Land degradation intensity mapping using environmental factors and Sentinel-2 derived spectral indices in an arid region in south of Iran

To mitigate land degradation and desertification as an environmntal issue, it is crucial to monitorland degradation intensities, identify influential factors, and implement necessary measures. Thisstudy utilized remote sensing data and logistic regression modeling to assess desertificationin Larestan County. Multiple indicators were considered in this study, encompassing climatefactors (such as rainfall, evapotranspiration, and aridity index), groundwater indicators (includingelectrical conductivity, chloride content, sodium absorption ratio, and groundwater level decline),soil indicators (such as EC, texture, and organic matter content), land use and land cover (LULC)type, and wind erosion. The logistic regression model was employed to identify the most influentialfactors contributing to desertification. The findings revealed different risk classes: a small low-riskclass in the eastern and southern regions covering 2.4% of the total area, a moderate-risk classin the foothill-plain areas covering 38.3% of the total area. The high-risk class of desertificationis mainly concentrated in the central part of the study area, adjacent to regions with moderaterisk. It is characterized by relatively large patches, particularly in the southwest of the interiorplains, covering approximately 1,980 hectares, which accounts for 7.9% of the total area. Theconcentration of high-risk desertification in specific areas highlights the urgent need for proactivemeasures to preserve the environmental balance and sustainability of the study area.