Tuesday, 5 May 2015

Pheno 2015 Liveblog: Day 1 Session 2

We return, refreshed from some good coffee (by which I mean it existed and contained caffeine) for the second session of the day.  Though someone stole my seat.  This session contains another Higgs talk, followed by a couple of cosmology/dark matter ones.

11:00 am: The Standard Model after the Higgs Discovery, Jens Erler

Electroweak precisions tests at high and low energy.  The more traditional examples are the W mass, Weinberg angle, and Z and h properties.  There are also low-energy (polarised) scattering and parity violation experiments.  In addition to the traditional results here and the obvious future at the LHC, there are a number of experiments at JLab, Mainz that will improve low energy searches.

The basic strategy is, of course, to identify SM relations (e.g. between Weinberg angle and W/Z masses) and compute them to high accuracy.  Currently this lives at the two-loop level, but a future precision machine such as ILC would likely require 3 loops.  Then compare precise predictions with precise measurements.

The newest examples involve the Higgs.  Using the Higgs branching ratios, you measure the Higgs mass to 123.7 GeV.  Compare this to the 125.6 average from CMS/ATLAS, and you see good agreement (within current uncertainties).

A "traditional" measurement is to use the W mass.  There is actually a small difference, the measured mass is slightly above expectations.  Currently this is only 1.5 σ, with no look-elsewhere effect.  Worth keeping an eye on.

An old anomaly is the discrepancy bewteen the most precise Z-pole asymmetries at LEP and SLC.  These are the left-right (SLC) and forward-backward (LEP) leptonic asymmetries.  The SLC result is low, the LEP high compared to each other and the whole data set average.

Some discussion on the anomalous magnetic moment of the muon, and the residual problem in hadronic light-by-light scattering.  Really, there's not much new that can be said here till the new experiment at Fermilab reports its results.  It's worth remembering that the original anomaly was found over ten years ago at Brookhaven, and noone has repeated the measurement since.

Too often, conference slides are packed with too much information, especially when the speaker insists on reading all of it.  Jens has gone the other way, with far too little information on his slides and he's talking for minutes on a slide just showing pictures of experiments (no information).  With almost 50 slides in total, that's not really feasible.

Theory error in Δρ: smaller than statistical uncertainty.
What about Z invisible width? Needs an ILC or similar to explore.

11:35 am: Cosmological Results from Planck 2015, Silvia Galli

"Only" 28 papers in last year from Planck.  Small compared to the collider collaborations, still quite impressive though.  One of the big pieces of news has been the progress made with polarisation data.  No comments on the whole BICEP fiasco.

2015 saw the release of data based on the full mission data.  Main previous release (in 2013) was based on roughly half the run time.  Three most important changes since 2013:
  • Change in calibration bringing Planck and WMAP into alignment
  • Improved systematics, removing anomalous feature at l = 1800
  • Polarisation
One notable feature is that best fit models just to temperature data do excellent job in describing polarisation data.

Worth noting that lensing of CMB by later time structure unambiguously detected.

Two notable changes in ΛCDM fit.  Change in calibration shifted initial amplitude by 3.5 σ but this is understood.  Optical depth to reionisation shifted down by 1 σ and errors actually got worse.  This is due to choice to use only one instrument on satellite, compared to WMAP data used previously (which may have underestimated dust contamination).  Polarisation data has some residuals that are likely systematics; under investigation.

Some tensions in ΛCDM parameters compared to other experiments, in σ8 and H0.  At one to two sigma, so probably not serious but under study.

Extensions of ΛCDM
Neutrino masses: Improved by about 50% compared to prior data.  Including BAO results we get a bound of 0.23 eV.

The number of relativistic species (infamously) fluctuated down, closer to 3.

Curvature constrained to sub-percent level.  This is despite CMB being insensitive to curvature.

Inflation and gravity waves constraints at r < 0.1 or so.

Annihilatin dark matter for e.g. AMS-02 ruled out unless cross section evolves with temperature.

What limits neutrino mass measurements? Constraints from large-scale temperature already cosmic variance limited.  Limits from lensing can be improved a lot, with future experiments looking at the small scale CMB fluctuations.  Could in principle eventually probe below the 0.1 eV point for distinguishing normal vs. inverted neutrino hierachies.  Timescale: 5 Years!

Bounds on Tensor-scalar range from polarisation data?  Do not improve limits.

Dark matter assumptions?  Bounds written as f times σ, where f is an efficiency in the range 0.2 - 0.6 for quarks - leptons.

12:10 pm: Physics from DESI, Daniel Eisensteinn 

These experiments are based on galaxy surveys at low energies, with the goal of probing dark energy.  The history involved SDSS, BOSS and now DESI in the future.  All explanations for dark energy are exotic, even the cosmological constant.

Baryon acoustic oscillaions refer to the sound waves propagating in the first 400,000 years after the Big-Bang.  This leads to a characteristic scale (500 million light years) for galaxy separation.  This enables accurate distance measurements.  This scale can actually be seen by eye in the Planck CMB data, the characteristic size of the blobs.  It is, of course, the first and largest CMB power spectrum peak.

BAO refers to compressional sound waves in ionised plasma; restoring force is radiation pressure.  After recombination, the waves become unstable and run away gravitationally, leading to structure.  Toy model: single density spike, pressure wave flies out.  At recombination, sound speed drops by orders of magniture and wave stalls.  This stall distance is the BAO scale.

Key point: this length is predictably based on other cosmological inputs (simple acoustic physics).  From Planck data, uncertainty only 0.3%.  Different cosmological distance measures complicate things but not insurmountable.

The experimental challenge is then to measure this scale, which can be done by galaxy surveys.  SDSS, almost 20 years of data and still running, largest sample of astronomical objects.  Looking to use BOSS to improve things further and measure spectroscopy for redshifts.  Experimental uncertainty on BAO scale aiming for 1%, so still much worse than prediction.

Galaxy correlation functions a simple if lengthy counting experiment.  

Use Lyman-α forest (quasar absorbtion data) to get galaxy correlation data at higher redshift, z ~ 2.

All of this ultimately goes to a distance-redshift plot with good accuracy due to knowing that length scale.  Everything agrees with Planck.  Still need some work to get galaxy survey data smaller than Planck.  Even with curvature and w different from -1, preferred region pushes us back to ΛCDM.

 Another important check comes from growth of structure.  If we misunderstand gravity this could fake ΛCDM.  Look at growth of structure in galaxy clusters.  One tool is to measure correlations from velocity distributions in cluster, which are different along and orthogonal to the line of sight.

Big gap in redshift between galaxy surveys, Lyman-α forest that needs to be studied.  eBOSS experiment running, next generation search will be DESI.  Will give important results on Hubble rate as function of redshift.  Construction underway (though final approval yet to come).

What constraints can DESI set? Errors on w of 2%.

No comments:

Post a Comment