The Spitzer 24-micron Photometric Light Curve of the Eclipsing M-dwarf Binary GU Bootis
Kaspar von Braun, Gerard T. van Belle, David Ciardi, Mercedes Lopez-Morales, D. W. Hoard, Stefanie Wachter
aa r X i v : . [ a s t r o - ph ] A ug **FULL TITLE**ASP Conference Series, Vol. **VOLUME**, **YEAR OF PUBLICATION****NAMES OF EDITORS** The
Spitzer µ m Photometric Light Curve of theEclipsing M-dwarf Binary GU Bo¨otis Kaspar von Braun, Gerard T. van Belle, David Ciardi
Michelson Science Center, California Institute of Technology, MC100-22, Pasadena, CA 91125; kaspar, gerard, [email protected]
Mercedes L´opez-Morales
Carnegie Fellow, Department of Terrestrial Magnetism, CarnegieInstitution of Washington, 5241 Broad Branch Rd. NW, Washington,DC 20015; [email protected]
D. W. Hoard, and Stefanie Wachter
Spitzer Science Center, California Institute of Technology, MC 220-6,Pasadena, CA 91125; hoard, [email protected]
Abstract.
We present a carefully controlled set of Spitzer 24 µ m MIPStime series observations of the low mass eclipsing binary star GU Bo¨otis (GUBoo). Our data cover three secondary eclipses of the system: two consecutiveevents and an additional eclipse six weeks later. The study’s main purpose isthe long wavelength characterization of GU Boo’s light curve, independent oflimb darkening and less sensitive to surface features such as spots. Its analysisallows for independent verification of the results of optical studies of GU Boo.Our mid-infrared results show good agreement with previously obtained systemparameters. In addition, the analysis of light curves of other objects in thefield of view serves to characterize the photometric stability and repeatabilityof Spitzer’s
MIPS-24 at flux densities between approximately 300–2,000 µ Jy. Wefind that the light curve root mean square about the median level falls into the1–4% range for flux densities higher than 1 mJy.
1. Why is GU Bo¨otis Important?
GU Bo¨otis is a nearby, low-mass detached eclipsing binary system, consistingof two nearly equal mass M-dwarfs (L´opez-Morales & Ribas 2005). It is one ofcurrently very few known nearby ( <
200 pc) double-lined, detached eclipsingbinary (DEB) systems composed of two low-mass companions (L´opez-Morales2007). Eclipsing binaries can be used as tools to constrain fundamental stellarproperties such as mass, linear radius, and effective temperature. Given the factthat over 70% of the stars in the Milky Way are low-mass objects with
M < M ⊙ (Henry et al. 1997), coupled with the considerable uncertainty over the mass-radius relation for low-mass stars, objects such a GU Bo¨otis are of particularinterest in exploring the low-mass end of the Hertzsprung-Russell Diagram.The characterization of the effects of limb darkening and star spots intro-duces free parameters and thus statistical uncertainty in the calculation of the1 von Braun et al. stellar radii and masses. Using the Spitzer Space Telescope , we obtained 24 µ m time series observations of three separate instances of GU Boo’s secondaryeclipse (see Table 1) to create a light curve far enough in the infrared to not becontaminated by the effects of limb darkening and star spots.A further goal of our study is to characterize the photometric stability ofthe Multiband Imaging Photometer (MIPS) on Spitzer at 24 µ m over short andlong time scales. Time-series observing is atypical (albeit increasingly common)for Spitzer which is the reason why there are very few published photometriclight curves based on
Spitzer observations. The recent spectacular observationsof primary and secondary eclipses of transiting planets are notable exception(see for instance Charbonneau et al. 2005; Deming et al. 2005). Of these, theDeming et al. (2005) study was performed at 24 µ m. We therefore observed twoconsecutive secondary eclipses of GU Boo about 12 hours apart (observing sets1 and 2), and then a third event about six weeks later (observing set 3). Table1 gives an overview.We describe our MIPS-24 observations and data reduction procedure in §
2, present our results concerning GU Boo’s light curve and the photometricstability of
Spitzer and MIPS-24 in §
3, and summarize in §
2. Observations and Data Reduction
We used MIPS-24 aboard the
Spitzer Space Telescope (Werner et al. 2004) toobserve GU Bo¨otis in February and April of 2006, as outlined in Table 1. MIPS-24, the 24 µ m array, is a Si:As detector with 128 ×
128 pixels, an image scale of2.55” pixel − , and a field of view of 5.4’ × µ m small field photometry pattern(for details, see, for instance, Richardson et al. 2006).Our goal was to observe three independent secondary eclipses of GU Boo:two consecutive ones and another one several weeks after the first two. Of ourtotal of nine of Spitzer’s
Astronomical Observation Requests (AORs), three wereused for each secondary eclipse event (see Table 1). Each AOR contained eightobserving cycles with 36 individual exposures each. The first exposure in eachcycle is 9s long, the subsequent 35 are 10s long. The first two exposures of everycycle were discarded due to a “first frames effect”. This procedure left 34 framesper cycle, 272 frames per AOR, 816 frames per secondary eclipse event, and 2448frames for the entire project (all 10s exposure time) .The MIPS-24 data are provided by the Spitzer Archive in the (flatfielded)Basic Calibrated Data (BCD) format. We applied further post-processing tothese data in order to correct for small scale artifacts, in particular using IRAF’s For background information on the
Spitzer and MIPS operations, we refer the reader to theSpitzer Observer’s Manual (SOM – http://ssc.spitzer.caltech.edu/documents/som/). For in-formation specifically related to MIPS data reduction, please consult the MIPS Data Handbook(MDH – http://ssc.spitzer.caltech.edu/mips/dh/) and Gordon et al. (2005). IRAF is distributed by the National Optical Astronomy Observatory, which is operated by theAssociation of Universities for Research in Astronomy, Inc, under cooperative agreement withthe National Science Foundation. µ m Light Curve of GU Bo¨otis Table 1. Spitzer MIPS-24 observations of GU Bo¨otisDate (2006) MIPS Campaign Obs. Set AORs Exposures a Feb 20
MIPS006500 Feb 21
MIPS006500 Apr 01
MIPS006700 a
10 seconds per exposure.
CCDRED package to remove the weak “jailbar” features in the images (as de-scribed in the MDH).The
Spitzer software package mopex (Makovoz & Khan 2005; Makovoz & Marleau2005) was used for co-adding the individual MIPS frames into mosaics of 17frames, using overlap correction and outlier rejection in the process. The choiceof 17 frames was made (1) to obtain a high signal-to-noise ratio (SNR) for ameasured stellar flux density in a resulting combined image and subsequentdata point in the respective star’s light curves, (2) to maintain a sufficientlyhigh effective observing cadence to temporally resolve elements of GU Boo’slight curve, and (3) not to be forced to combine frames from different cyclesinto a single light curve data point (see Table 1). The interpolated, remappedmosaics have a pixel scale of 2.45” pixel − . We show in Figure 1 the MIPS-24field of view of GU Bo¨otis.For photometric reductions of the mosaiced images, we utilized the apex component of mopex to perform point-source extraction as described in Makovoz & Marleau(2005) . This step included background subtraction of the images, and the fit-ting of a resampled point response function (PRF), derived from our own data.In order to match the PRF centroid as closely as possible to the centroid of thestellar profile, the first Airy ring was initially subtracted from the stellar profile,and detection was then performed on the resulting image.
3. Results3.1. Analysis of GU Boo’s Light Curve
We show in Fig. 2 the phased light curve of GU Boo. Based on preliminary lightcurve fitting to the relative flux density levels (scaled to magnitudes) shown inFigure 2, we find that our results are consistent with the system parameters Also see information on apex at http://ssc.spitzer.caltech.edu/postbcd/apex.html and theUser’s Guide at http://ssc.spitzer.caltech.edu/postbcd/doc/apex.pdf von Braun et al.
Figure 1. A
Spitzer
MIPS 24 µ m mosaic of GU Bo¨otis (marked with arrowat the center of the image). This mosaic was created using all 272 frames inone AOR and is about 5 arcmin on the side. North is up, east is to the left. µ m Light Curve of GU Bo¨otis Figure 2. Folded 24 µ m light curve for all 3 observed secondary eclipsesof GU Boo. The preliminary fit is overlaid. The ordinate is scaled to 24 µ m magnitudes with the zero point corresponding to the out-of-eclipse fluxdensity level. derived from the optical study of the system in L´opez-Morales & Ribas (2005).The orbital period and initial epoch of the primary eclipse were set to the valuesgiven in the ephemerides equation derived by L´opez-Morales & Ribas (2005).We further fixed the mass ratio and the radius ratio of the stars, as well as theeccentricity of the system ( e =0) to the values obtained in that work. We assumedno limb darkening effects in the light curves, as expected for observations that farinto the infrared (Claret et al. 1995; Richardson et al. 2006; Ciardi et al. 2007,and references therein), and no significant gravitational darkening or reflectioneffects, based on the spherical shape of the stars and the similarity in effectivetemperatures. All these are reasonable assumptions, based on the results of thestudy of GU Boo at visible wavelengths, and they are in fact hard to test indetail, given the photometric precision of the Spitzer light curve at this fluxdensity level. Table 2 gives our estimates of GU Boo’s system parameters.
Table 2. GU Boo System ParametersParameter ValueOrbital Period (days) a . ± . a a . ± . R ⊙ ) 0.66 ± a )Orbital Inclination i (degrees) 89.3 ± a ) a From L´opez-Morales & Ribas (2005). von Braun et al.
Figure 3 shows the fractional rms around median values for all objects with morethan 72 out of a total of 144 observational epochs for each individual observingset as well as for the three sets combined. Observing sets 1 and 2 were obtainedduring the MIPS006500 campaign, observing set 3 during MIPS006700 (Table1). We find that inter-set repeatability of
Spitzer’s
MIPS-24 is comparable toits repeatability within a set, both in terms of median flux density level as wellas rms values. For the objects with a flux density in excess of 1 mJy, the rmsvalues approach the 1–4 % level. The light curves of all objects in the field (otherthan GU Boo itself) are flat with different amounts of random scatter aroundthe median flux density level.
4. Summary
We used MIPS-24 onboard the
Spitzer Space Telescope to obtain time-seriesphotometry of the M-dwarf DEB GU Boo. Our observations cover three sec-ondary eclipse events, two consecutive ones and a further event six weeks later.Our mid-IR analysis of GU Boo’s light curve is less affected by stellar surfacefeatures than its optical counterpart. The results show good agreement withthe previously obtained system parameters based on optical and near-IR work.Finally, we find that the repeatability of MIPS-24 photometry is consistent overall temporal scales we sampled: within an observing set and on time scales of24 hours and six weeks.
Acknowledgments.
We gratefully acknowledge the availability of
Spitzer
Director’s Discretionary Time (DDT) for this project. We furthermore thankD. Frayer, S. Carey, J. Colbert, and P. Lowrance for their help with the MIPSdata reduction pipeline.
References
Charbonneau, D., et al. 2005, ApJ, 626, 523Ciardi, D. R., et al. 2007, ApJ, 659, 1623Claret, A., Diaz-Cordoves, J., & Gimenez, A. 1995, A&AS, 114, 247Deming, D., Seager, S., Richardson, L. J., & Harrington, J. 2005, Nat, 434, 740Gordon, K. D., et al. 2005, PASP, 117, 503Henry, T. J., Ianna, P. A., Kirkpatrick, J. D., & Jahreiss, H. 1997, AJ, 114, 388Makovoz, D., & Khan, I. 2005, Astronomical Data Analysis Software and Systems XIV,347, 81Makovoz, D., & Marleau, F. R. 2005, PASP, 117, 1113L´opez-Morales, M., & Ribas, I. 2005, ApJ, 631, 1120L´opez-Morales, M. 2007, ApJ, 660, 732Richardson, L. J., Harrington, J., Seager, S., & Deming, D. 2006, ApJ, 649, 1043Rieke, G. H., et al. 2004, ApJS, 154, 25Werner, M. W., et al. 2004, ApJS, 154, 1 µ m Light Curve of GU Bo¨otis Figure 3. A plot of median flux density versus fractional rms for the 24stars that have photometry for more than 72 out of 144 observational epochs.Shown are the individual 3 observing sets (see Table 1) to illustrate the re-peatability of