| In this paper, a comparative analysis between two successive torrential rain events occurred on 31 July and 1 august 2006 in Beijing was conducted with usage of a large number of available observations, such as high spatial-temporal resolution data from auto-weather station network, observations from microwave radiometer, vertical distribution of wind from a boundary wind profile, Doppler radar data, satellite images, TBB data, conventional weather observational data at a time interval of 3h, and NCEP reanalysis data with an interval of 6h, etc. Observational results from this study are listed as follows.1) It has been revealed that the dominant synoptic system of the case on 500hPa are a deepen trough and a cut-off one, both of which are developed from Hetao trough in its eastward movement but differ in developing phases. Although the two rainstorms show a similar large-scale circulation pattern, there are still differences in detail. Compared with the first heavy rainfall, the long distance vapor transportation of the second one is obviously weaker, and thus confronting situation between dry and cold atmosphere with warm and wet one died down. Besides, the environmental condition available for a greater perturbation above Beijing disappeared when the second rain occurred. The first rainfall can be concluded as a typical trough rainfall with a cold front in North China, while the other one is deduced from convections under the control of lower trough rather than a whole regional torrential rain.2) Analyses of the characteristics in the rain-masses in the two events show common in obvious meso-β-scale activities, short duration, severe intensity, and the maximum rainfall occurring in urban. Meantime, differences are distinct on their mesoscale characters in the aspects of influencing domains, shapes and moving paths, life times and influence times respectively. The former one in larger cover shaping is produced by a bigger air mass with a diameter of 60km moving from southwest to northeast continuously, and its torrential rain happened in the morning with a life time around 5-6 hours. The later one, shaped in line with a smaller domain, is brought by movement and new occurrence of isolated air mass with a diameter around 20-30km. Rainstorm in the second this event presents from afternoon to dusk, while the life-span for its air mass is shorter, around 3 hours.3) From an analysis of those data, the two rainfall events are found to be induced by meso-β-scale convective systems. The first storm's MCS initiated at the crossing of the upper level cold frontal cloud system and the cloud belt of the northwest edge of the subtropical high. The second storm's MCS had the characteristics of a squall line solicited by heat convection of cold eddy in the afternoon. The analysis of the radar data and information from auto-weather stations show that the developments of the convective echoes are associated with the development of the surface mesoscale convergence line.4) Analyzing the local physical conditions when the two storms happening by using Microwave Radiometer and wind profile data of Haidian and observatory, a phenomenon was found, which includes that the temperature of the bottom cloud increased suddenly several hours before the storm, relative humidity kept high to above 90%, and water vapor content and liquid water content were fold to increase before the storm. Variations of the above parameters happened all the same time in the first torrential rain, but not in the second one. Therefore, it can be concluded that cloud system is more systematical and grows larger with a non-local precipitation on 31st July, whereas cloud development and precipitation on 1st August is more local.5) Observed wind profiles in troposphere are used to analyze the vertical and temporal changes of horizontal winds in local convective layers before and after torrential rains. It is found that distinctions exist, even though the occurrences of the two events are both associated with the build-up of near east wind in a short time period in the local boundary layers. Near east flow came into exist and direction veered along height several hours before the first heavy rain, with two maximum values at around 1500-1800m and 300-600m. The former near east wind layer grew upward around 3000m just one hour before the torrential rain, which made the control of near east wind more intense in the vertical direction. While in the second event, southeast wind appeared 4-5 hours before the rainfall near the surface layer and developed from 900m up to 1200m. Therefore, the southeast flow in the second rainfall is less systematical, the vertical layer is relatively thin, and wind speed is smaller than the near east wind in the first one.6) The microwave radiometer data is used to calculate the value of CAPE with 20m intervals and the value of LCL with 1h intervals. It is found that LCL changed a little between the surface of 950hPa and that of 970hPa during the storm, rather than maintaining on the same height. The small increase was accompanied by the decline in precipitation rate, while severe increase by the end of a rainstorm. Positive CAPE with a large value (2000-3000J/kg) existed in both the two rainfalls under 300hPa, and the value was bigger in the first events which may be one reason for its stronger intensity. Release and reassociate processes of CAPE in deep convective layers in troposphere in the two torrential rain events. The whole CAPE release lasted about 1-1.5 hour.7) The direct reason for the first rainfall is analyzed to be the coupled occurrence of warm shear of 850hPa in front of trough approaching northward towards Beijing and the build-up ofβlow pressure circulation which was located along Shijiazhuang to Beijing. The second rainfall is deduced from a southeast moving squall line which was developed from isolated weather-side slope convective systems under a warm advection environmental and Beijing's special terrain condition, and then well organized from a positive mesoscale vorticity disturbance on 500hPa.8) From the evolvements of local physical variables when storm happened and before or after the storm, some forecasting characters for the approaching prediction of storm or heavy rainfall in Beijing are found. For example, the significant decrease of the cloud bottom down to 500-600m with a consistant near east flow in lower troposphere (at least boundary layer) gives a big possibility of severe precipitation in the future few hours. What's more, torrential rain fall in the following ten minutes is very likely if liquid water increases at fold. More attention should be paid on this aspect in weather detecting and forecasting.
|
PDF Full Text Request