IDENTIFYING SUITABLE SITES FOR RAINWATER HARVESTING USING RUNOFF MODEL (SCS-CN), REMOTE SENSING AND GIS BASED FUZZY ANALYTICAL HIERARCHY PROCESS (FAHP) IN KENITRA PROVINCE, NW MOROCCO

: The rising need for water coupled with an increasing scarcity in many parts across the world especially in the middle east and north african countries (MENA) requires more sustainable solutions for effective water conservation. In Morocco, water ressource is characterized by significant spatio-temporal variability. So, to ensure the availability of water for domestic and agro-industrial uses, it is advised to develop some alternatives that improve the local water resources management throughout the country. Rainwater harvesting (RWH) has been proven to be a very promising alternative to water shortage problem. However, identifying appropriate sites for RWH remains a complex task in the management of rainwater. The present study aims to identify optimal sites for RWH using GIS based Fuzzy Analytical Hierarchy Process (FAHP) method in the Kenitra province, NW Morocco. For preparing thematic layers, several data sources were used including remote sensing data (RS), digital elevation model (DEM), the soil and precipitation data were used to create the necessary database using ArcGIS software. Next, the model of the soil conservation service-curve number (SCS-CN) was adopted to generate the map of the annual potential runoff. Then, five thematic layers including runoff, slope, soil texture, land use/land cover (LULC) and drainage density were assigned apropriate weights for generating the RWH suitability map. The resultant map of runoff depth revealed that it ranges from 137 to 738 mm. Moreover, the RWH suitability map showed that Kenitra province can be classed into five RWH candidate areas: (i) unsuitable (12.7%), (ii) less suitable (10.9%), (iii) suitable (20.3%), (iv) very suitable (36%) and (v) extremely suitable (19.9%). The extremely suitable areas for RWH are distributed in the central and northeastern parts. Based on the area under curve (AUC) of the receiver operating caracteristics (ROC), the success rate for predicting suitable RWH sites was 51%.