Monitoring degradation in arid and semi-arid forests and woodlands: The case of the argan woodlands (Morocco)

Arid and semi-arid forests and woodlands (hereafter called «dryland forests»), in spite of their ecological and social importance, have received little attention in land change studies. Growing evidence shows that these forests have been receding at very high rates in many places, suggesting a need for a better understanding of the processes and causes of dryland forest degradation. Changes in the extent of dryland forests are debated in part because estimates of forest and woodland areas in drylands are uncertain. Causal explanations of the degradation tend to draw on the literature on desertification and tropical deforestation, and to emphasize either local or remote, and either social or biophysical drivers. This study contributes to a better understanding of dryland forest degradation as a basis for conservation policies. Firstly, we argue that monitoring arid and semi-arid forests and woodlands using area estimates may lead to an underestimation of the severity of change because tree density change often exceeds area change. Secondly, we argue that the analysis of degradation processes in these multifunctional landscapes should integrate both local and remote, and both social and biophysical factors. We use a case study of degradation in the argania woodlands in semi-arid to arid Southwest Morocco to test these two claims. We used gridded tree counts on aerial photographs and satellite images to estimate forest change between 1970 and 2007, and we tested several possible causes of change on the basis of original socio-economic field surveys and climatic and topographic data. We found that forest density declined by 44.5% during this period, a figure that is significantly underestimated if forest area change is used as a measure of degradation. Increasing aridity and, to a lesser extent, fuelwood extraction were related to forest decline. No effect of grazing by local livestock was found.