SurveySim

an MCMC-based code to constrain luminosity function evolution

Installation and use instructions are given in the User's Manual.

You can download the latest version of the code at GitHub.

If you use this code in full or in part, please cite: Kurinsky et al. 2016, ApJ, submitted

Overview

We present SurveySim: a new open source Markov Chain Monte Carlo (MCMC), galaxy evolution fitting and simulation package. The code is designed to be easily adaptable with the luminosity function and SED model assumptions fed into it externally. Here we adopt our state-of-the-art IR SED template library which includes templates for star-forming galaxies, composites, and AGN as well as their redshift and/or luminosity evolution. The code converges on a best-model by simulating the density of galaxies in diagnostic color-magnitude plots. The figure below shows an example of how this works in the case of the HerMES 250um-selected sources in the COSMOS field (Oliver et al. 2012).

Example results

The MCMC fit leads to posterior probability distributions on all parameters. When more than one dataset and/or diagnostic plot is used we have multiple probability distributions for the same parameters. These are multiplied together to obtain joined posterior probability distributions (red contours in the plot below) which necessarily show tighter constraints. In essence, this is a manifestation of the Bayesian concept of prior -- the more information we have to constrian a problem, the tighter our constraints on the model parameters. The best-fit parameters arrived at by SurveySim translate to direct predictions for the evolution of the IR luminosity function, the AGN luminosity function, IR number counts, redshift distributions and Cosmic Infrared Background (CIB) intensity. The results agree well on all fronts as presented in Kurinsky et al. (2016). The figure below show the results for the IR luminosity function where the multiple curves are the results of multiple SurveySim runs showing that the results are robust.