ERROR GROWTH AND ESTIMATES OF PREDICTABILITY FROM THE ECMWF FORECASTING SYSTEM

Examination has been made of the skill of ECMWF forecasts of the 500 h Pa height field produced daily out to ten days ahead, verifying in the period from 1 December 1980 to 31 May 1994. Over this time accuracy h as been improved substantially over the first half of the forecast ran ge. The systematic (seasonal-mean) component of the error has been gre atly reduced at all forecast times, but there has been little reductio n in the non-systematic (transient) component later in the range. The simple model proposed by Lorenz for the intrinsic growth of forecast e rror has been applied to the evolution of differences between consecut ive forecasts. The implied growth-rates of small forecast errors have increased significantly since 1981. They do not show much variation wi th season, and are a little lower in the southern than in the northern hemisphere. The most recent error-doubling times are around 1.5 days for the northern hemisphere and 1.7 days for the southern hemisphere. Error saturation levels are at present similar to or greater than thos e of the 1981 version of the model, having been significantly lower in intermediate years. The accuracy of recent short- and early medium-ra nge forecasts and realism of the climatology of the forecast model sup port the view that estimates of intrinsic error-growth parameters from the current forecasting system are more reliable than those obtained earlier. Forecast accuracy later in the medium range may thus not have benefited fully from improvements earlier in the range because of the faster error-growth associated with a more active, though more realis tic, forecast model. Overprediction of variance may nevertheless detri mentally affect present levels of skill and estimates of predictabilit y in all seasons other than summer. The error-growth model currently i ndicates that it is possible, in principle, to make deterministic medi um-range forecasts for the extratropical 500 hPa height field of the n orthern hemisphere that are as accurate five days ahead as present for ecasts are three days ahead, provided the one-day forecast error can b e reduced by the same factor in the future as has actually been achiev ed in the years since 1981. The level of error currently reached at da y seven would then be reached at around day ten. The scope for improve ment of forecasts for the southern hemisphere appears to be rather lar ger. Improvements seem to be possible throughout the spectral range st udied, up to total wave-number 40. This is found also for the rotation al and divergent wind components at 850 and 200 hPa. For these compone nts, particularly the divergent component, there is a quite pronounced error in the representation of the largest scales.