Impact of perturbation methods in the ECMWF ensemble prediction system on tropical cyclone forecasts



In the operational configuration of the ensemble prediction system of the European Centre for Medium-Range Weather Forecasts, different methods are applied to account for initial condition and model uncertainties. Singular vectors and an ensemble of data assimilations are used to generate perturbations to the initial conditions. Two stochastic tendency perturbation schemes aim to mimic model errors: the stochastic kinetic-energy backscatter scheme and the stochastically perturbed parametrization tendency scheme. In this study, the impact of the different perturbation methods on the ensemble spread during tropical cyclone (TC) events is compared and the time evolution of the spatial perturbation structure is investigated. The structure of perturbations due to different methods initially is quite different. However, our results show that they converge toward a TC displacement and an intensity-change pattern during the first two days of the forecast, on average. The perturbations generated by the stochastic tendency perturbation schemes grow rapidly and effectively excite growing modes of the flow. After 48 hours, a large part of the total energy of the singular vector perturbations in the vicinity of the TCs can be explained by the perturbations of other methods. The perturbations by the ensemble of data assimilations dominate the ensemble spread for a rather short lead time (around 24 hours). In about 40% of cases, the perturbations due to the tendency perturbation schemes or those generated from singular vectors produce a larger TC track and central pressure spread than the perturbations by the ensemble of data assimilations after two days forecast time. If all methods are applied, the average TC track spread of the ensemble matches the average error of the ensemble-mean well. In addition, the ensemble captures the anisotropy in the position uncertainty of the TCs. Copyright © 2012 Royal Meteorological Society