Remote sensing based on agricultural soil salinity assessment in Soc Trang province, Vietnam

This study aimed to assess the spatial distribution of agricultural land use on saline soils in Soc Trang province in 2024.Multi-temporal Sentinel-1A radar satellite imagery was utilized for land-use classification through a supervised classification approach using the Random Forest algorithm implemented on the Google Earth Engine platform.The resulting agricultural land use map was then overlaid with saline soil data from the Department of Agriculture and Rural Development of Soc Trang province to analyze spatial patterns concerning salinity levels.The classification output identified 11 land use categories: (1) water bodies, (2) built-up areas, (3) aquaculture, (4) rice-shrimp farming, (5) double rice crop, (6) triple rice crop, (7) other perennial crops, (8) other annual crops, (9) forest, (10) coconut, and (11) sugarcane.The classification demonstrated high accuracy assessment with an overall accuracy (T) at 91.26% and a Kappa coefficient at 0.89.The results revealed clear spatial differentiation in agricultural land use based on different soil types.Areas with low to moderate soil salinity had mainly none to low-tolerant crops such as rice, annual crops, coconut, sugarcane, and other perennials.In contrast, high-salinity areas were mainly used for shrimp aquaculture.Coastal areas with very high salinity were associated with protective mangrove forests.This study highlights the effectiveness of Sentinel-1 radar imagery in land use classification and monitoring in saline environments.Furthermore, it provides a scientific basis for land-use planning, crop cultivation, and soil salinity management toward the region's adaptive and sustainable agricultural development.