ACIS Recipes: Improve Astrometry

Improve astrometry

First read the CXC memo on astrometry problems to see if your dataset is affected by offsets introduced by processing errors or use of non-standard fiducial lights. Also read the CXC Caveat regarding randomization of event positions by +/- 0.25" which is incorporated into Level 1 processing. The original event positions can be recovered if desired using acis_process_events.

While most Chandra image have absolute positions accurate to 1"-2", a few mages have incorrect positions by several arcseconds. Improved precision may be possible. E. Feigelson suggests the following procedure for bring an ACIS image into the Hipparcos coordinate frame. Best results are expected from high-S/N wavelet sources in the inner portion of the field where the point spread function is narrow.

Check to see whether any ACIS sources are associated with in the Hipparcos/Tycho catalog with ~1 mas precision for the ~100,000 Hipparcos stars (V<9) and ~40 mas precision for the ~1M Tycho stars (V<11), and especially the new Tycho-2 catalogue with ~2.5M stars. The USNO UCAC2 catalogue with 27M stars with V<16 in the southen hemisphere should also now be available.
Note 1: Do not use the HST Guide Star Catalog Version 1 with ~20M stars due to its poor precision and pre-Hipparcos reference frame.
Note 2: It may be necessary to look at all columns of these databases; e.g., column 40 of the Tycho catalogues gives the astrometric accuracy of each star. See description of Tycho columns here. When printing out Hipparcos/Tycho tables from Netscape Navigator, you must "Save as" in "text" format; otherwise most columns are lost
Check the new 2-micron all-sky catalogues. 2MASS has >162M objects over half the sky available at IPAC. The astrometric accuracy of these objects a standard deviation of +/- 0.1" with respect to Tycho stars.

Check to see whether any ACIS sources have counterparts on the all-sky Schmidt photographic plates. For declinations >-20 deg, search the USNO-A2.0 catalogue of 526M objects from the Palomar Sky Survey, which is based on the Hipparcos frame and has positional precisions around 0.3". For more southerly declinations, try the ROE/NRL COSMOS catalog of ~500M objects from the ESO/UK Southern Sky survey plates, but note the frame is pre-Hipparcos.

Though unlikely, check to see whether any ACIS source has a counterpart in the FIRST radio survey covering 15% of the sky. Currently at 549M, these sources have precisions ranging from 0.05-1 arcsec on the VLBI reference frame (which is precise and nearly identical to the Hipparcos frame).

Check to see whether any ACIS sources are listed in NED or SIMBAD databases, but beware that their stated positions come from a variety of sources with various precisions and reference frames.

Once a set of astrometric counterparts has been identified it is convenient to convert their celestial positions to the (X,Y) coordinate system of your observation to compare to the x-ray positions you found. This can be done in IDL by reading the binary table header, extracting the astrometry information, and calling the routine ad2xy to convert the coordinates. These routines are in Landsman's IDL Astro Library. The code might resemble the following:


header = headfits(filename, EXTEN=1)

hprint, header

; Build an astrometry structure for this observation.

crpixX = fxpar( header, 'TCRPX11' ) 

crpixY = fxpar( header, 'TCRPX12' ) 

crvalX = fxpar( header, 'TCRVL11' ) 

crvalY = fxpar( header, 'TCRVL12' ) 

cdeltX = fxpar( header, 'TCDLT11' ) 

cdeltY = fxpar( header, 'TCDLT12' ) 

ctypeX = fxpar( header, 'TCTYP11' ) 

ctypeY = fxpar( header, 'TCTYP12' ) 

make_astr, astr, DELTA=[cdeltX,cdeltY], CTYPE=[ctypeX,ctypeY], $

; CRPIX=[crpixX,crpixY], CRVAL=[crvalX,crvalY]

; Convert from RA/DEC to X/Y sky coordinates. 

; Beware that ad2xy.pro output is off-by-1 dues to 0-based array indexing.

ad2xy, ra_coords, dec_coords, astr, x_index, y_index 

refx = x_index+1 

refy = y_index+1

S. Pravdo provides another IDL program, rd2xy1.pro, which asks for a filename and inputs of (ra,dec), and outputs image (x,y) coordinates (including the offset of 1). A MATLAB (or Numerical Recipes, or similar) program can then fit the (x,y) coordinates of ACIS sources to the (x,y) coordinates of astrometric standards with a (x,y)-offset and rotation. In all cases tested to date, the rotational correction is very small; e.g. 0.02 degrees corresponding to 1/3-pixel at the edge of the ACIS-I array. We generally are thus applying only a simple (x,y)-offset which can be determined from the average difference in (x,y) coordinates between ACIS sources and astrometric standards.
If several astrometric counterparts are available, one can quickly find the mean offset in (RA,Dec). The offsets should be inserted into the FITS header ACIS events or image files of interest. P. Broos advises this be done with the FTOOL fv: View the Header of the binary EVENTS table, and use the Edit Insert/Delete menu options to correct keywords TCRVL11 and TCRVL12 (for events files) or CRVAL1 and CRVAL2 (for image files).

If you wish to allow for a roll angle correction then the IDL program register_frames.pro will derive a least-squares coordinate transformation between two systems (e.g. the X/Y system used in your observation and the more accurate X/Y system of your astrometric counterparts), given the coordinates of a set of fiducial points in both systems. More information on register_frames is given elsewhere. Unfortunately, at this time a roll correction may NOT be applied to an event list simply by specifying FITS keywords (even though the FITS standards support that). You must instead change the actual (X,Y) values for the events themselves using the Perl script described below.

S. Koch has written a useful Perl script to apply translational and rotational shifts to an event list. It operates as follows:

NewX = ax + by + c NewY = dx + ey + f coordshift.pl Input.evt Output.evt a b c d e f

Note: Off-axis positions may have additional systematic positional errors of unknown amplitude due to the interaction between the centroiding algorithm of wavdetect/wrecon and the asymmetric off-axis point spread function.