DTIC ADA523842: Seasonal Variation of Gravity Wave Sources From Satellite Observation

2.6 MB·English

by  Defense Technical Information Center

AdvancesinSpaceResearch35(2005)1925–1932
www.elsevier.com/locate/asr
Seasonal variation of gravity wave sources from satellite observation
J.H. Jiang a,*, S.D. Eckermann b, D.L. Wu a, K. Hocke c, B. Wang d, J. Ma b, Y. Zhang d
aJetPropulsionLaboratory,CaliforniaInstituteofTechnology,Pasadena,CA,USA
bE.O.HulburtCenterforSpaceResearch,NavalResearchLaboratory,Washington,DC,USA
cNationalInstituteofInformationandCommunicationsTechnology,Japan
dSchoolofOceanandEarthScienceandTechnology,UniversityofHawaii,Honolulu,Hawaii,USA
Received8September2004;receivedinrevisedform24January2005;accepted25January2005
Abstract
Recent high-resolution satellite observations of gravity waves in the middle atmosphere have shown correlations with the
strength of thestratospheric jetstream,surface topography,and tropical convection. Seasonalvariations of wave-inducedstrato-
sphericradiancevariancesareoftenthemanifestationsofmodulationsofthesesourcesandrefractiveinfluences.Inthispaper,we
focusontheseasonalclimatologyofgravitywavesobservedbytheUARSMLS,whilealsoshowingsomenewresultsfromGPSand
AMSUinstruments.OuranalysisisaidedbyMWFMmodelingofmountainwavesathighlatitudesandCMAPprecipitationindi-
cesin the subtropicsto provide aclearer picture ofglobalgravity wavedynamics.
(cid:1)2005COSPAR. PublishedbyElsevier Ltd. Allrights reserved.
1. Introduction 2002), the second generation MLS instrument on board
Aura (launched in 2004), and a number of current fly-
Atmospheric gravity waves (GWs) are one of the ing low earth orbiters carrying GPS (Global Position
most important processes in the middle atmosphere System) receivers using radio occultation technology.
(Kim et al., 2003; Fritts and Alexander, 2003). Since Multiple-satellite data provide new opportunities for
the launch of the Upper Atmosphere Research Satellite data-model comparisons that can resolve gaps in our
(UARS) in 1991, the Microwave Limb Sounder (MLS) knowledge about stratospheric and mesospheric
instrument on board UARS, has provided a wealth of processes.
stratospheric GW data quasi-continuously from 1991 This study summarizes seasonal climatology of grav-
to 1997 (Wu and Waters, 1997; Jiang et al., 2003), en- ity waves observed by UARS MLS in the 1990s, and
abling compilation of multi-year climatologies and presents some examples of new GW measurements ex-
investigations of interannual variability and trends tractedfromAMSU andGPSinstrumentsinthe2000s.
(e.g., Jiang et al., 2004b). Most recently, a number of
new satellite instruments have been returning vast
amounts of new GW data with unprecedented resolu- 2. UARS MLS observations (1991–1997)
tion, accuracy and global coverage. Examples include
the Advanced Microwave Sounding Unit (AMSU) The UARS MLS GW measurements are most sen-
(Wu, 2004) and the Atmospheric Infrared Sounder sitive to waves with vertical wavelength >10km and
(AIRS) on board the Aqua satellite (launched in horizontal wavelength <100km. A review of the com-
putational procedure of UARS MLS GW radiance
* Correspondingauthor.Tel.:[phone];fax:[phone]. variance (including noise removal and recent updates)
E-mailaddress:[email protected](J.H.Jiang). has been given by Jiang et al. (2004a). The first global
0273-1177/$30 (cid:1)2005COSPAR.PublishedbyElsevierLtd.Allrightsreserved.
doi:10.1016/j.asr.2005.01.099
Report Documentation Page Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it

does not display a currently valid OMB control number.
1. REPORT DATE 3. DATES COVERED
24 JAN 2005 2. REPORT TYPE 00-00-2005 to 00-00-2005
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Seasonal variation of gravity wave sources from satellite observation
5b. GRANT NUMBER
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
Naval Research Laboratory,E.O. Hulburt Center for Space REPORT NUMBER
Research,Washington,DC,20375
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT
NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
14. ABSTRACT
see report
15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF
ABSTRACT OF PAGES RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE Same as 8
unclassified unclassified unclassified Report (SAR)
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
1926 J.H.Jiangetal./AdvancesinSpaceResearch35(2005)1925–1932
maps of GW radiance variance derived from UARS 2.1. Northern high-latitudes
MLS data were introduced by Wu and Waters
(1996, 1997). Since then, a number of detailed fol- Fig. 1 shows hemispheric maps of the 6-point limb-
low-up studies of MLS-observed GWs have been per- track ND normalized radiance temperature variances
formed, tracing many wave features to tropospheric ðT02=T2Þ at (cid:1)33km averaged for northern winter
b b
convection and surface orography through some de- (December–February, 1994–1997), spring (March–
tailed data-model analyses. For example, Jiang et al. May, 1995–1997), summer (June–August, 1995–1997),
(2004a) modeled MLS measurements of wintertime and fall (September–November, 1995–1997). Jiang
GW activity in the northern hemisphere using the etal.(2004a)haveshownthattheMLShasgreatestsen-
Mountain Wave Forecast Model (MWFM) in a hind- sitivity to mountain waves measured at high latitudes
cast mode and with a full limb-track MLS GW visibil- ((cid:1)50(cid:2)N–75(cid:2)N). Their combined data analysis and
ity function applied to the model output, to mimic MLS-filtered MWFM modeling showed that many
observation of these waves from space by MLS. MLSvarianceenhancementsinthenorthernhemisphere
McLandress et al. (2000) and Jiang et al. (2004b) have can be associated with mountain waves forced by flow
shown that both the geographical distribution and in- over specific mountainous terrain. These include moun-
ter-seasonal variation of subtropical GW variance tain ranges in Europe, North America, southeastern
measured by MLS are closely related to deep convec- Greenland and Iceland.
tion. These studies also showed that different MLS Duetovariationofthestratosphericjet-stream,these
observation modes (e.g., viewing directions) are sensi- orographicwaveshaveastrongreproducibleannualcy-
tive to different wave structures emanating from differ- cle,peakinginwinterandweakeninginotherseasonsas
ent sources, due to distinct wind and wave shown in Fig. 1.Since the MLS variances are related to
propagation directions at different latitude zones. the GW temperature amplitudes, these measurements
For example, MLS measurements conducted on can provide critical hemispheric data on mountain
north-looking descending (ND) orbits are most sensi- wave-induced temperature variability in the high-lati-
tive to wintertime orographic GWs at high northern tudestratospherethatimpactspolarstratosphericcloud
latitudes, while the convection-related subtropical (PSC) formation. Such data are critical for refining and
GWs are most prominent in MLS north-looking constraining the parameterizations for these processes
ascending (NA) and south-looking descending (SD) for global chemical transport models (CTMs) of strato-
data. The seasonal and vertical variations of these spheric ozone evolution (Pierce et al., 2002).
wave sources are therefore summarized according to The vertical variation of the wintertime GWs can be
various modes of observation spanning different lati- seen in Fig. 2, where the MLS variances are mapped at
tude regions. five different altitudes throughout the middle atmo-
December-February, 1994-1997 March-May, 1995-1997
>4.5
65 65
1
e 0
ud 45 45 7
atit 0
L
25 25
0 60 120 180 240 300 360 0 60 120 180 240 300 360
June-August, 1995-1997 September-November, 1995-1997
>4.5
65 65
1
0
e 7
d 45 45
u
atit 0
L
25 25
0 60 120 180 240 300 360 0 60 120 180 240 300 360
Longitude Longitude
Fig.1. Seasonalmeanfieldsofnormalizedradiancetemperaturevarianceat33kmcomputedfromMLSlimb-trackNDdata(asdescribedbyJiang
etal.(2004a)).ThedottedwhitecontoursareUKMOwindsataboutthesamealtitudesandaveragedoverMLSmeasurementdays.
J.H.Jiangetal./AdvancesinSpaceResearch35(2005)1925–1932 1927
sphere (28–80km). The geographical distributions of
GW variance look similar throughout the stratosphere,
butundergosomesignificantchangesatthestratopause.
For example, the wave variance enhancement over cen-
tral Eurasia seems to break apart at altitudes above
(cid:1)60km,butmid-latitudeGWvariancesacrosstheUni-
ted States increase above 60km. The normalized radi-
ance variances can be directly related to GW energy
and momentum fluxes. The relative amplitudes of vari-
ous wave patterns may reflect relative wave dissipation
or growth as it enters mesosphere from stratosphere,
or may reflect GWs entering/exiting the MLS viewing
windows as winds vary with height.
2.2. Sub-tropics
Fig.3showstheseasonalmapsofsubtropical(40(cid:2)S–
40(cid:2)N) MLS normalized radiance temperature variances
averaged for December–February 1991–1994, March–
May 1992–1994, June–August, 1992–1994, and Septem-
ber–November, 1992–1994, respectively. As in Jiang
et al. (2004b), the MLS 4-point limb-scan NA or SD
measurementsareusedtogeneratethesemapssincethey
are most sensitive to convectively-generated GWs in
sub-tropical latitudes.
To see the correlation of the subtropical waves to
convectiveactivities,Fig.4showsseasonalprecipitation
maps derived from the NOAA Climate Prediction Cen-
terMergedAnalysisofPrecipitation(CMAP),andaver-
aged for the MLS limb-scan observation days. The
CMAP blends station rain gauge observations and five
Fig.2. Mapsofwintertimenormalizedradiancetemperaturevariance different types of satellite products to estimate global
at28–80kmaltitudes,computedfromMLSlimb-trackNDdata.The precipitation amounts (Huffman et al., 1997; Xie and
dottedwhitecontoursbelow80kmareUKMOwinds.Noassimilated Arkin, 1997). Similar to the traditional OLR (outgoing
winddataareavailableat80km.
Fig. 3. Seasonal meanfieldsofMLSnormalizedradiancetemperaturevarianceat38km.ForDecember–February1991–1994andMarch–May
1992–1994,thevaluesarecomputedusingmeasurementsfromlimb-scanNAorbits;forJune–AugustandSeptember–November1992–1994,limb-
scanSDdataareused.ThedottedwhitecontoursareUKMOwindsataboutsamealtitudesandaveragedoverMLSmeasurementdays.
1928 J.H.Jiangetal./AdvancesinSpaceResearch35(2005)1925–1932
Fig.4. Seasonalmeanprecipitation(mm/day)mapsderivedfromtheNOAAClimatePredictionCenterMergedAnalysisofPrecipitation(CMAP),
andaveragedforthesameMLSlimb-scanobservationdays.
longwave radiation) data, the CMAP index is a good inFig.4areshowninFig.5,wherethenormalizedradi-
indicatorofdeepconvection.Itsharesthecommonpat- ancevariancesaremappedatfivedifferentaltitudes(28–
terns found in OLR data, but tends to include more 80km).Thewaveactivitypatternsforboththesouthern
small-scale details. (left-column) and northern (right-column) summer
Studying Fig. 4, we see that convection events in the monsoon seasons are mostly maintained throughout
tropicsappearin all four ofthe conventionalextratrop- the stratosphere, although some slight poleward shifts
icalseasons.Convectively-generatedGWs(Fig.3),how- are noticeablebetween 28 and 33km. This suggests fast
ever, are predominant in MLS data only in the low waves that propagate into the mesosphere without
latitudesofsummerhemispheres,i.e.,December–Febru- extensive lateral propagation, consistent with current
ary in southern sub-tropics, and June–August in north- theories and modeling studies (e.g., Horinouchi et al.,
ern sub-tropics. These are periods when summer 2002). However, the patterns undergo significant
monsoonwindsenhancelocalprecipitationsignificantly changes above the stratopause at 61 and 80km, with
in the southern and northern subtropics, respectively, theemergenceofhighlatitudewavepatterns.Themuch
yielding ‘‘wet seasons’’. Wind contours in Fig. 3 also broader vertical width of the channel 7 ((cid:1)80km)
show that these subtropical variance enhancements oc- weighting function (Wu and Waters, 1997) may also
cur where mean winds are largest. Studies by Jiang play a significant role in modifying geographical vari-
et al. (2004b) found that most convectively-generated ance distributions.
GWscanbeseenbyMLSinthestratosphereonlywhen
the background winds are westward, and exceeding 2.3. Southern high-latitudes
(cid:1)20m/s(seeFig.8ofJiangetal.(2004b)).Thiscriterion
is met only in the summer subtropics. Another notice- SeasonalMLSnormalizedradiancevariancemapsat
ablephenomenonistheasymmetriesbetweenthesouth- (cid:1)33km in the extratropical southern hemisphere (Fig.
ern summer (top-left) and northern summer (lower-left) 6) reveal enhancements at high southern latitudes.
convectively-generated waves seen by MLS. Comparing EnhancementsstarttoappearinMay,peakinsouthern
variancesinFig.3totheCMAPindicesinFig.4,wesee winter (June–August) and continue into the southern
that the centers of enhanced variances are often shifted spring (September–November). The large variance
(cid:1)10(cid:2) in latitude poleward from of the convection cen- enhancement overthesoutherntip oftheAndesMoun-
ters. Jiang et al. (2004b) suggested this observed pole- tains was analyzed previously using combined MWFM
ward shift may result from the background wind modeling and MLS data analysis (Jiang et al., 2002).
filtering of the GW propagating into the stratosphere Othervariancespeaksalsooccur,suchasalongitudinal
andisalsolikelyrelatedtotheMLSGWvisibilityfunc- broad variance enhancement over the Southern Ocean
tion, since local winds are weaker over the convection that follows the polar vortex edge region where zonal
centers. However, further modeling effort is needed to winds are strong. Thus, to first order these enhance-
fully understand this phenomenon. ments are consistent with dilated GW vertical wave-
The vertical variations of the December–February lengths in strong winds which make all GWs here
and June–August summertime subtropical wave fields more visible to the MLS instrument. Some of the
J.H.Jiangetal./AdvancesinSpaceResearch35(2005)1925–1932 1929
Fig.5. Mapsofnormalizedradiancetemperaturevarianceat28,38,48,61and80kmaltitudes.ThedataDecember–February(left)arecomputed
usingMLSlimb-scanSDmeasurements;fortheJune–Augustmaps(right),MLSlimb-scanNAdataareused.Thedottedwhitecontoursbelow
80kmareUKMOwindsataboutsamealtitudesandaveragedoverMLSmeasurementdays.Noassimilatedwinddataareavailableat80km.
variance enhancements seem to related to underlying much of these MLS wave patterns occurs well away
orography, such as those over the Palmer Peninsula from predicted orographic GWs and thus many vari-
(300(cid:2)E,65(cid:2)S)andthecoastalTransantarcticMountains ance enhancements are likely nonorographic in origin.
near McMurdo Sound ((cid:1)165(cid:2)E, 70–75(cid:2)S). However, Forexample,intenseevolvingweatherphenomenasuch
preliminarymodelingusingMWFM(Fig.7)revealsthat as the strong baroclinic storm systems develop along
Fig.6. Seasonalmeanfieldsofnormalizedradiancetemperaturevarianceat33kmcomputedfromMLSlimb-scanSAmeasurements.Thedotted
whitecontoursareUKMOwindsataboutsamealtitudesandaveragedoverMLSmeasurementdays.
1930 J.H.Jiangetal./AdvancesinSpaceResearch35(2005)1925–1932
As altitude increases, these GW enhancements in-
crease in magnitude but maintain almost the same geo-
graphical distribution until (cid:1)53km (Fig. 8(a)), where
the waves along the Antarctic edge over the Southern
Ocean start to level off (Fig. 8(b)). The orographic
GW variances, such as those over the southern Andes
and New Zealand regions, however, seem to maintain
their strength well into the mesosphere at 80km.
Fig. 7. MWFM simulated 5mb orographic waves (as temperature 3. Other satellite measurements (2001-present)
perturbation)averagedoverMLSlimb-scandays(June–August1992–
1994). No observational filter was applied in this simulation. Maxi-
Asuiteofnewsatelliteinstrumentsrecentlylaunched
mumcontourvalueis5K.
into orbit are beginning to return a wealth of radiance
and/or temperature data containing GW information
withhighresolution,accuracyandglobalcoverage.This
offers researchers an unprecedented, long-term global
stratospheric and mesospheric gravity wave product.
Some examples of these new datasets are briefly intro-
duced below.
3.1. GPS radio occultation
GPS radio occultation technology is developing into
apowerfultoolforcontinuouslymonitoringtheEarth(cid:1)s
atmosphere.TheGPSconstellationconsistsofabout30
satellites that are distributed in circular orbits at
(cid:1)20,000km altitude. Recently launched German
CHAMP and Argentinean SAC-C low-orbit satellites
each carry a new-generation GPS receiver, the ‘‘Black-
Jack.’’ These receivers have been collecting occultation
datatoretrieveprofilesofatmosphericrefractivity,tem-
perature, pressure, and water vapor since mid-2001
Fig.8. June–August1992–1994meanMLSnormalizedlimb-scanSA
(Hajj et al., 2002). As demonstrated by Tsuda et al.
radiancevariancemapfor53km(a)and80km(b).
(2000) using earlier GPS/MET data, GPS occultation
temperatures can be used to extract gravity wave infor-
this ‘‘roaring forties’’ region oftheSouthern Ocean and mation. As an example, Fig. 9 displays the temperature
can radiate gravity waves from jet-stream instabilities variancesderivedfromtheGPS/CHAMPmeasurements
(e.g., O(cid:1)Sullivan and Dunkerton, 1995; Guest et al., (left) and water vapor pressure obtained from NCEP
2000). reanalyses (right). The temperature variance data are
Fig.9. MeantemperaturevariancesofgravitywavesderivedfromGPS/CHAMPmeasurements(left)averagedforJuly–September2002between15
and25kmheight.TherightpanelistheNCEPwatervaporpressureat(cid:1)10kmandinterpolatedfortimeandplaceofthegivenoccultationevent.
J.H.Jiangetal./AdvancesinSpaceResearch35(2005)1925–1932 1931
Fig. 10. Mean seasonal gravity wave variance maps of the southern hemisphere derived from AMSU-A measurementsfor December–February
2002–2003(left)andJune–August2003(right).Thealtitudeis(cid:1)37km.
averaged for the period of July–September 2002 within Thesemapshavemanysimilaritieswiththecorrespond-
the height interval of 15–25km. During that period ingUARSMLSseasonalmaps(see,e.g.,Figs.3and6),
the GPS receiver onboard the CHAMP satellite col- including the three convection centers in southern sum-
lected (cid:1)10,000 temperature profiles over the globe. mersub-tropics,theirpolewardshift,andthelargewave
TheGWactivitysuperimposeduponatemperaturepro- enhancements over the southern Andes and along the
file is estimated by averaging the small-wavelength var- Antarctic edge over the Southern Ocean. The AMSU
iance of the vertical temperature fluctuations between GWmeasurementsaremostlysensitivetowaveshaving
15 and 25km. The temperature fluctuations are ex- vertical wavelength >10km and horizontal wavelength
tractedfromthetemperatureprofilesbymeansofanon- of 50–150km (Wu et al., this issue).
recursive digital filter (Kaiser filter) having a band pass
of 10–15km. The variance values are arranged as a
function of geographic longitude and latitude of the 4. Concluding remarks
occultationevents,andcomputedwithaboxcaraverage
having a box-size of 15(cid:2) longitude by 10(cid:2) latitude and We havepresenteda summary ofseasonal global cli-
moving with a step of 1(cid:2). From this variance map (left matologies of middle atmospheric GWs derived from
panel) and the correlated NCEP water vapor reanalysis UARS MLS measurements and have demonstrated
field (right panel), one can see two regions where GW examplesofsomenewerGWdataderivedfrommorere-
variance and moisture correlate well: one around the cent GPS and AMSU observations. Recent GW mea-
Bangladesh region (possibly related to the Indian mon- surements from other satellite instruments can also be
soon in summer of 2002), and another in the equatorial foundinanumberofreviewpapers(e.g.,Wuetal.,this
zone over the Western Pacific. As a measure of deep issue; Fritts and Alexander, 2003). In the past, conven-
water vapor convection, the averaged water vapor pres- tional gravity wave observations have mostly been con-
sure around 10km is depicted in the right panel of Fig. ducted from ground-based sounders (e.g., lidar, radar),
9.TheseGWactiveregionshavecomeunderincreasing rockets, balloons,and aircraft. Although these observa-
scrutinyinrecentyearsasapossiblemajorplayerinthe tions have given us important information on gravity
water vapor budget near the tropopause and lower wave properties, they provide only a local picture. This
stratosphere (e.g., Jensen et al., 2001). A recent study complicates the analysis of GWs: for example, waves
by Wu et al., this issue suggests that due to the vertical canpropagatebothverticallyandhorizontallyoverlong
and horizontal resolutions, GPS GW measurements are horizontal distances, and only a small portion of the
mostly limited to waves of short vertical (<10km) and wave(cid:1)s full life cycle may be measured by a ground-
long horizontal (>200km) wavelengths. based vertical profiling instrument as the wave propa-
gates obliquely across the instrument(cid:1)s vertical field of
3.2. AMSU observation view. Thus, truly global measurements of GWs at a
range of stratospheric and mesospheric altitudes from
A variance analysis technique has recently been high-resolution satellite instruments are especially valu-
developedbyWu(2004)toextractgravitywaveinduced able for understanding the full three-dimensional pic-
temperaturefluctuationsfromtheAMSUradiancemea- ture, such as the intensity and geographical variability
surements on board the Aqua satellite. The algorithm of the tropospheric sources of GWs. Nonetheless, these
has produced reliable gravity wave variance maps with space-based observations require support by appropri-
veryhighresolutionof0.5(cid:2)by0.5(cid:2)longitudebylatitude ate global modeling of anticipated wave signatures
grid, enabling geolocation of small-scale regional wave which account for the observational filtering properties
features. Shown in Fig. 10 areAMSU-A seasonal mean of the GW detection, to permit effective interpretation
radiance variance maps for December–February 2002– of the measurements (e.g., Jiang et al. (2004a)). Such
2003 (left) and June–August 2003 (right) at (cid:1)37km. data-model studies will provide and validate the most
1932 J.H.Jiangetal./AdvancesinSpaceResearch35(2005)1925–1932
complete long-term global data on gravity waves in the Jiang, J.H., Wu, D.L., Eckermann, S.D., Ma, J. Mountain waves in
middleatmosphere,andhelpresolvegapsinourknowl- themiddleatmosphere:microwavelimbsounderobservationsand
analyses.Adv.SpaceRes.32,801–806,2003.
edgeaboutstratosphericandmesosphericprocessesthat
Jiang,J.H.,Wu,D.L.,Eckermann,S.D.UpperAtmosphereResearch
are crucial to the advancement of atmospheric science.
Satellite (UARS) MLS observation of mountain waves over the
Andes.J.Geophys.Res.107,D22,2002.
Jiang, J.H., Eckermann, S.D., Wu, D.L., Ma, J., A search for
Acknowledgments mountain waves in MLS stratospheric limb radiances from the
winter Northern hemisphere: data analysis and global mountain
wavemodeling,J.Geophys.Res.109(D3),D03107,2004a.
The first author (J.H.J.) like to thank Prof. T.
Jiang,J.H,Wang,B.,Goya,K.,Hocke,K.,Eckermann,S.D.,Ma,J.,
NakamuraandDr.M.J.Taylorforinvitationtopresent Wu, D.L., Read, W.G. Geographical distribution and inter-
thispaperat35thCOSPARScientificAssemblyinParis, seasonal variability of tropical deep convection: UARS MLS
and for valuable comments and suggestions. This study observations and analyses. J. Geophys. Res. 109 (D3), D03111,
2004b.
was supported by the Jet Propulsion Laboratory, Cali-
Kim,Y.-J.,Eckermann,S.D.,Chun,H.-Y.Anoverviewofthepast,
fornia Institute of Technology, under contract with
present and future of gravity-wave drag parametrization for
NASA. S.D.E.(cid:1)s work was partially supported by The numerical climate and weather prediction models. Atmos. Ocean
OfficeofNavalResearchandbyNASA(cid:1)sGeospaceSci- 41,65–98,2003.
encesProgram.Theauthorsalsoappreciatethesupport McLandress,C.,Alexander,M.J.,Wu,D.L.Microwavelimbsounder
observations of gravity waves in the stratosphere: a climatology
of the University of Hawaii of USA, and the National
andinterpretation.J.Geophys.Res.105,11947–11967,2000.
Institute of Information and Communications Technol-
O(cid:1)Sullivan,D.,Dunkerton,T.J.Generationofinertia-gravitywavesin
ogy of Japan. a simulated life cycle of baroclinic instability. J. Atmos. Sci. 52,
3695–3716,1995.
Pierce,R.B.et al.Large-scalechemicalevolutionoftheArcticvortex
during the 1999/2000 winter: HALOE/POAM III Lagrangian
References
photochemical modeling for the SAGE III–Ozone Loss and
Validation Experiment (SOLVE) campaign. J. Geophys. Res.
Fritts,D.C.,Alexander,M.J.Gravitywavedynamicsandeffectsinthe 107,8317,2002.
middleatmosphere.Rev.Geophys.41(1),1003,2003. Tsuda, Nishida, T.M, Rocken, C. A global morphology of gravity
Guest,F.M.,Reeder,M.J.,Marks,C.J.,Karoly,D.J.Inertia-gravity waveactivityinthestratosphererevealedbytheGPSoccultation
wavesobservedinthelowerstratosphereoverMacquarieIsland.J. data(GPS/MET).J.Geophys.Res.105,7257–7274,2000.
Atmos.Sci.57,737–752,2000. Wu, D.L., Waters, J.W. Gravity-wave-scale temperature fluctuations
Hajj,G.A.,Kursinski,E.R.,Romans,L.J.,Bertiger,W.I.,Leroy,S.S. seenbytheUARSMLS.Geophys.Res.Lett.23,3289–3292,1996.
AtechnicaldescriptionofatmosphericsoundingbyGPSocculta- Wu,D.L.,Waters,J.W.ObservationsofgravitywaveswiththeUARS
tion.J.Atmos.Solar-Terr.Phys.64,451–469,2002. Microwave Limb Sounder, Gravity Wave Processes. NATO ASI
Horinouchi, T., Nakamura, T., Kosaka, J. Convectively generated SeriesI:GlobalEnvironmentChange50,103–120,1997.
mesoscale gravity waves simulated throughout the middle atmo- Wu, D.L. Mesoscale gravity wave variances from AMSU-A Radi-
sphere.Geophys.Res.Lett.29,2007,2002. ances.Geophys.Res.Lett.28(No.12),2004.
Huffman,G.J.,Adler,R.F.,Arkin,P.,Chang,A.,Ferraro,R.,Gruber, Wu,D.L.,Preusse,P.,Eckermann,S.D.,Jiang,J.H.,DelaT.Juarez,
A.,Janowiak,J.,McNab,A.,Rudolf,B.,Schneider,U.TheGlobal M.,Coy,L.,Wang,D.Y.,Remotesoundingofatmosphericgravity
PrecipitationClimatologyProject(GPCP)combinedprecipitation waves with satellite limb and nadir techniques, Adv. Space Res.,
dataset.Bull.Am.Meteor.Soc.78(1),5–20,1997. thisissue.
Jensen,E.J.,Pfister,L.,Ackerman,A.S.,Tabazadeh,A.,Toon,O.B.A Xie, P., Arkin, P.A. Global precipitation: A 17-year monthly
conceptualmodelofthedehydrationofairduetofreeze-dryingby analysis based on gauge observations, satellite estimates, and
thin,laminarcirrusrisingslowlyacrossthetropicaltropopause.J. numerical model outputs. Bull. Am. Meteor. Soc. 78 (11), 2539–
Geophys.Res.106,17,2001,237–17,252. 2558, 1997.