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

The DOI matches its Crossref record; 52 of the 52 references it cites were found in a registry or an index; none of the 41 sources its claims cite is checked for support yet.

ICESat‐2 Based River Surface Slope and Its Impact on Water Level Time Series From Satellite Altimetry

Open access

Abstract

The water surface slope (WSS) of rivers is essential for estimating flow velocity and discharge. It is also helpful as a correction applied to range measurements of satellite altimetry missions to derive water level time series at a virtual station. Using radar altimetry, WSS can only be roughly estimated and is limited to wide rivers because of its coarse spatiotemporal resolution. In contrast, the lidar sensor onboard Ice, Cloud, and Land Elevation Satellite 2 (ICESat‐2) can also observe small rivers. Using ICESat‐2's unique measurement geometry with six parallel laser beams, we derive instantaneous WSS along and across the satellite's ground track, time‐variable WSS (with an average of 5 days of records in the studied epoch between October 2018 and October 2021), and average WSS on reach‐scale. Although the method can be applied globally, this study is limited to 815 reaches in Europe and North America where sufficient validation data is available. We compare the ICESat‐2 WSS with time‐variable WSS derived from multiple gauges and constant data from the “SWOT River Database.” For 89% of the studied reaches, ICESat‐2 can be used to estimate the average WSS with a median absolute error of 23 mm/km. We also show the possible performance gain at multiple virtual stations (VS) in the “Database for Hydrological Time Series of Inland Waters” (https://dahiti.dgfi.tum.de), applying the WSS as a correction for altimetry satellites' ground track variability. We correct 137 VS for the derived ICESat‐2 WSS and yield improvements in the root mean square error by up to 30 cm or 66%.

Overview

CoChat Summary

TL;DR

The paper derives river water surface slope (WSS) from ICESat-2 ATL13 observations using across-track and along-track methods, combined into time-variable and average reach-scale WSS. It validates against gauge pairs and applies the WSS as a correction to radar-altimetry virtual stations, finding average WSS MAE of 23 mm/km and RMSE improvements up to 30 cm or 66%.

Key findings

  • Instantaneous across-track WSS had a median absolute error of 24 mm/km for all beams; along-track WSS had 57 mm/km after angle-dependent outlier rejection, versus 133 mm/km without it. Strong and weak beams showed no notable quality difference, though strong beams gave more estimates and coverage.
  • The combined daily WSS covered 89% of the 815 studied reaches with an MAE of 28 mm/km; the average combined reach-scale WSS had an MAE of 23 mm/km against median in-situ WSS, compared with 71 mm/km for SWORD WSS.
  • Time-variable WSS had median RMSE 49 mm/km and NRMSE 35% for the combined method, but the median correlation coefficient was only 0.35, and the median number of records per reach was 5 over October 2018–October 2021.
  • Applying the average combined WSS to 137 DAHITI virtual stations improved RMSE by up to 30 cm or 66%; the mean ICESat-2 WSS improvement was −2.20 cm RMSE (−8.24% NRMSE), while 64% of studied virtual stations had insignificant RMSE differences between −2 and 2 cm.

Methods

  • Used ICESat-2 ATL13 inland water surface height (`ht_water_surf`) over 815 SWORD reaches in Europe and North America, with in-situ gauge data from BfG, USACE, USGS, Hydroportail, and HIDMET for validation.
  • Derived instantaneous WSS by (a) across-track WSE differences between simultaneous beam intersections and (b) along-track linear fits projected to the river centerline; combined both into daily time-variable WSS by preferring across-track records and filling gaps with along-track records, then took the median as average reach-scale WSS.
  • Validated against time-variable in-situ WSS between gauge pairs and applied the average combined WSS as a correction to DAHITI Jason-2/3 and Sentinel-3A/B water level time series at 137 virtual stations.

Limitations

  • Stated: the study is limited to 815 reaches in Europe and North America where sufficient validation data are available; global applicability is uncertain because denser cloud cover in other climate zones may affect coverage.
  • Stated: time variability cannot be estimated with confidence because the median correlation is only 0.35 and temporal resolution is sparse, with a median of 5 records over the studied epoch; average WSS may be too coarse for local WSS variability or insufficient when temporal WSS variability is large.
  • Stated and evident: validation is limited by the spatial distribution of suitable gauges, epoch overlap, and metadata such as vertical datum, gauge zero, and exact location; pairs with negative slope, mismatching datums, or centerline errors were omitted, and only 137 virtual stations were selected, with many showing insignificant or degraded RMSE after correction.

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

Integrity signals

Registry

Crossref · 10.1029/2022wr032842

Cites

52 of 52 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Regions with studied reaches, in-situ gauges we use for validation, and virtual stations to which we apply the water surface slope correction. Top: Europe. Bottom: North America.
Figure 1 Regions with studied reaches, in-situ gauges we use for validation, and virtual stations to which we apply the water surface slope correction. Top: Europe. Bottom: North America.CC-BY-NC-ND · From the PDF
Image 2
CC-BY-NC-ND · From the PDF
Figure 2 Processing flow and data structure. After preprocessing (Section 3.1) the instantaneous water surface slope (WSS) is estimated using the across-track (blue, Section 3.2) and along-track (orange, Section 3.3) approach. The results are reduced to a time-variable WSS with daily resolution using a weighted average. Both approaches are combined (black, Section 3.4) using the across-track time-variable data supplemented by the along-track data while duplicate along-track data is discarded. An average value is derived for each approach by calculating the median of each time-variable WSS.
Figure 2 Processing flow and data structure. After preprocessing (Section 3.1) the instantaneous water surface slope (WSS) is estimated using the across-track (blue, Section 3.2) and along-track (orange, Section 3.3) approach. The results are reduced to a time-variable WSS with daily resolution using a weighted average. Both approaches are combined (black, Section 3.4) using the across-track time-variable data supplemented by the along-track data while duplicate along-track data is discarded. An average value is derived for each approach by calculating the median of each time-variable WSS.CC-BY-NC-ND · From the PDF
Image 4
CC-BY-NC-ND · From the PDF
Figure 3 Schematic of Ice, Cloud, and Land Elevation Satellite 2 ATL13 observations (circles) of one cycle within a reach area of interest (light-blue polygon), segmented as features (orange) per beam (green) with reference points (squares) at the centerline (dashed blue) intersections or the nearest centerline point. Features with multiple intersections (dashed orange) are split.
Figure 3 Schematic of Ice, Cloud, and Land Elevation Satellite 2 ATL13 observations (circles) of one cycle within a reach area of interest (light-blue polygon), segmented as features (orange) per beam (green) with reference points (squares) at the centerline (dashed blue) intersections or the nearest centerline point. Features with multiple intersections (dashed orange) are split.CC-BY-NC-ND · From the PDF
Figure 4 Schematic of an Ice, Cloud, and Land Elevation Satellite 2 beam ground track 𝐴𝐴 ⃗ 𝑏𝑏 crossing a river (blue polygon) with centerline tangent vector 𝐴𝐴 𝐴 at an angle γ , along-track water surface slope (WSS) (tan β ), and along-river WSS (tan α ).
Figure 4 Schematic of an Ice, Cloud, and Land Elevation Satellite 2 beam ground track 𝐴𝐴 ⃗ 𝑏𝑏 crossing a river (blue polygon) with centerline tangent vector 𝐴𝐴 𝐴 at an angle γ , along-track water surface slope (WSS) (tan β ), and along-river WSS (tan α ).CC-BY-NC-ND · From the PDF
Image 7
CC-BY-NC-ND · From the PDF
Figure 5 Along-Track water surface slope absolute errors by intersection angle and confidence interval with the applied angle-depended outlier threshold (black line). Estimates above the line are rejected.
Figure 5 Along-Track water surface slope absolute errors by intersection angle and confidence interval with the applied angle-depended outlier threshold (black line). Estimates above the line are rejected.CC-BY-NC-ND · From the PDF
Image 9
CC-BY-NC-ND · From the PDF
Figure 6 Left: Along-Track water surface slope (WSS) absolute errors by intersection length. Right: Across-Track WSS absolute errors by reach width. The orange lines show the rolling median within a 200 and 100 m window, respectively. The point color indicates the relative data point density.
Figure 6 Left: Along-Track water surface slope (WSS) absolute errors by intersection length. Right: Across-Track WSS absolute errors by reach width. The orange lines show the rolling median within a 200 and 100 m window, respectively. The point color indicates the relative data point density.CC-BY-NC-ND · From the PDF
Figure 7 Beam GT1R (strong) intersecting reach 22791100061 (Danube) on 29 February 2020. Top: All ATL13 observations, including rejected outliers. Lower-left: Along-track Elevation of the ATL13 observations after outlier rejection with fitted along-track water surface slope (WSS) (89 mm/km, confidence interval: 10 mm/km). Lower-right: Beam GT1R intersecting the river centerline vector at an angle of 34° with ATL13 observations after outlier rejection. The resulting along-river WSS is 107 mm/km.
Figure 7 Beam GT1R (strong) intersecting reach 22791100061 (Danube) on 29 February 2020. Top: All ATL13 observations, including rejected outliers. Lower-left: Along-track Elevation of the ATL13 observations after outlier rejection with fitted along-track water surface slope (WSS) (89 mm/km, confidence interval: 10 mm/km). Lower-right: Beam GT1R intersecting the river centerline vector at an angle of 34° with ATL13 observations after outlier rejection. The resulting along-river WSS is 107 mm/km.CC-BY-NC-ND · From the PDF
Image 12
CC-BY-NC-ND · From the PDF
Figure 8 Lower-right: Beam GT1L (weak) intersecting reach 22791100061 (Danube) on 29 February 2020 at an angle of 32°. Top: ATL13 observations including outliers. Lower-left: water surface slope (WSS) fitted to valid ATL13 data. The resulting along-river WSS of -18 mm/km (confidence interval: 28 mm/km) is rejected.
Figure 8 Lower-right: Beam GT1L (weak) intersecting reach 22791100061 (Danube) on 29 February 2020 at an angle of 32°. Top: ATL13 observations including outliers. Lower-left: water surface slope (WSS) fitted to valid ATL13 data. The resulting along-river WSS of -18 mm/km (confidence interval: 28 mm/km) is rejected.CC-BY-NC-ND · From the PDF
Figure 9 Lower-right: Beam GT2R (strong) intersecting reach 74282100031 (Mississippi) on 12 April 2020 at an angle of 19°. Top: ATL13 observations including outliers. Lower-left: water surface slope (WSS) fitted to valid ATL13 data. The resulting along-river WSS is 41 mm/km (confidence interval: 11 mm/km).
Figure 9 Lower-right: Beam GT2R (strong) intersecting reach 74282100031 (Mississippi) on 12 April 2020 at an angle of 19°. Top: ATL13 observations including outliers. Lower-left: water surface slope (WSS) fitted to valid ATL13 data. The resulting along-river WSS is 41 mm/km (confidence interval: 11 mm/km).CC-BY-NC-ND · From the PDF
Image 15
CC-BY-NC-ND · From the PDF
Figure 10 Water surface slope (WSS) time series for selected reaches of the rivers Elbe, Ohio, Loire, and Missouri. Blue dots show estimates using the combined approach. The in-situ WSS is shown in orange and the errors below. The correlation coefficient and root mean square error are given above each time series.
Figure 10 Water surface slope (WSS) time series for selected reaches of the rivers Elbe, Ohio, Loire, and Missouri. Blue dots show estimates using the combined approach. The in-situ WSS is shown in orange and the errors below. The correlation coefficient and root mean square error are given above each time series.CC-BY-NC-ND · From the PDF
Image 17
CC-BY-NC-ND · From the PDF
Figure 11 Histograms of the median in-situ water surface slope (WSS) per pair of gauges, the combined Ice, Cloud, and Land Elevation Satellite 2 WSS per reach, and the SWOT River Database WSS per reach.
Figure 11 Histograms of the median in-situ water surface slope (WSS) per pair of gauges, the combined Ice, Cloud, and Land Elevation Satellite 2 WSS per reach, and the SWOT River Database WSS per reach.CC-BY-NC-ND · From the PDF
Image 19
CC-BY-NC-ND · From the PDF
Figure 12 Histograms of the absolute (left) and relative (right) root mean square error difference applying the median combined (top), SWOT River Database (SWORD) (center), and median in-situ (bottom) water surface slope as a correction to the original Database for Hydrological Time Series of Inland Waters (DAHITI) water level time series. The in-situ histograms show all studied virtual stations in blue and a subset (orange) between pairs of gauges that are connected judging by SWORD data.
Figure 12 Histograms of the absolute (left) and relative (right) root mean square error difference applying the median combined (top), SWOT River Database (SWORD) (center), and median in-situ (bottom) water surface slope as a correction to the original Database for Hydrological Time Series of Inland Waters (DAHITI) water level time series. The in-situ histograms show all studied virtual stations in blue and a subset (orange) between pairs of gauges that are connected judging by SWORD data.CC-BY-NC-ND · From the PDF
Image 21
CC-BY-NC-ND · From the PDF
Figure 13 Root mean square error difference after applying the water surface slope (WSS)-correction compared to the original Database for Hydrological Time Series of Inland Waters time series for the 137 studied virtual stations. Negative values mean improvements due to the application of the WSS correction. The plots on the right show the regions of smaller WSS for more detail.
Figure 13 Root mean square error difference after applying the water surface slope (WSS)-correction compared to the original Database for Hydrological Time Series of Inland Waters time series for the 137 studied virtual stations. Negative values mean improvements due to the application of the WSS correction. The plots on the right show the regions of smaller WSS for more detail.CC-BY-NC-ND · From the PDF
Image 23
CC-BY-NC-ND · From the PDF
Figure 14 Top: Original Database for Hydrological Time Series of Inland Waters (DAHITI) and water surface slope (WSS)-corrected water level time series for virtual stations (VS) 37118 at the Platte River compared to the in-situ time series provided by U.S. Geological Survey (Station 06770200 near Kearney, Nebraska). Center: Crossing anomaly of the orbit ground tracks w.r.t. the VS nominal location. Bottom: Water level errors with and without the WSS correction applied.
Figure 14 Top: Original Database for Hydrological Time Series of Inland Waters (DAHITI) and water surface slope (WSS)-corrected water level time series for virtual stations (VS) 37118 at the Platte River compared to the in-situ time series provided by U.S. Geological Survey (Station 06770200 near Kearney, Nebraska). Center: Crossing anomaly of the orbit ground tracks w.r.t. the VS nominal location. Bottom: Water level errors with and without the WSS correction applied.CC-BY-NC-ND · From the PDF
Figure 15 Top: Original Database for Hydrological Time Series of Inland Waters (DAHITI) and water surface slope (WSS)-corrected water level time series for virtual stations (VS) 13443 at the Loire River compared to the in-situ time series provided by Hydroportail (Station K418001010 at Gien). Center: Crossing anomaly of the orbit ground tracks w.r.t. the VS nominal location. Bottom: Water level errors with and without the WSS correction applied.
Figure 15 Top: Original Database for Hydrological Time Series of Inland Waters (DAHITI) and water surface slope (WSS)-corrected water level time series for virtual stations (VS) 13443 at the Loire River compared to the in-situ time series provided by Hydroportail (Station K418001010 at Gien). Center: Crossing anomaly of the orbit ground tracks w.r.t. the VS nominal location. Bottom: Water level errors with and without the WSS correction applied.CC-BY-NC-ND · From the PDF
Image 26
CC-BY-NC-ND · From the PDF

Verification

Claims the paper makes

0 Valid · 7 Not valid · 33 Not checked · 122 Own results and interpretations

70 more claims

Create a free account to see all 162 claims and to verify the 33 references still open.

See all claims — free account
  • It may state two separate claims as one. 53
  • Probably not a substantial claim. 51

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

4 more claims about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

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: 3

  1. Jasinski M., 2021, Algorithm theoretical basis document (ATBD) for inland water data products, ATL13, version 5, technical report
  2. Julien P. Y., 2018, River dynamics, chapter 11
  3. Manning R., 1891, On the flow of water in open channels and pipes, Transactions of the Institution of Civil Engineers of Ireland, 20, 161

Cited by

See all 49

Metadata

Authors
Scherer, D., Schwatke, C., Dettmering, D., & Seitz, F.
Affiliations
Technical University of Munich (DE)
Corresponding author
Scherer, Daniel
Venue
Water Resources Research
Published
2022-11
Volume
58
Issue
11
ISSN
  • 0043-1397 (ISSN-L)
  • 1944-7973
OpenAlex
W4308149541
Cited by
49
References
50
Citations by year
2026: 10 · 2025: 15 · 2024: 17 · 2023: 7
FWCI
2.80
Type
journal article
Language
en
Open access
Locations
  • Water Resources Research (publishedVersion, cc-by-nc-nd)
  • Geo-Leo e-docs (Deutsche Initiative für Netzwerkinformation) (submittedVersion, cc-by-nd)
  • mediaTUM (Technical University of Munich) (submittedVersion, cc-by-nc-nd)
Funders
Deutsche Forschungsgemeinschaft (DE 2174/10‐2, 324641997, 2174/10-2)
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 › Earth and Planetary Sciences › Geophysics › Geophysics and Gravity Measurements
Keywords
ICESat-2, spaceborne lidar, water level time series, virtual stations, small rivers, river discharge estimation, flow velocity

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.