Spatiotemporal Dynamics and Multi-Factor Drivers of Land Surface Temperature in Andhra Pradesh, India (2020-2024): Integrating MODIS-Derived Indices, Urban Heat Islands, Hotspot Analysis and Trend Detection
M.R Goutham *
Department of Geology, Government College (Autonomous) Rajahmundry, AP, India.
Suneel Kumar Duvvuri
Department of Computer Science, Government College (Autonomous) Rajahmundry, AP, India.
Umamahesh Goudu
Department of Geology, Government College (Autonomous) Rajahmundry, AP, India.
*Author to whom correspondence should be addressed.
Abstract
Land Surface Temperature (LST) is recognized as a primary indicator of regional thermal shifts precipitated by rapid urban expansion and land-use modifications. In this research, a comprehensive five-year longitudinal assessment (2020-2024) of LST dynamics across Andhra Pradesh, India, was conducted. The methodology involved the integration of MODIS-derived products with topographic and land-cover datasets, processed via the Google Earth Engine platform. The temporal analysis indicated that the annual mean LST underwent a distinct reduction in 2022, falling by 0.82 °C compared to 2021 levels. Following this decline, a recovery phase was observed, culminating in a marginally significant monotonic increasing trend, as evidenced by the Mann-Kendall test results (S = +10, Z = 2.205, p = 0.051). Statistical evaluation through Pearson correlation for the year 2022 identified the Normalized Difference Built-up Index (NDBI) as the predominant positive driver of temperature (r = +0.794). Conversely, mitigating influences were attributed to the Normalized Difference Vegetation Index (NDVI, r = -0.654) and the Normalized Difference Water Index (NDWI, r = -0.541). Topographic factors also played a role in temperature regulation, with slope (r = -0.504) and elevation (r = -0.201) showing negative correlations with LST. Regarding localised climate phenomena, Urban Heat Island (UHI) intensities were found to be modest yet persistent, ranging between 0.006 °C and 0.010 °C. The application of Getis-Ord Gi* hotspot analysis further validated the existence of statistically significant thermal clustering within primary urban centers. Throughout the study period, built-up land covers consistently recorded the highest thermal values. It was concluded that urban growth exerts a dominant influence on regional warming, often offsetting the cooling contributions of vegetation and terrain. These findings offer a critical empirical basis for regional heat-mitigation and urban planning within the context of rapidly developing Indian states.
Keywords: Land surface temperature, MODIS, urban heat island, NDBI, Mann-Kendall trend, Getis-Ord Gi*, Andhra Pradesh, Google earth engine