Analyzing the implications of forest cover change on land surface temperature variations in protected areas; a case of Mpanga Central Forest Reserve, Uganda.

Abstract

In Uganda, tropical forests are critical to maintaining ecological stability, serving as carbon sinks, and regulating climate systems. Deforestation and land use changes have, however, caused significant forest degradation, intensifying local and global climate variability. This study aimed to; (1) quantify spatiotemporal changes in forest cover, (2) assess the variation in NDVI and LST over time, and (3) explore the impact of forest cover change using NDVI as a proxy on LST variations between 1993 and 2023. The study utilized satellite imagery from Landsat (5-TM, 7-ETM+, 8-OLI/TIRS) spanning 1993 to 2023 processed through Google Earth Engine and ArcGIS 10.8. All images were processed at 30m spatial and 16-day temporal resolutions, with cloud cover thresholds of 7%–9% to ensure high-quality data for temporal analysis. The results indicate significant land cover changes over the past three decades, with settlement expanding from 0.8% to 4.2%, while shrubland and natural forests declined (-12.7% and -10.5% respectively). Cropland expanded by 15.3%, while grassland and wetland exhibited minor fluctuations (3.0% and 0.5%, respectively). A Correlation analysis revealed negative relationships between forest cover and settlement (r = -0.623), cropland (r = -0.787) and wetland (r = -0.907), explaining forest loss due to urbanization and land-use conversion. Long term NDVI trends showed a gradual decline in mean values over 30 years, indicating vegetation degradation, while annual seasonal NDVI peaked between May and July (2023 showed the highest NDVI peak 0.80 in July) and 1993 had lower NDVI values between January and May. LST records revealed a warming trend, with January Mean Annual temperatures rising from 24.60°C (1993) to 26.73°C (2023) and October showing the most substantial increase (22.11°C to 28.25°C). However, the Pearson correlation analysis between NDVI proxy and LST (r = -0.090, p = 0.635) indicated a very weak relationship. Despite, enhanced forest protection combined with integrated monitoring of structural and ecological indicators (canopy, biomass, soil moisture) is recommended to effectively assess and manage local thermal dynamics. The findings call for integrated land management strategies to support Uganda’s climate action efforts under SDGs 13 and 15, and Vision 2040.