On The Forecast Inconsistency And Its Sensitivity Of The Surface Variables Prediction Over East Asia

Based on the ensemble prediction data of the surface air temperature and 10m wind speed data taken from 3 ensemble prediction systems in the TIGGE archive,namely, the European Centre for Medium-Range Weather Forecasts (ECMWF), the National Centers for Environmental Prediction (NCEP) and the China Meteorological Administration (CMA), the inconsistency characteristics of the control forecast and ensemble mean forecast of the surface air temperature and 10m wind has been investigated by using jumpiness index. In addition, the method of reducing the forecast inconsistency has been discussed. The results show that:The time mean inconsistency index of each model for surface air temperature is quite different. The time mean inconsistency index of ECMWF forecast system is the smallest, followed by NCEP,and CMA is the largest. In addition,the inconsistency indices of the control forecast in the NCEP forecast system and both the control forecast and the ensemble mean forecast in the ECMWF forecast system increase with the forecast lead time. In the ensemble forecast system, the ensemble mean forecast is superior to the control forecast, especially for longer term forecast. But in the CMA forecast system, the inconsistency indices of the control forecast and the ensemble mean forecast are substantially stable at a higher level.On the average, the time mean inconsistency index of each model has similar characteristics for 10m wind speed. In the ensemble forecast system, the inconsistency indices of the control forecast and ensemble mean forecast are gradually increasing with the forecast lead time. The difference between the control forecast and ensemble mean forecast is greater in the longer term forecast. In terms of the forecast inconsistency of the three ensemble prediction systems, the ECMWF forecast is the best, followed by NCEP forecast, and CMA forecast is the worst.Additionally, the time mean inconsistency is sensitive to the forecast areas and atmospheric average state. The larger the forecast area, the better the forecast consistency performance is. The more stable the atmospheric average state, the better the forecast consistency performance is. The difference between the control forecast and ensemble mean forecast inconsistency is more sensitive to the forecast area than the atmospheric average state, especially in the longer term forecast. For different variables, the time mean inconsistency is also different. The time mean inconsistency index of the surface air temperature is the smallest, followed by 10m meridional wind, the 10m zonal wind is the largest among the surface air temperature, zonal wind (u) and meridional wind (v). However, their sensitivity to the forecast area,time and forecast variables are limited.The multiple linear regression and bias-removed ensemble mean method for multi-lead time forecast help improving the inconsistency of 10m wind speed forecast, especially in the CMA forecast system, the improvement for zonal wind is notable. However, the improvement for ECMWF and NCEP forecast system is significant only for longer term forecasts. The multi-model ensemble mean and superensemble method may improve the forecast inconsistency for 10m wind speed variable, but it is significant only for the longer term forecast.