Seasonal forecasting of tropical storms
The prediction method relies on the use of coupled ocean-atmosphere dynamical models. Due to increased complexity, such a model is run at a coarser resolution than weather prediction models and thus is unable to produce tropical storms similar to those observed in the real world.
However, it is possible to identify features in the model which are indicative of tropical storms and hurricanes; for example, low central pressure and high relative vorticity. These features can be tracked and counted to arrive at a total number of storms and hurricanes for the season.
For each model storm, we also calculate the individual accumulated cyclone energy (ACE) index and total these values over the whole season to arrive at the seasonal ACE index. The ACE index is calculated as the square of the maximum wind speed for each six hourly period of the storms lifetime. The total number of model tropical storms, hurricanes and ACE index may differ from those values seen in the real world. To make these adjustments we use a calibration procedure based on a fixed historical period