Identifying ecosystem service bundles and the spatiotemporal characteristics of trade-offs and synergies in coal mining areas with a high groundwater table

Understanding the spatiotemporal characteristics of the interactions among ecosystem services (ESs) is a crucial but challenging task for maintaining human well-being and achieving sustainable regional development. However, understanding the spatiotemporal interactions of multiple ESs at different grid scales and within different ecosystem services bundles (ESBs) is relatively limited, particularly in coal mining areas with a high groundwater table (CMA-HGT) where the land use has drastically changed as a result of mining subsidence. This study examines CMA-HGT in Huainan, aiming to identify ESBs and explore the spatiotemporal characteristics of trade-offs/synergies among ESs at distinct grid scales and ESBs. Five ESs relating to provisioning, regulation, and maintenance, including food production (FP), water yield (WY), soil conservation (SC), carbon sequestration (CS), and biodiversity maintenance (BM) were quantified using different biological models during the period 1987–2018. Spatiotemporal trade-offs/synergies among ESs were explored using correlation analysis and significance tests at different scales. The spatiotemporal distributions and main characteristics of distinct ESBs were identified using a self-organizing map (SOM) and Calinski criterion. The interactions among ESs in different ESBs were detected. Relationships between ESs and land use or coal production (CP) were explored using redundancy analysis (RDA). The results demonstrate that spatiotemporal trade-offs were generally observed among provisioning services at distinct grid scales and within different ESBs. Meanwhile, spatiotemporal synergies generally appeared between regulation and maintenance services at distinct grid scales. Interactions among ESs presented temporal dynamic, spatial heterogeneity and scales dependence due to the relationships of FP–BM or SC–CS had changed with the increasing of research scales. Three ESBs—ESB1, ESB2, and ESB3—were identified at a grid of scale of 1000 m, and their spatial locations varied across different periods, but the areas of variation covered less than 24% of the study area. BM was synergistic with FP, WY, SC, and CS; while WY had only a trade-off relationship with FP in ESB1. WY had trade-off relationships with FP, SC, CS, and BM in ESB2. In ESB3, BM was synergistic with FP, SC, and CS; while it was in a trade-off relationship with WY. Cultivated land, construction land and CP were the main driving factors in the WSA, ESB1, ESB2 and ESB3. There was a certain degree of change in the relationships between ESs and land use/CP, and the relationships among ESs at different grid scales and ESBs over time and space, which indicates strong regional heterogeneity and scale dependence. These results can provide detailed guidelines for formulating spatially targeted ecosystem management, restoration programs and ES payment policies.