From ecological health indicators to priority zoning: A threshold-guided assessment framework for Northeast China
- 1University of Chinese Academy of Sciences, Beijing 100049, China; Key Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China; Geology Department, Faculty of Science, South Valley University, Qena 83523, Egypt
- 2Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China; University of Chinese Academy of Sciences, Beijing 100049, China
Abstract
Regional ecosystem health assessment requires composite indicators that capture spatial heterogeneity across land-use and climate gradients, yet many frameworks still rely on fixed-threshold classifications and linear driver analyses that may overlook nonlinear ecological responses. A key gap remains between detecting statistical breakpoints in ecological indicator responses and translating their uncertainty into spatially explicit priority zones for land management. This study develops an arithmetic Vigor–Organization–Resilience (VOR) Ecological Health Index (EHI) for Northeast China (NEC) using multi-source remote-sensing and geospatial data from 2000 to 2020 across Heilongjiang, Jilin, Liaoning, and Inner Mongolia. The framework combines LightGBM-based SHAP interpretation following candidate-model comparison and repeated spatial optimization, segmented regression calibrated using a spatial wild-cluster block bootstrap, and a confidence-interval-based zoning scheme defined by breakpoint bounds (α, τ^, β). EHI increased significantly across NEC from 2000 to 2020 (Mann–Kendall τ = 0.543, p < 0.001; Sen's slope = 0.00101 year −1 ). Liaoning recorded the highest mean EHI, while Inner Mongolia showed the most consistent monotonic increase. Land-use intensity showed the largest model contribution at the NEC scale and in Heilongjiang, Jilin, and Inner Mongolia, whereas precipitation was the dominant contributor in Liaoning. Spatially calibrated segmented regression identified significant threshold-like responses for elevation, slope, precipitation, and LUI across multiple spatial units, with considerable regional variation, while temperature showed more limited support. LUI breakpoints ranged from 0.9223 to 0.9375 and were interpreted as statistical warning signals rather than field-verified ecological tipping points. Arithmetic–geometric VOR sensitivity testing showed only small differences in LUI breakpoint locations, indicating that the identified LUI transitions were relatively robust to the choice of aggregation formulation. Additional LULC-control analysis showed uniformly weak residual EHI–LUI correlations after accounting for categorical land-cover class (|r| ≤ 0.068), suggesting that shared land-cover structure contributed to the observed association. Final integrated zoning was dominated by high-priority areas in Jilin (53.0%), Liaoning (52.4%), and Heilongjiang (48.06%), whereas Inner Mongolia was dominated by low- and medium-priority areas, with HP accounting for 15.3% and no VHP areas identified. Because EHI is derived from remote-sensing and modeled indicators, the proposed priority zones should be interpreted as decision-support outputs pending independent field-based ecological validation.
Related
- Process-dependent wetland degradation and ecological resilience in Northeast China revealed by a zone-weighted regional wetland composite index
- Scenario-based ecological zoning by coupling ecosystem service value and landscape ecological risk indicators in a rapidly urbanizing river–lake city
- Non-linear responses of ecological indicators to urban environmental drivers across Europe
- Hydrodynamic and ecological modeling for water quality assessment in Lake Qaroun, Egypt
- From Outpatient Consumption to Receiving Waters: A Retrospective Screening Assessment of Antibiotic Loads, Resistance Indicators, and Implications for the Recast Urban Wastewater Treatment Directive.
BibTeX
@article{rashad2026from,
title = {{From ecological health indicators to priority zoning: A threshold-guided assessment framework for Northeast China}},
author = {Eslam Rashad and Yujie Liu and Ahmed Refaee and Junjie Liu and Tao Pan},
journal = {Ecological Indicators},
year = {2026},
month = oct,
doi = {10.1016/j.ecolind.2026.115496},
url = {https://doi.org/10.1016/j.ecolind.2026.115496},
note = {CC BY},
}
RIS
TY - JOUR
TI - From ecological health indicators to priority zoning: A threshold-guided assessment framework for Northeast China
AU - Eslam Rashad
AU - Yujie Liu
AU - Ahmed Refaee
AU - Junjie Liu
AU - Tao Pan
PY - 2026
DA - 2026/10/23
JO - Ecological Indicators
DO - 10.1016/j.ecolind.2026.115496
UR - https://doi.org/10.1016/j.ecolind.2026.115496
AB - Regional ecosystem health assessment requires composite indicators that capture spatial heterogeneity across land-use and climate gradients, yet many frameworks still rely on fixed-threshold classifications and linear driver analyses that may overlook nonlinear ecological responses. A key gap remains between detecting statistical breakpoints in ecological indicator responses and translating their uncertainty into spatially explicit priority zones for land management. This study develops an arithmetic Vigor–Organization–Resilience (VOR) Ecological Health Index (EHI) for Northeast China (NEC) using multi-source remote-sensing and geospatial data from 2000 to 2020 across Heilongjiang, Jilin, Liaoning, and Inner Mongolia. The framework combines LightGBM-based SHAP interpretation following candidate-model comparison and repeated spatial optimization, segmented regression calibrated using a spatial wild-cluster block bootstrap, and a confidence-interval-based zoning scheme defined by breakpoint bounds (α, τ^, β). EHI increased significantly across NEC from 2000 to 2020 (Mann–Kendall τ = 0.543, p < 0.001; Sen's slope = 0.00101 year −1 ). Liaoning recorded the highest mean EHI, while Inner Mongolia showed the most consistent monotonic increase. Land-use intensity showed the largest model contribution at the NEC scale and in Heilongjiang, Jilin, and Inner Mongolia, whereas precipitation was the dominant contributor in Liaoning. Spatially calibrated segmented regression identified significant threshold-like responses for elevation, slope, precipitation, and LUI across multiple spatial units, with considerable regional variation, while temperature showed more limited support. LUI breakpoints ranged from 0.9223 to 0.9375 and were interpreted as statistical warning signals rather than field-verified ecological tipping points. Arithmetic–geometric VOR sensitivity testing showed only small differences in LUI breakpoint locations, indicating that the identified LUI transitions were relatively robust to the choice of aggregation formulation. Additional LULC-control analysis showed uniformly weak residual EHI–LUI correlations after accounting for categorical land-cover class (|r| ≤ 0.068), suggesting that shared land-cover structure contributed to the observed association. Final integrated zoning was dominated by high-priority areas in Jilin (53.0%), Liaoning (52.4%), and Heilongjiang (48.06%), whereas Inner Mongolia was dominated by low- and medium-priority areas, with HP accounting for 15.3% and no VHP areas identified. Because EHI is derived from remote-sensing and modeled indicators, the proposed priority zones should be interpreted as decision-support outputs pending independent field-based ecological validation.
SN - 1470-160X
SN - 1872-7034
ER -