Deciphering the nonlinear controls on water–sediment fluxes across scales in the Yellow River headwaters, Tibetan Plateau

Fluvial water – sediment fluxes contribute to material cycling, landform development and ecological change. Yet, the synergistic effects of multiple factors on water‑sediment fluxes and their respective contributions and pathways remain unclear in alpine regions. Here, complementary analytical methods were used to investigate the multiscale processes of water–sediment fluxes in the source region of the Yellow River Basin. Partial wavelet coherency (PWC) and multiple wavelet coherence (MWC) were used to assess individual and synergistic effects on water‑sediment fluxes, and the contributions and influence pathways of these factors were disentangled using XGBoost–SHAP and partial least squares structural equation modeling (PLS–SEM). Runoff increased significantly at 0.25 billion m 3 /decade, while sediment load showed a decreasing trend (1.95 million tons/decade). The water–sediment fluxes showed significant intermittent annual oscillations during 1982–2020. PWC analysis indicated that climate factors mask the explanatory power of vegetation for runoff, yet vegetation’s suppressive effect on sediment transport becomes evident once climatic influences are removed. MWC analysis showed that single factors were insufficient to explain the variability of water–sediment fluxes. The PRE (Precipitation)–TEP (Temperature)–SM (Soil Moisture) and PRE–TEP–NDVI combinations provided strong explanations of runoff and sediment load variability, with percentage area of significant coherence (PASC) values of 60.1% and 45.3%, respectively. XGBoost–SHAP identified precipitation and NDVI as the dominant controls on runoff and sediment load variability, with contributions of 38.22% and 37.98%, respectively. NDVI negatively affects runoff variability by mediating the precipitation–runoff relationship. Precipitation influences sediment transport via direct erosion and indirect runoff amplification, with path coefficients of 0.63 from precipitation to runoff and 0.64 from runoff to sediment. These findings clarify the inferred controls on water–sediment fluxes and inform soil erosion prevention strategies in cryospheric regions under climate change.

Authors

Institutions

Publication Details

Journal
Agricultural Water Management
Published
2026-11-13
DOI
https://doi.org/10.1016/j.agwat.2026.110947
Primary Topic
Soil erosion and sediment transport
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Deciphering the nonlinear controls on water–sediment fluxes across scales in the Yellow River headwaters, Tibetan Plateau

Yuning Chi, Panxing He, Shaoxiong Zhang, Xin Li et al.
Agricultural Water Management
Soil erosion and sediment transport
article

Deciphering the nonlinear controls on water–sediment fluxes across scales in the Yellow River headwaters, Tibetan Plateau

Yuning Chi, Panxing He, Shaoxiong Zhang, Xin Li, Jun Ma, Songyan Zhu, Yang Li, Junjie Wang, Bing Shi
article en

Abstract

Fluvial water – sediment fluxes contribute to material cycling, landform development and ecological change. Yet, the synergistic effects of multiple factors on water‑sediment fluxes and their respective contributions and pathways remain unclear in alpine regions. Here, complementary analytical methods were used to investigate the multiscale processes of water–sediment fluxes in the source region of the Yellow River Basin. Partial wavelet coherency (PWC) and multiple wavelet coherence (MWC) were used to assess individual and synergistic effects on water‑sediment fluxes, and the contributions and influence pathways of these factors were disentangled using XGBoost–SHAP and partial least squares structural equation modeling (PLS–SEM). Runoff increased significantly at 0.25 billion m 3 /decade, while sediment load showed a decreasing trend (1.95 million tons/decade). The water–sediment fluxes showed significant intermittent annual oscillations during 1982–2020. PWC analysis indicated that climate factors mask the explanatory power of vegetation for runoff, yet vegetation’s suppressive effect on sediment transport becomes evident once climatic influences are removed. MWC analysis showed that single factors were insufficient to explain the variability of water–sediment fluxes. The PRE (Precipitation)–TEP (Temperature)–SM (Soil Moisture) and PRE–TEP–NDVI combinations provided strong explanations of runoff and sediment load variability, with percentage area of significant coherence (PASC) values of 60.1% and 45.3%, respectively. XGBoost–SHAP identified precipitation and NDVI as the dominant controls on runoff and sediment load variability, with contributions of 38.22% and 37.98%, respectively. NDVI negatively affects runoff variability by mediating the precipitation–runoff relationship. Precipitation influences sediment transport via direct erosion and indirect runoff amplification, with path coefficients of 0.63 from precipitation to runoff and 0.64 from runoff to sediment. These findings clarify the inferred controls on water–sediment fluxes and inform soil erosion prevention strategies in cryospheric regions under climate change.

Agricultural Water ManagementVol. 338
Chinese Academy of Sciences (CN), Fudan University (CN), Northeast Institute of Geography and Agroecology (CN), University of Southampton (GB), Ocean University of China (CN), Shijiazhuang Tiedao University (CN)
Shijiazhuang Tiedao University, Fudan University
Clean water and sanitation
Soil erosion and sediment transport
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.

Laika access coming soon