Summary of Storm Surge Modeling in the Ai Era: Using Lstm-based Machine Learning For Enhancing Forecasting Accuracy, by Stefanos Giaremis et al.
Storm Surge Modeling in the AI ERA: Using LSTM-based Machine Learning for Enhancing Forecasting Accuracy
by Stefanos Giaremis, Noujoud Nader, Clint Dawson, Hartmut Kaiser, Carola Kaiser, Efstratios Nikidis
First submitted to arxiv on: 7 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Atmospheric and Oceanic Physics (physics.ao-ph)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A deep learning network-based machine learning architecture, utilizing Long Short-Term Memory (LSTM) recurrent neural networks, is proposed to predict and capture systemic error in storm surge forecast models. This approach aims to improve the accuracy of simulation results by correcting bias post facto. The model was trained on a dataset of 61 historical storms along the US coast and tested on hurricane Ian (2022) data. Results show consistent improvement in forecasting accuracy at all gauge station coordinates, with similar quality achieved using only six hurricanes from the initial training set. This work presents a transferable methodology for bias correction in various physics simulation scenarios. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way to make storm surge predictions more accurate is being developed. Scientists are creating a special kind of computer model that can learn from past storms and predict how well current simulations will match real-world data. The goal is to improve the accuracy of these simulations by correcting any mistakes made along the way. The team tested their approach on historical storm data and found it worked well, even when using only a small subset of the original data. This new method has the potential to make a big difference in predicting storm surges accurately. |
Keywords
* Artificial intelligence * Deep learning * Lstm * Machine learning
