Abstract
Agricultural suitability analysis using traditional methods is still arguable due to the uncertainty and subjectivity resulting from manual evaluations. The current work provides a novel framework for integrating the analytical hierarchy process (AHP) with fuzzy logic under the geographic information system (GIS) platform to generate suitability maps for cultivating wheat, broad bean, and maize under center pivot irrigation systems. The research was executed in an arid region (30229 ha) in the western Nile Delta fringes, Egypt. Meteorological data, digital elevation model, and samples collected from seventy soil profiles and fourteen artesian wells were analyzed to characterize local climate conditions, landscape characteristics, and irrigation water quality. The main and sub-criteria were prone to AHP to specify the relative importance (weight) of each factor. Using GIS tools, raster layers were created, assigned scores (fuzzy membership functions) according to crop requirements, and complied in accordance with the weighted sum algorithm to produce final crop suitability maps. Results revealed that climate conditions were highly (S1) and moderately (S2) suitable for winter crops (wheat and broad bean) but marginally suitable (S3) for summer crop (maize). Soil salinity, sodicity, and depth were the most important determinants of landscape suitability. Accordingly, the land resources in the studied region were suitable (S1, S2, and S3) for the selected crops; nevertheless, 193 and 275 ha were currently not suitable (N1) for broad bean and maize, respectively. Potential salinity and specific ion (sodium and chloride) toxicity hazards were the main constraints for groundwater irrigation. The center pivot irrigation would meet wheat and maize requirements but adversely affect broad bean yield. Overall, groundwater quality contributed to 46% of site suitability for crop production followed by landscape factors that contributed to 42% and climate conditions that accounted for 13%. The final suitability maps affirmed high priority for wheat cultivation in the studied region since the S1 and S2 classes encompassed 90 and 10%, respectively. Moreover, maize ranked as the second suitable crop with 55, 42, and 35 of the total area fitting the S1, S2, and S3 classes, respectively. The third place was due to broad bean with S2 and S3 classes representing 53 and 47% of the total area, respectively. Our study can offer a replicable framework to integrate AHP with GIS-fuzzy logic for sustainable food crop production in drylands.