Articles | Volume 28, issue 3
https://doi.org/10.5194/hess-28-479-2024
© 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-28-479-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
07 Feb 2024
Research article |
| 07 Feb 2024
| 07 Feb 2024
On the need for physical constraints in deep learning rainfall–runoff projections under climate change: a sensitivity analysis to warming and shifts in potential evapotranspiration
Sungwook Wi
CORRESPONDING AUTHOR
Department of Biological and Environmental Engineering, Cornell University, Ithaca, NY, USA
Scott Steinschneider
Department of Biological and Environmental Engineering, Cornell University, Ithaca, NY, USA
Viewed
Total article views: 6,680 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 09 Aug 2023)
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 5,007 | 1,485 | 188 | 6,680 | 417 | 251 | 305 |
- HTML: 5,007
- PDF: 1,485
- XML: 188
- Total: 6,680
- Supplement: 417
- BibTeX: 251
- EndNote: 305
Total article views: 4,029 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 07 Feb 2024)
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 3,179 | 743 | 107 | 4,029 | 165 | 129 | 154 |
- HTML: 3,179
- PDF: 743
- XML: 107
- Total: 4,029
- Supplement: 165
- BibTeX: 129
- EndNote: 154
Total article views: 2,651 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 09 Aug 2023)
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,828 | 742 | 81 | 2,651 | 252 | 122 | 151 |
- HTML: 1,828
- PDF: 742
- XML: 81
- Total: 2,651
- Supplement: 252
- BibTeX: 122
- EndNote: 151
Viewed (geographical distribution)
Total article views: 6,680 (including HTML, PDF, and XML)
Thereof 6,518 with geography defined
and 162 with unknown origin.
Total article views: 4,029 (including HTML, PDF, and XML)
Thereof 3,906 with geography defined
and 123 with unknown origin.
Total article views: 2,651 (including HTML, PDF, and XML)
Thereof 2,612 with geography defined
and 39 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
39 citations as recorded by crossref.
- An argument for parsimony in differentiable hydrologic models S. Poudel & S. Steinschneider https://doi.org/10.5194/hess-30-5173-2026
- Evaluation of Micrometeorological Models for Estimating Crop Evapotranspiration Using a Smart Field Weighing Lysimeter P. Ratshiedana et al. https://doi.org/10.3390/w17020187
- Deriving phase-contingent dynamic drought-limited water levels: An adaptive framework for managing megadrought evolution Y. Li et al. https://doi.org/10.1016/j.ejrh.2026.103309
- 3D U-Net correction of Northwest Pacific sea surface wind field forecasts incorporating wind–pressure relationship constraints Z. Kuang et al. https://doi.org/10.1016/j.atmosres.2026.108939
- Developing an alternative data-driven model to resemble geomorphologic rainfall-runoff models P. Huang & K. Lee https://doi.org/10.1080/19475705.2025.2516725
- Does MC-LSTM model improve the reliability of streamflow prediction in human-influenced watersheds? G. Sahu et al. https://doi.org/10.1016/j.jhydrol.2025.134711
- Investigating the streamflow simulation capability of a new mass-conserving long short-term memory (MC-LSTM) model across the contiguous United States Y. Wang et al. https://doi.org/10.1016/j.jhydrol.2025.133161
- Advancing streamflow prediction in data-scarce regions through vegetation-constrained distributed hybrid ecohydrological models L. Zhong et al. https://doi.org/10.1016/j.jhydrol.2024.132165
- Improving global soil moisture prediction based on Meta-Learning model leveraging Köppen-Geiger climate classification Q. Li et al. https://doi.org/10.1016/j.catena.2025.108743
- Climate change impact assessment on a German lowland river using long short-term memory and conceptual hydrological models A. Ley et al. https://doi.org/10.1016/j.ejrh.2025.102426
- Research on Wind Field Correction Method Integrating Position Information and Proxy Divergence J. Gan et al. https://doi.org/10.3390/biomimetics10100651
- Investigate the rainfall-runoff relationship and hydrological concepts inside LSTM Y. Hu et al. https://doi.org/10.1016/j.envsoft.2025.106527
- Tackling water table depth modeling via machine learning: From proxy observations to verifiability J. Janssen et al. https://doi.org/10.1016/j.advwatres.2025.104955
- Advancing Infiltration Rate Prediction in Algeria’s Mitidja Plain: A Machine Learning and Empirical Model Comparison A. Mazighi et al. https://doi.org/10.1007/s41748-025-00640-z
- Multi-scale drought analysis and machine learning-based completion of missing streamflow data in the Aras Basin B. Çırağ & C. Bozkurt https://doi.org/10.1016/j.pce.2026.104410
- Uncertainty in estimating the relative change of design floods under climate change: a stylized experiment with process-based, deep learning, and hybrid models S. Poudel et al. https://doi.org/10.1016/j.jhydrol.2025.134427
- Learning groundwater dynamics across heterogeneous aquifers: A clustering-integrated spatiotemporal deep learning framework S. Ho et al. https://doi.org/10.1016/j.ejrh.2026.103770
- Deep learning-based direct and indirect potential evapotranspiration prediction, a case of the Nakdong River basin, South Korea M. Waqas & S. Kim https://doi.org/10.1016/j.ecohyd.2026.100772
- Lake Titicaca water level forecasting using data augmentation and recurrent neural networks A. Flores et al. https://doi.org/10.3389/frwa.2026.1688939
- Machine learning advances and data model coevolution in geoscience A. Eltijnai & M. Mohammed https://doi.org/10.1007/s44288-026-00524-3
- Bayesian Hierarchical Pooling to Reduce Uncertainty in Projected Change of Design Floods S. Poudel & S. Steinschneider https://doi.org/10.1061/JHYEFF.HEENG-6837
- Multi-step ahead streamflow forecasting method using Embedding Multi-Layer Perceptron Y. Li & S. Yang https://doi.org/10.1016/j.ejrh.2026.103349
- Mass Conservative Time-Series GAN for Synthetic Extreme Flood-Event Generation: Impact on Probabilistic Forecasting Models D. Karimanzira https://doi.org/10.3390/stats7030049
- Multimodel analysis of climate-induced flow variability in transboundary tropical basins B. Ernest et al. https://doi.org/10.2166/wcc.2026.407
- A national-scale hybrid model for enhanced streamflow estimation – consolidating a physically based hydrological model with long short-term memory (LSTM) networks J. Liu et al. https://doi.org/10.5194/hess-28-2871-2024
- Applications of artificial intelligence models in seawater intrusion: progress, challenges, and future directions X. Wang et al. https://doi.org/10.1016/j.jhydrol.2026.136086
- Transformer based models with hierarchical graph representations for enhanced climate forecasting T. Ramu et al. https://doi.org/10.1038/s41598-025-07897-4
- An Approach for Future Droughts in Northwest Türkiye: SPI and LSTM Methods E. Taylan https://doi.org/10.3390/su16166905
- Coupling strategies of snowmelt runoff model and machine learning in the Lhasa River Basin T. Wang et al. https://doi.org/10.1016/j.ejrh.2026.103400
- Multi-site deep learning for groundwater level prediction across global datasets: toward scalable applications under data scarcity A. Nolte et al. https://doi.org/10.2166/hydro.2025.095
- Deep learning error post-processing improves stochastic watershed modeling B. Manoli et al. https://doi.org/10.1016/j.jhydrol.2026.135663
- Hybrid models generalize better to warmer climate conditions than process-based and purely data-driven models J. Bohl et al. https://doi.org/10.5194/hess-30-4667-2026
- The outcome prediction method of football matches by the quantum neural network based on deep learning Y. Sun & H. Chu https://doi.org/10.1038/s41598-025-91870-8
- Pooling local climate and donor gauges with deep learning for improved reconstructions of streamflow in ungauged and partially gauged basins S. Wi et al. https://doi.org/10.1016/j.jhydrol.2025.133764
- Predicting Forest Evapotranspiration Shifts Under Diverse Climate Change Scenarios by Leveraging the SEBAL Model Across Inner Mongolia P. Ji et al. https://doi.org/10.3390/f15122234
- Distinct hydrologic response patterns and trends worldwide revealed by physics-embedded learning H. Ji et al. https://doi.org/10.1038/s41467-025-64367-1
- A hybrid TFN–LSTM model for groundwater level forecasting in Rhode Island, USA H. Chu et al. https://doi.org/10.1016/j.jenvman.2026.130474
- Beyond prediction accuracy: using differentiable hybrid models as a tool for hydrological knowledge discovery Y. Hu et al. https://doi.org/10.1016/j.jhydrol.2026.136147
- Associations between deep learning runoff predictions and hydrogeological conditions in Australia S. Clark & J. Jaffrés https://doi.org/10.1016/j.jhydrol.2024.132569
39 citations as recorded by crossref.
- An argument for parsimony in differentiable hydrologic models S. Poudel & S. Steinschneider https://doi.org/10.5194/hess-30-5173-2026
- Evaluation of Micrometeorological Models for Estimating Crop Evapotranspiration Using a Smart Field Weighing Lysimeter P. Ratshiedana et al. https://doi.org/10.3390/w17020187
- Deriving phase-contingent dynamic drought-limited water levels: An adaptive framework for managing megadrought evolution Y. Li et al. https://doi.org/10.1016/j.ejrh.2026.103309
- 3D U-Net correction of Northwest Pacific sea surface wind field forecasts incorporating wind–pressure relationship constraints Z. Kuang et al. https://doi.org/10.1016/j.atmosres.2026.108939
- Developing an alternative data-driven model to resemble geomorphologic rainfall-runoff models P. Huang & K. Lee https://doi.org/10.1080/19475705.2025.2516725
- Does MC-LSTM model improve the reliability of streamflow prediction in human-influenced watersheds? G. Sahu et al. https://doi.org/10.1016/j.jhydrol.2025.134711
- Investigating the streamflow simulation capability of a new mass-conserving long short-term memory (MC-LSTM) model across the contiguous United States Y. Wang et al. https://doi.org/10.1016/j.jhydrol.2025.133161
- Advancing streamflow prediction in data-scarce regions through vegetation-constrained distributed hybrid ecohydrological models L. Zhong et al. https://doi.org/10.1016/j.jhydrol.2024.132165
- Improving global soil moisture prediction based on Meta-Learning model leveraging Köppen-Geiger climate classification Q. Li et al. https://doi.org/10.1016/j.catena.2025.108743
- Climate change impact assessment on a German lowland river using long short-term memory and conceptual hydrological models A. Ley et al. https://doi.org/10.1016/j.ejrh.2025.102426
- Research on Wind Field Correction Method Integrating Position Information and Proxy Divergence J. Gan et al. https://doi.org/10.3390/biomimetics10100651
- Investigate the rainfall-runoff relationship and hydrological concepts inside LSTM Y. Hu et al. https://doi.org/10.1016/j.envsoft.2025.106527
- Tackling water table depth modeling via machine learning: From proxy observations to verifiability J. Janssen et al. https://doi.org/10.1016/j.advwatres.2025.104955
- Advancing Infiltration Rate Prediction in Algeria’s Mitidja Plain: A Machine Learning and Empirical Model Comparison A. Mazighi et al. https://doi.org/10.1007/s41748-025-00640-z
- Multi-scale drought analysis and machine learning-based completion of missing streamflow data in the Aras Basin B. Çırağ & C. Bozkurt https://doi.org/10.1016/j.pce.2026.104410
- Uncertainty in estimating the relative change of design floods under climate change: a stylized experiment with process-based, deep learning, and hybrid models S. Poudel et al. https://doi.org/10.1016/j.jhydrol.2025.134427
- Learning groundwater dynamics across heterogeneous aquifers: A clustering-integrated spatiotemporal deep learning framework S. Ho et al. https://doi.org/10.1016/j.ejrh.2026.103770
- Deep learning-based direct and indirect potential evapotranspiration prediction, a case of the Nakdong River basin, South Korea M. Waqas & S. Kim https://doi.org/10.1016/j.ecohyd.2026.100772
- Lake Titicaca water level forecasting using data augmentation and recurrent neural networks A. Flores et al. https://doi.org/10.3389/frwa.2026.1688939
- Machine learning advances and data model coevolution in geoscience A. Eltijnai & M. Mohammed https://doi.org/10.1007/s44288-026-00524-3
- Bayesian Hierarchical Pooling to Reduce Uncertainty in Projected Change of Design Floods S. Poudel & S. Steinschneider https://doi.org/10.1061/JHYEFF.HEENG-6837
- Multi-step ahead streamflow forecasting method using Embedding Multi-Layer Perceptron Y. Li & S. Yang https://doi.org/10.1016/j.ejrh.2026.103349
- Mass Conservative Time-Series GAN for Synthetic Extreme Flood-Event Generation: Impact on Probabilistic Forecasting Models D. Karimanzira https://doi.org/10.3390/stats7030049
- Multimodel analysis of climate-induced flow variability in transboundary tropical basins B. Ernest et al. https://doi.org/10.2166/wcc.2026.407
- A national-scale hybrid model for enhanced streamflow estimation – consolidating a physically based hydrological model with long short-term memory (LSTM) networks J. Liu et al. https://doi.org/10.5194/hess-28-2871-2024
- Applications of artificial intelligence models in seawater intrusion: progress, challenges, and future directions X. Wang et al. https://doi.org/10.1016/j.jhydrol.2026.136086
- Transformer based models with hierarchical graph representations for enhanced climate forecasting T. Ramu et al. https://doi.org/10.1038/s41598-025-07897-4
- An Approach for Future Droughts in Northwest Türkiye: SPI and LSTM Methods E. Taylan https://doi.org/10.3390/su16166905
- Coupling strategies of snowmelt runoff model and machine learning in the Lhasa River Basin T. Wang et al. https://doi.org/10.1016/j.ejrh.2026.103400
- Multi-site deep learning for groundwater level prediction across global datasets: toward scalable applications under data scarcity A. Nolte et al. https://doi.org/10.2166/hydro.2025.095
- Deep learning error post-processing improves stochastic watershed modeling B. Manoli et al. https://doi.org/10.1016/j.jhydrol.2026.135663
- Hybrid models generalize better to warmer climate conditions than process-based and purely data-driven models J. Bohl et al. https://doi.org/10.5194/hess-30-4667-2026
- The outcome prediction method of football matches by the quantum neural network based on deep learning Y. Sun & H. Chu https://doi.org/10.1038/s41598-025-91870-8
- Pooling local climate and donor gauges with deep learning for improved reconstructions of streamflow in ungauged and partially gauged basins S. Wi et al. https://doi.org/10.1016/j.jhydrol.2025.133764
- Predicting Forest Evapotranspiration Shifts Under Diverse Climate Change Scenarios by Leveraging the SEBAL Model Across Inner Mongolia P. Ji et al. https://doi.org/10.3390/f15122234
- Distinct hydrologic response patterns and trends worldwide revealed by physics-embedded learning H. Ji et al. https://doi.org/10.1038/s41467-025-64367-1
- A hybrid TFN–LSTM model for groundwater level forecasting in Rhode Island, USA H. Chu et al. https://doi.org/10.1016/j.jenvman.2026.130474
- Beyond prediction accuracy: using differentiable hybrid models as a tool for hydrological knowledge discovery Y. Hu et al. https://doi.org/10.1016/j.jhydrol.2026.136147
- Associations between deep learning runoff predictions and hydrogeological conditions in Australia S. Clark & J. Jaffrés https://doi.org/10.1016/j.jhydrol.2024.132569
Saved (final revised paper)
Latest update: 31 Aug 2026
Short summary
We investigate whether deep learning (DL) models can produce physically plausible streamflow projections under climate change. We address this question by focusing on modeled responses to increases in temperature and potential evapotranspiration and by employing three DL and three process-based hydrological models. The results suggest that physical constraints regarding model architecture and input are necessary to promote the physical realism of DL hydrological projections under climate change.
We investigate whether deep learning (DL) models can produce physically plausible streamflow...
