Yuan et al. (2026) Baseflow in karst regions is significantly higher than the global average and exhibits spatial variability

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-03-27
• Authors: Ze Yuan, Qiuwen Zhou, Yuan Li, Yuluan Zhao, Shengtian Yang
• DOI: 10.5194/hess-30-1607-2026

Research Groups

• School of Geography and Environmental Science, Guizhou Normal University, Guiyang, China
• Karst Ecosystem Field Scientific Observation and Research Station of Guizhou Normal University & Guanling Autonomous County, Guanling, China
• Institute of Ecological Civilization, Guizhou Normal University, Guiyang, China
• College of Water Sciences, Beijing Normal University, Beijing, China

Short Summary

This study quantifies baseflow characteristics in 1375 global karst basins, revealing that baseflow constitutes approximately 78 % of streamflow, significantly higher than the global average of 60 %. It also identifies significant spatial variability and an increasing temporal trend in karst baseflow, primarily influenced by vegetation factors.

Objective

• To explore the baseflow characteristics and their internal differences across global karst regions, and to evaluate the influence of different environmental factors on these characteristics.
• To evaluate the applicability of twelve baseflow separation methods (including graphical and digital filter methods) in karst watersheds and identify the most suitable approaches.
• To reveal the unique hydrological signature of karst baseflow and analyze its spatiotemporal evolution characteristics under changing climate conditions.
• To quantify the heterogeneity of baseflow across different karst landform types and clarify the internal differences.
• To identify the dominant climatic, topographic, and geological drivers of baseflow variability using an XGBoost machine learning model.

Study Configuration

• Spatial Scale: Global, analyzing 1375 karst watersheds with areas smaller than 2500 square kilometers.
• Temporal Scale: Runoff data from 1960 to 2015 for baseflow index analysis; 2011–2012 for influencing factors analysis.

Methodology and Data

• Models used:
  • Baseflow separation: 12 distinct methods (4 graphical, 8 digital filtering), including Exponential Weighted Moving Average (EWMA) filter, Eckhardt filter, FIM, and Boughton.
  • Parameter estimation for digital filters: Brutsaert (2008) method for recession constant, multi-objective optimization approach by Arnold (Rammal et al., 2018) for secondary parameters.
  • Attributional analysis: XGBoost machine learning model.
  • Evaluation metrics: Nash-Sutcliffe Efficiency (NSE) and Kling-Gupta Efficiency (KGE).
  • Trend analysis: Mann–Kendall test and linear regression.
• Data sources:
  • Runoff data: Global Runoff Data Center (GRDC), The European Water Archive, National River Flow Archive (UK), Brazilian National Water Authority, The National Hydrological Data Archive of Canada, The Chinese Ministry of Water Resources, The National Hydrological Information System of the United States.
  • Climatic characterization: Köppen-Geiger climate classification system.
  • Influencing factors (12 factors): Temperature, rainfall, depth to bedrock, water storage in epikarst, slope, elevation, soil evaporation, runoff, population density, gross primary productivity (GPP), relative humidity, and surface radiation.

Main Results

• The baseflow index (BFI) of global karst areas is approximately 78 ± 6.9 %, which is significantly higher than the global average BFI of 60 %.
• Digital filter methods were most suitable for baseflow separation in karst regions (71 % of stations), with EWMA (24 %) and Eckhardt (21 %) being the most effective.
• BFI in arid karst regions (average 82 %) is significantly higher than in subtropical karst regions (77 %).
• Significant spatial variability in BFI exists even within the same climate zone, with differences exceeding 10 % in some regions.
• Vegetation factors, reflected in primary productivity, have the highest influence on karst BFI (14.8 %), while climatic factors (relative humidity, air temperature) have a lower influence (less than 5 %).
• The global karst BFI shows a statistically significant increasing trend of approximately 1.5 % from 1960 to 2015, stabilizing around 78.5 % ± 0.5 % since 2000.
• Runoff volume and epikarst water storage are identified as the most stable positive drivers of baseflow, while population density shows an inhibitory effect.

Contributions

• Provides the first systematic global characterization and quantification of baseflow in karst regions using a comprehensive dataset of 1375 basins.
• Evaluates and identifies the most suitable hydrograph separation methods for karst hydrology, highlighting the effectiveness of EWMA, Eckhardt, Boughton, and FIM.
• Reveals the unique hydrological signature of karst baseflow, demonstrating its significantly higher contribution to streamflow compared to non-karst regions.
• Quantifies the spatiotemporal variability of karst baseflow across different climatic zones and identifies key environmental drivers using an XGBoost model.
• Offers insights into the increasing trend of karst BFI over time, linking it to global groundwater depletion and the unique hydrological properties of karst aquifers.

Funding

• National Natural Science Foundation of China (grant nos. 42461004, U1812401, U1612441)
• Science and Technology Program of Guizhou Province (Qiankehejichu-ZK[2025] Zhongdian 045; Qiankehejichu-ZK[2025] Mianshang 268)

Citation

@article{Yuan2026Baseflow,
  author = {Yuan, Ze and Zhou, Qiuwen and Li, Yuan and Zhao, Yuluan and Yang, Shengtian},
  title = {Baseflow in karst regions is significantly higher than the global average and exhibits spatial variability},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-1607-2026},
  url = {https://doi.org/10.5194/hess-30-1607-2026}
}