Changes in vegetation cover and increasing land surface temperature (LST) are critical environmental issues, particularly in tropical regions vulnerable to land-use change. This study analyzes changes in the Normalized Difference Vegetation Index (NDVI) and LST, and examines their spatio-temporal relationship in Lokpaikat Sub-district, Tapin Regency. The research uses Landsat 5 TM (2009) and Landsat 8 OLI/TIRS (2016 and 2024) imagery processed via Google Earth Engine. Image processing involves creating median composites from the two best dates each year to reduce cloud effects. NDVI is calculated using the NIR and RED bands, while LST estimation follows a multi-step process: conversion of thermal radiance (Lλ), calculation of Brightness Temperature (BT), vegetation fraction (Pv), and surface emissivity (ε). The final LST in Celsius is derived from BT and ε. Statistical analysis uses pixel-based simple linear regression with Pearson’s correlation coefficient, coefficient of determination (R²), and p-value for significance testing. The results show a decrease in dense vegetation area from 8,280.05 hectares (2009) to 8,080.52 hectares (2024), accompanied by a significant rise in average LST from 22.50°C to 25.21°C. Pearson correlation between NDVI and LST reveals a very strong and significant negative relationship with r-values of -0.9265 (2009), -0.9993 (2016), and -0.9989 (2024), all with p-values < 0.001. These findings indicate that deforestation strongly contributes to land surface temperature increase, emphasizing the importance of remote sensing in environmental monitoring and sustainable land-use planning.