Steven Gratton

Postdoc


Institute of Astronomy, Cambridge, UK.

01223 765849

s-t-g-2-0-'at'-c-a-m-.-a-c-.-u-k



For the past few years I have been working full-time on the cosmological analysis of data from the Planck satellite. In March 2013 we released our first cosmology papers. They are available from the ESA website:

http://www.sciops.esa.int/index.php?project=PLANCK&page=Planck_Published_Papers.

Planck and the earlier WMAP satellite have given us exquisite views of the microwave sky at different wavelengths, along with ancillary data. We find out a wealth of information about the universe by carefully studying the intensity and polarization of the cosmic microwave background (CMB) radiation reaching us today.

Unfortunately there are some complications. A major one (particularly for polarization) is contamination from "foreground" signals, typically from the galaxy. These signals have different spectral and spatial correlation properties than the CMB, enabling us to extract the cosmological information nontheless. Before Planck launched, I spent a lot of time developing ways to do this and to fold in our incomplete knowledge of the foregrounds and their properties into uncertainties on cosmological predictions. The 2013 analysis only used Planck temperature data. The next analysis will include Planck polarization data and it will be interesting to see how these methods perform on real data.

I have done work on quantum cosmology, inflation and the landscape measure problem and continue to think about such matters. I looked again at eternal inflation from a novel standpoint, partially inspired by a minigrant from the Foundational Questions Institute. I was able to put a measure on fluctuating inflationary histories and to see how volume weighting (favouring the trajectories that inflated more) affected the way inflation might end. See the paper on the arXiv.

That minigrant enabled me to begin to investigate doing computing for cosmology on graphics cards, as GPGPU computing was just coming online. Click here for more details, now mainly for historic interest: Graphics Card Computing for Cosmology.

Another FQXi minigrant supported a fun meeting in 2009 entitled "New Thoughts About The Universe". See its webpage.