Summer Monsoon Experiment—A Review

T. N. Krishnamurti Department of Meteorology, Florida Stage University, Tallahassee, FL 32306

Search for other papers by T. N. Krishnamurti in
Current site
Google Scholar
PubMed

Print Publication:: 01 Sep 1985

Collections:: Review Articles in Monthly Weather Review

DOI:: https://doi.org/10.1175/1520-0493(1985)113<1590:SMER>2.0.CO;2

Page(s):: 1590–1626

Received:: 09 Aug 1984

Accepted:: 15 Apr 1985

Published Online:: Sep 1985

Displayed acceptance dates for articles published prior to 2023 are approximate to within a week. If needed, exact acceptance dates can be obtained by emailing amsjol@ametsoc.org.

Download PDF

Full access

Abstract

This paper presents a short summary of the Summer Monsoon Experiment (MONEX). The review is largely based on those papers that have made use of the summer MONEX observations during 1979. 0bservational aspects of this study emphasize the annual march of the monsoon rainfall belt from Indonesia to the foothills of the Himalayas, from the northern winter to the northern summer season and a reverse motion thereafter. The excellent FGGE/MONEX data sets have provided a detailed definition of the divergent wind; these are summarized with reference to the Hadley and the Walker circulations.

The manner in which monsoonal circulations respond to the evolving differential heating fields are presented via the mutual interactions among the rotational and divergent wind components. Specific examples of heat sources from the studies of Luo and Yanai highlight their contrast over different regions of the monsoon including the Tibetan Plateau. A problem of considerable interest in this context is the cooling of the Arabian Sea. A summary of results pertaining to this problem—especially the distribution of the wind stress curl—is highlighted.

The planetary boundary layer is another area of investigation which has drawn much interest, especially over the western Arabian Sea where the Somali jet exhibits interesting properties during summer monsoon. These studies cover modeling, theoretical and observational areas.

The onset and active monsoons were monitored by a large array of ship and research aircraft during MONEX. Studies in this area place an emphasis on observational, theoretical stability analysis and numerical weather prediction. The major results with respect to medium range prediction of the onset of monsoon and the formation and motion of a monsoon depression are summarized in the review.

A component of the MONEX observational program that is examined is the structure and maintenance of desert heat lows. A summary of these results includes the structure of the mixed layer, the day-night differences in the vertical motion profiles and the thermodynamic heat budget.

The final section of this review includes studies on low frequency modes—especially on the time scale of 30 to 50 days. It is becoming apparent that modulations of active and inactive spells of the monsoon are related to wave motions on this time scale. These MONEX data sets provide a strong signal for monitoring these waves. These wave motions on the planetary scale move eastward; on a more regional scale they move northward over the monsoon region. Their behavior is illustrated with respect to the onset, active and break monsoons.

Abstract

1590 MONTHLY WEATHER REVIEW VOLUME 113REVIEWSummer Monsoon Experiment--A Review T. N. KRISHNAMURTIDepartment of Meteorology, Florida State University, Tallahassee, FL 32306(Manuscript received 9 August 1984, in final form 15 April 1985)ABSTRACT This paper presents a short summary of the Summer Monsoon Experiment (MONEX), The review is largelybased on those papers that have made use of the summer MONEX observations during 1979. Observalionalaspects of this study emphasize the annual march of the monsoon rainfall belt from Indonesia to the foothillsof the Himalayas, from the northern winter to the northern summer season and a reverse motion thereafter.The excellent FGOE/MONEX data sets have provided a detailed definition of the divergent wind; these aresummarized with reference m the Hadley and the Walker circulations. The manner in which monsoonal circulations respond to the evolving differential heating fields are presentedvia the mutual interactions among the rotational and divergent wind components. Specific examples of heatsources from the studies of Luo and Yanai highlight their contrast over different regions of the monsoonincluding the Tibetan Plateau. A problem of considerable interest in this context is the cooling of the ArabianSea.. A summary of results pertaining to this problem---especially the distribution of the wind stress curl--ishighlighted. The planetary boundary layer is another area of investigation which has drawn much interest, especially overthe western Arabian Sea where the Somali jet exhibits interesting properties during summer monsoon. Thesestudies cover modeling, theoretical and observational areas. The onset and active monsoons were monitored by a large array of ship and research aircraR during MONEX.Studies in this area place an empfiasis on observational, theoretical stability analysis and numerical weatherprediction. The major results with respect to medium range prediction of the onset of monsoon and the formationand motion era monsoon depression are summarized in the review. A component of the MONEX observational program that is examined is the structure and maintenance ofdesert heat lows. A summary of these results includes the structure of the mixed layer, the day-night differencesin the vertical motion profiles and the thermodynamic heat budget. The final section of this review includes studies on low frequency modes--especially on the time scale of 30to 50 days. It is becoming apparent that modulations of active and inactive spells of the monsoon are relatedto wave motions on this time scale. These MONEX data sets provide a strong signal for monitoring these waves.These wave motions on the planetary scale move eastward; on a more regional scale they move northward overthe monsoon region. Their behavior is illustrated with respect to the onset, active and break monsoons.1. Introductio~ The monsoon experiment MONEXt was conductedwithin the FGGE year. The northern summer component of this experiment was conducted during themonths of May, June and July 1979. The purpose,motivation and goals of the Summer Monsoon Experiment have already been described by Fein andKuettner (1980), and hence they will not be repeatedhere. Studies on the Summer Monsoon Experiment haveranged from the low-frequency planetary scale motionsto those on local diurnal changes. The present reviewcannot be. ex. haustive within the limited pages and thussome om~sswns are unavoidable. Some of the salientlist of acronyms appears in Table I.findings from the FGGE/MONEX observations arereviewed here. The research on the various MONEXproposals is by no means complete at this stage; theprospects for further research in most of the followingareas appear quite promising.2. Planetary scale monsoon As a global problem, the monsoons have receivedconsiderable attention in recent years. The time averaged motion field around the global tropics, in thelower and the upper troposphere, describes the prominent place of the monsoon in the global circulations.Figures I and 2 show these climatological flow patternsat the 850 and 200 mb levels. In these illustrations wehave emphasized the lower-tropospheric confluentasymptote as well as the upper-level diffiuence patternsin the monsoon region. The lower-tropospheric conc 1985 American Meteorological SocietySEPTEMBER 1985 T.N. KRISHNAMURTI 1591TABLE I. List of useful acronyms.GARPFGGEMONEXTiros NNOAA-AGOESECMWFFGGE IIIbFSUJMARPNLMDNMCUKMOGlobal Atmospheric Research ProgramFirst GARP Global ExperimentMonsoon ExperimentU.S. Polar Orbiting SatelliteU.S. Polar Orbiting SatelliteU.S. Geostationary SatelliteEuropean Centre for Medium Range Weather ForecastsFinal gridded data produced by ECMWFFlorida State UniversityJapan Meteorological AgencyCanadian Numerical Weather Prediction GroupLaboratoire de Meteorologie Dynamique French Numerical Weather Prediction GroupNational Meteorological CenterMeteorological Office of the United Kingdomfluent asymptotes are well marked between Januaryand April over the eastern Indian Ocean. (Divergenceor convergence is not necessarily implied by diffiuenceor confluence.) This asymptote also marks climatological regions of monsoon rain and associated cloudcover. The principal belt of the monsoon rainfall performs an annual migration from Indonesia to thefoothills of the Himalayas, from the (northern) winterto the (northern) summer season. The reverse motionensues in the following months. Along this axis, regionsof maximum rainfall totals of the order of 300-500mm/month can be followed as they migrate polewardand equatorward seasonally. Although the interannualvariability along this axis is quite large, this can beidentified as a principal axis of the monsoon. The largestheat source of the monsoon is associated with condensation heating and it follows the course of this maximum rainfall belt. The behavior of the monsoon overIndia is largely dictated by the differential heating between the principal region of rainfall and a net coolingin the surrounding regions. The location of the axis ofdifferential heating can usually be approximated fromthe regions of divergent inflows to the divergent outflows at the 200 mb surface. The mode of migrationof the region of heating from northern Burma in themonth of May to the foothills of the Himalayas (Eastern Tibet) in the month of June is a critical factor indetermining the onset of the Indian Monsoon. TheIndian Monsoon can largely be viewed as a responseof the circulations to the differential heating betweenthe regions of cooling and heating. The onset of summer monsoon over eastern Asia (Burma, southern Indochina) occurs in May with a conspicuous change,from April to May, of the geometry of low-level flows.This feature was not apparent in the meridional motionof a clockwise gyre emphasized in the earlier analysisof Findlater (1969). This became evident from theMONEX data sets. In the upper troposphere (see Fig. 2), during thenorthern winter months, a pair of upper anticyclonesare located on the poleward sides of the near equatorial(lower tropospheric) convergence zone. Upper easterlies are found over the convergence zone and similaranomaly features have been noted in reference to theEl Nifio warming phenomenon over the central PacificOcean (Wallace and Gutzler, 1981). The climatologicalupper anticyclone makes a northward motion fromnorthern Malaysia in May towards the Tibetan Plateauby July. The anticyclone has its largest amplitude inthe month of August. Diffiuent upper level flows are acharacteristic feature above the regions of monsoonrainfall. As the withdrawal of the summer monsooncommences in September, the northern anticyclonemoves equatorward. There are, in fact, two upper anticyclones nearly all year round in each hemisphere.The annual cycles of the planetary scale monsoonshown in Figs. 1 and 2 are based on ten years of data.Within this period, large interannual variation in thecirculations were noted (Krishnamurti et al., 1984a). An important aspect of the planetary scale circulation is the divergent circulations (Krishnamurti, 1971).Usually the velocity potential at 200 mb is used todisplay this field. The depictions for the months ofJanuary and July are shown in Fig. 3. The arrow indicates wind direction. These are based on the FGGEdata sets. They illustrate a prominent role of the monsoon in these global circulations. In January the Hadleytype overturnings are more prominent, while in Julythe strong gradient of the velocity potential over theIndian Ocean is associated with the active monsoonwhich has more of an east-west orientation. It is ofinterest to note that a meridional belt of east-west circulations from roughly 20-S to 30-N is found over thecentral Pacific Ocean. It should be noted that there arehundreds of daily observations of cloud winds andcommercial aircraft wind reports that provide a basisfor this information over the Pacific Ocean. Thus, thenature of east-west circulations is far broader than theclassical picture of a Walker circulation along theequator. It is apparent that the monsoonal east-westoverturnings are a prominent part of the global divergent circulations. Zonal harmonic analysis of the velocity potential inthe tropical latitudes shows that a significant proportion(>60%) of the total variance is on the planetary scales(Krishnamurti, 1978). The covariance of the verticalvelocity and temperature (i.e., the T) is also distributedlargely on the planetary scale (Krishnamurti et al.,1984b). Thus, the east-west circulations have a majorrole in driving the planetary-scale tropical motions. Anumber of tropical depressions, storms and waves, withhorizontal scales are of the order of a few thousandkm, are usually present in the tropics. It is, however,very significant that these disturbances are asymmetrically placed in the tropics along planetary waves (i.e.,the zonally asymmetric ITCZ). This placing results ina substantial release of eddy available potential energyon the planetary scale, and only a small fraction of theenergy released drives the storm scales. 850MB TEN-YEAR MEaN STREAMLINES FOR~5 N~ - '_ _ ~-"CF',?;~~. ~ ~~ .~~~-~ 850MB TEN-YEAR MEAN STREAMLINES FOR FEBRUARY4 ~ N ~~-_2 2.~. -~-~'"~- ~,o, --~__~~.~. -~'-~,o~ . ~~~~C~~25S 850MB TEN -YEAR MLAN STREAMLINES FOR MARCHANUARY~ ' '~~/,.A.~45N30N _-------IONlOS25S 90W 60W 40W 20W OE 20E 40E 60E 80E IOOE 120E 140E 160E 180E 150W 140W 120W I00~'850MB TEN-YEAR MEAN STREAMLINES FOR APRIL30NION 850MB TEN-YEAR MEAN STREAMLINES FOR MAY5ONION25S ' 850MB TEN-YEAR MEAN STREAMLINES FOR JUNE 90W 60W 40W 20W OE 20E 40E 60E ~OE I~E 120E 140E I60E 180E 150W 140W 120W IOOWFiG. 1. Monthly mean streamlines and isotachs (m s-l) at 850 mb.45N50N 850MB TEN-YEAR MEAN STREAMLINES FOR JULY .~:~.~:~ ,, ~ .~..~_;.~ 5 2.5 ~ ~ __ ~ ~ -~IONlOS25 S 850MB TEN-YEAR MEAN STREAMLINES FOR AUI~UST45 N ~ 5 ~'""~'""~.'-'~ "' ~ 2'5"'~ IONlOS25S "~ -~'-~ -~~~~850MB TEN-YEAR MEAN STREAMLINES FOR SEPTEMBER45N30 NIONlOS25S 90w 60W 40W 20W OE 20E 40E 60E 80E IOOE 120E 140E 160E 180E 150W 140W 120W IOOW 850MB TEN-YEAR MEAN STREAMLINES FOR OCTOBER45 N F.5 .~_ - 5 ~ iT' J~'~' ION lOS25 S 850MB TEN-YEAR MEAN STREAMLINES FOR NOVEMBER45N30N ION 850MB TEN-YEAR MEAN STREAMLINES FOR DECEMBER45N30 N lOS 25S 90W 60W 40W 20W OE 20E 40E 60E 80E IOOE 120E 140E 160E 180E 150W 140W 120W IOOWFIG. 1. (Continued) 200MB TEN-YEAR MEAN STREAMLINES FOR JANUARY45N ION lOS25 S 2.00MB TEN-YEAR MEAN STREAMLINES FOR FEBRUARY45NSONiONlOS25S 200MB TEN-YEAR MEAN STREAMLINES FOR MARCH45N30 NIONlOS25S 90W 60W 40W ZOW OE 20E 40E 60E 80E IOOE 120E 'I40E 160E 180E 150W 140W I20W IOOW 2_OOMB TEN-YEAR MEAN STREAMLINES FOR APRIL45N30N ION 200MB TEN-YEAR MEAN STREAMLINES FOR MAY45N30N ION lOS25S 200.MB TEN-YEAR MEAN STREAMLINES FOR dUNE45N30 NIONlOS25S 90W 60W 40W 20W OE 20E 40E 60E $OE IOOE 120E 140E 160E 180E 150W 140W 120W IOOWFIG. 2. Monthly mean streamlines and isotachs (m s-l) at 200 mb. 200MB TEN-YEAR MEAN STREAMLINES FOR JULY45N I .... ~ I ~ ~'1'30N ION lOS25S '?.OOMB TEN-YEAR MEAN STREAMLINES FOR AUGUST45N 30N ION lOS 25S 200MB TEN-YEAR MEAN STREAMLINES FOR SEPTEMBER 30 N ION lOS 25S90w 60W 40W 20W OE 20E 40E 60E 80E IOOE 120E 140E 160E 180E 150W 140W 120W IOOW200MB TEN-YEAR MEAN STREAMLINES FOR OCTOBER45N30N ION25 S .... 200MB TEN-YEAR MEAN STREAMLINES FOR NOVEMBER45N30N ION lOS 25S 200MB TEN-YEAR MEAN STREAMLINES FOR DECEMBER45N30 NIONlOS25S 90W 60W 40W 20W OE 20E 40E 60E 80E IOOE 120E 140E 160E 180E 150W 140W 120W IOOWFIG. 2. (Continued)1596 MONTHLY WEATHER REVIEW VOLUME 113VELOCITY POTENTIAL 200MB JAN 1979 106M2S24,, '30N !' ~63os :.' ~:' '."i~:: , 90w120W 90w VELOCITY POTENTIAL ::>OOMB JUL 1979 106M2S24,N :::::~:,_2.~ 0 2' ;"' Y"'~ ' ' " 'ii!i: ' i "' q';'i'; '" .... ~-12~-'~~Q~";":~'~i~~~~ -.. 'i.: .-'- , t4.1 / f 90W60W SOW 0 30E 60E 90E 120E 150- 180 150W 120W 90WFIG. 3. Velocity potential at 200 mb for the months of (a) January and (b) July. Pasch (1983) examined the evolution and energeticsof the planetary scale motions of the monsoon. Hisstudy was based partly on observations (diagnostic) andpartly on the basis of global predictions. Several interesting aspects of the planetary scale emerged from thesestudies. He noted a marked increase (as a function oftime during the monsoon onset) of the planetary-scaleeddy kinetic energy over the global trdpics. Coincidentwith that was a marked increase of the eddy availablepotential energy during the period of onset of monsoonrains over India. In a further diagnosis of this systemhe emphasized the role of the organized planetary scaleheat sources in the generation of the eddy availablepotential energy. These in turn were converted to thestrong planetary scale motions via east-west overturnings. A key element in the planetary scale generationTABLE 2. List of useful symbols.xA.B.K,K,,,Kv, K~,,Kxzonal velocitymeridional velocityvertical velocitypotential temperaturespecific humiditystreamfunctionsvelocity potentialenergy exchange from A to Bkinetic energyfrictional force per unit mass of airof eddy available potential energy was identified withthe strong convection over the southeastern part of theTibetan Plateau. These results confirm the earlier observational findings summarized by Kanamitsu andKrishnamurti (1978).3. Differential heating and heat sources An important relation between differential .heatingand the maintenance of strong circulations can be deduced from simple energy considerations. That relationsimply states that in a closed system if the rotational,divergent and available potential energies are increasing(or in a statistical steady state), the maintenance ofsuch a system against frictional dissipation requires adifferential heating mechanism. The first applicationof this principle appeared in a zonally symmetric monsoon model developed by Murakami et al. (1970). Asimple zonally symmetric system may be described bythe following equations (A list of symbols is providedin Table 2).Momentum Equations Ou Ou Ou - v - --+fv+Fx (1) ot '- op Ov _ vOv wOv gOz Ot ~yy ~ fu--~--ffy + ry (2)SEPTEMBER 1985 T.N. KRISHNAMURTI 1597~ ~0mz 5oo_ 4oI~ 3oZ I0 0850 mb K~I I0 20 30 I0 20 3,0 I0 20 31MAY ,JUNE dULY DAYS "FIG. 4. Time evolutions of the rotational (K,) and divergent kinetic energy (Kx) over a MONEX domain at 850 mb; units m2 s-2.Hydrostatic relation gOz _ RT (3)Mass Continuity EquationOv + Ow 0 (4)Oy OpFirst Law of Thermodynamics - 1~ \R/cv O0 vO0 ~o00 ! vo H ot oy b~v +- - (5) ce \p/ 'Here H denotes diabatic heating. The energetics of this system may be expressed bythe relations: 0 O-~ Ku = (Kv. Ku) - Ox (6) 0 0-~ Kv = (e. Kv) - (Ko. Ku) - Dv (7) 0 0~ e = -(e' Ko) + G - De. (8)where Ku, Ko and P denote the zonal, meridional kinetic and available potential energy respectively, overthe entire mass of the atmosphere and Dx, Dy and Dedenote dissipation of the three aforementioned energyquantities. In a system that is either exhibiting an increase ofK,, Ko and P or is in a statistical steady state the following inferences can be made. If the dissipation termsD~, Dy and De are positive definite quantities, then (i)(Kv. Ku) must be positive, (ii) (e. Ko) must be positiveand (iii) a net generation (i.e., G > 0) is required tomaintain this system as stated above. The generationterm here is simply measured from a covariance of thediabatic heating and temperature. Murakami et al. (1970) utilized such a frameworkto simulate a zonally symmetric monsoon in a generalcirculation model. The model included oceans to thesouth, and land areas with mountains to the north.Other features were air-sea interaction, convective adjustment, detailed radiative processes and large scalecondensation. With the inclusion of mountains and amean July solar insolation input the model simulateda realistic monsoon including such features as: themonsoon trough, monsoon rainfall, warm troposphere,tropical easterly jet, strong Hadley circulation andlower tropospheric monsoon westerlies. A similar generalization on the role of differentialheating and monsoon circulation for a fully three-dimensional motion was presented by Krishnamurti andRamanathan (1982)- The analogous energetics of afully three-dimensional system may be written for therotational kinetic energy K~, divergent kinetic energyKx and available potential energy P over a dosed region: OKi _ (Kx. Ki) - D~ Ot OKx- (Kx.K~) + (P.Kx)- Dx Ot OP - (P.Kx)+G-De' OtHere G denotes the generation term and the dissipationterms are denoted by D~, Dx and De, respectively. Itis worth noting that the pressure interactions are absentin the zonal energy equation of a zonally symmetricsystem since Op/Ox = 0. When the fully three-dimensional system is in cast terms of the rotational and divergent motions, pressure interactions again vanishfrom the rotational kinetic energy equation. The energetics of the two systems and their interpretations arequite analogous. That is not too surprising since themeridional wind in a zonally symmetric system is entirely divergent and the zonal wind is entirely rotational. Thus, in a situation such as the onset period ofthe monsoon when the rotational, divergent and available potential energies of the monsoon region are in900~00?00 ~V~'V~~00500400$00~oo 100 0-IO0-~oo ,, ,b ,~o ~, k, ' ~'0 MAY dUNE30 I0 20 3,1 dULY 1979 FIG. 5a. The principal term in the energy exchange from the divergent to the rotational wind over a MONEX domain at 850 mb;units m2 s-3.1598 MONTHLY WEATHER REVIEW VOLUME 11320N ~.~~~~ .~I 0oN IO0 IOO30E 40 50 60 70 80 90 I00 I10 t20 130 140 150E4ONe:30N'~ IOS[~ $OE 40 50 60 70 80 90 I00 I10 120 130 140 150Eo x.../2os 8~,~-~OE 40 50 60 70 80 90 I00 I10 120 I.'50 140 150E FIG. 5b. The three panels show the distribution of the streamfunction (top), velocity potential (middle), and the energy exchangefrom the rotational to the divergent component (bottom) at 850 mb.The results here are based on a pre-onset period.creasing with time, a net differential heating is requiredto generate available potential energy. The availablepotential energy must transfer energy to the divergentmotions. Finally, we can make the statement that .divergent motions must transfer energy to the rotationalmotions. All of the aforementioned arguments presuppose that we are dealing with a closed system. Krishnamurti and Ramanathan (1982) examinedthe energetics of the three-dimensional problem in theabove context. Figure 4 illustrates the evolution of therotational and divergent kinetic energy over a MONEXdomain, at 850 mb, during the summer months. Similar evolution was also evident at 700 mb. This illustration shows a rapid rise of the rotational kinetic energy in the same one week prior to the onset of monsoon rains over central India. The magnitude of thedivergent kinetic energy also exhibits a small increase40 N~ ~/:50N-~IOS~40 N -,0( ~, '30N20N' ION- ~ -1lOS?OS130 S/ . , -~OE 40 50 60 '70 80 90 I00 I10 120 130 140 150E , 30E 40 50 60 70 80 90 i00 II0 120 130 140 150E40N .~ON 7 olOS 30E 40 50 60 70 80 90 100 II0 120 130 140 15~ FIG. 5c. As ~ Fig. 5b but ro~ a. active monsoon pe~.during the onset period. Figure 5a illustrates the timeevolution of the principal term in the exchange of energy from the divergent to the rotational kinetic energyduring this period. The large increase of this exchangeduring the onset and post-onset period is attributed toan evolution of the divergent wind. As the principalrainfall belt migrates from nothern Burma towards theTibetan Plateau, the divergent wind organizes itselfwith respect to the rotational wind and a rapid exchangeof energy from the divergent to the rotational windoccurs. This feature is illustrated in Fig. 5b, c wherethe fields of the streamfunction, velocity potential andthe energy exchangefV~b - Vx during the pre- and postonset periods are shown. The region where large exchange from the divergent kinetic to the rotational kinetic energy occurs shows an int6resting evolution ofthe monsoon. This same scenario was further expanded in a seriesof numerical weather prediction experiments bySEPTEMBER 1985 T.N. KRISHNAMURTI 1599I009080 ?0 60 50-, 4b~ 2o~. ~oo_ oX -I0 -20 -'50 -40 POST ONSET<Kx'K~ =+ VzXV~)~ + wd[~ ~x) LATE SPRINGTIME HOURS) --~ FtG. 6. Time evolution of the energy exchange from the divergentto the rotational component. The results shown here are based onthe time integrations of three separate runs of a multilevel primitiveequation model. This illustration is taken from Krishnamurti andRamanathan (1982). Units m2 s-3.Krishnamurti and Ramanathan (1982). In these experiments with a multilevel primitive equation model,they examined the evolutions of the flow fields usingidentical initial states for the rotational wind, the sealevel pressure and temperature fields. Experiments,however, differed from each other in their initial description of the divergent wind and the humidity fields.The different experiments described the pre-onset conditions of the large-scale flows, while the different configuration of the divergent wind and the humidity fieldsprovided different magnitudes of differential heatinginitially. In a three to four day prediction, the largestresponse of the initial state to the differential heatingwas noted where the initial heating was introduced overnortheastern India and the foothills of the Himalayas.In this instance a rapid monsoon onset occurred in amatter of three to four days. The detailed energetics ofthis experiment essentially confirmed the scenario presented above. Figure 6 shows the evolution of the energy exchange from the divergent to the rotational windin these three experiments. A differential heating corresponding to the post-onset period exhibits an explosive growth of the monsoon circulations. This experiment strongly suggests the need for a careful determination of the initial heat source in the monsoonprediction problem.4. Apparent heating and moistening Yanai et al.'s (1973) definition of the apparent heatsource Q~ and the apparent moisture sink Q2 providesa powerful tool for diagnostic studies of the atmosphere.Figure 7a-d illustrates Luo and Yanai's (1984) estimates of Q~ and Q2 over four relevant regions; theseare over the western and eastern regions of Tibet, theYangtze River valley over eastern China and the Assamvalley of northeastern India. These four regions arecharacterized by relative dry weather, moderate convective rain, steady stratiform rain and heavy convective rain respectively. The respective vertical distributions of Q~, and Q2 are shown here. Luo and Yanainoted that over western Tibet, Q2 is much smaller thanQ~ which suggests the absence of deep convection. Theyattribute the strong values of heating over western Tibetto the dry convective process. However, it is possiblethat dust and aerosols may contribute to a strongerradiative heating over this region. Over the region ofeastern Tibet about half of the net heating is attributedto latent heat release. Region III, which is over theYangtze River valley, shows apparent heating anddrying of nearly the same magnitude; this, accordingto Yanai and Luo, is a signature of steady stratiformrain. In region IV, which is over northeastern India, heavyconvective rain accounts for large heating rates, Q~~ 5-C day-l, and the apparent moisture sink Q2 hasan intensity of around 2.5-C day-~. Thus a variety ofdifferent profiles of Ql, Q2 are encountered over different regions of the monsoon. - In this context, it is worth mentioning some important recent studies done by Chinese authors on theformation of diurnal mixed layers and heat low on theTibetan Plateau (e.g., Yeh et al., 1981; Yeh, 1982; andGao et al., 1981). Further work remains in areas of geographical andtemporal variability of Q~, Q2 and their relationshipto the overall monsoon activity.5. Planetary boundary layer studies, cooling of the Arabian Sea and the humidity budget Two major areas of interest in this area of investigation are dynamics (i.e., the cross-equatorial flow dynamics within the PBL) and the thermodynamics (i.e.,those related to the air-sea exchanges). The crossequatorial low-level jet was monitored by some 80 lowlevel constant level balloons, launched from DiegoSuarez located near 15-S, at ~900 mb level (Cadet etal., 1981). The low-level cloud motion vector providedanother major data set for studies of the cross equatorialdynamics. This problem was addressed by Stout andYoung (1983), Sommeria et al. (1982) and Krishnamurti et al. (1983a). The transition in the balance offorces along the low-level flows from the SouthernHemisphere trades to the northern monsoon trough isdominated by advective acceleration in the near-equatorial latitudes. The transition from a near-Ekman typebalance, at -~20-S, to an accelerating flow along thelow jet, near the equator, results in a strong role forthe horizontal as well as the vertical advection terms.In this near-equatorial region the essential balance isamong frictional, advective and pressure gradientforces. Parcels of air within the PBL experience a strongbacking of winds with height at --~ 15-S. As these parcels1600 MONTHLY WEATHER REVIEW VOLUME 113 REGION Iloo1so2002~o3oo400500735MEAN-2 -1 0 1 -1 0 1 2 3 4 REGION TIT100 ~50 700 gSO g~o I I t -4 -3-,2 -1 0 1MEANI t t Lt ~ t I -2 -1 0 1 2 3 4 REGION TT MEAN100, ! L~ I -l I i I II II I ~ i i I 150 200250 3O0' 4005OO 610-4 -3 -2 -1 0 1 -1 0 I 2 3 4 REGION100150 - /200 300 400 700 - I~1835 ;55 I I I I I -4 -3 -2 -1 0 1 2 MEAI~ ,, I__, .~ 01 Q; I 'I III II I I ~ I-1 0 1 2 3 4 5 6 FIO. 7. Forty-day mean vertical distribution of the areal mean vertical p-velocity (mb h-~) heating rate Q#Ct, (K day-~) and dryingrate Q2/Cp (K day-~) over western Tibet, (Region I), eastern Tibet (Region II), Yangtze River Valley (Region III), Assam-Bengalregion (Region IV). These are based on Luo and Yanal.approach the equator the role ofadvective accelerationsbecomes quite important. With the diminished roleof the Coriolis force, the departure from Ekman balanceis prominently noted in the balance of forces. Theequatorial region is characterized by either weak backing or weak veering of wind with height depending onthe past history of the parcel speeds. The parcels arrivewith a stronger backing of wind with height over theSouthern Hemisphere subtropics. As they arrive overtropical latitudes this configuration results in a layeraveraged diffiuent wind and downward motions overthe eastern Indian ocean. Downstream from the maximum wind of the Somali jet a reverse picture in theboundary layer dynamics is found. Here the transitionfrom a near-equatorial advective to a more Ekmantype boundary layer dynamics results in a downstreamincrease of veering within the PBL. An asymptote ofconvergence and a gradual increase of.height of thePBL is noted in this region as the monsoon convectionincreases.a. Cooling of the Arabian Sea The Arabian Sea oceanographic experiment was animportant component of MONEX. The emphasis inthese studies has been on equatorial waves, the Somalicurrent and the mixed layer structure (over selectedsites). In this review we shall present some of the majorSEPTEMBER 1985 T.N. KRISHNAMURTI 1601 16Ib. lC:3 I0~b.I 8(3,.(n 6Z 4 o,;' 'do" '~'5' ;"'; ',;' MAY JUNE JUNE DAYS [1979) 2'~5 20 25 I 5 I0 15 MAY JUNE JUNE DAYS 0979) ~ FIG. 8. Time section of surface wind speed (m s-I) and sea surfacetemperature as measured by a research ship UMAY over the ArabianSea during the onset of monsoon 1979. Location of ship near 7-N,65-E.findings on the cooling of the Arabian Sea based onship observations during MONEX. An anomalous feature of the sea surface temperature(SST) over the monsoon region is the cooling of theArabian Sea during June. Between 10 and 20-N fromJune through August a warming of the SST nearly always occurs over the Atlantic and the Pacific Oceans.The Arabian Sea alone exhibits cooling during this period over comparable latitudes. Diiing and Leetmaa(1980) and Krishnamurti (1981) among others haveaddressed the problem. The magnitude of this coolingis of the order 3-4-C over large areas of the ArabianSea. The cooling starts to occur dramatically over aperiod of 5-6 days soon after the winds strengthen.Figure 8 illustrates the time evolution of the surfacewind speed (top panel) and of the SST (bottom panel)during the Monsoon Experiment. These measurementswere made by a Soviet research ship (with the call signUMAY) around 60-E, 6.7-N. The wind speeds buildup from the 10th to the 15th of June (from roughly 4to 14 m s-l), while the SST drops from '-~30 to~27.5-C at this location. Similar response of the oceanto the strengthening of the winds was noted at severalother locations. Several possible mechanisms for thecooling have been proposed (these are reviewed inKrishnamurti, 1981): 1) The onset of monsoons brings in extensive cloudcover with large amounts of high-level cirrus over theArabian Sea. This acts to diminish the incoming solarradiation. The cooling of the Arabian Sea is generallynot attributed to this effect because the oceanic responseto solar radiation has a lag of one to two months, whilethe observed cooling starts almost immediately afterthe onset of strong winds. 2) Southward flux of heat by ocean currents is contributed by clockwise oceanic eddies over the westernArabian Sea. This was estimated by Diiing and Leetmaa (1980); the related cooling of the northern ArabianSea by equatorward transport appears to be quite small. 3) Coastal upwelling and downstream shedding ofcold eddies: The premise that pockets of cold anomaliesform in the Somali and Arabian upwelling regions andare advected eastward by the broadscale Somali currents is a possibility. There is some evidence of thatfrom satellite observations. The rapid response of SSTto the rapid evolution of strong winds in central andeastern Arabian Sea during the onset precludes thispossibility since this requires an eastward advectivespeed of the cold pockets, from the upwelling regionsoff the east African coast, comparable to wind speedsin the low-level atmospheric jet. The oceanic advectionis known to be much smaller to account for a sufficientdivergence of flux and the known cooling of the ocean. 4) Strong evaporation in the region of strong windsis a possible candidate. However, the available budgetsof evaporative cooling again do not favor such an intense cooling. We feel that this effect needs to be assessed in detail with the boundary-layer humidity fluxmeasurements for disturbed and undisturbed conditions. The sea state over this region was highly turbulentduring the onset of monsoon. Flights made with lowlevel aircraft exhibited a very large number of whitecaps with a greenish color of water, indicating almosthurricane force surface winds during the passage of theonset vortex. With the large sea spray and strong winds,it is conceivable that evaporative cooling was underestimated in the past. During the onset of monsoons, the cooling of theArabian Sea also occurs over the interior of the ArabianSea where the winds increase explosively. Figure 9a-dillustrates the possible role of the wind stress over theinterior Arabian Sea on this problem. The parameters shown here are zonally averagedacross the Arabian Sea for 55-70-E (i.e., over the eastern Arabian Sea) and are based on once daily FGGE/MONEX observations. Surface wind speeds during theonset period developed explosively from 5 m s-t toalmost 25 m s-~ between 11 June and 17 June 1979around 10-N. Over the regions of the southern tradesaround 10-S, the surface winds lie between 5 and 10m s-~ during most of this period. The evolution of thestrong winds around 10-N arises from a superpositionof the westerlies to the south of the onset vortex andthe Somali jet. The surface wind stress increases sixfold from around 1 dyn cm-2 prior to the onset toaround 6 dyn cm-2. Over the southern trades the stressduring the onset period remains at around 1-1.5 dyn1602 MONTHLY WEATHER REVIEW VOLUME 113b.I:~ 2OD-~~.O>-15:50S 20 I0 EQ I0 20N LATITUDE bLATITUDE;508 20 I0 EQ I0 20N d212~2~24~?~4LATITUDE SST25 20 15 I0 5 I:O 5 I0 15 20N LATITUO-FIG. 9. Time-latitude evolution of (a) surface wind speed (m s-~), (b) surface wind stress, (c) wind stress curl, and (d) sea surface temperature (-C), averaged zonally from 50 to 70-E.cm-2. The wind stress curl shows a positive and a negative axis on either side of the axis of the low-levelwind maximum. The positive wind-stress curl regionnearly coincides with the region of the Arabian Seacooling illustrated in Fig. 9d. These calculations arestrongly suggestive of the role of the evolution of theinterior ocean stress in the cooling of the Arabian Sea.Further studies on this problem require observationsand modeling of the mixed layer. In summary, thereis a need for more work on this problem. Coupledmodels may be very helpful in our understanding ofthis problem.b. Humidity budgets Although research aircraft were deployed within thePBL during MONEX, the fluxes of sensible and latentheat and their vertical distribution have not been evaluated to date. Studies on the humidity budget over themonsoon region were evaluated by Greco (1984). Thesebudget studies are somewhat revealing of the monsoonactivity and the latent heat fluxes. Using Tiros-N precipitable-water estimates in three layers, 1000-700,700-500 and 500-300 mb pressure levels and theFGGE/MONEX winds at 1000, 850, 700, 500, 400SEPTEMBER 1985 T.N. KRISHNAMURTI 1603Q 30 2o I0 0 -I0 -20 -30b 3O 2O I0 0-I0-20-30 $0 '"~ , \~,q~..~ i ~ , ~t....~(..~ ,~ , ,-% - .2~ ~ ~' I L ~5)' - F' L/ I X I ~ I ! ! ~ , ,-~,lr- , r-..~uT--%T 0 40 50 60 7'0 80 90 I00 I I 0 120 , '~-. , , , ~ ~ r~..., '~.._j.~r , ,~, / - I 40 0 60 70 80 90 I00 I lO IZO 20 L L ,o 3o 4o ~o 6o ?o 80 9o ~oo t~od3- ~' ',~.~L/ ~.J V"~o~r'~ o 30 40 ,50 60 70 80 90 I00 I I0 120FIG. 10. Mean zonal water vapor ttux fields for: (a) 1-15 June; (b) 16-30 June; (c) 7-18 July; (d) 25 July-6 Aug. Units kg/m s-~ (from Greco, 1984).and 300 mb pressure levels, Greco estimated humiditybudgets over a MONEX domain. The results of meanzonal water vapor fluxes during the pre- and post-onsetperiods are of considerable interest. These are verticallyintegrated measures (Fig. 10). They show a substantialdivergence of flux of water vapor by the zonal winds,suggesting strong evaporative fluxes over the ArabianSea during the post-onset periods. A similar calculationof the divergence of flux of moisture by the meridionalwind across the Arabian Sea, between the equator and20-N, shows that the cross equatorial transport of humidity is substantial. The humidity transport acrossthe equator and the evaporation from the Arabian Seaseems to be equally important for the Indian Monsoonrainfall. Thermodynamic variations of the planetary boundary layer along the low-level flows over the ArabianSea deserve to be studied further. The surface fluxes,their vertical distributions, their interactions with thecloud layer, effects of vertical wind shear, effects oflarge scale vertical velocity and downstream variationsof the height of the planetary boundary layer are someof the presently unresolved areas. Much of our presentunderstanding of the planetary boundary layer hascome from studies over other oceans. The AtlanticTrade Wind Experiments (ATEX) and several experiments in the Pacific such as the Airmass ModificationExperiment (AMTEX) have provided relevant information in these areas. The Arabian Sea is unique inhaving the deserts to the north of a strong low-levelwesterly jet. In that sense, we believe that this regiondeserves a stronger focus on PBL investigation.6. Monsoon onset During the monsoon onset various dramatic changesare known to occur in the large-scale atmosphericstructure over the monsoon region. Some of the wellknown elements of the onset are a rapid increase ofdaily precipitation rate, an increase in the vertical integrated humidity (manifested as a meridionally propagating deep moist layer) and an increase in the kineticenergy, especially of the low-level flows. Although themode of onset does vary somewhat from one year tothe next, one or more of the aforementioned eventsare known to occur during the period of the onset. Asthese low-level winds grow in intensity, a cyclonic stormhas been known to form on the cyclonic shear side ofthis wind current near 10-N, 65-E. It appears that insome four out of six years there has been evidence,based on 80 years of observation, of the occurrence of1604 MONTHLY WEATHER REVIEW VOLUME 113 850 MB WINDS 12GMT II JUNE 1979~ON ~ ~ ,~-~.,~.~_ +.,~ /,""~",.'?:/:(.~1 ',, ~.'~~ON, "_~ t /~40E 50E 60E TOE 80E 90E850 MB WINDS 12 GMT 18 JUNE 1979 .-. . I40E 50E 60E 70E 80E 90EFiG. 11. 850 mb streamlines and isotachs (m s-') over the MONEX domain.(a) 1200 GMT I June 1979. (b) 1200 GMT 18 June 1979.. ,o.p ~-~,~,,/~ ~o??~/~//~//~ ~o~,~ -----_~>- : lOS 205 'sos'- - ----------~-~-- ;'6" ' ~5- ion i'~ - ~o MAY DATE JUNE FiG. I2. Time-longitude section of zonal velocity (ms-~) at 850 rob. These are based on zonal averages between 50 and 70-E.an onset vortex (Krishnamurti et aL, 1981 a). It usuallyforms on the cyclonic shear side of the low-level jet inthe lower troposphere over the eastern Arabian Sea;frequently, it moves meridionally towards the northernArabian Sea and subsequently its motions are morewestward usually toward the Arabian Coast, where itis known to dissipate. It has also been noted to firstform in the middle troposphere over the eastern Arabian Sea and subsequently cyclogenesis occurs in thelower troposphere. Figure 1 la, b illustrates two 850mb charts for the.formafive~ stage of the onset vortexand the dramatic increase of monsoonal southwesterlyflows over the Arabian Sea. Formation of the onsetvortex has been attributed to the instability arising fromSEPTEMBER 1985 T.N. KRISHNAMURTI 1605the explosive increase of horizontal shear of thebroadscale monsoon current.a. Barotropic dynamics and the onset vortex Figure 12 shows the time evolutions of the zonalshear flows as a function of latitude. Here the zonallyaveraged (between 50-E and 70-E) daily values at 850mb are shown. The onset vortex formed on 14 Juneat this level, and ~ 10-N over a region of marked cyclonic shear. Krishnamurti et al. (1981a)examinedvarious aspects of the barotropic dynamics to assessthe role of the horizontal shear flows. These included: i) An examination of the meridional gradient of absolute vorticity over the Arabian Sea in order to assessthe so-called inflection point instability (see Fig. 13b), ii) An examination of the linear barotropic growthrates as a function of scale following Yanai and Nitta(1968) (see Fig. 13c), iii) An evaluation of the nonlinear barotropic energyexchange from zonal flows to the eddies over the Arabian Sea. (see Fig. 13d), and iv) A barotropic forecast experiment during the formation of the onset vortex.Figure 13a also includes the profiles of the mean zonalflows over the Arabian Sea for this period, which exhibits a marked increase during this period. Figure 13bshows that the necessary condition for the existence ofbarotropic instability was satisfied by the above zonalflow profile--especially during the formative period ofthe onset vortex~ It furthermore shows that the barotropic growth rates for horizontal scales of the orderof 3000 km were indeed quite large during this period.These large growth rates, especially around 15 June,implied an e-folding time of the order of a couple ofdays. The results of nonlinear barotropic energy exchange show a significant exchange from the local zonalflows to the eddies throughout the onset as well as thepost-onset period. The lack of disturbance .activityduring the post-onset period is usually attributed to thecooling of the Arabian Sea and the buildup of strongeasterly vertical wind shear; here, it was of the orderof 40 m s-~ over a 650 mb depth of the troposphere.On the other hand, wind shears over the Bay of Bengalare usually quite weak during the post-onset period(Raman et al., 1981). Furthermore, the activity ofmonsoon over northeastern India results in a compensating general descent and dry troposphere over theArabian Sea, thus providing an additional stabilizinginfluence. These features can be deduced from the geometry of the velocity potential field at 200 mb. A barotropic forecast experiment (Krishnamurti etal., 1981a) resulted in the formation of a closed circulation near the region of the onset vortex. That wasbased on a conservation of absolute vorticity. The initial state was derived from the 850 mb motion field on13 June, some two days prior to the formation of theonset vortex. The formation of the closed circulationwas very clearly due to a transformation of shear vorticity into curvature vorticity in this experiment. The aforementioned analysis is strongly suggestiveof the importance of the barotropic dynamics duringthe formation of the onset vortex. This was also con~firmed by Mak and Kao (1982). It should be stated,however, that the onset vortex is not a necessary ingredient for the onset of monsoon rain over India. Itspresence in 1979 appeared to enhance the low-levelwesterlies on its equatorward side. The more importantaspect of the onset of monsoon is the establishment ofdeep moist westerlies that are evidently related to abroader scale pattern of differential heating discussedin Section 3.b. Numerical weather prediction Numerical weather prediction of the onset of monsoon in 1979 was addressed by a number of large scalemodeling groups. The modeling groups included theECMWF, LMD, FSU, RPN, JMA, UKMO and NMC.All of these groups utilized global models with differentdegrees of horizontal and vertical resolution. All of themodeling groups utilized identical initial states basedon the FGGE IIIb data (ECMWF multivariate optimalinterpolation and four-dimensional assimilation analysis). The results of the intercomparison experimentshave been documented by Tempert0n et al. (1983).The experimental forecasts started at 1200 GMT 11June 1979. That was almost a week prior to the onsetof monsoon rains over central India and was the periodof commencement of the explosive increase of lowlevel kinetic energy. The formation of the onset vortex,the establishment of deep moist westerlies, the prediction of rain and the evaluation of root mean squareerror statistics were the goals of these experiments. Weshall not go into an extensive review of these resultshere since they have been presented in some detailelsewhere. The most promising results were those produced by the ECMWF, LMD and the FSU models.The onset of monsoon was found to be most sensitiveto the cumulus parameterization procedures. The useof an unusually strong heating (a plume convectionmodel utilized by the UKMO) resulted in the formationof a hurricane-like vortex over the Arabian Sea. Thedrastic effect of this simulation was that low-level flowsover India were strong easterly--rather than thesouthwesterly monsoon. On the other hand, the use ofvery weak convection (classical Kuo's scheme) resultedin persistent dry northwesterly winds at the low levelsover India. An improved version of the Kuo's schemebased on Krishnamurti et al. (1983c) provided a reasonable simulation of the monsoon westerlies and theonset of monsoon rain. It required the use of steeperorography, called the Envelope mountains (Wallace etal., 1983), for a realistic prediction of the onset vortex,monsoon rains and the southwestefiies. Figure 14 illustrates the initial 850 mb flows for 1200 GMT 111606 MONTHLY WEATHER REVIEW VOLUME 113 9 dUNE 13 ,JUNE 25' *~n 20 I--~~l 20E 15 15I~,o io~ ~O ~ -5 -5 I -I0 I , * * , I, , , , -I0~ -20 0 20 40 -20 0 20 40'e ~ LATITUO[ '; 20 r 20 =~ ,5~- ,5 15 b 'oF ~o ,5?. / i~~i~1~ -,o~ -,o ~ -15~ -15 ~ -201 ....... ' - ' .... -~0 -20 0 20 40 -20 0 20 40 -20 0 20 40 '~ LATITUDE ;I , ~ 7~ 6 8'~ 5, 5~ 4 40 3~ 2 ~ I~- I I - I I ~ 0 0~ O' ' ' 0 I~ ~ 4 56 7 8 0 I 25456?8 0 I~ ~ 4 5 678 ( WAVE LENGTH (lO6m)850 mb 14 dUNE 17 dUNE';I ~'5 20 - 20 15 15 ; ~0 0 0 -5 I '5' -I0 -I0 -' -20 0 20 40 -20 0 20 40 II ;I I~ I0 ~ -I0 -I$ -tO ' ' ' ' ~ ~ ' ~ -~0 0 ~0 40 ~~ ~ I 4 I *J~ ~ ' I I 0 125456 78 . J FIG. 13. Top panel: Left to right shows the mean zonal velocity (m s-b as a function Of latitudefor' 9, 13, 14 and 17 June, respectively. Middle panel: Left to right shows the mean meridionalgradient of absolute vorticity as a function of latitude for 9, 13, 14 and 17 June respectively. Bottompanel: Left to right shows the growth rate of barotropically unstable disturbances as a function ofwavelength for 9, 13, 14 and 17 June, respectively.aJune for this experiment. The 6-day forecast and observed fields of the motion field at 850 mb are shownin Fig. 15a, b. The track and the formation of the onsetvortex were reasonably predicted by the global model.The onset of rains on 15 June was reasonably handledby this model; the observed and predicted precipitationfields on day 4 are shown in Fig. 16a, b. The FSUglobal model is described in Krishnamurti et al. (1983b,1984b). These experiments describe some of the firstsuccessful efforts on medium range numerical weatherprediction over the tropics.7. Monsoon depressions During the Summer Monsoon Experiment of 1979,a number of tropical depressions formed over the Arabian Sea and the Bay of Bengal. Two of these werewell documented in the literature largely because ofthe special observing systems that were deployed duringtheir life cycle. A number of previous' studies on suchdepressions dealt with descriptions when they had already formed and were located over a region well covered by the continental radiosonde network. The monsoon experiment provided a unique opportunity to examine two of these disturbances during the formativestage and during their entire life cycle as well. One ofthese was the disturbance over the Arabian Sea discussed in Section 6. In a recent study, Lindzen et al. (1983) ascribed thegrowth of a Bay of Bengal monsoon depression to thehorizontal shear flow instability mechanism. Lindzenexamined the asymptotic response to pulse perturbations. These are the instabilities generated from localized perturbations in horizontal shear flows. Lindzenconcludes that, in the month of July, the barotropicinstability appears to be the most likely mechanism.He also deemphasizes the role of vertical shear, statingthat it must be quite weak. The role of CISK is negatedfrom considerations of vertical momentum transport(cumulus friction) which he finds has a stabilizing influence. Observations over the Bay of Bengal do notseem to substantiate a barotropic vertical structure; thehorizontal temperature gradients are large and theequatorward slope of the disturbance towards cold airis quite substantial. Observationally, Douglas (1984) noted a strong vertical tilt of the depression toward colder air aloft, suggesting the importance of vertical shear. Figure 17a-c,obtained from a detailed.and careful analysis of Douglas, shows a pronounced equatorward tilt. Douglasmade use of nearly all of the available platforms of theFGGE/MONEX observing system to arrive at thisstructure: the data sets included marine ship observations, WWW, cloud winds, detailed collections ofcommercial aircraft, dropwindsonde data from research aircraft and soundings from FGGE/MONEXSEPTEMBER 1985 T.N. KRISHNAMURTI 1607~ / i Lii i i i i I t i i i i [ i i t i i t i I I I I I I I I i i i i i i i I I0(10'4) I(.lO's ) ~(10'~ ) ~ I(10-? ! P;6 18 20 22 24 26 28 ;~0 I :~ 5 9 II I$ 15 I1' 19 21 2:~ 25 21 29<Kz.K( - - MAY ]'- JUNE P(m~ s's) -I 17 19 21 2~ 25 27 28 51 2 4 6 8 I0 12 14 16 18 20 22 24 26 29 30(10-?) -I(10'~' 1-5 ol(tO-s) -5 -I0dFIG. 13. (Continued)research ships. Douglas' analysis appears to substantiatethe theoretical findings of Arakawa and Moorthi (1982)who have emphasized the importance of vertical shearand CISK in the growth of monsoon depressions. Arakawa and Moorthi (1982) examined the growth ratesfor the classical Green Modes in easterly shear. Weakeasterly vertical wind shear is a characteristic featureof the flows over the Bay of Bengal during the summermonsoon months. They noted that for a reasonableeasterly wind shear of the order of 20 m s-~/800 mb,large growth rates for horizontal scales of the order of1000-2000 km is possible with an e-folding time of acouple of days. Furthermore, they noted that in theenergetics of the amplifying modes the role of cumulusconvection is not only to generate eddy available potential energy, but also to provide an enhancement ofthe baroclinic energy conversion. Saha and Chang(1983) and Shukla (1978) have also examined thedepression/dynamics. Nearly similar results on the energy conversions,based on observational analysis of a depression (afterthe formative stage) were shown by Krishnamurti eta/. (1976). It is indeed surprising that during the formative stage, the role of vertical shear and cumulusconvection has been shown to be significant since thehorizontal shears are in fact quite large. A close examination of satellite imagery during the incipient stageof the depression does show some organization of cumulus convections over the eastern Bay of Bengal(Warner, 1984). Sanders (1984) examined the quasigeostrophic omega equation over this region and notedthe importance of the baroclinic forcing mechanismsin determining the field of quasi-geostrophic verticalmotions. That is perhaps also suggestive of the mechanism proposed by Arakawa and Moorthi (1982). A further step in such an analysis would be a detailednumerical weather prediction of a depression and asubsequent diagnosis of the results. That was carriedout in two recent studies by Krishnamurti et al. (1983a,1984a). These studies were based on a global spectral model.Medium range forecasts were made with real initialdata. The initial state for these experiments was takenfrom the FGGE/MONEX observations at 1200 GMTI July 1979. The depression formed over the Bay ofBengal on 5 July and arrived at the coast of India on7 July. The global spectral model is resolved via a triangulartruncation consisting of 42 waves and 11 vertical sigmalevels (Krishnamurti et al., 1984a). The use of a semiimplicit time differencing algorithm enables use of atime step of ~ 15 minutes providing a considerablesaving of time for medium range numerical weatherprediction. The mountains in these series of experi1608 MONTHLY WEATHER REVIEW VOLUME 113~ON30NON30S 30E0 HR FCST UV8SO MON JUN 11 78 I2Z (INT= .SOE+01) ,-k ,, $OE 90E 120E F~G. 14. The initial flow field at 1200 GMT 11 June 1979 at 850 mb over the MONEX domain. Speed m s-L Note that this is a part of the global analysis for the global model.ments are defined by the so-called "Envelope Orography" following Wallace et al. (1983). The basic datasets are the FGGE/MONEX observations available asof 1983. The FGGE IIIb analysis produced by the European Centre for Medium Range Weather Forecasts(ECMWF) were used as a first-guess field in our analysisof the MONEX observations. The analysis scheme usedby ECMWF is a multivariate optimal interpolationscheme and a four-dimensional assimilation schemeusing a 6-hour forecast assimilation. The MONEX datasets were added on via a successive correction method.The initialization for the present series of experimentsis based on physical and dynamical initialization; thisis described in some detail in Krishnamurti et al.(1984a). The cumulus parameterization method deployed here is a variant of Kuo's (1974) scheme andis described in detail by Krishnamurti et al. (1983c).Over regions of dynamical ascent of absolutely stablesaturated air, large-scale condensation is involved. Themodel includes a fairly detailed parameterization ofthe radiative processes where the diurnal change, cloudfeedback processes and energy balance of the earth'ssurface are included following Chang (1979). The surface layer fluxes are determined by similarity theorywhere the exchange coefficients are stability dependent.The vertical variation of the surface fluxes in the planetary boundary layer are defined by a mixing lengthK-theory. The model included a dry convective adjustment for the removal of superadiabatic lapse rates. The FSU model was adapted from the original version proposed by Daley et al. (1976) and extended bythe author and his colleagues. The prediction of themonsoon depression was successful on the mediumrange time frame of six to seven days. The success ofprediction was due to several factors such as data sets,improved analysis of the initial state and an improvedmodel. It is difficult to isolate any single major factorwithout a further series of carefully constructed numerical experiments. Figure 18a illustrates the initialflow field 1200 GMT (1 July 1979) at 850 mbs. It showsgeneral westerly flows over the Bay of Bengal. Thenorthern part of the Bay of Bengal exhibits a region ofcyclonic horizontal shear. At 200 mb the flows exhibitbroad eastefiies over the Northern Bay with an easterlyshear of around 30 m s-~ over a 650 mb depth of thetroposphere. Over the Eastern Bay and northern Burmadense convective clouds were apparent from satelliteimagery (Krishnamurti et al., 1979). The depressionthat formed on 4 July in the middle troposphere wasapparent at the 850 mb level by 5 July. Figure 18bshows the observed flow field for July. The 6-day forecast of the depression is illustrated in Fig. 18c. Severalvertical cross sections and time sections of the predictedstorms were also constructed. They revealed that theinitial formation of the storm did occur at the middletroposphere. It is usually assumed that the result of (i)linear stability analysis, (ii) observational energetics,and (iii) energetics based on numerical simulation orSEPTEMBER 1985 T.N. KRISHNAMURTI~ON OBSERVED UV8SO FOR dUN 17 79 12Z INT-160930NON3qS 30E 60E 1@.2 90E 120E1SOE~ON 1L~L HR FCST UV8SO MON JUN 1t 79 12Z 2 SIGMR lINT= .SOE+01I30NONb3OS 30E 8 60E 90E I 20E 1SOl~ FIG. 15. (a) Observed and (b) predicted 850 mb wind field (m s-~) at 1200 GMT 17 June 1979 over the MONEX domain. Speed m s-j. (Note that this is a 6-day forecast).1610 MONTHLY WEATHER REVIEW VOLUME 11320N24HOUR P, AINi::ALL (MM/DAY) ENDING OOZ 15 JUNE 1979 ( FINAL ANALYSIS) ":":'-'..,~i' '~..."~::::':':':':::':':':':':': 'i~ii!-'? :~:ii;;~ ,~,?:~ ~:~.:~i~ ,~,~..~.~ -,.~. ~-'-~. @IONONlOS20SSOS $OE40E 50E 60E TOE 80E 90E IOOE I IOE 120E 1:50E 140E 150E40N30NON30S ;50E96 HR FCST 2 SIG PCPMON JUNII 79 IZZ (INT=.5OE+OI)60E 90E 120E 150EbFIG. 16. (a) Observed and (b) Predicted rainfall rate at 1200 GMT June 15 1979. (Note that this is a 4-day prediction). Units: mm day-~.prediction are .three different stories. However, it appears that in some gross sense the mechanism suggestedby Arakawa and Moorthi (1982) is indeed borne outby the observational energetics and by the results basedon the aforementioned numerical weather predictionexperiment.SEPTEMBER 1985 T.N. KRISHNAMURTI 1611 - -:.' .....,:.,.."~G. 17. Sreamlines and i~chs m s-~ ba~d on Dou~' an~ysis ofa mon~on depression during MONEX. Note the strong vertical tilt. (a) 400 mb, (b) 500 mb and (c) 850 rob. The observational energetics for these same stormswere also carried out by Surgi (1984). Surgi averagedthe results of Lorenz (1960) energetics for two separateperiods; i.e., for the formative and the postformativeperiods. In order to address the interpretations andambiguity of the boundary fluxes she calculated opendomain energetics over 18 different-sized boxes. Fromthese she selected a domain that provided the smallestboundary fluxes of kinetic and potential energy. Thus,her results were considered as being valid for a closedsystem. Figure 19 illustrates the results of her calculations from 75 to 105-E, and between the equatorand 35 -N through the troposphere. These results showthat in the formative stage the eddy kinetic energy Kis enhanced by the barotropic as well as the ba~oclinicenergy conversions, while the depression stage showsa more dominant role for the baroclinic process. A1though Surgi did not explicitly evaluate the generationterms, it appears that the maintenance of the eddyavailable potential energy (which was slowly decreasingduring this entire period) required a generation mechanism. The most likely mechanism appears to be cumulus convection since there were no other apparentenergy sources on the scale of the depression. The results of the energetics from multilevel globalspectral model forecasts (Fig. 20) were evaluated byKrishnamurti e! al. (1983a) over a domain extendingfrom 10 to 30-N, 50 to 100-E through the depth ofthe troposphere. The abscissa in this diagram denotes the time axis(i.e., the ten days of the medium range forecast). Thebarotropic conversion (Kz. K/~) is considerably smallerand is even negative during the formative period compared to the baroclinic conversion (PE' KE). A sharp1612 MONTHLY WEATHER REVIEW VOLUME 113a OBSERVED UV 850 JUL OI 79 t2Z (INT=.5OE+OI)850 MB WINDS 12 $MT 7 JULY 1979COh144HR FCST UV850 SUN JUL I 79 12Z (INT-.50E+OI)~OE 60E 90E IL~OE 150E FIG. 18. (a) Initial 850 mb flow field at 1200 GMT 1 July 1979over the MONEX domain (note that this is a part of the initial statefor the global model). (b) Observed flow field at 1200 GMT 7 Julyt979. (c) Predicted flow field at 1200 GMT 7 July 1979. Speeds inra s-L (Note that this is a 6-day prediction).rise of the latter between 3-7 July 1979 lends supportto the hypothesis of Arakawa and Moorthi (1982).Furthermore, it was noted by Krishnamurti et al.(1983a) that the horizontal shear flow contributed toa net increase of eddy kinetic energy at the lower levelsinitially, especially at the 900 mb level. Thus the roleof the barotropic mechanism in the very lowest troposphere during the formative stage cannot be ruledout; a marked cyclonic shear was present over this region initially. Our understanding of the monsoondepression has increased considerably since the middle1970s. Further work is needed towards the understanding of the structure and life cycle of organizedconvective elements within the depressions. Warner(1984) has provided a detailed observational analysisof the convective elements as seen from satellite imagery and aircraft photogrammetry. Those data setsneed to be examined in the comext of the generationof eddy available potential energy and the scale of thedepression.8. Breaks in the monsoon During MONEX, a major break in the monsoonwas evident between 10 and 25 July 1979. The characteristics of this break were similar to those describedby Ramamurthy (1969). The principal rainfall belts'were located close to the equator, ~80-E, and nearthe foothills of the Himalayas. The time scale of breaksis usually of the order of several days or longer. Thatsuggests the possibility of a pressure rise over the regionof the monsoon trough from low-frequency phenomenaduring this period (Sadler and Harris, 1970; Krishnamurti and Ardanuy, 1980). Sadler and Harris identified meridionally propagating ridge lines from theequatorial latitudes toward central India, whileKrishnamurti and Ardanuy identified westward propagating systems on the time scale of 10-20 days. Morerecent studies by Yasunari (1981), Sikka and Gadgil(1980), Krishnamurti and Subrahmanyam (1982),Lorenc (1984), Krishnamurti et al. (1985), M. Murakami (1984) and T. Murakami et al. (1983) have alsoexamined the active-break cycle of the monsoon andtheir relationship to various aspects of the low-frequency oscillations. Krishnamurti et al. (1985) noteda superposition of the westward propagating ridge lineson the 30-50 day time scale. Power spectral analysisand wavenumber frequency spectra around latitutdecircles show that th6se are among the dominant lowfrequency modes of the tropical sea-level pressure field.The relevant scales for 30-50 days are the long waves,i.e., zonal wavenumbers 1, 2 and 3. Those relevant inthe 10-20 time scale are zonal wavenumbers 4, 5 and6. Figure 2 la, b shows x-t diagrams of these respectivepressure'waves covering the break period. The datasets include the entire 365 days of the FGGE year. Ofinterest is the passage of ridges over central Indian longitudes during the period of the Break, i.e., 10-25 July1979. A rise of sea level pressure by ~2 mb occursduring the superposition of these two families of lowfrequency systems. This feature is further emphasizedin Fig. 21c.9. Heat lows The southwesterly moist currents of the African andwest Asian monsoon lie to the south of a major beltof desert heat lows. The role of the heat lows with respect to the activity of the monsoon is not well understood at the present time. Most general circulationmodels do describe dry conditions over the deserts andSEPTEMBER 1985T. N. KRISHNAMURTIEQ-35N 75--105-1613ZONALP3'246:q0-5~IEDDIES.37ZONAL27.37EDDIES5'275:t10-6 I 10.99 I1.12xlO-48.664 x 10-5 7.235 a 10-51.986 ~ I0-5K z 2948~ K'36.7.___4 K /'4.9 21 :~ I0 -5 D' 9.953 x 10_6_~]IIK,.O7 I D'I PRE- DEPRESSION DEPRESSION I JUL- 4JUL 5JUL - 7JULFIG. 19. Energetics based on pre-onset and post-onset data sets during two different periods, July 1979. The units of energy quantities are m2 s-2 while the exchanges are in m2 s-3. This is based on Surgi (1984).the neighboring moist monsoon simulations are indeedquite realistic (e.g., Hahn and Manabe, 1975). Whathas been lacking is a systematic effort to define andcarry out sensitivity experiments on the relationship hO[ <Kz 'KE> .X IO-4 0.; %?~. , , , ~,~'5~?~. , ,,.,,2s-, / \,, JULY 1979 ~ <PE'KE X l0 -41,0rn2 -0.sL JULY 1979 FIG. 20. Results of energy exchanges from (a) zonal kinetic energyto eddy kinetic energy and (b) eddy available potential to eddy kineticenergy. These results are based on a 10-day global prediction of thelife cycle of monsoon depression. The results shown here are integralsfrom 100 mb to the earth's surface over a MONEX domain.of the heat low to the monsoon activity. Chamey (1975)pointed out a paradox: tropical and subtropical desertswere in fact heat sinks. That was noted from satellitemeasured estimates of the Earth's radiation budget.Over most of the tropics the incoming solar radiationexceeds the net outgoing radiation, while over desertsthe converse is the case. Since the thermal stratificationappears to remain nearly invariant from one day tothe next, its maintenance requires that energy be imported laterally to offset the energy loss at the top ofthe atmosphere. To address some of these questions from an observational perspective, a field experiment was conductedwithin MONEX in 1979 (Ackerman and Cox, 1982;Blake et al., 1983). The experiment provided directmeasurements of the up and down irradiances for thesolar and the longwave component via aircraft probesover the heat low of Saudi Arabia. The aircraft alsoprovided atmospheric soundings of temperature, pressure and humidity over this region for the entire troposphere below the 250 mb level. An analysis of thesedata sets has been carded out in the aforementionedstudies. They cover the day and night periods for selected days during May 1979. In Fig. 22a-c we showsome of the salient MONEX observations during thisperiod. The day and night vertical profiles of divergenceand vertical velocity are shown in Fig. 22a, b. Descending motions extend all the way to the surface at1614 MONTHLY WEATHER REVIEWa b MAY 29 MAY 29VOLUME 1 13JUN 12dUN 12JUN 27JUN 27 >.. ,~JUL 12 v-3 JUL 12JUL 27 JUL 27AUG I I AUG I AUG AUG 27 0 45 135 180 225 270 :515 360 45 60 75 90 105 120 135 LONGITUDE LONGITUDE FIG. 21. (a) x-t diagrams of planetary scale eastward propagating sea-level pressure perturbation (mb) along the equator on the 30-50 daytime scale. (b) x-t diagram of synoptic scale westward propagating sea level pressure perturbations (mb) along 10 to 20-N on the 10-20 daytime scale. (c) x-t diagram of the 30-50 day wave and the 10-20 day wave over the monsoon region.night while the lowest 100 mb shows upward motionwithin the heat low during the day time. The mixedlayer depth is determined from a composite data setand shows that the potential temperature and specifichumidity are nearly well mixed up to the 650 mb level(Fig. 22c). Since superadiabatic lapse rates only exist'in the lowest 100 meters over the desert it is apparentthat this mixed layer cannot be described by simpledry convective adjustment methods. In most large-scale numerical weather predictionmodels, dry convective adjustment replaces an unstablepotential temperature profile by a constant 0 (in thevertical) such that the integral of potential temperatureremains invarient during the adjustment. That tendsto describe a rather shallow mixed layer. It is apparentthat the vertical eddy flux of heat deserves a more sophisticated parameterization over the desert areas. Itis also noted that the observed depth of 350 mb duringMONEX is not a fixed entity. Examination of numerous soundings over Riyadh shows that the depthof mixed layer does vary considerably, and it appearsto have some relationship to the overall depth of thedust and aerosol layers. Further observational andmodeling studies are needed in these areas. The verticaldistribution of: up and down radiances is extremely interesting in this region. The downward directed shortwave irradiance is presented in Fig. 23a. The solar radiation reaching the ground is of the order of 800 Wm-2, which is much smaller than the solar constant(1375 W m-2). Thus, it is apparent that a considerablewarming of the troposphere occurs from the direct solarradiation. Usually the clear sky warming rates fromdirect solar radiation are of the order of 0.5-1 -C day-~,and the much larger warming rates over the desert region suggests a possible role of aerosols and dust (Ackerman and Cox, 1982). The upward directed short waveradiation (Fig. 23) is of the order of 350 W m-2 overmost of the troposphere; the earth's surface albedo isof the order of 375/850 ~ 45% and the earth's atmospheric albedo (at the top) is of the order of 300/1100m 27%. The downward directed longwave radiation,shown in Fig. 23d, decreases from around 400 W m-2SEPTEMISER 1985 T.N. KRISHNAMURTI 1615C OEMAY 29JUN 12JUN 27JUL 12JUL 27AUG I IAUG 2745ELONGITUDE 30-50 DAY WAVE 45 135 180 :::)25270 315W360 '~'~~b',-, I I I I I60 75 90 105 120E 135LONGITUDE 10-20 DAY WAVE I~G. 21. (Continued)at the earth's surface to about 100 W m-2 at around250 mb. The upward directed long wave radiation (Fig.23c) decreases from 500 to about 350 W m-2. A netdivergence of long wave irradiance of around 400 Wm-2 is experienced by the atmosphere. The long- andshortwave heating and cooling rate (Fig. 24), whichcorresponds to a net cooling of around 4-C day-1, isslightly smaller than the heating rate by solar radiationover the deep troposphere between 900 and 400 mb.The nighttime profiles of radiation are not presentedhere (see Blake et al., 1983). It is important to note,however, that over a 24-hour average a net longwavecooling dominates over the daytime net warming ofthe troposphere by solar radiation, thus resulting in theaforementioned net radiative losses as are measured atsatellite altitude. The stratification is thus largelymaintained by adiabatic warming associated with thedownward motions. The compensating upward massflux occurs over distant convective areas (see Blake etal., 1983). This can best be portrayed from an analysisof the divergent wind, which shows the active monsoonover Northern Malaysia, Southern Indochina and thewestern Pacific Ocean. The high-level divergentstreamlines emanate from this region, and exhibit aneventual confluence into the region of these desert heatlows as well as several other regions. The upper levelinflow of mass also brings in a steady supply of energy(Blake et al., 1983). Thus, the maintenance of the thermal stratification against radiative heat loss is attributedto the descent of air which in turn is supplied by lateralflux of enthalpy from the remote monsoon areas. It iswithin the well-mixed layer that the role of dry convection appears to be significant in the maintenanceof the stratification. FGGE IIIb analysis provides a useful synoptic structure of the heat low that can be found from the westcoast of Africa to the deserts of Pakistan. Observationalcomposite studies on the structure of the heat lowsover the Sahara, Arabia and Pakistan are worthy offurther investigations. Another dry region of equal, relevance is the westernpart of the Tibetan Plateau, which was addressed inSection 3.10. Low-frequency motions of the monsoon The monsoon season is only about 3-3 I/2 monthslong. Thus, within a season the relevant low-frequencymotions are those whose time scales are less than theseasonal time scale. There are, of course, several otherlower frequencies such as the semiannual, the annual,the quasi-biennial and the Southern Oscillation thatare evidently important for the monsoon problem. Inthis review we shall not address the latter but confineour attention to the major findings from the FGGE/MONEX observations and related theoretical findings.The most important low-frequency motions seem tobe those in the 30-50 day and the 10-20 day timescales. From the perspective of the monsoons, the 3050 day time scale was first emphasized by Yasunari(1981). He noted zonally oriented cloud lines thatpropagated meridionally from the equatorial latitudesto the Himalayas. Krishnamurti and Subrahmanyan(1982) identified distinct motion systems on this timescale. At the 850 mb surface, a train of zonally orientedtroughs and ridges were shown to exhibit a near-steadymeridional propagation. The troughs were associatedwith rising motions and clouds while the ridges wereessentially cloud free. The phenomena of onset, activeand break monsoon appeared to be related to the passage of these low-frequency systems. Figure 25 showsa sequence of the low-frequency wind analysis at 850mbs from Krishnamurti and Subrahmanyan (1982).The meridionally moving troughs and ridges tend toform near the equator, amplify as they arrive at ~ 10-Nand finally appear to dissipate as they arrive near theHimalayas. These low-frequency systems are thermallydirect in the sense that warmer air ascends in thetroughs and cold air descends in the ridges. Calculations ODIVERGENCE ( X 10-5 SEC-1 ) -2.0200.300.400.~ $00.r- 600.D~'~ 700.800.-I.O 0.0900.1.0 2.0 3.0 09 MAY 1979 -- OMEGA ---DIVERGENCE -2.0200.300.400.SO0.600.700. bDIVERGENCE ( X 10-5 -1 SEC )800.900.-1.0 0.0 1.0 2.0 3.012 MAY 197g--OMEGA---DIVERGENCE 30.0OMEGA ( X 10-4 MB SEC-1 )1000. I000. I I I I I I I I -20.0 -10.0 0.0 10.0 20.0 30.0 -20.0 -10.0 0.0 10.0 20.0 OMEGA ( X 10-4 MB SEC-1 ) c d --- MAY 10 VERTICAL PROFILE OF AVERAGE 300. - .... MAY 12 250 SPECIFIC HUMIDITY ---- U.S. STO ATM (15-N ANNUAL 400. 350 500, 450ua 60Q. ud 550~ ~ 650 900. / 850 290. 300. 310. 320. 330. 340. 350. / kg POTENTIAL TEMPERATURE I~G. 22. Vertical velocity ~o and horizontal divergence D over the Saudi ^mbian heat low during May 1979: (a) daytime vertical profiles, (b)nighttime vertical profiles, (c) vertical profiles of potential temperature and (d) specific humidity. They illustrate the depth of the mixed layer(Blake et al., 1983).SEPTEMBER 1985 T.N. KRISHNAMURTI 1617100.200.300.qO0.500.600.700,800. 900.lO00. "' 700.09 MAY 1979SHORTWAVE DOWN100.200.I I I I I900. 1000. 1100. 1200. 1300.WATTS/ma ~300.qO0.500.600.700.800.900. ~ 1000. ~800. 0. 100,09MAY 1979SHORTWAVE UPI I I I200. 300. qO0. 500.WATTS/m~ ,100.200.300.400.500.600.700.800. 900.1000.09 MAY 1979LONGWAVE DOWN100.200.I I I ItO0. 200. 300. 400. WATTS/ma300.400.500.600.700.800. 900.1000. I IO.500. 600. O. 100.09 MAY 1979LONGWAVE UPI I I I200. 300. 400. 500.WA'l-rS/mz ~dDo. 23. Vertical profiles of upward and downward irradiances during the daytime over the Saudi Arabian heat low: (a) shortwave up, (b) shortwave downward, (c) longwave downward, and (d) longwave upward. (Blake et al., 1983).1618 MONTHLY WEATHER REVIEW VOLUME 113200.300.'~ 5oo.n,' 600.I.d 700.900.1000. -10.a ' ' '-- TOTAL ! t--- LONGWAVE:--- SHORTWAVE i i ,# ~' ~ ~, t - ~1~ - 1 *, 1 * 1- I ~ I ~ ! i I I -8. -6. -~,. -2.HEATING RATE (~C DAY -1 ) F~o. 24. Vertical profiles of radiative heating rates during daytimeover Saudi Arabia. The long- and shortwave and the total heatingare shown here. Units -C day-'. of the energetics of this system were carried out by Krishnamurti et al. (1985) and Murakami et aL (1983). From a global perspective, the aforementioned regional description is not altogether complete. Krishnamurti and Gadgil (1985), Krishnamurti et al. (1985)and Lorenc (1984) have examined various aspects ofthe 30-50 day oscillations over the globe during theFGGE year. Krishnamurti and Gadgil examined theglobal FGGE data sets for the entire FGGE year. Fivevariables (U, V, T, Q, ps) at six levels (1000, 850, 700,500, 300, 200 rob) were examined on this time scale.They noted that a length of 365 days of data recordwas better suited for studies of 30-50 day oscillationsthan a short seasonal data set as was used previously.The surprising finding was that the amplitude of theoscillations on the 30-50 day time scale were not onlydominant over the region of the summer monsoon,but also at higher latitudes near 50-N and 50-S nearthe 200 mb level. From the time-filtered data sets of- the zonal wind on this time scale, it is possible to mapthe isotachs of the maximum wind in a given season(Krishnamurti and Gadgil, 1985). Figure 26a-d showsthese distributions for the winter, spring, summer andfall months, respectively. During northern winter, a distinct maximum velocity ~5.5 m s-~ is found near 160-W that turns out tobe another interesting source region for the low-frequency motion systems. The first appearance of amonsoonal low-frequency wind maximum occurs inthe spring season over Malaysia and Indochina. As wasstated earlier, this region encounters the onset of monsoon in early May. Thus, it appears that the oscillationover the monsoon region, which extends northward bythe summer season (see Fig. 26c), may be related tomonsoon activity. During these summer months thelarge amplitude of the zonal wind oscillation coversthe region between 10 and 20-N and from the ArabianSea to the Pacific Ocean. The amplitude lies between3 and 5 m s-I. In the fall months the amplitude of thelow-frequency activity over the monsoon region diminishes considerably (see Fig. 26d). Although it issomewhat apparent that low-frequency motion systemsmodulate the monsoon activity, the converse (i.e., howmonsoon activity might play a role in exciting the lowfrequency motions) is not altogether clear. Another major observational finding on the 30-50day time scale reich, ant to the monsoon problem arethe divergent motions on the planetary scale. Usingempirical orthogonal functions to represent the temporal behavior of the divergent wind, Lorenc (1984),identified a planetary scale (mostly zonal wavenumber1) wave that propagates from west to east in roughly30-50 days. Furthermore, he noted that this wave canbe identified nearly all year around. Krishnamurti etal. (1985) examined this wave in some detail by mapping it for the entire year. Figures 27-29 show a sequence of the low-frequency velocity potential ~at 200mb on this time scale around the period of the monsoononset in 1979. The divergence center arrived oversouthwestern India around 14 June close to the periodof commencement of rain over that region. Theoretical aspects of this problem have been addressed by Webster (1983), Dunkerton (1983) and several others. 11. Concluding remarks The data sets for the Summer Monsoon Experimentwere unique. They provided some of the best descriptions ever obtained of the planetary and regional scalemonsoon. Special observing systems over the IndianOcean provided high resolution cloud winds, soundingsfrom research ships, winds from constant level bat- loons, winds from commercial aircraft, enhanced worldweather watch and dropwindsonde measurements ofatmospheric soundings deployed from research aircraft.In addition, high resolution estimates of the earth'sradiation budget were obtained from geostationary(Indian Ocean GOES) and polar orbiting satellites(Nimbus-F, Tiros-N, NOAA-A), retrieval soundings oftemperature and precipitable water, and direct measurements of long- and shortwave irradiances from research aircraft. A special collection of marine data overthe Indian Ocean also provided some of the uniqueobservations that were complemented by FGGE andits special observing systems over the rest of the globe. a ON ('! ~~']~:50 $~ 30E JUNE 21,1979 - ~(A~t,~,,"~"~"~ ~ "- 2 I - ~,,,,~ ~ ! ~ . !60E 90E 120E 150E,40 N -~SON ~~ 3O[ JUNE 28,1979 ..-', ~., ~,-:,~.~, -~ .'~'- ~'.. '~'~60E 90E 120E 150E40NON50S 30EJULY 2,1979 120E 150Eb50E 60E 90E JULY 4 , 1979 JULY I0 , 1979 .40 N ,.~- .~ ~ 50N / ~ ON~ .~L[,_~(', ~'~'d ~os 120E 'ISOE 30E 60E 90E I~E I~E30S 50E 60E 90E JULY 8 , 1979 .40 N - 50 N? t. ~' ON120E JULY 14,1979 50 S150E 50E 60E 90E 120E 150EFIG. 25. A sequence of low-frequency streamline and isotach (m s-~) charts on the time scale of 30-50 days over the MONEX domain. (Krishnamurti and Subrahmanyam, 1983).1620 a~0~MONTHLY WEATHERREVIEWVOLUME 1137ONSON30NIONlOS30S __SOS 70S 9OS OE5OE 6OE . 9OE 12OE 150E 18OE 15OW 12OW 9OW 6OW SOWOEb90N70N50N30NIONlOS3OS5OS7OS9OS -~ ).,5-OE 3OE 6OE 9OE 12OE 15OE 180E 15OW 12OW 90W 60WFIG. 26. Seasonal maximum wind speed (m s-I) charts of the low-frequency motions on the time scale of 30-50 days over theglobe: (a) winter, (b) spring, (c) summer and (d) fall months during the FGG- year. (K.rishnamurti and Oadgil, 1984).OESEPTEMBER 1985 - OONKRISHNAMURTI162170N50N$ON ION lOS 50S 50S 70S 90SOE 30E 60E 90E 120E 150E 180E I50W I20W 90W 60W 30W OE d90N50 N ~5.5SON 3.O ION SOS 50S 70S 90SOE 30E 60E 90E 120E 150E 180E 150W 120W 90W 60W 50W FIG. 26. (Continued)OE1622 MONTHLY WEATHER REVIEW VOLUME 113 VELOCITY POTENTIAL (200MB)CONTOURS 30-50FILTER 06/01/7990N60N3ON EQ30 S60S t~64 30E90SoE 60E 90E 120E 150E 180E 150W 120W 90W 60W 30WIZZ INT=.64E+06 ~!~i~1!ii~11i ~ ..... I OE VELOCITY POTENTIAL (200MB)CONTOURS 30-50 FILTER 06/06/79 IP_Z INT= .61E+06~o. -~_z~~-'--~'c.o~3ON ' ~, '"'-~"~~~~~~'~~F '"'"~'~"~'~'~~~~90 N60 N30N EQ30 S60S90 .SoEVELOCITY POTENTIAL (200MB)CONTOURS 30-50FILTER 06/11/79 12Z INT=.51E+06~ ~ O-- ~-51 ,~ ~ ~_-~c~=.~-~-~~->o/...~~ ~?.;,'-~ ~o ~-'-;,,,~ ~ . ~e~~'~'?~'~ ....... ~!~$ i!i!::~:-' ':[ii :' ~t$~ ~ '~~- ~__.~ ~, o > ~~~~~~~~ ~ d/~~~ ~ ~,,. -61 0~5~ I I I I I --- I I I 30E 60E 90E 120E 150E 180E 150W 120W 90W 60W 30W OEFIG. 27. A sequence of low frequency charts of the 200 mb velocity potential on the 30-50 day time scale. These are based on the entire FGGE year of analysis.QbSEPTEMBER 1985 T.N. KRISHNAMURTI 1623VELOCITY POTENTIAL (2OOMB) CONTOURS 50-50FILTER 06/16/7'9 12Z INT=.62E+06OrE 30E 60E 90E 120E IOE 180E 150W 120W 90W 60W 30W OE VELOCITY POTENTIAL (2OOMB) CONTOURS 30-50FILTER 06/21/79 12Z INT=,66E+0690N ~ o'"""- 66 ~o , ~ H 132-60 N;3ON ~. ~-'-~ "'~',,.'~.'~-~-"~-"~--"'~'~'"~~~'~~~~~~~~"~\ /? N I L '-" k,---,'"~ '", ""30590S 66---'-------- O~ 30E 60E 90E 120E 150E 180E 150W 120W 90W 60W 30W OE90NVELOClTY. POTENTIAL (2OOMB) CONTOURS_ 30-50FILTER 06~26~79 IZZ INT=.69E+0660N3ON90S OE 30E 60E 90E 120E 150E 180E 150W 120W 90W 60W ~50W OEFIG. 28. A sequence of low-frequency charts of the 200 mb velocity potential on the 30-50 day time scale. These are based on the entire FGGE year of analysis.C1624 MONTHLY WEATHER REVIEW VOLUME 113 VELOCITY POTENTIAL (2OOMB) CONTOURS 30-50FILTER 07/01/79 IZZ INT= .61E+0690N . ' ~.~ 6~60N '~ o '90S~ , , , , ,../' OE 5OE 60E 90E 12'OE 150E 180E 150W 120W 90W 60W 30W OE VELOCITY POTENTIAL (2OOMB) CONTOURS :50-50FILTER 07/06/7912Z INT=.64E+0690S OE 50E 60E 90E 12'OE 150E 180E 150W 120W 90W 60W 30W OEb VELOCITY POTENTIAL (2OOMB) CONTOURS 30-50FILTER 07/ll/79 12_Z INT=.63E+0630 N EQ30SOE 30E 60E 90E 120E IOE 180E 150W IZOW 90W 60W 30W OEF~G. 29. A sequence of low frequency charts of the 200 mb velocity potential on the 30-50 day time scale. These are based on the entire FGGE year of analysis.SEPTEMBER I985 T.N. KRISHNAMURTI 1625 Some of the major achievements of MONEX are: - Use of an unprecedented volume of observationsfrom ships, aircraft, satellites, constant level balloonsand augmented WWW. - Most detailed four-dimensional descriptions of theevolution of the 1979 summer monsoon. - The planetary scale monsoon, the associated circulations, thermal structure and heating distributionsare better understood because of the unprecedenteddata sets. - The role of the Tibetan Plateau in its overall effecton the evaluation of the monsoon has been approachedby many scientists. The most promising accomplishments are in the areas of the definitions of the heatsources. - The onset scenario focussed on (i) the role of horizontal shear in the onset vortex dynamics and (ii) therole of differential heating in the establishment of broadscale zonal flows and divergent circulations. - The geostationary satellite located over the IndianOcean during MONEX provided imagery and cloudwinds that enabled the most complete descriptions ofthe monsoon over the oceans. This was simply notpossible prior to 1979. Use of these data sets, haveplayed a key role in the determination of air-sea interactions. - Major success is evident in the areas of numericalweather prediction, especially with global models. Theonset and active phases (depressions) have been reasonably predicted on the medium range time frame. - A number of major research groups participatedin FGGE intercomparison experiments on mediumrange numerical weather prediction. The experimentshave been most revealing on the sensitivity of the onsetof monsoon to different parameterizations of cumulusconvection. Further analysis of the monsoon during 1979 in mostof the aforementioned areas is necessary. A better synthesis of observational, theoretical and modeling studiesand finding is undoubtedly expected from further efforts. The future research trends in this area are alreadyevident from the current emphasis. Research on interannual variability of monsoon on several space-timescales is being pursued by several groups. Relationshipsand sensitivity of monsoon rain, floods and rain to avariety of parameters such as sea surface temperatures,antecedent climate anomalies, Himalayan snow coverand overall heat sources and sinks are important areasof future research. Acknowledgments. The research reported here wasjointly supported by the National Science FoundationGrant NSF ATM-83-04809 and by the NationalOceanic and Atmospheric Administration GrantNOAA NA 82AA-D-0004. The numerical computations were carried out at the CYBER 760 Computerof Florida State University and on the Cray at the National Center for Atmospheric Research, which issponsored by the National Science Foundation.REFERENCESAckerman, S. A., and S. K. Cox, 1982: The Saudi Arabian heat low: Aerosol distributions and thermodynamic structure. J. Geophys. Res., 87, 8991-9002.Arakawa, A., and S. Moorthi, 1982: Baroclinic (and barotropic) in stability with cumulus heating. Proc. WMO program on Research in Tropical Meteorology, Tsukuba, WMO, V-43-V-62. [Avail able from Word Meteorological Organization, Case Postale 5, CH- 1211, Geneva, Switzerland.]Blake, D. W., T. N. Kfishnamurti, S. V. Low-Nam and J. S. Fein, 1983: Heat low over the Saudi Arabian Desert during May 1979 (Summer MONEX). Mon. Wea. Rev., 111, 1759-1775.Cadet, D., H. Ovafiez and G. Sommeria, 1981: The BALSAMINE Experiment during the Summer MONEX. Bull. Amer. Meteor. Soc., 62, 381-393.Chang, C. B., 1979: On the influence of solar radiation and diurnal variation of surface temperature on African disturbance. Rep. No. 79-3, Dept. of Meteorology, Florida State University, Tal lahassee, FL, 1-157. [Available on request from Florida State University.]Charney, J. G., 1975: Dynamics of deserts and drought in the Sahel. Quart. J. Roy. Meteor. Soc., 101, 193-202.Daley, R. C., C. Girard, J. Henderson and L Simmonds, 1976: Short term forecasting with a multi-level spectral primitive equation model. Part 1: Model formulation. Atmosphere, 14, 98-116.Douglas, M., 1984: Synoptic structure of monsoon disturbances. Workshop on Monsoon Rainfall, 17-22. [Available from the International Center for Theoretical Physics, Trieste, Italy.]Diiing, W., and A. Leetmaa, t980: Arabian sea cooling; A preliminary heat budget. J. Phys. Oceanogr., 10, 307-312.Dunkerton, T. J., 1983: A non-symmetric equatorial inertial insta bility. J. Atmos. Sci., 40, 807-813.Fein, J. S., and J. P. Kuettner, 1980: Report of the summer MONEX field phase. Bull. Amer. Meteor. Soc., 61, 461--474.Findlater, J., 1969: A major low-level air current near the IndianOcean during the northern summer. Quart. J. Roy. Meteor. Soc.,95, 362-380.Gao, Y. X., M. C. Tang, S. W. Luo, Z. B. Shen and C. Li, 1981: Some aspects of recent research on the Qinghai-Xizang Plateau meteorology. Bull. Amer. Meteor. Soc., 62, 31-35.Greco, S., 1984: Water vapor transport over the Indian Ocean during the 1979 summer. Proc. 15th Conf on Hurricanes and Tropical Meteorology, Miami, Amer. Meteor. Soc., 394-397 pp.Hahn, D. G., and S. Manabe, 1975: The role of mountains in the south Asian monsoon circulation. J. Atmos. Sci., 32, 1515-1541.Kanamitsu, M., and T. N. Krishnamurti, 1978: Northern summer tropical circulations during drought and normal rainfall months. Mon. Wea. Rev., 106, 331-347.Kfishnamurti, T. N., 197 I: Tropical cast-west circulations during northern summer. J. Atmos. Sci., 28, 1342-1347.--, 1978: Large scale features of the tropical atmosphere. Meteo rology over the Tropical Oceans. Royal Meteorological Society, 31-56.--, 1981: Cooling of the Arabian Sea and the onset-vortex during 1979. Recent progress in equatorial oceanography: A report of the final meeting of SCOR WORKING GROUP 47 in Venice, Italy, 1-12. [Available from Nova Univ., Ocean Science Center, Dania, FL 33004.] , and P. Ardanuy, 1980: The 10-20 day westward propagating mode and breaks in the monsoon. Tellus, 32, 15-26. , and Y. Ramanathan, 1982: Sensitivity of monsoon onset of differential heating. J. Atmos. Sci., 39, 1290-1306. , and D. Subrahmanyam, 1982: The 30 to 50 day mode at 850 mb during MONEX. J. Atmos. Sci., 39, 2088-2095.1626 MONTHLY WEATHER REVIEW VOLUME 113 -, and S. Gadgil, 1985: On the structure of the 30 to 50 day mode over the globe during FGGE. Tellus, 36 (in press). , P. Ardanuy, Y. Ramanathan and R. Pasch, 1979: Quick look summer MONEX atlas. Part II. The onset phase. Florida State University Rep. No. 79-5, 312 pp. [Available from Dept. of Meteorology, Florida State University, Tallahassee, FL. 32306.] -- and .......... ,1981a: The onset-vortex of the summer monsoon. Mon. Wea. Rev., 109, 344-363. , R. J. Rasch, H. L. Pan, S. H. Chu and K. Ingles, 1983a: Details of low latitude medium range numerical weather prediction using a global spectral model. I: Formation of a monsoon depression. J. Meteor. Soc. Japan, 61, 188-207. , V. Wong, H. L. Pan, R. Pasch, J. Molinari and P. Ardanuy, 1983b: A three-dimensional planetary boundary layer model for the Somali jet. J. Atmos. Sci., 36, 1477-1484. , S. Low-Nam and R. Pasch, 1983c: Cumulus parameterization and rainfall rates. Mon. Wea. Rev., 111, 816-828. , K. Ingles, S. Cocke, T. Kitade and R. Pasch, 1984a: Details of low latitude medium range numerical weather prediction using a global spectral model. Part II: Effects of orography and physical initialization. J. Meteor. Soc. Japan, 62, 613-649.~, N. Surgi and J. Manobianco, 1984b: Recent results on divergent circulations over the global tropics. J. Meteor. Soc. Rep. China, (in press). ...... , P. K. Jayakumar, J. Sheng, N. Surgi and A. Kumar, 1985: Divergent circulations on the 30 to 50 day time scale. J. Atmos. Sci., 42, 364-375.Kuo, H. L., 1974: Further studies of the parameterization of theinfluence of cumulus convection of large scale flows. J. Atmos.Sci., 31, 1232-1240.Lindzen, R. S., B. Farrell and A. J. Rosenthal, 1983: Absolute baro tropic instability and monsoon depressions. J. Atmos. Sci., 40, 1178-1184.Lorenc, A., 1984: The evolution ofplanetary scale 200 mb divergence during the FGGE year. Quart. J. Roy. Meteor. Soc., 110, 427 442.'Lorenz, E. N., 1960: Energy and numerical weather prediction. Tellus, 12, 364-373.Luo, H., and M. Yanai, 1984: The large-scale circulation and heat sources over the Tibetan Plateau and surrounding areas during the early summer of 1979. Part II: Heat and moisture budgets. Mon. Wea. Rev., 112, 966-989.Mak, M., and C.-Y. J. Kao, 1982: An instability study of the onset vortex of the south-west monsoon, 1979. Tellus, 34, 358-368.Murakami, M., 1984: Analysis of deep convective activity over thewestern Pacific and southeast Asia. Part II. [Available from author, Dept. of Meteorology, University of Hawaii, Honolulu,HI 96844.]Murakami, T., R. V. Godbole and R. R. Kelkar, 1970: Numerical simulation of the monsoon along 80-E. Proc. of the Conference on the Summer Monsoon of Southeast Asia, C. S. Ramage~ Ed, Navy Weather Research Facility, Norfolk, VA, 39-51. , T. Nakazawa and Jin Hai He, 1983:40-50 day oscillations during the 1979 Northern Hemisphere summer. Tech. Rep. No. UHMET 83-02, 82 pp. [Available from the Dept. of Meteorology, University of Hawaii, Honolulu, HI 96844.]Parch, R. J., 1983: On the onset of the planetary scale monsoon. FSU Rep. No. 83-91, 187 pp. [Available from the Dept. of Me teorology, Florida State University, Tallahassee, FL 32306.]Ramamurthy, K., 1969: Some aspects of the break in the Indian south-west monsoon during July and August. Forecasting Manual, No. IV~18.3, 1-57. [Available from the India Meteor. Dept., Ramdurg House, Poona, India.]Raman, C. R. V., Y. P. Rao, S. K. Subramanian and J. A. Maliekal, 1981: Tropospheric wind shear oscillations as a characteristic of the southwest monsoon atmosphere. Mon. Wea. Re,., 109, 910 915.Sadler, I. C., and B. E. Harris, 1970: The mean tropospheric circu lation and cloudiness over southeast Asia and neighboring areas. Scientific Rep. No. 1, 1-38. [Available from Hawaii Institute ofGeophysics, University of Hawaii, Honolulu, HI 96844.]Saha, Kshudiram, and C.-P. Chang, 1983: The baroclinic processesof monsoon depressions. Mon. Wea. Rev., 111, 1506-1514.Sander, F., 1984: Quasi-geostrophic diagnosis of the monsoondepression of 5-8 July 1979. J. Atmos. Sci., 41, 538-552.Shukla, J., 1978: CISK-barotropic-baroclinic instability and the growth of monsoon depressions. J. Atmos, Sci., 35, 495-508.Sikka, D. R., and S. Gadgil, 1980: On the maximum cloud zone and the ITCZ over Indian longitudes during the Southwest Monsoon. Mon. Wea. Rev., 108, 1840-1853.Sommeria, G., G. Reverdin, H. Ovarlez and D. Cadet, 1982: Scientific. results of BASAMINE, a boundary layer experiment during summer MONEX. Proc. Int. Conf. on Scientific Results of the Monsoon Experiment. Denpasar Bali, 355-358. [Available from the World Meteorological Organization, Case Postale 5, CH 1211, Geneva, Switzerland.]Stout, J. E., and J. A. Young, 1983: Low-level monsoon dynamicsderived from satellite winds. Mon. Wea. Rev., 111, 774-798.Surgi, N., 1984: The structure and dynamics of a monsoon depression. Proc. 15th Con. on Hurricanes and Tropical Meteorology, Miami, Amer. Meteor. Soc., 354-358.Temperton, C., T. N. Kdshnamurti, T. Pasch and T. Kitade, 1983: WGNE forecast comparison experiments. World Climate Re search Program Rep. No. 6, 73-104. [Available from World Meteorological Organization, Case Postale 5, CH-1211, Geneva, Switzerland.]Wallace, J. M., and D. S. Gutzler, 198 I: Teleconnections in the geo potential height field during the Northern Hemisphere winter. Mon. Wea. Rev., 109, 784-812.--., S. Tibaldi and A. J. Simmons, 1983: Reduction of systematic forecast errors in the ECMWF model through the introduction of an envelope orography. Quart. J. Roy. Meteor. Soc., 109, 683-718.Warner, C., 1984: Core structure of a Bay of Bengal monsoon depression. Mort. Wea. Rev., 112, 137-152.Webster, P. J., 1983: Mechanisms of monsoon low-frequency vari ability: Surface hydrological effects. J. Atmos. Sci., 40, 2110 2124.Yanai, M., and T. Nitta, 1968: Finite difference approximation for the barotropic instability problem. J. Meteor. Soc. Japan, 46, 389-403.--, S. Esbensen and 1. H. Chu, 1973: Determination.ofbulk prop erties of tropical cloud clusters from large scale heat and moisture budgets. J. Atmos. Sci., 30, 611-627.Yasunari, T., 1981: Structure of an Indian summer monsoon system with around 40-day. J. Meteor. Soc. Japan, 59, 336-354.Yeh, T. C., and D. Zhu, 1982: Some aspects of the thermal influences of the Qinghai-Tibetan Plateau on the atmospheric circulation. Arch. Meteor. Geophys. Bioklim., 31, 205-220.--, Y. X. Gao and M. C. Tang, 1981: The average summer vertical circulation to the south of 45-N of Northern Hemisphere and its relation .to the distribution of heat sources and sinks in the atmosphere. Quart. J. Meteor. Soc. China, 39, 28-35.

Share Link

Copy this link, or click below to email it to a friend

Email this content

or copy the link directly:

https://journals.ametsoc.org/view/journals/mwre/113/9/1520-0493_1985_113_1590_smer_2_0_co_2.xml

Link copied successfully

Monthly Weather Review

Abstract

	All Time	Past Year	Past 30 Days
Abstract Views	0	0	0
Full Text Views	22902	21587	12270
PDF Downloads	919	218	28

The Impacts of Interannual Climate Variability on the Declining Trend in Terrestrial Water Storage over the Tigris–Euphrates River Basin

Authors:

Li-Ling Chang

and

Guo-Yue Niu

Evaluation of Snowfall Retrieval Performance of GPM Constellation Radiometers Relative to Spaceborne Radars

Authors:

Yalei You

George Huffman

Veljko Petkovic

Lisa Milani

John X. Yang

Ardeshir Ebtehaj

Sajad Vahedizade

, and

Guojun Gu

A 440-Year Reconstruction of Heavy Precipitation in California from Blue Oak Tree Rings

Authors:

Ian M. Howard

David W. Stahle

Michael D. Dettinger

Cody Poulsen

F. Martin Ralph

Max C. A. Torbenson

, and

Alexander Gershunov

Quantifying the Role of Internal Climate Variability and Its Translation from Climate Variables to Hydropower Production at Basin Scale in India

Authors:

Divya Upadhyay

Sudhanshu Dixit

, and

Udit Bhatia

Extreme Convective Rainfall and Flooding from Winter Season Extratropical Cyclones in the Mid-Atlantic Region of the United States

Authors:

Yibing Su

James A. Smith

, and

Gabriele Villarini

Summer Monsoon Experiment—A Review

Abstract

Abstract

Share Link

AMS Publications

Get Involved with AMS

Affiliate Sites

Follow Us

Contact Us

Sections

Cited By

Metrics

Related Content

Summer Monsoon Experiment—A Review

Abstract

Abstract

Share Link

Headings

Cited By

Metrics

Related Content

AMS Publications

Get Involved with AMS

Affiliate Sites

Follow Us

Contact Us