SSSC11 PURR Log

Stewart Gleadow, gleadows@unimelb.edu.au, University of Melbourne

We know the source used to create this simulation, an unpolarised source with with some spectral index. Earlier exercises had no spectral variation.

LSM shown below.

Checking for reasonable data.

Showing visibility data (from Bookmarks) only shows input signal in "0" (XX) and "3" (YY), nothing in the cross polarisations, consistent with an unpolarised signal.

First, make a single channel image to check IQUV. Signal should only be in Stokes I. Looks like there is a small amount of leakage into Stokes Q.

Also make a Stokes I image with the full 64 channels, to see if the effect of the spectral index on the input source can be seen (cut down to 64 pixels, and 15 acrmin field of view to reduce the data size since there are 64 channels). It hardly changes across the 64 channels (~5.01 to 4.96 Jy/beam), but doing some quick calculations using the spectral index of -0.7, this is the right order of magnitude for the variation we would expect to see across this bandwidth.

Calculate the gains by calibrating on the LSM we know was used in creating the simulation, with the same other settings as SSSC 1,6,7.

Set a solution sub-interval for time but not for freq, since the error is still only changing with time (so should be the same across all frequencies).

Imaged the residuals, but only saved every 10th channel (6 channels total), so that we could check that the solution held across all frequencies without having too large a data set.