INFORMATION TO USERS
This was produced from a copy of a document sent to us for microfilming. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the material submitted.
The following explanation of techniques is provided to help you understand markings or notations which may appear on this reproduction.
1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure you of complete continuity.
2. When an image on the film is obliterated with a round black mark it is an indication that the film inspector noticed either blurred copy because of movement during exposure, or duplicate copy. Unless we meant to delete copyrighted materials that should not have been filmed, you will find a good image of the page in the adjacent frame.
3. When a map, drawing or chart, ei.c., is part of the material being photo graphed the photographer has followed a definite method in "sectioning" the material. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again-beginning below the first row and continuing on until complete.
4. For any illustrations that cannot .be reproduced satisfactorily by xerography, photographic prints can be purchased at additional cost and tipped into your xerographic copy. Requests can be made to our Dissertations Customer Services Department.
s. Some pages in any document may have indistinct print. In all cases we have filmed the best available copy.
Uni~ MicrOfilms International 300 N. ZEEB ROAD, ANN ARBOR, MI 48106 18 BEDFORD ROW, LONDON WC1R 4EJ, ENGLAND 8012262
LIM, JooTICK
CHARACTERISTICS OF THE WINTER MONSOON OVER THE MALAYSIAN REGION
University ofHawaii PH.D. 1979
University Microfilms International 300 N. ZeebRoad, AnnArbor, MI 48106 18Bedford Row, London WCIR4EJ,England CHARACTERISTICS OF THE WINTER MONSOON
OVER THE MALAYSIAN REGION
A DISSERTATION SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HAWAII IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
IN METEOROLOGY
DECEMBER 1979
By
JOO TICK LIM
Dissertation Committee
Colin S. Ramage, Chairman Marie Sanderson Wan-Cheng Chiu James C. Sadler Thomas A. Schroeder iii ACKNOWLEDGEMENTS
The author is indebted to Professor Colin S. Ramage for
his continuous encouragement and guidance through0ut the
entire study. He would like to thank Professors Wan-Cheng
Chiu, James C. Sadler and Thomas A. Schroeder for their
constructive comments and helpful suggestions.
The World Meteorological Organization provided the
long~term fellowship to enable him to undertake graduate
study at the University of Hawaii. The Director-General of
the Malaysian Meteorological Service gave support and
encouragement, and had arranged for speedy delivery of requested data.
Finally, the study makes constant reference to the climatological works related to the Malaysian region by
Professors S. Nieuwolt, C. S. Ramage and J. C. Sadler and
Dr. W. L. Dale. iv
ABSTRACT
Two periods, a strong and weak monsoon periods, are chosen from the winter monsoon of 1974/75 for the study of some characteristics of the monsoon over the Malaysian region.
In the strong monsoon period, it is found that a time lag of about three days existed between the cold surge arrival at Hong Kong and the occurrence of heavy rainfall over the east coast of Peninsular Malaysia. Interaction between the quasi-stationary near-equatorial trough and monsoon surge caused the heavy rainfall over Malaysia.
Mean maps of various radiative parameters for South- east Asia for the two periods are analyzed. Of interest is the albedo map for the surge period, which indicates the maximum albedo (convective activity) areas were confined largely to the Peninsular Malaysia-South Thailand-South Bay of Bengal region.
Composite maps of surface wind speed, air temperature, dew-point and sea surface temperature (SST) over the South
China Sea are prepared. It is found that a strong wind band was aligned along the general wind direction during the monsoon surge. The strong wind band formed the main surface branch of the Hadley cell while the intense convective activity region, which lay downwind of the strong wind band, became the main ascending branch. In general, the distribu- v tions of SST, air temperature and dewpoint during the strong monsoon period showed large gradients north of 15N.
To south of 15N, they were relatively uniform. In the weak monsoon period, the distributions of these parameters, with the exception of SST, were influenced greatly by the presence of a tropical storm.
Surface heat exchanges are computed for several locations over the South China Sea using the bulk aero- dynamic equations. It is noted that during the strong monsoon period, the intake of latent heat and sensible heat by the atmosphere decreased by more than half when the northeasterlies traversed from the northern South China Sea to the Peninsular Malaysia-South Thailand region, although wind speed remained as strong as in the north.
A detailed study is made of the near-equatorial trough.
It is found that the trough has a double-convergence structure with primary convergence located in the trough northern flank. The trough thermal structure resembles that of the typhoon and evidence points to CISK as the principal mechanism for its development and maintenance.
Hourly surface wind observations along the Malaysian coasts are analyzed to determine the characteristics of the coastal winds and the extent of local influence. During the weak monsoon period, land/sea breeze circulations pre vailed everywhere, while during the strong monsoon period, vi the mesoscale circulations existed in most areas except northeastern Peninsular Malaysia. The coastal winds had three main diurnal variations which can be analyzed by a two-dimensional analytical model. Theoretical analyses indicate that the coastal wind characteristics are deter mined largely by the synoptic winds, Coriolis parameter, temperature difference between land and sea, and surface friction.
During the monsoon surge, the mountain ranges and near-equatorial trough restricted the strong monsoon conditions to northeastern Peninsular Malaysia, where a rains regime dominated. Here s,rong winds and cloudy skies suppressed all local effects and rainfall occurred mostly during the night and early morning. Elsewhere, a showers regime prevailed. The observed diurnal rainfall variations were generally the result of interactions among the various local factors and synoptic winds. In the weak monsoon period, a showers regime prevailed throughout Peninsular and East Malaysia. vii
TABLE OF CONTENTS
ACKNOWLEDGEMENTS • iii
ABSTRACT . iv
LIST OF TABLES xi
LIST OF ILLUSTRATIONS xii
1.0 INTRODUCTION 1
2.0 WINTER MONSOON SURGES. 4
2.1 Introduction . 4
2.2 Relation between monsoon surges and weather over Malaysia 6
2.3 Monsoon disturbances 7
2.4 The surge of January 1-8, 1975 . 9
2.5 The surges of Decembec 1973 10
2.6 Additional remarks. 10
3.0 LARGE-SCALE FEATURES OVER SOUTHEAST ASIA 12
3.1 Mean earth-atmospheric radiation components . 12
3.1.1 Radiation data 12
3.1.2 Radiative heating during the strong monsoon period . 12
3.1.3 Radiative heating during the monsoC'n break . 14
3.2 Surface characteristics of the South China Sea ....•. 14
3.2.1 Data sources 14
3.2.2 Strong monsoon characteristics 15
3.2.3 Weak monsoon characteristics 17 viii
TABLE OF CONTENTS (continued)
3.3 Heat exchange over the South China Sea. 18
3.3.1 Bulk aerodynamic equations 18
3.3.2 Results of heat transfer computation . 21
3.4 East Asian Hadley circulation 23
4.0 NEAR-EQUATORIAL TROUGH . 26
4.1 Climatology 26
4.1.1 Mean position of the gradient level trough 26
4.1.2 Rainfall distribution across the trough 27
4.2 Case study . 29
4.2.1 Thermodynamic structure of the trough during the strong monsoon period 29
4.2.2 Structure of the trough during the monsoon break . 34
4.2.3 Structure of the trough over East Malaysia . 34
4.2.4 Dynamic aspects. 35
4.3 Oceanic trough. 40
4.4 Comparison with earlier studies 42
4.5 Local changes in the intensity of the trough • 44
5.0 LAND/SEA BREEZES 46
5.1 Observations of coastal surface winds 46
5.1.1 Weak monsoon period. 46
5.1.2 Strong monsoon period. 47 ix
TABLE OF CONTENTS (continued)
5.2 Theoretical discussion. 48
5.2.1 Brief review of theoretical studies .. 48
5.2.2 Choice of coordinate system 52
5.2.3 Temperature difference between land and sea 53
5.2.4 Coefficient of surface resistance of the coast 55
5.2.5 Prevailing wind at the coast 59
5.2.6 Application of the model to the equatorial region . 59
5.2.7 Rate of change of wind direction 65
5.2.8 Further comments on the model. 67
5.3 Effect of the near-equatorial trough. 68
5.4 Weather associated with land/sea breezes 70
6.0 MONSOON RAINFALL CHARACTERISTICS 72
6.1 Rainfall of Peninsular Malaysia 72
6.1.1 Diurnal rainfall variation during the strong monsoon period 73
6.1.2 Diurnal rainfall variation during the monsoon break 75
6.2 Rainfall of East Malaysia 76
6.2.1 Diurnal rainfall variation during the strong and weak monsoon periods . 78
7.0 SUMMARY. 80 x
TABLE OF CONTENTS (continued)
APPENDIX A. ANALYTICAL SOLUTIONS TO LAND/SEA BREEZE EQUATIONS ... . 83
APPENDIX B. DA~A ~OURCES 86
REFERENCES . 87 xi
LIST OF TABLES
Table Page
I Computation of heat transfer at selected points over surface of the South China Sea 94
2 Temperatures at standard levels over Kota Bharu during the strong and weak monsoon periods (DC) 95
3 Mean resultant surface winds of Malaysian coastal stations (deg/m sec-I) . .... 96
4 Mean diurnal temperature range and mean temperature for Malaysian coastal stations . 97
5 Computed constants for various values of the coefficient of surface resistance (k) 98 xii
LIST OF ILLUSTRATIONS
Figure Page
1 Map of Southeast Asia 99
2 Peninsular Malaysia, location of rainfall stations and land above 500 feet . 101
3 East Malaysia, location of rainfall stations and land above 300 feet . 103
4 Monsoon surge over the South China Sea and rainfall over Malaysia during January 1-8, 1975 105
5 Upper-level winds over Malaysia during January 1-8, 1975 107
6 Monsoon surges of December 1973 109
7 Time cross-section of upper-level winds over Kota Bharu during December 1973 . 111
8a Albedo over Southeast Asia during the strong monsoon period 113
8b Outgoing long-wave radiation over Southeast Asia during the strong monsoon period . 115
8c Net radiation over Sour"east Asia during the strong monsoon period 117
9 850-mb streamline chart for 0000 GMT January 4, 1975 119
lOa Albedo over Southeast Asia during the weak monsoon period 121
lOb Outgoing long-wave radiation over Southeast Asia during the weak monsoon period . 123
10c Net radiation over Southeast Asia during the weak monsoon period 125
11 850-mb streamLine chart for 0000 GMT November 29, 1974 127 xiii
LIST OF ILLUSTRATIONS (continued)
Figure Page
12 Composite maps of surface wind speed and sea surface temperature for the strong monsoon period 129
13 Composite maps of surface air temperature and dewpoint temperature for the strong monsoon period 131
14 As in Figure 12 except for the weak monsoon period • 133
15 As in Figure 13 except for the weak monsoon period . 135
16 Vertical cross-section of relative humidity, potential temperature and winds during the strong monsoon period . 137
17 Mean gradient-level circulation over Southeast Asia for November and January 139
18 Mean monthly rainfall of Peninsular Malaysia for the winter monsoon 141
19 Mean monthly rainfall of East Malaysia for the winter monsoon . 143
20 Location of upper-air stations in Singapore, South Thailand and Peninsular Malaysia . 145
21 Cross-section of relative humidity and virtual equivalent potential tempera- ture for January 1, 1975 147
22 As in Figure 21 except for January 3 , 1975 149
23 As in Figure 21 except for January 4, 1975 151
24 As in Figure 21 except for January 8 , 1975 153 xiv
LIST OF ILLUSTRATIONS (continued)
Figure Page
25 Cross-section of surface pressure and pressure height across the trough for January 4 and for January 1-8, 1975 . 155
26 Daily pressure at 0730 LT along the east coast of Peninsular Malaysia during the strong monsoon period 157
27 As in Figure 21 except for November 28, 1974 . 159
28 As in Figure 21 except for December 4, 1974 . 161
29 Rainfall distribution along the coastal strip of East Malaysia which borders the South China Sea during the strong monsoon period . 163
30 Relative vorticity over the triangle Songkhla, Kota Bharu and Penang; the triangle Kota Bharu, Penang and Kuala Lumpur; and the triangle Kota Bharu, Kuala Lumpur and Paya Lebar . 165
31 As in Figure 30 except for divergence 167
32 As in Figure 30 except for vertical velocity . 169
33 Schematic model of a near-equatorial trough over Southeast Asia . 171
34 Mean 200-mb circulation over Southeast Asia for November and December . 173
35 Hourly surface pressures along the east coast of Peninsular Malaysia on January 4 and 5, 1975 175
36 Hourly weather observations at stations along the east coast of Peninsular Malaysia on January 4 and 5, 1975 177 xv
LIST OF ILLUSTRATIONS (continued)
Figure Page
37 Mean hourly surface resultant winds along the coasts of Peninsular Malaysia during the weak monsoon period . 179
38 Mean hourly surface resultant winds along the coasts of East Malaysia during the weak mdnsoon period . 181
39 As in Figure 37 except for the strong monsoon period 183
40 As in Figure 38 except for the strong monsoon period 185
41 Schematic diagram showing the observed and rotated wind directions for east and west coast stations 187
42 Mean hourly surface resultant winds along the coasts of Peninsular Malaysia during the weak monsoon period . 189
43 As in Figure 42 except for East Malaysia . 191
44 Mean hourly surface resultant winds along the coasts of Peninsular Malaysia during the strong monsoon period . 193
45 As in Figure 44 except for East Malaysia . 195
46 Theoretical surface winds for the weak monsoon period • 197
47 Theoretical surface winds for the strong monsoon period 199
48 Rotation rate of wind direction due to the mesoscale pressure term . 201
49 Hourly surface winds along the east coast of Peninsular Malaysia on January 4 and 5, 1975 203 xvi
LIST OF ILLUSTRATIONS (continued)
Figure Page
50 Theoretical winds along the southeast coast of Peninsular Malaysia on January 5, 1975 . 205
51 Diurnal rainfall variation of Peninsular Malaysia for the strong monsoon period 207
52 Diurnal rainfall variation of Peninsular Malaysia for the weak monsoon period 209
53 Monthly rainfall distribution of East Malaysia . 211
54 Diurnal variation of January rainfall of East Malaysia . 213
55 Diurnal rainfall variation of East Malaysia for the strong monsoon period 215
56 Diurnal rainfall variation of East Malaysia for the weak monsoon period 217 1
1.0 INTRODUCTION
In Malaysia,l the winter (northeast) monsoon begins
normally in mid-November and ends in early March. During
this period, the east coast of Peninsular Malaysia and the west coast of East Malaysia are shrouded regularly by rain-
laden clouds. Torrential rains drench these coasts two to
three times in a season, culminating almost always in the
flooding of the coastal plains. Many of these floods result
in light damage to crops and property but occasionally,
severe floods cause widespread devastation and loss of
lives.
The winter monsoon over subtropical Southeast Asia has been intensively studied by American meteorologists while that over equatorial Southeast Asia has barely been explored. Malaysia is undoubtedly aware of the need for better understanding of the monsoon, and has taken steps to encourage research in this field. Chiefly on account of 2 this, there was enthusiasm to have Winter MONEX 1978/79 based at Kuala Lumpur.
1 The location and topographical features of Malaysia are shown in Figures 1-3.
2 Winter Monsoon Experiment 1978/79 was part of the Global Atmospheric Research Program sponsored by the World Meteorological Organization. 2
This dissertation aims at achieving a broad compre- hension of the winter monsoon over the Malaysian region.
I open with discussion of East Asian monsoon surges (Chapter
2). In the following chapter, I look briefly into large- scale features such as the distribution of radiative para- meters over Southeast Asia and surface characteristics of the South China Sea. The near-equatorial trough and the land/sea breeze circulation are believed to play an important role in shaping the weather patterns in Malaysia and I deal with them in detail in Chapters 4 and 5.
Finally, I discuss the important features of monsoon rain- fall.
The study is essentially confined to two periods:
(i) a strong (active) monsoon period:
January 1-8, 1975
(ii) a weak monsoon period (monsoon break)
November 26-December 5, 1974
The strong and weak monsoon periods are defined in terms of 3 the Hadley circulation strength. During January 1-8, 1975, pressure over Central China rose above 1030 mb. Surface nnrtheasterlies over the South China Sea increased by
8-10 m/sec to speed greater than 15 m/sec within 48 hours
3 Generally, Hadley circulation strength is deduced from mean merioional motion averaged longitudinally over tropical latitudes. Here the term refers to the local Hadley circulation intensity over East Asia. 3 of passage of a cold front across the South China coast, indicating a major monsoon surge. Rainfall over the east coast of Peninsular Malaysia was about three times larger than the seasonal mean. In every respect, the period was typical of a period of strong monsoon circulation. In contrast, November 26-December 5, 1974 was a period of weak monsoon. Pressure at Hong Kong fell to a low of 1009 mb.
Northeaster1ies over Malaysia were weak and at times absent. Along the east coast of Peninsular Malaysia, relatively dry conditions prevailed. 4
2.0 WINTER MONSOC~ SURGES
2.1 Introduction
In late August, the monsoon trough over East Asia and
West Pacific begins to migrate southward to the equator.
At the same time, the Siberian anticyclone builds up gradually and becomes a stable feature by November. Cold surges begin to affect South China and Indochina during early October and by late October, cold air has reached the central South China Sea. Off the coasts of Malaysia, initial reports of gusty winds and stormy seas usually come in during the middle or end of November and this supports Bell's iuference from climatology (Ramage, 1971) that the first simultaneous freshening of northeasterlies throughout the South China Sea occurs between November 18 and 29. The surges then become inreasingly vigorous until January when the Asiatic winter monsoon is generally regarded to b~ at its maximum.
The monsoon surges are triggered by certain synoptic events in the northern latitudes. As reported by Thompson
(1951), the most commonly identified event is the develop ment of surface depression off the coast of China between
Taiwan and Sakhalin. These depressions originate from one of the following: (i) waves in the West Pacific polar S front, (ii) lee depressions south of Japan and (iii) short wave westerly troughs (identifiable at the 700-mb) which build down to the surface on reaching the coast.
In any event, the depression deepens and draws out cold air from the Siberian anticyclone which intensifies simultaneously. A sustained expansion of the Siberian anticyclone brings about a major cold air outbreak. The cold surge often follows at the rear of a cold front which sweeps southeastward into the South China Sea, and the passage of the front is indicated by rising pressure and falling dewpoint temperature at coastal stations.
Cold surge arrival at Hong Kong is most accurately predicted by inspecting the SOO-mb trough-ridge position in the middle latitudes. According to Chu (1978), a well marked SOO-mb trough positioned east of Lake Baikal with an associated ridge between 80E and 90E (west of Lake Baikal) forewarns a cold surge arrival at Hong Kong two days in advance.
In addition to westerly wave movement and development of East China coast depression, the outflow of cold air from China is frequently accompanied by intensification and southward displacement of the subtropical jet (Chin,
1969). 6
2.2 Relation between monsoon surges and weather over Malaysia
Deterioration of weather over Malaysia and Singapore corresponds closely to polar outbreaks over China. The pressure at Hong Kong has been recognized for quite sometime to be a good indicator of the pressure level over China and has been used for forecasting rain-spells over the east coast of Peninsular Malaysia. Gan (undated) examined all heavy rain-spells over Peninsular Malaysia during the period November-December 1949-58 and found that the majority of heavy rain-spells began only when Hong Kong had pressure above 1020-mb together with positive 24-hour pressure tendency. Based on this result, there should, by and large, exist a time lag between the first detection of the surge at Hong Kong and occurrence of torrential rainfall near the equator.
Conversely, Ramage (1970a) showed from a case study that as a cold front passed Hong Kong, wind strength over the South China Sea as well as rainfall over East Malaysia increased at almost the same time. This appears to be the manifestation of a near-simultaneous intensification of the
Hadley circulation which he stressed, could not be due to either frontal movement or advective processes. He was of the opinion that there was an increase in the south-north temperature gradient at the time when the surge was initiated and this accelerated the whole Hadley circulation leading to the observed increase in rainfall in East Malaysia. 7
2.3 Monsoon disturbances
Prior to 1970, it had been customary for Malaysian meteorologists to ascribe the cause of heavy rain-spells to vor.tices or e~uatoria1 waves as described by Palmer (1951).
Vortices identified over the South China Sea were assumed to have migrated subsequently from the West Pacific.
Recently, Cheang (1977) reemphasized the importance of these westward-moving vortices and their relation with severe flooding over Peninsular Malaysia. Evidence presented by him suggests that apart from spawning in the West
Pacific, the vortices can originate also from disturbances embedded in the zonal easterlies over the South China Sea.
As these disturbances are found to be associated with lateral shear, he attributed their formation to barotropic instability. A low-level easterly jet over Southeast Asia was observed during each surge period in his case study and because of the presence of vertical shear (due to the easterly jet) and large diabatic processes in the tropics, he suspected that the vortices are maintained by a combina tion of conditional instability of the second kind (CISK) and mixed barotropic and baroc1inic instability. Pre-
1iminary investigation of the energetics by Murakami (1977) does not entirely support this line of reasoning. He maintained that weak barotropic instability found in the 8
South China Sea plays an insignificant part in the mainte nance of eddy motion, which instead, could be accounted for by baroclinic processes.
The above view on the origin of tropical vortices has been opposed fervently by Sadler (1970). He contended that all low-level vortices form in the directional shear zone of a trough system, with the exception of vortices in the trade wind zone, which could be initiated by the tropical upper-tropospheric trough (TUTT). In recent years, more reliable analyses resulting from an increase of upper-air data in East Malaysia show that many vortices are quasi stationary and they develop in the part of the trough off the coast of Sarawak.
Often the cause of bad weather cannot be traced to any cyclonic disturbance. This was illustrated by Ramage
(1970b) with an example taken from the period March 6 through 14, 1967. Immediately after a cold outbreak, amorphous cloud masses formed over Sabah expanded and spread to Peninsular Malaysia on the 9th. Heavy rainfall was recorded at stations along the east coast of Peninsular
Malaysia on the 10th and 11th. The cloud masses began to break up on the 12th after the Hadley cell weakened. The cloud masses were related to the intensification of the stationary equatorial trough rather than to any travelling disturbance. 9
2.4 The surge of January 1-8, 1975
The synoptic situation over China on January 1, 1975
was not dissimilar to that described for a typical surge
condition in Section 2.1. Hong Kong pressure and dewpoint
temperature reacted accordingly following the passage of a
cold front during that day. The trends of these two
parameters continued until the 5th before they reversed.
Throughout the South China Sea (surface branch of Hadley
cell), surface winds accelerated to maximum force during
the 2nd and 3rd and thereafter they lapsed gradually
(Figure 4, top).
Over Peninsular Malaysia, the gradient-level trough
(Figure 5, top) moved to its southernmost position on the
3rd in quick response to the surface impulse. The East
Malaysian part of the trough was less responsive (Figure 5,
middle) due, perhaps, to the fact that the region was less
affected by the surge. The trough, nevertheless, moved to
its most southern point on the 5th.
The major part of the rainfall (Figure 4, bottom) took
place at least one day after the observed surge signal over
the South China Sea--between the 4th and 6th in Peninsular
Malaysia and between the 5th and 6th in East Malaysia. The
·midd1e and upper tropospheric winds (upper branch of Hadley
cell) were at their maximum on t 1,e 4th and 5th and by then
all signs indicate that the Hau1ey circulation had attained
its peak intensity. 10
Clearly, in this case, the various branches of the vertical circulation did not develop at the same time but in three or more brief stages. The principal cause of the increase in rainfall was again the fairly stationary trough which was activated by the converging northeasterlies.
2.5 The surges of December 1973
For comparison, I have chosen the surges of December
1973 which have been examined by Cheang (1977). From
Figure 6 which portrays several features of the monsoon during this period, it is to be noted that the surge-rainfall relationship applies quite well to Peninsular Malaysia but not to East Malaysia. This fact will be considered in detail in the next chapter. Four significant rainfall periods could be delimited: November 28-December 3,
December 7-9, December 20-22 and December 24-28 (Figure 6, bottom). Of the four periods, two (first and third) came about following a sequence of events resembling closely that mentioned in Section 2.4 while the other two seemed to occur under the circumstance of near-simultaneous intensification of the Hadley circulation (Figures 6 and 7).
2.6 Additional remarks
A report released by the International MONEX Management
Center for Winter MONEX (1978) disclosed that 63% of all the surges during the months of December from 1968 to 1977, took less than 24 hours to travel from the northern to the southern 11
part of the South China Sea. Rainfall statistics related
to the surges were not reported. But, unquestionably,
cold surges either accompany or precede major rainfall
events in the Malaysian region. The former may be more
frequent since the prediction of heavy rain-spells which
entails the use of predictors from South China or even
Central China is still far from satisfactory.
Controversy still remains as to whether most rain-spells
are linked to travelling disturbances, whose position and movement have so far been chiefly delineated through dubious
analyses over the data-sparse South China Sea and West
Pacific. Hopefully this would be resolved soon by the
continuous picture viewing capability of the recently
launched Japanese geostationary satellite GMS-l. 12
3.0 LARGE-SCALE FEATURES OVER SOUTHEAST ASIA
3.1 Mean earth-atmospheric radiation components
3.1.1 Radiation data
Daily values of various radiation components at each
2.5 x 2.5 degree grid have been derived from radiation measurements made by polar orbiting satellites since June
1974. These radiation components consist of outgoing long wave radiation, albedo, absorbed solar radiation and net radiation. Winston and Krueger (1977) have prepared mean maps of these components for the summer monsoon and have investigated the radiative heating in relation to the summer monsoon over Asia. More recently, Murakami (1978), using a larger data set, has discussed the climatology of the radiation components over East Asia and Southeast Asia.
In this study, the mean maps of radiative heating over
Southeast Asia for the chosen strong and weak monsoon periods were similarly prepared and analyzed.
3.1.2 Radiative heating during the strong monsoon period
The albedo field for the strong monsoon period is shown in Figure 8a. High albedo (greater than 50%) associated with frontal clouds occurred over South China and East China
Sea. Westw~rd of southern China, the albedo was much less--a feature that resembles the mean winter cloudiness map (Sadler and Harris, 1970). Over the equatorial region, 13 a belt of high albedo covered the South Bay of Bengal,
Isthmus of Kra and northeastern Peninsular Malaysia. The maximum albedo region was partly anchored by the near equatorial trough (Figure 9) and partly by the mountain ranges in Peninsular Malaysia and the Isthmus of Kra. The relatively high albedo was associated with intense convec tive activity over Southeast Asia and is certainly a characteristic feature of an early monsoon surge. In the late winter months, large-scale cross-equatorial flow often follows the monsoon surge. The resultant subsidence over the area north of the equator inhibits convective activity
(Lim, 1976). Maximum albedo is then restricted to the
Indonesian region south of the equator.
The long-wave radiation pattern (Figure 8b) was in general a reverse of that of the albedo. The outgoing 2) planetary radiation over Malaysia (between 180 and 220 Wm- was slightly less than over China (between 220 and 240 Wm -2) though the latter's albedo was higher. This is because of the presence of deep convective clouds in the equatorial region as opposed to the shallower stratiform clouds over the continent.
The net radiation field (Figure 8c) reveals that the area of cooling from the north extended southward between longitud~s 97E ~nd lOSE to envelop South Thailand and northeastern Peninsular Malaysia, where cloud cover not only 14 reduced long-wave radiation but also inhibited net absorp tion of short-wave radiation. In the seasonal mean, negative net radiation occurs to the north of l5N.
3.1.3 Radiative heating during the monsoon break
In comparison with the active period, the albedo during the monsoon break (Figure lOa) was less over Malaysia but much higher over Indonesia where the Southern Hemisphere near-equatorial trough was particularly active. The albedo was higher over Luzon Island and the neighboring South
China Sea. This is attributed to increase in cloudiness due to passage of a tropical st~rm across the area (Figure 11).
Consistent with the albedo pattern, there was more outgoing long-wave radiative cooling over Malaysia, and less over Indonesia and in the vicinity of Luzon Island
(Figure lOb). The pattern of net radiation (Figure IDe) assumed a more zonal characteristic over subtropical South- east Asia. Here cooling of the earth-atmospheric system occur~ed poleward of l3N. Over Indonesia, small areas of weak cooling coincided with areas of maximum albedo.
3.2 Surface characteristics of the South China Sea
3.2.1 Data sources
Over the South China Sea, the number of ship reports received at any particular reporting hour (0000 or 1200 GMT) is around ten. The ship data of the required periods were 15 extracted from observational reports published by the 4 PAGASA and Japan Meteorological Agency. Ship data from the region between ION and the equator are seldom found in these publications. They were extracted from the operational charts provided by the Malaysian Meteorological Service.
Besides ship reports, three island stations made daily surface observations.
Composite maps for temperature, wind speed, dewpoint temperature and sea surface temperature (SST) were construc- ted from the surface data for the selected monsoon periods.
Poleward of ION, only the 0000 GMT data were used; south of
ION, the 1200 GMT were employed to supplement the 0000 GMT data set as the latter was found to be inadequate.
3.2.2 Strong monsoon characteristics
The wind speed distribution is shown in Figure 12 (top).
Also shown in the same figure are the resultant vector wind direction and wind steadiness (within brackets) at certain points which were chosen because they were located mostly along a surge trajectory. As indicated by the wind steadi- ness, a near-steady state of the circulation existed. The winds were mainly north-northeasterly north of 20N, somewhat more easterly between 20N and ION, and east-northeasterly
4 Philippine Atmospheric, Geophysical and Astronomical Services Administration. 16 near Peninsular Malaysia and in the Gulf of Thailand. There was therefore indication of channelling of northeasterlies by the South Vietnam coast and the east-west aligned near equatorial trough.
A band of strong wind was aligned along the wind direction. The northeastern coast of Peninsular Malaysia and the Gulf of Thailand lay at the downstream end of this strong wind band. South of 4N and over East Malaysia, relatively light winds prevailed.
A moderately large north-south gradient in the SST
(Figure 12, bottom) was found north of l3N. To the south, the SST was more uniform. A minimum SST « 26°C) was located within the strong wind region. Strong winds (by enhancing mixing and evaporational cooling) and partly cloudy skies (by reducing insolation) were the principal causes of the lower SST.
Like SST, air temperature gradient (Figure 13, top) was large over the northern part of the South China Sea.
However, the coastal effect on air temperature was dis cernible as isotherms tended to parallel the Vietnam coast.
In the Malaysian-South Vietnam region, low air temperature was found to correspond to low SST. The air in this region was part of a cold outbreak moving rapidly southward at a small angle to the SST gradient. 17
On a horizontal surface, dewpoint temperature varies mainly with moisture content. Thus the narrow interval between dewpoint iso1ines north of l2N (Figure 13, bottom) indicates a high rate of absorption of moisture by dry southward flowing air. Between l2N and the equator, moisture intake by northeasterlies lessened greatly as dewpoint temperatures became fairly uniform (24°C).
3.2.3 Weak monsoon characteristics
An unsteady circulation existed during the monsoon break. This was caused by the rapid south-north movement of the equatorial trough and the presence of a tropical storm (Figure 11). As shown in Figure 14 (top), winds were strong near the path of the tropical storm which was roughly l7N. Over the equatorial region, winds were light and variable.
Near the Formosa Strait, the SST was found to be at least one degree higher than in the surge period (Figure 14, bottom). As the break period occurred one month earlier, the higher SST is attributed to the normal monthly difference.
Cooler seawater due to upwelling is expected along the path of the tropical storm, but this was not evident in the ship observations.
Air and dewpoint temperatures are sensitive to circula tion changes, which are reflected in their distribution shown in Figure 15. The tropical storm, by drawing warm moist 18 air from the south, had created a comparatively slack air temperature and dewpoint distributions between Vietnam and the Philippines. As usual, the air temperature and dewpoint in the Malaysian region were relatively uniform at about
26.5°C and 23°C respectively.
3.3 Heat exchange over the South China Sea
3.3.1 Bulk aerodynamic equations
An attempt was made to investigate the heat exchange at the surface of the South China Sea. Two methods are available for evaluating sensible and latent heat transfers
(Qs and Q at the sea surface--the energy budget method e) and the turbulent transfer method. For the energy budget method, the total heat exchange (Q + Q) can be computed s e from the equation (Malkus, 1962),
where O. is the amount of solar radiation reaching the sea ·1 surface, Q the amount of radiation reflected at the surface, r Q the amount of effective outgoing radiation, S the amount b of heat stored in the sea and Q the amount of heat advected v by sea currents. Sand Q are often neglected in the v computation of annual averages.
The turbulent transfer method makes use of the bulk aerodynamic equations given by 19
(3.1a)
= P c Cd (T -T)V (3.1b) o p sao where T and T are the sea surface and surface air tempera- s a tures;
is the specific humidity;
is the saturation specific humidity at sea surface;
L is the latent heat of vaporization;
V is the surface wind speed; o C is the drag coefficient; d is the specific heat of air at constant pressure; and
is the surface air density.
The above equations are derived under the assumption of neutral stability for the atmospheric layer close to the surface. Because the lowest tens of meters of the tropical atmosphere is usually neutrally stable and barotropic (Malkus,
1962; Garstang, 1967), the bulk aerodynamic equations are considered to be suitable for computing heat exchange over the tropics.
Calculations of sensible heat flux (Qs) and latent heat flux (Q ) were carried out for the same locations where e the wind steadiness had been determined (Figures 12 and 14).
The data for surface wind speed, sea surface temperature and surface air tempearture were estimated from the composite 20 maps in Figure 12-15. Saturation specific humidity at sea surface is related to saturation vapor pressure at sea surface and surface air pressure (P ) as follows o 0.622 e s (3.2)
Saturation vapor pressure (e ) for a particular sea surface s temperature can be obtained from the Smithsonian Meteorolo- gical Tables. Approximate mean surface pressures were deduced from surface charts published by the Japan Meteorolo- gical Agency.
Whereas many investigators assumed the drag coefficient
(Cd) over sea surface to be a constant (Riehl ~ al., 1951;
Riehl, 1973; Gray, 1972), some felt it more prudent to regard Cd as function of wind speed (Colon, 1964; Garstang,
1967; Clark, 1967). With the moderately large range of wind speed reported over the South China Sea, it may be advisable to include the variation of Cd with wind speed. A commonly applied expression is that proposed by Deacon and Webb
(1962) ,
(3.3) where V is in m/sec. Combining Eqs. (3.1) and (3.3), we o have
3 -2 1.819(1.0 + 0.07 V )(e - e ) V x 10 Wm (3.4a) o s a 0 p 0
-2 o 1.175(1.0 + 0.07 V)(T - T) V Wm (3.4b) 's o s a 0 21
3.3.2 Results of heat transfer computation
The heat transfer computations are summarized in
Table 1. In the strong monsoon period, points A, B, C, D,
E and F lay approximately along an air trajectory (see
Figure 12, top). At point A, off the coast of South China,
the atmosphere was receiving latent heat from the sea at -2 an average rate of 252 Wm . The rate of intake of latent heat differed very little at points A, B, C and D. This
is because both the relative humidity of surface air and wind speed which control the rate of evaporation of sea water remained almost constant (Table 1). At points E and F in the Malaysian region, the relative humidity along the trajectory increased to over 90%. Though winds were nearly 2 as strong, Q had decreased to about 100 Wm- at F off e the east coast of Peninsular Malaysia. At G near East -2 Malaysia, Q was about 63 Wm due chiefly to low wind e speed. Obviously, most of the latent heat transfer took place poleward of ION.
Sensible heat flux (Qs) from sea to atmosphere decreased 2) rapidly along the trajectory from point A (86 Wm- to -2 point C (23 Wm ). Equatorward of point C, Q was nearly s constant except in the low wind speed region near East
Malaysia.
Over most of the South China Sea, especially to south of lSN, where the sea was of uniform temperature and hori- zontal advection was weak, Q to Q ratio (Bowen ratio) was s e 22 small at around 0.13. Where the sea-air temperature difference was large, for example over the Formosa Strait, the Bowen ratio was twice as high.
It would be interesting to compare the above results with those obtained for the East China Sea. There the cold polar air mass leaving the continent, moves across the warm
Kuroshio current resulting in an extremely large sea-air temperature difference. For the southern part of the East
China Sea, Ninomiya (1975) used the heat energy budget -2 calculation to obtain a value of 279 Wm (575 1ang1eys/day) 2 for Q and 286 Wm- (591 1ang1eys/day) for Q for the e s AMTEX 1974 cold period. Meanwhile, Agee and Howley (1977), employing the bulk aerodynamic equations, arrived at an 2 areal average of 532 Wm- (1099 1ang1eys/day) for Q and e 2 235 Wm- (485 1ang1eys/day) for Q for the same area and s period. There is marked discrepancy in the two ca1cu1a- tions of Qe; the budget method might have underestimated Q e thus giving an unrealistically large Bowen ratio.
The relatively big intake of sensible heat by the atmos- phere over the East China Sea or the Sea of Japan is short- lived. The major proportion of heat energy transported by the monsoon toward the equator is actually latent heat in the form of water vapor. This heat energy is eventually released in the rising branch of the Hadley cell where water vapor condenses. 23
During the monsoon lull, the entrainment of both sensible and latent heat (Table 1) into the atmosphere was heightened by the presence of a tropical storm. In parti- cular, the total rate of heat transfer (Qe + Q in the s) area near its path (at B, C and D) was as large as in the surge period. In the Malaysian region where winds were -2 light, the total rate of transfer was around 85 Wm .
3.4 East Asian Hadley circulation
The main ascending motion near the equator during the active period was apparently confined to the active convec- tive region which encompassed the northeast of Peninsular
Malaysia, South Thailand and South Bay of Bengal. This region, as was noted earlier, lay at the downstream end of the strong wind band which probably formed the low-level branch of the meridional circulation.
A crude picture of the structure of the meridional circulation is shown in Figure 16. In this figure, a vertical cross-section of potential temperature and relative humidity determined from the mean soundings at Paya Lebar
(1020'N, l03°53'E), Kota Bharu (6°l0'N, l02°l7'E), Tanson
Hong Kong (22°l9'N, l14°l0'E), Fuzhou (26°05'N, l19°l7'E) and Nanjing (32°00'N, l18°48'E) was constructed. Except on
January 3, wind observations were often not available at the northern stations. The winds on this date were plotted instead of mean resultant wind~. 24
Between 20N and 30N, where the subtropical jet is usually located, strong westerlies prevailed above SOO-mb.
At low levels, northeasterlies were present, perhaps, indicating a cold outbreak from interior China. Low-level easterlies deepened toward the south, dominating the entire troposphere between ION and 2N, which was close to mean position of the near-equatorial trough. Above 400-mb in the equatorial region, winds were southeasterly represent- ing the return flow of the Hadley circulation. South- easterlies bordered southwesterlies alon b the subtropical ridge at about 15N. At 22N, southwesterlies merged with the subtropical jet.
At higher latitudes, the moisture content of the atmosphere was high near the surface but much less above.
To the south, the moisture spread upward and in the convec tive region, the whole troposphere was almost saturated due to large-scale rising motion.
Another characteristic of the Hadley circulation is revealed by the vertical distribution of potential tempera- ture (8). The isolines of 8 were closely spaced at the cold end of the circulation and widely spaced at the warm end.
Like the sea-breeze, the monsoon is initiated by the difference in heating between land and sea. However, the primary source of energy for maintaining the monsoon circula tion to observed strength must come from latent heat of 25 condensation. Even though the net radiation received by the earth-atmosphere system in the Malaysian region during the period January 1-8, 1975 was small or even negative, the huge quantity of latent heat released in the active convective area counteracted radiative cooling and drove the circulation. 26
4.0 NEAR-EQUATORIAL TROUGH
4.1 Climatology
4.1.1 Mean position of the gradient-level trough
Among the numerous terms applied to describe the low
level pressure zone in the low latitudes, the most popular are the intertropical convergence zone (ITCZ) and the near equatorial trough. The near-equatorial trough is normally difficult to identify as a surface pressure trough but is
readily detected as a directional shear-line in low-level
circulation charts.
As shown in the streamline charts (Figure 17, top)
prepared by Sadler and Harris (1970), the mean gradient level circulation in November over equatorial Southeast
Asia is characterized by the presence of two near-equatorial
troughs which are equidistant from the equator near 4N and
4S. The northern near-equatorial trough has moved southward
from its October position along ION. With the southward movement of the trough, the monsoon to the north of it
begins to affect Malaysia starting from November.
In December, there is a slight southward shift in the
northern trough. By January and February (Figure 17,
bottom), the trough has migrated to south of Malaysia and
is now close to the equator. The trough is usually weak
during the late monsoon. Northeasterlies often prevail over
the whole of Malaysia and consequently there is regular 27
occurrence of the cross-equatorial drift (Johnson and Morth,"
1960).
The trough returns to the November position during the
spring transition (March and April). In May, it is replaced by the southwesterly current which then dominates the
snyoptic pattern throughout summer.
4.1.2 Rainfall distribution across the trough
Mean monthly cloudiness charts compiled by Sadler and
Harris (1970) are, unfortunately, too coarse to provide any hint of the weather distribution across the trough. A rough indication may however be deduced from the monthly rainfall distribution.
In Peninsular Malaysia, rainfall maximum in November
(Figure 18) is located in the northeast which is east of the main mountain range (Figure 2) and north of the mean trough. From the northeast, rainfall decreases rapidly westward into the interior and mountainous region and south- ward along the coast. In December and January, the rain maximum extends southward along the east coast as the trough makes its climatological march toward the equator. During
February, rainfall diminishes sharply and lengthy dry spells occur in most parts of the peninsula.
Clearly, topography has a large influence on the rain- fall distribution when the trough is present. In the early monsoon, the mountain ranges and trough anchor the rainfall 28 maximum to the northeast of the peninsula. South of the trough and west of the mountain ranges are relatively dry.
But in the late monsoon, the trough becomes less persistent and topography loses its effect on rainfall.
In East Malaysia, the spatial rainfall pattern is less controlled by topography for the reason that the ranges, unlike those in West Malaysia, do not lie athwart the mon- soon flow. Further, the monthly rainfall is little related to the mean trough because its day-to-day movement is erratic. As a whole, rainfall increases from November to
January and declines in February (Figure 19).
When compared to the rainfall over east coast of
Peninsular Malaysia, much less rain falls over the west coast of East Malaysia during November and December, but slightly more so during January and February (cf. Figures
18 and 19). This proves once again that topography is not always the prime factor goveraing rainfall distribution
(Ramage, 1971). Presumably, that part of the trough over
Peninsular Malaysia is more active than its more eastern part during the early monsoon; during the late monsoon, the
East Malaysian part of the trough becomes more persistent and produces a rainfall maximum in the state of Sarawak. 29
4.2 Case study
4.2.1 Thermodynamic structure of the trough during the strong monsoon period
From January 1-8, 1975, trough movement was restricted to the length of the Malaysian peninsula. This provides an opportunity to closely examine the trough by means of a dense network of upper-air stations consisting of Paya Lebar at the southern tip of the peninsula, Kuala Lumpur and
Kuantan in the central part of the peninsula, and Penang,
Kota Bharu and Songkhla in the northern part of the peninsula (Figure 20).
Figures 21-24 show cross-sections of relative humidity and virtual equivalent potential temperature (8 ) across ve the trough at 0000 GMT (0730 LT) on certain selected days.
On January 1 (Figure 21), the surface trough was located to south of Kuantan. It extended upward to 400-mb. Whether there was a tilt in the trough is not apparent in the cross- section. In any case, the trough was drawn tilted slightly to north with height because on several occasions when its slope was well-defined by the wind structure, it was tilted in this way. Watts (1955) applied Margules 1 Equation in his analyses of the slope of a type of discontinuity called
"airstream boundary". As will be seen later, the structure of the trough is more complex and its slope cannot be so readily dealt with. Winds at and below 300 m at Kuantan were frequently north-northwesterly. This was the result 30
Gr the influence of land-breeze on synoptic winds. So the presence of near-surface north-northwesterlies at Kuantan does not necessarily mean that the surface trough must be located north of it.
The e pattern indicates that the atmosphere over ve Kuantan was potentially more stable than over Paya Lebar or
Kota Bharu. This was attributed to relatively dry surface layer and cooler early morning surface temperature at
Kuantan. The convective instability at Kuantan, however, underwent large diurnal variation, for as the sun rose, strong surface heating due to cloud-free skies rapidly increased the instability leading occasionally to the deve1op- ment of severe afternoon thunderstorms (Section 6.1.1).
Subsidence caused the lowering of relative humidity at the trough because the only source of dry air was from the upper atmosphere. Relative humidity was higher on either side of the trough, with larger gradients toward north than south.
The low-level trough was displaced southward of Singa- pore on January 3 (Figure 22). There was substantial cross- equatorial flow at the surface resulting in a weakened
trough. In this instance, atmospheric drying did not take place at the trough but north of it, especially in the middle
troposphere. As a consequence, a sharp increase in convec-
tive instability occurred over the central and northern part
of the peninsula. Surprisingly, the atmosphere over South
Thailand was moist. 31 The trough returned to its position near Kuantan on the
4th. From January 4-8 (Figures 23 and 24), it shifted very little. Its structure during this period was reminiscent of
January 1.
Cross-sections of surface pressure and pressure height for January 4 (at the height of the surge) as well as for the entire surge period are shown in Figure 25. The existence of the trough is not obvious in the cross-section for
January 4 but can be distinguished by a slight dip in the mean height of the 850 and 700-mb surfaces over Kuantan for the surge period (Figure 25, bottom).
In addition, pressure surfaces in the upper troposphere over Kota Bharu were considerably raised owing to condensation heating. The approximate atmospheric heating rate can be calculated from measured precipitation. The average rainfall rate over the northeast coast, estimated from three coastal stations, Kuala Trengganu (5°20'N, 103°08'E), Tenang Besut
(5°5l'N, 102°30'E) and Kota Bharu was about 60 mm/day during the surge period. As shown in Holton (1972), the average heating rate per unit mass of air can be calculated as follows:
& ~ Lm _CL _--=-_--;---,- (4.1) c c (P /g) p p 0
-15°C/day 32 where Q is the condensation heating rate per unit mass, L is the latent heat of condensation, m is the mass of water con 2 densed from an atmospheric column of 1 cm cross-section, and -2 finally, P /g ~l kgm cm ,P being the surface pressure and o 0 g the acceleration of gravity.
The computed heating (15°C/day) is, of course, assumed to be evenly distributed over the whole atmospheric column.
As a matter of fact, maximum warming was concentrated in a layer between 500 and lSO-mb where mean temperature exceeded that of the weak monsoon period by less than one degree
(Table 2).
Rainfall rates were considerably greater on January 4 averaging 167 mm/day. The corresponding heating rate was 4 42°C/day or equivalent to roughly 4.2 x 10 joules of heat 2 re 1 ease d lnto"t h e atmosp herlC e r I co 1 umn 0 f1 cm cross-sectlon" per day. But the quantity of heat that went to warming the atmosphere must be very limited as there was no significant temperature increase on the 4th. During this period, there was probably a near-balance between the latent heat generated on the one hand and the adiabatic cooling from ascending motion, radiative cooling from the atmosphere and clouds, and heat energy transported to middle latitudes on the other.
The rate of radiative cooling is in the order of I-2°C/day and in areas of high precipitation, it is much smaller than the adiabatic cooling as will be seen in Section 4.2.4. 33
Ramage (1968) felt that export of heat by Hadley circulation might be the prime regulator of the upper tropospheric
temperature at the heat source. In fact, by comparing the
intense winter circulation of 1963 to the much weaker circulation of 1964, Ramage has shown that the ZOO-mb temperature was generally lower in the tropics in 1963 and markedly more so over the maritime continent where rainfall difference between the two winters was largest.
He therefore speculated that the meridional circulation had siphoned heat from its heat source so much faster during the active period that less heat accumulated near the equator during this period than in the mild winter of 1964.
There is also evidence of warming in the middle tropo- sphere at Kuantan, particularly on the 4th. In an atmosphere where advective processes are unimportant, warm ing and drying must be due solely to adiabatic compression because of sinking motion. We thus find some similarities in the structure of the trough and that of the typhoon
(Miller, 1967): the trough itself corresponds to the eye of the typhoon and the convective zone on each side of the trough is likened to the wall clouds surrounding the eye.
A conspicuous dip was observed in the pressure heights in the middle and upper troposphere at Songkhla. The feature is rather unexpected as rainfall over southern 34
Thailand was not much less than over northeastern Peninsular
Malaysia during the strong monsoon period. The explanation probably lies in the different radiosonde systems used in
Malaysia and Thailand.
Tropospheric heating produces a lower tropospheric air density which in turn lowers surface pressure. This was exactly what had happened in the vicinity of the trough when surface pressure (Figure 26) fell and rose in harmony with rainfall fluctuations.
4.2.2 Structure of the trough during the monsoon break
Because of its large day-to-day displacement (Section
3.2.3), it has not been possible to obtain a steady-state trough structure during the weak monsoon period. Generally, when northeasterlies over the South China Sea were rela tively s~rong (Figure 27), more convective activity was found to the north of trough than to the south. When north- easterlies were weak, a relatively arid condition was noted near the trough and north of it (Figure 28).
4 . 2 . 3 Structure of the trough over East Malaysia
Owing to a lack of upper-air stations, a parallel study could not be made for East Malaysia. Trough struc- ture could only be inferred from rainfall distribution in relation to trough position. Daily rainfall amounts for five coastal stations facing the South China Sea were 35 analyzed in a time section depicted in Figure 29. It is easily seen that a rainfall minimum was centered at the trough, a primary maximum on the poleward side and a secondary maximum on the equator side of trough. There is possibly no outstanding difference between the East
Malaysian and Peninsular Malaysian trough.
4.2.4 Dynamic aspects
To investigate the dynamic characteristics, three triangles were formed from five upper-air stations in
Peninsular Malaysia (Figure 20). The vorticity, divergence and vertical velocity were computed for each triangle using the method devised by Bellamy (1949). It was rather unfortunate that radar winds were not measured at Kuantan, for otherwise, the dynamic variables for an area south of the trough could be determined also. As it is, I could only compute for triangles which were north of the trough and for one which spanned it.
(a) Relative vorticity
Relative vorticity was evaluated at eleven levels from
970-mb (about 300-meter level) to 100-mb. Since wind data were not available at 970-mb at Songkhla and Paya Lebar, I assumed that the winds at this level over the two stations were identical to those at 9l0-mb (900-meter level). 36
Over the two northern regions (Figure 30, top and middle), the shear component contributed mostly to the relative vorticity. Here except for December 31, January 1 and 4, negative vorticity dominated the lower troposphere.
The greatest positive vorticity occurred on January 4 coinciding with the peak intensity of the Hadley circulation.
Aside from January 5 and 6, negative vorticity prevailed in the upper troposphere.
Over the central region (Figure 30, bottom) where vorticity at the lower and middle troposphere was determined primarily by its curvature term, cyclonic vorticity clearly dominated the entire troposphere, with the possible exception of January 7-9. There was no sign of any signi- ficant increase in the lower tropospheric vorticity on the
4th, suggesting that large vorticity changes were concentra ted mainly in the easterly regime on the northern side of the trough.
(b) Divergence
Over the northernmost region (Figure 31, top), a two layer atmosphere existed in most of the days with conver gence controlling the lower troposphere and divergence the upper troposphere. On the 4th and 5th, however, very strong convergence was noted throughout the troposphere particularly the upper levels. Inspection of the data revealed that winds above 200-mb at Penang were too light 37
during these two days compared to those recorded at the three
neighboring stations. Measurement error in the winds above
200-mb at Penang is suspected.
Closer to the trough (Figure 31, middle), the main
divergence features resemble those in the northernmost region
except that divergence dominated the entire troposphere on
the 3rd. Within the central region (Figure 31, bottom),
divergence was present in most of the upper and middle
troposphere. Other than January 3 and 7, convergence was
generally found in the lower troposphere.
(c) Vertical velocity
Vertical velocity was derived from divergence using the kinematic technique. I neglected the contribution of verti-
cal velocity arising from the slope of the ground due to the
presence of mountain ranges. The usual boundary constraint
(zero vertical motion) was applied at 970-mb and vertical velocities at higher levels were arrived at by integrating the continuity equation upward from 970-mb.
Farthest to the north (Figure 32, top), ascending motion occurred virtually at every level and was especially
large from January 4 through 5. As a crude check to this result, the order of magnitude of vertical velocity was estimated independently from the thermodynamic equation.
Scale analysis of the thermodynamic equation in log-pressure 38 coordinates for deep tropical systems yields the following expression (Holton, 1972),
rw1~ JL (4.2) - c p where r , the static stability parameter, is close to 3°e/km for the tropical atmosphere, w*. the vertical velocity in log-pressure coordinates and Q is the diabatic heating. In areas of heavy rainfall, Eq. (4.2) carries the implication that condensation heating is offset almost entirely by adiabatic cooling . For January 4, the condensation heating • (Q/c ) has been estimated to be 42°e/day (Section 4.2.1). p Therefore the vertical velocity (W) is in the order of . Q rc -- 16 em/sec p
3 -3 The density of air at SOO-mb is about 0.7 x 10- gm em
(near the mean density of air in the troposphere). The order of magnitude of vertical velocity in units of mb/sec is then
- - pgW 4 -.." - 110 x 10- mb/sec (4.3)
The average vertical velocity between 600 and 300-mb over the two northern triangles (for January 4) as de~ermined from the continuity equation does not depart significantly from
(4.3). As can be recalled, Q/c was obtained from rainfall p recorded along the coast which was supposedly free from 39 terrain effect. Hence Figure 32 should provide a reasonable approximation to the vertical motion within the layer between the surface and 300-mb, which arose from synoptic influence.
In comparison, a subtropical cyclone over the west coast of
India (Miller and Keshavamurty, 1968) has vertical velocity -4 in the order of 210 x 10 mb/sec ( ~30 em/sec) between 600 and 300-mb.
Above 300-mb, the magnitude of the vertical velocity became unrealistic even granting that high quality data were used (O'Brien, 1970; Ramage, 1971). The reason is that error in the vertical velocity derived from errors in the divergence estimates adds up progressively at every integration step.
So vertical motion, instead of damping out, ends up very large near the tropopause. Certain schemes have been designed
(Lateef, 1967; O'Brien, 1970) for adjusting the velocity to more acceptable value. Such schemes are good if there is no serious problem in the wind data. In this study, for example, the huge error in the velocity near the tropopause would require excessive correction to be made throughout the troposphere.
Ascending motion was less toward the trough (Figure 32, middle and bottom) and reversed occasionally to subsidence near the trough. On the 3rd, when the trough was farthest south, subsidence prevailed over the Malaysian peninsula. 40 4.3 Oceanic trough
Trough structure over the South China Sea can be deduced from the preceding discussion. Over the sea, the trough is neither distorted by topography nor affected by significant diurnal variation. The convective zone on each side of the trough is more defined as evidenced in the GMS-l photographs compiled by Sadler (1979) during the Winter MONEX 1978.
Furthermore, mass convergence should be confined to the boundary layer. To verify the last statement, the geostro phic wind in the Gulf of Thailand (in which the pressure gradient was often directed from NNW to SSE) was calculated for the surface, 850 and 700-mb for January 4 using the pressure or height difference between Kota Bharu and Bangkok.
The values obtained are 19.4, 14.2 and 7.6 m/sec respectively.
At the surface, the actual wind speed was about half the geostrophic speed. On the other hand, the recorded and geostrophic winds at 850 and 700-mb were in good agreement in both direction and speed despite the fact that the region concerned is near the equator.
Thus, frictionally induced convergence in the boundary layer is the cause of vertical ascent of mass and moisture on either side of the trough, being in this case very much greater on the poleward side. It is immediately obvious that the conclusions arrived at by Charney (1968) on the mainte nance of the trade-wind trough, will be valid here. 41
Accordingly, a monsoon outbreak supplies the impetus to the following series of events north of the trough:
(1) increased low-level shear (positive vorticity) and
convergence,
(2) increased vertical flux of moisture,
(3) increased cumulus convection with release of
latent heat,
(4) a temperature rise especially in the middle and
upper troposphere,
(5) surface pressure falls,
(6) meridional circulation further intensifies, and
(7) further increase in positive vorticity and
convergence.
Equatorward of the trough, westerlies strengthen in response to falling surface pressure. This in turn gives rise to the same sequence of events i.e. events (1) to (7) along the southern convergence.
Near the latitude of the trough, enhanced subsidence leads to more warming and drying of the atmosphere. It is unclear, however, why sinking motion should exist at the trough. The mechanism involved may be identical to that operating within the typhoon eye.
The instability mechanism described above for the development and maintenance of the trough is the so-called 42
CISK. CISK has been introduced successfully in the theoreti cal study of tropical cyclones (Charney and Eliassen, 1964) and trade-wind trough (Charney, 1968; Holton and Wallace,
1971). Moreover, in the investigation of the global origin of tropical disturbances, Gray (1968) concluded that this mechanism accounts for the generation of most tropical perturbations.
An idealized model of the near-equatorial trough over the South China Sea is presented in Figure 33. Two zones of maximum convective activity are placed 2 degrees from the trough axis. Within 1 degree of the trough, low-level winds are weak and directed toward the trough axis. There is possibly weak rising motion near the surface at the trough and it is capped at approximately 700-mb by subsidence. Above
300-mb, air streams mostly northward along the upper branch of the Northern Hemisphere Hadley circulation. The flow often splits into east-southeasterly and east-northeasterly airstreams above the convergence south of the trough (Figure
34). Thus part of the air flows southward along the upper branch of the Southern Hemisphere Hadley circulation.
4.4 Comparison with earlier studies
In the forties and fifties, there was a growing interest in the low-latitude circulation which was largely brought about by the expansion of air traffic into the tropical and 43 equatorial regions. In a paper on the equatorial general circulation, Fletcher (1945) reasoned that multiple conver gence lines (lTC's) could be created and maintained through radiation processes. He pictured two main lTC's, one on each side of the equator separated by a zone of westerlies.
High pressure and general subsidence were assumed along the equator.
It was then widely held that the convergence and the trough coincide. Yet observational studies by Alpert (1945) and Simpson (1947) provided evidence to show otherwise.
Subsequently Flohn (1960) studied the trough over the equa torial Atlantic and uncovered a single convergence on the south side of the directional shear zone.
With the advent of satellites, Sadler (1963) combined conventional surface data with TIROS cloud data to build a model of the eastern North Pacific trough. In this model, the trough is associated with a relative cloudiness minimum. It separates two convergence zones with the more southern conver- gence much more active than the northern one. Sadler's model was supported by Ramage (1974) whose interest centered on the near-equatorial trough over the Indian Ocean.
Similarly, the Southeast Asian trough possesses a double-convergence structure. It differs from the eastern
North Pacific or Indian Ocean trough in the location of the primary convergence. It has also been ascertained that 44 appreciable subsidence exists at the trough and that its overall structure is analogous to that of the typhoon.
CISK has been cited to be the principal maintenance and development mechanism for the trough. A different theory has been advanced by Ramage (1974) who hypothesized that "The trough is a heat trough coinciding with a sea-surface temperature maximum which in turn is maintained and enhanced by insolation through relatively clear skies". If the trough is under such thermal control, its maintenance and development can be discussed in terms of Pettersen's development equation
(Falls, 1970; Ramage, 1970c). It is conceivable that the near-equatorial trough may be partly under thermal control, but once the circulation accelerates, the thermal factor may become less assertive. The following section focuses on the subject of radiative influence on the trough intensity and movement.
4.5 Local changes in the intensity of trough
Figure 35 shows the hourly surface pressures along the east coast of West Malaysia on January 4 and 5, 1975. Diurnal changes were detected in the meridional pressure distribution on the 4th, apparently related to uneven distribution of cloudiness. Because of the more cloudy skies over the northern stations, surface cooling during the night and surface heating during the day were greatly inhibited. Thermal effect there- fore produced lower pressures at Kuantan and Mersing than 45 those at Kota Bharu and Kuala Trengganu between 1000 LT and
1600 LT and a reversal of pressure gradient between 1800 LT and 2400 LT. These local pressure changes induced a discontinuous shift in the surface trough identical to those observed in East Africa (Johnson, 1962).
On the 5th, there was a gradual northward shift of the trough. Rain ceased between 1400 LT and 2200 LT at Kuala
Trengganu and from 2000 LT onward at Kota Bharu (Figur~ 36).
The movement of the trough exerted a noticeable influence on the surface winds but it is deemed more appropriate to cover this subject in conjunction with the land/sea breeze circulation. 46
5.0 LAND/SEA BREEZES
5.1 Observations of coastal surface winds
5.1.1 Weak monsoon period
Mean hourly surface winds for coastal stations in
Peninsular and East Malaysia are depicted in Figures 37-40.
In the weak monsoon period (Figures 37 and 38), land/sea breezes strongly influenced the surface winds. Generally, the winds were feeble during the night and morning and had a seaward component, whereas during the day, the winds increased in speed and had an onshore component.
Detailed examination of the winds reveals three main wind classifications according to the manner in which the wind direction changed. At several northern stations
(e.g. Bayan Lepas, Kota Kinabalu and Miri), the winds veered through a diurnal cycle. These stations lie between 4N and 7N and presumably the Coriolis parameter at these latitudes is sufficiently large to affect air motion.
Winds with almost pure land and sea breeze characteris- tlCS.( or antltrlp":lc....Wlnd s 5) occurredat Kota Bharu an d
Kuala Trengganu. These winds blew nearly perpendicular to the shoreline ~nd reversed direction near midday and midnight.
5 This term refers to winds which result from a balance between frictional and pressure gradient forces. 47
At several stations (Mersing, Malacca, Sibu, etc.), wind directions swung about the mean wind (Table 3). The directional oscillation was fixed by the synoptic wind and the position of land with respect to sea. For instance, the synoptic winds at Mersing (east coast of Peninsular
Malaysia) and Malacca (west coast) were northerly or northwesterly. The surface winds at Mersing veered from northwest to northeast in the morning and midday, and backed to northwest during the night, while at Malacca, the winds backed from northeast to west in the late morn ing and veered to northeast later during the night. In another example, the general winds were southwesterly along the central and western part of the East Malaysian coast (Bintulu and Sibu). There the surface winds tended to turn clockwise from southeast to northwest in the morning and afternoon, and rotated in the opposite direc tion in the evening and near midnight.
Large turnings in the wind direction were generally observed between 1000 LT and 1200 LT and between 1800 LT and 2400 LT. During these intervals, the winds changed from land to sea-breezes or vice versa. At other times, the rotation rate was relatively small.
5.1.2 Strong monsoon period
Only the northeast coast of Peninsular Malaysia was affected by vigorous monsoon conditions (Section 3.2.2). 48
The winds (Kota Bharu and Kuala Trengganu) exhibited
little diurnal variation (Figure 39). However, along the
sheltered or less exposed coasts, the three types of surface
wind mentioned earlier were present. At Miri (Figure 40),
winds veered over a 24-hour period, while those at Ma1acca
and Bintu1u displayed antitriptic wind features. Winds at
several other stations (e.g. Kuantan and Kota Kinaba1u)
seemed to belong to the third type.
5.2 Theoretical discussion
5.2.1 Brief review of theoretical studies
Early theoretical studies of sea-breeze were based on
analytical models which are given by
du 1 dP (5.1a) dt - fv = - p ax - H + Fx
dv dt + fu (5.1b) where
H represents the driving force for the sea-breeze
circulation and is a function of the difference
between the air temperature over land (T and L) that over sea (T S); F and F are the x and y components of the frictional x y force;
u and v are the x and y components of wind velocity;
f is the Corio1is parameter; 49
P is the large-scale pressure; and
p is the air density.
The x-axis is taken to be perpendicular to the coast. The large-scale pressure force in Eqs. (5.1) permits the investigation of the effect of the syno~-ic wind on the sea breeze but is omitted in most works.
For H, Schmidt (1947) assumed a horizontal and vertical temperature distribution in which the amplitude of the temperature wave at the surface reaches its maximum at a distance inland from the coast equals to half the wavelength of the temperature oscillation. The temperature wave is related to the density variation of the atmosphere and this together with the divergence and convergence of compressible air are supposed to be the only two causes of pressure changes. For the frictional force, Schmidt applied the Gu1dberg-Mohn hypothesis as follows:
F k(z)u (5.2a) x F k(z)v (5.2b) Y 1], where k(z), of dimension [t- is referred to as the co- efficient of resistance. Further simplification requires that the amplitude of the temperature wave and k(z) decrease exponentially with height.
Haurwitz (1947) demonstrated with the help of the circulation theorem that the presence of friction shortens the phase shift between (T -T maximum and the wind speed L S) 50 maximum. He also showed the effect of the Coriolis force on the diurnal shifting of the sea-breeze direction. The
Guldberg-Mohn formulations for F and F were again used. x y For the driving force, he chose
H(t) = C(T -T) = ~ + B cos Qt LS 1T 2 where Band C are constants, and Q is the angular velocity of earth rotation, on the basis that generally at Boston,
"the air over land does not become as much colder than the air over water at night as it becomes warmer in daytime".
Pierson (1950) improved on the theoretical considerations of Schmidt. He expressed (T -T in terms of a Fourier L S) series which thus closely approximates the true temperature variation. Additionally, he applied an improved form of the friction which involves the kinematic coefficient of eddy viscosity (v) as follows:
2 a u F V --2 (5.3a) x az 2 v F = v a (5.3b) y az 2
Probably the most involved analytical model was devised by Defant (1951). His model consists of the linearized equations of motion and heat transfer and the equation of continuity. A simple harmonic function dependent on the length of the circulation cell replaces (T -T and the L S) friction terms are introduced through Eqs. (5.2). 51
In all these analytical treatments, various restrictive assumptions have to be made in order to render the equations
tractable. As can be noted above, the temperature distribu-
tion has often been specified beforehand. This amounts to reducing the problem to a pure dynamical one. However, in spite of the necessity for simplification, analytical models have been successful in accounting for the broad features of the sea-breeze. Recently, Neumann (1977) showed that by neglecting the nonlinear terms in Eqs. (5.1), an expression can be obtained for the rotation rate of direction of land and sea-breeze which is consistent with observations.
In the age of computers, the nonlinear equations may be replaced by finite difference equations which are then handled conveniently by numerical techniques. One of the first numerical models was attempted by Pearce (1955), and many more theoretical investigations of increasing complexity have followed (Fisher, 1961; Estoque, 1961; McPherson, 1970;
Neumann and Mahrer, 1971).
As the principal aim here is to seek a simple theoretical explanation of the general characteristics of the surface winds along the Malaysian coasts, I have chosen an analytical approach. An elementary 2-dimensiona1 model analogous to
the one employed by Haurwitz would be suitable for this
purpose. The model is given as 52 dU 1 dP at - fv = - P ax - R(t) - ku (5.4a)
dV I dP at + fu = - P ay - kv (5.4b)
As in Pierson (1950), the advection terms in Eqs. (5.4) have been dropped and the final solutions for u and v will be restricted to surface winds near the coast.
5.2.2 Choice of coordinate system
In the model as given by Eqs. (5.4), the coastline is assumed to run south-north along the y-axis. For the east coast, a coordinate system is chosen such that positive x increases toward east and positive y toward north. The selected coordinate system for west coast has positive x and y increasing toward west and south respectively.
There is no question on what we mean by east and west coasts of Peninsular Malaysia. As regards to East Malaysia,
I have defined the coasts bordering the Sulu Sea and the
Celebes Sea as the east coast, and the long coastal stretch adjoining the South China Sea is referred to as the west coast.
The coast is normaliy not oriented south-north. To facilitate the comparison of observations with theoretical results, the wind direction at each station was rotated through an angle such that the coast at which the station 53
is located would then be aligned along the y-axis (Figure
41). The results are shown in Figures 42-45.
5.2.3 Temperature difference between land and sea
Before going on to the solution of Eqs. (5.4), we have to specify R(t) and k. For (T -T investigators fre- L S)' quently assume a simple sinusoidal function (e.g. Raurwitz,
1947; Chiu and Estoque, 1964). As indicated in Table 4, the mean air temperature over coastal areas was close to the air temperature over sea during the weak monsoon period and varied mostly within 1°C of the mean air temperature over sea during the strong monsoon period. Therefore I decided to approximate (T -T by the first two harmonics of a L S) Fourier series for temperature as
(5.5)
-5 -1 0.5 7.3 x 10 sec . Mean hourly temperatures n2 = at Miri in the weak monsoon period were fairly representative of the diurnal variation in this region. They were used to
Thus evaluate the constants aI' a Z' ¢l and ¢Z in Eq. (5.5).
(5.6)
In Eq. (5.6), t begins at 0000 LT and maximum temperature difference occurs near 0500 LT and 1400 LT respectively. The same equation was used for the strong monsoon period 54 because, basically, apart from the northeast coast of
Peninsular Malaysia, there was no significant difference in the diurnal range of temperature between the two periods.
A, the driving force per degree temperature difference c between the land and the sea, may be estimated as follows:
The air temperature over land at 1400 LT was around 29°e and this maximum temperature extended approximately over 50 km inland of the coastal plain (see mean maximum temperature maps prepared by Dale, 1963). The temperature over sea was about 26°e. Supposing that the sea-breeze circulation stretched to about 30 km seaward from the shoreline, then maximum horizontal temperature gradient was 3°e/80 km approximately. In the tropics, the sea-breeze may reach a height of 1300-1400 m (Defant, 1951). However, due to the relatively cloudy condition during the winter monsoon, the sea-breeze circulation could be shallower. Here it is assumed that the sea-breeze vanished at 900-mb (about 1 km).
The temperature at this level would be uniform over the coastal areas and in the mean about 2l oe (Clarkson, 1957).
If it is further assumed that constant lapse rate existed within the shallow layer between the surface and 900-mb at
1400 LT, then the lapse rate over land (Y is 8° per km L) and the lapse rate over sea (Y s ) is 5° per km. The driving force near the surface is
A 55 l',p where l',: is the mesoscale pressure gradient across the shore- line. That is
A
1 = (5.7) p /),x o where P and T denote the surface pressure and temperature L L over land at 1400 LT; P and T denote the surface pressure s L and temperature over sea at 1400 LT; and R is the specific gas constant. Thus
AA A = = (5.8) o Max I TS-TL 1- 3
Substituting in the appropriate values for temperature, pressure and lapse rate in Eq. (5.7), we have A = 2.065 x o -2 -2-1 10 cm sec deg .
5.2.4 Coe~ficient of surface resistance of the coast
In the investigation of air motion in low latitudes,
Gordon and Taylor (1975) suggested two methods for estimating the coefficient of resistance (k) for ocean surface. One procedure is to make use of tan a = vlu = kif where a is the angle between the isobars and the wind direction. In this equation, the x-axis is oriented along the isobars and balanced frictional flow is assumed. The equation can be 56 applied to synoptic or climatological charts of the wind and pressure fields. Their second method uses the follow- ing relationships
~ 1 at kv - (5.9) = - p az and
~ g ~ ~ TO JH pf(V-Vg) dz (5.10) 0
~ 7 where TO is the surface stress vector, Vg the geostrophic wind vector and Hg the height of the boundary layer. Eq. (5.10) is inconsistent in view of Eq. (5.9). A correct relationship between surface stress (TO) and winds was shown by Sheppard and Omar (1952) as
where v g and v n are the components of the geostrophic and actual wind normal to the surface wind direction. However, the values for k given by Gordon and Taylor seems to be in the right order of magnitude. With data obtained from the
Marshall islands, they estimated k to be 2.8 x 10-5 sec-l.
Another approach to determine an approximate value for k was furnished by Fujita et al. (1969) who made use of the equation
kJ v . VQ - QD (5.11) 57 where J is tI,e relative vorticity, Q the absolute vorticity and D the divergence. This equation is obtained from the vorticity equation by discarding the solenoid term and the horizontal gradient of vertical motion, and assuming steady- state flow. From the fields of cloud motion determined by 5 1 ATS-1 photographs, they estimated k = 3.0 x 10- sec- for the eastern Pacific region, which is not far from the value obtained by Gordon and Taylor.
In reality, k is rarely evaluated as compared to other surface frictional parameters such as the drag coefficient
Cd (a non-dimensional quantity), and the aerodynamic rough- ness parameter z (of dimension [t]). The latter two have o been extensively evaluated for certain large regions in the
Northern H~misphere as a function of latitude and season
(Kung, 1963).
The surface frictional characteristic, specifically, the drag coefficient, for shallow water about 7 meters deep
(e.g. lake and coastal water) has been known to behave similarly to that over open ocean (Hicks ~ a1., 1974).
Although k and Cd are not dimensionally compatible, it is not unreasonable to treat the value for k over coastal water to be the same as that over open sea, i.e. k = 2.8 x 10-5
Literature dealing with measurement of k over land surfaces appears to be unavailable. Nevertheless, since 58 and
1 b.T T o kV - - - = o p b.z o PHo g ' therefore
k = (5.12)
We shall apply this relationship in a crude estimate of k over the coastal region.
Sutton (1953) summarized the drag coefficients for several natural surfaces with reference velocity of 5 m/sec at a height of 2 m. Some of his values for Cd are 0.001 for mud flats or ice, 0.008 for thick grass up to 10 cm high and 0.016 for thick grass up to 50 cm high. In a micro- meteorological study at Melbourne Airport (about 25 km from
Port Phillip Bay), Brook reported Cd ranging from 0.0011 to
0.0115 for various directions of the 150-foot wind.
The height of the boundary surface above a fairly flat surface for moderate wind speed can be taken as 800 m (Brunt,
1939). If V = 8 m/sec, H = 800 m, then for Cd = 0.016, o g 5 1 -5 -1 k = 16 x 10- sec- and for Cd = 0.001, k 1.0 x 10 sec .
It is quite possible that the average k for the coast lies -5 -1 somewhere between 2.8 x 10 sec (sea surface) and 5 1 16 x 10- sec- (a fairly rough land surface). In this study, k is arbitrarily fixed as the average of these two -5 -1 values i.e. 9.4 x 10 sec . 59
5.2.5 Prevailing wind at the coast
The greater friction of a land surface results in the speed of the prevailing wind over the coastal plain to be less than over the sea. As an example, during January 4 and 5, the surface winds over the northeastern coast of
Peninsular Malaysia were not tempered by land/sea breezes.
The friction effect on the winds reduced the speed to 30-40% of that observed over the South China Sea between Peninsular and East Malaysia.
The synoptic winds at the coast are assumed to be steady; they satisfy the frictional flow relationship as given by
G - - fv + ku (5.13a) x s s
G = fu + kv (5.13b) y s s where u and v are the x and y components of the synoptic s s winds, G and G stand for the large-scale pressure gradients x y
1~ 1 _ap and respectively. p ax p ay
5.2.6 Application of the model to the equatorial region
With R(t) and k known, Eqs. (5.4) can be solved analytically following the procedure as detailed in Appendix
A. Thus
u
2 L M.sin(Q.t - ~. + X.) (5.14) i=l 1 1 1 1 60 where k tan x. = (5.15 ) l. Q. l.
(5.16)
and
2 v - - L N. cos(Q.t - ~. + ~.) (5.17) i=l l. l. l. l. where 2kQ. l. tan ~. = (5.18) l.
A. f N. l. (5.19) l.
It should be emphasized that the solutions as'given by
Eqs. (5.14) and (5.17) can be arrived at only by suppressing the transient term involving exp[-(if+k)t]. The omission of this term, as pointed out by Forsythe (1949), would result in the solution becoming infinite at latitude 30 0 when k = 0 i.e. in free air. In defense, Haurwitz (1950) argued that even if the transient term is retained as suggested, it would quickly be damped out. Further, friction has been shown to be an important factor in regulating the sea-breeze 61 mechanism and its inclusion in the model is essential at least in the theoretical analysis of sea-breeze at the surface layer. 5 -1 Bea r ingin mind 1.: ';1 at f.1 = 7. 3 x 10- sec , f.1 = 2f.1 l 2 1 5 -1 and k at the coast has been chosen to be 9.4 x 10- sec we have, for the zone between 3N and the equator, f < 0.8 x 5 l 10- sec- and is negligible compared to f.1. or k. J. 5 In the northern areas between 5N and 7N, f ~ 1.5 x 10- -1 sec , therefore it should not be ignored when compared to f.1. or k; but then, f2 is at least one order of magnitude J. 2 2 smaller than f.1. or k . J. 2/f.1. 2 Neglecting terms with f and f2/k2, J.
kG +fG 2 x y u ~ M. sin(f.1.t-~.+X.) (5.20) = 2 k i=l J. J. 1 J. where A. k 1 tan x. and M. = 1 st. J. J.
fG -kG 2 x y v ~ cos(st.t-~.+~.) (5.21) 2 N. k i=l J. J. 1 1 where 2kf.1. A.f 1 1 tan ljJ. = and N. = 1 1
Substituting G and G from Eq. (5.13) into Eq s , (5.20) and x y (5.21) gives 62 2 u = u L M. sin(n.t-cf>.+X.) (5.22) s 1 111 i=l
2 v = v L N. cos(n.t-cf>.+lJi.) (5.23) s 1 111 i=l
The constants, M., N., X., and lJi. were computed and the 111 1 results are shown in Table 5.
(a) Weak monsoon period
During the monsoon break, the synoptic winds over
Malaysia were light and variable.
Case (i): u = v = 0, f = (east and west coast) s s ° From Eqs. (5.22) and (5.23), we have
2 u L M. sin(n.t-cf>.+X.) i=l 1 1 1 1 and v = 0. In this case, the land and sea breezes blow perpendicular to the coast. This type of surface wind has been found in southern and northern Malaysia.
-~ -1 Case (ii): u = v = 0, f = 1.5 x 10 J sec (east and s s west coast)
The computed winds for east and west coasts are shown in
Figure 46 (a and b). These winds veer in the course of time.
Hence under light synoptic winds, surface winds at some northern stations (north of 4N) behave like sea-breezes in the mid-latitudes. The computed speed of sea-breeze has nearly the same magnitude as those recorded. The observed 63 land-breeze was, however, very much weaker. This is in view of the fact that over land, air motion is greatly curtailed by the increased stability of the boundary layer during the nigh t .
Over southern peninsula, the general winds were mostly light northerlies. The wind direction was almost parallel to the coasts near Kuantan (east coast) and Sitiawan (west coast).
Case (iii) u 0, v = -2 m/sec, f = (east coast) s s ° u 0, v 2 m/sec, f (west coast) s = s = = ° The direction of the computed wind for the east coast (Figure
46, c) turns clockwise in the morning and anticlockwise in the night. This agrees with the observed winds at Kuantan
(Figure 42). In the case of the west coast, the wind direction turns in the opposite sense (Figure 46, d).
Farther south along the east coast of the peninsula, the general winds had a large landward component. Apart from having an oscillation, the sea-breeze (see Mersing in
Figure 42) was clearly strengthened and prolonged by the synoptic winds. Over the west coast at Malacca, it was the land-breezes which were reinforced by the synoptic winds.
At Sibu and Bintulu in East Malaysia, the prevailing winds were southwesterly and parallel to the coast.
Case (iv): u v -2 m/sec, f (west coast) s = °'s = = ° 64
The theoretical winds (Figure 46, e) veer in the morning and
back in the evening.
(b) Strong monsoon period
In the South China Sea off the northeast coast of West
Malaysia, surface wind speed averaged about 9 m/sec. In
other parts of the South China Sea between East and
Peninsular Malaysia, wind speed was smaller with average
5 m/sec. 5 Case ( i) : u = -3 • 5 m sec, v =0 , f = 1.5 x 10- s / s -1 sec (east coast) -5 u 3.5 m/sec, v = 0, f 1. 5 x 10 s s sec-l (west coast)
Along the east coast, the sea-breezes, aided by the prevailing winds, appear much earlier in the morning and retreat much
later in the night (see Sandakan in Figure 45). For the west coast (Figure 47, b), the land-breezes are reinforced by
the prevailing winds as was reported by Bayan Lepas and
Sitiawan (Figure 44). - 5 Case (ii): u = O,v = 3 . 5 m sec, f 1.5 X 10 s s / = -1 sec (west coast)
The theoretical winds (Figure 47, c) back in the morning and
veer in the evening, in good agreement with the observed winds at Kota Kinabalu in East Malaysia (Figure 45).
At Miri which experienced weaker synoptic winds, the
surface winds veered in a diurnal cycle. Therefore, as the 65
Coriolis term is small, the surface winds would veer in a complete cycle if the synoptic winds are light and variable or blowing feebly perpendicular to the shoreline.
The surface winds at Malacca, Kuantan and Mersing
(Figure 44) in Peninsular Malaysia oscillated about a mean direction as was the case during the weak monsoon period.
5.2.7 Rate of change of wind direction
In view of Eqs. (5.13), Eqs. (5.4) can be rewritten as
au - R(t) + f(v-v ) + k(u -u) (5.24a) at s s
av = feu -u) + k(v -v) (5.24b) at s s
Following Neumann (1977), the rotation rate of wind direction can be derived from Eqs. (5.24) as follows: av au aex. u at - v at at v 2 o
f ( UU +vv )+k(uv -vu )] - f + s s s s + ~R(t) (5.25) [ V 2 V 2 o 0
2 2 2 where V = u + v and ex. arctan(v/u). o The rotation rate is thus dependent on the Coriolis parameter, the large-scale term and the mesoscale pressure term. For southern Malaysia close to the equator, the rate of turning should depend solely on the second and third term of Eq. (5.25). 66
Eq. (5.25) can be used to examine the variation of the theoretical winds. We consider the case of u v = 0 s s 5 -1 and f = 1.5 x 10- sec Then
dO. - f + --.:l- H(t) dt V 2 o
= - f - ~ N. cos (n . t - cf> • +tjJ • )] [ A. cos (n . t - cf> (5.26) V [ii=l 1 1 1 1 i=li 1 1 1.1 o
The rotation rate due to the mesoscale pressure term (second term in Eq. (5.26)) was plotted against time in Figure 48.
It shows that between 0200 LT and 0900 LT, and between
1400 LT and 1900 LT, there is positive (anticlockwise) but very small turning of the wind direction. During these periods, the small pressure term is opposed bv ~lie slightly larger Coriolis parameter which thereby turns the winds clockwise. Large rotation rate is found near 1100 LT and
2300 LT, and is primarily due to the pressure term which now acts in the same sense as the Coriolis term. The theoretical results concur with observations (weak synoptic wind cases) except that the large turning in the evening appeared slightly earlier.
With the presence of moderately large synoptic winds, the large-scale term has to be taken into account. The discussion embracing the three terms becomes more complicated.
But judging from the theoretical winds in Figures 46 and 47, 67 the times of maximum rotation rate would still be around
1100 LT and 2200 LT if the synoptic winds parallel the coast, that is, the times of maximum rotation rate in this case are determined largely by the mesoscale pressure term.
If the synoptic winds are perpendicular to the coast, the large-scale term would significantly affect the times of maximum rotation rate (Figure 47, a and b).
5.2.8 Further comments on the model
It is easily realized that the validity of the above theoretical analysis depends very much on the estimation of the driving force (A ) and the coefficient of surface o resistance (k). A affects proportionately the magnitude of o u and v components of the theoretical winds (see Eqs. (5.20) and (5.21)) and its accuracy does not really change much the result of my earlier discussion. More important is k which modifies not only the constants M. and N. of Eqs. ~ 1 (5.20) and (5.21) respectively but also X. and ~ .. In 1 ~ reality, X. and ~. determine the diurnal shift and the 1 ~ rotation rate of the wind direction.
As has been stated in Section 5.2.4, it is likely that -5 k for the coastal region falls between 2.8 x 10 and 5 -1 16 x 10- sec Using these two limits, I computed the constants M., N., ~. and the times of maximum rotation ~ ~ ~ rate based on Eq. (5.26). Table 5 summarizes the results.
For comparison, the values computed for the case k = 9.4 x 68 -5 -1 10 sec are also inserted in this table. It is seen that M. and N. (in other words, the wind speed) decrease as 1 1 k increases. The speeds computed for k between 9.4 x 10-5 and 16 x 10-5 sec-l are in the right magnitude as reported.
Besides, higher values of k are more desirable as they give the times of maximum rotation rate closer to the recorded.
On the other hand, N. is inclined to decrease at a faster 1 rate than M. as k becomes large, and therefore the diurnal 1 veering of the wind would become less visible if N is too small compared to M. Taking this into consideration, an acceptable value for k appears to be 9.4 x 10-5 sec-1 •
5.3 Effect of the near-equatorial trough
The modification of the diurnal variation of surface wind by trough movement is suitably exemplified by the hourly wind observations on January 4 and 5, 1975, along the east coast of the peninsula (Figure 49). In the morning and afternoon of January 4, the southward directed pressure gradient force combined with the prevailing northeasterlies to produce the northerly winds reported at Mersing and
Kuantan. The winds became southerly at Mersing on the evening of the 4th and at Kuantan on the morning of the 5th when the pressure gradient redirected toward north because of a northward jump in the trough position. The average pressure gradient (- %} ) between Mersing and Kota Bharu was roughly 0.5 mb/440 km between 0600 LT and 2400 LT on the 5th. 69
This pressure gradient force was the dominant large-scale
influence on the surface winds immediately south of the
trough. Putting f = 0 G = 0 and G .L ~ = , x Y Po ay -3 -2 9.71 x 10 ern sec into Eq s , (5.20) and (5.21), we find
that
2 u L: M. sin(n.t-~.+X.) (5.27a) i=l l. l. l. l.
G v = -X... 103 ern/sec (5.27b) k
The winds evaluated from Eqs. (5.27) are presented in
Figure 50. The observed winds at Kuantan and Mersing corres- pond approximately with the theoretical results, indicating
that the principal elements governing the diurnal variation of the winds over the southern part of east coast were the land/sea breezes and the northward directed pressure gradient set up by the trough movement to the north.
At Kuala Trengganu, the surface winds changed to south- southeasterly at around 1300 LT on January 5. They remained so until 2400 LT when northeasterlies reappeared. The winds nearly paralleled the coast and veered weakly with time.
This distinctly indicates the absence of the land/~ea breeze effect, in which case the wind direction was regulated by the large-scale pressure gradient force and to a smaller extent, the Coriolis force. During this period, the average pressure gradient between Kota Bharu and Kuala Trengganu 70
was 0.6 mb/140 km. The theoretical wind speed between
these two stations (G /k) is thus 3.9 m/sec which corres y ponds favorably with the measured speeds.
A shift in the wind direction signaled the passage of
the trough. At Kota Bharu~ this happened about two hours
later than at Kuala Trengganu.
5.4 Weather associated with the land/sea breezes
In the tropics, several well-known phenomena owe their
presence to land/sea breezes. In Florida, Byers and
Rodebush (1948) demonstrated that the low-level convergence
due to sea-breezes entering the peninsula from both sides is
responsible for the daily formation of afternoon thunder-
storms in the interior. A similar study conducted by Ramage
(1964) confirmed the existence of the phenomenon in the
Malaysian peninsula.
Apart from this, persistent afternoon and evening
thunderstorms are common along the leeward coast. Ramage
linked this weather pattern to convergence of prevailing winds with sea-breezes. Based on the results of a numerical
experiment by Estoque (1962), he visualized that a sea-breeze
front develops at the coast near noon. During the next six
to seven hours, the sea-breezes strengthen~ shift the front
30 to 50 km inland and weaken in the evening allowing the
front to be advected back to the coast by the prevailing winds. 71
The land-breeze is weaker and shallower than the sea breeze and is not expected to generate a convergence that is as vigorous as the sea-breeze front. However, th~ conver- gence of land-breezes from Peninsular Malaysia and Sumatra is sufficiently intense to encourage the formation of squall- lines called Sumatras. Sumatras are common only in the summer monsoon when they are said to contribute significantly to the nocturnal rainfall maximum along the west coast of the peninsula. Nocturnal thunderstorms are not restricted to narrow straits alone but could appear at any bay or any concave coast as was demonstrated by Neumann (1951) for the
East Mediterranean region. The Malaysian-Indonesian island complex also offers many such examples. The most striking is that of the Celebes Island, whose tentacle-like structure promotes convective activity over the slender lands during the day (sea-breeze convergence) and over the enclosed waters or bays during the night (land-breeze convergence;
Ramage, 1979). 72
6.0 MONSOON RAINFALL CHARACTERISTICS
6.1 Rainfall of Peninsular Malaysia
As in other tropical or equatorial areas, rainfall in
Malaysia is controlled largely by diurnal processes.
Several diurnal rainfall studies have been carried out for the peninsula. Ramage (1964) investigated the diurnal varia- tion of the August rainfall and noted the existence of five diurnal rainfall regimes which could be derived from inter actions among the synoptic winds, land/sea breezes, anabatic and katabatic winds and topography.
A more extensive study was pYesented by Nieuwolt (1968).
His study covered the diurnal rainfall of the summer and the winter monsoons, and the two transitional periods. He chose January as a typical winter monsoon month and found that there is little diurnal variation of rainfall at the east coast stations. He reasoned that since the prevailing northeasterlies are strong, local effects on rainfall along the east coast are minimal. Diurnal variation, however, exists inland and along the west coast as the synoptic winds are much weaker after passing over land and high mountain ranges.
The diurnal rainfall regimes obtained by Nieuwolt may be representative of the late winter monsoon but certainly not the early monsoon. As pointed out in Section 4.1.1, though the dominant flow over the peninsula is northeasterly 73 during January and February, the circulation pattern in
November and December is marked by the presence of a per-
sistent trough. It follows that the winds aloft differ in
the early and late monsoons. Additionally, the East Asian monsoon fluctuates substantially in a surge-lull cycle--rainy weather develops in concert with the monsoon surges and the monsoon breaks are relatively dry. These weather patterns
in Malaysia alternate in a time interval of one to three weeks.
Therefore, because of the large short-period fluctuations
in the early monsoon, rainfall could not be adequately investigated on a month-to-month basis, for doing so would mask the important features of the wet and dry periods.
6.1.1 Diurnal rainfall variation during the strong monsoon period
(a) Rainfall stations
Fifteen stations with autographic rainfall records
(Figure 2) were used. Most lie in the coastal plains. Very
few are sited in the interior especially in the mountainous
regions. Grik (between the Bintang Range and the Main Range) and possibly Sungei Tekam (near the southern end of the
central mountain-valley region) may be regarded as mountain- valley stations. As the Bintang Range and the East Coast
Range are low (mainly below 3000 feet), most of the valleys 74
in the peninsula are small or shallow. Local mountain winds
may exist but in most cases may not be sufficiently pro
nounced to create nocturnal rainfall.
(b) Diurnal rainfall regimes
Depicted in Figure 51 are the diurnal rainfall patterns
for the active monsoon period. The average 24-hour rainfall
is indicated within brackets. Grik is not shown because it
had two days of defective records.
In the northeast which lay open to the moist and strong
northeasterlies, rainfall was heaviest. Kuala Trengganu
displayed a prominent rainfall maximum in the morning.
Although a maximum occurred in the early afternoon at Kota
Bharu, most rain was recorded in the morning hours.
Many investigators (Worthley et al., 1967; Ramage ~ al.,
1969; Godbole, 1973; Schroeder ~ al., 1977) have discussed
at length the cause of the nocturnal rainfall peak. Briefly
stated, in a regime of strong winds and cloudy skies, local
circulations such as land/sea breezes and mountain-valley winds are of secondary importance. During the day, there is
radiational warming of the top of the extensive cloud layer
and limited insolation at the earth surface. This would
result in increased stability and reduced afternoon rain.
At night, radiational cooling of cloud top and warming of
cloud base would increase instability and rain.
To the leeward side of the Main Range, a completely
different diurnal regime emerged. The northeasterlies dried 75
up after passing over the mountain ranges and as such
comparatively less rain fell in the northwest region.
Rainfall (Bayan Lepas and A10r Star) occurred chiefly in the
afternoon and evening. The sky was mostly clear and after noon rain was caused by local convection and undercutting of the synoptic winds by the sea-breezes (Section 5.4).
The near-equatorial trough was a zone of little cloudi ness (Section 4.2.1). At the trough, the synoptic winds were light and local influence was strong. As such, the trough zone was characterized by a showers regime (Ramage,
1971). Inland stations (Cameron Highlands, Subang and
Sungei Tekam) and coastal stations (Sitiawan and Kuantan) which were often found near the trough showed a pronounced afternoon maximum. Rainfall was moderate except at Kuantan where it was as heavy as at stations in the northeast.
The southern part of the peninsula was influenced by the westerlies. Little cloudiness was observed and rainfall was small. Majority of the stations showed a slight after- noon maximum which was associated with local convection and convergence of sea-breezes.
6.1.2 Diurnal rainfall variation during the monsoon break
Under weak monsoon conditions, the peninsula was cloud
free most of the time (Section 3.1.3). Clear skies and light synoptic winds favored local convection inland and setting up of local circulations along the coasts. The weather acquired characteristics typical of a showers regime. 76
The diurnal rainfall variation is shown in Figure 52.
In this instance, the rainfall frequency instead of the rainfall amount was used. In recording an occasion of rain, a minimum fall of 0.1 mm per hour was taken.
More rain fell along the west coast than along the east coast, pointing to a reversal of the spatial distribution of the strong monsoon period. Most stations exhibited an after- noon or evening rainfall frequency maximum, which was related to local convection and sea-breeze convergence.
Both Sungei Tekam (central) and Grik (northeast) indicated a morning maximum, but Grik had also an afternoon maximum. To some extent, the observed feature suggests the existence of a mountain-valley regime.
Also showing nighttime and morning maxima were Kota
Bharu, Kuala Trengganu, Bayan Lepas and Sitiawan. The rain- fall pattern over the northeast coast may be attributed to the convergence of land-breezes and northeasterlies which still prevailed though weakly during the major part of this period. An explanation has yet to be found for the nocturnal rainfall at Bayan Lepas and Sitiawan.
6.2 Rainfall of East Malaysia
Published works on the rainfall of East Malaysia are indeed rare. The best account of the rainfall climatology of the area was presented by Nieuwolt et ale (undated). They mentioned most of the obvious features but appeared to have 77 overlooked some less visible though by no means unimportant ones.
A sketchy description of the northeast m~nsoon rainfall has been given in Section 4.1.2. The salient point is that the coastal region of Sarawak and Brunei, northeast coast of
Sabah and some parts of the interior have a rainfaJl peak in
January (Figure 53). This feature is not found in the west coast of Sabah (Labuan, Papar and Kota Kinabalu) which, on the contrary, is relatively dry and has a minimum in February.
Rainfall along this coast increases during the subsequent months and by the time of the southwest monsoon, it ~eceives more rainfall than most of the rest of East Malaysia. In my opinion, these features hint at the stress-differential induced effect (Bryson and Kuhn, 1961; Ramage, 1971). In
January and February, the near-equatorial trough is active in East Malaysia (Section 4.1.2) with its influence on rain fall greatest over Sarawak (across which the mean trough lies) and minimal over Sabah. The northeasterlies flowing parallel to the west coast of Sabah, when subjected to the stress difference between land and sea, diverge at the coast. Consequently, rainfall is depressed in the winter monsoon. In the summer monsoon, the opposite effect on rain fall is brought about by the resultant convergence of south- westerlies. This effect does not seem to operate along the coast south of 4N where rainfall diminishes from January to 78 a minimum in July (Bintulu, Mukah and Sarikei). Thus the equation relating differential wind stress and low-level divergence as developed by Bryson and Kuhn (1961) is not applicable in the vicinity of the equator.
Nieuwolt et al. have also discussed on the diurnal rainfall variation which they found to be primarily of two types: the afternoon and nocturnal maxima. Their study involved only coastal stations. In the deep interior it is likely that convergence of sea-breezes continues to promote an afternoon maximum (Flohn, 1970).
At those coasts exposed to the monsoon (Figure 54), night or morning rainfall is understandably dominant (Kuching and Sandakan). Along the coast from Bintulu to Labuan, the same nocturnal maximum is observed and this pattern shows up invariably throughout the year. Its persistence, I believe, is related to the concave coastline (Section 5.4) and the particularly mild synoptic conditions that prevail all year.
Along the same coast, a solitary station, Kota Kinabalu, experiences an afternoon maximum. This unusual feature was explained by Nieuwolt ~ al. as caused by the westward drift of thunderstorms developed on the slope of Mt. Kinabalu.
6.2.1 Diurnal rainfall variation during the strong and weak monsoon periods
It has been reiterated that East Malaysia was less affected than Peninsular Malaysia by the intensity changes 79 of the monsoon circulation. As a result, average daily rainfall in the active monsoon period (Figure 55) remained much the same as in the monsoon break (Figure 56).
Diurnal rainfall features generally conformed to the monthly mean pattern. The exception is Sandakan which for some unknown reasons, had mostly afternoon rain during the strong monsoon period. The afternoon maximum at Kuching can be explained by the convergence of the sea-breezes and the prevailing southwesterlies during both periods. 80
7.0 SUMMARY
In the surge of January 1-8, 1975, strong monsoon conditions were concentrated in the Peninsular Malaysia
South Thailand-South Bay of Bengal region which became the main rising branch of the East Asian Hadley circulation.
However, latent heat and sensible heat transfers over the seas within this region were small and represented less than half of the total heat transfer that took place in the northern South China Sea.
The Hadley circulation accelerated to peak strength in distinguishable stages. In particular, most of the rainfall in Malaysia occurred about three days after the passage of a cold front over Hong Kong. The heavy rainfall was definitely associated with the intense near-equatorial trough and not in any way linked to easterly disturbance. The intensity of the trough fluctuated with that of the Hadley circulation. It had a double-convergence structure with predominance of subsidence near its axis. The northern convergence was found to be more intense than the southern.
This constitutes the major difference between the Southeast
Asian trough and those of the Indian Ocean and eastern North
Pacific. In many ways the trough resembles a typhoon. These include having CISK as the principal mechanism in their development and maintenance. There was, nevertheless, some evidence of thermal control on the intensity and movement of the trough. 81
The trough and mountain ranges limited the monsoori influ ence to the northeast of Peninsular Malaysia. Here, the local circulations were suppressed greatly or totally absent.
Elsewhere including the whole of East Malaysia, local effects overwhelmed synoptic influence.
In the monsoon break of November 26 through December 5,
1974, mesoscale circulations prevailed at every coast and sizeable mountain-valley. In general, the coastal surface winds had three main diurnal variations. These variations can be subjected to satisfactory analysis by a two-dimensional analytical model. It is inferred from the model that the characteristics of the land/sea breezes are merely a function of synotpic winds, Coriolis parameter, temperature difference between land and sea and surface friction.
In the active period, a rains regime was established in northeastern Peninsular Malaysia. In compliance with such a regime, the diurnal rainfall took the form of a nocturnal maximum. Away from the northeast and within the trough zone itself, the meteorological conditions were typical of a showers regime. The observed diurnal rainfall variation may then be accounted for by one or a combination of the follow- ing causes: local convection, sea-breeze convergence and convergence of synoptic winds and sea-breezes. Over East
Malaysia, the monsoon was gentler and rainfall was not dependent on the relatively weak trough. Local effects 82 dominated everywhere. The diurnal rainfall characteristics did not deviate much from the monthly mean pattern.
During the monsoon break, the diurnal rainfall regimes in both East and Peninsular Malaysia were explainable by the interaction betweer. the various local factors plus synoptic winds.
An important fact brought out by this study is that the near-equatorial trough serves as an anchor to rainfall. The anchor is most effective for a stationary trough. Thus, a stationary trough over Malaysia interacting with a major monsoon surge may seem to be sufficient condition for an out break of heavy rain.
The trough is usually active in the early monsoon and with the mesoscale circulations prevalent in areas sheltered from the monsoon, controls the weather characteristics over
Malaysia. In the late monsoon, the trough weakens with frequent large-scale cross-equatorial flow. As a result, the weather patterns are shaped primarily by local factors. 83
APPENDIX A. ANALYTICAL SOLUTIONS TO LAND/SEA BREEZE EQUATIONS
The derived differential equations for land/sea breezes are given as (see Chapter 5),
au fv G H(t) ku (la) at --x -
av + fu = G kv (lb) at y - where
2 2 H(t) A ~ a. cos(Q.t-~.) ~ A. cos(Q.t-~.) o i=l 1 1 1 i=l 1 1 1
We introduce the following substitutions (in complex notation) to Eq s , (1)
W u + iv c =
G G + iG c= x y
Therefore Eq s , (1) become
aw C t (i f k ) WC G- H(t) (2) a + + C
The solution to this linear differential equation is
2 [Q. sin(Q. t-~ . )+(if+k) cos(Q. t-~ .)] ~ A 1 1 1 1 1 2 i=l i (if+k)2 + Q. 1
+ Cex p [ -(if+k)t] (3) 84
The transient term can be suppressed by assuming W vanishes c when both G = 0 and R(t) = 0 (see Raurwitz, 1947). c Separating the real and imaginary parts of Eq. (3) gives
u =
2 L A. (4) i=l ~
and
(5)
Eqs. (4) and (5) can be rewritten in the form
kG +fG 2 u = x Y L M. sin(Q.t-~.+X.) ~ ~ 1 ~ f2+k2 i=l where k tan X. =- ~ n. 1. 85
and
2 v = - L N. cos(Q.t-~.+ljJ.) i=l 1 1 1 1 where
2kQ. 1 tan 1JJ. 1
A.f 1 86
APPENDIX B. DATA SOURCES
1. Data for surface parameters over the South China Sea
were obtained from the following reports:
a. Surface marine observations, November 1974
January 1975 published by Philippine Atmospheric,
Geophysical and Astronomical Services
Administration (PAGASA).
b. Daily weather maps with synoptic data tabulations,
November 1974-January 1975 published by Japan
Meteorological Agency.
2. Rainfall data, upper-air data and hourly surface observa
tions for Malaysian stations were supplied by the Malaysian
Meteorological Service.
3. Professor T. Murakami kindly furnished the radiation data
which were derived from measurements made by NOAA polar
orbiting satellites. 87
REFERENCES
Agee, E. M. and R. P. Howley, 1977: Latent and sensible heat flux calculations at the air-sea interface during AMTEX 74. J. Appl. Meteor., 16, 443-447.
Alpert, L., 1945: The intertropical convergence zone of the eastern Pacific region (1). Bull. Amer. Meteor. Soc., 26, 426-432.
Bellamy, J. C., 1949: Objective calculations of divergence, vertical velocity and vorticity. Bull. Amer. Meteor. Soc., 30, 45-49.
Brook, R. R., 1970: Preliminary study of wind shear in the boundary layer at Melbourne Airport (Tullamarine). Meteor. Study No. 19, Bureau of Meteor., Australia, 57 pp.
Brunt, D., 1939: Physical and dynamical meteorology. Cambridge Univ. Press, London.
Bryson, R. A. and P. M. Kuhn, 1961: Stress-differential induced divergence with application to littoral precipitation. Erkunde, 15, 287-294.
Byers, H. R. and H. R. Rodebush, 1948: Causes of thunder storms of the Florida peninsula. J. Meteor., 5, 275-280.
Charney, J. G., 1968: The intertropical convergence zone and the Hadley circulation of the atmosphere. Proc. WMO-IUGG Symp. on numerical weather prediction, Tokyo, Japan Meteor. Agency, III 73-79.
Charney, J. G. and A. Eliassen, 1964: On the growth of the hurricane depression. J. Atmos. Sci., 21, 68-75.
Cheang, B. K., 1977: Synoptic features and structures of some equatorial vortices over the South China Sea in the Malaysian region during the Winter Monsoon December 1973. Pageoph, 115, 1303-1333.
Chin, P. C., 1969: Cold surges over South China. Technical Note No. 28, Royal Observatory, Hong Kong.
Chiu, W. C. and M. A. Estoque, 1964: Some selected studies of the atmospheric boundary layer. HIG-64-l9, Hawaii Inst. Geophys. 88 Chu, E. W. K., 1978: A method for forecasting the arrival of cold surges in Hong Kong. Technical Note No. 43, Royal Observatory, Hong Kong.
Clark, N. E., 1967: Report on an investigation of large-scale heat transfer processes and fluctuations of sea-surface temperature in the North Pacific Ocean. Solar Energy Research Fund, MIT, Cambridge, Massachusetts.
Clarkson, L. S., 1957: Note on the mean characteristic of the maritime equatorial troposphere over Singapore. Meteor. Mag., 86, 273-276.
Colon, J. A., 1964: On interactions between the southwest monsoon current and the sea surface over the Arabian Sea. Proc. Symp. Tropical Meteor., Rotorua, New Zealand Meteor. Serv., 216-229.
Dale, W. L., 1959: The rainfall of Malaya. Part 1, J. Trop. Geogr., 13, 23-37.
------::---,1963: Surface temperatures in Malaya. J. Trop. Geogr., 17,57-71.
Deacon, E. L. and E. K. Webb, 1962: Small-scale interactions. Vol. 1 of The Sea, Ed. M. N. Hill, Interscience Publishers, New York, 43-87.
Defant, F., 1951: Local winds in Compendium of meteorology. Amer. Meteor. Soc., Boston, 655-672.
Estoque, M. A., 1961: A theoretical investigation of the sea breeze. Quart. J. Roy. Meteor. Soc., 87, 136-146.
,1962: The sea breeze as a function of the ------:-:-- prevailing synoptic situation. J. Atmos. Sci., 19, 244-250.
Falls, R., 1970: On the formation, structure, and decay of an intense monsoon depression, North Australian region. WMO Proc. of the Regional Training Seminar, Singapore, 403-420.
Fisher, E. L., 1961: A theoretical study of the sea breeze. J. Meteor., 18, 216-233.
Fletcher, R. D., 1945: The general circulation of the tropical and equatorial atmosphere. J. Meteor., 2, 167-174. 89
Flohn, H., 1960: The structure of the intertropical conver gence zone. In "Tropical Meteorology in Africa", Ed. D. J. Bargman, Munitalp Foundation, Nairobi, 244-252.
------,1970: Climatic effects of local circulations in tropical and subtropical latitudes. In Proc. Symp. Tropical Meteor., Honolulu, Amer. Meteor. Soc., Boston, I-VI.
Forsythe, G. E., 1949: Exact particle trajectories for nonviscous flow in a plane with a constant Coriolis parameter. J. Meteor., 6, 337-346.
Fujita, T. T., K. Watanabe and T. Izawa, 1969: Formation and structure of equatorial anticyclones caused by large scale cross-equatorial flows determined by ATS-l photographs. J. Appl. Meteor., 8, 649-667.
Gan, T. L., (undated): A study of some heavy rain-spells on the east coast of Malaya during the northeast monsoon. Memoirs of the Malayan Meteor. Ser., No.6, Singapore.
Garstang, M., 1967: Sensible and latent heat exchange in low latitude synoptic scale systems. Tellus XIX, 492-508.
Godbole, R. V., 1973: On destabilization of clouds by radiative cooling. Mon. Wea. Rev., 101, 496-500.
Gordon, A. H. and R. C. Taylor, 1975: Computations of surface layer air parcel trajectories and weather in the ocean tropics. International Indian Ocean Expedition Meteor., Monographs No.7, Univ. Press of Hawaii, Honolulu.
Gray, W. M., 1968: Global view of the origin of tropical disturbances and storms. Mon. Wea. Rev., 96, 669-700.
_____~--~----~' 1972: A diagnostic study of the planetary boundary layer over the oceans. Dept. of Atmos. Sci., Colorado State Univ., Atmos. Sci. Paper No. 179.
Haurwitz, B., 1947: Comments on the sea-breeze circulation. J. Meteor., If, 1-8.
-----::---:-:-----' 1950: Particle dynamics and the sea breeze. J. Meteor. , 7, 164-165.
Hicks, B. B., R. L. Drinkrow and G. Grauze, 1974: Drag and bulk transfer coefficients associated with a shallow water surface. Boundary-Layer Meteor., 6, 287-297. 90 Holton, J. R., 1972: An introduction to dynamic meteorology. Academic Press, New York, 262-267.
Holton, J. R. and J. M. Wallace, 1971: On boundary layer dynamics and the ITCZ. J. Atmos. Sci., 28,275-280.
International MONEX Management Center for Winter MONEX, 1978: Report of pre-operational research. Fifth planning meeting for Winter MONEX, Manila.
Johnson, D. H., 1962: Rain in East Africa. Quart. J. Roy. Meteor. So c ; , 88, 1-19.
Johnson, D. H. and H. T. Marth, 1960: Forecasting research in East Africa. In "Tropical Meteorology in Africa". Ed. D. J. Bargman, Munitalp Foundation, Nairobi, 56-137.
Kung, E. C., 1963: Climatology of aerodynamic roughness parameter and energy dissipation in the planetary boundary layer over the Northern Hemisphere. Annual Rept. Contract DA-36-039-AMC-00878, Dept. of Meteor., Univ. of Wisconsin.
Lateef, M. A., 1967: Vertical motion, divergence and vorti city in the troposphere over the Carribbean, August 3-5, 1963. Mon Wea. Rev., 95, 778-790.
Lim, J. T., 1976: Rainfall minimum in Peninsular Malaysia during the Northeast Monsoon. Mon. Wea. Rev., 104, 96-99.
Malkus, J. S., 1962: Large scale interactions. Vol. 1 of The Sea, Ed. M. N. Hill, Interscience Publishers, New York, 88-294.
McPherson, R. D., 1970: A numerical study of the effect of a coastal irregularity on the sea breeze. J. Appl. Meteor., 9, 767-777.
Miller, B. 1., 1967: Characteristics of Hurricanes. Science, 157, 1389-1399.
Miller, F. R. and R. N. Keshavamurthy, 1968: Structure of an Arabian Sea summer monsoon system. Int. Indian Ocean Expedition, Meteorol. Monogr. 1.
Murakami, T., 1977: Regional energetics over the North Pacific, South China Sea and Indonesian seas during winter. Pageoph, 115, 1283-1301. 91
------,1978: Seminar, November 1978. Dept. of Meteor., Univ. of Hawaii.
Neumann, J., 1951: Land breezes and nocturnal thunderstorm. J. Meteor., 8, 60-67.
----.,.------....,.,1977: On the rotation rate of the direction of sea and land breezes. J. Atmos. Sci., 34, 1913-1917.
Neumann, J. and Y. Mahrer, 1971: A theoretical study of the land and sea breeze circulation. J. Atmos. Sci., 28, 532-542.
Nieuwo1t, S., 1968: Diurnal rainfall variation in Malaya. Annals, Assoc. of Amer. Geographers, 58, 313-326.
Nieuwo1t, S., Y. P. Leong, C. T. Sim (undated): Rainfall in East Malaysia and Brunei. Memoir No.8, Meteor. Sere Singapore.
Ninomiya, K., 1975: Large-scale aspects of air-mass trans formation over the East China Sea during AMTEX 74. J. Meteor. Soc. Japan, 53, 285-303.
O'Brien, J. J., 1970: Alternative solutions to the classical vertical velocity problem. J. App1. Meteor., 9, 197-203.
Palmer, C. E., 1951: Tropical meteorology. Compendium of Meteorology. Amer. Meteor. Soc., Boston, 859-880.
Pearce, R. P., 1955: The calculation of a sea-breeze circulation in terms of the differential heating across the coastline. Quart. J. Roy. Meteor. Soc., 81, 351-381.
Pierson, W. J., 1950: The effects of eddy viscosity, Corio1is deflection and temperature fluctuation on the sea-breeze as a function of time and height. Meteor. Papers, Vol. 1, No.2, College of Engineering, New York Univ., New York.
Plate, E. J., 1971: Aerodynamic characteristics of atmospheric bound~ry layers. U. S. Atomic Energy Commission.
Ramage, C. S., 1964: Diurnal variation of summer rainfall of Malaya. J. Trop. Geogr., 19, 62-68.
,1968: Role of a tropical "maritime conti ------::-:-...,---- nent" in the atmospheric circulation. Mon. Wea. Rev., 96, 365-370. 92
Ramage, C. S., 1970a: Variability of vertical monsoon circulations WMO Proc. of the Regional Training Seminar, Singapore, 254-269.
,1970b: Torrential rains. ------,,-:::- WMO Proc. of the Regional Training Seminar, Singapore, 294-307.
,1970c: Heat lows. In Proc. Symp. Trop. ----::-:------..,:::- Meteor., Honolulu, Amer. Meteor. Soc., Boston, I-VII.
, 1971: Monsoon meteorology. Academic Press, ----=----::"...---:--- New York.
,1974: Structure of an oceanic near-equatorial ------:----::---:- trough deduced from research aircraft traverses. Mon. Wea. Rev., 102, 754-759.
______, 1979: Personal communication.
Ramage, C. S., L. R. Brody, R. F. Adler and S. Brand, 1969: A diagnosis of the summer mo~soon of Southeast Asia. Navy Weather Research Facility; Norfolk.
Riehl, H., 1973: Surface interaction calculations over the Gulf of Tonkin. Tellus, XXV, 424-434.
Riehl, H., T. C. Yeh, J. S. Malkus and N. E. LaSeur, 1951: The northeast trade of the Pacific Ocean. Quart. J. Roy. Meteor. Soc., 77, 598-626.
Sadler, J. C., 1964: TIROS observations of the summer circu lation and weather patterns of the eastern North Pacific. Proc. Symp. Trop. Meteor., Rotorua, New Zealand Meteor. Ser., 553-571.
,1970: On the origin of the tropical vortices. ----:::-::-:,..,..-:---:::----- WMO Proc. of the Regional Training Seminar, Singapore, p • 65.
,1979: Synoptic scale quick-look for Winter ------MONEX--December 1978. Dept. of Meteor., Univ. of Hawaii.
Sadler, J. C. and B. E. Harris, 1970: The mean tropospheric circulation and cloudiness over Southeast Asia and neighboring areas. Hawaii Inst. Geophys. Scientific Re p t , No. 1.
Schmidt, F. H., 1947: An elementary theory of the land and sea-breeze circulation. J. Meteor., 4, 9-15. 93
Schroeder, T. A., B. J. Ki1onsky, B. N. Meisner, 1977: Diurnal variation in rainfall and cloudiness. Water Resources Research Center, Univ. of Hawaii, Honolulu.
Sheppard, P. A. and M. H. Omar, 1952: The wind stress over the ocean from observations in the Trades. Quart. J. Roy. Meteor. Soc., 78, 583-589.
Simpson, R. H., 1947: Synoptic aspects of the intertropical convergence near Central and South America. Bull. Amer. Meteor. Soc., 28, 335 -346.
Sutton, o. G., 1953: Micrometeoro1ogy. McGraw-Hill, New York.
Thompson, B. W., 1951: An essay on the general circulation of the atmosphere over Southeast Asia and the West Pacific. Quart. J. Roy. Meteor. Soc., 77, 569-597.
Watts, I. E. M., 1955: Equatorial weather. Oxford Univ. Press, London.
Winston, J. S. and A. F. Krueger, 1977: Diagnosis of the satellite-observed radiative heating in relation to the summer monsoon. Pageoph, 115, 1131-1144.
Worthley, L. E. and C. B. H. Lee, 1967: Weather phenomena in Hawaii. HlG-67-9, Hawaii lnst. Geophys. 94
TABLE 1. Computation of heat transfer at selected points over surface of the South China Sea
A. STRONG NONSOON PERIOD Re1 Air-Sea Temp SFC Wind Humidity Qe Qs I 2) Qs/Qe Diff (DC) Speed (m/s) (%) (Wm-2) I (Wm- A (23N 119E) 5.0 9 85 252 86 0.34
B (18N 117E) 2.0 12 86 291 52 0.18
C (15N 115j~) 1.0 11 86 228 23 0.10
D (12N 113E) 1.0 12 86 273 26 0.10
E (8N 110E) 1.5 10 91 178 30 0.17
F (6N lOSE) 1.0 9 94 100 17 0.17
G (8N 115E) 1.0 4 86 63 6 0.10
B. WEAK MONSOON PERIOD
A (23N 119E) 0.5 9 81 161 9 0.06
B (18N 117E) 2.0 11 91 217 46 0.21
C (15N 115E) 1.5 10 86 232 30 0.13
D (12N 113E) 2.0 9 86 228 34 0.15
E (8N 110E) 2.0 6 84 145 20 0.14
F (6N lOSE) O.S S 76 115 4 0.03
G (8N lISE) 0.5 3 81 49 2 0.04 95
TABLE 2. Temperature at standard levels over Kota Bharu during the strong and weak monsoon periods (OC)
Pressure Temperatures Mean Temperatures Level (mb) on Jan. 4, 1975 Strong Monsoon Weak Monsoon
1000 25.1 24.2 23.6
850 17 .3 16.9 17.3
700 9.4 9.1 9.0
600 1.2 2.7 2.8
500 - 4.8 - 5.5 - 5.4
400 -15.0 -15.2 -15.8
300 -29.6 -30.1 -31.0
200 -53.4 -53.7 -53.9
150 -67.1 -68.2 -68.5
100 -84.4 -84.8 -80.3 96
TABLE 3. Mean resultant surface winds of Malaysian coastal stations (deg/m sec-I)
Weak Monsoon Strong Monsoon
Peninsular Malaysia
Kota Bharu 063/0.6 093/1. 7 Kuala Trengganu 001/0.3 065/2.0 Kuantan 026/1.0 013/1.1 Mersing 343/1. 7 340/1.5 Bayan Lepas 328/0.4 129/0.1 Sitiawan 283/0.5 195/0.5 Malacca 013/0.7 350/0.6
East Malaysia
Sandakan 345/0.8 032/0.9 Kota Kinabalu 142/0.6 357/0.5 Miri 233/0.1 286/0.2 Bintu1u calm 259/0.1 Sibu 240/0.5 233/0.1 Kuching 300/0.6 calm 97
TABLE 4. Mean diurnal temperature range and mean temperature for Malaysian coastal stations
Mean diurnal temperature Mean temperature ( °C) range (OC) Weak Strong Weak Strong monsoon monsoon monsoon monsoon
Kota Bharu 6.2 2.8 26.5 25.3 Kuala Trengganu 5.5 2.0 25.7 24.6 Kuantan 6.5 4.1 25.1 24.1 Mersing 4.1 3.7 25.5 25.0 Bayan Lepas 6.6 7.1 26.5 26.5 Sitiawan 7.2 7.8 26.4 26.0 Malacca 6.3 7.1 26.1 26.0 Sandakan 6.5 4.1 26.6 26.3 Kota Kinabalu 5.1 5.3 26.3 25.9 Miri 5.8 5.9 25.8 26.5 Bintulu 5.6 6.1 25.6 26.1 Sibu 6.6 7.5 25.4 25.8 Kuching 6.4 7.0 25.5 25.7 TABLE 5. Computed constants for various values of the coefficient of surface resistance (k)
Time of k M1 N1 N1 M N2 maximum Xl X2 $1 $2 - 2 N2 - (x10-5sec-1) (deg) (deg) (deg) (deg) (em/sec) (em/sec) M1 (em/sec) (em/sec) M2 rotation rate (LT) 1300 LT 2.8 21 11 42 22 708 136 0.19 110 11 0.10 and 0132 LT 1116 LT 9.4 52 33 104 66 465 59 0.13 94 8 0.09 and 2243 LT 1032 LT 16 65 48 131 95 315 27 0.09 75 5 0.07 and 2129 LT
I.D 00 99
Figure 1. Map of Southeast Asia. 90E 100E 110E 130E I •• i ) '... '1"
BURMA '"l BAY OF BENGAL lqlfl:( A .~ ~~ 0 \) ~~PHILIPPINES 0 ~ (I tz?tJ~ 1°l A,.~ ~ ~MINDANAO 5ulll ~ O~ S.. Irlf
" Cole bes Sea ~
INDIAN OCI;AN
c» tP
I-' o o 101
Figure 2. Peninsular Malaysia, location of rainfall
stations and land above 500 feet. 102
,50 MILES I
3"
• Above 500 ft. 103
Figure 3. East Malaysia, location of rainfall stations
and land above 3000 feet. 0 0 0 8°~~~~-1090 111 113 115 1170 1190 I I I I I I »: ~~~
6 KINABAL~KOTA , t;,~~ ffI u~,\\~ O~ SABAH a ~~~ ~ t;,0 00
4°
CELEBES SEA SARAWAK eSlBu 20 l. o ~ KUC~NG.?\ )J"-...\~ .-.,-- 0 (-- ""...... _.r../"' -, ./ . ,....-._~,/ .....'-.:»'
LAND OVER 3000 ft...•••.•••• ~ 0'
I-' o .p. 105
Figure 4. Monsoon surge over the South China Sea and
rainfall over Malaysia during January 1-8,
1975. Top diagram: Surface wind speed (m/sec)
zonally averaged over the South China Sea.
Stippled area indicates speed greater than
10 m/sec. Bottom diagram: Average l2-hourly
rainfall over east coast stations of Peninsular
Malaysia and coastal stations of East Malaysia. 2uN ...... :~ /)7":" ""'5~ :·<~::o\~::.;;::;:,: 1(.' 18 N 00:0:;;0/ r;:o":O l,) 1~0:":.. ~ ~. 5 •• ' .: •••• 0° °0 .. '. ' .' .: : .: • 10 N ,: 00.°, .:.:. . -: :0:" .' ~ .... o o '. •••• 14 N w V 0 1·L;:o~o·i~::o:::O:;.;o7:0 ~ 12N j: .:: ..J 10N >-
10 ON ~:.~! ~ '-~'.::: ~,.O·:·:,·: 6N '- j' (5~ 4N
::e ::t 100 .... ~ PENINSULAR MALAYSIA ..J EAST MALAYSIA ...J <, ~ u. 50 Z ./ < a: ,.....,..-;-:::...~--..~.-? .. -' .--==-----,1- I ._;-"-' .- --, r------f I i 31 1 2 a .; 5 6 7 e 9 (-_DEC-7~ -.------JAN 1975 ~ 1&74
~ o 0\ 107
Figure S. Upper-level winds over Malaysia during
January 1-8, 1975. Top and middle diagram:
900-meter winds and the near-equatorial trough
over Peninsular and East Malaysia. Each long
barb in the wind plots represents 5 m/sec.
Bottom diagram: Time cross-section of upper-
level winds over Kota Bharu. Stippled area
denotes greater than 10 m/sec. 108
KOTA I »« f ~\ ~ KINABALU J /y "')..
BINTULU ~/y /' I v y >-* V ~ <{ ~ KUCHING yU-v j L'" 1...."<"-
31 1 2 3 4 5 II 7 a v 1974~tDEC .IAN 1175
150
-J:l !. 400 w a: :;:) en en ~ 500 Co 600
700
850 900
3\ 1 2 3 4 15 IS 7 e "'~Il~;-"41r------)t .IAN 'us ------~ 109
Figure 6. Monsoon surges of December 1973. Top diagram:
Daily pressure and dewpoint at Hong Kong.
Middle diagram: Meridional component of surface
winds (kts) over the South China Sea (Cheang,
1977). Bottom diagram: Average daily rainfall
over east coast stations of Peninsular Malaysia
and coastal stations of East Malaysia. 110
PRESSURE 1030 ...... 0 I .s 1020 ~ '" 1010 '"~ Po 1000
22 20 30 8 12 16 20 2. 28 ~ NOV 1973 -++-1------DEC 1973 ------4 20N
15N
10N
5N
150 EAST MALAYSIA ..... ! :j 100 \.,~ 0, ~ I. ...:z: .I ;a 1 . \ ! \ 50 \ / • ° \/
22 28 30 4 8 18 20 2. 28 ~ lIOV 1973_~~(,._------DEC 1973 ------_+_ 111
Figure 7. Time cross-section of upper-level winds over
Kota Bharu during December 1973 (Cheang,
1977). Each long barb in the wind plots
represents 5 m/sec. 112
2OOlIIl.
3OCMll.
IIOONII.
5OQlIl.
6ocNI.
70CN1l.
85ClCll.
30001'1'.
1 2 3 4 5 6 7 6 9 10 11 12 13 14 15 16 17 18
S'rIPL!D ARI'.AS .lRl!: II1'l'll IIUD 5P1:EI'S 20 lCl'J'l'S OR MORE 113
Figure Sa. Albedo (percent) over Southeast Asia during
the strong monsoon period. Hatching denotes
area with albedo greater than 50%. 114
w o. N,...
lU o o,...
lU \0 ~ \
z,...o 10 z "W 115
2) Figure 8b. Outgoing long-wave radiation (Wm- over
Southeast Asia during the strong monsoon
period. 95E 100E 105E 110E 115E 120E 125E I I , ~
~~ (~ ~ EQI2::? ts: ~Ii - 200, ---220 5S~ 240~
...... 0\ 117
2) Figure 8e. Net radiation (Wm- over Southeast Asia
during the strong monsoon period. 95E 100E 105E HOE 115E 120E 125E \. _120-
25NI-____-40) ~_120:J
20N
15N -, o 10N~
5N
EO I ,,>., '\ ;0' I ~ "I
~ 5S ---
I-' I-' 00 119
Figure 9. 850-mb streamline chart for 0000 GMT January 4,
1975. Each long barb in the wind plots
denotes 5 m/sec. Dashed line indicates near
equatorial trough. 120 121
Figure lOa. Albedo (percent) over Southeast Asia during
the weak monsoon period. Hatching denotes
area with albedo greater than 50%. 115E 120E
30 40_ / 25N 30
20N
15N'- 30 'VL:::3~~ 10N ~ 5N
~ ~ V :;£:)~= EQ 1------== " '" \C\ (L- \ ~,,_ 1 55~~ ~30~
_--40
I-' N N 123
2) Figure lOb. Outgoing long-wave radiation (Wm- over
Southeast Asia during the weak monsoon
period. 95E 100E 105E 110E 115E 1201! 125E ~240~ _ 260 25N
Eol ...... :7I >,. --.....:~~" 240\ '" CT/ / 7':J· -", .- 55~/ '"' ~ I
220
t-' N .p- 125
2) Figure lOco Net radiation (Wm- over Southeast Asia
during the weak monsoon period. 126 127
Figure 11. 850-mb streamline chart for 0000 GMT November
29, 1974. Each long barb in the wind plots
denotes 5 m/sec. Dashed line indicates near
equatorial trough. 128
w o Ol
o 129
Figure 12. Composite maps of surface wind speed (top)
and sea surface temperature (bottom) for the
strong monsoon period. In the top diagram,
vector mean wind direction and wind steadi
ness (within brackets) are plotted at
selected points. 130
100£ 1051 11DE 115£ 120E
SL~FACE WI~~ SPEED (m/sec)
20N
15N
10N
5N
SEA-SURFACE TEMPERATURE (OC)
20N
15 N
10N
5N 131
Figure 13. Composite maps of surface air temperature
(top) and dewpoint temperature (bottom) for
the strong monsoon period. 132
IDOl 1051 1101 1151 1201
AIR TEMPERATURE (OC) 133
Figure 14. As in Figure 12 except for the weak monsoon
period. 134
'OOE '05E "OE 'HiE 120E
SURFACE WIND SPEED (m/sec)
20N
15M
10N
5N
lOOE lOSE 135
Figure 15. As in Figure 13 except for the weak monsoon
period. 136
100E lOSE 110E 1'SE 120E
AIR TEMPERATURE (Oe)
20N
1SN
10N
SN
DEWPOINT TEMPERATURE
20N
1S N
10N
5N
100£ 105£ 120£ 137
Figure 16. Vertical cross-section of relative humidity,
potential temperature and winds during the
strong monsoon period. Dashed lines indicate
relative humidity in percent; thin continuous
lines, potential temperature in degrees
Kelvin and thick continuous lines are the
boundaries between the easterly and westerly
winds. PRESSURE (mb)
t-' o 00 0' .j:- N o o o o 0 o o o o 0 , --- TI I ------. I -- I ------. 10 III :e PIIYA LEBAR O'~~,~~ ~~~_t:: (1°20'N, 103°53'E) - 0\ d':':1}- v' ~ KOTA llHARU (6°10'N, 102°17'E) /j )! "\ ) ~0-j~'U ~t --J' J; '" I ;', - / " I I \ I I I \ \ I TANSON NHUT / ,/ J I (10049'N, 106°40'E) ~/ . .... -_.... /1 I ..:::\ 1m... I t\ I _----"" - ;/~-W J'T \ \ \ \ '. \\ / II HAIKOU If ~,' ~~ (20002'N, 110021'E) 'I/"A\~ I \ " \ \; I / _\ I I I ~ ...., \ KONG KONG /1//\.,\..... ,":---) /F f: (22°19'N, 114°10'E) I 1 "I ./ , -~. _..- ...... / " <,_-...., . /{' , =: ",.... ", "" -- -- " FUZHOU .... ", ~ " ~\\ i I -f, (26°05'N, 119°17E) " I -) '\ \ "'" ---_ ...' I \\ /' I NANJING ,\ \"~ \i~r.'\ " (32°QQ'N, 118°48'E) I ~ t-' N N I I l.J l.J l.J W w W l.J WI.) W III "-4 01"' III Cl III .. 0 .. N W .. III 01 ""Ill co 0000 0000 o o 0 o 00 00 139
Figure 17. Mean gradient-level circulation over
Southeast Asia for November (top) and
January (bottom). (After Sadler and
Harris, 1970.) 140
... ---_....--- i - .,;--;.-';'I ..---;;-_... :-... 141
Figure 18. Mean monthly rainfall of Peninsular
Malaysia for the winter monsoon (Dale,
1959). 142
NOVEMBER
IDO""
IOf 143
Figure 19. Mean monthly rainfall of East Malaysia for
the winter monsoon (Nieuwo1t et a1.,
undated). 144
~
OJ :! c: ... ~ ~ .. ",... 0 ,... 0 :E :E .. C '" E ~ 0 ~ 0 ~ a: E , > \ W a: E [!J e.. \\S:! < c: ~ ::E .. ::l ..'" w E co a: E co o a:l w '0 ~'{.' w '0 Q . .. /.; ... :: .. u. .c:.. .c: u ~ 0 .E 0 .~
:: c: UJ [!J ::E w o> z 145
Figure 20. Location of upper-air stations in Singapore,
South Thailand and Peninsular Malaysia.
The three triangles shown were used to
compute relative vorticity, divergence and
vertical velocity following Bellamy's
method. 146
7N
5N
3N
IN 147
Figure 21. Cross-section of relative humidity and
virtual equivalent potential temperature for
January 1, 1975. Relative humidity (%) is
shown in dashed line and virtual equivalent
potential temperature (OK) in continuous
line. The trough axis is denoted by thick
double lines. Each long barb in the wind
plots denotes 5 m/sec. 148
300 l I ,/ / / / / II --, / / / , / I 110 / 50 I I I/ " / I '\])" / I // ,/ / I / I I '" 50 .0 I ' ... _ 60 ...." I I I I , <, / \ \ 400 ~" , " ~ "."._-60..--." " , " /,,---, , 335 / / ...... '" -, --...... \. / / \ v"" I I ~ \ 500 I / ," 1"\ I " I 'go/ , / " ....3.0 ..a I .... E / 600 < / ~ »< LU a: I en:::I I en 335 LU a: / / a. 700 I /.-f\ \ I/f'" \ I/ \ 'I \ I I ..-'"'\ I I I I I 850 \ f .-""1\ <, I \,~11 I ,::~ ! I I \ 11-- .... t I \ 350 ~355 1000 9~\ \ 3.5 / 350 3.0 '" cr z ::;, oct a: 5 113 ~ oct :J: W Z :c ::,c .... ct a:I e :::I Z ~ :.:: 0 oct s en Q, :.::0 149
Figure 22. As in Figure 21 except for January 3, 1975. 150
~ '_ 345 345 300 ~ ~ ------.....~',1<, / I // <;-______, / / I I\ \.. 30_/",'" - <, -40- ( 40- ... ,,/ _.... ---- "/ '50 " ---~---- ...... <, '50---/ .""".' - -, /"" / 400 ".,--"...,--,\ \ \ -/;---,.,/y .r ->« "60-- 70/// A 110 340 / 500 I ...... l ..0 E - / w "" ...... I a: '<" 70 -- ) I :::3 600 'I." / } en ...... / I en w /1- / I a: I a. I IA 700 I I 340
850
1000
a: z ;:) -e c=: a: ..J lXl ~ ct ::l: W Z ::l: ~ ..J ct IX) C' ;:) Z ~ ~ 0 ~ 0 (J) c. ~ 151
Figure 23. As in Figure 21 except for January 4, 1975. 152
300 ," / 60 I I 50 I J I I 400 \ \ \ /> \ \ \ \ \ \. \ \
\\ , \ " I 500 I .Q I - I E J - \...... ,," w _ a: ::; 6"vv '" en (J) LLI a: a.
700
850
1000
<~ ....J: az o (J) 153
Figure 24. As in Figure 21 except for January 8, 1975. 154
340 300 I J , '" ,I I, I I -, I I I I / / I I I " ,/ I I I 340 40 " " / "... .> // , I I I " " " ,/ SO 50 40 I ,/'" ,/ ,"" I I I / \ , / / / '" " \ \ / ,/ \ / ,/ "" '<, \ 400 r / / \ / / \\ 'V\ / / , \ _..-40 I , " ,/ I ",,'" - " ,' I / '...... "' ... / I / ...... '_50 I / .... -....._- / .,.,.----- " ...... / / ------\, ,/ I I ..1...- sao I , »> ,.... I 80 ,Q \ I \. E ~ I - I U; , a: - \ :::l 600 \ 70 CJ) \., fn ... UJa: ...... - Do 700 V Y
335
850
~
1000 ~
a: z c( -e a::::l .... t: < ::r: =UJ z ::r: :ll:: .... < a1 CJ :) Z ~ :ll:: ~ 0 0 II) f :ll:: 155
Figure 25. Cross-section of surface pressure and
pressure height across the trough for
January 4 (top) and for January 1-8, 1975
(bot tom) . 156
16540 16480 --~-..-- 100 mb 16420 ------9680 6' 9650 ------300 mb c.. .!:! 9620 ------~--- l:: ~ 5831 !'J 5825 500 mb 1502 1490 1478 " 1501 1495 143. .0e Figure 26. Daily pressure at 0730 LT (OOOOGMT) along the east coast of Peninsular Malaysia during the strong monsoon period. 158 1012r------, KUANTAN \. 1010 ....... w a: 1008 ;:) C/) C/) w a: Co 1006 4 5 ~------JAN 1975 159 Figure 27. As in Figure 21 except for November 28, 1974. 160 300 ~ , .,.---..... , ", "." ..... , ...... """----- _.--""- .,--...... , " 340 '" , ,,/ \ 400 / "" _------50-___ -" " ------,- .... - ... _____110____ ,'"",/ <, ---- -...... ' \ I " / /-.... , \ ' ...... _,,'" / A\ \ 500 ,,/" ..r '\, ....--- \\\ \ '7 / \ \ -.t2 / E I ... \ ' I '"-- .... \, ' - I " , LLI a: I / " \ ' ..... ::l / , II) 600 / I II) / I \ -s-: LLI , I \ a: " I "', Do " \ '" I \ , <, 1"'/'" I 700 - LL...... __ I ...... ---70--' I " I 340 850 ---<; "gO, 1000 ' ... 350 a: Z ::l ~ ~ a: ~ llJ ;: ~ :J: W Z :J: ~ ~ ~ en e oct ::l Z ~ ::.c: ;3 0 a.; 0 II) ~ 161 Figure 28. As in Figure 21 except for December 4, 1974. 162 340 I 300 I '7 II"'" I 50 I I 20 340 I \ \ 335 I \ I \ I I 400 \ ~ \ \ , .... _---- "...... _--- 330 '- ...... , -- 60 '\ 500 40------50--_",1 ...... 335 .D ~25 ...... E /.,..._ -- --__2-0--'-= (/ w a:: 600 ' ...... ---320 ::J ...... CJ) CJ) " " ", .... ---...... ------w " '_..."" .",. .... --....., ...... a:: ...... "...... -- a. I <, ...... ,/1 ,(-...., '...... ----- ...... __ .... '\." .... 700 330'" I" ...... -, , I , " ...., I \ 40 ',I 50 ... _/ \ \ 850 .r 1000 a:: z ::J ct -e a:: ..J m ~ Figure 29. Rainfall distribution along the coastal strip of East Malaysia which borders the South China Sea during the strong monsoon period. The isohyets (thin continuous line) is in mm/day. The thick continuous line indicates the location of the gradient-level trough. KOTA KINABALU (SnS7'N ,116°3'E) I '\ , I '\ ,,10 , \ / , '\ \ I '\ , \ \ I \ , \ \ / \ 10 20 \ MIRI \ I \ I \ , (4°21'N .113°S9'E) I \ I \ I , I I »>:.... J , , I I I I --- BINTULU , \ I , , - (3"12'N,113°02'E) , \ I , \ , I I , \ \ I / .(c.... (0 10 , / I I / "'" .... ,,- SIBU \ ' ..... _-- "" / 10 I \ I '10 (2°20'N ,111·S0·E) \ '\, \ '-10 ...... \ '\ :ft:~~ \ '\ \ KUCHING \ ({60~ I (1·29'N ,110°2 O'E) \ " , 31 I 2 3 4 6 7 8 9 ~ DEC 1974 ~~(------JAN 1975 ) I-' 0'1 ~ 165 -5 -1 Figure 30. Relative vorticity in units of 10 sec over the triangle Songkhla, Kota Bharu and Penang (top); the triangle Kota Bharu, Penang and Kuala Lumpur (middle); and the triangle Kota Bharu, Kuala Lumpur and Paya Lebar (bottom). Stippled areas denote negative vorticity. 166 700 1000 31 1 2 3 • 5 II 7 a " +-DEC-" ~ JAN 1975 ) '87. 167 Figure 31. As in Figure 30 except for divergence. Stippled areas now denote divergence. 168 100 ~.~/: ~~ ~'- \0·.·. : ·2~·"".:.: ::: :: /::::::-2 ------2 '" .•. ' '.' '.' .. ~ @ ...... : ..... 200 ~ ...... n::-: :SI ,:'> . .: G .0 /' e --.....: .•..• ' .' .. 0-:-: '-' -- , :..y ' .. / ~2 o : . '0' 0 t>:l 300 ~ .. !5 en .. : :::: .. n 0 017 .tr.l t>:l IX: 500 \;.~iI ~ l:l- , ~.,,~ 700 -2~ ~20~: 1000 .•.•..••• _,.:- 100 " , 'x ". "j' \))))U \:O~". ::: -2 -4 _& -2 lE :<:"> n··.···. 200 '.~ ('~) " . " ...: ...... 0 :~c-'2: : : : ::. ':0:' .5 " '" ='" "...... : 300 .~' w ::. -J :.' !5 :: :': :. •:: ..•: : '0' tr.l en w 500 ::; ':0 .::,: .... c:: : : .: 0 -2 .:: : : .. l:l- ./ .... 700 :~32 tf:O~::: ~6>~ o. . . -2 .....•.: o : C- 0 '. 1000 /"": ~ : . '.' . . . -... 77~~ 100 \~/f(·:·::·~'i/>' :'\' .:':'(' ~':'; Y:': :.\<\~ >,~/(j,( L\,0 :,i .~...... /i*: 200 .;.... . i2U:"',...... : : . . . . - --/...... 0 . ' ! w IX: 300 ::> en en ~ l:l- 500 ~ ~ ~ : : ; ~ ~ =. ~ : ; ~: ',,;: . "0 .. : .:.'.'.:/ 700 - . 0 -2 :::: .'. /0 / <) 2 .~ .. OJ·:·:···········.. ~:~.~ .-:;~J ~'.'.':: :.~~~v: 1000 _-2, j r<; 31 2 3 4 5 6 7 II 9 <-DEC~f-(------JAN11175 -----~) 1974 169 Figure 32. As in Figure 30 except for vertical velocity in units of 10-4 mb/sec. Stippled areas now denote descending motion. 170 100' :':~ .' . (" . ~ ~ 200 \ \ 700 ~._------~C!~ / ..... 1000 l------_ 100 200 """..0 S '-' r.J 300 en!5 en ~ 500 j~' 0.. 0\:...... : .::')0 ...... ·:· 700 ...... -40 '0 /7":0 ...... 0 f: ::.j ...... 1000 100 ...... ' ./ ...... ( .) '; ...... '.'( •• ' '40' •• 0 .. " .. . 40 .\" ...... :: : : : : : . ( :: . ~ .... ' ...... o : ... :: •. :: ...... 0\:.:: :..·.0 ...... / .... J I « 2 3 4 5 6 7 8 9 ------JAN1975------)0) 171 Figure 33. Schematic model of a near-equatorial trough over Southeast Asia. Trough axis is denoted here by double lines. ¢ NORTH -. .- ~ 16 J- .....- t p 14 ,.. --+ 4- \ / ,..,...... ~ i i I I r 12 '" ~ I ~ :: ~\ v ,. ~ ~ J E 10 ~ \ { { t J ( 1 ( \ I J \ -~ ) \ I 300 mb ...... oX l- J: -t I {} ~ ~ Q a I I \ < UJ J: I t 't , f - ,. I IT" ~ ! - 6 . .... l •I ( I I I 500 mb 4 mb 2 850 t \ 1 -I '=-E:=::..I =t:=- 4-- 1N 2N 3N 4N 5N 6N 7N aN I-' -...J N 173 Figure 34. Mean 200-mb circulation over Southeast Asia for November (top) and December (bottom). (After Sadler and Harris, 1970.) 174 175 Figure 35. Hourly surface pressures along the east coast of Peninsular Malaysia on January 4 and 5, 1975. Numbers shown are actual pressures subtracted from 1000 mb. Circled numbers are the lowest pressures recorded among the four east coast stations at a particular hour. KQTA ~ ~ 7.3 G.48 8.0 7.9 G.7 6.2 5.4 7.0 6.9 e 8.8 9.2 8.3 8.5 BHAnu KUALA 7.0 6.5 G.3 @ 7.G 7.1 G.2 5.5 G.8 7.4 7.9 7.5 6.5 6.5 7.3 00 8.4 6.6 6.4 7.6 8.6 9.0 8.7 TRENGGANU KUANTAN f· 6.5 6.2 7.0 8.1 7.988 5. 6 7.0 6.4 8.4 7.9 e 6.2 7.6 9.0 9.4 e 6.1 5.7 7.1 8.1 8.& e ~ e~ MERSING 8.0 G 6.9 5.5 e 6.7 7.3 8.0 7.6 00 7.2 9.0 9.3 8.4 7.3 6.4 7.5 8.5 8.8 8.6 J I II111I1II11II I 1_ 1 _ 1 I __I 1_ _ 1 02 06 16 22 02 06 10 14 16 22 LT .r ------~) +(------5 Jan )' t-' ~ c- 177 Figure 36. Hourly weather observations at stations along the east coast of Peninsular Malaysia on January 4 and 5, 1975. Note that the weather observations at Mersing were carried out at certain hours of the day only. KOTA •• ~"':l •• .. •• •• •• BHARU -.- .. ... •• •• •• •• • •• •• •• •• • • •• •• 9 9 6 KUALA • •• ••• • • • • ~ TRENGGANU •• •• •• •• •• •• •.. •• )·e 6 Q • • •• ••• •••••• ~ c;> 6 9 .:. . . KUANTAN 6-0-9-0- • 6 ••• ~ R] ••• •• •• •• ~ 6 o • • 3 9 • 9 6 9 c I K MERSING 00 ClO H 00 (R) 02 06 14 18 22 02 06 10 14 18 22 LT t ~ ~ 5 Jan I-' '! co 179 Figure 37. Mean hourly surface resultant winds (at even hours) along the coasts of Peninsular Malaysia during the weak monsoon period. Numbers near the wind vectors indicate speed in m/sec (to nearest whole number). 180 KOTA @ @ BEARU 1 KUALA ~ .",." ... TRENGGANU 7/77 , 7 en \ // I I et 0 to) 1 1 0 2 2 2 2 1 1 1 1 ...en KUANrAN /-- et ; J \ t I I ; I J 1&1 r 2 2 2 2 2 3 4 3 2 1 2 2 MERSING --.". ~ , ; \ \ -- \ I I ; • \ 1 l' 1 2 2· 1 2 1 2 0 0 1 BAYAN \ ~ LEPAS , , //, / ---~ \ ~ rn -: " 0 1 1 3 1 0 "0 0 0 0 2 2 0 CJ ... SITIAWAN »<>: --..~~ -, ...... -- U) / I \ \ L:Is= 1 1 1 2 2, 1 t 0 t 1 -..,..., -'" ~ (§) MALACeA [ I I I I { I , / I I ! ! 1 04 08 12 '6 20 24LT 181 Figure 38. Mean hourly surface resultant winds (at even hours) along the coasts of East Malaysia during the weak monsoon period. Numbers near the wind vectors indicate speed in misec (to nearest whole number). 182 SANDAKAN KOTA· KINABALU o BINTULU "'- 2 2 1 0 to SIBU I ~------T ~ o KUCHING /~,--0 0 0 I ,.. -- @) 04 08 12 16 20 24LT 183 Figure 39. As in Figure 37 except for the strong monsoon period. 184 ,'222223222 KOTA BHARU ~-----~-----.---~-~ 0 1 2 3 4 3 2 3 2 KUALA ~ ~ /./~ ~ t; TRENGGANU / // ...- // ~ 0e , I- , ,, 2 2 2 2 2 1 1 , en KUANl'AN l I /' w< ; i ; t I I I I \ \ \ , 1 2 , 2 2 2 2 , 2 , MERSING \ " "'''''' J / 1 ; \ \ \ , 1 1 , 2 2 4 2 0 , 1 1 BAYAN LEPAS ; I I I \ / / \ I I I I- -- en < 0 0 0 0 1 2 4 3 0 0 0 CJ ~ ~, I- SITIAWAN /~/ ~ ...... -.- U) 't // UJ ~ 1 , 2 2 0 2 1 ,, , 2 0 , MALACCA /-",/, J r I I I I I I , 04 08 12 16 20 24LT 185 Figure 40. As in Figure 38 except for the strong monsoon period. 186 , 3 SANDAKAN I I - / / I i I iii 2 , , , " 0' KOTA / KINABALU ,'-----," " t ------ 0" 2 2 0 0 0 MIRI ',,,-- '" "<, ,..., -...... " '" " -- "'- , ,, , , 1 :2 2 , 0 0 BINTULU t \" \ /'\ \ \ \ \ ,0 0 0 ,, 0 0 0 0 SIBU @ /' I @ "- I ~" 1/ I 0 , "0 0 0 1 1 0 KUCHING @ " ""- "- \ I \ ~ "- \ -. 04 03 12 16 20 24LT 187 Figure 41. Schematic diagram showing the observed and rotated wind directions for east and west coast stations. The observed wind direction is rotated through an angle such that the coast at which the station is J.ocated is now aligned along the y-axis. 188 LAND (a) EAST COAST LAND (b) WEST COAST 189 Figure 42. Mean hourly surface resultant winds (at even hours) along the coasts of Peninsular Malaysia during the weak monsoon period. The coast lines are realigned as shown in Figure 41 so that the observed winds can be compared with the theoretical winds in Figure 46. 190 EAST COAST 0 0 2 2 2 1 0 0 0 KOTA @ @ @) BHARU ",,-'-.."'", --- 2 2 2 2 1 3 4 3 2 0 2 KUALA ~~--"'--- ~~---/ TRENGGANU I --' --~ , 1 0 2 2 2 2 2 1 1 1 1 1I.lJANrAN \ \ \ \ \ I ,/ ~ t \ \ \ 2 2 2 2 2 3 4 3 2 2 2 MERSING J "'''- t I / // I / I t WEST COAST 1 1 1 2 2 1 2 1 2 0 0 1 BAYAN ~ 1 I '- t l ~ I LEPAS f / -, f ~ .. I ; -----" " 0 0 0 1 2 3 2 1 0 0 0 SITIAWAN /~- / j -...... -- "'- \ I <, "<, 1 2 2 1 1 0 1 MA!JIC~\ ,..,. @ ------~ \ \ ----- 04 08 12 16 20 24LT 191 Figure 43. As in Figure 42 except for East Malaysia. 192 EAST COAST 1 1233220 SANDAKAN ''-...---'/~--~-----I- WEST COAST 2 1 1 0 2 KOTA ~-..- <, ) KINABALU --- I --- t ----...... -. -- 0 0 1 2 2 0 0 ~!IRI ,.,.., ~ - -- -- I ----'--~ -, o 1 2 2 o "o o BINTULU ----_e._----- I I / 2 1 0 0 SIBU @ ///'-.. @ 04 08 12 16 20 24LT 193 Figure 44. Mean hourly surface resultant winds (at even hours) along the coasts of Peninsular Malaysia during the stY0ng monsoon period. The coastlines are realigned as shown in Figure 41 so that the observed winds can be compared with the theoretical winds in Figure 47. 194 EAST COAST 1 1 2 2 2 2 2 3 2 2 2 1 KOTA BHARU \\ \ \ \ \ \ \ \ \ "1 2 3 3 2 0 4 3 2 " Ktl.o\I.A -~ 'l'RENGGANU ----..- -...... ---...... - ~--- 1 1 1 2 2 2 2 2 1 1 1 KUANrAN L \ \ \ \ + I I / I \ \ 1 2 1 2 2 2 2 2 1 "1 MERSING J \ \ \ /'/ I I I I t " WEST COAST 2- 4 2 0 1 1 BAYAN LEPAS ~- -.-. -...... ~ ...... --- \ /"""/ 0 0 0 1 2 1 4 3 0 0 0 SITIAWAN "<, ------"'-. I -: ---- " "'- 1 2 2 2 0 2 1 0 1 }!A!.ACCA ....------""" I ------04 08 12 16 20 24LT 195 Figure 45. As in Figure 44 except for East Malaysia. 196 SAST COAST 1 2 3 2 2 SANDAKAN --- \ - --- wssr COAST 1 0 1 2 1 KOTA ~ .- ,,- KL~Bl,LU "" ...... ~~--- \ I /" --- 0 1 ..,..., 1 2 2 1 0 0 0 MIRI ------..- / ------/- 1 1 1 1 2 2 o o o BnrrULU --~ ...... - o o o 0 SIBU I \ '\-- 04 08 12 15 20 24LT 197 Figure 46. Theoretical surface winds for the weak monsoon period. Numbers near the wind vectors indicate speed in m/sec (to nearest whole number). 198 -5 -1 (a) u =0, v =0, £=1. 5x10 sec (east coast) s s 2 4 5 4 2 1 4 5 4 3 1 ~ ------"- - -5 -1 (b) u =0, v =0, £=1.5xlO sec (west coast) s s 2 4 5 4 2 1 4 5 4 3 1 1 ------"""'"- " - (c) u =0, v =-2 rn/sec, £=0 (east coa st) s s 4 5 5 3 5 6 5 ~ ...... , ; ,,-6' ,.,. .,..; " " .> t \ (d) u =0, v =2 m/sec, £=0 (west coast) s s 3 4 5 5 3 2 5 6 5 3 2 2 ,..,- ,..., ~ ,/ .....- / \ '" '" " \ I (e) u =0, v =-2 m/sec, f=O (west coast) s s 4 5 5 5 6 5 3 ~ "- ...... " /' ~ ",- / / 1 \ 04 08 12 16 20 24LT 199 Figure 47. Theoretical surface winds for the strong monsoon period. Numbers near the wind vectors indicate speed in m/sec (to nearest whole number). 200 -5 -1 (a) u =-3.5 m/sec, v =0, £=1. 5x10 sec (east coa st) s s 0 2 5 8 9 e 6 4 3 - , ~ - - --....-. """""- --0------" -5 -1 (b) "u =3.5 m/sec, v =0, £=1. 5x10 sec (west coast) s s 6 7 e 8 6 2 2 1 3 4 ------~ -s- " .»: - - -5 -1 (c) u =0, v =3.5 rn/sec, £=1. 5xlO sec (west coast) s s 4 5 6 5 4 3 5 7 6 5 4 4 ~ I ,./ / ./ ,1 \, "- , "- \ \ I 04 08 12 16 20 24LT 201 Figure 48. Rotation rate of wind direction due to the mesoscale pressure term. 202 o o o N V I t 203 Figure 49. Hourly surface winds along the east coast of Peninsular Malaysia on January 4 and 5, 1975. 204 0 r .s <, ~ C" 0 - 0 - C't ..A '"'- - - .,.,- co r .rl' / .... r- -'1'" r: / r .,.....-\ /\ t ...."ff' c: ~ r r ~ ~'" It) r f 0 <, 0,... \" « )., , CD r \ 0 \ \' \ \ ~ C't \" \ 0 ( t: \ --. { C't r ..J-. C't r \ \ ~ r I r r .J.-- -.4.,. ....co r \ ~ ""'--. <, r r i: ..."ff' ~ c: r ...... '\ -- ...0 \ ~ ! - \ CD r \ 0 0 ~ I r 0 I j- C't Z. o ;/ 0 ,:, :::J Z o ~cr: z ~ z ocr ~cr iii :.::J: ...J(.:J z cr: CD cr ::lzce" ::l w :'::w :.:: ~ .-a: 205 Figure 50. Theoretical winds along the southeast coast of Peninsular Malaysia on January 5, 1975. Numbers near the wind vectors indicate speed in m/sec (to nearest whole number). 206 C'I -/ C'I MI 0N ~ I -co ~ I -to ~ I -~ N/ -N N\ ....0 co ~ \ 0 ~ to \ 0 207 Figure 51. Diurnal rainfall variation of Peninsular Malaysia for the strong monsoon period. Value within brackets denotes average 24-hour rainfall in mm. -"-'-'-"': 208 Q '-""- ALOR STAR • \ (3.3) : ) .MJ . r' .;:''', { KOTA BHARU \ ~ "'/~ (36.6) [] '-' SAYAN LEPASQ • (11.1 ) . HIGHLANDS (20.3) SITIAWAN SUNGEI TEKAM (14.3) • (4.7) KUANTAN l61.S} ~ G:JMERSING (14.6 ) oKLUANG (2.7) • • STATION MODEL :E :E 230 ~ l- ...I 20 I ....I if 10 ~ -e I I a: 0 I, MtOO'I<'IltIQ) ... ~ oOO ...... -<'IN III I II I' o t") to <:Tl N U'l Q) .;.. oOOO ...... _N 209 Figure 52. Diurnal rainfall variation of Peninsular Malaysia for the weak monsoon period. Value within brackets denotes average 24-hour rainfall in mm. 210 e KOTA BHARU (1.9) GRIK (2.4) e I '-I+"++~PW IPOH (14.4) BAYAN LEPAS CAMERON (15.7) e HIGHLANDS (4.8 ) SUNGEl TEKAM (5.3) e • WJKLUANG (8.S) e WI] • MALACCA (2.1 ) STATION MODEL SENAI t1.7) > U 5 Z W 4 a=:I 3 w 2 a: 1 u. I I o,-+-+-;-t-t-t--t-+-'I, 211 Figure 53. Monthly rainfall distribution of East Malaysia (Nieuwolt ~ al., undated). 212 % Q. a:-< 0 >- ~ - "" :. :. 213 Figure 54. Diurnal variation of January rainfall of East Malaysia (Nieuwolt et al., undated). 214 z -I .-.,~. \ ....."") ; \ -,., i ~. ~. ,:".,\ j i ( .... " ;:l • ~ -; :>;. . ..:::~., . 0·· • ,...'. ~-. ,"*,,0 :: _• 215 Figure 55. Diurnal rainfall variation of East Malaysia for the strong monsoon period. Value within brackets denotes average 24-hour rainfall in mm. lOgO 8°1-~'111° -~I113° -~,115° ~I~~117° 11g 0 I i I I °D 6°,_ [J] KOTA KINABALU (8.6 ) o .,', \ ~,\ I ',~ 4° 'IT]'...... ' " BINTULU (25.1) , MIAI ."." (16.6) .' I KUCHING (21.7) .'" " .[] STATION MODEL ,...... ,.- I SIBU - »: ::E ::E -, !(15.8) ;,.. ~ ; z 6 ..... _.,..-.-._._.".1 -. _./-.". ..J ..J 4 if ~ 2) IIIIIIII..II I I I a: MIOOlNU'lUl ... 'It OOO"',..,.-NN IIIIIIII OMIOOlNU'lUl'" 00009"'" .... ,...N °l ! I I L i N I-' 0\ 217 Figure 56. Diurnal rainfall variation of East Malaysia for the weak monsoon period. Value within brackets denotes average 24-hour rainfall in mm. 0 0 109 111 0 0 0 eO 113 115 117 119 0 °D [jJ SANDAKANo eO KOlA KINABALU (&.7) (6.2) ,,0 UBINTULU , (18.3 J ;,. ..,' ,." j' 111111111'" lliJKUCHING MIRI ; (10.8 , (21.0 J , • 2° • , STATION MODEL , -~_. ," :E== ruSIBU." ._• ..,,_ J' ,!(13.1) ." _J z ,,' ' .... , ...... -. ::l 3 it z 5 < ~t ~ ~ D o o "., I I I N I-' OJ