The AKARI NEP-Deep survey: a mid-infrared source catalogue

0.82 MB·English

by  T. Takagi

Astronomy&Astrophysicsmanuscriptno.draft˙final (cid:13)c ESO2012
January5,2012
⋆
The AKARI NEP-Deep survey: a mid-infrared source catalogue
T.Takagi1,H.Matsuhara1,T.Goto2,3,H.Hanami4,M.Im5,K.Imai6,T.Ishigaki4,H.M.Lee5,M.G.Lee5,M.Malkan7
Y.Ohyama8,S.Oyabu9,C.P.Pearson10,S.Serjeant11,T.Wada1,andG.J.White10,11
1 InstituteofSpaceandAstronauticalScience,JapanAerospaceExplorationAgency,Sagamihara,Kanagawa229-8510,Japan
e-mail:[email protected]
2 InstituteforAstronomy,UniversityofHawaii,2680WoodlawnDrive,Honolulu,HI,96822,USA
3 SubaruTelescope650NorthA’ohokuPlaceHilo,HI96720,USA
4 PhysicsSection,FacultyofHumanitiesandSocialSciences,IwateUniversity,Morioka020-8550,Japan
5 DepartmentofPhysicsandAstronomy,FPRD,SeoulNationalUniversity,Shilim-Dong,Kwanak-Gu,Seoul151-742,Korea
2
6 TOMER&DInc.Kawasaki,Kanagawa213-0012,Japan
1
7 DepartmentofPhysicsandAstronomy,UCLA,LosAngeles,CA,USA
0
8 AcademiaSinica,InstituteofAstronomyandAstrophysics,Taiwan
2
9 GraduateSchoolofScience,NagoyaUniversity,Furo-cho,Chikusa-ku,Nagoya,Aichi464-8602,Japan

n 10 RutherfordAppletonLaboratory,Chilton,Didcot,Oxfordshire,OX110QX,UK
a 11 AstrophysicsGroup,DepartmentofPhysics,TheOpenUniversity,MiltonKeynes,MK76AA,UK
J
4 Received22July2011;accepted8November2011

] ABSTRACT
O
Wepresentanewcatalogueofmid-IRsourcesusingtheAKARINEP-Deepsurvey.TheInfraRedCamera(IRC)onboardAKARIhas
C
acomprehensivemid-IRwavelengthcoveragewith9photometricbandsat2–24µm.Weutilizedallofthesebandstocoveranearly
h. circularareaadjacenttotheNorthEclipticPole(NEP).Wedesignedthecataloguetoincludemostofsourcesdetectedin7,9,11,
p 15and18µmbands,andfound7284sourcesina0.67deg2 area.Fromoursimulations,weestimatethatthecatalogueis∼ 80per
- centcompleteto200µJyat15–18µm,and∼10percentofsourcesaremissed,owingtosourceblending.Star-galaxyseparationis
o conductedusingonlyAKARIphotometry,asaresultofwhich10percentofcataloguedsourcesarefoundtobestars.Thenumber
r countsat11,15,18,and24µmarepresentedforbothstarsandgalaxies.Adrasticincreaseinthesourcedensityisfoundinbetween
t
s 11and15µmatthefluxlevelof∼ 300µJy.ThisislikelyduetotheredshiftedPAHemissionat8µm,givenourroughestimateof
a redshiftsfromanAKARIcolour-colourplot.Alongwiththemid-IRsourcecatalogue,wepresentoptical-NIRphotometryforsources
[ fallinginsideaSubaru/Sprime-camimagecoveringpartoftheAKARINEP-Deepfield,whichisdeepenoughtodetectmostofAKARI

mid-IRsources,andusefultostudyopticalcharacteristicsofacompletemid-IRsourcesample.
1
v Keywords.infrared:galaxies–surveys–catalogues–methods:dataanalysis
7
9
7
1. Introduction (Elbazetal. 2002; Serjeantetal. 2004; LeFloc’hetal. 2005;
0
Pe´rez-Gonza´lezetal. 2005; Babbedgeetal. 2006; Caputietal.
.
1 Over the last two decades, the space infrared missions have 2007; Magnellietal. 2009; Gotoetal. 2010; Rodighieroetal.
0
made deep extragalactic surveys, which produced a growing 2010; Gruppionietal. 2010; Magnellietal. 2011). At z ∼ 1,
2
1 sample of infrared-luminous galaxies. When IRAS discovered luminousinfrared galaxies (LIRGs) with LIR > 1011L⊙ are re-
a population of ultra-luminous infrared galaxies (ULIRGs – sponsibleformorethanhalfofthisinfraredenergydensity(e.g.
:
v see Sanders&Mirabel1996, fora review),theywere regarded LeFloc’hetal.2005).Thisalsoindicatesthatinfrared-luminous
Xi as very rare objects like QSOs. IRAS also provided good galaxies produced a significant fraction of stellar mass in the
evidence for number density evolution of infrared-luminous presentuniverse.Thus,they were indeedmajorgalaxypopula-
r
a galaxies, although the redshift range was limited to z <∼ 0.2 tioninthepast,givingimportantcluesastohowgalaxiesevolve.
(Hacking&Houck 1987; Lonsdaleetal. 1990; Saundersetal. Statistical samples of (U)LIRGs are vital for studies of galaxy
1990; Gregorichetal. 1995; Bertinetal. 1997). The next in- formationandevolution.
frared mission, ISO, provided further evidence of evolu- AKARI is the first Japanese space mission dedicated to in-
tion towards z ∼ 1 (Floresetal. 1999; Oliveretal. 1997; fraredastronomy(Murakamietal.2007).AKARIwaslaunched
Rowan-Robinsonetal. 1997; Xu 2000; Serjeantetal. 2004; in Feb 2006 with the M-V-8 rocket from the Uchinoura Space
Oyabuetal.2005).Thesemeasurementsaremostlysummarized Centre in Japan, as a second generation all-sky surveyor at in-
intermsofthecosmicstarformationratedensity,estimatedfrom frared wavelengths. The first point-source catalogues from the
theintegratedinfraredluminosityfunction.Probingdeeperand all-sky survey were released in March 2010. Along with this
wider areas, more advanced space telescopes, such as Spitzer, all-sky survey,AKARI conducted5088 pointed observationsin
AKARIandHerschel,firmlyestablishedstrongevolutionofin- selectedareasofsky,duringitsliquidheliumcoldphase.Using
frared galaxies, now in the form of evolving infrared luminos- 13 percent of the pointed observation opportunitiesin the cold
ity function. Results from various extragalactic survey fields phase, we conducted an extragalactic survey around the North
converge specifically in the redshift range z <∼ 1. That is, the Ecliptic Pole (NEP). ThisNEP surveyis two-tiered, consisting
comoving infrared energy density evolves as rapid as (1 + z)4 of the NEP-Deep and the NEP-Wide survey. A salient charac-
1
T.Takagietal.:TheAKARINEP-Deepsurvey:amid-infraredsourcecatalogue⋆⋆
teristic of this survey is its comprehensive mid-IR wavelength
S9W
coverage–weused9photometricbandstospanthewavelength
Outer Annulus
range from 2 to 24µm. The scientific advantage of obtaining
such comprehensive wavelength coverage include a capability
to reliably distinguish starburst-dominatedgalaxies from those Inner Annulus
withAGNcontributions,toobtainingrest-framemid-IRfluxes,
free from the uncertainty of complicated K-corrections. The Central
part
NEP survey providesa valuable input to the study of infrared-
luminousgalaxies.
Thispaperdescribesthe mid-IRsourcecatalogueproduced
with the NEP-Deep survey. The description of the data in the
NEP-Wideisgivenelsewhere(Leeetal.2009;Jeonetal.2010).
We give a brief summary of observations and data reduction
in section 2, including those for ancillary observations from
theground.Sourceidentificationandphotometryarepresented
in section 3, including performance checks using simulated Fig.1.CoveragemapoftheNEP-Deepsurveyat9µm,showing
sources. We describe the generation procedures for the cata- three regions. Square patterns indicate the field-of-view of the
logue,includingopticalidentificationandstar-galaxyseparation MIR-SchanneloftheIRC.Seetextfordetails.
insection4,andgivesomediscussioninsection5.Oursummary
isgiveninsection6.
Use of a concordance cosmology1 gives a scale of
146pointedobservationsforthecentralregion,theinnerandthe
1.8kpc/arcsecatz=0.1.Therefore,themajorityofthegalaxies
appear point-likeat z >∼ 0.1 at the resolution of AKARI, that is outerannuli,respectively.We adoptedanobservingmodewith
∼5′′. neitherfilterchangenordithering(i.e.so-calledAOT05),which
isoptimumfordeepsurveys.SinceeachchanneloftheIRChas
3 filters, we need at least 3 pointed observations to cover the
2. Observationsanddatareduction entire wavelength rangeof the IRC. On average, 4 pointed ob-
servationsperfilterwereconductedatagivenposition.
2.1.AKARIIRC
The observations were executed from 2006 May to 2007
The NEP-Deep survey covers a circular field with an area of August. Unfortunately, the quality of the resulting mid-IR im-
0.67squaredegrees2usingtheIRConboardAKARI,withafield- agedependsontheseasonwhentheobservationwasconducted.
of-viewof 10′×10′. Since detailsofobservationsand datare- From April to August, the observationstowards the NEP were
ductionaredescribedinWadaetal.(2008),wegiveonlyabrief affectedbystraylightfromtheEarthshine.Thisdegradationis
summaryhere. severerforlonger-wavelengthimages,andaffectsthenorth-east
The design of AKARI’s Sun shield and attitude determina- partofthefieldinanareaof∼0.15deg2.
tion system require that the optical axis of the telescope is al- We used the standard IRC imaging pipeline version
ways kept pointing 90◦ from the direction to the Sun with a 200710174foreachpointedobservation,inwhichdarksubtrac-
tolerance of only ±0.6◦. Due to this strong visibility constraint tion,subtractionofscatteredlightinsidethecamera,correction
in its sun-synchronous orbit, deep surveys are possible only for detector non-linearity, flat fielding, and correction for dis-
close to the Ecliptic Poles. Given the presence of the Large tortion are performed. For the MIR-S and MIR-L images, we
Magellanic Cloud near the South Ecliptic Pole, we chose the also removedthe diffuse backgroundby subtracting a median-
NorthEclipticPole(NEP)forAKARI’suniquedeepsurveyfield filteredimage.Astrometryinallofthenear-IRandS7andS9W
(Matsuharaetal.2006). images was determinedby the IRC pipeline using 2MASS ob-
The circular field of the NEP-Deep survey can be divided jectsasareference.Forimagesatlongerwavelengths,weused
intothreeregions–thecentralfield,andtheinnerandouteran- thecoordinatesofbrightsourcesintheimageone-bandshorter
nuliasshowninFigure1.Thesizeoftheinnerannulusisdeter- inwavelengththantheimageinquestion.Theindividualimages
minedbytheangularseparationbetweentheNIR/MIR-Schan- of a given band with calibrated astrometry were combined to
nel and the MIR-L channel3. During a pointed observation of produce a final mosaicked image by using a publicly available
theinnerannulus,theNIR/MIR-Schannelsalwaysobservethe software, Swarp5. The images were combinedby taking a me-
oppositesideoftheMIR-Lchannel.Theangularseparationbe- dian value of the corresponding individual images. Overviews
tweenthecentersoftheIRCchannelsis20′.Thisseparationand andclose-upsofthesefinalimageshavebeenpresentedinWada
thefield-of-viewofIRCresultinacentralhole∼10′indiameter. etal.(2008).
Thiscentralholedefinesthecentralregion(RA=17h56m,Dec
Wada et al. (2008)have measured the sky noise limit from
= 66◦37′) and was covered with one field-of-view of the IRC.
photometry of random positions containing no sources. The
Because of the differencein the field shape, i.e. a square field-
5σ detection limits of a point source are estimated to be
of-view for a circular area, some small gaps exist between the
9.6µJy in N2 band, 7.5µJy (N3), 5.4µJy (N4), 49µJy (S7),
centralregionandtheinnerannulus.SinceAKARIhasnocapa-
58µJy(S9W),71µJy(S11),117µJy(L15),121µJy(L18W)and
bilitytocontrolitsrollangle,weutilizedtheseasonalvariation 276µJy(L24)6.Hereafterweusethesevaluesasthenominalde-
ofthepositionangletocoverthesurveyfield.Theradiusofthe
tectionlimitsoftheNEP-Deepsurvey.
outer annuluswas set to 20′ so that there would be no gap be-
tweentheinnerandouterannuli.Intotalweallocated23,63,and
4 SeeAKARIIRCDataUserManualver1.3
1 Ωm =0.3,ΩΛ=0.7andH0 =70 kmsec−1Mpc−1 5 Availableathttp://www.astromatic.net/
2 AreacoveredwithL18Wbeforeanymasking. 6 CorrectedfortheerrorintheconversionfactorusedinWadaetal.
3 TheNIRandMIR-Schannelssharethesamefield-of-view. (2008)
2
T.Takagietal.:TheAKARINEP-Deepsurvey:amid-infraredsourcecatalogue⋆⋆
2.2.Ancillarydata fore we will give only a brief summary here. We conducted
FLAMINGOSobservationsin the J and K bandsto coverthe
For the NEP field, we performed various follow-up obser- s
S-cam field in the NEP-Deep field. We coveredthe target area
vations from X-ray to radio wavelengths (e.g. Whiteetal.
withobservationsat4differentpositions.Thetotaleffectivearea
2010a,b).Optical-NIRphotometryofmid-IRsourcesinthispa- is 750arcmin2, with stellar image sizes ranging from 1.1′′ to
perutilizesthefollowingdata–Subaru/Suprime-cam(BVRi′z′), 2′′ in FWHM. Data were reduced in the standard manner us-
CFHT/Megacam(u∗),KPNO2.1m/FLAMINGOS(JK ),which
s ingtheIRAFpackages.ThedeepestregionsreachJ =22.2and
cover ∼ 40 per cent of the NEP-Deep field, except for the u∗
K =21.3ABmagnitudein5σ.
s
band. An optical source catalogue covering the whole NEP-
Deep field with CFHT/Megacam (g′r′r′z′) is published in
Hwangetal.(2007),towhichreadersinterestedinopticalprop-
ertiesofmoresourcesarereferred. 3. Sourceidentificationandphotometry
3.1.SourcedetectioninMIR-SandMIR-L
2.2.1. Subaru/Suprime-cam
We combined the S7, S9W, and S11 images with the terapix
WeusedtheSubaruSuprime-cam(S-cam)toobtaindeepoptical
software ’SWarp’ to produce a detection image for the MIR-S
imagesforasinglefield-of-view,i.e.34′×27′,covering38per
channel,median-coaddingthreeseparateimages.FortheMIR-
centoftheNEP-Deepfield.ObservationsintheB,R,i′,z′bands
L,weusedtheaverageoftheL15andL18W imagesforsource
were conducted in June 2003. Additional V-band observations
detection, since the quality of L24 image was systematically
were performed in Oct 2003 and June 2004. The central posi-
lower. In the resulting MIR-S and MIR-L detection images,
tion of the field (RA=17h55m24s, Dec=+66◦37′31′′ [J2000])
theresidualfromtheskysubtractionwasreducedsignificantly,
andthepositionangle(PA=0◦)wereselectedtoavoidthepres-
whichprovidedmorereliablesourcecatalogs.
enceofbrightstars(V <10mag)inthefield.
For each detection image, we ran SExtractor
We obtainedintegrationtimes of 12960sec for the B band,
7421.2secforV band,7200secforRband,6900secfori′band, (Bertin&Arnouts 1996) to produce initial source lists of
and 10080sec for z′ band, with a typical seeing of 0.6′′ – 1′′. mid-IR sources. In order to make these initial source lists as
Frames which had seeing worse than 1′′ were not used in pro- complete as possible, we adopted a relatively low threshold
for source detection, i.e. requiring 5 connected pixels with
cessing the final mosaiced images. The data were reducedin a
>1.2σ. Thisresulted in 6746MIR-Ssourcesand 6719MIR-L
standard mannerusing SDFRED (Yagietal. 2002; Ouchietal.
sources. Mainly because of the backgroundfluctuation and the
2004).SourcesareextractedwithSExtractor(Bertin&Arnouts
1996) using the z′-band image as a detection image. The 5-σ lowthreshold,thesemayincludeanumberofspurioussources.
limitingmagnitudesmeasuredwitha2′′aperturewere28.4mag Thus, these sources were regarded as ‘candidates’, and are
for Bband,28.0magforV,27.4magforR,27.0magfori′ and screened with criteria described below. In the final catalogue,
the MIR-S and MIR-L catalogues were concatenated where
26.2magforz′ intheABmagnitudesystem.
duplicatedentriesareremoved.
2.2.2. CFHT/Megacam(u∗)
3.2.Photometryandband-merging
We obtained an UV image of the whole NEP-Deep field with
CFHT/Megacam. Observations in the u∗ band were taken in WefollowedtheproceduresofTakagietal.(2007)forphotom-
queuemodespreadover12nightsfromApril2007toSeptember etryandband-mergingforeachMIRchannel.Startingfromthe
2007, resulting in a total of 77 frames. The field of view of initial source position obtained with SExtractor, we searched
MegacamwaslargeenoughtocovertheentireNEP-Deepfield the centroid position for sources in each IRC image. The re-
withthecentralpositionofRA17h55h24sandDec+66◦37′32′′ evaluatedcentroidpositioncouldshiftfromtheSExtractorposi-
[J2000].Withtheintegrationtimeof600 secforeachframe(2 tionsignificantlyincaseofseveresourceblendingorlowsignal-
frameshave 680sec), we achievedthe total integrationtime of to-noise ratio. Therefore,we performedphotometryonly if the
773min.Mostofthesedata–56framesoutof77–weretakenat newcentroidislessthan3arcsecawayfromtheinitialposition,
airmasseslessthan1.5andtypicalseeingof1.0′′. whichcorrespondsto∼2σoftherelativeoffsetbetweenMIR-S
The data reduction was carried out in a standard manner, andMIR-Lsources.
using software developed for the large-format Megacam data. WeperformedaperturephotometryineachIRCimageatthe
Thepre-processing,includingbad-pixelmasking,overscanand centroid positions determined above, adopting a 2-pixel (2.9′′)
bias subtraction, and flat-fielding, were carried out before the radiusforNIRimagesand3-pixel(7.0′′)radiusforMIR-Sand
data delivery using the pipeline system ‘Elixir’. By stacking MIR-L images. We then applied the aperture corrections, esti-
the Elixir-processedimages,we produceda finalmosaicedim- mated from the average digitized point spread function (PSF)
ageusingasoftwarepackageprovidedbyTERAPIX7including createdfromstackingofbrightsourcesaroundtheNEP.Inthis
WeightWatcher, SExtractor, SCAMP and SWarp. The depth of stacking, we did not apply any weight as a function of fluxes,
the final image was estimated to be 24.6 mag [5σ; AB] from since high weight in bright sources also results in high weight
photometryof30000randompositionswith2′′aperture. forassociatedsourcesin theoutskirts.Thisreducedthesignal-
to-noiseratiooftheouterpartofthePSF,andincreasedtheun-
certaintyofthe aperturecorrection.We showthe averagePSFs
2.2.3. KPNO2.1m/FLAMINGOS
andthegrowthcurvesinFigure2and3,respectively.Fromthese
Imaietal. (2007) describe observations with KPNO2.1m growthcurves,wederivedtheaperturecorrectionfactors,which
FLAMINGOS and its data reduction in detail, and there- are tabulated in Table 1 along with other characteristics of the
PSFs.Theaperture-correctionfactorsderivedherearedefinedas
7 http://terapix.iap.fr theratiooffluxeswithtwodifferentaperturesizes,i.e.thesmall
3
T.Takagietal.:TheAKARINEP-Deepsurvey:amid-infraredsourcecatalogue⋆⋆
N2 N3 N4
S7 S9W S11
L15 L18W L24
Fig.2. Average PSFs obtained with stacking of bright sources, displayed with a linear grey scale covering 99% of the dynamic
range.Solidcurvesindicatetheareawithsignal-to-noiseratioof5orgreater.
Fig.3. Growth curves of enclosed fluxes as a function of aperture radius. The vertical axis reports relative enclosed fluxes, nor-
malized to the value at a 15-pixel radius. Dotted and dashed lines indicate adopted apertures for photometry and those for flux
calibrationinTanabe´etal.(2008).
Table1.SummaryofpointspreadfunctionsofIRC
N2 N3 N4 S7 S9W S11 L15 L18W L24
Referencewavelength[µm] 2.4 3.2 4.1 7.0 9.0 11.0 15.0 18.0 24.0
#ofpostagestampsuseda 168 114 43 45 28 24 56 66 43
FWHM(pixel)b 2.96 3.06 2.84 2.27 2.19 2.10 2.27 2.48 2.92
Apertureradius(pixel) 2 2 2 3 3 3 3 3 3
Aperturecorrectionfactorsc 1.97 1.96 1.95 1.18 1.21 1.25 1.42 1.53 1.70
Pixelscale 1.46×1.46 2.34×2.34 2.51×2.39
aThenumberofstackedpostagestampimagesforPSFcreation
bFWHMcomputedusingtheenclosedfluxradialprofile
cCorrectionfactorsforfluxes
4
T.Takagietal.:TheAKARINEP-Deepsurvey:amid-infraredsourcecatalogue⋆⋆
apertures adopted here and the IRC calibration apertures8 (10 3.4.1. Photometricerrors
pixelsfortheNIRchanneland7.5pixelsfortheMIRchannels,
We performedaperturephotometryforartificialsourcesaswas
seeTanabe´etal.2008).Thebackgroundlevelwasestimatedin
doneforthe realsources,andthenappliedthe aperturecorrec-
theskyannulusbetweena15-and20-pixelradius.
tions. The centroid positions of the artificial sources were de-
Althoughthisphotometricmethodis simple, it is relatively
terminedthroughGaussian-fittingofthesimulatedpointsource
robustagainstthe effectsof sourceblending,comparedtopho-
image,usingtheinputpositionasaninitialguess.Thus,theef-
tometrywith largeraperturesadoptedin the IRC calibrationor
fect of noise on the source position was included in the simu-
MAG AUTOphotometryinSExtractor.Asimilarapproachwas
lation.In Table 2,we havetabulatedthe relativeflux errorses-
successfully applied to photometry of blended sources in pre-
timatedfromtheinput-to-outputfluxratio.Wenotethatthear-
vious works (e.g. Koetal. 2009). In heavily confused images,
tificialsourceswereaddedtotheimageswithrealsources,and
photometry with simultaneous PSF-fitting for multiple sources
therefore the photometric errors reported here have some con-
maybea bettersolution.However,thereareargumentsagainst
tributionfromthesourceblending.Herethephotometricerrors
PSFfittinginourcase.Itisdifficulttoaccuratelydeterminethe
werecalculatedfromthescatteroftheoutputfluxeswiththere-
PSF in each frame, due to uncertainty in the pointing stability
jectionofoutliers,whichareduetothepresenceofnearbyreal
and the lens aberration whose direction is fixed in the detec-
brightsources.Althoughthecataloguegenerationmethodisop-
torarray.ThishampersrigorousPSF-fittinganalysis.Moreover,
timized for pointsources, the catalogueincludesnearbygalax-
faintsourceshaveonlyafewsignificantpixelsthatareusefulfor
ies,forwhichIRCfluxesarenotasaccurateasforpointsources.
fitting. Therefore,we adopted aperture photometry,rather than
Wealsocautionthatnear-IRfluxesoftheseveralbrighteststars
PSF-fittingphotometry.
aresaturatedandthereforenotaccurate.
Figure 4 compares aperture-corrected fluxes to those ob-
tained with the aperture photometry using the aperture size
adopted for the IRC calibration, i.e. 10 pixels for NIR and 7.5 3.4.2. Completenessandreliability
pixels for MIR-S and MIR-L. For bright sources, these fluxes
The completenessof the resulting catalogue is estimated again
areconsistentwitheachother,exceptforthebrightestonesthat
with the artificial-source method. To reduce the Poisson error
havesaturatedpixels.Thisconfirmsthattheaperturecorrection
of the completeness, we produced a large number of artificial
obtainedfromtheaveragePSFsisreasonable.Ontheotherhand,
sources–over 160,000. Since we combine multi-band images
fluxesoffainterobjects,butdetectedwith∼10σ,showsystem-
for source detection, the completeness depends on the colour
atic deviation–fluxeswiththe calibrationaperturearesystem-
of sources. We consider two cases for the MIR-S; a) flat spec-
aticallylarger.Forfaintersources,thenumberdensityincreases,
trumin f andb)redsources,forwhichweadoptthefollowing
andthereforetheprobabilitythatneighbouringsourcesfallinthe ν
fluxratios: f /f = 6 and f /f = 10in Janskyunits.
largeapertureforIRCcalibrationincreasesaswell.Theseasso- 9µm 7µm 11µm 7µm
This correspondsto the reddest coloursamong the sourcesde-
ciatedsourceswouldboostthefluxeswithlargeaperturesizeas
tected inall ofMIR-Sbands.9 The averageflux ratiosof MIR-
evidentlyseeninFigure4.
S–detected sources were found to be hf /f i = 1.8 and
9µm 7µm
hf /f i = 2.6. For the MIR-L channel,only the flat spec-
11µm 7µm
3.3.Astrometry trum case was considered, because we only combined 15 and
18µmimageswheremostofgalaxieshavesimilarfluxes.
TheastrometriccalibrationofIRCimageshasalreadybeendis-
WeranSExtractoroneachsimulatedimagewith200artifi-
cussed by Wada et al. (2008), and will not be repeated here.
cialsources.A sourcewasconsideredtoberecoveredifitsex-
We adopt the coordinate at the shortest wavelength band de- tractedpositionwaswithinaradiusof3′′.Insomeotherstudies,
tected as the final IRC coordinate of MIR sources. In Figure
the consistency of recovered fluxes is also taken into account
5, we compare the IRC coordinates with 2MASS coordinates
as a condition of the source recovery, in order to avoid mis-
for 1178 sources. The mean and σ of the coordinate shift are
identification with real sources. In our case, this flux check is
−0.0014′′and0.29′′forRightAscension,and0.011′′and0.32′′
notstraightforward,sincethedetectionimageswemadearenot
forDeclination,respectively.
calibrated.
Weestimatedtheeffectofsourceconfusioninthecomplete-
3.4.Simulationswithartificialsources nessanalysisasfollows.Weaddartificialsourcesinthenegative
detectionimage,createdbyinvertingthebackgroundsubtracted
Inordertoestimatethephotometricuncertaintiesandcomplete- detection image, and compare the results with the case for the
ness of our catalogue, we made a careful simulation to esti- normalpositive image. In Figure 6, we show the completeness
mate this. Artificial sources were randomly added to the final thus obtained for both positive and negative images. For both
NEP-Deepimagesandphotometrywas donein the same man- MIR-Sand MIR-Lchannels,we findthat the completenessfor
ner as for real sources. We consider only point sources here, thenegativeimageissystematicallyhigherthanthatforthepos-
sincethesourcedetectionandphotometryareoptimizedforsuch itive image. In the completeness analysis, we could wrongly
sources.WeusedthePSFsshowninFigure2foreveryartificial identify real sources as artificial sources we generated. If such
source, adding 200 artificial sources (less than ∼3 per cent of mis-identificationshavesignificantimpactonthecompleteness
true sources) at a time, and repeating this for a total of 40,000 analysis, we would expect that the completeness found for the
and160,000artificialsources,toestimatethephotometricerrors positive image with real sources would be higher than that for
andcompleteness,respectively. thenegativeimage.Sincethisisnotthecase,theeffectofmis-
8 IncalibratingtheIRCphotometry, Tanabe´etal.(2008) compared 9 Ourmainconcernforthesourceextractionprocedurewastheef-
the aperture photometry with the expected total fluxes from spectral fects of combining three MIR-S images taken at the different wave-
models of calibrationstars. Therefore, the IRCcalibration factors are lengths. In this procedure, sources with extreme colours may not be
supposed toinclude theaperture correctionfromthecalibrationaper- detectedinsomeof MIR-Sbands. Inorder toevaluateconservatively
turephotometrytothetotalfluxes. sucheffects,weconductedthesimulationwiththemostextremecolour.
5
T.Takagietal.:TheAKARINEP-Deepsurvey:amid-infraredsourcecatalogue⋆⋆
Fig.4. Fluxes with aperture correction adopted here versus those using the IRC calibration apertures, i.e. 10 pixels for NIR and
7.5pixelsfor MIR-SandMIR-L.Dashedlines indicate5σ sensitivity inWada etal. (2008).Solidlinescorrespondto thelinear
relation.
Fig.5. Comparisonof IRC coordinateswith the 2MASS coordinates.In the histogramplot, solid and dashedlines indicate ∆RA
and∆Dec,respectively.
identificationis likely to be negligible.The difference between Inordertoevaluatethecolourdependenceofthecomplete-
thecompletenessforthepositiveimageandforthenegativeim- ness, we calculate the completeness of red sources as a func-
ageimpliesanotherimportanteffect–sourceblending.Whenwe tion of the average flux of three MIR-S bands, i.e. hf i =
ν MIRS
addartificialsourcesinthepositiveimage,someofthemwould 5.67·f fortheassumedcolour,whichiscloseto f .There-
7µm 9µm
blend with real sources, which results in the positionalshift of sultingcompletenessisplottedinFigure6,andfoundtobeclose
theextractedsourceandnon-recovery.Atthe5σsensitivityof tothecasefortheflatspectrum.Thus,weconcludethatthecom-
IRC wide band images, i.e. S9W and L18W, the completeness pletenesshasweakcolourdependence,whentheaveragefluxes
forthenegativeimageis∼10%higherthanthatforthepositive ofobjectsareconsidered.Sinceweconsiderveryredsourcesin
image in both MIR-S and MIR-L channels. This indicates that thissimulation,itmightbeexpectedthatthecompletenessofthe
∼10% of mid-IRsourcesare missed, because of sourceblend- longest wavebands, 11µm, using the multi-band detection im-
ing. age,couldbeworsethanthecaseofusingtheS11imageasthe
6
T.Takagietal.:TheAKARINEP-Deepsurvey:amid-infraredsourcecatalogue⋆⋆
detectionimage.Thecompletenessofthesingle-bandsourceex-
tractionusingthesameimageisreportedbyWadaetal.(2008).
They obtained 80.7µJy for the 50% completeness at the S11
band.Ontheotherhand,the50%completenessweobtainedfor
red sources is 50.6µJy at the same band, which is comparable
to or evenbetter thanthe valuereportedby Wadaetal. (2008).
Thus,onaverage,weseenosignofdegradedcompletenessfor
redsources.
We estimate the reliability of the catalogue as described
below. Using the same method as for real sources, we made
a source catalogue for the negative detection images for both
MIR-S and MIR-L as described above. In band-merged cata-
logues,itisexpectedthatsourcesdetectedinfewerbandswould
havelowerreliability.InTable3,wesummarizethestatisticsof
sourceswith the multiplebanddetectionfor6 bandsin MIR-S
and MIR-L altogether. Although, from the negative image, we
obtain some spurious sources with multiple-band detection, it
turnsoutthatalmostallofthemarerarecasesproducedbythe
effect of invertedreal sources. Therefore,we do not take these
spurious sources into account, and conclude that sources with
multiple-banddetectionare highly reliable.In the negativeim-
ages,thereare113spurioussourceswithsinglebanddetection.
Theseareallfaintsourceswiththefluxesclosetothedetection
limits. On the other hand,in the positive images, we find 2138
sourceswithsinglebanddetection.Giventhesenumbers,wees-
timatethat∼5%ofsourceswithsingle-banddetectionarefake.
Thiscorrespondsto1.5%ofsourcesinthefinalcatalogue.
Fig.6.CompletenesssimulationfortheMIR-SandMIR-Lchan-
nels. Solid and dashed lines represent the completeness calcu-
lated with the positive and negative images, respectively. For
thesecases,weassumethatartificialsourceshaveaflatspectrum
4. Catalogue in fν.Thedot-dashedlineintheupperpanelindicatesthecom-
pletenessforredsources(see textin detail),wherewe adopted
4.1.MIR-mergedcatalogue the average of 7, 9, and 11µm flux for red sources. The verti-
caldashedlinesindicate5σsensitivityinWadaetal.(2008)at
Atthispoint,wehadtwocatalogues,basedonobservationswith S9WandL18WbandsforMIR-SandMIR-L,respectively.
theMIR-SandMIR-Lchannels,whichwethenmergedtogether
by eliminating duplicated entries. In order to choose the best
4.2.Opticalidentification
photometricresultsfromtheduplicatedentries,wecheckedthe
signal-to-noiseratio at each IRC band, and selected the source We made opticalidentificationswithin the Subaru/S-camfield.
withthe maximumnumberof> 3σdetections.Finally,were- AlthoughtheS-camfieldcoversonly38%oftheNEP-Deep,it
movedthe sources whose fluxes are lower than 5σ in all MIR isdeepenoughtodetectalmostalloftheAKARIMIR-detected
bands.Weadoptthe5σsensitivityfromWadaetal.(2008).We sources.Wefoundmultipleopticalcounterpartsfor29%ofMIR
obtained7284mid-IRsourcesinthefinalcatalogue. sources in our R-band S-cam image within 3′′ radius. In order
Theeffectsofsourceblendingarenotnegligibleinthepro- to find the best candidate, we adopt the maximum likelihood
method (Sutherland & Saunders 1992) for the optical identifi-
cessofcataloguegeneration,sinceweassumethatsourceswith
separations of < 3′′ are the same. There are multiple optical cation.Thelikelihoodratioisdefinedas
counterpartsfor29%(8%)ofMIRsourceswithin3′′ (2′′).The q(m)f(x,y)
L= , (1)
ratioofsourcedensityindicatesthatonly3%ofopticalsources
n(m)
intheSubaru/S-camimagearedetectableintheMIR,andthere-
forethefractionofsourceswithseriousblendingintheIRCim- whereq(m)istheinfinitesimalprobabilitythataMIRsourcehas
ageshouldbemuchlowerthanthesourcefractionwithmultiple anopticalmagnitudeofm, f(x,y)istheprobabilitydistribution
opticalcounterparts. functionofthepositionalerrorassumedtobeatwo-dimensional
Gaussian,andn(m)isthesurfacedensityofbackgroundobjects
Forsourceswithmoderateseparation,inspectionbyeyecan
with magnitudem. To derive the best estimate of q(m), we de-
easilyspottheproblemofblendingwhichaffectsboththesource
finedasub-sampleof1100IRCall-band-detectedsources,per-
detectionandphotometry.Inordertocallattentiontothem,we
formed optical identification with a simple nearest neighbour
setflags(groupIDandthenumberofgroupmember)toidentify
method, and visually checked all of the optical identification.
groups of MIR sources with separation of < 5′′. The sources
Wederivedq(m)usingthissub-samplewithvisually-confirmed
with group ID should be treated with caution. Some of them
optical identification. The resulting q(m) is shown in Figure 7,
haveacommonopticalcounterpart.
alongwithn(m)fromtheSubaruR-bandimage.Weassumethat
A small portionof this band-mergedcatalogue is shown in thisdistributionis notverydifferentfromthatforgeneralMIR
Table 4,which isprovidedin fullin the electricversionofthis sourcesdetectedintheNEP-Deepsurvey.For f(x,y),weadopt
paper. the astrometric dispersions described above, i.e. 0.29′′ for RA
7
T.Takagietal.:TheAKARINEP-Deepsurvey:amid-infraredsourcecatalogue⋆⋆
Table2.Summaryofsimulationforphotometricerrorsandcompleteness
Flux Photometricerrors Completenessa
(Jy) N2 N3 N4 S7 S9W S11 L15 L18W L24 MIR-S MIR-L
3.0e-06 ...b ... 0.817 ... ... ... ... ... ... 0.03 0.03
6.0e-06 ... 0.678 0.395 ... ... ... ... ... ... 0.04 0.03
1.2e-05 0.351 0.326 0.284 ... ... ... ... ... ... 0.10 0.04
2.4e-05 0.230 0.152 0.139 0.971 1.056 ... ... ... ... 0.34 0.06
4.7e-05 0.151 0.085 0.088 0.600 0.652 0.834 1.674 1.694 ... 0.70 0.19
9.4e-05 0.089 0.047 0.046 0.326 0.347 0.521 0.781 0.897 ... 0.85 0.54
1.9e-04 0.055 0.031 0.024 0.172 0.183 0.264 0.427 0.456 1.009 0.93 0.79
3.8e-04 0.027 0.015 0.011 0.087 0.094 0.143 0.244 0.261 0.504 0.95 0.89
7.5e-04 0.015 0.008 0.006 0.044 0.047 0.070 0.121 0.137 0.255 0.97 0.94
1.5e-03 0.007 0.004 0.003 0.022 0.024 0.034 0.066 0.069 0.134 0.99 0.95
3.0e-03 0.004 0.002 0.002 0.011 0.012 0.017 0.035 0.036 0.068 0.99 0.97
6.0e-03 0.002 0.001 0.001 0.006 0.007 0.009 0.018 0.017 0.035 0.99 0.98
1.2e-02 0.001 0.001 0.000 0.003 0.003 0.004 0.010 0.010 0.019 0.99 0.99
2.4e-02 0.001 0.000 0.000 0.002 0.002 0.002 0.005 0.005 0.010 0.99 0.99
4.7e-02 0.000 0.000 0.000 0.001 0.001 0.001 0.002 0.002 0.005 1.00 0.99
9.4e-02 0.000 0.000 0.000 0.001 0.000 0.001 0.001 0.002 0.003 1.00 1.00
aCompletenessinMIR-SandMIR-Ldetectionimagesforthecaseofconstantfluxes.
bPhotometricerrorsaregivenif|1−R |<0.2,whereR istheaverageoftheinput-to-outputfluxratio
ave ave
Table3.ThenumberofsourceswithmultipleMIRbanddetections
#ofdetectedbands 1 2 3 4 5 6 Total
#ofsources 2138 1736 1286 756 705 669 7284
Fraction 0.29 0.24 0.17 0.10 0.097 0.092 1
and0.32′′forDec.WecalculatedLforobjectswithinthesearch
radius of 3′′ and selected the object with the highest L as the
bestcandidatefortheopticalcounterpart.Asmallportionofthe
resultingcatalogueofopticalcounterpartsispresentedinTable
5.
Asexpected,theresultingopticalIDwithlikelihoodratiois
sometimes different from the nearest neighbour. For 915 MIR
sources with multiple optical counterparts in the S-cam field,
we foundthat 234 sources have an optical ID which is not the
nearestneighbour.For these sources,we visually inspectedthe
results, but it was not useful to identify correct counterparts.
Thus,wehaveseriousproblemsofopticalIDfor∼10%ofMIR
sources. Also, we spotted several pairs of MIR sources which
happentohaveacommonopticalcounterpart.Therefore,optical
Fig.7.Probabilitydistributionfunctionq(m)(shadedhistogram)
identificationofsourceswithcloseneighboursshouldbetreated
and n(m) (empty histogram). Probability densities estimated
withcaution.Inthecatalogue,thesesourcesareidentifiedwith
fromthesehistogramsareover-plottedwithsolidcurves.
thegroupIDflag.
Out of 7284mid-IRsources, we found3162sourcesin the
S-camfieldandopticalcounterpartsforallbut79sources.Most
of mid-IR sources with no optical ID were detected with only
a single mid-IR band, which means they are likely to be unre-
liable sources. However,we find that 9 sources out of 79 were Subaru/R S7 S9W S11
detectedinmorethan3mid-IRbands.Asaresultofthisvisual
inspection,we foundthattheir opticalcounterpartsare pairsof
Fig.8. Postage stamp of a mid-IRsource (MIRS4404)with no
brightsourcesorlieclosetoopticallybrightsourcesaffectedby
obviousopticalcounterpartsintheSubaru/S-camimage.Circles
spikes,whichhamperthecorrectidentification.Interestingly,we
indicatetheIRC/MIR-Spositionwith3′′radius.
findagenuineoptically-blankmid-IRsourceevenwiththedeep
Subaru/S-camimage,whichisshowninFigure8.TheR−L15
colourofthissourceis9.5mag(AB),whichisabout3magred-
derthanthecriteriontochoosethereddestinfraredsources,such
asfaintdust-obscuredgalaxies(DOGs–Deyetal.2008).
8
T.Takagietal.:TheAKARINEP-Deepsurvey:amid-infraredsourcecatalogue⋆⋆
4.3.Star-galaxyseparation
WemainlyuseN2−N3andN3−N4coloursforidentification
of stars, since these near-IR bands are the most sensitive IRC
bands.IntheVega-basedmagnitudesystem,thenear-IRcolours
of normalstars are close to zero. In this subsection, all magni-
tudesaregivenintheVega-basedmagnitudes.Wefirstlymadea
tentativelistofstars,basedonastellaritymeasuredintheCFHT
r′-band image (Hwangetal. 2007). In Figure 9, we show the
N3−N4versusN2−N3colour-colourplotforstar-galaxysep-
aration.Inthisplot,objectswithalargestellarity,i.e.stellarob-
jects, makeaclumparoundzero-colors,asexpected.Theaver-
agecoloursofthesestellarobjectsarehN2−N3i=0.049±0.057
andhN3−N4i=0.006±0.05610.Wedrawacircularboundary
in this colour-colourplotto separate stars and galaxiesdefined
with the radius ∆C(≡ p∆(N2−N3)2+∆(N3−N4)2) of 0.15,
wherewe adoptthemeancoloursofstellar objectsas a centre.
Thisclassificationresultsin673stars,outof7284sources.
Alarge∆Cforthecolourboundarywouldincreasethecom-
pletenessofstars, butcouldcausemis-classificationsofnearby
galaxiesasstars.Therefore,weadditionallyusefollowingcrite-
ria;−1< N2−S11<1andN2<13magforstars.Thiscolour
is useful to separate stars without mid-IR excess from nearby Fig.9. Near-IRcolour-colourplotin Vega magnitudesfor star-
galaxies. galaxyseparation.Objectswithlargestellarity(>0.9998)arein-
Furthermore,somebrightstarsweresaturatedatthecentral dicatedwithcrosses.Alargecircleindicatestheadoptedbound-
pixels,specificallyinthenear-IRbands.Thiscausesyetanother aryforstar-galaxysepation(seetextindetail).Zero-coloursare
mis-classificationifweusenear-IRbandsforstar-galaxysepara- markedwithdashedlines.
tion.Toremedythiseffect,weadoptadditionalcriteriaforstars,
i.e.S7 < 10magand−1 < S7−S11 < 1.Withthesecriteria,
sources at 11µm, where the prominent PAH emission around
wefinallyobtained720starsintotal.
8µmisredshiftedoutofthebandpassforz>∼0.4.
ThereisaprominentgalacticsourceintheNEP-Deepregion
A typical redshift of catalogued galaxies can be estimated
– a planetarynebula NGC6543.The IRC detectedfilaments of
fromAKARIcolours.Figure11showsthecolour-colourplotus-
this planetary nebula, which should be flagged. Inside this PN
area,photometryofdistantsourcesisseverelyaffectedbythese
filaments.Therefore,wesimplyflaggedoutallofsourceswithin
107
thePNarea.Thecentralpositionofa circularmaskisadjusted
toencompassthePNareawiththeradiusof194′′.
1]
-sr
5. Discussion 1.5·106
y
J
We calculated the number counts of mid-IR sources at 11, 15, m
18 and 24µm shown in Figure 10. In deriving the counts, we 2.5 [
S
first defined the area to be used for each image based on the S·
weightmap.We foundmostlow-weightareasin thenorth-east N/d105
partofthefield,inwhichAKARIobservationssufferedfromthe g d SL1115
Earth shine. To reduce this effect, we excluded the field with lo Stars LL1284W
theweightofthelowest20%.Asaresultofthiscut,weended Spitzer 24µm
up with areas of 0.46,0.53,0.51, and 0.46deg2 for 11, 15, 18, WSpIiStzEe r1 126µµmm
and24µm,respectively.Theexcludedareais0.167deg2atmost. 104-2 -1 0 1 2
Thecountsarecorrectedforthecompleteness.Wecompareour
log Flux [mJy]
L24 counts with Spitzer 24µm counts (Papovichetal. 2004),
andfinda reasonableagreement.However,evenwith the com-
pletenesscorrection,thecountsbelowthe5σdetectionlimitex- Fig.10. Number counts of galaxies and stars at 11, 15, 18,
hibit a significant under-estimation, compared to the Spitzer’s and 24µm. Large and small symbols with error bars indi-
counts.Therefore,inFigure10,weonlyshowthecountsabove cate galaxy and star counts, respectively. See legend for in-
the 5σ detection limits. For other mid-IR number counts in dividual bands. For the L24 band, the number of stars is not
AKARI bands, see Wadaetal. (2007, 2008); Leeetal. (2009); enough to derive statistical counts. Dashed lines with symbols
Pearsonetal.(2010)andPearsonetal.(2011,inpreparation). indicate Spitzer 24µm counts (Papovichetal. 2004), Spitzer
The wavelength-dependenceof number counts from 11-to- 16µm counts (Teplitzetal. 2011), and WISE 12µm counts
24µmcouldbeexplainedbytheSEDsofgalaxies,wherePAHs (Jarrettetal. 2011) from top to bottom. We show the number
and hot dust emission dominate. We find the least number of countsforfluxbinscontainingatleast30objectsandwithflux
greaterthanthe5σdetectionlimit.Allcountsarecorrectedfor
10 Withnocolourcorrections completeness.
9
T.Takagietal.:TheAKARINEP-Deepsurvey:amid-infraredsourcecatalogue⋆⋆
forthePromotionofScience(JSPS;grantnumber18·7747and
3 Sc template 21340042).H.M. Lee was supported by NRF grant No. 2006-
M82
Sb template 341-C00018.MIwassupporedbythegrantNo.2010-0000712
2 oftheNRFK/MEST.Thisresearchisbasedonobservationswith
AKARI,aJAXAprojectwiththeparticipationofESA,datacol-
B]
A 0.5 lected at Subaru Telescope, which is operated by the National
1 [ 1 Astronomical Observatory of Japan, observations at Kitt Peak
1
S NationalObservatory,NationalOpticalAstronomyObservatory,
S7- 0 and also observations obtained with MegaPrime/MegaCam, a
z=0
1 jointprojectofCFHT andCEA/DAPNIA.Theauthorswish to
1.5 2 thank the referee, whose comments are helpful to improve the
-1
contentsofthiswork.
-2
-1 -0.5 0 0.5 1
N2-N3 [AB] References
Babbedge, T.S.R.,Rowan-Robinson, M.,Vaccari, M.,etal.2006, MNRAS,
Fig.11.Colour-colourplotwithAKARIbands,usingN2,N3,S7 370,1159
Bertin,E.&Arnouts,S.1996,A&AS,117,393
andS11.Dots(black)andsmallcrosses(green)indicategalax-
Bertin,E.,Dennefeld,M.,&Moshir,M.1997,A&A,323,685
ies andstars, respectively,detectedatmorethan 3σ in allof 4 Caputi,K.I.,Lagache,G.,Yan,L.,etal.2007,ApJ,660,97
bands.Solidline(red)indicatesthecolourofScgalaxytemplate Dey,A.,Soifer,B.T.,Desai,V.,etal.2008,ApJ,677,943
as a functionofredshift.Dot-dashed(black)and dashed(blue) Elbaz,D.,Cesarsky,C.J.,Chanial,P.,etal.2002,A&A,384,848
Flores,H.,Hammer,F.,Thuan,T.X.,etal.1999,ApJ,517,148
linesareforSbgalaxyandM82template,respectively.TheSED
Goto,T.,Takagi,T.,Matsuhara,H.,etal.2010,A&A,514,A6+
templatesaretakenfromPollettaetal.(2007).
Gregorich,D.T.,Neugebauer,G.,Soifer,B.T.,Gunn,J.E.,&Herter,T.L.1995,
AJ,110,259
Gruppioni,C.,Pozzi,F.,Andreani,P.,etal.2010,A&A,518,L27+
ingAKARIbands,S7−S11versusN2−N3.Bothcolourshave
Hacking,P.&Houck,J.R.1987,ApJS,63,311
agooddynamicrangespecificallyatz <1,owingtothe1.6µm Hwang,N.,Lee,M.G.,Lee,H.M.,etal.2007,ApJS,172,583
bumpofthestellaremissionandthePAHemissionat8µm.At Imai,K.,Matsuhara,H.,Oyabu,S.,etal.2007,AJ,133,2418
z < 1, N2 − N3 almost continuously increases with increas- Jarrett,T.H.,Cohen,M.,Masci,F.,etal.2011,ApJ,735,112
Jeon,Y.,Im,M.,Ibrahimov,M.,etal.2010,ApJS,190,166
ing redshift, because of the 1.6µm bump. On the other hand,
Ko,J.,Im,M.,Lee,H.M.,etal.2009,ApJ,695,L198
S7−S11hasamaximumatz ∼ 0.5,sincetheredshiftedPAH LeFloc’h,E.,Papovich,C.,Dole,H.,etal.2005,ApJ,632,169
8µm feature is captured by S11 bands at that redshift. These Lee,H.M.,Kim,S.J.,Im,M.,etal.2009,PASJ,61,375
trends explain the arch-shaped distribution of galaxies in this Lonsdale,C.J.,Hacking,P.B.,Conrow, T.P.,&Rowan-Robinson, M.1990,
ApJ,358,60
colour-colourplot.Fromthiscolour-colourplotalone,itissafe
Magnelli,B.,Elbaz,D.,Chary,R.R.,etal.2009,A&A,496,57
to conclude that most of AKARI mid-IR sources in the NEP-
Magnelli,B.,Elbaz,D.,Chary,R.R.,etal.2011,ArXive-prints
Deepfieldlieatz<1.ThereddestS7−S11galaxiesmayhave Matsuhara,H.,Wada,T.,Matsuura,S.,etal.2006,PASJ,58,673
thestrongestPAH emissionfeaturesatz ∼ 0.5,andarestudied Murakami,H.,Baba,H.,Barthel,P.,etal.2007,PASJ,59,369
indetailbyTakagietal.(2010)as‘PAH-selected’galaxies. Oliver,S.J.,Goldschmidt,P.,Franceschini,A.,etal.1997,MNRAS,289,471
Ouchi,M.,Shimasaku,K.,Okamura,S.,etal.2004,ApJ,611,660
Oyabu,S.,Yun,M.S.,Murayama,T.,etal.2005,AJ,130,2019
Papovich,C.,Dole,H.,Egami,E.,etal.2004,ApJS,154,70
6. Summary
Pearson,C.P.,Oyabu,S.,Wada,T.,etal.2010,A&A,514,A8+
Pe´rez-Gonza´lez,P.G.,Rieke,G.H.,Egami,E.,etal.2005,ApJ,630,82
We have generated a mid-IR source catalogue based on im-
Polletta,M.,Tajer,M.,Maraschi,L.,etal.2007,ApJ,663,81
agesfromtheAKARINEP-DeepsurveypresentedbyWadaetal. Rodighiero,G.,Vaccari,M.,Franceschini,A.,etal.2010,A&A,515,A8+
(2008). This catalogue was designed to include most of the Rowan-Robinson,M.,Mann,R.G.,Oliver,S.J.,etal.1997,MNRAS,289,490
sources detected at 7, 9, 11, 15, or 18µm. In the catalogue we Sanders,D.B.&Mirabel,I.F.1996,ARA&A,34,749
report7284sourcesinthenearlycircularareaoftheNEP-Deep, Saunders,W.,Rowan-Robinson, M.,Lawrence,A.,etal.1990,MNRAS,242,
318
covering0.67deg2.Fromthesimulationofartificialsources,we
Serjeant,S.,Carramin˜ana,A.,Gonza´les-Solares,E.,etal.2004,MNRAS,355,
haveestimatedthephotometricerrorsineachbandandthecom- 813
pleteness in both MIR-S and MIR-L channels. The star-galaxy Takagi,T.,Ohyama,Y.,Goto,T.,etal.2010,A&A,514,A5+
separationisbasedsolelyonAKARIphotometry.Forsourcesin Tanabe´,T.,Sakon,I.,Cohen,M.,etal.2008,PASJ,60,375
Teplitz,H.I.,Chary,R.,Elbaz,D.,etal.2011,AJ,141,1
the Subaru/S-cam field, where the optical photometry is deep
Wada,T.,Matsuhara,H.,Oyabu,S.,etal.2008,PASJ,60,517
enoughtodetectmostofAKARImid-IRsources,weperformed
Wada,T.,Oyabu,S.,Ita,Y.,etal.2007,PASJ,59,515
opticalidentificationwiththelikelihoodratiomethod.Themid- White,G.J.,Pearson,C.,Braun,R.,etal.2010b,VizieROnlineDataCatalog,
IRnumbercountswederivedshowadrasticincreaseofsources 3517,79054
between 11µm and 15µm, owing to the effect of redshifted White,G.J.,Pearson,C.,Braun,R.,etal.2010a,A&A,517,A54+
Xu,C.2000,ApJ,541,134
PAH emission.Atthefluxlevelof1mJy,wefoundthehighest
Yagi,M.,Kashikawa,N.,Sekiguchi,M.,etal.2002,AJ,123,66
sourcedensityperfluxbininthe24µmband.Basedonredshift-
sensitivecolours,N2−N3andS7−S11,mostofAKARImid-IR
sourcesarefoundtolieatz<∼1.
Acknowledgements
We wouldliketo thankalltheAKARI teammembersfortheir
extensive efforts. This work is supported by the Japan Society
10