Mon.Not.R.Astron.Soc.000,1–5(2009) Printed8January2010 (MNLATEXstylefilev2.2) High resolution AMI Large Array imaging of spinning dust sources: m spatially correlated 8 m emission and evidence of a stellar wind in L675⋆ 0 1 0 AMI Consortium: Anna M. M. Scaife1,2†, David A. Green1, Guy G. Pooley1, 2 Matthew L. Davies1, Thomas M. O. Franzen1, Keith J. B. Grainge1,3, n a Michael P. Hobson1, Natasha Hurley-Walker1, Anthony N. Lasenby1,3, J 8 Malak Olamaie1, John S. Richer1,3, Carmen Rodr´ıguez-Gonza´lvez1, ] Richard D. E. Saunders1,3, Paul F. Scott1, Timothy W. Shimwell1, A G David J. Titterington1, Elizabeth M. Waldram1 & Jonathan T. L. Zwart4 . 1AstrophysicsGroup,CavendishLaboratory,JJThomsonAvenue,CambridgeCB30HE h 2DublinInstituteforAdvancedStudies,31FitzwilliamPlace,Dublin2,Ireland p 3KavliInstituteforCosmologyCambridge,MadingleyRoad,Cambridge,CB30HA - 4ColumbiaAstrophysicsLaboratory,ColumbiaUniversity,550West120thStreet,NewYork10027,USA o r t s a Accepted—;received—;inoriginalform8January2010 [ 2 ABSTRACT v We present 25′′ resolution radio images of five Lynds Dark Nebulae (L675, L944, L1103, 1 1 L1111 and L1246) at 16GHz made with the Arcminute Microkelvin Imager (AMI) Large 0 Array.TheseobjectswerepreviouslyobservedwiththeAMISmallArraytohaveanexcess 4 ofemissionatmicrowavefrequenciesrelativetolowerfrequencyradiodata.InL675wefind . a flat spectrum compact radio counterpart to the 850m m emission seen with SCUBA and 0 suggestthatitiscm-waveemissionfromapreviouslyunknowndeeplyembeddedyoungpro- 1 9 tostar.InthecaseofL1246thecm-waveemissionisspatiallycorrelatedwith8m memission 0 seenwithSpitzer.SincetheMIRemissionispresentonlyinSpitzer band4wesuggestthat : itarisesfromapopulationofPAH molecules,whichalsogiverisetothecm-waveemission v throughspinningdustemission. i X Keywords: Radiationmechanisms:general–ISM:general–ISM:clouds–stars:formation r a 1 INTRODUCTION cificcaseofdarkcloudstherecentAMIsample(AMIConsortium: Scaifeetal.2009;hereinafterPaperI)offourteenLyndsDarkNeb- Thecomplete characterization of microwave emission fromspin- ulaefoundanexcessinonlyfive. ningdustgrainsisakeyquestioninbothastrophysicsandcosmol- It has been suggested that cm-wave emission from spinning ogy. It probes a region of the electromagnetic spectrum where a dust is emitted by a population of ultra-small grains (Draine & number ofdifferent astrophysical disciplinesoverlap. Itisimpor- Lazarian1998).Theseultra-smallgrainsarethoughttoexistmainly tant for CMB observations in order to correctly characterise the in the form of single polycyclic aromatic hydrocarbon (PAH) contaminatingforegroundemission;forstarandplanetaryforma- molecules.PAHmoleculesaregenerallydetectedthroughtheirnar- tionitisimportantbecauseitpotentiallyprobesaregimeofgrain rowlineemissionfeaturesintheMIR.Fortheseemissionfeatures sizesthatisnototherwiseeasilyobservable. to be observed the PAH molecules must be exposed to a strong Althoughanumberofobjectshavenowbeenfoundtoexhibit source of UV flux. Sincethisflux isgenerally absent in the case anomalous microwave emission, attributed to spinning dust, it is ofdarkclouds,themicrowaveemissionfromtherotationofPAH stillunclearwhatdifferentiatesthoseobjectsfromthemanyother moleculesmaybetheonlywaytostudytheverysmallgrainpop- seeminglysimilartargetsthatdonot showtheexcess.Inthespe- ulationintheseobjects. Itisalsoknownthatradiocontinuumemissionindarkclouds ⋆ Werequestthatanyreferencetothispapercites“AMIConsortium:Scaife mayarisefromionizedgasassociatedwithastellaroutflow.When etal.2010” aluminousstarispresentthisariseseitherastheresultofacom- † E-mail:[email protected] pactHIIregionoranionizedstellarwind.Inthecaseofveryyoung 2 Scaifeetal. lowluminositystarsradiocontinuumemissionmaybealsobede- Table1.AMILALynds DarkNebulae. Column [1]Nameofcloud, [2] tected.Inthisinstanceitisgenerallyattributedtothepresenceof RightAscension,[3]Declination, [4]AMILAsynthesizedbeamFWHM apartiallyionized(0.026xe60.35;Bacciotti&Eislo¨ffel1999) majoraxis,[5]AMILAsynthesizedbeamFWHNminoraxis,and[6]r.m.s. stellar wind (Wright & Barlow 1975; Panagia & Felli 1975), or noisefluctuationsonthecombinedchannelmap. possiblyaneutralwindwhichhasbeenshock-ionizedfurtherfrom thecentralsource byimpactingonadenseobstacle(Curiel etal. Name RA Dec D qmaj D qmin s rms 1989). (J2000) (J2000) (arcsec) (arcsec) (mJy) bm In this paper we present follow-up observations of the five L675 192352.6 110739 49.9 27.4 35 AMISmallArray(SA)spinningdustdetections(PaperI)athigher L944 211740.8 431808 36.5 31.2 31 resolutionwiththeAMILargeArray(LA)overthesamefrequency L1103 214210.2 564344 32.0 26.3 25 range.Allco-ordinatesinthispaperareJ2000.0. L1111 214027.1 574810 39.4 30.8 29 L1246 232530.1 633830 31.2 26.9 25 2 OBSERVATIONS AMI comprises two synthesis arrays, one of ten 3.7m antennas (SA) and one of eight 13m antennas (LA), both sited at Lord’s Bridge, Cambridge (AMI Consortium: Zwart et al. 2008). The telescope observes in the band 13.5–17.9GHz from which eight 0.75GHzbandwidthchannelsaresynthesized.Inpractice,thetwo lowestfrequencychannels(1&2)arenotgenerallyusedduetoa lowerresponseinthisfrequencyrangeandinterferencefromgeo- Figure2.L675SourceA:datapointsarefluxdensitiesfromAMILAchan- stationarysatellites. nels3–8.Thebest-fitspectralindexofa =0.10±0.36isshownasadashed Observations of five Lynds dark nebulae selected from the line. originalAMISAsampleweremadein2009February–Marchus- ingtheAMILA.Theco-ordinatesofthesefieldsarelistedinTa- 3 RESULTS ble1alongwiththesizeoftheAMILAsynthesizedbeamtowards eachobjectandther.m.s.noisemeasuredoutsidetheprimarybeam L675: The AMI LA observations of L675 show two obvious re- ontheCLEANedmaps.WenotethattheAMILAobservationof gionsof compact emission, seeFig.1. Thefirstof these, slightly L1246 is towards the north–east of this cloud where anomalous offset from the pointing centre, is coincident with both the peak emissionwasdetectedbytheAMISAanddoesnotcoverthesame of the AMI SA emission and also the compact emission seen at areaastheoriginalSCUBAobservation. 850m mbytheSCUBAinstrument(Visseretal.2001;2002).We Data reduction was performed using the local software tool denote this source “A” (19h23m50.s5, +11◦07′44′′). The second, REDUCE, see Paper I for more details. Flux calibration was per- justoutsidetheLAprimarybeamFWHMtothenorth-east,isco- formedusingshortobservationsof3C286nearthebeginningand incident withtheprobableextragalacticpoint sourceidentifiedas end of each run. We assumed I+Q flux densities for this source “B”(19h24m02.s4,+11◦10′54′′)intheoriginalAMISAobserva- intheAMILAchannelsconsistentwiththefrequencydependent tions(PaperI). model of Baarset al.(1977), ≃3.3Jy at16GHz. AsBaarset al. SourceAiscompletelyunresolvedbytheAMILAandshows measureIandAMILAmeasuresI+Q,thesefluxdensitiesinclude aflatspectrumacrosstheAMIband,a 17.9=0.10±0.36,consis- 14.3 correctionsforthepolarisationofthecalibratorsourcederivedby tentwithfree–freeemission,seeFig.2.Thisspectralindexdiffers interpolatingfromVLAmeasurements.Acorrectionisalsomade considerablyfromthatmeasuredbyAMISA.Thisisbecausethe for the changing intervening air mass over the observation. The LAisnotsensitivetothelargescaleemissionseenwiththeSA.In- phasewascalibratedusinginterleavedobservations ofcalibrators deeditseemslikelythattheemissionseenbythetwoarraysarises selectedfromtheJodrellBankVLASurvey(JVAS;Patnaiketal. fromcompletelydifferentsources. 1992).Aftercalibration,thephaseisgenerallystableto5◦forchan- L944: The original AMI SA observations of L944 revealed nels4–7,and10◦forchannels3and8.TheFWHMoftheprimary acompactregionofemissiontothenorthofthecloud,coincident beamoftheAMILAis≈6′at16GHz. withonesideoftheprotostellaroutflow.AMILAobservations,see Reduced data were imaged using the AIPS data package. Fig.1,revealthisemissionarisesnotfromapoint-likeobjectbut CLEANdeconvolutionwasperformedusingthetaskIMAGRwhich ratherfromadiffuseregionofemission,thepeakofwhichoccurs appliesadifferentialprimarybeamcorrectiontotheindividualfre- at21h17m42.s6,+43◦20′08′′.Weestimatethefluxspectrumbyin- quency channels to produce the combined frequency image. De- tegratingthefluxdensityfromtheprimarybeamcorrectedchannel convolved maps weremade fromboth thecombined channel set, mapswithinatwoarcminuteradiusoftheLApointingcentre.This seeFig.1,andforindividual channels. Thebroadspectralcover- showsasteeplyrisingspectrumwitha 17.9=−2.1±0.5.Thisis 14.3 ageofAMIallowsarepresentationofthespectrumbetween14.3 consistentwiththatfoundfromtheAMISAdata,howeverthiscor- and 17.9GHz to be made independently of other telescopes and respondenceisnotmeaningfulasthelowsignaltonoiseintheSA inwhat followsweusetheconvention: S(cid:181) n −a ,whereSisflux dataprecludesapreciseestimate.Thefluxdensityfoundtowards density, n isfrequency anda isthespectralindex. Allerrorsare thisregionintheLAmapisonlymarginallylowerthanthatfound quotedto1s . from the comparatively coarser resolution SA map. This implies AMILA spinningdustobservations 3 L675 L944 L1103 L1111 L1246 Figure1.AMILAcombinedchanneldataisshownasgreyscaleinunitsofmJy/beam,greyconrtoursat−6,−3,±2,±1s andblackcontoursat3,6,12,24s etc.SCUBA850m mdataisshownasredcontourswithlevelsasinVisseretal.(2001;2002)fortheallcloudsexceptL1246.TheAMILAobservationof L1246doesnotcovertheregionobservedbySCUBA.AMISAdataisshownasbluecontourswithlevelsasinPaperI.TheAMILAprimarybeamFWHM isshownasacircleandthesynthesizedbeamasafilledellipseinthebottomleftcorner. 4 Scaifeetal. Table2.IntegratedfluxdensitiesinmJyforAMILAobservationsofL675,L944andL1246.Errorsarecalculatedass =p(0.05S)2+s r2ms,wheres rmsis ther.m.s.noiseintheindividualchannelmap. Freq.(GHz) Name 14.3 15.0 15.7 16.4 17.2 17.9 a L675 2.74 ±0.34 2.24 ±0.12 2.26 ±0.12 2.12 ±0.11 2.42 ±0.14 2.22 ±0.12 +0.10±0.36 L944 - 2.03 ±0.11 2.40 ±0.13 2.89 ±0.15 2.86 ±0.15 - −2.11±0.49 L1246 - 0.62 ±0.18 0.55 ±0.03 0.62 ±0.03 0.58 ±0.04 - −0.40±0.82 thattheemissioncomesnotfromonesmoothextendedregionthat ispartiallyresolvedoutbytheLAbaselines,butfromacollection ofsmallerfragmentsorfilaments.Thesefragmentsareunresolved byeitherarray,althoughthegranularitybecomesmoreevidentin thehigherfrequencychannelsoftheLA.Theamountoffluxlost inchannels5to8relativetochannel4issignificantlysmallerthan wouldbeexpectedfromaGaussiansourceofsimilardimensions. L1103 and L1111: AMI LA observations of L1103 and L1111donotshowanydistinctregionsofcompactcm-waveemis- sion.Thediffusepatchesoflowlevelemissionpresentwithinthe primary beam towards both sources are indicative of larger scale structureswhichhavebeenresolvedoutbythesynthesizedbeam. Wecanprovideanestimateofthefluxdensityseentowardsthese objectswiththeLAbyfittingandremovingatiltedplanebaselevel attheprimarybeamFWHM.Fromthecombinedchanneldatathis givesS16=5.1±0.6mJyandS16=2.4±0.3mJyforL1103and L1111,respectively.Thesevaluesindicatethatthefluxlossiscon- Figure3.L675:AMILAcombinedchanneldataisshownaswhitecontours siderable: approximately 45% and 96%. The sensitivity of these at3,6,12s etc.SpitzerBand4dataisshownasgreyscaleinMJy/sr,and issaturatedatbothendstoemphasisethestructurepresent.AMISAdatais LA observations is much greater than those of the SA and it is shownasbluecontoursasinFig.1. possiblethatthepatchyemissioninthesefieldscorrespondstoen- hancementsintheextendedemissionwhicharebelowthedetection thresholdintheSAdata. L1246: AMI SA observations towards L1246 did not show anyexcessemissioncoincidentwiththeSCUBAidentificationof thedarkcloud,butdidrevealaregionofemission≈2′tothenorth– east of the cloud, in a region not covered by the SCUBA map, whichhadnocounterpartinthelowerfrequencyobservations.AMI LAobservationsofthisNEregionshowanarcofemission(peak: 23h25m20.s4,+63◦38′40′′),seeFig.1.Weassessthespectralbe- haviour of this object in two ways. Firstly, we estimate the flux densityofthearcitself.Wefitandremoveatiltedplanebaselevel withinanirregularpolygondrawnaroundtheobjectandintegrate the remaining flux. Secondly, we fit a tilted plane baselevel to a circleattheprimarybeamFWHMandintegrateallthefluxabove thisbaselevelwithinthatradius.Bothmethodsgiveconsistentre- sults, as might be expected since the primary beam is relatively empty otherwise. From the first method we find a spectral index a 1157..09=−0.40±0.87,andfromtheseconda 1157..09=−0.47±0.82. Figure4.L1246:AMILAcombinedchanneldataisshownaswhitecon- toursat1,2,4,8s etc.SpitzerBand4dataisshownasgreyscaleinMJy/sr, saturatedatbothendsofthescaletoemphasisethestructurepresent.The AMILAprimarybeamisshownasacircleandthesynthesizedbeamasa 4 DISCUSSIONANDCONCLUSIONS filledellipseinthebottomleftcorner. L675andL1246havearchivalSpitzerIRACdata,whichshowsin bothcasesasignificantamountofemissioninBand4(6.4–9.4m m) InBand4itispresentasanarc,coincidentwiththatseenat16GHz andverylittleintheotherthree(3.2–3.9,4.0–5.0and4.9–6.4m m, intheAMILAdata. respectively).InthecaseofL675thisemissionispresentonavery Spitzer Band 4 contains two of the PAH emission lines, in- largescale,seeFig.3.Theemissionseenat16GHzwiththeAMI cluding the strongest (7.7m m). Of the three other Spitzer bands SAappearsonasimilarscale,howeverthesmallfieldofviewofthe only Band 1 contains an emission line (3.3m m) and for ionized Spitzer data precludes amore detailed comparison. L1246 shows PAHsthislineisexpectedtobesignificantlyweaker.Itisprobable anarcofemissionat16GHzwhichisalsoevidentinSpitzerIRAC therefore that the MIR correlated cm-wave data seen in the AMI Band4,seeFig.4.ThisemissionisagainnotpresentinBands1–3. maps isaconsequence of spinning dust emission froma popula- AMILA spinningdustobservations 5 tionofionizedPAHmolecules.NeutralPAHmoleculesdonotin Towards L675 wehaveobserved flatspectrum compact cm- generalpossessapermanentdipolemomentandarethereforenot waveemissioncoincidentwiththeSCUBA850m memissionfrom expectedtohaverotationalemission(Tielens2008).Thisemission, the same region. These characteristics suggest that this source is themechanismofwhichisdescribedindetailbyDraine&Lazar- associatedwithastellarwindfromadeeplyembeddedyoungpro- ian(1998),arisesfromtheintrinsicdipolemomentsofsmalldust tostar. grains,mostlikelytobePAHmolecules,whichemitpowerwhen We detect extended cm-wave emission to the North of the theyrotate.Thisrotationhasavarietyofcontributingfactors,the L944SMM-1protostarwhichdisplaysspectralbehaviour consis- relativeimportanceofwhichvarieswithgrainenvironment.How- tentwitheitherspinningdust,oralternativelyacollectionofultra- ever,inthemajorityofcasesexcitationthroughcollisionwithions compactHIIregions. ispredominant. L1246showsanarcofcm-waveemissionwhichiscoincident InthecaseofL675A,wemustconsiderthepossibilitythatwe with emission seen in Spitzer Band 4. We suggest that this is an areobserving acoincidental extragalacticradiosource. Usingthe exampleofemissionfromapopulationofPAHmolecules,seenin extended9Csurvey15GHzsourcecounts(Waldrametal.2009), emissionlinesintheSpitzer data,andemissionasaconsequence wheren(S)=51(S/Jy)−2.15Jy−1sr−1,theprobabilitythatasource ofrapidrotationofthemoleculesinthecm-wavedata. withfluxdensitygreaterthan2mJylieswithintheFWHMofthe AMILAprimarybeamis0.12,andonly0.01withintheSCUBA field.ItislikelythereforethattheradiosourceL675Aisassociated 5 ACKNOWLEDGEMENTS withtheSCUBAcore. Wethank the staff of the Lord’sBridge Observatory for their in- Afurtherquestioniswhetherthecm-waveemissionmightbe valuable assistance in the commissioning and operation of AMI. explainedbythermal(Planckian)dustemission.Asinglegreybody spectrumwithadusttemperature,T ≈27K,mightbeusedtoex- AMIissupportedbyCambridgeUniversityandtheSTFC.NH-W, d plaintheLAfluxdensity,howeveritwouldrequireab of0.6.Such CR-G, TWS,TMOF, MO and MLD acknowledge the support of PPARC/STFCstudentships.Thisworkisbasedinpartonarchival avaluewould beunusual evenfor objectsknown topossess flat- dataobtainedwiththeSpitzerSpaceTelescope,whichisoperated teneddusttails,suchasprotoplanetary disks.Thissimplefitalso bythe JetPropulsion Laboratory, CaliforniaInstituteof Technol- neglectsthefluxlostbytheAMILAbaselinedistribution.SAob- ogyunder acontractwithNASA.Supportforthisworkwaspro- servationshavealreadyshownthissourcetopossessasignificant videdbyanawardissuedbyJPL/Caltech. amountofextendedemissionwhichwouldmakethisscenarioeven moreunlikely. The presence of a neutral or partially ionized wind from an outflowsourcethathasbeenshockedthroughencounteringadense REFERENCES obstacle (Torrelles et al. 1985; Rodr´ıguez et al. 1986) is used to AMI Consortium: Hurley-Walker N., et al., 2009, MNRAS, 396, understand the spectral indices seen towards exciting sources in 365 theradioregime(Curieletal.1990;Cabrit&Bertout1992).This AMIConsortium:ScaifeA.M.M.,etal.,2009,arXiv:0908.1655 model allows a spectral index range of 0.1 (optically thin) to -2 AMIConsortium:ScaifeA.M.M.,etal.,2008,MNRAS,385,809 (optically thick), which explains results which deviate from the AMIConsortium:ZwartJ.T.L.,etal.,2008,MNRAS,391,1545 value of a =−0.6 required by a spherically symmetric ionized BaarsJ.W.M.,GenzelR.,Pauliny-TothI.I.K.,WitzelA.,1977, wind(Wright &Barlow 1975; Panagia &Felli1975). Usingthis A&A,61,99 model as described in Curiel et al. (1989; 1990) the radio emis- BacciottiF.,Eislo¨ffelJ.,1999,A&A,342,717 sionisexpectedtobeopticallythin(t =0.1),consistentwiththe CabritS.,BertoutC.,1992,A&A,261,274 spectralindexseenacrosstheAMIband. Assumingadistanceof ClemensD.P.,YunJ.L.,HeyerM.H.,1991,ApJS,75,877 300pcandastellarwindwithawindspeedof200kms−1,wecan CurielS.,RaymondJ.C.,MoranJ.M.,RodriguezL.F.,CantoJ., calculatethattheAMIfluxdensitiestowardsL675Aareconsistent 1990,ApJ,365,L85 withamasslossof3.5×10−7M⊙yr−1.Amasslosssuchasthis CurielS.,RodriguezL.F.,BohigasJ.,RothM.,CantoJ.,Torrelles impliesamechanicalluminosityfromthewindofLmech≈1.1L⊙, J.M.,1989,ApL&C,27,299 comparabletothevaluesfoundbyCurieletalforL1448. DraineB.T.,LazarianA.,1998,ApJ,508,157 ThenatureoftheemissionseentowardsL944withtheAMI LadaC.J.,1985,ARA&A,23,267 LA is uncertain. The spectral index of this emission is consis- PanagiaN.,FelliM.,1975,A&A,39,1 tentwithspinningdustemissionoralternativelytheopticallythick Patnaik A. R., Browne I. W. A., Wilkinson P. N., Wrobel J. M., componentoffree–freespectrum.Suchafree–freespectrummight 1992,MNRAS,254,655 be exhibited at 16GHzby ultra-compact HII regions. However a TorrellesJ.M.,HoP.T.P.,RodriguezL.F.,CantoJ.,1985,ApJ, turn-overfrequency above16GHzwouldhaveanextremelyhigh 288,595 massandshouldthereforebeobviousinsub-mmobservations.This RodriguezL.F.,CantoJ.,TorrellesJ.M.,HoP.T.P.,1986,ApJ, needs tobe confirmed by either higher radio frequency measure- 301,L25 mentsinordertomeasuretheopticallythinregionofthespectrum VisserA.E.,RicherJ.S.,ChandlerC.J.,2001,MNRAS,323,257 andtheturn-over,orsub-mmmeasurementstoplaceconstraintson VisserA.E.,RicherJ.S.,ChandlerC.J.,2002,AJ,124,2756 themassofsucharegion. WaldramE.M.,PooleyG.G.,DaviesM.L.,GraingeK.J.B.,Scott Inconclusion,wehaveusedtheAMILAtoobserveasample P.F.,2009,arXiv:0908.0066 offiveLyndsDarkNebulaeselectedascandidatesforspinningdust WrightA.E.,BarlowM.J.,1975,MNRAS,170,41 emissionfromtheAMISAsampleofLyndsDarkNebulae(Paper I).Towardstwooftheseclouds(L1103andL1111)wedetectonly patchy diffuse emission characteristic of the presence of a larger ThispaperhasbeentypesetfromaTEX/LATEXfilepreparedbythe author. structurewhichhasbeenmostlyresolvedout.