Predictive Simulation and Early Warning in Flood Digital Twins: A Systematic Review of AI Model Performance, Warning Lead Times, And Operational Deployment
DOI:
https://doi.org/10.65150/EP-gjetr/V2E8/2026-05Keywords:
Flood Digital Twin, LSTM, Early Warning Lead Time, Data Assimilation, Physics-informed Neural Networks, U.S. Flood RiskAbstract
Flooding is the most economically destructive natural hazard in the United States, with annual losses exceeding $60 billion and risk concentrated unevenly across communities and regions. This systematic review synthesizes evidence on AI-enhanced digital twin architectures for U.S. flood infrastructure management. Following PRISMA 2020 guidelines, a search across six academic databases (2020–2025) identified 3,847 records, of which 94 studies met inclusion criteria.
The findings are clear. Long Short-Term Memory (LSTM) networks reach Nash-Sutcliffe Efficiency (NSE) values of 0.88–0.97 in well-gauged U.S. basins, consistently outperforming the U.S. National Weather Service's physics-based operational models. Physics-informed neural networks (PINNs) work in data-scarce U.S. watersheds, improving NSE by 0.04–0.11 over standard baselines. AI-augmented systems push meaningful warning lead times from the conventional 1–3 hours out to 6–24 hours in medium-sized catchments; when paired with numerical weather prediction ensembles, operational deployments have reached 72-hour horizons. Data assimilation (DA) adds NSE gains of 0.05–0.12 over open-loop simulations, with particle filter approaches outperforming ensemble Kalman filters by up to 0.18 in non-linear inundation scenarios; satellite-based DA opens viable pathways in U.S. regions where ground gauges are sparse.
The gaps are equally clear. Only 31% of reviewed studies provide calibrated probabilistic predictions, and the evidence base is geographically concentrated in well-instrumented catchments, leaving data-scarce watersheds and socially vulnerable communities systematically underrepresented. The review closes with concrete, evidence-grounded recommendations for standardizing U.S. benchmarking frameworks and accelerating equitable operational deployment.
References
1) Adu, S. A. G., Gyang, P. A. E. M., & Yakin, Z. (2025). The role of GIS and spatial analysis in enhancing urban resilience and disaster response for vulnerable U.S. communities. World Journal of Advanced Research and Reviews, 27(1), 746–754.
2) Bartos, M. D., & Kerkez, B. (2021). Pipedream: An interactive digital twin model for natural and urban drainage systems. Environmental Modelling & Software, 144, 105120. https://doi.org/10.1016/j.envsoft.2021.105120
3) Bentivoglio, R., Isufi, E., Jonkman, S. N., & Taormina, R. (2022). Deep learning methods for flood mapping: A review of existing applications and future research directions. Hydrology and Earth System Sciences, 26(16), 4345–4378. https://doi.org/10.5194/hess-26-4345-2022
Cosgrove, B., Gochis, D., Flowers, T., et al. (2024). NOAA’s National Water Model: Advancing operational hydrology through continental-scale modeling. JAWRA Journal of the American Water Resources Association, 60(2), 247–272.
https://doi.org/10.1111/1752-1688.13184
4) Dasgupta, A., Hostache, R., Ramsankaran, R., Schumann, G. J.-P., Grimaldi, S., Pauwels, V. R. N., & Walker, J. P. (2021). A mutual information-based likelihood function for particle filter flood inundation data assimilation. Water Resources Research, 57(2), e2020WR027176. https://doi.org/10.1029/2020WR027176
5) Demir, I., Xiang, Z., Demiray, B., & Sit, M. (2022). WaterBench-Iowa: a large-scale benchmark dataset for data-driven streamflow forecasting. Earth System Science Data, 14, 5605–5616. https://doi.org/10.5194/essd-14-5605-2022
6) Donnelly, J., Daneshkhah, A., & Abolfathi, S. (2024). Physics-informed neural networks as surrogate models of hydrodynamic simulators. Science of the Total Environment, 912, 168814. https://doi.org/10.1016/j.scitotenv.2023.168814
7) Frame, J. M., Kratzert, F., Klotz, D., Gauch, M., Shalev, G., Gilon, O., Qualls, L. M., Gupta, H. V., & Nearing, G. S. (2022). Deep learning rainfall–runoff predictions of extreme events. Hydrology and Earth System Sciences, 26, 3377–3392.
https://doi.org/10.5194/hess-26-3377-2022
8) Fuller, A., Fan, Z., Day, C., & Barlow, C. (2020). Digital twin: Enabling technologies, challenges and open research. IEEE Access, 8, 108952–108971. https://doi.org/10.1109/ACCESS.2020.2990354
9) Gauch, M., Kratzert, F., Klotz, D., Nearing, G., Lin, J., & Hochreiter, S. (2021). Rainfall–runoff prediction at multiple timescales with a single Long Short-Term Memory network. Hydrology and Earth System Sciences, 25(4), 2045–2062. https://doi.org/10.5194/hess-25-2045-2021
10) Burke, P. C., Barnwell, J., Reagan, M., Rose, M. A., Galarneau, T. J., Jr., Otto, R., & Orrison, A. (2024). The 21 August 2021 catastrophic flash flood at Waverly, Tennessee: Harnessing the Warn-on-Forecast System for confident prewarning messaging of extreme rainfall. Bulletin of the American Meteorological Society, 105(10).
11) Ge, C., & Qin, S. (2025). Urban flooding digital twin system framework. Systems Science & Control Engineering, 13(1), 2460432.
https://doi.org/10.1080/21642583.2025.2460432
Gyang, P. A. E. M., Akomolafe, O., Lotsah, M., & Opoku, L. K. (2025b). Climate change and flood risk in the U.S.: A mini review of models, future scenarios and adaptation strategies. World Journal of Advanced Engineering Technology and Sciences, 14(1), 185–190.
12) Gyang, P. A. E. M., Akomolafe, O., Panful, B., & Yowetu, I. A. (2025a). A review of integrated use of machine learning algorithms, GIS and remote sensing techniques in the prediction of rainfall patterns and floods in the U.S. World Journal of Advanced Research and Reviews.
13) Gyang, P. A. E. M., Donkor, A., & Oware, D. (2024). A review of the use of GIS and remote sensing technologies in monitoring, prediction, and response to natural disasters in the U.S.. International Journal of Research Publication and Reviews, 5, 5976–5983.
14) IPCC. (2021). Climate change 2021: The physical science basis. Cambridge University Press.
15) Jadidoleslam, N., Mantilla, R., & Krajewski, W. F. (2021). Data assimilation of satellite-based soil moisture into a distributed hydrological model for streamflow predictions. Hydrology, 8(1), 52. https://doi.org/10.3390/hydrology8010052
16) Kratzert, F., Gauch, M., Klotz, D., & Nearing, G. (2024). HESS Opinions: Never train a Long Short-Term Memory (LSTM) network on a single basin. Hydrology and Earth System Sciences, 28(17), 4187–4201. https://doi.org/10.5194/hess-28-4187-2024
17) LaFraniere, S., & Levenson, E. (2021, August 30). Hurricane Ida leaves widespread devastation across the Gulf Coast. The New York Times. https://www.nytimes.com/2021/08/30/us/hurricane-ida-new-orleans.html
18) Ma, K., Feng, D., Lawson, K., Tsai, W.-P., Liang, C., Luo, X., Kifer, D., & Shen, C. (2021). Transferability of data-driven models for streamflow estimation. Water Resources Research, 57(4), e2020WR028982. https://doi.org/10.1029/2020WR028982
19) National Weather Service. (2022). Historic July 26–30, 2022 Eastern Kentucky flooding. Jackson, KY: NOAA/NWS.
https://www.weather.gov/jkl/July2022Flooding
20) NOAA repository version: The 21 August 2021 catastrophic flash flood at Waverly. NOAA/NWS.
https://repository.library.noaa.gov/view/noaa/69265
National Weather Service. (2022). Service assessment: The historic Middle Tennessee flash flood of August 21, 2021. NOAA. https://www.weather.gov/media/publications/assessments/Middle_TN_Flood_2021.pdf
21) National Weather Service. (2023). Significant river/flash flood in Southeastern Kentucky, July 2022 [Service Assessment/After-Action Report]. https://www.weather.gov/media/publications/assessments/July_2022_Significant_River_Flash_Flood_SE_KY.pdf
22) Nearing, G. S., Kratzert, F., Sampson, A. K., et al. (2021). What role does hydrological science play in the age of machine learning? Water Resources Research, 57(3), e2020WR028091. https://doi.org/10.1029/2020WR028091
23) Nearing, G. S., Klotz, D., Frame, J. M., et al. (2022). Technical note: Data assimilation and autoregression for using near-real-time streamflow observations in LSTM networks. Hydrology and Earth System Sciences, 26(21), 5493–5513. https://doi.org/10.5194/hess-26-5493-2022
24) Oddo, P. C., Bolten, J. D., Kumar, S. V., & Cleary, B. (2024). Deep convolutional LSTM for improved flash flood prediction. Frontiers in Water, 6, 1346104. https://doi.org/10.3389/frwa.2024.1346104
25) Page, M. J., McKenzie, J. E., Bossuyt, P. M., et al. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71
26) Xu, Q., Shi, Y., Zhao, J., & Zhu, X. X. (2025). FloodCastBench: A large-scale dataset and foundation models for flood modeling and forecasting. Scientific Data, 12, 431. https://doi.org/10.1038/s41597-025-04725-2
27) Tennessee flooding Sunday: Death toll tops 20 as towns begin recovery. (2021, August 22). The Tennessean.
28) Wasko, C., Nathan, R., Stein, L., & O'Shea, D. (2021). Evidence of shorter more extreme rainfalls and increased flood variability under climate change. Journal of Hydrology, 603, 126994. https://doi.org/10.1016/j.jhydrol.2021.126994
29) Wing, O. E. J., Lehman, W., Bates, P. D., et al. (2022). Inequitable patterns of U.S. flood risk in the Anthropocene. Nature Climate Change, 12(2), 156–162. https://doi.org/10.1038/s41558-021-01265-z
30) Xiang, Z., Yan, J., & Demir, I. (2020). A rainfall-runoff model with LSTM-based sequence-to-sequence learning. Water Resources Research, 56(1), e2019WR025326. https://doi.org/10.1029/2019WR025326
31) Xu, Y., Lin, K., Hu, C., Wang, S., Wu, Q., Zhang, L., & Ran, G. (2023). Deep transfer learning based on transformer for flood forecasting in data-sparse basins. Journal of Hydrology, 625, 129956. https://doi.org/10.1016/j.jhydrol.2023.129956
32) Zhou, H., Zhang, S., Peng, J., Zhang, S., Li, J., Xiong, H., & Zhang, W. (2021). Informer: Beyond efficient transformer for long sequence time-series forecasting. Proceedings of the AAAI Conference on Artificial Intelligence, 35(12), 11106–11115.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Lydia Dede Obeng (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
