Fei et al. (2026) Seasonality changes in the terrestrial water cycle under different vegetation types and their attribution in the Songliao River Basin, Northeast China

Identification

• Journal: Journal of Hydrology Regional Studies
• Year: 2026
• Date: 2026-01-06
• Authors: Wenli Fei, Lidu Shen, Eryuan Liang, Yage Liu, Yuan Zhang, Anzhi Wang, Jiabing Wu, Ling Zhu, Rongrong Cai, Jie Lai
• DOI: 10.1016/j.ejrh.2025.103100

Research Groups

• CAS Key Laboratory of Forest Ecology and Silviculture, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, China
• Key Laboratory of Wetland Ecology and Environment, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun, China
• Liaoning Provincial Climate Center, Liaoning Meteorological Bureau, Shenyang, China
• University of Chinese Academy of Sciences, Beijing, China

Short Summary

This study quantifies and attributes seasonality changes in terrestrial water cycle components (precipitation, snow water equivalent (SWE), runoff, evapotranspiration (ET), and soil moisture) across different vegetation types in the Songliao River Basin (1981-2020), revealing declining seasonality for most components (especially ET) but increasing SWE seasonality, primarily driven by rising air temperature and altered precipitation patterns.

Objective

• To analyze how the seasonality characteristics of the terrestrial water cycle (TWC) have changed under the main vegetation types of the Songliao River Basin (SRB) for the last four decades.
• To identify the main influencing factors (climatic and vegetation) driving these seasonality changes.

Study Configuration

• Spatial Scale: Songliao River Basin (SRB), Northeast China, covering approximately 1,249,200 square kilometers.
• Temporal Scale: 1981 to 2020 (40 years); runoff data from 1981 to 2018 (38 years).

Methodology and Data

• Models used:
  • Information-entropy-based Seasonality Index (SI)
  • Mann-Kendall trend test
  • SHapley Additive exPlanations (SHAP)
  • XGBoost (Extreme Gradient Boosting) machine learning model
  • CHTESSEL (used in ERA5-LAND reanalysis)
  • VIC (used in CNRD model simulation)
• Data sources:
  • China Meteorological Forcing Dataset (CMFD) (precipitation, air temperature, relative humidity, wind speed)
  • European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis version 5 for land (ERA5-LAND) (Snow Water Equivalent - SWE)
  • China Natural Runoff Dataset (CNRD) version 1.0 (runoff)
  • Global Land Evaporation Amsterdam Model (GLEAM) version 3.8a (evapotranspiration (ET), bare-soil evaporation, canopy evaporation, transpiration, surface soil moisture, root-zone soil moisture)
  • Daily gap-free Normalized Difference Vegetation Index (NDVI) dataset for China (satellite retrieval)
  • MODIS CMG (MCD12C1) Version 6.1 land-cover type product (satellite retrieval)

Main Results

• Precipitation seasonality: Generally decreased across the SRB, particularly near the main streams of the Songhua, Hailar, and Xilamulun Rivers. All land-cover types showed a slight downward trend, indicating more evenly distributed precipitation, primarily due to decreased summer and increased late-spring precipitation proportions.
• Snow Water Equivalent (SWE) seasonality: Generally increased across the SRB, with significant rises in the Songnen Plain and grassland areas. This is attributed to rising air temperature leading to reduced snow amounts, earlier snowmelt, and a shortened snow season, concentrating SWE in winter.
• Runoff seasonality: Generally declined across the SRB, particularly near the main streams of the Hailar, Songhua, and Liao Rivers, with reductions greater than those of precipitation seasonality. In snow-dominated croplands and forests, advanced spring snowmelt increased spring runoff peaks, shifting the runoff pattern to bimodal (spring and summer) and reducing seasonality. In arid grasslands, reduced summer precipitation and increased ET led to reduced summer runoff peaks and a slight decline in seasonality.
• Evapotranspiration (ET) seasonality: Exhibited a significant decreasing trend across the SRB, especially in the Hailar Basin, Songhua River Basin, and eastern Liao River Basin. This trend was observed across all land-cover types except forests. The decline is mainly due to the extended growing season from rising temperatures, increased ET proportions in winter and early spring, and decreased summer ET proportions, resulting in a more uniform monthly ET distribution.
  • Bare-soil evaporation seasonality significantly decreased in forests, savannas, and croplands.
  • Canopy evaporation seasonality decreased, particularly in grasslands, croplands, and savannas.
  • Transpiration seasonality decreased significantly in the northernmost savannas of the SRB.
• Soil moisture seasonality (surface and root-zone): Trends were much weaker than other variables. Surface soil moisture seasonality showed a significant downward trend near the Xilamulun River Basin and the Songnen Plain. Root-zone soil moisture seasonality exhibited significant decreasing trends in the Hailaer Basin, Xilamulun River Basin, Songnen Plain, and Sanjiang Plain. In grasslands, vegetation was the primary driver, while in croplands, air temperature was dominant.
• Overall Attribution: Rising air temperature is the primary driver for the significant increase in SWE seasonality and the significant decrease in ET seasonality and its components. Altered precipitation patterns and changes in vegetation cover also play significant roles in modifying the seasonality of TWC components.

Contributions

• Provides a comprehensive analysis of seasonality changes across multiple terrestrial water cycle components (precipitation, SWE, runoff, ET, and soil moisture) in the Songliao River Basin.
• Quantifies seasonality using an information-entropy-based index, offering a robust and universally applicable method.
• Employs a novel SHAP-based attribution analysis, integrated with mechanistic understanding, to identify the specific roles of climatic and vegetation factors in driving seasonality changes.
• Deepens the understanding of water cycle dynamics under different vegetation types in a crucial agricultural and ecological region of China.
• Offers scientific guidance for regional water resource management, agricultural planning, and disaster early warning systems in the SRB.

Funding

• National Key Research and Development Program of China (2022YFF1300501)
• Natural Science Foundation of Liaoning Province (2023-BSBA-312 and 2023-BSBA-319)

Citation

@article{Fei2026Seasonality,
  author = {Fei, Wenli and Shen, Lidu and Liang, Eryuan and Liu, Yage and Zhang, Yuan and Wang, Anzhi and Wu, Jiabing and Zhu, Ling and Cai, Rongrong and Lai, Jie},
  title = {Seasonality changes in the terrestrial water cycle under different vegetation types and their attribution in the Songliao River Basin, Northeast China},
  journal = {Journal of Hydrology Regional Studies},
  year = {2026},
  doi = {10.1016/j.ejrh.2025.103100},
  url = {https://doi.org/10.1016/j.ejrh.2025.103100}
}