Literature Review of Hybrid Neural–Physics Flood-Relevant Streamflow Forecasting with Cross-Event and Cross-Basin Generalization
DOI:
https://doi.org/10.58776/ijitcsa.v4i2.263Keywords:
Flood forecasting, Streamflow prediction, Hybrid neural–physics modeling, Ungauged basins, Cross-event generalizationAbstract
Reliable flood-relevant streamflow forecasting remains difficult at operational scale because physically based forecasting systems provide national coverage but cannot be directly verified at the overwhelming majority of ungauged river reaches. This study reconstructs and strengthens a hybrid neural–physics framework in which the National Water Model (NWM) supplies the physically constrained baseline forecast and an attribute-conditioned deep neural network learns forecast-error magnitude from gauged basins, then transfers that knowledge to ungauged basins. The emphasis is not on replacing process-based hydrology with a black-box predictor, but on improving forecast reliability under cross-event and cross-basin generalization. The empirical setting consists of 979 archived daily long-range forecasts, 30 forecast lead days, 16-member NWM ensembles, and 389 quality-controlled U.S. Geological Survey gauging stations across Alabama and Georgia, complemented by static watershed descriptors and gridded soil-moisture diagnostics. Exploratory analysis shows that forecast skill is strongly conditioned by basin properties: larger and forested watersheds are predicted more accurately than smaller and urban watersheds, while urban basins show severe positive bias, weaker anomaly correlation, and degraded hydrograph timing. These diagnostics motivate an error-learning postprocessor that maps NWM forecast magnitude and basin attributes to a non-negative error estimate and converts the deterministic physics forecast into a corrected predictive interval. Under a temporally disjoint evaluation intended to test cross-event transfer, observation coverage increases from 0.21±0.01 for the raw NWM ensemble range to 0.82±0.03 for the hybrid model; a repeated spatial holdout yields comparable performance (0.82±0.01). The gains are largest where the underlying physics model is weakest, especially in developed watersheds. The main limitation is that improved reliability is achieved through wider predictive intervals rather than sharper hydrographs. The study therefore supports a defensible conclusion of hybrid neural–physics postprocessing can materially improve flood-relevant streamflow reliability at ungauged basins, but interval sharpness and external cross-region validation remain open research needs.
References
. D. Gochis, W. Yu, and D. Yates, “The NCAR WRF-hydro technical description and user’s guide, version 1.0,” National Center for Atmospheric Research, 2013. Available: https://ral.ucar.edu/sites/default/files/public/projects/wrf_hydro/WRF_Hydro_Technical_Description_and_User_Guide_v1.0.pdf
. D. Gochis et al., “The NCAR WRF-hydro modeling system technical description, version 5.0,” National Center for Atmospheric Research, 2018. Available: https://ral.ucar.edu/sites/default/files/public/WRF-HydroV5TechnicalDescription.pdf
. M. Hrachowitz et al., “A decade of predictions in ungauged basins (PUB)—a review,” Hydrological Sciences Journal, vol. 58, no. 6, pp. 1198–1255, 2013, doi: 10.1080/02626667.2013.803183.
. T. Wagener and H. S. Wheater, “Parameter estimation and regionalization for continuous rainfall-runoff models including uncertainty,” Journal of Hydrology, vol. 320, no. 1–2, pp. 132–154, 2006, doi: 10.1016/j.jhydrol.2005.07.015.
. M. Sivapalan et al., “IAHS decade on predictions in ungauged basins (PUB), 2003–2012: Shaping an exciting future for the hydrological sciences,” Hydrological Sciences Journal, vol. 48, no. 6, pp. 857–880, 2003, doi: 10.1623/hysj.48.6.857.51421.
. R. Singh, S. A. Archfield, and T. Wagener, “Identifying dominant controls on hydrologic parameter transfer from gauged to ungauged catchments—a comparative hydrology approach,” Journal of Hydrology, vol. 517, pp. 985–996, 2014, doi: 10.1016/j.jhydrol.2014.06.030.
. G. Nearing, D. Cohen, V. Dube, et al., “Global prediction of extreme floods in ungauged watersheds,” Nature, vol. 627, no. 8004, pp. 559–563, 2024, doi: 10.1038/s41586-024-07145-1.
. F. Kratzert, D. Klotz, C. Brenner, K. Schulz, and M. Herrnegger, “Rainfall–runoff modelling using long short-term memory (LSTM) networks,” Hydrology and Earth System Sciences, vol. 22, no. 11, pp. 6005–6022, 2018, doi: 10.5194/hess-22-6005-2018.
. F. Kratzert, D. Klotz, G. Shalev, G. Klambauer, S. Hochreiter, and G. S. Nearing, “Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets,” Hydrology and Earth System Sciences, vol. 23, no. 12, pp. 5089–5110, 2019, doi: 10.5194/hess-23-5089-2019.
. F. Kratzert, D. Klotz, S. Hochreiter, and G. S. Nearing, “A note on leveraging synergy in multiple meteorological data sets with deep learning for rainfall–runoff modeling,” Hydrology and Earth System Sciences, vol. 25, no. 5, pp. 2685–2703, 2021, doi: 10.5194/hess-25-2685-2021.
. J. M. Frame et al., “Deep learning rainfall–runoff predictions of extreme events,” Hydrology and Earth System Sciences, vol. 26, no. 13, pp. 3377–3392, 2022, doi: 10.5194/hess-26-3377-2022.
. D. Klotz et al., “Uncertainty estimation with deep learning for rainfall–runoff modeling,” Hydrology and Earth System Sciences, vol. 26, no. 6, pp. 1673–1693, 2022, doi: 10.5194/hess-26-1673-2022.
. J. M. Frame, F. Kratzert, A. Raney, M. Rahman, F. R. Salas, and G. S. Nearing, “Post-processing the national water model with long short-term memory networks for streamflow predictions and model diagnostics,” Journal of the American Water Resources Association, vol. 57, no. 5, pp. 885–905, 2021, doi: 10.1111/1752-1688.12964.
. Y. Zhang et al., “Generalization of an encoder-decoder LSTM model for flood prediction in ungauged catchments,” Journal of Hydrology, vol. 614, p. 128577, 2022, doi: 10.1016/j.jhydrol.2022.128577.
. L. Schmidt, F. Hesse, S. Attinger, and R. Kumar, “Challenges in applying machine learning models for hydrological inference: A case study for flooding events across germany,” Water Resources Research, vol. 56, no. 5, p. e2019WR025924, 2020, doi: 10.1029/2019WR025924.
. V. Moreido et al., “How well can machine learning models perform without hydrologists? Application of rational feature selection to improve hydrological forecasting,” Water, vol. 13, no. 12, p. 1696, 2021, doi: 10.3390/w13121696.
. J. G. Arnold, R. S. Muttiah, and G. Bernhardt, “Automated base flow separation and recession analysis techniques,” Ground Water, vol. 33, no. 6, pp. 1010–1018, 1995, doi: 10.1111/j.1745-6584.1995.tb00046.x.
. J. G. Arnold and P. M. Allen, “Automated methods for estimating baseflow and groundwater recharge from streamflow records,” Journal of the American Water Resources Association, vol. 35, no. 2, pp. 411–424, 1999, doi: 10.1111/j.1752-1688.1999.tb03599.x.
. R. M. Vogel and C. N. Kroll, “Estimation of baseflow recession constants,” Water Resources Management, vol. 10, no. 4, pp. 303–320, 1996, doi: 10.1007/BF00508898.
. C. Shen et al., “A transdisciplinary review of deep learning research and its relevance for water resources scientists,” Water Resources Research, vol. 54, no. 11, pp. 8558–8593, 2018, doi: 10.1029/2018WR022643.
. P. Parisouj et al., “Physics-informed data-driven model for predicting streamflow: A case study of the voshmgir basin, iran,” Applied Sciences, vol. 12, no. 15, p. 7464, 2022, doi: 10.3390/app12157464.
. H. Solanki et al., “Improving streamflow prediction using multiple hydrological and hydrodynamical models with machine learning,” Water Resources Research, vol. 61, no. 1, 2025, doi: 10.1029/2024WR038192.
. H. Wu et al., “Runoff modeling in ungauged catchments using machine learning algorithm-based model parameters regionalization methodology,” Engineering, vol. 28, pp. 93–104, 2023, doi: 10.1016/j.eng.2021.12.014.
. A. Tursun et al., “Streamflow prediction in human-regulated catchments using data-driven models and multiscale static attributes,” Water Resources Research, vol. 60, no. 7, p. e2023WR036853, 2024, doi: 10.1029/2023WR036853.
. V. Nair and G. E. Hinton, “Rectified linear units improve restricted boltzmann machines,” in Proceedings of the 27th international conference on machine learning, 2010, pp. 807–814.
. T. Tieleman and G. Hinton, “Lecture 6.5—RMSprop: Divide the gradient by a running average of its recent magnitude.” COURSERA: Neural Networks for Machine Learning, 2012.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Yanan Elliot Moko
This work is licensed under a Creative Commons Attribution 4.0 International License.
Attribution 4.0 International
You are free to:
- Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
- The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
- Attribution — You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Notices:
You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation .
No warranties are given. The license may not give you all of the permissions necessary for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.
