The PAH emission characteristics of the reflection nebula NGC2023

9.5 MB·

by  Els Peeters

DraftversionJanuary25,2017
PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
THEPAHEMISSIONCHARACTERISTICSOFTHEREFLECTIONNEBULANGC2023
ElsPeeters1,2,CharlesW.Bauschlicher,Jr.3,LouisJ.Allamandola4,AlexanderG.G.M.Tielens5,AlessandraRicca2,MarkG.Wolfire6
DraftversionJanuary25,2017
ABSTRACT
Wepresent5-20µm spectralmapsofthereflectionnebulaNGC2023obtainedwiththeInfraredSpectro-
graphSLandSHmodesonboardtheSpitzerSpaceTelescopewhichrevealemissionfrompolycyclicaromatic
hydrocarbons (PAHs), C , and H superposed on a dust continuum. We show that several PAH emission
60 2
7 bandscorrelatewitheachotherandexhibitdistinctspatialdistributionsrevealingaspatialsequencewithdis-
1 tancefromtheilluminatingstar.Weexplorethedistinctmorphologyofthe6.2,7.7and8.6µm PAHbandsand
0 findthatatleasttwospatiallydistinctcomponentscontributetothe7–9µm PAHemissioninNGC2023. We
2 reportthatthePAHfeaturesbehaveindependentlyoftheunderlyingplateaus. Wepresentspectraofcompact
ovalPAHsranginginsizefromC toC ,determinedcomputationallyusingdensityfunctionaltheory,and
n 66 210
investigatetrendsinthebandpositionsandrelativeintensitiesasafunctionofPAHsize,chargeandgeometry.
a
J BasedontheNASAAmesPAHdatabase, wediscussthe7–9µm componentsintermsofbandassignments
andrelativeintensities. Weassigntheplateauemissiontoverysmallgrainswithpossiblecontributionsfrom
3
PAHclustersandidentifycomponentsinthe7–9µm emissionthatlikelyoriginatesinthesestructures. Based
2
ontheassignmentsandtheobservedspatialsequence,wediscussthephotochemicalevolutionoftheinterstel-

larPAHfamilyastheyaremoreandmoreexposedtotheradiationfieldofthecentralstarintheevaporative
]
A flowsassociatedwiththePDRsinNGC2023.
Subjectheadings:Astrochemistry-Infrared: ISM-ISM:molecules-ISM:moleculardata-ISM:lineand
G
bands-Line: identification-techniques: spectroscopy
.
h
p
- 1. INTRODUCTION etal.(2013)andStocketal.(2014)recentlyreportedthatthe
o strong correlation between the 6.2 and 7.7 µm PAH bands
The mid-infrared (IR) spectra of many astronomical ob-
r breaks down on small spatial scales towards the giant star-
t jectsaredominatedbystrongemissionbandsat3.3,6.2,7.7,
s forming region N66 in the Large Magellanic Cloud and to-
8.6,11.3and,12.7µm,generallyattributedtopolycyclicaro-
a wards the massive Galactic star-forming region W49A. This
[ matichydrocarbonmolecules(PAHs)andrelatedspecies.De-
spite the significant progress made in our understanding of suggeststhatinadditiontoPAHcharge,otherparameterssuch

1 theseemissionbands(furtherreferredtoasPAHbands), the asmolecularstructureinfluencetheastronomical6.2and7.7
v specifics of their carrier remain unclear. Indeed, no single µm PAHemission.AdependenceonPAHparameterssuchas
5 charge,molecularstructure,sizecanbeprobedbyinvestigat-
PAH molecule or related species has been firmly identified
8 ing theoretical calculations of PAH spectra (Langhoff 1996;
todate. Thislackofdetailedknowledgehampersourunder-
5 Bauschlicher & Langhoff 1997; Ellinger et al. 1998; Pauzat
standing of the PAH emission bands and their use as a diag-
6 & Ellinger 2002). Advances in computing power and com-
nostictoolforprobingthelocalphysicalconditions.
0 putational methods now allow spectra for larger PAHs and
ThePAHemissionbandsshowclearvariationsinpeakpo-
. for a much wider range of PAH species to be calculated and
1 sitions, shapes and (relative) intensities, not only between
0 sources,butalsospatiallywithinextendedsources(e.g.Hony thusforabetterandlargerscanoftheparameterspace(Mu-
7 et al. 2001; Peeters et al. 2002; Smith et al. 2007b; Galliano lasetal.2006;Mallocietal.2007;Bauschlicheretal.2008,
1 et al. 2008). In particular, it is well established that the 3.3 2009;Riccaetal.2010,2012;Candianetal.2014;Candian&
v: and 11.2 µm PAH bands correlate with each other and that Sarre2015). Hence,asystematicstudyoffullyresolvedPAH
emissionathighspatialscaleoverthefullmid-IRbandwidth
i the6.2,7.7and8.6µm PAHbandscorrelatetightlywitheach
X other. Laboratory and theoretical PAH studies have long in- ofalargesampleofobjectsincombinationwithasystematic
investigationoftheoreticalPAHspectrapromisestoreveala
r dicated that the main driver behind these correlations is the
a PAHchargestate: emissionfromneutralPAHsdominatesat more detailed and nuanced view of the characteristics of the
emittingPAHpopulation.
3.3 and 11.2 µm while ionized PAHs emit strongest in the
Thehighsensitivityandbroadwavelengthcoverageofthe
5-10µm andareresponsibleforthe11.0µm band(e.g.Hud-
Spitzer Space Telescope has provided a unique opportunity
gins et al. 2004, and references therein). However, Whelan
to study the variations of the PAH features within extended
sources that have been spectrally mapped. In particular, re-
1DepartmentofPhysicsandAstronomy,UniversityofWesternOntario, flection nebulae (RNe) are very interesting in this respect as
London,ONN6A3K7,Canada;[email protected]
they are known to be excellent laboratories to determine the
2CarlSaganCenter,SETIInstitute,189N.BernardoAvenue,Suite100,
characteristics of interstellar dust and gas due to their high
MountainView,CA94043,USA
3Entry Systems and Technology Division, Mail Stop 230-3, NASA surfacebrightnessandrelativelysimplegeometry. Moreover,
AmesResearchCenter,MoffettField,CA94035,USA in contrast to H ii regions, observations of RNe are not con-
4NASA-Ames Research Center, Space Science Division, Mail Stop fusedbydiffuseemissionduetorecombinationandcoolingof
245-6,MoffettField,CA94035,USA
5LeidenObservatory,POBox9513,2300RALeiden,TheNetherlands ionizedgas,simplifyingtheiranalysisgreatly. Consequently,
6AstronomyDepartment, UniversityofMaryland, CollegePark, MD thewell-knownreflectionnebulaeNGC7023andNGC2023
20742,USA have been prime targets for PAH studies (e.g. Abergel et al.
2 Peetersetal.
2002; Werner et al. 2004b; Sellgren et al. 2007; Compie`gne
et al. 2008; Fleming et al. 2010; Peeters et al. 2012; Pilleri
etal.2012;Boersmaetal.2013,2014;Shannonetal.2015;
Stocketal.2016).
In this paper, we present Spitzer-IRS spectral maps of
NGC 2023 in the 5-20 µm region and theoretical data from
oval, compact PAHs in order to investigate and interpret the
spatialbehaviourofthePAHemissioncomponents,inpartic-
ular in the 6 to 9 µm region. Section 2 discusses the obser-
vations,thedatareduction,typicalspectraandthecontinuum
andfluxdetermination. Thedataareanalyzedintermsoffea-
tures’emissionmapsandcorrelationsinSection3andtheo-
reticaldatafromoval,compactPAHsarepresentedinSection
4.ThesearecombinedanddiscussedinSection5.Finally,we
endwithashortsummaryinSection6.
2. THEDATA
2.1. Observations
TheobservationsweretakenwiththeInfraredSpectrograph
(IRS,Houcketal.2004)onboardtheSpitzerSpaceTelescope
(Werneretal.2004a)andwerepartoftheOpenTimeObser-
vationsPIDs20097,30295and50511.
We obtained spectral maps for two positions in the reflec-
tion nebula NGC 2023 (see Fig. 1): towards the dense shell Figure1. TheIRAC[8.0]imageofNGC2023withtheSLandSHFOV
southsouth-west(-11(cid:48)(cid:48), -78(cid:48)(cid:48))oftheexciting starHD37903 shown(pinkandwhitelinesrespectively)forboththenorthandsouthposi-
correspondingtotheH emissionpeak,andtowardsaregion tionsstudiedhere.ThestarHD37903isindicatedbyablackcircle,sources
2
tothenorth(+33(cid:48)(cid:48),+105(cid:48)(cid:48))oftheexcitingstar(Burtonetal. CandDarefromSellgren(1983), andthewhitecirclesindicate2MASS
pointsourceslocatedinsidetheapertures. Sreferstothesouthernridgelo-
1998). The north position is characterized by a much lower catedinthemiddleoftheFOVofthesouthmap,S’tothesouthernridgeat
density (∼104 cm−3) than that at the south position, where thetopoftheFOVofthesouthSLmap,SEtotheeasternmostridgeintheSL
densities exceed 105 cm−3 (Burton et al. 1998; Sheffer et al. FOVandSSEtothesouth-southeasternridge. MovingtothenorthernFOV,
NreferstothenorthridgeandNWtothenorthwesternridge. Mapsshown
2011). Likewise, the radiation field between 6 and 13.6 eV
inthispaperusethepixelorientationdenotedbythevectors(X,Y).Figure
inthenorthernregion(∼500G07)islowerthantheUVfield adaptedfrompaperI.
wheretheH peaksinthesouthernregion(∼104 G ,Burton
2 0
etal.1998;Shefferetal.2011).
The spectral maps are made with the short-low and short- Table1
Observationlog
high modes (respectively SL and SH). The SL mode covers
awavelengthrangeof5-15µm ataresolutionrangingfrom
60 to 128 in two orders (SL1 and SL2) and has a pixel size northposition southposition
of1.8(cid:48)(cid:48), aslitwidthof3.6(cid:48)(cid:48)andaslitlengthof57(cid:48)(cid:48)(resulting
in2pixelsperslitwidth). TheSHmodecoversawavelength mapcoordinatesa 5:41:40.65,-2:13:47.5 5:41:37.63,-2:16:42.5
rangeof10-20µm ataresolutionof∼600andhasapixelsize mode SL SH SL SH
of2.3(cid:48)(cid:48),aslitwidthof4.7(cid:48)(cid:48)andaslitlengthof11.3(cid:48)(cid:48)(resulting PID 50511 30295 20097
in2pixelsperslitwidth). Weobtainedbackgroundobserva- AORs 26337024 17977856 14033920
tionsinSHforthenorthpositionandinSLforthesouthpo- cyclexramptime 2x14s 2x30s 2x14s 1x30s
sition. Table1givesadetailedoverviewoftheobservations. pointings(cid:107) 1 7 3 12
The15-20µm SHspectrahavebeenpresentedinPeetersetal. stepsize(cid:107) 26(cid:48)(cid:48) 5(cid:48)(cid:48) 26(cid:48)(cid:48) 5.65(cid:48)(cid:48)
(2012),hereafterpaperI. pointings⊥ 20 15 18 12
stepsize⊥ 1.85(cid:48)(cid:48) 2.3(cid:48)(cid:48) 3.6(cid:48)(cid:48) 4.7(cid:48)(cid:48)
2.2. Reduction Backgrounda,b 5:42:1.00,-2:6:54.5 5:40:26.21,-2:54:40.4
The SL raw data were processed with the S18.7 pipeline
aα,δ(J2000)ofthecenterofthemap;unitsofαarehours,
version by the Spitzer Science Center. The resulting bcd-
minutes,andseconds,andunitsofδaredegrees,arcminutes,and
products are further processed using cubism (Smith et al. arcseconds.Theilluminatingstar,HD37903,hascoordinatesof
2007a)availablefromtheSSCwebsite8. 05:41:38.39,-02:15:32.48.
AsdiscussedindetailinpaperI,thebackgroundaddsonlya binSHforthenorthpositionandinSLforthesouthposition.
smallcontributiontotheon-sourcePAHfluxexceptforsource
D from Sellgren (1983). Consequently, we did not apply a
background subtraction and exclude source D from the PAH bad pixels and record bad pixels. We excluded the spurious
analysisfortheremainderofthispaper. data at the extremities of the SL slit by applying a wavsamp
We applied cubism’s automatic bad pixel generation with of 0.05 to 0.95 for the north map and of 0.06 to 0.94 for the
σ =7andMinbad-fraction=0.50and0.75fortheglobal southmap.Remainingbadpixelsweresubsequentlyremoved
TRIM
manually.
7 G0 istheintegrated6to13.6eVradiationfluxinunitsoftheHabing Spectrawereextractedfromthespectralmapsbymoving,
field=1.6x10−3ergs/cm2/s. in one pixel steps, a spectral aperture of 2x2 pixels in both
8http://ssc.spitzer.caltech.edu directionsofthemaps. Thisresultsinoverlappingextraction
PAHemissioninNGC2023 3
Figure2. AtypicalSLspectrumtowardsNGC2023shownwiththediffer- Figure3. AtypicalSL+SHspectrumtowardsNGC2023(SLinblueand
entSLcontinuumextractionmethodsdiscussedhere. Thetoppanelexem- SHinblack;bothtakenwiththesameaperture)shownwiththeappliedSH
plifiesthesplinedecompositionmethodsappliedinthispaper: theplateau continuum(toppanel): thelocalsplinecontinuum(LS,red)andtheplateau
continuum(darkgreen),theglobalcontinuum(GS,magenta),andthelocal continuum(green). NotethatSLisscaledtomatchtheSHfluxbetween10
continuum(LS,red). The5-10and10-15µm plateausareshownbythe and13µm(excludingtheH2line)andnocorrectionforthedifferentspatial
light-green-linedregionsandthe8µm bumpbytheblue-linedregion. The PSFofSLandSHismade.Themiddlepanelshowstheresultingcontinuum
middlepanelshowstheresultingcontinuumsubtractedspectrawhenusing subtractedspectrawhenusingtheplateau,andlocalcontinuum(respectively,
theplateau,globalandlocalcontinuum(respectively,indarkgreen,magenta indarkgreenandred). Forcomparison, thedecompositionobtainedwith
andred).Forcomparison,thedecompositionobtainedwithPAHFITisshown PAHFIT is shown in the bottom panel: fit (red), PAH features (blue), H2
inthebottompanel:fit(red),PAHfeatures(blue),H2lines(green)andcon- lines(green)andcontinuum(magenta).
tinuum(magenta).
µm. ThesePAHbandsareperchedontopofbroademission
apertures. Slightmismatchesinfluxlevelwereseenbetween plateaus from roughly 5–10, 10–15 and, 15–18 µm. C ex-
60
thedifferentordersoftheSLmodule(SL1andSL2). These hibit bands at 7.0, 8.6, 17.4 and 19 µm (Cami et al. 2010;
werecorrectedbyscalingtheSL2datatotheSL1data. The Sellgren et al. 2010). The latter two are clearly present in
appliedscalingfactoraveragedto5and7%forthesouthand thesespectra(paperI).
north SL map respectively. Around 8% of the spectra in the NGC 2023 contains a cluster of young stars (Sellgren
southSLmaphavescalingfactorsofmorethan20%, which 1983). YSO source D is located in the southern quadrant
are all located at high x-values (i.e. at the south side of the of the south SL map and just outside the south SH map (see
map,belowtheSridge). Hence,theappliedscalingdoesnot Fig. 1). The spectra of regions close to source D show PAH
influenceourresults. emission features as well as characteristics typical of YSOs,
We’ve noticed an apparent small (of the order of a few i.e., a strong dust continuum and a (strong) CO ice feature
2
× 0.01 µm) wavelength shift in some 2x2 SL spectra, in near15µm. Furthermore,thebackgroundcontributiontothe
particular for the south map (see Appendix A). While its 11.2 PAH flux is significant for a large fraction of the spec-
effects can be noticed in the top row of the north map and tra containing ice-features. As in paper I, we therefore ex-
in the bottom two rows of the south map in most feature cludedthesespectraintheanalysis. The15-20µm spectrum
intensitymapspresentedinthispaper(Figs.5,6,9,10,A.4, ofYSOsourceCisfoundinboththeSLandSHsouthmap
A.5, and A.11), it does not change the conclusions of this (see Fig. 1). This YSO has the same spectral characteristics
paper. as the spectra across NGC 2023 but with enhanced surface
brightness. However, its SL2 spectrum suffers from instru-
For the reduction of the SH data, we refer to paper I for a mental effects (i.e. a sawtooth pattern likely due to the un-
detaileddiscussion. Spectrawereextractedfromthespectral dersampledIRSPSF).Hence,weexcludedthissourceinthe
mapsbymoving,inonepixelsteps,aspectralapertureof2x2 analysisoftheSLdatabutincludeditfortheanalysisofthe
pixels in both directions of the maps; the same procedure as SHdata.
fortheSLdata.
2.4. Continuumsubtractionandbandfluxes
2.3. Thespectra
TheextinctiontowardsNGC2023isestimatedtobesmall:
Figs. 2and3showtypicalspectratowardsNGC2023. The A = 0.2 − 0.65 mag (for a detailed overview, see Sheffer
K
complete 5-20 µm spectra reveal a weakly rising dust con- et al. 2011). Moreover, little variation in the extinction is
tinuum, H emission lines, C emission, and a plethora of found across the south map except around source C and D
2 60
PAH emission bands. In addition to the main PAH bands at (Pillerietal.2012;Stocketal.2016). AnextinctionA =0.4
K
6.2, 7.7, 8.6 and 11.2 µm, weaker bands are detected at 5.7, magresultsinextinctioncorrectionsof11%forthe8.6and
6.0, 11.0, 12.0, 12.7, 13.5, 14.2, 15.8, 16.4, 17.4 and, 17.8 11.2µm PAHbands(becauseofitsoverlapwiththesilicate
4 Peetersetal.
tinuumpointatroughly8.2µm (Fig.2).Thisaffectstheband
profilesandintensitiesofthe7.7and8.6µm PAHbandsand
theunderlyingplateau. Thisalsocreatesanewplateauunder-
neathonlythe7.7and8.6µm PAHbandsthatisthendefined
bythedifferenceofthelocalsplinecontinuumandtheglobal
continuumandisfurtherreferredtoasthe8µm bump. The
plateau continuum (Fig. 2) is obtained by using continuum
Figure4. AtypicalGaussiandecompositiontoextractthe6.0µm PAHband
pointsatroughly5.5,9.9,10.2,10.4,and14.7µm.Theunder-
(left)andthe11.0µm PAHband(SL:middle,SH:right).Thedataareshown
byasolidblackline,thefitbyasolidredline,thetwoindividualGaussians lying plateaus between 5-10 and 10-15 µm are then defined
bystripedcyanlines,andtheresidualsbyasolidgoldline(offsetby-100 bythedifferenceofthiscontinuumwiththeGSandLScon-
MJy/sr).SeeSect.2.4fordetailsonthecomposition. tinuarespectively. Thefluxesareobtainedinthesamewayas
discussedabovefortheLScontinuum.
band) and between 4-7 % for the other features considered Finally,thethirdapproachemploysPAHFITtoanalyzethe
here. In this paper, we determined the PAH fluxes assuming data(Fig.2,Smithetal.2007b). ThePAHFITdecomposition
zeroextinction. results in components representing the dust continuum
emission (which is a combination of modified blackbodies),
WeappliedthreedifferentdecompositionmethodstotheSL the H emission and the PAH emission. In particular, the
2
data.Forthefirstmethod,wesubtractalocalspline(LS)con- PAH emission is fit by a combination of Drude profiles. A
tinuumfromthespectra,consistentwiththemethodofHony detailed discussion of this decomposition method and its
etal.(2001)andPeetersetal.(2002)asshowninFig.2. This resultscanbefindinAppendixC.
continuum is determined by using anchor points at roughly
5.4, 5.5, 5.8, 6.6, 7.0, 8.2, 9.0, 9.3, 9.9, 10.2, 10.5, 10.7, FortheSHdata,weappliedthelocalsplinecontinuumby
11.7, 12.1, 13.1, 13.9, 14.7, and 15.0 µm. The fluxes of the using anchor points at roughly 10.2, 10.4, 10.8, 11.8, 12.2,
mainPAHbandsarethenestimatedbyintegratingthecontin- 13.0,13.9,14.9,15.2,15.5,16.1,16.7,16.9,17.16/17.19(for
uum subtracted spectra while those of the weaker bands and the south and north maps, respectively), 17.6, 18.14/18.17
theH2 linesaremeasuredbyfittingaGaussianprofiletothe (for the north and south maps, respectively), 18.5, 19.3, and
band/line. The central wavelength of the Gaussian used to 19.4µm andtheplateaucontinuumbyusinganchorpointsat
measuretheH2 linefluxeswasallowedtovary,tocorrectfor 10.4, 15.2, 18.5 and 19.4 µm. The fluxes of the 11.0, 11.2,
the wavelength shifts. The 6.0, 11.0, and 12.7 µm features 12.7µm PAHbandsandthe12.3µm H linearedetermined
2
needspecialtreatmentbecauseofblending. Sincethe6.0and in the same way as for the SL data. The parameters for the
6.2µm PAHfeaturesareblendedandgiventhelowresolution 11.0 and 11.2 Gaussians are λ (FWHM) of 11.018 (0.1205)
oftheSLdata, weextractedthe6.0bandintensitybyfitting and11.243(0.144)µm respectively(Fig.4,rightpanel). The
two Gaussians with λ (FWHM) of 6.026 (0.099) and 6.229 fluxesoftheweaker12.0and13.5µm bandsaredetermined
(0.1612)µm respectivelytothedata(excludingtheredwing by fitting a Gaussian. The 14.2 µm PAH displays a weaker
of the 6.2 PAH band). These values were obtained by tak- blue shoulder. These two components were fitted by a
ingtheaverageoverallspectrawhenfittedbytwoGaussians Gaussian(λ(FWHM)of13.99(0.1178)and14.230(0.1830)
havingpeakpositionsandFWHMthatwerenotfixed.Ascan µm respectively).
be seen in Fig. 4, this decomposition works reasonably well
except, ofcourse, fortheredwingofthe6.2µm band. The TheSLandSHfluxesforfeaturesinthe10-14µm region
6.0 µm PAH flux is then subtracted from the integrated flux tend to differ. This difference arises from the following rea-
ofthe6.0+6.2µm PAHbandstoobtainthe6.2µm PAHflux. sons. Firstly, wedonotregridtheSLdatatothecoarserSH
A similar decomposition method can be applied for the 11.0 grid and do not correct for the different spatial PSFs. How-
and11.2µm bandsbyfittingtwoGaussianswithλ(FWHM) ever, when the SL data are regridded to the SH grid and the
of 10.99 (0.154) and 11.26 (0.236) µm respectively (Fig. 4, SL data are scaled to the SH data but no correction for the
middlepanel). Analogouslytothe6µm region,thisdecom- different spatial PSFs are made, it is clear that the features’
position works remarkably well except for the red wing of strengthdiffertovaryingdegreesintheSLandSHdata(see
the 11.2 µm band. In order to obtain the fluxes of the 12.7 Fig.3).Secondly,thecontinuumislessaccuratelydetermined
µm PAH band and the 12.3 µm H2 line, we fit a second or- intheSLdataduetoblendingoftheemissionfeatures(11.2
der polynomial to the local ‘continuum’ (i.e. the blue wing PAH,12.0PAH,H and12.7PAH)whichstronglyinfluences
2
of the 12.7 µm PAH band) and a Gaussian profile to the H2 the fluxes of the weaker bands. Hence, we will refrain from
line. The flux of the 12.7 µm PAH band was then obtained comparingtheSLandSHfluxesdirectlywitheachotherand
by subtracting the H2 flux from that obtained by the (com- wewillnotusetheweaker12.0, 13.5and14.5µm bandsin
bined) integrated flux of the 12.7 µm PAH band and the H2 theSLdata.
line. Weestimatedthesignal-√to-noiseratioofthefeaturesas Theappliedcontinuumisclearlynotunique.Henceneither
follows: S/N = F / (rms× N ×∆λ) with F the feature’s isthedecompositionofthePAHemission,northecalculated
flux[W/m2/sr],rmsthermsnoise,Nthenumberoffluxmea- bandstrengths.Thus,forcomparison,wedidafullanalysisof
surementswithinthefeatureand∆λthewavelengthbinsize theSLdatausingthesethreemethods.Intheremainderofthe
as determined by the spectral resolution. Although the rms paper,thedefaultappliedcontinuumisthelocalsplinecontin-
noiseisameasureofhowaccuratethecontinuumcanbede- uumfortheintensitiesofthePAHbands(excludingthe5–10
termined,thismethoddoesnottakeintoaccounttheerrorin µm plateau9)unlessstatedotherwise(e.g. Section3.1.2). We
thecontinuummeasurement. will discuss the influence of the continuum and decomposi-
The second method is a modification of the first method.
A global spline (GS) continuum is determined by using the 9Definedasthedifferencebetweentheplateaucontinuumandtheglobal
sameanchorpointsasinthefirstmethodexceptforthecon- splinecontinuum.
PAHemissioninNGC2023 5
Figure5. Southmap:Spatialdistributionoftheemissionfeaturesinthe5-15µm SLdatatowardsNGC2023(usingacutoffvalueof2sigmaandapplyinga
localsplinecontinuumexceptforthe5–10µm plateau).ΣPAHsreferstothecombinedfluxofallPAHfeaturesandthe8µm bump(i.e.excludingthe5-10and
10-15µm plateaus)andVSGreferstothecontinuumfluxat14.7µm. Bandintensitiesaremeasuredinunitsof10−8Wm−2sr−1andcontinuumintensitiesin
unitsofMJysr−1.Asareference,theintensityprofilesofthe11.2and7.7µm emissionfeaturesareshownascontoursinrespectivelyblack(at3.66,4.64,5.64
and,6.7810−6Wm−2sr−1)andpink(at1.40,1.56,1.70and,1.9010−5Wm−2sr−1).ThemapsareorientatedsoNisupandEisleft.Thewhitearrowinthetop
leftcornersindicatesthedirectiontowardsthecentralstar. ArrowsareshownseparatelyfortheS,SE,andSSEridgesinthebottomrightmap. Thewhiteline
acrosstheFOVrepresentsthelinecutusedinFigs16,A.6,andA.12. Theaxislabelsrefertopixelnumbers. RegionsnearsourceCandDexcludedfromthe
analysisaresettozero.ThenomenclatureandtheFOVoftheSHmaparegiveninthebottomrightpanel(seealsoFig.1).
6 Peetersetal.
Figure6. Northmap:Spatialdistributionoftheemissionfeaturesinthe5-15µm SLdatatowardsNGC2023(usingacutoffvalueof2sigmaandapplyinga
localsplinecontinuumexceptforthe5–10µm plateau).ΣPAHsreferstothecombinedfluxofallPAHfeaturesandthe8µm bump(i.e.excludingthe5-10and
10-15µm plateaus)andVSGreferstothecontinuumfluxat14.7µm. Bandintensitiesaremeasuredinunitsof10−8Wm−2sr−1andcontinuumintensitiesin
unitsofMJysr−1.Asareference,theintensityprofilesofthe11.2and7.7µm emissionfeaturesareshownascontoursinrespectivelyblack(at1.42,1.57,1.75
and,1.9910−6Wm−2sr−1)andpink(at5.54,6.30,6.80and,7.2010−6Wm−2sr−1). ThemapsareorientatedsoNisupandEisleft. Thewhitearrowinthe
bottomrightcornersindicatesthedirectiontowardsthecentralstar. ArrowsareshownseparatelyfortheNandNWridgesinthebottomrightmap. Thewhite
lineacrosstheFOVrepresentsthelinecutusedinFigs16,A.6,andA.12.Theaxislabelsrefertopixelnumbers.ThenomenclatureandtheFOVoftheSHmap
aregiveninthebottomrightpanel(seealsoFig.1).
tionontheresultswherenecessary. restrictourselvestothesouthmap.
3. DATAANALYSIS 3.1. SLdata
Here we investigate the relationship between individual Thespatialdistributionofthevariousemissioncomponents
PAH emission bands, the underlying plateaus, H emission, inthe5-15µm SLdataareshowninFigs. 5and6(therange
2
C emission,andthedustcontinuumemissionpresentinthe in colours is set by the minimum and maximum intensities
60
5-20µm region.Thenorthmapischaracterizedbylowerflux present in the map, and a local spline continuum is applied
levelsandthereforelargerscatterispresentinthemaps/plots except for the 5–10 µm plateau) and feature correlations in
tracing weaker PAH features. Hence, when discussing the Fig.7(withalocalsplinecontinuaappliedexceptforthe5–
spatial distribution of the 14.2, 15.8 and 17.8 µm bands, we 10µm plateauandtheGaussiancomponentsasdiscussedin
PAHemissioninNGC2023 7
Section 3.1.2). To exclude the influence of PAH abundance N ridge. These S(2) and S(3) distributions are confirmed by
and column density in the correlation plots, we normalized themapsobtainedwithPAHFITforboththenorthandsouth
thebandfluxestothatofanotherPAHband. FOVs(seeAppendixC).
Fig. 7 shows observed intensity correlations; their fit pa-
3.1.1. Overallappearance rametersandcorrelationcoefficientscanbefoundintableD
The following trends are derived from the south spectral and their line cuts in Fig. A.7. The well known, very tight
maps shown in Fig. 5. The 11.2 µm feature, the 5-10 and correlation between the 6.2 and 7.7 µm PAH bands is also
10-15 µm plateaus and the continuum emission show very observedwithinoursample,consistentwiththeirsimilarspa-
similar spatial morphology with distinct peaks at the S and tialmorphology.Notethattheobservedcorrelationiscloseto
SSEpositions.The8µm bumpexhibitsasimilarmorphology a1:1correlation(i.e. through(0,0)). Surprisingly,the8.6µm
butshowsmoreenhancedemissionattheSEridgeandwestof band correlates with the 6.2 and 7.7 µm bands despite their
theS’ridge.ThePDRfrontiswelltracedbytheH emission, differingspatialdistributions. However,thiscorrelationisnot
2
which also clearly peaks at the S and SSE positions and is astightasthatbetweenthe6.2and7.7µm bands,butoverall,
heavilyconcentratedalongthesetworidgesonly. Incontrast, inkeepingwiththeirclosebutdifferingspatialdistributions.
thedistributionofthe6.2,7.7,12.7µm featuresaredisplaced The enhanced scatter in the correlation of the 8.6 µm band
towardstheilluminatingstarandawayfromtheSridge,with with the 6.2 and 7.7 µm bands has regularly been attributed
the loss of emission at the S position accompanied by a rise totheinfluenceofextinctionand/orthelargeruncertaintyin
at the S’ and SE ridges. Put another way, the 6.2, 7.7 and determining the 8.6 µm band intensity. However, we found
12.7µm featuresshowverysimilarspatialmorphologywith thatlargerdeviationsfromthelineofbestfitarefoundinlo-
distinctpeaksattheSSEandSEridgesandwithweakerpeaks cationswherethespatialdistributionofthebandsisdifferent,
attheSandS’positions.Inaddition,theyshowbroad,diffuse i.e.,theobservedscatteroriginatesinthedistinctspatialdis-
emission NW of the line connecting the S and SSE ridges. tributions. The 11.0 µm PAH band correlates best with the
The8.6µm bandisfurtherdisplacedtowardstheilluminating 8.6 µm PAH band although not as tight as the 6.2 with the
star: it peaks at the SE and S’ position but does not peak at 7.7µm PAHbands. The11.0vs. 8.6correlationexhibitsex-
theSandSSEridgesasdothe6.2,7.7and12.7µm emission. actlya1:1dependence(i.e.thebestfitgoesthrough(0,0);see
Infact,itlacksemissionintheSridge. Similartothe6.2,7.7 Fig.7)andhasahighcorrelationcoefficient. Thismaynotbe
and12.7µm emission, itdoesshowabroad, diffuseplateau immediatelyclearfromtheirspatialdistributionforthesouth
NWofthelineconnectingtheSandSSEridges. Asdoesthe mapasshowninFig.5whichisattributedtothebottomtwo
8.6µm PAHemission,the11.0µm PAHemissionalsopeaks rows (in y-direction; see Appendix B for details). The 11.0
attheSEandS’ridgesandlacksemissionintheSridge. But µm PAH band also correlates with the 6.2 and 7.7 µm PAH
itisalsoverystrongatthebroad, diffuseplateauNandNW bandsbutwithslightlymorescatter.Hereaswell,the11.0µm
ofthelineconnectingtheSandSSEridges.The6.0µm PAH PAHemissionhasadifferentspatialdistributionthanthe8.6,
emissionpeaksattheSandSSEridgesbutitspeaksseemto 6.2and7.7µm PAHemissionresultinginenhancedscatterin
bedisplacedtowardstheeastcomparedtothe11.2µm PAH thesecorrelationplotsandthusalowercorrelationcoefficient
emission. It has weaker emission maxima in the form of an (ranging from 0.949 to 0.958 compared to 0.978 for the 6.2
arcsouthoftheS’ridgeandwestoftheSEridge. Hence,6.0 vs. 7.7correlation). The12.7µm PAHbandalsoshowsade-
µm PAHemissionseemstobesomewhatunique. pendenceonthe6.2,7.7,8.6and11.0µm PAHbandsbutthis
Thevarietyinthespatialdistributionofthedifferentemis- connectionisclearlyweakerthanthoseamongstthe6.2,7.7,
sion components is even more pronounced in the north map 8.6and11.0µm bands(withcorrelationcoefficientsranging
(Fig.6). The11.2µm PAHemissionpeaksintheNWridge. between0.930to0.945).
In contrast with the south map, the continuum flux and the
10-15µm plateauaredisplacedfromthe11.2µm PAHemis- 3.1.2. Decompositionofthe7to9µm region
sionandpeaksintheNridgewhilethe5-10µm plateauand
Thedistinctspatialdistributionofthe7.7and8.6µm PAH
the8µm bumptraceboththeNand(partof)theNWridge.
bands prompts further investigation. As discussed in Sect.
The6.2and7.7µm PAHemissionareagainverysimilarand
2.4,thechosenlocalsplinecontinuumclearlyinfluencesband
differfromthe11.2µm PAHemission. Whiletheyalsopeak
intensities. However,ifanemissionfeatureisduetoasingle
intheNWridgeasdoesthe11.2µm PAHfeature,theyshow
carrier or distinct subset of the PAH population, the spatial
decreased emission in the northern part of this ridge and ex-
distributionsofitssub-componentsshouldallbeidentical,in-
tendtowardsthewestinthesouthernpartofthisridge(note
dependent of how these sub-components have been defined.
thedifferenceintheblackandpinkcontourswhichtracethe
Hence,thedistinctspatialdistributionofthe7.7and8.6µm
11.2and7.7µm PAHbandsrespectively). Themorphology
PAHemissionindicatesthattheyoriginateinmultiplecarri-
ofthe12.7µm PAHemissionisabitofboththatofthe11.2
ers or loosely related PAH subpopulations. In an attempt to
and 7.7 µm PAH emission. The 8.6 µm emission is distinct
resolvethesub-componentsofthe7to9µm PAHemission,
from the 6.2 and 7.7 µm PAH emission and peaks slightly
each related to a single carrier or subset of the PAH popu-
westofthesouthernpartoftheNWridge.Thisisalsoseenin
lation, we subtracted the global spline continuum (GS, see
thespatialdistributionofthe11.0µm PAHwhichpeakseven
Fig.2,magentaline)fromthespectraanddecomposedthere-
further west of the southern part of the NW ridge compared
maining PAH emission into four Gaussians with λ (FWHM)
to the 8.6 µm PAH emission. The 6.0 µm PAH emission in
of 7.59 (0.450), 7.93 (0.300), 8.25 (0.270), and 8.58 (0.344)
thenorthmap,asforthesouthmap,hasauniquemorphology
µm respectively(seeFig.8)10.Thesevalueswereobtainedby
whichishighlyconcentratedandpeaksattheintersectionof
the N and NW ridge. The spatial distribution of the H line
2 10NotethattheG8.2componentisnotpresentwhenusingthelocalcon-
intensities varies as well: the S(3) [9.7 µm] intensity peaks
tinuum(LS,seeFig.2,orangeline)asforthiscontinuum,ananchorpointat
betweenthemaximainthe11.2and7.7µm PAHintensities 8.2µm istakenandthiscomponentthusbecomespartofthe8µm bump.
along NW ridge and the S(2) at 12.3 µm intensity peaks in Whenusingtheplateaucontinuum(seeFig.2,greenline),theG8.2compo-
8 Peetersetal.
Figure7. CorrelationsinPAHSLintensityratiosacrossNGC2023.Theredsquaresareforthesouthmapandthebluetrianglesforthenorthmap.Weapplied
acutoffof3sigmaontheindividualfeatures.Theweightedcorrelationcoefficientisgiveninthetopleftcornerforboththenorthandsouthmap(black),forthe
southmap(red)andforthenorthmap(blue). TheLevenberg-Marquardtleast-squaresminimizationfittotheN+Sdataisshownasasolidline. Thedashed
linerepresentsafittothedataforcedthrough(0,0).ThefitparameterscanbefoundinTableA.1.NotethatwechangedthenormalizationfortheG7.8andG8.2
componentsfromthe11.2µm bandtotheG8.6componentsincetheformertwocomponentsexhibitaspatialmorphologysimilartothatofthe11.2µm band
inthesouthmap. 7.7(LS)and8.6(LS)refertothe7.7and8.6µm PAHbandswhenapplyingalocalspline(LS)continuumandplat. 7referstothe5–10µm
plateau.
PAHemissioninNGC2023 9
Figure9. SpatialdistributionofthefourGaussiancomponentsusedinthedecompositionofthe7to9µm regionwhenapplyingaglobalsplinecontinuumfor
thesouthmap(toppanels)andnorthmap(bottompanels).Asareference,theintensityprofilesofthe11.2and7.7µm emissionfeaturesareshownascontours
inblackandpink,respectively.Units,maporientationandsymbolsarethesameasinFigs.5and6.
needed to obtain a good fit in the 7 – 9 µm region. We did
notincludeaGaussianat7.4µm becausenoneofthespectra
inthemapshowa’feature’near7.4µm likethatobservedin
the ISO-SWS data of NGC 7023 (Moutou et al. 1999). The
‘nominal’7.7µm PAHcomplexisdominantlycomprisedof
theG7.6componentandonlyoriginatesbyarelativelysmall
fraction in the G7.8 component (see Appendix E). The G7.6
andG8.6µm componentsexhibitalmostidenticalspatialdis-
tributionsindicatingthereisnosignificantcontributionfrom
an additional, spatially distinct component in either of the
G7.6 and G8.6 µm components (Fig. 9). Both components
peak at the SE, SSE and S’ ridges in the south map. In the
Figure8. A typical Gaussian decomposition of the continuum subtracted
spectrum(usingtheglobalsplinecontinuum)inthe7–9µm regioninto northern FOV, they peak at the southern and centre portion
theG7.6,G7.8,G8.2andG8.6components. Thedataareshownbyasolid of the NW ridge, with an extension towards the west in the
blackline,thefitbyasolidredline,thefourindividualGaussiansbystriped southern part of this ridge. The G7.8 and G8.2 µm compo-
cyanlines,andtheresidualsbyasolidgoldline. Thelatterareoffsetby-
100MJy/sr. Greenstripedlinesindicate7.35and8.1µm. SeeSect. 3.1.2 nentsalsohaveasimilarspatialdistribution,thoughdiscrep-
for details on the composition. The four Gaussians have a peak position ancies are present (Fig. 9). They peak at the S and SSE
(FWHM)of7.59(0.45), 7.93(0.3), 8.25(0.27), and8.58(0.344)µm re- ridgesinthesouthmap,similartothe11.2µm emissionand
spectively. the 5-10 and 10-15 µm plateau emission. In the north map,
theytraceboththeN,centreandnorthernportionoftheNW
takingtheaverageoverallspectrawhenfittedbyfourGaus-
ridge, whichissimilartothe10-15µm plateauandthe10.2
sians having peak positions and FWHM that were not fixed
µm continuumemission. Asreportedfortheratioofthe7.6
but were constrained to fit these four bands. These compo-
and7.8µm subcomponents(Bregman&Temi2005;Rapaci-
nents are further referred to as G7.6, G7.8, G8.2 and G8.6
oli et al. 2005; Boersma et al. 2014), the ratio of G7.6/G7.8
bands. We chose four components to represent the 8.6 µm
tracesthedifferentenvironmentsverywell:itisclearlysmall-
PAHbandandthe7.6and7.8µm subcomponentsofthe7.7
µm complex, whose ratio, 7.6/7.8, exhibit spatial variation est where H2 emission peaks and inside the molecular cloud
(seeAppendixE).
withdistancefromthestar(Bregman&Temi2005;Rapaci-
Theseresultsarealsofoundinthecorrelationplots(Fig.7).
olietal.2005;Boersmaetal.2014). Afourthcomponentis
TheG7.6andG8.6µm componentsexhibitaverytightcorre-
lation(bestcorrelationcoefficientof0.987)whichcloselyre-
nentsitsontopoftheplateauemission.
10 Peetersetal.
map12. For the south map, the extreme ∼7.35 µm emission
isspatiallyverysimilartothatofthe11.0µm PAHemission
(Fig. 5): it traces the S’, SE and SSE ridges, the horizontal
filamentsinthenorthernregionofthemapandthebroad,dif-
fuseplateauNWofthelineconnectingtheSandSSEridges.
The extreme ∼8.1 µm emission is similar to the G8.2 com-
ponent (Fig. 9) and peaks at the S and SSE ridges. So does
the11.2µm PAHemission(Fig.5)buttheextreme∼8.1µm
emissionhasenhancedemissioninthehorizontalfilamentsin
thenorthernregionofthemapandthebroad,diffuseplateau
comparedtothe11.2µm PAHemission. Forthenorthmap,
the extreme distribution at ∼7.35 µm peaks at the southern
part of the NW ridge and the extension towards the west of
theNWridgelikethe8.6and11.0µm PAHemission(Fig.6)
and the G8.6 component (Fig. 9). In contrast, the extreme
∼8.1µm emissionpeaksintheNridgeasdoesthe10.2µm
continuum(Fig.6)andtheG7.8component(Fig.9).
3.1.3. ImplicationsforotherPAHbands
The morphology of the PAH emission changes continu-
ouslywithwavelengthforallmajorPAHbands(i.e. the6.2,
11.2 and, 12.7 µm bands), as for the south map, albeit to a
considerablelesserextent(Fig.10): noneofthewavelengths
exhibitamorphologyasextremeasthatofthe∼7.35µm or
the∼8.1µm extremes.
At ∼6.0 µm, the morphology is similar to that of 6.0 PAH
showninFigs.5and6forbothmaps.Forthesouthmap,with
increasingwavelengthfrom6.0to6.2µm,emissionincreases
in the SE ridge, followed by increased diffuse emission NW
of the line connecting the SSE and S ridge, and increased
Figure10. Asetofstillframesfromananimationshowingspatialmaps emission in the S’ ridge. This is accompanied by decreased
oftheintensityIν ofthecontinuum-subtractedspectra(applyingtheglobal emissionintheSridgeresultinginaspatialdistributionvery
splinecontinuum)asafunctionofwavelengthforthesouthmap(toppanels) similartothatoftheintegrated6.2PAHemissionmap. Sub-
andnorthmap(bottompanels).Thetwoimagesshowthetwoextremespatial sequently, for longer wavelengths (>∼6.25 µm), the diffuse
distributionsfoundinthe7to9µm region.Asareference,the11.2and7.7
µm bandintensitiesareshownascontoursinblackandpink,respectively. emissionandtheemissionintheS’ridgedecreasesandtheS
Units,maporientationandsymbolsarethesameasinFigs. 5and6. The ridge becomes a tiny bit stronger while the peak emission is
fullanimatedsouthandnorthmapsareavailableintheonlineversionofthe foundintheSEandSSEridge. Consideringthenorthmap,at
Journal.Intheanimatedmapstheindividual6-7,7-9and10.8-13µm maps
beginatthet=0,t=14andt=50sectimestamps,respectively. ∼6.1µm,theemissionpeaksintheNWridge. Withincreas-
ing wavelength up to 6.2 µm, emission in the southern part
semblesa1:1relation(i.e. itgoesthrough(0,0)). Incontrast, andwestofthesouthernpartoftheNWridgeincreaseswhile
the correlation of the G7.8 and G8.2 µm Gaussian compo- thenorthernpartoftheNWridgesfadesalittle, resultingin
nentsshowsmorescatterresultinginacorrelationcoefficient a morphology somewhat between the 6.2 PAH and the G8.6
of0.833.Remarkably,thisenhancedscatterlargelyoriginates emission. Subsequently,theNWridgebecomesabitstronger
in the south map. Indeed, the correlation coefficient for the and eventually, the emission west of the southern part of the
southmapis0.817whilethatforthenorthmapisawhopping NWridgefadesaway.
0.96511. Thisscatteroriginatesintheslightmismatchoftheir The most dramatic change in the 11 µm region is found
spatial distributions and either indicates the shortcomings of from11.07to11.14µm withtheswitchfromthe11.0to11.2
ourdecompositionorisduetothefactthattheyariseindif- PAHemission. Themorphologiesnear11.0and11.2µm,as
ferentPAHsub-populations. in the 6 µm region, are well represented by that of the 11.0
Despitethearbitrarinessofthisdecomposition(byassum- PAHemissionand11.2PAHemissionrespectively. Oncebe-
ing a decomposition into four Gaussians), we can conclude yond the peak of the 11.2 PAH band, the morphology con-
that at least 2 spatially distinct components contribute to the tinues to vary, albeit to a less dramatic extent, and becomes
PAH emission in the 7 to 9 µm region. In this respect, it is more sharply peaked. In the south map, the emission peaks
veryenlighteningtowatchananimationshowingthechange sharplyattheSandSSEridges(andweakeratS’ridge),very
in spatial distribution, as a function of wavelength, of the similarly to the H emission. In contrast, for the north map,
2
continuum-subtracted emission (applying the global spline theemissionpeaksinthenorthernpartoftheNWridgewhile
continuum, Fig.10; animationisavailableonline). Thespa- theremainingoftheNWridgehassimilarintensityastheN
tialdistributionofthePAHemissioninthe7to9µm region
continuously varies between two extremes which are found 12 Notethatthemorphologyofthetwoextremesdependsonthechosen
at around ∼7.35 and ∼8.1 µm in both the north and south continuum(here,theglobalsplinecontinuum).Usinginsteadthelocalspline
continuumortheplateaucontinuumwillremovethecontributioncaptured
bythe8µm bumpintheformercaseandaddthecontributionofthe5-10µm
11Thesameholdswhennormalizedtothe6.2µm bandwithcorrelation plateauinthelattercase. Thesecomponentshaveadifferentmorphology
coefficients of 0.823, 0.787 and 0.961 for respectively both the north and asthatofthe7.35µm extremeandaremoresimilartothatofthe8.1µm
southmaps extreme.