Physics-Aware AI-Initialized Dynamical Downscaling for Regional Extreme-Weather Forecasting with Open Climate Data
DOI:
https://doi.org/10.58776/ijitcsa.v4i2.256Keywords:
extreme precipitation forecasting, AI weather prediction, regional downscaling, WRF, Pangu-Weather, physics-aware machine learning, open climate dataAbstract
Accurate regional forecasting of extreme precipitation remains difficult because the scales that control disaster-producing rainfall are neither fully resolved by global numerical weather prediction nor reliably preserved by current global artificial intelligence weather models. Global AI systems such as Pangu-Weather, GraphCast, GenCast, and related models have transformed medium-range forecasting skill and computational efficiency, yet they remain fundamentally constrained by coarse training targets, regression-induced smoothing, and limited direct representation of terrain-locked convection and local hydrometeorological extremes. This paper reconstructs and substantially extends an event-based manuscript on AI-driven regional forecasting into a submission-oriented framework centered on the more defensible idea of physics-aware regional extreme-weather forecasting or downscaling with open climate data. The core argument is that AI should not be treated as a wholesale substitute for high-resolution regional physics; rather, it should be used as a skillful large-scale predictor whose state can be physically harmonized and injected into a regional nonhydrostatic model. We therefore formalize an AI-initialized, physics-aware dynamical downscaling pipeline in which open global reanalysis and observation products are used to generate, constrain, and evaluate regional forecasts of extreme rainfall. The framework is instantiated using the published North China July–August 2023 extreme precipitation case, for which the original study compared WRF simulations driven by Pangu forecasts against WRF simulations driven by NCEP GFS forecasts across lead times of 0.5, 3.0, and 5.5 days. This paper contributes in three ways. First, it repositions the original study within the modern literature on AI weather forecasting, regional downscaling, and physically constrained machine learning. Second, it formulates the coupling problem mathematically, clarifies the state alignment needed to make AI forecasts dynamically usable by WRF, and introduces a coherent reliability-oriented evaluation logic based on error growth, threshold skill, and event-structure consistency. Third, it reorganizes the experiments and results into a rigorous narrative grounded in reproducibility. Using the published event-level metrics, the AI-initialized regional system outperforms the GFS-initialized counterpart at extended lead times. For the North China case, the maximum precipitation threshold retaining a Threat Score of at least 0.1 is 400 mm at 5.5-day lead for Pangu-initialized WRF, whereas the GFS-driven counterpart retains comparable skill only at 50 mm. At 0.5-day lead, both systems perform competitively, but the AI-driven system still exhibits stronger spatial correlation (0.76 versus 0.68) and lower RMSE (86.2 mm versus 96.4 mm). The evidence supports a restrained but important conclusion: physics-aware AI-initialized regional modeling is a promising route for long-lead extreme-weather forecasting, yet current evidence remains case-limited and should be interpreted as a strong event-based demonstration rather than universal proof of general superiority.
References
P. D. Dueben and P. Bauer, “Challenges and design choices for global weather and climate models based on machine learning,” Geoscientific Model Development, vol. 11, no. 10, pp. 3999–4009, 2018, doi: 10.5194/gmd-11-3999-2018.
J. A. Weyn, D. R. Durran, and R. Caruana, “Can machines learn to predict weather? Using deep learning to predict gridded 500-hPa geopotential height from historical weather data,” Journal of Advances in Modeling Earth Systems, vol. 11, no. 8, pp. 2680–2693, 2019, doi: 10.1029/2019MS001705.
G. H. Bryan, J. C. Wyngaard, and J. M. Fritsch, “Resolution requirements for the simulation of deep moist convection,” Monthly Weather Review, vol. 131, no. 10, pp. 2394–2416, 2003, doi: 10.1175/1520-0493(2003)131<2394:RRFTSO>2.0.CO;2.
J. Vergara-Temprado, N. Ban, D. Panosetti, L. Schlemmer, and C. Schär, “Climate models permit convection at much coarser resolutions than previously considered,” Journal of Climate, vol. 33, no. 5, pp. 1915–1933, 2020, doi: 10.1175/JCLI-D-19-0286.1.
K. Bi, L. Xie, H. Zhang, X. Chen, X. Gu, and Q. Tian, “Accurate medium-range global weather forecasting with 3D neural networks,” Nature, vol. 619, no. 7970, pp. 533–538, 2023, doi: 10.1038/s41586-023-06185-3.
R. Lam, A. Sanchez-Gonzalez, M. Willson, et al., “Learning skillful medium-range global weather forecasting,” Science, vol. 382, no. 6677, pp. 1416–1421, 2023, doi: 10.1126/science.adi2336.
I. Price, A. Sanchez-Gonzalez, F. Alet, et al., “Probabilistic weather forecasting with machine learning,” Nature, vol. 637, no. 8044, pp. 84–90, 2025, doi: 10.1038/s41586-024-08252-9.
A. Allen, S. Markou, W. Tebbutt, et al., “End-to-end data-driven weather prediction,” Nature, vol. 641, no. 8065, pp. 1172–1179, 2025, doi: 10.1038/s41586-025-08897-0.
K. Chen, T. Han, F. Ling, et al., “The operational medium-range deterministic weather forecasting can be extended beyond a 10-day lead time,” Communications Earth & Environment, vol. 6, p. 518, 2025, doi: 10.1038/s43247-025-02502-y.
C.-C. Liu, K. Hsu, M. S. Peng, et al., “Evaluation of five global AI models for predicting weather in eastern asia and western pacific,” npj Climate and Atmospheric Science, vol. 7, p. 221, 2024, doi: 10.1038/s41612-024-00769-0.
G. Camps-Valls, M.-Á. Fernández-Torres, K.-H. Cohrs, et al., “Artificial intelligence for modeling and understanding extreme weather and climate events,” Nature Communications, vol. 16, p. 1919, 2025, doi: 10.1038/s41467-025-56573-8.
H.-Y. Liu, J.-F. Gu, Y. Wang, and J. Xu, “What controlled the low-level moisture transport during the extreme precipitation in henan province of china in july 2021?” Monthly Weather Review, vol. 151, no. 5, pp. 1347–1365, 2023, doi: 10.1175/MWR-D-22-0200.1.
D. Zhao, H. Xu, Y. Li, et al., “Locally opposite responses of the 2023 beijing–tianjin–hebei extreme rainfall event to global anthropogenic warming,” npj Climate and Atmospheric Science, vol. 7, p. 38, 2024, doi: 10.1038/s41612-024-00584-7.
H. Xu, Y. Zhao, D. Zhao, Y. Duan, and X. Xu, “Improvement of disastrous extreme precipitation forecasting in north china by pangu-weather AI-driven regional WRF model,” Environmental Research Letters, vol. 19, no. 5, p. 054051, 2024, doi: 10.1088/1748-9326/ad41f0.
D. C. Dowell, C. R. Alexander, E. P. James, et al., “The high-resolution rapid refresh (HRRR): An hourly updating convection-allowing forecast model. Part I: Motivation and system description,” Weather and Forecasting, vol. 37, no. 8, pp. 1371–1395, 2022, doi: 10.1175/WAF-D-21-0151.1.
M. Zhang and D.-L. Zhang, “Subkilometer simulation of a torrential-rain-producing mesoscale convective system in east china. Part I: Model verification and convective organization,” Monthly Weather Review, vol. 140, no. 1, pp. 184–201, 2012, doi: 10.1175/MWR-D-11-00029.1.
T. Vukicevic, D. J. Posselt, and A. Stankovic, “Sensitivity of modeled microphysics to stochastically perturbed parameters,” Journal of Advances in Modeling Earth Systems, vol. 14, no. 1, p. e2021MS002933, 2022, doi: 10.1029/2021MS002933.
L. Chen, X. Zhong, H. Li, et al., “A machine learning model that outperforms conventional global subseasonal forecast models,” Nature Communications, vol. 15, p. 6425, 2024, doi: 10.1038/s41467-024-50714-1.
M. Bonavita, “On some limitations of current machine learning weather prediction models,” Geophysical Research Letters, vol. 51, no. 12, p. e2023GL107377, 2024, doi: 10.1029/2023GL107377.
Y. Sha, J. S. Schreck, W. Chapman, et al., “Improving AI weather prediction models using global mass and energy conservation schemes,” Journal of Advances in Modeling Earth Systems, vol. 17, no. 11, p. e2025MS005138, 2025, doi: 10.1029/2025MS005138.
L. Deng, J. Feng, Y. Zhao, et al., “The remote effect of binary typhoon in-fa and cempaka on the ‘21.7’ heavy rainfall in henan province, china,” Journal of Geophysical Research: Atmospheres, vol. 127, no. 10, p. e2021JD036260, 2022, doi: 10.1029/2021JD036260.
Z. Ding, K. Zhao, K. Zhu, et al., “Assimilation of polarimetric radar observation with GSI cloud analysis for the prediction of a squall line,” Geophysical Research Letters, vol. 49, no. 12, p. e2022GL098253, 2022, doi: 10.1029/2022GL098253.
H. Xu, Y. Duan, Y. Li, and H. Wang, “Indirect effects of binary typhoons on an extreme rainfall event in henan province, china, from 19 to 21 july 2021: 2. Numerical study,” Journal of Geophysical Research: Atmospheres, vol. 127, no. 21, p. e2021JD036083, 2022, doi: 10.1029/2021JD036083.
H. Xu, D. Zhao, J. Yin, et al., “Indirect effects of binary typhoons on an extreme rainfall event in henan province, china, from 19 to 21 july 2021. 3. Sensitivities to microphysics schemes,” Journal of Geophysical Research: Atmospheres, vol. 128, no. 7, p. e2022JD037936, 2023, doi: 10.1029/2022JD037936.
S. Hobeichi, N. Nishant, Y. Shao, et al., “Using machine learning to cut the cost of dynamical downscaling,” Earth’s Future, vol. 11, no. 3, p. e2022EF003291, 2023, doi: 10.1029/2022EF003291.
N. Rampal, S. Hobeichi, P. B. Gibson, J. Baño-Medina, et al., “Enhancing regional climate downscaling through advances in machine learning,” Artificial Intelligence for the Earth Systems, vol. 3, no. 2, p. e230066, 2024, doi: 10.1175/AIES-D-23-0066.1.
L. Harris, S. Rasp, et al., “A generative deep learning approach to stochastic downscaling of precipitation forecasts,” Journal of Advances in Modeling Earth Systems, vol. 14, no. 10, p. e2022MS003120, 2022, doi: 10.1029/2022MS003120.
D. Quesada-Chacón et al., “Downscaling CORDEX through deep learning to daily 1 km multivariate ensemble in complex terrain,” Earth’s Future, vol. 11, no. 8, p. e2023EF003531, 2023, doi: 10.1029/2023EF003531.
T. Yoshikane and K. Yoshimura, “A downscaling and bias correction method for climate model ensemble simulations of local-scale hourly precipitation,” Scientific Reports, vol. 13, p. 9412, 2023, doi: 10.1038/s41598-023-36489-3.
J. González-Abad, J. Baño-Medina, and J. M. Gutiérrez, “Using explainability to inform statistical downscaling based on deep learning beyond standard validation approaches,” Journal of Advances in Modeling Earth Systems, vol. 15, no. 10, p. e2023MS003641, 2023, doi: 10.1029/2023MS003641.
J. Pathak, S. Subramanian, P. Harrington, S. Raja, A. Chattopadhyay, and M. Mardani, “FourCastNet: A global data-driven high-resolution weather model using adaptive Fourier neural operators,” arXiv preprint arXiv:2202.11214, 2022, doi: 10.48550/arXiv.2202.11214.
J. Baño-Medina, A. Sengupta, D. Steinhoff, et al., “A regional high resolution AI weather model for the prediction of atmospheric rivers and extreme precipitation,” npj Climate and Atmospheric Science, vol. 8, p. 385, 2025, doi: 10.1038/s41612-025-01265-9.
H. Hersbach, B. Bell, P. Berrisford, et al., “The ERA5 global reanalysis,” Quarterly Journal of the Royal Meteorological Society, vol. 146, no. 730, pp. 1999–2049, 2020, doi: 10.1002/qj.3803.
D. A. Lavers, A. Simmons, F. Vamborg, et al., “An evaluation of ERA5 precipitation for climate monitoring,” Quarterly Journal of the Royal Meteorological Society, vol. 148, no. 748, pp. 3152–3165, 2022, doi: 10.1002/qj.4351.
T. Beucler, M. Pritchard, S. Rasp, J. Ott, P. Baldi, and P. Gentine, “Enforcing analytic constraints in neural networks emulating physical systems,” Physical Review Letters, vol. 126, no. 9, p. 098302, 2021, doi: 10.1103/PhysRevLett.126.098302.
N. D. Brenowitz et al., “A practical probabilistic benchmark for AI weather models,” Geophysical Research Letters, vol. 52, no. 4, p. e2024GL113656, 2025, doi: 10.1029/2024GL113656.
Y. Shen, P. Zhao, Y. Pan, and J. Yu, “A high spatiotemporal gauge-satellite merged precipitation analysis over china,” Journal of Geophysical Research: Atmospheres, vol. 119, no. 6, pp. 3063–3075, 2014, doi: 10.1002/2013JD020686.
National Centers for Environmental Prediction, “NCEP ADP Global Upper Air and Surface Weather Observations (PREPBUFR format).” 2008. doi: 10.5065/Z83F-N512.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Mina Annetta, Shanti Purohit, Viljar Vagle, Deepak Deo
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.
