Evaluation And Analysis On Dynamic And Physical Performance Of GRAPES-Meso Model At Convection-permitting Resolution

Development of kilometer-scale models,also known as CPMs?convection-permitting models,which is generally considered that the horizontal grid is less than 4km?has become one of the active research fields and international hotspots in recent years.The evaluation and analysis of CPMs performance is an important basic work to improve and apply CPMs.In this paper,the dynamic and physical precipitation performance of GRAPES-Meso4.0 model at convection-permitting resolution is evaluated thoroughly in the view of horizontal kinetic energy spectrum,updraft vertical velocity,cumulant,frequency,intensity,and diurnal cycle,by using real-time forecast products of the GRAPES-Meso4.0 experimental system at 3km horizontal resolution covering the southeastern China?GRAPES-Meso4.0₃km?,against the observation precipitation data and the typical observed distribution of atmospheric kinetic energy spectrum during the summer of 2015.The forecast products over the same period and region from the operational model in the same version but at 10 km resolution?GRAPES-Meso4.0₁0km?is also used to compare and diagnose the forecast biases.Results of this study show that:1)GRAPES-Meso4.0 model does a fine job of simulating the horizontal kinetic energy spectrum in the scale of sub-synoptic scale and ?-scale.GRAPES-Meso4.0₃km makes significant improvement on the resolving capacity of multi-scale,especially in ?-scale as the resolution is increased form 10 km to 3km.6?35?X is the dissipative scale of GRAPES-Meso4.0,the dissipative scale of GRAPES-Meso4.0₁0km and GRAPES-Meso4.0₃km is about 60 km and 18 km,respectively.2)The vertical velocity range of GRAPES-Meso4.0₃km increased by an order of magnitude,the frequency spectrum of vertical velocity is broadened and the frequency extents in the direction of the vertical velocity,meanwhile,the forecast frequency of large-scale system about vertical velocity?0.5m/s below,generally corresponding to ?-scale above?,model can provide more favorable dynamic environmental conditions for meso-scale and small-scale systems compared with GRAPES-Meso4.0₁0km.The mean vertical velocity distribution profile of GRAPES-Meso4.0₃km is very similar to GRAPES-Meso4.0₁0km,but the value is significantly increased,and the vertical position center of gravity in maximum occurrence area slightly downward,closer to the observation height.3)GRAPES-Meso4.0₃km perfectly captures the characteristics of size and spatial distribution of observed mean daily precipitation and precipitation freq uency,which indicates that the higher resolution model can conspicuously correct the positive forecasting deviation of GRAPES-Meso4.0₁0km.4)The advantage of GRAPES-Meso4.0₃km in precipitation intensity is mainly manifested as the fine description of the detailed spatial distribution c haracteristics in precipitation intensity as well as the frequency and distribution of short-time strong rainfall?precipitation intensity ? 10mm/h?.However,the predicted heavy rain?general precipitation?intensity is strong?weak?,GRAPES-Meso4.0₁0km is the opposite.5)Hourly precipitation and diurnal cycle of precipitation frequency which predicted by GRAPES-Meso4.0₃km could reflect an observed bimodal characteristic of precipitation and a closer relationship between precipitation and its frequency in diurnal cycle,is significantly better than the performance of GRAPES-Meso4.0₁0km despite of the weaker peak in the afternoon?16:00BT?.6)The model resolution is increased to convection-permitting as well as the model explicitly describes cloud and precipitation process are important reasons for the comprehensive improvement of GRAPES-Meso4.0₃km's performance compared to GRAPES-Meso4.0₁0km,and the difference in the initial field of model is also a factor that can't be ignored.