During 2010 Astronomy Field Camp Robert Arlt Jr. worked with Travis Barman on two projects. The first, a pointed search of 2MUCD 10529, an M9-type star, for Mars-sized-extrasolar planets. The second, a search for an ejected planet or stellar companion of HR 8799. Focus was given to the second project, as data for it was available upon the camps arrival at Lowell Observatory.
Data for both projects was obtained with the 1.8-meter-Lowell-BU Perkins Telescope on Anderson Mesa. Mimir, a facility-class near-infrared (NIR) instrument, was used for both projects.
The Perkins Telescope equipped with Mimir.
“Mimir, is multifunction, performing wide-field (F/5) and narrow-field (F/17) imaging, long-slit spectroscopy,” imaging polarimetry, and capable of switching between modes in a few seconds (Clemens et al. 2007). Mimir’s F/5 mode was used to expose HR 8799 and 2MUCD 10529 for 30 seconds onto a 1024 x 1024 pixel ALADDIN II InSb array detector with a plate scale of approximately .579 degree arcseconds per pixel. Data for the second project was also supplemented by the 2MASS catalog.
Data was obtained for the first project January 8 and 9, 2010. 2MUCD 10529 was the target for both nights.
2MUCD 10529 is circled in red.
The star was targeted because of its position during the half nights it was observed (above AP 2.00 for all but an hour of observable time and within the safe-slew limits of the telescope), its H-band magnitude (12.266), and its spectral type of M9. A total of just under six hours of data was obtained between the two nights. The data was calibrated, but then focus was given to the second project.
“[I]n 2008 November, Marois et al. (2008) announced the discovery of three planets orbiting the A5V star HR 8799, based on near-IR imaging at the Keck and Gemini telescopes.”
HR 8799's three planets are visable.
“The extrasolar planetary system around HR 8799 is the first multiplanet system ever imaged… This is a remarkable system with no analog in any other known planetary system,” (Close and Males 2009). For this reason, any information that can be gained about the system is highly valuable.
If a stellar companion of HR 8799 was found it could be studied to constrict the age of the entire system. The second project was designed with this in mind. A similar project was conducted by Close and Males of Steward Observatory, University of Arizona. They used visible and NIR data from HST/NICMOS and Gemini-North/Altair AO/NIRI NIR in order to obtain proper motions for server object near HR 8799.
Close and Males data.
It was hoped that by using only NIR data obtained with Mimir a deeper and more accurate search for a stellar companion could be conducted due to much better seeing in the NIR band.
Data for the search was obtained with Mimir on October 20, 21, and 22, 2008 and January 16, 2010. The data was calibrated with the IDL based Mimir Software Package – Basic Data Reduction (BDP) (the BDP is developed by Dan Clemens and was available from http://people.bu.edu/clemens/mimir/software.html at time of writing). An astrometric solution was then obtained for the data by cross-correlating the data with the 2MASS catalog using an IDL routine developed by Mike Pavel.
The data was further calibrated and stacked using the IRAF packages ccdmask, ccdproc, imarith, imcombine, and mscimage. All objects in the stacked image were marked by hand with SAOImage DS9 and the IRAF package tvmark.
The final 2008 stacked image.
Accurate centroid positions were calculated using the Moffat algorithm with the IRAF package phot. The IRAF package wcsctran was then used to calculate the Right ascension and Declination of each object. Approximately 40% of the objects in the frame had to be rejected due to inaccurate centroiding (logical position errors greater than half a pixel). Meaning that search could not be completed on a level as deep as anticipated.
Originally, the positions from the 2008 data were planned to be compared to the 2010 data in order to obtain proper motions. However, instrument problems and sky conditions made the 2010 data scientifically unusable.
The final stacked 2010 image.
So, the 2008 data was compared to the 2MASS catalog. Proper motions were calculated for nearly all of the objects in the frame. However, the centroiding errors caused radial position errors for the objects as high as 100 milliarcseconds (mas) on proper motions that were no higher than 150 mas per year. By comparing the calculated proper motion of several stars to their known proper motions, it was determined that any object within 100 mas per year in either direction of HR 8799’s proper motion could potentially be a companion. Despite the inaccuracy of the calculated proper motions and the wide rage of possible companion proper motion values, no possible companions were identified in the search.
In order for a more accurate and telling search to be made much more accurate astrometry must be obtained for the objects. The astrometry could possibly be improved by using a different centroiding algorithm; by taking longer exposures; or by using a camera with a smaller plate scale, such as Mimir’s F/17 mode. The next planned step in the project is to test different centroiding algorithms to determine if better astrometry and be obtained from the existing data, until new, longer exposure time data can be obtained. However, switching to the F/17 mode will likely be avoided, if possible, as the field of view limits the number of stars that can be imaged in the same amount of time.
Leave a Reply