GIS-Based Integration of Landsat-Derived Indicators for Environmental Assessment of the Shurtan Gas Field, Uzbekistan

Abstract

Arid hydrocarbon landscapes are challenging to assess because surface temperatures, sparse vegetation, and exposed soils may reflect natural dryland conditions and local surface modification. This study applied a reproducible, sensitivity-tested Landsat-based framework to characterize surface environmental conditions within and around the Shurtan gas field, Uzbekistan, during 2014–2025. Annual June–September composites were generated from 484 Landsat 8 and Landsat 9 Collection 2 Level-2 scenes in Google Earth Engine (GEE). Land-surface temperature (LST), the Modified Soil Adjusted Vegetation Index (MSAVI), and the Bare Soil Index (BSI) were normalized and integrated into the Surface Environmental Condition Index (SECI), a relative composite in which higher values indicate warmer surfaces, lower soil-adjusted vegetation signal, and stronger bare-surface expression. Indicator weights were derived using the CRiteria Importance Through Intercriteria Correlation (CRITIC) method. Patterns were evaluated through zonal statistics, land-cover stratification, high-SECI frequency and persistence, the Mann–Kendall trend test, Sen slope estimation, and sensitivity analyses of normalization, spatial extent, and indicator selection. CRITIC assigned weights of 0.428 to LST, 0.324 to inverse MSAVI, and 0.248 to BSI. Mean SECI was higher in the analytical field core (0.863) than outside the field core within the 10-km analysis extent (0.698). Persistent high-SECI conditions occupied 19.8% of the field core, compared with 10.2% outside it. High SECI values occurred mainly in dryland natural vegetation and bare or sparse surfaces, underscoring land-cover context. Significant increasing SECI trends (p < 0.05) covered 51.8% of the field core and 44.0% of the 10-km analysis extent, whereas significant decreasing trends occupied only a limited area. Sensitivity tests showed stability across alternative normalization schemes and spatial extents, while LST was the most influential component. The framework provides a spatially explicit baseline for monitoring changing surface environmental conditions in arid gas-field landscapes and for directing future studies that integrate operational, climatic, and field observations.