Recent Biennial Variability of Meteorological ... - Semantic Scholar
GEOPHYSICAL RESEARCH LETTERS, VOL. 27, NO. 15, PAGES 2185-2188, AUGUST 1, 2000
Recentbiennial variability of meteorologicalfeatures in the Eastern Highland Himalayas Laura Bertolani
and Massimo
Bollasina
ClimateResearchDivision,CentroEpsonMeteo,Milan, Italy.
Gianni
Tartari
IRSA-CNR, Milan, Italy.
Abstract. With meteorologicaldata from high altitude surface stations and gridded dataset from NCEP/NCAR Reanalysis,a biennial oscillationover the EasternHighland Himalayasfrom 1980 to 1998 is described.This variability concernsair temperature, precipitation, geopotential andwind speed.Evidenceis given on the connectionsbetweenlocal dataandlarge-scale circulationpatterns. The mostremarkable oscillatingfeaturesare found duringwinter and, in general, the signalsare particularlymarkedon the southernslopeof the HimalayanRange.A possiblemechanism is explainedin terms of a periodicityin surfaceheat and moisturefluxes. Finally, the peculiarityof the region as a climatic change observatoryis underlined.
Centerfor Atmospheric Research(NCAR). The resultspermit to supportandto extendto highaltitudesone of the possible mechanisms of TBO, usuallydescribed by meansof modeling
Chang and Li, 1999]. Some other works focus on the importanceof land-surface processes, especiallysnowcover and soil moistureover Eurasia[e.g., Yasunariet al., 1991; Vernekaret al., 1995;Ose, 1996]. Moreover,the importance
87.1øE, 4302 m), Xigatse(29.3øN 88.9øE,3837 m) and Xainza (31.0øN 88.6øE, 4671 m). The daily data of the
studies.
2. Data
and Methods
The possibilityof a biennialperiodicityin the courseof meteorologicalparametersis first investigatedconsidering stationtime seriesfrom 1980 to 1998. The synopticpatterns of circulationabovethe Asian Continent(15øN-45øN,65øE110øE) are studiedto explain the large-scaleorigin of the phenomenaobservedat the stations;numericalvaluesand mapsare plottedby meansof daily/monthlydata available from CDAS-NCAR Reanalysis on a 2.5ø x 2.5ø grid [Kalnay 1. Introduction et al., 1996]. Precipitation griddeddatahavebeencollected from the monthlydatasetof ClimatePredictionCenter(CPC) The TBO (Tropospheric BiennialOscillation)is a quasi- MergedAnalysisof Precipitation[Xie andArkin, 1996]. biennialperiodicityin the courseof many variables(e.g., Four stationsare considered:the Laboratory-Observatory precipitation, atmospheric pressure, seasurfacetemperature) called "Pyramid",locatedin the Nepal HighlandHimalayas, occurring in theIndianandPacificregions.It is recognized to andthreeWMO stations,placedin thesouth-eastern regionof be the resultof large-scaleinteractionamongsea,land and Tibet. atmosphere,includingtropical-mid-latitude interaction.Its The Pyramidmeteorological stationwassetup in 1990 by originis a matterof interesting debatesand severaltheories the Water Research Institute of the Italian National Research exist[e.g., Websteret al., 1998].The interannual variability Council(IRSA/CNR) and sinceDecember1993 it has run of the Asian summer monsoonand its relationshipwith continuouslyyear round. The Pyramid is locatedin the tropical Sea SurfaceTemperatureAnomalies(SSTA) has Khumbu Valley at 5050 m next to Mt. Everest. Its beenoutlinedby manyauthors[e.g.,Nicholls,1978;Meehl, geographical coordinates are28.0øN86.8øE. 1987, 1997; Shen and Lau, 1995; Yang and Lau, 1998; The Tibetan stations consideredare: Tingri (28.6øN
of the Tibetan Plateau as an elevated heat source/sink is well
described[Yanai et al., 1992; Murakami, 1987] and a biennialsignal in the tendencies of heat fluxes has been recognized[Yanaiand Tomita,1998]. In this paper,a biennialoscillationfound in the Eastern Himalayasis highlightedby meansof daily (from 1994 to 1998) andmonthly(from 1980 to 1998) datarecordedat four stationslocatedat high altitudes(above3800 m). The local climaticfeaturesare confirmedandexplainedby the analysis of large-scalesynopticdata availablefrom the National
Chinese stationsare collectedfrom NOAA global surface summaryof daily observations for the period 1994-1998. Monthly summarydatasetsfrom NCAR (CAC GLOBAL CEAS summary)arealsousedfor theperiod1980-1998.
The followingmeteorological variablesare considered in this analysis:air temperature, precipitation, geopotential and wind speed.
3. Analysis and Results A variabilitywith a periodof 24 monthsappearsin many parameters, suchas winter temperatures and wind speed, winterandsummerprecipitation andgeopotential.
Copyright2000 by the AmericanGeophysical Union.
The mostremarkablefeatureappearsin wintertemperature values: during even years, January mean temperaturesare
Papernumber1999GL011198.
warmerthanduringodd yearsfor all the stationsandfor the
greaternumberof cases. NCEP/NCARReanalysis datapoint
0094-8276/00/1999GL011198505.00
2185
2186
BERTOLANI
ET AL.' TBO IN THE EASTERN
HIGHLAND
HIMALAYAS
PyramidStation o
o..... -7.0
a) -8.0
•
-9.0
• -10.0
-11.0 -12.0 94
95
96
97
98
Years
Figure 3. Air meantemperature (øC) at Pyramidfrom 1994 to 1998.The solidanddashedlinesshowdatafor Januaryand February,respectively. 65E
70E
75E
80E
85E
90E
95E
100E
105E
110E
from the Arabian Sea, reachesthe Range. Anyway, the monthlymeananomalyresultsslightlynegative(-0.5øC).The strongest episodeappearsin thefirst threeweeksof January, whenwintercirculationis fully established. At thebeginning of Februarythe anomalymovessoutheastward; thus,overthe Himalayas,it weakensand, within few days, it reverses out that this event has occurredwith the highestfrequency (monthlymeanof-0.5øC). To confirmthisanalysis, the time (from80% to 100%)exactlyovertheconsidered area(Fig. 1). seriesof temperature anomalyat 500 hPa averagedoverthe The analysisof the differencein the air temperature field at mesh 87.5øE-92.5øE, 27.5øN-32.5øN has been defined as the 500 hPafor January(evenyearsminusoddyears)highlightsa referencetime series.Lead-lagcorrelationswith analogous positiveareaof about1.5øCover the EasternHimalayasand time seriesaveragedover all the other 5øx5ø mesheshave the Tibetan Plateau(Fig. 2). Surprisingly,at the Nepalese beencalculated.At-10 daysa positivesignificantcorrelation station (Pyramid) and for the entire five-year period (0.4) appearswith a centerplacedin theArabianSeamoving considered,Januaryis warmerthan Februaryin even years, northeastward to the Himalayas.The +10 dayslag mapshows while the oppositeoccursin the oddones(Fig. 3). This event thatthepositiveanomalyhasreachedthe Bay of Bengalwith is also confirmedby the monthlymean griddedvaluesof a correlationcoefficientof 0.6; at +15 days,the correlation temperatureat 500 hPa and appears,over the southernslope overtheHimalayasis completelyreversed (-0.4). of theHimalayas,eastof 80øEwith a frequency of occurrence A biennial variability is also found in the seasonal higherthan85%. To investigatein detailthe evolutionof the precipitationamounts,especiallyin the 90's. In evenyears, temperature positiveanomaly,the dailygriddeddatasetat 500 wintertimeprecipitation is usuallymoreabundant thanduring hPahasbeenconsidered andthefive-daysaverages havebeen odd yearsalongthe southernslopeof the Himalayasand its calculated.Three main periodsof strongpositiveanomaly vicinity.This is confirmedbothby Pyramidand Tingri data over the Himalayasare evident.The first appearsduringthe and by the Xie-Arkin griddedanalysis.The sameoccursin secondhalf of December,when the air coolingand sinking summer monsoon precipitation (Fig. 4). However, the motion are establishingover the Continent,the westerly oscillation is not as regular as the winter temperature circulationis developingand a positiveanomaly,coming anomaly,probablydue to the contributionof local convective phenomena, particularlystrongat highaltitudes. To studythe circulationfeaturesat a synopticscale,the
Figure 1. Numberof cases(%) Januarymeantemperature at 500 hPaon an evenyearis higherthanthoseon thefollowing andprecedingoddyearsfor theperiod1980-1998.
geopotentialfield at 500 hPa has been analyzedboth for
45N
42N
•0•_•
39N
•
winterand summermonsoonseasons. The map reproducing ____jthe winter anomaly(even yearsminusodd years)showsa
-••
• •.• • • • • • •-._._...•
0.3•
36N
0.9
33N
1.2
30N
1.5
27N
1.5
24N
largeareaof positivevaluescenteredovertheHimalayas(Fig. 5). Monthly mapsrevealthat Januarymostlycontributes to thisconfiguration, with a highcenterof about21 m abovethe Himalayas.Decemberdoesnot showsignificantvaluesof the difference, while February has negative values over the Himalayas(-2 m). The meanconfigurationfor the summer monsoonmonths points out that during even years the "TibetanHigh" resultsslightly stronger(about 5 m) than duringodd years. As a consequence of the winter events,the Sub Tropical Jet Stream undergoesa southernshift and a variationin its
21N 18N 15N
65E
70E
75E
80E
85E
90E
95E
100E
105E
110E
Figure 2. Differencein the air temperature field (øC) at 500 hPa for January(even years minus odd years),calculated considering all theyearsin theperiod1980-1998.
intensity.On thesynopticmapsat 500 hPa,duringevenyears the Jet Stream,usuallylocatedjust southof the Himalayas, movesat higherlatitudesandweakens,bringingits axiscloser to the Rangeand,thus,causingan increasein wind speedat Pyramidand Tingri (for the last five yearsavailable).The
BERTOLANI
ET AL.' TBO IN THE EASTERN
HIGHLAND
HIMALAYAS
2187
(a) Winter 32
E
28
E 24 t.o_
20
•
16
o
8
•_
4 0
i
i
i
i
i
!
i
!
i
i
i
i
i
i
i
i
(b) Summer 450
•- 400 E 350
v
300
._o
250
• ß• •
200 150 100
._
Q-
50 0
o
oj
o•
(D
o3
o
oj
,•-
(D
03
Years
Figure 4. Seasonaltotalprecipitation(mm) for winter(a) andsummer(b). The light grey(left) and dark grey(right) rectangles showPyramiddata(1994-1998) andTingri data(1980-1998), respectively.
anomaloussignal of upper-levelwind field on South Asia duringwinterwasalsoobservedby Websterand Yang[1992].
in January. The yearby yearvariabilityin theoccurrence of eachstrongeventdetermines thespreadof theanomaly over thethree-weeks period.Stationdailyrecords of temperature validatethis large-scale view, moreevidentat the highest 4. Discussion and Conclusions stations (mainlyat thePyramid,beingon thewindwardside The analysisof stationdata hasrevealedthe existenceof a of theRange).Forall stations, in theevenyears,it is possible biennial oscillationover the EasternHighland Himalayas. to recognize at leastoneunusual warmepisode in January. At These resultsare confirmedby synopticmaps,which point the end of winter (February),the albedo due to the out that the maximaleffectsof TBO occurin the Himalayan accumulated new snowproduces a coolingeffecton the air Rangeandits surroundings duringthewinterseason. On the basis of the inverse correlation
between
Eurasian
and a consequentconsolidationof the subsidentmotion.
Griddeddatarevealthatin Marchtheexceeding wintersnow
meltedand both the negativetemperature snowcoverandsummermonsoonrainfall [Sankar-Raoet al., is completely anomalydisappear. 1996; Bamzai and Shukla, 1998], a possiblemechanism anomalyand the negativegeopotential activityis facilitatedandin thesummer explaining the observed oscillation, at least over the After that,convective Himalayas and the Tibetan Plateau, could be describedas follows.
After a summerseasonwith scarceprecipitation (odd 4SN
year), the ground ofthe highlands iscovered by thin layer of 42N snow. The following winter the surface has aa lower albedo,
'•'-'----4.•
there isalower ofthe solar heat to melt snow 'L••.•/-•O. therefore, the airconsumption coolingdue to surface heat sink andand, the 39N consequent subsident motion are less intense. This determines 36N '42-the weakeningof the thermalblockinganticyclone at ground 33N
levelovertheHimalayas andtheTibetan Plateau, therising of 30N the geopotentialat 500 hPa and the weakening and the
northerly shiftoftheSubTropical JetStream. Westerly zonal 27N circulation is modified and an anomaloustrough appears, 24N
whichdrives warmandhumidaircoming fromtheArabian 21N Sea toward the EasternHimalayas,increasingprecipitation.
Thewarmevents beginat theendof December andbecome •SN particularly intense during January, when the winter •SN
circulation iscompletely established and,consequently, the 65E 70E 75E 80E 85E 90E 95E 100E105E110E anomalies in thewesterlies aremorepronounced. Daily Figure 5. Difference inthegeopotential field(m)at500hPa
synoptic mapsof geopotential at 500hPashowat leastone for winter (even yearsminusodd years),calculated deeptroughpassing overthesouth-eastern Himalayas region considering all theyearsin theperiod1980-1998.
2188
BERTOLANI
ET AL.: TBO IN THE EASTERN
monsoonseasonthe "TibetanHigh" resultsslightly stronger than the precedingand the following year, with its higher positiveanomalyover the Range.This bringsmoreabundant precipitation,mainlyon the windwardsideof the Range.The abundanceof snow cover and soil moisturepreventsthe heatingof the groundduringthe followingwinter(odd year), the strongerthermal anticycloneplaced over the Tibetan Plateauprotectsthis area from southernwarm maritimeair, winter precipitationis lessabundant,the "TibetanHigh" is weakerand the summermonsoonprecipitationis lessertoo. So the cyclerepeats. The biennialoscillationobservedover the Himalayasand its surroundings was not found in the ?O's.It startedin 1982 and gaineda strongperiodicityonly in the last decade,with the winter anomaliespunctuallyrepeated.In 1982-1983 a particularlyanomalouswarm episodeis evidentin NINO 3 SSTA time series[Websterat al., 1998], as well as in the
HIGHLAND
HIMALAYAS
commentsandsuggestions. Pyramiddataareprovidedby IRSA/CNR
aspartof theEv-K2-CNR Project. References Bamzai, A., and J. Shukla, Relation between Eurasian snow cover,
snow depth and the Indian summermonsoon:an observational study,COLA Tech.Rep. n. 53, 42 pp., Centerfor Ocean-LandAtmosphereStudies,Calverton,Md., 1998. Chang, C.-P., and T. Li, A theory for the tropical tropospheric biennialoscillation,J. Atmos.Sci., accepted,1999. Kalnay,E., et al., The NCEP/NCAR 40-year reanalysisproject,Bull. Amer. Meteor. Soc., 77, 437-471, 1996.
Meehl, G. A., The annual cycle and interannualvariability in the tropical Indian and Pacific Oceanregions,Mon. Wea. Rev., 115, 27-50, 1987.
Meehl, G. A., The South Asian monsoonand the tropospheric biennial oscillation,J. Climate, 10, 1921-1943, 1997.
Murakami,T., Orographyand monsoons, in Monsoons,editedby J. S. Fein and P. L. Stephens,pp. 331-364, JohnWiley and Sons,
New York, 1987. tropicalcirculation[Ropelewskiet al., 1992]. This event could have given the startingphaseto the oscillation.Of Nicholls, N., Air-sea interactionand a quasi-biennialoscillation, Mon. Wea. Rev., 106, 1505-1508, 1978. course,the oscillationcouldhavea differentperiodicityon a Ogasawara,N., A. Kitoh, T. Yasunari,and A. Noda, Tropospheric longerperiod(near to the peakof 2-3 years,seeTomiraand biennial oscillation of ENSO-Monsoon system in the MRI
Yasunari,1996; Meehl, 1997). The biennialvariabilityunder coupledGCM, J. Meteor. Soc.Japan, 77, 1247-1270, 1999. discussionis clearly modulatedby land-surfaceprocesses, Ose, T., The comparisonof the simulatedresponseto the regional snowmassanomaliesover Tibet, EasternEurope,and Siberia,J. thoughlargelydependingon synopticforcingas SSTA, which Meteor. Soc.Japan, 74, 845-866, 1996. provide more or less moistureand influencethe meridional Ropelewki, C. F., M. S. Halpert, and X. Wang, Observed temperature contrast between land and ocean. Thus, SSTA
troposphericbiennial variability and its relationshipto the
southernoscillation,J. Climate, 5, 594-614, 1992. would give rise to an oscillationof the same sign of that Sankar-Rao,M., K. M. Lau, and Song Yang, On the relationship proposed.These resultssuggestthat TBO is an intrinsic between Eurasian snow cover and the Asian summer monsoon, oscillationof monsooncirculationand emphasizethe role lnt. J. Clim., 16, 605-616, 1996. playedby the extra-tropicalinteraction. Shen,S., and K. M. Lau, Biennial oscillationassociatedwith the east Asian summermonsoonand tropical sea surfacetemperature,J. Meehl [1997] recognized different structuresof the Meteor. Soc.Japan, 73, 105-124, 1995. geopotentialfield at 500 hPa and surfaceheatingfor three Tomita, T., and T. Yasunari, Role of the Northeast winter monsoon years(1987-1989) centeredon a strongmonsoonyear(1988). on the biennial oscillation of the ENSO/Monsoon system,J. Many featuresof his work can be found in the theoryexposed Meteor. Soc.Japan, 74, 339-413, 1996. above, mainly the configurationof geopotentialand the Vernekar, A.D., J. Zhou, and J. Shukla,The effect of Eurasiansnow variation of positionof the 500 hPa trough in successive cover on the Indian monsoon,J. Climate, 8, 248-266, 1995. years.
Webster,P. J., V. O. Magafia,T. N. Palmer,J. Shukla,R. A. Tomas, M. Yanai, andT. Yasunari,Monsoons: processes, predictability,
General Circulation Model (GCM) experiments[e.g., and the prospectsfor the prediction,J. Geophys.Res., 103, Vernekaret al., 1995; Ogasawaraet al., 1999] assessed the 14451-14510, 1998. great influence of snow cover on surface heat fluxes and Webster,P.J., and Song Yang, Monsoonand ENSO: selectively interactivesystems,Q. J. R. Meteorol.Soc.,118, 877-926, 1992. precipitation.Vernekaret al. [1995] alsofoundthatthe effect Xie, P., and P. A. Arkin, Analysisof global monthlyprecipitation on air temperature wasparticularlyenhanced overtheTibetan usinggaugeobservations, satelliteestimates, andnumerical model Plateau.
Fromthis studyemergesthe importanceof stationdatasets in the comprehension of the mechanisms of TBO, thatreveals clear featuresalso at high altitudes.On the other hand,data provided by a GCM have an overall intrinsic limit of reliabilityand detailin areawith complexorography. Point measuresof snowcoverdepthare necessary to assess the role of surface processeswith more certainty. Unfortunately,the operatingmeteorological stationslocated in that remotesite are very scarce,especiallyon the southern slopeof theRange,anda detailedandcomplete description of the phenomenon is not simple.Anyway,thisattemptcouldbe an incentivefor the planningof future researchesto be developedat high altitudes.One of the aims couldbe the study of the effectsof TBO on the long-rangetransportof pollutants deposited on snow cover in the Highland Himalayas.
predictions,J. Climate, 9, 840-858, 1996. Yanai, M., and T. Tomita, Seasonaland interannualvariability of
atmospheric heatsourcesandmoisturesinksas determined from NCEP/NCAR reanalysis, J. Climate,11, 463-482, 1998. Yanai, M., C. Li, and Z. Song, Seasonalheatingof the Tibetan Plateau and its effects on the evolution of the Asian summer
monsoon, J. Meteor. Soc.Japan, 70, 319-351, 1992. Yang, S., and K. M. Lau, Influenceof SST and groundwetnesson the Asian summermonsoon,J. Climate, 11, 3230-3246, 1998. Yasunari,T., A. Kitoh, and T. Tokioka, Local and remoteresponses to excessivesnowmassover Eurasiaappearingin the northern
springand summerclimate:a studywith the MRI-GCM, J. Meteor. Soc.Japan, 69, 473-487, 1991. L. Bertolani and M. Bollasina, Climate ResearchDivision, Centro
EpsonMeteo, via Pisa 250, 20099 SestoS. Giovanni,Milan, Italy. (e-mail: [email protected]) G. Tartari, Istituto di Ricerca sulle Acque, ConsiglioNazionale delle Ricerche,via della Mornera25, 20047 Brugherio,Milan, Italy. (e-mail: [email protected])
Acknowledgments.The authorsthankM. Yanai(UCLA) for his (ReceivedNovember4, 1999;revisedMay 18, 2000; encouragement,and two anonymousreferees for their useful accepted June12,2000.)
Recentbiennial variability of meteorologicalfeatures in the Eastern Highland Himalayas Laura Bertolani
and Massimo
Bollasina
ClimateResearchDivision,CentroEpsonMeteo,Milan, Italy.
Gianni
Tartari
IRSA-CNR, Milan, Italy.
Abstract. With meteorologicaldata from high altitude surface stations and gridded dataset from NCEP/NCAR Reanalysis,a biennial oscillationover the EasternHighland Himalayasfrom 1980 to 1998 is described.This variability concernsair temperature, precipitation, geopotential andwind speed.Evidenceis given on the connectionsbetweenlocal dataandlarge-scale circulationpatterns. The mostremarkable oscillatingfeaturesare found duringwinter and, in general, the signalsare particularlymarkedon the southernslopeof the HimalayanRange.A possiblemechanism is explainedin terms of a periodicityin surfaceheat and moisturefluxes. Finally, the peculiarityof the region as a climatic change observatoryis underlined.
Centerfor Atmospheric Research(NCAR). The resultspermit to supportandto extendto highaltitudesone of the possible mechanisms of TBO, usuallydescribed by meansof modeling
Chang and Li, 1999]. Some other works focus on the importanceof land-surface processes, especiallysnowcover and soil moistureover Eurasia[e.g., Yasunariet al., 1991; Vernekaret al., 1995;Ose, 1996]. Moreover,the importance
87.1øE, 4302 m), Xigatse(29.3øN 88.9øE,3837 m) and Xainza (31.0øN 88.6øE, 4671 m). The daily data of the
studies.
2. Data
and Methods
The possibilityof a biennialperiodicityin the courseof meteorologicalparametersis first investigatedconsidering stationtime seriesfrom 1980 to 1998. The synopticpatterns of circulationabovethe Asian Continent(15øN-45øN,65øE110øE) are studiedto explain the large-scaleorigin of the phenomenaobservedat the stations;numericalvaluesand mapsare plottedby meansof daily/monthlydata available from CDAS-NCAR Reanalysis on a 2.5ø x 2.5ø grid [Kalnay 1. Introduction et al., 1996]. Precipitation griddeddatahavebeencollected from the monthlydatasetof ClimatePredictionCenter(CPC) The TBO (Tropospheric BiennialOscillation)is a quasi- MergedAnalysisof Precipitation[Xie andArkin, 1996]. biennialperiodicityin the courseof many variables(e.g., Four stationsare considered:the Laboratory-Observatory precipitation, atmospheric pressure, seasurfacetemperature) called "Pyramid",locatedin the Nepal HighlandHimalayas, occurring in theIndianandPacificregions.It is recognized to andthreeWMO stations,placedin thesouth-eastern regionof be the resultof large-scaleinteractionamongsea,land and Tibet. atmosphere,includingtropical-mid-latitude interaction.Its The Pyramidmeteorological stationwassetup in 1990 by originis a matterof interesting debatesand severaltheories the Water Research Institute of the Italian National Research exist[e.g., Websteret al., 1998].The interannual variability Council(IRSA/CNR) and sinceDecember1993 it has run of the Asian summer monsoonand its relationshipwith continuouslyyear round. The Pyramid is locatedin the tropical Sea SurfaceTemperatureAnomalies(SSTA) has Khumbu Valley at 5050 m next to Mt. Everest. Its beenoutlinedby manyauthors[e.g.,Nicholls,1978;Meehl, geographical coordinates are28.0øN86.8øE. 1987, 1997; Shen and Lau, 1995; Yang and Lau, 1998; The Tibetan stations consideredare: Tingri (28.6øN
of the Tibetan Plateau as an elevated heat source/sink is well
described[Yanai et al., 1992; Murakami, 1987] and a biennialsignal in the tendencies of heat fluxes has been recognized[Yanaiand Tomita,1998]. In this paper,a biennialoscillationfound in the Eastern Himalayasis highlightedby meansof daily (from 1994 to 1998) andmonthly(from 1980 to 1998) datarecordedat four stationslocatedat high altitudes(above3800 m). The local climaticfeaturesare confirmedandexplainedby the analysis of large-scalesynopticdata availablefrom the National
Chinese stationsare collectedfrom NOAA global surface summaryof daily observations for the period 1994-1998. Monthly summarydatasetsfrom NCAR (CAC GLOBAL CEAS summary)arealsousedfor theperiod1980-1998.
The followingmeteorological variablesare considered in this analysis:air temperature, precipitation, geopotential and wind speed.
3. Analysis and Results A variabilitywith a periodof 24 monthsappearsin many parameters, suchas winter temperatures and wind speed, winterandsummerprecipitation andgeopotential.
Copyright2000 by the AmericanGeophysical Union.
The mostremarkablefeatureappearsin wintertemperature values: during even years, January mean temperaturesare
Papernumber1999GL011198.
warmerthanduringodd yearsfor all the stationsandfor the
greaternumberof cases. NCEP/NCARReanalysis datapoint
0094-8276/00/1999GL011198505.00
2185
2186
BERTOLANI
ET AL.' TBO IN THE EASTERN
HIGHLAND
HIMALAYAS
PyramidStation o
o..... -7.0
a) -8.0
•
-9.0
• -10.0
-11.0 -12.0 94
95
96
97
98
Years
Figure 3. Air meantemperature (øC) at Pyramidfrom 1994 to 1998.The solidanddashedlinesshowdatafor Januaryand February,respectively. 65E
70E
75E
80E
85E
90E
95E
100E
105E
110E
from the Arabian Sea, reachesthe Range. Anyway, the monthlymeananomalyresultsslightlynegative(-0.5øC).The strongest episodeappearsin thefirst threeweeksof January, whenwintercirculationis fully established. At thebeginning of Februarythe anomalymovessoutheastward; thus,overthe Himalayas,it weakensand, within few days, it reverses out that this event has occurredwith the highestfrequency (monthlymeanof-0.5øC). To confirmthisanalysis, the time (from80% to 100%)exactlyovertheconsidered area(Fig. 1). seriesof temperature anomalyat 500 hPa averagedoverthe The analysisof the differencein the air temperature field at mesh 87.5øE-92.5øE, 27.5øN-32.5øN has been defined as the 500 hPafor January(evenyearsminusoddyears)highlightsa referencetime series.Lead-lagcorrelationswith analogous positiveareaof about1.5øCover the EasternHimalayasand time seriesaveragedover all the other 5øx5ø mesheshave the Tibetan Plateau(Fig. 2). Surprisingly,at the Nepalese beencalculated.At-10 daysa positivesignificantcorrelation station (Pyramid) and for the entire five-year period (0.4) appearswith a centerplacedin theArabianSeamoving considered,Januaryis warmerthan Februaryin even years, northeastward to the Himalayas.The +10 dayslag mapshows while the oppositeoccursin the oddones(Fig. 3). This event thatthepositiveanomalyhasreachedthe Bay of Bengalwith is also confirmedby the monthlymean griddedvaluesof a correlationcoefficientof 0.6; at +15 days,the correlation temperatureat 500 hPa and appears,over the southernslope overtheHimalayasis completelyreversed (-0.4). of theHimalayas,eastof 80øEwith a frequency of occurrence A biennial variability is also found in the seasonal higherthan85%. To investigatein detailthe evolutionof the precipitationamounts,especiallyin the 90's. In evenyears, temperature positiveanomaly,the dailygriddeddatasetat 500 wintertimeprecipitation is usuallymoreabundant thanduring hPahasbeenconsidered andthefive-daysaverages havebeen odd yearsalongthe southernslopeof the Himalayasand its calculated.Three main periodsof strongpositiveanomaly vicinity.This is confirmedbothby Pyramidand Tingri data over the Himalayasare evident.The first appearsduringthe and by the Xie-Arkin griddedanalysis.The sameoccursin secondhalf of December,when the air coolingand sinking summer monsoon precipitation (Fig. 4). However, the motion are establishingover the Continent,the westerly oscillation is not as regular as the winter temperature circulationis developingand a positiveanomaly,coming anomaly,probablydue to the contributionof local convective phenomena, particularlystrongat highaltitudes. To studythe circulationfeaturesat a synopticscale,the
Figure 1. Numberof cases(%) Januarymeantemperature at 500 hPaon an evenyearis higherthanthoseon thefollowing andprecedingoddyearsfor theperiod1980-1998.
geopotentialfield at 500 hPa has been analyzedboth for
45N
42N
•0•_•
39N
•
winterand summermonsoonseasons. The map reproducing ____jthe winter anomaly(even yearsminusodd years)showsa
-••
• •.• • • • • • •-._._...•
0.3•
36N
0.9
33N
1.2
30N
1.5
27N
1.5
24N
largeareaof positivevaluescenteredovertheHimalayas(Fig. 5). Monthly mapsrevealthat Januarymostlycontributes to thisconfiguration, with a highcenterof about21 m abovethe Himalayas.Decemberdoesnot showsignificantvaluesof the difference, while February has negative values over the Himalayas(-2 m). The meanconfigurationfor the summer monsoonmonths points out that during even years the "TibetanHigh" resultsslightly stronger(about 5 m) than duringodd years. As a consequence of the winter events,the Sub Tropical Jet Stream undergoesa southernshift and a variationin its
21N 18N 15N
65E
70E
75E
80E
85E
90E
95E
100E
105E
110E
Figure 2. Differencein the air temperature field (øC) at 500 hPa for January(even years minus odd years),calculated considering all theyearsin theperiod1980-1998.
intensity.On thesynopticmapsat 500 hPa,duringevenyears the Jet Stream,usuallylocatedjust southof the Himalayas, movesat higherlatitudesandweakens,bringingits axiscloser to the Rangeand,thus,causingan increasein wind speedat Pyramidand Tingri (for the last five yearsavailable).The
BERTOLANI
ET AL.' TBO IN THE EASTERN
HIGHLAND
HIMALAYAS
2187
(a) Winter 32
E
28
E 24 t.o_
20
•
16
o
8
•_
4 0
i
i
i
i
i
!
i
!
i
i
i
i
i
i
i
i
(b) Summer 450
•- 400 E 350
v
300
._o
250
• ß• •
200 150 100
._
Q-
50 0
o
oj
o•
(D
o3
o
oj
,•-
(D
03
Years
Figure 4. Seasonaltotalprecipitation(mm) for winter(a) andsummer(b). The light grey(left) and dark grey(right) rectangles showPyramiddata(1994-1998) andTingri data(1980-1998), respectively.
anomaloussignal of upper-levelwind field on South Asia duringwinterwasalsoobservedby Websterand Yang[1992].
in January. The yearby yearvariabilityin theoccurrence of eachstrongeventdetermines thespreadof theanomaly over thethree-weeks period.Stationdailyrecords of temperature validatethis large-scale view, moreevidentat the highest 4. Discussion and Conclusions stations (mainlyat thePyramid,beingon thewindwardside The analysisof stationdata hasrevealedthe existenceof a of theRange).Forall stations, in theevenyears,it is possible biennial oscillationover the EasternHighland Himalayas. to recognize at leastoneunusual warmepisode in January. At These resultsare confirmedby synopticmaps,which point the end of winter (February),the albedo due to the out that the maximaleffectsof TBO occurin the Himalayan accumulated new snowproduces a coolingeffecton the air Rangeandits surroundings duringthewinterseason. On the basis of the inverse correlation
between
Eurasian
and a consequentconsolidationof the subsidentmotion.
Griddeddatarevealthatin Marchtheexceeding wintersnow
meltedand both the negativetemperature snowcoverandsummermonsoonrainfall [Sankar-Raoet al., is completely anomalydisappear. 1996; Bamzai and Shukla, 1998], a possiblemechanism anomalyand the negativegeopotential activityis facilitatedandin thesummer explaining the observed oscillation, at least over the After that,convective Himalayas and the Tibetan Plateau, could be describedas follows.
After a summerseasonwith scarceprecipitation (odd 4SN
year), the ground ofthe highlands iscovered by thin layer of 42N snow. The following winter the surface has aa lower albedo,
'•'-'----4.•
there isalower ofthe solar heat to melt snow 'L••.•/-•O. therefore, the airconsumption coolingdue to surface heat sink andand, the 39N consequent subsident motion are less intense. This determines 36N '42-the weakeningof the thermalblockinganticyclone at ground 33N
levelovertheHimalayas andtheTibetan Plateau, therising of 30N the geopotentialat 500 hPa and the weakening and the
northerly shiftoftheSubTropical JetStream. Westerly zonal 27N circulation is modified and an anomaloustrough appears, 24N
whichdrives warmandhumidaircoming fromtheArabian 21N Sea toward the EasternHimalayas,increasingprecipitation.
Thewarmevents beginat theendof December andbecome •SN particularly intense during January, when the winter •SN
circulation iscompletely established and,consequently, the 65E 70E 75E 80E 85E 90E 95E 100E105E110E anomalies in thewesterlies aremorepronounced. Daily Figure 5. Difference inthegeopotential field(m)at500hPa
synoptic mapsof geopotential at 500hPashowat leastone for winter (even yearsminusodd years),calculated deeptroughpassing overthesouth-eastern Himalayas region considering all theyearsin theperiod1980-1998.
2188
BERTOLANI
ET AL.: TBO IN THE EASTERN
monsoonseasonthe "TibetanHigh" resultsslightly stronger than the precedingand the following year, with its higher positiveanomalyover the Range.This bringsmoreabundant precipitation,mainlyon the windwardsideof the Range.The abundanceof snow cover and soil moisturepreventsthe heatingof the groundduringthe followingwinter(odd year), the strongerthermal anticycloneplaced over the Tibetan Plateauprotectsthis area from southernwarm maritimeair, winter precipitationis lessabundant,the "TibetanHigh" is weakerand the summermonsoonprecipitationis lessertoo. So the cyclerepeats. The biennialoscillationobservedover the Himalayasand its surroundings was not found in the ?O's.It startedin 1982 and gaineda strongperiodicityonly in the last decade,with the winter anomaliespunctuallyrepeated.In 1982-1983 a particularlyanomalouswarm episodeis evidentin NINO 3 SSTA time series[Websterat al., 1998], as well as in the
HIGHLAND
HIMALAYAS
commentsandsuggestions. Pyramiddataareprovidedby IRSA/CNR
aspartof theEv-K2-CNR Project. References Bamzai, A., and J. Shukla, Relation between Eurasian snow cover,
snow depth and the Indian summermonsoon:an observational study,COLA Tech.Rep. n. 53, 42 pp., Centerfor Ocean-LandAtmosphereStudies,Calverton,Md., 1998. Chang, C.-P., and T. Li, A theory for the tropical tropospheric biennialoscillation,J. Atmos.Sci., accepted,1999. Kalnay,E., et al., The NCEP/NCAR 40-year reanalysisproject,Bull. Amer. Meteor. Soc., 77, 437-471, 1996.
Meehl, G. A., The annual cycle and interannualvariability in the tropical Indian and Pacific Oceanregions,Mon. Wea. Rev., 115, 27-50, 1987.
Meehl, G. A., The South Asian monsoonand the tropospheric biennial oscillation,J. Climate, 10, 1921-1943, 1997.
Murakami,T., Orographyand monsoons, in Monsoons,editedby J. S. Fein and P. L. Stephens,pp. 331-364, JohnWiley and Sons,
New York, 1987. tropicalcirculation[Ropelewskiet al., 1992]. This event could have given the startingphaseto the oscillation.Of Nicholls, N., Air-sea interactionand a quasi-biennialoscillation, Mon. Wea. Rev., 106, 1505-1508, 1978. course,the oscillationcouldhavea differentperiodicityon a Ogasawara,N., A. Kitoh, T. Yasunari,and A. Noda, Tropospheric longerperiod(near to the peakof 2-3 years,seeTomiraand biennial oscillation of ENSO-Monsoon system in the MRI
Yasunari,1996; Meehl, 1997). The biennialvariabilityunder coupledGCM, J. Meteor. Soc.Japan, 77, 1247-1270, 1999. discussionis clearly modulatedby land-surfaceprocesses, Ose, T., The comparisonof the simulatedresponseto the regional snowmassanomaliesover Tibet, EasternEurope,and Siberia,J. thoughlargelydependingon synopticforcingas SSTA, which Meteor. Soc.Japan, 74, 845-866, 1996. provide more or less moistureand influencethe meridional Ropelewki, C. F., M. S. Halpert, and X. Wang, Observed temperature contrast between land and ocean. Thus, SSTA
troposphericbiennial variability and its relationshipto the
southernoscillation,J. Climate, 5, 594-614, 1992. would give rise to an oscillationof the same sign of that Sankar-Rao,M., K. M. Lau, and Song Yang, On the relationship proposed.These resultssuggestthat TBO is an intrinsic between Eurasian snow cover and the Asian summer monsoon, oscillationof monsooncirculationand emphasizethe role lnt. J. Clim., 16, 605-616, 1996. playedby the extra-tropicalinteraction. Shen,S., and K. M. Lau, Biennial oscillationassociatedwith the east Asian summermonsoonand tropical sea surfacetemperature,J. Meehl [1997] recognized different structuresof the Meteor. Soc.Japan, 73, 105-124, 1995. geopotentialfield at 500 hPa and surfaceheatingfor three Tomita, T., and T. Yasunari, Role of the Northeast winter monsoon years(1987-1989) centeredon a strongmonsoonyear(1988). on the biennial oscillation of the ENSO/Monsoon system,J. Many featuresof his work can be found in the theoryexposed Meteor. Soc.Japan, 74, 339-413, 1996. above, mainly the configurationof geopotentialand the Vernekar, A.D., J. Zhou, and J. Shukla,The effect of Eurasiansnow variation of positionof the 500 hPa trough in successive cover on the Indian monsoon,J. Climate, 8, 248-266, 1995. years.
Webster,P. J., V. O. Magafia,T. N. Palmer,J. Shukla,R. A. Tomas, M. Yanai, andT. Yasunari,Monsoons: processes, predictability,
General Circulation Model (GCM) experiments[e.g., and the prospectsfor the prediction,J. Geophys.Res., 103, Vernekaret al., 1995; Ogasawaraet al., 1999] assessed the 14451-14510, 1998. great influence of snow cover on surface heat fluxes and Webster,P.J., and Song Yang, Monsoonand ENSO: selectively interactivesystems,Q. J. R. Meteorol.Soc.,118, 877-926, 1992. precipitation.Vernekaret al. [1995] alsofoundthatthe effect Xie, P., and P. A. Arkin, Analysisof global monthlyprecipitation on air temperature wasparticularlyenhanced overtheTibetan usinggaugeobservations, satelliteestimates, andnumerical model Plateau.
Fromthis studyemergesthe importanceof stationdatasets in the comprehension of the mechanisms of TBO, thatreveals clear featuresalso at high altitudes.On the other hand,data provided by a GCM have an overall intrinsic limit of reliabilityand detailin areawith complexorography. Point measuresof snowcoverdepthare necessary to assess the role of surface processeswith more certainty. Unfortunately,the operatingmeteorological stationslocated in that remotesite are very scarce,especiallyon the southern slopeof theRange,anda detailedandcomplete description of the phenomenon is not simple.Anyway,thisattemptcouldbe an incentivefor the planningof future researchesto be developedat high altitudes.One of the aims couldbe the study of the effectsof TBO on the long-rangetransportof pollutants deposited on snow cover in the Highland Himalayas.
predictions,J. Climate, 9, 840-858, 1996. Yanai, M., and T. Tomita, Seasonaland interannualvariability of
atmospheric heatsourcesandmoisturesinksas determined from NCEP/NCAR reanalysis, J. Climate,11, 463-482, 1998. Yanai, M., C. Li, and Z. Song, Seasonalheatingof the Tibetan Plateau and its effects on the evolution of the Asian summer
monsoon, J. Meteor. Soc.Japan, 70, 319-351, 1992. Yang, S., and K. M. Lau, Influenceof SST and groundwetnesson the Asian summermonsoon,J. Climate, 11, 3230-3246, 1998. Yasunari,T., A. Kitoh, and T. Tokioka, Local and remoteresponses to excessivesnowmassover Eurasiaappearingin the northern
springand summerclimate:a studywith the MRI-GCM, J. Meteor. Soc.Japan, 69, 473-487, 1991. L. Bertolani and M. Bollasina, Climate ResearchDivision, Centro
EpsonMeteo, via Pisa 250, 20099 SestoS. Giovanni,Milan, Italy. (e-mail: [email protected]) G. Tartari, Istituto di Ricerca sulle Acque, ConsiglioNazionale delle Ricerche,via della Mornera25, 20047 Brugherio,Milan, Italy. (e-mail: [email protected])
Acknowledgments.The authorsthankM. Yanai(UCLA) for his (ReceivedNovember4, 1999;revisedMay 18, 2000; encouragement,and two anonymousreferees for their useful accepted June12,2000.)
