Abstract

Heavy metal contamination in fish from urban aquatic ecosystems poses significant ecological and public health risks. Bioaccumulation of toxic metals such as mercury, lead, and cadmium threatens aquatic life and human consumers. Existing assessment methods often rely on limited sampling and outdated detection techniques, which fail to provide real-time, spatially representative data. To address these limitations, this study proposes an integrated framework combining geospatial analysis, advanced biosensors, and machine learning algorithms for dynamic monitoring of heavy metal levels in fish tissues. The proposed method enables continuous, location-specific tracking of contaminants and enhances prediction accuracy through adaptive learning models. Utilizing this approach, we assessed fish samples from three urban water bodies and observed improved detection sensitivity and spatial mapping of contamination hotspots. The findings indicate that the proposed system can effectively identify areas of concern, guide environmental management, and protect food safety.

Keywords: Heavy metals, Urban aquatic ecosystems, Fish contamination, Biosensors, machine learning, Geospatial analysis