HAL Id: hal-01347199
https://hal.sorbonne-universite.fr/hal-01347199
Submitted on 20 Jul 2016
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Distributed under a Creative Commons Attribution| 4.0 International License
Mid-infrared mapping of Jupiter’s temperatures, aerosol
opacity and chemical distributions with IRTF/TEXES
Leigh N. Fletcher, T.K. Greathouse, G.S. Orton, J.A. Sinclair, R.S. Giles,
P.G.J. Irwin, T. Encrenaz
To cite this version:
ContentslistsavailableatScienceDirect
Icarus
journalhomepage:www.elsevier.com/locate/icarus
Mid-infrared
mapping
of
Jupiter’s
temperatures,
aerosol
opacity
and
chemical
distributions
with
IRTF/TEXES
Leigh
N.
Fletcher
a,∗,
T.K.
Greathouse
b,
G.S.
Orton
c,
J.A.
Sinclair
c,
R.S.
Giles
d,
P.G.J.
Irwin
d,
T.
Encrenaz
ea Department of Physics & Astronomy, University of Leicester, University Road, Leicester LE1 7RH, UK b Southwest Research Institute, Division 15, 6220 Culebra Road, San Antonio, TX 78228, USA
c Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91109, USA
d Atmospheric, Oceanic & Planetary Physics, Department of Physics, University of Oxford, Clarendon Laboratory, Parks Road, Oxford OX1 3PU, UK e LESIA, Observatoire de Paris, CNRS, UPMC, Univ. Paris Diderot, 92195 Meudon, France
a
r
t
i
c
l
e
i
n
f
o
Article history: Received 21 March 2016 Revised 3 June 2016 Accepted 13 June 2016 Available online 18 June 2016 Keywords: Jupiter Atmospheres composition Atmospheres dynamics
a
b
s
t
r
a
c
t
GlobalmapsofJupiter’satmospherictemperatures,gaseouscompositionandaerosolopacityarederived fromaprogrammeof5–20μmmid-infraredspectroscopicobservationsusingtheTexasEchelonCross EchelleSpectrograph(TEXES)onNASA’sInfraredTelescopeFacility (IRTF).Image cubesfromDecember 2014ineightspectralchannels,withspectralresolutionsofR∼2000−12,000andspatialresolutionsof 2–4° latitude,areinvertedtogenerate3Dmapsoftroposphericandstratospherictemperatures,2Dmaps ofuppertroposphericaerosols,phosphineandammonia,and2Dmapsofstratosphericethaneand acety-lene.Theresultsarecomparedtoare-analysisofCassiniCompositeInfraredSpectrometer(CIRS) obser-vationsacquiredduringCassini’sclosestapproachtoJupiterinDecember2000,demonstratingthatthis newarchiveofground-basedmappingspectroscopycanmatchandsurpassthequalityofprevious inves-tigations,andwillpermitfuturestudiesofJupiter’sevolvingatmosphere.Thevisibilityofcoolzonesand warmbeltsvariesfromchanneltochannel,suggestingcomplexverticalvariationsfromthe radiatively-controlleduppertropospheretotheconvectivemid-troposphere.Weidentifymid-infraredsignaturesof Jupiter’s5-μmhotspotsviasimultaneousM,NandQ-bandobservations,whichareinterpretedas tem-peratureandammoniavariationsinthenorthernEquatorialZoneandontheedgeoftheNorthEquatorial Belt(NEB).EquatorialplumesenrichedinNH3gasarelocatedsouth-eastofNH3-desiccated‘hotspots’on
theedgeoftheNEB.Comparisonofthehotspotlocationsinseveralchannelsacrossthe5–20μmrange indicatethattheseanomalousregions tiltwestwardwithaltitude.Aerosolsand PH3 arebothenriched
attheequatorbutarenot co-locatedwiththeNH3 plumes.Theequatorialtemperatureminimumand
PH3/aerosolmaximahavevariedinamplitudeovertime,possiblyasaresultofperiodicequatorial
bright-eningsandthefreshupdraftsofdisequilibriummaterial.Temperatemid-latitudesdisplayacorrelation betweenmid-IRaerosolopacityandthewhitealbedofeaturesinvisiblelight(i.e.,zones).Wefind hemi-sphericasymmetriesinthedistributionoftroposphericPH3,stratospherichydrocarbonsandthe2Dwind
field(estimatedviathethermal-windequation)thatsuggestadifferingefficiencyofmechanicalforcing (e.g.,verticalmixingandwavepropagation)betweenthetwohemispheresthatweargueisdrivenby dynamicsratherthanJupiter’ssmallseasonalcycle.Jupiter’sstratosphereisnotablywarmeratnorthern mid-latitudesthaninthesouthinboth2000and2014,althoughthelattercanbelargelyattributedto strongthermalwaveactivitynear30°Nthatdominatesthe2014stratosphericmapsandmaybe responsi-bleforelevatedC2H2inthenorthernhemisphere.Avertically-variablepatternoftemperatureand
wind-shearminimaandmaximaassociatedwithJupiter’sQuasiQuadrennialOscillation(QQO)isobservedat theequatorinbothdatasets,althoughthecontrastsweremoresubduedin2014.Large-scale equator-to-polegradientsinethaneandacetylenearesuperimposedontopofthemid-latitudemechanically-driven maxima,withC2H2decreasingfromequatortopoleandC2H6showingapolarenhancement,consistent
witharadiatively-controlledcirculationfromlowtohighlatitudes.Coldpolarvorticesbeyond∼60° lat-itudecanbeidentifiedintheuppertroposphericandlowerstratospherictemperaturemaps,suggesting
∗ Corresponding author.
E-mail address: leigh.fletcher@leicester.ac.uk
(L.N. Fletcher).
http://dx.doi.org/10.1016/j.icarus.2016.06.008
enhancedradiative cooling frompolaraerosols. Finally,compositional mapping oftheGreat RedSpot confirmsthelocalenhancementsinPH3and aerosols,thenorth–southasymmetryinNH3 gasandthe
presenceofawarmsouthernperipherythathavebeennotedbypreviousauthors.
© 2016TheAuthors.PublishedbyElsevierInc. ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).
1. Introduction
Thermal infrared sounding of Jupiter provides a rich resource for investigation of the dynamical, chemical and cloud-forming processes shaping thethree-dimensional structureof theplanet’s atmosphere. The 5–25
μ
m region provides access to a host of spectral absorption and emission features, superimposed onto a continuumofhydrogen-heliumemissionandaerosolopacity,from which we can determine the horizontal and vertical distribu-tions of temperature, composition and aerosol structures from the churning cloud tops to the overlying stratosphere. Spatially-resolved thermal mapping from Voyager, Galileo and Cassini al-lowedustoexploretheconnectionbetweenthedynamicactivity observedinthecloud-formingregionandtherelativelyunexplored circulation and chemistry of the middle atmosphere (upper tro-posphere and stratosphere).However, instrumentsto exploit this spectralrangeareabsentfromfuturemissionstoJupiter,including theupcoming Junospacecraft.Inthisstudywereport ona regu-larprogrammeofspectroscopicmappingobservationsfromNASA’s InfraredTelescopeFacility(IRTF),aimingtomatchandsurpassthe capabilitiesofpreviousspacecraftthermal-IRobservationsto pro-vide anewdatabaseforinvestigatorsstudyingjovianclimate, dy-namics andchemistry. Ouraimisto bridgetheobservationalgap inIRspectroscopybetweentheCassiniandJunoepochs(2000and 2016,respectively).Multi-wavelength imagingin narrow-band filters covering the 5–25
μ
m spectral range (including those from the Galileo photopolarimeter-radiometer instrument, Ortonet al., 1996) have proven highly effectivein constrainingatmospheric temperatures atdiscretepressurelevels,anddataamassedoverseveraldecades haverevealed:(i) tropicalvariabilityassociatedwithstratospheric wind and temperature oscillations (analogous to Earth’s quasi-biennial oscillation, Orton et al., 1991; Leovy et al., 1991; Or-ton et al., 1994; Friedson, 1999; Simon-Miller et al., 2006b); (ii) belt/zone variabilitycausedby the lifecycle ofjovian‘global up-heavals’ (Rogers, 1995), particularly the fade and revival cycleof theSouthEquatorial Belt(SEB, Fletcheretal., 2011b);(iii)a char-acterisationoftheGalileoprobeentrysiteasaregionofuniquely powerfulatmosphericsubsidenceanddesiccation(Friedson,2005; Ortiz et al., 1998; Orton et al., 1998); (iv) understanding of the thermal aftermath of large impact events (e.g., Harrington et al., 2004, andreferences therein); and(v) the thermal structure and variability of Jupiter’s large anticyclones like the Great Red Spot (Fletcheretal., 2010b).Despitethesesuccesses, temperatures de-rived fromthermalimaging observationsare subjectto large de-generacieswithchemicalcomposition andcloud opacity(Fletcher etal., 2009b), renderingthequantitative results highlyuncertain. Spatial mapping of tropospheric and stratospheric gases, in par-ticular, requires us to spectrally resolve the forest of absorption andemissionfeaturestoderiveabundances.Itisthisdeficiencyin spectroscopythatourIRTFspectralprogrammeseekstoaddress.Spatially resolved spectralmapsofJupiter havebeenprovided by Voyager/IRIS (Infrared Radiometer and Spectrometer, Hanel et al., 1977) and Cassini/CIRS (Composite Infrared Spectrometer,
Flasar et al., 2004a), butthese were limited to snapshots during brief flybys, so they failed to explore the temporal variability of the thermal emission. Voyager-1 and-2 spectra (March andJuly
1979,respectively)havebeenpresentedaszonallyaveraged spec-tra forinterpretation (Carlson et al., 1992; Conrathand Gierasch, 1984;Conrathetal.,1998;ConrathandPirraglia,1983;Flasaretal., 1981; Griffith et al., 1992; Sada et al., 1996; Simon-Miller etal., 2000),althoughsparselongitudinally-resolvedcoveragewas avail-ableandhasbeenusedtoinvestigatetheGreatRedSpot(Griffith et al., 1992; Read et al., 2006a; Sada et al., 1996; Simon-Miller et al., 2002) and the spatialdistribution of water ice signatures (Simon-Milleretal.,2000).Onlythespectral mapsofCassini/CIRS betweenDecember2000–January2001canclaimtohaveprovided near-globalcoveragebysweepingitsdetectorsfromnorthtosouth to generate multiple maps over approximately two weeks. The Cassini datasets have provided us with tropospheric and strato-spheric temperature maps (Flasar et al., 2004a; Li et al., 2006), distributions ofthe disequilibrium speciesphosphine (PH3, Irwin etal.,2004; Fletcheretal.,2009a; 2010b),distributionsof ammo-nia,thekeycondensibleinJupiter’suppertroposphere(Achterberg etal.,2006),cloud opacity(Fletcheretal.,2009a;Matchevaetal., 2005; Wong etal., 2004) andstratospheric hydrocarbons (Kunde etal.,2004;Nixonetal.,2007;2010;Zhangetal.,2013a). Tempo-ralvariabilityoftemperatureswasobservedduringthiscloseflyby (Flasaretal.,2004a;Lietal.,2006),butnoorbitalmissionhasever providedalong-termdatabasetostudythisfourthdimension.
Thebesthopeforcharacterisationofthevariabilityofthe ther-mal and chemical environment is therefore ground-based spec-troscopy, albeit limited to regions free of terrestrial contamina-tion(the M,NandQbandsnear5,10 and20
μ
m, respectively). Ground-based spectroscopy permits the high spectral resolutions requiredto resolve spectral line shapes. However, previous stud-ieshavefocussedondiscrete regionssothatspatio-spectral map-pingisrareandglobalcoveragehasnotbeenpreviouslypublished (Fastetal.,2011;Fletcheretal.,2011a;Kostiuketal.,1987; Liven-goodetal.,1993;Sadaetal.,1998).ObservationsfromtheIRSHELL spectrometer on the IRTF (achieving spectral resolutions of R ≈ 10, 000, Lacy et al., 1989) were employed to map Jupiter’s tem-peratures,cloudsanddistributions ofphosphineandammonia in the 10− 36◦ S domain (Lara et al., 1998) in 1991, although onlyzonal-mean cross-sections are shown. IRSHELL was subsequently used in 1994 to map emission surrounding the Shoemaker-Levy 9impact sites (Bezard etal., 1997; Griffith etal., 1997). IRSHELL wasretiredin1994asasuccessor,TEXES(theTexasEchelonCross Echelle Spectrograph, Lacy et al., 2002) was developedas a visi-torinstrumentfortheIRTF. TEXEShasbeen previouslyemployed totracethefateofHCNandH2OrelatedtotheShoemaker-Levy9
Fig. 1. Synthetic spectrum of Jupiter calculated at 1 cm −1 spectral resolution compared to the Earth’s transmission spectrum from ATRAN ( Lord, 1992 ) (right hand axis, red line, calculated for Mauna Kea using an airmass of unity and precipitable water vapour column of 3 mm). The nine TEXES channels are shown as grey vertical bars, selected for their high telluric transmission and sensitivity to key atmospheric gases. The M, N and Q bands are labelled. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
2. Data
The TEXES instrument (Texas Echelon Cross Echelle Spectro-graph, Lacy etal.,2002) isa cross-dispersedgratingspectrograph abletorecordspatially-resolvedspectrathroughouttheM(5
μ
m), N (7–13μ
m) andQ (17–24μ
m) bands. Fig. 1 compares a syn-theticspectrumofJupitertotheEarth’stransmissionwindows:the Qbandisshapedbythecollision-inducedabsorptionofH2andHefromwhich we can determine upper tropospheric temperatures; the N-band features broad absorption features of ammonia and phosphine,plus emissionfeatures of methane(aprobe of strato-spherictemperatures),ethaneandacetylene(productsofmethane photolysis); and the M band senses thermal emission from the mid-troposphereattenuatedbyoverlying clouds,hazes,PH3,NH3,
CH3Dand other minorspecies. We focus onthe N andQ bands
inthisstudy,withinitialresultsfromtheMbandtobepresented elsewhere(Encrenazetal.,2016).
Given that our primary targets are lines formed in the upper troposphere and lower stratosphere, pressure broadening domi-nates andthe maximum TEXES spectral resolution (R ≈ 80, 000 in cross-dispersed mode) is not required. Our programme uses medium(R∼15,000)andlow(R∼2000)spectralresolutions, em-ployingthe1.4× 45-arcsecond slitto covertheentirejoviandisc ataseriesofdistinctwavelengthsettings,asdescribedbelow.The lower spectral resolutions bypass theechelon grating andsimply usetheechelleorfirst-ordergrating asthedisperser(Lacyetal., 2002),allowing thefull slit length tobe imaged ontothe 256× 256SiAsdetectorarray.
TEXES was used in ‘scan mode,’ whereby we alignedthe slit along the celestial north-south and stepped from west to east acrosstheplanetin0.7” increments(Nyquistsamplingthe1.4” slit width),with2-sintegrationsateachstep.Wedidnotaligndirectly withJupiter’scentralmeridiansothatanyrow/column defectson thedetectorcouldbereadilydistinguishedfromthebanded struc-tureof Jupiter. These scansstarted andfinished on blankskyto permitbackgroundsubtractionfromtheon-sourcemeasurements. Scansinaparticularsettingwererepeated2–4timesinquick suc-cessiontobuildupsignal-to-noiseandminimisetherisksofdata lossdue tocosmic rays or detectordefects, before moving onto thenext spectralsetting.Unlikemid-infraredimaging, we donot usethechoppingsecondary,usingtheoff-targetscanstepsinstead ofnoddedpairstoremovethebackground.GiventhatJupiter’s 10-hrotationislongerthanitsvisibilityfromtheIRTF,werequested groups of 2–3 consecutive nights in order to cover as many jo-vianlongitudes aspossible ina short spaceof time, buildingup
near-completemaps ofthe planet.Typically 5–10individual scan maps were obtained for each of the nine spectral settings used in this study. This combination of the long TEXES slit, efficient scan-mapping and calibration routines developed by the TEXES instrumentscientists, andcarefully-selected spectral settings per-mits the globaltemperature andcomposition mappingdescribed in Section5.
Thisglobalmappingprogrammehas,todate,providedmapsin February 2013, October andDecember 2014, March and Novem-ber 2015,andJanuaryandApril 2016.Eachobserving runused a standardsetofTEXESsettingsatlowandmediumspectral resolu-tions,detailedin Table1.Settingswerechosenbasedontheir sen-sitivitytotemperaturesina particularaltituderangeorthe pres-enceofabsorption/emissionfeatures inrelatively clearregions of thetelluric transmission spectrum.Twochannelsofthe February 2013datasetsensingtropospheric NH3 were previouslypublished
by Fletcher etal. (2014).In thisstudy we usethe full December 2014 datasetof nine channels,detailed in Tables A.3 and A.4 in
AppendixA,acquiredovertwonights(December8thand9th),as an excellent example ofthe quality of themaps that can be de-rived fromTEXES data.A suiteof9 channelstook approximately 70minto acquire,andwascycledrepeatedly overapproximately 6–8h. December 8thfocussedon 0− 180◦ W andDecember9th
focussedon180− 360◦ W.
2.1. TEXESdataprocessing
2.1.1. Radiometricandwavelengthcalibration
Target spectra were radiometrically calibrated and flat-fielded usingtwo observationsoftheskyemission andtwo observations ofaroom-temperatureblack body(ahigh-emissivitymetal chop-perbladejustabovetheentrancetotheDewar,temperatureTblack ), observedimmediatelypriortoeach scan.Ifweassume thatTblack is approximatelyequal to thesky (Tsky) and telescope(Ttel)
tem-peratures,thenthedifferencebetweentheblackbodyandthesky observationscan beused asthe flatfield toaccount forboththe telluric and instrument emission. The calibrated target intensity Iν(target)isthereforegivenby Lacyetal.(2002):
Iν
(
target)
=Sν(
target-sky)
Bν(
Ttel)
Sν
(
black-sky)
(1)whereSν(target-sky)isthe measuredflux differencebetweenthe target and the sky, Sν(black-sky) is the measured flux difference betweentheblackandthesky,andBν(Ttel)istheblackbodyflux
Table 1
Nine spectral channels considered in this study, showing the spectral resolution, coverage, diffraction-limited spatial resolution and key spectral features.
Central wavenumber (cm −1 ) Resolving power Coverage (cm −1 ) Resolution (cm −1 ) Diffraction limit Key features/objectives
538 7907 537–541 0.068 1.56” H 2 –He tropospheric T 586 5836 584–589 0.101 1.43” H 2 –He tropospheric T 744 10,292 742–747 0.072 1.12” C 2 H 2 819 7724 815–823 0.106 1.02” C 2 H 6 901 2896 885–915 0.311 0.93” NH 3 960 2664 945–975 0.360 0.87” NH 3 & PH 3 1161 2157 1138–1170 0.538 0.72” PH 3 , CH 3 D and Aerosols 1248 12,358 1243–1252 0.101 0.67” CH 4 stratospheric T
2137 12,366 2131–2142 0.173 0.39” Deep cloud opacity
fills the TEXES field of view, we need not account for the FOV-filling corrections that are typically required if standard divisors (e.g.,mid-IRbrightstarsorasteroids)areused,providingahighly efficientcalibrationschemethathasbeenfound tomatchthe ac-curacy ofmorestandard absolutecalibrationtechniques.This sky subtractioncannot remove thetelluricabsorption completely, do-ingabetterjobwithgasesinEarth’swarmertroposphere(e.g.,CO2
andH2O) thanthose inthe coldandhighstratosphere(e.g., O3).
Variable water vapour and clouds (especially thin cirrus clouds) between each step of the scan are partially accounted forusing the small portions of skyavailable at the ends of the slit away fromthetarget.However,asweshallseein Section3,the calibra-tion becomeslessaccurate inregions where Tsky andTblack differ substantially (i.e.,wheretheskyemission islow),andwherethe TEXESsystemresponse(Fig.5of Lacyetal.,2002)becomessmall. Giventhehighsensitivityofspectralinversionstothisradiometric accuracy, westill requirecross-calibrationwithspace-based mea-surementsforthepurposeofthisstudy.
The TEXES data reduction package (Lacy et al., 2002) per-forms the required sky subtraction, flat fielding and radiometric calibration, as well as corrections for optical distortions within the instrument and the removal of dead pixels on the detector. The measured sky scans were correlated with a model for the Earth’stransmissionspectrumtoassignwavelengthstoeachpixel, although thistoorequired finetuning prior tospectral inversion. A custom-designed IDL pipeline was created to assign latitudes, longitudes, Doppler shifts and emission angles (observing zenith angles)toeach pixelusingavisualfittothelocationofthe plan-etarylimb.Eachindividualscanmapwastheninterpolatedontoa regular1× 1° gridandradianceswereDopplershiftedbacktothe rest frame for subsequent analysis (i.e., removing redshifts from the dusk limb and blueshifts from the dawn limb). To improve further on the wavelength calibration in each spectral setting, a forward modelbased onCassini/CIRS determinations of tempera-tures,compositionandaerosolopacity (Fletcheretal.,2009a)was used toidentify spectral features.This wascomparedto a TEXES spectrum averaged within ±30° of latitude andlongitude of the sub-observer pointforevery individual scan map,andany differ-enceswereusedtoimprovetheaccuracyofthespectralcalibration via ashift-and-stretch method(Fletcher etal., 2014). Theaverage skytransmission(usingthesameDopplershiftastheJupiterdata) wasusedtoidentifycontaminatedregionsofeachspectrum.
2.1.2. Inter-cubevariability
The individual wavelength-corrected and absolutely-calibrated scan maps were combined into global maps, with the raw data foreachspectralsettingshownin Fig.2.Tocreatethesemaps,we averaged all data ata particular latitude/longitude witha zenith angle within 10° of the minimum (i.e.,as close to nadir as pos-sible foreach location).Although empirically corrected usingthe zenithangle,themapssometimesshowdiscontinuitiesinradiance asverticalstripes,duetothe mismatchofzenithangles between adjacent longitudes. Upon initial inspection, we discovered small
radianceoffsets fromcube to cubein a particularsetting, poten-tially correlated with changes to the sky background during the observing run. Variablewater humidity or cirrus cloud over the course of the two nights would change the effectiveness of the absolutecalibration,andproduced starksteps inthe absolute ra-diance in theglobal maps ofthe order 5–15% depending on the specificsetting.Whilstthislevelofvariabilityiswithinthe conser-vative20% uncertainty envelopeusually quoted forcalibration of ground-baseddata,itisinsufficientlyaccuratetopermit spatially-resolvedretrievals.
We therefore extracted averaged radiances from within ±10° longitudeofthecentralmeridianforeachscanmap,averagedover thespectralchannel,andnormalisedthemalltothemedianvalue withinaspecificlatituderange.Wechose latituderangesthatare relativelyunaffectedby Jupiter’sintrinsiclongitudinalvariability -equatorialregionsforstratosphericchannels(CH4,C2H2andC2H6)
andlimb-darkenedhighlatitudesfortroposphericchannels.These corrected centralmeridian radiances are shown foreach channel in Fig.3,comparedtoCassini/CIRSzonally-averagedradiances, av-eragedoverthesamespectralrangeaseachTEXESchannel.Thisis notaquantitativelyaccuratecomparison,giventhatCIRSradiances havealowerspectral resolution(both0.5cm−1 and2.5cm−1 ob-servationsare shown)andare notaffected byterrestrial contam-ination.Nevertheless,they reveal that large-scale offsetsbetween theTEXES andCIRSabsolute calibrations are present, which will bedealtwithin Section3.
OneunfortunatefeatureoftheTEXESimagecubesisa‘column noise’due toblemisheson the filter,whichmanifests asa verti-cal stripeon theimages thatappears to havea lower brightness than the rest of the image. As Jupiter’s central meridian was at an angleto the detectorrows andcolumns (the slit wasaligned along the celestial north-south),this translatesto diagonal strip-inginthecylindricalmaps.SuchstripescanbeseenintheQ-band images(Fig. 2f,g)butarealso presentat819cm−1 (Fig. 2i).The combinationoftheblemishes,Jupiter’sintensebrightnessatthese wavelengths, and the relative clarity of the telluric atmosphere (i.e.,very littleskyflux)means that we haveno directmeans to removethem fromthe datavia flatfielding.This addsadditional uncertainty to the retrieved products which will be assessed in
Section3.
2.1.3. Spatialresolution
The highest spatial resolution of the Cassini/CIRS maps of Jupiter was 2700 km from 137 RJ using its 273 × 273
μ
rad detectors, equivalent to 2.2° at Jupiter’s equator. The TEXES ob-servations occurred when Jupiter was at a distance of 4.83 AU (7.2 × 108 km), so that the spatial resolution varies betweenFig. 3. Central-meridian averaged radiances extracted from each TEXES spectral cube and averaged over the wavelength range of the setting (where the transmission exceeds 80% of the clearest region in the setting). Each cube has been normalised to a specific latitude range as described in the main text, and the colours correspond to the central meridian longitudes shown in the inset key. These are compared to Cassini/CIRS spectra averaged over the same spectral range for (a) 0.5 cm −1 data averaged from November 15th 20 0 0 to February 15th 20 01 with a low spatial resolution (black dotted); and (b) 2.5 cm −1 data acquired at closest approach on December 31st 20 0 0 (red dotted). This provides an indication of the spatial resolution of the TEXES data when compared to CIRS, and also highlights systematic differences between CIRS and TEXES that will be explored in Section 3 . Note that CIRS spectra do not cover the 538 cm −1 and 2137 cm −1 channels, so are not shown in panels (a) and (i). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
datasetsisthereforecomparableatwavelengthsbelow10
μ
m.For the remainder ofthispaper we exploreboth zonal-mean spectra andspatiallyresolvedspectrafrombothTEXESandCIRS.2.2. Inspectionofimages
Before proceeding with inversion of the TEXES, we describe someofthefeaturesrevealedin Fig.2incomparisonwitha mon-tage of visible light images (Fig. 2a), kindly provided by M. Ve-dovatobasedonobservationsbyamateurobservers.Imagesofthe 0− 180◦Wwereacquiredapproximately10h(onerotation)before
theTEXESmaps,whereasimagesof180− 360◦ Wcoincided with
the TEXES maps.It is importantto note that the TEXES maps of aparticularregionwereallacquiredwithinapproximately70min ofoneanother(TablesA.3and A.4),solongitudinalmotionswould havebeennegligibleduringthisinterval.Thereaderisreferredto
TableB.5fornomenclatureforthebelt/zonestructureusedinthe textthatfollows.Alllatitudesinthisstudyareplanetographic.
Thedominantfeatures ofthemapsare thecool,cloudyzones andwarm,cloud-freebelts, punctuatedbydramaticwave activity andlargeanticyclonicvortices(theGreatRedSpotnear120°Wand OvalBAnear80°W).Thevisibilityofthewarmemissionfromthe beltsvaries asa function ofwavelength (and therefore altitude), withtroposphericbeltsintheNorthTemperateDomain (20− 50◦
N) being most prominent in NH3-sensitive channels (10.4, 11.1,
12.2
μ
m in Fig. 2d, e andi) and hard to distinguish in aerosol-sensitive channels(4.7 and 8.6μ
m, Fig. 2b,c). In particular, the warmbandat27°N(theNorthTemperateBelt(NTB), borderedby a prograde jet at 24°N and a retrograde jet at 31°N, Table B.5) that isvisiblein Fig.2d andedoesnot appear tohavea readily distinguishablecounterpart inthe visiblelight image -a thermal anomalypotentiallymaskedbyoverlyingaerosols.Fig. 4. Focusing on dynamic activity in the North Equatorial Belt (NEB) between 0 and 20 °N, covering all longitudes. Visible-light images show the ‘hotspots’ as features of extremely low albedo on the southern edge of the red-brown belt on the prograding jet at 7 °N. The visible-light images are the same as those in Fig. 2 : images between 0 − 180 ◦W we’re taken approximately 10 h prior to the TEXES images; the images between 180 − 360 ◦ W coincided with TEXES. The TEXES observations of a particular longitude were taken within 70 min of one another. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
the visible (known as ‘brown barges’). These barges are at the limit of the resolution of the IRTF, but can be seen as bright patches at 4.7, 8.7, 10.4 and11.1
μ
m (Fig. 2b–e), indicating that they are depleted in both ammonia and aerosols (Orton et al., 2015).Theycannotbeseenintheupper-tropospheresensitive fil-tersfrom13− 18μ
m,suggestingdeep-seatedfeatures. TheNH3-sensitivechannelsalsorevealupto threedistincttemperatebelts inthe southern hemisphere between 30− 50◦ S, the most
equa-torwardofwhichispartiallydisruptedby thepassageoftheGRS andOvalBA. A chain of anticyclonic whiteovals (AWOs) can be seeninthevisible-lightimageinthe SouthSouthTemperateBelt (SSTB),butareatthe limitofthe spatialresolution oftheTEXES observations - they can be seen as darker patches in the 10– 11
μ
mmaps.Thesesamefiltersreveal non-uniformitywithinthe equatorialzone,whereregions ofbrighteremission coincidewith visibly-darkalbedostructures,suggestingsmallgapsinthe other-wisethickreflectiveclouds.Fig.2i,j(12.2and8.0
μ
m) showthemostsensitivityto strato-spherictemperatures via emission fromethaneandmethane, re-spectively. The 8-μ
m map is unlike any other, showing banded structures(awarmequatorandcoolneighbouringlatitudes;warm mid-latitudebands) that have no counterpart in the deeper tro-posphere. The mid-latitude stratospheric bands exhibit dramatic waveactivity,particularlyinthenorthernhemisphereinthe180− 270◦ W region.Thisstratosphericwave impacts boththe temper-ature andcomposition of the mid-stratosphere, and will be dis-cussed in Section 5. Heating associated with the northern auro-ralovalisevident between180− 210◦ W (asobservedpreviouslyinground-based observations,e.g., Livengood etal., 1993;Kostiuk etal.,1993),althoughhigh-spectralresolutionTEXESobservations (Sinclairetal.,2015)are requiredtodeterminethevertical struc-tureofthisenergydeposition(fromacombinationofJouleheating inresponse to currentsflowing downwardsfrom thehomopause levelanddirectdepositionbyprecipitatingelectrons).There isno evidenceofheatingassociatedwiththesouthernaurora,butgiven thetimingoftheTEXESobservations(northernsummer)thismay beduetoa poorobservinggeometryforsouthern highlatitudes. Furthermore,thesouthernauroralovaloccursatahigherlatitude (∼75°S)thanthatinthenorth.
Besides the large-scale banded structures, the TEXES dataset alsoallowsustoprobetheverticalstructureofsmaller-scales.Two examplesareshownin Figs.4and 5,fortheNorthEquatorialBelt (NEB)andtheregionsurroundingtheGRSandOvalBA.The bright-nessvariationsintheNEBare relatedtoRossby waveactivityon
theprogradingjetat7°N.Visibly-darkstructuresinthevisible im-agesare associatedwithcloud-freeregionsat4.7
μ
m,wherethe dearth of aerosol opacity permits emission fromdeeper, warmer layers.These‘5-μ
mhotspots’areinfactvisiblethroughouttheM and N-bands, showingthat they are perturbing the temperature, aerosol andpossibly the composition field in the 400–600 mbar region. They are harder to observe in the Q-band, although this may simply be related to the lower spatialresolution. The most interesting featureof Fig.4 istheoffsetsobserved inthehotspot locations as a function of wavelength, primarily in the eastern hemisphere (observations acquiredonDecember 8th2014).FromTableA.3,we seethatTEXES scansatdifferentwavelengths were taken ina strict sequence, so that the same spatiallocations on the planet wouldhave beencovered withno more than 70min separationbetweenonewavelengthandthenext,anditwasoften muchfaster– forexample,imagesat901,960,1161and2137cm−1 focussedon 90°Wlongitude wereacquired within 30min.Could this represent a real tilt of the hotspots westward with height, fromthedeepestsensing2137-cm−1filter(Fig.4b) tothehighest sensing960cm−1channel(Fig.4f)?Thistiltisnotobserved every-where within theNEB, withhotspotsin the western hemisphere generallymoreco-alignedasafunctionofdepth,andwespeculate thatthiscouldbeduetothedifferencesinthethicknessofNH4SH
cloudsbetweentheeasternandwesternhotspots.Theoffsetwith respecttothevisiblelightobservationsnear90°W(Fig.4a)maybe atemporaloffsetduetotenhours separationbetweenthe TEXES andamateur images,duringwhichfeatures ontheNEBsjet could moveeastby∼3.4° longitude.Nevertheless,thereisacloser align-ment of the albedo patterns with the N-band observations than there is with the M-band observations, supporting the idea that theM-bandprobeslevelsbeneaththetop-mostclouddecks.
Fig. 5. TEXES maps focusing on the Great Red Spot and Oval BA, demonstrating the co-alignment of the images from 5–18 μm. The visible light images are there same as those in Fig. 2 .
in the200-400mbar region observed at586cm−1; anda warm and aerosol-free SEB (particularly associated with rifting activity in thenorthwestwakeof theGRS)contrasted againsta cold and cloudySouthTropical Zone(STropZ). Fig.5showsasuperior spa-tialresolution tothosemaps oftheGRSacquiredby Cassini/CIRS (see Fig.4 of Fletcher et al., 2010b). Recalling that each pixelin theseimages representsa fullTEXES spectrum ofeight channels,
Figs.4and 5highlightthecapabilityfortemperature,composition andaerosolsounding within thegiant vortices andother regions ofinterestonJupiter.
3. TEXESretrievalpipeline 3.1. Spectralmodelandinversion
Fig.6showstheeightTEXESNandQ-bandchannelsconsidered inthiswork,withkeyspectralfeatureslabelled.Thecorresponding vertical sensitivity is shownin Fig. 7. Zonal-mean TEXES spectra at901 and960 cm−1 were previously analysed by Fletcheret al. (2014)usingtheradiative-transferandspectral-retrievalalgorithm, NEMESIS(Irwinetal.,2008).Thisworkextendsthatprevious anal-ysistoincludeafurthersixspectralsettingsat538,586,744,819, 1161 and1247 cm−1 asshownin Table1, performing simultane-ous retrievalsfromall eightchannels.We developedtheretrieval pipeline one channel at a time, starting from the troposphere-sensing N-band channelsandsubsequently adding incapabilities for stratospherictemperatures (CH4), uppertropospheric
temper-atures (538 an 586 cm−1) and stratospheric composition (C2H2
and C2H6). The addition of M-band channels to sound the
mid-troposphere willbe thesubjectof futurework. Ateach stage we performedteststodetermine theretrievalsensitivityto theprior andtheimplicationsofaddingthenewchannels.
The forward model calculation uses the correlated-k method (Goody et al., 1989; Lacis and Oinas, 1991), which required the pre-tabulation of smooth k-distributions (ranking absorption co-efficients k according to their frequency distributions) based on a variety of sources ofspectral linedata (Table 2). Isotopologues for methane(12CH
4, CH3D and13CH4) andammonia (14NH3 and
15NH
3) were treated separately, but hydrocarbon isotopologues
were combined into single tables. The k-distributions for each
channel are pre-convolved withan instrumentfunction with the spectralresolutionsshownin Table1,calculateddirectlyfromthe gratingequation dependingon thegrating angleandthe angular sizeoftheTEXES slit.Thesedistributionsarethen combinedinto a single tabulationfor each of the species listed in Table 2. The TEXES instrument function is expected to be a convolution of a Gaussian anda Lorenztian,but testingofa variety ofinstrument functionsby Fletcheretal.(2014)showedthattheuseofasimple GaussianwassufficientforanalysisoftheTEXES dataatlowand moderateresolutions.Thesek-distributions(withdifferentspectral resolutions for each of the eight channels), combined withboth thecollision-induced absorptionin Table2andaerosolabsorption describedbelow,constitutetheforwardmodel.
The NEMESIS optimal estimation retrieval algorithm allows us to fit the TEXES spectra via a Levenburg-Marquardt iterative scheme, whilst using smooth a priori state vectors to ensure physically-realisticsolutions(see Rodgers,2000;Irwinetal.,2008, fora full discussion ofthistechnique). The apriori jovian atmo-sphere was specified on 120 levels from 10 bar to 1
μ
bar, us-ingreferenceprofilesoftemperature,ammonia,phosphine,ethane and acetylene from a low-latitude mean of Cassini/CIRS results (Fletcher etal.,2009a; Nixonetal.,2007). TheCIRS-derived tem-perature profile originally used the T(p) from the Galileo Atmo-spheric Structure Instrument (ASI, Seiff et al., 1998) as a prior. The deep helium andmethane mole fractions were set to 0.136 and1.81× 10−3,respectively,basedontheGalileoprobemeasure-ments of Niemann et al. (1998) that were used to constrain the photochemical model of Moses et al. (2005). Methane then de-creasedwithaltitudefollowingthediffusivephotochemicalmodel of Moses etal. (2005),which wasalso usedas theprior forthe C2H2 and C2H6 measurements of Nixon et al. (2007). Ethylene
(C2H4) is includedbased onthe photochemical modelof Romani (1996)asitmayhaveaminoreffectnear950cm−1.Weassumed isotopologueratiosofD/H(CH
4)=2× 10
−5(the valueinthe
proto-solarcloud, GeissandGloeckler,2003),aterrestrialratioof13C/12C
anda15N/14Nratioof2.3× 10−3 (Owenetal.,2001)thatwas
pre-viouslyconfirmedby modellingoftheTEXES 901-and960-cm−1 spectra(Fletcheretal.,2014).
Fig. 6. Two examples of zonally-averaged TEXES spectra at the equator (blue) and the North Equatorial Belt (black). The data are shown as circles with error bars, the best-fitting model is shown as the solid line of the same colour. The red line denotes the TEXES transmission spectrum (taking both the sky emission and losses within the telescope into account), indicating the presence of telluric contamination. Gaps in the spectrum show regions that were not used in the spectral inversions. Approximate locations of key features are indicated, but the features shown in panels f and g are blends of multiple lines of NH 3 , PH 3 and CH 3 D. Q-band spectra in panels a and b should be flat (as shown by the solid lines), so variations are artefacts related to the poor telluric transmission. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
temperature,ortherateofchangeofradiancewithrespecttothe T(p) profile) in Fig. 7to show how the vertical sensitivityof the TEXES data varies as a function of wavelength. Note that these were computed for a nadir geometry – the greater atmospheric path at higher zenith angles would cause these contribution functionstomovetohigheraltitudes.Thisfigureintroducessome ofthecomplexityofmodellingtheTEXESspectra.Firstly,the con-tributionfunctionsassociatedwiththefinehydrocarbonemissions areoftenmulti-lobed,withsensitivityinthe1–10mbarrangeand atailofsensitivityinthelinecoresprobingthe5–15
μ
barrange. Therelativeweightofthesetworegions isacomplexfunction of theverticaltemperatureandcomposition structure,with observa-tionsathigherspectralresolutionprovidingmoredatapoints(andhence more retrieval sensitivity) for the lowest pressures sensed in the line cores. Cassini/CIRS 2.5-cm−1 resolution spectra, by contrast,donotprovidesufficientsensitivitytoprobep<1mbar inthis nadirgeometry. Theethane linecores,forexample,sense a broadrangefrom0.1to 20mbar, making inferencesofvertical T(p) and composition gradients extremely degenerate with the limiteddataavailable.
(a) 538 cm-1 537.5 538.0 538.5 539.0 539.5 Wavenumber (cm-1) 1000 100 10 Pressure (mbar) (b) 586 cm-1 586 587 588 Wavenumber (cm-1) 1000 100 10 1 Pressure (mbar) (c) 744 cm-1 743 744 745 746 Wavenumber (cm-1) 1000.000 100.000 10.000 1.000 0.100 0.010 0.001 Pressure (mbar) (d) 819 cm-1 817 818 819 820 821 822 Wavenumber (cm-1 ) 1000.00 100.00 10.00 1.00 0.10 0.01 Pressure (mbar) (e) 901 cm-1 895 900 905 Wavenumber (cm-1 ) 1000 100 Pressure (mbar) (f) 965 cm-1 960 965 970 Wavenumber (cm-1 ) 1000 100 Pressure (mbar) (g) 1161 cm-1 1140 1145 1150 1155 1160 1165 1170 Wavenumber (cm-1) 1000 100 Pressure (mbar) (h) 1247 cm-1 1246 1247 1248 1249 1250 Wavenumber (cm-1) 1000.000 100.000 10.000 1.000 0.100 0.010 0.001 Pressure (mbar)
Fig. 7. Contribution functions (Jacobians of temperature) computed for the TEXES channels used in this study. The vertical axis changes from panel to panel, depending on how much of the stratosphere is being sampled. Grey contours are given in steps of 0.05 from zero (white) to one (black), normalised to the strongest contribution for this spectral channel.
and ∼200 mbar at586 cm−1. The cores of the NH3 lines probe
up towards the 150-300 mbar level in Fig. 7e, where tempera-tureconstraintmustcomefromtheQ-bandchannels(e.g., Fig.7b). The continuum between the C2H2 features at 744 cm−1 senses
the 400–700 mbar level, significantly overlapping the continuum in Fig.7(d–g).Anyinconsistenciesbetweenthesecontinuum radi-ances wouldresultin difficulties inselecting representative tem-peraturesforthesealtitudelevels,asweshallseebelow.
The informationcontent oftheTEXES spectra issuch that we deriveafullprofileofatmospherictemperature,butwe retrievea singlescalingfactorforthehydrocarbonsandaparameterised pro-file forNH3 andPH3 (aconstant molefractionup toa transition
pressure p0,above whichthe abundancedeclines dueto
conden-sationand/orphotolyticdestructionwithafractionalscaleheight, f), for reasons we discuss in Section 4. The abundance of NH3
andPH3 is forcedtozeroforaltitudes above thetropopause.We
alsoderive ascale factorfortheoptical depthofa singleaerosol
layer, modelled asa simple grey absorber at the 800-mbar level withacompactscaleheight0.2× thegasscaleheight(Achterberg etal., 2006; Fletcher et al., 2009a; Matcheva et al., 2005; Wong etal.,2004),andlatertesttheTEXESsensitivitytodifferentcloud parameterisations.Eachspectral inversiontherefore provides esti-matesofthe3Dthermalprofileand2D distributionsofPH3,NH3,
C2H6,C2H2 and∼800mbaraerosolopacity.
3.2.RadiometriccomparisontoCassini
Table 2
Sources of spectroscopic linedata (all isotopologues taken from the same sources). Exponents for temperature dependence T n given in the final column.
Gas Line Intensities Broadening Half Width Temperature Dependence T n
Collision-induced absorption
(CIA) additional H H 2 -H 2 opacities from 2 -He, H 2 -CH Orton et al. (2007) 4 and CH 4 -CH 4 , plus opacities from Borysow et al. (1988) , Borysow and Frommhold (1986) and Borysow and Frommhold (1987) , respectively.
– -
CH 4 , CH 3 D Brown et al. (2003) H 2 broadened using a half-width of 0.059 cm −1 atm −1 at 296 K
n = 0 . 44 ( Margolis, 1993 ) C 2 H 6 Vander Auwera et al. (2007) (also found in
GEISA 2009, Jacquinet-Husson et al., 2011 ) 0.11 cm
−1 atm −1 at 296 K ( Halsey
et al., 1988; Blass et al., 1987 , for H 2 and He, respectively)
n = 0 . 94 ( Halsey et al., 1988 )
C 2 H 2 GEISA 2003 ( Jacquinet-Husson et al., 2005 ) (unchanged in GEISA 2009 at 13.6 μm,
Jacquinet-Husson et al., 2011 )
Fits to data in Varanasi (1992) -
PH 3 Kleiner et al. (2003) Broadened by both H 2 and He
using γH2 = 0 . 1078 − 0 . 0014 J cm −1 atm −1 and
γHe = 0 . 0618 − 0 . 0012 J cm −1 atm −1 ( Bouanich et al., 2004; Levy et al., 1993 )
n = 0 . 702 − 0 . 01 J ( J is the rotational quantum number) ( Salem et al., 2004 )
NH 3 Kleiner et al. (2003) (also found in GEISA 2009,
Jacquinet-Husson et al., 2011 )
Empirical model of Brown and Peterson (1994)
Empirical model of Brown and Peterson (1994)
C 2 H 4 GEISA 2003 ( Jacquinet-Husson et al., 2005 ) Fits to data in Bouanich et al.
(20 03) ; 20 04 ) (Bezard, personal communication )
n = 0 . 73 ( Bouanich et al., 2004 )
H 2 Quad. HITRAN 2012 ( Rothman et al., 2013 ) 0.0017 cm −1 atm −1 ( Reuter and
Sirota, 1994 )
n = 0 . 75 ( Rothman et al., 2013 )
in the radiometric calibration. In their analysis of TEXES spectra ofSaturn’sstratosphericvortex, Fouchetetal.(2016)found TEXES-derivedstratospherictemperaturestobesystematicallycoolerthan thosederivedfromCIRS.Theyattributedthistothesignificant dif-ferenceinspatialresolutionfoundwhenconvolvingCassini’s high-spatial-resolutionthermalmapswithaseeing-limitedFWHMthat wasreasonable for the IRTF at the time of their measurements. However,astheTEXES andCIRSJupiterdatasetshavea compara-blespatialresolution,we cannot attribute theradiometric offsets observedin Fig.3tothesameeffect.
To assess the magnitude of the CIRS-TEXES discrepancy, we comparezonally-averagedTEXES spectratoCassini-based forward modelsateverylatitude.CIRSspectraat2.5-cm−1 spectral resolu-tionwereextractedfromthelatestcalibrationoftheCIRSdatabase (version4.2)fortheATMOS02Amap acquiredonDecember 31st, 2000. We replicated the work of Fletcher et al. (2009a), fitting temperatures,PH3,NH3andaerosolsasdescribedabove.Thiswas
extendedby simultaneously fittingscalefactors forthe hydrocar-bondistributions(C2H6 andC2H2).Theresultingzonal-mean
tem-perature, composition and aerosol opacity will be presented in
Section5,butthesewereusedtoforwardmodeltheTEXES chan-nels,usingtheobservinggeometry(latitudesandemissionangles) ofthe TEXES spectra themselves. Fig. 8 shows that the required radiancescalefactorforeachchannelislargelyconstantasa func-tion of latitude (the mean offsets and their standard deviations are also shown), withthe exception of the tropical region (NEB, EZ and SEB) that displays the largest atmospheric variability as a function of time, andthus thelargest differences betweenthe Cassini(2000)andTEXES(2014)observations. Fig.9comparesthe rawspectrabothwithandwithoutthemultiplicativescalingfactor applied.
The mean scale factors for the radiance in each channel are broadly consistent with the offsets observed in the zonal-mean spectralaverages in Fig.3,butare moreaccuratebecause weare abletocomparedataandmodelsatthesamespectral resolution. Onlythoseregions unaffectedby telluriccontaminationandaway fromnarrowjovianemissionfeatures(i.e.,thecontinuumbetween hydrocarbonlines) are used to estimate the scale factor, andwe
findthat systematicoffsets betweenTEXES andtheCIRSforward modelaredetectedforalmosteverychannel.Intriguingly,the ma-jorityoftheTEXES observationsoverpredict theflux andneedto be scaled downwards, whereas those at 1247 cm−1 need to be scaledup tomatchCIRS.We notethat previously-reportedTEXES observationsof Jupiter(Fletcheretal., 2014) andSaturn(Fouchet et al., 2016) have shown offsets in the same direction as those foundhere.
Giventhat Jupiter’s temperature andcomposition are intrinsi-cally variable,theseglobally-averagedscale factorsare highly un-certain– indeed,low-latitude1247-cm−1observationsareactually consistentwithCIRS-derivedtemperatures.Althoughthescale fac-tors maylooksizeable(in one caseup to 45%),it ismore mean-ingfultoviewtheseasbrightnesstemperatureoffsets– the equiv-alentchangeinblackbodytemperaturerequiredtoreproducethe difference,andanestimateoftheatmospherictemperatureatthe altitudeprobedbyaparticularchannel.WefindthatQ-band tem-peratures only need to be decreased by ∼3 K, 700–1200 cm−1 temperaturesneedtobedecreasedby∼3to∼8K,andthe 1248-cm−1spectrumneedstobeincreasedby∼3K(allvaluesgivenin
Fig. 8). Indeed, the largest discrepancybetween CIRS and TEXES occurs where the sky is at its most transparent (i.e., Tsky is at its smallest), at819 and901 cm−1,and we wouldexpect to de-rive global tropospheric temperatures some 5–8 K warmer from the TEXES data than fromthe CIRS data.The lack of skyflux at these wavelengths means that the TEXES flat (the difference be-tweenthereferenceblackbodycardandtheskyemission)is sub-ject to larger uncertaintiesin the most transparent regions. Fur-thermore, the TEXES system response is smallest in the regions showingthelargestoffsets(Fig.5of Lacyetal.,2002).Future ob-servationshavebeenscheduledtobettercharacterisethe system-atic offsets inTEXES Jupiter spectra over consecutivenights as a resourceforotherusersofTEXES.
Fig. 8. The latitudinal variation of the multiplicative scale factor (circles) that must be applied to the TEXES spectral radiances to match the forward-modelled radiances based on Cassini/CIRS atmospheric retrievals. The solid and dashed horizontal line show the mean scale factor and its standard deviation, respectively (also given as the topmost number in each panel). Overlapping with the dotted horizontal line would imply that no scaling is necessary, as is the case for the 744 cm −1 channel. The bottom number in each panel is the equivalent change in the temperature of a black body required to reproduce these radiance offsets.
mustaccountfortheseoffsetsinsubsequentmodelling.Notethat smaller-scale latitudinal differencesbetween CIRS andTEXES are real,andareinvestigatedin Section5.
3.3. Errorhandling
TEXESspectraareaffectedbysourcesofbothrandomand sys-tematic uncertainty, with the latter being the hardest to quan-tify. The inspection ofcentral-meridian radiances from individual TEXES cubes (Section 2) revealed a high level of precision from cubetocubeandnighttonight,withradiancesreproduciblefrom cubetocubeatthe5–15%level,althoughsomeofthiscan be at-tributedtoJupiter’sownintrinsiclongitudinalvariability.However, cross-comparisonwithCIRSobservations in Section 3.2suggest a radiometric accuracy that varies with wavelength by up to 50%. The implications for this accuracy on spectral inversions will be discussedin Section5.
Precision uncertainties on TEXES spectra were explored in
Fletcher et al. (2014), providingseveral approaches to estimating the measurement noise. The uncertainty in a particular spectral channel varieswith time (due to variable skyemission and sta-bilityduringanight) andwavelength(withlargervaluescloseto telluricfeatures).Foreachcubeconsideredinthisstudy,we calcu-latethestandarddeviationoftheradianceforeachwavelengthin 10× 10pixelsquaresfromthefourcornersofthearray(i.e.,away fromJupiter).Wethenaveragethisoverthewavelengthrangeand compare to the radiancein the centreof the cube (i.e.,the
cen-tre ofJupiter), andfinally averagethisover all cubesina partic-ular spectral setting. This allows us to estimate the background flux variation as follows: 1.1% at 538 cm−1, 1.5% at 586 cm−1, 3.8% at744 cm−1, 0.6% at 819 cm−1, 0.3% at901 cm−1, 1.2% at 960cm−1,0.9%at1161cm−1and8.2%at1248cm−1.Asexpected, thisstandarddeviationissmallestwheretheatmosphereismost transparent.
When TEXES spectra were zonally or spatially averaged, we compare these ‘background uncertainties’ to the standard devia-tionofthemeanspectrumandtakethemostconservativeasour initial estimate of therandom uncertainty. However, if thiswere tobeapplied uniformlyacross theTEXESspectrum,we wouldbe assigningequalweighttobothclearandtelluric-contaminated re-gions inthe inversions.Following Fletcher etal., 2014,we there-foreweightourmeasurementuncertaintyusingthemeasuredsky emissionspectra(ratherthanamodelledtellurictransmission fol-lowing Greathouse etal., 2005), accounting forthe Doppler shift that was applied to each pixel ofthe Jupiter cubesto bring the wavelengthsto their reststates.The wavelength-dependent stan-dard deviation wasinflated by a factor of two in the vicinity of strongtelluric features so that they would be effectivelyignored inthespectral inversion.Furthermore,the worst-affectedspectral regionswereremovedfromthefitentirely,producinggapsinthe spectralcoverageofasinglechannel.
Fig. 9. TEXES zonally-averaged spectra before (black) and after (blue) application of a scaling factor ( Fig. 8 ) to correct the radiometric calibration to match a synthetic model based on Cassini (this example is for a spectrum at 15 °N). The red line shows the sky transparency (accounting for both sky transmission and losses in the telescope). Differences are most notable in the centre of the N-band (panels d–f) where the sky transparency was largest. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
werebeingmisinterpretedbythespectralinversionalgorithm. For-wardmodelssuggestthatthecontinuumshouldeffectivelybeflat awayfromtheH2S(1)andC2H2emissionfeatures,sowe
empiri-callycorrectedthedatabydividingthroughbya
τ
n ,whereτ
isa normalisedskyspectrum topreservetheabsolutefluxcalibration andnisatuningparametertoflattenthespectrum.The measure-mentuncertaintywasincreasedintheseregionsbythedifference betweentheoriginalandflattenedspectrum.Theserandom preci-sionuncertaintieswerefixedfortheremainderofthespectral in-versions,andtheinfluenceofsystematicuncertaintiesinaccuracy areconsideredinthefollowingsections.4. Retrievalsensitivity
Before applying the NEMESIS spectral retrieval algorithm to zonal averages and spatially-resolved spectra, we first created a
zonal-mean spectrum ofJupiter’s tropical domain (±20° latitude) fromtheTEXES cubesto demonstratethe influenceofthe radio-metricaccuracyanda prioritemperature,gas andcloud distribu-tions on the robustness of the retrievals. We compared our re-trieved properties to those from the Cassini/CIRS ATMOS2A map (December 31, 2000) at 2.5-cm−1 spectral resolution,which was spatiallyaveragedinthesamewayastheTEXEScubes.
4.1. Influenceofcalibrationuncertainties
Assuming initially that the TEXES radiometric calibration was accurate,weattemptedtofittheeightchannelssimultaneouslyby varying T(p), parameterised NH3 and PH3 distributions, scale
fac-tors for C2H2, C2H6 and the opacity of a 800-mbar grey cloud.
and901cm−1channelsshowedsuchinconsistencythatthe tropo-spheric temperatures that were required to matchthe 901 cm−1 channel caused significant overestimation of the continuum at 744 cm−1. Fitting these two channels independently, we found that the744 cm−1 continuum required440-mbartemperaturesof 132–135K,whereasthe901-cm−1 channelrequiredtemperatures between 141–143 K atthe same altitude.These ∼10-K tempera-turedifferenceshaveanorder-of-magnitudeeffectontheretrieved abundancesofNH3,whichdominatestheN-bandabsorption
spec-trum. The higherthe tropospherictemperature, the more ammo-nia was required to reproduce the spectrum. We note that 440-mbar temperatures in the 130–135-K range (i.e., those from the 744 cm−1 continuum)are more consistent withthe independent CIRS analysis of Achterberg et al. (2006) and the ∼130-K tem-perature derived from the Voyagerradio science experiment and Galileo probe Atmospheric Structure instrumentfor thispressure level(Lindal,1992;Seiff etal.,1998).Furthermore,ourestimatesof the NH3 abundanceare anorderof magnitudehigherthanthose
of Achterbergetal.(2006)ifweassume theTEXES calibrationto beaccurate.Thisqualitativelyconfirmstheneedtoscalethe 901-cm−1radiancedownwards.
As a second example,the stratospheric temperaturesrequired to fit the 819 cm−1 channel significantly overestimated the flux in the CH4 lines at1247 cm−1.Stratospheric temperature fits to
the819cm−1channel (whilealsopermittingethanetovary) sug-gested1-mbartemperaturesinthe167–178Krange,dependingon thelatitude.Conversely,fitstothe1247-cm−1CH4 linessuggested
1-mbartemperaturesinthe159–169Krange.TheCassini/CIRS re-sults derived from a full 600–1400 cm−1 spectrum favoured 1-mbartemperaturesbetweenthesetwoextremes,qualitatively sup-portingadecreaseinthe819-cm−1channelandanincreaseinthe 1248-cm−1channeltomakethingsconsistent.Althoughindividual channelscanbereproducedinisolation,thisdifficultyinfittingall channelssimultaneouslywasevidenteverywhere,despiteourbest attemptstodosoafterathoroughexplorationofthepriors.
AlthoughtheradiometricaccuracyofCassini/CIRSisitself sub-ject to uncertainty and would certainly benefit from indepen-dent ground-based confirmation, the quantitative similarity be-tweenCIRSandVoyagerobservationsofJupiter(e.g., Simon-Miller etal.,2006b)givesuscausetotrusttheCIRScalibration.We there-foreapplytheglobal-meanscalefactorsshownin Fig.8globallyto the TEXES data (equivalent tobrightness temperaturedifferences of8-Kin theworst case)andrerun theinversion. Thisallows us to reproducetheeight TEXESchannelswitha temperature struc-turethatlooksreasonableandisquantitativelysimilartothat de-rived from Cassini. In Fig. 10 we demonstrate how the retrieved T(p) structure is modified by changing the scale factor for each channel, one ata time, to return it to the original TEXES values (i.e.,ascalefactorofone).Insome channelstheeffectsarerather straightforward -scaling the1248-cm−1 from1.0to 1.3(Fig.10f) has theeffect ofwarming the mid-stratospherictemperatures by ∼4 K whilst improving the quality ofthe fit to the data, consis-tentwiththeexpectedchangesinblackbodytemperaturefroma 30% changeinradiance (Fig. 8).Changingthe scalefactorforthe 901, 960and1161 cm−1 channels hasa subtleeffecton the T(p) (Fig.10c–e) buta dramatic effecton the retrieved ammonia and cloudabundancesandtheabilitytofitthespectrum.
More complicated effects occur when the contribution func-tionsforaTEXESchanneloverlaptheuppertroposphereandlower stratospherein Fig.10a,b.The 819cm−1 channel senses boththe troposphere and stratosphere, but retaining the unscaled TEXES datacausesafailure ofourmodeltoconvergetoareasonable so-lution-thehighlyoscillatorystructurein Fig.10bisanexampleof aretrievalstrugglingtofitinconsistentdata,resultinginthe over-allgoodness-of-fit
χ
2/N(whereNisthenumberofspectralpoints)increasing from∼0.8inthe bestcaseto∼4.7 fortheworst case.
Theshapeofthetropopauseregiondemonstratesahighsensitivity tothescalingsappliedintheQ-bandduetothelimitedconstraint providedbythe contributionfunctionsfromtheother channels– hencetemperaturesvarywildlyhereinanattempttoimprovethe qualityofthespectralfitwithlimitedsuccess(the
χ
2/Nvariesbe-tween0.8and0.9forthefourcases shown).Thispresentsa sig-nificantproblemfortherobustnessofretrievalsinthisregion, par-ticularlyastheQ-bandspectrasufferfromsignificanttelluric con-tamination.
Insummary,theuncertaintyintheTEXES radiometric calibra-tion impacts the quantities that can be derived fromthese data, particularlygiventhe degeneracies inherentin spectral inversion. We proceed withthe CIRS-derived scaling factors (Fig. 8), which allowustofittheTEXESdatawiththebestgoodness-of-fitanda smoothT(p)structure.
4.2.Sensitivitytothetemperatureprior
Fig.11showsthesensitivityoftheretrievedtemperature struc-turein Jupiter’stropics to changesin oura priori thermal struc-ture.Fivedifferentpriorswereused,varying±20Kfromthe refer-enceatmospheredescribedin Section3.TheretrievedT(p)remains consistent throughout the troposphere and lower stratosphere, onlybeginningtodiffersignificantly inthemid-stratosphere near 1mbar.Forp<0.5mbartheretrievedT(p)differencesbeginto ex-ceedtheuncertaintyonournominalretrievals,implyingthateven withthehighspectral resolution ofTEXES we still havea signif-icantsensitivity tothe prior inthe upperstratosphere.Notethat theuncertaintiesshownin Fig.11aretheformalerrorsonthe op-timal estimate derived from the measurement, smoothing anda priorierrorcovariancematrices(Eq.(22)of Irwinetal.,2008),but they underestimatethe trueuncertainty shownin Fig.11 forthe upperstratosphere.
Asethaneandacetyleneemissionbandsbothhavemulti-lobed contribution functions that probe these low pressures, this will haveimplicationsfor ourabilityto determinehydrocarbon abun-dances. For the five cases in Fig. 11, the scaling factors for the priorC2H2andC2H6abundancesvaryby40%and6%,respectively.
Acetylene’sstrongdependenceontheupperatmospheric tempera-tureisunsurprisinggiventhehigh-altitudepeaksofthe contribu-tionfunctionsin Fig.7,whereasethanesensesthedeeper strato-spherewhere temperaturesare betterconstrained.The quality of the fits varies from
χ
2/N=0.79− 0.81 for this particulartropi-cal spectrum, with marginally better fits for the warmest upper stratosphere. Repeating this test for a tropical-mean of the CIRS 2.5cm−1 observationsshowsexactlythesameproblem, withT(p) profiles becoming dependent onthe prior forp < 0.2mbar, and corresponding uncertaintiesin theC2H2 andC2H6 abundances of
25%and3%,respectively.Thehydrocarbonabundancesarestrongly sensitivetothisupperatmospherictemperatureuncertainty,sowe mustidentifysomewaytoconstrainthepriorusingprevious mea-surements.Wenote thatournominalpriorhasa1-mbar temper-atureof167K(consistentwiththeestimateof168-Kfrom Lindal, 1992)andthat Seiff etal.(1998)showedaquasi-isothermal struc-ture up to p<2
μ
bar, consistent with the warmer retrievals inFig. 11. However, this is only representative of one location on Jupiter, and higher spectral resolutions (with more high-altitude informationcontent)willbeneededtoproperlyconstrain temper-aturesandacetyleneabundancesinJupiter’supperatmospherefor p<1mbar.
4.3.Sensitivitytothehydrocarbonpriors
Fig. 10. Demonstration of the effects on the T ( p ) of scaling the TEXES channels relative to one another. Panel a shows the tropopause effects of scaling the Q-band channels; panels b–f show the results from scaling the 819, 901, 960, 1161 and 1248-cm −1 channels. The effects on the composition and the quality of the spectral fit are discussed in
Section 4 .
peak inthe 15–20
μ
bar region (asshown in Fig. 12), where our thermaluncertaintiesare extremelylarge.Toexplorethese uncer-tainties, we constructed a grid of stratospheric temperature and hydrocarbon priors in the following way: (a) using isotherms in theupperstratospherebetween150and190K,smoothly connect-ing to our nominal temperature prior in the lower stratosphere; (b)anethanedistributionwithzeroabundanceforp>100mbar, arange offractional scaleheights for100 > p > 0.1mbar; con-stantmolefractionsof0.5–5.0 pmbetween0.1>p>0.01mbar; andadeclinewithaltitudeforp<0.01mbar;and(c)anacetylene distributionwithsimilarparametersbutwithahigher-altitude re-gionofconstantabundancebetween0.05>p>0.005mbar.With five variables defining the temperature and hydrocarbon priors, we reran the tropical retrievals for both CIRS and TEXES over a thousandtimes,recording thebest-fitting2-mbarabundance and thegoodness-of-fitto the744, 819and 1247-cm−1 channels.The scaledprofilesareshownin Fig.12.ThisexperimentdemonstratedthatbothCIRSandTEXES spec-tracontaininformationontheverticaldistributionsoftemperature andhydrocarbons,withsomepriorsleadingtobetterfitsthan oth-ers.Thescatterinthebest-fittingC2H2 profiles(Fig.12a)islarger
thanthoseforC2H6(Fig.12b),giventhatonlythreenarrow
acety-lenelinesareobservedintheTEXESchannels.Theretrievalisstill abletoconvergeonaverticalhydrocarbonprofileeventhoughthe upperatmospherictemperaturesareuncertain(Fig.12c).Allofthe profiles,despiteradicallydifferentabundancesintheupper atmo-sphere,convergeinthe0.1–5.0mbarregionwhereTEXESismost
sensitive.Inthistropical-meanspectrum,wefound2-mbar acety-leneabundancesbetween0.02ppmand0.07ppm(column densi-tiesintegratedforp<10 mbarof2.5− 4.5× 1015 molecules/m2)
and2-mbarethaneabundancesbetween1.5–2.5ppm(1.9− 2.3× 1016 molecules/m2 at p < 10 mbar), depending on the vertical
profileandretaining onlythose modelsthat reproduced thedata within 1
σ
. These abundances fall within the range of previous studies(seetheexcellentsummaryinFig.1of Zhangetal.,2013a). Themostradicaldeviationsoftheabundanceprofilesfrom previ-ousworkfailedtoreproducethedatasatisfactorily.Fig. 11. Testing the retrieval sensitivity to the T ( p ) prior. Five different priors were used, varying ±20 K from our reference atmosphere (dotted lines). The correspond- ing retrieved T ( p ) (solid lines) are remarkably consistent until p < 0.5 mbar, when their deviations begin to exceed the formal uncertainty on the optimal estimate (dashed lines).
altitude,achangeintheverticalabundancegradient,orachange in upper atmospheric temperature. Our inversions therefore as-sume that horizontaltemperature changes at microbar pressures mirror thoseatmillibarpressures (fromsmooth relaxationtothe upperatmosphericprior setbyVoyagerandGalileoradio science experiments) and that the hydrocarbon profiles have the same vertical shapeseverywhere.Ultimatelya combinationofradiative andphotochemicalmodelswillberequiredtosetbetterpriorsfor temperatureandhydrocarbons inthe upperatmospheretobreak thisextremelychallengingdegeneracy.
4.4.Sensitivitytotroposphericpriors
Degeneratesolutions alsoexist when we consider thevarious contributors-temperature, aerosols andthe verticaldistributions ofammonia andphosphine– thatshape spectra inthe819, 901, 960and1161cm−1channels.Adoptingasimilarstrategy tothose employed inthe stratosphere,we explored thisdegeneracyvia a grid, allowing the PH3 and NH3 transition pressures p0 and the
cloud base pressure to vary between 600 and 1000 mbar; the cloudscaleheighttovarybetween0.1and0.5(i.e.,compactor ex-tended);andrepeatedthisfortwo cloudcross-sectionmodels:(i) aspectrally-uniformabsorber,and(ii)acloudofsphericalNH3ice
particles(Martonchiketal., 1984) witha standard gamma distri-butionofmeanparticleradii10
μ
mandvariance5μ
m(i.e.,large particlesconsistent with Fletcher etal., 2009a). We thenallowed the NH3 and PH3 deep abundances and fractional scale heights,plus the temperatures and opacity of our aerosol layer, to vary freely during the tropical-mean retrieval, and the resulting NH3,
PH3 andtemperatureprofilesareshownin Fig.13.
We found negligible difference in the fitting quality between thecompactandextendedaerosollayers,limitedsensitivitytothe basepressureofthecloudsthemselves,andnodifferenceinfitting qualitybetweenthegreycloudorlargeNH3 iceparticles.Theuse
oftheextendedcloudfavouredT(p)profiles(Fig.13c)withwarmer 1-bartemperatures(∼172K)whereasthecompactcloudfavoured ∼161K(theGalileoprobemeasured∼166 Katthesamealtitude,
Seiff etal., 1998). Giventhe broad-band effectof aerosolopacity onJupiter’sspectrum,awider spectralrangewouldbeneededto providenewconstraintsontheuppertroposphericaerosols,sowe use a p=800 mbar compact cloud oflarge NH3 ice spheres for
theremainderofthisstudy,representativeofpreviousliteraturein thisrange(Achterbergetal.,2006;Fletcheretal.,2009a;Matcheva etal., 2005; Wong etal., 2004). This choicehas verylittle effect on the 440-mbar temperatures and gas abundances, buta more substantial impact near800 mbar – temperatureshereare ∼2 K warmer for higher-altitude cloud bases, but this is smaller than the∼10 K scatter intemperatures dueto thepoorconstraint on thedeepammoniaabundance.Thisreflectsthesubstantial degen-eracybetweenallthreeparameters(temperatures,aerosolsandgas abundances)atp>800mbar.
Retrieved vertical profiles of ammonia and phosphine (Fig. 13a,b) overlap in the 400–600 mbar range despite large
Fig. 13. Testing the sensitivity of TEXES inversions to the chosen priors for vertical distributions of (a) PH 3 and (b) NH 3 , and (c) the corresponding effects on the T ( p ) retrieval. Models fitting the data within a χ2 /N = 1 are shown as black lines, models with poorer fits are shown as the grey lines.
differences in the location of the profile transition pressures p0 and the deep mole fraction q0. The sensitivity to the deep
abundances islimited in theTEXES data,resulting inthe scatter of results spanning an order of magnitude for both species. For ammonia,thebest-fittingprofileshadp0near800mbar,whichis
alsoconsistentwiththealtitudeoftheputativeNH3condensation
cloud. At higher pressures, the Galileo probe indicated a deep NH3 molefractionof570 ±220 ppm forp> 8bar (Wong etal., 2004),butJupiterobservationsatmicrowavewavelengthssupport a depletion for p < 4 bar to reach 100–200 ppm levels in the 1–2 bar region (de Pater et al., 2001; Showman and de Pater, 2005). The deep abundances estimated by TEXES fall between thesetwoextremes. Athigheraltitudes, theTEXES fits supporta steepdecline inNH3 to∼5 ppm near 440mbar, consistent with
the 2− 10 ppm range reported by Achterberg et al. (2006). For theremainder ofthisstudy,we fix theNH3 p0 to 800 mbarand
vary both the deep abundance and fractional scale height to fit thedata.
ThedeepabundanceofPH3ispoorlyconstrainedbytheTEXES
data.ThePH3 profilesin Fig.13ademonstratethatthefitting
qual-ityisonly weakly sensitive top0,with valuesinthe range 600–
800mbarreproducingthedatawithin 1
σ
anda best-fitforp0=750mbar.PH3 abundances forthebest-fittingmodels all overlap
near400 mbarwhere TEXES hasthe mostsensitivity, withmole fractionsintherange 0.35–0.45ppm dependingon thechoice of vertical profile. Ifwe fix p0 to 750 mbar, we derive deep
abun-dancesthat areconsistentwiththeq0 ∼2ppmestimatedforp>
1barbyprevious mid-infraredstudies(see Fletcheretal.,2009a, andreferencestherein),butlargerthanestimatesofq0 ∼0.7ppm
usingthedeeper-sounding5-
μ
mwindow(Gilesetal.,2015,using thesamespectralinversiontechniques).Resolvingthisdiscrepancy requires simultaneous modelling of both the 5- and 10-μ
m PH3bandsandisbeyondthescopeofthecurrentstudy,sowefixthe PH3 to q0=2 ppm for p > 750mbar forthe remainder of this
study.
In summary, despite theexcellent spatialandspectral resolu-tionof the TEXES Jupiter dataset, one significant challenge ham-pers its analysis – the radiometric calibration. If the calibration ofthe eight channels were accurate, then the exploration of pa-rameterspacedescribedabove wouldhaveprovidedsomeinsight intotheverticaldistributionsoftemperature,hydrocarbons, tropo-sphericgasesandaerosols.Instead,we havesystematically tuned theabsoluteabundancesandtemperaturestobroadlyreflect previ-ousinvestigations.Wearenowabletoexplorerelativespatial
vari-abilityineachofthesepropertiesinthenextSection,butwiththe caveatthatsystematicuncertaintiesarelarge.
5. Resultsanddiscussion
In this Section, we presenta comparisonof Jupiter’s temper-atures, composition and aerosol opacity from both CIRS (2000, Ls =110◦) and TEXES observations (2014 Ls =175◦). Zonal-mean spectrawerecomputedfromall TEXESandCIRSdataona2° lat-itudinalgrid witha4° width.Spatially resolved spectraare com-putedonthesamelatitudinalgrid,butwithalongitudinalstepof 2° andawidthof4°,resultinginapproximately11,000spectrafor aglobalmap between60°Nand60°S.Forthezonal-meanspectra we retrievedvertical temperatureprofiles atevery location along with(i) the optical depthof the800-mbar compact cloud of
10-μ
mradius NH3 icespheres;(ii)thescale heightforPH3 above awell-mixedmolefractionof2ppmforp>750mbar;(iii)thedeep molefractionandfractionalscaleheightforNH3withatransition
pressure of p=800 mbar; and(iv) scale factors for low-latitude meanprofilesofC2H2 andC2H6 from Nixonetal.(2007).The
re-trievalstrategyforthespatially-resolvedmapswassimilar,except thatwe simplyscaledalow-latitudemeanofthePH3,NH3,C2H2
andC2H6 profiles derived fromthe zonal-mean spectra.We
cau-tionthereaderthat thechoiceoftemperatureandgaseouspriors doesindeed influencethe retrieval, andthat alternative distribu-tions are often able to reproduce the data equally well. In par-ticular, the hydrocarbon vertical gradients are held constant and all variationsare assumedto be horizontal. This isthe first time thatglobalmapsofthesespecieshavebeenpresentedfrom mid-infraredspectroscopy.
5.1. Temperatures
Although several previous authors have used the December 2000CIRS2.5cm−1observationsofJupitertodeterminethe zonal-mean temperatures (e.g., Flasar et al., 2004b; Matcheva et al., 2005;Simon-Milleretal.,2006b;Achterbergetal.,2006;Fletcher etal.,2009a),werepeatedthesemeasurementstoensurethatthe retrievalparameterisationswereidenticalbetweentheTEXESand CIRS inversions. Fig. 14 compares Jupiter’s zonal-mean tempera-tures in2000and2014 at fivedifferentpressures; Fig. 15 shows latitude-pressurezonalcross-sectionsofthetemperaturefield;and
