0807.1766/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: 
3: \usepackage{emulateapj5}
4: 
5: 
6: 
7: \begin{document}
8: \title{The HD 163296 Circumstellar Disk in Scattered Light: Evidence of 
9: Time-Variable Self-Shadowing}
10: \author{John P. Wisniewski\altaffilmark{1,2}, Mark Clampin\altaffilmark{1}, 
11: Carol A. Grady\altaffilmark{1,3}, David R. Ardila\altaffilmark{4}, Holland C. 
12: Ford\altaffilmark{5}, David A. Golimowski\altaffilmark{5}, Garth D. 
13: Illingworth\altaffilmark{6}, \& John E. Krist\altaffilmark{7}}
14: 
15: \altaffiltext{1}{NASA Goddard Space Flight Center, Code 667, 
16: Greenbelt, MD 20771 John.P.Wisniewski@gmail.com, Mark.Clampin@nasa.gov, 
17: Carol.A.Grady@nasa.gov}
18: \altaffiltext{2}{NPP Fellow}  
19: \altaffiltext{3}{Eureka Scientific, 2452 Delmer Street Ste 100, Oakland, CA 94602}
20: \altaffiltext{4}{Spitzer Science Center, Pasadena, CA 91125, ardila@ipac.caltech.edu}
21: \altaffiltext{5}{Johns Hopkins University, 3400 North Charles Street, Baltimore, MD 20218, 
22: ford@pha.jhu.edu,dag@pha.jhu.edu}
23: \altaffiltext{6}{Lick Observatory, University of California, Santa Cruz, CA 95064, gdi@ucolick.org}
24: \altaffiltext{7}{Jet Propulsion Laboratory, 4800 Oak Grove Drive, MS183-900, Pasadena, CA 
25: 91109, john.e.krist@jpl.nasa.gov}
26: 
27: \begin{abstract} We present the first multi-color view of the scattered 
28: light disk of the Herbig Ae star HD 163296, based on coronagraphic observations from the
29: \textit{Hubble Space Telescope} Advanced Camera for Surveys (ACS).  
30: Radial profile fits of the surface brightness along the disk's semi-major axis indicates 
31: that the disk is not continuously flared, and extends to $\sim$540 AU.  The disk's 
32: color (V-I)=1.1 at a radial distance of 3$\farcs$5 is redder than the observed stellar color 
33: (V-I)=0.15.  This red 
34: disk color might be indicative of either an evolution in the grain size distribution (i.e. grain 
35: growth) and/or composition, both of which would be consistent with the observed 
36: non-flared geometry of the outer disk.  We also identify a single ansa 
37: morphological structure in our F435W ACS data, which is absent from 
38: earlier epoch F606W and F814W ACS data, but corresponds to one of the two ansa 
39: observed in archival HST STIS coronagraphic data.
40: 
41: Following transformation to similar band-passes, we find that the scattered light disk of HD 163296  
42: is 1 mag arcsec$^{-2}$ fainter at 3$\farcs$5 in the STIS data than in the ACS data.  Moreover, 
43: variations are seen in (i) the visibility of the ansa(e) structures, in (ii) the relative surface brightness 
44: of the ansa(e) structures, and in (iii) the (known) intrinsic polarization of the system.  These results 
45: indicate that the scattered light from the HD 163296 disk is variable.  We 
46: speculate that the inner disk wall, which Sitko et al. suggests has a variable scale height as diagnosed 
47: by near-IR SED variability, induces variable self-shadowing of the outer disk.  We further 
48: speculate that the observed surface brightness variability of the ansa(e) structures 
49: may indicate that the inner disk wall is azimuthally asymmetric. 
50:   
51: \end{abstract}
52: 
53: \keywords{circumstellar matter --- stars: individual (HD 163296) --- planetary systems: 
54: formation --- planetary systems: protoplanetary disks}
55: 
56: \section{Introduction} \label{intro}
57: 
58: Investigations of the nature and evolution of dust grains in proto-planetary systems are  
59: motivated in part by our desire to understand the birth and evolution of planetary bodies, 
60: which originate from these systems.  It is well established that Herbig 
61: Ae/Be stars \citep{her60} are 
62: intermediate mass pre-main-sequence stars, analogous to the more familiar low mass T Tauri 
63: stars, which contain copious amounts of circumstellar gas and dust
64: \citep{wat98}.  A wealth of observational evidence \citep{man97,oud99,man00,
65: vin02,eis04} 
66: suggests that the geometry of this gas and dust takes the form of a circumstellar disk.
67: By contrast, our understanding of fundamental properties such as grain composition and 
68: size distribution, as well as the time evolution of these parameters, are much less understood.
69: 
70: Analysis of ISO spectra led \citet{mee01} to develop a 2-part evolutionary classification 
71: of Herbig systems;  
72: Group I sources were characterized as being slightly younger systems with flared outer 
73: disks, while Group II sources were characterized as slightly more 
74: evolved systems which have flatter disks 
75: owing to grain growth and settling.  While \citet{mee01} invoked dust settling to explain 
76: the differences in flaring between Group I and II sources, it has also been suggested 
77: that the outer regions of some Herbig disks may experience self-shadowing, 
78: owing to a ``puffed-up'' inner circumstellar disk rim \citep{nat01,dul01,dul04,ise05}.  
79: Recent models which include inflated inner disk rims have proven to be successful in 
80: explaining the near-IR and interferometric properties of several Herbig Ae stars \citep{ise06}.  Conclusive confirmation of the geometry of the outer disk regions of 
81: Herbig Ae/Be stars inferred by SED-based studies has generally not occurred.
82: 
83: HD 163296 is a young (4 Myr; \citealt{van00}), nearby (122 pc; \citealt{van98}) 
84: Herbig Ae star.  While the star is not deeply embedded in a natal star formation 
85: dust cloud \citep{the85}, it still displays clear signs of active accretion via the presence 
86: of jets and Herbig Haro knots \citep{gra00,dev00,was06}.  Hence, HD 163296 appears 
87: to be in a transition phase between optically thick, extremely young pre-main-sequence 
88: stars and the much more optically thin, near zero-age-main-sequence debris disk type 
89: stars \citep{ard04,kal05,kal06,gol06}.  Indirect techniques \citep{bjo95} and resolved 
90: imaging \citep{man97,gra00,ise07} have confirmed the presence of a circumstellar disk associated 
91: with HD 163296.  HST STIS white-light coronagraphic observations spatially resolved the outer 
92: regions of this disk, and detected evidence of disk structure including an inner annulus of 
93: reduced scattering and a bright outer ring or ansae of material \citep{gra00}.  The 
94: inner region of HD 163296's 
95: circumstellar disk, as diagnosed by near-IR (0.8 - 5 $\mu$m) SED monitoring, exhibits 
96: evidence of variability, possibly owing to changes in the inner disk wall \citep{sit06}.
97: 
98: Attempts to link processes which dominate the inner regions of Herbig Ae disks to 
99: morphological features observed in the outer disk regions, i.e. determining the relative 
100: roles of self-shadowing, dust 
101: settling, and disk flaring \citep{dul04}, requires a wealth of observational data which diagnose
102: the inner and outer disk regions.  In this paper, we present a multi-epoch, multi-color 
103: view of the HD 163296 scattered light disk, providing a significant improvement in the 
104: documented behavior of the outer regions of HD 163296's circumstellar disk.  We describe the 
105: observational details and data reduction procedures applied to our ACS data in 
106: Section \ref{obs-red}.  In Section \ref{acs-charact}, we document the basic 
107: features of the HD 163296 scattered light disk in our ACS data, and compare these results 
108: to earlier epoch scattered light imaging in Section \ref{stis}.  The possible mechanisms 
109: which might explain the observed scattered light variability are discussed in 
110: Section \ref{origin}.  We provide a summary of our main results in Section \ref{summary}.
111: 
112: 
113: \section{Observations and Data Reduction} \label{obs-red}
114: 
115: We obtained coronagraphic imaging of HD 163296 in the HST ACS F435W filter via the
116: cycle 12 GTO program 9987, and in the ACS F606W and F814W filters via the cycle 11
117: GTO program 9295.  All observations utilized 
118: the coronagraphic mode of the High Resolution 
119: Channel (HRC) camera, along with its 1$\farcs$8 diameter occulting spot, yielding a 
120: 29$\farcs$2 x 26$\farcs$2 field of view at a pixel scale of 0$\farcs$028 x 
121: 0$\farcs$025.  The FWHM of point spread function (PSF) of the HRC is 50 mas in V.  
122: Our science target data were obtained at two different HST roll angles to  
123: better facilitate the separation of PSF artifacts from real structures.  The 
124: A3IV star HD 145570, selected for its spectral match to HD 163296, 
125: was used as a PSF reference, and was observed  
126: in orbits immediately following our HD 163296 exposures.  A summary of the 
127: basic parameters of our dataset is presented in Table \ref{datatable}.  
128: 
129: The standard HST pipeline was used to perform basic initial data reduction 
130: procedures.  We used the best available calibration files for this processing, 
131: including a correction for the different flat field illumination of the region nearby the 
132: coronagraphic spot.  CR-split images were combined using standard pipeline techniques.
133: 
134: Techniques to identify and remove PSF signatures from HST ACS coronagraphic data are 
135: well documented in the literature \citep{cla03,gol06}.  Following these established 
136: procedures, we first normalized and aligned the non-distortion corrected 
137: images of our PSF reference star to
138: HD 163296 in an iterative manner, using a cubic convolution interpolation 
139: function.  Residual alignment errors in the resultant best subtraction are on the order
140: of $\pm$0.05 pixels \citep{pav05} and our normalization is accurate to 
141: $\sim$2\% \citep{cla03}.  Following the subtraction 
142: of a fully registered and scaled PSF, we used MultiDrizzle to correct our 
143: data for the known geometric distortion of the HRC field of view and rotate all images 
144: into a common orientation.
145: 
146: Next, \textit{SYNPHOT}, the synthetic photometry package within STSDAS, was 
147: used to determine the correction factor needed to calibrate our data 
148: to an absolute photometric scale.  As a template for HD 163296, we used a  
149: synthetic spectrum of 69 Her (A2V), normalized to the V-band magnitude of HD 163296, 6.899; 
150: a synthetic spectrum of 59 Her (A3III) normalized to a V-band magnitude of 4.940 served as a template for 
151: HD 145570.  The ratio of the stellar flux between HD 163296 and HD 145570, extracted
152: both from our coronagraphic images and our synthetic representations of these data,  
153: were consistent to $<$ 2\% over multiple filter combinations, indicating our choice 
154: of synthetic spectra were reasonable.   All PSF-subtracted, distortion corrected images 
155: were normalized to the synthetic flux of HD 163296, and the different roll angle images 
156: were combined to increase the effective signal to noise ratios (SNR).  Note that all 
157: surface brightnesses reported in the native ACS filter bandpasses are based on the STMAG 
158: photometric system, and have been corrected for the known achromatic 52.5\% reduction of flux 
159: induced by the coronagraph.
160: 
161: \section{Disk Characteristics Derived from HST ACS Observations} \label{acs-charact}
162: 
163: Figures \ref{combB}, \ref{combV}, and \ref{combI} present the PSF-subtracted, roll angle 
164: combined, calibrated images of HD 163296 in the F435W, F606W, 
165: and F814W filters respectively, using a square-root intensity stretch and a 3 pixel 
166: Gaussian kernel smoothing function.  We have masked regions interior to a radial distance 
167: of 2$\farcs$0, which are dominated by PSF-subtraction residuals.  
168: Note that several apparent features in these figures 
169: are actually well known instrumental artifacts, whose characteristics and origin are discussed in depth elsewhere \citep{cla03,kr05a,gon05}.  All narrow radial spikes 
170: are residuals due to incomplete PSF subtraction.  Additionally, image imperfections 
171: arising from the 3$\farcs$0 diameter 
172: occulting spot and its occulting finger are clearly seen in all images (WNW of the central
173: source in the F435W filter, ESE of the central source in the F606W and F814W filters).  
174: This occulting finger and spot also projects ``horn-shaped'' artifacts located 
175: at a distance of 2$\farcs$25-3$\farcs$0 from the central source.
176: 
177: Three Herbig-Haro knots and a bipolar jet are clearly visible in our multi-epoch images; 
178: these features have been identified in previous investigations \citep{gra00,dev00}, and 
179: \citet{was06} has used some of these data to estimate both the proper motion of the knots 
180: and the mass-loss rate of the HD 163296 system.  Having identified these known features and 
181: artifacts of our data, we now focus on describing the radial and azimuthal surface brightness, color, 
182: and morphological features of the HD 163296 disk.
183: 
184: \subsection{Disk Surface Brightness} \label{acs-sb}
185: 
186: While the surface brightness profile of portions of the HD 163296 disk were characterized 
187: via white-light STIS coronagraphic imaging \citep{gra00}, the large occulting wedges used 
188: by this coronagraph prevents an investigation of the full radial and azimuthal structure 
189: of the disk.  Our ACS images do not suffer from such artifacts, hence we are able to provide the 
190: first complete, multi-color assessment of the HD 163296 disk.  
191: 
192: We first calculate the disk surface brightness 
193: along the semi-major axis, as reported by \citet{gra00} and \citet{was06}.  We extracted 10 pixel 
194: wide (0$\farcs$25) radial profile cuts at position angles of 132$^{\circ}$ and 312$^{\circ}$ measured 
195: north to east, and subsequently combined these data in 4 pixel (0$\farcs$10) bins (see 
196: Figure \ref{azradprofile}).  As illustrated in Figure \ref{allfit} for our F606W data, these radial profiles 
197: followed three distinct power-law trends.  Between $\sim$2$\farcs$2 to $\sim$2$\farcs$8, 
198: evidence of scattered light is apparent; however, it is likely that PSF-subtraction residuals at least 
199: partially contribute to this measured flux.  Clear evidence of the scattered light disk is seen in all filters from 
200: $\sim$2$\farcs$9 - $\sim$4$\farcs$4 (350-540 AU).  The abrupt turnover in the profiles at radial 
201: distances $>$4$\farcs$4 (540 AU) represents the termination of our detection of the disk; this boundary
202:  could either indicate the physical outer radius of the disk's upper atmosphere or merely the radial 
203: location at which the disk becomes extremely optically thin.    
204: 
205: Table \ref{tabmajor} lists the exponents of the best power law fits derived along the disk major axis 
206: in each of our three filters.  Concentrating on the $\sim$2$\farcs$9 - $\sim$4$\farcs$4 regions over 
207: which we are clearly seeing the scattered light disk, several features and trends are apparent.  At the 
208: level of the SNR of our data, we do not detect clear evidence of 
209: differences between the surface brightness behavior in the SE versus NW quadrants of the disk in the 
210: F606W and F814W filters.  Note that the SE quadrant of our F435W disk major axis radial profile cut 
211: is ``contaminated'' 
212: by the presence of an ansa feature, which will be discussed in Section 3.3.
213:    
214: 
215: \subsubsection{Disk Size}
216: 
217: A large range of sizes have been suggested for the HD 163296 circumstellar disk.  \citet{dom03} 
218: assumed an outer radius of 50 AU in their SED fitting of the system.  \citet{ise07} reported that CO 
219: emission, likely produced in the upper layers of the disk, extended out to a radial distance of 540 AU; 
220: moreover, they reported an unexpected deficit of mm continuum emission beyond 200 AU which they 
221: interpreted as evidence that the disk was strongly depleted of dust outside of 200 AU.  \citet{ise07} 
222: speculated that this deficit could be due to the inward migration of large bodies in the disk or due to 
223: clearing by a low mass companion.
224: 
225: Our spatially resolved imagery reveals that light scattering off of dust residing in the upper layers of the 
226: HD 163296 disk extends out to a radial distance of $\sim$540 AU.  Given our detection, we suggest that there is no 
227: physical gap of dust at radial distances greater than 200 AU as hypothesized by \citet{ise07}.  Rather, 
228: we suggest that some of the alternate interpretations discussed in \citet{ise07} might provide more 
229: feasible explanations for the non-detection of dust mm continuum emission beyond 200 AU, including 
230: grain growth and/or a change in the disk geometry beyond 200 AU (assumed by \citealt{ise07} 
231: to be a fully flared disk).  
232: 
233: \subsubsection{Disk Geometry}
234: 
235: Power-law fits to scattered light imagery provide one means to diagnose the geometry of Herbig 
236: Ae circumstellar disks (see e.g. \citealt{gra07}).  As seen in Table \ref{tabmajor}, the surface 
237: brightness of the HD 163296 disk in the ACS F435W, F606W, and F814W filters follow a 
238: $\sim$r$^{-4}$ dependence.  These data clearly deviate from the expected behavior of a 
239: continuously flared disk, which would follow a r$^{-2}$ dependence, or a geometrically 
240: thin disk, which would follow a r$^{-3}$ dependence \citep{whi92}.  
241: \citet{dom03} also reached the conclusion that the HD 163296 disk was not continuously flared based 
242: on SED modeling, though some of the parameters adopted in their model which produced this 
243: result (i.e. a disk outer radius of 50 AU) are dubious.   
244: 
245: 
246: \subsection{Disk Color} \label{acs-color}
247: 
248: Our multi-filter HST ACS dataset also allows us to provide the first measurement of 
249: the color of HD 163296's scattered light disk.  We first calculated the surface 
250: brightness at radial distances of 3$\farcs$0 and 3$\farcs$5 along the disk semi-major 
251: axis (PAs of 132$^{\circ}$ and 312$^{\circ}$).  The median flux within 1$\farcs$0 x 1$\farcs$0 
252: apertures were determined at each location, and complimentary measurements (PAs of 132 
253: and 312) were averaged to produce the surface brightnesses listed in Table \ref{colors}.  Given our 
254: previous assessment that PSF-subtraction residuals could be contributing to the observed scattered light 
255: inside a radial distance of 2$\farcs$9, we caution that our aperture photometry at 3$\farcs$0 similarly 
256: might be \textit{marginally} contaminated: by contrast, the aperture photometry at 3$\farcs$5 
257: should be free 
258: from such effects.  Background sky surface brightnesses (Table \ref{colors}) were determined by 
259: averaging the median pixel values in two 1$\farcs$0 x 1$\farcs$0 apertures located 
260: 9$\farcs$0 from the central star.  The standard deviation of these background sky apertures 
261: were used to provide an estimate of the photometric uncertainties within our science 
262: apertures at 3$\farcs$0 and 3$\farcs$5 along the disk semi-major axis.  
263: 
264: The surface brightnesses in each of our ACS filters (Table \ref{colors}) were 
265: converted to the standard BVI system by iteration of the transformation equation 
266: outlined by \citet{sir05}; TMag = SMag + c0 + c1 * 
267: TC + c2 * TC$^{2}$, where TMag is the Johnson B, V, or I magnitude,  
268: SMag is the ACS filter magnitude in the OBMag 
269: photometric system, TC 
270: is the OBMag color difference of the filters used, and c0, c1, and c2 are 
271: transformation coefficients tabulated by \citet{sir05}.  Note that the B-band 
272: surface brightness and (B-V) color were calculated using multi-epoch 
273: data, while V- and I-band surface brightnesses and (V-I) color were calculated using 
274: single epoch data (see Table \ref{johncolors}).
275: 
276: The observed disk (V-I) color 3$\farcs$5 from the central star along the semi major axis, 1.1, 
277: is significantly redder than literature measurements of the stellar (V-I) color, 0.15 \citep{the85}.  
278: To better assess the global behavior of the disk's color, we transformed our F606W and F814W 
279: imagery to the Johnson V- and I-band filters using the aforementioned techniques.  The resultant 
280: (V-I) image of the scattered light disk, plotted on a linear scale and smoothed with an 
281: 8 pixel Gaussian kernel, is shown in Figure \ref{colorfig}.  From inspection of Figure \ref{colorfig},  
282: we conclude that our data provide no clear evidence of a spatial dependence in the (V-I) color 
283: of the scattered light disk.
284: 
285: With the exception of AU Mic \citep{kr05a}, the colors of many other resolved disk systems are 
286: similarly redder than 
287: their host stars (HR 4796A, \citealt{sch99}; HD 141569A, \citealt{cla03}; 
288: GG Tauri, \citealt{kr05b}; $\beta$ Pic, \citealt{gol06}).  As discussed for these 
289: aforementioned systems, disk colors have been used to provide tentative 
290: constraints on the relative size distribution and/or chemistry of grain populations, 
291: although degeneracies clearly exist in such an analysis (e.g. \citealt{mcc02}).   
292: One possible interpretation of a red disk color is that some amount of grain 
293: processing has occurred (i.e. grain growth; chemistry evolution).  Grain growth 
294: and grain settling is a phenomenon which is believed to occur in Herbig Ae 
295: disks \citep{van03,ack04,dul04,du04b}, and is generally associated with systems 
296: whose outer disks are not continuously flared.  Previous analysis of the IR SED of HD 163296 
297: has led to suggestions that this disk system has experienced grain growth and 
298: settling \citep{mee01}.  The detection of a non-continuously flared disk and red (V-I) disk color 
299: from our spatially resolved scattered light imagery is thus consistent (though not conclusive) 
300: with such phenomenon occurring.  We also note that our HD 163296 imagery does not exhibit 
301: significant evidence of a scattering asymmetry due to differences in forward and back 
302: scattering of grains in the disk, as is seen in systems such as GG Tauri \citep{kr05b}.  Future 
303: modeling efforts which utilize both the known SED properties \citep{mee01} and scattered 
304: light properties (the present study) of the HD 163296 disk would help to 
305: better constrain the grain size distribution in a more quantitative manner.  
306: 
307: \subsection{Morphological Features} \label{acs-morph}
308: 
309: The superlative spatial resolution of the HST ACS allows for the search of both radial 
310: and azimuthal morphological structures within the HD 163296 scattered light disk.  
311: Simple visual inspection of our F606W and F814W images 
312: (Figure \ref{alldata}) 
313: reveals no clear evidence of disk structure.  By contrast, our later epoch F435W data  
314: (Figure \ref{alldata}) clearly exhibits evidence of an ansa-like structure in the 
315: southeast (SE) quadrant of the disk, located at the same position of one of two ansae noted 
316: in archival HST STIS coronagraphic observations \citep{gra00}.  The ansa is 
317: visible and stationary in both of the roll angle images at F435W, whereas PSF-subtraction 
318: artifacts rotate with the roll angle; moreover, this structure is not associated with the 
319: ``horn-shaped'' artifacts attributable to the ACS large occulting spot and finger.
320: 
321: We extracted 10 pixel wide (0$\farcs$25) median averaged radial profile cuts from our 
322: data to parameterize this disk structure.  The bottom panels of Figure \ref{radprof1} depict 
323: the radial surface brightness profiles, in 3 pixel (0$\farcs$075) radial bins, at position 
324: angles of 140$^{\circ}$ N to E (SE disk) and 320$^{\circ}$ N to E (NW disk) for our 
325: three band-passes.  At its brightest 
326: point, a radial distance of $\sim$3$\farcs$1-3$\farcs$2, the structure has a F435W surface 
327: brightness of $\sim$19.5 mag arcsec$^{-2}$, and is preceded by sudden dimming in 
328: surface brightness.  At slightly different position angles, 130$^{\circ}$ 
329: N to E and 310$^{\circ}$ N to E (see top panel, Figure \ref{radprof1}), the feature 
330: is slightly brighter (F435W $\sim$19.0 mag arcsec$^{-2}$), characterized by a 
331: less steep dimming in
332: surface brightness, and reaches a peak flux at a slightly more interior radial distance.  
333: At both position angles, no corresponding structures are seen either in the NW cut 
334: of the F435W data or the SE and NW cuts of the F606W and F814W data.  
335: Note that the apparent ``feature'' in the F814W filter (and less apparent in the F606W 
336: filter) data at 3$\farcs$5 in the NW disk quadrant is merely a background star.  
337: It is quite clear that the overall surface brightness of the F435W data 
338: at the radial distance of the feature ($\sim$3$\farcs$0) is brighter in the SE region of the 
339: disk than in NW region of the disk.
340: 
341: 
342: \section{The Multi-Epoch Behavior of the HD 163296 Scattered Light Disk} \label{stis}
343: 
344: \citet{gra00} presented white-light HST STIS coronagraphic observations of HD 163296, 
345: obtained in September 1998.  The STIS observations were characterized by a featureless signal 
346: from 1$\farcs$5 to 2$\farcs$5, a darker zone at 2$\farcs$65 of width 0$\farcs$4, and two bright
347: ansae peaking at 2$\farcs$9 in the SE quadrant and 3$\farcs$2 in the NW quadrant (see 
348: Figure \ref{alldata}).  \citet{gra00} 
349: considered two primary explanations for the 
350: observed bright ansae and dark lanes in their STIS imagery, changes in the dust grain composition and 
351: dynamical clearing by planetary-like bodies, although they noted other possibilities could 
352: not be excluded.  
353: 
354: Comparison of these archival white-light ($\lambda_{central}$ = 5850 \AA, $\Delta\lambda$ 
355: = 4410 \AA) STIS data with our multi-epoch F435W ($\lambda_{central}$ = 4297 \AA, 
356: $\Delta\lambda$ = 1038 \AA), F606W ($\lambda_{central}$ = 5907 \AA, 
357: $\Delta\lambda$ = 2342 \AA), and F814W ($\lambda_{central}$ = 8333 \AA, $\Delta\lambda$ = 
358: 2511 \AA) ACS observations requires caution, given the different band-passes sampled by each 
359: individual observation.  Our single-epoch F606W and F814W observations span most of 
360: the STIS bandpass.  To serve as a crude, \textit{approximate} template 
361: for comparison to the archival STIS images, we have combined our ACS F606W and F814W data after
362: first normalizing the images to account for the different throughput of these filters, $\sim$0.16 for F814W and $\sim$0.26 for F606W.  Hereafter we refer to this combination as our ACS F606W+F814W dataset.  
363: Similarly, \textit{under the assumption that the HD 163296 disk surface brightness is 
364: not variable}, we have combined our different epochs of ACS data, again after normalizing the images 
365: to account for the different throughput of the filters, $\sim$0.16 for F814W, $\sim$0.26 for F606W, and $\sim$0.19 for F435W.  We will refer to this latter combination, which covers the majority of the STIS 
366: clear filter bandpass, as our ACS F435W+F606W+F814W dataset.
367: 
368: \subsection{Disk Surface Brightness} \label{stis-sb}
369: 
370: Mirroring the analysis performed in Section \ref{acs-color}, we calculated the 
371: average disk surface brightness along the semi-major axis of the disk, at radial distances of 
372: 3$\farcs$0 and 3$\farcs$5 from the central star.  These surface brightness measurements are listed in 
373: Table \ref{colors}; we also report the surface brightness of the STIS epoch data in the 
374: Johnson V-band in Table \ref{johncolors}.   We stress that this quoted V-band 
375: surface brightness, calculated via a transformation equation which incorporated our 
376: observed (V-I) disk color, \textit{is only a crude approximation}.  
377: 
378: The 1998 STIS epoch surface brightness along the major axis of the HD 163296 scattered light 
379: disk is significantly fainter than all other measurements in all other filter band-passes.  In the 
380: Johnson V-band system, the 1998 STIS observation at a radial distance of 
381: 3$\farcs$5 from the central star (20.1 mag arcsec$^{-2}$) is $\sim$1 mag arcsec$^{-2}$ fainter 
382: than the 2003 ACS V-band surface brightness (19.2 mag arcsec$^{-2}$; see Table \ref{johncolors}).  
383: In the instrumental filter system, the 1998 STIS clear filter 
384: surface brightness at a radial distance of 3$\farcs$5 from the central star 
385: (20.3 mag arcsec$^{-2}$) is similarly $\sim$1.0 mag arcsec$^{-2}$ fainter 
386: than that measured in both of our ACS ``white-light'' filter approximation datasets (19.3 and 19.1 
387: mag arcsec$^{-2}$ respectively; see Table \ref{colors}).  As the 1998 
388: epoch STIS data are significantly fainter than that measured in each bandpass of our ACS data, 
389: it is clear that no combination of the ACS data can reproduce the behavior of the 
390: STIS scattered light surface brightnesses.  We suggest that the observed multi-epoch behavior 
391: of the disk's surface brightness is \textit{one} piece of evidence which suggests that 
392: the system's scattered light disk is variable.
393: 
394: \subsection{Morphological Features} \label{all-morph}
395: 
396: \citet{gra00} reported the presence of two ansae in the white-light 
397: STIS coronagraphic image of HD 163296 obtained in 1998, with the brightest ansa located in 
398: the NW quadrant of the disk.  In Section \ref{acs-morph}, we reported a 
399: null detection of any morphological features in our 2003 epoch data (F606W and F814W filters) and the  
400: detection of a single ansa structure in the SE quadrant of our HST ACS F435W 
401: data.  The variability in the visibility of these morphological features are easily seen 
402: in Figure \ref{alldata}.  
403: 
404: To compare and contrast these features in a quantitative manner, we
405: have followed the technique outlined in Section \ref{acs-morph} and extracted radial 
406: surface brightness profiles for the ACS F606W+F814W 
407: and STIS data (see Figure \ref{radprof2}). 
408: Note that our ACS radial profile cuts correspond to the median average of a 
409: 10 pixel wide (0$\farcs$25 arcsec) aperture which were subsequently averaged in 3 
410: (radial) pixel bins, while our STIS radial profile cuts correspond to the median average 
411: of a 5 pixel wide (0$\farcs$25 arcsec) aperture which were subsequently averaged in 
412: 3 (radial) pixel bins.  
413: 
414: It is clear that the ansa structure we reported in our ACS F435W data in the SE quadrant 
415: of the disk (top left panel, Figure \ref{radprof2}) occurs at the same radial distance as 
416: the ansa in the SE quadrant reported by \citet{gra00}; hence it is plausible to suggest that 
417: these morphological features are in fact the same phenomenon.  Interestingly, 
418: while the ansa is brighter in the NW quadrant than in the SE quadrant in the 
419: STIS data, a nearly opposite scenario characterizes the ACS F435W data, 
420: such that the ansa is present in the SE quadrant and absent in the NW quadrant.  The ansa 
421: features which characterize the HD 163296 scattered light disk not only exhibit variability in 
422: the epochs in which they are discernible, but also exhibit different surface brightnesses in 
423: the epochs in which they are visible.  
424: 
425: The non-detection of these morphological features in our 2003 epoch data are unlikely to be 
426: due to the sensitivity limits of our data.  Ignoring the obvious differences in band-passes, the 
427: observed surface brightness of our ACS data which fail to detect any 
428: morphological features (2003; F606W, F814W) are actually brighter than that observed in our 
429: ACS data in which structure is detected (2004; F435W).  Moreover, the limiting magnitude of 
430: our 2003 ACS data, depicted as dashed horizontal lines in Figure \ref{radprof1}, is 
431: one magnitude arcsec$^{-2}$ below the measured surface brightness of the morphological feature in 
432: our 2004 epoch F435W data, inside of a radial distance of 3$\farcs$5.  
433: 
434: One might also reason that, given the different band-passes probed by our multi-epoch ACS and STIS data, 
435: perhaps the visibility of the morphological structure is an optical depth effect, and is 
436: extremely color dependent (i.e. produced by small particle size scatterers).  Given the very red (V-I) 
437: color of the disk (Section \ref{acs-color}), we suggest that this scenario is dubious.  Moreover, 
438: it is difficult to envision how an optical depth phenomenon could simultaneously explain both the visibility 
439: pattern (visible in 1998 white-light; 2004 F435W) and surface brightness variability (NW ansa brighter than 
440: SE ansa in 1998; SE ansa brighter than (non-detected) NW ansa in 2004) of these morphological 
441: features.  Rather, we suggest that the visibility and surface brightness variability of these morphological 
442: features are \textit{two} additional pieces of evidence which suggest that the system's scattered light 
443: disk is variable.   
444: 
445: 
446: We also note that unresolved linear spectropolarimetric monitoring of HD 163296 has 
447: revealed evidence of a time variable intrinsic polarization component (Bjorkman 2007, 
448: personal communication).  Such data indicate variability in the net (unresolved) scattered light 
449: behavior of the system.  As the integrated scattered light of HD 163296 exhibits variability, 
450: we suggest it is perhaps not surprising to find evidence that when this scattered light is 
451: resolved via coronagraphic imaging, that one finds multiple avenues of evidence indicating 
452: variability.  
453: 
454: 
455: \section{Origin of the Observed Scattered Light Variability} \label{origin}
456: 
457: Our analysis of multi-epoch, multi-color spatially resolved imagery of HD 163296 
458: suggests that its scattered light disk is variable; we observe a $\sim$1 mag arcsec$^{-2}$ 
459: change in the overall surface brightness of the disk between epochs of observations, variability in the 
460: visibility of ansa structure(s), 
461: and azimuthally asymmetric variability of the surface brightness of these ansa(e), when they are visible.  
462: We now consider the possible origins 
463: of this variability, in the context of what is already known about the nature of the HD 163296
464: disk.  
465: 
466: Previous interpretations of the bright ansae and dark zones which 
467: characterize the STIS scattered light data \citep{gra00} outlined numerous mechanisms  
468: which could explain the origin of the observed phenomenon, 
469: including dynamical clearing, grain composition and/or size distribution differences,
470: changes in the degree of disk flaring, and changes in the number density of grains. 
471: \citet{gra00} also postulated that, if the darker zones were in fact a partially cleared 
472: annulus, then a $\sim$0.4 M$_{Jupiter}$ gas giant planet might be responsible for  
473: producing the disk clearing.  While our 2003 epoch ACS F606W and F814W data clearly 
474: detected the HD 163296 scattered light disk, no evidence of any dark lanes were 
475: observed.  Thus it appears unlikely that the morphological ``dark lane'' identified 
476: by \citet{gra00} corresponds to a cleared region of the disk; we suggest that the 
477: scattered light data present no clear evidence of the presence of a planetary mass 
478: body.  
479: 
480: We consider two classes of mechanisms which could induce variability in our scattered 
481: light observations, changes in the fundamental properties of the scatterers and changes 
482: in the illumination of these scatterers.  Our data indicate that the scattered light disk is 
483: variable on \textit{at least} a time-scale of several years, which strongly suggests that 
484: we can exclude changes in the fundamental properties of the scatterers (i.e. grain composition, 
485: grain size distribution, grain number density, and degree of disk flaring) as a plausible 
486: mechanism.  
487: 
488: Changes in the illumination of the disk, from the perspective of the scatterers, 
489: could include either (or both) stellar variability or variations in the number of stellar photons 
490: which reach the scatterers.  We are not aware of any evidence which suggests that HD 163296 
491: experiences UX-Ori-like photometric outbursts \citep{wat98,dul03}; rather, the star  
492: appears to have a generally stable photometric brightness of V=6.88, 
493: with a maximum variability range (high to low) of 0.42 magnitudes \citep{her99}.  
494: Direct imaging of HD 163296 was obtained in each of our ACS filters at the epoch of our coronagraphic 
495: observations; however, these data were heavily saturated and did not use a gain 
496: setting which sampled the full well depth of the HRC, which would have permitted  
497: accurate photometry to be extracted \citep{gil04,sir05}.  Thus, we do not have any 
498: observational evidence which could link HD 163296's observed scattered light variability with 
499: stellar variability.  Moreover, it is not clear that brightening/dimming of the central star 
500: could explain the visibility and surface brightness variability patterns observed in the 
501: disk ansae features.  Rather, we suggest that HD 163296's scattered 
502: light variability is related to changes in the number of photons which reach the 
503: disk scatterers.
504: 
505: Self-shadowing of the outer regions of Herbig Ae disks is 
506: a phenomenon predicted by theory \citep{nat01,dul01,dul04,ise05}, and variable illumination 
507: of envelopes has been reported in observations of GG Tau \citep{kr05b} and Hubble's 
508: Variable Nebula \citep{lig89}. One suggested 
509: method to produce a self-shadowed disk is by inflation of the inner 
510: disk wall, which prevents copious amounts of photons from reaching at least 
511: part of the outer (flatter) disk.  \citet{sit06} has noted that the infrared spectral energy 
512: distribution (IR SED) of HD 163296 can vary by $\sim$30\% on time-scales of 3 years 
513: or less, which they interpret as evidence of structural changes in the region of the disk near 
514: the dust sublimation zone.  The mechanism driving this inner disk variability is not known, although 
515: \citet{sit06} discuss several possibilities: $\alpha$-disk instabilities, magneto-rotational 
516: instabilities, an X-wind origin, and perturbations by a planetary-mass body.  
517: Given the non-continuously flared geometry of the HD 163296 outer 
518: disk, as inferred from our ACS 
519: scattered light data (Section \ref{acs-sb}), it is plausible that even small-scale changes 
520: in the scale height of the inner disk wall could induce a variable amount of self-shadowing 
521: of the outer disk.  Thus we suggest that variable 
522: self-shadowing is one plausible mechanism which might explain the observed variability of 
523: the outer scattered light disk.  Coordinated, contemporaneous observations of the inner and 
524: outer disk regions of HD 163296, namely follow-up IR SED and coronagraphic monitoring, 
525: would provide a robust means of investigating this proposed mechanism.
526: 
527: Recall that the bright ansa(e) features which sometimes characterize the 
528: HD 163296 scattered light disk were only observed during epochs in which the overall disk 
529: surface brightness was low (1998, 2004; see Figure \ref{radprof2} and Tables \ref{colors} 
530: and \ref{johncolors}).  Thus, in our working picture of the HD 163296 scattered light disk, 
531: with admittedly few data points to work with, these morphological features are only detected 
532: during periods of enhanced self-shadowing.  We suggest 
533: that the source of these observed bright features is not a density enhancement of scatterers
534: (i.e. a clump), as one would expect such a clump would contribute a localized 
535: enhancement of scattered light not only during enhanced self-shadowing periods, but 
536: also during epochs characterized by less self-shadowing.  Rather, 
537: we speculate that these bright 
538: features signify a change in the spatial distribution of the scatterers, i.e. the ansa(e) 
539: represent a region in the disk which is modestly less flattened than other 
540: surrounding locations.  During periods of increased self-shadowing these 
541: structures would partially escaped the shadow created by the inner disk wall.   
542: During periods of less self-shadowing, this region would be illuminated similar to its 
543: neighboring disk regions, thus as long as the structures do not have a significantly higher 
544: density of scatterers, the morphological feature would not be conspicuous in scattered light profiles. 
545: While the origin of the perturbations which give rise to these localized features is unknown, 
546: structure in the tenuous upper layers of protoplanetary disks is predicted to occur by theory.  
547: For example, \citet{jan07} predicted such a phenomenon would occur in the inner regions of 
548: protoplanetary disks as a byproduct of planet formation via the disk instability model.  
549: 
550: 
551: Finally, we consider the origin of the surface brightness variability of the ansa(e) 
552: structures, during the epochs in which they were visible (Section \ref{all-morph}; 
553: Figure \ref{radprof2}).  If the visibility of these features is linked to the behavior
554: of the inner disk wall, as we have suggested, then it is likely that the surface 
555: brightness variability of these features is 
556: also linked to events in the inner disk wall region.  We speculate that  
557: the inner disk wall might be azimuthally asymmetric, hence cast an azimuthally 
558: asymmetric shadow on the outer disk.  Given the 
559: orbital period of material located at the inner wall sublimation radius, $\leq$1 week 
560: (Sitko 2007, personal communication), the time-scale for such a asymmetry to produce 
561: a brighter NW versus SE illumination (and vice-versa) is much less than the maximum 
562: time-scale of the observed ansa variability, several years.  We suggest that a 
563: series of scattered light images of HD163296, obtained in time steps of $\sim$1 to 
564: several weeks \textit{and} during a period of enhanced self-shadowing, would provide 
565: a direct test of our interpretation of the system.
566: 
567: \section{Summary} \label{summary}
568: 
569: We have presented the first multi-color, multi-epoch analysis of resolved optical 
570: scattered light imaging of the Herbig Ae star HD 163296.  To summarize the 
571: observational properties of these data: \begin{enumerate}
572: 
573: \item We spatially resolved the HD 163296 scattered light disk over radial 
574: distances of 2$\farcs$9-4$\farcs$4 (350-540 AU) in the HST/ACS F435W, F606W, and 
575: F814W filters.
576: 
577: \item Radial profiles of the surface brightness along the semi-major axis of the disk 
578: follow a $\sim$r$^{-4}$ power law behavior, which is indicative of 
579: a non-continuously flared disk.
580: 
581: \item The (V-I) color of the disk at 3$\farcs$5, $\sim$1.1, 
582: is significantly redder than the stellar (V-I) 
583: color, 0.15.  This red disk color appears to be spatially uniform at the SNR of our data, and 
584: is consistent with that expected for a disk which has 
585: experienced grain growth and is at least partially self-shadowed.
586: 
587: \item A single ansa structure is present in the SE disk quadrant of the 
588: 2004 epoch ACS data (F435W filter), while no structure was observed 
589: in the 2003 epoch ACS data (F606W and F814W filters), despite the fact that the 
590: limiting detection magnitude of these latter data was sufficient to probe structure 
591: at the surface brightness observed in 2004.  This morphological feature is spatially  
592: coincident with the fainter ansa reported in 1998 STIS observations \citep{gra00} of 
593: HD 163296; the brighter ansa seen in the NW disk quadrant of the STIS data are 
594: absent from our ACS epoch data.  
595: 
596: \item The scattered light disk was observed to be significantly brighter in each of three 
597: filters of ACS observations from 2003-2004, as compared to white-light STIS observations 
598: in 1998.  Combining the multi-color ACS data to crudely approximate 
599: the STIS bandpass suggests that the ACS epoch data are $\sim$1 mag 
600: arcsec$^{-2}$ brighter than the STIS data.  Along with the variability of 
601: the visibility and surface brightness of ansa structure(s) in the disk, these results 
602: \textbf{\textit{suggest the scattered light 
603: disk of HD 163296 is variable}}.  
604: 
605: \end{enumerate}
606: 
607: We expect that the characterization of the basic behavior of HD 163296's 
608: scattered light disk that we have provided in this study will serve as an 
609: important foundation to future efforts to model the multi-wavelength behavior 
610: of the system.  Although speculative, we suggest that one plausible 
611: explanation for the origin of the observed variability is: \begin{enumerate}
612: 
613: \item The scale height of the inner disk wall is believed to inflate and 
614: deflate on time-scales of less than 
615: a few years, based on IR SED monitoring \citep{sit06}.  The variable scale 
616: height of the inner disk 
617: wall could induce variable self-shadowing of the outer disk, hence produce the observed overall 
618: variability of the scattered light disk.
619: 
620: \item We suggest that the ansa structure(s), which appear to be discernible only 
621: during periods of enhanced self-shadowing, represent a localized region of scatterers 
622: which are at a scale height (at least marginally) above the projected shadow.  As the 
623: features are not discernible during periods of more complete illumination of the outer 
624: disk, we believe that a localized enhancement in the scale height of the disk is 
625: more likely to produce the observed phenomenon than a localized density enhancement 
626: of the scatterers (i.e. a clump). 
627: 
628: \item The relative surface brightness variability of the NW versus SE ansa 
629: structures during epochs in which they are visible suggests that the inner disk wall might 
630: be azimuthally asymmetric, hence produce an azimuthally asymmetric shadow on the 
631: outer disk.  Based on the short ($\leq$1 week, Sitko 2007 personal communication) orbital
632: time-scale of material located at the inner wall, we suggest that a series of resolved 
633: scattered light images of HD 163296, obtained during an epoch of enhanced self-shadowing, 
634: would provide a test of this particular suggested phenomenon.
635: 
636: \end{enumerate}
637: 
638: 
639: \acknowledgements 
640: 
641: We thank Mike Sitko, Aki Roberge, Nick Collins, and Karen Bjorkman for helpful discussions 
642: regarding this paper.  We also thank our referee for providing helpful suggestions which 
643: improved the quality of this paper.  Support for this project was provided by NASA NPP fellowship 
644: NNH06CC03B (JPW).  We acknowledge use of the SIMBAD database operated at CDS, 
645: Strasbourg, France, and the NASA ADS system.
646: 
647: 
648: \begin{thebibliography}{}
649: 
650: \bibitem[Acke et al.(2004)]{ack04} Acke, B., van den Ancker, M.E., Dullemond, C.P., 
651: van Boekel, R., \& Waters, L.B.F.M. 2004, A\&A, 422, 621
652: 
653: \bibitem[Ardila et al.(2004)]{ard04} Ardila, D.R. et al. 2004, ApJL, 617, 147
654: 
655: \bibitem[Bjorkman et al.(1995)]{bjo95} Bjorkman, K.S. et al. 1995, BAAS, 27, 1319
656: 
657: \bibitem[Clampin et al.(2003)]{cla03} Clampin, M. et al. 2003, AJ, 126, 385
658: 
659: \bibitem[Devine et al.(2000)]{dev00} Devine, D., Grady, C.A., Kimble, R.A., Woodgate, 
660: B., Bruhweiler, F.C., Boggess, A., Linsky, J.L., \& Clampin, M. 2000, ApJL, 542, 115
661: 
662: \bibitem[Dominik et al.(2003)]{dom03} Dominik, C., Dullemond, C.P., Waters, L.B.F.M., 
663: \& Walch, S. 2003, A\&A, 398, 607
664: 
665: \bibitem[Dullemond et al.(2001)]{dul01} Dullemond, C.P., Dominik, C., \& Natta, A. 
666: 2001, ApJ, 560, 957
667: 
668: \bibitem[Dullemond et al.(2003)]{dul03} Dullemond, C.P., van den Ancker, M.E., Acke, B., 
669: \& van Boekel, R. 2003, ApJL, 594, 47
670: 
671: \bibitem[Dullemond \& Dominik(2004a)]{dul04} Dullemond, C.P. \& Dominik, C. 2004a, A\&A, 417, 159
672: 
673: \bibitem[Dullemond \& Dominik(2004b)]{du04b} Dullemond, C.P. \& Dominik, C. 2004b, 
674: A\&A, 421, 1075
675: 
676: \bibitem[Eisner et al.(2004)]{eis04} Eisner, J.A., Lane, B.F., Hillenbrand, L.A., Akeson, 
677: R.L., \& Sargent, A.I. 2004, ApJ, 613, 1049
678: 
679: \bibitem[Gilliland(2004)]{gil04} Gilliland, R.L. 2004, Instrument Science Report 
680: (ACS 04-01; Baltimore: STScI)
681: 
682: \bibitem[Golimowski et al.(2006)]{gol06} Golimowski, D.A. et al. 2006, AJ, 131, 3109
683: 
684: \bibitem[Gonzaga et al.(2005)]{gon05} Gonzaga, S. et al. 2005, ACS Instrument 
685: Handbook, Version 6.0, (Baltimore:STScI)
686: 
687: \bibitem[Grady et al.(2000)]{gra00} Grady, C.A., Devine, D., Woodgate, B., 
688: Kimble, R., Bruhweiler, F.C., Boggess, A., Linsky, J.L., Plait, P., Clampin, M., 
689: \& Kalas, P. 2000, ApJ, 544, 895
690: 
691: \bibitem[Grady et al.(2005)]{gra05} Grady, C.A. et al. 2005, ApJ, 630, 958
692: 
693: \bibitem[Grady et al.(2007)]{gra07} Grady, C.A. et al. 2007, ApJ, 665, 1391
694: 
695: \bibitem[Habart et al.(2006)]{hab06} Habart, E., Nata, A., Testi, L., \& Carbillet, M. 2006, 
696: A\&A, 449, 1067
697: 
698: \bibitem[Herbig(1960)]{her60} Herbig, G. 1960, ApJS, 4, 337
699: 
700: \bibitem[Herbst \& Shevchenko(1999)]{her99} Herbst, W. \& Shevchenko, V.S. 
701: 1999, AJ, 118, 1043
702: 
703: \bibitem[Isella \& Natta(2005)]{ise05} Isella, A. \& Natta, A. 2005, A\&A, 438, 899
704: 
705: \bibitem[Isella et al.(2006)]{ise06} Isella, A., Testi, L., \& Natta, A. 2006, A\&A, 451, 951
706: 
707: \bibitem[Isella et al.(2007)]{ise07} Isella, A., Testi, L., Natta, A., Neri, R., Wilner, D., 
708: \& Qi, C. 2007, A\&A, 469, 213
709: 
710: \bibitem[Jang-Condell \& Boss(2007)]{jan07} Jang-Condell, H. \& Boss, A.P. 2007, ApJL, 659, 169
711: 
712: \bibitem[Kalas et al.(2005)]{kal05} Kalas, P., Graham, J.R., \& Clampin, M. 2005, Nature, 
713: 435, 1067
714: 
715: \bibitem[Kalas et al.(2006)]{kal06} Kalas, P., Graham, J.R., Clampin, M.C., \& Fitzgerald, 
716: M.P. 2006, ApJL, 637, 57
717: 
718: \bibitem[Krist et al.(2005a)]{kr05a} Krist, J.E. et al. 2005a, AJ, 129, 1008
719: 
720: \bibitem[Krist et al.(2005b)]{kr05b} Krist, J.E. et al. 2005b, AJ, 130, 2778
721: 
722: \bibitem[Lecavelier des Etangs et al.(1993)]{lec93} Lecavelier des Etangs, A. et 
723: al. 1993, A\&A, 274, 877
724: 
725: \bibitem[Lightfoot (1989)]{lig89} Lightfoot, J.F. 1989, MNRAS, 239, 665
726: 
727: \bibitem[Mannings \& Sargent(1997)]{man97} Mannings, V. \& Sargent, A.I. 1997, ApJ, 490, 792
728: 
729: \bibitem[Mannings \& Sargent(2000)]{man00} Mannings, V. \& Sargent, A.I. 2000, ApJ, 529, 391
730: 
731: \bibitem[McCabe et al.(2002)]{mcc02} McCabe, C., Duchene, G., \& Ghez, A.M. 2002, 
732: ApJ, 575, 974
733: 
734: \bibitem[Meeus et al.(2001)]{mee01} Meeus, G., Waters, L.B.F.M., Bouwman, J., 
735: van den Ancker, M.E., Waelkens, C., \& Malfait, K. 2001, A\&A, 365, 476
736: 
737: \bibitem[Natta et al.(2001)]{nat01} Natta, A., Prusti, T., Neri, R., Grinin, V.P., 
738: \& Mannings, V. 2001, A\&A, 371, 186
739: 
740: \bibitem[Oudmaijer \& Drew(1999)]{oud99} Oudmaijer, R.D. \& Drew, J.E. 1999, MNRAS, 
741: 305, 166
742: 
743: \bibitem[Paresce \& Burrows(1987)]{par87} Paresce, F. \& Burrows, C. 1987, 
744: ApJL, 319, 23
745: 
746: \bibitem[Pavlovsky et al.(2005)]{pav05} Pavlovsky, C., et al. 2005, ACS Data 
747: Handbook, Version 4.0, (Baltimore: STScI)
748: 
749: \bibitem[Schneider et al.(1999)]{sch99} Schneider, G. et al. 1999, ApJL, 513, 127
750: 
751: \bibitem[Sirianni et al.(2005)]{sir05} Sirianni, M. et al. 2005, PASP, 117, 1049
752: 
753: \bibitem[Sitko et al.(2008)]{sit06} Sitko, M. L. et al. 2008, ApJ, 678, 1070
754: 
755: \bibitem[Th\'e et al.(1985)]{the85} Th\'e, P.S., Felenbok, P., Cuypers, H., \& Tjin A Djie, 
756: H.R.E. 1985, A\&A, 149, 429
757: 
758: \bibitem[van Boekel et al.(2003)]{van03} van Boekel, R., Waters, L.B.F.M., Dominik, C., 
759: Bouwman, J., de Koter, A., Dullemond, C.P., \& Paresce, F. 2003, A\&A, 400, L21
760: 
761: \bibitem[van den Ancker et al.(1998)]{van98} van den Ancker, M.E., de Winter, D., \& 
762: Tjin A Djie, H.R.E. 1998, A\&A, 330, 145
763: 
764: \bibitem[van den Ancker et al.(2000)]{van00} van den Ancker, M.E., Bouwman, J., Wesselius, 
765: P.R., Waters, L.B.F.M., Dougherty, S.M., \& van Dishoeck, E.F. 2000, A\&A, 357, 325 
766: 
767: \bibitem[Vink et al.(2002)]{vin02} Vink, J.S., Drew, J.E., Harries, T.J., \& Oudmaijer, R.D. 
768: 2002, MNRAS, 337, 356
769: 
770: \bibitem[Wassell et al.(2006)]{was06} Wassell, E.J., Grady, C.A., Woodgate, B., 
771: Kimble, R.A., \& Bruhweiler, F.C. 2006, ApJ, 650, 985
772: 
773: \bibitem[Waters \& Waelkens(1998)]{wat98} Waters, L.B.F.M. \& Waelkens, C. 1998, 
774: ARA\&A, 36, 233
775: 
776: \bibitem[Whitney \& Hartmann(1992)]{whi92} Whitney, B.A. \& Hartmann, L. 1992, ApJ, 
777: 395, 529
778: 
779: \end{thebibliography}
780: 
781: 
782: \clearpage
783: \begin{table}
784: \begin{center}
785: \footnotesize
786: \caption{Summary of Observations\label{datatable}}
787: \begin{tabular}{lcccccc}
788: \tableline
789: Object & Date & Instrument & Filter & Expos. Time & PA & Comment \\
790: \tableline
791: HD 163296 & 16 Aug 2004 & HST ACS HRC & 435W & 2275 sec & 281 & roll angle 1 \\
792: HD 163296 & 16 Aug 2004 & HST ACS HRC & 435W & 2480 sec & 251 & roll angle 2 \\
793: HD 145570 & 16 Aug 2004 & HST ACS HRC & 435W & 2320 sec & 280 & PSF star \\
794: HD 163296 & 25 Mar 2003 & HST ACS HRC & 606W & 2350 sec & 76 & roll angle 1 \\
795: HD 163296 & 25 Mar 2003 & HST ACS HRC & 606W & 2350 sec & 104 & roll angle 2 \\
796: HD 145570 & 25 Mar 2003 & HST ACS HRC & 606W & 900 sec & 94 & PSF star \\
797: HD 163296 & 25 Mar 2003 & HST ACS HRC & 814W & 2100 sec & 76 & roll angle 1 \\
798: HD 163296 & 25 Mar 2003 & HST ACS HRC & 814W & 2260 sec & 104 & roll angle 2 \\
799: HD 145570 & 25 Mar 2003 & HST ACS HRC & 814W & 900 sec & 94 & PSF star \\
800: \tableline
801: \end{tabular}
802: \end{center}
803: \vspace{-0.3in}
804: \tablecomments{Note that the PA listed in column six refers to the position angle 
805: of the V3-axis (roll-angle) of the HST.}
806: \end{table}
807: 
808: 
809: \clearpage
810: \begin{table}
811: \begin{center}
812: \footnotesize
813: \caption{Power Law Exponents of Disk Major Axis Radial Profiles \label{tabmajor}}
814: \begin{tabular}{lccc}
815: \tableline
816: Filter-Disk Quadrant & Region 1 & Region 2 & Region 3 \\
817:               & 2$\farcs$2-2$\farcs$8 & 2$\farcs$9-4$\farcs$4 & 4$\farcs$5-6$\farcs$1 \\
818: \tableline
819: F435W-NW & 1.4 $\pm$1.1 & 3.7 $\pm$0.3 & 2.2 $\pm$0.6 \\
820: F435W-SE & 4.9 $\pm$1.1 & 5.3 $\pm$0.4 & 1.5 $\pm$ 0.5 \\
821: F606W-NW & 0.5 $\pm$0.8 & 4.6 $\pm$0.2 & 1.1 $\pm$0.2 \\
822: F606W-SE & -1.0 $\pm$0.4 & 3.9 $\pm$0.1 & 1.3 $\pm$0.2 \\
823: F814W-NW & 2.4 $\pm$0.5 & 3.7 $\pm$0.3 & 0.4 $\pm$0.3 \\
824: F814W-SE & -0.7 $\pm$0.6 & 3.2$^{1}$ $\pm$ 0.4 & 0.7$^{2}$ $\pm$0.3 \\
825: \tableline
826: \end{tabular}
827: \end{center}
828: \vspace{-0.3in}
829: \tablecomments{Power law fits to the surface brightness along the south-east (Left) and 
830: north-west (Right) regions of the semi-major axis in the F606W filter.  The radial 
831: profiles were determined from the median of 10 pixel-wide (0$\farcs$25) cuts across 
832: the disk major axis, which were subsequently combined in 4 pixel (0$\farcs$10) radial 
833: bins.  Radial profile cuts at a disk position angle of 132$^{\circ}$, 
834: measured north to east, sampled the southeast (SE) disk quadrant while cuts at a disk position angle of 
835: 312$^{\circ}$ sampled the northwest (NW) disk quadrant.  Fit regions which were truncated from that 
836: listed in columns 2-4 include: $^{1}$ 3$\farcs$1 - 3$\farcs$7, 4$\farcs$4 - 4$\farcs$5; $^{2}$ 
837: 4$\farcs$6 - 6$\farcs$1.}
838: \end{table}
839: 
840: 
841: \clearpage
842: \begin{table}
843: \begin{center}
844: \footnotesize
845: \caption{Surface Brightnesses \label{colors}}
846: \begin{tabular}{lccc}
847: \tableline
848: Filter & Disk Major Axis & Disk Major Axis & Background Sky S.B. \\
849:           & 3$\farcs$0 & 3$\farcs$5 &  9$\farcs$0 \\
850:           & (mag arcsec$^{-2}$) & (mag arcsec$^{-2}$) & (mag arcsec$^{-2}$) \\
851: \tableline
852: ACS F435W & 19.6$\pm$0.1 &  20.1$\pm$0.2 &  22.3 \\
853: ACS F606W & 18.6$\pm$0.2 & 19.2$\pm$0.2 &  20.6 \\
854: ACS F814W & 19.1$\pm$0.2 &  19.4$\pm$0.2 &  20.8 \\
855: ACS (F606+F814)$_{ave}$ & 18.9$\pm$0.2 &  19.3$\pm$0.2 &  20.7 \\
856: ACS (F435+F606+F814)$_{ave}$ & 18.7$\pm$0.2 &  19.1$\pm$0.2 &  20.7 \\ 
857: STIS Clear & 20.0$\pm$0.1 &  20.3$\pm$0.2 &  21.8 \\
858: \tableline
859: \end{tabular}
860: \end{center}
861: \vspace{-0.3in}
862: \tablecomments{Disk surface brightnesses, calculated by median averaging 
863: 1$\farcs$0 x 1$\farcs$0 apertures placed at radial distances of 3$\farcs$0 and 
864: 3$\farcs$5 along the disk major axis (PA = 132$^{\circ}$, 312$^{\circ}$ 
865: measured north to east) are cited in the STMAG photometric reference frame.  
866: The background sky surface brightnesses were determined by averaging the median 
867: pixel values within two 1$\farcs$0 x 1$\farcs$0 apertures located 9$\farcs$0 from 
868: the central star.  As discussed in Section \ref{stis}, the 
869: ACS (F606W+F814W)$_{ave}$ and ACS (F435W+F606W+F814W)$_{ave}$ data 
870: correspond to weighted averages of single- and multi-epoch ACS data respectively 
871: which were combined to crudely approximate the STIS clear filter bandpass.}
872: \end{table}
873: 
874: 
875: 
876: \clearpage
877: \begin{table}
878: \begin{center}
879: \footnotesize
880: \caption{Johnson Filter Disk Major Axis Surface Brightnesses and Colors \label{johncolors}}
881: \begin{tabular}{lcc}
882: \tableline
883: Filter & 3$\farcs$0 & 3$\farcs$5  \\
884:           & (mag arcsec$^{-2}$) & (mag arcsec$^{-2}$)  \\
885: \tableline
886: B (ACS 2004)$^{1}$ & 20.2$\pm$0.1 & 20.7$\pm$0.2  \\
887: V (ACS 2003) & 18.5$\pm$0.2 & 19.2$\pm$0.2   \\
888: I (ACS 2003) & 17.8$\pm$0.2 & 18.1$\pm$0.2   \\
889: \tableline
890: V (STIS 1998) & 19.6$\pm$0.1 & 20.1$\pm$0.2  \\
891: \tableline
892: (V-I) & 0.7$\pm$0.3 & 1.1$\pm$0.3 \\
893: \tableline
894: \end{tabular}
895: \end{center}
896: \vspace{-0.3in}
897: \tablecomments{The disk surface brightnesses quoted in Table \ref{colors} 
898: were converted to the standard Johnson UBVRI system.  Transformations of the  ACS data were 
899: achieved using the iterative technique outlined by \citet{sir05}.  Note that all 
900: data characterized by a $^{1}$ reference were derived via multi-epoch ACS 
901: observations.}
902: \end{table}
903: 
904: 
905: \clearpage
906: \begin{figure}
907: \begin{center}
908: \includegraphics[width=10cm]{f1.jpg}
909: \caption{The PSF subtracted, 2 roll angle combined image of HD 163296 in the F435W
910: filter.  The image, which has a field of view of 20$\farcs$0 x 20$\farcs$0, has been plotted using 
911: a square root scale, smoothed with a 3 pixel Gaussian kernel.  Regions inside a radial distance of 2$\farcs$0 were dominated by PSF-subtraction residuals and were masked 
912: from the displayed data.  \label{combB}}
913: \end{center}
914: \end{figure}
915: 
916: 
917: \clearpage
918: \begin{figure}
919: \begin{center}
920: \includegraphics[width=10cm]{f2.jpg}
921: \caption{HD 163296 in the F606W filter, using the same parameters as in 
922: Figure \ref{combB}.  \label{combV}}
923: \end{center}
924: \end{figure}
925: 
926: 
927: \clearpage
928: \begin{figure}
929: \begin{center}
930: \includegraphics[width=10cm]{f3.jpg}
931: \caption{HD 163296 in the F814W filter, using the same parameters as in
932: Figure \ref{combB}.  \label{combI}}
933: \end{center}
934: \end{figure}
935: 
936: 
937: \clearpage
938: \begin{figure}
939: \begin{center}
940: \includegraphics[width=12cm]{f4.jpg}
941: \caption{Ten pixel wide (0$\farcs$25) median radial surface brightness cuts along the semi-major 
942: axis (PA = 132$^{\circ}$ and 312$^{\circ}$) of the HD 163296 scattered light disk.  The 
943: data were subsequently combined into 4 pixel (0$\farcs$10) wide bins.  The colored horizontal 
944: lines denote the background sky surface brightness in each filter. \label{azradprofile}}
945: \end{center}
946: \end{figure}
947: 
948: 
949: \clearpage
950: \begin{figure}
951: \begin{center}
952: \includegraphics[width=7.5cm]{f5a.jpg}
953: \includegraphics[width=7.5cm]{f5b.jpg}
954: \caption{Power law fits to the surface brightness along the south-east (Left) and 
955: north-west (Right) regions of the semi-major axis in the F606W filter.  The radial 
956: profiles were determined from the median of 10 pixel-wide (0$\farcs$25) cuts across 
957: the disk major axis, which were subsequently combined in 4 pixel (0$\farcs$10) radial 
958: bins.  Three distinct fitting regions, 2$\farcs$2-2$\farcs$8, 2$\farcs$9-4$\farcs$4, and 
959: 4$\farcs$5-6$\farcs$1, denoted in this figure by different color schemes, 
960: yielded the power law parameters listed in Table 
961: \ref{tabmajor}.  A similar analysis was performed for the F435W and F814W data. \label{allfit}}
962: \end{center}
963: \end{figure}
964: 
965: 
966: \clearpage
967: \begin{figure}
968: \begin{center}
969: \includegraphics[width=7cm]{f6.jpg}
970: \caption{A 14$\farcs$5 x 14$\farcs$5 (V-I) color image of the HD 163296 scattered 
971: light disk plotted on a linear scale, smoothed with a 8 pixel Gaussian kernel.  
972: Regions inside a radial distance of 2$\farcs$0 are masked, and the large white circle 
973: corresponds to a radial distance of 4$\farcs$0.  The color of the scattered light disk is 
974: similar to that of the 3 H-H knots visible, and exhibits no significant evidence of 
975: systematic spatial variations at the SNR level of our data. \label{colorfig}}
976: \end{center}
977: \end{figure}
978: 
979: 
980: \clearpage
981: \begin{figure}
982: \begin{center}
983: \includegraphics[width=12cm]{f7.jpg}
984: \caption{1998 epoch HST STIS clear filter (upper left), 2003 epoch HST ACS F606W (upper right), 2003 epoch HST ACS F814W (lower left), and 2004 epoch HST ACS F435W (lower right) coronagraphic 
985: observations of HD 163296.  The field of view for all images is 12$\farcs$0 x 12$\farcs$0; all images are 
986: plotted on a square root stretch scale smoothed by a second order Gaussian function, except for the 
987: STIS image which is smoothed with a first order Gaussian.  Regions inside a radial distance of 2$\farcs$0 are masked, and the large black circle corresponds to a radial distance of 4$\farcs$0.    \label{alldata}}
988: \end{center}
989: \end{figure}
990: 
991: 
992: \clearpage
993: \begin{figure}
994: \begin{center}
995: \includegraphics[width=10cm]{f8.jpg}
996: \caption{\textbf{Bottom Panels} The median average of ten pixel wide 
997: (0$\farcs$25) radial profile cuts, subsequently averaged in bins 3 radial 
998: pixel wide, were calculated at position angles of 140$^{\circ}$ and 320$^{\circ}$,
999:  measured N to E.  These cuts clearly depict the behavior of the spiral arm structure 
1000: observed in the SE disk quadrant of the F435W data (at a radial distance of 
1001: $\sim$3$\farcs$0-3$\farcs$5).  The dashed horizontal lines represent the limiting surface 
1002: brightness magnitudes for each bandpass, while the vertical lines depict the 
1003: three regions characterized by distinctive power laws in Table \ref{tabmajor}.  
1004: \textbf{Top Panels}.  The 
1005: median average of ten pixel wide radial profile cuts, subsequently average in 
1006: 3 (radial) pixel bins, calculated at position angles of 130$^{\circ}$, 140$^{\circ}$,
1007: 310$^{\circ}$, and 320$^{\circ}$ for the F435W data.  \label{radprof1}}
1008: \end{center}
1009: \end{figure}
1010: 
1011: 
1012: 
1013: 
1014: \clearpage
1015: \begin{figure}
1016: \begin{center}
1017: \includegraphics[width=10cm]{f9.jpg}
1018: \caption{\textbf{Bottom Panels} The median average of 0$\farcs$25 arcsec wide radial profile 
1019: cuts are shown for all datasets, calculated at position angles of 140$^{\circ}$ and 
1020: 320$^{\circ}$, and subsequently averaged in bins 3 radial pixels wide.  Note the epochs 
1021: characterized by low overall surface brightness profiles (STIS - pink, ACS F435W - blue) 
1022: are also the epochs in which spiral arm structure(s) are observed in the disk.  The dashed horizontal 
1023: lines represent the limiting surface 
1024: brightness magnitudes for each bandpass, while the vertical lines depict the 
1025: three regions characterized by distinctive power laws in Table \ref{tabmajor}. 
1026: \textbf{Top Panels}.  The median average of 0$\farcs$25 arcsec wide radial profile cuts 
1027: are shown for all epochs which exhibit spiral arm structure(s).  \label{radprof2}}
1028: \end{center}
1029: \end{figure}
1030: 
1031: 
1032: 
1033: 
1034: \end{document}
1035: