SURFS: Riding the waves with Synthetic UniveRses For Surveys

We present the Synthetic UniveRses For Surveys (SURFS) simulations, a set of N-body/Hydro simulations of the concordance Lambda Cold Dark Matter (Lambda CDM) cosmology. These simulations use Planck cosmology, contain up to 10 billion particles, and sample scales and halo masses down to 1 kpc and 10(8) M-circle dot. We identify and track haloes from z = 24 to today using a state-of-the-art 6D halo finder and merger tree builder. We demonstrate that certain properties of haloes merger trees are numerically converged for haloes composed of greater than or similar to 100 particles. Haloes smoothly grow in mass, V-max, with the mass history characterized by logM(a) proportional to exp [-(a/beta)(alpha)], where a is the scale factor, alpha(M) approximate to 0.8 & beta(M) approximate to 0.024, with these parameters decreasing with decreasing halo mass. Subhaloes follow power-law cumulative mass and velocity functions, i.e. n(> f) proportional to f(-alpha) with alpha(M) = 0.83 +/- 0.01 and alpha(Vmax) = 2.13 +/- 0.03 for mass and velocity, respectively, independent of redshift, as seen in previous studies. The halo-to-halo scatter in amplitude is 0.9 dex. The number of subhaloes in a halo weakly correlates with a halo's concentration c and spin lambda: haloes of high c and low lambda have 60 per cent more subhaloes than similar mass haloes of low c and high lambda. High cadence tracking shows subhaloes are dynamic residents, with 25 per cent leaving their host halomomentarily, becoming a backsplash subhalo, and another 20 per cent changing hosts entirely, in agreement with previous studies. In general, subhaloes have elliptical orbits, e approximate to 0.6, with periods of 2.3(-1.7)(+2.1) Gyrs. Subhaloes lose most of their mass at pericentric passage with mass loss rates of similar to 40 per cent Gyr(-1). These catalogues will be made publicly available.