An Optical/near-infrared investigation of HD 100546 b with the Gemini Planet Imager and MagAO

TL;DR

This study uses optical and near-infrared observations to analyze the protoplanet HD 100546b, revealing that its detected emission is likely a disk feature rather than a planet or accreting object, with implications for understanding planet formation.

Contribution

It provides new spectroscopic and photometric data showing the emission is consistent with scattered light, challenging previous interpretations of HD 100546b as an accreting protoplanet.

Findings

H band emission is likely scattered light, not a planet.

Emission appears stationary over 4.6 years, inconsistent with orbit.

Upper limits on accretion rate suggest no active accretion at detection level.

Abstract

We present H band spectroscopic and Halpha photometric observations of HD 100546 obtained with GPI and MagAO. We detect H band emission at the location of the protoplanet HD 100546b, but show that choice of data processing parameters strongly affects the morphology of this source. It appears point-like in some aggressive reductions, but rejoins an extended disk structure in the majority of the others. Furthermore, we demonstrate that this emission appears stationary on a timescale of 4.6 yrs, inconsistent at the 2sigma level with a Keplerian clockwise orbit at 59 au in the disk plane. The H band spectrum of the emission is inconsistent with any type of low effective temperature object or accreting protoplanetary disk. It strongly suggests a scattered light origin, as it is consistent with the spectrum of the star and the spectra extracted at other locations in the disk. A non detection at the 5sigma level of HD 100546b in differential Halpha imaging places an upper limit, assuming the protoplanet lies in a gap free of extinction, on the accretion luminosity and accretion rate of 1.7E-4 Lsun and MMdot<6.4E-7Mjup^2/yr for 1Rjup. These limits are comparable to the accretion luminosity and rate of TTauri-stars or LkCa 15b. Taken together, these lines of evidence suggest that the H band source at the location of HD 100546b is not emitted by a planetary photosphere or an accreting circumplanetary disk but is a disk feature enhanced by the PSF subtraction process. This non-detection is consistent with the non-detection in the K band reported in an earlier study but does not exclude the possibility that HD 100546b is deeply embedded.

Figures

Tables

Table 1. Table 1 Observing log of HD 100546.

Instrument	UT-date	Camera	Mode	Filter	Exposure tine	Nexp	FoV rotation	$⟨$Airmass$⟩$	$⟨\varpi ⟩$
					(s)		(deg)		$″$
GPI	2016/02/27	IFS	Spec.	H	60	120	51.6	1.36	0.87
MagAO	2014/04/12	VisAO	SDI	H$\alpha$/Cont.	2.27	4939	72	–	0.58

Full text

I. Introduction

The evolution of circumstellar disks is driven by accretion/ejection (Lynden-Bell & Pringle, 1974; Armitage, 2011; Alexander et al., 2013), photoevaporation (Johnstone et al., 1993; Armitage, 2011), dust growth, and planetary formation (Crida & Morbidelli, 2007; Dullemond et al., 2009; Testi et al., 2014; Espaillat et al., 2014). These mechanisms act simultaneously, albeit with different efficiencies, over timescales of up to a few million years. They are responsible for dramatic changes in disk properties and observational signatures, and might ultimately create the centrally-cleared disk cavities that are the signature feature of the transitional disk subclass.

The transitional disk phase was first identified by near infrared flux deficits in the spectral energy distributions of young objects compared to those of class II objects (Strom et al., 1989), and was later interpreted as the formation of an optically thin gap or cavity at the disk center (Espaillat et al., 2011). Resolved images of transition disks in a broad range of wavelengths have since revealed rich spatial structures in both dust and gas tracers, including: gaps, cavities, spirals, and vortices (e.g., Fukagawa et al., 2006; Andrews et al., 2011; Muto et al., 2012; Debes et al., 2013; Casassus et al., 2013; van der Marel et al., 2013; Quanz et al., 2013b; Follette et al., 2015). Several recent observational results have refuted photoevaporation and viscous evolution as the main mechanisms responsible for transition disks (Pinilla et al., 2012; Garufi et al., 2013; van der Marel et al., 2013), as they cannot reproduce asymmetric structures (Owen & Clarke, 2012). Thus, planetary formation is emerging as a very likely mechanism for the creation of observed transition disk structures (Crida & Morbidelli, 2007; Dodson-Robinson & Salyk, 2011; Birnstiel et al., 2013; Espaillat et al., 2014; Zhu, 2015), in particular the dust segregation as seen in different wavelength regimes (e.g., Pinilla et al., 2015b).

Observationally, protoplanetary disks are important in that they allow us to directly observe the first stages of planet formation. Recent discoveries of a small number of low-mass companions inside transition disk gaps (Kraus & Ireland, 2012; Biller et al., 2012; Close et al., 2014; Sallum et al., 2015) are aiding in our understanding of planet formation.

The detection of point-like emissions near observed disk features such as inner rims (e.g., Huélamo et al., 2011) require cautious interpretation, as they could be image artifacts stemming from the circumstellar disk and not actual companions (Olofsson et al., 2013; Thalmann et al., 2016). The cospatial nature of planets and disks therefore makes the distinction between disk structures and real companions nontrivial. This is particularly true when searching for self-luminous companions embedded in an optically thick circumstellar disk, and is especially complicated when using adaptive optics (AO) imaging with the Angular Differential Imaging technique (ADI, Marois et al., 2006). During an ADI sequence, the field of view rotates with respect to the detector while the optics of the telescope and the instrument remain stable. Any rotating astrophysical source can then be separated from the fixed point spread function (PSF) and from quasi-static speckles. This observing mode, and the associated data reduction process, induces flux loss of the astrophysical signal and produces morphological hence measurement biases. For extended and inclined structures like disks, PSF subtraction techniques may also create asymmetries and blobs (Milli et al., 2012), which can mimic point sources and be interpreted as embedded companions.

HD 100546 (spectral type of B9V, mass of $2.4\pm 0.1$ $M_{\odot}$, distance of $109\pm 4$ pc, age of $5-10$ Myr, Levenhagen & Leister 2006; van den Ancker et al. 1997; Gaia Collaboration et al. 2016; Guimarães et al. 2006; van den Ancker et al. 1997) hosts an extensively studied transition disk that extends out to $300$ au and exhibits a number of peculiar and potentially planet-induced morphological features, including: a gap, spiral arms, and asymmetries. (e.g., Augereau et al., 2001; Liu et al., 2003; Grady et al., 2005; Ardila et al., 2007; Benisty et al., 2010; Quanz et al., 2011; Mulders et al., 2013; Boccaletti et al., 2013; Avenhaus et al., 2014; Panić et al., 2014; Pineda et al., 2014; Pinilla et al., 2015a; Garufi et al., 2016). A protoplanet, HD 100546 b, has been detected by AO direct imaging at L ’ ($3.8$ $\micron$) and M ’ ($4.8$ $\micron$) with VLT/NaCo, and is embedded in the Northen part of the disk at a physical separation111The original separation in the discovery paper was $53$ au but it has been updated and propagated throughout the paper based on the new Gaia parallax measurement (Gaia Collaboration et al., 2016). of $59\pm 2$ au (Quanz et al., 2013a, 2015). Its mass is not well constrained ($5-15$ $M_{\mathrm{Jup}}$ ) because the origin of the emission is unclear (planet photosphere or circumplanetary disk) and also because the age of the star and of the (perhaps much younger) protoplanet is not well determined. However, the emission has a black body effective temperature of $\approx 900$ K, and a luminosity of $\approx 2.6\times 10^{-4}$ L⊙ (Quanz et al., 2015). Analysis of its morphology showed that it is composed of a point source (the protoplanet and an unresolved circumplanetary disk) surrounded by a warm component of the surrounding circumstellar disk (Quanz et al., 2015). However, beyond the extent of the resolved emission, the rest of the field of view is clear of any disk emission, and HD 100546 b lies in an weakly polarized region of the disk (Quanz et al., 2013a; Avenhaus et al., 2014), favoring a protoplanet scenario compared to a disk hot spot. Recently, with shorter wavelength ($\lambda_{c}=1.64$ $\micron$), a detection of HD 100546 b was reported with the Gemini Planet Imager (GPI, Macintosh et al., 2014) along a very bright disk arc extending from the protoplanet location to the coronagraph edge (Currie et al., 2015).

To follow-up this peculiar system, we obtain considerably longer dataset with large field rotation as previously reported. HD100546 was observed with the GPI Instrument at Gemini South as part of the Gemini Planet Imager Survey (GPIES) and with the Visible AO (VisAO) camera at the Magellan Clay Telescope at Las Campanas Observatory. In this paper, we present images of a source detected at the location of HD 100546 b in H band but not in H$\alpha$. We describe four independent arguments assessing the nature of this source based on its morphology (Sec. III), astrometry (Sec. IV), H band spectrum (Sec. IV.2), and accretion luminosity (Sec. IV.3). Numerous disk structures are also resolved in these data but their description and interpretation are presented in a companion paper (Folette et al., submitted).

II. Observing strategy and image processing

II.1. Gemini Planet Imager

HD 100546 was observed on UT 2016-02-27 (program ID: GS-2015A-Q-501). These data are fundamentally new observations that gave us the capacity to present the first spectrum in H band. A total of $120\times 60$ second IFU images were taken in $H$ band ($1.49-1.79$ $\micron,\lambda/\Delta\lambda\approx 45$) in coronographic mode using ADI. A total of $51.6$ deg of field rotation (FoV) was obtained over the sequence. Observing conditions were stable and good with an average DIMM seeing of $0.8\arcsec$ and coherence time of $4$ ms. The observing parameters are reported in Table 1. Observations of an argon arc-lamp were acquired at the target elevation immediately prior to the sequence and were combined with reference arcs taken at zenith to calculate flexure compensation (Wolff et al., 2014). The $\theta_{1}$ Ori field was observed to calibrate the platescale ($14.166\pm 0.007$ mas.pixel*-1*) and the correction to the detector position angle of ADI observations between the North and the vertical axis ($0.1\pm 0.13$ deg, Konopacky et al. 2014).

Another set of observations of HD 100546 was obtained on UT 2014-12-17 (GS-2014B-Q-500) but with only $12.9$ deg of FoV rotation. Therefore, this dataset was not used in the present paper.

The raw 2-D images were initially processed with the GPI Data Reduction Pipeline v1.3.0 (DRP, Perrin et al., 2014) which performs dark subtraction, bad pixel identification and removal, instrument flexure correction, extraction of spectra to create 3-D $(x,y,\lambda)$ data cubes (Maire et al., 2014), interpolation over a common wavelength axis, and distortion correction as measured with a pinhole mask (Konopacky et al., 2014). The positions of the four satellite spots - attenuated replicas of the central occulted star - at each wavelength in each data cube were also measured with the DRP (Wang et al., 2014). The barycenter of each set of spots was then used to align all the images to the location of the star.

The reduced data cubes were then processed to subtract the stellar point spread function (PSF). Large-scale structures (e.g., temporal variation of the residual turbulence affecting the stellar halo and background) slowly varying from one frame to another are not fully subtracted by ADI algorithms and must be removed before the PSF subtraction. The data were therefore high-pass filtered. An apodized Fourier filter was used to attenuate, following a Hanning profile, spatial frequencies lower than a parametrized cutoff. Different sizes were tested from 4 pixels up to 16 pixels. To subtract the PSF, three ADI algorithms were applied and the results were compared: the standard cADI (Marois et al., 2006), LOCI (Lafrenière et al. 2007;$dr=5$ pixels, $N_{A}=500$ full width at half maximum, FWHM, $3.6$ pixels in $H$ band, $g=1$, and $N_{\delta}=0.75$ FWHM), and PCA (Soummer et al. 2012; Amara & Quanz 2012; 1-3 Karhunen-Loève (KL) modes kept, a single region from $3$ to $100$ pixels in radius). For each algorithm, the residual frames were rotated to align North with the vertical axis and combined with a trimmed mean (10%) in both temporal and spectral directions, resulting in a single image.

The data were also processed using the Reference Differential Imaging (RDI) technique, which is much less susceptible to self-subtraction of disk features, and was not available in the previous works since we benefit from the GPIES reference library. First, the data were processed with pyKLIP (Wang et al., 2015), a python implementation of the KLIP algorithm, to produce a RDI PSF subtracted broad-band image. Second, since the RDI pyKLIP pipeline does not handle spectral datacubes, the data were processed with TLOCI to produce 37 RDI PSF subtracted spectral images. For the first reduction, a library of reference images was created from all the 6265 individual spectroscopic $H$ band observations obtained with GPI as part of GPIES,at the time were the data were reduced, but rejected targets with known disks and/or companions to ensure star subtraction only (Draper et al. 2016, Millar-Blanchaer et al. in prep). Each observation was reduced with the GPI DRP as described previously and collapsed over the wavelengths to populate a homogeneous library of broadband images. HD 100546 data cubes were processed through pyKLIP, using the 1000 most correlated reference images in a 10-70 pixel annulus to avoid the mask edges and the satellite spots. This process produces a broad-band image subtracted from the star PSF that can be compared to the ADI reduced images. For the second reduction, the TLOCI quicklook pipeline was used. The library and the HD 100546 datacubes were spatially scaled to align speckles based on the satellite spot locations in each wavelength channel and flux normalized with the spots. Each reference library datacube was median-combined to create 426 achromatic reference images. For each frame of HD 100546, the 20 most correlated images of the library were median-combined and subtracted to remove the star PSF. No high-pass filter was applied to preserve large scale disk structures. The RDI-processed images of HD 100546 were scaled back to the original spatial resolution, and rotated to align the North with the vertical axis. We ended up with a 37 channel RDI-reduced datacube that was used to extract the spectrum of the disk.

II.2. MagAO

Magellan Adaptive Optics observations of HD100546 were taken on UT 2014-04-12 in better than median conditions (average seeing 0$\farcs$58). Total integration time for the image set was 187 minutes, and individual exposures were limited to 2.3 seconds to minimize saturation effects. The data were taken in Angular Differential Imaging mode, and a total of 72*∘* of rotation were achieved (see Table 1).

We utilized the Simultaneous Differential Imaging (SDI) mode of MagAO’s visible light camera VisAO to image the disk simultaneously in H$\alpha$ and in the neighboring continuum as part of the Giant Accreting Protoplanet Survey (GAPlanetS, Follette et al., in prep). Under this mode, the continuum channel serves as a sensitive and simultaneous probe of both the stellar PSF and any scattered light features.

The SDI data reduction and analysis procedures are described in detail in the companion paper to this study (Follette et al., accepted). Briefly, a custom IDL pipeline performed bias subtraction, flat field correction, splitting in two channels corresponding to H$\alpha$ and continuum and star registration. The PSF in each channel was estimated and removed using pyKLIP with 30 pixel wide annuli and a movement criterion of 12 pixels, a paramater silimar to $N_{\delta}$ in LOCI and defined by the number of pixels a source would move azimuthally and radially due to ADI and SDI. The non-coronagraphic images were saturated out at 8 pixels in radius, and so a 10 pixel radius mask was applied to all images. The region near the AO control radius, which is dominated by quasi-static speckles, was also masked. SDI images were created by scaling and subtracting each continuum frame from its H$\alpha$ counterpart, and these were then processed with pyKLIP. SDI processing serves to remove both PSF artifacts common to both channels and any scattered light structures present in the image; excess emission in the H$\alpha$ channel is thus preserved.

Contrast curves were also generated. The SDI residual image was convolved with an estimate of the PSF, the noise was measured in annuli of 1 FWHM in width and corrected for small sample statistics (Mawet et al., 2014). Algorithm throughput at the location of any accreting source was computed by injecting fake planets into the raw H$\alpha$ datasets and processing through SDI and pyKLIP. We measure only the contrast limit for an H$\alpha$ excess source like LkCa15 b (Sallum et al., 2015). Therefore, we are only looking for accreting objects considering that any forming planet surrounded by a massive circumstellar disk would be heavily fed by gas. We refer to the paper of Follette et al. (accepted) for additional details and to Section IV.3 for a discussion of several hypotheses.

III. Morphology at the position of HD 100546 b

PCA images, with two different high pass filter sizes, and RDI images are shown in Figure 1. The key feature is a smooth disk arc that is resolved to the North West of the star, which is also seen in the SPHERE IRDIS image as the “West wing” (Garufi et al., 2016). Moreover, a point-source lying along this structure is revealed at the expected location of HD 100546 b with aggressive reduction parameters. We present several arguments regarding the nature of this source in the most aggressive reductions.

No negative lobes flank the source although they are inherent to ADI processing of point sources. They are created by the point source being partly present at different parallactic angles in the reference PSF images built with ADI algorithms (Marois et al., 2006; Lafrenière et al., 2007; Soummer et al., 2012).

When using a larger high-pass filter as displayed in the middle panel of Figure 1, the source rejoins the arc and appears as its natural extension. This behavior is even more evident with RDI as shown in the right panel, the disk being extended to the East with the brightest part being consistent with the arc seen in ADI.

Sharp disk features are highlighted by both the ADI process, which acts as a frequency filter, and by further high pass filtering. The more aggressive the filter and ADI algorithm parameters are, the more prominent these sharp structures appear. Milli et al. (2012) demonstrated similar effects on observations of disks with inclination less than $50$ deg, like HD 100546: clumps appear and can lead to misinterpretation of the data. In our data, this is most easily seen in the spiral arm resolved on the East side of the disk that extends nearly up to the location of HD 100546 b, clearly seen in the left panel of Figure 1. The very end of this spiral is broken into blobs by aggressive data processing, and just like our source, these blobs are not robust across algorithms and algorithmic parameters.

Quantitatively, we measured the FWHM of the source as a function of the filter size from $16$ to $4$ pixels, i.e., from $4.5$ to higher than $1$ GPI $H$ band FWHM, for all three ADI algorithms. A Gaussian function was fit to the position, size, and flux of the source in each residual image in a wedge of $3\times 3$ GPI $H$ band FWHM centered on the position of the source. Errors were estimated by varying the size of the wedge. To make sure that the fitting process does not subtract both source and disk, the procedure adopted by Quanz et al. (2015) was followed. Briefly, the fitting is done such that the subtraction of the best fit function to the data does not over-subtract the underlying disk arc. Briefly, the determinant of the Hessian matrix computed at that location would be negative due to an oversubtraction, showing that the surrounding disk has been affected by the process. The result of the analysis is shown in Figure 2. The size of the source strongly depends on the filter size, becoming more compact with a smaller filter. Even for the smallest filter, the source is still larger than a point source FWHM.

While it is possible that a true object embedded in the extended disk might be revealed with this variety of aggressive processing, the morphology of the H band source at the location of HD100546 b, together with the impact of filter size and reduction algorithm, are most consistent with a disk artifact rather than a standalone point source. Note however that this argument does not formally rule out the presence of an astrophysical object, which is why we then pursue other lines of inquiry.

IV. Physical properties at the position of HD 100546 b

If the source detected within our GPI H band data is consistent with HD 100546 b seen in L ’ and M ’, it should have the characteristics of a protoplanet (or protoplanet plus circumplanetary disk) with effective temperature of $\approx 900$ K orbiting close to 60 au from its host star as derived by Quanz et al. (2015). In light of the previous H band detection (Currie et al., 2015), we investigate the source characteristics in detail and propose an alternative interpretation.

IV.1. Astrometry

Considering that the planet is still in formation, it is fair to assume that it orbits in the disk plane, with a circular orbit. Eccentricity growth, if any, is expected to occur after the gas dissipation, which has not yet occurred in this system. With the caveat that inward migration is possible (Crida & Morbidelli, 2007), the Keplerian motion of HD 100546 b at a deprojected distance of $59\pm 2$ au around the $2.4\pm 0.1$ $M_{\odot}$entral star is $1.4\pm 0.1$ deg/yr, thus $6.8\pm 0.4$ deg projected on the disk plane between the first detection in June 2011 and our observations in February 2016. Using the position angle of $8.9\pm 0.9$ deg from Quanz et al. (2013a), HD 100546 b should be $3.0\pm 1.0$ deg North East of the star in 2016 if rotating clockwise or $14.8\pm 1.0$ deg if rotating counterclockwise since its rotation direction is unknown.

The best fit astrometry of the source in our GPI data gives a projected separation of $471\pm 9$ mas and a position angle of $7.4\pm 1.7$ deg, where the uncertainties are combined in quadrature from the errors of the measurement, the star center, the plate scale, and the position angle offset. These measurements are consistent with those of Currie et al. (2015).

All of the reported projected separations remain unchanged to within $1\sigma$ over the four year time-span, which is consistent with both the stationary or a Keplerian orbit (only $\approx 14$ mas of total decrease). However, if the H band emission corresponds to a physical objet, then its position angle is $1.9\sigma$ away from the expected position assuming clockwise rotation of the system, and $3.7\sigma$ away from the counterclockwise prediction.

Therefore, the location of the source seen in our GPI data is consistent with an absence of motion within 1 $\sigma$. (see Figure 3).

Non-Keplerian orbital motion might be expected for HD 100546 b if the planet undergoes migration through disk-planet interaction (Crida & Morbidelli, 2007; Baruteau et al., 2016). Given the estimated mass of the planet (several Jupiter masses), classical inward type II migration would be expected on timescales of $10^{4}-10^{5}$ yrs. However, this scenario predicts a fast decrease of the separation of the protoplanet and a cleared gap (a necessary condition for fast type II migration) at levels that would likely already have been detected (see IV.3 for further details on the gap non-detection). The migration rate and direction can be significantly altered if the gap is only partially cleared due to a high viscosity of the disk (Baruteau et al., 2016). Dramatic slow-down of the migration rate or even slow outward migration cannot be rejected based on the astrometry collected so far.

IV.2. Emission

If the source detected in our GPI data is indeed the protoplanet HD 100546 b with an estimated effective temperature of $\simeq 900$K (Quanz et al., 2015), its H band spectrum should be different from that of the surrounding disk. It should exhibit either water absorption if it resembles planets like HR 8799 b (e.g., Barman et al., 2011), or a very red slope without molecular absorption if it is very dusty like the hotter 2M1207 b (Patience et al., 2010). Alternatively, if the protoplanet is accreting and the emission mostly originates from the circumplanetary disk, the H band spectrum should have either a red slope or a triangular shape, depending on the accretion rate and the inner disk radius (Zhu, 2015). The best fit contrast of the source in our GPI data is $\Delta H=13.2\pm 0.3$ mag, where the uncertainties are combined in quadrature with the errors from the measurement and the star-to-spot ratio (Maire et al., 2014). Assuming the source is embedded in the disk, the dust extinction at this location is $A_{H}=3.4$ mag (Currie et al., 2015), leading to $H=15.8\pm 0.3$ mag, and $H-L\,^{\prime}=3.1\pm 0.3$ mag. These values are consistent with the predictions of an accreting protoplanet with $M_{\mathrm{p}}\dot{M}=3\times 10^{-6}$ $M^{2}_{\mathrm{Jup}}.yr^{-1}$ and an inner disk radius of $1$ $R_{\mathrm{Jup}}$ (Currie et al., 2015). Therefore, the predictions for the source from Zhu (2015) suggest a red featureless H band spectrum.

Our GPI data can test the above scenario by comparing the spectrum of the source with that of the disk. The most aggressive $4$ pixels high-pass filtered PCA reduction was used for that purpose (Fig. 1, left) since it shows the resolved source and many bright disk structures. Like the astrometry, the contrast of the source was extracted in each wavelength channel at the best fit position and FWHM ($6.8$ pixels). For comparison, contrasts of the disk were also extracted, using aperture photometry with a diameter of $6.8$ pixels, from a location South East of the star at the same projected separation. The regions of these extractions are shown in Figure 4. Since ADI might bias the extracted disk contrasts, they were extracted at the candidate and opposition disk locations in the RDI reduced residual wavelength channels. All contrasts were then multiplied by the spectrum of the $10500$ K central star (van den Ancker et al., 1997), which was obtained by averaging BT-NextGen models (Allard et al., 2012) at $10400$ and $10600$ K and binning to the resolution of GPI.

To estimate the uncertainties of the measurements, each ADI wavelength channel image was convolved by the same aperture. The standard deviation of the pixels at the separation of the extractions was then taken in a region free of astrophysical signal, giving the final error. Typical noise is $14\pm 3\%$ of the source contrast across the band. The same exercise in the RDI reduced images provides typical $18\pm 2\%$ typical noise.

We do not flux calibrate the spectra but simply perform a comparison by eye since only the relative slopes and absorption features are of interest. Indeed, calibrating the self-subtraction of the disk due to ADI requires precise modelling (e.g., Esposito et al., 2014) that is beyond the scope of this paper. Nevertheless, since the reduction parameters are the same for all wavelength channels, ADI self-subtraction should be achromatic and only a function of the separation, which is why the measurements were done at the same angular distance.

The final spectra are plotted in Figure 4. The H spectrum of the source at the location of HD 100546 b exhibits no absorption feature and has a blue slope. The comparison with the star spectrum proves the detected emission is pure scattered light as seen when compared to the disk spectrum at the same separation. The spectrum is also compared to that of several young substellar objects with similar mass in Figure 5: GSC6214-0021 (M9$\gamma$, Lachapelle et al. 2015), $\beta$ Pictoris b (L2$\gamma$, Chilcote et al. 2017), HR8799 b (L/Tpec, Barman et al. 2011), and 2M1207 b (L/Tpec, Patience et al. 2010). These spectra are binned to the resolution of GPI in H band and interpolated over the same wavelength grid. This comparison further emphasizes the dissimilarities with the emission from a young planet.

This pure stellar-like spectrum is therefore inconsistent with a planetary photosphere, dust-obscured companion, or accreting circumplanetary disk source.

IV.3. Accretion

MagAO H$\alpha$ images in Figure 6 reveal structure in the inner $\sim$0$\farcs$2 of the disk, and nothing of note outside of this radius. The signal in the inner region is clearly scattered light, as scaling by the stellar H$\alpha$/continuum ratio and subtracting creates structure-free SDI images. When these SDI images are KLIP processed, no H$\alpha$ excess is apparent at the location of HD 100546 b.

Correcting the 5$\sigma$ ASDI contrast limit at this radius for KLIP throughput reveals that the maximum H$\alpha$ contrast at the location of HD 100546 b is $\sim$1$\times$10*-4*, under the assumption that no flux is present in the continuum channel. The contrast limit at this location can be used to place upper limits on the accretion properties of the protoplanet, given several assumptions as described below. The non-detection of an H$\alpha$ source at this location is indeed not necessarily inconsistent with the existence of an accreting protoplanet.

Accreting protoplanets, and giant planets in particular, are expected to clear circular regions in the disk, each several Hill radii in extent (about 7 au in our case) (Dodson-Robinson & Salyk, 2011) and extending azimuthally as they orbit. Therefore, the protoplanet should lie in a region free of extinction by the dust disk and the accretion luminosity would be directly detectable down to the sensitivity limit. Following Close et al. (2014) and Sallum et al. (2015), we use the contrast limit in H$\alpha$ to derive an upper limit on the accretion luminosity onto the protoplanet. Assuming the extinction towards the protoplanet is the same as the star, we derive $A_{R}=0.11$ mag from $A_{V}=0.15$ mag (Sartori et al., 2003) and the standard interstellar dust extinction law (Cox, 2000). Then, given the width and zero point of the H$\alpha$ filter, the distance and magnitude of HD 100546, $A_{R}$, and the contrast limit, we derive $\mathrm{L}_{\mathrm{H}\alpha}<1.4\times 10^{-5}~{}\mathrm{L}_{\odot}$ and, considering that accretion laws from T-Tauri stars also apply to planetary mass objects (Rigliaco et al., 2012),  $\mathrm{L}_{\mathrm{acc}}<1.7\times 10^{-4}~{}\mathrm{L}_{\odot}$. The accretion rate $\dot{\mathrm{M}}$ can be computed from the accretion luminosity with the mass ($\mathrm{M}$) and radius of the accreting object (Gullbring et al., 1998). We consider radii of 1 $\mathrm{R}_{\mathrm{Jup}}$ and 2 $\mathrm{R}_{\mathrm{Jup}}$ to bracket plausible values from evolutionary models (Baraffe et al., 2003) but treat the mass as a free parameter as it is very uncertain. Therefore, we place upper limits on $\mathrm{M}\dot{\mathrm{M}}<6.3\times 10^{-7}~{}\mathrm{M}_{\mathrm{Jup}}^{2}.\mathrm{yr}^{-1}$ and $<1.3\times 10^{-6}~{}\mathrm{M}_{\mathrm{Jup}}^{2}.\mathrm{yr}^{-1}$, for 1 and 2 $\mathrm{R}_{\mathrm{Jup}}$ respectively, under the assumption that the protoplanet is in a clear region. For comparison, with a possible mass from a few to 15 $\mathrm{M}_{\mathrm{Jup}}$ for HD 100546 b, the accretion flow on the protoplanet, if any, remains very small in comparison to that of on the central star ($\mathrm{M}\dot{\mathrm{M}}\sim 2\times 10^{-1}~{}\mathrm{M}_{\mathrm{Jup}}^{2}.\mathrm{yr}^{-1}$, Mendigutía et al. 2015). The upper limit on the accretion rate rules out the previously proposed scenario in which the H band emission is coming from an accreting protoplanetary disk with an inner radius of 1 $\mathrm{R}_{\mathrm{Jup}}$ and $\mathrm{M}\dot{\mathrm{M}}=3.6\times 10^{-6}~{}\mathrm{M}_{\mathrm{Jup}}^{2}.\mathrm{yr}^{-1}$ (Currie et al., 2015).

The upper limits on the accretion rate and luminosity are typical for T-Tauri stars (Gullbring et al., 1998; Rigliaco et al., 2012), or the young planet LkCa 15 b (Sallum et al., 2015). On the other hand, the protoplanet may be temporarily in a quiet phase and remains undetectable since accretion is known to be stochastic on TTauri stars (Bouvier et al., 2007).

Alternatively, it has been hypothesized that HD 100546 b is still embedded in the disk (Quanz et al., 2013a; Currie et al., 2015). The non-detection of any gap larger than few au around the protoplanet, despite a resolution of 2 au in the optical (Garufi et al., 2016), does not reject the hypothesis of a deeply embedded planet. With $A_{H}=3.4$ mag (Currie et al., 2015), the extinction due to the disk in H$\alpha$ is $A_{R}=22$ mag. Therefore, the accreting protoplanet would likely not be detectable in H$\alpha$. We can still do the same exercise and place some constraints on the accretion luminosity and rate. We find $\mathrm{L}_{\mathrm{acc}}<8\times 10^{3}~{}\mathrm{L}_{\odot}$ and $\mathrm{M}\dot{\mathrm{M}}<1.5\times 10~{}\mathrm{M}_{\mathrm{Jup}}^{2}.\mathrm{yr}^{-1}$ for a 1 $\mathrm{R}_{\mathrm{Jup}}$ planet. As expected, these upper limits place no meaningful constraint on the accretion properties of a possible deeply embedded protoplanet.

V. Conclusions

HD 100546 hosts a well-studied transition disk with a protoplanet detected in L ’ and M’ band with VLT/NaCo (Quanz et al., 2013a, 2015) and a recently proposed detection in H band with Gemini/GPI (Currie et al., 2015). With new GPI and Magellan/VisAO data, we present in this paper a source detected only in H band at the expected location of HD 100546 b. Detailed analyses reveal that:

• •

The source is located at the tip of a bright disk arc and non-aggressive high-pass filtering and reduction processes reveal that the source is contiguous with the underlying disk feature, similarly to blobs aligned along other spiral arms;

• •

The size of the source, point-like only in aggressive reductions, steadily decreases with that of the high-pass filter. This demonstrates that the source itself is extended;

• •

The astrometry of the source is consistent with stationary motion with respect to the discovery epoch and inconsistent at a $2\sigma$ level with a Keplerian circular orbit in the disk plane;

• •

The spectrum of the source is consistent with that of the disk and with pure scattered-light, proving the emission is not coming from a planet photosphere or an accreting circumplanetary disk;

• •

The non detection of a source at the $5\sigma$ level at this location in H$\alpha$ places upper limits, with the hypothesis the protoplanet is a cleared disk region, on the accretion luminosity of $1.7\times 10^{-4}\mathrm{L}_{\odot}$ and on the accretion rate of and $6.3\times 10^{-7}~{}\mathrm{M}_{\mathrm{Jup}}^{2}.\mathrm{yr}^{-1}$ for a radius of 1 $\mathrm{R}_{\mathrm{Jup}}$. This rules out the presence of the accreting circumplanetary disk previsouly proposed (Currie et al., 2015) but instead does not exclude the embedded hypothesis.

Therefore, the source detection in H band benefits from visual inspections of its morphology and spectrum, a 2$\sigma$ rejection of clockwise keplerian orbit, and a 5$\sigma$ non-detection of accretion light in H$\alpha$. Altogether these lines of evidence enable us to suggest that the source detected in H band with GPI is not related to HD 100546 b but is a disk feature, enhanced by the data reduction process, and the bulk of the detected flux comes from the disk. However, none of them refute the embedded hypothesis: planet-disk interaction (migration) might make the orbital motion non-keplerian and emissions in H band and H$\alpha$ from the protoplanet might be blocked by the surrounding circumstellar disk.

Considering that the detected source is pure disk emission in H-band that happens to be at the location of HD 100546 b, one wonders whether disk emission can contribute to the detection at longer wavelenghts. The source has $H=19.2\pm 0.3$ mag without dust extinction. From the H-L ’$=-1.08\pm 0.35$ mag color of the disk (Avenhaus et al., 2014), the source that we detect would have $\Delta$L ’$=15.4\pm 0.5$ mag assuming a grey albedo, compared to $\Delta$L ’$=9.4\pm 0.1$ mag of HD 100546 b reported by Quanz et al. (2013a). Therefore the disk alone cannot produce the the detected L$\,{}^{\prime}$ emission and a local source of heat is necessary to explain it. The non-detection of HD 100546 b in H band in our data is also consistent with the non-detection in K band in VLT/SPHERE-IRDIS data (Garufi et al., 2016). The most logical explanation for these multiwavelength data is that the optically thick circumstellar disk could be blocking thermal emission from the deeply embedded protoplanet, preventing any detection in the visible/near-infrared. The most problematic issue is the absence of polarization enhancement coming from the disk itself at the location of the protoplanet (Quanz et al., 2013a; Avenhaus et al., 2014; Garufi et al., 2016), which is hard to reconcile with the scattered-light emission seen inH band. To resolve this problem, the disk architecture and properties of the dust have to be better understood. ALMA observations at high resolution might reveal the presence of the circumplanetary disk or a peculiar morphology at the location of HD 100546 b. The size of such structures could be used to infer the mass of the protoplanet and help to understand disk clearing mechanisms and planet formation. JWST promises discerning thermal spectroscopy of the protoplanet that could reveal absorption features placing stronger constraints on its physical properties.

Based on observations obtained at the Gemini Observatory, which is operated by the Association of Universities for Research in Astronomy, Inc., under a cooperative agreement with the National Science Foundation (NSF) on behalf of the Gemini partnership: the NSF (United States), the National Research Council (Canada), CONICYT (Chile), the Australian Research Council (Australia), Ministério da Ciência, Tecnologia e Inovação (Brazil) and Ministerio de Ciencia, Tecnología e Innovación Productiva (Argentina). This work has made use of data from the European Space Agency (ESA) mission Gaia (http://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, http://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement. J.R., R.D. and D.L. acknowledge support from the Fonds de Recherche du Québec. KBF and JRM’s work was performed in part under contract with the California Institute of Technology (Caltech)/Jet Propulsion Laboratory (JPL) funded by NASA through the Sagan Fellowship Program executed by the NASA Exoplanet Science Institute. Supported by NSF grants AST-1411868 (K.B.F. and B.M.), AST-141378 (G.D.), and AST-1411868 (JP). Supported by NASA grants NNX14AJ80G (E.L.N., S.C.B., B.M., F.M., and M.P.), NNX15AD95G (BM, JEW, TME, RDR, GD, JRG, PGK) and NNX16AD44G (K.M.M, T.S.B., and L.M.C.). KWD is supported by an NRAO Student Observing Support Award SOSPA3-007. Portions of this work were performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. Facility: Gemini:South (GPI) and Magellan:Clay (MagAO)