Exploring the nexus between water quality and land use/land cover change in an urban watershed in Uruguay: a machine learning approach.

The expansion of urban areas contributes to the growth of impervious surfaces, leading to increased pollution and altering the configuration, composition, and context of land covers. This study employed machine learning methods (partial least square regressor and the Shapley Additive exPlanations) to explore the intricate relationships between urban expansion, land cover changes, and water quality in a watershed with a park and lake. To address this, we first evaluated the spatio-temporal variation of some physicochemical and microbiological water quality variables, generated yearly land cover maps of the basin adopting several machine learning classifiers, and computed the most suitable landscape metrics that better represent the land cover. The main results highlighted the importance of spatial arrangement and the size of the contributing watershed on water quality. Compact urban forms appeared to mitigate the impact on pollutants. This research provides valuable insights into the intricate relationship between landscape characteristics and water quality dynamics, informing targeted watershed management strategies aimed at mitigating pollution and ensuring the health and resilience of aquatic ecosystems.

Enhancing environmental data imputation: A physically-constrained machine learning framework.

In water resources management, new computational capabilities have made it possible to develop integrated models to jointly analyze climatic conditions and water quantity/quality of the entire watershed system. Although the value of this integrated approach has been demonstrated so far, the limited availability of field data may hinder its applicability by causing high uncertainty in the model response. In this context, before collecting additional data, it is recommended first to recognize what improvement in model performance would occur if all available records could be well exploited. This work proposes a novel machine learning framework with physical constraints capable of successfully imputing a high percentage of missing data belonging to several environmental domains (meteorology, water quantity, water quality), yielding satisfactory results. In particular, the minimum NSE computed for meteorologic variables is 0.72. For hydrometric variables, NSE is always >0.97. More than 78 % of the physical-water-quality variables is characterized by NSE > 0.45, and >66 % of the chemical-water quality variables reaches NSE > 0.35. This work's results demonstrate the proposed framework's effectiveness as a data augmentation tool to improve the performance of integrated environmental modeling.