The earliest phases of star formation - A Herschel key project. The thermal structure of low-mass molecular cloud cores

4.5 MB·English

by  R. Launhardt

Astronomy&Astrophysicsmanuscriptno.epos˙lm˙Jan2013˙acc˙astroph (cid:13)c ESO2013
December11,2013
The earliest phases of star formation - A Herschel(cid:63) key project
The thermal structure of low-mass molecular cloud cores(cid:63)(cid:63)
R.Launhardt1,A.M.Stutz1,A.Schmiedeke2,1,Th.Henning1,O.Krause1,Z.Balog1,H.Beuther1,S.Birkmann3,1,M.
Hennemann4,1,J.Kainulainen1,T.Khanzadyan5,H.Linz1,N.Lippok1,M.Nielbock1,J.Pitann1,S.Ragan1,C.
Risacher6,5,M.Schmalzl7,1,Y.L.Shirley8,B.Stecklum9,J.Steinacker10,1,andJ.Tackenberg1
1 Max-Planck-Institutfu¨rAstronomie(MPIA),Ko¨nigstuhl17,D-69117Heidelberg,Germany
e-mail:[email protected]
2 Universita¨tzuKo¨ln,Zu¨lpicherStrasse77,D-50937Ko¨ln,Germany
3 ESA/ESTEC,Keplerlaan1,Postbus299,2200AGNoordwijk,TheNetherlands
4 Laboratoire AIM Paris-Saclay, Service d’Astrophysique, CEA/IRFU – CNRS/INSU – Universite´ Paris Diderot, Orme des
3
MerisiersBat.709,91191Gif-sur-YvetteCedex,France
1
5 Max-Planck-Institutfu¨rRadioastronomy(MPIfR),AufdemHu¨gel69,D-53121Bonn,Germany
0
6 SRONNetherlandsInstituteforSpaceResearch,POBox800,9700AVGroningen,TheNetherlands
2
7 LeidenObservatory,LeidenUniversity,POBox9513,2300RA,Leiden,TheNetherlands

n 8 StewardObservatory,933NorthCherryAvenue,Tucson,AZ85721,USA
a 9 Thu¨ringerLandessternwarteTautenburg,Sternwarte5,d-07778Tautenburg,Germany
J 10 InstitutdePlane´tologieetd’AstrophysiquedeGrenoble,Universite´deGrenoble,BP53,F-38041GrenobleCe´dex9,France

8
ReceivedOctober1,2012;accepted

]
R ABSTRACT
S
Context.Thetemperatureanddensitystructureofmolecularcloudcoresarethemostimportantphysicalquantitiesthatdetermine
.
h the course of the protostellar collapse and the properties of the stars they form. Nevertheless, density profiles often rely either on
p thesimplifyingassumptionofisothermalityoronobservationallypoorlyconstrainedmodeltemperatureprofiles.Theinstrumentsof
- theHerschelsatelliteprovideusforthefirsttimewithboththespectralcoverageandthespatialresolutionthatisneededtodirectly
o
measurethedusttemperaturestructureofnearbymolecularcloudcores.
r
t Aims.Withtheaimofbetterconstrainingtheinitialphysicalconditionsinmolecularcloudcoresattheonsetofprotostellarcollapse,
s inparticularofmeasuringtheirtemperaturestructure,weinitiatedtheGuaranteedTimeKeyProject(GTKP)“TheEarliestPhasesof
a
StarFormation”(EPoS)withtheHerschelsatellite.Thispapergivesanoverviewofthelow-masssourcesintheEPoSproject,the
[
Herschelandcomplementaryground-basedobservations,ouranalysismethod,andtheinitialresultsofthesurvey.

1 Methods.Westudythethermaldustemissionof12previouslywell-characterized,isolated,nearbyglobulesusingFIRandsubmm
v continuummapsatuptoeightwavelengthsbetween100µmand1.2mm.Oursamplecontainsbothglobuleswithstarlesscoresand
8 embeddedprotostarsatdifferentearlyevolutionarystages.ThedustemissionmapsareusedtoextractspatiallyresolvedSEDs,which
9 arethenfitindependentlywithmodifiedblackbodycurvestoobtainline-of-sight-averageddusttemperatureandcolumndensitymaps.
4 Results.Wefindthatthethermalstructureofallglobules(meanmass7M(cid:12))isdominatedbyexternalheatingfromtheinterstellar
1 radiationfieldandmoderateshieldingbythinextendedhalos.Allglobuleshavewarmouterenvelopes(14-20K)andcolderdense
. interiors(8-12K)withcolumndensitiesofafew1022cm−2.Theprotostarsembeddedinsomeoftheglobulesraisethelocaltem-
1
peratureofthedensecoresonlywithinradiiouttoabout5000AU,butdonotsignificantlyaffecttheoverallthermalbalanceofthe
0
globules.Fiveoutofthesixstarlesscoresinthesamplearegravitationallyboundandapproximatelythermallystabilized.Thestarless
3
coreinCB244isfoundtobesupercriticalandisspeculatedtobeonthevergeofcollapse.Forthefirsttime,wecannowalsoinclude
1
externallyheatedstarlesscoresintheL /L vs.T diagramandfindthatT <25Kseemstobearobustcriteriontodistinguish
: smm bol bol bol
v starlessfromprotostellarcores,includingthosethatonlyhaveanembeddedverylow-luminosityobject.
i
X Keywords.Stars:formation,low-mass,protostars–ISM:clouds,dust–Infrared:ISM–Submillimeter:ISM
r
a
1. Introduction and scientific goals
(cid:63) Herschel is an ESA space observatory with science instruments Theformationofstarsfromdiffuseinterstellarmatterisoneof
providedbyEuropean-ledPrincipalInvestigatorconsortiaandwithim- the most fundamental and fascinating transformation processes
portantparticipationfromNASA. intheuniverse.Starsformthroughthegravitationalcollapseof
(cid:63)(cid:63) Partially based on observations carried out with the IRAM 30m the densest and coldest cores inside molecular clouds. The ini-
Telescope,withtheAtacamaPathfinderExperiment(APEX),andwith
tialtemperatureanddensitystructureofsuchcoresarethemost
the James Clerk Maxwell Telescope (JCMT). IRAM is supported
important physical quantities that determine the course of the
by INSU/CNRS (France), MPG (Germany) and IGN (Spain). APEX
collapse and its stellar end product (e.g., Larson 1969; Penston
is a collaboration between Max Planck Institut fu¨r Radioastronomie
(MPIfR),OnsalaSpaceObservatory(OSO),andtheEuropeanSouthern
Observatory (ESO). The JCMT is operated by the Joint Astronomy Council of the United Kingdom, the Netherlands Association for
Centre on behalf of the Particle Physics and Astronomy Research ScientificResearch,andtheNationalResearchCouncilofCanada.
1
R.Launhardtetal.:Thestructureoflow-masscloudcores
1969; Shu 1977; Shu et al. 1987; Commerc¸on et al. 2010); 70µm (Bacmann et al. 2000; Stutz et al. 2009a,b; Tobin et al.
however, deriving the physical properties of such cores from 2010). One of the main results from these studies is that the
observations and thus constraining theoretical collapse models LoS-projectedgeometryoftendrasticallydepartsfromspherical
is a nontrivial problem. More than 98% of the core mass is geometrybothintheprestellarandprotostellarphases.
locked up in H molecules and in He atoms that do not have Observations of scattered interstellar radiation in the form
2
a permanent electric dipole moment and that therefore do not of cloudshine, although permitting certain diagnostics of cloud
radiate when cold. The excitation of ro-vibrational and elec- structure (Padoan et al. 2006; Juvela et al. 2006), also do not
tronic states of H requires temperatures much higher than are trace the interiors of dense cores. Scattered MIR interstellar
2
presentinthesecores(e.g.,Burton1992).Thisproblemisusu- light can penetrate most low-mass cores producing coreshine
ally overcome by observing radiation from heavier asymmetric (Steinackeretal.2010).But,theeffectrequiresthepresenceof
molecules, which are much less abundant, but have rotational larger dust grains for scattering to be efficient and is visible in
transitions that are easily excited via collisions with hydrogen only half of the cores (Pagani et al. 2010). Beside the advan-
molecules (e.g., CO, CS; Bergin & Tafalla 2007). However, tage of an only weak dependence on temperature effects, scat-
at the high densities (n ≥ 106cm−3) and low temperatures tered light is also sensitive to the 3D structure of the core and
H
(T ≤ 10K; e.g., Crapsi et al. 2007) inside such cloud cores, couldallowtotracetheactualdensitystructure.Buttousethis
most of these heavier molecules freeze out from the gas phase information, the outer radiation field and the scattering phase
and settle on dust grains, where they remain ’invisible’ (e.g., functionofthedustparticlesneedstobeknownpreciselywhich
Bergin&Langer1997;Charnley1997;Casellietal.1999;Hily- posesaproblemformanycoresinthecomplexenvironmentof
Blant et al. 2010). The few remaining ’slow depleters’ (e.g., star-formationregions.
NH3 or N2H+, ncrit(1−0) ∼ 105cm−3) or deuterated molecular Hence, the thermal emission from dust grains remains as a
ionsthatprofitfromthegasphasedepletionofCO(e.g.,N2D+ superior tracer of matter in the coldest and densest molecular
and DCO+; Caselli et al. 2002) can be used to identify prestel- cloudcoreswherestarsform.Consequently,mostofourcurrent
lar cores and infer their kinematical and chemical properties. informationonthedensitystructureofprestellarcloudcoresand
However,therespectiveobservationaldata,inparticularforthe protostellarenvelopescomesfromsubmmandmmdustcontin-
deuterated species, are still sparse and their complex chemical uummaps(e.g.,Ward-Thompsonetal.1994,1999;Launhardt&
evolution is not yet understood sufficiently well to derive reli- Henning 1997; Henning & Launhardt 1998; Evans et al. 2001;
able density profiles. A very promising and robust alternative Motte & Andre´ 2001; Shirley et al. 2002; Andre´ et al. 2004;
tracer for the matter in such cores is therefore the dust, which Kirketal.2005;Kauffmannetal.2008;Launhardtetal.2010).
constitutes about 1% of the total mass of the interstellar matter Nevertheless, ultimately we will need to combine information
(ISM)inthesolarneighborhood. from different tracers and methods to calibrate the dust opaci-
Interstellar dust can be traced by its effect on the attenua- ties as well as to derive quantities that cannot be derived from
tion of background light (Lada et al. 1994), scattering of am- thedustemissionalone(seediscussionandoutlookattheendof
bient starlight (”cloudshine”: Foster & Goodman 2006; ”core- Sect.7).
shine”:Steinackeretal.2010;Paganietal.2010),orbyitsther- To relate the thermal dust emission at a given wavelength
malemission(e.g.,Smithetal.1979).Inprinciple,itisprefer- to the mass of the emitting matter, three main pieces of infor-
abletouseextinctionmeasurementstotracethecolumndensity mationareneeded:thetemperatureofthedustgrains,T ,their
d
because they are to first order independent of the dust temper- mass absorption coefficient, κ , and the gas-to-dust mass ratio.
ν
ature (unlike emission). However, extinction measurements re- Themeanopacityofthedustmixtureisafunctionofgrainop-
quirethatatleastsomemeasurableamountoflightfromback- tical properties, wavelength, and temperature (e.g., Henning &
ground sources passes through the cloud, which is no longer Stognienko 1996; Agladze et al. 1996; Draine 2003; Boudet et
the case at the highest column densities in the core centers al.2005)andmayactuallyvarylocallyandintimeasafunction
(N > few×1022cm−2). Another limitation of this technique of the physical conditions. However, observational constraints
H
istheinabilitytoexactlymeasuretheextinctiontowardindivid- ofsuchopacityvariationsareverydifficulttoobtainduetovar-
ualsourceswithwithaprioriunknownSEDs,andthatitrelies ious degeneracies and often insufficient data (e.g., Shetty et al.
on statistically relevant ensembles of background stars, which 2009a,b; Shirley et al. 2011; Kelly et al. 2012). Therefore, the
restrictsthismethodtofieldswithhighbackgroundstardensity conversionoffluxintomassisusuallydonebyadoptingoneor
(i.e., close to the galactic plane) and limits the effective angu- the other ’standard’ dust model that is assumed to be constant
larresolution.Consequently,mostattemptstoderivethedensity throughouttheregionofinterest.
structure of cloud cores from extinction maps have been done The definition of a common temperature for the dust re-
in the near-infrared (NIR) and have targeted less-opaque cloud quires that the grains are in local thermal equilibrium (LTE),
coresandenvelopes(AV ≤20mag;e.g.,Alvesetal.2001;Lada whichisfortunatelythecase,tofirstorder,inthewell-shielded
etal.2004;Kandorietal.2005;Kainulainenetal.2006).Going dense interiors of molecular cloud cores where collisions with
to wavelengths longer than optical or NIR, where the opacity H molecules occur frequently and stochastic heating of indi-
2
islower,alsousuallydoesnothelpmuchsincethebackground vidual grains (timescales τ << τ ), e.g., by UV photons,
cool heat
starsalsobecomedimmeratlongerwavelengths. doesnotplayasignificantrole(e.g.,Siebenmorgenetal.1992;
InadditiontotheconventionalNIRextinctionmappingtech- Pavlyuchenkovetal.2012).However,withtypicaldusttemper-
niques,midandfar-IR(MIR,FIR)shadows,orabsorptionfea- aturesinsidesuchdensecoresbeingintherange8−15K(e.g.,
tureswherethedensestportionsofcoresareobservedinabsorp- di Francesco et al. 2007; Bergin & Tafalla 2007), the submm
tion against the background interstellar radiation field (ISRF), emission represents only the Rayleigh-Jeans tail of the Planck
can be used to map structure at high resolution. Provided the spectrum. Therefore, the dust temperature is only very poorly
absolute background level in the images can be accurately cal- constrained by these data, which results in large uncertainties
ibrated, they provide good measurements of the line-of-sight in the derived total masses and in the radial density profiles.
(LoS) projected structure and column density in very dense re- E.g., at λ = 1mm, where the thermal emission of low-mass
gions.Suchfeatureshavebeenusedtostudycoresat8,24,and cores is practically always optically thin, more than twice the
2
R.Launhardtetal.:Thestructureoflow-masscloudcores
amount of 8K dust is needed to emit the same flux density as
12K dust. Hence, a dust temperature estimate of 10±2K still
leaves the mass and the (local) column density uncertain to a
factor of two. The effect of unaccounted temperature gradients
onthederiveddensityprofileisevenmoresignificantandleads
to large uncertainties in the observational constraints on proto-
stellar collapse models. Earlier attempts to reconstruct the dust
temperatureprofilesofstarlessandstar-formingcoresfrom,e.g.,
SCUBA450/850µmfluxratiomapsorself-consistentradiative
transfermodelingarediscussedinSect.6.3,butremainobserva-
tionallypoorlyconstrainedanduncertain.
Fig.1. All-sky map (gnomonic projection), showing the mean
stellar K-band flux density distribution of the Milky Way
(grayscale,derivedfromCOBE-DIRBENIRall-skymaps2)and
This situation is currently improving as new observing ca-
thelocationofourtargetsources(seeTable1).
pabilities start to provide data that can potentially help to
better constrain the dust temperatures. The Herschel Space
Observatorywithitsimagingphotometersthatcoverthewave-
lengths range from 70 to 500µm (Pilbratt et al. 2010) provides
forthefirsttimetheopportunitytofullysampletheSEDofthe
thermal dust emission from these cold objects1 at a spatial res-
olutionthatisbothappropriatetoresolvethesecoresandcom-
parable to that of the largest ground-based submm telescopes,
which extend the wavelength coverage into the Rayleigh-Jeans
regimeoftheSEDs.
TomakeuseofHerschel’snewcapabilitiesandimproveour
knowledge on the initial conditions of star formation by over-
comingsomeoftheabove-mentionedlimitations,wehaveiniti-
atedtheGTKP”TheEarliestPhasesofStarFormation”(EPoS).
The main goal of our Herschel observations of low-mass pre-
and protostellar cores in the framework of this project is to de-
rivethespatialtemperatureanddensitystructureofthesecores
with the aim of constraining protostellar collapse models. For
this purpose, we selected 12 individual, isolated, nearby, and
previously well-characterized cores and obtained spatially re-
Fig.2. IRAS 100µm dust continuum emission map, showing
solved FIR dust emission maps at five wavelengths around the
the ρOph region and the loaction of two of our target sources
expected peak of the emission spectrum. This paper gives an
(B68andCB68).TheplaneoftheMilkyWaybecomesvisible
overviewoftheobservations,datareduction,analysismethods,
atthelowerleftcornerofthemap.ComparetoFigs.18andA.7.
andinitialresultsonthetemperatureandcolumndensitystruc-
tureoftheentirelow-masssample.Initialresultsononeofthe
sources(CB244)werealreadypublishedbyStutzetal.(2010). 2. Sources
Afirstdetailedfollow-upstudyofanothersourcefromthissam-
ple (B68) is published by Nielbock et al. (2012). Initial results Based on the results of earlier studies (e.g., Launhardt &
onthehigh-masscoresobservedintheframeworkofEPoSare Henning 1997; Launhardt et al. 1997; Henning & Launhardt
publishedbyBeutheretal.(2010,2012),Henningetal.(2010), 1998; Stutz et al. 2009a; Launhardt et al. 2010), we have se-
and Linz et al. (2010). An overview on the high-mass part of lected 12 well-isolated low-mass pre- and protostellar molec-
EPoSisgiveninRaganetal.(2012). ular cloud cores in regions of exceptionally low cirrus confu-
sion noise. In order to obtain deep maps at 100µm, which is
essential for a precise estimate of the dust temperature, the ab-
solute background levels and the point source confusion noise
(PSCN)wereimportantselectioncriteria.Forthispurpose,con-
This paper is organized as follows. Section2 describes
tributions from spatial fluctuations of the extragalactic back-
the target selection criteria and the sample of target sources.
ground were derived from Negrello et al. (2004). Estimates of
Section3 describes the Herschel and complementary ground-
the Galactic cirrus confusion noise are based on the ISOPHOT
based observations and the corresponding data reduction. In
confusionnoisemeasurements,scaleddowninthepowerspec-
Sect.4 we introduce our method of deriving dust temperature
and column density maps from the data. Section5 gives an 1 ThepeakofthefluxdensityofablackbodywithT = 10−20K
overview of the initial results from the survey, and in Sect.6 isatλ ≈ 290−145µm.Thepeakoftheenergydensity(νS )ofthe
ν
wediscusstheuncertaintiesandlimitationsofourapproachand opticallythinglobulesatthesetemperaturesisatonlyslightlyshorter
compare our results to earlier work by other authors. Section7 wavelengths(230−130µm).
summarizesthepaper. 2 http://lambda.gsfc.nasa.gov/product/cobe/dirbe overview.cfm
3
R.Launhardtetal.:Thestructureoflow-masscloudcores
Table1.Sourcelist
Source Other R.A.,Dec.(J2000) Region Dist. Ref. Evol. Ref.
names [h:m:s,◦:(cid:48):(cid:48)(cid:48)] [pc] class
CB4 ... 00:39:03,+52:51:30 CasA,GB 350±150 18 starless 1
CB6 LBN613 00:49:29,+50:44:36 CasA,GB 350±150 18 Cl.I 2
CB17 L1389 04:04:38,+56:56:12 Perseus,GB 250±50 2,18 starlessa 2
CB26 L1439 05:00:09,+52:04:54 Taurus-Auriga 140±20 2,11 starlessb 2,3,4
CB27 L1512 05:04:09,+32:43:08 Taurus-Auriga 140±20 1,11 starless 1,4
BHR12 CG30 08:09:33,−36:05:00 Gumnebula 400±50 2,10 Cl.0c 2,10
CB68 L146 16:57:16,−16:09:18 Ophiuchus 120±20 2,11 Cl.0 2
B68 L57,CB82 17:22:35,−23:49:30 Ophiuchus,Pipenebula 135±15 15,16,17 starless 5
CB130 L507 18:16:15,−02:32:48 Aquilarift,GB 250±50 12,14 Cl.0d 2
B335 CB199 19:36:55,+07:34:24 Aquila,isolated 100±20 8 Cl.0 2,7
CB230 L1177 21:17:39,+68:17:32 Cepheusflare,GB 400±100 2,13 Cl.I 2
CB244 L1262 23:25:47,+74:17:36 Cepheusflare,GB 200±30 2,13 Cl.0e 2,9
Notes. (a)CB17:additionallow-luminosityClassIYSO25(cid:48)(cid:48)fromstarlesscore.(b)CB26:additionalClassIYSO3.6(cid:48)south-westofthestarless
core.(c)BHR12:additionalClassIcore∼20(cid:48)(cid:48)southofClass0source.(d)CB130:additionallow-massprestellarcore∼30(cid:48)(cid:48)westandClassIYSO
∼15(cid:48)(cid:48)eastofClass0core.(e)CB244:additionalstarlesscore∼90(cid:48)(cid:48)westofClass0source.
References. (1)thispaper;(2)Launhardtetal.(2010);(3)Launhardt&Sargent(2001);(4)Stutzetal.(2009a);(5)Alvesetal.(2001);(7)Stutz
etal.(2008);(8)Olofsson&Olofsson(2009);(9)Stutzetal.(2010);(10)Bourkeetal.(1995);(11)Loinardetal.(2011);(12)Launhardt&
Henning(1997);(13)Kun(1998);(14)Straizˇysetal.(2003);(15)deGeusetal.(1989);(16)Lombardietal.(2006);(17)Alves&Franco(2007);
(18)Perrot&Grenier(2003)
trumtotheresolutionofHerschel/PACS(Kissetal.2005).Our tionalClassIcoreataprojectedseparationof∼20(cid:48)(cid:48)(∼8000AU).
selectedglobulesourceshavetypical100µmbackgroundlevels CB244containsanadditionalstarlesscoreataprojectedsepa-
of 1mJy/(cid:3)(cid:48)(cid:48) or lower and PSCN≤ 1mJy/beam. For compari- ration of ∼90(cid:48)(cid:48) (∼18000AU; Launhardt et al. 2010; Stutz et al.
son, typical 100µm background levels within the Taurus cloud 2010).Twooutofthe12targetglobules(CB6andCB230)are
are about 2−4mJy/(cid:3)(cid:48)(cid:48), within Ophiuchus about 5mJy/(cid:3)(cid:48)(cid:48), and dominated by embedded ClassI YSOs, i.e., the extended emis-
reach values of > 100mJy/(cid:3)(cid:48)(cid:48) in regions of high-mass star for- sion in them supposedly arises from remnant post-collapse en-
mation and infrared dark clouds (IRDCs). This specific source velopes.Ofthese,CB230isknowntobeabinarysourcewitha
selectionstrategyenabledustoobtaindeeper100µmmapsthan projectedseparationof∼10(cid:48)(cid:48)(∼4000AU;Launhardtetal.2010).
most large-area surveys and to derive robust dust temperature
We have also re-evaluated the distance estimates toward all
estimates.
globules, confirming the earlier used distances for 7 sources
Ourtargetlistcontainsonlyestablishedandpreviouslywell-
(e.g., Launhardt et al. 2010), and adjusting the distances for 5
characterizedsourcesanddoesnotcoverregionswithunknown
sources. CB4 and CB6, which we earlier associated with the
source content. All sources are nearby (100–400pc, with a
mean distance of 240 ± 100pc), have angular diameters of 3(cid:48) so-called “−12km/s” HI clouds at 600–800pc, are unlikely
to6(cid:48),linearsizesbetween0.2and1.0pc,andtotalgasmassesof to be that far away, as suggested by the lack of foreground
stars within a 3(cid:48) diameter area toward the cores. Despite their
1 to 25M (see Table6). Coordinates, distances, spectral (evo-
(cid:12) v ∼−12km/s,theyaremorelikelyassociatedwiththeCasA
lutionary) classes, and references are listed in Table1. Figure1 LSR
darkcloudsinGould’sBeltat∼350pc(Perrot&Grenier2003).
showsthedistributionofthetargetsourcesonthesky.Figure2
CB68 is associated with the ρOph dark clouds, for which we
illustrates that even those globules in our source list that are
adopt the new precise trigonometric distance of 120pc from
looselyassociatedwithlargerdarkcloudcomplexes(likeρOph)
Loinard et al. (2011). B68 is located within the Pipe nebula
arestillisolatedandlocatedoutsidetheregionsofhighextinc-
andsomewhatfartherawayfromρOph.Variousdirectandindi-
tionandsourceconfusion.
rectestimatessuggestadistanceof135pc±15pc(deGeusetal.
Seven out of the 12 target globules were known to con-
1989;Lombardietal.2006;Alves&Franco2007).ForCB130,
tainstarlesscores(Ward-Thompsonetal.1994).Ofthese,only
which is associated with the Aquila Rift clouds, we adopt the
CB17 was already shown before to be prestellar3 in nature
distanceestimateof250pc±50pcfromStraizˇysetal.(2003).
(Pavlyuchenkovetal.2006),whilethestar-formingpotentialof
the other sources is not known. CB17 contains, in addition to
theprestellarcore,alow-luminosityClassIyoungstellarobject
(YSO)ataprojectedseparationof∼25(cid:48)(cid:48)(∼6000AU;Launhardt
etal.2010).CB26contains,inadditiontothestarlesscore(Stutz
et al. 2009a), a well-studied ClassI YSO at a projected separa-
3. Observations and data reduction
tionof∼3.6(cid:48)(∼0.15pc;Launhardt&Sargent2001;Stecklumet
al.2004;Launhardtetal.2009;Sauteretal.2009).Fiveoutof
the12targetglobuleswereknowntohostClass0protostars.Of TheHerschelFIRcontinuumdataarecomplementedbysubmm
these, three cores contain additional sources with other evolu- dustcontinuumemissionmapsat450µm,850µm,and1.2mm,
tionaryclassifications.CB130containsanadditionallow-mass obtainedwithground-basedtelescopes.Observationsandreduc-
starlesscoreaswellasaClassIYSO.BHR12containsanaddi- tionofboththeHerschelandcomplementaryground-baseddata
are described in the following two subsections. Representative
3 Weusetheterm“prestellar”onlyforthosestarlesscoresthathave general observing parameters are listed in Table2, and the ob-
beenshowntobegravitationallyunstableoratthevergeofcollapse. servationsofindividualsourcesaresummarizedinTable3.
4
R.Launhardtetal.:Thestructureoflow-masscloudcores
Table3.SummaryofFIRandsubmmobservations
Source PACS SPIRE 450µm 850µm 1.2mm
ObsID ObsID
CB4 134221614(5,6) 1342188689 – – M11
CB6 134222272(3,4) 1342189703 L10 L10 L10/M11
CB17 134219100(8,9) 1342190663 – L10 L10/M11
CB26 134221739(7,8) 1342191184 L10a L10a L10/M11
CB27 134219197(5,6) 1342191182 DF08 DF08 M11
BHR12 13421985(39,40) 1342193794 L10 L10 L10
CB68 134220428(6,7) 1342192060 L10 L10 L10
B68 134219305(5,6) 1342191191 DF08 DF08/N12 B03
CB130 134220598(2,3) 1342191187 L10 L10 L10
B335 134219603(0,1) 1342192685 L10 L10 M11
CB230 134218608(3,4) 1342201447 L10 L10 L10/K08
CB244 134218869(4,5) 1342199366 L10 L10 L10
Notes. (a)TheSCUBAmapsofCB26coveronlythewesterncorewiththeembeddedClassIYSO,butnottheeasternstarlesscore.
References. (L10)Launhardtetal.(2010);(M11)dedicatedMAMBO-2observations,describedinthispaper(seeSect.3.2);(DF08)DiFrancesco
etal.(2008,SCUBALegacyCatalogues);(B03)Bianchietal.(2003,SEST-SIMBAobservations);(N12)Nielbocketal.(2012,APEX-Laboca
observations);(K08)Kauffmannetal.(2008,IRAM-MAMBO-2observations)
Table2.Generalobservingparameters othersources(OD230-511)andthedatawerereducedwiththe
respectivelatestHIPEversionandcalibrationtreethatwasavail-
λ Instr. (cid:104)HPBW(cid:105) (cid:104)mapsize(cid:105) (cid:104)rms(cid:105) ableatthattime.SinceprocessingofthePACSscanningdatais
0
[µm] [arcsec] [arcmin] [mJy/beam] verytime-consuming,are-reductionofalldatawitheverynew
100 PACS 7.1a 10 3 HIPEversionwithmodificationsrelevantforourdataisnotfea-
160 PACS 11.2a 10 13 sible.Instead,were-reducedthedataforonesourcewiththelat-
250 SPIRE 18.2a 16-20 15 estversionofHIPEtoverifyhowmuchthedatareductionand
350 SPIRE 25.0a 16-20 17
post-processing modifications affect our results on the derived
450 SCUBA 8.5-9.5b 2 200
dusttemperatureandcolumndensitymaps.Sincethiscompari-
500 SPIRE 36.4a 16-20 14
son, which is presented and discussed in Sect.6.1, showed that
850 SCUBA 14.4-15.1b 3 20
the effects are well within the range of the other uncertainties,
870 LABOCA 19.2c 40 8
we do not re-reduce all data, but incorporate these effects into
1200 MAMBO-2 11 5-10 6
ouruncertaintyassessmentinSect.6.1.
Notes. (a) Anianoetal.(2011)(b) Launhardtetal.(2010)(c) Nielbock Apart from the standard reduction steps, we applied ’2nd
etal.(2012) level deglitching’ to remove outliers in the time series data
(’time-ordered’option)byapplyingaclippingalgorithm(based
on the median absolute deviation) to all flux measurements in
3.1. HerschelFIRObservations
thedatastreamthatwillfinallycontributetotherespectivemap
FIR continuum maps at five wavelength bands were obtained pixel.Forourdatasetsweappliedan’nsigma’valueof20.
withtwodifferentinstrumentsonboardtheHerschelSpaceob- Afterproducinglevel1data,wehavegeneratedfinallevel2
servatory:PACSat100and160µm,andSPIREat250,350,and mapsusingtwodifferentmethods:high-passfilteringwithphot-
500µm. project(withinHIPE)andScanamorphos(Roussel2012),which
doesnotusestraightforwardhigh-passfiltering,butinsteadap-
pliesitsownheuristicalgorithmstoremoveartifactscausedby
3.1.1. PACSdata
detector flickering noise as well as spurious bolometer temper-
All 12 sources were observed with the Photodetector Array ature drifts. Based on the high-pass median-window subtrac-
CameraandSpectrometer(PACS; Poglitschetal.2010)inthe tion method, the photproject images turned out to suffer from
scan map mode at 100 and 160µm simultaneously. For each more missing flux and striping than the Scanamorphos images.
source, we obtained two scan directions oriented perpendicular Becausethecorrectrecoveryofextendedemissioniscriticalfor
toeachothertoeliminatestripinginthefinalcombinedmapsof accurate temperature mapping, high-pass filtering was clearly
effectivesize∼10(cid:48)×10(cid:48).Thescanspeedwassetto20(cid:48)(cid:48) sec−1, disadvantageous for our science goals. We also note that in a
and a total of 30 repetitions were obtained in each scan direc- previousreduction,wefoundthattheMADmap4 processingof
tion, resulting in a total integration time of ∼ 2.6hrs per map. ourbrightersources(e.g.,B335)usedtoproducelargeartifacts5
We emphasize that the accurate recovery of extended emission in the final maps due to the relatively bright central protostar;
isthemaindriverforthefollowingdiscussionandexplorationof thereforewedonotutilizethisreductionschemeinthiswork.
reductionschemes.ThePACSdataat100µmand160µmwere ForthesereasonswehavedecidedtouseScanamorphosv.9
processed in an identical fashion. They were reduced to level1 asourstandardlevel2datareductionalgorithm.Acomparison
using HIPE v.6.0.1196 (Ott 2010), except for CB4 (processed
withHIPEv.6.0.2044),CB26(processedwithv.6.0.2055),and 4 http://herschel.esac.esa.int/twiki/pub/Pacs/PacsMADmap/-
CB6 (processed with v.7.0.1931). The rationale for using dif- UM MADmapDOC.txt
ferent versions of HIPE for different sources is that the latter 5 This problem has eventually been solved in the most recent
threesourceswereobservedmuchlater(OD660-770)thanthe MADmapimplementation.
5
R.Launhardtetal.:Thestructureoflow-masscloudcores
offinalmapsprocessedwithdifferentScanamorphosv.9options tionataspeedof8(cid:48)(cid:48)/swhilechoppingwiththesecondarymirror
showed that the “galactic” option recovered the highest levels along the scan direction at a rate of 2Hz. Most sources were
of extended emission and was thus the best-suited for our data mappedtwice,withdifferentprojectedscanningdirections(ide-
setandscientificgoals.Weuseauniformfinalmappixelscale allyorthogonal)anddifferentchopperthrows(46(cid:48)(cid:48) and60(cid:48)(cid:48))to
of 1(cid:48).(cid:48)6 pix−1 at 100 µm and 3(cid:48).(cid:48)2 pix−1 at 160 µm for all ob- minimize scanning artifacts in the final maps. For one source
jects.Furthermore,thesedatawereprocessedincludingthenon- (CB27),wecouldobtainonlyonecoverage.Effectivemapsizes
zero-accelerationtelescopeturn-arounddata,withnoadditional (central pixel coverage) were adapted to the individual source
deglitching(’noglitch’setting). sizes and were in the range 5(cid:48)–10(cid:48), resulting in total mapping
timespercoverageof40minto1.5hrs.
The raw data were reduced using the standard pipeline
3.1.2. SPIREdata ”mapCSF”providedwiththeMOPSICsoftware6.Basicreduc-
tion steps include de-spiking from cosmic ray artifacts, cor-
All12sourceswerealsoobservedat250,350,and500µmwith
related signal filtering to reduce the skynoise, baseline fitting
theSpectralandPhotometricImagingReceiver(SPIRE;Griffin
andsubtraction,restoringsingle-beammapsfromthedual-beam
etal.2010).Thescanmapsofeachsourcewereobtainedsimul-
taneouslyatallthreebandsatthenominalscanspeedof30(cid:48)(cid:48)/sec scan data using a modified EKH algorithm (Emerson et al.
andovera(16(cid:48)−20(cid:48))2 area,resultinginatotalintegrationtime 1979),andtransformationandaveragingoftheindividualhori-
zontalmapsintotheequatorialsystem.Correlatedskynoisecor-
of10−16minspermap.Thesedatawereprocesseduptolevel1
rectionwasperformediteratively,startingwithnosourcemodel
with HIPE v.5.0.1892 and calibration tree 5.1, which was the
in the first run. The resulting map is smoothed and the region
most up-to-date version at the time all data became available.
withsourcefluxisselectedmanuallytorepresentthefirstsource
In Sect.6.1, we show that using the newest HIPE v.9.1.0 only
model,whichistheniterativelyimprovedin20successivecorre-
leads to nonsignificant minor changes in the SPIRE maps that
lated skynoise subtraction runs. Since our sources are faint, we
arewellbelowallotheruncertainties.Forthisreason,weconsid-
follow the recommendation to apply this algorithm only up to
ereditunnecessarytore-reducethedata.Uptolevel1(i.e.,the
level1,exceptforthebrightestsourceB335,wherewegoupto
level where the pointed photometer timelines are derived), the
level3.Athirdorderpolynomialbaselinefittothetime-ordered
stepsoftheofficialpipeline(POF5 pipeline.py,dated2.3.2010)
datastreamaswellasfirstorderfitstotheindividualscanlegs
provided by the SPIRE Instrument Control Center (ICC) have
are subtracted from the data, after masking out the region with
been performed. The level1 frames were then processed with
source flux, to correct for slow atmospheric and instrumental
theScanamorphossoftwareversion9(Roussel2012).Allmaps
drifts. Despite some unsolved electronic problems during these
were reduced using the ’galactic’ option. The final map pixel
sizesare6,10,and14(cid:48)(cid:48)at250,350,and500µm,respectively. observing runs, which limited the efficiency of the correlated
skynoisesubtraction,theresultingnoiselevelsinthefinalmaps
Photometric color corrections for both PACS and SPIRE
are ∼6mJy/beam, except for CB27, where it is ∼11mJy/beam
data, that account for the difference in spectral slopes between
becauseonlyonecoveragecouldbeobtained.
flux calibration (flat spectrum) and actually observed source
The assumption of a flat-spectrum source in the definition
spectrum,areappliedonlyinthedataanalysisandaredescribed
of the nominal wavelength of the MAMBO detectors in com-
inSect.4.1.
bination with the broad bandpass made it necessary to ap-
ply a ”color correction”. In contrast to the Herschel FIR ob-
3.2. (Sub)mmcontinuumObservations servations, (sub)mm wavelengths are sufficiently close to the
Rayleigh-Jeansregimeandtheemissionisopticallythinforall
Complementary submm dust continuum emission maps at our sources, such that the slope of the SED follows S ∝ λ−4
ν
450µm,850µm,and1.2mmfromground-basedtelescopesare (for the dust opacity submm spectral index β = 2). Therefore,
availableorwereobtainedthroughdedicatedobservationsforall we follow Schneeet al. (2010) andcorrect the reference wave-
12 sources. References to the existing and published data used length for all MAMBO data to 1100µm. Color corrections for
herearegiveninTable3. the SCUBA data are negligibly small (cf. Schnee et al. 2010)
Additional dedicated maps of the 1.2mm dust emission andwerenotapplied.
from 6 sources were obtained with the 117-pixel MAMBO-2 Wealsoverifiedthepointinginthefinalmapsbycomparing
bolometerarrayoftheMax-Planck-Institutfu¨rRadioastronomie the peak positions with interferometric positions where possi-
(Kreysa et al. 1999) at the IRAM 30m-telescope on Pico ble(CB26,CB68,CB230,CB244,B335,BHR12)andfound
Veleta (Spain) during two pool observing runs in October and thatdeviationsareinallcasessmallerthan2(cid:48)(cid:48),henceconfirming
November2010.Themeanfrequency(assumingaflat-spectrum thepointingstabilityandmakingadditionalpointingcorrections
source) was 250GHz (λ0 ∼1.2mm) with a half power (HP) unnecessary.Thefinalmapsofthosesourcesforwhichwehad
bandwidth of ∼80GHz. The HP beam width (HPBW) on sky botholdandnew1.2mmdata(CB6,CB17,CB26,andCB230)
was 11(cid:48)(cid:48), and the FoV of the array 4(cid:48) (20(cid:48)(cid:48) pixel spacing). weregeneratedbyweightedaveraging,afteradaptingbeamsizes
Weather conditions were good, with zenith optical depths be- andcorrectingsmallpointingoffsets.
tween 0.1 and 0.35 for most of the time and low sky noise. Additional complementary NIR extinction maps as well as
Pointing,focus,andzenithopticaldepth(byskydip)weremea- different molecular spectral line maps were also obtained with
suredbeforeandaftereachmap.Thepointingstabilitywasbet- the aim of studying the relation between density, temperature,
ter than 3(cid:48)(cid:48) rms. Absolute flux calibration was obtained by ob- and dust properties on the one side, and gas phase abundances
servingUranusseveraltimesduringtheobservingrunsandmon- and chemistry on the other side. These latter data will be de-
itoredbyregularlyobservingseveralsecondarycalibrators.The scribed and analyzed in forthcoming papers and are not pre-
flux calibration uncertainty is dominated by the uncertainty in sentedhere.
theknowledgeoftheplanetfluxesandisestimatedtobe∼20%.
The sources were observed with the standard on-the-fly dual- 6 CreatedandcontinuouslyupdatedbyR.Zylka,IRAM,Grenoble;
beam technique, with the telescope scanning in azimuth direc- seehttp://www.iram.es/IRAMES/mainWiki/CookbookMopsic
6
R.Launhardtetal.:Thestructureoflow-masscloudcores
Table4.DC–fluxlevelsintheHerscheldata
Source R.A.,Dec.a f (σ ) f (σ ) f (σ ) f (σ ) f (σ )
DC f DC f DC f DC f DC f
(J2000) 100µm 160µm 250µm 350µm 500µm
[h:m:s,◦:(cid:48):(cid:48)(cid:48)] [µJy/(cid:3)(cid:48)(cid:48)] [µJy/(cid:3)(cid:48)(cid:48)] [µJy/(cid:3)(cid:48)(cid:48)] [µJy/(cid:3)(cid:48)(cid:48)] [µJy/(cid:3)(cid:48)(cid:48)]
CB4 00:38:55,+52:56:06 8.4E2(4.4E1) 5.9E2(1.0E2) 2.8E2(3.9E1) 1.4E2(2.1E1) 6.9E1(1.0E1)
CB6 00:49:02,+50:42:34 1.1E3(3.0E1) 4.6E2(3.7E1) 1.8E2(2.2E1) 8.7E1(1.1E1) 3.8E1(5.0E0)
CB17 04:04:14,+56:53:07 1.3E3(4.2E1) 7.5E2(4.5E1) 2.6E2(3.1E1) 1.3E2(1.6E1) 6.5E1(7.0E0)
CB26 05:00:28,+52:00:36 7.8E2(4.7E1) 7.7E2(1.1E2) 1.4E2(2.7E1) 6.8E1(1.0E1) 3.3E1(7.0E0)
CB27 05:04:16,+32:38:19 1.0E3(4.7E1) 6.7E2(1.3E2) 1.9E2(2.8E1) 9.4E1(1.7E1) 5.0E1(7.0E0)
BHR12 08:09:10,−36:07:44 1.3E3(5.7E1) 6.7E2(8.5E1) 3.1E2(2.9E1) 1.7E2(1.3E1) 7.7E1(5.0E0)
CB68 16:57:07,−16:14:27 1.3E3(5.4E1) 5.1E2(6.9E1) 2.2E2(3.5E1) 1.3E2(2.0E1) 4.2E1(1.1E2)
B68 17:22:36,−23:44:11 9.1E2(8.4E1) 8.9E2(7.2E1) 1.2E2(2.9E1) 5.9E1(1.4E1) 1.3E1(5.0E0)
CB130 18:16:30,−02:29:23 1.4E3(7.1E1) 8.0E2(1.5E2) 3.0E2(7.4E1) 1.6E2(3.9E1) 7.7E1(1.7E2)
B335 19:36:48,+07:30:37 1.5E3(4.3E1) 7.2E2(7.9E1) 1.5E2(2.1E1) 8.7E1(9.0E0) 3.9E1(6.0E0)
CB230 21:17:00,+68:14:35 1.3E3(4.8E1) 5.1E2(1.8E2) 1.8E2(4.3E1) 1.4E2(7.1E1) 7.4E1(2.8E1)
CB244 23:24:55,+74:23:01 1.3E3(4.0E1) 5.1E2(7.5E1) 3.0E2(2.7E1) 1.7E2(1.3E1) 7.7E1(7.0E0)
Notes. (a)CenteroftheDCreferencefield(seeSect.4.1).
4. Modeling approach corrected properly. For the long-wavelength SPIRE maps with
beamsizes∼18(cid:48)(cid:48)–36(cid:48)(cid:48) andnoreadilyavailableastrometricref-
4.1. Strategyanddatapreparation
erence, the pointing errors are fortunately negligible. For the
The calibrated dust emission maps at the various wavelengths PACSmapsweutilizethefactthatmany100µmpointsources
wereusedtocompilespatiallyandspectrallyresolveddatacubes are also detected in the Spitzer MIPS 24µm images. We select
that cover the full extent of these relatively isolated sources on in each field three stars detected in both the PACS 100µm and
bothsidesofthepeakoftheirthermalSEDs.Ideally,onewould MIPS 24µm images to align the PACS astrometry in both the
want to compare such data to synthetic maps from radiative 100µmand160µmimagestotheSpitzerimages.Thepointing
transfer models convolved with the respective beams (forward- correctionswerefoundtobe≤ 3(cid:48)(cid:48) inallcases,i.e.,smallcom-
modeling, virtual observations) in order to constrain the phys- paredtotheSPIRE500µmbeam.
ical properties of the sources. However, to avoid circular aver- Surfacebrightnesscalibrationcorrections:SincethePACSim-
agingwithitswell-knowncaveats,onewouldneed3-Dmodel- agesarecalibratedtoJypix−1,theconversiontoJy/(cid:3)(cid:48)(cid:48) doesnot
ing to account for the complex structure present even in these involve an assumption of the beam sizes. The SPIRE data, on
relatively simple and isolated sources. Furthermore, not all of theotherhand,arecalibratedtoJybeam−1;toconverttosurface
the observed features (like, e.g., the core-envelope temperature brightness units we adopt the FWHM beam sizes from Aniano
contrast,seeSects.5.6and5.7)maybeeasilyreproduciblewith et al. (2011) listed in Tab.2. The (sub)mm data have also been
existingself-consistentmodels,makingthefullyself-consistent convertedtoJy/(cid:3)(cid:48)(cid:48)assumingtheappropriateeffectivebeamsizes
forward-modeling approach very time-consuming. This will be fortherespectiveobservations(seereferencesinTable3).Since
dealtwithinforthcomingpapersmodelingindividualcores. thestandardcalibrationisdoneonpointsources,butweareus-
ingsurfacebrightnessestocompilespatiallyresolvedSEDs,we
Forthissurveyoverviewpaper,wethereforetakeasimpler
have applied the recommended extended emission calibration
and more direct approach, giving up some of the spatial infor-
corrections to the SPIRE data (see SPIRE Observer’s Manual:
mation in the short-wavelength maps and directly recovering
HERSCHEL-DOC-0798,version2.4,June7,2011).
and modeling beam and LoS optical-depth-averaged (hereafter
LOS-averagedforshort)SEDs,thusintroducingasfewaspos- PACS and SPIRE map background subtraction: The
sible model-dependent assumptions into our analysis. For this Scanamorphos “galactic” option preserves the large spatial
purpose,thecalibratedmapsarepreparedinthefollowingway: frequency modes of the bolometer signal in the Herschel
scan-maps. However, the absolute flux level of the background
1. allmapsareregisteredtoacommoncoordinatesystem, and spatial structures more extended than the scan-maps
2. pointing corrections are applied where necessary (see be- cannot be recovered since the exact level of the strong thermal
low), backgroundfromtheonlypassivelycooledmirrorsM1andM2
3. all maps are converted to the same physical surface bright- of the observatory is unknown. Therefore, it is not clear how
ness units (here Jy/(cid:3)(cid:48)(cid:48)) and extended emission calibration the extended, large-scale emission levels in the final Herschel
correctionsareappliedwherenecessary(seebelow), maps are related to the absolute flux level of the background,
4. background levels are determined and subtracted from the which is composed of emission features larger than the map
maps(seebelow),andfinally, size, the general galactic background ISRF, the cosmic mi-
5. allmapsareconvolvedtotheSPIRE500µmbeam(FWHM crowave background (CMB), and other possible contributions,
36(cid:48).(cid:48)4). all varying differently with wavelength. Since uncertainties in
additive flux contributions affect flux ratio measurements and
PACSpointingcorrections:SincetheHerschelspacecraftabso- therespectivederivationofphysicalparametersfromtheSEDs,
lute1σpointingerrorwas∼2(cid:48)(cid:48) andthePACSandSPIREmaps in particular at low flux levels, we subtract the background
were not obtained simultaneously, random relative pointing er- levels from all Herschel maps, thus making them consistent
rors between PACS and SPIRE maps can add up to ∼3(cid:48)(cid:48)–4(cid:48)(cid:48), with the chopped (sub)mm data (although chopping is a much
whichisabouthalfthePACS100µmFWHMbeamsize(Aniano more aggressive filtering than that present in the Herschel
et al. 2011). This could severely hamper the usefulness of flux maps). The consequences and uncertainties of this background
ratio (SED) maps, in particular around compact sources, if not removalonthedataanalysisarediscussedinSects.4.2and6.1.
7
R.Launhardtetal.:Thestructureoflow-masscloudcores
Weemphasizethatthisapproachwasfeasibleonlybecauseour 4.2. Derivingdusttemperatureandcolumndensitymaps
sources were already initially selected to be relatively isolated
Thesubtraction(orchoppingout)ofaflatbackgroundlevelfrom
onthesky(seeSect.2).
the emission maps implies that the remaining emission in the
To accomplish this re-zeroing, we adopt a method similar mapateachimagepixelisgivenby
to that applied to Spitzer MIPS images in Stutz et al. (2009a).
(cid:16) (cid:17) (cid:16) (cid:17)
For each object we identify a 4(cid:48) × 4(cid:48) region in the PACS and S (ν)=Ω 1−e−τ(ν) B (ν,T )−I (ν) (1)
ν ν d bg
SPIREimagesthatisrelativelyfreefromspatiallyvaryingemis-
with
sion and appears “dark” relative to the globule extended emis-
sion levels. We impose the additional requirement that this re- M
τ(ν)= N m d κ (ν) (2)
gionisinorneararegioninourcomplementarymolecularline H H M d
mapswhichisrelativelyfreeof12CO(2–1)emission.Wealways H
use the same region in all five Herschel scan-maps. For each where Sν(ν) is the observed flux density at frequency ν, Ω
band,wethencalculatetherepresentativefluxlevel, f ,inthe the solid angle from which the flux arises (here normalized to
DC
4(cid:48)×4(cid:48) regionbyimplementinganiterativeGaussianfittingand 1arcsec2),τ(ν)theopticaldepththroughthecloud,Bν(ν,Td)the
σ-clippingschemetothepixelvaluedistributionateachwave- Planckfunction,Td thedusttemperature,Ibg(ν)thebackground
length.Wedonotconsiderpixelsbelow2σfromthemeanand flux level, NH = 2×N(H2)+N(H) the total hydrogen column
fitthemainpeakofthepixelvaluedistribution.The fDCvalueis density,mHtheprotonmass,Md/MHthedust-to-hydrogenmass
thendeterminedbyiterativelyfittingaGaussianfunctiontothe ratio, and κd(ν) the dust mass absorption coefficient. Td and κd
histogramofpixelvalues,whereateachiterationweincludeone may vary along the LoS, in which case Eq.1 has to be written
moreadjacenthigherfluxbin.Thefinaladoptedvalueof f is indifferentialformandintegratedalongtheLoS.However,for
DC
defined as the mean value of the best-fit Gaussian that has the this overview paper we assume both parameters to be constant
minimumσvalueforalliterations.Inthiswayweexcludepix- along the LoS and discuss the uncertainties and limitations of
els with higher flux which effectively broaden the distribution thisapproachinSect.6.2.
and would cause a bias in the fDC calculation. The fDC values For κd(ν), we assume for all sources in this paper the
arethensubtractedfromthecorrespondingHerschelmapsatthe tabulated values listed by Ossenkopf & Henning (1994) for
respectivewavelength(seeTable4forthe f andσvaluesand mildly coagulated (105yrs coagulation time at gas density
coordinatesofthereferenceregion). DC 106cm−3)compositedustgrainswiththinicemantles(column5
in their Table1, usually called “OH5”), logarithmically inter-
Imageconvolution:Finally,allmapsareconvolvedtothebeam
polated to the respective wavelength where necessary. For the
of the SPIRE500µm map, using the azimuthally averaged
hydrogen-to-dustmassratiointheSolarneighborhood,weadopt
HerschelconvolutionkernelsprovidedbyAnianoetal.(2011). M /M =110 (e.g., Sodroski et al. 1997). Note that the total
H d
For the convolution of the (sub)mm maps, we use a Gaussian
gas-to-dust mass ratio, accounting for helium and heavy ele-
convolutionkernelofwidthequaltothedifferenceinquadrature ments,isabout1.36timeshigher,i.e.,M /M ≈150.
g d
betweentheeffectiveFWHMofthe(sub)mmobservationsand
The exact background flux levels at the location of the in-
theSPIRE500µmFWHMof36(cid:48).(cid:48)4.
dividual clouds, I (ν), are a priori unknown as explained in
bg
Afterre-griddingtoacommonNyquist-sampledpixelgrid, Sect.4.1. At the frequencies relevant for this paper, this back-
we thus compile pixel-by-pixel SEDs for the wavelength range ground radiation is composed mainly of the CMB, the ex-
from100µmthrough1.2mm.Photometriccolorcorrectionsfor tragalactic cosmic infrared background (CIB), and the diffuse
each pixel and each band are iteratively derived from the 100– galacticbackground(DGB).Whilethefirstcontribution(CMB)
160µmand250–500µmspectralindicesandpolynomialfitsto iswell-knownanddominatesatwavelength>500µm,themean
colorcorrectionfactorsprovidedforPACSinTable2ofMu¨ller levelsoftheCIBhavebeencompiledby,e.g.,Hauser&Dwek
et al. (2011) and in Table5.3 of the SPIRE Observers Manual, (2001)fromfluxmeasurementsinthe”LockmanHole”byvari-
andareappliedinthesubsequentSEDmodelingonly. ousspaceobservatories(e.g.,COBEandISO)andground-based
(sub)mm instruments (e.g., SCUBA). However, at wavelengths
The nominal point-source calibration uncertainty of the
< 350µm,theDGB,whichstronglyvarieswithposition,domi-
PACS photometers listed in the latest calibration notes is 3%
natesoverCMBandCIB.HerewefollowtheapproachofStutz
at100µmand5%at160µm.However,additionaluncertainites
et al. (2010) and use the Schlegel et al. (1998) 100µm IRAS
thatarehardtoquantifyareintroducedby,e.g.,theconversionto
mapsandtheISOSerendipitySurveyobservationsat170µmto
surfacebrightnessunitsforextendedemission,thebeamconvo-
extrapolateapproximatemeanfluxlevelsatthetypicalgalactic
lution (imperfect kernels), and the color corrections. Likewise,
locations of our sources. Since the uncertainty in the resulting
thefinalrecommendedcalibrationuncertaintyof7%forSPIRE
temperatureandcolumndensityestimatesintroducedbytheun-
does not account for some of the uncertainties introduced in
certaintyintheexactknowledgeoftheI levelsisnegligible,as
thepost-processing.Recentsystematiccomparisonsbetweenthe bg
discussedquantitativelyinSect.6.1,wedonotattempttoderive
HerschelandSpitzersurfacebrightnesscalibrationsshowedthat
thelocalvaluesoftheDGBforeachsourceseparately.Instead,
they agree to within about 10%, with intrinsic uncertainties on
weadoptthefollowingmeanvaluesforallsources:0.3,0.8,0.5,
bothsides7.Therefore,weadoptforallHerscheldatatheconser-
0.3, 0.2, 2.9, and 6.5mJy/arcsec2 at λ 100, 160, 250, 350, 500,
vativevalueof15%fortherelativecalibrationuncertaintyinthe
850,and1100µm,respectively.
final maps. This number is used for calculating the weights of
From the calibrated and background-subtracted dust emis-
the individualdata points and withrespect to the ground-based
sion maps, we extract for each image pixel the SED with up
datainthesubsequentfittingprocedure(Sect.4.2).
to 8 data points between 100µm and 1.2mm. These individ-
ualSEDsareindependentlyfit(χ2 minimization)withasingle-
temperaturemodifiedblackbodyoftheformofEqs.1and2with
7 see PICC-NHSC-TR-034 and PICC-NHSC-TN-029 under Td and NH being the free parameters. The individual flux data
https://nhscsci.ipac.caltech.edu/. points are weighted with σ−2, where σ is the quadratic sum of
8
R.Launhardtetal.:Thestructureoflow-masscloudcores
fluxtimestherelativecalibrationuncertainty(seeSect.4.1)and tail-likeextensionslike,e.g.,inCB17(Fig.14),the∼18(cid:48)SPIRE
the mean rms noise in the respective map measured in regions maps usually cover the entire extent of the detectable FIR dust
of zero or lowest emission outside the sources (see Table4). In emission down to the more or less flat background levels. The
ordertominimizetheeffectofT − N degeneraciesinthefit- slightlysmallerPACSmaps(∼10(cid:48))alsousuallycovertheentire
H
ting,becauseatthelowtemperaturesinthesecores,T ≤ 10K, extentoftheclouds,butinsomecasescutoffabitmoreofsome
eventhe1.2mmemissiondoesnotrepresenttheRayleigh-Jeans filamentaryortail-likeextensions(e.g.,Fig.14).
regime where the SED slope is independent of T, the weight The PACS 100µm maps often show significantly fainter or
of data points with S < σ was set to zero, i.e., these points lessextendedemissionthanthemapsatlongerwavelengths,ow-
ν
werenotconsideredinthefitting.Tofurtheravoiderroneousex- ing to the fact that 100µm samples the steep short-wavelength
trapolationsoftheSEDfittounconstrainedshorterwavelengths, sideoftheSEDatthelowtemperatureofthedustintheseclouds
onlypixelswithvalidfluxesinall5Herschelbandswerefitted. (Fig.4). In some cases, the diffuse 100µm emission even ex-
ThislattercriterionwasusuallyconstrainedbythePACS100µm hibitsa“hole”or“shadow”atthelocationofthecolumndensity
maps.Thebest-fittingparameterswerederivedbyemployinga peak(e.g.,CB27;Fig.16),hintingatextremelycoldanddense
leastsquaresfittoallfluxvaluesbetween100µmand1.2mmat (highcolumndensity)coreswiththinwarmerenvelopes,which
agivenimagepixel,usinga“robust”combinationoftheclassi- arethusgoodcandidatesofcoresatthevergeofprotostellarcol-
calsimplexamoebasearchandamodifiedLevenbergMarquardt lapse(seeStutzetal.2009a,foradiscussionof24µmand70µm
methodwithadaptivesteps,asimplementedinthe’mfit’toolof shadowsindensecores).
GILDAS8. On the other hand, the 100µm maps, which also have the
This procedure yields LoS-averaged dust temperature and highestangularresolutionofallourHerschelmaps(≈ 7(cid:48)(cid:48)),are
column density maps of the sources, which are presented in mostsensitivetoembeddedheatingsourceslike,e.g.,protostars.
Sect.5.2. The temperature maps provide a robust estimate of Althougheventhepreviouslyknownverylowluminosityobject
the actual dust temperature of the envelope in the projected (VeLLO) in CB130 (Kim et al. 2011; see also Dunham et al.
outer regions, where the emission is optically thin at all wave- 2008)iswell-detected(Fig.20),ourmapsdonotrevealanyob-
lengths and LoS temperature gradients are negligible. Toward vious previously unknown warm compact source in any of the
the source centers, where cooling and shielding or embedded globules, confirming that all our presumably starless cores are
heatingsourcescanproducesignificantLoStemperaturegradi- indeedstarless.TheonlypossibleexceptionmightbeCB17(see
ents and the observed SEDs are therefore broader than single- Fig.14), where we find hints of an extremely low-luminosity
temperature SEDs, the central dust temperatures will be over- (L < 0.04L ) cold embedded source very close (in projec-
bol (cid:12)
estimated (in the case of a positive gradient in cold sources) or tion)toCB17-IRS(Chenetal.2012;Schmalzletal.2013).
underestimated(negativegradientininternallyheatedsources).
For these reasons, the column density maps, which come out
5.2. Dusttemperatureandcolumndensitymapsand
usually very smooth, are corrupted and exhibit artifacts in re-
radialprofiles
gionsof∼20(cid:48)(cid:48)–40(cid:48)(cid:48)radiusaroundthewarmprotostars(see,e.g.,
Figs.30 or 29). These caveats are discussed and quantified in The LoS-averaged dust temperature and hydrogen column
Sect.6.2,includingsomealreadyworked-onsolutions. density maps derived with the fitting procedure described in
Sect.4.2 are presented in Figs.24 through 35. All sources
show systematic dust temperature gradients with cold interi-
5. Results
ors(11–14K)andsignificantlywarmerouterrims(14–20K).
Embeddedheatingsourceslikeprotostars,includingtheVeLLO
5.1. HerschelFIRmaps
inCB130(Fig.32;Kimetal.2011),showupveryclearlyinthe
Theresultingcalibrateddustemissionmapsatλ100,160,250, temperaturemaps.
350,and500µmandatoriginalangularresolutionarepresented To characterize the column density profiles quantitatively,
in Figs.12 through 23. The Herschel maps are accompanied we fit circular profiles around the column density peak to the
by optical (red) images from the second Digitized Sky Survey maps, using a “Plummer-like” profile (Plummer 1911; see also
(DSS29),whichwereobtainedthroughtheSkyView10 interface. Whitworth & Ward-Thompson 2001), modified by a constant
Theseopticalimagesclearlyshowtheregionsofhighestextinc- termtoaccountfortheobservedoutercolumndensity“plateau”
tionaswellas,inseveralcases,extendedcloudshinestructures. (seediscussioninSect.5.7):
Allmapsareoverlaidwithcontoursofthe(sub)mmdustcontin- 
uumTehmei1s6si0oµnmobtsherrovuegdhw5i0th0gµrmoumnda-pbsasaerdetuesluesaclloypevse.ry similar NH(r)=0(1+(r∆/rN1)2)p/2 +Nout iiffrr≤>rrout . (3)
out
toeachotherinappearance,outliningthethermaldustemission
from the dense cores along with the often filamentary or tail- Thepeakcolumndensityisthen
like, more tenuous envelopes. Except for some filamentary or N = N (r=0)=∆N+N . (4)
0 H out
8 http://www.iram.fr/IRAMFR/GILDAS This profile accounts for an inner flat (column) density core at
9 TheDigitizedSkySurveyswereproducedattheSpaceTelescope r<r where
1
ScienceInstituteunderU.S.GovernmentgrantNAGW-2166.Theim-
∆N
agesofthesesurveysarebasedonphotographicdataobtainedusingthe N = N (r )= +N , (5)
OschinSchmidtTelescopeonPalomarMountainandtheUKSchmidt 1 H 1 2p/2 out
Telescope.Theplateswereprocessedintothepresentcompresseddig-
approachesapower-lawwithindex patr (cid:29) r ,turnsoverinto
italformwiththepermissionoftheseinstitutions. 1
aflatoutercolumndensity“plateau”outside
10 SkyViewhasbeendevelopedwithgeneroussupportfromtheNASA
(cid:115)
AISRandADPprograms(P.I.ThomasA.McGlynn)undertheauspices (cid:32)∆N(cid:33)2/p
of the High Energy Astrophysics Science Archive Research Center r =r −1 (6)
(HEASARC)attheNASA/GSFCAstrophysicsScienceDivision. 2 1 N
out
9
R.Launhardtetal.:Thestructureoflow-masscloudcores
Table5.Columndensityanddusttemperatureprofiles
Sourcea N b N b r b r b r b pb T b T c T b
0 out 1 2 out in peak out
[cm−2] [cm−2] [AU] [AU] [AU] [K] [K] [K]
CB4 7.5E21 6.0E19 2.5E4 6.2E4 9.5E4 5.0 14.0 – 19.5
CB17-SMM 2.5E22 1.0E21 1.0E4 3.1E4 3.5E4 3.0 11.7 – 13.1
CB26-SMM2 1.4E22 1.2E21 8.4E3 2.2E4 3.4E4 2.4 12.8 – 14.0
CB27 1.9E22 2.1E21 1.6E4 2.1E4 2.8E4 4.0 11.9 – 14.2
B68 2.5E22 4.0E20 1.0E4 2.1E4 4.6E4 5.0 11.9 – 18.5
CB244-SMM2 4.6E22 1.1E21 7.0E3 6.0E4 9.0E4 1.8 11.3 – 14.5
CB130-SMMd 2.6E22 2.0E21 7.2E3 2.8E4 4.7E4 1.8 12.1 – 13.5
CB6e,f 5.0E21 5.0E20 ... ... 1.4E5 ... 14.5 18.4 15.7
CB26-SMM1e 1.0E22 9.0E20 ... ... ... ... ... 19.0 14.6
BHR12f 5.4E22 1.0E21 1.4E4 3.5E4 1.1E5 4.0 14.8 18.5 17.3
CB68 2.4E22 5.0E20 1.1E4 2.7E4 2.9E4 4.0 12.7 19.0 16.5
B335 3.1E22 6.2E20 3.6E3 1.8E4 3.8E4 2.3 14.7 18.8 16.2
CB230-SMMe,f 2.7E22 5.0E20 ... ... 1.4E5 ... 14.5 19.6 15.6
CB244-SMM1 1.5E22 1.0E21 1.2E4 5.5E4 9.0E4 1.7 11.5 19.6 14.2
starlesscores 2.3±1.0E22 1.1±0.6E21 1.2±0.6E4 3.5±1.6E4 5±2E4 3.3±1.2 12.2±0.8 ... 15.3±2.2
protostellarcores 2.4±1.4E22 0.7±0.2E21 0.9±0.5E4 3.4±1.2E4 9±4E4 3.0±0.9 13.6±1.2 19.0±0.4 15.7±0.9
all 2.3±1.3E22 0.9±0.5E21 1.1±0.5E4 3.5±1.5E4 7±4E4 3.2±1.2 13.0±1.2 ... 15.5±1.8
Notes.Incontrasttotheothertables,sourcesinthistablehavebeengroupedintostarlessandprotostellar.(a) CoordinatesarelistedinTable6.
(b)Derivedfromcircularfitstothedusttemperatureandcolumndensitymaps;seeSect.5.2,Eqs.3through9.(c)Peaktemperatureatthepositionof
theembeddedheatigsource(protostar),severelyunderestimatesthetruecentraldusttemperaturebecauseofbeamsmoothingandLoSaveraging.
(d) CB130:theembededVeLLOcausesonlyaverysmalltemperatureincreaseinthesmootheddusttemperaturemap.(e) Badfitduetolarge
unresolvedLoSgradients,nomeaningfulvaluesforsomeoftheparameters.(f) TheembeddeddoublesourcesinBHR12andCB230arenot
resolvedinthesmootheddusttemperaturemaps.
where To illustrate the quality of these radial profile fits and the
meaningofthevariousparameters,weshowinFig.3theradial
N = N (r )=2×N , (7)
2 H 2 out distribution of N and T values in the resulting maps for B68
H d
andiscutoffatr .Thisouterboundaryofthethinenvelopeor alongwiththebest-fitprofilesandtheparameterslabeledonthe
out
haloisnotwell-recoveredinmostcasesbecausefilamentaryex- diagramaxes.Thebest-fitparametervaluesfor N , N ,r ,r ,
0 out 1 2
tensionsemphasizedeviationsfromcircular/sphericalgeometry r , p,T ,T (forcoreswithembeddedheatingsources),and
out in peak
withincreasingdistancefromthecorecenterandthefluxlevels T ofallsourcesarelistedinTable5.Wedonotlistparameters
out
graduallydeclinebelowthenoisecut-off(seeSect.4.2).Forthe r andq,butnotethatfromthecorecentersoutward,theLoS-
T
same reason, the flatness of the outer column density profile at averageddusttemperaturetypicallyraisesby1Kataradiusof
level N might be partially an artifact of circularly averaging (1±0.3)104AU.
out
noncircularstructureataleveljustabovethenoise. Notethatinthenextsection,wedefinecoreandcloudsizes
To empirically fit and parameterize the radial temperature via the mean extent of certain column density contours in the
profilesofpurelyexternallyheatedcores,weuseasimilarprofile maps,andnotviatheradialprofilefitparametersr andr .In
1 out
oftheform: particular for very elliptical sources and sources with extended
∆T halos, the derived mean diameters are therefore not necessarily
T(r)=T − . (8)
out (cid:18)1+(cid:16)r (cid:17)2(cid:19)q/2 identicaltotwicethecorrespondingradialprofile-fitradii.
rT
Thecentraltemperatureminimumisthengivenby 5.3. Sourcemorphologies,integralproperties,andSEDs
T =T −∆T . (9)
in out In addition to the pixel-by-pixel SEDs and the column density
Before fitting radial profiles to the maps, we mask some of anddusttemperaturemaps,wealsoderiveintegratedsourcepro-
the spurious low-SNR edge features and tail-like (asymmetric) pertiesusingcertaincolumndensitylevelstodefinetheintegra-
extensions, always applying the same spatial mask in both the tionareas.Totalcloudsizesarederivedbymeasuringthemajor
columndensityanddusttemperaturemaps.Forcoreswithem- andminorextentoftheN contourinthecolumndensitymaps,
out
bedded heating sources (protostars), we also mask the region whichismarkedastheouteryellowcontourinFigs.24through
withlocaltemperatureincreasebeforefittingtheradialprofiles 35.Theresultinglineardiameters(invokingthedistanceslisted
withEqs.3and8andalsolistinTable5thevalueofthecentral inTab.1)arelistedinTab.6.Themeanradiioftheglobulesin
temperature maximum T . The radius of the masked region oursamplerangefrom0.1pc(CB130)to0.5pc(CB230),with
peak
varied between 20(cid:48)(cid:48) and 40(cid:48)(cid:48). This extrapolation for N and T ameanof0.22±0.1pc(4.8±2.2104AU),whichisabouttwice
0 in
oftheouterprofileintothecorecenter,whereboththelocaltem- the typical Jeans lengths of these clouds (∼ 0.13pc, assuming
peratureandthecolumndensityestimatesareveryuncertaindue themeanradiusandmassandameantemperatureof13K;see
to large and unresolved LoS temperature gradients, leads to an alsostabilitydiscussioninSect.5.5).Themeanprojectedaspect
increaseduncertaintyinparticularofthevalueforN inthepro- ratio of the clouds is 0.66±0.16, with the most extreme cases
0
tostellar cores. This and other shortcomings of this model and CB6 and BHR12 (0.4) and B335 (0.98). These measures do
theinterpretationofitsparametersarediscussedinSect.6.2. not account for the long tails, e.g., in CB17 or CB68, nor do
10