A torrential rain struck Qingyang on July 15, 2022, which is located in semi-arid region of eastern Northwest China, causing the daily rainfall and hourly rainfall at several stations to exceed the historical extreme values. The formation mechanism of the torrential rain in semi-arid region of Northwest China is analyzed based on multi-source observation data and ERA5 reanalysis data, so as to provide some useful reference for rainstorm forecast in arid and semi-arid areas. The results show that the torrential rain process occurred under the background of weak synoptic scale baroclinic forcing, weak unstable energy and deep wet layer in the complex terrain of the Loess Plateau, with characteristics of strong locality and long duration of heavy precipitation, which is a warm-sector torrential rain. The special circulation configuration of South Asian high, western Pacific subtropical high and pressure system at the lower level is conducive to the occurrence and development of mesoscale convective system. Convective initiation and development were triggered by surface wind convergence line and low-level southerly jet. Development and long-time maintenance of the low-level jet intensified surface convergence line continuously. The left side of low-level jet (rainstorm area) formed two stable secondary circulations with the right side of the exit and entrance of it, respectively, which is the key factor for the maintenance of the convective system. The release of condensation latent heat caused local frontogenesis and low-level positive vorticity development, which is another important factor for development and maintenance of convective systems, and it is also an important reason for maintenance of atmospheric instability. The mesoscale convective system exhibited deep, low center of mass and quasi-stationary characteristics under the combined effects of the above mentioned factors, the radar echoes were characterized by backward propagation and train effect.
Based on daily observation data and disaster data from automatic weather stations in Hedong region of Gansu Province from 2008 to 2017, the 75 hail cases were selected with some standards and divided into three types according to main weather influencing system of hail and mesoscale diagnostic method, including the northwestern airstream, the low trough and the low vortex types. Then statistical characteristics of radar products and radar echo feature under the three weather types of hail were compared, and the typical hail cases were chosen to verify the results. The results show that the maximum reflectivity (Zmax) and the height of maximum reflectivity ( H Z m a x) had relatively little differences under the three weather types of hail, Zmax was above 50 dBZ and H Z m a x was above 2.0 km. There were significant differences in echo top height (ET), core area thickness (H), center height of echo reflectivity above 45 dBZ (H45 dBZ), the maximum height of storm body echo reflectivity equal to or more than 30 dBZ (TOP), vertically integrated liquid water content (VIL), and the density of vertically integrated liquid water content (VILD). The H, H45 dBZ, TOP and VIL of the northwestern airstream type were significantly higher than those of other types, and ET of the low trough type and VILD of the low vortex type were lower. The frequencies of overhang echo for the three types of hail weather were all higher than 61.0%. The frequencies of three body scattering and side-lobe echo for the the low trough type were 35.5% and 48.4%, respectively, and the probability of bounded weak echo was 12.9%. The frequencies of bounded weak echo area for both of the northwestern airstream type and the low vortex type were higher than 38.5%. The indicators of three body scattering and side-lobe echo could predict the hail weather 18 to 30 minutes in advance, which had some significance in hail weather forecast.
Based on the refined numerical prediction products of the European Center of Medium-range Weather Forecast (ECMWF), precipitation guidance products from China Meteorological Administration (TP_CMA) and precipitation observation data at 340 meteorological stations of Gansu Province, the objective divisions of precipitation at 340 meteorological stations of Gansu Province from April to September during 2017-2019 were done by using spatial cluster and Tyson polygon (SCTP) approach. On this basis, the physical quantity factors related to precipitation were selected and used to build prediction model by using random forest (RF) algorithm, and the correction experiment of short-term quantitative precipitation objective forecast in Gansu Province was carried out, the forecast effect was verified. The results are as follows: (1) There were 7, 6, 14, 13, 14 and 11 precipitation regions in sequence from April to September in Gansu Province. (2) In terms of rain probability forecast, the forecast ability of SCTP-RF products in flood season (from June to August) in Gansu Province was better than that of TP_CMA guidance products and ECMWF model products, and the prediction accuracy improved by 6.1% and 4.2%, respectively. In space, SCTP-RF products had a certain ability to correct rain probability forecast at all stations of Gansu Province, and the prediction accuracy at most stations improved by 5%, especially in the east of Yellow River in Gansu (Hedong area). (3) For graded precipitation forecast, the forecast ability of SCTP-RF products to moderate rain and heavy rain was superior to TP_CMA guidance products and ECMWF model products, and the correction effect at most stations was better, especially in the middle part of Hedong and southeastern Gansu. However, the correction ability to light rain and rainstorm forecast was unstable during the heavy rainfall processes, especially to light rain in southeastern Gansu.
