Point source detection and extraction in APEX rely upon a good measurement of the point response function (PRF). The term PRF is not just an alternative way of saying point-spread function (PSF). PSF-fitting is a commonly used term. However, PRF and PSF are two different objects. A PSF is an image of a point-source, and it is often oversampled; i.e., the pixel size of the PSF image is a fraction of the pixel size of the detector array or the mosaic image for which the the PSF is applicable. A PRF, however, is not an image of a point source. One way to think of a PRF image is as a kind of “table” of values of the responses of detector pixels to point source “hits” at different sub-pixel positions of a central pixel, stored in a convenient, interleaved form that resembles an image. Normally, a PRF is oversampled, so that there are n x n samples stored, with n being the resampling factor. This means that the PRF holds pixel response images for hits in each n x n grid point in a pixel. You can find more information in the document entitled “How to make the PRF from the PSF”, available at:
NOTE: For Spitzer data, standard PRFs may be used in APEX, or they may be measured from the mosaic with which they are to be used. Standard PRFs are available from the website. Users may produce their own PRFs using the PRF Estimate pipeline (prf_estimate.pl), which is part of the MOPEX distribution. DO NOT use PRF Estimate with and undersampled data, for example IRAC Channels 1 and 2. For IRAC PRF fitting you should use the provided PRFs.
IMPORTANT: APEX normalizes the PRF before fitting. If Normalization Radius is specified in the Source Estimate module, the normalization is performed over the pixels within that radius (IN PRF PIXEL UNITS) from the center of the PRF. If Normalization Radius is not specified, APEX will normalize over the available range that the central PRF covers in the PRF FITS file. Getting correct PRF-fitted absolute fluxes requires having the correct PRF normalization for your data!
Variable PRF
A provision is made to use a variable in the field of view PRF using the so-called PRF maps. It is assumed that several areas in the detector field of view will be identified, such that PRF variation within each area will be negligible and a constant PRF can be used for each area. A PRF map identifies the areas of constant PRF.
A PRF map maps an image with respect to the PRF image used for any pixel of the image. In the PRF map file header, the keyword Number_PRF gives the number of different PRF's that should be used for an image. The names of the files containing the PRF's are set in the keywords PRF_Filename_?, where "?" runs from 1 to Number_PRF. The PRF's can be in the form of a fits file (.fits) or an IPAC table (.tbl). The keyword HaveSigma is used to indicate whether PRF Sigma images are available. If they are, HaveSigma = 1; HaveSigma = 0 otherwise. The names of the images with the Sigmas are given by the keywords PRFSigma_Filename_?. The keywords ImageX and ImageY give the sizes of the images for which the PRF's are to be used. Here is a sample:
