Parameter estimation of an atmospheric model using geostationary satellite observation to improve prediction of tropical cyclones: an idealized experiment
Randomized trial estimates model parameters using satellite observations to enhance tropical cyclone predictions, suggesting significant improvements in forecasting accuracy.
Key Points
This study aims to reduce parametric uncertainty in atmospheric modeling by calibrating parameters with geostationary satellite data to enhance tropical cyclone predictions.
Developed a calibration method for meso-scale atmospheric model parameters using satellite brightness temperature observations.
Utilized machine-learning-based surrogate models and an image-processing inspired evaluation index.
Evaluated the estimated parameters' impact on the accuracy of tropical cyclone intensity predictions.
Effectively estimated parameters in cloud microphysics and boundary layer schemes from satellite observations.
Improved prediction accuracy of satellite image simulations and reduced errors in tropical cyclone intensity forecasts.
Demonstrated the advantages of using multiple parameter adjustments based on geostationary satellite data.