Speaker
Description
We present an early-stage effort toward the first end-to-end reanalysis of IRAS (1983), the first space telscope to perform a survey of the entire sky at infrared wavelengths. Our re-analysis is built directly from the raw time-ordered data (TOD), aiming to produce new public maps for cosmological use. Currently, the raw uncalibrated IRAS data does not exist publicly, however, we now have access to custom-compressed backups made by members of the original team in Groningen, which we have recovered.
IRAS presents numerous difficulties, both in terms of the raw data we have access to, which is partly incomplete and lossy, and in terms of original design choices. I will highlight one key challenge: the asymmetric rectangular detector masks, for which standard beam-deconvolution don’t apply. I will show preliminary efforts at reconstructing response functions, along with deconvolved images from said response functions.