Abstract: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 absolute 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.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to improve the track and intensity prediction of tropical cyclones (such as hurricanes). Specifically, the authors propose a new multimodal machine - learning framework (called Hurricast), aiming to combine multiple machine - learning techniques and different data sources to improve the accuracy of tropical cyclone track and intensity prediction within 24 hours. ### Main problems: 1. **Limitations of existing models**: Current operational forecasting models (including dynamical models, statistical models, and statistical - dynamical models) can provide effective predictions to a certain extent, but there is still room for improvement in terms of computational efficiency and accuracy. 2. **Integration of multiple data sources**: Existing methods usually rely on a single type of data (such as historical data or reanalysis data) and fail to fully utilize the combined advantages of multiple data sources. ### Solutions: The Hurricast framework solves the above problems in the following ways: - **Multimodal data fusion**: Hurricast combines spatio - temporal data (such as atmospheric reanalysis maps) and statistical data (such as historical storm data), and extracts features through a deep - learning encoder - decoder architecture, and uses Gradient Boosting Trees (XGBoost) for prediction. - **Efficient prediction**: This framework can complete the prediction within seconds while achieving a mean - square error (MAE) and skill score comparable to existing operational forecasting models. - **Performance improvement**: Experimental results show that incorporating Hurricast into the consensus model can further improve the forecasting performance, surpassing the official forecasts of the National Hurricane Center. ### Experimental verification: The authors conducted 24 - hour - ahead forecasting experiments from 2016 to 2019 in the North Atlantic and the East Pacific basins, verified the effectiveness of Hurricast, and demonstrated its advantages over existing models. ### Significance: This research proves that using machine - learning techniques in combination with different types of data sources can bring new opportunities in tropical cyclone forecasting, not only improving the accuracy of short - term forecasting, but also providing the possibility of improvement for long - term forecasting, thereby reducing forecasting uncertainty and benefiting society. ### Formula representation: - **Mean - square error (MAE)**: \[ \text{MAE}=\frac{1}{N}\sum_{i = 1}^{N}\left|y_{\text{true}, i}-y_{\text{pred}, i}\right| \] where \(N\) is the number of predictions, \(y_{\text{pred}, i}\) is the wind speed predicted 24 hours in advance, and \(y_{\text{true}, i}\) is the true value of the 1 - minute maximum sustained wind speed provided by the WMO. - **Geographical distance error (Haversine formula)**: \[ d = 2r\arcsin\left(\sqrt{\sin^{2}\left(\frac{\phi_{2}-\phi_{1}}{2}\right)+\cos(\phi_{1})\cos(\phi_{2})\sin^{2}\left(\frac{\lambda_{2}-\lambda_{1}}{2}\right)}\right) \] where \(r\) is the radius of the earth, \(\phi_{1},\phi_{2}\) are latitudes, and \(\lambda_{1},\lambda_{2}\) are longitudes. Through these methods and formulas, the Hurricast framework demonstrates its potential and superiority in tropical cyclone forecasting.

Hurricane Forecasting: A Novel Multimodal Machine Learning Framework

A multimodal deep learning approach for hurricane tack forecast based on encoder-decoder framework

Enhanced Rainfall Nowcasting of Tropical Cyclone by an Interpretable Deep Learning Model and Its Application in Real-Time Flood Forecasting

HurriCast: An Automatic Framework Using Machine Learning and Statistical Modeling for Hurricane Forecasting

Storm Surge Modeling in the AI ERA: Using LSTM-based Machine Learning for Enhancing Forecasting Accuracy

Tropical Cyclone Track Forecasting Using Fused Deep Learning From Aligned Reanalysis Data

Structural Forecasting for Tropical Cyclone Intensity Prediction: Providing Insight with Deep Learning

A Neural Network Framework for Fine-Grained Tropical Cyclone Intensity Prediction

A Review of Application of Machine Learning in Storm Surge Problems

Harnessing Diverse Data for Global Disaster Prediction: A Multimodal Framework

A Novel Data-Driven Tropical Cyclone Track Prediction Model Based on CNN and GRU with Multi-Dimensional Feature Selection

A Bayesian Hierarchical Model Framework to Quantify Uncertainty of Tropical Cyclone Precipitation Forecasts

A Novel Tropical Cyclone Track Forecast Model Based on Attention Mechanism

Global Flood Prediction: a Multimodal Machine Learning Approach

A Spatio-Temporal Graph-Guided Convolutional LSTM for Tropical Cyclones Precipitation Nowcasting.

Tropical Cyclone Intensity Prediction by Inter- and Intra-Pattern Fusion Based on Multi-Source Data

A real time prediction methodology for hurricane evolution using LSTM recurrent neural networks

A Novel Hybrid Machine Learning Model for Rapid Assessment of Wave and Storm Surge Responses Over an Extended Coastal Region

A Convolutional Neural Network-based Ensemble Post-processing with Data Augmentation for Tropical Cyclone Precipitation Forecasts

A Novel Deep Learning Approach for Tropical Cyclone Track Prediction Based on Auto-Encoder and Gated Recurrent Unit Networks