Land Use and Land Cover Change from 1998–2024 in the Shumba Watershed, Peru: A Remote Sensing-Based Assessment.

Abstract

The Shumba watershed, located in the province of Jaén, played an essential role in the ecological balance and water supply of the region. The objective of this study was to analyze changes in land cover and land use between 1998 and 2024, using spectral indices of vegetation, soil, and water derived from Landsat 5 and 8 images. The processing was performed in Google Earth Engine and ArcGIS 10.5, applying the Random Forest algorithm for supervised classification. The results indicate an increase of 2708.11 ha in the category “Mosaic of crops, pastures and natural spaces” and 122.50 ha in “Continuous urban fabric”. In contrast, reductions were recorded in “Shrub/herbaceous vegetation” (- 1867.32 ha), “High dense forest” (-462.86 ha), “Transient crops” (-446.25 ha), and “Bare land” (-54.17 ha). Validation of the classification yielded an overall accuracy of 0.90 and a Kappa coefficient of 0.88, which supports the reliability of the results. These changes show significant transformations in the landscape, providing key information for territorial planning and the implementation of environmental conservation policies.