Nascimento et al. (2025) How do geological map details influence the identification of geology-streamflow relationships in large-sample hydrology studies?

Identification

• Journal: Hydrology and earth system sciences
• Year: 2025
• Date: 2025-12-15
• Authors: Thiago Victor Medeiros do Nascimento, Julia Rudlang, Sebastian Gnann, Jan Seibert, Markus Hrachowitz, Fabrizio Fenicia
• DOI: 10.5194/hess-29-7173-2025

Research Groups

• Eawag: Swiss Federal Institute of Aquatic Science and Technology, Dübendorf, Switzerland
• Department of Water Management, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Delft, the Netherlands
• Chair of Hydrology, Faculty of Environment and Natural Resources, University of Freiburg, Freiburg, Germany
• Department of Geography, University of Zurich, Zurich, Switzerland

Short Summary

This study investigates how the level of detail in geological maps (global, continental, regional) influences the identification of geology-streamflow relationships across 4469 European catchments using a multi-scale, nested-catchment approach. It finds that while large-scale analyses show inconsistent map performance, increasing geological detail at intermediate and small scales consistently strengthens correlations with streamflow signatures, particularly for baseflow, aligning better with hydrological process understanding.

Objective

• To investigate how the level of detail in geological maps (global, continental, regional) influences the identification of relationships between geology and streamflow signatures across multiple spatial scales in European catchments.
• To quantify the relative influences of landscape and climate attributes on streamflow signatures and evaluate how geological map detail affects the interpretation of dominant controls.
• To assess if increased geological detail leads to correlations more consistent with physical hydrological understanding.

Study Configuration

• Spatial Scale: Multi-scale approach:
  • Large scale: 63 river basins across Europe (comprising 4469 catchments), with total areas between 7000 and 35000 square kilometers.
  • Intermediate scale: The Moselle basin (27100 square kilometers, 152 sub-catchments).
  • Small scale: Five nested catchments within the Moselle basin (e.g., Moselle-Toul, Moselle-Saar), each with at least nine gauged sub-catchments.
  • Individual catchment areas ranged from 50 to 35000 square kilometers.
• Temporal Scale:
  • Streamflow data: At least 10 years of daily streamflow data (not necessarily consecutive) between 1950 and 2020.
  • Climatic attributes: Derived from the E-OBS time series dataset between 1950 and 2022.

Methodology and Data

• Models used: Exploratory statistical analysis using the Spearman correlation coefficient (r_s) to identify relationships between catchment attributes and streamflow signatures.
• Data sources:
  • EStreams dataset (do Nascimento et al., 2024a): Hydrometeorological time series and landscape attributes for 4469 selected European catchments.
  • Global geological map: Global Lithological Map (GLiM) (Hartmann et al., 2012) at 1:3750000 scale, classified into 16 bedrock types.
  • Continental geological map: International Hydrogeological Map of Europe (IHME) (Duscher et al., 2019; Günther and Duscher, 2019) at 1:1500000 scale, classified into 31 bedrock types.
  • Regional geological map (for Moselle basin): Combined data from BD LISA database (France, 1:250000), Geologische Übersichtskarte der Bundesrepublik Deutschland (GÜK200) (Germany, 1:200000), Administration de la gestion de l'eau (Luxembourg, 1:250000), and IHME (Belgium). Total 31 classes for the Moselle.
  • Climate data: E-OBS time series dataset (Cornes et al., 2018).
  • Other landscape attributes: Topography (e.g., mean elevation, mean slope), soils (e.g., mean fraction of sand/silt/clay, rooting depth), and land use (e.g., CORINE Land Cover, mean NDVI).
  • All geological maps were reclassified into four qualitative permeability classes: low, medium-low, medium-high, and high.

Main Results

• Large-scale analysis (63 nested basins): Dominant controls on streamflow signatures varied widely between basins, with landscape attributes frequently showing stronger correlations with baseflow-related signatures than climate attributes (51 out of 63 basins). Neither global nor continental geology maps consistently outperformed the other for all streamflow signatures, though the continental map showed a slight advantage for flow extremes and surface water dynamics.
• Intermediate-scale analysis (Moselle basin): A clear increase in Spearman correlation coefficient (|rs|) values was observed when transitioning from global (average max |rs| = 0.45) to continental to regional (average max |r_s| = 0.68) geology maps for all streamflow signatures. With the regional map, geology became the most correlated attribute group for four out of six streamflow signatures, particularly those related to baseflow and flow persistency.
• Small-scale analysis (five Moselle sub-catchments): Correlation patterns varied considerably across individual sub-catchments. However, the regional geology map consistently produced positive correlations with the high-permeability class for the baseflow index (r_s > 0.40) across all five catchments, aligning with process understanding.
• General findings: Sufficient geological heterogeneity is crucial for detecting strong landscape-hydrology relationships. Reclassifying geological units into hydrologically relevant permeability categories is essential for identifying meaningful correlations. Global maps (e.g., GLiM level 1) may lack sufficient detail to disentangle hydrological behaviors, suggesting the need for higher detail levels.

Contributions

• Systematically investigates the impact of geological map detail (global, continental, regional) on identifying geology-streamflow relationships in large-sample hydrology studies across multiple scales.
• Demonstrates that the perceived "uniqueness of place" or weak landscape influence in large-sample studies can be partly attributed to the insufficient detail of landscape data, particularly geology.
• Highlights the benefit of integrating detailed, region-specific geological data and a nested basins design for robust assessment of streamflow generation processes.
• Provides insights into the complementary strengths and weaknesses of global versus continental geological maps (e.g., contour detail versus class diversity).
• Underscores the importance of reclassifying raw geological data into hydrologically meaningful permeability classes for effective analysis.
• Offers practical recommendations for future large-sample hydrology studies, including using higher detail levels of global datasets (e.g., GLiM level 3) and prioritizing regionalization or clustering approaches.

Funding

• "Money Follows Cooperation" project (project no. OCENW.M.21.230) between the Netherlands Organization for Scientific Research (NWO) and the Swiss National Science Foundation (SNSF).
• TU Delft Climate Action Research and Education seed funds.

Citation

@article{Nascimento2025How,
  author = {Nascimento, Thiago Victor Medeiros do and Rudlang, Julia and Gnann, Sebastian and Seibert, Jan and Hrachowitz, Markus and Fenicia, Fabrizio},
  title = {How do geological map details influence the identification of geology-streamflow relationships in large-sample hydrology studies?},
  journal = {Hydrology and earth system sciences},
  year = {2025},
  doi = {10.5194/hess-29-7173-2025},
  url = {https://doi.org/10.5194/hess-29-7173-2025}
}