A Spatial Model of Land Surface Temperature Based on the Integration of Satellite Data and IoT Sensors for Urban Heat Island Mitigation in the Climate Village Program Banjarmasin

The study investigates the linear regression relationship between urban IoT sensor temperatures and satellite-derived land surface temperatures (LST) to predict the effects of the Urban Heat Island (UHI) phenomenon. Satellite data from Landsat 9, combined with real-time IoT sensor data, facilitates the analysis of urban temperature variations across Banjarmasin, South Kalimantan. The regression analysis, using a model based on IoT and satellite temperature data, computes the regression coefficient and intercept, providing an equation for temperature prediction. The coefficient of determination (R^²) is employed to evaluate the model's explanatory power, revealing that 95.17% of the temperature variation can be explained by satellite data. The high R^² value indicates a strong correlation between urban and rural temperatures. Furthermore, the Root Mean Square Error (RMSE) was calculated to quantify the prediction accuracy, yielding an RMSE of 0.714℃, which suggests the model's high reliability. The findings demonstrate that satellite data can significantly enhance UHI mitigation strategies by informing decisions on green infrastructure implementation. The integration of IoT and satellite data offers a scalable solution for urban planners to better address the UHI effect and improve urban resilience to climate change. The study underscores the importance of combining technological tools for more accurate and efficient climate action planning in urban areas.