Spatio-temporal Dynamics of Ecosystem Service Trade-offs/Synergies and Ecological Zoning in the Yellow River Floodplain over a 30-year Period

Understanding the dynamic trade-offs and synergies among ecosystem services (ESs) and their underlying drivers is essential for implementing differentiated ecological zoning and promoting regional sustainability. This study analyzed the spatio-temporal evolution and driving mechanisms of ES relationships in the Yellow River Floodplain from 1990 to 2020 at both 1 km and county scales. Correlation analysis and geographical detector method were employed to quantify ES interactions, and a self-organizing feature map (SOFM) was used to delineate cross-scale ecological zones based on the status and long-term trajectories of ecosystem service bundles (ESBs). The results show: (1) Carbon sequestration (CS) and soil conservation (SC) services maintained relatively stable spatial patterns over the study period, while food production (FP) and water yield (WY) services exhibited significant spatial variability. CS generally declined, FP increased sharply after 2000, and both WY and SC showed fluctuating upward trends. (2) Since 2000, most ES pairs shifted toward synergies, with trade-offs mainly involving SC. From 1990 to 2000, trade-offs between FP-CS and WY-CS transitioned to synergies across both scales, with synergistic regions accounting for 23% and 19% at 1 km scale, and 19% and 14% at county scale, respectively. Between 2000 and 2010, 13% of the regions experienced synergy enhancement for the WY-SC pair. (3) From 1990 to 2020, five of six ES pairs at 1 km scale and four of six at county scale showed a trend of synergy strengthening, except for FP-SC at 1 km scale and WY-CS and CS-SC at county scale. (4) Interaction detection revealed that two-factor interactions exerted stronger effects on ES patterns than single drivers, with natural factors interacting with urbanization-related factors (landscape composition, socioeconomic attributes, e.g.) exhibiting particularly strong explanatory power. Land-use composition had greater explanatory strength at the county scale. (5) Based on ESB patterns and trends, nine ecological functional zones were identified across scales. Tailored management priorities were proposed for each zone according to the dominant driving factors at the corresponding scale. This research provides a robust multi-scale zoning framework that supports the coordinated optimization of ESs and offers practical guidance for long-term spatial planning and differentiated ecological management in floodplain systems.