EUVIMEDSkopri EUVIMED
Crossref

Performance of Different Change-Point Detection Methods for Ecosystem Services at Multiple Scales

Zhengang Wang, Ruifang Hao, Yunqi Wang, Yue Lu, Yajing Zhang, Yanan Chang, Zhaolei Xue, Yahui Zhang

Ecosystem Health and Sustainability · 2026-01-01 · 12

DOI: 10.34133/ehs.0548 ↗

At a glance

Journal article in Ecosystem Health and Sustainability, published online on 2026-06-01. Written by Zhengang Wang and 7 co-authors. 63 references. Funded by: National Key R&D Program of China.

AstrattDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Ecosystem services and ecosystem stability have a close relationship. Change-point detection in long-term ecosystem services is critical for ecosystem management and sustainability. However, the efficiency and suitable application scope of change-point methods for ecosystem-service time series remain unclear. In this study, the contrasting regions of the Yangtze River Basin and the Yellow River Basin were selected as the study areas. We compared 2 nonparametric approaches of the Mann–Kendall (MK) test and the Pettitt test to detect the abrupt shift of annual ecosystem services of water yield, carbon storage, and soil conservation from 2000 to 2023. The ecosystem services were estimated with the Integrated Valuation of Ecosystem Services and Tradeoffs model and the revised Universal Soil Loss Equation. Change-point detection was assessed at 3 spatial scales (basin, sub-basin, and pixel) to evaluate the performance and interpretability of different change-point detection methods. Both methods detected clear shift years at the basin scale. At the sub-basin scale, detected change years became more dispersed, suggesting weaker temporal synchrony among hydrological units. At the pixel scale, spatial heterogeneity was strongest. The MK test identified broader water-yield shifts, whereas the Pettitt test highlighted more spatially discrete changes in the carbon storage. Overall, both methods were sensitive and robust across scales. The MK test better captured broad and spatially extensive changes, while the Pettitt test was more effective for localized and isolated shifts. These findings provide a reference for long-term ecosystem-service change-point monitoring and support basin management and policy development.

Funding

  • National Key R&D Program of China

Cite

APA

Wang, Z., Hao, R., Wang, Y., Lu, Y., Zhang, Y., Chang, Y., Xue, Z., & Zhang, Y. (2026). Performance of Different Change-Point Detection Methods for Ecosystem Services at Multiple Scales. Ecosystem Health and Sustainability, 12. https://doi.org/10.34133/ehs.0548

Vancouver

Wang Z, Hao R, Wang Y, Lu Y, Zhang Y, Chang Y, et al. Performance of Different Change-Point Detection Methods for Ecosystem Services at Multiple Scales. Ecosystem Health and Sustainability. 2026;12. doi:10.34133/ehs.0548

BibTeX
@article{Wang2026,
  author = {Wang, Zhengang and Hao, Ruifang and Wang, Yunqi and Lu, Yue and Zhang, Yajing and Chang, Yanan and Xue, Zhaolei and Zhang, Yahui},
  title = {Performance of Different Change-Point Detection Methods for Ecosystem Services at Multiple Scales},
  journal = {Ecosystem Health and Sustainability},
  year = {2026},
  volume = {12},
  doi = {10.34133/ehs.0548}
}
Start for free →