Spatial analysis of slum areas on the riverbanks of Palembang City using the Anselin Local Moran’s I analysis

Slums are closely associated with riverbanks. The same situation can be found in one of the urban regions in Indonesia, that is, Palembang City. This research was intended to spatially analyze slum areas along the edge of the river traversing the city using the Anselin Local Moran’s I statistic. In the quantitative method used, samples of slum houses were taken by area-proportional random sampling, and data on their locations were collected with a GPS and then spatially analyzed using the Anselin Local Moran’s I. Results showed families living on flood-prone riverbanks in two districts, non-flood-prone banks in 15 districts, and flood-prone areas outside the banks in one district. Most slum areas are located in one district, Kertapati. Slums with the highest number of houses, i.e., more than 1000, are spread across Jakabaring, Kertapati, Seberang Ulu II, Plaju, and Gandus, and the most occupied slums with more than 1000 families are found in seven districts: Seberang Ulu I, Ilir Timur I, Kertapati, Ilir Timur III, Seberang Ulu II, Plaju, and Gandus. The Anselin Local Moran’s I analysis found high-low (HL) outliers in six districts and low-low (LL) clusters of slums in five districts. The Local Moran’s I index values were positive for neighboring population densities and negative for farther-spaced population densities. In addition, this analysis produced z-scores and p-values that suggested the riverbank slums in Palembang formed a clustered and random pattern.