Very Small Scale | Galaxy Clustering
​
Dark matter outnumbers baryons by more than a factor of five. Our current theories of galaxy clustering assume that the density distribution of dark matter drives the distribution of galaxies in the universe. However, what if there are scales at which galaxy clustering is determined not by just the dark matter, but by some intrinsic property of the galaxy-such as luminosity, or the observed epoch? The answer to these questions can be found by studying galaxy clustering within an individual dark matter halo. Measuring and modeling galaxy clustering on these scales less than about 200 kpc requires careful analysis of large data sets, in particular the Sloan Digital Sky Survey (SDSS) (York et al., 2000). It also requires comparing this data set to a large suite of cosmological N-body simulations of different volumes using a sophisticated massively parallel Bayesian fitting engine powered by Kraken petascale computing cluster, part of the XSEDE network.
Project | 01
Luminosity Dependence of Very Small Scale Clustering
We measure the very small-scale angular clustering of galaxies in volume-limited luminosity samples drawn from the SDSS DR7. We model this clustering using mock galaxy catalogues produced from the LasDamas simulations and the Halo Occupation Distribution (HOD) framework. In order to probe the spatial distribution of galaxies within dark matter halos, we adopt a flexible HOD that allows galaxies to have a biased density profile with respect to the dark matter. We find that luminous galaxies have a steep correlation function, and are thus more centrally concentrated in halos than the underlying dark matter. Lower luminosity galaxies, however, have s density profile that is consistent with that of dark matter.
Project | 02
Evolution in the Very Small Scale Clustering of Galaxies
In order to see if the trend of lower luminosity galaxies having less steep density functions continues to higher redshift, we also measure the projected correlation function of SDSS-III BOSS CMASS galaxies on similar scales. An evolution between the first and last redshift bin is not surprising. Fixed mass objects with changing number densities should be more clustered when measured at higher redshift - a trend we see.
Project | 03
LasDamas Simulations:
Adding Fiber Collisions
I added fiber collisions, an observational systematic that causes galaxies to not have measured redshifts, to the LasDamas simulations using a group finding algorithm that is meant to mimic the targeting of SDSS galaxies. We used the Conditional Luminosity Function (CLF) to create a flux limited sample in addition to the normal, volume limited, mock galaxy catalogues. I can then compare what the relationship between the assigned redshift of the galaxy and its true redshift.
AAS Poster |
Click for PDF (LARGE)
Just a sample of my work. To see more or discuss possible work >>