Zheng Zheng Zheng Zheng Institute for Advanced Study Institute for Advanced Study Cosmology and Galaxy Evolution from Galaxy Clustering
Jan 17, 2016
Zheng ZhengZheng Zheng
Institute for Advanced StudyInstitute for Advanced Study
Cosmology and Galaxy Evolution from Galaxy
Clustering
Halo Occupation Distribution (HOD)Halo Occupation Distribution (HOD)
Breaking the Degeneracies between Breaking the Degeneracies between Cosmology and Galaxy BiasCosmology and Galaxy Bias
With David Weinberg (Ohio State University)
(Zheng & Weinberg, astro-ph/0512071)
Evolution of Galaxies from HOD Modeling of Evolution of Galaxies from HOD Modeling of DEEP2 and SDSS Galaxy ClusteringDEEP2 and SDSS Galaxy Clustering
With
Alison Coil (University of Arizona)
Idit Zehavi (Case Western Reserve University)
Outline:
Galaxies from SDSS
Snapshot @ z~0Light-Mass relation not well
understood
Snapshot @ z~1100Light-Mass relation well
understood
CMB from WMAP
Cosmological Modelinitial conditions
energy & matter contents
Galaxy Formation Physicsgas dynamics, cooling
star formation, feedback
m 8 ns
Dark Halo Population n(M)
(r|M) v(r|M)
Halo Occupation Distribution P(N|M)
spatial bias within halosvelocity bias within halos
Galaxy ClusteringGalaxy-Mass Correlations
Halo Occupation Distribution (HOD)Halo Occupation Distribution (HOD)
• P(N|M)P(N|M)
Probability distribution of finding N galaxies in a halo of virial mass MProbability distribution of finding N galaxies in a halo of virial mass M
mean occupation <N(M)>mean occupation <N(M)> + higher moments+ higher moments
• Spatial bias within halosSpatial bias within halos
Difference in the distribution profiles of dark matter and galaxies within Difference in the distribution profiles of dark matter and galaxies within haloshalos
• Velocity bias within halosVelocity bias within halos Difference in the velocities of dark matter and galaxies within halosDifference in the velocities of dark matter and galaxies within halos
e.g., Jing & Borner 1998; Ma & Fry 2000; Peacock & Smith 2000; Seljak 2000; Scoccimarro et al. 2001; Berlind & Weinberg 2002;Yang, Mo, & van den Bosch 2003; …
Cosmological Modelinitial conditions
energy & matter contents
Galaxy Formation Physicsgas dynamics, cooling
star formation, feedback
m 8 ns
Dark Halo Population n(M)
(r|M) v(r|M)
Halo Occupation Distribution P(N|M)
spatial bias within halosvelocity bias within halos
Galaxy ClusteringGalaxy-Mass Correlations
Cosmological Modelinitial conditions
energy & matter contents
Galaxy Formation Physicsgas dynamics, cooling
star formation, feedback
m 8 ns
Dark Halo Population n(M)
(r|M) v(r|M)
Halo Occupation Distribution P(N|M)
spatial bias within halosvelocity bias within halos
Galaxy ClusteringGalaxy-Mass Correlations
CosmologyA
Halo PopulationA
HODA
Galaxy ClusteringGalaxy-Mass Correlations
A
CosmologyB
Halo PopulationB
HODB
Galaxy ClusteringGalaxy-Mass Correlations
B
=
Cosmology from Galaxy Clustering?
Halo populations from distinct cosmological models
Zheng, Tinker, Weinberg, &
Berlind 2002
Changing m
with 8, ns, and Fixed
CosmologyA
Halo PopulationA
HODA
Galaxy ClusteringGalaxy-Mass Correlations
A
CosmologyB
Halo PopulationB
HODB
Galaxy ClusteringGalaxy-Mass Correlations
B
=
Flexible HOD parameterization
• P(N|M)
Motivated by galaxy
formation models
• Spatial bias within halos Different concentrations of galaxy distribution
and dark matter distribution (c)
• Velocity bias within halos vg= vvm
Kravtsov et al. 2004; Zheng et al. 2005
Observational quantities
• Spatial Clustering Galaxy overdensity g(r)
Group multiplicity function ngroup(>N)
2-point and 3-point correlation function of galaxies
• Dynamically Sensitive Observables m
0.6/bg
Pairwise velocity dispersion v(r)
Average virial mass of galaxy groups <Mvir(N)>
• Galaxy-mass cross-correlation function
mgm(r)
Constraints on HOD and cosmological parameters
Changing m
with 8, ns, and Fixed
Constraints on HOD parameters
Changing m
with 8, ns, and Fixed
Constraints on cosmological parameters
Changing m only
Changing 8 only
Cluster-normalized
Halo MF matched
Influence Matrix
Constraints on cosmological parameters Forecast :
~10% on m
~10% on 8
~5% on 8 m0.75
From 30 observables
of 8 different statistics
with 10% fractional errors
Abazajian et al. 2005
Conclusion
Galaxy bias and cosmology are not degenerate with respect to galaxy clustering.
* HOD modeling can greatly increase the cosmological power of galaxy redshift survey by taking the advantage of high-precision clustering measurements at small and intermediate scales.
* Simultaneously, using galaxy clustering data, we can constrain the HODs for different classes of galaxies, which provide valuable tests of galaxy formation models.
Cosmological Modelinitial conditions
energy & matter contents
Galaxy Formation Physicsgas dynamics, cooling
star formation, feedback
m 8 ns
Dark Halo Population n(M)
(r|M) v(r|M)
Halo Occupation Distribution P(N|M)
spatial bias within halosvelocity bias within halos
Galaxy ClusteringGalaxy-Mass Correlations
Galaxy Evolution from Galaxy Clustering
Galaxy SamplesGalaxy Samples
DEEP2, z~1 (Coil et al 2006)DEEP2, z~1 (Coil et al 2006)
SDSS, z~0 (Zehavi et al 2005)SDSS, z~0 (Zehavi et al 2005)
Measurements of two-point correlation functions Measurements of two-point correlation functions as a function of luminosityas a function of luminosity
Two-point correlation function of galaxies
1-halo term
2-halo term
Excess probability w.r.t. random distribution of finding galaxy pairs at a given separation
Central
Satellite
Halo Occupation DistributionHalo Occupation Distribution
For a sample of galaxiesmore luminous than Lmin
Mmin - characteristic minimum mass
of halos hosting Lmin galaxies
M1 - mass of halos on
average hosting one satellite galaxy above Lmin
Scatter between galaxy luminosity and host halo mass
Modeling resultsModeling results
L
DEEP2 galaxies
Distribution of central galaxy luminosityDistribution of central galaxy luminosity
Mass scales of host halosMass scales of host halos
Establishing an evolution link Establishing an evolution link between DEEP2 and SDSS galaxiesbetween DEEP2 and SDSS galaxies
MergingStar
Formation
z~0
z~1
Merging
z~1
z~0
Stellar mass evolution
Star formation efficiency vs Halo massStar formation efficiency vs Halo mass
Stellar mass evolution (z~1 to z~0) Stellar mass evolution (z~1 to z~0) as a function of halo massas a function of halo mass
Merging
z~1
z~0
Star Formation
Tentative conclusion:
For central galaxies in z~0 M<1012 h-1Msun halos,~80% of their stars form after z~1
For central galaxies in z~0 M>1012 h-1Msun halos,~20-40% of their stars form after z~1
Fardal et al. 2006
Summary & Future WorkSummary & Future Work HODs at z~1 and z~0 from modeling two-point correlation functions ofHODs at z~1 and z~0 from modeling two-point correlation functions of DEEP2 and SDSS galaxiesDEEP2 and SDSS galaxies
Evolution link through halo evolutionEvolution link through halo evolution
Stellar mass evolution from z~1 to z~0 for central galaxies as a function Stellar mass evolution from z~1 to z~0 for central galaxies as a function of halo mass (pure merger vs star formation)of halo mass (pure merger vs star formation)
Useful constraints to galaxy formation modelsUseful constraints to galaxy formation models Clustering measurements for galaxy samples based on stellar massClustering measurements for galaxy samples based on stellar mass
Galaxy samples at different redshiftsGalaxy samples at different redshifts
Evolution of satellite galaxiesEvolution of satellite galaxies