Abstract
AbstractTropical cyclones are responsible for large-scale loss of life and property1–4, motivating accurate risk assessment and forecasting. These objectives require accurate reconstructions of storms’ wind and pressure fields which assimilate real-time observations5–9, but current methods used for these reconstructions remain computationally expensive and limited10. Here, we show that a physics-informed neural network11,12 can be a promising and computationally efficient algorithm for tropical cyclone data assimilation. Using synthetic training data sparsely sampled from hurricanes simulated in a forecast model, a physics-informed neural network is able to reconstruct full realistic 2- and 3-dimensional wind and pressure fields which capture key features of the cyclone. We also demonstrate how a set of sparse, real-time observations, can be used to accurately reconstruct Hurricane Ida. Our results highlight how recent advances in deep learning can augment data assimilation schemes. The methods are also general and can be applied to other flow problems.
Publisher
Springer Science and Business Media LLC
Subject
General Earth and Planetary Sciences,General Environmental Science
Reference56 articles.
1. NOAA. Hurricane costs https://coast.noaa.gov/states/fast-facts/hurricane-costs.html#:~:text=Of%20the%20310%20billion%2Ddollar,6%2C697%20between%201980%20and%202021 (2022).
2. Willoughby, H., Rappaport, E. & Marks, F. Hurricane forecasting: the state of the art. Nat. Hazards Rev. 8, 45–49(2007).
3. Klotzbach, P. J., Bowen, S. G., Pielke, R. & Bell, M. Continental u.s. hurricane landfall frequency and associated damage: Observations and future risks. Bull. Am. Meteorol. Soc. 99, 1359–1376 (2018).
4. Weinkle, J. et al. Normalized hurricane damage in the continental United States 1900-2017. Nat. Sustain. 1, 808–813 (2018).
5. Kurihara, Y., Tuleya, R. E. & Bender, M. A. The GFDL hurricane prediction system and its performance in the 1995 hurricane season. Mon. Weather Rev. 126, 1306–1322 (1998).