简介：

简介：Anoptimizationmethodisbasedtodesignasnowfallestimatemethodbyradarforoperationalsnowwarning,anderrorestimationisanalyzedthroughacaseofheavysnowonMarch4,2007.Threemodifiedschemesaredevelopedforerrorscausedbytemperaturechanges,snowflaketerminalvelocity,thedistancefromtheradarandcalculationmethods.Duetotheimprovements,thecorrelationcoefficientbetweentheestimatedsnowfallandtheobservationis0.66(exceedingthe99%confidencelevel),theaveragerelativeerrorisreducedto48.74%,andthemethodisabletoestimateweaksnowfallof0.3mm/handheavysnowfallabove5mm/h.Thecorrelationcoefficientis0.82betweentheestimatedsnowfallfromthestations50to100kmfromtheradarandtheobservation.Theimprovedeffectisweakwhentheinfluenceofthesnowflaketerminalvelocityisconsideredinthosethreeimprovementprograms,whichmayberelatedtotheuniformecho.Theradarestimateofsnow,whichisclassifiedbythedistancebetweenthesampleandtheradar,hasthemostobviouseffect:itcannotonlyincreasethedegreeofsimilarity,butalsoreducetheoverestimateandtheundervaluationoftheerrorcausedbythedistancebetweenthesampleandtheradar.Theimprovedalgorithmfurtherimprovestheaccuracyoftheestimate.Theaveragerelativeerrorsare31%and27%fortheheavysnowfallof1.6to2.5mm/handabove2.6mm/h,respectively,buttheradaroverestimatesthesnowfallunder1.5mm/handunderestimatesthesnowfallabove2.6mm/h.Radarechomaynotbesensitivetotheintensityofsnowfall,andtheconsistencyshownbytheerrorcanbeexploitedtoreviseandimprovetheestimationaccuracyofsnowforecastintheoperationalwork.

简介：Athree-dimensional(3D)globaladiabaticspectralprimitiveequationmodelhasbeendesigned.Themainfeaturesareasfollows.(1)AdoptionofsphericalharmonicsandTschebyscheffpolynomialsasthebasisfunctionsinthehorizontalandverticalrespectively,buttheunknownsinthespectralequationsaretwo-dimensional;(2)Inclusionofthetropopause,whichmayvarywithtimeandspace;(3)Suggestionofaspectralmethodforrepresentingtheverticalstructureoftheatmosphereapplicabletotheunsmoothedprofilecase;(4)Inconsiderationofnonlinearverticalaliasingatechniqueisproposedtoavoiditandnonlinearcomputationalinstability.Basedonrealdataforecastsupto48hourshavebeenperformed.TheresultsshowthatthestatisticalverificationswiththemodelaresuperiorontheaveragetothosewiththeT42L9usedoperationallybefore1995atNMCofChinaatthesamemeanresolution.

简介：宇如聪，金向泽，张学洪Ｄｅｓｉｇｎ　ａｎｄ　Ｎｕｍｅｒｉｃａｌ　Ｓｉｍｕｌａｔｉｏｎ　ｏｆ　ａｎ　Ａｒｃｔｉｃ　Ｏｃｅａｎ　Ｃｉｒｃｕｌａｔｉｏｎ　ａｎｄ　Ｔｈｅｒｍｏｄｙｎａｍｉｃ　Ｓｅａ－Ｉｃｅ　ＭｏｄｅｌＤｅｓｉｇｎａｎｄＮｕｍｅｒｉｃａｌＳｉ...

简介：Sometypicalsamplesareusedtoexplorethequantitativecorrelationwiththeirfeaturesbetweenaconvectivecloudanditsrainfallfield,withwhichtodeveloptwomorphologicalfunctionsforthecorrelationandbysinglingouttheirmostsuitablegroupsofparametersweproposeamodelforquantitativelyestimatingprecipitationinthecontexto{thein-advancerecognitionofmeso-αconvectivesystempropertiesanditsprecipitatingcenter.Fromthemodelfittingprecisionandforecastingaccuracywefindthatitisfeasibletoutilizegeostationarymeteorologicalsatellite(GMS)digitalizedimageryforestimatingshort-termrainfallinaquantitativemanner.Also,evidencesuggeststhatthemodelissupposedtoberestrictedinitsapplicabilityduetothefactthattheemployedsamplesarefromrathertypicalrainfalleventsthatarelarge-scale,slow-movingandhavewell-definedgenesisanddissipativestages.