Assessment of Urban Mapping Index Accuracy in Relation to Physical Land Characteristics in Humid Tropical Areas

Settlements and built-up areas can lead to the degradation of ecological systems. Good quality and efficient regional planning is therefore needed for urban areas. Spatial data and satellite imagery can be used for mapping and monitoring urban growth. Unfortunately, mapping urban areas can sometimes be difficult due to local variations, and different algorithms can provide varying results. Urban indices often rely on remote sensing reflectance, the accuracy of which can be influenced by land characteristics. No studies have examined the impact of land characteristics on the accuracy of remote sensing urban indices in the humid tropics. The purpose of this study was to compare urban and built area indices, namely EBBI, NDBI, UI, and IBI, in two climatically and topographically different cities. This study also examined the stability and relationship between these indices with environmental factors such as slope, elevation, and temperature. The results showed that EBBI was the index with the highest accuracy in both study areas: 85% for Batu City and 89.17% for Pasuruan City. Also, EBBI was the most stable index for the temporal studies. Environmental factors, especially slope and elevation, had a strong relationship with the index value applied. Therefore, these findings need to be considered in applying the index in areas that have topographical variations. Keywords: built-up land, landsat, EBBI, NDBI, UI, IBI, topography