Using dynamic Bayesian belief networks to infer the effects of climate change and human activities on changes in regional ecosystem services

Abstract

In the context of the Anthropocene, the unparalleled degradation of ecosystem services (ES) has been driven by the intricate nature of ecosystems and their responsive interplay with human actions. This underscores the paramount importance of forecasting and managing these services for regional sustainability. However, a comprehensive understanding of the correlated impacts and driving factors of the spatial patterns of ES remains elusive. In this study, we utilize the dynamic Bayesian belief network (DBN) to infer the probabilistic changes in ecosystem services (biodiversity conservation, carbon storage, soil retention, water regulation, net primary productivity, and crop production) under the influence of climate change and human activities in the Yangtze River Delta urban agglomeration. Through this investigation, we discern conspicuous discrepancies in the impacts of climate and human activities on various ecosystem services and their spatial differentiations in the Yangtze River Delta urban agglomeration. The dynamic Bayesian belief network model reveals that geographical factors and land use play pivotal roles in shaping the spatial patterns of ecosystem services. Concerning spatial differentiation, Water yield (WY) and Biodiversity maintenance (BM) exhibit heightened sensitivity to changes in precipitation, while Net primary productivity (NPP) is significantly influenced by variations in vegetation cover and temperature. Moreover, land use and land cover, as reflections of human activities, greatly affect Carbon storage (CF) through the expansion of construction land and the loss of forested areas. Soil retention (SR), on the other hand, is predominantly influenced by rainfall. Spatial variations in Crop production (CP) are found to be contingent on the extent of vegetation cover. In summary, our findings unveil the non-linear relationships among ecosystem services and their direct and indirect responses to climate change and human activities. These insights hold crucial implications for supporting land-use planning based on ecosystem services.

Keywords

• Ecosystem service management
• Dynamic Bayesian models
• Generalized additive models
• Human activities
• Climate change