NUMERICAL SIMULATION OF MICROPHYSICAL PROCESSES IN CUMULONIMBUS——PART Ⅰ:MICROPHYSICAL MODEL

A comprehensive parameterized model of microphysical processes in cumulonimbus clouds is presented.The transformation rates of the water contents and number concentrations of the cloud droplets,rain drops,ice crystals,graupels and hails are deduced on theoretical and experimental results for 26 kinds of micro-physical processes,which include condensation,deposition,evaporation,collection,ice nucleation,ice mul-tiplication,melting,freezing,and autoconversions of cloud to rain,ice to graupel and graupel to hail.     

A Numerical Investigation on Microphysical Properties of Clouds and Precipitation over the Tibetan Plateau in Summer 2014

In order to improve our understanding of microphysical properties of clouds and precipitation over the Tibetan Plateau (TP), six cloud and precipitation processes with different intensities during the Third Tibetan Plateau Atmospheric Science Experiment (TIPEX-Ⅲ) from 3 July to 25 July 2014 in Naqu region of the TP are investigated by using the high-resolution mesoscale Weather Research and Forecasting (WRF) model. The results show unique properties of summertime clouds and precipitation processes over the TP. The initiation process of clouds is closely associated with strong solar radiative heating in the daytime, and summertime clouds and precipitation show an obvious diurnal variation. Generally, convective clouds would transform into stratiform clouds with an obvious bright band and often produce strong rainfall in midnight. The maximum cloud top can reach more than 15 km above sea level and the velocity of updraft ranges from 10 to 40 m s-1. The simulations show high amount of supercooled water content primarily located between 0 and -20℃ layer in all the six cases. Ice crystals mainly form above the level of -20℃ and even appear above the level of -40℃ within strong convective clouds. Rainwater mostly appears below the melting layer, indicating that its formation mainly depends on the melting process of precipitable ice particles. Snow and graupel particles have the characteristics of high content and deep vertical distribution, showing that the ice phase process is very active in the development of clouds and precipitation. The conversion and formation of hydrometeors and precipitation over the plateau exhibit obvious characteristics. Surface precipitation is mainly formed by the melting of graupel particles. Although the warm cloud microphysical process has less direct contribution to the formation of surface precipitation, it is important for the formation of supercooled raindrops, which are essential for the formation of graupel embryos through heterogeneous freezing process. The growth of graupel particles mainly relies on the riming process with supercooled cloud water and aggregation of snow particles.     

ALLOCATION DIFFERENCE ANALYSES OF WATER SUBSTANCES DURING TYPHOON LANDING PROCESSES

Based on a successful simulation of Typhoon Haikui (2012) using WRF (Weather Research & Forecasting) model with the WSM6 microphysics scheme, a high-resolution model output is presented and analyzed in this study. To understand the cause of the average gridded rainfall stability and increases after Haikui’s landfall, this research examines the fields of the physical terms as well as the vapor and condensate distributions and budgets, including their respective changes during the landing process. The environmental vapor supply following the typhoon landfall has no significant difference from that before the landfall. Although Haikui’s secondary circulation weakens, this circulation is not conducive to typhoon rainfall stability or increases, although the amounts of the six kinds of water substances (vapor, cloud water, cloud ice, snow, rain, and graupel) increase in the outer region of the typhoon. This reallocation of water substances is essential to the maintenance of rainfall. The six kinds of water substances are classified as vapor, clouds (cloud water and ice) and precipitation (snow, rain, and graupel) to diagnose their budgets. This sorting reveals that the changes in the budgets of different kinds of water substances, caused by the reduced mixing ratios of snow and ice, the water consumption of clouds, and the transformation of graupel, induce increased concentrations of precipitation fallout, which occur closer to the ground after typhoon landfall. In addition, this pattern is an efficient way for Haikui’s rainfall to remain stable after its landfall. Thus, the allocation and budget analyses of water substances are meaningful when forecasting the typhoon rainfall stability and increases after landfall.     

Microphysical Responses to Catalysis During a Stratocumulus Aircraft Seeding Experiment over the Sanjiangyuan Region of China

This study explores the microphysical responses to a cloud seeding operation in the Sanjiangyuan region, China. The cloud seeding was performed using a zigzag flight pattern, while the detection phase was accomplished using a back-and-forth flight pattern through the top of a stratocumulus layer. Global Position System（GPS） and Particle Measuring System（PMS） data obtained during the operation are used to determine the efective cloud area before and after the operation, diferentiate the phase states of cloud particles, and analyze changes in the concentrations of liquid cloud particles and ice crystals, the evolution of the cloud particle spectrum, and the content of supercooled water. The median diameter of liquid cloud particles in the area of the cloud-seeding operation was 3.5–18.5 μm, most cloud particles observed in the 21.5–45.5-μm size regime were ice crystals, while all particles of size 50 μm and above were in the ice phase. Changes in the concentration and typical diameter of cloud particles within 36 km downwind of the cloudseeding operation did not exceed natural fluctuations in the cloud area before the operation; however, the concentration of liquid cloud particles decreased substantially in areas with high concentrations of supercooled water（concentrations of supercooled water exceeding 0.01 g m 3）. The concentration of ice crystals within the measuring range of the Forward Scattering Spectrometer Probe（FSSP） increased substantially, the water content of ice-phase particles increased, and the average supercooled water content in the cloud decreased from（68.3± 23.1）% before the operation to（34.2± 12.4）%. The efects of cloud seeding were more pronounced in parts of the cloud where the content of supercooled water was higher. Little to no efects were observed in parts of the cloud with low concentrations of supercooled water.     

A MODELING STUDY OF PARAMETERIZATION SCHEMES FOR DEPOSITIONAL GROWTH OF ICE CRYSTAL: FOUR RAINFALL CASES OVER TROPICS AND MIDLATITUDES

Depositional growth of ice crystal is one of the major processes for development of precipitation systems and can be represented by depositional growth of cloud ice from cloud water (PIDW) and depositional growth of snow from cloud ice (PSFI) in cloud-resolving model. Four parameterization schemes are analyzed in the cloud-resolving model simulations of four rainfall cases over the tropics and midlatitudes. The comparison of time and model domain mean data shows that Shen’s scheme produces the closest rainfall simulation to the observation. Compared to Zeng’s scheme, Shen’s scheme improves the mean rain-rate simulation significantly through the dramatic decrease in depositional growth of cloud ice from cloud water. Compared to other schemes, Shen’s scheme produces the better rainfall simulation via the reduction in the mean rain rate associated with the enhanced gain of cloud water and ice.     

Simulation of the Microphysical Processes and Effect of Latent Heat on a Heavy Rainfall Event in Beijing

An extraordinary rainstorm that occurred in Beijing on 21 July 2012 was simulated using the Weather Research and Forecasting model. The results showed that：（1） The two precipitation phases were based on a combination of cold cloud processes and warm cloud processes. The accumulated conversion amount and conversion rate of microphysical processes in the warm-area phase were all much larger than those in the cold front phase.（2） 72.6% of rainwater was from the warm-area phase. Rainwater mainly came from the melting of graupel and the melting of snow, while the accretion of cloud water by rain ranked second.（3） The net heating rate with height appeared as an overall warming with two strong heating centers in the lower and middle layers of the troposphere and a minimum heating center around the melting layer. The net heating effect in the warm-area phase was stronger than that in the cold front phase.（4） Warm cloud processes contributed most to latent heat release, and the thermal effect of cold cloud processes on the storm in the cold front phase was enhanced compared to that in the warm-area phase.（5） The melting of graupel and snow contributed most to latent heat absorption, and the effect of the evaporation of rainwater was significantly reduced in the cold front phase.     

NUMERICAL SIMULATION STUDY OF HAIL CLOUD—PART Ⅰ:THE NUMERICAL MODEL

In order to study mechanisms of hailstone formation and hail suppression with seeding and toobtain optimum seeding technique for hail cloud,a 3-D compressive numerical seeding model forhail cloud is developed.The water substance in hail cloud is divided into 8 categories,i.e.,watervapor,cloud droplet,raindrop,ice crystal,snow.graupel,frozen drop and hail,and the detailedmicrophysical processes are described in a spectrum with two variable parameters and morereasonable particle number/size distributions.Then,the model is able to predict concentration andwater content of various particles.Especially.it can calculate the number of hailstones whosecores are graupel or frozen drop and apply to study mechanism of hailstone formation.Additionally,a conservative equation of AgI as seeding or glacigenous agent is found andnucleation by condensation of artificial nucleus,and nucleation by freezing of cloud droplet or raindrop which contact with AgI particle are considered.The dynamic energy flux of hail shooting onground is used to verify seeding effect.Therefore the model is also used to study mechanism of hailsuppression with seeding and the seeding technique,     

NUMERICAL SIMULATION OF ACIDIFICATION PROCESSES IN COLD CUMULUS CLOUDS

In this paper, a 1-D time dependent cold cumulus chemistry model is presented. In the coldcumulus model, 4 categories of hydrometeors: cloud drops, raindrops, cloud ice crystals, graupelparticles, and 18 microphysical interactions are considered, In the chemical model, the source andsink terms for pollutants include: the complicated interactions between pollutants (gases andaerosol particles) and the hydrometeors (especially ice crystals and graupel particles), the ex-changes of chemical compounds between two hydrometeors accompanying microphysical processesand the aqueous oxidations of S (IV) to S (VI). The two models are combined to study the acidifi-cation processes in cold cumulus clouds.     

The cloud processes of a simulated moderate snowfall event in North China

The understanding of the cloud processes of snowfall is essential to the artificial enhancement of snow and the numerical simulation of snowfall. The mesoscale model MM5 is used to simulate a moderate snowfall event in North China that occurred during 20–21 December 2002. Thirteen experiments are performed to test the sensitivity of the simulation to the cloud physics with different cumulus parameterization schemes and different options for the Goddard cloud microphysics parameterization schemes. It is shown that the cumulus parameterization scheme has little to do with the simulation result. The results also show that there are only four classes of water substances, namely the cloud water, cloud ice, snow, and vapor, in the simulation of the moderate snowfall event. The analysis of the cloud microphysics budgets in the explicit experiment shows that the condensation of supersaturated vapor, the depositional growth of cloud ice, the initiation of cloud ice, the accretion of cloud ice by snow, the accretion of cloud water by snow, the deposition growth of snow, and the Bergeron process of cloud ice are the dominant cloud microphysical processes in the simulation. The accretion of cloud water by snow and the deposition growth of the snow are equally important in the development of the snow.     

The Role of Initial Cloud Condensation Nuclei Concentration in Hail Using the WRF NSSL 2-moment Microphysics Scheme

The effects of the initial cloud condensation nuclei(CCN) concentrations(100–3000 mg~(-1)) on hail properties were investigated in an idealized non-severe hail storm experiment using the Weather Research and Forecasting(WRF) model, with the National Severe Storms Laboratory 2-moment microphysics scheme. The initial CCN concentration(CCNC) had obvious non-monotonic effects on the mixing ratio, number concentrations, and radius of hail, both in clouds and at the surface, with a CCNC threshold between 300 and 500 mg~(-1). An increasing CCNC is conducive(suppressive) to the amount of surface hail precipitation below(above) the CCNC threshold. The non-monotonic effects were due to both the thermodynamics and microphysics. Below the CCNC threshold, the mixing ratios of cloud droplets and ice crystals increased dramatically with the increasing CCNC, resulting in more latent heat released from condensation and frozen between 4 and 8 km and intensified updraft volume. The extent of the riming process, which is the primary process for hail production, increased dramatically. Above the CCNC threshold, the mixing ratio of cloud droplets and ice crystals increased continuously, but the maximum updraft volume was weakened because of reduced frozen latent heating at low level. The smaller ice crystals reduced the formation of hail and smaller clouds, with decreased rain water reducing riming efficiency so that graupel and hail also decreased with increasing CCNC, which is unfavorable for hail growth.     

过冷雨水低含量条件下冰雹形成和增长机制及其催化效果的数值模拟

利用中国科学院大气物理研究所开发的三维全弹性冰雹云模式，对美国对流降水协作试验（CCOPE）期间观测的1981年8月1日雹云进行模拟，讨论在过冷雨水低含量条件下冰雹形成和增长机制及其碘化银催化效果。结果表明:(1) 自然云的模拟与观测事实一致，如最大上升气流速度、云顶高度、流场结构以及雹胚组成等。(2) 雹胚以霰为主，霰主要来自冰雪晶与过冷小水滴的碰冻，其次来自雪的积聚转化；霰、冻滴和冰雹在形成后主要靠碰并过冷云水增长。(3)人工催化试验表明，碘化银主要以凝华核（包括凝结－冻结）的作用产生大量的人工冰晶，加速了过冷水向冰晶的转化，过冷云水因而大量减少；催化后霰和冻滴的数浓度增大，对过冷云水的竞争增强，其平均尺度减小导致转化成雹的数量减少；冰雹碰冻过冷云水的增长在催化后也被削弱，导致冰雹总质量进一步减少。此外，催化后降雨量也显著减少。     

A Simulation Study on the Characteristics of Cloud Microphysics of Heavy Rainfall in the Meiyu Front

A heavy rainfall in the Meiyu front during 4--5 July 2003 is simulated by use of the non-hydrostatic mesoscale model MM5 (V3--6) with different explicit cloud microphysical parameterization schemes. The characteristics of microphysical process of convective cloud are studied by the model outputs. The simulation study reveals that: (1) The mesoscale model MM5 with explicit cloud microphysical process is capable of simulating the instant heavy rainfall in the Meiyu front, the rainfall simulation could be improved significantly as the model resolution is increased, and the Goddard scheme is better than the Reisner or Schultz scheme. (2) The convective cloud in the Meiyu front has a comprehensive structure composed of solid, liquid and vapor phases of water, the mass density of water vapor is the largest one in the cloud; the next one is graupel, while those of ice, snow, rain water and the cloud water are almost same. The height at which mass density peaks for different hydrometeors is almost unchangeable during the heavy rainfall period. The mass density variation of rain water, ice, and graupel are consistent with that of ground precipitation, while that of water vapor in the low levels is 1--2 h earlier than the precipitation. (3) The main contribution to the water vapor budget in the atmosphere is the convergence of vapor flux through advection and convection, which provides the main vapor source of the rainfall. Besides the basic process of the auto-conversion of cloud water to rain water, there is an additional cloud microphysical process that is essential to the formation of instant heavy rainfall, the ice-phase crystals are transformed into graupels first and then the increased graupels mix with cloud water and accelerates the conversion of cloud water to rain water. The positive feedback mechanism between latent heat release and convection is the main cause to maintain and develop the heavy precipitation.     

区域冷锋降水微物理机制研究

20世纪利用一维层状云模式对2002年4月4~5日河南省冷锋降水过程进行了模拟。数值模拟结果显示,此次冷锋降水属于冷云降水过程,冷锋前后云中主要以冰相粒子为主,云中水质粒自上而下的空间分布依次为冰晶、雪、云水、霰、雨水。冷锋前后,各种水质粒有着不同的含量及数密度,但形成水质粒的主要微物理过程都表现为:冰晶数密度的增加主要依靠核化、繁生,大部分雪主要靠凝华、撞冻过冷云水和冰晶增长,霰的质量增加主要靠撞冻雪、过冷云水和雪自动转化而来,大部分的雨水是由霰融化而来,因而此次冷锋降水机制表现为“水汽—雪—霰—雨水”。     

冻滴微物理过程的分档数值模拟试验研究

霰和冻滴是深对流降水的主要来源。由于二者密度差异造成的不同下落末速度必然会导致云微物理过程的变化以及降水时空分布的改变。我们在以色列特拉维夫大学二维轴对称对流云全分档模式的基础上,将水成物粒子从34档增加到40档,修改了霰和雪的密度,加入冻滴分档处理的微物理过程,发展了一个包括液滴、冰晶、雪、霰和冻滴更为详细的云微物理分档模式。利用改进后的模式模拟了一次理想的强对流天气过程,分析了改进模式与原模式模拟的云微物理量场以及水成物粒子的时空分布特征,模拟结果表明:（1）由于冻滴的产生,较大的下落末速度导致在云内-3℃至-8℃较早地出现了冻滴,并造成了大量的冰晶繁生。（2）冻滴形成前期,液态水中心区域位于垂直上升速度大值中心上方,形成液态水累积区;冻滴形成期,液态水累积区位于0℃层以上,雨滴冻结生成冻滴,霰与半径大于100 μm的液滴碰并生成冻滴;冻滴增长期,在垂直上升气流的支撑下,冻滴碰并过冷水增长,导致冻滴含量增大,液态水含量减小。因此,改进模式能较好的模拟冻滴的形成过程,可以将该分档处理的微物理方案耦合到三维WRF（Weather Research and Forecasting model）模式中,更深入地研究强雷暴风切变在冰雹生成过程中的作用。     

东北冷涡中尺度云系降水机制研究 II: 数值模拟

在利用卫星、雷达和机载PMS(粒子测量系统)等观测资料对2003年7月8日东北冷涡积层混合云系的降水形成机制分析的基础上,将观测分析与数值模拟研究相结合,用中尺度数值模式对积层混合云系做数值模拟,并结合观测资料进一步分析了积层混合云系的微物理结构、粒子形成过程和降水形成机制,获得如下结果:(1)混合云中对流云具有分层的微物理结构.冰晶含水量最大值出现的高度最高,其次由高到低的排序是雪、云水、霰和雨;雨水主要出现在云的暖区;各种粒子中以雨水含水量最高,其次是霰.对流云体生命期较长,微物理结构基本稳定.(2)粒子形成增长过程有差异.冰晶通过凝华过程增长.雪主要来源于冰晶,产生后主要通过撞冻、收集冰晶和凝华过程增长,其中撞冻过冷云水增长对雪质量贡献最大,其产生率极大值高度与过冷云水相当.丰富的过冷云水,给雪的撞冻增长提供了有利条件.在高、中和低层雪的形成有着不同的机制,高层雪收集冰晶长大后,下落到低层又以雪撞冻过冷云水的结淞增长为主要过程.霰主要由雨滴冻结和雪的转化产生,过冷雨滴与冰晶接触冻结成霰;过冷雨滴收集雪,雪随着雨滴的冻结而转化成霰.因此霰的产生与过冷雨滴关系极大.霰主要撞冻云水、收集雪和冰晶增长,其中撞冻是霰的重要增长过程.雨水主要由霰的融化形成,降水主要是由冷云过程产生的.在过冷层,霰撞冻增长占优势.云上部的冰晶和雪对云的中部具有播撒作用,过冷层中存在丰富的过冷水,对冰相粒子的撞冻增长有利.对云水消耗的分析表明,雨滴对云滴的收集、霰和雪对云水的撞冻增长是消耗云水的主要过程.(3)从各种粒子的形成和增长过程可以看出,大部分雨水由霰融化形成,暖云过程贡献要小得多.可见,降水主要是由冷云过程产生的,这与观测分析的结果一致.     

锋面云系降水中的增雨潜力数值研究

采用中尺度模式MM5模拟研究2002年4月4～5日出现在河南省境内的一次冷锋云系宏观和微观结构特征、云系降水机制以及区域内水分循环收支、"拟降水效率"与云系增雨潜力.模拟结果显示,这次大范围冷锋降水云系分层、非均匀性结构明显,高层主要由冰相粒子构成,中层是冰-水混合层,低层为液水层.云系内同时存在较强的冷、暖云过程,低层和中层首先由暖云过程生成大量云水,而高层冰晶出现后不断向雪转化,大量的冰雪晶由高层降落至中层高云水含量区,促使雨水大量产生,云系存在明显的"播种-供给"云降水机制.此外,模拟显示云系降水主     

一次梅雨锋暴雨过程数值模拟的云微物理参数化敏感性研究

梅雨锋暴雨中的云微物理过程对降水的演变有着重要影响。本文通过WRF模式（3.4.1版本）,针对2018年6月29～30日一次梅雨锋背景下的暴雨过程进行数值模拟,分别采用了Morrison、Thompson和MY云微物理参数化方案进行对比分析,结果发现:（1）三个方案模拟的背景场在天气尺度上,都与ERA5再分析资料一致,能够模拟出有利于强降水发生的环流场。云微物理过程对梅雨期暴雨的局地环流有着显著影响,不同方案存在明显差异,本次过程中,Thompson方案模拟出更强的局地环流系统变率和上升气流。三个方案的模拟降水均有所夸大,小时降水率始终大于观测值。冰相粒子融化或雨滴搜集云滴的高估可能是造成降水模拟值偏强的重要原因之一,总体来看,Morrison方案的模拟效果相对最优。（2）冰相粒子融化、雨滴搜集云滴是雨滴增长的关键源项,蒸发则是其最重要的汇项。总的来说,雨滴对云滴的搜集量大于冰相粒子融化。但上述过程在不同方案中存在空间上的差异,从而使得模拟降水的空间分布存在差异。（3）Thompson方案中,冰相粒子融化量最大,雨滴蒸发项显著大于其它两个方案,在底层表现得最为明显。同时,该方案水汽凝结效应最强,使得雨滴搜集更多云滴。该方案模拟的雨滴最多,降水最强。该方案中凝华的主要产物为雪,且其在与过冷水碰并增长过程中占主导地位,故模拟的雪最多。（4）Morrison方案中,水汽主要凝华为雪和少量霰（冰晶忽略不计）;Thompson方案中水汽基本凝华为雪,其它冰相粒子极少;MY方案中,水汽主要凝华为雪和冰晶,冰晶总量略少于雪,但显著大于其它方案。（5）云滴在凇附过程中的总体贡献大于雨滴。Morrison和MY方案中,霰粒子搜集云滴增长的量均最大。Morrison方案中,其它凇附过程不同程度发挥作用,而MY方案中,其它凇附过程几乎可忽略不计。并且,霰粒子搜集云滴的增长量大于凝华过程产生的雪粒子总量。贝吉龙及凇附效应的差异,是不同方案中冰相粒子分布差异的关键原因之一。     

基于飞机观测的华北积层混合云降水微物理特征的数值模拟研究

基于2009年5月1日积层混合云降水2架飞机观测数据分析,使用中尺度模式WRFV3对此次过程积云区和层云区的微物理特征和转化过程进行数值模拟比较研究。飞机观测数据分析表明,此次积层混合云中的层云区和积云区冰粒子形状和形成过程有明显差别,层云区的粒子形状组成比较复杂,包含针状、柱状和辐枝状等,而积云区主要以辐枝状粒子为主,聚并、凇附过程明显。数值模式能较好地模拟出此次积层混合云降水过程的基本特征,包括回波分布、飞行路径上降水粒子的数浓度和液态水含量等。数值模拟结果表明,云水相对丰富、上升气流强的层云区凇附过程较强,产生的雪在低层融化为雨水,为后期高层形成的雪和霰提供丰富的液态水,能发展成对流较强的积云区,存在播种—供给机制。在积云区,水成物的比例从大到小依次为雪（51.9%）、霰（31.0%）和雨水（16.0%）;雪的主要源项包括淞附增长（56.8%）和凝华增长（40.1%）,霰的主要源项包括凇附增长（46.6%）、雨水碰并雪成霰（42.6%）和凝华增长（16.1%）,雨水的主要源项是霰（77.6%）和雪（22.4%）的融化。而相对云水较少、上升气流较弱的层云区将保持层云的状态,层云区水成物的比例从大到小依次为雪（90.4%）、雨水（6.1%）、冰晶（3.5%）;高层冰晶和雪通过凝华过程增长,雪在零度层下融化为弱的降水。     

利用GRAPES模式研究气溶胶对云和降水过程的影响

在GRAPES中尺度模式的双参数微物理方案中加入了气溶胶活化参数化过程,实现了对云滴数浓度的预报。选取不同季节两个降水过程进行模拟,并分别开展了不同气溶胶背景下的两个试验进行对比分析,研究气溶胶对云和降水可能的影响。结果表明:气溶胶浓度增加后,因为活化产生了更多尺度较小的云滴,抑制了云雨的自动转化,使大气中滞留了更多的云水,暖云降水减小;另一方面,云水的增加会使冰相粒子,尤其是雪和霰通过碰并云水等过程而增大,最后融化成雨增加冷云降水,同时冰相粒子增加会释放更多的潜热,促进上升气流的发展,进一步增加冷云降水。气溶胶对降水的影响存在空间不一致性,暖云较厚的地方暖雨过程受到的抑制明显,使地面降水减小,冷云厚度相对较厚时,冷云降水的增加会大于暖云降水的抑制,使地面降水增加。同时由于在云降水发展的不同阶段冷暖云的变化,气溶胶对降水的影响也存在着时间不一致性。     

广东一次飑线过程中一个雷暴单体成熟阶段的电荷结构演变特征的数值模拟

为了进一步印证以往观测反推得到的广东地区雷暴云多偶极性电荷结构的结论,利用加入了起放电参数化方案的WRF模式,模拟了广东在2017年5月8日发生的一次飑线过程,并对这次飑线过程中一个雷暴单体成熟期的电荷结构演变特征进行分析,通过分析动力、云水含量、各水成物粒子混合比及携带电荷情况,讨论了电荷结构的形成及演变机制。结果表明,成熟阶段的单体,电荷结构从三极性逐渐演变为偶极性。这是因为在成熟初期,霰粒子在有效液态水含量适中且温度较高的地方与冰晶/雪花粒子发生了非感应碰撞,因此底部霰粒子携带正电,雷暴云底部形成次正电荷区,电荷结构为三极性。而在成熟后期,由于丰富的云水含量,使冰粒子的凇附过程增强,霰不断增加,冰晶和雪花不断被消耗,温度较暖区域与霰共存的冰晶和雪花急剧减少,使得该区域大小冰粒子的非感应碰撞起电急剧减少,此处霰粒子不能再通过非感应碰撞获得正电荷,底部次正电荷区随之消失,雷暴云的电荷结构转变为偶极性。此结果和以往观测反推得到的结论不同,这表明,对南方雷暴电荷结构还需继续深入认识。