URBAN ENERGY MODELING USING REMOTE SENSING APPROACHES

Abstract

This research aims to discover the urban energy spatial pattern based on the analysis of urban heat
island (UHI) and the land surface temperature (LST) in Yogyakarta city. The vegetation cover was analyzed using
the vegetation index with the normalized difference vegetation index (NDVI) formula. The satellite imagery used
Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS), recorded on April 22, 2019, for
the dry season images, and June 25, 2019, for the wet season images. The processing results showed that the
maximum LST in the dry season was 34.09°C, and the minimum was 27.22°C; meanwhile, in the wet season, the
maximum LST was 31.27°C, and the minimum was 23.67°C. The presence of vegetation indicated by NDVI
values had a strong correlation with LST, i.e. R = 0.74 for the dry season and R = 0.60 for the wet season. The
high NDVI had a low temperature, whereas the low NDVI had a high temperature. Based on the analysis results,
UHI increases in temperature (7°C) from the temperature threshold. The difference in the spatial distribution of
LST and UHI in the dry and wet seasons is strongly influenced by the image recording time and the atmospheric
conditions. The percentage of non-UHI area is deficient at around 0.35% in the dry season and 0.59% in the wet
season, indicating that the Yogyakarta city's energy needs are very high. Therefore, it requires urban planning that
considers urban green space area.