Monitoring desertification in Ordos City using integrated vegetation and soil indicators

Abstract

Desertification is a major environmental challenge in arid and semi-arid regions. This study examines the spatiotemporal evolution of desertification in Ordos City, China (2001–2020) using long-term remote sensing monitoring and the Geographical Detector Model (GDM). To improve classification accuracy, we developed a multi-indicator framework integrating Fractional Vegetation Cover (FVC), Modified Soil-Adjusted Vegetation Index (MSAVI), and Soil Organic Carbon (SOC). Results indicate an overall decline in desertification, with severely desertified land (SED) decreasing from 32.65 % in 2001 to 18.48 % in 2020, while moderately desertified land (MD) increased by 15.81 %. The most significant improvements occurred between 2015 and 2020, particularly in northwestern and central Ordos, including Hanggin Banner, Otog Banner, and Dalad Banner. Analysis of climate driving factors identified land surface temperature (LST) (q = 0.26), wind speed (WS) (q = 0.24), and precipitation (PRE) (q = 0.20) as the primary contributors, while topographic factors (elevation and slope) had minimal influence. LST and WS had the strongest impacts, positively correlating with desertification, whereas PRE exhibited a negative correlation. Interaction analysis revealed LST and WS had the most significant combined effect, followed by PRE and LST, while SLO and DEM interactions had the weakest influence. Those findings, combined with environmental policy analysis, provide a deeper understanding of desertification processes and highlight the need for continuous monitoring and targeted measures in vulnerable regions.