Detecting desertification in the ancient oases of southern Morocco

Abstract

Understanding what led to desertification in the long-term is crucial for adaptation to climate change and pressures on resources in North Africa, but existing maps do not accurately show the extent of degraded land or the traditional water systems which underpinned cultivation. These products rely on recent vegetation trends and hindcasted statistical data. Desertification which occurred prior to the later twentieth century is poorly represented, if at all. However, large areas of abandoned fields are distinctive in satellite imagery as brightly reflectant and smooth surfaces. We present a new and open-source machine-learning workflow for detecting desertification using satellite data. We used Google Earth Engine and the random forest algorithm to classify five landcover categories including a class representing desertified fields. The input datasets comprised training polygons, a 12-band Sentinel-2 composite and derived tasselled cap components, and a Sentinel-1 VV-polarisation composite. We test our approach for a case study of Skoura oasis in southern Morocco with a resulting accuracy of 74–76% for the desertification class. We used image interpretation and archaeological survey to map the traditional irrigation systems which supply the oasis.

A high-precision oasis dataset for China from remote sensing images

“…Clarifying the distribution location and area change of oasis is of great significance for scientific research in arid zones. Although there have been many research results on oases in individual regions 9 , 10 or certain areas 11 – 13 , they are generally characterized by fragmentation and a lack of comprehensive and macroscopic understanding of oases on a large scale, such as the national and global scales. Unlike geographic units with detailed distribution datasets, such as glaciers 14 , 15 and lakes 16 , 17 , no high-precision oasis-related datasets are available in the world to date.…”