Short-range ensemble forecasting of an explosive cyclogenesis with a limited area model

摘要

Since the atmosphere is a chaotic system, small errors in the initial condition of any numerical weather prediction (NWP) model amplify as the forecast evolves. To estimate and possibly reduce the uncertainty of NWP associated with initial-condition uncertainty (ICU), ensemble forecasting has been proposed which is a method of, differently from the traditional deterministic forecasting, running several model forecasts starting from slightly different initial states. In this dissertation, the impact of ICU and short-range ensemble forecasting (SREF) on quantitative precipitation forecasts (QPFs), as well as on sea-level cyclone position and central pressure, is examined for a case of explosive cyclogenesis that occurred over the contiguous United States. A limited-area model (the PSU/NCAR MM4) is run at 80-km horizontal resolution and 15 layers to produce a 25-member, 36-h forecast ensemble. Lateral boundary conditions for the MM4 model are provided by ensemble forecasts from a global spectral model (the NCAR CCM1). The initial perturbations of the ensemble members possess a magnitude and spatial decomposition which closely match estimates of global analysis error, but they were not dynamically-conditioned. Results for 80-km ensemble forecast are compared to forecasts from the then operational Nested Grid Model (NGM), a single 40-km MM4 forecast, and a second 25-member MM4 ensemble based on a different cumulus parameterization and slightly different initial conditions. Acute sensitivity to ICU marks ensemble QPF and the forecasts of cyclone position and central pressure. Ensemble averaging always reduces the rms error for QPF. Nearly 90% of the improvement is obtainable using ensemble sizes as small as 8-10. However, ensemble averaging can adversely affect the forecasts related to precipitation areal coverage because of its smoothing nature. Probabilistic forecasts for five mutually exclusive, completely exhaustive categories are found to be skillful relative to a climatological forecast. Ensemble sizes of --, 10 can account for 90% of improvement in probability density function. Our results indicate that SREF techniques can now provide useful QPF guidance and increase the accuracy of precipitation, cyclone position, and cyclone's central pressure forecasts. With current analysis/forecast systems, the benefit from simple ensemble averaging is comparable to or exceed that obtainable from improvement in the analysis/forecast system.