There is significant field distortion in ISAAC images. At the edge, it amounts to two pixels and in the corners, it amounts to 2.5 pixels. This affects observations in two significant ways.
The individual tiles of mosaics have to be corrected before they are tiled together.
It is quite common in spectroscopic mode to offset from bright reference starsonto a faint target that cannot be seen in direct imaging. If these offsets have been calculated from ISAAC images and if they are large enough (greater than 30 arc seconds), then it is possible for the faint target to be missed, or, at the very least, to be near the edge of the slit. To avoid this, one should choose reference targets that are no more than 30" away from the target of interest.
As of 2004, the ISAAC field distortion will be determined regularly (at leastonce per year) and after every intervention. This is done by observing globular clusters which have previously been observed by HST/ACS.Currently we are using public NGC 5139 data as reference for the measurement of ISAAC field-distortion. After having obtained a matched ISAAC/ACS catalogof stars we use IRAF / IMMATCH to fit thrid order distortion (GEOMAP). The resulting transformation files are given in the table below.
|Object||Observed||GEOMAP file||Distortion image|
|NGC 1850||November 1998||isaac-dist-nov1998.db||IMG|
|NGC 5139||December 2004||isaac-dist-dec2004.db||IMG|
|NGC 5139||May 2005||isaac-dist-20050513.db||IMG|
|NGC 5139||May 2009|
|NGC 5139||August 2009 (at UT3)|
|NGC 5139||March 2010|
|NGC 5139||July 2010|
|NGC 5139||January 2011|
|NGC 5139||July 2011 (after the intervention)|