Coordinate Systems

The IRAF environment provides three coordinates systems:

• physical: image pixel coordinates used in the reference (original) image.
• logical: image pixel coordinates, relative to the current section and size of the image. These are the units usually returned by tasks that read the image cursor (both in Imtool and SAOimage). These are the original (physical) image coordinates modified by transformations (such as sectioning, blocking) performed on the image.
• world: these are sky coordinates, typically celestial; however, other coordinate systems are also available. The PROS software supports Celestial, Ecliptic, Galactic and Supergalactic coordinates.

The PROS units for sky coordinates can be: decimal degrees, radians, $\prime$, $\prime\prime$, hours:minutes:seconds (hms), or degrees:arcmin:arcsec (dms). The default epoch is that of the reference image (i.e., B1950 for Einstein images, and J2000 for ROSAT images). However, the sky system may be changed by explicitly declaring the new system, as in the examples below.

The following syntax, (for the ROSAT test data target, AR Lac) can be used to specify source positions in the imcontour task (cf. ``help imcontour''):

• 22:13:53.7 44:28:53.9 - the usual representation of RA, Dec
• 333.4737d 44.48163d - in decimal degrees
• 5.820213r 0.776351r - in radians
• b1950; 22:11:49.3 44:13:59.3 - in 1950.0 coordinates, Besselian equinox (i.e.. FK4 system).
• J2000; 22:13:53.7 44:28:53.9 - in 2000.0 coordinates, Julian equinox (i.e.. FK5 system).

Logical, physical and world coordinate systems are all supported in the PROS software, and regions can be specified with any combination of them. Sky coordinates are identified by the input format (i.e., :, d, r, $^{\prime}$, $^{\prime\prime}$). Coordinates in pixels are logical by default (this is also the IRAF default). To use physical pixel coordinates, you need to declare them as physical, as in the examples below (unique abbreviations are accepted).

For example, to specify a circular region at the center of xdata$rp110590.qp [block=30] at the peak position :

• logical: numbers (real or integers) : l; c 231 313 8
• logical: numbers (real or integers) : c 231 313 8
  
• physical: numbers (real or integers) : ph; c 6930 9390 240
• world: any of the units described above: c 22:13:53.7 44:28:53.9 2$^{\prime}$
• mixed units can also be used. Any of the following combinations will be accepted and interpreted correctly:
  • ph; c 6930 9390 2$^{\prime}$
  • c 231 313 2$^{\prime}$
  • c 22:13:53.7 44:28:53.9 8
  • ph; c 22:13:53.7 44:28:53.9 240

N.B.: When a qpoe file is displayed and no blocking or subsectioning has been applied, the physical and logical units are the same. However, if the input file is, for example, xdata$rp110590.qp[6675:7187,9135:9647] (selects a region from the central portion of the file only, reducing the file size to 512$\times$512 while preserving full resolution), the values in the ``physical'' example above should be:

• physical: 6930 9390 240
• logical: 255 255 240
• world: 22:13:53.7 44:28:53.9 2$^{\prime}$

In an example using the Einstein data, xdata$snr.qp[block=2], the values would be:

• physical: 503 513 22.5
• logical: 251.5 256.5 11.25
• world: 22:59:04.9 58:36:55.4 3$^{\prime}$

N.B.: Input can be in any of the above syntaxes, in physical, logical and world coordinates. If SAOimage is used to make a region file the output is in logical pixels, regardless of the input units. There is a task, fixsaoreg, which will convert these logical coordinates to physical coordinates, which free the region definition from the particular section used in SAOimage. Then a PROS task like imcnts can use the converted region file with any image or section derived from the original source image.

For more information see `help regions', `help regcoords', and `help coords'. Also, see fixsaoreg, skypix, and wcscoords for information regarding coordinate conversion.