The PDS 66 Circumstellar Disk as seen in Polarized Light with the Gemini Planet Imager

3.7 MB·

by  Schuyler G. Wolff

Draft version January 28, 2016
PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
THE PDS 66 CIRCUMSTELLAR DISK AS SEEN IN POLARIZED LIGHT WITH THE GEMINI PLANET
IMAGER
Schuyler G. Wolff1, Marshall Perrin2, Maxwell A. Millar-Blanchaer3,4, Eric L. Nielsen5,6, Jason Wang7,
Andrew Cardwell8,9, Jeffrey Chilcote4, Ruobing Dong7, Zachary H. Draper10,11, Gaspard Ducheˆne7,12,
Michael P. Fitzgerald13, Stephen J. Goodsell14, Carol A. Grady15, James R. Graham7, Alexandra Z.
Greenbaum1, Markus Hartung9, Pascale Hibon9, Dean C. Hines2, Li-Wei Hung13, Paul Kalas7, Bruce
Macintosh5, Franck Marchis6, Christian Marois11, Laurent Pueyo2, Fredrik T. Rantakyro¨9, Glenn
Schneider16, Anand Sivaramakrishnan2,17, and Sloane J. Wiktorowicz18,19
1 DepartmentofPhysicsandAstronomy,JohnsHopkinsUniversity,Baltimore,MD21218,USA
2 SpaceTelescopeScienceInstitute,Baltimore,MD21218,USA
6 3 DepartmentofAstronomy&Astrophysics,UniversityofToronto,Toronto,ON,M5S3H4,Canada
1 4 DunlapInstituteforAstronomyandAstrophysics,UniversityofToronto,Toronto,ON,M5S3H4,Canada
0 5 KavliInstituteforParticleAstrophysicsandCosmology,StanfordUniversity,Stanford,CA94305,USA
2 6 SETIInstitute,CarlSaganCenter,189BernardoAvenue,MountainView,CA94043,USA
7 AstronomyDepartment,UniversityofCalifornia,Berkeley;Berkeley,CA94720,USA
n 8 LBTObservatory,UniversityofArizona,933N.CherryAve,Room552Tucson,AZ85721,U.S.A.
a 9 GeminiObservatory,Casilla603,LaSerena,Chile
J 10 UniversityofVictoria,3800FinnertyRd,Victoria,BC,V8P5C2,Canada
11 NationalResearchCouncilofCanadaHerzberg,5071WestSaanichRd,Victoria,BC,V9E2E7,Canada
7 12 Univ. GrenobleAlpes/CNRS,IPAG,F-38000Grenoble,France
2 13 DepartmentofPhysics&Astronomy,430PortolaPlaza,UniversityofCalifornia,LosAngeles,CA90095,USA
14 GeminiObservatory,670N.A’ohokuPlace,Hilo,HI96720,USA
] 15EurekaScientific,2452Delmer,Suite100,Oakland,CA96002,USA
P 16 StewardObservatoryandtheDepartmentofAstronomy,UniversityofArizona,933NorthCherryAvenue,Tucson,AZ85721,USA
E 17 DepartmentofAstrophysics,AmericanMuseumofNaturalHistory,NewYork,NY10024,USA
18 DepartmentofAstronomy&Astrophysics,UniversityofCalifornia,SantaCruz,CA95064,USAand
h. 19 RemoteSensingDepartment,TheAerospaceCorporation,ElSegundo,CA90245,USA
p Draft version January 28, 2016
-
o ABSTRACT
r We present H and K band imaging polarimetry for the PDS 66 circumstellar disk obtained during
t
s thecommissioningoftheGeminiPlanetImager(GPI).Polarizationimagesrevealacleardetectionof
a the disk in to the 0.12(cid:48)(cid:48) inner working angle (IWA) in H band, almost 3 times as close to the star as
[
thepreviousHST observationswithNICMOSandSTIS(0.35(cid:48)(cid:48) effectiveIWA).Thecentro-symmetric

1 polarization vectors confirm that the bright inner disk detection is due to circumstellar scattered
v light. A more diffuse disk extends to a bright outer ring centered at 80 AU. We discuss several
8 physical mechanisms capable of producing the observed ring + gap structure. GPI data confirm
4 enhanced scattering on the East side of the disk which is inferred to be nearer to us. We also detect
2 a lateral asymmetry in the South possibly due to shadowing from material within the inner working
7 angle. This likely corresponds to a temporally variable azimuthal asymmetry observed in HST/STIS
0 coronagraphic imaging.
.
1 Subject headings: instrumentation: adaptiveoptics,protoplanetarydisks,stars: individual(PDS66),
0 techniques: high angular resolution, techniques: polarimetric
6
1
: 1. INTRODUCTION et al. 2015).
v
PDS 66 (MP Muscae) is one of the closest T Tauri
i Classical T Tauri stars (cTTS) with optically-thick,
X gas-richprotoplanetarydisksprovidevaluableknowledge stars. It was identified as part of the Pico Dos Dias Sur-
vey(Gregorio-Hetemetal.1992). Mamajeketal.(2002)
r of the precedent conditions for planet formation. By
a comparing the observed intensity in scattered light to classified PDS 66 as a member of the Lower Centaurus
Crux (LCC) subgroup with a mean age of 17±1 Myrs.
radiative transfer models, we can infer the grain proper-
Mamajek et al. (2002) list PDS 66 as a K1 spectral type
ties (size, density, composition) and the geometry at the
surface of the disk (e.g. Graham et al. 2007; Murakawa star with a kinematic parallax distance of 86+−87 pc and
2010). ageestimatesrangingfrom7–17Myrs. PDS66wasthe
TheGeminiPlanetImager(GPI)wasdesignedtoover- only cTTS found in their sample of over 100 pre-main-
come the contrast problem inherent in the detection of sequence stars. Torres et al. (2008) first suggested that
circumstellar material within ∼ 1.5(cid:48)(cid:48) from their host PDS 66 is more likely a member of the (cid:15) Cha Associa-
stars. GPI combines an advanced adaptive optics (AO) tion. Murphy, Lawson, & Bessell (2013) reinvestigated
system, an apodized coronagraph, and an IR integral the membership of PDS 66 and found that the proper
field spectrograph with both spectral and polarimetric motion is more consistent with (cid:15) Cha (age: 5 – 7 Myrs,
modes (Macintosh et al. 2014; Larkin et al. 2014; Perrin kinematic distance: 101 ± 5 pc). The membership of
PDS 66 remains somewhat uncertain between LCC and
(cid:15) Cha. Given that the disk properties of PDS 66 appear
swolff[email protected]
2 Wolff et al. 2015
Table 1
GeminiPlanetImagerobservationsofPDS66
Exp. #of Coronagraph Field
Date Mode Band Time(s) Exposures SpotSize(”) Rotation(◦) Notes
2014May14 Spectral H 59.6 10 0.246 3.5 0.5–0.8(cid:48)(cid:48) seeing,highwinds
2014May15 Polarization H 59.6 32 0.246 13.3 0.7–1.0(cid:48)(cid:48) seeing
2014May15 Polarization K1 59.6 16 0.306 11.4 0.7–1.0(cid:48)(cid:48) seeing
to be inconsistent with the LCC, and the younger age of in high wind conditions with the AO system operating
(cid:15) Cha is below the typical disk dissipation timescale of at 500Hz. For the spectral mode data, the raw frames
10 Myr (Haisch et al. 2001), in this paper, we adopt the weredarksubtracted, correctedforbadpixels, destriped
age and distance appropriate for eps Cha. to correct for variations across read-out channels, and
ThePDS66diskisinaninterestingevolutionarystage. Fourier filtered to remove microphonics noise. A wave-
The spectral energy distribution (SED) lacks signs of length calibration using arc lamp data taken before the
large-scale evolution or an inner clearing. Based on the sequenceisusedtoconverttherawimagesto3Dspectral
1.2mmcontinuumfluxCarpenteretal.(2005)estimated datacubes (Wolff et al. 2014). The location of the star
a total dust mass of 5.0×10−5M . Schu¨tz et al. (2005) behind the coronagraphic mask was measured from the
(cid:12)
modeled the PDS 66 mid-infrared spectra and SED and satellite spots (Sivaramakrishnan & Oppenheimer 2006;
infer an inner disk radius consistent with the dust subli- Wangetal.2014)andthedatawerecorrectedforspatial
mation radius of 0.1 AU. However, Cortes et al. (2009) distortion.
comparethePDS66SEDtothemedianTaurusSEDand In polarimetry mode, frames are taken in sets of four
find a flux decrement between 4-20 microns, indicating differenthalf-waveplaterotationsandcombinedtoform
a partial clearing of material in the disk. CO measure- Stokes cubes with slices I, Q, U, and V. Data are dark
mentsbyKastneretal.(2010)showamoleculargasdisk subtracted, destriped, and a thermal sky background is
extending out to 120 AU with a lower limit for the gas subtracted(inK1 bandonly). Theindividualframesare
massof9.0×10−6M . Althoughuncertain,thissuggests converted into two orthogonal polarization states using
(cid:12)
alowergas-to-dustratiolimitof≥0.2. Eventhoughthe a spot location calibration file that has been corrected
accretion rate inferred for PDS 66 is small for a cTTS for elevation-induced flexure. Each cube is divided by
(estimates range from 5 × 10−9 to 1.3 × 10−10M /yr a low pass filtered flat field to correct for low frequency
(cid:12)
Pascuccietal.2007;Inglebyetal.2013), theimpliedac- variations(Millar-Blanchaeretal.2015). Themeanstel-
cretion timescale is short, < 105 yrs. (based on the disk lar polarization and instrumental polarization are sub-
mass inferred from CO). tracted and the polarization pairs are cleaned via a dou-
The PDS 66 circumstellar disk was first resolved in bledifferencealgorithm(Perrinetal.2015). Thesatellite
HST/NICMOS imaging by Cortes et al. (2009). They spots are again used to determine the location of the oc-
detected a disk with an outer radius of 170 AU, and an culted star and to calibrate the flux of the disk using a
inclinationof32◦±5◦. Theauthorsalsoprovideevidence conversionfactorof1ADUcoadd−1 s−1 =7.4±2.6mJy
for grain growth through an analysis of the spectral en- arcsec−2inH bandand31±10mJyarcsec−2inK1 band
ergy distribution. Likewise, Bouwman et al. (2008) ob- (Hung et al. 2015).
tained Spitzer spectroscopy (8−13µm) and found that Figure 1 shows the H and K1 band polarimetry for
the dust grain properties are well fit by a model con- PDS 66 with the Stokes vectors giving the orientation.
sisting of amorphous olivine and pyroxene with average HeretheStokesparametershavebeentransformedtora-
particlesizesofafewmicrons. Schneideretal.(2014)ob- dial Stokes parameters (Schmid et al. 2006). The +Qr
taineddeepHST/STIScoronagraphyshowingconsistent imagecontainsthepolarizationorientedinthetangential
geometry, albeit with detection of faint halo extending direction in the disk, −Qr contains the radial polariza-
out to beyond 520 AU. tionandUr containsthepolarizationoriented±45◦ from
PDS 66 was observed during the commissioning of Qr. For an optically thin disk, the Ur image should con-
GPI to test instrument performance on a typical bright, tain no polarized flux from the disk and can be treated
nearby disk. Our GPI observations are described in Sec- as a noise map. For an optically thick disk like PDS
tion 2. The morphology of the PDS 66 disk as seen in 66,multiplescatteringeventscanresultinnon-negligible
polarized light with GPI is discussed in Section 3. We brightness,atafew%oftheQr signalforlow-inclination
place limits on our sensitivity to planetary companions disks (Canovas et al. 2015). Given this small amplitude,
in Section 4. Section 5 discusses the results. we adopt the Ur channel as a measure of our errors, rec-
ognizingthatthecontributionofbothnoiseandpotential
signal renders it a conservative estimate.
2. OBSERVATIONS
ThespectralmodedatawerePSF-subtractedusingthe
Coronagraphic imaging polarimetry and spectroscopy pyKLIP software (Wang et al. 2015)1. pyKLIP com-
of PDS 66 were obtained in 2014 May (Table 1). The
bines both Spectral Differential Imaging (SDI: for spec-
GPI Integral Field Spectograph (IFS) has a plate scale
tralmodedata)andAngularDifferentialImaging(ADI)
of 0.014 arcseconds/pixel, a FOV of 2.8 X 2.8 arcsec2, using the Karhunen-Loeve Image Projection (KLIP) al-
and an angular resolution of ∼0.05(cid:48)(cid:48) in H band (Macin- gorithm (Soummer et al. 2012). Due to the face-on na-
tosh et al. 2014; Larkin et al. 2014). Data were reduced
using the GPI Data Reduction Pipeline; see Perrin et
al. (2014) and references therein. Data were obtained 1 https://bitbucket.org/pyKLIP/pyklip
PDS 66 Polarimetry with GPI 3
1.0 GPI H Band Qr 1.0 GPI H Band
25 AU 101
s
Offset [Arcsec] 000...055 000...055 100 urface Brightnes2[mJy/arcsec]
S
1.0 N 1.0 10-1
E P = 1 count sec−1 spaxel−1
1.0 0.5 0.0 0.5 1.0 1.0 0.5 0.0 0.5 1.0
Offset [Arcsec] Offset [Arcsec]
1.0 GPI K1 Band Qr 1.0 GPI K1 Band
25 AU
101
s
Offset [Arcsec] 000...055 000...055 100 urface Brightnes2[mJy/arcsec]
S
N
1.0 1.0
E P = 1 count sec−1 spaxel−1 10-1
1.0 0.5 0.0 0.5 1.0 1.0 0.5 0.0 0.5 1.0
Offset [Arcsec] Offset [Arcsec]
103
HST/STIS HST/STIS
1.0 25 AU 4 50 AU
Offset [Arcsec] 000...055 022 110012 rface Brightness2[mJy/arcsec]
u
100 S
N
1.0 4
E 10-1
1.0 0.5 0.0 0.5 1.0 4 2 0 2 4
Offset [Arcsec] Offset [Arcsec]
Figure 1. PolarimetrydataforPDS66inH band(top)andK1 band(middle),andwhitelightopticalSTISdataattwospatialscales
forcomparison(bottom;datafromSchneideretal. 2014). PolarizedintensityisshownontheleftfortheGPIdata,whiletherightpanels
showthesamepolarizedintensityover-plottedwithpolarizationvectors. Thevectororientationgivesthepositionangleforthepolarized
electricfield. Greyinnerregionsrepresentthecoronagraphicspotsize.
4 Wolff et al. 2015
tureofthedisk,recoveryofthetotalintensityisdifficult Figure 3 shows the azimuthal brightness variations for
via ADI. We leave forward modeling of the disk’s total two disk annuli (35 – 50 AU and 70 – 90 AU) computed
intensitysurfacebrightness,andcalculationofthepolar- fromthemeanandstandarddeviationin12◦ wedges. In
ization fraction, to future work. the35to50AUregion,thereisa∼35%decreaseinthe
surfacebrightnessfromPA160◦−220◦(measuredEfrom
3. DISKMORPHOLOGY N). Schneider et al. (2014) also saw brightness asymme-
The GPI data reveal a bright disk interior to a more tries of ∼ 30% between two epochs of data spaced three
diffusediskextendingtoanouterring,andanazimuthal months apart. Though at a different parallactic angle,
asymmetry indicative of interesting structure close in to the drop in brightness subtends approximately the same
the central star (Figure 1). We also show the STIS angular fraction of the disk.
dataprovidedbySchneideretal.(2014)toillustratethe
fainter outer halo outside the field of view of GPI. The 4. PLANETARYCOMPANIONLIMITS
inner disk likely extends from the sublimation radius to Our spectral data constrain planetary companions of
the change in the power law slope at 45 AU. The region a given mass and age. We compute a 5σ contrast
between the inner disk and outer ring (45 – 80 AU) is curve assuming a methane dominated planetary spec-
not entirely cleared, as evidenced by the azimuthal ori- trum. We achieve a contrast of ∼ 10−5 outside of 0.3(cid:48)(cid:48)
entation of the polarization vectors. and ∼2×10−6 outside of 0.4(cid:48)(cid:48). We detect no planetary
We fit an ellipse to the brightness contours in the candidates, but we recover a bright source in the north
outerdiskringusingtheconstrained,linear,leastsquares at 50σ, which was previously confirmed as a background
method described in Fitzgibbon et al. (1996). We find a source (Schneider et al. 2014; Cortes et al. 2009).
position angle for the disk major axis of 10◦ ±3◦ E of Planet sensitivities are calculated following Nielsen &
N, an axial ratio of 0.86±0.02, and a disk inclination of Close (2010) and Nielsen et al. (2008) (Figure 4). The
31◦±2◦ from a face on viewing geometry. These values contrastcurveisusedtosetcompanionbrightnesslimits
agree well with the STIS results (minor:major axial ra- withradius. Thebrightnessofaplanetwithagivenmass
tio 0.889±0.026, inclination 27.3±3.3 degrees: Schnei- and age are set by the hot start evolutionary tracks of
der et al. 2014). We measure no stellocentric offset to Baraffeetal.(2003). Foranageof7Myrandadistance
within 30mas, consistent within errors with the offset in of100pc((cid:15)Chamembership)thereisa90%confidence
the STIS observations of 33±10 mas (Schneider et al. thatwewouldhavedetecteda8M planetat∼20AU
Jup
2014). Low SNR in the satellite spots of these observa- or a 3 M planet outside of 40 AU. At 17 Myrs and
Jup
tionslimitsourknowledgeoftheobscuredstar’slocation 86 pc (LCC membership), the 90 % confidence limits
to within ∼2 pixels. increase to a 10 M planet at ∼ 20 AU. Planetary
Jup
We deproject the disk and calculate a radial bright- companions may exist, but lack a methane absorption
ness profile (Figure 2) separately for the East and West feature, or could be low-mass enough to remain hidden
sides of the disk. Note that the peak in surface bright- below the opaque disk surface.
ness is slightly offset from the edge of the coronagraphic
mask. This is likely due to a lower throughput from 5. DISCUSSION
an instrumental effect rather than a decrease in the sur- PDS66joinstheclassofpre-transitionaldisks(Espail-
face density of the disk (See also Rapson et al. 2015). lat et al. 2010) with an optically thick inner disk sepa-
The East side of the disk is brighter in both total in- rated from an outer disk by a dip in surface brightness
tensity (STIS/NICMOS) and polarized intensity (GPI). around 0.5(cid:48)(cid:48) that could indicate a partial clearing of the
Since we expect the dust particles in the disk to be pre- disk. The gap/ring structure observed in our GPI data,
dominatelyforwardscattering,weconcludethattheEast combined with the detection of orbiting CO (Kastner
side is the nearer side. We fit power laws ∝ r−γ to the et al. 2010) confirm that PDS 66 closely resembles the
surface brightness profile in the inner disk, the central V4046 Sgr and TW Hya systems. All are nearby cTTS
region, and the outer ring (see Figure 2). The power that have retained their molecular gas to late ages and
lawslopeintheinnerdiskisconsistentwithanoptically showmulti-ringedstructures. GPIpolarimetrywasused
thick, gas-rich disk. For the outer component (80 – 105 to confirm the presence of scattering dust in the gaps
AU), the GPI power law fit agrees well with the STIS of the V4046 Sgr multi-ringed structure (Rapson et al.
and NICMOS result (Cortes et al. 2009). 2015a). TWHyaismulti-ringedwithpartiallyfilledgaps
After correcting for extinction (AV = 0.7 ± 0.2 as well (Debes et al. 2013; ?).
mag; Cortes et al. (2009)) and stellar color (assum- If the disk is optically thick, the ring/gap structure is
ing a K1 spectral type with intrinsic H-K = 0.14), the a result of a variation in the disk surface that could be
azimuthally-averaged apparent color of the disk is H-K caused by a change in the surface density, the local scale
= 0.45±0.17 in polarized intensity, implying that the height, or the dust properties of the sub-micron sized
dust in the disk is ∼ 50% more effective at reflecting K1 grains in the disk. Here we discuss possible sources for a
band light than H band light. In H band, the East side change in the disk surface properties:
ofthediskis2.1timesbrighterthantheWestside,while
the East side is only 1.6 times brighter than the West in (a) GapOpeningPlanets: Aplanet/(s)inthelowsurface
K1 band. The E/W flux ratio is much lower than seen brightnessregioncouldinduceagapinthedustdisk
in total intensity in the visible (Schneider et al. 2014), and deplete the gas (Dong et al. 2015). Dust filtra-
which suggests either more isotropic scattering and/or a tionisefficientatpilinguplargerdustparticles(mm-
high polarization fraction on the (fainter) W side. sized) into a ring at the pressure bump outside of a
A region within the south side of the disk appears de- gasgap(Zhangetal.2015), whilesmallergrains(re-
pleted in polarized intensity in both H and K1 bands. sponsible for scattered light) could still populate the
PDS 66 Polarimetry with GPI 5
Deprojected Separation (AU) Deprojected Separation (AU)
100 80 60 40 20 20 40 60 80 100
y H Band γ=1.9 0.2
nsit K1 Band γ=−0.±4±0.3
nte2c)101 γ=4.3±0.3
d Ise 17 - 45 AU 45 - 80 AU 80 - 105 AU
ec
arizy/ar
olmJ
P
an (100
e
M East West
1.0 0.8 0.6 0.4 0.2 0.2 0.4 0.6 0.8 1.0
Deprojected Separation (arcsec) Deprojected Separation (arcsec)
Figure 2. RadialbrightnessprofileofthetangentialpolarizedintensityforH andK1 bandsfortheEast(left)andWest(right)sidesof
thedisk. VerticaldashedlinesindicatetheouteredgeofthecoronagraphicspotinH andK1 bands. Fitstothepowerlawslopes(γ)are
giveninthelegend(E/Wslopesagree). ErrorbarsaredrawnfromtheUr errormaps. Theprofileshowsthebrightinnerringofmaterial
andapeakat∼0.8(cid:48)(cid:48) (80AU)correspondingtotheouterring.
Figure 3. The azimuthal variation of the median polarized intensity as measured for the H band GPI data in an annulus in the gap
from35to50AUandforthebrightringfrom70to90AU.TheNorthandSouthpolarizedintensitiesareplottedseparatelytoemphasize
thedropinfluxseenontheSouthsideofthedisk. Polarizedintensityvalueshavebeennormalizedtothemeanseparatelyforeachring.
TheinsetshowstheH bandimage(Northup). Thedashedblue/redlinesrepresenttheannulususedtomeasuretheazimuthalvariation.
Theblackdottedlinegivesthelocationofthediskminoraxis. Thegrayshadedregioncorrespondstotheshadedinsetwedge.
6 Wolff et al. 2015
gap. Given the observed width (∼ 35 AU) and the timescales at the relevant orbital separations, Schneider
shallow depth (ring:gap = 1.4), this is most likely a etal.(2014)hypothesizedthatthe changescouldbedue
planetarysystemwithseveralsub-jupitermassplan- to either time-variable shadowing from material in the
ets. inner disk hidden behind the coronagraphic mask, or lo-
calizedaccretionhotspotsonthestellarphotosphere. It
(b) Disk Shadowing: A scale height enhancement in the ispossiblethattheazimuthalasymmetryseenintheGPI
inner part of the disk shadows the outer disk, until data(at∼40AU)isduetosuchaneffect,rotatedaround
the flaring of the disk eventually brings the disk sur- to affect the illumination over a different range of posi-
face above the penumbra. Dong (2015) find that a tionangles. Densityenhancementsinthediskcausedby
puffed up inner wall can create a three part broken accreting protoplanets might cast shadows on the outer
power law in the radial brightness profile, as seen in regionsofthedisk,thoughtheshadowedareaspredicted
the GPI data. A shadow cast out to 80 AU would by simulations for planets as massive as 50 M are
Earth
require a flat disk and/or a low flaring exponent. only ∼ 7AU2 (Jang-Condell 2009). Alternatively, cold
spotsonthestellarsurfacewhicharedarkerduetomag-
(c) Dust Particle Properties: A localized change in the netic suppression of convection typically cover 5 - 30%
dustpropertieswouldchangetheopacityofthedisk. of the stellar surface and could cause an azimuthal mod-
Dust settling due to grain growth could induce a ulation of the stellar illumination incident on the outer
change in the scale height, which would change the disk on stellar rotation timescales (Venuti et al. 2015).
height of the scattering surface relative to the disk If the azimuthally variable disk surface brightness dis-
midplane,producingthebrightring. GapsintheHL tribution is due to nonuniform brightness on the stellar
Tau disk have been ascribed to the effects of snow- surface, it will change on timescales of the rotation pe-
lines (Zhang et al. 2015). However, given the large riod (5 days). If instead it is due to material orbiting at
radius of the observed ring (80 AU), this seems un- the estimated inner radius (10.5 days at 0.1 AU) or em-
likely. bedded in the bright inner ring the shadowing will vary
over a longer period. More data is needed to elucidate
For an inclined disk that is optically thick vertically the timescales of the azimuthally variable disk surface
and axisymmetric, a ring with a higher surface height brightness.
would appear as an offset structure relative to the cen- Acknowledgements: We thank the referee, Joel
tralstar(Lagageetal.2006). Combiningtheoffsetmea- Kastner, for his advice that helped strengthen this
sured in the STIS image with the ring radius, we infer paper. We acknowledge financial support of Gem-
that the scattering in the ring occurs 4 AU above the ini Observatory, the NSF Center for Adaptive Op-
disk midplane. The expected scale height for gas in ver- tics at UC Santa Cruz, the NSF (AST-0909188; AST-
ticalhydrostaticequilibriumatthelocationoftheringis 1211562;AST-1413718),NASAOrigins(NNX11AD21G;
about 4–5AU (assuming T = 5000K, L = 1.1L and NNX10AH31G), the University of California Office of
eff (cid:63)
T =10−15K),i.e. similartotheheightwherescat- thePresident(LFRP-118057),andtheDunlapInstitute,
80AU
tering occurs. In optically thick disks, the disk surface UniversityofToronto. ThisworkissupportedbytheNa-
is typically located 2–4 times higher than the gas scale tionalScienceFoundationGraduateResearchFellowship
height(e.g.,D’Alessioetal.1999). Thissuggeststhatthe under Grant No. DGE-1232825. Portions of this work
PDS 66 disk is flatter and/or less flared than primordial were performed under the auspices of the U.S. Depart-
disks, i.e., possibly significantly settled as was originally ment of Energy by Lawrence Livermore National Labo-
suggestedbyCortesetal.(2009). Aflatteneddiskcould ratory under Contract DEAC52-07NA27344 and under
favor the ”shadowing” scenario above, but only a more contract with the California Institute of Technology/Jet
complete SED+image modeling effort can confirm this. Propulsion Laboratory funded by NASA through the
From this dataset, no clear conclusions can be drawn Sagan Fellowship Program executed by the NASA Ex-
on the origin of the gap + ring structure. ALMA dust oplanet Science Institute. This study is based in part on
continuumobservationswouldhelpdistringuishbetween observations made with the NASA/ESA Hubble Space
the scenarios above. For scenario (a), we would expect Telescope (program GO 12228), obtained at the Space
to see a significant pile up of mm-sized grains right out- TelescopeScienceInstitute,whichisoperatedbytheAs-
sidetheNIRring,duetothedustfiltrationeffect,which sociationofUniversitiesforResearchinAstronomy,Inc.,
wouldgenerateatleastafactorof∼10orhigherincon- under NASA contract NAS 526555. Based on observa-
tinuum flux. In scenario (b), the shadowed region would tions obtained at the Gemini Observatory.
have a slightly lower temperature, which would result
in less flux in the optically-thin mm continuum as well,
REFERENCES
though only on the order of (cid:46)50%. ALMA gas observa-
tions may be able to detect gas depletion in the scenario
(a), however given the shallowness of the gap this may Baraffe,I.,Chabrier,G.,Barman,T.S.,Allard,F.,&Hauschildt,
P.H.2003,A&A,402,701
notprovidesufficientcontrastbetweenshadowedandun-
Bouwman,J.,Henning,T.,Hillenbrand,L.A.,etal.2008,ApJ,
shadowed regions.
683,479
We detect an azimuthal departure from axisymmetry, Canovas,H.,M´enard,F.,deBoer,J.,etal.2015,A&A,582,L7
seen as a dimmer region in the disk’s southern side at Carpenter,J.M.,Bouwman,J.,Silverstone,M.D.,etal.2008,
around a 40 AU radius. Schneider et al. (2014) observed ApJS,179,423
Carpenter,J.M.,Wolf,S.,Schreyer,K.,Launhardt,R.,&
that the east/west asymmetry of the disk is variable on
Henning,T.2005,AJ,129,1049
timescales as short as three months (Schneider et al.
Cortes,S.R.,Meyer,M.R.,Carpenter,J.M.,etal.2009,ApJ,
2014). Since this is much shorter than the dynamical 697,1305
PDS 66 Polarimetry with GPI 7
!
!
!
!
!
!
7 Myrs 17 Myrs
100 pc 86 pc
Figure 4. Companion sensitivity as a function of separation and mass for membership in (cid:15) Cha (Left) and the LCC (Right) with ages
anddistancesasshown.
D’Alessio,P.,Calvet,N.,Hartmann,L.,Lizano,S.,&Cant´o,J. Nielsen,E.L.,Close,L.M.,Biller,B.A.,Masciadri,E.,&
1999,ApJ,527,893 Lenzen,R.2008,ApJ,674,466
Debes,J.H.,Jang-Condell,H.,Weinberger,A.J.,Roberge,A.,& MurphyS.J.,LawsonW.A.,BessellM.S.,2013,MNRAS,435,
Schneider,G.2013,ApJ,771,45 1325
deJuanOvelar,M.,Min,M.,Dominik,C.,etal.2013,A&A,560, Pascucci,I.,Hollenbach,D.,Najita,J.,etal.2007,ApJ,663,383
A111 Pecaut,M.J.,Mamajek,E.E.,&Bubar,E.J.2012,ApJ,746,
Dong,R.,Zhu,Z.,&Whitney,B.2015,ApJ,809,93 154
Dong,R.2015,ApJ,810,6 Perrin,M.D.,Duchene,G.,Millar-Blanchaer,M.,etal.2015,
Espaillat,C.,D’Alessio,P.,Herna´ndez,J.,etal.2010,ApJ,717, ApJ,799,182
441 Perrin,M.D.,Maire,J.,Ingraham,P.,etal.2014,Proc.SPIE,
Fitzgibbon,A.W.,Pilu,M.,andFischerR.B.,1996,Proc.ofthe 9147,91473J
13thInternationConferenceonPatternRecognition,253 Rapson,V.A.,Kastner,J.H.,Andrews,S.M.,etal.2015,ApJ,
Gr¨afe,C.,&Wolf,S.2013,A&A,552,A88 803,L10
Graham,J.R.,Kalas,P.G.,&Matthews,B.C.2007,ApJ,654, Rapson,V.A.,Kastner,J.H.,Millar-Blanchaer,M.A.,&Dong,
595 R.2015,ApJ,815,L26
Gregorio-Hetem,J.,Lepine,J.R.D.,Quast,G.R.,Torres, Schmid,H.M.,Joos,F.,&Tschan,D.2006,A&A,452,657
C.A.O.,&deLaReza,R.1992,AJ,103,549 Schneider,G.,Grady,C.A.,Hines,D.C.,etal.2014,AJ,148,59
Haisch,K.E.,Jr.,Lada,E.A.,&Lada,C.J.2001,ApJ,553, Schu¨tz,O.,Meeus,G.,&Sterzik,M.F.2005,A&A,431,165
L153 Silverstone,M.D.,Meyer,M.R.,Mamajek,E.E.,etal.2006,
Hung,L.-W.,Duchˆene,G.,Arriaga,P.,etal.2015,ApJ,815,L14 ApJ,639,1138
Ingleby,L.,Calvet,N.,Herczeg,G.,etal.2013,ApJ,767,112 Sivaramakrishnan,A.,&Oppenheimer,B.R.2006,ApJ,647,620
Jang-Condell,H.2009,ApJ,700,820 Soummer,R.,Pueyo,L.,&Larkin,J.2012,ApJ,755,L28
Kastner,J.H.,Hily-Blant,P.,Sacco,G.G.,Forveille,T.,& Torres,C.A.O.,Quast,G.R.,Melo,C.H.F.,&Sterzik,M.F.
Zuckerman,B.2010,ApJ,723,L248 2008,HandbookofStarFormingRegions,VolumeII,5,757
Lagage,P.-O.,Doucet,C.,Pantin,E.,etal.2006,Science,314, Venuti,L.,Bouvier,J.,Irwin,J.,etal.2015,A&A,581,A66
621 Wang,J.J.,Rajan,A.,Graham,J.R.,etal.2014,Proc.SPIE,
Larkin,J.E.,Chilcote,J.K.,Aliado,T.,etal.2014,Proc.SPIE, 9147,914755
9147,91471K Wang,J.J.,Ruffio,J.-B.,DeRosa,R.J.,etal.2015,
Macintosh,B.,Graham,J.R.,Ingraham,P.,etal.2014, AstrophysicsSourceCodeLibrary,recordascl:1506.001
ProceedingsoftheNationalAcademyofScience,111,12661 Wolff,S.G.,Perrin,M.D.,Maire,J.,etal.2014,Proc.SPIE,
Mamajek,E.E.,Meyer,M.R.,&Liebert,J.2002,AJ,124,1670 9147,91477H
Millar-Blanchaer,M.A.,Graham,J.R.,Pueyo,L.,etal.2015, Zhang,K.,Blake,G.A.,&Bergin,E.A.2015,ApJ,806,L7
arXiv:1508.04787
Murakawa,K.2010,A&A,518,A63
Nielsen,E.L.,&Close,L.M.2010,ApJ,717,878