1: \documentclass[12pt,preprint]{aastex}
2: \usepackage{epsfig}
3: \usepackage{emulateapj5}
4: \begin{document}
5:
6:
7: \title{A Young Planet Search in Visible and IR Light: DN Tau, V836 Tau, and V827 Tau$^1$}
8:
9:
10: \author{L. Prato\altaffilmark{2,3}, M. Huerta\altaffilmark{4},
11: C. M. Johns--Krull\altaffilmark{3,5}, N. Mahmud\altaffilmark{3,5},
12: D. T. Jaffe\altaffilmark{6}, P. Hartigan\altaffilmark{5}}
13:
14:
15: \altaffiltext{1}{This paper includes data taken at The McDonald Observatory
16: of The University of Texas at Austin.}
17: \altaffiltext{2}{Lowell Observatory, 1400 West Mars Hill
18: Rd. Flagstaff, AZ 86001; lprato@lowell.edu}
19: \altaffiltext{3}{Visiting Astronomer at the Infrared Telescope Facility,
20: which is operated by the University of Hawaii under cooperative agreement
21: NCC 5-538 with the National Aeronautics and Space Administration, Office of
22: Space Science, Planetary Astronomy Program.}
23: \altaffiltext{4}{American Astronomical Society, 2000 Florida Avenue Suite 400,
24: Washington, DC 20009; marcos.huerta@aas.org}
25: \altaffiltext{5}{Department of Physics and Astronomy, Rice University,
26: MS-108, 6100 Main Street, Houston, TX 77005; cmj@rice.edu, naved@rice.edu, hartigan@rice.edu}
27: \altaffiltext{6}{Department of Astronomy, University of Texas, R. L. Moore
28: Hall, Austin, TX 78712; dtj@astro.as.utexas.edu}
29:
30: \begin{abstract}
31:
32: In searches for low-mass companions to late-type stars,
33: correlation between radial velocity variations and
34: line bisector slope changes indicates contamination
35: by large starspots. Two young stars
36: demonstrate that this test is not sufficient to rule
37: out starspots as a cause of radial velocity variations.
38: As part of our survey for
39: substellar companions to T Tauri stars, we identified the
40: $\sim$2 Myr old planet host candidates DN Tau
41: and V836 Tau. In both cases, visible light radial velocity modulation
42: appears periodic and is uncorrelated with line bisector span variations,
43: suggesting close companions of several M$_{Jup}$ in these
44: systems. However, high-resolution, infrared spectroscopy
45: shows that starspots cause the radial velocity
46: variations. We also report unambiguous results for V827 Tau,
47: identified as a spotted star on the basis of both visible light
48: and infrared spectroscopy. Our results suggest that infrared follow up
49: observations are critical for determining the source of radial
50: velocity modulation in young, spotted stars.
51:
52: \end{abstract}
53:
54: \keywords{planetary systems: formation -- stars: individual (DN Tau,
55: V836 Tau, V827 Tau) -- stars: spots -- techniques: radial velocities}
56:
57:
58: \section{Introduction}
59:
60: Extrasolar planets are common; over 300 systems have
61: been discovered (e.g., exoplanet.eu).
62: Recent studies have targeted
63: higher mass \citep{sat07, joh07}, lower mass \citep{but06, end06},
64: and younger objects \citep{pau06, set07}.
65: Identifying young planets is important to define the time scale for
66: planet formation and thus distinguish the possible formation process(es).
67:
68: Young stars still surrounded by the
69: circumstellar material which forms planets
70: are typically located at distances of $>$100~pc and are thus
71: inherently faint and often obscured.
72: They also manifest strong magnetic activity \citep[e.g.,][]{cmj07}
73: and are highly spotted. Numerous, large spots complicate detection of
74: extrasolar planets through radial velocity (RV) monitoring \citep{saa97}
75: because a spot that is partially visible at all times on the
76: surface of an inclined star mimics RV modulation
77: \citep[e.g.,][]{bou07, hue08}.
78:
79: \citet{pau06} studied 12$-$300~Myr old nearby
80: stars and found no evidence for planets
81: with masses $>$1$-$2~M$_{Jup}$ at the 3~$\sigma$ level.
82: \citet{set07} identified a minimum mass 6.1~M$_{Jup}$ planet in
83: a 852~day period orbit around the 100~Myr old G1$-$G1.5~V star HD~70573.
84: More recently, \citet{set08} reported a $\sim$10~M$_{Jup}$ planet
85: in a 3.56 day orbit around the 10~Myr old star TW Hya, although
86: \citet{huel08} identify this result as attributable to spots.
87:
88: In this letter we present our observations of the young stars
89: DN Tau, V836 Tau, and V827 Tau. While the V827 Tau visible light data
90: clearly implicate spots as the cause of the apparent RV
91: variability, corresponding data for DN Tau and V836 Tau suggest the
92: presence of giant planets. Our infrared (IR) observations show
93: that spots cause the RV variations seen in all
94: three stars. In \S 2 we describe the
95: observations, in \S 3 present our data analysis and
96: the evidence for spots, and in \S 4 provide a brief discussion.
97: We summarize in \S 5.
98:
99: \section{Observations and Data Reduction}
100:
101: \subsection{Visible Light Spectroscopy}
102:
103: Visible light spectra of DN Tau (M0), V836 Tau (K7), and V827 Tau (K7)
104: were taken at the McDonald Obervatory 2.7~meter
105: Harlan J. Smith telescope, between November, 2004 and January, 2008,
106: with the Coud\'e echelle spectrograph
107: \citep{tul95}. A 1.2$''$ slit yielded R$\sim$60,000.
108: Integration times were $\sim$1800~s; average seeing
109: was $\sim$2$''$. ThAr exposures taken immediately before and after
110: each spectrum provided wavelength calibration; typical
111: RMS values for the dispersion solution precision were $\sim$4~m~s$^{-1}$.
112: RV standards \citep{nid02, but96, cum99} were observed on
113: every night of every run; their overall RMS scatter
114: is 140~m~s$^{-1}$. We obtained 43 spectra of DN Tau,
115: 21 of V836 Tau, and 20 of
116: V827 Tau and applied standard IRAF reduction routines.
117: Details are given in \citet{hue07} and \citet{hue08}.
118:
119:
120: \subsection{Infrared Spectroscopy}
121:
122: We observed DN Tau, V836 Tau, V827 Tau, and the RV
123: standards HD 65277 (K5) and
124: GJ~281 (M0) on UT 2008 February 13$-$20 with CSHELL \citep{tok90, gre93},
125: the high-resolution, IR spectrograph at the NASA IRTF 3-m telescope.
126: The seeing was 0.4$-$0.8$''$. Each object was observed on 6$-$8 nights.
127: The 0.5$''$ slit yielded R$\sim$46,000. We obtained data
128: in 10$''$ nodded pairs. Spectra were centered at 2.298~$\mu$m (vacuum).
129: Integration times were $\sim$1 hour for the T~Tauri stars and $\sim$8 minutes
130: for the standards. The signal to noise ratio (SNR) was $\sim$70$-$120.
131: Data were reduced as described in \citet{joh99}.
132:
133:
134: \section{Analysis}
135:
136: \subsection{Radial Velocities: Visible Light}
137:
138: Relative RVs were determined by cross correlating
139: a high SNR, fiducial spectrum against
140: all other spectra for the same target. We used six orders
141: spanning $\sim$5700~\AA\ to 6800~\AA. Uncertainties
142: were estimated from the standard deviation of the mean for the
143: 6 orders, added in quadrature with the 140~m~s$^{-1}$ uncertainty derived
144: from the RV standards \citep[\S 2;][]{hue08}. RVs were
145: corrected for the Earth's barycentric motion.
146:
147: Optimum periods and uncertainties for phasing the RV data were selected based
148: on power spectra \citep{hue08}. For DN Tau we found
149: P$=$6.33~$\pm$0.20 days and a false alarm probability (FAP) of $< $0.001, for
150: V836 Tau, P$=$2.48~$\pm$0.49 days and FAP$=$0.10, and for
151: V827 Tau, P$=$3.76~$\pm$0.06 days and FAP$<$0.001.
152: We also checked for periodicity using the discrete Fourier transform
153: plus CLEAN method of Roberts et al. (1987). The strongest power
154: spectrum peaks for DN Tau and V836 Tau occur at the
155: same periods. The CLEAN method recovered a best
156: period of 3.61 d for V827 Tau, within $\sim$2~$\sigma$
157: of the above estimate. The phased RV data
158: are shown in Figures 1$-$3.
159:
160: \subsection{Line Bisector Spans}
161:
162: The presence of a starspot will distort the
163: line profile at the RV that corresponds to the stellar velocity at the
164: location of the spot. This distortion is
165: in proportion to the ratio of quiescent photosphere surface
166: brightness and surface brightness within the spot, at the observing
167: wavelength, and
168: the fraction of stellar surface covered by the spot \citep[e.g.,][]{que01}.
169: Thus, asymmetries in the line profiles originating from spots will be
170: present for all lines in the spectrum of a young star and are typically
171: correlated with the RV measured from the same spectrum. These
172: asymmetries have become the
173: standard criterion for rejecting starspots as the cause of false RV signals
174: \citep[e.g.,][]{que01, bou07, hue08, set08}.
175: For each of the six orders used to determine
176: the RVs, we cross-correlated all absorption lines and measured the
177: cross-correlation function (CCF) for that order. The average of
178: these six CCFs was used to measure the bisector spans
179: (lower panels of Figures 1$-$3). The linear correlation coefficient
180: and associated FAP \citep{bev92} is listed in the captions.
181: As expected for a spotted star, a clear correlation between bisector
182: span and RV is observed for V827 Tau. DN Tau and V836 Tau show no
183: correlation, suggesting that the variability is not the result
184: of spots.
185:
186:
187: \subsection{Infrared Radial Velocity Signatures}
188:
189: The contrast between a 4000~K photosphere and a 3000~K spot
190: \citep{bou89} is greater in visible than in IR light
191: because flux scales as a steeper function of temperature at
192: wavelengths shorter than the black body peak
193: \citep[e.g.,][]{car01}. Given
194: the decreased spot to photosphere contrast in the near-IR,
195: the amplitude of the RV modulation will be smaller.
196: Conversely, if a planet drives the RV modulation, the
197: amplitude should be the same in visible and IR light.
198:
199: \citet{bla07, bla08} used near-IR observations to search for companions to
200: low-mass objects, exploiting telluric absorption lines for
201: high-precision RV measurements. Figure 4
202: shows an example of a GJ~281 K-band spectrum and illustrates our
203: similar approach. We created models by combining high
204: resolution telluric absorption \citep{liv91} and cool stellar spectra
205: (the sunspot atlas of Wallace \& Livingston 1992), applying a range
206: of velocity shifts relative to the telluric lines.
207: Other free parameters are $v$sin$i$,
208: a Gaussian FWHM for the spectrometer line spread function,
209: scale factors for line depths,
210: and a first order continuum normalization function.
211: We employed the Marquardt method for
212: non-linear least squares fitting \citep{bev92} of
213: each model to an observed spectrum. The difference
214: between the stellar and telluric velocities in the best fit
215: model spectrum yields the RV,
216: which was then corrected for barycentric motion.
217:
218: Figures 1$-$3 show the IR-derived relative RVs. The
219: data are phased to the periods
220: given in \S3.1. The RV standard deviation in the IR data
221: for HD~65277 is 127~m~s$^{-1}$ and
222: for GJ~281 is 98~m~s$^{-1}$. Internal errors, measured from
223: the least squares fitting, are $\sim$40~m~s$^{-1}$ for both
224: standards. For the young stars, internal errors were
225: 100$-$300~m~s$^{-1}$, depending on the SNR achieved. Random errors,
226: which we assume add in quadrature with our internal errors to give
227: the overall scatter in velocities, are 120~m~s$^{-1}$ for HD~65277 and
228: 90~m~s$^{-1}$ for GJ~281. We use 110~m~s$^{-1}$ as our final value
229: for the random errors. The uncertainties shown for the IR data in
230: Figures 1$-$3 represent the sum, in quadrature, of the individual
231: internal error and the 110~m~s$^{-1}$ random error. Within our
232: measurement precision, we are unable to detect any IR RV variability in
233: DN Tau and V836 Tau. V827 Tau shows significant IR RV
234: variations but at a reduced amplitude from those observed in
235: visible light.
236:
237:
238: \section{Discussion}
239:
240: Figures 1 and 2 show the visible light RV modulation for DN Tau
241: and V836 Tau, with full amplitudes of $\sim 1500$ and $\sim 2700$~m~s$^{-1}$,
242: respectively. Within the 1~$\sigma$ uncertainties, all but one point in the IR RVs
243: of DN Tau are consistent with zero; all six IR RVs of V836 Tau also show
244: no variation. These results indicate that no planets
245: are present around DN Tau or V836 Tau with masses greater than a few
246: M$_{Jup}$ at $<$0.5~AU or $\sim$10~M$_{Jup}$ at $\sim$1~AU, despite the
247: absence of a correlation between the visible light RVs and bisector spans.
248: \citet{mat89} identify V836 Tau and V827 Tau as RV variables with
249: peak-to-peak amplitudes of 7$-$8~km~s$^{-1}$. Apparently the density and
250: size of spots on V836 Tau vary; historical data may
251: serve as an additional criterion for heavily spotted young stars.
252:
253: The primary conclusion from our visible light data is that
254: the lack of a correlation between the line bisector span
255: and the RV is not proof of a reflex motion companion. In addition, the
256: RV period can change significantly with new
257: data. Initial analysis of 20 visible light RVs for DN Tau,
258: taken over 2.5 years, indicated convincing modulation with P$=$7.5 days and
259: FAP$=$0.002. Although not markedly different from rotation period
260: estimates of 6.0$-$6.4 days \citep{bou93, per06}, $\chi^2$ for P$=$7.5 days
261: was more favorable than that for a secondary
262: peak at 6.3 days. Removing the fit
263: to either the 7.5 or 6.3 day period from the RV data and recalculating
264: the power spectrum yielded no significant peak, suggesting that the
265: data were best represented with only one period. With the
266: addition of RVs measured in winter 2007$-$2008, the integrity of the
267: 7.5 day period was diminished and the 6.3 day period came to
268: dominate the power spectrum.
269: The rotation period of V836 Tau has been stable for decades at 6.76 days
270: \citep{gra08}. We find a different RV period, 2.48 days, and
271: substantially reduced power in the RV modulation near 6.76 days.
272: The RV period for V827 Tau, 3.76 days, is indistinguishable
273: from previously determined values of the rotation period \citep{gra08}.
274:
275: If the visible light RV modulation originates in spots,
276: why don't the bisector spans correlate (Figures 1 and 2)? Figure 3
277: clearly shows correlation for V827 Tau; many other stars
278: show a correlation as well \citep[e.g.,][]{que01, bou07, hue08}.
279: We do not believe that this can be attributed to a
280: stronger impact of spots on particular spectral lines because the same
281: echelle orders were used to determine the bisector spans for all targets.
282: \citet{des07} show that when $v$sin$i$ is smaller than the spectrometer
283: resolution, RV and bisector variations originating in spots can mimic the
284: behavior expected from short period giant planets. The $v$sin$i$ of
285: DN Tau and V836 Tau, $\sim$10~km~s$^{-1}$,
286: are not much larger than our visible light spectral resolution, 5~km~s$^{-1}$,
287: while V827 Tau and LkCa 19 (Huerta et al. 2008) have $v$sin$i$ values of
288: $\sim$20~km~s$^{-1}$, suggesting qualitative agreement
289: with the simulations of \citet{des07}.
290:
291:
292: \section{Summary}
293:
294: We have measured the RVs of the $\sim$2~Myr old T~Tauri stars DN~Tau, V836~Tau,
295: and V827~Tau in visible and IR light; the variations
296: we see in all three systems are likely the result of
297: starspots. Furthermore: (1) Periodogram analysis can
298: reveal the presence of a period but not necessarily a reliable value
299: until RV measurements densely sample the full phase of the periodic
300: signal. (2) The lack of correlation of
301: line bisector spans with RVs for some stars with spots
302: is an important problem to address in the search
303: for planets around young stars. (3) High-resolution, IR spectroscopy is
304: critical for the verification of young planet candidates;
305: without the contrast in RV modulation amplitude
306: between the visible light and IR data, it is unclear whether
307: spots or a companion cause variability.
308: (4) In the case of the spotted young star, V827 Tau (Figure 3), we observe
309: a substantial reduction in RV amplitude between the visible
310: light and IR data. This result can be exploited to improve our
311: understanding of starspot temperature and filling factors.
312: (5) T~Tauri stars are virtually guaranteed to have spots at some
313: level; if planets are also present, their detection will likely
314: require disentanglement of the blended RV signals and will benefit greatly
315: from measurements that span a broad range in wavelength.
316:
317: \bigskip
318: \bigskip
319:
320:
321: The authors acknowledge the SIM Young Planets Key Project
322: (PI C. Beichman) for research support; funding was also
323: provided by NASA grant 05-SSO05-86.
324: We thank G. Blake, C. Salyk, and E. Schaller for sharing
325: Keck time (our first attempt at IR observations of DN Tau),
326: D. Gies and A. Fullerton for their assistance
327: implementing the CLEAN power spectrum estimate, and the
328: referee, J. Eisloeffel, for a prompt and helpful report.
329: LP thanks P. Bodenheimer and T. Barman for informative discussions.
330: This work made use of the SIMBAD database, the NASA
331: Astrophysics Data System, and the Two Micron All
332: Sky Survey, a joint project of the Univ. of Massachusetts
333: and IPAC/Caltech, funded by NASA and the NSF.
334: We recognize the significant cultural role that Mauna Kea
335: plays in the indigenous Hawaiian community and are
336: grateful for the opportunity to observe there.
337:
338:
339: \begin{thebibliography}{}
340:
341: \bibitem[Bevington \& Robinson(1992)]{bev92} Bevington, P.\ R. \& Robinson,
342: D.\ K. 1992, New York: McGraw-Hill, c1992, 2nd ed.
343:
344: \bibitem[Blake et al.(2008)]{bla08} Blake, C.~H., Charbonneau, D., White, R.~J.,
345: Torres, G., Marley, M.~S., \& Saumon, D.\ 2008, \apjl, 678, L125
346:
347: \bibitem[Blake et al.(2007)]{bla07} Blake, C.~H., Charbonneau, D., White, R.~J.,
348: Marley, M.~S., \& Saumon, D.\ 2007, \apj, 666, 1198
349:
350: \bibitem[Bouvier et al.(2007)]{bou07} Bouvier, J., et al.\ 2007, \aap, 463, 1017
351:
352: \bibitem[Bouvier et al.(1993)]{bou93} Bouvier, J., Cabrit, S., Fernandez, M.,
353: Martin, E.~L., \& Matthews, J.~M.\ 1993, \aap, 272, 176
354:
355: \bibitem[Bouvier \& Bertout(1989)]{bou89} Bouvier, J., \& Bertout, C.\ 1989, \aap, 211, 99
356:
357: \bibitem[Butler et al.(1996)]{but96} Butler, R. P., Marcy, G. W., Williams,
358: E., McCarthy, C., Dosanjh, P., \& Vogt, S. S. 1996, PASP, 108, 500
359:
360: \bibitem[Butler et al.(2006b)]{but06} Butler, R. P., Johnson, J. A., Marcy,
361: G. W., Wright, J. T., Vogt, S. S., \& Fischer, D. A. 2006, PASP, 118, 1685
362:
363: \bibitem[Carpenter et al.(2001)]{car01} Carpenter, J.~M.,
364: Hillenbrand, L.~A., \& Skrutskie, M.~F.\ 2001, \aj, 121, 3160
365:
366: %\bibitem[Catala et al.(2007)]{cat07} Catala, C., Donati,
367: %J.-F., Shkolnik, E., Bohlender, D., \& Alecian, E.\ 2007, \mnras, 374, L42
368:
369: \bibitem[Cumming et al.(1999)]{cum99} Cumming, A., Marcy, G. W., \& Butler, R. P.
370: 1999, ApJ, 526, 890
371:
372: \bibitem[Desort et al.(2007)]{des07} Desort, M., Lagrange, A.-M., Galland, F.,
373: Udry, S., \& Mayor, M.\ 2007, \aap, 473, 983
374:
375: \bibitem[Endl et al.(2006)]{end06} Endl, M., Cochran, W.~D.,
376: K{\"u}rster, M., Paulson, D.~B., Wittenmyer, R.~A., MacQueen, P.~J.,
377: \& Tull, R.~G.\ 2006, \apj, 649, 436
378:
379: \bibitem[Forrest et al.(2004)]{for04} Forrest, W. J., et al. 2004, ApJS, 154, 443
380:
381: \bibitem[Grankin et al.(2008)]{gra08} Grankin, K.~N., Bouvier, J., Herbst, W.,
382: \& Melnikov, S.~Y.\ 2008, \aap, 479, 827
383:
384: \bibitem[Greene et al.(1993)]{gre93} Greene, T.~P., Tokunaga,
385: A.~T., Toomey, D.~W., \& Carr, J.~B.\ 1993, \procspie, 1946, 313
386:
387: \bibitem[Huelamo et al.(2008)]{huel08} Huelamo, N., et al. 2008, A\&A, in press
388:
389: \bibitem[Huerta(2007)]{hue07} Huerta, M.\ 2007, Ph.D.~Thesis, Rice University
390:
391: \bibitem[Huerta et al.(2008)]{hue08} Huerta, M., Johns--Krull, C. M., Prato, L.,
392: Hartigan, P., \& Jaffe, D. T. 2008, ApJ, 678, 472
393:
394: \bibitem[Ireland \& Kraus(2008)]{ire08} Ireland, M.~J., \& Kraus, A.~L.\ 2008,
395: \apjl, 678, L59
396:
397: \bibitem[Johns--Krull(2007)]{cmj07} Johns--Krull, C. M. 2007, ApJ, 664, 975
398:
399: \bibitem[Johns--Krull et al.(1999)]{joh99} Johns--Krull, C.~M.,
400: Valenti, J.~A., \& Koresko, C.\ 1999, \apj, 516, 900
401:
402: \bibitem[Johnson et al.(2007a)]{joh07} Johnson, J. A., et al. 2007, ApJ, in press
403:
404: %\bibitem[Kurtz \& Mink(1998)]{kur98} Kurtz, M.~J., \& Mink, D.~J.\ 1998, \pasp, 110, 934
405:
406: %\bibitem[Lin et al.(1996)]{lin96} Lin, D.~N.~C., Bodenheimer,
407: %P., \& Richardson, D.~C.\ 1996, \nat, 380, 606
408:
409: \bibitem[Livingston
410: \& Wallace(1991)]{liv91} Livingston, W., \& Wallace, L.\ 1991, NSO Technical Report,
411: Tucson: National Solar Observatory, NOAO
412:
413: \bibitem[Mathieu et al.(1989)]{mat89} Mathieu, R.~D., Walter,
414: F.~M., \& Myers, P.~C.\ 1989, \aj, 98, 987
415:
416: %\bibitem[Mekkaden(1998)]{mek98} Mekkaden, M.~V.\ 1998, \aap, 340, 135
417:
418: \bibitem[Najita et al.(2007)]{naj07} Najita, J. R., Strom, S. E., \& Muzerolle, J.
419: 2007, MNRAS, 378, 369
420:
421: \bibitem[Nidever et al.(2002)]{nid02} Nidever, D. L., Marcy, G. W., Butler, R. P.,
422: Fischer, D. A., \& Vogt, S. S. 2002, ApJS, 141, 503
423:
424: \bibitem[Paulson et al.(2006)]{pau06} Paulson, D. B., \& Yelda, S. 2006, PASP,
425: 118, 706
426:
427: \bibitem[Percy et al.(2006)]{per06} Percy, J.~R., Gryc,
428: W.~K., Wong, J.~C.-Y., \& Herbst, W.\ 2006, \pasp, 118, 1390
429:
430: %\bibitem[Qi et al.(2004)]{qi04} Qi, C., et al.\ 2004, \apjl, 616, L11
431:
432: \bibitem[Queloz et al.(2001)]{que01} Queloz, D., et al.\ 2001, \aap, 379, 279
433:
434: \bibitem[Roberts et al.(1987)]{rob87} Roberts, D.~H., Lehar,
435: J., \& Dreher, J.~W.\ 1987, \aj, 93, 968
436:
437: \bibitem[Saar \& Donahue(1997)]{saa97} Saar, S.~H., \& Donahue, R.~A.\ 1997, \apj, 485, 319
438:
439: \bibitem[Sato et al.(2007)]{sat07} Sato, B., et al. 2007, ApJ, 661, 527
440:
441: \bibitem[Setiawan et al.(2008)]{set08} Setiawan, J., Henning,
442: T., Launhardt, R., M{\"u}ller, A., Weise, P., K{\"u}rster, M.\ 2008, \nat, 451, 38
443:
444: \bibitem[Setiawan et al.(2007)]{set07} Setiawan, J., Weise, P., Henning, Th.,
445: Launhardt, R., M?ller, A., \& Rodmann, J. 2007, ApJ, 660, L145
446:
447: \bibitem[Tokunaga et al.(1990)]{tok90} Tokunaga, A.~T.,
448: Toomey, D.~W., Carr, J., Hall, D.~N.~B.,
449: \& Epps, H.~W.\ 1990, \procspie, 1235, 131
450:
451: %\bibitem[Torres et al.(2004)]{tor04} Torres, G., Konacki, M.,
452: %Sasselov, D.~D., \& Jha, S.\ 2004, \apj, 609, 1071
453:
454: \bibitem[Tull et al.(1995)]{tul95} Tull, R. G., MacQueen, P. J., Sneden, C., \&
455: Lambert, D. L. 1995, PASP, 107, 251
456:
457: \bibitem[Wallace
458: \& Livingston(1992)]{wal92} Wallace, L., \& Livingston, W.~C.\ 1992, NSO Technical
459: Report \#92-001; Tucson: National Solar Observatory
460:
461: \end{thebibliography}
462:
463:
464: \clearpage
465: \begin{figure}
466: \plotone{f1.eps}
467: \caption{{\it Top:} Relative RV versus phase for DN Tau.
468: Visible light (open squares) and IR (filled circles) data have been
469: phased to a priod of 6.33~$\pm$0.20 days.
470: {\it Bottom:} Line bisector spans as a function of RV for the visible
471: light data. The linear correlation coefficient
472: is $r = -0.27$ and the associated FAP, $f_p$, is 0.09.
473: \label{fig1}}
474: \end{figure}
475:
476:
477: \clearpage
478: \begin{figure}
479: \plotone{f2.eps}
480: \caption{Same as Figure 1 for V836 Tau, phased to a period of
481: 2.48~$\pm$0.49 days ($r = -0.30$, $f_p = 0.20$).
482: \label{fig2}}
483: \end{figure}
484:
485:
486: \clearpage
487: \begin{figure}
488: \plotone{f3.eps}
489: \caption{Same as Figure 1 for V827 Tau, phased to a period of
490: 3.76~$\pm$0.06 days ($r = -0.88$, $f_p = 2.8 \times 10^{-7}$).
491: \label{fig3}}
492: \end{figure}
493:
494:
495: \clearpage
496: \begin{figure}
497: \plotone{f4.eps}
498: \caption{{\it Middle:} UT 20 February 2008 CSHELL spectrum (solid line) of the RV
499: standard GJ~281 and a model (dash-dot line) consisting of a stellar photosphere
500: component ({\it Top:} from the NSO sunspot atlas) and a
501: telluric component ({\it Bottom:} from the NOAO spectral atlas).
502: \label{fig4}}
503: \end{figure}
504:
505:
506:
507:
508: \end{document}
509: