ANALYZING THE RELATIONSHIP BETWEEN LAND COVER AND LAND SURFACE TEMPERATURE (LST) DISTRIBUTION IN THE MOUNT PAPANDAYAN AREA, WEST JAVA

摘要

Land surface temperature (LST) is an essential indicator for detecting climate change due to its role in providing information regarding the energy balance process on the earth's surface. In understanding the climate process of an area, it is important to study the influence of land cover on LST. However, over mountainous areas, such studies are limited mostly because of the strong effect of topography on LST. In this study, the relationship between land cover and LST distribution in the Mount Papandayan area, West Java, Indonesia was analyzed using Landsat 8 OLI/TIRS imagery. To estimate LST, single-channel algorithm and land surface emissivity (LSE) correction through NDV1-based method (NBEM) were used. The effect of topography on LST was then normalized using a multi-linear regression model from digital elevation model (DEM) and solar irradiance. Land cover classification was done using object-based image analysis (OBI A) through multi-resolution segmentation and threshold classification combined with nearest neighbor (NN) classification. The results reveal that the effect of topography on LST was reduced after correction. LST in the Mount Papandayan area ranged from 14.27 to 40.01°C with a mean value of 23.77°C The distribution of mean LST across land cover types was found to be the highest in crater area (29.59°C), followed by built-up (26.91°C), cropland (25.64°C), tea plantation (24.16°C), pasture (23.56°C), shrub (22.40°C), and forest area (20.63°C). The difference between the mean LST of each land cover and the mean LST of the whole study area indicates the heating or the cooling effect of each land cover type on the area. By using the mean LST of the study area as the reference LST, it can be inferred that crater, built-up, cropland, and tea plantation land cover types contribute to heating the area, whereas pasture, shrub, and forest area contribute to cooling the area. From this study, it can be concluded that in the Mount Papandayan area, the distribution of LST across land cover types was the highest in crater, followed by built-up, cropland, tea plantation, pasture, shrub, and forest.