Assessing the ecological risk induced by PM_2.5 pollution in a fast developing urban agglomeration of southeastern...

High PM_2.5 concentration threats ecosystem functions but limited quantitative studies have recognized PM_2.5 pollution as an individual stressor in evaluating ecological risk. In this study, we applied a machine-learning-based simulation model incorporating full-coverage satellite-driven PM_2.5 dataset to estimate high-resolution ground PM_2.5 concentration for the Golden Triangle of Southern Fujian Province, China (GTSF) in 2030 under two Representative Concentration Pathways (RCPs). Based on the simulation output, the ecological risk's spatiotemporal change and the risk for different land cover types, which were caused by PM_2.5 pollution, were assessed. We found that the PM_2.5 levels and ecological risk in the GTSF under RCP 4.5 would be reduced while those under RCP 8.5 would continue to increase. The regions with the highest ecological risk under RCP 4.5 are the most urbanized and industrialized districts, while those with the highest ecological risk under RCP 8.5 are of the highest rate in urbanization and the greatest decrease in planetary potential layer height. For both base years and 2030 under two RCPs, the ecological risk on developed land is the highest, while that on the forest is the lowest. Our study can provide useful information for environmental policy risk assessment.