Spatiotemporal Assessment of Water Quality around Western Australian Central Business District Using Satellite Multispectral Imaging

Authors:
Preprint
Publisher website
Google Scholar
Abstract
Monitoring water quality in urban environments is increasingly vital due to growing pressures from urbanisation, climate change, and stormwater runoff. Multispectral satellite remote sensing offers a promising, scalable approach for assessing water quality across large spatial and temporal scales compared to conventional in-field sampling and laboratory analysis. This research assesses the capability of Sentinel-2 satellite multispectral imagery to track the water quality index (WQI) of the Swan-Canning River system and nearby coastal areas around the central business district (CBD) of Western Australia (WA) from 2018 to 2025. Satellite-derived optically active parameters such as chlorophyll-a, turbidity, and absorption of coloured dissolved organic matter were used to compute the trophic state index, the turbidity index, and the absorption by coloured dissolved organic matter index, subsequently generating a WQI raster output using a fuzzy inference system (FIS). Spatiotemporal analysis revealed that poor water quality (WQI 40-60) dominated the study area, particularly in stagnant urban lakes, while localised zones of good quality persisted near estuarine and coastal regions. Seasonal trends showed that rainfall and temperature fluctuations were key drivers of short-term deterioration, while dissolved oxygen concentrations were inversely correlated with poor WQI scores. Despite limitations in data availability due to cloud cover, the study demonstrates that satellite-based multispectral imaging combined with FIS modelling offers a robust, scalable framework for urban freshwater quality monitoring. The findings emphasise the urgent need for targeted catchment management and real-time adaptive strategies in rapidly urbanising regions.

This publication has 26 references indexed in Scilit: