Cao et al. (2025) Integrated modeling of blue and green water evolution in a headwater region of Chaohu Lake: Impacts of climate and surface environmental factors

Identification

• Journal: Ecological Indicators
• Year: 2025
• Date: 2025-12-22
• Authors: Yang Cao, Weijia Xu, Xin Guan, Jingjing Yang, Yingruyi Feng, Le Li, Miaomiao Zhang
• DOI: 10.1016/j.ecolind.2025.114356

Research Groups

• Research Center for Weathering and Carbon Cycle, School of History Culture and Tourism, Fuyang Normal University, Fuyang, China

Short Summary

This study analyzed the multi-decade dynamics of blue water (surface and groundwater) and green water (soil-stored rainfall used by plants) in the Hangbu River Basin using a semi-distributed hydrological model, revealing their "decline-recovery" trends and the spatially heterogeneous impacts of climate and surface environmental factors.

Objective

• To analyze the characteristics of spatial and temporal changes of blue water (BW) and green water (GW) at different time scales in the Hangbu River Basin (HRB), including long-term trends and phased changes.
• To explore the effects of climate change, as well as environmental factors such as slope and land use, on the changes of BW and GW.

Study Configuration

• Spatial Scale: Hangbu River Basin (HRB), approximately 4,246 square kilometers. The basin was divided into 254 sub-basins, further subdivided into Hydrological Response Units (HRUs).
• Temporal Scale: Long-term analysis from 1959 to 2023 (65 years) for BW-GW dynamics. Model calibration (2007–2014) and validation (2015–2020) for stream flow. Land use data available for ten sets from 1980 to 2020. Model operates on a daily scale.

Methodology and Data

• Models used:
  • Hydrological model: Physically-based, semi-distributed Soil and Water Assessment Tool (SWAT).
  • Parameter optimization: SUFI-2 algorithm in SWAT-CUP software.
  • Trend analysis: Mann-Kendall (MK) mutation testing, Theil-Sen slope estimator.
  • Spatial autocorrelation: Global Moran’s I and Local Moran’s I.
  • Driving factor analysis: Principal Component Analysis (PCA), Geographically Weighted Regression (GWR) model.
• Data sources:
  • Spatial data: Digital Elevation Model (DEM) (90 meter spatial resolution, National Earth System Science Data Center of China), Land use data (30 meter spatial resolution, 1980–2020, China Multi-Period Land Use/Cover Remote Sensing Monitoring Dataset - CNLUCC), Soil data (Digital Soil Map of the World (DSMW) by FAO, 1:5,000,000 scale).
  • Attribute data: Daily stream flow records (2007–2020, Xiaotian and Taoxi hydrological stations, Hydrological Yearbook of the Yangtze River Basin), Long-term daily meteorological data (1959–2023, 4 national benchmark meteorological stations, National Meteorological Information Center of China (NMIC)), including precipitation, maximum and minimum temperature, relative humidity, wind speed, and sunshine duration.
  • Derived data: Evapotranspiration (ET) estimated by the Penman-Monteith equation within the SWAT model.

Main Results

• Temporal Trends: Both BW and GW generally exhibited a "decline–recovery" trend over the 1959–2023 period.
  • BW decreased rapidly from 1959 to 1968 (slope = −37.61 mm/year) and gradually recovered from 1969 to 2023 (slope = 1.48 mm/year), with a mutation around 1968.
  • GW declined from 1959 to 1995 (slope = −1.54 mm/year) and recovered after 1996 (slope = 2.17 mm/year), with a distinct turning point in 1995.
• Seasonal Variability: GW volumes were significantly higher than BW across all seasons.
  • The GW-to-BW ratios were 4.6 in spring, 2.3 in summer, 4.4 in autumn, and 11.4 in winter, indicating significant seasonal differences.
  • BW showed the greatest interannual variability in summer (amplitude 282.6 mm) and the highest change rate in autumn (59.4%).
  • GW showed the largest interannual fluctuations in autumn (amplitude 166.7 mm) and the highest change rate in autumn (6.5%).
• Spatial Patterns: Both BW and GW exhibited significant positive spatial autocorrelation (Global Moran’s I: BW = 0.85, GW = 0.58, both p < 0.01), with BW showing more pronounced spatial agglomeration.
  • BW High-High (HH) clusters were mainly distributed in upstream forested regions, while Low-Low (LL) clusters were concentrated in the midstream and downstream areas.
  • GW HH clusters were primarily found in reservoir areas (upstream and midstream) and dense river network regions (downstream), while LL clusters were widespread across the entire basin.
  • BW consistently followed a spatial pattern of upstream > midstream > downstream.
  • GW generally followed a spatial pattern of upstream > downstream > midstream.
  • Regions with negative BW trends were mainly concentrated in upstream cropland areas, while positive trends were primarily located in upstream forest land.
  • For GW, negative trends were predominantly found in the entire upstream region, whereas positive trends were mainly observed in the downstream region.
• Driving Factors (Temporal): Precipitation was identified as the primary driver of both BW (r = 0.99) and GW (r = 0.77) variability. Forest land positively reinforced water storage, while cropland showed a significant negative correlation with both BW (r = -0.72) and GW (r = -0.58). Urban land was positively correlated with BW (r = 0.75) and GW (r = 0.62).
• Driving Factors (Spatial):
  • For BW, precipitation exerted the strongest positive influence (mean coefficient: 155), while ET showed the strongest negative effect (mean coefficient: -43). Water bodies contributed positively (13), while slope, cropland, and forest land generally constrained BW.
  • For GW, ET was the dominant positive driver (mean coefficient: 165), and slope had the strongest negative effect (mean coefficient: -23). Forest land showed an average positive effect (23), but its influence was complex due to negative slope effects. Urban land exerted a strong negative effect on GW (-19).

Contributions

• Provides a deeper understanding of the distribution patterns and formation mechanisms of blue water (BW) and green water (GW) in complex river basins.
• Reveals the distinct spatiotemporal dynamics and driving factors of BW-GW, including their "decline-recovery" trends and seasonal variability.
• Offers scientific support for sustainable water resource management and land-use policy in the Hangbu River Basin and similar regions, emphasizing the need for differentiated water management strategies based on local conditions.

Funding

• Fuyang Normal University University-Level Scientific Research Projects (NO. 2024HHWH01; 2022FSKJ0300)
• Fuyang Normal University Doctoral Research Start-up Fund (NO. 2024KYQD0034)

Citation

@article{Cao2025Integrated,
  author = {Cao, Yang and Xu, Weijia and Guan, Xin and Yang, Jingjing and Feng, Yingruyi and Li, Le and Zhang, Miaomiao},
  title = {Integrated modeling of blue and green water evolution in a headwater region of Chaohu Lake: Impacts of climate and surface environmental factors},
  journal = {Ecological Indicators},
  year = {2025},
  doi = {10.1016/j.ecolind.2025.114356},
  url = {https://doi.org/10.1016/j.ecolind.2025.114356}
}