Research

This page describes a subset of the various projects I am currently involved in, or that have recently been completed…


Dynamical Friction

 

 

The movie shows how a single near-resonant horse-shoe orbit (in blue) changes its energy as it approaches a massive object (in black). When integrated over a sufficiently long period, the net energy gain/loss is zero (due to conservation of the Jacobian energy E_J). Dynamical friction arises if there is an assymetry in the number of stars that loose and gain angular momentum (due to a gradient in the distribution function), which then causes the orbit of the perturber to change (loose orbital energy), which in turn causes an evolution in the resonant frequencies. We are examining various aspects of dynamical friction, including core stalling, using this resonant picture of dynamical friction.


Galaxy Assembly Bias and Galactic Conformity

screen-shot-2016-10-13-at-8-48-02-pmWe are using a variety of different approaches to study whether and how galaxies display assembly bias, in that their properties depend on the assembly history of their host halo at fixed halo mass. Using mock galaxy catalogs, some with and some without galaxy assembly bias, we are testing how this phenomenon manifests itself observationally. In addition to having an impact on the clustering of galaxies, we have also shown that 2-halo galactic conformity is a smoking gun of galaxy assembly bias.  Using Decorated HODs, we have detected evidence for assembly bias in the clustering of SDSS galaxies.  We also tested to what extent group finders can be used to probe galactic conformity, and have shown that galactic conformity has its origin in the conformity in halo assembly histories.


Precision Cosmology with Galaxies

screen-shot-2013-10-30-at-8-38-32-amUsing a combination of galaxy clustering and galaxy-galaxy lensing data from the Sloan Digital Sky Survey, we have put tight constraints on the cosmological matter density, Ωm, and the power spectrum normalization, σ8.  Whereas the results are in good agreement with those from the 7-year data release from WMAP, they are inconsistent with the more recent Planck constraints. We are currently investigating the cause of this discrepancy, focussing on assembly bias and massive neutrinos. The results are published in a series of three papers:  Paper I  presents the models,  Paper II presents a Fisher matrix analysis, and  Paper III presents the actual results with the application to the SDSS data.


Statistics of Dark Matter Substructure

screen-shot-2016-10-13-at-8-06-55-pmWe have developed a new model for the substructure of dark matter haloes. Using accurate halo merger trees, and an analytical model for the average mass loss rate of subhalos as they orbit their host, the model yields the subhalo mass and velocity functions for subhalos of all different orders. Comparison with simulation results shows that the model is extremely accurate. We have used this model to revisit the too-big-to-fail problem, and to investigate to what extent the abundance of subhaloes obeys Poisson statistics. We are also using it to investigate subhalo disruption. I also write a paper on the segregation of dark matter substructure, showing that subhaloes are segregated by a variety of properties, in a complex way.


Coming of Age in the Dark Sector

screen-shot-2016-10-13-at-8-37-07-pmUsing a combination of semi-analytical techniques and numerical simulations we have studied  how dark matter halos  assemble their mass and grow their central potential wells. We have shown that the average mass accretion histories of dark matter haloes have a truly universal form, which we use to develop a new and improved analytical model for the mass accretion rates. Click here for the paper.

A Fortran code to compute the average and median mass assembly history, M(z)/M0, and potential well growth history, Vmax(z)/Vvir,0, for a dark matter halo of mass M0 at redshift z0 in a LCDM cosmology is made publicly available. In addition, the code also computes the NFW halo concentration parameter, c, of the main progenitor as function of redshift, as well as the average and median mass accretion rates.


Constraining Halo Mass with Satellite Kinematics

The kinematics of satellite galaxies can be used to probe the potential of their host halo. We use this technique to constrain the galaxy-dark matter connection. We have developed new techniques to combine satellite-weighting and host-weighting to infer not only the mean, but also the scatter of the relation between halo mass and galaxy luminosity or stellar mass. We have applied our methods to both 2dFGRS and SDSS. We have shown that red centrals reside in more massive halos than blue centrals of the same stellar mass, a result that has recently been confirmed using galaxy-galaxy lensing.

clickMore recently, we developed a novel, Bayesian hierarchical approach to satellite kinematics, called Baselisk. This method has the advantage that it leaves the data in raw form, and can be applied to the full flux-limited sample (as opposed to a volume-limited sample as before). We are in the process of applying Basilisk to SDSS data in order to constrain galaxy asssembly bias and cosmological parameters.