Next: X-Ray Source Detection Using the Wavelet Transform
Previous: The ASC Fitting Environment
Table of Contents --- Search --- PS reprint


Astronomical Data Analysis Software and Systems V
ASP Conference Series, Vol. 101, 1996
George H. Jacoby and Jeannette Barnes, eds.

Photometric Calibration of the Digitized Sky Survey

J. Doggett, M. Postman, B. M. Lasker, M. Meakes

Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218

Abstract:

The ST ScI Digitized Sky Survey (DSS) is based on scans of the photographic plates of surveys taken with the UK and the Palomar Schmidt Telescopes. Preliminary analyses have shown that photometry to within 0.5 magnitudes can be achieved by using a single calibration function for each photographic survey of the DSS. Photometric calibrations derived on a plate-by-plate basis will be good to better than 0.2 magnitudes.

1. Introduction

The Digitized Sky Survey (DSS) is based on scans of an all-sky set of plates taken with the UK and Palomar Schmidt Telescopes. (See Lasker and McLean (1994) for more information about the DSS.) The initial distribution of the DSS included a set of astrometric calibrations. Here we present the current status of the photometric calibration of the DSS.

2. Calibration System

  A software system called the DSS Calibration System (DSS-CS) has been developed to perform and archive the photometric calibrations. It consists of a database and associated software which controls the calibration procedures. The database is stored as a directory tree organized by plate in order to be simple to implement and to optimize access to the calibration data by plate.

The DSS-CS software is implemented as an IDL subsystem layered on top of the Special Plate Analysis and Measurement program (SPAM; McLean 1994). SPAM contains tools and data structures designed to facilitate interactive image analysis and calibration from within IDL. The DSS-CS software supplements SPAM with the abilities to non-interactively drive calibration pipelines, and to archive and retrieve calibration data.

3. Photometric Calibration

  The calibration of a plate from the DSS requires three basic steps: 1) Select the photometric reference stars, 2) Determine the image parameters for the reference stars, and 3) Model the calibration by fitting a function to the relationship of image parameter vs. magnitude of the reference stars.

The reference star catalog used for calibrating the DSS is the the Guide Star Photometric Catalog (GSPC; Lasker et al. 1988). The GSPC is particularly well-suited because it is an all-sky photometric catalog whose sequences are located at the centers of the plates of the DSS. The thresholding algorithm of Lutz (1980) as implemented in SPAM is used to detect objects on the plate and compute image parameters for the objects including size, area, and integrated signal.

Three analyses of the photometric calibration of the DSS are presented here. The first explores which image parameter is best suited for performing the photometric calibrations. The second illustrates the consistency of the calibration function for the plates among a single photographic survey. The third examines the calibration of an individual plate. Only analyses of the E-bandpass plates of the Palomar Observatory Sky Survey (POSS-E) are presented here. The results from the analyses of plates from the other photographic surveys in the DSS are similar to that of the POSS-E plates.

3.1. Image Parameters

Different image parameters may be used for deriving photometric calibrations. Trial calibration runs were performed to select the image parameter that yielded the smallest internal rms magnitude residuals.

Three image parameters which are computed by the Lutz algorithm were selected for these trials: integrated signal, ellipse area, and pixel area. The integrated signal is the sum of the counts above the first threshold in an object. The ellipse area is based on using the approximate major and minor axes of an object to compute the object's area. The pixel area is the number of pixels above the first threshold of an object.

Figure 1 shows the distributions of the internal residuals from the trial calibrations of the POSS-E plates for each of the image parameters. The modes of the distributions show that on average the integrated signal gives the smallest residuals.

  
Figure 1: Distribution of POSS-E Internal Residuals Using Various Photometric Parameters.
Figure 1: PS 27 Kb

3.2. Global Photometric Calibration

The global calibration curve, log(integrated intensity) vs. magnitude, for 2941 GSPC stars from the POSS-E plates is shown in Figure 2. This global calibration curve has an rms scatter of about 0.5 magnitudes. Because the calibration curve is well-behaved, it may be used to derive a single coarse calibration for the entire POSS-E survey.

  
Figure 2: Global Calibration for the POSS-E Plates.
Figure 2: PS 155 Kb

3.3. Sample Calibration

The calibration of the POSS-E plate XE 322 is show in Figure 3. The GSPC stars used to calibrate XE 322 are plotted with . The internal rms magnitude residual of a second order polynomial fit (solid line) to the GSPC stars is 0.09 magnitudes. Also plotted are stars from the composite photometric catalog of Mermilliod and Mermilliod (1994; + symbols) which, unlike the GSPC stars, are distributed across the whole plate. The external rms residual from the polynomial fit of the Mermilliod and Mermilliod stars is 0.17 magnitudes.

  
Figure 3: Photometric Calibration of the POSS-E Plate XE 322.
Figure 3: PS 40 Kb

4. Summary and Availability

  We have developed a database and software system for calibrating the DSS. Preliminary analyses indicate that global calibration functions for each photographic survey in the DSS can yield photometry good to 0.5 magnitudes, and that calibrations of plates on an individual basis are good to better than 0.2 magnitudes.

Global calibration functions for each photographic survey of the DSS will be available by the Spring of 1996, and more precise plate-by-plate calibrations will be ready in the the Fall of 1996. Announcements will be posted on the DSS WWW home page (http://www-gsss.stsci.edu/img_cdrom/img_top.html) as the calibrations become available.

Acknowledgments:

This work was supported in part by NASA grant NAG W-2166.

References:

Lasker, B. M., & McLean, B. J. 1994, ST ScI Newsletter, 11, No. 2, 39

Lasker, B. M., Sturch, C. R., et al. 1988, ApJS, 68, 1

Lutz, R. 1980, Computer Journal, 23, No. 3, 262

McLean, B. J. 1994, private communication

Mermilliod, J.-C., & Mermilliod, M. 1994, Catalogue of Mean UBV Data on Stars, (Springer-Verlag New York, Inc.)


Next: X-Ray Source Detection Using the Wavelet Transform
Previous: The ASC Fitting Environment
Table of Contents --- Search --- PS reprint
Wed Jul 3 07:38:52 MST 1996