|
+ Contact NASA
|
||
|
|
|||
|
|
|||
| |||
|
| |||
|
|||
|
||
|
|
||
| Kepler Calibration | |||
|
|
|||
|
|
|
ASTROMETRY Precise coordinates for Kepler targets are essential, both in the target assignment phase and during subsequent photometric analysis. Source coordinates are provided by the Kepler Input Catalog (KIC), searchable via the MAST interface. Astrometric calibration of the source centroids occurs in the Photometric Analysis (PA) pipeline module. Precise centroids of sources within the target aperture permits correction of the photometry for spacecraft motions during the observing season. The focal plane array contains significant gaps between the detectors, designed to exclude very bright stars from falling photo-active pixels. Too-bright stars will induce excessive bleeding along the read-out columns, and exhibit significant PSF wings, corrupting photometry of surrounding stars. These gaps will cause some sources, which appear to lie on the FOV, to actually be not observable. The Kepler FOV is shown to the right, where some of the excluded bright stars are indicated. Observers should carefully check to see if their proposed sources land on active silicon. |
|
|
|
|
Source Astrometry
Guest observers obtain coordinates for their sources from the Kepler Input Catalog (KIC). The source of the astrometry depends on which catalog (or catalogs) contain data for that source. The order for the choice of astrometric values is described in the release notes for the Kepler Input Catalog. Those catalogs are listed here, with an estimate of their positional accuracy.
1. Kepler Stellar Classification
Program; 50 milliarcseconds, data obtained closer to the
Kepler epoch, minimizing proper motion offsets The values listed in the KIC arise from the first catalog in the above list that contains a measure for that source. The source catalogs should be consulted for a detailed discussion of astrometric accuracy. The Kepler Project required positions to be accurate to 200 milliarcseconds, and proper motions to 20 millarseconds per year.
Observers will generally want to know where their sources lie within the assigned target aperture. Image centroids are calculated as part of the calibration pipeline, within the Photometric Analysis (PA) module. PA computes flux-weighted mean centroids for all stars, which are tabulated in the light curve files, expressed as row and column pixels values. These data provide an image centroid time series, and enable observers to assess target placement during the observing sequence. On each channel, a set of bright (but not saturated), relatively isolated stars are chosen to provide a reference grid for astrometry. Photometry for this set of reference stars is processed within the Photometer Performance Assessment (PPA) pipeline module to provide metrics of the photometric and astrometric stability of the instrument. Image centroids for these stars are used to create a "plate" solution specific for each CCD channel. This solution is then interpolated to convert detector coordinates (row, column) to celestial coordinates (RA, Dec). In the currently released data (Q0-Q3), both public and proprietary Guest Observer data, the derived celestial coordinates are not provided in the light curve tables. Analysis of centroid motion is not encumbered by the lack of this calibration; both significant and more subtle motions can be discerned in the centroid time series (detailed in the paper mentioned below). In particular, the flux-weighted centroid will respond to photometric variability when two or more stars are present within the aperture. The centroid will appear to move in a systematic manner as one star varys in brightness; the resulting motion time series can be used to help identify false positive transit signatures. We anticipate that astrometric coordinates will be included within Pipeline Version 7.0, approximately coincident with the first Cycle 2 data release. FFIs: The full frame images were designated as engineering data by the Project, and no astrometric calibration was initally intended. MAST developed an astrometric solution for the 8 Golden full-frame images, using the public astrometry.net tool, developed by Blanton, Hogg, Lang, Mierle & Rowies. These 8 FFIs were taken under ideal pointing and thermal stability at the start of the mission, and are avaiable here. Subsequent FFIs do not yet contain astrometry. Plans for the possible astrometric calibration of all FFIs is under review.
The high signal-to-noise ratios achieved with Kepler permit a high level of astrometric accuracy, despite the large pixel scale and large field of view. In principle the Kepler data can be used to determine parallaxes and proper motion for tens of thousands of stars, and explore more subtle motions, hinting at planetary companions. The Kepler Project is working towards understanding the astrometric precision of the data, and its potential applications. Based on analysis of the Q0+Q1 data, Monet provides an estimate of Kepler's astrometric precision of ~4 milliarcseconds over a single 30 minute observation (1 long cadence). Additional details are provided in an initial report on astrometric results from Kepler. The abtract from this paper (Monet etal 2010): |
|
Although not designed as an astrometric instrument, Kepler is expected to produce astrometric results of a quality appropriate to support many of the astrophysical investigations enabled by its photometric results. On the basis of data collected during the first few months of operation, the astrometric precision for a single 30 minute measure appears to be better than 4 milliarcseconds (0.001 pixel). Solutions for stellar parallax and proper motions await more observations, but the analysis of the astrometric residuals from a local solution in the vicinity of a star have already proved to be an important tool in the process of confirming the hypothesis of a planetary transit. |
Questions concerning Kepler's science opportunities and open programs, public archive or community tools? Contact us via the email address.
