简介:TheensemblebasedforecastsensitivitytoobservationmethodbyLiuandKalnayisappliedtotheSPEEDY-LETKFsystemtoestimatetheobservationimpactofthreetypesofsimulatedobservations.Theestimationresultsshowthatalltypesofobservationshavepositiveimpactonshort-rangeforecast.ThelargestimpactinNorthernHemisphereisproducedbyrawinsondes,followedbysatelliteretrievedprofilesandclouddriftwinddata,whichinSouthernHemisphereisproducedbysatelliteretrievedprofiles,rawinsondesandclouddriftwinddata.SatelliteretrievedprofilesinfluencemoreontheSouthernHemispherethanontheNorthernHemisphereduetofewobservationsfromrawinsondesintheSouthernHemisphere.Atthelevelof200to300hPa,thelargestimpactisattributedtowindobservationsfromrawinsondesandclouddriftwind.
简介:PreliminaryresultsofthewindvelocityestimationusingtheMaximumEntropyMethod(MEM)toMUradarobservationdatasetsarepresented.ThecomparisonoftheresultsfromtheperiodogrammethodandtheMEMshowsthattheMEMestimationisreliable,andhashigheraccuracy,resolutionanddetectabilitythantheestimationfromperiodogrammethod.ThehighaccuracypowerspectrumobtainedbytheMEMisveryusefultostudyingtheatmosphericturbulencestructure.However.theMEMneedsthelongercomputingtimeforobtainingthehighaccuracyspectrum.Particularly,theestimationofMEMwillbringseriousdevia-tionatlowersignal-to-noiseratio.
简介:Thispaperdescribesandteststwomodelsforestimatingnetradiation(ortheradiationbalance)onslopingsurfacesofalpineenvironments.Theyareanempiricalmethodbasedonthelinearrelationshipbetweennetradiationandglobalsolarradiationandaflux-by-fluxmethodinvolvingtheestimationofalltheindividualcomponentsofradiationbudgetindependently.Theresultsshowthattheempiricalmethodiscapableofpredictinghourlynetradiationonslopingsur-facestowithinabout±53Wm-2underallskyconditions.Duringclearskyconditions,itcouldpredictnetradiationonslopestowithin±58Wm-2or16%ofthemeasuredvalues.Theflux-by-fluxmethod,althoughitdidnotperformaswellastheempiricalmethod,performedadequatelyandcouldgiveestimatesofnetradiationonslopeswithrootmeansquareerroroflessthan74Wm-2(20%)andameanbiaserrorof27Wm-2(7%).
简介:在最近的年里,全球航行卫星系统反射计(GNSS-R)被开发作为一个新遥感工具估计土壤潮湿内容(SMC)。全球放的系统(GPS)bistatic雷达的信号错误是影响SMC评价的精确性的一个重要因素。在这份报纸,直接、反映的信号被介绍的包含两个的GPS信号刻度的二个方法,和如此的方法的理论基础的详细解释被给。利用校准的GPSL乐队信号的一个改进SMC评价模型被建议,并且评价精确性在2002从土壤潮湿实验用在空中的GPS数据被验证(SMEX02)。我们为确认在US的核桃溪区域与大豆和玉米选择21个地点。这些地点根据他们的植被盖子被划分成三个范畴:赤裸的土壤,中间植被的盖子(中间蔬菜),并且高植被的盖子(高蔬菜)。SMC评价的精确性为赤裸的土壤是11.17%并且8.12%为中间蔬菜的地点,比传统的模型的好一些。为高蔬菜的地点,信号变细的效果preliminarily由于植被盖子被考虑,与规范的差别植被索引(NDVI)有关的一个线性模特儿被收养为修正在刻度上,和错误获得一个因素因为高蔬菜的地点最后被归结为3.81%。
简介:ESTIMATIONOFVERTICALWINDFIELDFROMSINGLE-DOPPLERRADARRHIOBSERVATIONSPengHong(彭红)andGeRunsheng(葛润生)InstituteofMesoscaleMeteorol...
简介:Drylandsareamongthoseregionsmostsensitivetoclimateandenvironmentalchangesandhuman-inducedperturbations.Themostwidelyaccepteddefinitionofthetermdrylandisaratio,calledtheSurfaceWetnessIndex(SWI),ofannualprecipitationtopotentialevapotranspiration(PET)beingbelow0.65.PETiscommonlyestimatedusingtheThornthwaite(PETTh)andPenman–Monteithequations(PETPM).ThepresentstudycomparedspatiotemporalcharacteristicsofglobaldrylandsbasedontheSWIwithPETThandPETPM.ResultsshowedvastdifferencesbetweenPETThandPETPM;however,theSWIderivedfromthetwokindsofPETshowedbroadlysimilarcharacteristicsintheinterdecadalvariabilityofglobalandcontinentaldrylands,exceptinNorthAmerica,withhighcorrelationcoefficientsrangingfrom0.58to0.89.Itwasfoundthat,during1901–2014,globalhyper-aridandsemi-aridregionsexpanded,aridanddrysub-humidregionscontracted,anddrylandsunderwentinterdecadalfluctuation.Thiswasbecauseprecipitationvariationsmademajorcontributions,whereasPETchangescontributedtoamuchlesserdegree.However,distinctdifferencesintheinterdecadalvariabilityofsemi-aridanddrysub-humidregionswerefound.ThisindicatedthattheinfluenceofPETchangeswascomparabletothatofprecipitationvariationsintheglobaldry–wettransitionzone.Additionally,thecontributionofPETchangestothevariationsinglobalandcontinentaldrylandsgraduallyenhancedwithglobalwarming,andtheThornthwaitemethodwasfoundtobeincreasinglylessapplicableunderclimatechange.
简介:从卫星导出的大气的不稳定性信息在预报的短期的天气起一个重要作用,特别预报严重对流暴风雨。为为朝鲜多种用途的地球同步的卫星节目的天气卫星的下一代,一台新成像仪器被开发了。尽管这台成像仪器没被设计执行完整的发出声音的使命,它的能力是有限的,它的多光谱的红外线的隧道在垂直发出声音上提供信息。为了从大部分充分利用观察数据,改进了成像器的空间与时间的分辨率,为大气的不稳定性的推导的一条人工的神经网络途径的可行性被调查。有前馈控制和背繁殖训练算法的多层的视感控器模型显示出相当敏感回答到训练数据集和模型建筑学的选择。通过有7197独立侧面的小心地选择的训练数据集的广泛的性能测试,模型体系结构被选择分别地为隐藏的节点,时代的数字,和学习的率的数字是12,5000,和0.3。选择模型给330Jkg的一个吝啬的绝对错误,RMSE,和关联系数-1,420Jkg-1,并且0.9分别地。可行性进一步从一台类似的仪器经由模型的申请被表明到真实观察数据与计划成像器有可比较的观察隧道。