An updated statistical correction scheme for tropical cyclones in the ECMWF ensemble prediction system — Australian Meteorological and Oceanographic Society

An updated statistical correction scheme for tropical cyclones in the ECMWF ensemble prediction system (#41)

Ying Jun Chen 1 , Jeffrey Kepert 1 , Saima Aijiz 1
  1. Bureau of Meteorology, Melbourne, VIC, Australia

For the past three tropical cyclone seasons, the Bureau of Meteorology has supplied bias-corrected ensemble forecasts of winds and waves for offshore regions of north west Australia to help industry manage their exposure to tropical cyclones. In contrast to most bias-correction methods, which correct for systematic errors in a field such as temperature, this system corrects for biases in a specific weather system, namely tropical cyclones. These biases occur largely because of the relatively coarse resolution of current global ensemble prediction systems (EPS).

We have recently updated the statistical scheme which underpins the bias correction, with substantial improvements in the accuracy and spread of the results. For example, intensity predictions are now highly competitive with high-resolution specialized NWP, with the added advantages of providing seamless prediction to ten days, and probabilistic forecasts.

This talk will present an overview of the tropical cyclone wind-wave ensemble system, together with verification results for the recalibrated system.

#amos2020