The Impact of Initial Condition Uncertainty on Numerical Simulations of Large-scale Explosive Cyclogenesis

Abstract

The impact of initial condition uncertainty on short-range (up to 48 h) forecasts of large-scale explosive cyclogenesis is examined. Predictability experiments are conducted on 11 cases of rapid oceanic cyclogenesis that occurred in a long-term, perpetual January integration of a global, high-resolution, spectral model. Results are derived from the 11-case ensemble average. The perturbation used to represent the initial condition error in this study has a magnitude and spatial distribution that closely matches estimates of global analysis error. Results from the predictability experiments are compared to a set of physics sensitivity experiments which are used to represent an estimate of a “typical” modeling, error. Compared to the control simulations, the inclusion of initial error produces a composite cyclone with maximum deepening rate that is slightly reduced and a 24 h period of most rapid deepening that is somewhat delayed. The absolute position error in the surface cyclone is approximately 100 km the first +36 h of the forecast then abruptly increases to 300 km by +48 h. We estimate that, on the average, the forecast error due to initial condition uncertainty is as large as that due to the modeling error associated with today's best operational models, whereas five years ago modeling error was much more important. The relative importance of initial condition uncertainty for explosive cyclogenesis is compared to that for the entire midlatitude flow in general. Error growth rates in an explosive cyclogenetic environment are 50% greater in the upper troposphere (500 mb and above) and two times faster near the surface (850 mb and below). The rapid growth rates indicate that short-range forecasts of explosive cyclogenesis are much wore sensitive to initial error than those for ordinary flows. The case-to-case variability exhibited by the 11-member ensemble is examined. Noteworthy departure from the aggregate results are evident In individual cases, initial condition error can lead to short-range forecast differences which can be either greater than those due to a typical modeling error or much less. This variability implies a strong sensitivity to initial condition perturbation location and structure.