The shadow knows: using shadows to investigate the structure of the pretransitional disk of HD 100453
Zachary C. Long, Rachel B. Fernandes, Michael Sitko, Kevin Wagner, Takayuki Muto, Jun Hashimoto, Katherine Follette, Carol A. Grady, Misato Fukagawa, Yasuhiro Hasegawa, Jacques Kluska, Stefan Kraus, Satoshi Mayama, Michael W. McElwain, Daehyeon Oh, Motohide Tamura, Taichi Uyama, John P. Wisniewski, Yi Yang
DDraft version April 11, 2017
Preprint typeset using L A TEX style AASTeX6 v. 1.0
THE SHADOW KNOWS: USING SHADOWS TO INVESTIGATE THE STRUCTURE OF THEPRETRANSITIONAL DISK OF HD 100453
Zachary C. Long , Rachel B. Fernandes , Michael Sitko , Kevin Wagner , TakayukiMuto , Jun Hashimoto , Katherine Follette , Carol A. Grady , Misato Fukagawa ,Yasuhiro Hasegawa , Jacques Kluska , Stefan Kraus , Satoshi Mayama , MichaelW. McElwain , Daehyeon Oh , Motohide Tamura , Taichi Uyama , John P.Wisniewski , Yi Yang (Accepted 02/03/2017) Department of Physics, University of Cincinnati, Cincinnati, OH 45221, USA Center for Extrasolar Planetary Studies, Space Science Institute, 4750 Walnut St, Suite 205, Boulder, CO 80301 Department of Astronomy/Steward Observatory, The University of Arizona, 933 N. Cherry Avenue, Tucson, AZ 85721 Division of Liberal Arts, Kogakuin University, 1-24-2 Nishi-Shinjuku, Shinjuku-ku, Tokyo, 163-8677, Japan National Astronomical Observatory of Japan, 2-21-1, Osawa, Mitaka, Tokyo, 181-8588, Japan Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, Stanford, California 94305, USA NASA Sagan Fellow Eureka Scientific, 2452 Delmer St. Suite 100, Oakland CA 96402, USA Division of Particle and Astrophysical Science, Graduate School of Science, Nagoya University, Nagoya, Japan Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109, USA Institute of Astronomy and Astrophysics, Academia Sinica (ASIAA), Taipei 10641, Taiwan Division of Theoretical Astronomy, National Astronomical Observatory of Japan, Osawa, Mitaka, Tokyo 181-8588,Japan a r X i v : . [ a s t r o - ph . E P ] A p r University of Exeter Astrophysics Group, School of Physics, Stocker Road, Exeter, Devon EX4 4QL UK Department of Astronomical Science, The Graduate University for Advanced Studies (SOKENDAI), Shonan Village,Hayama, Kanagawa 240-0193 Japan Exoplanets and Stellar Astrophysics Laboratory, Code 667, NASA’s Goddard Space Flight Center, Greenbelt, MD20771, USA Department of Astronomy and RESCUE, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan Astrobiology Center of NINS, 2-21-1, Osawa, Mitaka, Tokyo, 181-8588, Japan Homer L. Dodge Department of Physics, University of Oklahoma, Norman, OK 73071, USA
ABSTRACTWe present GPI polarized intensity imagery of HD 100453 in Y-, J-, and K1 bandswhich reveals an inner gap (9 −
18 au), an outer disk (18 −
39 au) with two prominentspiral arms, and two azimuthally-localized dark features also present in SPHERE totalintensity images (Wagner et al. 2015). SED fitting further suggests the radial gapextends to 1 au. The narrow, wedge-like shape of the dark features appears similarto predictions of shadows cast by a inner disk which is misaligned with respect to theouter disk. Using the Monte Carlo radiative transfer code HOCHUNCK3D (Whitneyet al. 2013), we construct a model of the disk which allows us to determine its physicalproperties in more detail. From the angular separation of the features we measurethe difference in inclination between the disks (45 ◦ ), and their major axes, PA = 140 ◦ east of north for the outer disk and 100 ◦ for the inner disk. We find an outer diskinclination of 25 ± ◦ from face-on in broad agreement with the Wagner et al. (2015)measurement of 34 ◦ . SPHERE data in J- and H-bands indicate a reddish disk whichpoints to HD 100453 evolving into a young debris disk. INTRODUCTIONImages of SAO 206462 (Stolker et al. 2016) and HD 142527 (Marino et al. 2015) have revealedazimuthally-localized dark features in their outer disks. Both studies interpret the features as shadowscast by an optically thick, non-coplanar inner disk. Such an inner disk may be indicative of theexistence of planets or large dynamical changes in the disk’s history. Modeling of the disk structurescan provide a predictive tool for where low mass companions may be hiding. Sequential follow-upobservations can then detect these worlds, as was the case with the young gas giant β Pic b, whichdue to its inclined orbit is driving a similar gravitational warp in the inner disk that was seen beforethe planet (Apai et al. 2015).Another disk that exhibits azimuthally-localized dark features is associated with HD 100453A (A9V,luminosity L ∼ (cid:12) , mass M ∼ (cid:12) (Dominik et al. 2003), d = 103 ± . (cid:48)(cid:48) ± . (cid:48)(cid:48)
02 (Chen et al. 2006). Totalintensity (TI) imaging at Y-K2 bands, obtained by Wagner et al. (2015) with the extreme AO imagerSpectro-Polarimetric High-contrast Exoplanet REsearch (SPHERE), revealed two spiral arms, darkfeatures, and the outer extent of the large radial gap inferred from the IR spectral energy distribution(SED) (Khalafinejad et al. 2016; Maaskant et al. 2013).We present Gemini Planet Imager (GPI) polarized intensity (PI) imagery (Macintosh et al. 2014)of HD 100453. This dataset, the SPHERE data, and the infrared SED are modeled using the 3DMonte Carlo radiative transfer code HOCHUNK3D (Whitney et al. 2013). This code allows fortwo, independent and radially separated two-layer disks which need not be coplanar, and the useof multiple grain opacity models. Model images are compared to image data, while simultaneouslyfitting the SED, to find the cause of the dark features and investigate how structural changes withinthe disk affect their shape and locations. OBSERVATIONS AND DATA REDUCTION2.1.
GPI Observations and Data Reduction
HD 100453 was observed with GPI on 2015 April 10-11th UT using polarimetric differential imagingmode in J-, Y-, and K1-bands. GPI’s “direct” mode utilizes polarization to suppress stellar light,but not a coronagraphic mask, thereby sacrificing contrast in favor of a tighter inner working angle.We used the shortest exposure time (1.49 seconds) to minimize saturation and co-added 10 frames toincrease the signal-to-noise ratio. The half-wave plate angle was rotated from 0 to 67.5 ◦ with 22.5 ◦ steps to obtain linear polarization. This sequence was repeated 41, 35, and 36 times, resulting intotal exposure time of 41, 35, and 36 minutes for Y-, J-, and K1-bands, respectively. The averagedairmass values were 1.21 and 1.19 in the J - and K < r <
10 pixels, which shows nosignificant disk emission). Finally, a custom pipeline fix was implemented to allow image alignmentin unblocked mode. GPI image alignment typically relies on well-calibrated satellite spots injectedby the GPI apodizers which are not present in unblocked mode, so a Gaussian stellar centroid wasused instead. Though this alignment method is imperfect due to stellar saturation, it worked well inthis case. The resulting images were transformed to “radial” Stokes parameters (Schmid et al. 2006).Images used here represent the Q R component, which holds all linear polarized flux oriented eitherparallel (negative) or perpendicular (positive) to the line connecting that pixel to the central star.Positive flux in these images therefore represents singly-scattered photons from the circumstellar disk.HIP 56071 was observed as a flux standard, with the same procedures, except without closing theAO loop, thereby avoiding saturation. This allowed us to derive conversion factors between ADUsec − pixel − and mJy asec − . Since the K1-band magnitude in HIP 56071 was unavailable, wetranslated the 2MASS K s -band magnitude into this band by relating the stellar flux to the Vega fluxassuming box passbands in two bands (the GPI K1-band with 1.9-2.19 µ m and the 2MASS K s -bandwith 1.989-2.316 µ m) and blackbody radiation with a T eff of 9200 K for HIP 56071 (A1V) and 9700 Kfor Vega (A0V). The color correction of K s 2MASS − K1 GPI was − GPI magnitude is 7.860. We derived conversion factors of 1 ADU sec − pixel − are 0.846 and 0.853mJy asec − in the J and K1-bands, respectively. Note that Y-band flux was not converted to mJyasec − due to no literature value for the Y-band.We find the projected separation between the image center and the companion to be 1 . (cid:48)(cid:48)
05, inagreement with Chen et al. (2006). The region within ∼ ∼ . (cid:48)(cid:48)
06) from thecentral star is saturated. We conservatively estimate the features outside 6-7 pixels from the centerto be real while the area interior to this radius is washed out by speckle residuals. We detect theouter disk and spirals, as in Wagner et al. (2015). The total PI within 0 . (cid:48)(cid:48) < r < . (cid:48)(cid:48) Archival Total Intensity Imagery and Assembly of the IR SED
Wagner et al. (2015) also observed HD 100453 with VLT/SPHERE on 2015 April 10. The ob-servations were carried out in IRDIFS extended mode (Girard et al. 2016) using IRDIS to takedual-band TI images in K1- and K2-bands and simultaneously using the IFS to obtain low-resolutionspectra from Y- to H-bands. Data reduction is described in Wagner et al. (2015). Photometry usedin constructing HD 100453’s SED includes all sources mentioned in Khalafinejad et al. (2016) as well Very Large Telescope SPHERE User Manual as Herschel
PACS data at 70, 100, and 160 µ m (Pascual et al. 2016), and 2MASS at J-, H-, andKs-bands (Cutrie et al. 2003). RESULTS3.1.
SPHERE and GPI Imagery
A radial gap can clearly be seen from 9 ± ± ∼
19 au using d = 103 ± ± ± . (cid:48)(cid:48) ± . (cid:48)(cid:48) . (cid:48)(cid:48) < r < . (cid:48)(cid:48) fdiskftotal of 0 . ± .
001 for J-band and 0 . ± .
002 forH-band which may suggest that the outer disk is reddish and therefore comprised of large, compactgrains (Mulders et al. 2013). By contrast we calculate a fractional luminosity for the spiral arms of0 . ± .
001 for J-band and 0 . ± .
001 for H-band which, though slightly red, is more bluethan the rest of the outer disk indicating they may be comprised of smaller grains (Mulders et al.2013).Two distinct azimuthally-localized dark features are seen at the same position angle (PA) in bothTI and PI images, and are therefore not artifacts of PI imagery. They also have a similar outer diskto feature contrast in all available bands (Figure 1).3.2.
Determining the Likely Cause of the Dark Features
We discuss three possible physical scenarios for the origin of the dark features: a physical gap duethe dynamical clearing of a large body, grain growth and settling, or shadows cast from an inner diskcomponent.The SPHERE Y-band PSF for HD 100453A has a FWHM of 5 pixels ( 3.8 au). This suggeststhe dark features are resolved and have an azimuthal extent of 14 ± ± . (cid:12) . The M-type companion, HD 100453B, is visible in SPHERE, GPI, and Chandra(Collins et al. 2009) imagery. An M-type star with a mass of 0.4 M (cid:12) therefore would certainly bevisible in the location of the dark features if it existed. Thus, we reject the hypothesis of local clearingto explain the dark features.Grain growth and settling has been suggested as a source of dark regions at NIR wavelengths(Dullemond & Dominik 2004a,b; Birnstiel et al. 2012), and should also produce bright rings inthe sub-mm. This would require resolution of at least 0 . (cid:48)(cid:48)
03, which is reachable by interferometrictelescopes such as ALMA. The differential rotation of the disk would cause the dark features todeform over time however, which suggests that grain growth and settling is not the cause of thefeatures.An inner disk would cast two shadows which have a large radial extent and have similar outerdisk to shadow contrast over the wavelength range in which it is optically thick (Stolker et al. 2016),similar to what is seen in the GPI and SPHERE images. We will show that shadows cast by an innerdisk with a suitable inclination is fully capable of producing such dark features. In the followingsections we present a model that is capable of generating the dark features as well as reproducingthe observed SED. 3.3.
Development of Preliminary Model
Literature Values for Outer Disk Inclination
Dong et al. (2016) adopted i ∼ ◦ , where i is the outer disk inclination from face on with respectto the observer, to reproduce the spiral arm morphology in hydrostatic modeling of the HD 100453system. If we accept the Dong et al. (2016) assumption of a completely coplanar disk, we can alsoassume the equatorial plane of the star is coplanar because, according to Greaves et al. (2014), moststars rotate in the same plane as their disks. Using the 5 ◦ inclination for the outer disk and a v sin i of48 ± -1 for HD 100453A (Guimar˜aes et al. 2006) we find the equatorial velocity of HD 100453Awould be ∼
550 km s -1 , which is 150 km s -1 above the break-up velocity of the star (Slettebak et al.1966). Because HD 100453A is still intact, a 5 ◦ inclination for the star is non-physical which implies,by extension, that the inclination of the outer disk also cannot be 5 ◦ . Moreover in GPI images wecan clearly see the major and minor axes of the disk which would not be possible given a nearly faceon disk as in TW Hydrae (Andrews et al. 2016). If the spiral arms are close to face on as Dong etal. (2016) suggest, they are not coplanar with the outer disk.Wagner et al. (2015) measured an outer disk inclination of i ∼ ◦ from face on through fitting anellipse to the peak intensity along the center of outer disk and assuming circular geometry. However,because the outer disk has finite thickness, this method will overestimate the outer disk’s inclination.We will review this inclination in Section 3.3.3.3.3.2. Initial Modeling Parameters
In order to reduce the degeneracies in our model, literature values, SPHERE and GPI imagery, andthe SED (Figure 2) are used in combination to determine initial modeling parameters. For the star weused a PHOENIX stellar atmosphere model (Brott & Hauschildt 2005) with T = 7400 K (appropriatefor an A9 star) and a distance of 103 pc (Gaia Collaboration et al. 2016a). HOCHUNK3D uses theLucy (1999) method for calculating temperature in our model. Box filters were created in Y-K2 toallow for direct comparison of the model to data.Through SED fitting Khalafinejad et al. (2016) suggested that the the inner disk extends from0.25 to 1.7 au. Confirmation of the inner edge radius is provided by H-band interferometry withVLTI/PIONIER (Lazareff et al. 2016) who find an upper limit of 0.27 au. The inner disk extends toat least 0 . ± . d = 103 ± ≤ µm . Forthe less-settled grain opacities we used Model 1 from Wood et al. (2002). Similarly this contains amixture of amorphous carbon and astronomical silicates, a power law size distribution with a = 3.5and 3.0, respectively, plus an exponential cutoff with a turnover at 50 µ m, a maximum particle size ≤ µm .3.3.3. Inclination of the Outer Disk ◦ ≤ i ≤ ◦ in 5 ◦ increments andcompared the resulting model images to observed images and find a best fit of 25 ± ◦ . From thismodel we also find a PA for the outer disk’s major axis of 140 ± ◦ east of north.3.4. Adopted Model Parameters
After initial modeling of the outer disk, we constructed a more complete disk model to fit the SED(Figure 2). We find that our model closely matches the observed SED including the NIR region, whichwas ignored by Benisty et al. (2017), to be discussed later. It consists of an inner disk ( ∼ . − ± . . ± . −
39 au).The inner disk has a vertical inner-edge thickness of 0.11 ± z = Cr b where zis the density scale height (thickness) of the disk, C is a constant, r is the radial distance from thestar, and b is the flaring exponent (Whitney et al. 2003a). To match mid-far IR emission we find b= 1.28 ± α = 2 .
30 and α = 2 .
25 respectively.3.5.
Difference in Inclination for Inner Disk
Definition of ∆ i In this section we test the validity of a difference in inclination between the inner and outer diskas the cause of the dark features. We define this difference in inclination as ∆ i = i − β where i is1the inclination of the outer disk from face on and β is the inclination of the inner disk from face on.While we define i as positive, β can be positive or negative depending on the inner disk’s directionof tilt (Figure 3).3.5.2. Azimuthal Separation of Dark Features and Comparison to Model
Model images were produced with the adopted i = 25 ◦ while ∆ i is varied in 10 ◦ increments from0 ◦ to 70 ◦ (Figure 4). These are convolved with the PSF of SPHERE J-band images and thereforeproduce clearer shadows than longer wavelength bands. No azimuthally-localized dark features wereobserved in model images with ∆ i = 0 ◦ despite a good fit to the SED, excluding a coplanar disksystem. At ∆ i = 20 ◦ the inner disk shadows the northern section of the image and predicts too muchflux between 1-10 µ m because of its nearly face-on orientation with respect to the observer (Figure 4).Moving to ∆ i = 70 ◦ the shadows narrow and predicts too little flux between 1-10 µ m because theinner disk is nearly edge-on to the observer. We find ∆ i = 45 ± ◦ matches the general appearanceof the imagery, to be quantified below.Azimuthal intensity profiles of the GPI images and model imagery were generated for the outerdisk (Figure 5) and measure an azimuthal separation of 140 ± ◦ for the dark features which arein the same locations in each band. The additional local minimum in intensity profile plots is dueto polarization effects along the minor axis of the outer disk. We find that as ∆ i is increased, theshadow separation also increases and produces a best fit to the dark features at ∆ i = 45 ± ◦ whichcorresponds to β = − ± ◦ . This deduced inner disk inclination angle is significantly different fromthe one found in Lazareff et al. (2016), β = − ◦ , using H-band interferometry with VLTI/PIONIER.The main source of uncertainty in our measurement comes from azimuthal extent of the shadowedregion. To match the dark features’ azimuthal location we rotate the inner disk by -40 ◦ with respectto the outer disk which places the major axis of the inner disk at 100 ± ◦ east of north. This differs2from the PA for the major axis found in Lazareff et al. (2016), 81 ◦ , and will be discussed later.A recent study by Benisty et al. (2017) also proposed a misaligned inner disk as the cause of theouter disk shadowing in HD 100453. In their paper they generate a model to test this hypothesisusing measurements from Lazareff et al. (2016). In order to test the validity of their model wegenerated our own using their value for the outer disk inclination, i = 38 ◦ , and their quoted ∆ i of72 ◦ . This model however does not produce shadows in the same locations as in the data (Figure 6),similar to what was found earlier in this section.In our modeling we find that the inclination of the outer disk had little effect on the separationof the shadows on it’s own unless the near side of the outer disk is reversed. We also find that theseparation of the shadows is less than 180 ◦ on the side of the outer disk which corresponds to thenear side of the inner disk. Essentially the inner disk major axis divides the outer disk into twosemicircles. The 180 ◦ constraint implies that the centers of the shadows cannot exist in separatesemicircles and this strongly constrains the orientation of the inner disk. In Figure 6 we see that thequoted orientation of the inner disk major axis places the shadows in separate semicircles in the GPIimage which cannot occur if the inner disk is the source of the shadows. In addition the gap has avisibly more elliptical structure in the Benisty et al. (2017) model than in the data which is likelydue to the thickness of the outer disk as discussed in Section 3.4 as well as the larger inclination ofthe outer disk. It is important to note that this ∆ i corresponds to β = − ◦ instead of the inner diskinclination proposed separately in Benisty et al. (2017) of β = − ◦ . Additionally in our modeling wefind that the NIR region of the SED is fairly sensitive to the inclination of the inner disk (Figure 4).Because Benisty et al. (2017) did not fit the NIR excess of the SED, the Benisty et al. (2017) modelis not as tightly constrained as our simultaneous image and SED fitting. We found the inclinationof the inner disk proposed Benisty et al. (2017) did not produce a good match to the 1-10 micronSED nor the constraints on the shadows that that we found (Figure 7). Differences in the inner disk3orientation which produce shadows along an axis other than the major axis may be possible, thoughwe did not observe this in our modeling.3.6. Effects of Inner Disk Thickness and Outer Disk Flaring on Shadows
To examine how inner disk thickness affects the morphology of the shadows, we generated modelimages at inner-edge inner disk thicknesses from 0.07 - 0.14 au in increments of 0.014 au withoutfitting the SED. We find that as the thickness increases, the width of the shadows increases with nochange in the location of the shadows. (Figure 8). The differences in width become small at smallerthicknesses however, suggesting that shadows can act as an upper constraint of inner disk thickness.Model images were also generated at flaring exponents (Section 3.4) of 1.00, 1.20, and 1.40 (Figure 9)in order to examine the effect of outer disk structure on shadow morphology, also without fitting theSED. We find that the width of the shadows’ inner edge decreases with increasing b, while the outeredge remains largely unaffected. In addition we observe that the flaring of the disk causes the outerdisk to appear thicker on the far side as seen in Figure 9 where the SW side is the near side of theouter disk. This effect is most prevalent in the rightmost panel where the SW side of the disk isapproximately half the thickness of the NE side. In the case of HD 100453, the SW side of the outerdisk is narrower in GPI images (Figure 1) indicating that this is the near side. The ratio of thicknessesbetween the near and far side of the disk, coupled with the inclination of the outer disk, could allowus to quantitatively describe the degree of flaring of the outer disk. When taken in conjunction withthe inner disk thickness, this should allow us to strongly constrain the disk’s physical structure.3.7.
Using Shadows to Determine Near Side of Inner and Outer Disk
Due to thickness of the outer disk, we find that any ∆ i (cid:54) = | ◦ | will offset the apparent locationof the shadows on the outer disk in the direction of the tilt of the inner disk (Figure 10). Theshadows are cast along the major axis of the inner disk and create a shadow, which is not coplanar4with the outer disk, on the inner edge of the outer disk. The shadow will therefore be shifted by anamount which depends both on the thickness of the outer disk and the value of ∆ i . This suggeststhe separation of shadows must be less than 180 ◦ on the side of the outer disk which corresponds tothe near side of the inner disk as seen in Figure 10. Examination of the shadow location thereforeallows for a simple, effective method of determining the near side of the inner disk. In the case ofHD 100453 it is clear that the northern side of the inner disk is the near side because the shadowseparation is less than 180 ◦ on that side. In contrast, it is much more difficult to discern the nearside of the outer disk via examination of the shadows as this depends on both the tilt of the outerdisk and ∆ i . In this case, because we know ∆ i = 45 ◦ , the inclination of the outer disk is | ◦ | fromSection 3.3.3, and from the NIR portion of the SED we are not looking along the edge of the innerdisk, the SW side of the disk must be the near side. DISCUSSION4.1.
Dropoff in Spiral Arm Intensity with Wavelength
We observed a dropoff in spiral arm intensity at longer wavelengths in SPHERE TI imagery, mostprevalent in K1 (Figure 1), which suggests they are made up of small, compact grains. Though thereis a low gas to dust ratio in the disk of HD 100453 (Kama et al. 2016) there is also no detection ofa small dust grain dominated tail by HST ACS (Collins et al. 2009). The lack of a tail, as seen inHD 141569 (Konishi et al. 2016) and young debris disks, suggests the gas to dust is not low enoughfor radiation pressure to dominate. If gravitational interactions with the M-type companion are thesource of the spiral arms, the small grains, which are more tightly coupled to the gas than the largegrains, would be pulled more easily with the gas and may lie in a slightly different plane than the restof the disk. This could explain why the arms are bluer than the rest of the outer disk and accordingto (Dong et al. 2016) appear to be almost face on. When coupled with the difference in optical depth5between the arms and the ring of the outer disk, this suggests that there is a gradient in the particlesize distribution of the disk. 4.2.
Rejection of a Coplanar Disk System
A coplanar inner and outer disk cannot reproduce the shadows seen in the SPHERE and GPI data.From the shadow separation and comparison of model images to data we find a misaligned innerdisk with ∆ i = 45 ± ◦ and a rotation of the inner disk major axis of -40 ± ◦ with respect to theouter disk or 100 ± ◦ East of North. This disagrees however with Lazareff et al. (2016) who find amajor axis PA for the inner edge of the inner disk of 81 ± ◦ East of North. This measurement wasfound using simplified ellipse and ring models of the inner disk based on VLTI/PIONIER H-banddata taken from 2012 December 19 to 2013 February 20.4.3.
Investigation into Shadow Location Over Time
At this time there has been no observed change in the shadow pattern over the ∼ . ◦ change in major axis position angle between the Lazareff et al. (2016)measurement and the Wagner et al. (2015) measurement which spans ∼ −
27 months. If this is achange in the major axis it could be due to precession of the inner disk, orbital motion of materialin the inner disk, or the discrepancy could be a warp between the shadowing structure and the innerregion of the inner disk.Though the cause of the misalignment between the inner and outer disk is unknown, it has beenshown in β Pic (Lagrange et al. 2010) that a planet can warp the inner disk. If a planet did cause themisalignment of the inner disk in HD 100453 as suggested above, it would likely cause precession ofthe inner disk, and by extension the shadows. However, this precession would occur on a timescaleon the order of 10 times the orbital timescale expected from Newtonian dynamics (Rawiraswattana6et al. 2016). This suggests we should not observe a change in the location of the shadows due toprecession for timescales less than a decade. A change of 19 ± ◦ in ∼ −
27 months ( ∼ ◦ yr − ∼
10 au. This is much larger thanthe proposed outer radius of the inner disk and just outside the region excluded by speckle residualsin GPI data. The large radius and lack of any disk detection in GPI data suggest this is also likelynot the cause of the discrepancy between major axis PAs.This leaves us with a warp between the inner region of the inner disk (traced by VLTI/PIONIER)and a shadowing region further out. Though this cannot be ruled out using existing GPI or SPHEREdata, N-band measurements using VLTI/MATISSE would allow for the angular resolution necessaryto detect any difference in the inner and outer portions of the inner disk.4.4.
Frequency of Transitional Disks with Misaligned Inner Disks
To date, three of the fourteen, ∼ ∼ ∼ G (Pffeifer and Dong 2004). A close binary system can cause a misaligned inner disk howeverCollins et al. (2009) did not observe a strong X-ray source at the location of HD 100453A whichwould be indicative of another M-type companion, and brighter stars would be visible in the SED.Alternatively a wide binary system, where the companion lies outside of circumstellar disk (as in thecase of HD 100453), can also cause misaligned disks. However, this causes large outer disk eccen-7tricity at higher misinclinations (Martin et al. 2014) which is not seen here. The possibility of anunseen planet however, makes HD 100453 a promising candidate for future planet searches (Gratiaand Fabrycky 2016). To date however, there have been no radial velocity measurements taken forthe star with the intent of planet detection. In addition this system resembles β Pic in having a twobelt architecture and a misaligned inner disk, indicating that similar structure in young debris disksis likely inherited from the transitional disk phase. CONCLUSIONSWe have carried out GPI PI imaging of HD 100453 in Y-, J-, and K1-bands, and have examinedthe circumstellar disk using this data, the SED, and archival SPHERE TI imagery. With the helpof the Monte Carlo Radiative tranfer code HOCHUNK3D, we have generated models which allowedus to probe the inner and outer disk morphology and make several testable predictions about thestructure of the disk. Our conclusions are: • The circumstellar disk of HD 100453 contains an inner disk which SED fitting suggests extendsfrom 0.13 - 1 au, followed by a large radial gap (1 - 18 au), and an outer disk (18 - 39 au). • The outer disk is red in TI imagery, with fdiskftotal = 0 . ± .
002 in J-band and fdiskftotal = 0 . ± . • Both TI and PI images exhibit azimuthally-localized dark features at similar PAs. The sizeof the features is sufficiently large that any single body clearing them would be detected as abright source (Janson et al. 2012). Given the sharpness and narrow size of the features and theirconsistent appearance in PI and TI light we also exclude inhomogeneities in grain properties.We therefore suggest that they are shadows cast by an optically thick, misaligned inner disk.8 • An outer disk inclination of 5 ◦ (Dong et al. 2016) from face on is not supported by the data.In contrast we measure an inclination of 25 ± ◦ , in broad agreement with the 34 ◦ inclinationreported in Wagner et al. (2015). • The degree of misalignment between the inner and outer disk can be determined using theseparation of the shadows. We find this separation is best reproduced with a ∆ i = 45 ± ◦ which also gives us the best fit to the NIR region of the SED. • Examination of the shadows can constrain the thickness of the inner disk. As the inner diskthickness increases the width of the shadows it causes also increases. • There is a difference in the color of the spiral arms and the rest of the outer disk. This suggeststhat they are likely made of small grains which couple to gas more easily and are separated fromthe outer disk through interaction with the M-type companion. • There is a discrepancy of ∼ ◦ between the major axis PA for the inner disk found by Lazareff etal. (2016) and the one found in our study. Because of the long timescale required for precessionand the lack of a visible blob at 10 au, we suggest that the cause of this discrepancy is a warpbetween the inner portion of the inner disk and an outer shadowing portion.We have shown that shadow morphology and location constrains both the thickness and orientationof the inner disk as well as the flaring of the outer disk. This offers an independent means ofmeasuring structure in transitional disks. In addition the possibility of differences between the innerand outer edges of the inner disk is intriguing and should be investigated further. Predictions of theshadowing model will be tested for this and other Herbig Ae/Be stars with VLTI/MATISSE andother interferometric instruments in the future.Based in part on data obtained at the Gemini Observatory via the time exchange program between9Gemini and the Subaru Telescope (GS-2015A-C-1). The Gemini Observatory is operated by theAssociation of Universities for Research in Astronomy, Inc., under a cooperative agreement withthe NSF on behalf of the Gemini partnership: the National Science Foundation (United States),the National Research Council (Canada), CONICYT (Chile), Ministerio de Ciencia, Tecnolog´ıa eInnovaci´on Productiva (Argentina), and Minist´erio da Ciˆencia, Tecnologia e Inova¸c˜ao (Brazil). MTis partly supported by JSPS KAKENHI 2680016. CAG is supported under NASA Origins of SolarSystems Funding via NNG16PX39P. Y.H. is supported by Jet Propulsion Laboratory, CaliforniaInstitute of Technology under a contract from NASA. MS is supported by NASA Exoplanet ResearchProgram NNX16AJ75G. J.K. acknowledges support from Philip Leverhulme Prize (PLP-2013-110,PI: Stefan Kraus). S.K. acknowledges support from an ERC Starting Grant (Grant Agreement No.639889). We also thank the referee for their comments and suggestions which added clarity to thispaper. REFERENCES Khalafinejad, S., Maaskant, K., M., Mari˜nas, N., &Tielens, A.G.G.M., 2016, A&A, 587, A62Konishi, M., Grady, C. A., Schneider, G., et al. 2016,ApJL, 818, 2Lazareff, B., Berger, J.-P., Kluska, J., et al. 2016,A&A, submittedLagrange, A. M., Bonnefoy, M., Chauvin, G., et al.2010, Science, 329, 5987Lucy, L. B. 1999, A&A, 344, 282Maaskant, K. M., Honda, M., & Waters, L. B. F. M.,et al. 2013, A&A, 555, A64Macintosh, B., Graham, J. R., Ingraham, P., et al.2014, PNAS, 111, 35Maire, J., Perrin, M. D., Doyon, R. et al., 2010,Proceedings of the SPIE, 7735, id. 773531Marino, S., Perez, S., & Casassus, S., 2015, ApJL, 198,2Martin, R., Nixon, C., Lubow, S., et al., 2014, ApJL,792, L33Meeus, G., Bouwman, J., & Dominik, C., et al. 2002A&A, 392, 1039-1046Meeus, G., Bouwman, J., & Dominik, C., et al. 2002A&A, 402, 767Menu, J., van Boekel, R., Henning, Th., et al. A&A,581, A107 Mouillet, D., Larwood, J. D., Papaloizou, J. C. B., &Lagrange, A. M. 1997, MNRAS, 292, 896Mulders, G. D., Min, M., & Dominic C., 2013, A&A,549, A112Pascual, N., Montesinos, B., Meeus, G., et al., 2015,A&A, 585, A6Perrin, M. D., Maire, J., Ingraham, P. et al., 2014,Proceedings of the SPIE, 9147, id. 91473JPffeifer, H. P. & Dong, L. 2004, ApJ, 604, 766Rawiraswattana, K., Hubber, & D. A., Goodwin, S. P.,MNRAS, 460, 4Schneider, G., Weinberger, A. J., Becklin, E. E., et al.,2009, ApJ, 137, 53-61Slettebak A., ApJ, 145, 126SStolker, T., Dominik, C., Avenhaus, H., et al., 2016,A&A, http://arxiv.org/abs/1603.00481Wagner, K., Apai, D., Kasper, M., & Robberto, M.,2015, ApJL, 813, L2Whitney, B.A., Wood,K., Bjorkman, J.E., & Wolff, M.,2003, ApJ, 591, 1049Whitney, B.A., Robitaille, T.P., Bjorkman, J.E., Dong,R., Wolff, M.J., Wood, K., & Honor, J., 2013Wood, K., Wolff, M.J., Bjorkman, J.E., & Whitney, B.,2002, ApJ, 564, 887 Figure 1 . The r -scaled GPI polarized intensity images in Y, J, and K1 bands, plus a schematic of thedisk marking the locations of the dark features (top row), and r -scaled SPHERE total intensity images ofHD 100453 in Y, J, H, and K1 bands (bottom row). The images have a field of view of 0 . (cid:48)(cid:48) × . (cid:48)(cid:48) Figure 2 . Best model fit (solid line) of the SED of HD 100453 after initial modeling with the stellaratmosphere (dotted line) also visible. Major Axis of the Inner Disk
Major Axis of the Outer Disk i = 25° β Δi Observer i
Figure 3 . LEFT: A cross section of the disk structure with ∆ i = i − β where i is the positive inclinationof the outer disk from face on with respect to the observer while β is the positive or negative inclination ofthe inner disk from face on. RIGHT: A schematic view of the overall disk structure of HD 100453 in theobserved frame, as deduced from our best-fit model (Section 3). The major axes of the inner and outer disksare marked. Δ i = 70° Δ i = 60° Δ i = 50° Δ i = 40° Δ i = 30° ≈
30 AU Δ i = 20° Δ i = 0° Δ i = 10° Figure 4 . SEDs between 1 and 10 µ m coupled with r -scaled, total intensity model images at J-band,convolved with the PSF of SPHERE total intensity imagery, of various ∆ i ’s. Here the ∆ i = 0 does notproduce dark features and therefore HD 100453 cannot have two coplanar disks. We find a best fit of∆ i = 45 ± ◦ . Locations of Dark Features
Figure 5 . The colored lines represent azimuthal intensity profiles of GPI polarized intensity (top) and totalintensity, J-band model (bottom) images where 0 ◦ represents due North and we trace counterclockwisealong the outer disk. GPI traces are normalized by the maximum intensity in each band. Model images areconvolved with the PSF of SPHERE imagery and unscaled. The vertical black lines mark the position ofthe dark features. Figure 6 . LEFT: The r -scaled, total intensity Benisty et al. (2017) model recreation at J-band, convolvedwith the PSF of SPHERE total intensity imagery. CENTER: Observational GPI polarized intensity r -scaledJ-band image. RIGHT: r -scaled, total intensity model image at J-band of our best-fit model outlined inSection 3.5, convolved with the J-band PSF of SPHERE total intensity image. Here the yellow dottedline shows the major axis PA of the inner disk in the Benisty et al. (2017) model recreation and the reddotted line represents the major axis PA of the inner disk in our model. It is apparent from the figure thatthe Benisty et al. (2017) model recreation that neither the major axis location nor the shadow locationsthemselves match the locations of the shadows in the data. In addition simple visual examination of thismodel reveals that both the azimuthal separation of the shadows and the ellipticity are too large to matchthe observational data. Figure 7 . The colored lines represent azimuthal profiles of GPI polarized intensity, our best fit J-bandmodel, and our recreation of the Benisty et al. (2017) model. The vertical black lines represent the locationof the shadows in the GPI image. In our model recreation using ∆ i = 72 ◦ and i = 38 ◦ neither the locationof the shadows nor their separation agree with the GPI imagery. Figure 8 . Graph showing the effects of inner disk thickness on the width of the dark features. The coloredlines represent intensity traces derived from unscaled J-band total intensity model images convolved withthe PSF of SPHERE total intensity imagery. 0 ◦ represents due North and we trace counterclockwise alongthe outer disk. We find that as the thickness increases the width of the dips associated with the shadowsalso increases while their intensity decreases. ≈
30 AU
Figure 9 . Figures showing total intensity model generated images at J-band with flaring exponents (Section3.4) of 1.00, 1.20, and 1.40. Model images are convolved with the PSF of SPHERE imagery and unscaled.The SW side is the near side of the outer disk and it is clear that the near edge is thinner in the flared disksthan the far edge. Figure 10 . Figure showing both a face on schematic of the outer disk (left) and a vertical cross-section ofthe outer disk (right). The red dashed line represents the major axis of the inner disk and the solid linedenotes the vertical slice along which the cross section is produced. The inner disk is tilted with the top edgecloser to the observer in this schematic and the dark regions represent the shadows it casts as it intersectsthe light coming from the central star (represented by the yellow star). In the cross-section we can see howthe shadow is shifted vertically in the direction of the tilt of the inner disk. The face-on view of the outerdisk shows how the center of these shadows (yellow dashed line) can be offset from the view of the observer.Because of this offset, the separation of the shadows will always be less than 180 ◦◦