An Meso-scale Ensemble Forecasting Research Based On Perturbation Of Physics Progress

According to the uncertain of cumulus convective parameters in meso-scale model forecasting, using WRP model do physical process perturbation ensemble forecasting experiments which occurred in Huaihe river during July 4th and 5th.The sensitive and ensemble researches based on mass-fluxed cumulus convective parameterization of Kain-Fritsch-eta and adjusted cumulus convective parameterization of Betts-Miller-Janjic, which are popular used by meso-scale models. The main results are:(1) In KFE sensitive experiments, the minimum net environment entrainment rate and cumulus radius have more effect on explicit precipitation than implicit precipitation. And the scope and strength of jet center and the total energy have direct relationship with the above parameters. The complicit difference of variable vertical profile exits between 800hPa and 150hPa.(2) In BMJ sensitive experiments, the weighting coefficient and the deep convective adjustment time have effect on precipitation. They also reveal that the total energy have directive relationship with the above parameters. The linear difference of variable vertical profile exits between 800hPa and 150hPa.(3) The predictability of position and scope of precipitation are improved by "KFE" ensemble mean . The ensemble mean also meliorate the bias of "too dry" and "too moistening". The spread of precipitation field shows that the most uncertain predictability centralized on the area with complicated terrain or strongest rainfall.(4) The "BMJ" ensemble mean improves the skill of predictability by decrease the amount of rainfall, and meliorate the bias of "dry".(5) The results of bias score and ETS score reveal that, the predictability of ensemble system is generally better than single seldom forecast or several parameterizations averaged ensemble forecast.