Abstract
Human activities, which affect ecosystem dynamics, pose a continuous challenge to individuals and communities trying to survive in arid and semi-arid regions. The development of a method to employ rainwater harvesting (RWH) in the management of agricultural drought in arid and semi-arid regions comprised two phases: (i) detection of agricultural drought in Egypt’s El-Beheira governorate using a normalized difference vegetation index differencing technique and (ii) the delineation of RWH locations potentially suitable for the management of agricultural drought in the region using a GIS decision support system (DSS). Temporal vegetation cover analysis showed significant spatio-temporal changes that have occurred in the last 40 years: a general decrease in vegetation cover reflecting a trend towards ecosystem degradation, contrasted by a greening trend in some pockets within the region. Potentially suitable rainwater harvesting areas for agricultural drought management and attendant vegetation recovery were delineated in the region using DSS. The model generated a RWH map with five categories of suitability: excellent, good, moderate, poor and unsuitable. On average, 10.9 % (1104.17 km
2
) and 12 % (1215 km
2
) of the study area was classified as excellent and good for RWH, respectively, while 11.7 % (1185.21 km
2
), 15.4 % (1560 km
2
) and 50 % (5065 km
2
) of the area were classified as moderate, unsuitable and poor, respectively. Most of the areas with excellent to good suitability predominantly lie in areas which faced severe drought between 2010 and 2014. To successfully implement the drought management plan, a number of RWH sites within the excellent areas must be developed.