna region can be accurately approximated by these mathematical models and therefore aid in decision-making.