June 4, 2003 Morning session Bob Nichol Galaxy Evolution in Clusters and Groups Study effect of environment on galaxy props in SDSS models must explain morph-dens/T-Sigma rel'n, etc. Measuring environment: distance to nth nearest neighbor Cluster finding cross-correlate spectro/photo data F-O-F Can SFR & morph of galaxy be affected by environment? Yes, at around 1 gal/Mpc^2, which suggests there are different physical mechanisms at play above/below density. Adjusting for value of h, Postman & Geller found 2-3 gals/Mpc^2 plot normalized SFR vs. surface density to 10th nearest neighbor at >1/Mpc^2 , median and width of distribution of SFRs change (get lower) ram pressure stripping? of warm halo gas from infalling galaxies via tidal interactions? Dave' cosmo simulation finds SFR break at 1/Mpc^2 "passive spirals": select spirals from SDSS via conc. index. redder ones show no emission lines, blue show lines. passive spirals live at right density no correlation between AGN fraction - density - galaxy type 40% of galaxies show AGN and live everywhere -->AGN are a bulge phenomenon, black hole formed early, AGN has no effect on SFR If all galaxies have AGN, why only see 40%? duty cycle? bursting? compare H-alpha EW vs. local density no change in SFR-density relation with velocity dispersion of host system! ========================================================================================== ========================================================================================== M-A. Bernardi Early type galaxies and their environment: constraints on galaxy formation models Semi-analytic gal. form. models age/metallicity vs.mass environmental dependence Bias: where are galaxies SPH simulations for gas dynamics, but do not resolve SF, problems with multiphase gas Semianalytic models, but have many parameters -N-body+ 1. Gas cooling + ang. mom transfer 2. SFR (IMF, timescale, efficiency) 3. Feedback 4. UV background (z-dep) 5. spectro-photometric evol'n (Bruzual-Charlot, etc.) 6. Mergers & morhp. evoln. Successes: TF, LF, clustering, morph vs. env, sizes of galaxies (spirals from ang. mom, ell. from size) Failures: Number counts vs. z dist, alpha-enhancement vs. mass, age/metallicity vs. mass, environmental dep. in CDM hierarchical clustering, most massive gals are last to form. if burst @each merger, biggest systems should have ~young pop. semi-analytic models predict increase in metallicity w/ mass, no increase in age w/ mass test these w/ SDSS use Lick indices, compare w/ stellar-pop models combine SDSS spectra of galaxies with similar props/environ to get high S/N Bernardi et al. 03d -no age-luminosity correlation -strong age-veldisp correlation (higher veldisp=older) (large scatter in lum-veldisp rel'n which is why there is no age-L reln) -take 40000 early types, make 1500 composite spectra. -in high density regions, also get same reln. (older=bigger) contrasts with SA models see alpha/Fe increases w/ mass Thomas,Maraston & Bender 02: shortening of SF timescale with increasing mass but not much change in scatter of SFRs at a given veldisp --maybe flattening of IMF? more massive galaxies make more massive stars. ->more SNII, higher alpha enhancements SA models assume same IMF ! SA models predict older early type gals in clusters, no metallicity dep. on environment -observe weak dependence of age on environment (~2Gyr max diff) -early type SHOULD be younger in field; not really observed Use FP rel'ns in four bands. Compute residual of galaxy from FP rel. to environment Weak environmental dependence: cluster galaxies have somewhat lower SB, higher veldisp, but no mag dependence. ->suggests early type cluster galaxies are slightly older, but only by 1-1.5Gyr ========================================================================================== ========================================================================================== A. Cooray Linear -> non-linear power spectrum large scales (small k) gives primordial tilt then turnover @matter-radiation equality Omega_m h^2 baryon wiggles from LSS constrain other params 2dF measures P(k) out to k~1/Mpc but cut off analysis at 0.15/Mpc to stay in ~linear regime halo model as way to go to non-linear P(k) -one-halo term goes as power law (see talks from Monday) fit P(k)=P(linear) + Ak^p A,p contain useful info. Also reduces biases in other measurements. When you place a hard cutoff in P(k), and don't add nonlinear term, will have bias in results (overestimate linear component) in redshift space, need to correct for peculiar motions, velocity dispersions ========================================================================================== ========================================================================================== L. Moscardini Modelling the ICM and DM distributions in high-res simulations goals: analytical fits to DM radial profiles -improve NFW fits -extend fitting to get analytical expressions veldisp vel. anisotropy phase space density analytical fits to ICM radial profiles veldisp entropy etc. useful for testing cluster mass estimates density profiles impt. for Xray, SZ emission and lensing LCDM Nbody sims (ZIC, Tormen et al 1997) use to get initial conditions to resimulate high density regions (of all types, randomly selected) resim w/ GADGET (Springel et al 01) nonradiative only particle mass~2-5*10^9 Msun high force resln 5kpc/h 17 sim clusters 3.6*10^14 - 1.5*10^15 Msun/h DM profiles -NFW fits inadequate -fit relation for radial profile of coarse grained phase-space denisty dens/sig_r^3 ~ r^1.95 -add velocity profile: at small r, vel. field is nearly isotropic, at large r radial motions dominate -velocity anisotropy B(r)=1-sig_t^2/2sig_r^2 -see infall motions on average, so not fully relaxed -new DM profile dens ~ r^-1 (1+r_p)^(-3/2) Mass estimates: -integrate DM profile or -use analytic expressions inside Jeans eqn. compare their model, NFW, and Jeans. Find NFW overestimates mass at small radii by up to 40%. Jeans & their model are good. Gas density profile gas & dm self-consistent at r>0.1r_vir Beta-model inappropriate dens(r) ~ (r+r_p)^-2.5 with r_p=0.04r_vir Gas temp profile nearly constant at r<0.3r_vir T~r^0.016 / (1+ r^4/r_T)^0.013 ; r_T = 0.009 r_vir Gas vel profile still infalling Gas mass estimates compare different methods. Find good agreement between their model, Beta model (NOT at center) Questions asked about issues at r>r_vir ========================================================================================== ========================================================================================== M. Valluri Substructure in Hydra I cluster (A1060) from HI w/ van Gorkom, McMahon A1060 looks relaxed in x-ray, z=0.011 L similar to Virgo, R=1 46%spiral, 34%S0, 19%E 190 redshifts , highly isolated cluster. spirals in clusters are HI deficient Solanes et al. 2001 HI imaging of Virgo, Coma show shrunken disks at centers of clusters but in Hydra, no evidence of HI deficiency! Even found 11 unidentified dwarfs look at redshifts in radial bins; see large difference betw. E/S0 and Sp in central region use mixture models to fit multiple gaussians (KMM) in velocity distribution find two components within 90' , with 1000 km/s offset can also try 3-component; in that case all HI deficient guys are in center, and find two groups in tails of velocity distribution. Central veldisp more consistent with Xray temp