Extended-range probability forecasts based on dynamical model output

A probability forecast has advantages over a deterministic forecast as the former offers information about the probabilities of various possible futur e states of the atmosphere. As physics-based numerical models find their su ccess in modern weather forecasting, an important task is to convert a mode l forecast, usually deterministic, into a probability forecast. This study explores methods to do such a conversion for NCEP's operational 500 mb-heig ht forecast and the discussion is extended to ensemble forecasting. Compare d with traditional model-based statistical forecast methods such as Model O utput Statistics, in which a probability forecast is made from statistical relationships derived from single model-predicted fields and observations, probability forecasts discussed in this study are focused on probability in formation directly provided by multiple runs of a dynamical model-eleven 00 00 UTC runs at T62 resolution. To convert a single model forecast into a strawman probability forecast (si ngle forecast probability or SFP), a contingency table is derived from hist orical forecast-verification data. Given a forecast for one of three classe s (below, normal, and above the climatological mean), the SFP probabilities are simply the conditional (or relative) frequencies at which each of thre e categories are observed over a period of time. These probabilities have g ood reliability (perfect for dependent data) as long as the model is not ch anged and maintains the same performance level as before. SFP, however, doe s not discriminate individual cases and cannot make use of information part icular to individual cases. For ensemble forecasts, ensemble probabilities (EP) are calculated as the percentages of the number of members in each cat egory based on the given ensemble samples. This probability specification m ethod fully uses probability information provided by the ensemble. Because of the limited ensemble size, model deficiencies, and because the samples m ay be unrepresentative, EP probabilities are not reliable and appear to be too confident, particularly at forecast leads beyond day 6. The authors hav e attempted to combine EP with SFP to improve the EP probability (referred to as modified forecast probability). Results show that a simple combinatio n (plain average) can considerably improve upon both the EP and SFP.