Articles | Volume 28, issue 15
https://doi.org/10.5194/hess-28-3665-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/hess-28-3665-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Technical note
| 
13 Aug 2024
Technical note |
| 13 Aug 2024
| 13 Aug 2024
Technical Note: The divide and measure nonconformity – how metrics can mislead when we evaluate on different data partitions
Department of Compound Environmental Risks, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany
Martin Gauch
Google Research, Zurich, Switzerland
Frederik Kratzert
Google Research, Vienna, Austria
Grey Nearing
Google Research, Mountain View, California, USA
Jakob Zscheischler
Department of Compound Environmental Risks, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany
Department of Hydro Sciences, TUD Dresden University of Technology, Dresden, Germany
Viewed
Total article views: 4,283 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 29 Feb 2024)
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 3,233 | 896 | 154 | 4,283 | 195 | 207 | 199 |
- HTML: 3,233
- PDF: 896
- XML: 154
- Total: 4,283
- Supplement: 195
- BibTeX: 207
- EndNote: 199
Total article views: 3,033 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 13 Aug 2024)
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 2,442 | 490 | 101 | 3,033 | 195 | 152 | 143 |
- HTML: 2,442
- PDF: 490
- XML: 101
- Total: 3,033
- Supplement: 195
- BibTeX: 152
- EndNote: 143
Total article views: 1,250 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 29 Feb 2024)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 791 | 406 | 53 | 1,250 | 55 | 56 |
- HTML: 791
- PDF: 406
- XML: 53
- Total: 1,250
- BibTeX: 55
- EndNote: 56
Viewed (geographical distribution)
Total article views: 4,283 (including HTML, PDF, and XML)
Thereof 4,024 with geography defined
and 259 with unknown origin.
Total article views: 3,033 (including HTML, PDF, and XML)
Thereof 2,790 with geography defined
and 243 with unknown origin.
Total article views: 1,250 (including HTML, PDF, and XML)
Thereof 1,234 with geography defined
and 16 with unknown origin.
| Country | # | Views | % |
|---|
| Country | # | Views | % |
|---|
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
Cited
16 citations as recorded by crossref.
- A variational approach at uncertainty estimation in data-driven rainfall-runoff modeling M. Álvarez Chaves et al. https://doi.org/10.1088/3049-4753/ae89ba
- Ensemble agroecosystem modeling enhances predictions of crop yields and soil carbon across the United States S. Gautam et al. https://doi.org/10.5194/soil-12-821-2026
- EARLS: a runoff reconstruction dataset for Europe D. Klotz et al. https://doi.org/10.5194/essd-18-5485-2026
- Setting expectations for hydrologic model performance with an ensemble of simple benchmarks W. Knoben https://doi.org/10.1002/hyp.15288
- Physics-informed gated transformer for robust flash flood forecasting in data-scarce small catchments via variance-constrained learning S. Wei & X. Wang https://doi.org/10.1016/j.jhydrol.2026.135706
- Technical note: Benchmarking large-domain model performance under sampling uncertainty G. Gründemann et al. https://doi.org/10.5194/hess-30-3439-2026
- Investigation and modeling of land use effects on water quality in two NYC water supply streams R. Mukundan et al. https://doi.org/10.1016/j.jenvman.2024.123993
- CAMELS-FI: hydrometeorological time series and landscape properties for 320 catchments in Finland I. Seppä et al. https://doi.org/10.5194/essd-18-4745-2026
- Comparing multi-model mosaic and multi-model combination methods to simulate streamflow across the contiguous USA C. Thébault et al. https://doi.org/10.5194/hess-30-3945-2026
- Spatially-varying parametrization of the Total Runoff Integrating Pathways (TRIP) scheme for improved river routing at the global scale A. Tsilimigkras et al. https://doi.org/10.1016/j.jhydrol.2025.133477
- Hydrological Model Calibration in Data-Scarce Mediterranean Catchments: A Comparative Assessment of Three Strategies A. Jahanshahi et al. https://doi.org/10.3390/hydrology13020066
- Introducing the Model Fidelity Metric (MFM) for robust and diagnostic land surface model evaluation Z. Wu et al. https://doi.org/10.5194/hess-30-2651-2026
- Rethinking Evaluation Metrics in Hydrological Deep Learning: Insights from Torrent Flow Velocity Prediction W. Chen et al. https://doi.org/10.3390/su17198658
- HydroGRAF: Hybrid discharge reconstruction and basin-aware streamflow forecasting in the Himalayas A. Gul et al. https://doi.org/10.1016/j.jhydrol.2026.135479
- Comment on Williams (2025): “Friends don't let friends use NSE or KGE for hydrologic model accuracy evaluation: A rant with data and suggestions for better practice” M. Clark et al. https://doi.org/10.1016/j.envsoft.2026.106869
- Spatially resolved rainfall streamflow modeling in central Europe M. Vischer et al. https://doi.org/10.5194/hess-29-5233-2025
16 citations as recorded by crossref.
- A variational approach at uncertainty estimation in data-driven rainfall-runoff modeling M. Álvarez Chaves et al. https://doi.org/10.1088/3049-4753/ae89ba
- Ensemble agroecosystem modeling enhances predictions of crop yields and soil carbon across the United States S. Gautam et al. https://doi.org/10.5194/soil-12-821-2026
- EARLS: a runoff reconstruction dataset for Europe D. Klotz et al. https://doi.org/10.5194/essd-18-5485-2026
- Setting expectations for hydrologic model performance with an ensemble of simple benchmarks W. Knoben https://doi.org/10.1002/hyp.15288
- Physics-informed gated transformer for robust flash flood forecasting in data-scarce small catchments via variance-constrained learning S. Wei & X. Wang https://doi.org/10.1016/j.jhydrol.2026.135706
- Technical note: Benchmarking large-domain model performance under sampling uncertainty G. Gründemann et al. https://doi.org/10.5194/hess-30-3439-2026
- Investigation and modeling of land use effects on water quality in two NYC water supply streams R. Mukundan et al. https://doi.org/10.1016/j.jenvman.2024.123993
- CAMELS-FI: hydrometeorological time series and landscape properties for 320 catchments in Finland I. Seppä et al. https://doi.org/10.5194/essd-18-4745-2026
- Comparing multi-model mosaic and multi-model combination methods to simulate streamflow across the contiguous USA C. Thébault et al. https://doi.org/10.5194/hess-30-3945-2026
- Spatially-varying parametrization of the Total Runoff Integrating Pathways (TRIP) scheme for improved river routing at the global scale A. Tsilimigkras et al. https://doi.org/10.1016/j.jhydrol.2025.133477
- Hydrological Model Calibration in Data-Scarce Mediterranean Catchments: A Comparative Assessment of Three Strategies A. Jahanshahi et al. https://doi.org/10.3390/hydrology13020066
- Introducing the Model Fidelity Metric (MFM) for robust and diagnostic land surface model evaluation Z. Wu et al. https://doi.org/10.5194/hess-30-2651-2026
- Rethinking Evaluation Metrics in Hydrological Deep Learning: Insights from Torrent Flow Velocity Prediction W. Chen et al. https://doi.org/10.3390/su17198658
- HydroGRAF: Hybrid discharge reconstruction and basin-aware streamflow forecasting in the Himalayas A. Gul et al. https://doi.org/10.1016/j.jhydrol.2026.135479
- Comment on Williams (2025): “Friends don't let friends use NSE or KGE for hydrologic model accuracy evaluation: A rant with data and suggestions for better practice” M. Clark et al. https://doi.org/10.1016/j.envsoft.2026.106869
- Spatially resolved rainfall streamflow modeling in central Europe M. Vischer et al. https://doi.org/10.5194/hess-29-5233-2025
Saved (final revised paper)
Latest update: 15 Sep 2026
Short summary
The evaluation of model performance is essential for hydrological modeling. Using performance criteria requires a deep understanding of their properties. We focus on a counterintuitive aspect of the Nash–Sutcliffe efficiency (NSE) and show that if we divide the data into multiple parts, the overall performance can be higher than all the evaluations of the subsets. Although this follows from the definition of the NSE, the resulting behavior can have unintended consequences in practice.
The evaluation of model performance is essential for hydrological modeling. Using performance...
