In order to assess the practical feasibility of a dynamical one-month forecast, we performed several sets of ensemble forecasts using the lagged average forecast method. The results obtained by verifying the eight wintertime cases are summarized as follows: (1) The eight-case average anomaly correlation coefficient of the ensemble forecast for the 1-30 day mean 500 hPa geopotential height in the Northern Hemisphere is 0.54, and its average root mean square error is 55 m. The skill of the ensemble forecast is better than that of both the ordinary dynamical forecast starting from the latest intitial state and the climate forecast. So it seems that the one-month forecast using the lagged average method is effective for practical use. (2) For the 10-day mean 500 hPa geopotential field, the merit of ensemble averaging is apparent after the second 10-day mean (11-20). But in the third 10-day mean (21-30), the average root mean square error is slightly larger than that of the climate forecast. So its skill is marginal for practical use.
ASJC Scopus subject areas