Abstract:To investigate the effect of the ensemble transform Kalman filter with rescaling (ETKF_R) initial perturbation method on typhoon forecasting, we perform retrospective experiments from 18 to 29 July 2021 using China Meteorological Administration-Regional Ensemble Prediction System (CMA-REPS). The effect of the initial perturbation structure on the track and intensity forecasts of Typhoon In-Fa is analyzed and compared with the ECMWF and NCEP global ensemble forecasts. The results can be summarized as follows. The ETKF_R method improves the amplitude and structure of initial three-dimensional wind field perturbations, but the initial ensemble spreads of typhoon location and intensity are small. By reasonably reducing the ensemble spread of weather system which significantly influences the forecast of typhoon track, ETKF_R can constrain the excessive dispersion of typhoon translation speed and direction. This further improves the ensemble mean track forecast skill for the whole life of Typhoon In-Fa and the relationship between ensemble mean error and ensemble spread of typhoon track. In ETKF_R, the ensemble spreads of typhoon structure and intensity grow rapidly in the first 24 h, and the performance of ensemble mean intensity forecast after 24 h is comparable to that of the ETKF method without rescaling. Compared with the international advanced global ensemble forecasts, ETKF_R has the best landfalling forecast of Typhoon In-Fa. The statistically averaged 0-2 d track forecast error of ETKF_R is comparable to that of ECMWF ensemble. Although NCEP ensemble has the smallest 0-2 d track forecast error, its overdispersed feature is obvious. Meanwhile, ECMWF ensemble generally underestimates the intensity of Typhoon In-Fa, while NCEP ensemble has a high accuracy in predicting the maximum intensity of Typhoon In-Fa, with a slower intensification speed than ETKF_R. Our results suggest that the forecast of typhoon track and intensity by CMA-REPS has operational significance of reference.

Abstract:Based on multi-source observation data, the radar-echo characteristics of wide-range damaging thunderstorm gale events and the associated convective systems in Shandong Province and surrounding areas from 2005 to 2021 are analyzed. The results show that there were 41 wide-range damaging thunderstorm gale events in the 17 years, with an annual average frequency of 2.4 times, and they mainly occurred in June. Before the occurrence of damaging thunderstorm gale, there tends to be remarkable conditional instability and moderate-to-slight humidity condition in the middle and lower troposphere. Meanwhile, there is a prominent dry layer in the middle level, and the vertical wind shear is moderate-to-slightly strong. The convective systems that lead to wide-range damaging thunderstorm gales can be divided into four categories: type Ⅰ squall line (cell resolvable type), type Ⅱ squall line (banded-echo type), multicell storm cluster and weak echo squall line. The rear inflow jet carries dry-cold air into the squall line and enhances the negative buoyancy through evaporative cooling effect, which is an important formation mechanism of damaging thunderstorm gale generated by type Ⅰ and type Ⅱ squall lines. The backward (or rightward) propagating multicell storm clusters all contain supercells, whose gust front can both produce damaging thunderstorm gale and trigger new storms. Due to the fast moving speed of the severe storm and the possibly-existing rear inflow jet, as well as the downward transport of momentum from high levels caused by the downdraft in the precipitation, the asymmetric downburst can occur, which increases the possibility of extreme thunderstorm gales.The damaging thunderstorm gales generated by the weak echo squall lines are most likely to be ignored. The convective subsystems that directly cause damaging gales can be classified into five categories, namely bow echo, strong single cell storm, supercell storm, gust front and mixed type, accounting for 30%, 26%, 6%, 23% and 16%, respectively. The average values of the maximum wind speed caused by bow echo and supercell are the largest, reaching 28.2 m·s-1 and 29.9 m·s-1 respectively. The formation of bow echo can be predicted about 20 minutes in advance according to strong rear inflow jet and remarkable mid altitude radial convergence (MARC). The damaging thunderstorm gale mainly occurs in the central and left parts of the bow echo relative to its moving direction. Extreme thunderstorm gales with wind speed over 32.6 m·s-1 are generated by the line-echo wave pattern embedded with bow echoes, the combination of bow echo and mesoscale vortex, and supercells. Thunderstorm gales at the speed about 30 m·s-1 can be produced by the strong single cell storms with deep MARC, the gust fronts in strongly developed squall lines, and the superposition of surface cold front and gust front which is often accompanied by downward momentum transportation.

Abstract:Based on the ECMWF ensemble prediction system (EC-EPS) and the potential vorticity tendency (PVT) diagnosis, this paper examines the contributions of physical factors to track forecast of Typhoon Chanthu (2114) by comparing two distinct ensemble members. It is found that the typhoon centers of two ensembles both move towards the area with the largest wave number one component (WN1) of PVT. The contribution of the horizontal advection term is dominant most of the time, followed by the diabatic heating term, and the contribution of the vertical advection term is relatively small. The PVT discrepancy of the two members at bifurcation time is mainly attributed to the diabatic heating term, while that after the bifurcation time is still dominated by the horizontal advection term (i.e., steering flow). The asymmetric convective activities of two ensemble members also show different distributions at bifurcation time, which is consistent with the direction of diabatic heating, that is, the asymmetric convective structure and the vertical gradient of diabatic heating may play a key role in the movement of typhoon when the steering flow is weak, which is largely affected by the environmental factors such as water vapor, potential instability, and low-level jet. This research emphasizes the impact of diabatic heating processes (including development and asymmetric structure of convective activities) in typhoon track forecast for more accurate prediction when steering flow is weak.

Abstract:Based on the upper-air, surface observation and ERA5 hourly reanalysis data, the circulation situation and environmental conditions of two snow and rain events in Hubei Province in 2022 are compared and analyzed, and the errors and causes of the two events in the prediction of precipitation type, snowfall and snow depth are discussed. The results show that the trough of the 27-29 January 2022 event was westward, the jet stream was relatively weak and the maintenance duration was short. The temperature in the middle and lower troposphere was high, dominated by warm advection. The ice crystal content was low, so the precipitation type was dominated by rainfall. The main errors of snowfall were due to the deviation of precipitation type forecast, the forecast of strong low-level jet stream by the model, which led to the total snowfall amount less than that of the forecast, and the insufficient forecast of snow melting mechanism led to snow depth lower than that of the forecast. In the 6-7 February 2022 event, the trough was easterly, the jet was relatively strong and maintained for a long time, the temperature in the middle and lower troposphere was lower, the ice crystal content was high, and the relative humidity in the lower layer was low, so the precipitation was mainly dominated by snowfall. The prediction of snowfall amount and precipitation type by the ECMWF model were close to obsevations, but the high boundary layer temperature prediction and insufficient snow melting ability prediction were the main reasons for the errors of snow depth change. Therefore， correction of model snowfall amount and snow depth should focus on ice crystal content, boundary layer temperature and snow melting capacity by model.

Abstract:Based on automatic weather station, dual-polarization radar, millimeter-wave cloud radar, temperature profiler and ERA5 hourly reanalysis data, the characteristics of the multi-source observation data of the precipitation phase variation during the extremely heavy snowstorm in the western area of Shandong on 7 November 2021 are analyzed. The results show that this process was commonly influenced by the westerly trough, low vortex shear line and cold front. The system had strong baroclinicity, with low-level jet. Though the snowfall duration was not long, the hourly intensity of snow precipitation was high, resulting in extremely heavy snowfall in the western part of Shandong. The multi-source observational data captured the precipitation phase and change information well in this process. Wind and vertical velocity from wind profiler radar, correlation coefficient and 0℃ layer bright band of dual-polarization radar, reflectivity, vertical liquid water content, velocity and spectral width of millimeter-wave cloud radar, boundary layer temperature of temperature profiler and the two-dimensional images and information of particles from the 2D video raindrop spectrometer were used as indicators for judging the phase of the precipitation, so these information provided a reference for the 0-1 h short-term nowcasting. During this process, air temperature above the ground cooled very rapidly, the rain quickly turned to snow with the sleet only lasting for a short time. The cooling of air temperature above the ground was caused by the combination of strong negative temperature advection, vertical motion and non-adiabatic effect, of which, the effect of negative temperature advection was the largest, followed by precipitation diabatic effect, and the vertical motion effect in order. The strong negative temperature advection and the negative temperature advection intensity increasing with the decrease of height, led to the rapid cooling of air temperature above the ground.

Abstract:The spatio-temporal statistical downscaling of precipitation has always been a difficult research point in intelligent grid forecasting. By comparing several spatio-temporal downscaling methods, we obtained the optimal scheme suitable for the Beijing Region in this study. In terms of spatial statistical downscaling, spatial distribution characteristics of five methods are compared, namely inverse distance weighting, ordinary Kriging, Nearest, Bilinear, 3-D ordinary Kriging method, and the results show that Bilinear interpolation method has the best application effect in Beijing Region, with the smallest error and highest ETS. For temporal downscaling, two kinds of allocation schemes based on regional numerical model are compared, including hourly allocation schemes (RMAPS and CMA-MESO) and average allocation. The results show that there are no significant differences in RMSE and MAE for the three methods. The hourly allocation by RMAPS is more preponderant than by observation in ETS score, and it performs also better in heavy rainfall cases, which means the allocation by RMAPS has better advantages from the perspective of forecast accuracy. The schemes of bilinear interpolation and RMAPS hourly allocation are taken as the spatio-temporal downscaling schemes in the objective forecasting technique of Beijing Meteorological Observatory, and can support intelligent grid forecasting for providing refined forecast products. The results can provide some references for forecasting and associated researches.

Abstract:One of the critical challenges in the retrieval of three-dimensional wind fields in precipitation systems is the accurate estimation of the final falling velocity (Wt) of precipitation particles. To assess the ability of dual-polarization radar in estimating Wt, the relationship between Wt and dual-polarization radar in the S and X bands is established in this paper based on the raindrop spectrum data collected from the Longmen Area of Guangdong Province. This relationship is then applied to the wind field retrieval of the Guangzhou and Shaoguan radars. In addition, an experiment is conducted to invert a squall line event that occurred in South China in April 2019. The wind field structure of the squall line process is analyzed, and the difference in the wind field structure of Wt in retrieval performance is estimated by different methods. The results indicate that the functions of Wt estimated by S and X band radars using echo intensity (ZH) and differential reflectivity factor (ZDR) are in the forms of a power function and a primary function, respectively. The root mean square error of Wt estimated by ZDR is smaller than by ZH, and the correlation coefficient is larger, which suggests that it is better to estimate Wt by ZDR. The squall line process primarily developed from northwest to southeast, with the wind field dominated by westerly and southwesterly winds. There was a distinct convergence zone in the arcuate echo area in the front of the squall line, which had a perpendicular structure with low-level convergence and high-level divergence. Compared to the three-dimensional wind field obtained by estimating Wt with different methods, the changes in the horizontal wind field are primarily concentrated in the range of ±1 m·s-1. Specifically, the change in horizontal meridional wind speed (Δu) is mainly positive, while the variation of horizontal zonal wind speed (Δv) is negative. In the vertical direction, the wind speed (Δw) varies within ±0.15 m·s-1, but is positive on the whole. Additionally, the values of Δu, Δv, and Δw in lower levels are smaller than in higher levels. These research findings could offer valuable references for retrieving the three-dimensional wind field and vertical velocity of precipitation systems.

Abstract:Based on NCEP/NCAR reanalysis data and conventional meteorological and hydrological observation data, the meteorological and hydrological characteristics as well as circulation anomalies and possible causes of the autumn floods in Hanjiang River in 2021 are analyzed. The results show that the precipitation in Hanjiang River in 2021 was more in the west and less in the east. The accumulated rainfall in the upper reaches of Hanjiang River was 400-985 mm, and the accumulated areal rainfall ranked the first since 1961. During the autumn flood season, the Danjiangkou Reservoir experienced the flood peak processes of more than 10000 m3·s-1 for 7 times. The accumulated alarm time of the major hydrological control stations in the middle and lower reaches of the Hanjiang River got to 151-580 h, with the maximum amplitude beyond warning line being 0.29-1.56 m, showing the characteristics of the longer alarm time, the higher amplitude over warnig line in the lower reaches. The comparison with the hydrological conditions in 2011 and 2017 shows that in 2021 the flood peak in the upper reaches of Hanjiang River came the earliest, the number of flood peaks was the largest, the duration of high water level in the upper reaches was the longest, the accumulated alarm time in the middle and lower reaches of Hanjiang River was also the longest, and the abnormal degree of water regime was significantly much stronger. In terms of atmospheric circulation, the low trough of Lake Balkhash and Lake Baikal region, the high ridge from the east of mainland China to the Northwest Pacific, the low trough in the areas to the east of Japan Island, the subtropical high, the South Asian high and the East Asia subtropical westerly jet were all abnormally stronger than those in the same period in history. The formations of abnormal lowlevel flow field and water vapor convergence center, combined with the divergence and suction of highlevel flow field, were the largescale circulation background favorable for the continuous and enhancement of precipitation. When the inflow of Danjiangkou Reservoir was close to 2000 m3·s-1, the 5 d accumulated areal rainfall exceeding 60 mm in the upper reaches of Hanjiang River was closely related to the appearance of flood peaks above 10000 m3·s-1. After the upstream inflow increased, the narrowing river coarse of the middle and lower reaches of the Hanjiang River was the main reason for the overalarmline floods in this area where the precipitation was not so heavy.

Abstract:In the winter of 2022/2023, the average temperature in China was 0.2℃ higher than normal, and the temperature in most of China was warmer than normal. The average precipitation was 24.6% less than normal, showing a spatial distribution pattern of more in the north and less in the south of China. The standardized East Asia winter monsoon index was 0.25, slightly stronger than normal. The standardized Siberian high intensity index was 0.47, which was also slightly stronger than normal. The Arctic oscillation index of the winter was -0.6 with a significant phased characteristic feature of being negative in the earlier winter and positive in the later winter.The intensity index of Northwest Pacific subtropical high was -45.6 gpm, which was weaker than normal. In winter 2022/2023, the 500 hPa geopotential height field in the middle and high latitudes of Eurasia was high in the west and low in the east. The geopotential height field in North China and Northeast China was at a lower altitude, but the geopotential height field in other regions of China was higher. The winter atmospheric circulation had significant intraseasonal variation characteristics, with the strong meridional circulation in December 2022 and the zonal circulation in February 2023. The corresponding winter monsoon also showed the trend of strong in earlier winter but weak in later winter. Correlation analysis of sea surface temperature and height fields shows that there was a significant negative correlation between the North Atlantic sea surface temperature (NAT) in August and September and the height fields of Ural Mountains Region in December. The abnormal negative phase (positive phase) of NAT in the early stage was conducive to the strengthening (weakening) of the meridional circulation in East Asia in the earlier stage of winter.

Abstract:The main characteristics of the general atmospheric circulation at 500 hPa in April 2023 are as follows. There were two polar vortex centers in the Northern Hemisphere, stronger than usual. The circulation in middle-high latitudes showed an anomalous four-wave pattern, and the Asian circulation was relatively flat and straight. The northern part of China had a negative anomaly in the geopotential height field, the strength of western Pacific subtropical high was weaker than that in normal years, and the south branch was a little weaker, more westward than usual. The monthly mean temperature was 11.4℃, 0.1℃ lower than normal. The monthly mean precipitation was 48.7 mm, 11.7% more than the value (43.6 mm) in normal periods, of which the precipitation in the Northwest China and North China was significantly more than that in normal years. In addition, cold air was very active during this month, and four cold air processes occurred, bringing large amplitude of drop in temperature. There were four large-scale torrential rain processes across China and six regional severe convection processes with strong local intensities. In addition, six sand-dust events occurred in northern China, and drought maintained and developed in Yunnan and other places.