Ecological Indicators Reveal Contrasting Post-Earthquake Recovery Trajectories and Resilience Across Mountain Ecosystems

Net primary productivity (NPP) is a useful ecological indicator for evaluating ecosystem functioning and recovery after natural disturbance. Here, we used multi-source remote-sensing and environmental data from 2010 to 2024 to compare post-earthquake NPP recovery in the 2017 Jiuzhaigou and Milin earthquake regions. We integrated trend analysis, temporal persistence assessment, co-seismic landslide overlay, eXtreme Gradient Boosting (XGBoost) interpreted using SHapley Additive exPlanations (SHAP), and locally weighted scatterplot smoothing (LOWESS)-based contribution-transition detection to compare NPP dynamics and environmental associations in the two earthquake-affected regions. The two regions exhibited contrasting responses. The slope of regional mean annual NPP decreased from 9.69 to 8.21 g C m−2 yr−2 between the pre- and post-earthquake periods in Jiuzhaigou, whereas it increased from 3.69 to 9.37 g C m−2 yr−2 in Milin. Hurst-based trend-persistence analysis indicated a higher proportion of sustained improvement and a lower proportion of degradation-related classes in Jiuzhaigou, whereas Milin showed more degradation-related classes. Within co-seismic landslide zones, sustained degradation dominated in both regions and was more extensive in Milin. XGBoost–SHAP results showed that pre-earthquake NPP was the most influential predictor in both regions, but the contributions of secondary predictors differed. Elevation made a larger relative contribution to the Jiuzhaigou model, whereas pre-earthquake NDVI, precipitation, and temperature made larger relative contributions to the Milin model. LOWESS analysis further revealed distinct nonlinear response transitions in temperature, precipitation, and elevation. These results demonstrate the value of combining NPP trends, temporal persistence, and model interpretation to compare vegetation productivity dynamics and inform assessments of post-earthquake functional recovery in mountain ecosystems.

Authors

Institutions

Publication Details

Journal
Remote Sensing
Published
2026-10-09
DOI
https://doi.org/10.3390/rs18203451
Primary Topic
Remote Sensing in Agriculture
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Ecological Indicators Reveal Contrasting Post-Earthquake Recovery Trajectories and Resilience Across Mountain Ecosystems

Wensong Wang, Xin Wang, Lanxin Dai, Qiang Xu et al.
Remote Sensing
Remote Sensing in Agriculture
article

Ecological Indicators Reveal Contrasting Post-Earthquake Recovery Trajectories and Resilience Across Mountain Ecosystems

Wensong Wang, Xin Wang, Lanxin Dai, Qiang Xu, Li Tian, Xuanmei Fan
article en

Abstract

Net primary productivity (NPP) is a useful ecological indicator for evaluating ecosystem functioning and recovery after natural disturbance. Here, we used multi-source remote-sensing and environmental data from 2010 to 2024 to compare post-earthquake NPP recovery in the 2017 Jiuzhaigou and Milin earthquake regions. We integrated trend analysis, temporal persistence assessment, co-seismic landslide overlay, eXtreme Gradient Boosting (XGBoost) interpreted using SHapley Additive exPlanations (SHAP), and locally weighted scatterplot smoothing (LOWESS)-based contribution-transition detection to compare NPP dynamics and environmental associations in the two earthquake-affected regions. The two regions exhibited contrasting responses. The slope of regional mean annual NPP decreased from 9.69 to 8.21 g C m−2 yr−2 between the pre- and post-earthquake periods in Jiuzhaigou, whereas it increased from 3.69 to 9.37 g C m−2 yr−2 in Milin. Hurst-based trend-persistence analysis indicated a higher proportion of sustained improvement and a lower proportion of degradation-related classes in Jiuzhaigou, whereas Milin showed more degradation-related classes. Within co-seismic landslide zones, sustained degradation dominated in both regions and was more extensive in Milin. XGBoost–SHAP results showed that pre-earthquake NPP was the most influential predictor in both regions, but the contributions of secondary predictors differed. Elevation made a larger relative contribution to the Jiuzhaigou model, whereas pre-earthquake NDVI, precipitation, and temperature made larger relative contributions to the Milin model. LOWESS analysis further revealed distinct nonlinear response transitions in temperature, precipitation, and elevation. These results demonstrate the value of combining NPP trends, temporal persistence, and model interpretation to compare vegetation productivity dynamics and inform assessments of post-earthquake functional recovery in mountain ecosystems.

Remote SensingVol. 18(20)
Chengdu University of Technology (CN), State Key Laboratory of Geohazard Prevention and Geoenvironment Protection
Openalex Percentile: Top 15%
Remote Sensing in Agriculture
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.