Hurricane Forecasting: A Novel Multimodal Machine Learning Framework
This paper describes a novel machine learning (ML) framework for tropical cyclone intensity and track forecasting, combining multiple ML techniques and utilizing diverse data sources. Our multimodal framework, called Hurricast, efficiently combines spatial-temporal data with statistical data by extr...
Main Authors: | , , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
arXiv
2020
|
Subjects: | |
Online Access: | https://dx.doi.org/10.48550/arxiv.2011.06125 https://arxiv.org/abs/2011.06125 |
id |
ftdatacite:10.48550/arxiv.2011.06125 |
---|---|
record_format |
openpolar |
spelling |
ftdatacite:10.48550/arxiv.2011.06125 2023-05-15T17:33:08+02:00 Hurricane Forecasting: A Novel Multimodal Machine Learning Framework Boussioux, Léonard Zeng, Cynthia Guénais, Théo Bertsimas, Dimitris 2020 https://dx.doi.org/10.48550/arxiv.2011.06125 https://arxiv.org/abs/2011.06125 unknown arXiv Creative Commons Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/legalcode cc-by-4.0 CC-BY Machine Learning cs.LG Artificial Intelligence cs.AI Atmospheric and Oceanic Physics physics.ao-ph FOS Computer and information sciences FOS Physical sciences Article CreativeWork article Preprint 2020 ftdatacite https://doi.org/10.48550/arxiv.2011.06125 2022-03-10T15:23:44Z This paper describes a novel machine learning (ML) framework for tropical cyclone intensity and track forecasting, combining multiple ML techniques and utilizing diverse data sources. Our multimodal framework, called Hurricast, efficiently combines spatial-temporal data with statistical data by extracting features with deep-learning encoder-decoder architectures and predicting with gradient-boosted trees. We evaluate our models in the North Atlantic and Eastern Pacific basins on 2016-2019 for 24-hour lead time track and intensity forecasts and show they achieve comparable mean average error and skill to current operational forecast models while computing in seconds. Furthermore, the inclusion of Hurricast into an operational forecast consensus model could improve over the National Hurricane Center's official forecast, thus highlighting the complementary properties with existing approaches. In summary, our work demonstrates that utilizing machine learning techniques to combine different data sources can lead to new opportunities in tropical cyclone forecasting. : Spotlight talk at NeurIPS 2021, Tackling Climate Change with AI Under revision by the AMS' Weather and Forecasting journal Article in Journal/Newspaper North Atlantic DataCite Metadata Store (German National Library of Science and Technology) Pacific |
institution |
Open Polar |
collection |
DataCite Metadata Store (German National Library of Science and Technology) |
op_collection_id |
ftdatacite |
language |
unknown |
topic |
Machine Learning cs.LG Artificial Intelligence cs.AI Atmospheric and Oceanic Physics physics.ao-ph FOS Computer and information sciences FOS Physical sciences |
spellingShingle |
Machine Learning cs.LG Artificial Intelligence cs.AI Atmospheric and Oceanic Physics physics.ao-ph FOS Computer and information sciences FOS Physical sciences Boussioux, Léonard Zeng, Cynthia Guénais, Théo Bertsimas, Dimitris Hurricane Forecasting: A Novel Multimodal Machine Learning Framework |
topic_facet |
Machine Learning cs.LG Artificial Intelligence cs.AI Atmospheric and Oceanic Physics physics.ao-ph FOS Computer and information sciences FOS Physical sciences |
description |
This paper describes a novel machine learning (ML) framework for tropical cyclone intensity and track forecasting, combining multiple ML techniques and utilizing diverse data sources. Our multimodal framework, called Hurricast, efficiently combines spatial-temporal data with statistical data by extracting features with deep-learning encoder-decoder architectures and predicting with gradient-boosted trees. We evaluate our models in the North Atlantic and Eastern Pacific basins on 2016-2019 for 24-hour lead time track and intensity forecasts and show they achieve comparable mean average error and skill to current operational forecast models while computing in seconds. Furthermore, the inclusion of Hurricast into an operational forecast consensus model could improve over the National Hurricane Center's official forecast, thus highlighting the complementary properties with existing approaches. In summary, our work demonstrates that utilizing machine learning techniques to combine different data sources can lead to new opportunities in tropical cyclone forecasting. : Spotlight talk at NeurIPS 2021, Tackling Climate Change with AI Under revision by the AMS' Weather and Forecasting journal |
format |
Article in Journal/Newspaper |
author |
Boussioux, Léonard Zeng, Cynthia Guénais, Théo Bertsimas, Dimitris |
author_facet |
Boussioux, Léonard Zeng, Cynthia Guénais, Théo Bertsimas, Dimitris |
author_sort |
Boussioux, Léonard |
title |
Hurricane Forecasting: A Novel Multimodal Machine Learning Framework |
title_short |
Hurricane Forecasting: A Novel Multimodal Machine Learning Framework |
title_full |
Hurricane Forecasting: A Novel Multimodal Machine Learning Framework |
title_fullStr |
Hurricane Forecasting: A Novel Multimodal Machine Learning Framework |
title_full_unstemmed |
Hurricane Forecasting: A Novel Multimodal Machine Learning Framework |
title_sort |
hurricane forecasting: a novel multimodal machine learning framework |
publisher |
arXiv |
publishDate |
2020 |
url |
https://dx.doi.org/10.48550/arxiv.2011.06125 https://arxiv.org/abs/2011.06125 |
geographic |
Pacific |
geographic_facet |
Pacific |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_rights |
Creative Commons Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/legalcode cc-by-4.0 |
op_rightsnorm |
CC-BY |
op_doi |
https://doi.org/10.48550/arxiv.2011.06125 |
_version_ |
1766131536826990592 |