The Coordinat...g Late Summer_Yitong LIN.pdf

The Coordinated Influence of Indian Ocean Sea Surface Temperatureand Arctic Sea Ice on Anomalous Northeast China Cold VortexActivities with Different Paths during Late SummerYitong LIN1,2,Yihe FANG*1,2,Chunyu ZHAO1,2,Zhiqiang GONG3,4,Siqi YANG5,and Yiqiu YU1,21Regional Climate Center of Shenyang,Liaoning Provincial Meteorological Administration,Shenyang 110016,China2Key Opening Laboratory for Northeast China Cold Vortex Research,China Meteorological Administration,Shenyang 110016,China3College of Physics and Electronic Engineering,Changshu Institute of Technology,Changshu 215500,China4Laboratory for Climate Studies,National Climate Research Center,China Meteorological Administration,Beijing 100081,China5Inner Mongolia Climate Center,Hohhot 010051,China(Received 27 December 2021;revised 6 April 2022;accepted 11 May 2022)ABSTRACTThe Northeast China cold vortex(NCCV)during late summer(from July to August)is identified and classified intothree types in terms of its movement path using machine learning.The relationships of the three types of NCCV intensitywith atmospheric circulations in late summer,the sea surface temperature(SST),and Arctic sea ice concentration(SIC)inthe preceding months,are analyzed.The sensitivity tests by the Community Atmosphere Model version 5.3(CAM5.3)areused to verify the statistical results.The results show that the coordination pattern of East Asia-Pacific(EAP)and LakeBaikal high pressure forced by SST anomalies in the North Indian Ocean dipole mode(NIOD)during the preceding Apriland SIC anomalies in the Nansen Basin during the preceding June results in an intensity anomaly for the first type ofNCCV.While the pattern of high pressure over the Urals and Okhotsk Sea and low pressure over Lake Baikal during latesummerwhich is forced by SST anomalies in the South Indian Ocean dipole mode(SIOD)in the preceding June and SICanomalies in the Barents Sea in the preceding Aprilcauses the intensity anomaly of the second type.The third type isatypical and is not analyzed in detail.Sensitivity tests,jointly forced by the SST and SIC in the preceding period,can wellreproduce the observations.In contrast,the results forced separately by the SST and SIC are poor,indicating that theNCCV during late summer is likely influenced by the coordinated effects of both SST and SIC in the preceding months.Key words:machine learning method,Northeast China cold vortex,path classification,Indian Ocean sea surfacetemperature,Arctic sea ice,model sensitivity testCitation:Lin,Y.T.,Y.H.Fang,C.Y.Zhao,Z.Q.Gong,S.Q.Yang,and Y.Q.Yu,2023:The coordinated influence ofIndian Ocean sea surface temperature and Arctic sea ice on anomalous Northeast China cold vortex activities with differentpaths during late summer.Adv.Atmos.Sci.,40(1),6277,https:/doi.org/10.1007/s00376-022-1415-9.Article Highlights:The NCCV during late summer is identified and categorized into three types based on their movement paths by amachine learning method.The NCCV is related to the coordinated influence of Indian Ocean sea surface temperature and Arctic sea ice in thepreceding months.1.IntroductionAs an important part of the atmospheric circulations inthe middle-and high-latitudes of the Northern Hemisphere,the Northeast China cold vortex(NCCV)has an importantinfluence on the persistence and adjustment of atmosphericcirculation anomalies in Northeast China(NEC)and in EastAsia.This pattern can occur all year round,with the highestfrequency of occurrence in the spring and summer(Sun et*Corresponding author:Yihe FANGEmail:ADVANCES IN ATMOSPHERIC SCIENCES,VOL.40,JANUARY 2023,6277 Original Paper Institute of Atmospheric Physics/Chinese Academy of Sciences,and Science Press and Springer-Verlag GmbH Germany,part of Springer Nature 2023al.,2000;Huang and Li,2020).The frequent NCCV activitiesare the primary cause of the meteorological disasters in theNEC(i.e.,rainstorms,short-term heavy precipitation,andlow-temperature damage),which have significant impactson agricultural production,transportation,and peoples lives(Sun et al.,2002).Therefore,it is very important to diagnoseand accurately predict the NCCV anomaly.Previous research mainly included the objective identifi-cation and climatological characteristics of the NCCV andthe causes of the NCCV anomaly.In terms of climatologicalcharacteristics,many statistical analyses have been conductedon the occurrence frequency,duration,change trend,period,and spatial distribution of the cold vortex in previous studies(Sun,1997;Liang et al.,2009;Hu et al.,2011;Xie andBueh,2012;Liu et al.,2015).As for the cause of the NCCVanomaly,some studies have been performed from the perspec-tive of atmospheric circulation factors(Liu et al.,2002;Heet al.,2006;Miao et al.,2006;Xie and Bueh,2012).Lian etal.(2010)analyzed the relationship between NCCV activityin early summer and the atmospheric circulations in the North-ern Hemisphere and pointed out that anomalous NCCV activ-ity in the early summer is related to the persistent circulationanomalies near the Ural Mountains and Japan.Xie andBueh(2017)revealed the dynamic features of the coordinatedinfluence between the NCCV and the blocking high on the cli-mate in NEC and its relation with the climatic circulationbackground on intra-seasonal time scales.Liu et al.(2016)and Yang et al.(2015a)explored the close relationshipbetween the NCCV and the Okhotsk high and its responseto early signals of El Nio-Southern Oscillation(ENSO).Inaddition,the NCCV anomaly is also related to the westernPacific subtropical high and the upper-level jet(Parker et al.,1989;Sun,1997;Sun et al.,2000).Moreover,it is also influ-enced by sea surface temperature(SST)and land surface ther-mal anomalies (Bueh and Xie,2013;Chen et al.,2018;Wang et al.,2018).Fang et al.(2018)conducted diagnosticanalyses and numerical simulations on the precipitationanomalies in NEC in early summer.They point out that theNorth Atlantic tripole(NAT)SST and the SST anomalies inthe Kuroshio area induce circulation anomalies,such as theanomalous blocking high located upstream and downstreamof the NCCV through the air-sea interaction,further affectingthe cold vortex precipitation in NEC in early summer.Lu etal.(2020)analyzed spring precipitation in NEC andrevealed the physical mechanisms related to the interannualvariabilities of the tropical North Atlantic and NorthwestPacific SST affecting the low-pressure system over theNEC in spring.Fang et al.(2020)analyzed the north-southposition of the NCCV in early summer and pointed out thatthe SST in the northwestern Pacific and the southeasternPacific in the preceding spring are the dominant factors caus-ing more vortexes in the north,while the NAT SST in the pre-ceding period is responsible for more vortexes in the south.Most existing research mainly concentrates on theNCCV in early summer,while studies on the NCCV anoma-lies in late summer are quite rare.Previous studies have sug-gested that the summer monsoon is the dominant factor forthe late summer precipitation in NEC,accounting for 70%80%of the summer total(Shen et al.,2011).Current studieshave revealed that the late summer precipitation is alsoaffected by the circulation systems in middle and high lati-tudes(Shen et al.,2011)and that the influence of circulationsystems such as the NCCV and blocking high on the late sum-mer precipitation in NEC was enhanced after the mid-1980s(Lin et al.,2021).Sun et al.(2002)conducted a diagnosticanalysis of nine typical heavy rainfall cases over the Songhua-jiang-Nenjiang River Basin(SNRB)in the late summer of1998 and found that the NCCV activity was abnormally fre-quent during that period.In addition,the configuration ofthe NCCV,the blocking high over East Asia and the WestPacific subtropical high,both in intensity and position,pro-vide a favorable large-scale circulation background for persis-tent heavy rainstorms over the SNRB.Gao and Gao(2018)pointed out that the NCCV activity played a major role inthe anomalously abundant precipitation over NEC duringthe 2013 summer.It can be seen that the diagnosis and the pre-diction of the NCCV anomaly in late summer are particularlyimportant.The anomalous position of the NCCV couldaffect the temperature and precipitation in different regionsof NEC (Fang et al.,2021).To meet the demand for anincreasingly precise prediction of the NCCV,it is necessaryto perform an objective classification for the NCCV in latesummer in terms of its path and carry out research on thecauses of various cold vortex anomalies.In this study,theNCCV is objectively identified based on the geopotentialheight and temperature data with a high spatiotemporal resolu-tion(Fang et al.,2018),and the machine learning method ofK-means is used to objectively classify the paths of theNCCV(Wang et al.,2018;Fang et al.,2021).The relation-ships of the NCCV along different paths in late summerwith the atmospheric circulations during the same period,aswell as the SST and Arctic sea ice concentration(SIC)inthe preceding period,are investigated by using statistical anal-ysis methods,and the possible mechanisms are alsoexplored and discussed.Moreover,sensitivity tests aredesigned based on Community Atmosphere Model version5.3(CAM5.3)to verify the conclusions obtained by statisticalanalysis methods.The remainder of this paper is organized as follows.Sec-tion 2 describes the data and methods used in this study.Sec-tion 3 presents the characteristics of the trajectories forNCCV.Section 4 presents the atmospheric circulations inlate summer corresponding to various NCCV.Section 5 dis-cusses the relationships of the NCCV in late summer withthe precedent SST and SIC,and section 6 gives the resultsof numerical model sensitivity test.Finally,the findings ofthe present study are summarized and discussed in section 7.2.Data and methods2.1.DataThe research period in this study covers the late summersJANUARY 2023LIN ET AL.63(from July to August)from 1979 to 2018.The data includedthe geopotential height and temperature data with spatiotem-poral resolutions of 6 h and 1 1 from the European Centrefor Medium-Range Weather Forecasts Re-Analysis Interim(ERA-Interim,Dee et al.,2011),the 500-hPa geopotentialheight with a resolution of 1 1 in the monthly reanalysisdata came from ERA-Interim,the monthly SST data,andthe monthly sea ice area fraction data with a horizontal resolu-tion of 1 1 was provided by the Hadley Centre(Rayneret al.,2003).2.2.MethodsIn this study,the NCCV during mid-summer is objec-tively identified using the ERA-Interim geopotential heightand temperature data from 1979 to 2018.See details of theidentification method in Fang et al.(2021).2.2.1.Calculation Method of the Indexes of the NCCVThe NCCV index is defined as the absolute value of thesum of the geopotential height anomaly of the lowest-valuegrid point within the inner equipotential height line of thecold vortexes in all the NCCV occurrence days in late summerfrom 1979 to 2018.In this way,the adopted index calculationmethod was not only able to reflect the number of cold vortexdays but also represented the intensities of cold vortex cen-ters.2.2.2.NCCV Movement Path Classification based on theK-means MethodThe objective classification method of the NCCV isgiven in the three steps below.First,characteristic parameters were collected.Accord-ing to the identification method of the NCCV,when weread the geopotential height and temperature data for eachtime into the script,the first time that meets the identificationconditions is printed as the starting time of the cold vortexprocess,and the last time that meets the conditions is the end-ing time.The center position is defined as the grid point ofthe lowest geopotential height value within the inner equipo-tential height line of the cold vortex,while the generationand extinction positions are defined as the center position atthe starting and ending time,respectively.Based on the objec-tive identification results of the NCCV,the characteristicparameters for the NCCV processes in late summer are col-lected,including the latitudes and longitudes of the centerpositions from the starting times to the ending times.Second,the trajectory description parameters were calcu-lated.The generation position was collected to represent thesource information of the cold vortex.Then,the meridionaland zonal mean from the starting to ending times are calcu-lated to describe the location of the cold vortex process.Themeridional,zonal,and diagonal variances from startingtimes to ending times are calculated to describe the movingcurvature and distance(e.g.,a large diagonal variance corre-sponds to a straighter motion trajectory and a longer movingdistance;see Zheng et al.,2013).The center positions of thecold vortex along its trajectory within the geographicalscope of NEC are also collected to represent the relationshipbetween its trajectory and the geographical location of NEC.The above parameters are preliminarily used to describe thetrajectory of the NCCV.Third,the cluster number is determined.The K-meansclustering method is applied to the objective classificationof the trajectories of cold vortexes.Before this,the value ofthe classification number,K,should be determined.Basedon the dataset of trajectory describing parameters obtainedin the previous step,two to seven initial classification num-bers are preliminarily determined,and the shape coefficientscorresponding to these classification numbers are calculated.The closer the shape coefficient is to 1,the more reasonablethe classification number is,and thus an optimal classificationnumber is determined.In this study,a classification numberof three is selected.Thus,the NCCV is classified into threetypes in terms of trajectories.2.2.3.Definition Indexes Used in This StudyThe SST or SIC indices of the key areas were definedaccording to the regions where the correlation coefficientbetween the NCCV indexes in late summer and the SSTs orSICs in earlier months passed the 90%confidence level(see section 5).The specific definitions are given inTable 1.2.2.4.Numerical modelThe CAM5.3,as part of the Community Earth SystemModel version 1.2.2(CESM1.2.2)modeling framework bythe National Center for Atmospheric Research(NCAR),isused in this study to further verify the physical linkagesbetween the previous SST (or SIC)and the late summer500-hPa geopotential height field.This model has been usedin previous studies of climate over East Asia(Huang et al.,2019;Zhang et al.,2020).The finite-volume dynamicalcore configured with a horizontal resolution of 1.9 latitude 2.5 longitude (f19_f19)and 30 vertical hybrid levels isselected.More details about this model can be found inNeale et al.(2012).Nine sets of numerical simulation experiments are con-ducted to examine the correlations of the SST and SIC inthe preceding period with the 500-hPa geopotential heightfield in late summer.In the control experiment,the monthlyclimatological data is used as the initial field,while themodel integration starts on 1 January of th