Copy the page URI to the clipboard
Goodfellow, Mark Damian Dominic
(2013).
DOI: https://doi.org/10.21954/ou.ro.0000f117
Abstract
The topic of this thesis is to gather evidence for a link between star formation in galaxies that contain a quasar in their nuclei. For quasars I argue that the best technique to measure the star formation rates is to use the far infrared.
I used the Herschel PACS 100 and 160 micron infrared data from the Herschel ATLAS programme to examine star formation rates in quasar host galaxies. I performed a stacking analysis of the PACS 100 and 160 micron data of a sample of 411 quasar candidates from the SDSS and 2SLAQ catalogues. The quasars are detected at up to 32σ, which I infer is from star formation in the galaxies that contain these quasars.
In order to obtain this data I have written a new timeline filtering subprogramme, plus a cross-scan filtering subprogramme, for the PACS data analysis pipeline. The outputs of this programme, using the early science phase Herschel data, were used to do the quasar stacking analysis.
I was also able to derive the formal minimuin-variance point source filter using the calculus of variations with undertermined Langrangian multiplers. This kernel optimises the signal to noise but leads to false positive signals. I opted to use a modified kernel that whitens the power spectrum of the noise, and gives us a more reliable point source detection. However, the SNR penalty, for different PACS pixels and data sets, lies in the range 30% to 40%.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)
