An Experimental Study on Monthly Numerical Weather Prediction
月数值天气预报的试验研究

Experiments of monthly dynamical extended-rang prediction with the global spectralmodel developed by Lab II, IAP are performed. The results show that the 30-day mean 500 hPa heightforecasts are obviously superior not only to the persistence but also to climate forecasts. In this paper,we also discuss how to get a better monthly mean height forecast from day-by-day prediction, whichindicates that forecast skill is enhanced markedly when weighted mean forecast is used instead of arithmetic one.