Re-examination of Tropical Cyclone Formation in Monsoon Troughs over the Western North Pacif i c

ZONG Huijunand WU Liguang∗

1Pacif i c Typhoon Research Center and Key Laboratory of Meteorological Disaster of Ministry of Education, Nanjing University of Information Science and Technology,Nanjing 210044

2State Key Laboratory of Severe Weather,Chinese Academy of Meteorological Sciences,Beijing 100081

Re-examination of Tropical Cyclone Formation in Monsoon Troughs over the Western North Pacif i c

ZONG Huijun1,2and WU Liguang∗1,2

1Pacif i c Typhoon Research Center and Key Laboratory of Meteorological Disaster of Ministry of Education, Nanjing University of Information Science and Technology,Nanjing 210044

2State Key Laboratory of Severe Weather,Chinese Academy of Meteorological Sciences,Beijing 100081

The monsoon trough(MT)is one of the large-scale patterns favorable for tropical cyclone(TC)formation over the western North Pacif i c(WNP).This study re-examines TC formation by treating the MT as a large-scale background for TC activity during May–October.Over an 11-year(2000–10)period,8.3 TC formation events on average per year are identif i ed to occur within MTs,accounting for 43.1%of the total TC formation events in the WNP basin.This percentage is much lower than those reported in previous studies.Further analysis indicates that TC formation events in monsoon gyres were included at least in some previous studies.The MT includes a monsoon conf l uence zone where westerlies meet easterlies and a monsoon shear line where the trade easterlies lie north of the monsoon westerlies.In this study,the large-scale f l ow pattern associated with TC formation in the MT is composited based on the reference point in the conf l uence zone where both the zonal and meridional wind components are zero with positive vorticity.While previous studies have found that many TCs form in the conf l uence zone,the composite analysis indicates that nearly all of the TCs formed in the shear region,since the shear region is associated with stronger low-level relative vorticity than the conf l uence zone.The prevailing easterly vertical shear of zonal wind and barotropic instability may also be conducive to TC formation in the shear region,through the development of synoptic-scale tropical disturbances in the MT that are necessary for TC formation.

monsoon trough,tropical cyclone formation,summer monsoon,vertical wind shear

1.Introduction

The summertime monsoon trough(MT)over the western North Pacif i c(WNP)is a convergence zone in the lower troposphere,where westerly monsoon winds are found in its equatorward region,while easterly trade winds exist poleward of the trough.A number of studies have shown that the MT is an important large-scale circulation pattern associated with tropical cyclone(TC)formation in the WNP basin [e.g.,Gray,1968;Ramage,1974;Holland,1995;Lander, 1996;Briegel and Frank,1997;Ritchie and Holland,1999; see Li(2012)for a review].The MT can provide both dynamical and thermodynamical large-scale conditions that are favorable for TC formation,including strong low-level cyclonic relative vorticity,high mid-level relative humidity,and weak vertical wind shear(Sadler,1967;Gray,1968;Ramage,1974;Gray,1975;Frank,1987;Chia and Ropelewski, 2002;Harr and Chan,2005;Chen et al.,2006).Although many studies have documented the percentages of TCs that form in the MT(e.g.,Frank,1987;McBride,1995;Briegel and Frank,1997;Ritchie and Holland,1999),Molinari and Vollaro(2013)noticed that the results of these observational studies were examined mainly in a subjective manner,even without a common def i nition of the MT.As mentioned in Molinari and Vollaro(2013),for example,the MT was subjectively def i ned in Briegel and Frank(1997)and Ritchie and Holland(1999)and details about how a TC was considered to be associated with the MT were not mentioned.

As an alternative,Wu et al.(2012)and Molinari and Vollaro(2013)recently def i ned the MT as a geographical region in terms of monthly mean 850 hPa relative vorticity.Wu et al.(2012)used the mean relative vorticity averaged over 5°–20°N to def i ne the zonal variation of the MT intensity,and it was not allowed to exist beyond the specif i ed latitude range. In Molinari and Vollaro(2013),the MT was def i ned for each month as the contiguous region with positive monthly mean 850 hPa relative vorticity.Considering signif i cant movement and structural changes of the MT on multiple time scales (Harr and Wu,2011),uncertainty still exists in Wu et al. (2012)and Molinari and Vollaro(2013).That is,some formationevents that were not associated with the MT may havebeen counted,while some formation events that formed beyond the specif i ed latitude ranges may have been missed.In addition,the positiverelativevorticity ofthe MT in theirdef initionalso includesthe contributionfromothersystems,such as monsoongyres.In other words,TCs that formin monsoon gyres were included in Wu et al.(2012)and Molinari and Vollaro(2013).Wu et al.(2013)showed that 19.8%of all May–October TCs over the WNP formed in monsoon gyres over the period 2000–10.

Although the MT is a large-scale summertime f l ow pattern over the WNP,it includes TCs and wave trains on the synoptic timescale that can alter convection,wind,and vorticity,which leads to changes in the structure and position of the MT(Holland,1995;Lander,1996;Harr and Wu,2011). For example,northwestward-propagating wave trains have a typical wavelengthof 2500–3000km anda timescale of 6–10 days,consisting of alternating regions of cyclonic and anticyclonic circulation(Lau and Lau,1990;Chang et al.,1996; Sobel and Bretherton,1999;Li et al.,2003,2006;Li and Fu, 2006;Fu et al.,2007;Chen and Huang,2009).Numerical studies suggest that the presence of the MT is important for thegenerationofsynopticwavetrains,andthusTCformation (Kuo et al.,2001;Aiyyer and Molinari,2003;Li,2006;Gall et al.,2010;Gall and Frank,2010).In addition to providing large-scale conditions favorable for TC formation,these studiessuggestthatthesynoptic-scalewavetrainsplayanimportant role in how the MT affects TC formation in the WNP basin.

In order to investigate how the MT affects the development of synoptic-scale wave trains and subsequent TC formation,this study proposes a new approach for identifying the MT and the associated TC formation events.That is,the MT is treated as the background for the activity of synoptic systems including wave trains and TCs.Unlike previous studies in which large-scale patterns were composited with respect to TC centers,the dynamic composite in this study is conducted with respect to the reference point or the easternmost end of the MT.The reference point here refers to the point in the conf l uence zone where both the zonal and meridional wind components are zero with positive vorticity.It is conceivable that the identif i ed TC formation events and the associated structure of the MT should be more reasonable due to a lack of contamination of the synoptic-scale circulation.The TC formationevents associated with the MT in May–October during the period 2000–10 are examined in this study.

2.Data and methodology

2.1.Data

Three datasets are used in this study:(1)National Centers for EnvironmentalPrediction(NCEP)Final(FNL)Operational Global Analysis data with 1°×1°grids at six hour intervals(http://rda.ucar.edu/datasets/ds083.2/);(2)European Centre for Medium-Range Weather Forecasts(ECMWF)Interim Re-Analysis(ERA-Interim)data with 1.5°×1.5°grids (Simmons et al.,2007);and(3)Joint Typhoon Warning Center(JTWC)best track data.In this study,the FNL wind fi elds are utilized in identifying the MT and compositing the MT structure.The 850 hPa monthly winds from ERA-interim data are also used to compare the identi fi ed TC formation events with those in Molinari and Vollaro(2013).TC formation is de fi ned when the maximum sustained wind of a TC fi rst exceeds 34 knots in the JTWC best track data.

2.2.Identi fi cation of monsoon troughs and the associated TC formation

In the American MeteorologicalSociety(AMS)Glossary ofMeteorology(http://amsglossary.allenpress.com/glossary), an MT is de fi ned as a line of the minimum sea level pressure in monsoonal regions.However,most previous studies treated a MT as a circulation pattern in the lower troposphere with convergence/con fl uence between westerly monsoon winds and easterly trade winds(e.g.,Frank,1982;Lander, 1996;Briegel and Frank,1997;Ritchie and Holland,1999). Ritchie and Holland(1999)further categorized a MT as a monsoon con fl uence zone where westerlies meet easterlies and a monsoon shear line where the trade easterlies lie north of the monsoon westerlies.

Following these studies,the 850 hPa low-pass fi ltered winds and the associated relative vorticity are used to identify MTs.As we know,it is not unusual that multiple TCs are observedsimultaneously in the WNP basin.Hsu et al.(2008) revealed that extreme events such as TCs contribute signi ficantly to the seasonal mean and the intraseasonal and interannual variance of the 850 hPa vorticity along the TC tracks in the tropical WNP because the time averaging and fi ltering processes are unable to remove higher-frequencysignals. Following Hsu et al.(2008)and Wu et al.(2013),the TC circulation in the FNL data is fi rst removed with the procedure proposed by Kurihara et al.(1993,1995).

After removal of the TC circulation,a 10-day low-pass Lanczos fi lter is applied to the wind fi eld to obtain the lowfrequency background wind fi eld(Duchon,1979).Figure 1 shows an example of the MT with the 850 hPa low-pass fi ltered wind fi eld.At 0000 UTC 7 July 2006(Fig.1a),the MT can be roughly identi fi ed with the equatorial westerly fl ows that extend all the way to about 160°E and easterly trade winds on the poleward side.Typhoon Ewiniar(2006) and a distinct cyclonic circulation to the southeast of Ewiniar can be seen in the trough.About two and a half days later (Fig.1b),the cyclonic circulation of Typhoon Bilis(2006) becomes dominant in the wind fi eld and it is hard to visually identify the trough.In the corresponding10-daylow-pass fi ltered wind fi elds(Figs.1c and d),however,the MT is well de fi ned.Figures 1e and f are the synoptic-scale wave trains that distort the MT in Figs.1a and b.Here,the synopticscale fl ow is de fi ned as the difference between the un fi ltered winds and the low-pass fi ltered winds.Figure 1 suggests that the low-pass fi ltering can effectively separate the large-scale MT from the synoptic-scale systems,such as TCs and wave trains,consistent with results from Fu et al.(2007,2012)and Xu et al.(2013).

The MT mainly prevails in the WNP basin from early or middle May to late September(e.g.,Murakami and Matsumoto,1994;Lau and Yang,1997;Wu and Zhang,1998; Hsu et al.,1999;Zhang et al.,2002).In this study,we focus on the MT activity and the associated TC formation(0°–30°N,105°E–180°)in May–Octoberduringthe period2000–10.80.8%of TCs(211 out of 261)were observed during the period.Following Ritchie and Holland(1999),an MT in this study includes two parts.The fi rst part consists of westerly monsoon fl ows on the equatorward side and easterly trade fl ows on the poleward,which is called a monsoon shear line or an MT shear region.The second part is a con fl uence zone to the easternmost end of the shear line,where westerly monsoon fl ows from the west meet easterly trade fl ows from the east.First,to identifythe changes of the zonal winds,the MT axis is determined with the line of the zero-zonal wind and positive relative vorticity at 850 hPa.The successive westerly winds(easterly winds)to the west(east)of the MT axis and positiverelativevorticity at 850hPa are then examinedto determine the western(eastern)part of the MT.Considering that the easterlies to the east of the MT axis can extend to the eastern Pacif i c,the negative zonal stretching deformation in the predetermined area is also checked to determine the eastern boundary of the MT.Generally,the MT region extends up to 1500 km from its axis.Once the MT is determined,if a TC forms within this area,it is identif i ed as a TC formation event within the MT.

We should mention that the low-pass Lanczos f i lter cannotcompletelyremovethesynoptictimescalecomponent,especially in the vicinity of the cut-off period(10 days).As a result,the low-pass f i ltered winds are capable of capturing the migration of MTs on the timescale of several days.Fig-ure 2 shows an exampleof the migrationof a MT in the zonal and meridional directions.Note that the MT axis was identif i ed with the line of the zero contour of the low-frequency 850 hPa zonal wind in the monsoon trough.At 0600 UTC 23 June 2006,the MT extended from Indonesia to 140°E to the south of the Philippines.A week later(30 June 2006),the trough migrated northeast by about 10°of latitude,extending from Hainan Island to 150°E.At 0600 UTC 7 July 2006, the trough moved further northeast and the east end reached 170°E.As shown in Fig.2,the MT subsequently retreated southwest.The migration was accompanied by an orientation change of the MT,as well as the MT shape,which can beseeninthelow-passf i ltered850hPawindf i elds(f i gurenot shown).Such a low-frequency variation of the MT is closely related to the atmospheric intraseasonal oscillation(ISO)(Li and Wang,2005;Li,2010;Hsu and Li,2011),and also the quasi-biweekly oscillation(QBWO)in the region(Kikuchi and Wang,2009).Some studies have shown that TC formation can be inf l uenced by the ISO and QBWO modes through their modulations of the MT(e.g.,Ko and Hsu,2009;Li and Zhou,2013).

2.3.Dynamic composite analysis

We noticed that the composite analyses in previous studies were basedmainlyonTC ortropicaldepression(TD)centers(Briegel and Frank,1997;Ritchie and Holland,1999;Fu et al.,2007;Xu et al.,2013).Given the large variation in the relative positions of TC formation within MTs,the composited large-scale patterns may be a poor representation of the large-scale background for TC formation.In this study,the compositeis conductedbasedona referencepoint.Thereference point is in the conf l uence zone where both the zonal and meridional wind components are zero with positive vorticity. The longitude of the reference point can be used to indicate the eastward extension/westward retreat of the MT.

Two typical types of MTs are identif i ed in our analysis.Seventy-three TC formation events are found within southeast–northwest(SE–NW)oriented MTs,while 18 TC formation events occurred in west–east(W–E)oriented troughs.The SE–NW oriented MTs are dominant during summer in the WNP,as shown in Figs.1c and d.This type of MT is very similar to the one in the monthly mean 850 hPa wind f i eld.The other type of MT is largely zonally oriented,whose axes are nearly parallel to the equator.Lander (1996)showed that there are reverse-orientedMTs with their western ends in the tropics extendingnortheastwardinto subtropical latitudes.Such reverse-orientedMTs are excluded in this study.

Although the SE–NW oriented MTs in our study are predominant,accounting for 80.2%,the orientation of these MTs varies signif i cantly.To reduce the uncertainty in our composite analysis,the MT orientation is rotated to the mean orientation based on 91 troughs around the reference point.First,as mentioned before,the MT axis is identif i ed by the connecting line of the zero-zonal wind in the monsoon trough.Second,a straight line is used to f i t the MT axis,and the angle between the axis and the equator is calculated.Finally,the MT axis is rotated to the mean angle and all of the variables in the following discussion are also adjusted based on the rotation.

3.TC formation within monsoon troughs

TC formation events within MTs are identif i ed every six hours in this study.During the period 2000–10,91 TCs are foundto formin MTs in May–October,accountingfor43.1% of the total TC formation events.On average,8.3 TC formation events are associated with MTs per year,with a maximum of 11 TCs in 2008 and a minimum of 6 TCs in 2006. Gao andLi(2011)def i neda multipleTC formationeventas a process in which two successive TCs form within three days, with the distance between them being less than 4000 km, andtheyfoundthattheeastward-extendingWNP MTfavored multiple TC formation events.Based on their def i nition,36 TCs are involved in such multiple TC formation events,accounting for 39.6%of the TCs associated with MTs.

During the six-month season,the monthly frequency of TC formation within the MT reaches its maximum in August(Fig.3a).However,compared to the total TC formation frequency,the percentage of TC formation within the MT increases from May(44.4%)to June(70.6%)and then shows a decreasing trend after July.While more than 50%of TCs formed in MTs during June–July,the percentage becomes less than 50%during August–October.In particular,only 12.9%of TCs formed within MTs in October.This agrees with previous studies that MTs occur mainly from early or middle May to late September(e.g.,Murakami and Matsumoto,1994;Lau and Yang,1997;Wu and Zhang,1998; Hsu et al.,1999;Zhang et al.,2002).We also calculate MT activedaysforeachmonth(Fig.3b).MTactivedaysaredays withMTactivityintheirlow-frequencywindf i elds.Itisclear that the decreasing percentage of TC formation in MTs after July is accompanied by a decreasing number of MT active days.

We f i nd that the percentages of TC formation in MTs are much lower than those in previous studies(Briegel and Frank,1997;Ritchie and Holland,1999;Molinari and Vol-laro,2013).Ritchie and Holland(1999)found that 71%of TCs formed within MTs,including 42%within the monsoon shearregionand29%withintheconf l uencezone.InMolinari and Vollaro(2013),73%of all July–November TCs formed within MTs duringthe period 1988–2010,andthe percentage varied interannually from as low as 50%to nearly 100%.

As mentionedabove,thedifferenceinthepercentagemay result from the different approaches for identifying monsoon troughs and the associated TC formation.Since we have diff i culty in reproducing the results from Briegel and Frank (1997)and Ritchie and Holland(1999),we focus here on a comparison with Molinari and Vollaro(2013).

Molinari and Vollaro(2013)def i ned an MT as the contiguous region with positive long-term mean July–November 850 hPa relative vorticity.Their analysis region ranged from 122°E to 180°.To understand how the different approaches can affect the results,we f i rst use the approach in Molinari and Vollaro(2013)to identify the numberof TCs that formed in MTs.To avoid a possible inf l uence of the difference in datasets,both the ERA-Interim reanalysis data and the FNL data from 2000 to 2010 are used.The former dataset was used in Molinari and Vollaro(2013).

When we use the approach in Molinari and Vollaro (2013),the TC formation percentages based on the ERAInterim reanalysis data and the FNL data are 58.8%and 58.3%,respectively,both of which are higher than the 43.1% established by our approach.Table 1 shows the monthly frequencyofTC formationeventsidentif i edwiththeapproaches usedinthisstudyandinMolinariandVollaro(2013).TheTC formation events identif i ed from the two datasets using the method of Molinari and Vollaro(2013)are nearly the same (Table 1),suggesting that the difference in the TC formation percentage is not a result of the different datasets.

However,the monthly percentages of TC formation in MTs from our approach are very different from the approach in Molinari and Vollaro(2013),especially in September and October.While 34 and 19 TCs are identif i ed to form within the climatological mean MT by the approachof Molinari and Vollaro(2013),our approach identif i es only 17 and 4 TCs in September and October,respectively.

Figure 4 shows a comparison of TC formation events identif i ed with the two different approaches.While some TC formation events(black)are identif i ed by both approaches, a considerable number of TC formation events are counted only by one or the other of the two approaches.It is clear that TC formation events in the South China Sea(west of 120°E)are not counted by the approach of Molinari and Vollaro(2013).In addition,Fig.4 indicates that TC formation events(green)that are counted by the approach of Molinari and Vollaro(2013)are excluded in our approach,in particular after August.Among the excluded TC formation events, further analysis shows that half of them formed in monsoon gyres.

Wu et al.(2013)recently indicated that,over the WNP, 19.8%of TCs formed in monsoon gyres in May–October during the period 2000–10,but the contribution of monsoon gyres cannot be distinguished from the MT def i ned in Molinari and Vollaro(2013).That is,the TC formation events from the approach of Molinari and Vollaro(2013)include those within monsoon gyres.If the percentage of TC formation eventswithin monsoongyresis added,we have 62.9%of TCs that form in both MTs and monsoon gyres.As shown in Fig.3a,the percentage of TC formation in monsoon gyres is morethan20%inAugust–October.Theincreasedpercentage is consistent with the increase in monsoon gyre active days in these months(Fig.3b).Except for October(35.5%),the combined percentage ranges from 61.1%in May to 82.4%in June,which is comparable to those in previous studies.

Figure 5 shows the composited 850 hPa low-frequency wind f i eld for the MTs at the TC formation time.Ritchieand Holland,(1999)found that 29%of TCs formed within the conf l uence zone,while 42%of the TC formation events occurred within the monsoon shear region.As shown in Fig. 5,nearly all of the TC formation events occurred in the monsoon shear region.The difference may result from the contamination of synoptic-scale systems,such as TCs and wave trains,which were included in Ritchie and Holland(1999). In the following section,we show that the region favorable for TC formation is associated with the favorable large-scale conditions for TC formation.

Table 1.Monthly frequency of TC formation in monsoon troughs identi fi ed with low-pass fi ltered fi elds and monthly mean fi elds in May–October during 2000–10.For comparison,the monthly frequency of TC formation in monsoon gyres(Wu et al.,2013)and the monthly frequency of TC formation in monsoon troughs identi fi ed from the ERA-Interim dataset are also shown in parentheses in the fi rst and second rows,respectively.

4.Large-scale environment for TC formation

In order to understand why the monsoon shear region is favorable for TC formation,we conduct a composite analysis of the MT structure of the selected 91 samples during the period 2000–10.In agreement with previous studies(Sadler, 1967;Gray,1968;Ramage,1974;Gray,1975;Frank,1987; Chia and Ropelewski,2002;Harr and Chan,2005),the monsoonshearregionis associatedwith stronglow-levelcyclonic relative vorticity,high mid-level relative humidity,and rela-tively weak vertical wind shear,which are favorable for TC formation.

Figure 6a shows the relative vorticity and zonal wind component across the axis of the composited MT(A1–A2). As shown in Fig.5,the cross section represents the most favorable region for TC formation,about 1000 km west of the reference point.The positive relative vorticity associated with the MT is conf i ned to a narrow region of about 1200 km in width,extending to about 350 hPa.On the equatorward side are the westerly monsoon winds,covering about 1200 km in width below 600 hPa.On the poleward side,the easterly trade winds cover a wider region and extend throughout the troposphere.In agreement with observations,it is suggested that TCs that formwithin the MT are generallysteered westward by the mean deep-layer mean easterly f l ow.

Figure 6b shows the cross section of relative vorticity and relative humidity along the MT axis at the time of TC formation(B1–B2 in Fig.5).The maximum relative vorticity occurs around 850 hPa and the positive relative vorticity is conf i ned in the middle and lower troposphere(below 500 hPa).Consistent with the high TC formation rate,as shown in Fig.6b,the region with high relative vorticity and relative humidity occurs in the east part of the shear region.

Figure 7a shows the composited vertical wind shear f i eld between 200 hPa and 850 hPa at the TC formation time.The environmental vertical wind shears are weak near the MT axis.Most of the TCs formed in weak vertical wind shear of less than 12.5 m s−1.To further understand the spatial distribution pattern of TC formation in the MT,we compare the vertical prof i les of relative vorticity,zonal wind,and relative humidity in three selected regions.As shown in Fig. 5,Regions A and B are in the shear-line region,where there is the most TC formation events,while Region C is in the conf l uence region,featuringless TC formation.Compared to Regions A and B,Region C is characterized by weaker lowlevel relative vorticity(Fig.8a),which is unfavorable for TC formation(Gray,1968,1975).But there is not much difference in the mid-level relative humidity between Regions A, B and C(Fig.8c).

A signif i cant feature of Region C is the westerly zonalvertical shear(Fig.8b).In a well-known idealized modeling study,Kurihara and Tuleya(1981)showed that mean easterly vertical wind shear is more favorable for TC formation than westerly vertical wind shear.In order to examine the favorableness of TC formation in the easterly vertical wind shear, we plot a histogram of frequencies of zonal wind shear in bins of 5.0 m s−1for all of the formation events in MTs(Fig. 7b).The shear between 200 hPa and 850 hPa is calculated based on individual TC events at TC formation time with the f i ltered winds.While the TCs can form only in westerly zonal shear of less than 5.0 m s−1,most formation events occur in easterly zonal shear,even as high as 10.0 m s−1. Figure9showsthecompositedthree-dimensionalstructureof MTs.Thelow-levelMTis overlaidbythenortheasterlyf l ows associated with the South Asia high at 200 hPa.This is why most of the formation events occurred in the easterly zonal shear.

Although this result is in agreement with the idealized modeling result in Kurihara and Tuleya(1981),the favorableness of easterly zonal shear in MTs may be associated with other large-scale conditions and the initial synopticscale tropicaldisturbancesthatalways precedeTC formation. Using a multilevel,nonlinear baroclinic model,Li(2006) suggested that the synoptic wave train in an MT may result from instability of the environmental fl ow.His numerical experiments indicate that an easterly shear can lead to a faster growth and northwestward propagation of synoptic wave trains than a westerly shear.Ge et al.(2007)examined the effect of vertical shear on wave trains induced by TC energy dispersion and found that an easterly(westerly)shear favorsthe growthofthe wavetrainin thelower(upper)troposphere.Wang and Xie(1996)found that vertical wind shear can signi fi cantly affect the Rossby wave and westward Yanai wave.While westerly shear favors trapping of these waves in the upper troposphere,easterly shear tends to con fi ne them to the lower troposphere.In all of the 91 TC formation events, we fi nd that 38 cases were clearly associated with synoptic wave trains.We shouldmentionthat the resulting wave trains or tropical disturbances may provide a synoptic environment to counterthe low-frequencyvertical wind shear and enhance the other favorable conditions for TC formation.Further investigation is called for in this respect.

FerreiraandSchubert(1997)andLi(2006)suggestedthat the initial disturbances for TC formation may arise from the instability of the environmental fl ow.Based on numerical simulations,Li(2006)argued that the origin of the synoptic wave train arises from both the dynamic effect of the MT fl ow and the thermodynamic effect of the convective heatingfeedback.In order to examine the possible role of barotropic instability,the meridional gradients of the absolute vorticity of the composited MT were calculated.There are indeed isolated regions with the sign reversal of the absolute vorticity gradient(f i gure not shown).This suggests that the barotropic instability of the MT f l ow may play a role in TC formation in MTs.

5.Summary

Unlike previousstudies in whichthe MTwas subjectively def i ned(BriegelandFrank,1997;Ritchie andHolland,1999) or treated as a geographical region in terms of monthly mean 850 hPa relative vorticity(Wu et al.,2012;Molinari and Vollaro,2013),this study re-examines the TC formation in the MT over the WNP by treating it as a large-scale background fortheactivityofsynopticsystems,includingwave trainsand TCs.Given the possible distortion of the large-scale pattern compositedwith respecttoTCcentersinpreviousstudies due to large variation in the relative positions of TC formation within MTs(e.g.,Briegel and Frank,1997;Ritchie and Holland,1999),the composite in this study is conducted based on a reference point in the conf l uence zone where both the zonal and meridional wind components are zero with positive vorticity.

During the period 2000–10 in May–October,91 TC formation events are identif i ed in MTs,with a seasonal average of 8.3 formation events or 43.1%of all TC formation events in the WNP.The percentage of TC formation in MTs ranges from 12.9%in October to 70.6%in June.The resulting percentage of TC formationevents in MTs is much lower than those in previous studies(e.g.,Briegel and Frank,1997; Ritchie and Holland,1999;Molinari and Vollaro,2013).The difference may result from the different approaches used in identifying MTs and the associated TC formation events.In addition,we argue that previous studies may have included TC formation events in monsoon gyres(Molinari and Vollaro,2013).Wu et al.(2013)foundthat 19.8%ofTCs formed in monsoon gyres in May–October during the period 2000–10.Except for October(35.5%),the combined percentage in MTs andmonsoongyresrangesfrom61.1%inMay to82.4% in June,which is comparable to those in previous studies.

Nearly all of the TC formation events occur in the MT shear region,while previousstudies indicatedthat a considerablenumberofTCsformedintheconf l uencezone.Theshear line region is associated with stronger low-level relative vorticity,easterly vertical shear of zonal wind,and barotropicinstability.The favorableness of easterly shear agrees with the observationalanalysis in Lee(1989)andthe modelingresults in Kurihara and Tuleya(1981).We argue that the favorableness in the WNP may be associated with the development of precursor tropical disturbances in the MT that are necessary for TC formation.

In the last decade or so,some studies have foundTC genesis frequency to be related to long-term atmospheric oscillations,such as the North Pacif i c Oscillation(NPO)and the Antarctic Oscillation(AAO),and the sea ice cover over the North Pacif i c(e.g.,Wang and Fan,2006;Wang et al.,2007a, 2007b).In those studies,the inf l uences were associated with the changes in the large-scale environmental conditions for TC formation.It is possible that they may affect TC formation through changing the orientation and zonal extension of the monsoon trough.Further investigation in this f i eld is needed.

Acknowledgements.This research was jointly supported by the National Basic Research Program of China(Grant Nos. 2013CB430103,2015CB452803),the National Natural Science Foundation of China(Grant No.41275093),and the project of the“Specially-Appointed Professorship”of Jiangsu Province.

REFERENCES

Aiyyer,A.R.,and J.Molinari,2003:Evolution of mixed Rossbygravity waves in idealized MJO environments.J.Atmos.Sci., 60,2837–2855.

Briegel,L.M.,and W.M.Frank,1997:Large-scale inf l uences on tropical cyclogenesis in the western North Pacif i c.Mon.Wea. Rev.,125,1397–1413.

Chang,C.P.,J.M.Chen,P.A.Harr,and L.E.Carr,1996: Northwestward-propagating wave patterns over the tropical western North Pacif i c during summer.Mon.Wea.Rev.,124, 2245–2266.

Chen,G.H.,and R.H.Huang,2009:Interannual variations in mixed Rossby–gravity waves and their impacts on tropical cyclogenesis over the western North Pacif i c.J.Climate,22, 535–549.

Chen,T.-C.,S.-Y.Wang,and M.-C.Yen,2006:Interannual variation of the tropical cyclone activity over the western North Pacif i c.J.Climate,19,5709–5720.

Chia,H.H.,and C.F.Ropelewski,2002:The interannual variability in the genesis location of tropical cyclones in the Northwest Pacif i c.J.Climate,15,2934–2944.

Duchon,C.E.,1979:Lanczos f i ltering in one and two dimensions. J.Appl.Meteor.,18,1016–1022.

Ferreira,R.N.,and W.H.Schubert,1997:Barotropic aspects of ITCZ breakdown.J.Atmos.Sci.,54,261–285.

Frank,W.M.,1982:Large-scale characteristics of tropical cyclones.Mon.Wea.Rev.,110,572–586.

Frank,W.M.,1987:Tropical cyclone formation.A Global View of Tropical Cyclones,R.L.Elsberry,Ed.,University of Chicago Press,53–90.

Fu,B.,T.Li,M.S.Peng,and F.Z.Weng,2007:Analysis of tropical cyclogenesis in the western North Pacif i c for 2000 and 2001.Wea.Forecasting,22,763–780.

Fu,B.,M.S.Peng,T.Li,and D.E.Stevens,2012:Developing versus nondeveloping disturbances for tropical cyclone formation.Part II:Western north Pacif i c.Mon.Wea.Rev.,140, 1067–1080.

Gall,J.S.,and W.M.Frank,2010:The role of equatorial Rossby waves in tropical cyclogenesis.Part II:Idealized simulations in a monsoon trough environment.Mon.Wea.Rev.,138, 1383–1398.

Gall,J.S.,W.M.Frank,and M.C.Wheeler,2010:The role of equatorial Rossby waves in tropical cyclogenesis.Part I:Idealized numerical simulations in an initially quiescent back-ground environment.Mon.Wea.Rev.,138,1368–1382.

Gao,J.Y.,and T.Li,2011:Factors controlling multiple tropical cyclone events in the western North Pacif i c.Mon.Wea.Rev., 139,885–894.

Ge,X.Y.,T.Li,and X.Q.Zhou,2007:Tropical cyclone energy dispersion under vertical shears.Geophys.Res.Lett.,34, L23807,doi:10.1029/2007GL031867.

Gray,W.M.,1968:Global view of the origin of tropical disturbances and storms.Mon.Wea.Rev.,96,669–700.

Gray,W.M.,1975:Tropical cyclone genesis.Paper No.234,Dept. of Atmos.Sci.,Colo.State Univ.,Ft.Collins,CO,121 pp.

Harr,P.A.,and J.C.L.Chan,2005:Monsoon impacts on tropical cyclone variability.WMO Tech.Doc.1266,TMPR Rep.70, WMO,512–542.

Harr,P.A.,and C.-C.Wu,2011:Tropical cyclone characteristics and monsoon circulations.The Global Monsoon System:Research and Forecast,2nd ed.,C.-P.Chang et al.,Eds.,World Scientif i c Publishing,357–372.

Holland,G.J.,1995:Scale interaction in the western Pacif i c monsoon.Meteor.Atmos.Phys.,56,57–79.

Hsu,H.-H.,C.-T.Terng,and C.-T.Chen,1999:Evolution of large-scale circulation and heating during the f i rst transition of Asian summer monsoon.J.Climate,12,793–810.

Hsu,H.-H.,C.-T.Terng,A.-K.Lo,C.-C.Wu,and C.-W.Hung, 2008:Inf l uence of tropical cyclones on the estimation of climate variability in the tropical western North Pacif i c.J.Climate,21,2960–2975.

Hsu,P.C.,andT.Li,2011:Interactionsbetween boreal summer intraseasonal oscillations and synoptic-scale disturbances over the western North Pacif i c.Part II:Apparent heat and moisture sources and eddy momentum transport.J.Climate,24, 942–961.

Kikuchi,K.,and B.Wang,2009:Global perspective of the quasibiweekly oscillation.J.Climate,22,1340–1359.

Ko,K.C.,and H.H.Hsu,2009:ISO modulation on the submonthly wave pattern and recurving tropical cyclones in the tropical western North Pacif i c.J.Climate,22,582–599.

Kuo,H.C.,J.H.Chen,R.T.Williams,and C.-P.Chang,2001: Rossby waves in zonally opposing mean f l ow:Behavior in northwest Pacif i c summer monsoon.J.Atmos.Sci.,58,1035–1050.

Kurihara,Y.,and R.E.Tuleya,1981:A numerical simulation study on the genesis of a tropical storm.Mon.Wea.Rev.,109, 1629–1653.

Kurihara,Y.,M.A.Bender,and R.J.Ross,1993:An initialization scheme of hurricane models by vortex specif i cation. Mon.Wea.Rev.,121,2030–2045.

Kurihara,Y.,M.A.Bender,R.E.Tuleya,and R.J.Ross,1995:Improvements in the GFDL hurricane prediction system.Mon. Wea.Rev.,123,2791–2801.

Lander,M.A.,1996:Specif i c tropical cyclone track types and unusual tropical cyclone motions associated with a reverseoriented monsoon trough in the western North Pacif i c.Wea. Forecasting,11,170–186.

Lau,K.H.,and N.C.Lau,1990:Observed structure and propagation characteristics of tropical summertime synoptic scale disturbances.Mon.Wea.Rev.,118,1888–1913.

Lau,K.M.,and S.Yang,1997:Climatology and interannual variability of the southeast Asian summer monsoon.Adv.Atmos. Sci.,14,141–162,doi:10.1007/s00376-997-0016-y.

Lee,C.S.,1989:Observational analysis of tropical cyclogenesis in the western North Pacif i c.Part I:Structural evolution of cloud clusters.J.Atmos.Sci.,46,2580–2598.

Li,R.C.Y.,and W.Zhou,2013:Modulation of western North Pacif i c tropical cyclone activity by the ISO.Part I:Genesis and Intensity.J.Climate,26,2904–2918.

Li,T.,2006:Origin of the summertime synoptic-scale wave train in the western North Pacif i c.J.Atmos.Sci.,63,1093–1102.

Li,T.,2010:Monsoon climate variabilities.Climate Dynamics:Why Does Climate Vary?Geophys.Monogr.Ser.,D.-Z. Sun and B.Frank,Eds.,American Geophysical Union,doi: 10.1029/2008GM000782.

Li,T.,2012:Synoptic and climatic aspects of tropical cyclogenesis in western North Pacif i c.Cyclones:Formation,Triggers and Control,K.Oouchi and H.Fudeyasu,Eds.,Nova Science Publishers,Inc.,61–94.

Li,T.,and B.Wang,2005:A review on the western North Pacif i c monsoon:Synoptic-to-interannual variabilities.Terrestrial,Atmospheric and Oceanic Sciences,16,285–314.

Li,T.,and B.Fu,2006:Tropical cyclogenesis associated with Rossby wave energy dispersion of a pre-existing typhoon. Part I:Satellite data analyses.J.Atmos.Sci.,63,1377–1389. Li,T.,B.Fu,X.Ge,B.Wang,and M.Peng,2003:Satellite data analysis and numerical simulation of tropical cyclone formation.Geophys.Res.Lett.,30,2122,doi:10.1029/2003GL 018556.

Li,T.,X.Y.Ge,B.Wang,and Y.T.Zhu,2006:Tropical cyclogenesis associated with Rossby wave energy dispersion of a pre-existing typhoon.Part II:Numerical simulations.J.Atmos.Sci.,63,1390–1409.

McBride,J.L.,1995:Tropical cyclone formation.Global Perspectives on Tropical Cyclones,R.L.Elsberry,Ed.,World Meteorological Organization,63–105.

Molinari,J.,and D.Vollaro,2013:What percentage of western north pacif i c tropical cyclones form within the monsoon trough?Mon.Wea.Rev.,141,499–505.

Murakami,T.,and J.Matsumoto,1994:Summer monsoon over the Asian continent and western North Pacif i c.J.Meteor.Soc. Japan,72,719–745.

Ramage,C.S.,1974:Monsoonal inf l uences on the annual variation of tropical cyclone development over the Indian and Pacif i c oceans.Mon.Wea.Rev.,102,745–753.

Ritchie,E.A.,and G.J.Holland,1999:Large-scale patterns associated with tropical cyclogenesis in the western Pacif i c.Mon. Wea.Rev.,127,2027–2043.

Sadler,J.C.,1967:On the origin of tropical vortex.Proc.Working Panel on Tropical Dynamic Meteorology,Naval Weather Research Facility,Norfolk,VA,39–75.

Simmons,A.,S.Uppala,D.Dee,and S.Kobayashi,2007:ERAInterim:New ECMWF reanalysis products from 1989 onwards.ECMWF Newsletter,110,25–35.

Sobel,A.H.,and C.S.Bretherton,1999:Development of synoptic-scale disturbances over the summertime tropical Northwest Pacif i c.J.Atmos.Sci.,56,3106–3127.

Wang,B.,and X.S.Xie,1996:Low-frequency equatorial waves in vertically sheared zonal f l ow.Part I:Stable waves.J.Atmos. Sci.,53,449–467.

Wang,H.J.,and K.Fan,2006:Relationship between the Antarctic Oscillation in the western North Pacif i c typhoon frequency. Chinese Science Bulletin,51,561–565.

Wang,H.J.,J.Q.Sun,and K.Fan,2007a:Relationships between the North Pacif i c Oscillation and the typhoon/hurricane frequencies.Science in China(D),50,1409–1416.

Wang,H.J.,K.Fan,J.Q.Sun,X.M.Lang and M.J.Lin,2007b:Some advances in the researches of the western North Pacif i c typhoon climate variability and prediction.Chinese Journal of Atmospheric Sciences,31,1076–1081.(in Chinese)

Wu,G.X.,and Y.S.Zhang,1998:Tibetan Plateau forcing and the timing of the monsoon onset over South Asia and the South China Sea.Mon.Wea.Rev.,126,913–927.

Wu,L.,Z.P.Wen,R.H.Huang,and R.G.Wu,2012:Possible linkage between the monsoon trough variability and the tropical cyclone activity over thewestern North Pacif i c.Mon.Wea. Rev.,140,140–150.

Wu,L.G.,H.J.Zong,and J.Liang,2013:Observational analysis of tropical cyclone formation associated with monsoon gyres. J.Atmos.Sci.,70,1023–1034.

Xu,Y.M.,T.Li,and M.Peng,2013:Tropical cyclogenesis in the western North Pacif i c as revealed by the 2008–09 YOTC Data.Wea.Forecasting,28,1038–1056.

Zhang,Y.S.,T.Li,B.Wang,and G.X.Wu,2002:Onset of the summer monsoon over the Indochina Peninsula:Climatology and interannual variations.J.Climate,15,3206–3221.

:Zong,H.J.,and L.G.Wu,2015:Re-examination of tropical cyclone formation in monsoon troughs over the western North Pacif i c.Adv.Atmos.Sci.,32(7),924–934,

10.1007/s00376-014-4115-2.

(Received 4 June 2014;revised 27 October 2014;accepted 5 December 2014)

∗Corresponding author:WU Liguang Email:liguang@nuist.edu.cn

©Institute of Atmospheric Physics/Chinese Academy of Sciences,and Science Press and Springer-Verlag Berlin Heidelberg 2015