H41L-1316 Spatial Identification and Dynamics of Groundwater-Dependent Ecosystems in the North China Plain: An Integrated Random Forest and Multi-Criteria Decision Analysis Approach

Groundwater-dependent ecosystems (GDEs) play a crucial role in maintaining ecological stability and biodiversity, particularly in arid and semi-arid regions. This study investigates terrestrial GDEs within the North China Plain (NCP), a region critically affected by groundwater depletion due to rapid urbanization and intensive agriculture. By integrating multi-source remote sensing datasets, Random Forest (RF) modeling, and GIS-based Multi-Criteria Decision Analysis (MCDA), we mapped the spatial distribution and assessed the dynamics of GDEs from 2004 to 2023. Predictor variables included vegetation indices (NDVI, FPAR), the evapotranspiration-to-precipitation ratio (ET/P), ET components (vegetation transpiration, soil evaporation, canopy interception), groundwater depth, land surface temperature, and topographic factors (elevation, slope). The RF model, trained on over 10,000 georeferenced points and optimized through hyperparameter tuning, achieved an overall accuracy of 0.87. The spatiotemporal analysis revealed notable changes in GDE distribution. Between 2004–2008, the average GDE area was 1,366 km². A moderate expansion occurred in 2009–2013, reaching 1,520 km², mainly in the northern and southeastern regions. This was followed by a decline to 1,081 km² during 2014–2018, likely driven by groundwater stress or land use changes. A significant recovery was observed in 2019–2023, with GDE area expanding to 1,983 km². Comparative analysis between 2004–2008 and 2019–2023 indicates that 57% of recent GDEs are newly emerged, 17% transitioned from non-GDE to GDE, and 26% remained stable. Overlap analysis showed that 79% of moderate-to high-probability GDE zones identified by the MCDA approach were consistently classified as GDEs by the RF model, highlighting the robustness of our framework. Groundwater shallowing and increasing vegetation activity have supported GDE expansion in recent years, likely driven by improved recharge and land use changes. These findings offer insights into the evolving distribution of GDEs and provide a decision-support framework for ecological monitoring and groundwater management in the NCP. Keywords: Groundwater-dependent ecosystems; Terrestrial GDE; North China Plain; Random Forest; Multi-Criteria Decision Analysis; GDE dynamics