1990s 1920s (from Friedmann to Freedman)
theoretical technology available, but no data 20th century: birth of observational cosmology
Hubble's law ~1930 Development of astrophysics 1940s 1950s Discovery of the CMB 1965 Inflation 1981 CMB anisotropies: COBE ~1990
1920s 1990s (from Friedmann to Freedman)
theoretical technology available, but no data 20th century: birth of observational cosmology
Hubble's law ~1930
from antiquity Universe had been assumed to be static relativity naturally expects universe to expand or contract, but very few people took this literally Alexander Friedmann Georges Lematre not Einstein!
expansion eventually discovered by observation
At z << 1 all cosmological models expect a linear behaviour, z d
first evidence: Edwin Hubble 1929
"the possibility that the velocity-distance relation may represent the de Sitter effect" slope of graph 46550 km/s/Mpc or 51360 km/s/Mpc (individual vs grouped) assumption of linearity no centre to expansion established by 1931
(Hubble & Humason)
1
Timeline
1907: Bertram Boltwood dates rocks to 0.4 2.2 Gyr (U-Pb) 1915: Vesto Slipher demonstrates that most galaxies are redshifted 1925: Hubble identifies Cepheids in M31 and M33 1927: Arthur Holmes "age of Earth's crust is 1.6 3.0 Gyr" 1929: Hubble's constant value of 500 km/s/Mpc implies age of Universe ~2.0 Gyr
Hubble's law systematics
distances mostly depend on m M = 5 log(d/10) (luminosity distance) getting M wrong changes d by a factor of which does not affect linearity (just changes slope) typical systematic error: very difficult to spot
Jan Oort expressed doubts very quickly (1931) no-one else till 1951!
10 ( M - M est ) 5
potential problem here...
Hubble used
Cepheid variables as calibrated by Shapley (1930) brightest stars in galaxies as calibrated by Cepheids total luminosities of galaxies calibrated by Cepheids and brightest stars
Wrong by factor of 2!
Wrong by factor of ~4!
Wrong because calibration wrong
Shapley (1930):
calibration of extragalactic Cepheids based on assumption of consistency with RR Lyrae variables in globular clusters
Baade (1952):
Cepheids in Magellanic Clouds ( Cephei stars or classical Cepheids) are different from "Cepheids" in globular clusters (W Vir stars or Type II Cepheids)
Typical classical Cepheid and W Vir light curves from the HIPPARCOS database
2
Period-luminosity relation
RR Lyrae stars
period < 1 day M ~ 0.7 (on horizontal branch) little evidence of dependence on period (does depend on metallicity) period > 10 days post-horizontal-branch low mass stars
MB = -4.35 log P + 3.98 MB = -1.33 log P + 0.24
W Vir stars
DH McNamara, AJ 109 (1995) 2134 Ngeow & Kanbur, MNRAS 349 (2004) 1130 MB = -2.59 log P - 0.67
classical Cepheids
period > 1 day post-main-sequence stars of a few solar masses
Distance error
from +0.7 to -0.7: ~ factor 2
Idea: brightest in stars all galaxies are about the same absolute magnitude
not unreasonable: tip of red giant branch is still used as distance indicator might worry about age and metallicity effects but first be sure you are looking at a star!
Hubble wasn't: he was seeing H II regions (ionised gas around young massive stars) these are much brighter than individual stars difference ~2 mag
M74/NOAO
!
M100 spiral arm
" "
blue plate; star marked
red plate; H II regions marked
Allan Sandage, ApJ 127 (1958) 123
3
#
1200 1000
Compilation by John Huchra
H 0 (km/s/Mpc)
800 600 400 200
Jan Oort Baade identifies Pop. I and II Cepheids "Brightest stars" identified as H II regions
0 1920 1930 1940 1950 1960 1970 1980 1990 2000
Date
$
Distance indicators
Stars, clusters, etc.
classical Cepheids novae globular clusters planetary nebulae supernovae Ia and II
Sources of uncertainty
calibration
zero point dependence on age, metallicity, galaxy type, etc. reddening corrections
bias
Malmquist bias
at large distances, you tend to detect brighter than average objects
Galaxies
Tully-Fisher Fundamental plane
Bigger things
Sunyaev-Zeldovich effect gravitational lensing
personal biases too!
Allan Sandage: low Gerard de Vaucouleurs: high
$
140 120
general cosmology dependent Key project Sandage camp de Vaucouleurs cam p
H 0 (km/s/Mpc)
100 80 60 40 20 0 1970 1980
Date
1990
2000
reasonable convergence only in last decade see later
4
%
Does Hubble's law mean universe is expanding (i.e. a(t) in RW metric not constant)?
Alternative hypotheses
real explosion at some past time
over time t galaxies travel distance d=vt, so build up Hubble law don't expect to be at centre of expansion, so don't expect isotropy
"...