GIS-Based PCA–MDS Soil Fertility Index (SFI) Assessment for Coastal Soil Management

Coastal agroecosystems are increasingly vulnerable to soil degradation due to intensive agricultural practices, salinity stress, and climate-driven pressures, threatening long-term food security. The main objective was to assess soil fertility status and management suitability by combining multivariate soil diagnostics with geospatial and decision-analysis tools in the Mnasra region of the Gharb Plain, Morocco. Thirty surface soil samples (0–20 cm) were collected and analyzed for key physical and chemical properties. Multivariate statistical analyses including principal component analysis (PCA) were applied to identify dominant soil factors and derive a Minimum Data Set (MDS). A Soil Fertility Index (SFI) was then constructed using PCA-based weighting and standardized scoring, and spatially interpolated using Ordinary Kriging (OK) in GIS. Results revealed substantial variability among soil parameters, particularly in clay content (ranging from sandy to silty-clay textures), organic matter, and nutrient concentrations. Pearson correlation analysis showed strong texture-driven relationships. PCA explained 49.9% of total variance in the first component, identifying texture, carbonate content, and nutrient retention as dominant fertility controls. Based on PCA loadings, a MDS was established to construct a SFI. SFI values ranged from 0.175 to 0.836, classifying soils into low (30%), moderate (23%), good (27%), and high fertility (20%) classes. Spatial interpolation using OK revealed a clear fertility gradient from low-fertility coastal sandy soils to higher-fertility inland and river-influenced zones. The PCA–MDS–SFI framework proved effective in synthesizing complex soil variability into a quantitative, spatially interpretable fertility assessment, supporting sustainable soil management in coastal Mediterranean agroecosystems.