THE LAND SURFACE TEMPERATURE IMPACT TO LAND COVER TYPES

Land cover type is an important signature that is usually used to understand the interaction between the ground surfaces with the local temperature. Various land cover types such as high density built up areas, vegetation, bare land and water bodies are areas where heat signature are measured using remote sensing image. The aim of this study is to analyse the impact of land surface temperature on land cover types. The objectives are 1) to analyse the mean temperature for each land cover types and 2) to analyse the relationship of temperature variation within land cover types: built up area, green area, forest, water bodies and bare land. The method used in this research was supervised classification for land cover map and mono window algorithm for land surface temperature (LST) extraction. The statistical analysis of post hoc Tukey test was used on an image captured on five available images. A pixel-based change detection was applied to the temperature and land cover images. The result of post hoc Tukey test for the images showed that these land cover types: built up-green, built up-forest, built up-water bodies have caused significant difference in the temperature variation. However, built up-bare land did not show significant impact at p<0.05. These findings show that green areas appears to have a lower temperature difference, which is between 2° to 3° Celsius compared to urban areas. The findings also show that the average temperature and the built up percentage has a moderate correlation with R2 = 0.53. The environmental implications of these interactions can provide some insights for future land use planning in the region.