Using XGBoost-SHAP for understanding the ecosystem services trade-off effects and driving mechanisms in ecologically fragile areas DOI Creative Commons
Peiyu Du,

Heju Huai,

Xiaoyang Wu

et al.

Frontiers in Plant Science, Journal Year: 2025, Volume and Issue: 16

Published: April 28, 2025

Introduction Understanding the spatial and temporal variability of Ecosystem services (ES), along with trade-offs synergies among different services, is crucial for effective ecosystem management sustainable regional development. This study focuses on Wensu, Xinjiang, China, as a case to address these challenges. Methods ES their were systematically assessed from 1990 2020. Explainable machine learning models (XGBoost-SHAP) employed quantify nonlinear effects threshold trade-offs, specific attention identifying driving factors. Results (1) From 2020, water yield (WY) soil conservation (SC) exhibited an inverted "N"-shaped downward trend in Wensu County: mean annual WY decreased 22.99 mm 21.32 mm, SC per unit area declined 1440.28 t/km² 1351.3 t/km². Conversely, windbreak sand fixation (WS) showed increase 2.32×10⁷ t 3.11×10⁷ t. Habitat quality (HQ) initially improved then deteriorated, values 0.596, 0.603, 0.519, 0.507 sequentially. (2) Relationships between WY-WS, WY-HQ, WS-HQ, SC-WS, SC-HQ primarily tradeoffs, whereas WY-SC interactions synergistic. Trade-offs SC-HQ, WS-HQ stronger, while weaker. (3) The XGBoost-SHAP model revealed land use type (Land), precipitation (Pre), temperature (Tem) dominant drivers demonstrating responses effects. For instance, intensified when exceeded 17 thresholds governed WY-HQ trade-off/synergy transitions. Discussion advances identification trade-off drivers. model's interpretability capturing complexities clarifies mechanisms underlying dynamics. Findings are generalizable other ecologically vulnerable regions, offering critical insights strategies comparable environments.

Language: Английский

Assessment and multi-scenario prediction of ecosystem services in the Yunnan-Guizhou Plateau based on machine learning and the PLUS model DOI Creative Commons
Yuan Li, Yuling Peng,

H. P. Peng

et al.

Frontiers in Ecology and Evolution, Journal Year: 2025, Volume and Issue: 13

Published: Feb. 18, 2025

Introduction Machine learning techniques, renowned for their ability to process complex datasets and uncover key ecological patterns, have become increasingly instrumental in assessing ecosystem services. Methods This study quantitatively evaluates individual services—such as water yield, carbon storage, habitat quality, soil conservation—on the Yunnan-Guizhou Plateau years 2000, 2010, 2020. A comprehensive service index is employed assess overall capacity, revealing spatiotemporal variations services exploring trade-offs synergies among them. Additionally, machine models identify drivers influencing services, informing design of future scenarios. The PLUS model used project land use changes by 2035 under three scenarios—natural development, planning-oriented, priority. Based on simulation results these scenarios, InVEST applied evaluate various Results During 2000-2020, exhibited significant fluctuations, driven synergies. Land vegetation cover were primary factors affecting with priority scenario demonstrating best performance across all Discussion research integrates model, providing more efficient data interpretation precise design, offering new insights methodologies managing optimizing Plateau. These findings contribute development effective protection sustainable strategies, applicable both plateau similar regions.

Language: Английский

Citations

1
Unraveling supply-demand relationship of urban agglomeration's ecosystem services for spatial management zoning: Insights from threshold effects DOI

Mutian Xu,

Chao Bao

Sustainable Cities and Society, Journal Year: 2025, Volume and Issue: unknown, P. 106239 - 106239

Published: Feb. 1, 2025

Language: Английский

Citations

0
Using XGBoost-SHAP for understanding the ecosystem services trade-off effects and driving mechanisms in ecologically fragile areas DOI Creative Commons
Peiyu Du,

Heju Huai,

Xiaoyang Wu

et al.

Frontiers in Plant Science, Journal Year: 2025, Volume and Issue: 16

Published: April 28, 2025

Introduction Understanding the spatial and temporal variability of Ecosystem services (ES), along with trade-offs synergies among different services, is crucial for effective ecosystem management sustainable regional development. This study focuses on Wensu, Xinjiang, China, as a case to address these challenges. Methods ES their were systematically assessed from 1990 2020. Explainable machine learning models (XGBoost-SHAP) employed quantify nonlinear effects threshold trade-offs, specific attention identifying driving factors. Results (1) From 2020, water yield (WY) soil conservation (SC) exhibited an inverted "N"-shaped downward trend in Wensu County: mean annual WY decreased 22.99 mm 21.32 mm, SC per unit area declined 1440.28 t/km² 1351.3 t/km². Conversely, windbreak sand fixation (WS) showed increase 2.32×10⁷ t 3.11×10⁷ t. Habitat quality (HQ) initially improved then deteriorated, values 0.596, 0.603, 0.519, 0.507 sequentially. (2) Relationships between WY-WS, WY-HQ, WS-HQ, SC-WS, SC-HQ primarily tradeoffs, whereas WY-SC interactions synergistic. Trade-offs SC-HQ, WS-HQ stronger, while weaker. (3) The XGBoost-SHAP model revealed land use type (Land), precipitation (Pre), temperature (Tem) dominant drivers demonstrating responses effects. For instance, intensified when exceeded 17 thresholds governed WY-HQ trade-off/synergy transitions. Discussion advances identification trade-off drivers. model's interpretability capturing complexities clarifies mechanisms underlying dynamics. Findings are generalizable other ecologically vulnerable regions, offering critical insights strategies comparable environments.

Language: Английский

Citations

0