Geographical Information System for Mapping Flood-Prone Areas in Manado City Using the K-Means Clustering Method

Abstract

Floods are natural events or phenomena that can cause environmental damage, loss of property, psychological effects, and death or casualties. One way to control flooding non-structurally is by mapping areas that are prone to flooding. This study builds a geographic-based information system to map flood-prone areas in Manado City using the K-Means Clustering algorithm. The main objective of this research is to identify and map areas with a high risk of flooding using spatial data. Slope, land cover type, soil type, water discharge (discharge), and rainfall are independent variables that will be used and processed using the K-Means Clustering algorithm. There are four clusters in the mapping results of flood-prone areas, namely: high vulnerability, medium vulnerability, low vulnerability, and not vulnerable. By using the K-Means method, the results obtained are Paal Dua and Wenang sub-districts are high-vulnerability groups, followed by Mapanget, Tuminting, and Singkil subdistricts with medium vulnerability groups. Tikala District is the only area with low vulnerability. Meanwhile, Bunaken, Sario, Wanea, and Malayayang sub-districts are areas that are not potentially prone to flooding.