Climate change effects on animal presence in the Massaciuccoli Lake basin

Abstract

Big-data mining approaches based on Artificial Intelligence models can help forecast biodiversity changes before they happen. These approaches can predict macroscopic species distribution patterns and trends that can inform preventive measures to avoid the loss of ecosystem functions and services. They can, therefore, help study and mitigate climate change implications on biodiversity conservation in fragile ecosystems. Wetlands are particularly fragile ecosystems where climate change poses severe risks and has dramatically reduced their size over the past century, with profound consequences on biodiversity and ecosystem services. Through big-data mining approaches, we can predict future wetland biodiversity trends in the context of climate change. This paper proposes such predictive analysis for a specific wetland: The Massaciuccoli Lake basin in Tuscany, Italy. This basin is a critical tourist attraction due to its rich biodiversity, making it an area of interest for citizens, tourists, and scientists. However, the region's suitability for native and non-native species is at risk due to climate and land-use change. Using machine-learning models, we predict the potential effects of climate change on animal spatial distribution in the basin under different greenhouse gas emission scenarios. The results suggest that habitat suitability has generally improved from 1950 to today, presumably owing to the targeted conservation strategies adopted in the area, but climate change will severely reduce bird biodiversity by 2050 while favouring several insect species' proliferation and other species' habitat change, even under a medium-emission scenario. This will lead to significant changes in the basin's biodiversity. Our methodology is adaptable to other wetland basins, being fully based on open data and models. The spatially explicit modelling used in this research provides valuable information for policymakers and spatial planners, complementing traditional biodiversity trend analyses.