Exploring the interplay between land use and surface water quality across Italy's watersheds

The sustainability of freshwater availability and quality is seriously threatened by climate change (CC) and land-use/land-cover change (LULC). On the other hand, extreme weather and climate-related events depend strongly on LULC. Multiple evidence suggests that rapid global development is the main driving factor altering all the fundamental processes that control the hydrologic cycle and temporal and spatial variations of river basins. Moreover, the interaction between the upstream and downstream of a basin significantly impacts the overall status of the basin. Understanding the “source-to-sink” effect is crucial for managing water quality in river basins. This study aims to understand the impacts of LULC on water quality at the river basin scale in Italy, providing a baseline model for predicting the probability of achieving good ecological status for each watershed under different Shared Socioeconomic Pathways (i.e., SSP2 and SSP5) and Representative Concentration Pathways (i.e., RCP4.5 and RCP8.5) for mid- and long-term timeframe (i.e., 2050 and 2100). To fulfill this bold objective, this study integrates Principal Component Analysis (PCA) and several regression models to explore the influence of various landscape metrics on the ecological status of each watershed, taking into account the effect of changes in land use from upstream watersheds to downstream ones. The outcomes reveal that conserving natural areas is essential for improving water quality across the territory. However, conservation efforts alone are insufficient without restoring places that were natural previously but are now used for agriculture and urban development. Implementing agricultural practices that promote harmony and links between natural regions and farmed areas may effectively reduce the harm from unsustainable farming practices. Future work will focus on integrating climate change variables and spatio-temporal occurrence of extreme events, such as flood and drought hotspots. Moreover, advanced probabilistic models (e.g., Bayesian Network) and Machine Learning will be employed to assess the possible interactions between LULC and CC and their impacts on water quality. The outcomes of this analysis contribute to developing adaptive strategies that safeguard water resources and ensure the long-term sustainability of freshwater ecosystems.