Performance Evaluation Of GRAPES?Meso 3km Model And Its Cloud Analysis System On Weather Prediction And The Improvement Scheme For Moisture Adjustment In The Northwest China

The performance of GRAPES?Meso model with 3 km horizontal resolution and its cloud analysis system were tested and evaluated through numerical prediction experiments with 13 heavy rainfall cases in the summer of 2018 and one-month batch test in July 2018 in the Northwest China.Considered that the importance of moisture(q_v) accuracy on model predictions,two moisture adjustment schemes in the cloud analysis system were evaluated with 8 heavy rainfall cases.On the basis of this,a new moisture adjustment scheme was reconstructed for numerical weather predictions(NWPs)in the Northwest China,whose effectivity and efficiency were tested with eight heavy rainfall cases and one-month batch test.The main conclusions are as follows:(1)The GRAPES?Meso 3 km model has a good and stable skill in precipitation forecasting in the Northwest China.The result of 13 heavy rainfall cases and one-month batch test showed that the model can reproduce the precipitation process,which can provide data support for short-term weather forecasting and nowcasting.Meanwhile,the model can predict the 2 m temperature well and 10 m wind to some extent,but the forecast skill of 10 m wind is not as good as that of the 2 m temperature.(2)The introduction of cloud analysis system into the model can improve the weather forecast ability of the model in Northwest China.Using radar and satellite observation data,cloud analysis system can reasonably adjust the hydrometeor and water vapor field,and consequently reduce spin-up problem and improve the forecasting skill of precipitation,radar reflectivity,2 m temperature and 10 m wind.However,the use of cloud analysis system may strengthen reflectivity intensity.(3)A new moisture adjustment scheme(named NewRH scheme)was constructed for NWPs in the Northwest China.The use of different moisture adjustment scheme in cloud analysis system has a significant effect on precipitation forecast.The sensitivity experiments also showed that precipitation forecast is sensitive to parameters in the moisture adjustment scheme,even for in the same scheme.It implies that the accuracy of water vapor plays an important role on NWPs.The reconstruction method provides an effective approach to obtaining an appropriate moisture adjustment scheme in a cloud analysis system for other regions or NWP models.(4)The use of NewRH scheme further improves the forecasting skill of precipitation in the Northwest China.Compared with the original moisture adjustment scheme,NewRH scheme can obtain the spatial pattern of variable such as hydrometeor more reasonablely.The precipitation particle content and water vapor field correspond to observed precipitation,and consequently the precipitation pattern can be described in detail.The results of eight heavy rainfall cases and one-month batch tests showed that the NewRH scheme overperformed the original moisture adjustment scheme in precipitation forecast.(5)The use of NewRH scheme can further improve the forecasting skill of 2 m temperature and 10 m wind.The results of 8 heavy rainfall cases showed that the correlation coefficients between observations and the forecasted by NewRH scheme for2 m temperature and 10 m wind are obvious higher than those by the original moisture adjustment scheme,and their corresponding error are lower.The one-month batch results with NewRH scheme are slightly worse than that of 8 heavy rainfall cases.Both for 2 m temperature and 10 m wind,the correlation coefficient obtained from NewRH scheme is higher than that from the original scheme during 0-12 h.However,the error of 2 m temperature is slightly higher than the original,and the error of 10 m wind speed is similar to that from the original scheme,which may be due to the fact that the data samples used to build NewRH scheme are derived from the heavy rainfall case.