A critical review of mathematical, statistical, and AI approaches to river pollution modelling for climate-resilient water management

This is the authors' abstract. We don't add key points for this paper.

River pollution has become a pressing global concern, affecting ecosystems, human health, and long-term water security. The challenge is further intensified by climate change, which alters rainfall patterns, raises temperatures, and increases the frequency of floods and droughts, all of which worsen water quality. This review critically examines the role of mathematical, statistical, and artificial intelligence (AI) models in understanding and predicting river pollution. Mathematical models describe the physical transport of pollutants, statistical tools provide insights into trends and forecasts, and AI methods uncover hidden patterns in complex datasets. Despite these advances, current approaches often overlook climate-driven factors, limiting their ability to address future risks. We highlight the need for integrated modelling frameworks that combine traditional methods with modern data-driven techniques to strengthen climate resilience. By identifying research gaps and proposing future directions, this study contributes to advancing sustainable water management in line with global priorities such as Clean Water and Sanitation (SDG 6) and Climate Action (SDG 13).