|
Документ взят из кэша поисковой машины. Адрес
оригинального документа
: http://www.adass.org/adass/proceedings/adass94/chenj.html
Дата изменения: Sat Nov 4 01:46:25 2000 Дата индексирования: Tue Oct 2 01:07:14 2012 Кодировка: Поисковые слова: п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п |
240 kB PostScript reprint
J. C. Chen, M. A. Conroy, J. DePonte, and F. A. Primini
Smithsonian Astrophysical Observatory, 60 Garden St., Cambridge, MA 02138
Approximate count rates for point sources observed with the ROSAT HRI
can easily be calculated by using the IRAF/PROS imcnts task to
accumulate X-ray events in a (usually) circular aperture about the
source, with background derived from either a model background map or
a region adjacent to the source. However, more accurate determinations
require corrections for various instrumental effects, such as
reduction in effective area for off-axis sources (vignetting),
fraction of total source counts in the source aperture (encircled
energy), and variations in detector quantum efficiency with event
position on the detector. Although level-1 processing properly
corrects for such effects, PROS users who wish to apply non-standard
data screens, or who wish to study sources not detected in standard
processing, are required to compute and apply such corrections by
hand. The task is complicated by the fact that the correction factors
include contributions from a range of off-axis angles or detector
positions, due to the induced 
3
satellite wobble
present in most observations.
Srcinten is a new PROS task that computes and applies the above correction factors to the net counts in a user-specified aperture. If significance of the net counts is less than a user-specified threshold, an upper limit is calculated. As a pedagogical exercise, the task was designed to minimize additional coding and maximize the use of other existing tools in PROS and IRAF. It will be fully integrated into IRAF and available in the PROS xray.xspatial package in the next PROS release.
Both the vignetting and encircled energy corrections are primarily
functions of source off-axis angle. The existing PROS task QPSPEC is
used both to compute net counts in the source aperture and to compile
an off-axis angle histogram, whose entries comprise the fraction, 
,
of counts in the aperture corresponding to off-axis angle 
.
For each angle 
, the vignetting correction 
at 1 keV
is evaluated using the function in Figure 1 (David et al. 1993).
The average vignetting correction is then
The encircled energy correction is simply the integral of the point
response function (PRF) within the source aperture. Although in general
the ROSAT HRI PRF exhibits significant azimuthal asymmetries for
off-axis sources, the azimuthally-averaged function 
may be
used to estimate encircled energy (David et al. 1993). For each angle
in the off-axis angle histogram, the encircled energy within
aperture radius R is given by
for various off-axis angles 
is shown in Figure 1.
The average encircled energy is given by
Relative quantum efficiency variations of approximately 
10% across the
detector are present in the ROSAT HRI, and their correction requires
knowledge of the event positions in detector coordinates (i.e., a
coordinate system fixed to the detector). These coordinates are
present in the event structure of QPOE files generated with PROS
version 2.3, and may be added to earlier QPOE files using the PROS
task upqpoerdf.
The PROS task qpcopy is used to generate an image of the events in the source aperture, blocked at a resolution equal to that of the calibration quantum efficiency map (QE). The average quantum efficiency correction is then
where 
is the fraction
of total counts in binned detector pixel (i, j).
Srcinten uses the IRAF CL script language and, as described above, takes advantage of a number of existing IRAF/PROS tools. The structure chart in Figure 2 shows all the tools used in srcinten, and a brief overview is given below.
qpspec extracts the total counts from the source and background regions and computes net counts and error. It also produces an off-axis angle histogram file in table format.
qpcopy copies one QPOE file to another through region filtering, to generate a source image in detector coordinates.
imcalc multiplies the source and QE images.
imcnts computes the total counts in the source image.
tabpar, keypar, tinfo get keywords and values from tables, needed in computing correction factors.
tcalc performs arithmetic operations on columns to compute corrections.
Figure: ROSAT XRT vignetting function at the energy 1 keV
( left). ROSAT HRI encircled energy at various off-axis
angles ( right).
Original PostScript figures
(27 kB),
(127 kB)
Figure: Structure Chart for srcinten.
Original PostScript figure (5 kB)
Before using srcinten, users are required to load the xspectral, xspatial, ximages, images, and TABLES packages. An example srcinten run is shown below, and the output table is displayed in Table 1. Both the count rate and upper limit columns are included in the output, but for sources above the user-specified significance threshold, only the count rate column is filled.
xs> srcinten source qpoe file: xdata$rh110267n00.qp source region descriptor: circle 2427. 2806. 82. background qpoe file: xdata$rh110267n00.qp bkgd region descriptor: annulus 2427. 2806. 82. 100. output root name (_ccr.tab): src -- Creating aspect histogram table from qpoe -- -- Computing vignetting correction factor -- -- Computing prf correction factor -- -- Computing qe correction factor -- -- SRCINTEN created Output table: src_ccr.tab --
Table: Output table src_ccr.tab
Although srcinten is a simple tool, it reproduces much of the functionality of its HRI level-1 processing counterpart and demonstrates the utility of assembling existing PROS tasks into CL scripts to address real analysis problems. Possible future enhancements include computing an energy-averaged vignetting function weighted by the source spectrum, allowing apertures of arbitrary shape, and correcting for spatially-varying exposure through an HRI exposure map.
This work is partially supported by NASA contract NAS5--30934.
240 kB PostScript reprint