Modelling spatiotemporal patterns in surface water quality in northwestern Argentina: the importance of potential pollution sources and climatic conditions.

Good surface water quality is essential for environmental health and human well-being. Over the last decades, developing economies have experienced worsening water quality conditions, impairing their sustainable development. This is the case for South America in general and for Argentina in particular, where water stress poses a significant risk to socio-environmental integrity. There, the scarcity of data to monitor water quality and assess the potential effect of pollution sources undermines effective water management. Global water quality models can help overcome these challenges, yet need to be validated to ensure they accurately represent local conditions. Here, we combined a global water quality model (DynQual 1.0) with sparse local data using Approximate Bayesian Computation to investigate broad spatial and temporal trends in organic water pollution, measured through the mean annual biochemical oxygen demand (BOD), in the province of Tucumán (northwestern Argentina). Our model revealed that the two most populous departments and the six departments harbouring most of the industries had the highest mean annual BOD. The density of industries and the mean annual temperature correlated positively with the mean annual BOD across decades (1980 to 2019–2022). Our findings suggest that industrial effluents in conjunction with rapid oxygen depletion due to warm ambient conditions result in poor water quality. Faced with a warming climate, the rigorous control of pollution sources is essential for ameliorating water quality in Tucumán. More broadly, our modelling approach demonstrates the value of integrating global models and local data to foster effective water management in data-poor regions.