Enhancing water quality monitoring in a coastal shallow lake using GIS and multi-criteria decision analysis

Water quality decline is a critical global issue affecting both aquatic ecosystems and human health. This study focuses on Vrana Lake, a coastal shallow lake in Dalmatia, Croatia, which is subject to seasonal water level fluctuations, salinization from the Adriatic Sea, and various anthropogenic and environmental pressures. To evaluate and improve water quality monitoring, a spatial assessment was conducted using a Geographic Information System-based Multi-Criteria Decision Analysis (GIS-MCDA) integrated with the Fuzzy Analytical Hierarchy Process (F-AHP). The analysis incorporated multiple criteria including electrical conductivity, turbidity, dissolved oxygen, water temperature, distance from water resources, distance from land nutrient runoff, distance from environmental pollutants, wind, precipitation, and air temperature, using data collected from July 2023 to June 2024. The methodology included standardization of criteria in GIS software, weighting through F-AHP, and a sensitivity analysis using 1000 Monte Carlo simulations to assess model robustness. Results revealed spatial variations in water quality, with lower quality observed in the northwestern and southern regions, influenced by anthropogenic activities and seawater intrusion. An optimized monitoring network of seven stations was developed based on Jenks natural breaks classification, ensuring comprehensive lake coverage and efficient representation of all water quality classes. Three existing stations and two from previous networks were incorporated into the new network, with two additional stations proposed to maintain consistent spatial distribution. This study provides an adaptable decision support framework for improving water quality monitoring and management in coastal shallow lakes, offering a strategic approach to early eutrophication warning systems and informed resource planning. • GIS-MCDA integrated with F-AHP maps water quality for Vrana Lake in Croatia. • Monte Carlo sensitivity analysis confirms model stability under uncertainty. • Final map shows overall good water quality; hotspots are tied to human impact. • Optimized monitoring network includes seven strategically placed stations.