More advanced processing

Most of the likely requirements for an advanced dynamic processing system stem from the necessary combinatorial operations that will be required on images/catalogues derived from basic calibrated science survey products.

The more important and obvious of these have been grouped into clearly defined interrelated packages and are outlined below.

• Many advanced survey products are predicated on the ability to automatically generate detailed well-sampled PSFs as a function of position over each basic product image. Example uses of accurate PSFs are: optional automated PSF fitting to improve s:n of point sources particularly in crowded stellar regions; and adaptive kernel matching of PSFs for difference imaging for transient event detection (SNe, planetary transits, solar system objects).
• A related, but functionally separate issue, is how to optimally stack images taken at different times in (necessarily) different seeing conditions. This will involve assessing various stacking techniques where optimality may well depend on the science aims for the stacked data and may well delve deeply into (CPU intensive) image restoration techniques.
• Producing contiguous tiled regions and seamless overlap-calibrated large area catalogues. These and the preceding item could be initiated automatically from a DataBase-driven list of survey fields and survey progress eg. is it worth (re)stacking (re)mosaicing (re)calibrating this region ?
• Detection and parameterisation of Low Surface Brightness objects and an investigation of more general purpose image detection methodology that is not just optimised to find faint PSF-like images.
• Matching/merging frames in different passbands and for use in defining master catalogues for any given field for list-driven co-located photometry (cf. SDSS and their ability to measure fluxes and limits to fluxes in a consistent way).

Several of these items particularly PSF-based operations involving optimal stacking are research areas in their own right and could profitably be pursued as collaborative research projects. We have asked for a further 1 FTE/yr of research effort in our bid mainly for progressing these and related issues. As a further example of this we have already held fruitful meetings with the Subaru Supercomputing Group in Hilo who are keen to investigate alternative (and necessarily compute intensive) methods for optimal stacking of images taken in varying seeing conditions.