In the Dark

It’s getting even warmer in Cape Town as we approach the Easter vacation. The few clouds to be found in the sky over the last couple of days have now disappeared and even the mountain behind the campus has lost its white fluffy hat:

It’s going to be a busy weekend in these parts over the forthcoming weekend. As in the UK, tomorrow (Good Friday) is a national holiday and there will be a 5K fun run around the campus. The temporary stands and marquees you can see in the picture are associated with that. On Saturday there’s a really big event finishing there too – the Two Oceans Marathon – which will finish on the University of Cape Town campus. At the moment it’s 30 degrees, but the forecast is to cool down a bit over the holiday weekend. Good news for the runners, but not I suspect for everyone who’s disappearing off for a weekend at the beach!

Anyway, I did my talk this morning which seemed to go down reasonably well. It was followed by a nice talk by Roberto Trotta from Imperial College in a morning that turned out to be devoted to statistical cosmology. I didn’t get the chance to coordinate with Roberto, but suspected he would focus on in the ins and outs of Bayesian methods (which turned out to be right), so I paved the way with a general talk about the enormous statistical challenges cosmology will face in the era after Planck. The main point I wanted to make – to an audience which mainly comprised theoretical folk  – was that we’ve really been lucky so far in that the nature of the concordance cosmology has enabled us to get away with using relatively simple statistical tools, i.e. the power spectrum.This is because the primordial fluctuations from which galaxies and large-scale structure grew are assumed to be the simplest possible statistical form, i.e. Gaussian.  Searching for physics beyond the standard model, e.g. searching for the  non-Gaussianities which might be key to understanding the physics of the very early stages of the evolution of the Universe,  will be more difficult  by an enormous factor and will require much more sophisticated tools than we’ve needed so far.

Anyway, that’s for the future. Cosmological results from Planck won’t be freely available until next year at the earliest, so I think I can still afford to take the long weekend off  without endangering the “Post-Planck Era” too much!