Integrating remote sensing, GIS, and statistical analysis for assessing ecotourism sustainability

Ecotourism ecosystems in protected dryland landscapes are increasingly affected by climate variability and land-use dynamics; however, long-term spatially explicit evidence integrating vegetation greenness, land surface temperature, rainfall variability, and land-use change remains limited, particularly in data-scarce protected areas such as Mago National Park, Ethiopia. This gap restricts comprehensive understanding of how multi-decadal environmental changes influence ecological suitability for ecotourism development. This study evaluates the environmental sustainability of ecotourism in Mago National Park, Ethiopia, using an integrated remote sensing and GIS approach implemented in Google Earth Engine. MODIS NDVI, CHIRPS rainfall, and Landsat surface reflectance data were used to examine vegetation dynamics, rainfall variability, land-use/land-cover (LULC) change, and land surface temperature (LST) over the period 2000–2025. The results reveal a pronounced greening trend and substantial ecological improvement across the park. Dense vegetation increased from 171.23 km 2 (16.82%) in 2000 to 753.15 km 2 (73.97%) in 2025, whereas sparse/bare land declined from 198.94 km2 (19.54%) to 5.21 km 2 (0.51%). The most dominant land-cover transition was from moderate vegetation to dense vegetation, accounting for 518.87 km 2 (50.96%), indicating widespread vegetation recovery and intensification. Mean annual rainfall increased from 759.86 mm to 1378.99 mm, while representative land surface temperature declined from 42.06 °C to 39.69 °C, reflecting improved hydro-ecological and thermal conditions. Statistical analysis showed a strong negative relationship between NDVI and LST, whereas the NDVI-rainfall relationship remained weak at the pixel level. The Ecotourism Sustainability Index also demonstrated marked improvement, with the mean ESI increasing from 0.474 in 2000 to 0.771 in 2025, and the proportion of land in high- to very-high sustainability classes rising from 38.74% to 74.56%. Overall, the findings indicate a substantial increase in satellite-derived vegetation greenness and improved biophysical suitability for ecotourism; however, these results should be interpreted as remote-sensing-based ecological indicators rather than direct field-confirmed evidence of full ecosystem recovery or multidimensional ecotourism sustainability.