Musings

Thaler

We have Jesse Thaler visiting us from Harvard. The first thing I got him to do when he arrived was to guide me in installing the software for the LHC Olympics. If Golem ever seems slow, in the next few months, it may be because I’m mucking about with the Pythia event generator.

His seminar was a more detailed version of one of points expounded in Nima Arkani-Hamed’s talk at Strings 2005. The problem of extracting new Physics from the LHC will be a tricky one. The map from “parameter space” to “signal space” is many-to-one, with radically dissimilar “bins” in parameter space mapping to the same “bin” in signal space.

What they’ve been doing is exploring this systematically in a class of MSSM models. They allow the gluinos and squarks to range in mass from 600 GeV to 1 TeV, the other gauginos, the higgsino and the sleptons to range in mass from 100 GeV to 1 TeV, and $\tan\beta$ to range between 2 and 50. They then generated 1 year (10 $\text{fb}^{-1}$) of “LHC data” for each of 39137 MSSMs and study all the LHC signals they could think of for these models.

Changing the parameters continuously, it is easy to distinguish nearby models. Change, say, the wino mass by a few 10s of GeV, and it is easy to distinguish that model from the original one. However, changing the parameters by a lot, one runs into pairs of models with nearly identical (statistically indistinguishable) signatures, but which occupy radically different positions in parameter space.

While they were looking at a particular class of MSSM models, the phenomenon seems to be true more generally, whenever the dominant signatures involve missing transverse energy (as is true in SUSY models with R-parity, large-extra dimension models, …), so that you don’t completely reconstruct all the decay chains.

Unless someone get very clever, we may turn on the LHC, see that there’s new physics, and have no way to pin down any of the details of what’s going on for several decades, until the ILC turns on. And, since that’s only a 500 GeV COM machine — at least initially — we will, likely still have trouble even after it turns on.

Anyway, though I have no illusions about being the one clever enough to crack this nut, I do want to play around and really get a feel for the tough interpretational issues we’re gonna face in 3 years time.

I’d love to expound more on this, but I have a lecture to write for our Geometric Langlands seminar, tomorrow, and I still have some calculations to finish for that…