A possible dipole in the laws of physics: How ALMA can help

John Webb, School of Physics, University of New South Wales, Australia

UNSW team:Matthew BainbridgeJulian BerengutSteve CurranVictor FlambaumJulian KingElliott KochAnant TannaMike WilcynzskaJohn WebbAngus WrightElsewhere:Bob Carswell (Cambridge)Michael Murphy (Swinburne)Wim Ubachs (Amsterdam)

€

= e2

hc

Motivation:

relies entirely on himThis: plus this

Let’s check the basic assumptions to the best of our ability

Is the cosmological principle right, or just a very good approximation?

To Earth

CIVSiIVCIISiIILyaem

Lya forest

Lyman limit Lya

NVem

SiIVem

CIVem

Lybem

Lyb SiII

quasar

Quasars: physics laboratories in the early universe

Exaggerated illustration of how transitions shift in different directions by different amounts – unique pattern

First calculated, relativisitic Hartree-Fock, 1999.Typicallly ~100<Q<1500

α 1999-2004, HIRES/KeckΔα

/α (1

0-5)

• Previous largest sample of quasar absorber constraints on Δα/α, Murphy et al (2004), from Keck/HIRES• Δα/α = ( -0.57 ± 0.11 ) x 10-5

• Obvious question: what would a different telescope find?

α 2011: Comparing new UVES/VLT data with Keck

VLT + Keck

Keck

VLT

KeckVLT

Different patterns in different directions

Lines on Earth

Redshifted lines

Looking “South”

Looking “North”

Looks pretty shifty…

4.1σ departure from α=constant (variable)

Δα/α = c + A cos(θ)

Keck & VLT dipoles independently agree, p=4%

VLT Keck Combined

Low and high redshift cuts are consistent in direction.Effect is larger at high redshift.

z > 1.6 z < 1.6 Combined

Distance dependence

∆α/α vs BrcosΘ for the model ∆α/α=BrcosΘ+m showing the gradient in α along the best-fit dipole. The best- fit direction is at right ascension 17.4 ± 0.6 hours, declination −62 ± 6 degrees, for which B = (1.1 ± 0.2) × 10−6 GLyr−1 and m = (−1.9 ± 0.8) × 10−6. This dipole+monopole model is statistically preferred over a monopole-only model also at the 4.1σ level. A cosmology with parameters (H0 , ΩM , ΩΛ ) = (70.5, 0.2736, 0.726).

Are a few high S/N outliers responsible for the signal, by chance?

• Alternative to growing error bars• Robustness check – iterative trimming• Adopt statistical-only errors and iteratively clip most deviant point• How much data do we need to discard to remove the dipole?

• c2 n = 1 reached

when ~10% clipped• Dipole significance ~5.5s at c2

n = 1 • Dipole significance stays above 3 s until ~60% of data discarded

Hints that this result might be real

Two internal consistencies:

1 Keck and VLT dipoles agree. Independent samples, different data reduction procedures, different instruments and telescopes.

2 High and low redshift dipoles also agree - different species used at low and high redshift – and different transitions respond differently to the same change in .a

Other suggestive points:

3 Scatter in data exceeds statistical-only error bars (expected). Keeping all points and growing errors is conservative. Trimming increases significance and shows signal is present in the majority or all of the data.

4 Monopole. Predominant in Keck. Mg isotopes? Early enrichment by very massive stars?

How can ALMA help?

• High-z sub-mm galaxies• Radio lines very sensitive to e.g. • Strong CI and CO and other lines• Select gravitationally lensed objects and pick out

individual massive molecular cores• Minimise radiative transfer effects, line offsets (carbon

isotopes?)• Statistical sample to randomise residual species offsets

€

2gpμ