复杂下垫面环境上海局地强对流天气研究
|
摘要
本文通过上海现有的城市综合观测网，对上海一次典型局地强对流0731观测进行分析，并提出上海局地强对流发生发展的概念模型。在此基础上，设计了快速更新同化试验，验证了概念模型的合理性。并对海陆风、城市热岛效应对局地强对流的促发机制进行进一步的探讨。利用快速更新同化数值试验，验证了各种观测资料在强对流模拟中的作用。最后，按照相同的技术路线，对局地强对流个例0813进行综合分析，证明上述的研究结果及技术路线是合理、有效的。本研究主要结论如下：
1)随着城市化进程的发展，上海夏季热岛呈现扩大化趋势。上海热岛效应在午后（14时）出现频率最高，热岛平均强度也最大，这可能是局地强对流多发的诱因。上海夏季强对流天气发生次数呈逐年递增趋势，2001年后强对流发生频次有明显的增加。从空间上看，表现为中心城区多，郊县少的特点，黄浦江沿岸及淀山湖边频次较高。
2)上海三面环水的城市下垫面特征，导致夏季午后东北、东南两支海陆风及湖陆风的爆发；城市热岛效应显著时，南北两支海陆风在中心城区形成地面辐合线时，极易导致局地强对流的爆发和发展；而湖陆风在城市热岛环流叠加下，也容易激发强对流。
3）快速更新同化数值试验成功地再现了2011年7月31日午后强对流天气过程。通过对降水、近地面风场、沿上海城市的垂直剖面环流场、垂直速度场等方面的分析，揭示了本次强对流天气过程的发生发展机制。上海市与海洋之间的热力差异产生海陆风；由于上海以锲形形状伸入东海，因而有南北两支海陆风的生成；而中心城区热岛效应，加剧了海陆风发展，两支海陆风在城市上空叠加，近地面的水平风速加大，在城区近地层形成了中尺度地面风场辐合线；两支海风带来水汽与高层云叠加，带来了水汽并加剧了对流的发展。而西侧湖陆风与大尺度背景场的偏西风叠加，加剧了近地层风场的辐合，辐合线附近产生强烈的上升运动，有利于暴雨的产生。
4）敏感性试验表明，海陆风的形成对强对流天气的发生和发展有着重要的影响。当把整个上海市下垫面更换为水体后，海风深入上海，因此形成的辐合线较ctrl试验更偏西，造成垂直抬升运动的位置也偏西，因此强降水的位置也会发生相应的西移；对流不稳定能量的积累受到抑制，不利于强对流的发展。当不考虑太阳的日变化时，陆面增温明显慢于ctrl试验，稳定的大气不利于能量的累积，因此不利于触发大气产生湍流运动，强对流天气不易发生；此外，地面温度的变化导致地面辐合线位置的变化，从而导致垂直方向上环流场的变化，继而降水的强度和位置发生了明显的变化。
5）通过观测资料的敏感性试验，得到以下认识：
去除雷达资料后，降水落区和量级都不能很好的把握，控制试验将雷达通过云分析调整优化了初始场中的水物质含量（云水、云冰、雨水、雪、冰雹等混合比），增加了模式中的低层云量，同时剔除了虚假的高云，尤其对东方体育中心上空的云冰进行了剔除，使云的分布更加合理，从而对中小尺度对流系统的结构和强度起到了关键性作用，这些因素是预报降水场改进的直接原因。
去除探空资料后，500hPa、700hPa和850hPa高度场和风场均不同程度变差，继而模拟的地面降温范围和强度均发生了变化。尽管探空资料仅在试验中使用了一次，但其影响可持续影响到后续时刻的试验。
AMDAR资料试验表明：没有同化AMDAR资料，模拟的弱冷空气南压稍快，16时~17时强降水中心明显偏南；同化AMDAR资料后，500hPa和850hPa的高空槽的位置和强度都有所调整，此次强对流天气的形势场得到了很好的修正，使得分析场更接近实况，这是对预报降水改进的主要原因。更重要的是，AMDAR资料每个时次均可获得，这是对探空资料最好的补充。
地面观测资料（常规地面站和自动站）试验表明：加入地面观测资料后，即近地层的温度、湿、风等要素的调整，有效地调整了地面温湿场；试验中心城区的热岛效应明显，海陆风的发展模拟更有效；此外，温度场梯度分布影响着风场，决定了地面辐合线的位置，直接影响降水的落区。因此，地面观测资料在快速更新同化试验中，也是不可或缺的重要高频次资料。
6）采用相同的技术路线，可再现2013年8月13日强对流天气过程。与实况相比，数值试验模拟的降水落区及降水随时间演变与实况基本一致，降水起止时间略有提前，但模拟的降水范围略偏大。模拟的边界层中尺度辐合线、海陆风的时空演变以及垂直环流特征与强对流发生的时空变化基本一致。该个例分析再次证明上海局地强对流的发生、发展与城市热岛效应及海陆风有密切关系。海陆温差是导致东北海陆风形成并加剧、东南海陆风增强的直接原因，同时边界层急流提供了充沛的水汽条件、加剧了地面的中尺度辐合和局地对流的发展。
Using Shanghai observational network, the development of the “0731” localsevere convection case was analyzed and discussed, then conceptual model of theearly-afternoon summer convections in Shanghai was put forward. By rapidupdated cycling forecasting experiments, the rationality of the conceptual modelwas validated. The future studied the role of UHI, sea breeze in the initiation anddevelopment of the local severe convection case. By experiments, the effects of allkinds of observation data in experiments were verified. Finally, using the samemethods, the “0813”local severe convection case was studied, and the rationalityof the research results and technical route was validated. The main conclusions canbe drawn as follows:
1) With the development of urbanization, the zone of UHI in Shanghai isexpanding. the original UHI is in the urban area, then expanded to the suburbancounty, formed a more large UHI. The frequency of UHI is highest in theafternoon(14Z), and the intensity of UHI is greatest, these may be the causes oflocal severe convection cases. The frequency of severe convection in Shanghaisummer was increasing, obvious increase was appeared after2001. The severeconvections mostly occurred in downtown area and regions along the rivers orlakes.
2) The peculiar geography of Shanghai led to the break out of the sea breezeand the lake breeze. When the intensity is greater, the surface convergence line indowntown area was formed due to the sea breeze effects, which cause the localconvection initialization and development. While the lake breeze can cause thesevere convection under the background of the UHI circulation.
3) Rapid updated cycling forecasting system successfully reveals the severeconvection case, which occurred in Shanghai on July312011. Through thediagnostic analysis of rainfall, surface wind, vertical circulation and verticalvelocity field and so on, the results revealed the mechanisms of the triggering anddeveloping. It was also found that due to the thermal difference between landand sea, and the peculiar geography of Shanghai, two sea breeze from differentnorth and south directions met in shanghai area. At the same time low levelconvergence line formed, as a results, weak updraft in boundary layer appeared,combined with the local urban heat island effect, created favorable conditions fortriggering the server convections; and the high gradient of moisture between lowlevel and middle level and its unstable vertical structure provided favorableconditions for moisture vertical transport. The coupling of lake breeze on the westside and the environment west wind, caused strong low level convergence line andstrong upward motion, which result the heavy rainfall.
4)The simulation of sensitivity experiments show that the sea breeze has impact on genesis and development of server convection weather. In underlyingsurface test which using water body as the surface in Shanghai, sea breezeextended to Shanghai area, as the result the low level convergence line and theupward motion is more westward, and so heavy rainfall occurred westly,. All ofthese can restrain the unstable energy, and impede severe convection development.In diurnal variation test, the warming of the land surface temperature issignificantly slower than CTRL test, the steadiness of the atmosphere is notbeneficial to accumulate energy. And so the turbulent motion and the severeconvection are not beneficial. In addition, the change of surface temperature leadto the change of the location of low level convergence and the vertical circulation,and result that the change of the intensity and location of rainfall obviously.
5) The sensitivity experiment was performed to investigate the effects ofobservation data. There are some conclusions as below.
The initial radar data assimilation is able to significantly improve theprecipitation location and intensity prediction. Cloud analysis scheme that utilizesradar observations adjusted the initial state of the model on water substance, suchas cloud water mixing ratio, cloud ice mixing ratio, rain water mixing ratio, icemixing ratio, snow mixing ratio, graupel mixing ratio, et al. And lower cloudamount increased, false high clouds were removed. Cloud ice over was orientalsports center removed to improve distribution of cloud. It is found that cloudanalysis adjusted structure and intensity of a mesoscale convection system directly and the precipitation was a close relation to distribution of cloud water substance.
The results show that the simulation results whose height fields and windfields at500hPa,700hPa and850hPa assimilating the radiosonde data are moreclose to the environment than the ones which not. This, region and strength of adrop in temperature have changed. However, the radiosonde data used once insimulation, it affected forecasting.
The simulation of assimilated AMDAR data show that, without assimilatingAMDAR data, the intrusion of north wind accompanied by cold advection morequicker than actual weather and rainfall center locating by north at16:00~17:00.Location and strength of upper trough at500hPa and850hPa assimilating theradiosonde data are more close to the environment than the ones which not. Mostimportantly, every time can get AMDAR data, it supply and improve soundingsdata.
The simulation results of assimilated surface meteorological observation datashow that temperature fields, humidity fields and wind fields were adjusted atsurface boundary layer, and an obvious heat island and breeze were morereasonable. Moreover, precipitation location affected by surface wind convergenceline which influenced distribution of temperature gradient. Surface observationdata was indispensable and important data in rapid refresh system.
6) Severe convection weather on August132013can reproduce by using thesame way. Compare actual weather, the simulation precipitation location andprecipitation time is close to the real case, started and stopped of precipitationadvance in time, and range of precipitation is more wider than actual. Moreover,spatial-temporal distribution of PBL (planetary boundary layer) mesoscaleconvergence line and breeze, variation characteristics of scale vertical circulationwere similar to spatial-temporal distribution of server convection weather. Againthis weather case proved that the occurrence and development of the severeconvection weather in Shanghai were closely related to urban heat island effectand breeze. The direct reason for formation and strengthen of breeze wasdifference in heating between the land and sea. The southeasterly andnortheasterly boundary layer jet stream brought abundant plentiful moisture, itstrengthened the mesoscale convergence near surface and the development of localconvection weather.
1)随着城市化进程的发展，上海夏季热岛呈现扩大化趋势。上海热岛效应在午后（14时）出现频率最高，热岛平均强度也最大，这可能是局地强对流多发的诱因。上海夏季强对流天气发生次数呈逐年递增趋势，2001年后强对流发生频次有明显的增加。从空间上看，表现为中心城区多，郊县少的特点，黄浦江沿岸及淀山湖边频次较高。
2)上海三面环水的城市下垫面特征，导致夏季午后东北、东南两支海陆风及湖陆风的爆发；城市热岛效应显著时，南北两支海陆风在中心城区形成地面辐合线时，极易导致局地强对流的爆发和发展；而湖陆风在城市热岛环流叠加下，也容易激发强对流。
3）快速更新同化数值试验成功地再现了2011年7月31日午后强对流天气过程。通过对降水、近地面风场、沿上海城市的垂直剖面环流场、垂直速度场等方面的分析，揭示了本次强对流天气过程的发生发展机制。上海市与海洋之间的热力差异产生海陆风；由于上海以锲形形状伸入东海，因而有南北两支海陆风的生成；而中心城区热岛效应，加剧了海陆风发展，两支海陆风在城市上空叠加，近地面的水平风速加大，在城区近地层形成了中尺度地面风场辐合线；两支海风带来水汽与高层云叠加，带来了水汽并加剧了对流的发展。而西侧湖陆风与大尺度背景场的偏西风叠加，加剧了近地层风场的辐合，辐合线附近产生强烈的上升运动，有利于暴雨的产生。
4）敏感性试验表明，海陆风的形成对强对流天气的发生和发展有着重要的影响。当把整个上海市下垫面更换为水体后，海风深入上海，因此形成的辐合线较ctrl试验更偏西，造成垂直抬升运动的位置也偏西，因此强降水的位置也会发生相应的西移；对流不稳定能量的积累受到抑制，不利于强对流的发展。当不考虑太阳的日变化时，陆面增温明显慢于ctrl试验，稳定的大气不利于能量的累积，因此不利于触发大气产生湍流运动，强对流天气不易发生；此外，地面温度的变化导致地面辐合线位置的变化，从而导致垂直方向上环流场的变化，继而降水的强度和位置发生了明显的变化。
5）通过观测资料的敏感性试验，得到以下认识：
去除雷达资料后，降水落区和量级都不能很好的把握，控制试验将雷达通过云分析调整优化了初始场中的水物质含量（云水、云冰、雨水、雪、冰雹等混合比），增加了模式中的低层云量，同时剔除了虚假的高云，尤其对东方体育中心上空的云冰进行了剔除，使云的分布更加合理，从而对中小尺度对流系统的结构和强度起到了关键性作用，这些因素是预报降水场改进的直接原因。
去除探空资料后，500hPa、700hPa和850hPa高度场和风场均不同程度变差，继而模拟的地面降温范围和强度均发生了变化。尽管探空资料仅在试验中使用了一次，但其影响可持续影响到后续时刻的试验。
AMDAR资料试验表明：没有同化AMDAR资料，模拟的弱冷空气南压稍快，16时~17时强降水中心明显偏南；同化AMDAR资料后，500hPa和850hPa的高空槽的位置和强度都有所调整，此次强对流天气的形势场得到了很好的修正，使得分析场更接近实况，这是对预报降水改进的主要原因。更重要的是，AMDAR资料每个时次均可获得，这是对探空资料最好的补充。
地面观测资料（常规地面站和自动站）试验表明：加入地面观测资料后，即近地层的温度、湿、风等要素的调整，有效地调整了地面温湿场；试验中心城区的热岛效应明显，海陆风的发展模拟更有效；此外，温度场梯度分布影响着风场，决定了地面辐合线的位置，直接影响降水的落区。因此，地面观测资料在快速更新同化试验中，也是不可或缺的重要高频次资料。
6）采用相同的技术路线，可再现2013年8月13日强对流天气过程。与实况相比，数值试验模拟的降水落区及降水随时间演变与实况基本一致，降水起止时间略有提前，但模拟的降水范围略偏大。模拟的边界层中尺度辐合线、海陆风的时空演变以及垂直环流特征与强对流发生的时空变化基本一致。该个例分析再次证明上海局地强对流的发生、发展与城市热岛效应及海陆风有密切关系。海陆温差是导致东北海陆风形成并加剧、东南海陆风增强的直接原因，同时边界层急流提供了充沛的水汽条件、加剧了地面的中尺度辐合和局地对流的发展。
Using Shanghai observational network, the development of the “0731” localsevere convection case was analyzed and discussed, then conceptual model of theearly-afternoon summer convections in Shanghai was put forward. By rapidupdated cycling forecasting experiments, the rationality of the conceptual modelwas validated. The future studied the role of UHI, sea breeze in the initiation anddevelopment of the local severe convection case. By experiments, the effects of allkinds of observation data in experiments were verified. Finally, using the samemethods, the “0813”local severe convection case was studied, and the rationalityof the research results and technical route was validated. The main conclusions canbe drawn as follows:
1) With the development of urbanization, the zone of UHI in Shanghai isexpanding. the original UHI is in the urban area, then expanded to the suburbancounty, formed a more large UHI. The frequency of UHI is highest in theafternoon(14Z), and the intensity of UHI is greatest, these may be the causes oflocal severe convection cases. The frequency of severe convection in Shanghaisummer was increasing, obvious increase was appeared after2001. The severeconvections mostly occurred in downtown area and regions along the rivers orlakes.
2) The peculiar geography of Shanghai led to the break out of the sea breezeand the lake breeze. When the intensity is greater, the surface convergence line indowntown area was formed due to the sea breeze effects, which cause the localconvection initialization and development. While the lake breeze can cause thesevere convection under the background of the UHI circulation.
3) Rapid updated cycling forecasting system successfully reveals the severeconvection case, which occurred in Shanghai on July312011. Through thediagnostic analysis of rainfall, surface wind, vertical circulation and verticalvelocity field and so on, the results revealed the mechanisms of the triggering anddeveloping. It was also found that due to the thermal difference between landand sea, and the peculiar geography of Shanghai, two sea breeze from differentnorth and south directions met in shanghai area. At the same time low levelconvergence line formed, as a results, weak updraft in boundary layer appeared,combined with the local urban heat island effect, created favorable conditions fortriggering the server convections; and the high gradient of moisture between lowlevel and middle level and its unstable vertical structure provided favorableconditions for moisture vertical transport. The coupling of lake breeze on the westside and the environment west wind, caused strong low level convergence line andstrong upward motion, which result the heavy rainfall.
4)The simulation of sensitivity experiments show that the sea breeze has impact on genesis and development of server convection weather. In underlyingsurface test which using water body as the surface in Shanghai, sea breezeextended to Shanghai area, as the result the low level convergence line and theupward motion is more westward, and so heavy rainfall occurred westly,. All ofthese can restrain the unstable energy, and impede severe convection development.In diurnal variation test, the warming of the land surface temperature issignificantly slower than CTRL test, the steadiness of the atmosphere is notbeneficial to accumulate energy. And so the turbulent motion and the severeconvection are not beneficial. In addition, the change of surface temperature leadto the change of the location of low level convergence and the vertical circulation,and result that the change of the intensity and location of rainfall obviously.
5) The sensitivity experiment was performed to investigate the effects ofobservation data. There are some conclusions as below.
The initial radar data assimilation is able to significantly improve theprecipitation location and intensity prediction. Cloud analysis scheme that utilizesradar observations adjusted the initial state of the model on water substance, suchas cloud water mixing ratio, cloud ice mixing ratio, rain water mixing ratio, icemixing ratio, snow mixing ratio, graupel mixing ratio, et al. And lower cloudamount increased, false high clouds were removed. Cloud ice over was orientalsports center removed to improve distribution of cloud. It is found that cloudanalysis adjusted structure and intensity of a mesoscale convection system directly and the precipitation was a close relation to distribution of cloud water substance.
The results show that the simulation results whose height fields and windfields at500hPa,700hPa and850hPa assimilating the radiosonde data are moreclose to the environment than the ones which not. This, region and strength of adrop in temperature have changed. However, the radiosonde data used once insimulation, it affected forecasting.
The simulation of assimilated AMDAR data show that, without assimilatingAMDAR data, the intrusion of north wind accompanied by cold advection morequicker than actual weather and rainfall center locating by north at16:00~17:00.Location and strength of upper trough at500hPa and850hPa assimilating theradiosonde data are more close to the environment than the ones which not. Mostimportantly, every time can get AMDAR data, it supply and improve soundingsdata.
The simulation results of assimilated surface meteorological observation datashow that temperature fields, humidity fields and wind fields were adjusted atsurface boundary layer, and an obvious heat island and breeze were morereasonable. Moreover, precipitation location affected by surface wind convergenceline which influenced distribution of temperature gradient. Surface observationdata was indispensable and important data in rapid refresh system.
6) Severe convection weather on August132013can reproduce by using thesame way. Compare actual weather, the simulation precipitation location andprecipitation time is close to the real case, started and stopped of precipitationadvance in time, and range of precipitation is more wider than actual. Moreover,spatial-temporal distribution of PBL (planetary boundary layer) mesoscaleconvergence line and breeze, variation characteristics of scale vertical circulationwere similar to spatial-temporal distribution of server convection weather. Againthis weather case proved that the occurrence and development of the severeconvection weather in Shanghai were closely related to urban heat island effectand breeze. The direct reason for formation and strengthen of breeze wasdifference in heating between the land and sea. The southeasterly andnortheasterly boundary layer jet stream brought abundant plentiful moisture, itstrengthened the mesoscale convergence near surface and the development of localconvection weather.
引文
[1]田晨，周伟灿，苗峻峰.中国地区下垫面特征对强对流天气影响研究进展[J].气象科技，2012，40(2):207-212.
[2]翟国庆，丁华君，高坤．中尺度下垫面特征对气旋暴雨影响的数值试验[J].杭州大学学报，1995，22(2):185-190．
[3]刘永强，叶笃正，季劲钧．土壤湿度和植被对气候的影响Ⅰ：短期气候异常持续性的理论分析[J]．中国科学（B辑），1992，(4):441-448．
[4]刘永强，叶笃正，季劲钧．土壤湿度和植被对气候的影响Ⅱ：短期气候异常持续性的数值试验[J]．中国科学（B辑），1992，(4):554-560．
[5] Roger A, Pielke Sr. Influence of the spatial distribution of vegetation and soild on theprediction of cumulus convective rainfall [J]. Reviews of Geophysics,2001,39（2）：151-177.
[6]张弛，束炯.土地利用类型变化对城市大气边界层特征影响的数值模拟[J].华东师范大学学报（自然科学版），4:83-93.
[7]殷健，梁珊珊.城市化对上海市区域降水的影响[J].水文，2010，30(2):66-72.
[8] Huff, F. A., and S. A. Changnon. Urban effects on daily rainfall distribution [C]. Preprints,Second National Conf. on Weather Modification, Santa Barbara, CA, Amer. Meteor. Soc.,1972,215–220.
[9] Huff, F. A., and S. A. Changnon. Precipitation modification of major urban areas [J]. Bull.Amer. Meteor. Soc.,1973,54,1220–1232.
[10] Bornstein, R., and Q. Lin. Urban heat islands and summertime convective thunderstormsin Atlanta: Three case studies [J]. Atmos. Environ.,2000，34,507–516.
[11] Diem, Jeremy E., Thomas L. Mote. Interepochal changes in summer precipitation in thesoutheastern United States: Evidence of possible urban effects near Atlanta, Georgia [J]. J.Appl. Meteor.,2005，44,717–730.
[12] Chen Tsingchang, Wang Shihyu, Yen Mingcheng. Enhancement of afternoonthunderstorm activity by urbanization in a valley: Taipei [J]. Journal of AppliedMeteorology and Climatology,2007,46:1324-1340.
[13]蒙伟光，闫敬华，扈海波．城市化对珠江三角洲强雷暴天气的可能影响[J]．大气科学，2007，31（2）：364-376.
[14]杨露华，叶其欣，邬锐等.基于GPS/Pwv资料的上海地区2004年一次夏末暴雨的水汽输送分析[J].气象科学，2006,25(5):502-508
[15]尹红萍，曹晓岗.盛夏上海地区副热带高压型强对流特点分析仁[J].气象，2010，36(8):19-25.
[16]漆梁波，陈雷.上海局地强对流天气临近预报要点[J].气象，2009，35(9):11-18
[17]顾问，路璐，施春红等.上海盛夏连续清晨对流天气过程的边界层结构分析[J].大气科学研究与应用，2011，2:55-62
[18]姚建群，戴建华，姚祖庆.一次强飑线的成因及维持和加强机制分析[J].应用气象学报，2005，16(6):746-753
[19]张德林，马雷鸣.“0730”上海强对流天气个例的中尺度观测分析及数值模拟[J].气象，2010，36(3):62-69
[20]孙继松.北京地区夏季边界层急流的基本特征及形成机理研究[J].大气科学，2005，29(3):445-452.
[21]孙继松.城市边界层过程在北京2004年7月10日局地暴雨过程中的作用[J].大气科学，2006，30(2):221-234.
[22]孙继松,杨波.地形与城市环流共同作用下的β中尺度暴雨.大气科学，2008，32(6):1352-1364.
[23]郑祚芳，刘红燕，张秀丽.局地强对流天气分析中非常规探测资料应用[J].气象科技，2009，37(2)：243-247.
[24]丁青兰，王令，卞素芬等.北京局地降水中地形和边界层辐合线的作用[J].气象科技，2009，37(2)：152-156.
[25]赵金霞，徐灵芝，卢焕珍等.盛夏渤海湾大气边界层辐合线触发对流风暴对比分析[J].气象，2012，38(3):336-343.
[26]卢焕珍，赵玉洁，俞小鼎等.雷达观测的渤海湾海陆风辐合线与自动站资料的对比分析[J].气象科技，2008，34(9)：57-64.
[27]王彦，于莉莉，李艳伟等.边界层辐合线对强对流系统形成和发展的作用[J].应用气象学报，2011，22(6):724-731.
[28]陈明轩，俞小鼎，谭晓光等.对流天气临近预报技术的发展与研究进展[J].应用气象学报，2004，15(6):754-766.
[29]胡胜，曾沁，冯业荣.北京2008奥运会天气预报示范项目基本情况简介[J].广东气象，2008，12(6):1-4.
[30]刘英，柳崇健，周海光.突发性强灾害天气预警系统：理论与应用[J].地球科学进展，2008，23(11):1141-1149.
[31]陈明轩，高峰，孔荣等.自动临近预报系统及其在北京奥运期间的应用.应用气象学报：2010，21(4):395-404.
[32]Wilson, J. W. Precipitation Nowcasting: Past, Present and Future.6th InternationalSymposium on Hydrological Applications of Weather Radar.2011.
[33] Benjamin, S.G., D. Devenyi, S. Weygandt, etal. An hourly assimilation-forecast cycle:The RUC. Mon. Wea. Rev.,2004,132,495-518.
[34] Benjamin, S.G., G. Grell, J.M. Brown, T.G Smirnova etal.. Mesoscale weather predictionwith the RUC hybrid isentropic-terrain-following cordinate model. Mon. Wea. Rev.,2004,132,473-494.
[35] Benjamin, S.G., D. Devenyi, T. Smirnova, etal.. From the13-km RUC to the RapidRefresh.12th Conf. on Aviation, Range, and Aerospace Meteorology (ARAM), Atlanta,GA, Amer. Meteor. Soc.2006, CD-ROM,9.1.
[36] Benjamin, S. G., S. Weygandt, J. M. Brown, etal.. From the radar-enhanced RUC to theWRF-based Rapid Refresh.22nd Conf. Wea.Anal.and Fcstg./18th Conf. Numer. Wea.Pred., Park City, Amer. Meteor.Soc.,2007, Paper J3.4.
[37] Benjamin S. G., S. S. Weygandt, C. R. Alexander, etal.. NOAA's hourly-updated3kmHRRR and RUC/Rapid Refresh-recent (2010) and upcoming changes toward improvingweather guidance for air-traffic management.2011.
[38] Benjamin, S. G., S. S. Weygandt, J. M. Brown, etal.. Assimilation of METAR cloudandvisibility observations in the RUC.11thConference on Aviation, Range, Aerospaceand22nd Conference on Severe Local Storms, Hyannis, MA, Amer. Meteor. Soc.2004,Paper9.13.
[39] Weygandt, S., S. G. Benjamin, D. Dévényi, etal.. Cloud and hydrometeor analysis usingmetar, radar, and satellite data within the RUC/Rapid-Refresh model.12th Conference onAviation Rang and Aerospace Meteorology, Atlanta, GA.2006.
[40] Brown, J.M., S. G. Benjamin, T. Smirnova, etal.. Rapid-Refresh Core Test: aspects ofWRFNMM and WRF-ARW forecast performance relevant to the Rapid-Refreshapplication.18thConf. Num. Wea.Pred., Park City, UT, Amer. Meteor.Soc.,2007,4A.1.
[41] Hu, M., S. Weygandt, S. G. Benjamin, M. Xue. Ongoing development and testing ofgeneralized cloud analysis package within GSI for initializing Rapid Refresh, Preprints,13thConf. on Aviation, Range and Aerospace Meteorology. New Orleans, LA, Amer.Meteor. Soc.,2008, Paper7.4.
[42] Brown, J. M., T. G. Smirnova, S. G. Benjamin, etal. Rapid-Refresh testing: example offorecast performance.13th Conference on Aviation Rang and Aerospace Meteorology,New Orleans, LA, Amer. Meteor. Soc.,2008, Paper7.1.
[43] Bratseth, A. M. Statistical interpolation by means of successive corrections. Tellus,1986,38A,439-447.
[44] Sashegyi, K.D., D.E. Harms, R.V. Madala, S. Raman. Application of the Bratseth schemefor the analysis of GALE data using a mesoscale model. Mon. Wea. Rev.,1993,121,2331-2350.
[45] Cole, R.E. and F. W. Wilson. ITWS gridded wind product, Preprints,6th Conf. onAviation Wetaher Systems, AMS, Boston.1995
[46] Zhang J. Moisture and diabatic initialization based on radar and satelliteobservations, Ph.D. dissertation, University of Oklahoma, School of Meteorology,1999,194pp
[47] Brewster, K.. Recent advances in the diabatic initialization of a non-hydrostaticnumerical model. Preprints,21st Conf. on Severe Local Storms, and Preprints,15th Conf.Num. Wea.2002.
[48] Hu, M..3DVAR and cloud analysis with WSR-88D level-II data for the prediction oftornadic thunderstorms, Ph. D. Dissertation, School of Meteorology, University ofOklahoma,2005,217pp.
[49]杨艳蓉.面向雷电预报的雷达、卫星遥感资料同化及其应用初步研究，南京信息工程大学博士学位论文.2008.
[50] Zhang, J., f. Carr, K. Brewster. ADAS cloud analysis, Preprints,12th conference onNumerical Weather Prediction, Boston,1998,185-188
[51] Haines P. A., J. K. Luers, C. A Cerbus. The role of the Smith-Feddes model in improvingthe forecasting of aircraft icing. Preprints, Third Conf on Aviation Weather Systems,Anaheim, Cal, Amer Meteor Soc,1989,258-263.
[52] Benjamin S. G., Weygandt S S, etal. An hourly assimilation-forecast cycle: The RUC.Mon Wea Rev,2004,132(2):495-518
[53] Lynch. P. and X.-Y. Huang. Initialization of the HIRLAM model using a digital filter.Mon. Wea. Rev.1992,120,1019-1034
[54] WRF user guide V3.3available from website:http://www.mmm.ucar.edu/wrf/users/pub-doc.html
[55]吴浩云.太湖流域典型年梅雨洪涝灾害比较分析[J].水文，2000，20（4）：54-57.
[56]陈艺敏，钱永甫.116a长江中下游梅雨的气候特征[J].南京气象学院学报，2004,27(1):65-72.
[57]杨义文.长江中下游梅雨与中国夏季旱涝分布[J].气象，2002,28(11):11-15.
[58]张运林，罗潋葱.江苏省近47年来的梅雨特征及灾害影响[J].灾害学，2003,18(2):58-62.
[59]田心如，姜爱军，高苹，等.江苏省梅雨及其灾害影响评估分析[J].灾害学，2004,19(1):40-45.
[60]李峰.东亚季风和中国梅雨暴雨研究的评述[J].成都气象学院学报，2000，15(1):72-77.
[61]张小玲，陶诗言，张顺利.梅雨锋上的三类暴雨[J].大气科学，2004,28(2):187-205.
[62]孙彭龄.浙江的梅雨和台风年际变化[J].浙江气象科技，1993，20(3):18-21.
[63]张艳，鲍文杰，余琦等．超大城市热岛效应的季节变化特征及其年际差异[J]．地球物理学报，2012，55(4),1121-1128.
[64]陶诗言.中国之暴雨.北京：科学出版社，1980:8-11.
[65] Wilson, James W., Wendy E. Schreiber. Initiation of convective storms at radar-observedboundary-layer convergence lines [J]. Mon. Wea. Rev.,1986,114,2516–2536.
[66] Wilson, James W., G. Brant Foote, N. Andrew Crook, James C. Fankhauser, Charles G.Wade, John D. Tuttle, Cynthia K. Mueller, Steven K. Krueger. The role of boundary-layerconvergence zones and horizontal rolls in the initiation of thunderstorms: A Case Study [J].Mon. Wea. Rev.,1992，120,1785–1815.
[67] Wilson, James W., Cynthia K. Mueller. Nowcasts of thunderstorm initiation andevolution [J]. Wea. Forecasting,1993,8,113–131.
[68] Roger A. Pielke. A Three-Dimensional Numerical Model of the Sea Breezes Over SouthFlorida. Monthly Weather Review，1974，Vol.102, No.2, pp.115–139.
[69]杨薇，苗俊峰，谭哲敏.太湖地区湖陆风对雷暴过程影响的数值模拟.第九届长三角气象科技论坛论文集.
[70]王谦谦，钱永甫.太阳辐射日变化对夏季风模拟特征的影响.气象学报，1997,55（3）：334-345.
[71] Hu M. and Xue M.. Initializing convection using cloud analysis and radar data ingrid-point statistical interpolation(GSI) system and impact on the forecast of advancedresearch WRF Accepted. Geophys. Res. Letters,2007b..
[72] Xue M., Wang D., Hou D., etal. Prediction of the7May1995squall line over the centralU S with intermittent data assimilation12th conf on numerical weather prediction.Phoenix AZ, Amer Meteor Soc,1998，191-194.
[73] Haase G., Grewell S., Simmer C, etal. Assimilation of radar data in mesoscale models:Physical initialization and latent heat nudging. Phys Chem Earth (B),2003,25:1237-1242.
[74] Hu M., Xue M., Brewster K.3DVAR and cloud analysis with WSR-88D Level-Ⅱ Datafor the prediction of the Fort Worth, Texas, Tornadic Thunderstorms. Part Ⅰ: CloudAnalysis and Its Impact. Mon. Wea. Rev.,2006,134:675-698.
[75] Hu M, Xue M, Gao J., etal.3DVAR and cloud analysis with WSR-88D Level-Ⅱ Data forthe prediction of the Fort Worth, Texas, Tornadic Thunderstorms. Part Ⅱ: Impact ofRadial Velocity Analysis via3DVAR. Mon. Wea. Rev.,2006,134:699-721.
[76]张利红，蒋丽娟，陈朝平等.探空观测资料在西南暴雨中的同化试验.2009，高原山地气象研究，29(3):31-38.
[77]舒守娟，王元，宋金杰.西北太平洋台风“海棠”结构的GPS下投式探空仪观测分析.2011，气象学报，69(6):933-944.
[78]黄燕燕，万齐林，陈子通等.加密探空资料在华南暴雨数值预报的应用试验.2011，热带气象学报，27(2):179-188.
[79]梁科，万齐林，丁伟钰等.飞机报资料在0506华南致灾暴雨过程模拟中的应用.2007，热带气象学报，23(4):313-325.
[80]黄卓，李延香，王慧等. AMDAR资料在天气预报中的应用.2006，气象，32（9）：42-48.
[81]邵长亮. WRF-EnSRF同化地面自动站资料的实现与技术研究.2011，南京信息工程大学硕士学位论文.
[82]陈新梅、段旭、王曼等.使用于云南地区的地面资料同化方案设计及个例分析.高原气象，2010，29（4）：907-917.
[83]良敏.自动站资料质量控制及其三维变分同化研究.南京信息工程大学硕士学位论文.
[84] Zhang D L, J M Fritcsh. A case study of the sensitivity of numerical model simulation ofmesoscale convective systems to varying initial condition. Mon. Wea. Rev.,1986,114:2481-2431.
[85]吴海英，曾明剑，尹东屏等.一次苏皖特大暴雨过程中边界层急流结构演变特征和作用分析.高原气象，2010，29(6)：1431-1440.
[86]何立富，陈涛，周庆亮，等.北京“7.10”暴雨β中尺度对流系统分析.应用气象学报，2007，18(5):655-665.
[2]翟国庆，丁华君，高坤．中尺度下垫面特征对气旋暴雨影响的数值试验[J].杭州大学学报，1995，22(2):185-190．
[3]刘永强，叶笃正，季劲钧．土壤湿度和植被对气候的影响Ⅰ：短期气候异常持续性的理论分析[J]．中国科学（B辑），1992，(4):441-448．
[4]刘永强，叶笃正，季劲钧．土壤湿度和植被对气候的影响Ⅱ：短期气候异常持续性的数值试验[J]．中国科学（B辑），1992，(4):554-560．
[5] Roger A, Pielke Sr. Influence of the spatial distribution of vegetation and soild on theprediction of cumulus convective rainfall [J]. Reviews of Geophysics,2001,39（2）：151-177.
[6]张弛，束炯.土地利用类型变化对城市大气边界层特征影响的数值模拟[J].华东师范大学学报（自然科学版），4:83-93.
[7]殷健，梁珊珊.城市化对上海市区域降水的影响[J].水文，2010，30(2):66-72.
[8] Huff, F. A., and S. A. Changnon. Urban effects on daily rainfall distribution [C]. Preprints,Second National Conf. on Weather Modification, Santa Barbara, CA, Amer. Meteor. Soc.,1972,215–220.
[9] Huff, F. A., and S. A. Changnon. Precipitation modification of major urban areas [J]. Bull.Amer. Meteor. Soc.,1973,54,1220–1232.
[10] Bornstein, R., and Q. Lin. Urban heat islands and summertime convective thunderstormsin Atlanta: Three case studies [J]. Atmos. Environ.,2000，34,507–516.
[11] Diem, Jeremy E., Thomas L. Mote. Interepochal changes in summer precipitation in thesoutheastern United States: Evidence of possible urban effects near Atlanta, Georgia [J]. J.Appl. Meteor.,2005，44,717–730.
[12] Chen Tsingchang, Wang Shihyu, Yen Mingcheng. Enhancement of afternoonthunderstorm activity by urbanization in a valley: Taipei [J]. Journal of AppliedMeteorology and Climatology,2007,46:1324-1340.
[13]蒙伟光，闫敬华，扈海波．城市化对珠江三角洲强雷暴天气的可能影响[J]．大气科学，2007，31（2）：364-376.
[14]杨露华，叶其欣，邬锐等.基于GPS/Pwv资料的上海地区2004年一次夏末暴雨的水汽输送分析[J].气象科学，2006,25(5):502-508
[15]尹红萍，曹晓岗.盛夏上海地区副热带高压型强对流特点分析仁[J].气象，2010，36(8):19-25.
[16]漆梁波，陈雷.上海局地强对流天气临近预报要点[J].气象，2009，35(9):11-18
[17]顾问，路璐，施春红等.上海盛夏连续清晨对流天气过程的边界层结构分析[J].大气科学研究与应用，2011，2:55-62
[18]姚建群，戴建华，姚祖庆.一次强飑线的成因及维持和加强机制分析[J].应用气象学报，2005，16(6):746-753
[19]张德林，马雷鸣.“0730”上海强对流天气个例的中尺度观测分析及数值模拟[J].气象，2010，36(3):62-69
[20]孙继松.北京地区夏季边界层急流的基本特征及形成机理研究[J].大气科学，2005，29(3):445-452.
[21]孙继松.城市边界层过程在北京2004年7月10日局地暴雨过程中的作用[J].大气科学，2006，30(2):221-234.
[22]孙继松,杨波.地形与城市环流共同作用下的β中尺度暴雨.大气科学，2008，32(6):1352-1364.
[23]郑祚芳，刘红燕，张秀丽.局地强对流天气分析中非常规探测资料应用[J].气象科技，2009，37(2)：243-247.
[24]丁青兰，王令，卞素芬等.北京局地降水中地形和边界层辐合线的作用[J].气象科技，2009，37(2)：152-156.
[25]赵金霞，徐灵芝，卢焕珍等.盛夏渤海湾大气边界层辐合线触发对流风暴对比分析[J].气象，2012，38(3):336-343.
[26]卢焕珍，赵玉洁，俞小鼎等.雷达观测的渤海湾海陆风辐合线与自动站资料的对比分析[J].气象科技，2008，34(9)：57-64.
[27]王彦，于莉莉，李艳伟等.边界层辐合线对强对流系统形成和发展的作用[J].应用气象学报，2011，22(6):724-731.
[28]陈明轩，俞小鼎，谭晓光等.对流天气临近预报技术的发展与研究进展[J].应用气象学报，2004，15(6):754-766.
[29]胡胜，曾沁，冯业荣.北京2008奥运会天气预报示范项目基本情况简介[J].广东气象，2008，12(6):1-4.
[30]刘英，柳崇健，周海光.突发性强灾害天气预警系统：理论与应用[J].地球科学进展，2008，23(11):1141-1149.
[31]陈明轩，高峰，孔荣等.自动临近预报系统及其在北京奥运期间的应用.应用气象学报：2010，21(4):395-404.
[32]Wilson, J. W. Precipitation Nowcasting: Past, Present and Future.6th InternationalSymposium on Hydrological Applications of Weather Radar.2011.
[33] Benjamin, S.G., D. Devenyi, S. Weygandt, etal. An hourly assimilation-forecast cycle:The RUC. Mon. Wea. Rev.,2004,132,495-518.
[34] Benjamin, S.G., G. Grell, J.M. Brown, T.G Smirnova etal.. Mesoscale weather predictionwith the RUC hybrid isentropic-terrain-following cordinate model. Mon. Wea. Rev.,2004,132,473-494.
[35] Benjamin, S.G., D. Devenyi, T. Smirnova, etal.. From the13-km RUC to the RapidRefresh.12th Conf. on Aviation, Range, and Aerospace Meteorology (ARAM), Atlanta,GA, Amer. Meteor. Soc.2006, CD-ROM,9.1.
[36] Benjamin, S. G., S. Weygandt, J. M. Brown, etal.. From the radar-enhanced RUC to theWRF-based Rapid Refresh.22nd Conf. Wea.Anal.and Fcstg./18th Conf. Numer. Wea.Pred., Park City, Amer. Meteor.Soc.,2007, Paper J3.4.
[37] Benjamin S. G., S. S. Weygandt, C. R. Alexander, etal.. NOAA's hourly-updated3kmHRRR and RUC/Rapid Refresh-recent (2010) and upcoming changes toward improvingweather guidance for air-traffic management.2011.
[38] Benjamin, S. G., S. S. Weygandt, J. M. Brown, etal.. Assimilation of METAR cloudandvisibility observations in the RUC.11thConference on Aviation, Range, Aerospaceand22nd Conference on Severe Local Storms, Hyannis, MA, Amer. Meteor. Soc.2004,Paper9.13.
[39] Weygandt, S., S. G. Benjamin, D. Dévényi, etal.. Cloud and hydrometeor analysis usingmetar, radar, and satellite data within the RUC/Rapid-Refresh model.12th Conference onAviation Rang and Aerospace Meteorology, Atlanta, GA.2006.
[40] Brown, J.M., S. G. Benjamin, T. Smirnova, etal.. Rapid-Refresh Core Test: aspects ofWRFNMM and WRF-ARW forecast performance relevant to the Rapid-Refreshapplication.18thConf. Num. Wea.Pred., Park City, UT, Amer. Meteor.Soc.,2007,4A.1.
[41] Hu, M., S. Weygandt, S. G. Benjamin, M. Xue. Ongoing development and testing ofgeneralized cloud analysis package within GSI for initializing Rapid Refresh, Preprints,13thConf. on Aviation, Range and Aerospace Meteorology. New Orleans, LA, Amer.Meteor. Soc.,2008, Paper7.4.
[42] Brown, J. M., T. G. Smirnova, S. G. Benjamin, etal. Rapid-Refresh testing: example offorecast performance.13th Conference on Aviation Rang and Aerospace Meteorology,New Orleans, LA, Amer. Meteor. Soc.,2008, Paper7.1.
[43] Bratseth, A. M. Statistical interpolation by means of successive corrections. Tellus,1986,38A,439-447.
[44] Sashegyi, K.D., D.E. Harms, R.V. Madala, S. Raman. Application of the Bratseth schemefor the analysis of GALE data using a mesoscale model. Mon. Wea. Rev.,1993,121,2331-2350.
[45] Cole, R.E. and F. W. Wilson. ITWS gridded wind product, Preprints,6th Conf. onAviation Wetaher Systems, AMS, Boston.1995
[46] Zhang J. Moisture and diabatic initialization based on radar and satelliteobservations, Ph.D. dissertation, University of Oklahoma, School of Meteorology,1999,194pp
[47] Brewster, K.. Recent advances in the diabatic initialization of a non-hydrostaticnumerical model. Preprints,21st Conf. on Severe Local Storms, and Preprints,15th Conf.Num. Wea.2002.
[48] Hu, M..3DVAR and cloud analysis with WSR-88D level-II data for the prediction oftornadic thunderstorms, Ph. D. Dissertation, School of Meteorology, University ofOklahoma,2005,217pp.
[49]杨艳蓉.面向雷电预报的雷达、卫星遥感资料同化及其应用初步研究，南京信息工程大学博士学位论文.2008.
[50] Zhang, J., f. Carr, K. Brewster. ADAS cloud analysis, Preprints,12th conference onNumerical Weather Prediction, Boston,1998,185-188
[51] Haines P. A., J. K. Luers, C. A Cerbus. The role of the Smith-Feddes model in improvingthe forecasting of aircraft icing. Preprints, Third Conf on Aviation Weather Systems,Anaheim, Cal, Amer Meteor Soc,1989,258-263.
[52] Benjamin S. G., Weygandt S S, etal. An hourly assimilation-forecast cycle: The RUC.Mon Wea Rev,2004,132(2):495-518
[53] Lynch. P. and X.-Y. Huang. Initialization of the HIRLAM model using a digital filter.Mon. Wea. Rev.1992,120,1019-1034
[54] WRF user guide V3.3available from website:http://www.mmm.ucar.edu/wrf/users/pub-doc.html
[55]吴浩云.太湖流域典型年梅雨洪涝灾害比较分析[J].水文，2000，20（4）：54-57.
[56]陈艺敏，钱永甫.116a长江中下游梅雨的气候特征[J].南京气象学院学报，2004,27(1):65-72.
[57]杨义文.长江中下游梅雨与中国夏季旱涝分布[J].气象，2002,28(11):11-15.
[58]张运林，罗潋葱.江苏省近47年来的梅雨特征及灾害影响[J].灾害学，2003,18(2):58-62.
[59]田心如，姜爱军，高苹，等.江苏省梅雨及其灾害影响评估分析[J].灾害学，2004,19(1):40-45.
[60]李峰.东亚季风和中国梅雨暴雨研究的评述[J].成都气象学院学报，2000，15(1):72-77.
[61]张小玲，陶诗言，张顺利.梅雨锋上的三类暴雨[J].大气科学，2004,28(2):187-205.
[62]孙彭龄.浙江的梅雨和台风年际变化[J].浙江气象科技，1993，20(3):18-21.
[63]张艳，鲍文杰，余琦等．超大城市热岛效应的季节变化特征及其年际差异[J]．地球物理学报，2012，55(4),1121-1128.
[64]陶诗言.中国之暴雨.北京：科学出版社，1980:8-11.
[65] Wilson, James W., Wendy E. Schreiber. Initiation of convective storms at radar-observedboundary-layer convergence lines [J]. Mon. Wea. Rev.,1986,114,2516–2536.
[66] Wilson, James W., G. Brant Foote, N. Andrew Crook, James C. Fankhauser, Charles G.Wade, John D. Tuttle, Cynthia K. Mueller, Steven K. Krueger. The role of boundary-layerconvergence zones and horizontal rolls in the initiation of thunderstorms: A Case Study [J].Mon. Wea. Rev.,1992，120,1785–1815.
[67] Wilson, James W., Cynthia K. Mueller. Nowcasts of thunderstorm initiation andevolution [J]. Wea. Forecasting,1993,8,113–131.
[68] Roger A. Pielke. A Three-Dimensional Numerical Model of the Sea Breezes Over SouthFlorida. Monthly Weather Review，1974，Vol.102, No.2, pp.115–139.
[69]杨薇，苗俊峰，谭哲敏.太湖地区湖陆风对雷暴过程影响的数值模拟.第九届长三角气象科技论坛论文集.
[70]王谦谦，钱永甫.太阳辐射日变化对夏季风模拟特征的影响.气象学报，1997,55（3）：334-345.
[71] Hu M. and Xue M.. Initializing convection using cloud analysis and radar data ingrid-point statistical interpolation(GSI) system and impact on the forecast of advancedresearch WRF Accepted. Geophys. Res. Letters,2007b..
[72] Xue M., Wang D., Hou D., etal. Prediction of the7May1995squall line over the centralU S with intermittent data assimilation12th conf on numerical weather prediction.Phoenix AZ, Amer Meteor Soc,1998，191-194.
[73] Haase G., Grewell S., Simmer C, etal. Assimilation of radar data in mesoscale models:Physical initialization and latent heat nudging. Phys Chem Earth (B),2003,25:1237-1242.
[74] Hu M., Xue M., Brewster K.3DVAR and cloud analysis with WSR-88D Level-Ⅱ Datafor the prediction of the Fort Worth, Texas, Tornadic Thunderstorms. Part Ⅰ: CloudAnalysis and Its Impact. Mon. Wea. Rev.,2006,134:675-698.
[75] Hu M, Xue M, Gao J., etal.3DVAR and cloud analysis with WSR-88D Level-Ⅱ Data forthe prediction of the Fort Worth, Texas, Tornadic Thunderstorms. Part Ⅱ: Impact ofRadial Velocity Analysis via3DVAR. Mon. Wea. Rev.,2006,134:699-721.
[76]张利红，蒋丽娟，陈朝平等.探空观测资料在西南暴雨中的同化试验.2009，高原山地气象研究，29(3):31-38.
[77]舒守娟，王元，宋金杰.西北太平洋台风“海棠”结构的GPS下投式探空仪观测分析.2011，气象学报，69(6):933-944.
[78]黄燕燕，万齐林，陈子通等.加密探空资料在华南暴雨数值预报的应用试验.2011，热带气象学报，27(2):179-188.
[79]梁科，万齐林，丁伟钰等.飞机报资料在0506华南致灾暴雨过程模拟中的应用.2007，热带气象学报，23(4):313-325.
[80]黄卓，李延香，王慧等. AMDAR资料在天气预报中的应用.2006，气象，32（9）：42-48.
[81]邵长亮. WRF-EnSRF同化地面自动站资料的实现与技术研究.2011，南京信息工程大学硕士学位论文.
[82]陈新梅、段旭、王曼等.使用于云南地区的地面资料同化方案设计及个例分析.高原气象，2010，29（4）：907-917.
[83]良敏.自动站资料质量控制及其三维变分同化研究.南京信息工程大学硕士学位论文.
[84] Zhang D L, J M Fritcsh. A case study of the sensitivity of numerical model simulation ofmesoscale convective systems to varying initial condition. Mon. Wea. Rev.,1986,114:2481-2431.
[85]吴海英，曾明剑，尹东屏等.一次苏皖特大暴雨过程中边界层急流结构演变特征和作用分析.高原气象，2010，29(6)：1431-1440.
[86]何立富，陈涛，周庆亮，等.北京“7.10”暴雨β中尺度对流系统分析.应用气象学报，2007，18(5):655-665.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |