Cold Wave in East and Southeast Asia and Their Effect on Atmospheric Circulation at Lower Latitudes

Cold Wave in East and Southeast Asia and Their

Effect on Atmospheric Circulation at Lower

Latitudes

著者

Nomura Chihiro

雑誌名

The science reports of the Tohoku University.

7th series, Geography

巻

50

号

1

ページ

59-74

発行年

2000-06

URL

http://hdl.handle.net/10097/45248

59

Cold Waves in East and Southeast Asia and Their Effect

on Atmospheric Circulation at Lower Latitudes

Chihiro NOMURA*

Abstract Cold waves in East and Southeast Asia based on the spatial pattern

of 850 hPa temperature anomalies and their effect on lower-level atmospheric

circulation at lower latitudes are examined in this paper. First, by applying

rotated principal components analysis to daily 98-station data in the area 100

—150°E , 0°-45°N, for 4 winters (1983/84-1986/87), five cold anomaly patterns

of 850 hPa temperature are extracted ; Type I (central China), Type II

(north-eastern China), Type III (Japan), Type IV (southern China), and Type V

(north-ern Japan). Secondly, the effect of these cold anomaly patterns and cold wave

events on cloud activity and lower-level atmospheric circulation at lower

latitudes is investigated. As a result, based on the moving direction of tropical

disturbance in the area near the equator, three kinds of lower-level atmospheric

condition at lower latitudes are regarded ; one usual eastward phase pattern

(EP pattern) and two unusual westward phase patterns (WP-N pattern and

WP-S pattern). The features are summarized as follows ;

WP-N pattern : In this pattern, cold waves have closest relation with

lower-level atmospheric condition at lower latitudes. Cross equatorial flow

derived from a part of cold wave flow in the Northern Hemisphere and

remark-able equatorial westerly appear near not only maritime continent but also the

Indian Sea, and these condition in latter site are associated with appearance of

easterly wave in the Bay of Bengal. These process leads to unusual westward

moving of tropical disturbance in the area near the equator.

WP-S pattern : Unusual westward moving of tropical disturbance in the

area near the equator in this pattern is due to appearance of a cyclonic

circulation in Southern Hemisphere.

EP pattern : The flow derived from a part of cold wave does not reach to

lower latitudes in this pattern.

In Particular, WP-N pattern appears to be accompanied with the

emer-gence of Type IV of cold anomaly patterns. Same tendency is somewhat

regarded in Type III.

Key words : cold wave, East Asia, Southeast Asia, rotated principal

nents analysis, midlatitude-tropical interaction

* Graduate Student_{, Institute of Geography, Tohoku University, Sendai 980-8578}

Science Reports of Tohoku University, 7th Series (Geography) Vol. 50, No. 1, June, 2000, 59-74

1. Introduction

In most of studies on midlatitude-tropical interaction through Asian monsoon in

winter, the monsoonal character may be represented by seasonal or monthly mean

data (e.g., Tomita and Yasunari, 1996). However, in subtropical region over the

western Pacific, daily diabatic heating at the lowest atmospheric layer is balanced with

the horizontal temperature advection by heat loss of the ocean due to air-sea heat

exchange induced by cold wave flow (Kawamura, 1988). Therefore, cold waves that

occur in shorter term than a season or a month are remarkably associated with

midlatitude-tropical interaction.

In particular, cold waves reached to lower-latitudes have studied as "cold surges".

Chang and Lau (1980) showed the meridional process that occurrence of a cold surge

strengthened the local Hadley circulation and activated the near-equatorial

convec-tion, but details of its zonal process in Asia is not clarified. In various studies on cold

surges, its appearance (or onset) is defined by the local weather change at a specific key

location, and its temporal development has examined by composite analysis based on

its onset (e.g., Wu and Chang, 1995). But with these method, the spatial pattern of the

influence by the cold waves and various temporal continuance of the cold waves can

not be clarified.

In this study, cold waves are analyzed by daily temperature anomalies at 850 hPa

that are standardized in each station. With this method, cold waves at

lower-lati-tudes where daily temperature change is comparatively _{small can be clarify . By}

applying rotated principal components analysis to these temperature anomalies, cold

waves are investigated from the spatial aspect as cold anomaly patterns and from the

temporal aspect with time series of component score. Nomura (1999) examined the

classification of cold waves in East and Southeast Asia based on these spatial pattern

of 850 hPa temperature anomalies and their relation to cloud activity at lower

lati-tudes. The present paper is a further study that the effect on disterbance near the

equator and on lower-level atmospheric circulation at lower latitude by appearance of

their cold anomaly patterns or cold waves is investigated.

2. Data and method

2.1. Data

The calculation of temperature anomalies were carried out by use of the sounding

data (SD tape) from Japan Meteorological Agency. The reason for using this data in

this study was to extract cold anomaly patterns by observation data.

To investigate the activity of disturbance _{near the equator , time-longitude}

Cold Waves in East and Southeast Asia and Their Effect on Atmospheric Circulation at Lower Latitudes 61

was utilized. The original of this data is "Monthly Retort on Climate System,

Technical Note No. 32", pubrished by the Japan Meteorological Agency.

In the study of lower-level atmospheric circulation, wind at 1,000 hPa is

investigat-ed by using NCEP/NCAR (National Centers for Environmental Prediction/National

Center for Atmospheric Research) global reanalysis data (5-day mean), cloud activity

is examined by using 5-day mean OLR data.

The period that is examined in this study is from 1983/84 to 1986/87 (4 winters : from November to March (missing data : March 1987)).

2.2. Method

To extract cold anomaly patterns, temperature at 850 hPa was used. Since it is

impossible to remove the influence of the difference between each station and this level

was identified as the level of maximum horizontal heat flux in the East Asian region

by Blackmon et al. (1977), therefore temperature at 850 hPa is best to represent

depression of temperature by cold waves.

Analysis of temperature anomalies was performed to the daily data (at 00 GMT),

because the days before and after the onset of cold surges have been investigated by

using daily data in most of the previous studies (e.g. Wu and Chan, 1995). But 5-day

running mean data is used for making the influence of synoptic disturbances like

travelling anticyclones or cyclones at midlatitudes small.

Daily temperature anomaly from climatological value at 850 hPa was divided by

standard deviation at each station to normalize. For this calculation, climatological

mean temperature and standard deviation are necessary on each calendar day, but the

period of analysis is only 4 winters. In this study, average and standard deviation of

temperature at 850 hPa on each calendar day and its 15 days before and after during

4 winter (124 days in total) were substituted for 'climatological mean temperature and

standard deviation'. Using the data from November to March'), 'climatological mean

temperature and standard deviation' of the date from 16 November to 16 March were

calculated and therefore 467 distribution rnaps2) were made. The reason why this

process of normalization by these 'climatological mean temperature and standard

deviation' was applied in this study is that distribution of cold anomalies not affected

by the fluctuation of temperature in winters was tried to extract.

To investigate the cold anomalies at midlatitude and lower latitude in East and

Southeast Asia, the analyzed region is 0°-45°N, 100-150°E (Fig. 1). On applying

princi-pal components analysis to extract the cold anomaly patterns, succeeding data is

needed. Therefore 98 stations (• in Fig. 1) where the missing data is below one third

of all analyzed day are adopted in this statistics analysis3). However, there are few

stations at lower-latitude near the Philippine. To Make the influence of these

4crN•4'1',:.;..

, fp' : •, -- --•.,. •,,•

• I •

_{rr----11%,c,:r?-4}

ti

•irOea: I •--) 1 • - - - -'-•_, ' • 7 • • ^ •

3"

-- •i

_{-7 --0:-• _,Sili*:' .}

1

•

• 1 :'• _{, • ,} •411: ..- ••___ 4 •,• ----—

IIP-' •4Tr---•V

IL\

i>,,i

i,

20°N_,1.4,,Or • ---- - , - - 7 --- ' •

qir

• , 10°N _ _ -

l?

V-51.0

• •

)1

al N I

)/7

I • 1

CF.4

1 _ - 40°N I I I 4,, 300 • I II I I II _ - _ 30°N 0° 0°N V • I I 0°N /00-E Fig. 1 • 0 , 1, 50 • / 10°E120°E 130°E140°E

The location of aerological stations used in this study Stations selected for rotated principal components analysis Stations compared with the results of the analysis

(1987) described that stable solutions could be obtained generally by the rotated

principal components analysis to large spatial degree of freedom compared with the

domain. This analysis have often been applied for the station data which are

dis-tributed unequally in the study area (e.g., Yasunari et al., 1996).

For applying the rotated principal components analysis in this study,

variance-covariance matrix was used. The five principal components were rotated by using the

varimax method. The number of components for rotation is decided according to the

maximum difference of each eigenvalue.

3. Spatial patterns of cold waves in East and Southeast Asia

Fig. 2 shows the geographical distributions of factor loading of five components

and these percentage variances from rotated principal components analysis for the 850

hPa temperature anomaly. The cumulative percentage variance of the 5 components

is 79.1%.

Cold Waves in East and Southeast Asia and Their Effect on Atmospheric Circulation at Lower Latitudes ₆₃

a

40 _°,1

_{,_•:;1,L}

_-

SO

V 7

20.N 10J . . 401`1 3014 • J 100E

407v.r)

30°N 4 4 1 20.N

----4:

107V 0.

d

401

Ig./• L._ 201 iO N - COMP1 28. 0% 0° 0. r.; 40°N Q • r • 30°N 11 CL,. (:17E 110•E 120°E

C

40°N 130'E 140°E 30-N I 1 11 "II 2014 COMP2 16. 8% 10°N

110°E _{120°E 130°E 140°E} 150'

40'N 20'N

a3

I. • 30Is) 30°N

L

Fr'r4—ai

_ 4,-. orArN,. —,-...- •;'' .1' L ---' GOMP3 20°N 10'N

Er:

A40°N

rrrqit;:S

;JW(41707

, . 107vi! 30°N 16. 4% 110°E 120E 140E 0-, 1.50 .V.:4 100° • /10°E ic 130E

e

4" 120°E 20°N

• COMP4

14CE 130°E 4.cr 150° 30w 20°N 0 low ±-100 `E .• !_]

6 .2.

• • 40t4 1.30°N 4_,

"7OMP5

7.0% 201•1 7FH 10°N

110°E • 120°E 130°E 140rE

Fig. 2 Geographical distributions of factor loading of five components derived from

rotated pricipal components analysis for the 850 hPa temperature anomaly in winter

(1983/84-1986/87)

solid line : -F, dashed line : — 4.0°, 150°m

first component (Fig. 2-a), in northeastern China on the second component (Fig. 2-b), over the south of Japan on the third component (Fig. 2-c), over the South China Sea on

the fourth component (Fig. 2-d), in northern Japan on the fifth component (Fig. 2-e).

The negative score of each anomaly pattern (the lst---5th components) represents

each cold anomaly pattern. Since dealing with cold waves in this study, these cold

anomaly patterns are examined. Incidentally, the positive score of each anomaly

pattern represents the condition that a cold wave does not occur on the continental

area or that an extratropical cyclones passes on the oceanic area, but this is not my

present concern.

Therefore, these cold anomaly patterns were named Type I (central China) for the

first component, Type II (northeastern China) for the second component, Type III

(Japan) for the third component, Type IV (south China) for the fourth component, Type V (northern Japan) for the fifth component.

4. Effect on low-level atmospheric circulation at lower latitudes by cold waves

4.1. Cloud activity near the equator in relation to these cold anomaly patterns

In this chapter, cloud activity near the equator in relation to these cold anomaly

patterns is examined. Fig. 3 shows the time series of component scores (Fig. 3-a, b, c,

d, e) and time-longitude sections of 5-day mean OLR averaged from 10°N to 10°S (Fig.

3-f). Fig. 3-g is explained in next chapter.

Atmosphere near the equator has the feature of the eastward-moving wave with

30-50-day oscillation (Madden and Julian, 1972). But the period that this

eastward-moving wave is not remarkable exists. In Fig. 3-f, tropical disturbance in the area

near the equator by OLR move eastward in 1984/85 winter and 1986/87 winter, on the

other hand, the eastward-moving of the tropical disturbance is not definite in 1983/84

winter and 1985/86 winter. To investigate the relation between the period that

eastward-moving of the tropical disturbance is not remarkable and a cold wave,

disterbances area near the South China Sea (100-120°E), where Chang et al. (1979)

studied cold surges and their effect, is examined in detail. In Fig. 3-f, rectangles with

solid lines and dashed lines indicate westward and weak westward phase of the

tropical disturbance in this region, respectively. From the relation between these

rectangles and each component scores, it is clarified that the appearance of Type IV

coincides with the westward phase of the tropical disturbance in the area near the

equator rather than the usual eastward phase. In addition to this, the minimum value

of the third components tends to precede the minimum value of the fourth components.

This means that tropical disturbance is likely to move westward when Type IV and

Cold Waves in East and Southeast Asia and Their Effect on Atmospheric Circulation at Lower Latitudes 1983/84 NOV DEC

f0

in in b, 'o

in

JAN FEB MAR 1984/85 NOV I II _IIE T-- 7 DEC JAN FEB MAR 1985/86 NOV

86E 80E 120E 150E 1

_ I I >

3

1.; 1 1 a 0

T g

Case1 III IV <WP-N>

I Case2

I —.IV <WP-N> -Case3 - —.11[

- 'I Case4

V - T Case5 <WP-N> <WP-N> <EP>

I

I —.11 <EP>

Case6

- Case7

60E DOE 120E 150E 180 150W

DEC JAN FEB MAR 1986/87 NOV I I I I 1 1r_ •A^e •

I Casa

II—'IV

ITCase9

U —.11[ T Case10 1 _--.1IIV <WP-S> <WP-N> <WP-N> <EP>

ICasell

111--41V <WP-N> DEC

JAN

_{r_}

FEB IF 7 - < ) _ _ cl-k. 1 1 .1 1

I Casie12

Case13 - 1V 1---; 4-1

COMP1 COMP2 COMP3 COMP4 COMP5 80E 90E 120E 150E 180 150W

Fig. 3 Time series of the lst-5th component scores for 850 hPa temperature an

winter and fluctuation of tropical disturbances

Rectangles with solid (dashed) line in westward (weak westward) pha

disturbance area from 100°E to 120°E.

a : the 1st component, b : the 2nd component, c : th 3rd component, d :

component, e : the 5th component, f : Time-longitude sections of 5-day nu

(Outgoing Longwave Radiation) averaged from 10°N to 10°S (hatch : below 2( (Original : Monthly Report on Climate System, Technica Note No. 32, publish,

Japan Meteorological Agency.), g : Cases of cold waves and their types

anomaly patterns and transition patterns

<WP-N>

<WP-N>

Time series of the lst-5th component scores for 850 hPa temperature anomaly in

dicate westward (weak westward) phase of the

e 3rd component, d : the 4th

sections of 5-day mean OLR 10°S (hatch : below 00 W/m2) l Note No. 32, published by the Ives and their types of cold

4.2. Case study on the effect of cold waves on low-level atmospheric circulation at lower latitudes

To clarify variation of tropical disturbance in the area near the equator associated

with cold waves, the case study on the effect of various cold waves on low-level

atmospheric circulation at lower latitudes is performed.

Cases of each cold wave are determined based on cold anomaly patterns. Since

NCEP/NCAR global reanalysis data for wind at 1,000 hPa and OLR data for cloud

activity are both 5-day mean value, a suitable cold anomaly pattern need to be

assigned to each pentad. Practically, for each day in the period of analysis, the

component that have maximum absolute value in 5 component scores is assigned. If

the component value is negative, temperature at 850 hPa on the day is identified as a

suitable cold anomaly pattern. The pentad that this component value below —1.0

with same component succeed for more than 4 days or that below — 1.5 for more than

3 days is assigned to its suitable cold anomaly pattern. Through these processes, 13

cases of cold waves that have successive cold anomaly patterns are extracted. For

each case of cold waves, tropical disturbance below 200 W/m2 of OLR and wind field

at 1,000 hPa are investigated.

As a result, based on the moving direction of tropical disturbance in the area near

the equator in Fig. 3-f, two kinds of lower-level atmospheric transition patterns at

lower latitudes are divided roughly ; one usual eastward phase pattern (EP pattern)

and another unusual westward phase patterns (WP pattern). In WP patterns, due to

the origin of the westward tropical disturbance, two kinds of patterns are further

subdivided ; one origin from the Northern Hemisphere (WP-N pattern) and another

origin from the Southern Hemisphere (WP-S pattern). Date on 13 cases of cold

waves, their transition patterns, and their composing cold anomaly patterns are given

in Fig. 3-g.

(1) WP-N pattern : Fig. 4 shows a typical case of WP-N patterns from the 73th

pentad in 1983 to the 2nd pentad in 1984. In Fig. 4-a, northeasterly by a cold wave

prevails over the South China Sea. Tropical disturbance exists mainly on the area of

10°N-10°S, 100-130°E. On the next pentad (Fig. 4-b), northeasterly over southern part

of the South China Sea further accelerates, a part of the northeasterly turns into cross

equatorial flow near 105°E (maritime continent) and intensifies equatorial westerly

from 0° to 10°S. Also near 125°E, Cross equatorial flow from the Northern Hemisphere

and strengthened equatorial westerly appears weakly. The other part of the

north-easterly passes over the Malayan Peninsula, and further strengthens easterly over the

Bay of Bengal. On the following pentad (Fig. 4-c), tropical disturbance emerges at the

remarkable easterly zone over the Bay of Bengal. This easterly zone further turn into

cross equatorial flow and equatorial westerly near 80°E over the Indian Sea, minimum

Cold Waves in East and Southeast Asia and Their Effect on Atmospheric Circulation at Lower Latitudes 67

(a)

1000hPa Wind & OLR

the 73th pentad / 1983 20°N 0 20°S t,ei, Pe 4 .4

4,\K

4,,e,c A,0 .,, A,494 *4 1, 11 A' )1 It 3/ 4— : 10 m/s 1/4 .4. 441..

rk *11

4- ynr 4444444 1,-•411,4,4 4 r 4 N. 4 r 4 4-10 yrr rfrr /r1 AK Y1-4 0- r+

1"C

3r)/ 4 .4,4. eelelebeti VA' e i v.. b. • I. 1.1b."*." 'AV* sbi 41/ KKK K W.JOZzis,*1 ^ CItra t tf ehew ie. e IC *" la la b. ,4 r 4 4 4,4 • --• 80°F 100°F the 1st pentad / 1984 120°F

(b)

140°E 4— : 10111/S 20°N 0 20°S 10.44 *44,4 * •4, r.`I kkk 4 4- 4. it It j4•-770,:a_{_} .* 4 ^"*Ps% 20 44 4-4- + Ye ‘1, -te-ek,/41 ed 4. k"4--4-qr-4.--4- kkkkr*"*% •• • •4-40

41;1

't

ep°

:ff44

4- it A A .1, * * 'Tr 9- * + + * • • 1 •••FO PE 4**-44...**Ki _t... 14. 4. le he EIC t'ehe'1,7 7 he k- V,ete ^klk Vo • k ,1,v-Ir.4141 41,1, 111 ac 444 k • • 0. 40.1. t

414

0;

80°E 100°E the 2nd pentad / 1984 4 a, 4,4 F

(c)

120°E 20°N 0 20°S 140°E : 10 m/s 4.* fakAki<kne° 410

014 No. rt

:

1\

74:10

••• 4.4f 1. • • • IL ve ^yyII '11011 `0".^C•1 49 •* 144. 4

3

^r + eie iikr4." ticket* le la • III, 4 9...srev.v.•^ ^4 v44-11V11 • 4 et.1,10 11f; 4-44• 4-4"4-4 k k 4 it 44 ^ * / • -• • 3 fryrritit f-f-f-I K.1‘ %ZS rs.'\ 11. IOC* i n4SI"*".4 1, 4A P. I. rr 4

80°F 100°E 120°E 140°E

Fig. 4 1,000 hPa wind and OLR in the case of WP-N pattern

1000hPa Wind & OLR

(a)

20°N

0

20°S

the 1 ith pentad / 1985 : 10 m/s

444 * k AKA

004'

q

741

#4.W4.-ArA eetetell4'"-

tivig,'"

4,,A11,4,4 4 * 0-.0 t ' t I

4*,_

...i. arA-lekei reek, s P, *_{, rev} reek- 11 &reek",

44;peeee

Jim' If ,

.

' ,' 444

r):CS;\

1 4 4, vie e he ~yyy444 V.%

yl4e

k v.** _St-rA vw49 4`,7404('

A

4

A

tit

10- A/C.4 4 A -4 4.1k 4 4, 1. -A 0 4 4 r 21 ZAL CA"; 4r 1,0 * 4,

(b)

80°E 20°N 100°E 0 20°S 120°E the 12th pentad / 1985 w, 4

V

k

JI ' allw 4101,711

_,10,";

* 4m4.10.,VH4* I 4, 114 x ge 4A 4), .., ,.. '... 4 b. •A t... A A A 1 4 4 KRKwwf 140°F 4— : 10 m/s

X 411

4 44 I IL•44^ 4 ‘,41V 4 4 I r' A' 4 4' -•*".,

, 'rI' irittc ,.* fJrhe It '44re4- 0-4- _-; 4-4-4

4

4

4-

4-ef;441t

44,4 4 %.•4... !I ;LW 41.# AL, irRf. ff. * wwwwwaw 11,1•111' o4.0,AA A A 0, *Val • • -4 9 4 • 4444 f I KI 444 ALZO in k.- oc,t 4 * ' .4'1411,4 A AAAAN44 rrrkr „ir 11 •-A -A-A -0 4 4.30.4 4. *OA rr1 4-4-4- 4-4 11+. 4 1 4

(c)

80°E 20°N 100°E the 13th pentad / 1985 0 120°E 44 4 ^4 4 .op 4^.

q;

*A: II"1.. t' 4 V V'44 4 140°E zigr 4'V ,44, 544044 /1% Al 444,1, 4— : 10 mis 20°S ke 5,..*M41? ^ I 1 'I ) 4 f 4 f f rrr keer 4 L4.44. ***icier A. A. rr4U.NA 4^ ".../.4

^A

A AAS

tt~i A AI 441,11A . .0 A, II L „Air At it tt K 04-416,4/0g Vitieem ieseWe ebeekel AI-4,34,11cg s4, 7'4 4.4 lk 444 4-420 44.-2--•-v •^ ^ 0 -0 r•Nik Aa. 14 4 4 +Wi 4 4 11 `A `A k_-11Ad 4, 80°F 1 00°E 120°E yf 10,41, 140°F Fig 5 1,000 hPa wind and OLR in the case of WP-S pattern

Cold Waves in East and Southeast Asia and Their Effect on Atmospheric Circulation at Lower Latitudes 69

disturbance tend to arise from appearance of equatorial westerly, the cause of this

remarkable westerly is to appear cross equatorial flow from the Northern Hemisphere

near not only maritime continent but also the Indian Sea. These processes seem to

lead to unusual westward moving of tropical disturbance in the area near the equator

in Fig. 3-f. According to Fig. 3-g, the latter period of cold waves with WP-N patterns tend to be type III or type IV of cold anomaly patterns.

(2) WP-S pattern : Fig. 5 shows the only case of WP-S pattern from the 11th

pentad to the 13th pentad in 1985. In Fig. 5-a, northeasterly associated with cold

waves predominates over the South China Sea. Cross equatorial flow exists near 80°

E, but it disappears on the next pentad (Fig. 5-b). Cross equatorial flow and

strength-ened equatorial westerly do not exist near 105°E or 125°E by this pentad. On the

following period (Fig. 5-c), a cyclonic circulation in the Southern Hemisphere centered

at 10°S, 95°E emerges suddenly, this leads to intention of equatorial westerly at 5°S.

Tropical disturbance also stretch to 90°E. Therefore, since abrupt development of

cyclonic circulation in the Southern Hemisphere occurs on the latter period of the cold

wave with this pattern, unusual westward moving of tropical disturbance in the area

near the equator also seems to appear in Fig. 3-f.

(3) EP pattern : Fig. 6 shows a typical case of 3 EP patterns, from the 13th pentad

to the 15th pentad in 1984. Northeasterly by a cold wave predominates over the South

China Sea in Fig. 6-a, but it weakens after the next pentad (Fig. 6-b, c). Cross

equatorial flow and strengthened equatorial westerly are not found. Tropical

distur-bance by OLR moves eastward as well as usual condition shown in Fig. 3-f.

There-fore the influence of cold wave on lower latitudes does not be seen in this pattern.

4.3. Characteristics of wind field at low-level atmosphere when each cold anomaly

pattern emerged

In the previous chapter, three kinds of lower-level atmospheric transition pattern

at lower latitudes are divided and emerging tendency of several cold anomaly patterns

in these transition patterns is regarded. In this chapter, composite 1,000 hPa wind

anomaly charts for each cold anomaly pattern are shown in Fig. 7, characteristics of

wind field at lower-level atmosphere when each cold anomaly pattern emerged is

confirmed as the summary of this study. Type I of cold anomaly pattern (Fig. 7-a) has

northerly wind anomaly over the East China Sea and northeasterly wind anomaly over

the South China Sea, Type II (Fig. 7-b) has westerly wind anomaly east of the East

China Sea. Flows of both types are influenced by cold waves, but cross equatorial flow

does not appear. In Type III (Fig. 7-c), on the contrary, strong northwesterly wind

anomaly by cold waves that reaches east of the Philipines exists, a part of this is

weakly connected with cross equatorial flow at 125°E. Northeasterly wind anomaly

1000hPa Wind & OLR

(a)

20°N 0 20°S the 13th pentad / 1984 4— : 10 m/s A IS*. Oak, 1p4*.0 41-74 4 a + 1r le it 444.4ekk 41

V.14r,

• 4'; ^1 -0/41

K;p44

414

kt .2 2 2 2 he

rt 4^;"

f I ) .• 141 044.. +1 r.< < 044 147/pi

fiCiri.V4..11(e

It

At

•

r

r NC

o 4. +4-C.-4 ^#•1,...K. 1--.4...K..C.N. 0 44,4W^Wq' 110 Wie'V'e +4 4. t tr± le& 4-41te /Vie* a A 4 • .4 .1T

it

(b)

80°E 20°N 0 100°E 20°S the 14th pentad / 1984 120°E 140°F 4— : 10 m/s *0 4 4**4 .,ski,b.4c 4 4 J0-0 A *4,4 4 .1 .4.4" ) ^ * -444 o 444 ^ 464 4 A • Al r 4 4 41,4-0-A.4.4 4I 42.2. *-4. 4

4\

0.1 ^ ^ 4 .0 .1 4 , • .4 f.. r..4. 4 f flICKR+71

0..

K.

N

1

\

V L.../...AN% ...°1/41`....K..SINs.. l. * 4 4. 4. 1.4 ^Lot:„.12 keitrakk krir4r 4- is 4 +44-4 4 4 ' 440 • 21 11..• Ar 4 4 4-•k t 4-so Lk'. I it • F r ^`a Alt it, 4 i7Zr•

(c)

80°E 100°E 20°N 0 20°S the 15th pentad / 1984 120°E

*

WO! 44 • ,* a 14 • rr 444 140°F 4— : 10 m/s * r 4 k ^ 4 ic 4. .1,441 •k k ^6'4 'r•Z

?4,4

or). le. 4r, 1* MA 14 41 .1 I • rvr 4 ^••i• • • • • • /1 A>vv-A 44 kpr4444-,

F 44-4-4-4-4

t

I':

4.4

A.I 4 *

;. 4 4r

44

46.

I 4,4 ‘ f 4 .. -•^4- -4 i, • ^ 041tf.4.4 k^k** In4 kk 104,40 4 pif-t44-4- • - G:4.. ,4\- 44 *2471 ear-:11fr,,,1 4" •• • ). 4 . k 4- *car rrr ,4Ack

.5,41wi,47

r r " 1 4^I," PC or t •44-1, 4La

1\1c:z

4.41 80°F 100°E 120°F 140°F

Fig. 6 1,000 hPa wind and OLR in the case of EP pattern

Cold Waves in East and Southeast Asia and Their Effect on Atmospheric Circulation at Lower Latitudes 71 •

(a)Type I

1-- :

5

m/sb

T pe

if

'— :

5

m/s

40°N5-'.11e4w,:,;:i„,:.;-••,v'.„,

P.-2,1,h.4„,,f,,..-,e40°Nrio2--;!ft.,0.-.-7—,-...!..,!:-„,-4-:.:

-1=7,,t,

_'0-.'11,:

.4WkA41..,";,,'r-4".'44'1`.

:

„,tr''S':.-.A1:Z.'4:',.:',,-!144.1011;01

...,,.

..„„

_{_.,,----iii}

_{ilto,tr".6,Vii}

_A

_A,

_{. 44}

_.4N..40.

_^4.-1-"'S

_{±ii-'V10.'4',A,4}

::*.•*4.,•7al'il"...",Y^'It!:::.4...o.Pk,,i,„,...•..„,,,,C1+It.-1,,,,o, 4-..A444.,,,,.._

°A4

1k

_{.WNW ,:44,,:pdp'-iftz:.-='}

20°N/r

Will'1,4`.pisame'—••-...,,,mit..A.,.

,,,,,10!,F,..,"

Ira::

:::20°NIr...,„ ,,

r40.4._,,,,,,.1!" .,•44,>

0.40',‘,",g•CV'....,,,(1(1,...4.,44,r.•4V *V..v.',tit

.0',,f/44,•p,....if,^.'`,.,,,,a'1,ENO.{ ',.. _{.4 4 4..,,,.., 4 . ,',A,•V *4 IA4.}

''',4r".".'.sfiewit'"7ev.r. Mu.r C 4 V.IL. , ,Ai^ 44*^+•".

^

4.v,•^•^

p.1,Ittkieje

,:hif0

k r

1>

:4,;;..:^:.1:!t.1..!15.4'.:{1:4y,AA4.41'.:...a

v. 4 1. 4 . -..i ' ..-• q ,.., 1 . . Aff I 4+...

.,,,,,,.

, A

.1

_{v 4 , e qt,„„i}

_{. 1146.4}

_{"x-.A..1",=:IN}

_,,,

_{,,,, ,,,, A}

_A

_A

_{a 4,‘}

_At:,

_{. w 4. ,,,,..k...m,}

AlYIII

401,4-4,,,,, ••^ ...•Or...,.t4 r +a , .. .4, V .1 t ^ ^ Ili 4 v V. .., , ^ 4y

.A.41.

.41,4,

a•a.,4

4

i

gii"'I144.-aaall.*...4,-.111'

+

4--/.-4,1111

. . . '1.4

f.&f4.‘.

f4,1-

A0444 40A44. .•-+

1 0 0 vu.a.,...- " ... ..., v 0144 •,,,,,, • - '

20°S 20°S

80°E 100°E 120°E 140°E 80.E 100°E 120°E 140°E

(c)TypeIa

4—

:

5 tri/s

•,

1(d)

TypeN4—:

5

m/s

40°N thli0004-•-*"lra;...41,"./t,40°NM•

iffIM..„,,,,;,--I4

4,1,1;...,

la

,,.,.,1::,.:.11,...,., :;:kr-A:-.'q7;k4,.7*-7,1:::1,44,;J-47:.:

>...-•^„,.

_{It0.*tiptitAi,I.,}

_{.L,44i.A4, 4 r.S.:"'i"vs.}

_{. , A}

_{,[. A}

_,10,\,:)

_{*0-V.$''...-".-0.""A.#4,",a,t,.47.ir....-+„..,,...„.}

,:,,,..aa.,,,,.„„,...**1.11:v.a

4,...o.r,..::

t. &it

_rrt..40;:3,

_{N,N,lt,•7:11:..,&}

,*:agi-t1,,-i.,

20°N

"' 4Elk'8411114 Iv ALNk20°N

_{,'.:1,',114111:4b::..f}

..4k..4ir•'''•'y

.r:7$

1'!0

T

.. _{; "} 0 ^ ^ ^ 4, . ,'' s• A, a ' ,.:-. 1 , , _{7,... , t 4; .."Co..., eil A. , ',. , . a• ....4 : : :} .. '0, ^ : : Z RA- hr4 a!, 1 I'414.,, • f r .444 • ^1r•-..4-4,.40 My •4,

I.,. ,4akca44,

k,,,,,-.4.4 (4)!7I

,,l'_1

f• 4-

0.."

r , 1 v.1,,

4404,.ixt..0^...••• I

fa

.

44 V farie y 4-0, _{.41;,rr ,...r,-....1,,..4.4. 4L.. V V,,,,i4^2.,:.."..4, 0° 0°}

4,

.4.4,,,,,,10

_{..!:.j,..yrd4V.„:411.tr4,40.,}

_*^:

_{::::::‘q'4Vrallignil}

1,tr,: : :2;0'',;p,*...I....k.-:rtit.:teN.",In/^..,44fr,...0_ 1Ils'iNti ..4,,,,,.-..4-*.,,:iik...7--,...

00,,,I.•

_{4,41,N4-^•••••••••-,004r,....„}

•-0•vio.,00.4',V174.":4411,

,...:,,.,.,..

,!Y,

,7:r„„"4-4

.iti.0r

.., y ,4.4-4-0.-0-044 4-4,--4-4-0-044 20S4-4. 20S

80°E 100°E 120rE 140°E 80°E 100°E 120°E 140°E

(e) T peV

4--

:

5

m/s

40°N cim54,404— ...4;,-„.4-;

4 ' 41 - 11

F',4,WF t..= *ttr,Ag• 44 !,4-*-- ,tirr"' 4 i„....s1„,?^04`sOt.$ 01,./V.,* ** • ?. i.: r.;`,.1 ; ,,, ^_{" .} • Jo, --.4,, 777,. „ 1 "" . . Art 447,..k.tiiris

V • . inp.-.1:„

,. !I -P.p..,

, . ,

... l f . 4 . , . ..I , .,.. t • • - '", ,,,,, . k ^

4 • i .4

,,,

IL'. r,•4 ypti

K.4414

4 x•

.0.0. 4 „,...,.1.1[ .„., • 'I 44.4 4,,

0° .. i•

, " , 0

4

-04-^I't,,N

intr,1--

.,,‘„,71-cimi

— , .1.4 _{. , , 4 ..., -7. } + 4.3 ' •ariV, 7,....,...11*,.'„ ,} ,..,71."-m..1„,t . socv• 37,X24,- _{t .}

. )10101. 10,,,,,,Nes. 4.

20°S .. . 41,,•,14 4 4,,,,

80°E 100°E 120°E 140°E

Fig 7 Composite 1,000 hPa wind anomaly chart for cold anomaly patterns

a . Type I, b : Type II, c . Type III, d . Type IV, e : Type V

easterly wind anomaly intensifies over the south part of the Malayan Peninsura, but

cross equatorial flow at 105°E is not clear. Concerning Type IV (Fig. 7-d), strong

northerly wind anomaly by cold waves prevails over the South China Sea, it forms

anomaly that zone from 5°S to 10°S. A part of flow near the South China Sea passes

over the Malayan Peninsula, and change into easterly or southeasterly wind anomaly

over the Bay of Bengal. These conditions seem to be associated with appearance of

easterly wave in the Bay of Bengal. Type V has only northerly wind anomaly east of

Japan, flow by cold wave is not seen over south of Japan, the East China Sea and the South China Sea.

From the results above, It may be concluded that appearance of Type IV of cold

anomaly patterns particularly give rise to connection between midlatitude and the

tropics in the wide range from the western Pacific to the Indian Sea, and same

tendency is somewhat regarded in Type III. This means that link between cold waves

and lower-level atmosphere at lower latitudes is not the simple meridional process that

strengthen local Hadley circulation shown by Chang and Lau (1980), has the zonal

structure that cause easterly wave in the Bay of Bengal.

5. Concluding remarks

Cold waves in East and Southeast Asia based on the spatial pattern of 850 hPa

temperature anomalies and their effect on lower-level atmospheric circulation at

lower latitudes are examined in this paper.

First, by applying rotated principal components analysis to daily 98-station data

in the area 100-150°E, 0°-45°N, for 4 winters (1983/84-1986/87), five cold anomaly

patterns of 850 hPa temperature are extracted ; Type I (central China), Type II

(north-eastern China), Type III (Japan), Type IV (southern China), and Type V (northern

Japan).

Secondly, cloud activity at lower latitudes in relation to these cold anomaly

patterns is examined. In particular, the appearance of Type IV coincides with the

westward phase of the tropical disturbance in the area near the equator rather than the

usual eastward phase. To clarify the effect of these cold anomaly patterns and cold

wave events on cloud activity, lower-level atmospheric circulations at lower latitudes

for 13 case of cold waves are investigated. As a result, based on the moving direction

of tropical disturbance in the area near the equator, three kinds of lower-level

atmospheric condition at lower latitudes are regarded one usual eastward phase

pattern (EP pattern) and two unusual westward phase patterns (WP-N pattern and

WP-S pattern). The features are summarized as follows ;

WP-N pattern : In this pattern, cold waves have most relation with lower-level

atmospheric circulation at lower latitudes. A part of the northeasterly turns into

cross equatorial flow near 105°E and 125°E and intensifies equatorial westerly from 0°

to 10°S. On the other hand, the other part of the northeasterly passes over the

Cold Waves in East and Southeast Asia and Their Effect on Atmospheric Circulation at Lower Latitudes 73

to cross equatorial flow and equatorial westerly near 80°E over the Indian Sea. These

process leads to unusual westward moving of tropical disturbance in the area near the

equator.

WP-S pattern : Unusual westward moving of tropical disturbance in the area near

the equator in this pattern is due to appearance of a cyclonic circulation in Southern

Hemisphere.

EP pattern : The flow derived from a part of cold wave does not reach to lower

latitudes in this pattern.

In Particular, WP-N pattern seems to be accompanied with the appearance of

Type IV of cold anomaly patterns. Same tendency is somewhat regarded in Type III.

These features is also confirmed by composite 1,000 hPa wind anomaly chart for cold

anomaly patterns.

Acknowledgements

I am indebted to Dr. R. Suzuki, Frontier Research System for Global Change, for

providing the SD data used in this study. Special thanks are due to Professor T.

Maruyama at Tokyo Gakugei University, for editing this SD data. I woud like to

express my gratitude to Dr. S. Miyazaki at Tsukuba University for critical reading of

the manuscript.

Notes

1) For the calculation of 'climatological mean temperature and standard deviation' from 16 February to 16 March, the data of March 1983 was substituted for the missing data of

March 1987.

2) In 1986/87 winter, distribution maps were made until 26 Feb 1987.

3) The average of each two days before and after the missing data was substituted for the missing data.

4) When applying rotated principal components analysis to the data, the fourth component that have the maximum area of factor loading at lower latitudes where observing station

is few was extracted. To confirm the reliability of the fourth component, 850 hPa

temperature of 8 observing stations at lower latitudes (a in Fig. 1) that have many missing

data and are not used for statistics analysis were compared with the score of the fourth

component. As a result, variations of both data tend to resemble each other. Therefore,

it is appropriate to deal with the fourth component as well as other components.

References (*in Japanese, **in Japanese with English abstract)

Barnston, A.G. and Livezey, R.E. (1987) : Classification, seasonality and persisternce of

frequency atmospheric circulation patterns. Mon. Wea. Rev., 115, 1083-1126.

the Northern Hemisphere wintertime circulation. J. Atmos. Sci., 34, 1040-1053. Chang, C.P., Erickson, J.E., and Law, K.M. (1979) : Northeastely cold surges and

rial disturbances over the winter MONEX during December 1974, Part 1 : Synoptic

aspects. Mon. Wea. Rev., 107, 812-829.

Chang, C.P. and Law, K.M. (1980) : Northeastely cold surges and near-equatorial disturbances

over the winter MONEX during December 1974, Part 2 : Planetary scale aspects. Mon.

Wea. Rev., 108, 298-312.

Japan Meteorological Agency (1989) : Monthly report on climate system : Technical Note No.

32.* Japan Meteorological Society. Tokyo.

Kawamura,R. (1988) : The interaction between winter monsoon activities in East Asia and sea

surface temperature variations over the western Pacific Ocean.** Geographical Review

of Japan, 61A, 469-484.

Madden, R.A. and Julian, P.R. (1972) : Description of global-scale circulation cells in the

tropics with a 40-50 day period. J. Atmos. Sci., 29, 1109-1123.

Nomura, C. (1999) : Classification of cold waves in East and Sontheast Asia and their relation

to cloud activity at lower latitudes.** Geographical Review of Japan, 72A, 808-828.

Tomita, T. and Yasunari, T. (1996) : Role of the northeast winter monsoon on the biennial

oscillation of the ENSO/monsoon system. J.Meteor. Soc. Japan, 74, 399-413.

Wu, M.C, and Chan, J.C.L. (1995) : Surface features of winter monsoon surges over south

China. Mon. Wea. Rev., 123, 662-680.

Yasunari, T., Nishimori, M. and Mito, T. (1998) : Trends and inter-decadal variations of the

surface and lower-tropospheric temperature in the Northern Hemisphere from 1964 to

93. J. Meteor. Soc. Japan, 76, 517-531.