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

The DOI matches its Crossref record; 70 of the 71 references it cites were found in a registry or an index; none of the 16 sources its claims cite is checked for support yet.

SABER: A Model-Agnostic Postprocessor for Bias Correcting Discharge from Large Hydrologic Models

Open access

Abstract

Hydrologic modeling is trending toward larger spatial and temporal domains, higher resolutions, and less extensive local calibration and validation. Thorough calibration and validation are difficult because the quantity of observations needed for such scales do not exist or is inaccessible to modelers. We present the Stream Analysis for Bias Estimation and Reduction (SABER) method for bias correction targeting large models. SABER is intended for model consumers to apply to a subset of a larger domain at gauged and ungauged locations and address issues with data size and availability. SABER extends frequency-matching postprocessing techniques using flow duration curves (FDC) at gauged subbasins to be applied at ungauged subbasins using clustering and spatial analysis. SABER uses a “scalar” FDC (SFDC), a ratio of simulated to observed FDC, to characterize biases spatially, temporally, and for varying exceedance probabilities to make corrections at ungauged subbasins. Biased flows at ungauged locations are corrected with the scalar values from the SFDC. Corrected flows are refined to fit a Gumbel Type 1 distribution. We present the theory, procedure, and validation study in Colombia. SABER reduces biases and improves composite metrics, including Nash Sutcliffe and Kling Gupta Efficiency. Recommendations for future work and a discussion of limitations are provided.

Overview

CoChat Summary

TL;DR

SABER is a model-agnostic postprocessor that extends frequency-matching bias correction to ungauged subbasins using scalar flow duration curves, clustering, and spatial/network analysis. In a Magdalena River case study with GEOGloWS ECMWF Streamflow hindcasts, it reduced median bias and improved KGE/NSE, though some locations worsened and several limitations remain.

Key findings

  • In the Magdalena River validation set (n = 109 reserved gauge locations), median mean error fell from 23.61 m³/s for raw GES results to 2.36 m³/s with SFDC corrections and 1.16 m³/s with direct frequency matching; median MAPE fell from 258.62% to 92.40% and 88.40%, respectively.
  • Median KGE improved from −0.66 (raw) to 0.04 (SFDC corrected) and 0.16 (frequency matched); median NSE improved from −8.22 to −0.92 and −0.82.
  • The raw-vs-SFDC mean-error t-test had p ≈ 0.12, not statistically significant at the 0.05 threshold, but the paper reports reduced variance and narrower interquartile range after correction, with about 20 outliers skewing statistics.
  • The optimal number of clusters for the Magdalena watershed was four; 1164 of 2326 subbasins were stream order 1 with average drainage area 153 km², which the paper links to large percent errors. Some validation points decreased in KGE, especially near the outlet and in the south-central area.

Methods

  • SABER uses frequency matching with flow duration curves at gauged subbasins and a scalar flow duration curve (SFDC), the ratio of simulated to observed FDC, applied monthly to ungauged subbasins. Ungauged subbasins are paired to gauges using hydraulic connectivity, FDC clusters from k-means with dynamic time warping and the knee method, stream order, drainage area, and proximity; corrected flows are compared to a Gumbel Type 1 distribution, with flows of 100-year return period or larger replaced by the 100-year Gumbel estimate.
  • Case study: GEOGloWS ECMWF Streamflow (GES) hindcast for the Magdalena River Basin, Colombia, using 233 IDEAM gauges, 2326 subbasins, and daily discharge from 1 January 1980 through 31 December 2021. Approximately half of gauge locations were randomly reserved for validation; metrics included ME, MAPE, MAE, NRMSE, NSE, and KGE for raw, SFDC-corrected, and frequency-matched conditions.

Limitations

  • The paper states: monthly temporal grouping may be too fine or coarse; clustering may not scale to very large models such as the National Water Model retrospective dataset (~17 TB); minimum data requirements and sensitivity to data amount are not quantified; SABER does not enforce conservation of mass between subbasins; clustering and pairing depend heavily on spatial relationships and only indirectly account for land use, soil, or slope; and SABER does not specially treat subbasins with dams, diversions, reservoirs, or other flow controls.
  • Evident from the text: the raw-vs-SFDC mean-error difference was not statistically significant (p ≈ 0.12) due to outliers; MAPE remained about 90% after correction and NSE remained negative; and some validation locations decreased in KGE, including near the watershed outlet and in the south-central area.

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

Integrity signals

Registry

Crossref · 10.3390/hydrology9070113

Cites

70 of 70 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Image 1
CC-BY · From the PDF
Figure 2 Sample calculation of the scalar flow duration curve using observed and simulated flow duration curves from the same location.
Figure 2 Sample calculation of the scalar flow duration curve using observed and simulated flow duration curves from the same location.CC-BY · From the PDF
Figure 3 Simulated flow duration curves clustered in six groups using k-means and dynamic time warping measures. Cluster centroids are shown in red, and individual flow duration curves are shown in black.
Figure 3 Simulated flow duration curves clustered in six groups using k-means and dynamic time warping measures. Cluster centroids are shown in red, and individual flow duration curves are shown in black.CC-BY · From the PDF
Figure 4 Plot of k-means cluster residuals (error) vs. the number of clusters with the knee point labeled.
Figure 4 Plot of k-means cluster residuals (error) vs. the number of clusters with the knee point labeled.CC-BY · From the PDF
Figure 5 Magdalena River basin from the GES model's stream network with IDEAM gauge locations shown.
Figure 5 Magdalena River basin from the GES model's stream network with IDEAM gauge locations shown.CC-BY · From the PDF
Figure 6 Box plots of mean error in the Magdalena River basin for each test group (outliers above 400 m 3 /s excluded).
Figure 6 Box plots of mean error in the Magdalena River basin for each test group (outliers above 400 m 3 /s excluded).CC-BY · From the PDF
Figure 7 Box plots of MAPE results in the Magdalena River basin for each test group.
Figure 7 Box plots of MAPE results in the Magdalena River basin for each test group.CC-BY · From the PDF
Figure 8 Box plots of KGE results in the Magdalena River basin for each test group. A horizontal red line indicates a KGE of -0.41, which corresponds to the KGE of a constant mean observed value simulation.
Figure 8 Box plots of KGE results in the Magdalena River basin for each test group. A horizontal red line indicates a KGE of -0.41, which corresponds to the KGE of a constant mean observed value simulation.CC-BY · From the PDF
Figure 9 Mapof subbasins for the GEOGloWS ECMWF Streamflow model's stream segments colored in groups that share the same assignment justification.
Figure 9 Mapof subbasins for the GEOGloWS ECMWF Streamflow model's stream segments colored in groups that share the same assignment justification.CC-BY · From the PDF
Figure 10 Maps of validation gauge locations colored by KGE metric values for uncorrected, SFDCcorrected, and frequency-matched corrected conditions.
Figure 10 Maps of validation gauge locations colored by KGE metric values for uncorrected, SFDCcorrected, and frequency-matched corrected conditions.CC-BY · From the PDF
Image 11
CC-BY · From the PDF

Verification

Claims the paper makes

0 Valid · 3 Not valid · 9 Not checked · 92 Own results and interpretations

49 more claims

Create a free account to see all 104 claims and to verify the 15 references still open.

See all claims — free account
  • It may state two separate claims as one. 26
  • Probably not a substantial claim. 31

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 71 references

Research works printed but not linked: 2

  1. Sanchez, J.L., Nelson, J., Williams, G.P., Hales, R., Ames, D.P., and Jones, N. (2020, January 1–17). A Streamflow Bias Correction and Validation Method for GEOGloWS ECMWF Streamflow Services. Proceedings of the AGU Fall Meeting Abstracts, Virtual.
  2. Hales, R., Sanchez, J.L., Nelson, J., Williams, G.P., Ames, D.P., and Jones, N. (2020, January 1–17). A Post-Processing Method to Calibrate Large-Scale Hydrologic Models with Limited Historical Observation Data Leveraging Machine Learning and Spatial Analysis. Proceedings of the AGU Fall Meeting Abstracts, Virtual.

Cited by

See all 19

Metadata

Authors
Hales, R. C., Sowby, R. B., Williams, G. P., Nelson, E. J., Ames, D. P., Dundas, J. B., & Ogden, J.
Affiliations
Brigham Young University (US)
Corresponding author
Hales, Riley C.
Venue
Hydrology
Publisher
MDPI AG
Published
2022-06-22
Volume
9
Issue
7
Pages
113
ISSN
2306-5338 (ISSN-L)
OpenAlex
W4283317086
Cited by
19
References
71
Citations by year
2026: 6 · 2025: 5 · 2024: 3 · 2023: 4 · 2022: 1
FWCI
0.81
Type
journal article
Language
en
Open access
Locations
  • Hydrology (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • NASA (80NSSC20K0157, 80NSCC18K0440)
  • MDPI (80NSSC20K0157, 80NSCC18K0440)
Topics
  • Physical Sciences › Environmental Science › Water Science and Technology › Hydrology and Watershed Management Studies
  • Physical Sciences › Environmental Science › Management, Monitoring, Policy and Law › Environmental Monitoring and Data Management
  • Physical Sciences › Environmental Science › Environmental Engineering › Soil Moisture and Remote Sensing
Keywords
bias correction, frequency matching, flow duration curve, spatial bias, clustering, SABER, model calibration and validation

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 2026-11-19, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-19 (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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: SABER: A Model-Agnostic Postprocessor for Bias Correcting Discharge from Large Hydrologic Models. https://app.cochat.ai/papers/10.3390/hydrology9070113
The paper
Hales, R. C., Sowby, R. B., Williams, G. P., Nelson, E. J., Ames, D. P., Dundas, J. B., & Ogden, J. (2022). SABER: A Model-Agnostic Postprocessor for Bias Correcting Discharge from Large Hydrologic Models. Hydrology. https://doi.org/10.3390/hydrology9070113

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.