Tropical cyclone dataset for a high-resolution global nonhydrostatic atmospheric simulation

Area of Science:

• Atmospheric Science
• Climate Modeling
• Meteorology

Background:

• Tropical cyclones pose significant threats globally.
• Accurate simulation and prediction of tropical cyclones are crucial for disaster preparedness.
• Existing datasets may lack the resolution or comprehensive life-cycle data needed for advanced modeling.

Purpose of the Study:

• To create a comprehensive dataset of tropical cyclone tracks and associated atmospheric data.
• To facilitate the development of machine learning models for tropical cyclone analysis.
• To support research in tropical cyclone detection, intensity, and cyclogenesis prediction.

Main Methods:

• High-resolution Nonhydrostatic Icosahedral Atmospheric Model (NICAM) simulations were conducted over 30 years.
• Tropical cyclone tracks were identified and extracted, encompassing precursor (pre-TC), tropical cyclone (TC), and post-tropical cyclone (post-TC) stages.
• Atmospheric field data, including wind, pressure, and temperature, were compiled for each track.

Main Results:

• A dataset of 2,463 tropical cyclone tracks was generated.
• Each track includes detailed information on time, location, intensity (wind speed, pressure), and life stage.
• Associated atmospheric data covers a 1000 km² area centered on the cyclone for each time step.

Conclusions:

• The generated dataset provides a valuable resource for tropical cyclone research.
• It enables the training and validation of machine learning algorithms for improved tropical cyclone forecasting.
• This dataset can advance our understanding of tropical cyclone dynamics and prediction capabilities.