Spatiotemporal Analysis of Land Surface Temperature, Vegetation, and Water Index Correlations Using MODIS Data (2003-2024) (Case Study: Palangkaraya City, Indonesia)
Soni Darmawan, Rika Hernawati, Muhammad Faiza Abdurrahman, Lintan Bening Nurulhakim

Institut Teknologi Nasional Bandung

Abstract

Rapid urbanization and land cover transformation have led to increased land surface temperatures (LST), particularly in dense city areas, contributing to the Urban Heat Island (UHI) phenomenon. This study investigates the spatiotemporal patterns and correlations between Land Surface Temperature (LST), Normalized Difference Vegetation Index (NDVI), and Normalized Difference Water Index (NDWI) in Palangkaraya City from 2003-2024 using MODIS satellite data and Google Earth Engine. We used an 8-day composite MODIS dataset (MOD11A2 for LST, MOD13A2 for NDVI, and MYD09A1 for NDWI calculations) covering the period 2003 to 2024. Data preprocessing included cloud masking, data filtering, and gap filling using temporal interpolation. The dataset was then clustered into annual averages using Google Earth Engine over the administrative area of Palangkaraya City. Time series analysis was performed to detect temporal trends, while spatial analysis used distribution mapping and Pearson correlation analysis through scatter plots to quantify relationships between variables. Over the 21-year period, Palangkaraya City showed significant LST increase of 0.3 degrees Celcius, concurrent with declining vegetation cover (NDVI: -0.02) and water content (NDWI: -0.002). Correlation analysis revealed a moderate negative relationship between LST and NDVI with r = -0.54, and strong negative correlation between LST and NDWI with r = -0.87.

Keywords: Land Surface Temperature- NDVI- NDWI- MODIS- Urban Heat Island- Google Earth Engine- Palangkaraya

Topic: Topic B: Applications of Remote Sensing