| Speaker: | Takahiro Nishimichi (Kavli IPMU) |
|---|---|
| Title: | Modeling the nonlinear growth of large scale structure with perturbation theories and N-body simulations: implications to on-going and future surveys |
| Date (JST): | Wed, Feb 27, 2013, 16:00 - 17:00 |
| Place: | Seminar Room A |
| Related File: | 894.pdf |
| Abstract: |
In the standard \Lambda CMD cosmology, primordial Gaussian fluctuations generated during inflation are evolved driven mainly by gravity ending up with nearby large scale structure (LSS) observed in forms of galaxy clustering. Thus, the LSS is expected to have rich information ranging from the initial condition of the universe to the late time acceleration of the cosmic expansion. Unlike cosmic microwave background, however, LSS is a nonlinear process and thus more demanding to model accurately so as not to suffer from severe systematic error in cosmological tests using huge datasets expected from next generation survey projects. After briefly highlighting some promising cosmological probes using the LSS, I will review the current status and future prospects of the theoretical modeling of the LSS. I first discuss developments of "renormalized" perturbation theory techniques along with the roles played by numerical simulations. I then show the limitations of these PT techniques and introduce our new method that combines PTs with halo model in a consistent manner. I will then turn to two topics related to galaxy bias w.r.t. the underlying matter density field, through which I will show the difficulties and the possible advantages using galaxies as a cosmological probe. In the first topic, I will demonstrate our method to model the observed clustering patten of Luminous Red Galaxies including their velocity structure using N-body simulations. The second one is about new tests of primordial non-Gaussianities using the scale dependence in the biasing relation. I will show how we can distinguish primordial fluctuations sourced by multiple independent fields (e.g., multi-field inflation models)from models based on a single field. |
