1: \documentclass[12pt,preprint]{aastex}
2:
3:
4: \newcommand{\be}{\begin{enumerate}}
5: \newcommand{\ee}{\end{enumerate}}
6: \newcommand{\dg}{$^{\circ}$}
7: \newcommand{\um}{$\mu$m}
8: \newcommand{\rsun}{R_{\odot}}
9: \newcommand{\msun}{M_{\odot}}
10:
11:
12: \shorttitle{GCIRS 16SW: A Massive Eclipsing Binary}
13: \shortauthors{Peeples et al.}
14:
15: \title{The Nature of the Variable Galactic Center Source GCIRS~16SW
16: Revisited:\\ A Massive Eclipsing Binary}
17:
18:
19: \author{Molly~S.~Peeples\altaffilmark{1}, A.~Z.~Bonanos\altaffilmark{2},
20: D.~L.~DePoy\altaffilmark{1}, K.~Z.~Stanek\altaffilmark{1},
21: J.~Pepper\altaffilmark{1}, Richard~W.~Pogge\altaffilmark{1},
22: M.~H.~Pinsonneault\altaffilmark{1}, K.~Sellgren\altaffilmark{1} }
23:
24: \altaffiltext{1}{Department of Astronomy, Ohio State University, 140 W.\
25: 18th Ave., Columbus,~OH~43210} \altaffiltext{2}{Department of
26: Terrestrial Magnetism, Carnegie Institution of Washington, 5241 Broad
27: Branch Road, NW, Washington, DC 20015}
28: \email{molly@astronomy.ohio-state.edu}
29:
30: \begin{document}
31:
32:
33: \begin{abstract}
34: We present a re-analysis of our $H$- and $K$-band photometry and
35: light-curves for GCIRS~16SW, a regular periodic source near the Galactic
36: center. These data include those presented by \citet{depoy04}; we
37: correct a sign error in their reduction, finding GCIRS~16SW to be an
38: eclipsing binary with no color variations. We find the system to be an
39: equal mass overcontact binary (both stars overfilling their Roche lobes)
40: in a circular orbit with a period $P=19.4513$~days, an inclination angle
41: $i=71^{\circ}$. This confirms and strengthens the findings of
42: \citet{martins06} that GCIRS~16SW is an eclipsing binary composed of two
43: $\sim 50\msun$ stars, further supporting evidence of recent star
44: formation very close to the Galactic center. Finally, the calculated
45: luminosity of each component is close to the Eddington luminosity,
46: implying that the temperature of 24400~K given by \citet{najarro97}
47: might be overestimated for these evolved stars.
48: \end{abstract}
49:
50: \keywords{Galaxy: center --- stars: individual (GCIRS 16SW) --- stars:
51: binaries: eclipsing}
52:
53: \section{Introduction}\label{sec:intro}
54:
55: Standard star formation modes are thought to break down near a
56: supermassive black hole (SMBH), raising the question of whether or not
57: star formation near a SMBH is possible, and if so, through what
58: mechanism \citep{nayakshin05}. Our own Galaxy provides us with a unique
59: opportunity to study individual stars in the presence of a SMBH, namely,
60: Sgr~A*. Direct observations of massive, and therefore young, stars
61: close to Sgr~A* indicate that there has been recent star formation
62: at the Galactic center \citep{lebofsky82}.
63:
64: GCIRS~16SW (hereafter IRS16SW) is a variable source near the Galactic
65: center ($\alpha = 17^{\rm h}45^{\rm m}40\fs 1$, $\delta = -29^{\circ} 00'
66: 29''$, J2000.0). \citet{ott99} reported that the source is regularly
67: variable and suggested that it could be a binary star with very massive
68: components. \citet{depoy04} confirmed the period of the object, but
69: argued that the source was more likely a pulsating variable. Recently,
70: however, \citet{martins06} reported spectroscopic observations of
71: IRS16SW that showed radial velocity variations consistent with a
72: binary composed of two massive stars.
73:
74: Prompted by the convincing nature of the radial velocity variations seen
75: by \citet{martins06} we have re-analyzed the data presented by
76: \citet{depoy04} as well as additional data from the same observing
77: campaign. We find that the original data reduction process was seriously
78: flawed. In particular, the color variation, light-curve asymmetry, and
79: sign of the brightness variations that \citeauthor{depoy04}\ presented
80: are artifacts of the data reduction process.
81:
82: In this Letter, we report on the re-analysis of the
83: \citeauthor{depoy04}\ data. We find that there is no color change in
84: IRS16SW over its variations and that the shape of the light curve is
85: consistent with an eclipsing binary system. The new results are
86: consistent with \citet{martins06} and confirm that IRS16SW is a binary
87: composed of two massive stars. In \S\ref{sec:obs} we describe the
88: observations and present the data, in \S\ref{sec:analysis} we describe
89: the best-fit model to the light-curve, and in \S\ref{sec:conc} we
90: discuss and summarize the results.
91:
92:
93: \section{Observations}\label{sec:obs}
94: Observations of the Galactic center in the $H$ (1.6\um) and $K$ (2.2\um)
95: bands were made at the Cerro-Tololo Inter-American Observatory
96: (CTIO)/Yale 1-m telescope using the facility optical/infrared imager
97: (ANDICAM; see \citealt{depoy03} for details). ANDICAM has a pixel scale
98: of $0\farcs 22$~pix$^{-1}$ on a $1024\times 1024$ array. Both $H$- and
99: $K$-band images were taken in the 2001 and 2002 observing seasons;
100: $H$-band data were also obtained in 2000. (The \citealt{depoy04} analysis
101: includes only the 2001 data.) The observing campaign consists of every
102: usable night from UTC 2000 August 13 (HJD 2451769.5) through UTC 2000
103: October 14 (HJD 2451831.5), UTC 2001 May 20 (HJD 2452049.5) through
104: UTC 2001 November 3 (HJD 2452216.5), and UTC 2002 June 9 (HJD
105: 2452434.5) through UTC 2002 September 25 (HJD 2452542.5). Each night, a
106: set of seven slightly offset images were obtained and then combined and
107: trimmed to form a final nightly image. The $H$-band images consist of
108: 30~s exposures, and it took about four minutes to obtain the group of
109: seven images; the $K$-band images consist of 10~s exposures and took
110: about two minutes to obtain.
111:
112: The final $512 \times 512$ pixel images, corresponding to a field of
113: view of $112 \times 112$ arcseconds, are approximately centered on the
114: Galactic center. After image quality cuts were made, there are a total
115: of 144 $H$-band and 137 $K$-band images. The seeing ranges from
116: $0\farcs 93$ to $1\farcs 93$ full-width half maximum; in general, the
117: $H$-band images are of higher quality (typical seeing $\sim 1\farcs 3$)
118: than the $K$-band (with typical seeing $\sim 1\farcs 45$).
119:
120: Because the field is crowded, we reduced the data using the ISIS
121: difference image analysis package \citep{alard00,hartman04}. This
122: analysis revealed a sign error in the original \citet{depoy04}
123: reduction; IRS16SW is clearly an eclipsing binary. Because IRS16SW is
124: subject to significant blending---there are roughly half a dozen sources
125: in the \citet{ott99} catalog within about one arcsecond of IRS16SW---we
126: calibrated our light-curves with the \citet{ott99} data as presented by
127: \citet{martins06}. The \citet{martins06} $K$-band photometry gives a
128: mean magnitude of 0.2~mag higher than the \citet{ott99} mean, and
129: includes two more seasons of data.
130:
131: \citeauthor{ott99}\ found a variability amplitude of 0.55~mag; using
132: DAOPhot photometry to scale the ISIS fluxes, we find an amplitude of
133: $\sim 0.35$~mag in both $H$ and $K$ \citep[see][Appendix B]{hartman04}.
134: This substantial amplitude difference is indicative of significant
135: blending in our data. We used the period of 19.45~days reported by
136: \citet{depoy04} and \citet{martins06} to scale our $K$-band light-curve
137: to have the same mean magnitude and amplitude as the
138: \citeauthor{martins06}\ data. There were 110 nights for which data were
139: obtained in both the $H$- and $K$-bands, providing contemporaneous
140: measurements of the $H-K$ color. Using the DAOPhot photometry, we find
141: a constant $H-K$ color with an rms of 0.05~mag; this color does not vary
142: with time, phase, $H$-, or $K$-band magnitude, as shown in
143: Figure~\ref{fig:color}. Lacking properly calibrated $H$-band data, we
144: scaled the similarly blended $H$-band data to have the same amplitude
145: and mean magnitude as the $K$-band data. These scaled light-curves form
146: the basis of our analysis; Table~\ref{tbl:data} gives the final scaled
147: time-series photometry.
148:
149:
150:
151:
152: \section{Light Curve Analysis}\label{sec:analysis}
153: We simultaneously fit the 144 $H$-band and 184 $K$-band points
154: (including the \citeauthor{martins06}\ $K$-band points, except for two
155: noisy points at HJD 2498704 and 2499908) using the October 2005 version
156: of the Wilson-Devinney (WD) code \citep{Wilson71, Wilson79, wilson90} in
157: the overcontact mode (MODE 3). We fixed $T_{\rm eff1}=24400$~K as
158: estimated by \citet{najarro97} and used the square-root limb darkening
159: law, taking the values of the limb darkening coefficients from \citet
160: {claret00} for a LTE ATLAS9 \citep{kurucz93} stellar atmosphere model
161: with $T_{\rm eff}=24000$~K, $\log(g)=3.0$ (cgs), turbulent velocity of
162: 2~$\rm km \; s^{-1}$ and solar metallicity. We fixed gravity brightening
163: exponents and albedos to unity from theoretical values for stars at such
164: temperatures. We assumed equal masses, circular orbits, and synchronous
165: rotation, fitting for 7 parameters: the period $P$, time of primary
166: eclipse $T_{\rm prim}(\rm HJD)$, inclination $i$, $T_{\rm eff2}$, the
167: luminosity of the primary in each band ($L_{1H},L_{1K}$) and the surface
168: potential ($\Omega_{1}=\Omega_{2}$, \citealp[see][eq.~1]{Wilson79}).
169: We defined convergence to be when the corrections for all adjusted
170: parameters were smaller than their respective standard or statistical
171: errors after three consecutive iterations. The best fit parameters are
172: shown in Table~\ref{tbl:WD}. The ephemeris is
173: \begin{equation}
174: T_{\rm prim}=2451775.102 \pm 0.032 + 19.4513 \pm 0.0011 \times E\; (\rm HJD).
175: \end{equation}
176:
177:
178: A good fit required that the stars overfill their limiting Roche lobes,
179: which justifies using the overcontact mode of WD. \citet{martins06}
180: adopted the largest filling factor allowed by NIGHTFALL (1.3); we
181: calculate a larger fill-out factor (as defined by \citealt{mochnacki72})
182: of $F=1.44$. The critical surface potentials for the inner and outer
183: surfaces under the above assumptions are $\Omega_{\rm in}=3.75$ and
184: $\Omega_{\rm out}=3.21$. Assuming different values for the mass ratio
185: $q$ also produced good fits. However, a photometric mass ratio is not
186: well constrained by our photometry; therefore, we did not attempt to
187: solve for it. The inclination we derive is $i=71$\dg, in agreement with
188: \citet{martins06}; however, if there is still unaccounted for blending,
189: then the inclination angle could be larger.
190:
191: In Figures~\ref{fig:lch} and \ref{fig:lck} we show the $H$- and $K$-band
192: light curve model fit for IRS16SW. The \citet{martins06} data are
193: plotted with unfilled symbols in Figure~\ref{fig:lck}. Error bars for
194: our data are set to 0.04~mag, corresponding to the typical variation
195: seen for a constant star of similar magnitude (Peeples et al., in
196: preparation). The fact that the data are fit well under the assumption
197: of circular orbits is an indication that this assumption is sound; these
198: data give no evidence for an eccentric orbit. The eccentricity of
199: $e=0.09$ found by \citet{martins06} is derived from their radial
200: velocity curve, which tends to yield nonzero eccentricities
201: \citep{lucy71}. Furthermore, the circularization time for a system with
202: these physical characteristices (discussed below) is only tens of
203: thousands of years \citep{zahn75, zahn77}, making it unlikely that we
204: are observing IRS16SW pre-circularization.
205:
206: From the definition of $\Omega$, WD calculates the following best fit
207: fractional radii for both stars in units of the orbital separation: the
208: polar radius, $r_{\rm pole}=0.39$; the radius in the plane of revolution
209: and perpendicular to the line connecting the stars' centers, $r_{\rm
210: side}=0.41$, and the radius in the direction of L2, $r_{\rm
211: back}=0.47$. The orbital separation of \citet{martins06}, $(a_1+a_2)\sin
212: i = 132.8 \pm 4.4\,\rsun$, yields physical radii of $R_{\rm pole}=54.5
213: \pm 1.8\, \rsun$, $R_{\rm side}=58.2 \pm 1.9\, \rsun$, and $R_{\rm
214: back}=62.7 \pm 2.1\,\rsun$. The WD visualization for this system is
215: shown in Figure~\ref{fig:peanut}.
216:
217:
218: \section{Discussion and Conclusion}\label{sec:conc}
219:
220: Using Kepler's law, \citet{martins06} find $M_1 \approx M_2 \approx
221: 50\msun$, placing the components of IRS16SW among the most massive stars
222: known. Until recently, the most massive stars measured in binaries were
223: R136-38 in the Large Magellanic Cloud \citep[$57\;\msun$,][]{massey02}
224: and WR~22 ($55\;\msun$, \citealt{rauw96}, \citealt{schweickhardt99}), an
225: evolved star in our Galaxy. The current heavyweight champion is a
226: Wolf-Rayet binary, WR~20a (82 \& 83 $\msun$, \citealt{rauw04},
227: \citealt{bonanos04}), in the young Galactic cluster Westerlund~2.
228:
229: The luminosity of IRS16SW poses a problem. Using a radius of
230: $R=59.7\rsun$ (the mean radius given by WD for an orbital separation of
231: $140.6\rsun$) and an effective temperature of $T_{\rm eff}=24400$~K
232: \citep{najarro97}, we can estimate the luminosity $L=4\pi R^2 \sigma
233: T_{\rm eff}^4$ of each component as $4.4\times 10^{39}\,\rm erg \; s
234: ^{-1}$. The non-sphericity of IRS16SW will only drive this luminosity
235: higher. For comparison, the Eddington luminosity of a $50\msun$ star is
236: $L_{\rm edd} = 1.3 \times 10^{38} (M/\msun) = 6.5 \times 10^{39}\,\rm
237: erg \; s ^{-1}$. It is highly unlikely that each component of IRS16SW
238: has been radiating stably at nearly their Eddington luminosities for
239: eleven years \citep{humphreys94}; the combined photometry of
240: \citet{ott99} and this work span 1992--2002. Assuming the orbital
241: separation as calculated by \citet{martins06} is correct (if it is
242: smaller, then $L/L_{\rm edd}$ will be even larger), this calculation
243: implies that the temperature of \citet{najarro97} is an over-estimate.
244: A change in the assumed $T_{\rm eff}$ affects the WD model parameters;
245: specifically, a decrease in $T_{\rm eff}$ by a few thousand degrees
246: Kelvin will decrease the inclination angle $i$, and thus increase the
247: masses of the stars, by more than the formal $1\sigma$ uncertainties
248: given by WD.
249:
250: A radial velocity curve for the secondary is necessary to determine the
251: value of the mass ratio $q$. It remains a puzzle as to why IRS16SW does
252: not appear to be a double-line spectroscopic binary. It is readily
253: apparent from the depths of the eclipses that the two stars have
254: near-equal fluxes and from the depth ratio that they have near-equal
255: surface brightnesses, yet \citet{martins06} see only one set of
256: spectroscopic lines. However, since the spectral features used by
257: \citet{martins06} are wind lines, with strong characteristic P~Cygni
258: profiles \citep{najarro97}, differences in wind strength or small
259: differences in the effective temperatures of the stars could easily
260: conspire to make detection of the second set of lines difficult.
261:
262: We confirm that GCIRS~16SW is a massive eclipsing binary with both stars
263: overflowing their Roche lobes. We find a refined orbital period of
264: $19.4513 \pm 0.0011$~days and an inclination of 71\dg\ with an assumed
265: mass ratio of 1, supporting the findings by \citet{martins06} that the
266: masses of the two stars are both $\sim 50\msun$. The projected distance
267: between IRS16SW and Sgr~A* is 0.05~pc $\sim 11000$~AU (assuming a
268: distance to the Galactic center of 7.6~kpc, \citealt{eisenhauer05}); in
269: fact, IRS16SW is part of a moving group that is likely bound to Sgr~A*
270: \citep{lu05, paumard06}. As the lifetime of $50\msun$ stars is $\sim
271: 4$~Myr \citep{schaller92}, these observations are strong evidence that
272: IRS16SW was formed within 0.1~pc of Sgr~A* despite the tidal shear from
273: the black hole which creates problems in star formation models.
274:
275:
276:
277: \acknowledgements
278: We thank F.\ Martins for providing us with the \citet{ott99} $K$-band
279: light-curve, Slavek Rucinski, Andy Gould, and John Beacom for useful
280: discussions, and the anonymous referee for helpful suggestions.
281: A.~Z.~B. acknowledges research and travel support from the Carnegie
282: Institution of Washington through a Vera Rubin Fellowship. KS gratefully
283: acknowledges support from NSF grant AST-0206331.
284:
285:
286: %%% \bibliographystyle{apj} \bibliography{$HOME/latex/references}
287:
288: \begin{thebibliography}{28}
289: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
290:
291: \bibitem[{{Alard}(2000)}]{alard00}
292: {Alard}, C. 2000, \aaps, 144, 363
293:
294: \bibitem[{{Bonanos} {et~al.}(2004){Bonanos}, {Stanek}, {Udalski},
295: {Wyrzykowski}, {{\.Z}ebru{\'n}}, {Kubiak}, {Szyma{\'n}ski}, {Szewczyk},
296: {Pietrzy{\'n}ski}, \& {Soszy{\'n}ski}}]{bonanos04}
297: {Bonanos}, A.~Z., {Stanek}, K.~Z., {Udalski}, A., {Wyrzykowski}, L.,
298: {{\.Z}ebru{\'n}}, K., {Kubiak}, M., {Szyma{\'n}ski}, M.~K., {Szewczyk}, O.,
299: {Pietrzy{\'n}ski}, G., \& {Soszy{\'n}ski}, I. 2004, \apjl, 611, L33
300:
301: \bibitem[{{Claret}(2000)}]{claret00}
302: {Claret}, A. 2000, \aap, 363, 1081
303:
304: \bibitem[{{DePoy} {et~al.}(2003){DePoy}, {Atwood}, {Belville}, {Brewer},
305: {Byard}, {Gould}, {Mason}, {O'Brien}, {Pappalardo}, {Pogge}, {Steinbrecher},
306: \& {Teiga}}]{depoy03}
307: {DePoy}, D.~L., {Atwood}, B., {Belville}, S.~R., {Brewer}, D.~F., {Byard},
308: P.~L., {Gould}, A., {Mason}, J.~A., {O'Brien}, T.~P., {Pappalardo}, D.~P.,
309: {Pogge}, R.~W., {Steinbrecher}, D.~P., \& {Teiga}, E.~J. 2003, in Instrument
310: Design and Performance for Optical/Infrared Ground-based Telescopes. Edited
311: by Iye, Masanori; Moorwood, Alan F. M. Proceedings of the SPIE, Volume 4841,
312: pp. 827-838 (2003)., ed. M.~{Iye} \& A.~F.~M. {Moorwood}, 827--838
313:
314: \bibitem[{{DePoy} {et~al.}(2004){DePoy}, {Pepper}, {Pogge}, {Stutz},
315: {Pinsonneault}, \& {Sellgren}}]{depoy04}
316: {DePoy}, D.~L., {Pepper}, J., {Pogge}, R.~W., {Stutz}, A., {Pinsonneault}, M.,
317: \& {Sellgren}, K. 2004, \apj, 617, 1127
318:
319: \bibitem[{{Eisenhauer} {et~al.}(2005){Eisenhauer}, {Genzel}, {Alexander},
320: {Abuter}, {Paumard}, {Ott}, {Gilbert}, {Gillessen}, {Horrobin}, {Trippe},
321: {Bonnet}, {Dumas}, {Hubin}, {Kaufer}, {Kissler-Patig}, {Monnet},
322: {Str{\"o}bele}, {Szeifert}, {Eckart}, {Sch{\"o}del}, \&
323: {Zucker}}]{eisenhauer05}
324: {Eisenhauer}, F., {Genzel}, R., {Alexander}, T., {Abuter}, R., {Paumard}, T.,
325: {Ott}, T., {Gilbert}, A., {Gillessen}, S., {Horrobin}, M., {Trippe}, S.,
326: {Bonnet}, H., {Dumas}, C., {Hubin}, N., {Kaufer}, A., {Kissler-Patig}, M.,
327: {Monnet}, G., {Str{\"o}bele}, S., {Szeifert}, T., {Eckart}, A.,
328: {Sch{\"o}del}, R., \& {Zucker}, S. 2005, \apj, 628, 246
329:
330: \bibitem[{{Hartman} {et~al.}(2004){Hartman}, {Bakos}, {Stanek}, \&
331: {Noyes}}]{hartman04}
332: {Hartman}, J.~D., {Bakos}, G., {Stanek}, K.~Z., \& {Noyes}, R.~W. 2004, \aj,
333: 128, 1761
334:
335: \bibitem[{{Humphreys} \& {Davidson}(1994)}]{humphreys94}
336: {Humphreys}, R.~M. \& {Davidson}, K. 1994, \pasp, 106, 1025
337:
338: \bibitem[{{Kurucz}(1993)}]{kurucz93}
339: {Kurucz}, R. 1993, ATLAS9 Stellar Atmosphere Programs and 2 km/s grid.~Kurucz
340: CD-ROM No.~13.~ Cambridge, Mass.: Smithsonian Astrophysical Observatory,
341: 1993., 13
342:
343: \bibitem[{{Lebofsky} {et~al.}(1982){Lebofsky}, {Rieke}, \&
344: {Tokunaga}}]{lebofsky82}
345: {Lebofsky}, M.~J., {Rieke}, G.~H., \& {Tokunaga}, A.~T. 1982, \apj, 263, 736
346:
347: \bibitem[{{Lu} {et~al.}(2005){Lu}, {Ghez}, {Hornstein}, {Morris}, \&
348: {Becklin}}]{lu05}
349: {Lu}, J.~R., {Ghez}, A.~M., {Hornstein}, S.~D., {Morris}, M., \& {Becklin},
350: E.~E. 2005, \apjl, 625, L51
351:
352: \bibitem[{{Lucy} \& {Sweeney}(1971)}]{lucy71}
353: {Lucy}, L.~B. \& {Sweeney}, M.~A. 1971, \aj, 76, 544
354:
355: \bibitem[{{Martins} {et~al.}(2006){Martins}, {Trippe}, {Paumard}, {Ott},
356: {Genzel}, {Rauw}, {Eisenhauer}, {Gillessen}, {Maness}, \&
357: {Abuter}}]{martins06}
358: {Martins}, F., {Trippe}, S., {Paumard}, T., {Ott}, T., {Genzel}, R., {Rauw},
359: G., {Eisenhauer}, F., {Gillessen}, S., {Maness}, H., \& {Abuter}, R. 2006,
360: \apjl, 649, L103
361:
362: \bibitem[{{Massey} {et~al.}(2002){Massey}, {Penny}, \& {Vukovich}}]{massey02}
363: {Massey}, P., {Penny}, L.~R., \& {Vukovich}, J. 2002, \apj, 565, 982
364:
365: \bibitem[{{Mochnacki} \& {Doughty}(1972)}]{mochnacki72}
366: {Mochnacki}, S.~W. \& {Doughty}, N.~A. 1972, \mnras, 156, 51
367:
368: \bibitem[{{Najarro} {et~al.}(1997){Najarro}, {Krabbe}, {Genzel}, {Lutz},
369: {Kudritzki}, \& {Hillier}}]{najarro97}
370: {Najarro}, F., {Krabbe}, A., {Genzel}, R., {Lutz}, D., {Kudritzki}, R.~P., \&
371: {Hillier}, D.~J. 1997, \aap, 325, 700
372:
373: \bibitem[{{Nayakshin} \& {Sunyaev}(2005)}]{nayakshin05}
374: {Nayakshin}, S. \& {Sunyaev}, R. 2005, \mnras, 364, L23
375:
376: \bibitem[{{Ott} {et~al.}(1999){Ott}, {Eckart}, \& {Genzel}}]{ott99}
377: {Ott}, T., {Eckart}, A., \& {Genzel}, R. 1999, \apj, 523, 248
378:
379: \bibitem[{{Paumard} {et~al.}(2006){Paumard}, {Genzel}, {Martins}, {Nayakshin},
380: {Beloborodov}, {Levin}, {Trippe}, {Eisenhauer}, {Ott}, {Gillessen}, {Abuter},
381: {Cuadra}, {Alexander}, \& {Sternberg}}]{paumard06}
382: {Paumard}, T., {Genzel}, R., {Martins}, F., {Nayakshin}, S., {Beloborodov},
383: A.~M., {Levin}, Y., {Trippe}, S., {Eisenhauer}, F., {Ott}, T., {Gillessen},
384: S., {Abuter}, R., {Cuadra}, J., {Alexander}, T., \& {Sternberg}, A. 2006,
385: \apj, 643, 1011
386:
387: \bibitem[{{Rauw} {et~al.}(2004){Rauw}, {De Becker}, {Naz{\'e}}, {Crowther},
388: {Gosset}, {Sana}, {van der Hucht}, {Vreux}, \& {Williams}}]{rauw04}
389: {Rauw}, G., {De Becker}, M., {Naz{\'e}}, Y., {Crowther}, P.~A., {Gosset}, E.,
390: {Sana}, H., {van der Hucht}, K.~A., {Vreux}, J.-M., \& {Williams}, P.~M.
391: 2004, \aap, 420, L9
392:
393: \bibitem[{{Rauw} {et~al.}(1996){Rauw}, {Vreux}, {Gosset}, {Hutsemekers},
394: {Magain}, \& {Rochowicz}}]{rauw96}
395: {Rauw}, G., {Vreux}, J.-M., {Gosset}, E., {Hutsemekers}, D., {Magain}, P., \&
396: {Rochowicz}, K. 1996, \aap, 306, 771
397:
398: \bibitem[{{Schaller} {et~al.}(1992){Schaller}, {Schaerer}, {Meynet}, \&
399: {Maeder}}]{schaller92}
400: {Schaller}, G., {Schaerer}, D., {Meynet}, G., \& {Maeder}, A. 1992, \aaps, 96,
401: 269
402:
403: \bibitem[{{Schweickhardt} {et~al.}(1999){Schweickhardt}, {Schmutz}, {Stahl},
404: {Szeifert}, \& {Wolf}}]{schweickhardt99}
405: {Schweickhardt}, J., {Schmutz}, W., {Stahl}, O., {Szeifert}, T., \& {Wolf}, B.
406: 1999, \aap, 347, 127
407:
408: \bibitem[{{Wilson}(1979)}]{Wilson79}
409: {Wilson}, R.~E. 1979, \apj, 234, 1054
410:
411: \bibitem[{{Wilson}(1990)}]{wilson90}
412: ---. 1990, \apj, 356, 613
413:
414: \bibitem[{{Wilson} \& {Devinney}(1971)}]{Wilson71}
415: {Wilson}, R.~E. \& {Devinney}, E.~J. 1971, \apj, 166, 605
416:
417: \bibitem[{{Zahn}(1975)}]{zahn75}
418: {Zahn}, J.-P. 1975, \aap, 41, 329
419:
420: \bibitem[{{Zahn}(1977)}]{zahn77}
421: ---. 1977, \aap, 57, 383
422:
423: \end{thebibliography}
424:
425: \clearpage
426:
427: \begin{deluxetable}{crr}
428: \tablewidth{0pc}
429: \tablecaption{\sc $H$- and $K$-band Photometry of GCIRS~16SW
430: \label{tbl:data}
431: }
432: \tablehead{
433: \colhead{Band} &
434: \colhead{HJD} &
435: \colhead{Scaled} \\
436: \colhead{} &
437: \colhead{$-2450000.$} &
438: \colhead{Magnitude}}
439: \startdata
440: H & 1769.6620 & 9.630 \\
441: H & 1772.5868 & 9.730 \\
442: \dotfill & \dotfill & \dotfill \\
443: H & 2542.5417 & 10.041\\ \hline
444: K & 2048.9194 & 9.886 \\
445: K & 2051.7765 & 9.527 \\
446: \dotfill & \dotfill & \dotfill \\
447: \enddata
448:
449: \tablecomments{Both $H$- and $K$-bands are scaled to have the same
450: amplitude and mean magnitude as the \citet{ott99} $K$-band data
451: presented by \citet{martins06}. All errors are set to 0.04~mag,
452: corresponding to the typical variation seen for a constant star of
453: similar magnitude (Peeples et al., in preparation). Table
454: \ref{tbl:data} is published in its entirety in the electronic edition
455: of the Astrophysical Journal. A portion is shown here for guidance
456: regarding its form and content. }
457:
458: \end{deluxetable}
459:
460: \clearpage
461:
462: \begin{deluxetable}{lc}
463: \tablewidth{0pc}
464: \tablecaption{\sc Best-Fit Parameters From Combined $H$\&$K$ Light-Curve
465: Analysis With Wilson-Devinney Program\label{tbl:WD}
466: }
467: \tablehead{
468: \colhead{Parameter} & \colhead{Value}}
469: \startdata
470: Period, P & 1$9.4513 \pm 0.0011$~days \\
471: Time of primary eclipse, $T_{\rm prim}$ & $2451775.102 \pm 0.032$ \\
472: Inclination, $i$ & $70.85$\dg $\pm 0.6$\dg\\
473: Temperature ratio, $T_2 / T_1$ & $0.96$ \\
474: Surface potential, $\Omega$ & 3.51 \\
475: Light ratio in $H$, $L_{2}/L_{1}$ & 0.936\\
476: Light ratio in $K$, $L_{2}/L_{1}$ & 0.939\\
477: Radius, $r_{\rm pole,\, 1} = r_{\rm pole,\, 2}$& 0.39 \\
478: ............ $r_{\rm side,\, 1} = r_{\rm side,\, 2}$ & 0.41 \\
479: ............ $r_{\rm back,\, 1} = r_{\rm back,\, 2}$ & 0.47 \\
480: & \\
481: Secondary temperature, $T_2$ & 23500~K\\
482: Radius, $R_{\rm pole}$ & $54.5 \pm 1.8\, \rsun$ \\
483: ............ $R_{\rm side}$ & $58.2 \pm 1.9\,\rsun$ \\
484: ............ $R_{\rm back}$ & $62.7 \pm 2.1\,\rsun$
485: \enddata
486:
487: \tablecomments{First ten parameters are best-fit parameters from a
488: combined $H$\&$K$ light-curve analysis with the Wilson-Devinney (WD)
489: program. The $1\sigma$ uncertainties given by WD are unrealistically
490: small, and thus not listed. The radii $r_{\rm pole}$, $r_{\rm side}$,
491: and $r_{\rm back}$ are in units of the orbital separation.
492: The final four (physical) parameters are
493: based on the orbital separation, $(a_1 +a_2)\sin i = 140.6 \pm
494: 4.7\rsun$ and the assumed effective temperature of 24400~K for $T_1$
495: \citep{najarro97}. }
496:
497: \end{deluxetable}
498:
499: \clearpage
500:
501: \begin{figure}
502: \plotone{f1.eps}
503: \caption{\label{fig:color} $H-K$ color residuals for DAOPhot photometry
504: versus calibrated $K$-band magnitude. No clear trend between color and
505: magnitude is observed. The rms variation about a constant color is
506: 0.05~magnitudes. The errorbar on the left shows the typical
507: uncertainty in $H-K$ color for a constant star of similar magnitude.
508: See \S~\ref{sec:obs} for further discussion.}
509: \end{figure}
510:
511: \clearpage
512:
513: \begin{figure}
514: \plotone{f2.eps}
515: \caption{\label{fig:lch}Wilson-Devinney fit and residuals of an
516: overcontact binary to the $H$-band light-curve of IRS16SW. The period
517: is 19.4513~days; the model parameters have zero eccentricity and an
518: inclination of 71\dg. The rms variation about the model fit is
519: 0.06~magnitudes. }
520: \end{figure}
521:
522: \clearpage
523:
524: \begin{figure}
525: \plotone{f3.eps}
526: \caption{\label{fig:lck}Wilson-Devinney fit and residuals of an
527: overcontact binary to the $K$-band light-curve of IRS16SW. The filled
528: circles are the data presented here; the open squares are the
529: \citet{martins06} data. The period is 19.4513~days; the model
530: parameters have zero eccentricity and an inclination of 71\dg. The rms
531: variation about the model fit is 0.06~magnitudes.}
532: \end{figure}
533:
534: \clearpage
535:
536: \begin{figure}
537: \plotone{f4.eps}
538: \caption{\label{fig:peanut}Wilson-Devinney visualization of IRS16SW
539: at an orbital phase of 0.12. Axes are in units of $\rsun$.}
540: \end{figure}
541:
542: \end{document}
543: