annual cycle of southeast asia—maritime continent … cycle of southeast asia—maritime continent...

Annual Cycle of Southeast Asia—Maritime Continent Rainfall and the AsymmetricMonsoon Transition

C.-P. CHANG AND ZHUO WANG

Department of Meteorology, Naval Postgraduate School, Monterey, California

JOHN MCBRIDE

Bureau of Meteorology Research Centre, Melbourne, Australia

CHING-HWANG LIU

Department of Atmospheric Sciences, Chinese Culture University, Taipei, Taiwan

(Manuscript received 21 October 2003, in final form 9 June 2004)

ABSTRACT

In general, the Bay of Bengal, Indochina Peninsula, and Philippines are in the Asian summer monsoonregime while the Maritime Continent experiences a wet monsoon during boreal winter and a dry seasonduring boreal summer. However, the complex distribution of land, sea, and terrain results in significant localvariations of the annual cycle. This work uses historical station rainfall data to classify the annual cycles ofrainfall over land areas, the TRMM rainfall measurements to identify the monsoon regimes of the fourseasons in all of Southeast Asia, and the QuikSCAT winds to study the causes of the variations.

The annual cycle is dominated largely by interactions between the complex terrain and a simple annualreversal of the surface monsoonal winds throughout all monsoon regions from the Indian Ocean to theSouth China Sea and the equatorial western Pacific. The semiannual cycle is comparable in magnitude tothe annual cycle over parts of the equatorial landmasses, but only a very small region reflects the twice-yearly crossing of the sun. Most of the semiannual cycle appears to be due to the influence of both thesummer and the winter monsoon in the western part of the Maritime Continent where the annual cyclemaximum occurs in fall. Analysis of the TRMM data reveals a structure whereby the boreal summer andwinter monsoon rainfall regimes intertwine across the equator and both are strongly affected by thewind–terrain interaction. In particular, the boreal winter regime extends far northward along the easternflanks of the major island groups and landmasses.

A hypothesis is presented to explain the asymmetric seasonal march in which the maximum convectionfollows a gradual southeastward progression path from the Asian summer monsoon to the Asian wintermonsoon but experiences a sudden transition in the reverse. The hypothesis is based on the redistributionof mass between land and ocean areas during spring and fall that results from different land–ocean thermalmemories. This mass redistribution between the two transition seasons produces sea level patterns leadingto asymmetric wind–terrain interactions throughout the region, and a low-level divergence asymmetry in theregion that promotes the southward march of maximum convection during boreal fall but opposes thenorthward march during boreal spring.

1. Introduction

The region of Southeast Asian landmasses, which in-cludes Indochina, the Malay Peninsula, and the Mari-time Continent, is situated between the Asian (Indian)summer monsoon and the Asian winter (Australiansummer) monsoon in both space and time. This area

forms the “land bridge” along which maximum convec-tion marches gradually from the Asian summer mon-soon to the Asian winter monsoon during boreal fall(e.g., Lau and Chan 1983; Meehl 1987; Yasunari 1991;Matsumoto 1992; Matsumoto and Murakami 2002;Hung et al. 2004). The seasonal march is not symmetric.During boreal spring the convection tends to stay nearthe equator as if it were blocked from moving north-ward. It stays until the reversed meridional thermal gra-dient is established when it jumps over the northernequatorial belt to mark the onset of the Asian summermonsoon.

Indochina is often considered as part of the Asian

Corresponding author address: Dr. C.-P. Chang, Department ofMeteorology, Naval Postgraduate School, Code MR/Cp,Monterey, CA 93943.E-mail: [email protected]

15 JANUARY 2005 C H A N G E T A L . 287

© 2005 American Meteorological Society

JCLI3257

summer monsoon region while rainfall over most loca-tions in the Maritime Continent tends to reach maxi-mum during the boreal winter. This wet season is oftenrelated to the Australian summer monsoon (e.g.,McBride 1987) due to the proximity of the two regions.However, significant geographical variations of the sea-sonal march have been recognized since Braak (1921–29, see Ramage 1971). A main reason for these varia-tions is the complex terrain due to islands of differentsizes interspersed among the surrounding seas. Thevariations motivated Wyrtki (1956) to divide the searegion into 12 climate subregions that were presentedin Ramage (1971, Fig. 5.9). Other studies on the Indo-nesian rainfall climatology include Schmidt and Fergu-son (1951), Sukanto (1969), and McBride (1998). Morerecently, Hamada et al. (2002) documented the sea-sonal variations of rainfall at 46 Indonesian stationsduring a 30-yr period (1961–90) and classified the sta-tions objectively into five climatological types depend-ing on the phase and relative amplitudes of the annualand semiannual cycles of rainfall. Another approachwas used by Aldrian and Susanto (2003) and Aldrian etal. (2003), who used spatial correlation among stationrainfalls to divide the Maritime Continent into threesubregions. All these studies showed complex geo-graphical variations of the annual cycle of rainfall in theregion.

The large variation in local rainfall over small dis-tances indicates that high-resolution wind data are re-quired to analyze the wind–terrain interaction of theMaritime Continent. The lack of historical high-resolution and consistent wind observations made suchan analysis difficult. Furthermore, since the rainfalldata used in the previous studies are all from land sta-tion reports and often concentrate only in Indonesia, agap exists in our knowledge of the distribution of theannual cycle of rainfall in the inner and surroundingseas of the Maritime Continent as well as regions northof Indonesia.

Recently available satellite-based observations makeit possible to alleviate both problems. In this study,we analyze the annual cycle of rainfall by using twotypes of satellite data together with historical stationrainfall datasets. The Tropical Rainfall MeasuringMission (TRMM) precipitation radar data, availablesince late 1997, are used to enhance the analysis ofthe geographical variations of the annual cycle of rain-fall in a domain encompassing most of Southeast Asiaand the Maritime Continent. The National Aeronauticsand Space Administration’s Quick Scatterometer(QuikSCAT) winds, available since July 1999 at the seasurface, are used to help examination the relationshipbetween the monsoon–terrain interaction and the geo-graphic variations. The result of the analysis will beused to address the cause of the asymmetric seasonalmarch of maximum convection during the transitionalseasons.

2. Data

Two datasets of monthly station rainfall are used inthis study. The first is an extension of the Indonesianrainfall dataset prepared by Kirono et al. (1999) andused in Haylock and McBride (2001). This dataset,hereinafter referred to as the INDO dataset, containsrainfall during 1950–97 at 63 Indonesian stations. Thesecond is the Association of Southeast Asian Nations(ASEAN) Climatic Atlas Project (ACAP) data, fur-nished by the Malaysian Meteorological Service. Thisdataset covers 935 rainfall stations in all the membernations of ASEAN during the data collection phase.The beginning dates of the stations vary, with the ear-liest date in each country as follows: Singapore, 1875;Malaysia, 1876; Indonesia, 1879; Philippines, 1902; andThailand, 1911. The majority of the stations start from1951 or earlier, and almost all start from 1958 or earlier.All stations end in 1975. The two sets have overlappingobservations over Indonesia between 1950 and 1975.

The TRMM microwave precipitation data (Simpsonet al. 1996) became available November 1997. In thisstudy, we use the data of January 1998–December 2002(data for 7–24 August 2001 are missing). The originaldata at resolution of 4 km � 4 km are smoothed into agrid of 0.5° � 0.5° using the two-step filter of Leise(1982).

The QuikSCAT scatterometer winds (Liu 2002)cover the period January 1999–December 2002. Theseare scatterometer winds at the sea surface at 25 km �25 km resolution. On a typical day the available datacovers 75% of the equator and increased percentage ofarea away from the equator. These data are used toproduce monthly mean winds at the sea surface.

3. Analysis based on station rainfall reports

a. Monthly mean rainfalls

As background, the large influence of the annualcycle can be seen clearly in Fig. 1, which includesmonthly mean rainfall from both the INDO and ACAPdata for January, April, July, and October (Chang et al.2004). For this figure all station rainfall data havebeen interpolated to a 0.5° � 0.5° grid using the meth-odology of Cressman (1959). This figure should be in-terpreted with reference to the data locations or sta-tions in Fig. 2. The stations north of 10°N experience awet season during July and October and those south of5°S have their wet season during January and April.Rainfall amounts are high throughout the region withindividual monthly totals on the order of 300 to 500mm. Most of the region experiences a distinct dry sea-son at some time of year, with the exception being partsof Borneo and New Guinea, which have high rainfallyear-round. (Place names are shown in Fig. 2.) Theseasonal shift is such that in January the center of massof the heavy rainfall is south of the equator while in

288 J O U R N A L O F C L I M A T E VOLUME 18

July the rainfall is northward of 10°N. This is associatedwith the seasonal migration of the intertropical conver-gence zone (ITCZ) in the region, as has been docu-mented by Johnson and Houze (1987), Waliser andGautier (1993), McBride et al. (1995), and Qian andLee (2000).

The other major feature in Fig. 1 is the existence ofstrong rainfall gradients existing at all times of the year;for example, notice the east–west gradients of rainfallacross the Philippines in January and across the MalayPeninsula in April. As will be discussed below, thesepatterns result largely from the interaction between thehigh topography in the region and the moisture-bearinglow-level monsoon flow, whereby rainfall is enhancedwhen the flow is lifted on the upstream side of a moun-tainous island or peninsula. Conversely the flow on thedownstream side experiences a rainfall minimum asso-ciated with a lee or rain-shadow effect.

b. Annual cycles

The annual cycle and semiannual cycle modes are thefirst two harmonics of the climatologically averaged an-nual rainfall variation at each location. Because onlymonthly mean station rainfall data are available, each

time series has only 12 data points. Figure 2 shows theannual cycle mode at land stations, with the amplituderepresented by the length of each arrow and the phaseshown as a 12-month clock with a northward arrowindicating maximum rainfall in January. The arrow ro-tates clockwise with eastward, southward, and west-ward arrows indicating April, July, and October, re-spectively. This figure is examined along with the meanQuikSCAT winds in January and July shown in Fig. 3,which also includes the distribution of topography.

North of the Maritime Continent, data are availablein two Southeast Asian regions, Indochina (Thailandstations) and the Philippines (Fig. 2). In Indochina, theeffect of the Asian summer monsoon is clearly indi-cated with most Thailand stations showing maximumrainfall around June. Over the Philippines, the Asiansummer monsoon rainfall is defined at most stations inthe south and west where the rainfall maximum occursaround July. These are stations on the windward sideduring the southwest monsoon (Fig. 3). On the otherhand, most northern and eastern stations show maxi-mum rainfall in late summer or early fall. There are twofactors influencing this. A major factor for this varia-tion is that during the peak of the southwest monsoonin July these stations are on the leeward side of the high

FIG. 1. Monthly mean rainfall and topography for (a) Jan, (b) Apr, (c) Jul, and (d) Oct (from Chang et al. 2004).


Fig 1 live 4/C

topography on the islands. Because of the shelteringeffect of the topography, they do not experience a wetmonsoon at this time of year. However, during Septem-ber through December when the southwest monsoonbegins its retreat, northeasterly flow occurs and thesesame stations are now on the windward side. At thattime of year (fall–summer) there is still upward motionassociated with the retreating ITCZ; so there is a latewet season associated with terrain-lifted rainfall. Thesecond factor is that the number of typhoons peaks inSeptember; and tropical cyclones are considered to bemajor contributors to rainfall in the northeastern Phil-ippines (Coronas 1912; Masó 1914; Flores and Balagot1969). The relative contributions of these two factorsare not known without further study.

The southern Philippine island of Mindanao is con-sidered part of the Maritime Continent, based on Ra-mage’s (1968) definition. Here, the seasonal cycle overmost of the region is characterized by the summer mon-soon rainfall. The exception is in its northeastern cor-ner where maximum rainfall occurs around Novemberand December, again as a result of the prevailing north-

easterly onshore winds (Fig. 3) during the northernwinter monsoon.

In Fig. 2 the distribution of the annual cycle over thePhilippines appears as a counterclockwise pattern inthe phase diagram, associated with a progression froma November–December maximum on the southeasternside, an August–September maximum on the northeast,and a July maximum along the western coastline. Incontrast, over the western Maritime Continent fromIndochina to the Malay Peninsula, phase of the annualcycle moves in a clockwise manner reflecting the sea-sonal march of the deep convection that follows the sun(Lau and Chan 1983). Thus, the annual cycle maximumoccurs mostly during northern fall in the Malay Penin-sula and northern Sumatra and changes to around De-cember in southern Sumatra. Over the rest of Indonesiathe boreal winter maximum in the annual cycle occursin most places, especially in southern Borneo, Java, andother islands near 10°S.

As pointed out by previous investigators (e.g., Ra-mage 1971; Hamada et al. 2002), the majority of thelocations south of the equator have monsoonal rainfall

FIG. 2. The annual cycle mode at rainfall stations. The phase of the cycle is shown as a 12-month clock with anorthward arrow indicating maximum rainfall in Jan. The arrows rotate clockwise with eastward, southward, andwestward arrows indicating Apr, Jul, and Oct, respectively. The length of the arrow defines the amplitude of thecycle. This figure also serves to show the locations of the observation stations for the monthly rainfall datasets usedin this study.


with a wet boreal winter and a dry boreal summer, al-though there are distinct exceptions. The most note-worthy exception is over the central Celebes and theMolucca area, where the maximum in the annual cycle ofrainfall may occur during boreal spring and early borealsummer (Fig. 2). This has been noted by many previousinvestigators (e.g., Wyrtki 1956; Ramage 1971; Hamada etal. 2002; Aldrian and Susanto 2003; Aldrian et al. 2003).

This “out of phase” effect is restricted to the south-ern coastline of the Molucca region whereas the north-ern coastline has its annual cycle maximum in January–February, which is closer to the phase of the larger-scale region (Fig. 2). The topography of the islands issuch that a mountainous ridge lies along the islands inan east–west direction. The east and south sides of theislands in the Molucca area are sheltered from thenortheast monsoon winds during boreal winter, butthey face the southeasterly monsoon winds that pro-duce convergence against the terrain during borealsummer (Fig. 3). Thus, the cause of this out-of-phasebehavior with the majority of the Maritime Continentappears to be the result of sheltering by the mountains

during the boreal winter and orographic uplift duringthe boreal summer. The large variation in phasing ofthe annual cycle across the islands presents an excellentexample of the importance of local and orographic ef-fects at low latitudes, as was discussed by Riehl in hisclassical text on tropical meteorology (Riehl 1954).

For reasons that are not entirely obvious, the up-wind–lee effect seems not to apply along the Indone-sian archipelago (Java and points eastward on Fig. 2).Part of the reason is that at this more southern locationthe boreal winter monsoon flow has a more dominantwesterly component, so the flow is parallel to the long-island topography. This is not the entire explanation,however, as the island of Timor, for example, has anorientation whereby there should be a rain-shadow ef-fect for some stations; but according to the data in Fig.2, this is not observed.

c. Areas without strong annual cycles

The discussion of Figs. 1 and 3 has brought out thevery large role played by the annual cycle in Maritime

FIG. 3. Mean QuikSCAT wind for Jan (black) and Jul (red), and topography (m).


Fig 3 live 4/C

Continent monthly rainfall. However, there are loca-tions where the annual cycle is less dominant. Theselocations are identified in Fig. 4, in which the annualcycle amplitude and phase are plotted (black arrows)against the maximum monthly rainfall (gray arrows) forthe objectively analyzed gridded-rainfall field, with theresults plotted every other grid point (1° � 1° resolu-tion). The maximum monthly rainfall arrows are scaledto 1:4 of the annual cycle arrows. The locations wherethe annual cycle is less important are delineated by ablack dot, which is plotted if the amplitude of the an-nual cycle is no more than 40% of the maximum rainfalland if the phase difference is at least 2 months.

The locations where the annual cycle plays a lesserrole are the eastern and central Philippines, northeastBorneo, southern Thailand, and parts of the northwestend of New Guinea. Inspection of the plots for theindividual months indicates the reasons for this geo-graphic variation. In southern Thailand the wet seasonextends through the boreal summer (as indicated by theJune orientation of the annual cycle arrow); but thisculminates in a period of intense rainfall at the end ofthe wet season around September (as indicated by theorientation of the “maximum month” arrow). In con-

trast, in northeast Borneo the reason for the less promi-nent annual cycle is that the rainfall tends to be welldistributed over all months of the year, so the magni-tude of the annual cycle arrow is small compared to thatof the maximum month.

d. Areas with significant semiannual cycles

Following the approach of Hamada et al. (2002), wealso examined the distribution of the semiannual modeof station rainfall (Fig. 5). The semiannual mode ismore likely to be meaningful if it is comparable in mag-nitude to that of the annual cycle. In Fig. 5, the semi-annual cycle is plotted at grid points where its ampli-tude is at least 80% of the annual cycle amplitude. Theannual cycle at all grid points is also plotted for com-parison. Here for the semiannual cycle mode, a vertical(north–south) bar indicates rainfall maximum in winterand summer and a horizontal (east–west) bar indicatesrainfall maximum in spring and fall. Because the semi-annual cycle has two maxima, each bar is centered atthe grid point and points in the two (opposite) direc-tions. The length of the bar from the center to each ofthe two end points has the same scale as the length ofthe vectors in the annual cycle mode. Thus, the entire

FIG. 4. The annual cycle mode (black) and the month and amount of the maximum monthly rainfall (gray) forthe objectively analyzed gridded rainfall. Black dots indicate grid points where the amplitude of the annual cyclemode is no more than 40% of the maximum rainfall and the phase difference is at least 2 months.


length of the bar representing the semiannual mode istwice the length of an annual cycle vector with the sameamplitude.

In general, the semiannual cycle tends to be concen-trated over three areas within the equatorial zone of3°S–7°N. The first is west of 110°E, which is the tran-sitional zone between summer and winter monsoonwith the annual mode peaking in boreal fall. North ofthe equator in western Malay Peninsula–northernSumatra, the semiannual indicators orient mostly June/December versus the annual mode that peaks aroundSeptember. Farther southeast in the narrow equatorialsea region between Sumatra and Borneo, the semian-nual indicators have a February/August orientationversus the annual mode peak of November. Thesesemiannual mode indicators are normal to the annualmode vectors and suggest the influence of both borealsummer and boreal winter rainfall.

The second main area is in northeastern Borneo be-tween 114°E and 120°E. The annual modes are small

and the semiannual mode indicators again have a bo-real summer/winter orientation, with a gradual clock-wise shift from north to south. The orientation is Janu-ary–February/July–August at about 3°N–4°N, March/September near 1°N–2°N, and April/October at theequator.

The few equatorial grid points define the third mainarea farther to the east, which is the equatorial zonebetween 120°E and 130°E. This region covers the vi-cinity of the Molucca Sea and Molucca Passage, includ-ing northern Celebes and Halmahera. Here the semi-annual mode indicators are oriented horizontally(March–April/September–October), suggesting thepossible effect of the twice-a-year crossing of the sunover the equator. The double rainfall peak associatedwith the twice overhead crossing of the sun is some-times considered as the classical mode for the annualcycle of tropical rainfall. One of the most significantaspects of Fig. 5 is how rarely the double-rainfall peakoccurs in the Maritime Continent, as it is effectively

FIG. 5. The semiannual cycle mode (dark bar), overlapped with the annual cycle mode, for the objectively analyzedgridded rainfall. Data are plotted if the amplitude of the semiannual cycle is at least 80% of that of the annual cycle. Eachbar is centered at the grid point and pointing in the two (opposite) directions of the semiannual cycle peaks. The entirelength of the bar is twice the length of an annual cycle vector with the same amplitude. See text for details. Example ofrainfall time series (mm day�1) at grid points where the semiannual cycle is important (5°N, 100°E), the annual cycle isimportant (8°S, 110°E), and neither is important (3°S, 133.5°E) are shown in the upper-right insert.


restricted to those few equatorial grid points where thesemiannual mode indicators orient horizontally.

4. Monsoon regimes and transitions

The TRMM data provide complete spatial coveragefor the domain, but the available data period is only 5yr and the data are derived from remotely sensed vari-ables. This very small sample size makes it difficult forany composite values to attain a high level of signifi-cance. The confidence in the composite results willhave to depend on whether the results can be consis-tently explained by physical reasoning over many dif-ferent locations. To assess the quality of the data, it isdesirable to compare the TRMM data with the stationrainfall data. However, a direct comparison is not pos-sible because the two datasets do not overlap. Since thepurpose of this study is analysis of the mean annualcycle, our interest is whether the annual cycle derivedfrom the TRMM dataset is consistent with that of thelong-term station rainfall. Therefore, the comparison is

done by first calculating the annual cycle mode of theTRMM Precipitation Radar (PR) data and then com-puting the difference (Fig. 6) between it and the annualcycle mode of the analyzed rainfall data (Fig. 4) at allland grid points. In Fig. 6 the vector direction shows thephase difference in a 12-month clock. A northward-pointing arrow means the two annual cycles are exactlyin phase, and a westward (eastward)-pointing arrowmeans the TRMM PR annual cycle mode leads (lags)the analyzed rainfall mode by 3 months. Gray arrowsmean the amplitude of TRMM PR mode is larger andblack arrows mean it is smaller. In general, the phaseangles are quite small (within two months). The ampli-tude difference is sometimes large, particularly overthe Indochina Peninsula north of 10°N. Since the twodata periods are not overlapping, it is difficult to con-clude that any difference represents errors of theTRMM mode. On the other hand, the generally smallphase differences suggest that the TRMM data are use-ful for the study of the rainfall annual cycle over theregion.

FIG. 6. Comparisons of annual cycles deduced from TRMM PR vs analyzed rainfall. The direction of vector showsthe phase difference in a 12-month clock. A northward-pointing arrow means the two annual cycles are exactly inphase, and a westward (eastward)-pointing arrow means the TRMM PR annual cycle mode leads (lags) the analyzedrainfall mode by 3 months. Gray arrows mean the amplitude of TRMM PR mode is larger and black arrows meanit is smaller.


a. Regimes of boreal summer and boreal wintermonsoon rainfall

Figure 7 shows the difference in TRMM PR databetween December–February (DJF) and June–August(JJA). Here the yellow–red colors show areas of morerainfall in JJA and the green–purple colors show areasof more rainfall in DJF. Thus, this chart can be viewedas the demarcation of the boreal summer and wintermonsoon rainfalls. Because of the limited duration (15months for each season) and the narrow individualTRMM data swaths, some small-scale features may beartificial, but the general patterns should reflect themean seasonal differences. Also plotted on Fig. 7 arethe DJF minus JJA QuikSCAT winds. From the rain-fall difference, the boreal summer rainfall regime domi-nates north of the equator, and the boreal winter rain-fall regime dominates south of the equator. The tworegimes become mixed near the equator, but the extentof mixing is not symmetric. The boreal winter regimeextends far northward into the boreal summer regime,whereas the southward extension of the boreal summerregime south of the equator is much more limited. OverSoutheast Asia, the intrusions beyond 5°N occur in thefollowing areas: east of the Philippines, northeast andnorthwest of Borneo in the South China Sea, east ofVietnam, eastern coast of Malay Peninsula, and northof Sumatra. In most of these areas the high boreal win-ter rainfall is due to the onshore northeasterly wintermonsoon winds from the northwest Pacific and the

South China Sea. These northeasterly monsoon windsare stronger than the southeasterly winds because ofthe intense baroclinicity over the cold Asian continentin boreal winter; a similar counterpart does not exist inthe Southern Hemisphere in boreal summer. There arealso very few coastal areas between 5°S–10°S that facethe prevailing seasonal wind as is the case in the north-ern Tropics.

The fact that most of the monsoon rainfall is a resultof wind–terrain interactions is also readily apparent inthe summer monsoon regime (warm colors in Fig. 7).The heaviest rainfall occurs on the summertime wind-ward side of the terrain, which includes the northeast-ern Bay of Bengal, the eastern Gulf of Thailand, west ofthe north–south-oriented mountain ridge along Viet-nam, and the northern South China Sea (off the west-ern coast of the northern Philippines and the southerncoast of China’s Guangdong Province). It is worth not-ing that while heavy summer rainfall is often associatedwith tropical cyclones such as monsoon depressions inthe Bay of Bengal, the heaviest rainfall is not along theaveraged track of the cyclones, which extends north-westward to the eastern coast of India (Neumann1993), but is near the western coast of Burma wheremuch fewer cyclones have made landfall. In fact, therelatively lower amount of monsoon rainfall over theIndia subcontinent compared to the Bay of Bengal maybe partly due to the wind–terrain interaction in theeastern Arabian Sea where heavy rainfall accumulates

FIG. 7. Differences of TRMM PR rainfall and QuikSCAT winds between boreal winter and boreal summer (DJF minusJJA). Warm colors are the boreal summer monsoon regime and cool colors are the boreal winter monsoon regime. Seetext for details.


Fig 7 live 4/C

off the west coast of India (not shown). Another ex-ample of the importance of the wind–terrain interactionin the summer monsoon can be seen southwest of theBay of Bengal, where Fig. 7 shows another extension ofthe boreal winter monsoon regime to the northernTropics. This winter regime just east of Sri Lanka be-tween 5°N and 10°� is in the middle of the easternIndian Ocean where the DJF rainfall does not stand outas being particularly high in the boreal winter TRMMrainfall distribution, but rather is a minimum JJA rain-fall area in the eastern Indian Ocean (not shown). Thisminimum is apparently due to the sheltering of thesouthwest monsoon by the island of Sri Lanka. As aresult, the DJF � JJA difference (Fig. 7) indicates thisregion to be a winter monsoon regime.

The intrusion of the boreal winter rainfall regimeinto the southern South China Sea northwest of Borneodoes not occur as a result of direct onshore winds,where the northeast monsoon is parallel to the coast-line. This is the vicinity of the low-level quasi-stationaryBorneo vortices that are associated with the heating ofthe island and can develop any time of the year. Duringboreal winter, the northeast cold surge winds increaseperiodically for periods of one to several weeks andenhance the low-level cyclonic shear vorticity off thenorthwest coast of Borneo. As a result, the Borneovortices are particularly active during boreal winter(e.g., Johnson and Houze 1987; Chang et al. 2003, 2004,2005). Due to the increased frequency of the Borneovortex, the deep convection and heavy rainfall fre-quently extend offshore several hundred kilometersinto the South China Sea during boreal winter, butmuch less so during boreal summer.

Over the Maritime Continent region the only signifi-cant intrusion of the boreal summer rainfall regime intothe southern Tropics beyond 5°S is on the south side ofwestern New Guinea between 135° and 138°E. Duringboth winter and summer, this area faces some onshorewinds (Fig. 3). The southeasterly winds during borealsummer are stronger and have a larger angle with thesouthern coastline; the westerly winds during borealwinter are weaker and nearly parallel to the coastline.

Within the equatorial belt 5°S–5°N, the mixing of theboreal winter and boreal summer rain regimes is almostentirely the result of the sheltering by the mountainsand the direct onshore winds onto the complex islandgeography. Here, the TRMM data available over bothland and water show this wind–terrain interactionwithin the Molucca region more completely than thatidentified by the station rainfall data (Fig. 2).

b. Regimes of the transitional seasons and theasymmetric seasonal march

In this section we present the monsoon regimes dur-ing the transition seasons, boreal spring [March–April–May (MAM)] and boreal fall [September–October–November (SON)]. Figure 8 is a summarized view ofthe monsoon rainfall in these seasons. Two steps are

taken to create this chart. In the first step, at each gridpoint where the MAM rainfall (green) is the maximumamong the four seasons, the difference between theMAM rainfall and the higher of the summer or winterrainfall is computed. The same procedure is followed atpoints where the SON rainfall (yellow) is the maximumof all four seasons. The resultant values are then plottedtogether in Fig. 8. The QuikSCAT winds for both sea-sons are also plotted, with the MAM wind in black andthe SON wind in red.

In Fig. 8 the SON monsoon rainfall dominates areasnorth of the equator, particularly the western side ofthe domain. The MAM monsoon rainfall dominates ar-eas south of the equator, particularly the eastern side ofthe domain. Within the equatorial belt 10°S–10°N, thearea west of 114°E, or about the longitude of centralBorneo, is mainly in the SON regime while the area tothe east is mainly in the MAM regime. Away from theequator, significant Northern Hemisphere SON mon-soon rainfall can also be found in the South China Seaand east of the Philippines. Similarly significant mon-soon rainfall does not exist in the Southern HemisphereMAM regime (not shown).

As has been noted by previous investigators (e.g.,Lau and Chan 1983; Meehl 1987; Yasunari 1991; Mat-sumoto and Murakami 2000, 2002; Hung and Yanai2004; Hung et al. 2004), the distribution of the transi-tion season convection indicates that the advance of theannual cycle in space is not symmetric. The Asian sum-mer monsoon has its maximum convection located inSouth Asia around India and the Bay of Bengal. Duringboreal fall, the maximum convection moves southeast-ward through the eastern Indian Ocean and the SouthChina Sea along a track that roughly follows the South-east Asia land bridge, to reach southern Indonesia andnorthern and eastern Australia during boreal winter.This pattern is reflected in the annual cycle diagram(Fig. 2). However, during boreal spring the monsoonconvection does not return to the Northern Hemi-sphere along the same path. The maximum rainfall re-mains mostly south of the equator. In the western sec-tor of the domain in the eastern Indian Ocean, it occursto the south of 10°S and northwest of Australia, whichis farther south than the boreal winter convection belt.In the eastern sector of the domain it moves slightlynorthward in the New Guinea region. So there is nonorthwestward progression to retrace the path of theAsian summer to Asian winter monsoon progression.

This spring–fall convection asymmetry is apparentlyrelated to the interaction between winds and the terrainof Southeast Asia. In the northern South China Sea, themean seasonal wind in boreal fall is northeasterly andits speed is stronger than the weak easterly during bo-real spring. This contrast was noticed by Matsumotoand Murakami (2000) who related the cold surges dur-ing boreal fall to the enhanced convection in theBorneo area. The seasonal winds lead to strong con-


vection off the coast of Vietnam in boreal fall but not inspring (Fig. 8). Similarly, strong convection occurs eastof the Philippines only in boreal fall due to strongernortheasterly onshore winds. The convection in theseareas is much stronger than both boreal summer andwinter, even though during boreal winter the northeast-erly winds are stronger. This is because during winterthe colder and drier air and the lower sea surface tem-perature (SST) make deep convection less likely to de-velop north of 10°N, so the strong winter cold surgesactually produce drying conditions in northern andmiddle South China Sea (Chang et al. 2004, 2005).

This fall�spring difference in winds may be traced tothe difference in the sea level pressure (SLP) distribu-tion, which tends to have opposite signs between landand ocean areas because of differences in surface tem-peratures (e.g., Van den Dool and Saha 1993). Thelong-term mean SON-minus-MAM SLP during 1968–96 computed from the National Centers for Environ-mental Prediction–National Center for AtmosphericResearch (NCEP–NCAR) reanalysis is shown in Fig. 9.Compared to boreal spring, SLP in the Northern Hemi-sphere during boreal fall tends to be higher over theland and lower over the ocean, and the reverse is truein the Southern Hemisphere. This pattern is seenthroughout most of the global domain except for themidlatitude storm tracks over oceanic areas, whose ef-

fect in reversing the difference sign can be confirmed bya map of spring�fall 500-hPa height difference (notshown). In this map the most conspicuous difference isthe lower heights during the local fall season due to theactive storms near the midlatitude jet streaks.

The SLP difference shown in Fig. 9 may be caused bytwo effects. Over oceans, the delayed thermal responsewill give a relatively warm SST in fall, which contributesto lower SLP, and a relatively cool SST in spring, whichcontributes to higher SLP. Over land areas, the thermalmemory is very short; thus an asymmetric land–sea SLPcontrast will occur between fall and spring. This asym-metry is further enhanced by the timing of the equinox.Because it occurs at approximately the 22d day of therespective 3-month seasons, the land surface tempera-ture tends to be cooler and SLP higher during fall thanduring spring. In Fig. 9, this land SLP response in lowand middle latitudes is largest in East Asia, with themaximum SLP difference centered in the northern Chi-na–southern Mongolia Plain. Another interesting fea-ture in Fig. 9 is that both polar regions have the samesign because the Arctic is water and Antarctic is land.1

1 In the southern high latitudes, the semiannual oscillation ofSLP surrounding Antarctica (e.g., van Loon 1967) may also playa role in the lower SLP in the SON minus MAM difference.

FIG. 8. Monsoon regimes during transition seasons deduced from TRMM PR rainfall. A grid point is identified if therainfall during one of the two transition seasons is the maximum in the annual cycle, and the value plotted is the differencebetween this transition-season rainfall and the boreal winter and or boreal summer, whichever is highest. Warm colors arethe boreal fall monsoon regime and cool colors are the boreal spring monsoon regime. There are large areas of zero valueindicating that maximum rainfall occurs outside of spring and fall. These areas are plotted with the lightest warm color.The difference of QuikSCAT winds between the two transition seasons (SON minus MAM) is plotted for the entiredomain. See text for details.


Fig 8 live 4/C

An enlargement of the Southeast Asia and MaritimeContinent domain shown in the insert to Fig. 9 is dis-played in Fig. 10, with schematic sea level winds basedon the SLP gradient depicted in key areas for discus-sion. In these areas the pressure gradient is less suscep-tible to the errors introduced in the conversion of pres-sure from land surface to sea level. Schematic windsbased on broader-scale pressure patterns instead of ac-tual spring�fall wind differences make the discussionsomewhat qualitative rather than quantitative, but thepurpose here is to derive a hypothesis for the source ofthe asymmetric seasonal march that is not blurred bythe results of the asymmetric march. The actual winddifferences already include the effect of the asymmetrywithin the tropical belt and it would be harder to sortout the cause–effect. The large SLP difference overEast Asia leads to the stronger SON northeasterlywinds in the northern South China Sea and northwest-ern Pacific (area A in Fig. 10), which causes the deepconvection east of Vietnam and the Philippines in bo-real fall. In the southern South China Sea, the SLPgradient favors cyclonic flow and therefore deep con-vection in SON (area B in Fig. 10). This widespreadconvection in the southern South China Sea during bo-real fall is not very obvious in Fig. 8 because at manypoints convection in boreal summer is even stronger.The convection is further enhanced by the interactionof the wind with terrain on the east coast of Sumatraand west coast of Borneo.

The enhanced convection in the eastern equatorialIndian Ocean may also be explained, at least partly, bythe spring�fall SLP difference. Here the Bay of Bengaland the northern Indian Ocean have higher SLPs inboreal fall than in boreal spring, which suggests thatatmosphere–ocean interaction is involved to increasethe SST faster during boreal spring. One possibility isthat the initially anticyclonic flow with weak winds

FIG. 10. Differences of sea level pressure between boreal falland boreal spring (SON minus MAM), for the insert region in Fig.9. Negative isobars are dotted and the zero line is dashed. Unit:hPa. Schematics of sea level wind differences based on the differ-ences in the sea level pressure pattern are indicated. The elliptic-shaped area indicates preferred belt of convergence in fall anddivergence in spring. See text for details.

FIG. 9. Differences of sea level pressure between boreal fall and boreal spring (SON minus MAM, warm colors: positive,cool colors: negative). Unit: hPa. Higher values over land areas at high latitudes are truncated because they may beaffected artificially by the surface to sea level conversion. The rectangular box is an insert region covering the domain ofFig. 10. Areas where values do not meet the 5% significance test are indicated by black dots.


Fig 9 live 4/C

cause less evaporation and more solar heating at the seasurface and downwelling in the upper ocean, so asspring progresses the SST becomes higher and SLP be-comes lower than during fall. However, the land–searedistribution of mass still contributes to a lower SLPover the Bay of Bengal in boreal fall compared to sur-rounding land areas. The resulting difference in pres-sure gradient during boreal fall gives rise to cyclonicflow in the Bay of Bengal and favors increased cross-equatorial flow from the southern Indian Ocean.

The convection in and around the middle and south-ern South China Sea in boreal fall helps to inducesouthwesterly winds west of Sumatra (area C in Fig.10). These southwesterly winds are enhanced by thetendency for cross-equatorial flow and the cyclonic flowin the Bay of Bengal. Other atmospheric and oceanicfactors, such as the east–west pressure gradient acrossthe equatorial Indian Ocean, may also contribute to thedevelopment of equatorial westerly winds. These windshave two effects, both of which lead to more convec-tion. The first is the onshore flow that causes conver-gence along the western coasts of northern Sumatraand the Malay Peninsula. The second is the beta effectthat produces convergence in the equatorial westerliesin the boundary layer. The increased convection mayfurther enhance the westerlies off the coast of Sumatraand make a positive feedback possible.

South of the equator, the SLP difference betweenAustralia and the south Indian Ocean favors counter-clockwise flow toward the equator (area D in Fig. 10)in boreal fall. This also enhances the cross-equatorialflow that becomes a westerly wind north of the equator,and thus increases the wind–terrain interactions on thewest coast of the adjacent land areas, such as westernBorneo (area B). Furthermore, the convergence ofthese winds from south of the equator and the north-easterly winds in the northern South China Sea and thenorthwestern Pacific (area A) give rise to a broad-scalebelt of convergence between the equator and 20°N dur-ing boreal fall (marked by the elliptic-shaped area inFig. 10), which also favors convection. During borealspring, this belt becomes a zone of low-level diver-gence, so convection is suppressed. The results of thesevarious effects are that maximum convection is pre-vented from marching continuously from the Asianwinter monsoon to the Asian summer monsoon.

5. Summary and concluding remarks

In this study, historical station rainfall data were usedto classify the geographical distribution of the annualcycle of rainfall in the Southeast Asia–Maritime Con-tinent region into several types, which depends on thewet season, the amplitude and phase of the annualcycle, and its importance relative to the maximum rain-fall and season, and the relative importance of the semi-annual cycle. Over Indonesia, the results are consistent

with the gross features of Hamada et al.’s (2002) five-type classifications using 1961–90 data.

The annual cycle dominates rainfall over most of theregion and its phase and amplitude vary significantly.The QuikSCAT sea surface winds show that most ofthese variations are due to interactions between thecomplex terrain and a simple annual reversal of thesurface monsoonal winds. The most notable example isthe effect of seasonal-varying onshore winds and theactivity of low-level vortices in the South China Seaalong the northwest coast of Borneo. The semiannualcycle is comparable in magnitude to the annual cycleover parts of the large islands of Sumatra and Borneoand the Malay Peninsula. Only a very small part ofeastern Indonesia close to the equator exhibits a mod-est semiannual variation that may be attributed to thetwice-yearly crossing of the sun.

Analysis of the data reveals a clear demarcation ofthe boreal summer and winter monsoon regimes overthe entire region, showing the two regimes intertwiningacross the equator. There are stronger intrusions of thewinter regime northward into the summer regime dueto strong northeasterly monsoon winds in the SouthChina Sea and the western Pacific east of the Philip-pines. These monsoon winds produce strong onshoreflow and excess rainfall along the eastern flanks of themajor island groups and landmasses. During borealsummer, the interaction between the southwest mon-soon wind and terrain also affects strongly the distribu-tion of the summer monsoon rainfall, with maximumrainfall occurring on the windward side of terrain in theIndian Ocean, the Indochina Peninsula, and the SouthChina Sea.

The analysis also clearly delineates the asymmetry ofthe season march between boreal fall and boreal spring.During boreal fall the convection maximum occurs inthe eastern Indian Ocean, Malay Peninsula andSumatra, and southern South China Sea. This generalarea forms the midpoint of the path of the southeast-ward progression of convection from the Asian summermonsoon to the Asian winter monsoon. During borealspring the maximum convection does not retrace theprocess but stays near and south of the equator. Thisasymmetric feature is a manifestation of the asymmet-ric locations of the spring and fall ITCZ noted by manyinvestigators (e.g., Lau and Chan 1983; Meehl 1987;Yasunari 1991; Matsumoto and Murakami 2000, 2002;Hung et al. 2004).

The cause of this asymmetry has not been fully un-derstood. A possible factor lies in the difference in thestrength of broad-scale Walker circulation driven byzonal SST gradients over both the equatorial Pacificand Indian Oceans. Over the Pacific, the Walker circu-lation that gives rising motion over the Maritime Con-tinent is affected by the intensity of the eastern equa-torial Pacific cold tongue (Wang 1994; Li and Philander1996). The cold tongue is most intense in boreal fall, sothe Pacific Walker cell may favor convection over the


Maritime Continent in boreal fall rather than in borealspring. Across the equatorial Indian Ocean the zonalSST gradient is more favorable for a local Walker-typecirculation that supports convection over the MaritimeContinent in boreal fall than in boreal spring, as theannual cycle of the western equatorial Indian Ocean iswarmest in boreal spring (Webster et al. 1998). As aresult, the planetary scale circulation may favor en-hanced convection over the land bridge of SoutheastAsia–Maritime Continent in boreal fall and thus a con-tinuous southeastward march from Asian summer mon-soon to Australian summer monsoon, and may makethe reverse trip harder during boreal spring.

More direct hypotheses to explain the asymmetryhave been proposed by several investigators. Matsu-moto and Murakami (2000) suggested the asymmetry isdue to the fact that cold surges that originate from thehigh terrain of the Asian continent during boreal fallare stronger than the cold surges that originate fromAustralia during boreal spring. Matsumoto and Mu-rakami (2002) also proposed that this asymmetry is re-lated to the different equatorial basic flows in the west-ern Pacific and Indian Ocean, and their respective an-nual variations. They suggested that the westernequatorial Pacific with a strong Kelvin-type mean flowduring boreal fall allows the convection to move south-ward, and that the flow weakens during boreal springsuch that it inhibits the return journey during borealfall. They also suggested that the variation of a complexRossby-type mean flow in the equatorial Indian Oceanallows convection to move northward in boreal springbut inhibits the southward movement in boreal fall.

Hung et al. (2004) proposed that the northwardmarch of the ITCZ during boreal spring is blocked bysubsidence in oceanic regions to the west of heatsources (India, Indochina, and the Philippines). Theysuggested that this subsidence, which is similar to thatoccurring west of Tibetan Plateau during the summermonsoon (Yanai et al. 1992), is the result of the Rossbywave response in the monsoon–desert mechanism sug-gested by Rodwell and Hoskins (1996). T. Li (2003,personal communication) proposed that the asymmetryis attributed to an internal atmospheric dynamicsmechanism by which equatorial oceanic convectiontends to propagate eastward due to the production of aboundary layer convergence to the east of the deepconvection. Such a process will favor the southeastwardseasonal march from the Asian summer monsoon to theAsian winter monsoon, but not the northwestwardmarch from the Australian region to the Asian summermonsoon.

Our analysis suggests that at least a part of this asym-metry is due to a combination of two effects: (i) thelow-level wind and terrain interactions throughout theregion during boreal fall that do not have counterpartsin boreal spring, and this includes the effect of borealfall cold surges noted by Matsumoto and Murakami(2002); and (ii) an intervening region between the

equator and 20°N in the longitudes of Southeast Asiaand the Maritime Continent that in the low levels has aconvergence bias in boreal fall and a divergence bias inboreal spring. Over the eastern Indian Ocean, this ef-fect is enhanced by the beta effect of the equatorialzonal wind near the surface. Both effects can be relatedto the SLP differences between spring and fall that ap-pear to be the result of the global-scale mass redistri-bution between the land and ocean regions, which isdriven by their different thermal memories. Because ofthe orientation of the Asian and Australian landmasses,this redistribution facilitates the southeastward marchof maximum convection from the Asian summer mon-soon to the Asian winter (Australian summer) mon-soon, but it deters the reverse march in boreal spring.

Acknowledgments. We wish to thank Tim Li forstimulating discussions and Russ Elsberry and Pat Harrfor reading the manuscript. This work was supported bythe National Oceanic and Atmospheric Administrationunder Grant NA01AANRG0011 and the National Sci-ence Foundation under Grant ATM-0101135 at the Na-val Postgraduate School.

REFERENCES

Aldrian, E., and D. Susanto, 2003: Identification of three domi-nant rainfall regions within Indonesia and their relationshipto sea surface temperature. Int. J. Climatol., 23, 1435–1452.

——, L. Dumenil-Gates, and F. H. Widodo, 2003: Variability ofIndonesian rainfall and the influence of ENSO and resolutionin ECHAM4 simulations and in the reanalyses. Rep. 346,Max Planck Institute for Meteorology, Hamburg, Germany,30 pp.

Braak, C., 1921–29: Het Climaat van Nederlandsch Indie. Magn.Meteor. Observ. Batavia, Verhand, 8, Parts I and II.

Chang, C.-P., C. H. Liu, and H. C. Kuo, 2003: Typhoon Vamei: Anequatorial tropical cyclone formation. Geophys. Res. Lett.,30, 1150, doi:10.1029/2002GL016365.

——, P. A. Harr, J. McBride, and H. H. Hsu, 2004: MaritimeContinent monsoon: Annual cycle and boreal winter. EastAsian Monsoon, World Scientific Series on Meteorology andEast Asia, Vol. 2, World Scientific, 107–150.

——, ——, and H. J. Chen, 2005: Synoptic disturbances over theequatorial South China Sea and western Maritime Continentduring boreal winter. Mon. Wea. Rev., in press.

Coronas, Rev. J., 1912: The extraordinary drought in the Philip-pines, October 1911 to May 1912. Government of the Phil-ippine Islands, Weather Bureau, Manila Bureau of Printing,19 pp.

Cressman, G. P., 1959: An operational objective analysis system.Mon. Wea. Rev., 87, 367–374.

Flores, J. F., and V. F. Balagot, 1969: Climate of the Philippines.Climates of Northern and Eastern Asia, H. Arakawa, Ed.,World Survey of Climatology, Vol. 8, Elsevier, 159–213.

Hamada, J.-I., M. D. Yamanaka, J. Matsumoto, S. Fukao, P. A.Winarso, and T. Sribimawati, 2002: Spatial and temporalvariations of the rainy season over Indonesia and their link toENSO. J. Meteor. Soc. Japan, 80, 285–310.

Haylock, M., and J. McBride, 2001: Spatial coherence and pre-dictability of Indonesian wet season rainfall. J. Climate, 14,3882–3887.

Hung, C.-W., and M. Yanai, 2004: Factors contributing to the


onset of the Australian summer monsoon. Quart. J. Roy. Me-teor. Soc., 130, 739–761.

——, X. Liu, and M. Yanai, 2004: Symmetry and asymmetry of theAsian and Australian summer monsoons. J. Climate, 17,2413–2426.

Johnson, R. H., and R. A. Houze Jr., 1987: Precipitating cloudsystems of the Asian monsoon. Monsoon Meteorology, C.-P.Chang and T. N. Krishnamurti, Eds., Oxford UniversityPress, 298–353.

Kirono, D., N. Tapper, and J. McBride, 1999: Documenting In-donesian rainfall in the 1997/1998 El Niño event. Phys.Geogr., 20, 422–435.

Lau, K. M., and P. H. Chan, 1983: Short-term climate variabilityand atmospheric teleconnections from satellite-observed out-going longwave radiation. Part II: Lagged correlations. J. At-mos. Sci., 40, 2751–2767.

Leise, J. A., 1982: A multidimensional scale-telescoped filter anddata extension package. NOAA Tech. Memo. ERL WPL-82,19 pp.

Li, T., and S. G. H. Philander, 1996: On the annual cycle of theeastern equatorial Pacific. J. Climate, 9, 2986–2998.

Liu, W. T., 2002: Progress in scatterometer application. J. Ocean-ogr., 58, 121–136.

Masó, Rev. M. S., 1914: Annual amount and distribution of rain-fall in the Philippines. Government of the Philippine Islands,Weather Bureau, Manila Bureau of Printing, 42 pp.

Matsumoto, J., 1992: The seasonal changes in Asian and Austra-lian monsoon regions. J. Meteor. Soc. Japan, 70, 257–273.

——, and T. Murakami, 2000: Annual changes of tropical convec-tive activities as revealed from equatorial symmetric OLRdata. J. Meteor. Soc. Japan, 78, 543–561.

——, and ——, 2002: Seasonal migration of monsoons betweenthe Northern and Southern Hemisphere as revealed fromequatorially symmetric and asymmetric OLR data. J. Meteor.Soc. Japan, 80, 419–437.

McBride, J., 1987: The Australian summer monsoon. MonsoonMeteorology, C.-P. Chang and T. N. Krishnamurti, Eds., Ox-ford University Press, 203–231.

——, 1998: Indonesia, Papua New Guinea, and tropical Australia:The Southern Hemisphere monsoon. Meteorology of theSouthern Hemisphere, Meteor. Monogr., No. 49, Amer. Me-teor. Soc., 89–99.

——, N. E. Davidson, K. Puri, and G. C. Tyrrell, 1995: The flowduring TOGA COARE as diagnosed by the BMRC tropicalanalysis and prediction system. Mon. Wea. Rev., 123, 717–736.

Meehl, G. A., 1987: The annual cycle and interannual variability

in the Tropical Pacific and Indian Ocean regions. Mon. Wea.Rev., 115, 27–50.

Neumann, C. J., 1993: Global overview. Global Guide to TropicalCyclone Forecasting, WMO Tech. Doc. 560, World Meteo-rological Organization, Geneva, Switzerland, 1.1–1.56.

Qian, W., and D. K. Lee, 2000: Seasonal march of Asian summermonsoon. Int. J. Climatol., 20, 1371–1386.

Ramage, C. S., 1968: Role of a tropical “maritime continent” inthe atmospheric circulation. Mon. Wea. Rev., 96, 365–369.

——, 1971: Monsoon Meteorology. Academic Press, 296 pp.Riehl, H., 1954: Tropical Meteorology. McGraw-Hill, 392 pp.Rodwell, M. J., and B. J. Hoskins, 1996: Monsoons and the dy-

namics of deserts. Quart. J. Roy. Meteor. Soc., 122, 1385–1404.

Schmidt, F. H., and J. H. A. Ferguson, 1951: Rainfall types basedon wet and dry period ratios for Indonesia with western NewGuinea. Verhandelingen, 42, 77 pp. plus figures.

Simpson, J., C. Kummerow, W.-K. Tao, and R. F. Adler, 1996: Onthe Tropical Rainfall Measuring Mission (TRMM). Meteor.Atmos. Phys., 60, 19–36.

Sukanto, M., 1969: Climate of Indonesia. Climates of Northernand Eastern Asia, H. Arakawa, Ed., World Survey of Clima-tology, Vol. 8, Elsevier, 215–229.

Van den Dool, H. M., and S. Saha, 1993: Seasonal redistributionand conservation of atmospheric mass in a general circulationmodel. J. Climate, 6, 22–30.

van Loon, H., 1967: The half-yearly oscillation in middle and highsouthern latitudes and the coreless winter. J. Atmos. Sci., 24,472–486.

Waliser, D. E., and C. Gautier, 1993: A satellite-derived climatol-ogy of the ITCZ. J. Climate, 6, 2162–2174.

Wang, B., 1994: On the annual cycle in the tropical eastern andcentral Pacific. J. Climate, 7, 1926–1942.

Webster, P. J., T. Palmer, M. Yanai, R. Tomas, V. Magana, J.Shukla, and T. Yasunari, 1998: Monsoons: Processes, predict-ability and the prospects for prediction. J. Geophys. Res., 103,14 451–14 510.

Wyrtki, K., 1956: The rainfall over the Indonesian waters. Kemen-terian Perhubungan Lembaga Meteorologi dan Geofisik Ver-hand, 49.

Yanai, M., C. Li, and Z. Song, 1992: Seasonal heating of theTibetan Plateau and its effects on the evolution of the Asiansummer monsoon. J. Meteor. Soc. Japan, 70, 319–351.

Yasunari, T., 1991: The monsoon year—A new concept of theclimatic year in the Tropics. Bull. Amer. Meteor. Soc., 72,1331–1338.


annual cycle of southeast asia—maritime continent … cycle of southeast asia—maritime continent...

Documents