应用气象学报

2022, 33(4)

[FullText](432) [PDF 789KB](40)

Abstract:

Development Mechanisms of the Yellow Sea and Bohai Sea Cyclone Causing Extreme Snowstorm in Northeast China

He Lifu, Chyi Dorina, Yu Wen

2022, 33(4): 385-399. DOI: 10.11898/1001-7313.20220401

[FullText HTML] (190) [PDF 6324KB](211)

Abstract:
The structure evolution and explosive development mechanisms of the Yellow Sea and Bohai Sea cyclone causing the extreme snow in Northeast China from 7 November to 9 November in 2021 are analyzed with high-resolution observations and the fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) atmospheric reanalysis data (ERA5) with a 0.25° by 0.25° spatial resolution. Results show that the extreme snowstorm occurs under the background of high-altitude cold vortex collocated with surface cyclone. After the formation of the surface cyclone in the Yellow Sea, it strengthens rapidly and moves northward along the eastern part of Northeast China. The snowfall area is mainly distributed on the west side of the cyclone, and the snowfall intensity is closely related to the occurrence and development of the surface cyclone. Its explosive development stage corresponds to the strongest period of the extreme blizzard process. The Yellow Sea and Bohai Sea cyclone is generated from a ground inverted trough which gradually strengthens with eastward shift into the sea. During its explosive developing and occluding stages, the leaf cloud system evolved into hook comma cloud system and vortex cloud system. The horizontal structure shows frontal fracture and the warm front back bending and wrapping, while the vertical structure shows high-altitude frontal fracture, the emergence of dry and warm center, the formation of neutral occluded front, and deep low value system from inclined vortex column. Wave activity flux analysis shows that the ridge in Siberian, the trough in North China and the ridge in Northeast China at 500 hPa devote to Rossby wave train. With the continuous eastward movement and the wave energy dispersion downstream of the positive anomaly center in the upper reaches of Siberia, the wave activity flux from the northwest in the North China trough is rapidly enhanced, and therefore the cold vortex enhances rapidly. The sharp enhancement of vorticity factor over surface cyclones is beneficial to the explosive development of cyclones. The potential vorticity diagnosis on the isobaric surface shows that the abnormal area of positive potential vorticity gradually approaches and superimposes on the middle and low-level system, with the continuous southward development and downward propagation of stratospheric high-level vorticity along the isentropic surface, resulting in the rapid development and downward extension of middle-level cold vorticity and thereby the explosive enhancement of surface cyclones. In addition, the slow downward propagation of potential vorticity is also conducive to the maintenance of occluding stage in the frontal cyclone.

Upper Wind Difference Characteristics and Forecast Within 3.5 Hours Before and After Rocket Launch

Cheng Huhua, Cheng Wei, Shen Hongbiao, Zhao Liang

2022, 33(4): 400-413. DOI: 10.11898/1001-7313.20220402

[FullText HTML] (123) [PDF 1885KB](78)

Abstract:
The upper wind has a major impact on the safety of the launch vehicle, and the wind speed 3 hours before launching is a determinant for the eventual schedule. If the maximum aerodynamic load generated by the upper wind does not exceed the threshold, the rocket can be launched as planned, otherwise it will be considered to postpone the launch. The deviation of the upper wind 3 hours before and at the launch time should be investigated. Taking the upper wind dataset with an interval of 3.5 hours (December 2014 to December 2020) to analyze the wind difference, and a forecast model for the wind of 0.5 hours after the launch is established using WRF model and the wind data of 3 hours before the launch. It shows that the characteristics of upper wind speed and direction difference within 3.5 hours are related to altitude, season, and the average upper wind speed. Although the maximum wind speed deviation range is -24.00-26.00 m·s^-1, for two-thirds of the cases the deviation is within 10 m·s^-1, which mainly occur in the middle and upper troposphere with altitude of [6.5 km, 11.5 km). The absolute value of the maximum wind direction deviation range is 10.00°-180°, mainly in the [30°, 60°) interval, and this mostly occurs in the middle and lower troposphere with altitude of [1.5 km, 6.5 km). The average absolute deviation of the upper wind speed within 3.5 hours shows an increasing trend with the increase of average upper wind speed 3 hours before the launch, but the relative error and the wind direction deviation decreases, indicating that when the upper wind is strong, the wind direction is less prone to short-term changes. The upper wind deviation within 3.5 hours varies with the seasons, for example, the absolute deviation of winter wind speed is greater than that of summer, but the absolute deviation of wind direction in winter is smaller than that of summer. Using the forecast model results of wind 0.5 hours after the launch can help to avoid the risk of rocket launch in advance.

Comparative Analysis on Dual Polarization Features of Two Severe Hail Supercells

Diao Xiuguang, Li Fang, Wan Fujing

2022, 33(4): 414-428. DOI: 10.11898/1001-7313.20220403

[FullText HTML] (182) [PDF 8620KB](210)

Abstract:
Using S-band dual-polarization weather radar data, sounding and ground meteorological observations, and disaster investigation reports, the similarity and difference of dual polarization parameters between Lixian and Zhangqiu supercells with hails above 50 mm are analyzed. Lixian supercell occurred at Lixian, Heibei Province on 25 June 2020, and Zhangqiu supercell occurred at Zhangqiu, Shandong Province on 9 July 2021. The results show that two supercells occurred in similar weather pattern (northwest flow) and large vertical wind shear environmental conditions which is conducive to the generation and maintenance of supercell storms, but Zhangqiu supercell is with stronger convective effective potential energy, larger humidity, and higher wet bulb 0℃ layer height. The main similarities include obvious differential reflectivity (Z_DR) arcs along the forward flank of supercell storms, Z_DR rings distributed around the updraft in the middle layer, and obvious Z_DR columns and specific differential phase (K_DP) columns above the 0℃ level. Z_DR arcs are associated with large raindrops or small melting hail particles, Z_DR columns mark the location of convective updrafts as large raindrops or wet ice particles are lofted to subfreezing temperatures, and K_DP columns are dominated by large concentrations of small and medium-sized raindrops or melting ice particles. The similarity of the updraft structure plays a key role in the commonness or similarity of the polarization characteristics. The main differences are stronger reflectivity factor Z_H, but lower height of Z_DR column and K_DP column in Lixian supercell. The strong overhang echo above the weak echo area in Lixian supercell contains large hail particles generated by cumulated growth. After the overhanging large hail particles enters the descending channel, they will produce obvious growth again and become more irregular, resulting in stronger horizontal polarization reflectivity factor Z_H and smaller correlation coefficient. The obvious differential attenuation signature and nonuniform beam filling are observed in low level of Lixian supercell. The differential attenuation caused a decrease in the differential reflectivity as the beam propagates through large hail cores. Nonuniform beam filling is generated by inhomogeneous filling of different hydrometeor particles in the sampling volume. Under similar weather patterns, the distribution characteristic of humidity vertical profile is one of the key environmental factors of storm intensity. Lixian supercell storm occured in very low humidity vertical distribution environment, while Zhangqiu supercell storm occured in wetter environment.

Retrieval of Air Vertical Velocity and Droplet Size Distribution in Squall Line Precipitation Using C-FMCW Radar

Chen Shaojie, Zheng Jiafeng, Yang Ji, Che Yuzhang, Ren Tao, Huang Xuan

2022, 33(4): 429-441. DOI: 10.11898/1001-7313.20220404

[FullText HTML] (148) [PDF 3510KB](87)

Abstract:
The zenith C-band frequency modulation continuous wave(C-FMCW) radar has good detection capability with high temporary-spatial resolution and large dynamic range. The Doppler spectral density data of two squall lines precipitation cases at Longmen of Guangdong are utilized to retrieve the air vertical velocity (V_a) in clouds and droplet size distribution (DSD). The empirical relation method (checking relationship between mean particle falling velocity (V_t) and reflectivity factor) and the small-particle-trace method are explored to retrieve the air vertical velocity in clouds. And then the droplet size distribution is retrieved from the translated Doppler spectral density by a velocity-diameter relation. The retrieved DSD of two squall lines are then compared and validated with the observation of K-band micro rain radar and second-generation Parsivel disdrometer. The retrieved V_a by the empirical relation method is slightly smaller for strong monomer than that by the small-particle-trace method and slightly larger for weak convective precipitation, but V_t is the opposite. The absolute value of V_t negative velocity by the empirical relation method and the small-particle-trace method corresponds to the moment of large particles and heavy rain observed by Parsivel disdrometer, indicating that V_a of two methods is basically reliable. The comparison in DSD retrieval show that the number of small droplets observed by radar is higher, but Parsivel disdrometer may underestimate it. The results of the empirical relation method are closer to micro rain radar and Parsivel disdrometer when rain rate is below 1 mm·h^-1. The medium droplets obtained by radar retrieval are consistent with Parsivel disdrometer measurements, but the concentration of large droplets is low when rain rate is stronger than 10 mm·h^-1. The retrieval results of both methods are close to Parsivel disdrometer and micro rain radar when rain rate is between 1 mm·h^-1 and 10 mm·h^-1. The strong convection makes droplets rupture severer in the peak area of heavy precipitation in the squall line, resulting in smaller mass-weighted mean diameter (D_m) and larger generalized intercept parameter N_w of the empirical relation method and the small-particle-trace method retrieval. For weak convective precipitation at the back of the squall line, the value of empirical relation method is quite close to Parsivel disdrometer. Under different rain rate, μ value of C-FMCW radar is less than 10 and the fluctuation is smaller, indicating that the results of C-FMCW radar is even more reliable than Parsivel disdrometer and micro rain radar.

Macro and Micro Characteristics of a Fog Process in Changbai Mountain in Summer

Wang Yufei, Qi Yanbin, Li Qian, Li Jian

2022, 33(4): 442-453. DOI: 10.11898/1001-7313.20220405

[FullText HTML] (186) [PDF 2948KB](144)

Abstract:
In the summer of 2021, the fog droplet spectrum observation is carried out on the main peak of Changbai Mountain for the first time. From 31 July to 1 August, there is a fog process that lasts for 19 hours, and the minimum visibility in the extremely dense fog stage is less than 100 m. Using the observations of laser fog droplet spectrometer, combined with the ground automatic weather station, GPS balloon sounding, Himawari-8 satellite and ERA5 data, the macro and micro physical characteristics of the fog are studied, the causes of the fog are discussed, and the microphysical characteristics evolutions of the extremely dense fog period are analyzed.The results show that the fog process lasts for a long duration with occasional short dissipation, and during the process the ambient wind speed is high, the visibility is low, the number concentration of the droplets is low, with small effective diameter and low liquid water content. The wind speed is always high in the period of extremely dense fog, which is significantly different from that of plain fog. In the early stage the fog is arisen from windward slope, which is a typical topographical cloud and fog on the main peak of Changbai Mountain in summer. It is formed by the continuous southwest warm and humid airflow climbing along the terrain under the condition of stable temperature inversion stratification. While the latter process of the fog is generated by the advection to the main peak of Changbai Mountain. The temporary dissipation of fog is related to the intensity and movement of the jet core at 700 hPa. The average effective diameter of fog droplets is 5.7 μm, the average number concentration is 246.4 cm^-3, and the average liquid water content is 0.05 g·cm^-3. The microphysical characteristics are similar to those of sea fog.For the extremely dense fog, the minimum visibility is less than 100 m. The extremely dense fog is characterized by explosive enhancement. Due to the rapid expansion of the droplets through the turbulent collision process, a single peak structure is formed. The peak diameter of the droplet particles is 6.0 μm, which has a significant contribution to the formation of the summer fog on the main peak of Changbai Mountain. In the formation, development and weakening stages of the extremely dense fog, the changes of droplet number concentration, liquid water content and effective diameter have a good corresponding relationship, but it is not obvious in the mature stage.

Freezing Injury of Winter Wheat in Northern China and Delaying Sowing Date to Adapt

Song Yanling, Zhou Guangsheng, Guo Jianping, Li Yong, Pan Yaru, Fu Yan, Yang Rongguang, Bai Xiaoying, Xu Jinxia

2022, 33(4): 454-465. DOI: 10.11898/1001-7313.20220406

[FullText HTML] (173) [PDF 1847KB](136)

Abstract:
Under the background of global warming, whether freezing injury is still the main meteorological disaster in northern winter wheat growing region of China becomes uncertain, and whether delaying sowing date is an effective measurement to adapt to climate change becomes an urgent scientific problem to solve. It is found that the correlation coefficient between winter freezing injury index and winter wheat yield reduction rate is 0.62 in the northern winter wheat region from 1981 to 2000, which indicates that winter freezing injury is one of the main disasters before the year of 2000. However, this correlation becomes very low after the year of 2000, indicating that winter freezing injury is no longer the main factor for yield reduction of winter wheat. Experiments are carried out at Taian and Xianyang stations, showing that the accumulated temperature before winter and the accumulated temperature of the whole growth season of winter wheat are significantly reduced. The plant height, total aboveground dry weight and leaf area index will decrease when sowing date of winter wheat is delayed for 10 days and 20 days. Furthermore, the delay of sowing date has an adverse impact on the yield structure, the effective panicles and grains per panicle are decreased by 5% and 10.2% respectively when the sowing date of winter wheat is delayed for 10 days from 2018 to 2021 at Taian Station, and they are decreased by 17.2% and 11.9% respectively when the sowing date is delayed for 10 days at Xianyang Station. Overall, the average yield of winter wheat is reduced by 22% and 40% when sowing dates are delayed for 10 days and 20 days respectively, which indicates that the delayed sowing date of winter wheat have no positive effects. The possible cause is that the local winter wheat varieties have changed at Xianyang and Taian, and farmers have appropriately adjusted the sowing date according to experience. The current winter wheat sowing date and the main winter wheat varieties have adapted to local climate change.

Comparison of Drought Recognition of Spring Maize in Northeast China Based on 3 Remote Sensing Indices

Chen Yuye, Wang Peijuan, Zhang Yuanda, Yang Jianying

2022, 33(4): 466-476. DOI: 10.11898/1001-7313.20220407

[FullText HTML] (329) [PDF 1621KB](107)

Abstract:
Drought is a complex and widespread natural disaster, which has brought serious environmental and social problems and caused huge economic losses to China. For nearly half a century, the trend of aridification in Northeast China has been very significant, the area of influence has increased, and the degree of drought has also intensified significantly. Drought index is the basis of judging the occurrence of drought events, evaluating the degree of drought, clarifying the spatiotemporal characteristics of drought, and formulating measures for drought prevention and mitigation. Numerous studies indicate that solar-induced chlorophyll fluorescence(SIF), normalized difference vegetation index(NDVI), enhanced vegetation index(EVI), and normalized difference water index(NDWI) can be used to identify agricultural drought, but the research on comparing the ability of SIF index, NDWI and NDVI for identifying agricultural drought has not been reported publicly. Taking spring maize in Northeast China as the research object, NDWI and NDVI are calculated using the surface reflectance data MOD09A1. Combined with SIF index, NDWI and NDVI, the time series dataset of remote sensing drought index is constructed, respectively, and the accuracy and sensitivity of these three indices for identifying the drought is further explored. It shows that the accuracy of three indices in indentifying maize drought are all higher than 80%, and the accuracy of SIF index is the highest, reaching 89.27%. The accuracy for identifying severe drought is higher than mild and moderate drought for three indices, all reaching more than 94%, and the accuracy of SIF index exceeds 95%. From the perspective of different developmental stages of spring maize, the monitoring accuracy is the highest at seedling stage, reaching more than 90%, and is the lowest at jointing-booting stage and grain filling-maturity stage. The drought identifying accuracies of SIF index during four developmental periods of spring maize are all better than those of NDWI and NDVI. The sensitivities of SIF index, NDWI and NDVI to the identification of maize drought are different, and the SIF index has the highest sensitivity to drought identification, followed by NDWI, and NDVI is slightly lower. In terms of drought grades, the identifying sensitivities of three indices to severe drought are all higher than those of mild and moderate drought. Above all, compared with NDWI and NDVI, SIF index has better accuracy and sensitivity in identifying the drought of spring maize in Northeast China, and can make timely and accurate response to maize drought in Northeast China. The results have important practical significance for accurately identifying and predicting drought of spring maize in Northeast China, and taking effective drought-resistant measures in a timely and objective manner to minimize the damage to crops.

The Interaction Between Intensity and Rainfall of Typhoon Rammasun(1409)

Qin Hao, Zheng Fengqin, Wu Liquan

2022, 33(4): 477-488. DOI: 10.11898/1001-7313.20220408

[FullText HTML] (136) [PDF 2426KB](135)

Abstract:
Based on the methods of Fourier decomposition, correlation analysis and Liang-Kleeman information flow, the interaction between intensity and rainfall of Typhoon Rammasun (1409) is studied using the best track data of Shanghai Typhoon Institute (STI) of China Meteorological Administration (CMA), Tropical Rainfall Measuring Mission (TRMM) satellite 3B42 rainfall estimation data of National Aeronautics and Space Administration (NASA) and ERA5 reanalysis data of European Centre for Medium-Range Weather Forecasts (ECMWF) with 0.25°×0.25° grids. The results show that the rainfall of Typhoon Rammasun (1409) has obvious characteristics of asymmetry, and it is mainly located in the west side of the typhoon center. The rainfall of Typhoon Rammasun (1409) increases significantly twice during the whole life cycle, which corresponds to the intensification period. The information flow analysis shows that the rainfall of Typhoon Rammasun (1409) is affected by the intensity of typhoon itself, whereas the rainfall can also feedback on the latter. Compared with the influence of typhoon intensity on rainfall, the information flow from rainfall to typhoon intensity decreases by nearly an order of magnitude, indicating that the typhoon intensity plays a dominant role in the interaction relationship. The possible mechanism by which the typhoon intensity affects rainfall are analyzed by diagnosing the water vapor and dynamic conditions respectively. In terms of water vapor conditions, the convergence area of vertical integral of moisture flux corresponds well to the rainfall. The intensification (reduction) of Typhoon Rammasun (1409) partly results in the enhanced convergence (divergence) of vertical integral of moisture flux in the southwest of the typhoon center, which brings more (less) rainfall in this region. In addition, the South China Sea and the Western Pacific water vapor transport channel have obvious response to the changes of Typhoon Rammasun (1409) intensity. In terms of dynamic conditions, the strong center of vertical helicity in the middle and lower layers is mainly located in the west side of the typhoon center, indicating the rainfall area. The intensification (reduction) of Typhoon Rammasun (1409) leads to the increase (decrease) of absolute vertical helicity on the west side of the typhoon center, thus promotes (inhibits) the development of upward movement to a certain extent, resulting in more (less) water vapor condensation and rainfall.

Machine Learning Correction of Wind, Temperature and Humidity Elements in Beijing-Tianjin-Hebei Region

Han Nianfei, Yang Lu, Chen Mingxuan, Song Linye, Cao Weihua, Han Lei

2022, 33(4): 489-500. DOI: 10.11898/1001-7313.20220409

[FullText HTML] (177) [PDF 1550KB](244)

Abstract:
Weather conditions have an important impact on agricultural production, transportation, economic activities, so the improvement of forecast accuracy has been a constant concern of the society. After more than 100 years of continuous development, the accuracy of numerical weather model has been continuously improved, but there are still inevitable forecast errors. Therefore, it is an important issue worthy of study to improve the prediction accuracy by studying various error correction methods and post-processing the results of numerical weather prediction.Machine learning method is applied to revise four meteorological elements forecasted by RMAPS-RISE(rapid-update multi-scale analysis and prediction system-rapid integration and seamless ensemble) system developed by Beijing Institute of Urban Meteorology. First, the data are preprocessed by interpolating the system forecast data and extracting the data of each element site from the grid data. The observations of automatic weather stations and forecast data are processed to establish unified datasets for the application and modeling of machine learning. Second, linear regression method, gradient boosting regression method, XGBoost method and Stacking method are designed to combine various machine learning algorithms to improve the generalization ability of the model. In addition, an error analysis model is constructed according to four correction methods, and the correction technology research and experimental application of the forecast errors of each station's initial time under the complex terrain of Beijing-Tianjin-Hebei are carried out. Finally, the improvement of the revised forecast of different machine learning methods compared with the original RMAPS-RISE system forecast accuracy is compared.In the experimental part, two modeling ideas are proposed, and four machine learning methods are used to conduct correction and comparison experiments. It shows among the modeling ideas based on error analysis, the Stacking method has the best effect, effectively reducing the forecast error of the original system for the next 3-12 hours for 24 initial times. Among the other three single machine learning method, XGBoost method performs the best, followed by the gradient boosting regression method and linear regression method, and all of them have a significant positive effect on the prediction accuracy. Overall, the forecast error correction model based on machine learning methods can effectively reduce the original forecast error of RMAPS-RISE system, and they have broad application prospects in forecast correction. It is helpful to further improve the forecast accuracy of the objective interpretation product of the site under complex terrain.

A Low Visibility Recognition Algorithm Based on Surveillance Video

Liu Dongwei, Mu Haizhen, He Qianshan, Shi Jun, Wang Yadong, Wu Xueqin

2022, 33(4): 501-512. DOI: 10.11898/1001-7313.20220410

[FullText HTML] (258) [PDF 1935KB](197)

Abstract:
Low visibility has significant influences on highways, ferries, civil aviation, and other modes of transportation, and the visibility observation of meteorological departments is not dense enough to meet the monitoring needs of low visibility weather. Using existing video surveillance equipment to extract visibility data can save a significant amount of money on visibility instrument deployment and maintenance, improve data density, and provide finer data support for traffic and urban safety operations. Based on video live image conversion, a simple convolutional neural network classification approach is suggested to extract visibility levels. The algorithm assumes that the video devices are installed horizontally and have an open view, and it creates a new fixed-size image by dividing the original video image into horizontal chunks and extracting the gradient, color saturation, and brightness information from each horizontal chunk. A simple convolutional neural network is used to learn and develop a visibility level recognition model from the converted images. The model is trained by 29668 video images of Yangshan Port Weather Station in Shanghai from September 2019 to December 2020, and then tested with 5757 video images from January to May in 2021. The comparison indicates the recognition model generated with this technique has a greater accuracy than the recognition model built directly with AlexNet network. The model has an overall accuracy of 87.99% during daytime and 81.32% during nighttime when the observed visibility is classified into five levels of fog-free, light fog, fog, dense fog, and thick fog according to the fog forecasting level. The model's identification ability for no fog and light fog is high. However, because the scenery becomes nearly indistinguishable once dense fog appears at night, the model's recognition ability for dense fog level at night is poor, and it is easy to categorize it as a fog level mistakenly. Taking 1000 meters as the criterion of low visibility weather, the algorithm's accuracy is 96.18% during daytime and 96.14% during nighttime. The algorithm features a quick learning rate and ease of application, making it suitable for low visibility video image recognition in most open-field scenarios. The model is applied during a radiation fog in Shanghai on 13 April 2021. The video images of the sparse area of the automatic weather station installation are collected for visibility identification, and the visibility distribution map formed together with the existing automatic weather station visibility meter data is more complete and accurate, which demonstrates that the model established by this algorithm can effectively compensate the problem of insufficient density of the existing automatic station visibility meter data, and has certain application value in meteorological operations.

Vol.33, NO.4, 2022