Analyzing the mahakam river water quality using the geographically weighted panel regression model

Area of Science:

• Environmental Science
• Spatial Statistics
• Water Quality Management

Background:

• Mahakam River faces challenges with biochemical oxygen demand (BOD).
• Understanding spatial and temporal variations in water quality is crucial.
• Existing models may not fully capture spatial heterogeneity in riverine ecosystems.

Purpose of the Study:

• To apply the Geographically Weighted Panel Regression (GWPR) model to Mahakam River water quality data.
• To identify and map key factors influencing biochemical oxygen demand (BOD) across different locations and times.
• To compare the performance of GWPR against a global Fixed Effects Model (FEM).

Main Methods:

• Utilized panel data on Mahakam River water BOD from 2022-2024.
• Employed the GWPR model with FEM as the global model, incorporating a demeaning transformation for temporal effects.
• Performed spatial analysis and statistical modeling using R, GNU Octave, QGIS, and Google Earth.

Main Results:

• GWPR model demonstrated superior performance over FEM, evidenced by a lower AIC and higher R-squared (80.321%).
• Key factors influencing BOD were identified as temperature, water pH, color degree, nitrate, ammonia, total suspended solids, and sulfate.
• The study successfully mapped the spatial distribution of these influencing factors.

Conclusions:

• The GWPR model is a powerful tool for analyzing spatially heterogeneous panel data in environmental studies.
• Accurate identification of BOD influencing factors enables targeted water quality management strategies for the Mahakam River.
• Local analysis provided a more nuanced understanding of water quality dynamics compared to global models.