Predictive Digital Mapping of Surface Soil Properties using Remote Sensing and Multivariate Statistical Analysis.

Abstract

Accurate prediction of surface soil properties is crucial for agricultural and environmental purposes. This study aimed to utilize geoinformatics approaches and Landsat OLI-8 data to predict specific physicochemical properties of the surface soil in Sulaimaniyah, Kurdistan Region of Iraq (KRI). It also examined the statistical relationships between these properties and spectral reflectance, vegetation cover, soil/vegetation moisture contents, and elevation. The study made use of the Normalized Difference Vegetation Index (NDVI) and Normalized Difference Moisture Index (NDMI), as well as seven bands of the OLI image for the statistical analysis. The results demonstrated a statistical connection between organic matter (O.M.) and vegetation cover based on NDVI. It was observed that the northern parts of Sulaimaniyah exhibited dense vegetation, albeit covering a small area. Generally, mountainous regions had a higher proportion of canopy cover compared to other parts of the arid zone, with moisture availability being the most influential factor on vegetation. Moreover, the majority of the research area showed the highest CaCO3 content and a significant negative relationship was found between vegetation (NDVI) and soil moisture (NDMI) with organic matter (O.M.) and clay. Using geoinformatics datasets and techniques proved valuable in identifying, mapping, and investigating specific surface physicochemical properties in the study area.