Photometry and Astrometry

The photometry was done for all fields using HSTPHOT (Dolphin 2000). Procedure:

1. Mask known bad pixels
2. Use header information to determine which sets of frames are taken at the same pointing, and to get initial estimate of offsets for images taken at slightly different pointings
3. Reject cosmic rays in stacks of exposures at the same pointing
4. Combine frames taken at same pointing
5. Measure sky level
6. Remove remaining hot pixels
7. Simultaneously do photometry on set of combined frames, using a model PSF, but allowing for empirical residuals from the frames being reduced to be added onto the PSF model. Correct for CTE effects during the fitting process, and automatically determine aperture corrections during the process; apply photometric calibrations as well.

For a subset of fields, photometry was also performed using completely independent routines implemented within the XVISTA package. Procedure:

1. Determine image shifts from header information
2. Mask known bad pixels and areas around saturated stars
3. Reject cosmic rays in stacks of exposures at the same pointing
4. Create PSF for frames being reduced, using a custom model for each frame; the custom model allows for the focus of the telescope to be fit for each frame being reduced
5. Simultaneously fit stack of individual frames, allowing for frame shifts, and correcting for CTE effects in the process.
6. Interactively determine aperture corrections, and apply them and photometric calibration.

The independent reductions give us a way to estimate systematic uncertainties in the photometry that depend on the reduction technique. Initial comparison suggests there are uncertainties at the level of several percent; efforts are underway to pinpoint the exact cause and, if possible, to modify the techniques to minimize these effects. For example, we are in the process of attempting to improve PSF characterization.

Astrometry

Given the output photometry (brightness for stars at different pixel locations), we correct for the distortion within the WFPC2 cameras, including the 34-row correction, using the latest distortion prescription (Anderson & King 2003). We then calculate offset positions of each object from the target position of the field (for a specified reference images), the relative position from the center of the galaxy, and the absolute position of each object. Relative positions are good to $\sim$ 0.05 arcsec, while the accuracy of the absolute positions is limited by the accuracy of the HST coordinates, which are limited by the accuracy of the HST guide star catalog to $\sim$ 1 arcsec.

Simulated star tests

While doing the HSTPHOT photometry, we include an extensive set of artificial star tests for understanding completeness and possible systematic errors in the photometry. Aritificial stars are placed at random locations in the color-magnitude diagram, with magnitudes and colors that are distributed in the same way as the real stars. Tables of input and recovered magnitudes for all of the artificial stars are provided as one of the photometry products.

Quality control

We implement careful quality control of the photometry, which includes visual inspection of the color-magnitude diagrams, aperture corrections, and artificial star results. In particular, we look closely for any systematic differences from one chip to another. All steps performed in the course of the photometry are logged in a file that is accessible to any user, so results should be able to be reproduced by anyone.

We plan to provide notes for each field about any special considerations that need to be made when using the data.