Assessment of Watershed Restoration Effectiveness from the Perspective of Water-Related Ecosystem Service Trade-Offs

<sec><title>Objective</title> Mountains, forests, grasslands, and other landscape elements are all intricately connected by hydrological processes in watersheds, which are essential ecological communities. Theoretically, watersheds are the best geographical scale for effectiveness of ecological restoration since they are whole ecological units with cohesive biological processes. As ecosystem services having the most direct impact on human civilizations and serving as the primary determinants of the effectiveness of watershed restoration, water-related ecosystem services (WES) are vital connections between ecological restoration processes and human well-being. Additionally, one crucial metric for assessing the relationship between ecosystem services is the trade-offs between WES. Not merely the main source of WES, the ecological spatial pattern of watersheds is also the physical expression of coupled natural-anthropogenic processes, making it an essential analytical viewpoint for restoration ecology. In addition to improving effectiveness assessment, examining watershed restoration from the perspective of WES trade-offs may also help guide strategic approaches to integrated watershed management. </sec><sec><title>Methods</title> An integrated methodological approach is used in this research to assess the effectiveness of ecological restoration in the mainstream watershed of the Liaohe River from 2000 to 2020. To thoroughly assess restoration results, the research employs a multi-model approach that includes geospatial analysis, landscape ecology measures, and ecosystem service modeling. In order to measure changes in the composition and layout of ecological spaces, satellite imagery processed in ArcGIS 10.8 is used to create land use transition matrices and landscape pattern indices. Four important WES are assessed using the InVEST 3.14.1 model, namely water purification (nutrient delivery ratio module), water conservation (annual water yield module), soil conservation (sediment delivery ratio module), and habitat quality (habitat quality module). Root mean square deviation (RMSD) is used to calculated the trade-off intensity between various services, and Origin 2024 is used for data standardization and statistical analysis. Additionally, the research adopts multiscale geographically weighted regression (MGWR 2.2 software) to distinguish between natural elements (driven by climate) and anthropogenic elements (driven by land use) affecting WES trade-offs in order to pinpoint the driving processes. Through spatial explicit modeling, this analytical methodology makes it possible to diagnose the root causes of restoration effectiveness and quantify them across several dimensions (spatial pattern, individual service, and ecosystem service trade-off). To guarantee region-specific accuracy, local hydrological and ecological data are used to calibrate all model parameters. </sec><sec><title>Results</title> Research results are summarized as follows (covering the period from 2000 to 2020). 1) Landscape transformation: The conversion of agricultural production space (2,765.45 km²) creates 1,873.06 km² of new ecological space (including 292.67 km² of forests, 980.10 km² of grasslands, 382.96 km² of wetlands, and 217.33 km² of water bodies), and produces unique spatial patterns, such as aggregated growth down the mainstream (AI increases by over 17%), and dispersed expansion in upper tributaries (PD and LSI increase by 11.77% and 2.64%, respectively). 2) Despite regional variation, all the four WES display quantifiable improvements: There is a 9.14% improvement (with nitrogen output decreasing from 1.74×10⁷ kg to 1.58×10⁷ kg) in water purification (WP), mostly along the mainstream of the Liaohe River and upper reaches of its tributaries; a remarkable 184% increase (from 9.81×10⁷ m³ to 27.86×10⁷ m³) in water conservation (WC); a significant gain of 85.73×10⁶ tons in soil conservation (SC), representing a 74.7% improvement from the baseline in 2000; and a modest but ecologically significant progress in habitat quality (HQ), with the watershed-wide mean index increasing from 0.315 to 0.321 (a 1.9% increase). 3) Two of the six trade-off connections under investigation indicate a decline in trade-off intensity (WC-WP: RMSD decreases by 0.0339; WC-HQ: RMSD decreases by 0.0035), while the other four show the reverse pattern (WP-SC: RMSD increases by 0.0219; WP-HQ: RMSD increases by 0.0192; WC-SC: RMSD increases by 0.0515; SC-HQ: RMSD increases by 0.0039). 4) In particular, the landscape composition is advantageous for WP, SC, and HQ but detrimental for WC, the landscape fragmentation is advantageous for WP but detrimental for SC, while the landscape aggregation is opposite. These ecological spatial patterns have opposite effects on WES, which is the primary cause of the increase in WES trade-offs. 5) In addition, the ecological spatial layout plan previously centered on water purification is a significant factor in the rise in WES trade-offs. </sec><sec><title>Conclusion</title> From the perspective of WES, this research has verified that ecological restoration in the mainstream watershed of the Liaohe River from 2000 to 2020 is a traditional single-objective ecological restoration mode, which is beneficial for single-objective local restoration, but detrimental for multi-objective coordinated restoration. Optimizing the ecological spatial pattern is a crucial tactic to raise the overall effectiveness of ecological restoration of territorial space and human well-being in watersheds. In the future, integrating the trade-off intensity of WES into the effectiveness assessment system will support the multi-objective coordinated development of ecological restoration in watersheds. This research provides factual support for the shift from single-objective to multi-functional watershed restoration strategies, as well as a replicable assessment framework. There are new avenues for operationalizing “ecological civilization” principles in real-world watershed management through the scientific fusion of landscape ecology, ecosystem service research, and spatial statistics. </sec>