Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 41 of the 41 references it cites were found in a registry or an index; none of the 11 sources its claims cite is checked for support yet.

Global‐scale river network extraction based on high‐resolution topography and constrained by lithology, climate, slope, and observed drainage density

Open access
Schneider, A., Jost, A., Coulon, C., Silvestre, M., Théry, S., Ducharne, A. · 2017 · Geophysical Research Letters

Abstract

To improve the representation of surface and groundwater flows, global land surface models rely heavily on high‐resolution digital elevation models (DEMs). River pixels are routinely defined as pixels with drainage areas that are greater than a critical drainage area (Acr). This parameter is usually uniform across the globe, and the dependence of drainage density on many environmental factors is often overlooked. Using the 15″ HydroSHEDS DEM as an example, we propose the calibration of a spatially variable Acr as a function of slope, lithology, and climate, to match drainage densities from reference river networks at a 1:50,000 scale in France and Australia. Two variable Acr models with varying complexities were derived from the calibration, with satisfactory performances compared to the reference river networks. Intermittency assessment is also proposed. With these simple tools, river networks with natural heterogeneities at the 1:50,000 scale can be extracted from any DEM.

Overview

CoChat Summary

TL;DR

The paper calibrates spatially variable critical drainage area (A_cr) values for extracting river networks from the 15″ HydroSHEDS DEM as functions of lithology, climate, and slope, using 1:50,000 reference networks in France and Australia. The resulting LC and LCS models reproduce reference drainage densities reasonably in calibration countries and predict a global mean drainage density of about 0.70 km⁻¹, much higher than classical uniform-A_cr global networks.

Key findings

  • Calibrated A_cr values are much smaller than those used in global databases (25–1000 km²). The LC model uses 11 A_cr values from 0.3 to 97 km² across 35 environmental classes; the LCS model uses 29 A_cr values from 0.3 to 193 km² across 120 classes.
  • In calibration countries, LCS performs slightly better than LC: Australia mean δ is 0.43 km⁻¹ for reference, 0.42 for LC, and 0.45 for LCS; France mean δ is 0.75 km⁻¹ for reference, 0.68 for LC, and 0.71 for LCS. Biases approach zero in the U.S., France, and Australia (less than 5% absolute value).
  • Global mean drainage density is predicted at approximately 0.70 km⁻¹ (LC 0.69, LCS 0.74), compared with 0.08 km⁻¹ for the FAO/AQUAMAPS network. Global intermittent-stream proportions are 34% for FAO, 27% for LC, and 29% for LCS.
  • A_cr increases and reference δ decreases when precipitation decreases, for permeable rocks, and when slope decreases. In Australia, a 500 mm yr⁻¹ precipitation threshold classifies 42% of LC/LCS rivers as intermittent, versus 69% in FAO and 98% in Geofabric.

Methods

  • Used the 15″ HydroSHEDS DEM for flow directions, flow accumulation, and slope; global lithology from Hartmann and Moosdorf (2012) with 12 classes; and WFDEI precipitation (1980–2009) at 0.5°. Reference networks were Australia Geofabric and France CARTHAGE at 1:50,000 for calibration, with U.S. NHD at 1:24,000 and Brazil IBGE at 1:25,000–1:250,000 for independent validation.
  • Drainage density was calculated on a 7.5° grid (about 150 km² at the equator). Environmental classes combined 12 lithology, 5 precipitation, and 5 slope classes (300 classes); A_cr was calibrated by minimizing bias between modeled δ and reference δ, testing A_cr from 0.3 to 200 km², and calibration was restricted to grid cells where dominant lithology covered more than 70% of the cell (81% of continents).
  • Intermittency was assessed with a calibrated precipitation threshold of 500 mm yr⁻¹: rivers initiated where precipitation is below this threshold are intermittent until reaching a pixel where precipitation exceeds it; downstream pixels then remain perennial.

Limitations

  • The paper states that calibration and validation strongly depend on reference data quality, with main uncertainties from unnatural channels and scale inconsistencies; DEM and flow-accumulation quality are important but difficult to evaluate; lithological input can induce uncertainty; and the first-order intermittency classification needs better description of multiple controls. It also notes that models still underestimate some very high δ values and cannot reproduce disrupted connectivity.
  • The paper states that vegetation and other control factors were not included for simplicity, and calibration was limited to France and Australia.
  • Evident from the text: validation outside the calibration countries is weak, with U.S. correlations of 0.09–0.11 and Brazil correlations of 0.14, plus a high positive Brazil bias of about 0.40–0.41 km⁻¹; the text attributes this to reference-network scale inconsistencies, U.S. lithological map discontinuities, and artificial drainage in southwestern France.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1002/2016gl071844

Cites

41 of 41 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 Valid · 0 Not valid · 11 Not checked · 22 Own results and interpretations

5 more claims

Create a free account to see all 33 claims and to verify the 5 references still open.

See all claims — free account
  • It may state two separate claims as one. 19
  • Probably not a substantial claim. 7

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 41 references

Research works printed but not linked: 1

  1. Dingman S. L., 2015, Physical Hydrology

Cited by

Metadata

Authors
Schneider, A., Jost, A., Coulon, C., Silvestre, M., Théry, S., & Ducharne, A.
Affiliations
Centre National de la Recherche Scientifique (FR)
Corresponding author
Schneider, A.
Venue
Geophysical Research Letters
Published
2017-03-21
Volume
44
Issue
6
Pages
2773-2781
ISSN
  • 0094-8276 (ISSN-L)
  • 1944-8007
OpenAlex
W2805938370
Cited by
110
References
41
Citations by year
2026: 7 · 2025: 9 · 2024: 9 · 2023: 9 · 2022: 23 · 2021: 21 · 2020: 12 · 2019: 10 · 2018: 9 · 2017: 1
FWCI
4.19
Type
journal article
Language
en
Open access
Locations
  • Geophysical Research Letters (publishedVersion)
  • HAL (submittedVersion, other-oa)
Funders
  • European Institute of Innovation and Technology
  • Agence Nationale de la Recherche (ANR-14-CE01-00181-01)
Topics
  • Physical Sciences › Environmental Science › Water Science and Technology › Hydrology and Watershed Management Studies
  • Physical Sciences › Environmental Science › Environmental Engineering › Groundwater and Watershed Analysis
  • Physical Sciences › Environmental Science › Environmental Engineering › Soil Moisture and Remote Sensing
Keywords
river network extraction, drainage density, lithology, climate, slope, intermittency, HydroSHEDS, digital elevation model, groundwater flow, surface flow

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-11-18, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-18 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.