1: %
2: %
3: %
4: %
5: %
6: %
7: %
8:
9: %
10: %
11: %
12:
13: %
14: %
15: %
16: %
17:
18: \documentclass{aastex}
19: \usepackage{natbib}
20: \bibliographystyle{apj}
21: \usepackage{emulateapj5}
22: %
23:
24: %
25:
26: %
27: %
28: %
29: %
30: %
31: %
32:
33: %
34: %
35: %
36: %
37: %
38: %
39: %
40: %
41:
42: \newcommand{\vdag}{(v)^\dagger}
43: \newcommand{\myemail}{suvrath@astro.ufl.edu, vaneyken@astro.ufl.edu, jge@astro.ufl.edu, dewitt@astro.ufl.edu, scfleming@astro.ufl.edu, cohenr@astro.ufl.edu, jcrepp@astro.ufl.edu}
44:
45: %
46:
47: %
48:
49: %
50: %
51: %
52: %
53: %
54: %
55:
56: \shorttitle{Measuring Stellar Radial Velocities with DFDI}
57: \shortauthors{Mahadevan et al.}
58:
59: %
60: %
61:
62: \begin{document}
63:
64: %
65: %
66: %
67:
68: \title{Measuring Stellar Radial Velocities with a Dispersed Fixed-Delay Interferometer }
69:
70: %
71: %
72: %
73: %
74: %
75: %
76:
77: \author{Suvrath Mahadevan\altaffilmark{1}, Julian van Eyken, Jian Ge, Curtis DeWitt, Scott W. Fleming, Roger Cohen, Justin Crepp \& Andrew Vanden Heuvel }
78: \affil{Astronomy Department, University of Florida, 211 Bryant
79: Space Science Center P.O. Box 112055 Gainesville, FL 32611-2055}
80: \email{suvrath@astro.ufl.edu}
81:
82:
83: %
84: %
85: %
86: %
87:
88: \altaffiltext{1}{Visiting Astronomer, Kitt Peak National
89: Observatory, National Optical Astronomy Observatory. KPNO is
90: operated by AURA, Inc.\ under contract to the National Science
91: Foundation.}
92: %
93: %
94: %
95: %
96: %
97:
98:
99: %
100: %
101: %
102: %
103: %
104:
105:
106: \begin{abstract}
107: We demonstrate the ability to measure precise stellar barycentric
108: radial velocities with the dispersed fixed-delay interferometer
109: technique using the Exoplanet Tracker (ET), an instrument
110: primarily designed for precision differential Doppler velocity
111: measurements using this technique. Our barycentric radial
112: velocities, derived from observations taken at the KPNO 2.1 meter
113: telescope, differ from those of Nidever et al. by 0.047 km
114: s$^{-1}$ (rms) when simultaneous iodine calibration is used, and
115: by 0.120 km s$^{-1}$ (rms) without simultaneous iodine
116: calibration. Our results effectively show that a Michelson
117: interferometer coupled to a spectrograph allows precise
118: measurements of barycentric radial velocities even at a modest
119: spectral resolution of R $\sim 5100$. A multi-object version of
120: the ET instrument capable of observing $\sim$500 stars per night
121: is being used at the Sloan 2.5 m telescope at Apache Point
122: Observatory for the Multi-object APO Radial Velocity Exoplanet Large-area
123: Survey (MARVELS), a wide-field radial velocity survey for
124: extrasolar planets around TYCHO-2 stars in the magnitude range
125: $7.6<V<12$. In addition to precise differential velocities, this
126: survey will also yield precise barycentric radial velocities for
127: many thousands of stars using the data analysis techniques
128: reported here. Such a large kinematic survey at high velocity
129: precision will be useful in identifying the signature of accretion
130: events in the Milky Way and understanding local stellar kinematics
131: in addition to discovering exoplanets, brown dwarfs and
132: spectroscopic binaries.
133: \end{abstract}
134:
135: %
136: %
137: %
138: %
139:
140: \keywords{ techniques: radial velocities --- techniques:
141: spectroscopic --- instrumentation: interferometers ---
142: instrumentation: spectrographs --- stars: kinematics --- methods:
143: data analysis}
144:
145: %
146: %
147: %
148: %
149: %
150: %
151: %
152: %
153:
154: \section{INTRODUCTION}
155: The discovery of various tidal streams in our galaxy (Ibata et al.
156: 1994; Yanny et al. 2006; Belokurov et al. 2006) challenges the
157: monolithic collapse formation model (Eggen 1962), but the streams
158: discovered so far form only a small fraction of the mass of the
159: galaxy. Large kinematic surveys are necessary to determine if such
160: merger and accretion events are more common. Radial velocity
161: surveys using multi-object spectrographs, like RAVE (Steinmetz et
162: al. 2006), are already producing barycentric radial velocities for
163: many thousands of stars at a precision of $2-3$ km s$^{-1}$,
164: enabling studies of galactic dynamics (e.g. Veltz et al. 2008). In
165: the northern hemisphere the magnitude limited Geneva-Copenhagen
166: Survey (Nordstrom et al. 2004) has provided radial velocities for
167: over 14000 bright nearby F and G dwarfs ($V < 8.6$) at a precision
168: of $100-300$ m s$^{-1}$. Coupled with gravity and chemical
169: composition derived from the observed spectra, such datasets of
170: precise barycentric radial velocities are very valuable in
171: identifying past accretion events and substructure in the galactic
172: disk (Helmi et al. 2006), finding members of moving groups (Jones
173: 1970) and determining cluster membership.
174:
175: The ability to determine precise differential radial velocities
176: has also improved dramatically in the last few years, and planet
177: search surveys are currently at a precision of $1-3$ m s$^{-1}$
178: (Rupprecht et al. 2004; Butler et al. 1996). Such planet search
179: surveys, however, use an arbitrary zero point for the stellar
180: velocities since only differential velocities are measured. The
181: true radial velocity of the center-of-mass of a star is a
182: difficult quantity to measure accurately using spectroscopy since
183: it involves additional modelling (or assumptions) of effects like
184: convective blue shift and gravitational redshift (Lindegren \&
185: Dravins 2003). A quantity that is easier to obtain, in terms of
186: observables and known physical constants, is the apparent Doppler
187: shift of the lines in the stellar spectra, as measured in the
188: solar system barycenter, with the Sun setting the velocity zero
189: point. This Doppler shift can be converted into a velocity that we
190: define as the barycentric radial velocity of the star. Very
191: precise barycentric radial velocities have been measured using
192: high resolution echelle spectrographs by Udry et al. 1999 (UD99
193: hereafter) for a number of stable F, G, K dwarfs, as well as by
194: Nidever et al. 2002 (ND02 hereafter) for a larger sample.
195:
196: In this paper we demonstrate the ability of a dispersed
197: fixed-delay interferometer (DFDI) to measure precise barycentric
198: radial velocities. The Exoplanet Tracker (ET) is a prototype of
199: this class of instruments, which use a Michelson interferometer
200: followed by a medium resolution spectrograph working in the first
201: grating order. Fixed delay interferometers have long been used by
202: solar astrophysicists to measure solar oscillations (Gorskii \&
203: Lebedev 1977; Kozhevatov 1983; Harvey et al. 1995). In 1997 D. J.
204: Erskine proposed adding a spectrograph as a post disperser for
205: increasing the measurement precision of differential radial
206: velocities (Erskine \& Ge 2000; Ge, Erskine \& Rushford 2000).
207: When using this technique sinusoidal interference fringes are
208: formed in the slit direction at the position of the stellar
209: absorption lines. A shift in the dispersion direction of the
210: underlying spectra due to a Doppler shift manifests itself as the
211: phase shift of the sinusoidal fringes in the slit direction (Ge
212: 2002).
213:
214: Precise radial velocities can be measured with DFDI even at low
215: wavelength dispersions since information is also being encoded as
216: sinusoidal fringes in the slit direction. This results in each
217: stellar spectrum taking up only a small fraction of the available
218: number of pixels on the CCD detector used to record the spectrum.
219: This feature of the DFDI technique is a great advantage since it
220: has the potential to enable a multi-object instrument (similar to
221: ET) to simultaneously obtain precise radial velocities for a large
222: number of stars (Ge 2002; Ge et al. 2002) without the need for an
223: expensive cross-dispersed high spectral resolution instrument. One
224: such instrument, the W. M. Keck Exoplanet Tracker, is being used
225: at the Sloan 2.5 m telescope for the magnitude limited
226: Multi-object APO Radial Velocity Exoplanet Large-area Survey (MARVELS), a
227: survey for extrasolar planets around TYCHO-2 stars in the
228: magnitude range $7.6<V<12$ that is being conducted as part of SDSS
229: III (Weinberg et al. 2007). This instrument can observe 59 stars
230: simultaneously, and approximately 500 stars per night. Designed
231: primarily as a wide-field precision radial velocity survey to
232: discover short and intermediate period planets, MARVELS will also
233: provide a dataset of precise barycentric radial velocities for
234: thousands of late F, G, and K TYCHO-2 stars.
235:
236: \section{Observations}
237: The observations presented in this article were conducted with the
238: single object ET instrument at the Kitt Peak 2.1 meter telescope
239: in January 2006 as part of an ongoing planet search program. The
240: ET instrument is linked to the telescope by a 200 $\mu$m optical
241: fiber, which is fed by the f/8 beam of the 2.1 meter telescope.
242: The instrument itself is housed in an isolated room in the
243: basement and consists of a Michelson interferometer in series with
244: a medium resolution spectrograph operating in the wavelength range
245: $5000-5640$ \AA , and with a spectral resolution of R $\sim 5100$.
246: An iodine cell can be inserted into the optical path and serves to
247: calibrate out instrument drifts. This instrument has successfully
248: demonstrated its short-term stability by confirming known planets
249: (e.g. van Eyken et al. 2004) and discovering a short period planet
250: around the star HD102195 (Ge et al. 2006).
251:
252: The observing procedure for the ET planet search program is
253: similar to other precision radial velocity programs that use an
254: iodine cell to calibrate out the instrument drifts (e.g. Butler et
255: al. 1996). For each star in the program a single exposure is
256: acquired without the iodine cell inserted in the beam path. An
257: exposure of the iodine cell, illuminated by a tungsten lamp, is
258: also acquired immediately before or after this stellar exposure.
259: These exposures are referred to as the star and iodine template
260: respectively. The iodine template is acquired at high signal to
261: noise (S/N), typically S/N = 150-200 per pixel. The
262: signal-to-noise on the star template depends on the brightness of
263: the star and the exposure time and is generally in the range of
264: S/N = 50-200 per pixel. All subsequent exposures are taken with
265: the iodine cell inserted in the path of the stellar beam. It is
266: from these exposures that precision radial velocities are obtained
267: since the velocity drift of the iodine component tracks the
268: instrument drift, allowing it be calibrated out. Exposure times
269: are similar to those used on the stellar template, typically
270: yielding a S/N $\sim 1/\sqrt{2}$ of that obtained on the stellar
271: template (since the iodine cell absorbs almost 50\% of the light).
272:
273: Stable stars and known planet-bearing stars are routinely observed
274: for calibration as part of the ET planet search program. During
275: the observing run we were able to obtain regular observations of
276: the RV stable stars $\eta$ Cas (small linear trend in velocity),
277: $\tau$ Ceti, and 36 Uma as well as the known planet bearing stars
278: 51 Peg (Mayor \& Queloz 1995), 55 Cnc (McArthur et al. 2004), and
279: HD4967 (Butler et al. 2003). For six stars, chosen because they
280: had rms velocity scatter less than 100 m s$^{-1}$ reported in ND02
281: (HD3674, HD3861, HD12414, HD105405, HD130087, HD134044), we also
282: acquired stellar templates and iodine templates immediately before
283: or after the respective stellar templates, although no
284: star-with-iodine data points were obtained for these six stars.
285: Typical exposure times varied from 3 minutes for the brightest
286: star ($\eta$ Cas, $V=3.45$) to 10 minutes for the faintest
287: (HD130087, $V=7.52$), leading to a photon noise limited radial
288: velocity error of $3-10$ m s$^{-1}$ for most observations.
289:
290: The advantage of an isolated, fiber-fed instrument is the
291: potential for high stability. The total instrument velocity drift
292: during the observing run (as tracked by the iodine templates) can
293: be seen in Figure 1. The instrument velocity drift is only 2.5 km
294: s$^{-1}$ over the entire 12 day observing run and less than 0.5 km
295: s$^{-1}$ over the first night. This short term stability is
296: essential for the data analysis technique described in Section
297: 3.1.
298:
299: %
300: %
301: %
302: %
303: %
304: %
305: %
306: %
307: %
308: %
309:
310: \section{Data Analysis Procedure}
311: The standard ET data analysis pipeline was designed to extract
312: differential radial velocity for the planet search program. It
313: does not give the barycentric radial velocity of the star since
314: the zero point itself is arbitrary. The pipeline processing steps
315: are described in more detail in van Eyken et al. 2004, and Ge et
316: al. 2006. Here we discuss the various processing steps briefly,
317: but we concentrate on the modifications that allow barycentric
318: radial velocities to be measured.
319:
320: The data produced by ET were processed using standard IRAF
321: procedures, as well as software written in Research System Inc.'s
322: IDL software. The images were corrected for biases, dark current,
323: and scattered light and then trimmed, illumination corrected, and
324: low-pass filtered. The visibilities ($V$) and the phases
325: ($\theta$) of the fringes were determined for each channel by
326: fitting a sine wave to each column of pixels in the slit
327: direction, with the fringe visibility, $V$, being defined as
328: \begin{equation}
329: V= (I_{max} - I_{min})/(I_{max} + I_{min})
330: \end{equation}
331: where $I_{max}$ and $I_{min}$ are the maximum and minimum values
332: of the fitted sine wave.
333: A wavelength calibration was applied using the observed
334: Thorium-Argon spectra, and the visibility and phase data were
335: dispersion corrected and re-binned to a log-lambda wavelength
336: scale. To determine differential velocity shifts in the
337: star+iodine data (starlight passing through the iodine cell) the
338: data can, to a good approximation, be considered as a summation of
339: the complex visibilities (${\bf V}=Ve^{i\theta}$) of the relevant
340: star ($V_Se^{i\theta_{S_0}}$) and iodine ($V_Ie^{i\theta_{I_0}}$)
341: templates (van Eyken et al. 2004; Erskine 2003). For small
342: velocity shifts the complex visibility of the data can be written
343: as
344: \begin{equation}
345: V_D e^{i\theta_D} = V_S
346: e^{i\theta_{S_0}}e^{i\theta_S-i\theta_{S_0}}
347: +V_Ie^{i\theta_{I_0}}e^{i\theta_I-i\theta_{I_0}}
348: \end{equation}
349: where $V_D, V_S,$ and $V_I$ are the fringe visibilities for a
350: given wavelength in the star+iodine data, star template and iodine
351: template respectively, and $\theta_D, \theta_{S_0}$, and
352: $\theta_{I_0}$ the corresponding measured phases.
353: In the presence of real velocity shift of the star and instrument
354: drifts the complex visibilities of the star and iodine template best match the
355: data with a phase shift of $\theta_S - \theta_{S_0}$ and $\theta_I -
356: \theta_{I_0}$ respectively. The iodine is a stable reference and the iodine phase shift tracks the instrument drift. The
357: difference between star and iodine shifts is the real phase shift
358: of the star, $\Delta \phi$, corrected for any instrumental drifts
359: \begin{equation}
360: \Delta \phi = (\theta_S - \theta_{S_0}) - (\theta_I -
361: \theta_{I_0})
362: \end{equation}
363: This phase shift can be converted to a velocity shift ($\Delta v$)
364: by a known phase-to-velocity scaling factor that is a function of
365: the optical delay in the Michelson interferometer ($d$), the
366: wavelength ($\lambda$) and the speed of light ($c$). These
367: parameters are connected by the relation
368: \begin{equation} \Delta v= \frac{c \lambda}{2 \pi d}
369: \Delta \phi \label{eqn:delayfringe2}
370: \end{equation}
371:
372: In reality the phase shifts of the star and iodine templates are
373: not the only free parameters needed to fit the observed data. The
374: Doppler velocity shift of the star induces a wavelength shift of
375: the entire spectrum and instrument instabilities can lead to pixel
376: shifts relative to the templates. Finding the correct phase shifts
377: requires that these wavelength shifts (or pixels shifts) for the
378: star and iodine templates also be fit simultaneously with the
379: phase shift parameters.
380:
381: To determine barycentric radial velocities we used
382: the stellar template of a star with known barycentric radial
383: velocity ($\eta$ Cas) in place of the target star template in
384: Equation 2. The visibility of the reference star was allowed to
385: scale to partially compensate for the template mismatch. The
386: stellar visibility depends on the line depth and width (Ge 2002),
387: and allowing it to scale is in effect similar to attempting to
388: match the line depths and widths of the stellar template with that
389: of the observed star. Similar spectral-morphing techniques have
390: been used (Johnson et al. 2006) with high-resolution echelle data
391: from the N2K survey to achieve short term differential velocity
392: precision of $\sim 5$ m s$^{-1}$ without the need for a stellar
393: template.
394:
395: \subsection{Radial Velocity using Templates Alone}
396:
397: The stability of the ET instrument (Figure 1) makes it possible to
398: attempt to extract the radial velocity without using simultaneous
399: calibration with the iodine cell. When the cell is not inserted
400: into the optical path, the resulting observations of the star can
401: explicitly be modelled as the template spectrum with a phase shift
402: added (the template spectrum may also be the stellar template of a
403: reference star)
404: \begin{equation}
405: V_D e^{i\theta_D} = V_S
406: e^{i\theta_{S_0}}e^{i\theta_S-i\theta_{S_0}}
407: \end{equation}
408: In such a case the phase shift, $\theta_S-\theta_{S_0}$, is a
409: combination of the velocity shift of the star compared to the
410: template and the instrument drift. When using this technique the
411: ET observing procedure requires all star-only exposure to have an
412: iodine template taken immediately before or after the exposure. If
413: the instrument is stable enough, these iodine exposures can be
414: used to determine and calibrate out the instrument drifts for the
415: star exposures. This is only possible when the instrumental
416: velocity drifts are small in the time interval between the
417: exposure mid-points of the star and iodine templates.
418:
419: Using the entire spectral range of $\sim640$ \AA\ a
420: cross-correlation was performed in log-lambda space using the
421: visibilities ($V$) of the reference star and the target star. The
422: purpose of this cross-correlation was to yield an approximate
423: velocity shift, accurate to $\sim 1-2$ km s$^{-1}$, which is
424: useful as a first guess starting point to calculate a more precise
425: velocity shift using both the visibility and phase information.
426:
427: The velocity extraction algorithm that is part of the ET pipeline
428: was used to minimize the residuals between the complex
429: visibilities (${\bf V}$) of the reference star and the target star
430: by varying the phase shift applied to the reference template
431: complex visibility. The visibility of the reference star was also
432: allowed to scale in the fitting procedure in an attempt to
433: minimize the template mismatch, as described in the previous
434: section. This technique yields a phase shift
435: $\theta_S-\theta_{S_0}$ that is a measure of the difference in
436: velocity between the stellar template used and the target star.
437:
438: To account for instrument drifts between the reference star
439: template and the target star, a similar analysis procedure was
440: followed for the iodine templates (taken very close in time to the
441: reference and target star exposure) with the only difference being
442: that the iodine visibility was not allowed to scale (since there
443: is no template mismatch between iodine cell spectra taken at
444: different times). The phase drift between the two iodine templates
445: is a direct measure of the instrument drift. Subtracting off the
446: iodine phase shift corrects for the instrument drift, giving the
447: real phase shift of the target star with respect to the reference
448: star.
449:
450: %
451: %
452: %
453: %
454: %
455: %
456: %
457: %
458:
459: \subsection{Radial Velocity from Star+Iodine Data}
460: In the normal data acquisition mode the stellar beam passes
461: through the iodine cell. Cross-correlation with the visibilities
462: from a reference stellar template alone is not appropriate since
463: the iodine spectrum is also present in the data. Using Equation 2
464: we can write
465: \begin{equation}
466: V_D^2 = V_S^2 +V_I^2 +2V_SV_I \cos(\theta_S-\theta_I)
467: \end{equation}
468: The term $2V_SV_I \cos(\theta_S-\theta_I$) has an expectation value close to zero over the
469: 640 \AA\ wavelength band since the cosine term fluctuates
470: randomly. We ignored this term and approximated the visibilities
471: as adding in quadrature ($V_D^2 \approx V_S^2 +V_I^2$). This
472: allowed us to use the TODCOR algorithm (Zucker \& Mazeh 1994) on
473: the ensemble of measured visibilities to determine the approximate
474: velocity shifts of the star and iodine templates that are a best
475: fit to the data. The TODCOR algorithm is a two dimensional
476: cross-correlation technique generally used to determine the
477: individual radial velocities of both components of a spectroscopic
478: binary, and can be applied in this case since the two components
479: are approximately additive. Since a reference stellar template for
480: a different star ($\eta$ Cas) was being used, a visibility scaling
481: was derived using mean visibilities and Equation 6, and used to
482: scale the stellar visibilities for TODCOR. The scaling factor in
483: visibilities accounts partially for the mismatch in stellar
484: templates. The approximate radial velocity values derived using
485: TODCOR were useful in determining the correct phase shift when the
486: shift was more than $ \pm 2\pi$. These radial velocities were used
487: as a first guess to determine the correct wavelength-shift and
488: phase-shift of the stellar and iodine templates that minimized the
489: residuals in the complex visibility between the data and the best
490: fit solution. Subtracting off the iodine phase shift gives the
491: phase shift of the target star with respect to the reference star,
492: corrected for instrument drift.
493:
494: \section{Radial Velocity Results}
495: The differential radial velocity precision of one of the reference
496: stars, $\eta$ Cas, is shown in Figure 2. These velocities were
497: extracted using the techniques described in Section 3, but with an
498: arbitrary zero-point. In the case of $\eta$ Cas the velocity
499: results are identical to those obtained using the standard
500: pipeline since this object is the reference star. The velocity rms
501: of $\sim6.4$ m s$^{-1}$ is due to a combination of photon noise
502: errors in the data and the templates and systematic errors in the
503: velocity extraction. We used this star as the stellar reference
504: template to determine barycentric radial velocities. Velocities
505: for all the stars are calculated relative to $\eta$ Cas, requiring
506: prior knowledge the barycentric radial velocity of $\eta$ Cas
507: itself to apply the correct velocity offset. We used the $\eta$
508: Cas mean barycentric radial velocity from ND02 ($8.314$ km
509: s$^{-1}$). $\eta$ Cas is known to be a long-period binary, whose
510: velocity is well fit with a linear trend. We transformed the ND02
511: velocity to our epoch of observation by applying a correction term
512: for its known $9.3$ m s$^{-1}$ year$^{-1}$ linear velocity drift
513: (Cummings et al. 1999). Any error in the determination of this
514: velocity is a error in the zero-point itself, and will affect all
515: target stars in a similar way. The phase difference of each star
516: from the $\eta$ Cas stellar template was converted to velocity
517: using the phase-to-velocity scaling parameter (${c \lambda}/{2 \pi
518: d}$). This radial velocity was then corrected to the solar system
519: barycenter by removing the barycentric velocity contribution of
520: the telescope, calculated with software written in IDL that use
521: the JPL ephemeris. These algorithms are accurate to better than
522: $\sim 2$ m s$^{-1}$. The corrected barycentric radial velocity of
523: $\eta$ Cas was added to obtain the barycentric radial
524: velocity of the star.%
525:
526: \subsection{Radial Velocity Results from Templates Alone}
527: Barycentric radial velocities were derived using the procedure
528: outlined in Section 3.1. A comparison of these velocities with
529: those published by ND02 is shown in Figure 3.
530: No attempt has been made to correct the velocities for the known
531: planetary companions to 51 Peg, 55Cnc and HD49674, which may cause
532: additional velocity scatter. Our velocities agree well with those
533: of ND02 and show a scatter of 120 m s$^{-1}$, with a mean offset
534: of $-79$ $\pm$ 25 m s$^{-1}$. The agreement of our results with
535: those of ND02 confirms the validity of using the iodine templates
536: to remove the instrument drift when such drifts are small (Figure
537: 1). For example, the typical instrument drift over the first night
538: is approximately 10 m s$^{-1}$ over a 15-minute exposure. Such
539: exposure times with the ET instrument allow the radial velocity of
540: an 11$^{th}$ magnitude (V=11) star to be determined at this level
541: of precision (100 m s$^{-1}$).
542:
543: \subsection{Radial Velocity Results from Star+Iodine Data}
544: When the
545: required precision is high, it is perhaps more appropriate to use
546: the star+iodine data since the iodine tracks the instrument drift
547: exactly. For all the reference and planet bearing stars, the
548: barycentric radial velocities obtained for all data points (using
549: $\eta$ Cas as template), derived using the procedure outlined in
550: Section 3.2, were averaged together to give a mean barycentric
551: radial velocity for each star. These radial velocities are shown
552: in Figure 4 and display a rms scatter of 47 m s$^{-1}$ with a mean
553: offset of -3 $\pm$ 19 m s$^{-1}$ in comparison with ND02.
554:
555: %
556: %
557: %
558: %
559: %
560: %
561: %
562:
563: \subsection{Systematic Errors and Velocity Zero Point}
564: Tests with twilight solar spectra as the reference stellar template show similar
565: levels of rms scatter to those using $\eta$ Cas, but the twilight
566: observation of the solar spectrum (instead of day sky) introduced
567: an offset in the velocities when the solar spectrum was used as
568: the template. Our results depend on the velocity of $\eta$ Cas
569: derived by ND02 who use the National Solar Observatory (NSO) solar
570: spectrum (Kurucz et al. 1984) as the stellar template with a 522 m
571: s$^{-1}$ offset to make the radial velocity of the Sun zero. Due
572: to the use of the same zero point
573: our estimated velocities will exhibit similar systematic errors with spectral type as ND02,
574: in addition to any template mismatch effects introduced due to our use of
575: $\eta$ Cas (G0V) instead of the Sun (G2V).
576:
577: %
578: %
579: %
580: %
581: %
582: %
583:
584: A number of systematic errors can occur in velocity determination
585: using this technique. Fast rotators ($v \sin{i} >$ 12 km s$^{-1}$)
586: have very low fringe visibilities at our chosen interferometer
587: optical delay (7 mm), leading to a significantly reduced velocity
588: precision for such stars. Incorrect calibration of the
589: interferometer delay can lead to errors in the phase-to-velocity
590: scaling, causing errors in the calculated radial velocity. Our
591: reference ($\eta$ Cas) has a small linear drift in velocity that
592: we have accounted for using the mean epoch of observation in ND02.
593: An error in this determined barycentric radial velocity of $\eta$
594: Cas will appear as a shift in the mean offsets reported here. The
595: stars with known planets have additional velocity variability due
596: to their companions. This variability is reduced when using
597: star+iodine data (Section 4.2) and averaging all the observed
598: velocities, especially for the short period planets 51 Peg and
599: HD49674 that have orbital periods less than our 12 day observing
600: run. The multi-planet 55 Cnc system (McArthur et al. 2004),
601: though, has long period planets which affect the calculated radial
602: velocity. For each star in their sample ND02 report only the mean
603: of the observed barycentric radial velocities, and to ensure a
604: consistent comparison we have made no attempt here to correct our
605: data for these known planetary companions. In general, velocity
606: corrections for known companions is easy to implement if the
607: orbital parameters are known, but we have not done so for
608: consistency with ND02.
609:
610: \section{DISCUSSION}
611: We have demonstrated the ability to measure barycentric radial
612: velocities at a high level of precision (rms = $50 -120$ m
613: s$^{-1}$) using a dispersed fixed delay interferometer instrument.
614: Our results also show that the radial velocities derived using a
615: very different technique and an instrument at substantially lower
616: resolution are not very different in precision from those derived
617: using high resolution spectroscopy and cross-correlation with
618: stellar (ND02; UD99) or Fe I line templates (Gullberg \& Lindegren
619: 2002). Simultaneous calibration with the iodine cell yields more
620: precise velocities, but the data analysis using only the templates
621: (Section 3.1, 4.1) may be more appropriate for faint stars where
622: the low signal to noise may introduce significant errors when
623: extracting velocities from the star+iodine data. We have corrected
624: the observed radial velocities to the barycenter of the solar
625: system. Using the corrected $\eta$ Cas radial velocity from ND02
626: to set the velocity scale for our radial velocities then
627: automatically corrects all velocities for the gravitational
628: redshift and convective blueshift of the Sun (used by ND02 as the
629: zero point). By adopting the radial velocity of $\eta$ Cas from
630: ND02 we have automatically adopted ND02 as our velocity reference
631: for the absolute calibration of a zero-point and performed a
632: secondary measurement to determine the velocities of all our other
633: target stars. No attempt is made to further correct the stellar
634: radial velocities for spectral type and luminosity class dependent
635: effects. This results in our derived velocities having typical
636: accuracies of $0.2-0.3$ km s$^{-1}$ (derived from ND02, our
637: velocity reference) for late F, G and K dwarfs, and precisions of
638: $\sim 50-100$ m s$^{-1}$ when comparing stars of similar spectral
639: types. With the ET instrument scheduled to be a facility
640: instrument at the KPNO 2.1m, this velocity measurement capability
641: will be available to the wider astronomical community. These
642: techniques will also be applied to data from the MARVELS survey
643: that will observe $\sim$ 10,000 stars as part of SDSS III, leading
644: to precise barycentric radial velocities being measured for the
645: majority of the stars in the sample.
646:
647: Future work with these instruments will focus on using the day sky spectra of the Sun as a stellar
648: template to determine the velocity zero point, allowing
649: independent comparison with the radial velocities of ND02 and UD99
650: without having to adopt either of these works as the velocity
651: reference, and a better understanding of any zero point
652: differences that may arise due to the much lower spectral
653: resolution (Dravins \& Norlund 1990) of ET compared to high
654: resolution echelle spectrographs.
655:
656: \acknowledgments We are grateful to Richard Green, Skip Andree,
657: Daryl Wilmarth and the KPNO staff for their generous support.
658: We would also like to thank the anonymous referee for a careful reading of the manuscript and for very constructive suggestions.
659: This work is supported by National Science Foundation grant AST
660: 02-4309, JPL, the Pennsylvania State University, and the
661: University of Florida. J. v. E. and S.M. acknowledge travel
662: support from KPNO and the Michelson Fellowship. This research has
663: made use of the SIMBAD and Vizier databases, operated at ADC,
664: Strasbourg, France. This work was performed in part under contract
665: with the Jet Propulsion Laboratory (JPL) funded by NASA through
666: the Michelson Fellowship Program. JPL is managed for NASA by the
667: California Institute of Technology. This work is based on data
668: obtained at the Kitt Peak 2.1m telescope.
669:
670: \begin{thebibliography}{}
671: \bibitem[Belokurov et al.(2006)]{2006ApJ...637L..29B} Belokurov, V., Evans,
672: N.~W., Irwin, M.~J., Hewett, P.~C., \& Wilkinson, M.~I.\ 2006,
673: \apjl, 637, L29
674: \bibitem[Butler et al.(1996)]{1996PASP..108..500B} Butler, R.~P., Marcy,
675: G.~W., Williams, E., McCarthy, C., Dosanjh, P., \& Vogt, S.~S.\
676: 1996, \pasp, 108, 500
677: \bibitem[Butler et al.(2003)]{2003ApJ...582..455B} Butler, R.~P., Marcy,
678: G.~W., Vogt, S.~S., Fischer, D.~A., Henry, G.~W., Laughlin, G., \&
679: Wright, J.~T.\ 2003, \apj, 582, 455
680: \bibitem[Cumming et al.(1999)]{1999ApJ...526..890C} Cumming, A., Marcy,
681: G.~W., \& Butler, R.~P.\ 1999, \apj, 526, 890
682: \bibitem[Dravins \& Nordlund(1990)]{1990A&A...228..203D} Dravins, D., \&
683: Nordlund, A.\ 1990, \aap, 228, 203
684: \bibitem[Erskine \& Ge(2000)]{2000ASPC..195..501E} Erskine, D.~J., \& Ge,
685: J.\ 2000, ASP Conf.~Ser.~195: Imaging the Universe in Three
686: Dimensions, 195, 501
687: \bibitem[Erskine(2003)]{2003PASP..115..255E} Erskine, D.~J.\ 2003, \pasp,
688: 115, 255
689: \bibitem[Erskine et al.(2005)]{2005SPIE.5905..249E} Erskine, D.~J.,
690: Edelstein, J., Harbeck, D., \& Lloyd, J.\ 2005, \procspie, 5905,
691: 249
692: \bibitem[Ge(2002)]{2002ApJ...571L.165G} Ge, J.\ 2002, \apjl, 571, L165
693: \bibitem[Ge et al.(2002)]{2002PASP..114.1016G} Ge, J., Erskine, D.~J., \&
694: Rushford, M.\ 2002, \pasp, 114, 1016
695: \bibitem[Ge et al.(2006)]{2006astro.ph..5247G} Ge, J., et al.\ 2006, ArXiv
696: Astrophysics e-prints, arXiv:astro-ph/0605247
697: \bibitem[Gorskii \& Lebedev(1977)]{1977IzKry..57..228G} Gorskii, S.~M., \&
698: Lebedev, V.~P.\ 1977, Izvestiya Ordena Trudovogo Krasnogo Znameni
699: Krymskoj Astrofizicheskoj Observatorii, 57, 228
700: \bibitem[Gullberg \& Lindegren(2002)]{2002A&A...390..383G} Gullberg, D., \&
701: Lindegren, L.\ 2002, \aap, 390, 383
702: \bibitem[Harvey \& The GONG Instrument Team(1995)]{1995ASPC...76..432H}
703: Harvey, J., \& The GONG Instrument Team 1995, ASP Conf.~Ser.~ 76:
704: GONG 1994.~Helio- and Astro-Seismology from the Earth and Space,
705: 76, 432
706: \bibitem[Helmi et al.(2006)]{2006MNRAS.365.1309H} Helmi, A., Navarro,
707: J.~F., Nordstr{\"o}m, B., Holmberg, J., Abadi, M.~G., \&
708: Steinmetz, M.\ 2006, \mnras, 365, 130
709: \bibitem[Ibata et al.(1994)]{1994Natur.370..194I} Ibata, R.~A., Gilmore,
710: G., \& Irwin, M.~J.\ 1994, \nat, 370, 194
711: \bibitem[Johnson et al.(2006)]{2006ApJ...647..600J} Johnson, J.~A., et al.\
712: 2006, \apj, 647, 600
713: \bibitem[Jones(1970)]{1970AJ.....75..563J} Jones, B.~F.\ 1970, \aj, 75, 56
714: \bibitem[Kozhevatov(1983)]{1983IGAFS..64...42K} Kozhevatov, I.~E.\ 1983,
715: Issledovaniia Geomagnetizmu Aeronomii i Fizike Solntsa, 64, 42
716: \bibitem[Kurucz et al.(1984)]{1984sfat.book.....K} Kurucz, R.~L., Furenlid,
717: I., Brault, J., \& Testerman, L.\ 1984, National Solar Observatory
718: Atlas, Sunspot, New Mexico: National Solar Observatory, 1984,
719: \bibitem[Lindegren \& Dravins(2003)]{2003A&A...401.1185L} Lindegren, L., \&
720: Dravins, D.\ 2003, \aap, 401, 1185
721: \bibitem[McArthur et al.(2004)]{2004ApJ...614L..81M} McArthur, B.~E., et
722: al.\ 2004, \apjl, 614, L81
723: \bibitem[Mahadevan et al.(2003)]{2003SPIE.5170..184M} Mahadevan, S., Ge,
724: J., van Eyken, J.~C., DeWitt, C., \& Shaklan, S.~B.\ 2003,
725: \procspie, 5170, 184
726: \bibitem[Marcy et al.(1997)]{1997ApJ...481..926M} Marcy, G.~W., Butler,
727: R.~P., Williams, E., Bildsten, L., Graham, J.~R., Ghez, A.~M., \&
728: Jernigan, J.~G.\ 1997, \apj, 481, 926
729: \bibitem[Nidever et al.(2002)]{2002ApJS..141..503N} Nidever, D.~L., Marcy,
730: G.~W., Butler, R.~P., Fischer, D.~A., \& Vogt, S.~S.\ 2002, \apjs,
731: 141, 503
732: \bibitem[Nordstr{\"o}m et al.(2004)]{2004A&A...418..989N} Nordstr{\"o}m,
733: B., et al.\ 2004, \aap, 418, 989
734: \bibitem[Rupprecht et al.(2004)]{2004SPIE.5492..148R} Rupprecht, G., et
735: al.\ 2004, \procspie, 5492, 148
736: \bibitem[Steinmetz et al.(2006)]{2006astro.ph..6211S} Steinmetz, M., et
737: al.\ 2006, ArXiv Astrophysics e-prints, arXiv:astro-ph/0606211
738: \bibitem[Udry et al.(1999)]{1999ASPC..185..367U} Udry, S., Mayor, M., \&
739: Queloz, D.\ 1999, ASP Conf.~Ser.~185: IAU Colloq.~170: Precise
740: Stellar Radial Velocities, 185, 367
741: \bibitem[van Eyken et al.(2004)]{2004ApJ...600L..79V} van Eyken, J.~C., Ge,
742: J., Mahadevan, S., \& DeWitt, C.\ 2004, \apjl, 600, L79
743: \bibitem[Veltz et al.(2008)]{2008arXiv0801.2120V} Veltz, L., et al.\ 2008,
744: ArXiv e-prints, 801, arXiv:0801.2120
745: \bibitem[Weinberg et al.(2007)]{2007AAS...21113210W} Weinberg, D.~H., et
746: al.\ 2007, American Astronomical Society Meeting Abstracts, 211,
747: \#132.10
748: \bibitem[Yanny et al.(2003)]{2003ApJ...588..824Y} Yanny, B., et al.\ 2003,
749: \apj, 588, 824
750: \bibitem[Zucker \& Mazeh(1994)]{1994ApJ...420..806Z} Zucker, S., \& Mazeh,
751: T.\ 1994, \apj, 420, 806
752:
753: \end{thebibliography}
754:
755: \clearpage
756:
757: \begin{figure}
758: \includegraphics*[scale=0.8,angle=90]{f1.eps}
759: \caption{Velocity drift of the ET instrument. The solid line is
760: the velocity drift as tracked by the iodine templates (filled
761: circles). The dashed line tracks the temperature drift of the
762: Michelson interferometer (asterisks). A loose correlation between
763: the temperature drifts and the velocity drifts is evident.
764: Instrument stability is essential for obtaining radial velocities
765: if the simultaneous iodine calibration is not used.}
766: \end{figure}
767:
768: \begin{figure}
769: \includegraphics*[scale=0.8,angle=90]{f2.eps}
770: \caption{Differential radial velocity observations of the star
771: $\eta$ Cas with ET. The rms velocity scatter of $\sim6.4$ m
772: s$^{-1}$ is caused by a combination of photon noise errors and
773: systematic errors. This star is used as the stellar template to
774: determine barycentric radial velocities (see Section 4).}
775: \end{figure}
776:
777: \begin{figure}
778: \includegraphics*[scale=0.8,angle=90]{f3.eps}
779: \caption{Comparison of radial velocities obtained by ET, from
780: stellar and iodine templates alone, with ND02. The comparison of
781: ET velocities with ND02 velocities yields an rms velocity scatter
782: of 120 m s$^{-1}$.}
783: \end{figure}
784:
785: \begin{figure}
786: \includegraphics*[scale=0.8,angle=90]{f4.eps}
787: \caption{Residuals of the average ET radial velocities derived
788: from star+iodine data from the velocities of ND02. Residuals show
789: an rms velocity scatter of 47 m s$^{-1}$.}
790: \end{figure}
791:
792:
793: \end{document}
794: