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

The DOI matches its Crossref record; 72 of the 73 references it cites were found in a registry or an index, and 1 are not checked yet; none of the 54 sources its claims cite is checked for support yet.

Multivariable Integrated Evaluation of Hydrodynamic Modeling: A Comparison of Performance Considering Different Baseline Topography Data

Open access
Modi, Prakat, Revel, Menaka, Yamazaki, Dai · 2022 · Water Resources Research

Published version of a preprint

Abstract

Continental‐scale river hydrodynamic modeling helps to understand the global hydrological cycle, and model evaluation is essential for calibration and performance assessment. Although many models have been evaluated using several variables separately, methods for the integrated multivariable assessment have yet to be established. We proposed an integrated multivariable evaluation method and an overall basin skill score (OSK) metric, which integrates the sub‐basin skill score, calculated by combining multiple variables' normalized Nash‐Sutcliffe efficiency values. The sub‐basin approach and OSK assess the model performance at the sub‐basin and basin scales and overcome the observation‐related limitations of different spatial dimensions. Evaluation of the CaMa‐Flood global river model for the Amazon Basin considering three variables (discharge, water surface elevation, and flooded area) and two baseline topography data (Shuttle Radar Topography Mission (SRTM) and Multi‐Error‐Removed Improved‐Terrain (MERIT) digital elevation models (DEMs)), shows that the proposed methodology can overcome the limitation of single‐variable evaluation in identifying multiple best parameter sets due to low sensitivity. CaMa‐Flood considering MERIT DEM achieves a single optimal parameter set with a maximum OSK of 0.57 compared with 0.52 for the SRTM DEM, and it shows the importance of accuracy of topography data for improvement in the performance of river hydrodynamic modeling. The proposed multivariable integrated evaluation with better topography data proved to help reduce equifinality and estimate the best parameter by maintaining the physical relationships among variables.

Checked 2026-11-19 · 1 reference not checked yet

Overview

CoChat Summary

TL;DR

The paper proposes an integrated multivariable evaluation method and overall basin skill score (OSK) that combines normalized Nash-Sutcliffe efficiency for discharge, water surface elevation, and flooded area across sub-basins to evaluate the CaMa-Flood hydrodynamic model in the Amazon Basin. Using MERIT DEM instead of SRTM DEM yields a single optimal parameter set with a higher maximum OSK (0.57 vs. 0.52), indicating better topography reduces equifinality and improves overall performance.

Key findings

  • Single-variable evaluation could not identify one optimal parameter set: discharge (Q) contours had no peaks, water surface elevation (WSE) and flooded area (FA) performed well over wide parameter ranges, and best-parameter locations differed among variables.
  • Multivariable OSK with equal weights gave maximum OSK 0.52 for SRTM and 0.57 for MERIT; best parameters were SRTM a = 0.0087, b = 0.70 (45 m river-mouth depth) and MERIT a = 0.144, b = 0.45 (35 m). SRTM had similar optimal performance at two zones (35 and 45 m), whereas MERIT had one optimal zone at 35 m.
  • MERIT improved OSK by ∼10% over SRTM; average NNSE for Q and WSE increased by 3% and 66%, while FA decreased by 6%. WSE improved notably (e.g., sub-basin 18 NNSE 0.15 SRTM vs. 0.82 MERIT; sub-basin 3 0.07 vs. 0.48), while Q was similar (sub-basin 18 0.84 both; sub-basin 3 0.23 vs. 0.24).
  • The method was robust to FA buffer (10–70 km), Manning’s coefficient (0.02–0.04), and adding flood extent; best depth parameters remained unchanged. With weights 30% Q, 30% WSE, 40% FA, maximum OSK was 0.49 at a = 0.349, b = 0.35 (25 m depth) for MERIT; FA NNSE increased from 0.44 to 0.47 while Q and WSE decreased from 0.74 to 0.72 and from 0.53 to 0.41.

Methods

  • The study evaluated the CaMa-Flood global river routing model at 0.1° (∼10 km) resolution with a 10-min time step and local inertial equation, forced by WRR2/HTESSEL runoff (0.25°, daily) over the Amazon Basin (∼7,050,000 km²; annual average discharge 200,000 m³/s at the river mouth).
  • Two DEMs were compared: SRTM (3 arc-sec, ∼90 m, 6 m mean absolute vertical error) and MERIT (3 arc-sec, ∼90 m, multiple error components removed). Observations included 418 daily discharge stations (401 ANA, 17 HYBAM) for 2001–2009; ENVISAT altimetry WSE for 2002–2010; and GIEMS-2 monthly flooded area for 1992–2008 at ∼25 km, downscaled to 0.1°.
  • The basin was divided into 25 sub-basins (threshold 1.5 × 10⁵ km²); Q and WSE were evaluated along the mainstream, and FA within a 50-km buffer on each side. Skill score used NNSE = 1/(2 − NSE); OSK averaged weighted sub-basin NNSE across variables. Depth was varied via the power law H = aQ^b, with b from 0.35 to 0.85 at 0.05 intervals and river-mouth effective depth 25–55 m; Manning’s coefficient was initially 0.03, then tested at 0.02 and 0.04.

Limitations

  • Stated: uncertainty consideration was simplified by perturbing mainly river channel depth and Manning’s coefficient for the best parameter set; multiple error sources could be considered simultaneously in future work.
  • Stated: weights did not account for sub-basin sizes or corresponding river sizes; only SRTM and MERIT DEMs were compared, though other DEMs could be used if hydrography data are available; OSK can be affected by the number of sub-basins (sensitivity test with 15, 25, and 35 sub-basins showed maximum and range of OSK decreased with more sub-basins, but best depth parameters remained unchanged).
  • Stated: FA performance was limited by coarse GIEMS-2 spatial (∼25 km) and monthly temporal resolution, larger uncertainty than directly measured Q and WSE, and the simple nearest-neighbor downscaling; OSK improvement was within 10% due to sub-basin observation limitations. Evident: the evaluation uses one basin and one model, so generalizability to other basins or models is not demonstrated in the text.

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

Integrity signals

Registry

Crossref · 10.1029/2021wr031819

Cites

72 of 73 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Study framework showing the flowchart of calculation of the overall basin skill score from sub-basin skill scores calculated using the normalized Nash-Sutcliffe efficiency values associated with discharge ( Q ), water surface elevation, and flooded area.
Figure 1 Study framework showing the flowchart of calculation of the overall basin skill score from sub-basin skill scores calculated using the normalized Nash-Sutcliffe efficiency values associated with discharge ( Q ), water surface elevation, and flooded area.CC-BY-NC · From the PDF
Image 2
CC-BY-NC · From the PDF
Figure 2 Biased and corrected digital elevation models showing the effect of tree bias on discharge ( Q ), water surface elevation (WSE), and flood extent estimations.
Figure 2 Biased and corrected digital elevation models showing the effect of tree bias on discharge ( Q ), water surface elevation (WSE), and flood extent estimations.CC-BY-NC · From the PDF
Image 4
CC-BY-NC · From the PDF
Figure 3 River depth (H) versus discharge ( Q ) plot representing the power law equation ( H = a Q b ). Lines were drawn considering the depth at the river mouth (H; same color represents the same depth at the river mouth), and the gradient along the river was changed by adjusting the range of exponent ' b ' and coefficient ' a ' accordingly.
Figure 3 River depth (H) versus discharge ( Q ) plot representing the power law equation ( H = a Q b ). Lines were drawn considering the depth at the river mouth (H; same color represents the same depth at the river mouth), and the gradient along the river was changed by adjusting the range of exponent ' b ' and coefficient ' a ' accordingly.CC-BY-NC · From the PDF
Image 6
CC-BY-NC · From the PDF
Image 7
CC-BY-NC · From the PDF
Figure 4 (a) Map of the Amazon Basin with discharge ( Q ), water surface elevation (WSE), and flood extent available observation locations (circles, triangles, and shaded areas, respectively). (b) The zoomed-in area shows sub-basin 13, with red circles and yellow triangles indicating the considered observations of Q and WSE, respectively, along the sub-basin main river channel and the pink-shaded area indicating the zone (50-km on each side of the main river) considered for evaluation of model using the flooded area criterion. (c) Decomposed sub-basins with identification numbers.
Figure 4 (a) Map of the Amazon Basin with discharge ( Q ), water surface elevation (WSE), and flood extent available observation locations (circles, triangles, and shaded areas, respectively). (b) The zoomed-in area shows sub-basin 13, with red circles and yellow triangles indicating the considered observations of Q and WSE, respectively, along the sub-basin main river channel and the pink-shaded area indicating the zone (50-km on each side of the main river) considered for evaluation of model using the flooded area criterion. (c) Decomposed sub-basins with identification numbers.CC-BY-NC · From the PDF
Image 9
CC-BY-NC · From the PDF
Figure 6 Overall basin skill score contour plot with equal weighting of all three variables (discharge ( Q ), water surface elevation, and flooded area) for (a) the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) and (b) the Multi-Error-Removed Improved-Terrain (MERIT) DEM.
Figure 6 Overall basin skill score contour plot with equal weighting of all three variables (discharge ( Q ), water surface elevation, and flooded area) for (a) the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) and (b) the Multi-Error-Removed Improved-Terrain (MERIT) DEM.CC-BY-NC · From the PDF
Image 11
CC-BY-NC · From the PDF
Figure 7 Sub-basin skill scores for (a) discharge ( Q ), (b) water surface elevation (WSE), and (c) flooded area (FA) (or flood extent) with the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) and (d) Q , (e) WSE, and (f) FA with the Multi-Error-Removed Improved-Terrain (MERIT) DEM, and skill scores representing combined multiple variables with the (g) SRTM and (h) MERIT DEMs for the best set of parameters. The gray color indicates sub-basins for which observations of the variables were unavailable.
Figure 7 Sub-basin skill scores for (a) discharge ( Q ), (b) water surface elevation (WSE), and (c) flooded area (FA) (or flood extent) with the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) and (d) Q , (e) WSE, and (f) FA with the Multi-Error-Removed Improved-Terrain (MERIT) DEM, and skill scores representing combined multiple variables with the (g) SRTM and (h) MERIT DEMs for the best set of parameters. The gray color indicates sub-basins for which observations of the variables were unavailable.CC-BY-NC · From the PDF
Figure 8 Time series plots of simulated and observed discharge ( Q ) for (a) sub-basin 18 and (b) sub-basin 3, simulated and observed water surface elevation (WSE) for (c) sub-basin 18 and (d) sub-basin 3, and simulated and observed flooded area (FA) for (e) sub-basin 18 and (f) sub-basin 3. Values in the bracket show the latitude and longitude of the observation location and normalized Nash-Sutcliffe efficiency values considering Shuttle Radar Topography Mission (SRTM) and Multi-ErrorRemoved Improved-Terrain (MERIT) DEMs, respectively. Refer Figure 4c for the sub-basin locations.
Figure 8 Time series plots of simulated and observed discharge ( Q ) for (a) sub-basin 18 and (b) sub-basin 3, simulated and observed water surface elevation (WSE) for (c) sub-basin 18 and (d) sub-basin 3, and simulated and observed flooded area (FA) for (e) sub-basin 18 and (f) sub-basin 3. Values in the bracket show the latitude and longitude of the observation location and normalized Nash-Sutcliffe efficiency values considering Shuttle Radar Topography Mission (SRTM) and Multi-ErrorRemoved Improved-Terrain (MERIT) DEMs, respectively. Refer Figure 4c for the sub-basin locations.CC-BY-NC · From the PDF
Image 14
CC-BY-NC · From the PDF

Verification

Claims the paper makes

0 Valid · 1 Not valid · 34 Not checked · 80 Own results and interpretations

42 more claims

Create a free account to see all 115 claims and to verify the 46 references still open.

See all claims — free account
  • It may state two separate claims as one. 46
  • Probably not a substantial claim. 28

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

Research works printed but not linked: 4

  1. Fahrland E., 2020, Copernicus digital elevation model—Product handbook
  2. Hess L. L., 2015, LBA‐ECO LC‐07 JERS‐1 SAR Flooded Wetlands and Vegetation, Amazon Basin: 1995‐1996
  3. 10.5194/hess-2018-225 · doi:10.5194/hess-2018-225
  4. 10.5194/nhess-2020-272 · doi:10.5194/nhess-2020-272

Cited by

See all 32

Metadata

Authors
Modi, P., Revel, M., & Yamazaki, D.
Affiliations
  • The University of Tokyo (JP)
  • Institute of Industrial Science, The University of Tokyo (JP)
Corresponding author
Modi, Prakat
Venue
Water Resources Research
Published
2022-08
Volume
58
Issue
8
ISSN
  • 0043-1397 (ISSN-L)
  • 1944-7973
OpenAlex
W4291583371
Cited by
32
References
72
Citations by year
2026: 8 · 2025: 7 · 2024: 8 · 2023: 9
FWCI
1.71
Type
journal article
Language
en
Open access
Locations
Water Resources Research (publishedVersion, cc-by-nc)
Funders
Japan Society for the Promotion of Science (20H02251, 16H06291)
Topics
  • Physical Sciences › Environmental Science › Water Science and Technology › Hydrology and Watershed Management Studies
  • Physical Sciences › Environmental Science › Global and Planetary Change › Flood Risk Assessment and Management
  • Physical Sciences › Environmental Science › Environmental Engineering › Soil Moisture and Remote Sensing
Keywords
model evaluation, Nash-Sutcliffe efficiency, CaMa-Flood, Amazon basin, digital elevation model, SRTM, MERIT DEM, equifinality, parameter sensitivity, discharge, water level, flood inundation area

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

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.