<![CDATA[Rivers in East Kalimantan serve as the primary raw water source for most of the population, making water quality monitoring an urgent priority. This study aims to model Dissolved Oxygen (DO) as a river water quality indicator in East Kalimantan, and to identify the factors influencing DO levels through the Mixed Geographically Weighted Regression (MGWR) approach. MGWR is a useful tool for modeling regression relationships where the impact of some explanatory variables on the response variable is global, while the influence of others varies spatially. This study uses secondary data, namely the 2025 surface water quality monitoring analysis report sourced from the Environmental Agency of East Kalimantan. The sample size consists of 27 observation points of Class I river water quality in East Kalimantan in 2025. The spatial weighting is computed using an adaptive Bi-Squares kernel function, and the optimal bandwidth is selected using Generalized Cross-Validation (GCV) criterion . The results showed that the MGWR model outperformed both the linier regression and Geogrpahically Weighted Regression (GWR) models, yielding a GCV of 0,003, R² of 0,940, an AIC of −75,762, and RMSE of 0,175. BOD, sulfate concentration, and TDS were found to globally influence DO levels, while pH exhibited a spatially varying local influence across observation sites. The results of this study can serve as a basis for policy-making in river water quality monitoring in East Kalimantan. These findings are expected to assist the government in formulating more effective and targeted policies for river water quality management in East Kalimantan.]]>
Copyrights © 2026
