Introducing SAGUARO -- Simulating IGM Evolution and Environments At High Resolution: Setup and First Results

Abstract

Small-scale physics in the intergalactic medium (IGM) plays a crucial role in shaping the progress of cosmic reionization and several high-redshift observables that probe this period. Several recent studies have characterized the complex, dynamical response of the IGM to reionization down to kilo-parsec scales, including its effect on observables such as the Ly$α$ forest. However, there has been no concentrated attempt to simulate and characterize these effects across the full parameter space of realistic large-scale IGM environments during reionization. To meet this need, we introduce the SAGUARO simulation suite, sub-titled ``Simulating IGM Evolution and Environments At High Resolution''. SAGUARO is a suite of over two hundred high-resolution, coupled radiative-hydrodynamics simulations of IGM gas dynamics during and after reionization. The suite spans a grid of photoionization rates, redshifts of reionization, and box-scale densities. We also simulate other physical effects, such as X-ray pre-heating, recombination radiation, baryon-dark matter free-streaming, and alternative dark matter cosmologies. Our suite includes box sizes of $2$ and $0.25$ $h^{-1}$Mpc, extending to volumes large enough to begin capturing halos above the atomic cooling limit and resolutions high enough to fully resolve the IGM Jeans scale in the cold, neutral universe. We present a detailed description of the setup and first results from SAGUARO, descriptions of the IGM gas dynamics and thermal structure, opacity, self-shielding properties, the effect of the IGM on the reionization photon budget, and the halo mass function, and Ly$α$ transmission properties. SAGUARO will help facilitate detailed studies of small-scale IGM structure and its effects that will help inform the next generation of reionization simulations and data interpretation.

Figures (31)

• Figure 1: Visualization of the gas density in slices through different simulations and redshifts from the "Core" set of saguaro simulations. The upper left panel (A) shows the simulation with $z_{\rm re} = 7$, $\delta/\sigma = 0$, and $\Gamma_{-12} = 0.3$ at $z = 5.5$, which we use as a reference point for comparison. Panel B shows the same slice at $z = 6.9$, highlighting the initially clumpy state of the IGM just after reionization, which is followed by considerable pressure smoothing. This comparison is made again for a lower $z_{\rm re} = 5$ in Panels D and E. Panel C shows that earlier reionization results in less surviving structure at fixed redshift. Panels F and J show the effect of increasing and decreasing $\Gamma_{\rm HI}$, respectively. Higher (lower) $\Gamma_{\rm HI}$ results in less (more) pronounced, compact density peaks due to the interplay between self-shielding and pressure smoothing. Panels G and H highlight the enhancement of structure formation in over and under-dense boxes, respectively.
• Figure 2: Gas temperature for the same snapshots and layout as in Figure \ref{['fig:example_density']}. The thermal structure proceeds from mostly uniform (B) just after ionization to a complex structure with small-scale fluctuations driven by pressure smoothing several hundred Myr later (A). These fluctuations disappear if reionization was early and happened long ago (C), and temperatures are higher and fluctuations stronger if reionization occurs later (D and E). The gas is hotter and displays stronger $T$ fluctuations if $\Gamma_{\rm HI}$ is higher (F and J). Regions with densities higher (G) and lower (H) than the cosmic mean display stronger and weaker $T$ fluctuations, respectively.
• Figure 3: Visualization of self-shielding properties of the Core suite. The maps show $\tau_{912}$ integrated along one axis of the box, with dark red regions intersecting self-shielding absorbers along the integrated axis. Self-shielding depends on photo-evaporation of self-shielding systems (A and B), the redshift of reionization (C-E). The strength of the ionizing background modulates the self-shielding density, with higher $\Gamma_{\rm HI}$ shrinking the sizes of absorbers (F and J). The abundance of systems massive enough to self-shield increases dramatically with box-scale density (G and H).
• Figure 4: Effect of other physical parameters (rows 4-8 of Table \ref{['tab:simulation_summary']}) on the gas density at $z = 5.5$. Panel A is the same as that in Figure \ref{['fig:example_density']}. Assuming Case A recombinations has almost no effect on the density field (B) as does a non-zero streaming velocity (F). X-ray pre-heating to $T = 10^3$ K (C) erases the smallest structures before reionization starts, and has a modest effect on the subsequent evolution. Free-streaming of DM particles (D) prevents these structures from forming in the first place, and has a larger effect than pre-heating for our choice of parameters. We see that the effects of a harder ionizing spectrum (E) and streaming velocities (F) are both small, similar to panel B.
• Figure 5: Gas temperature for the same snapshots shown in Figure \ref{['fig:example_density_phase_2']}. As expected, the models with case A recombinations (B) and streaming velocity (F) are very similar to the fiducial case, and even the X-ray heated model (C) displays only modestly less small-scale thermal structure. The WDM model (D) shows considerably less evidence of compression heating around expanding filaments, indicating that the complex thermal structure in our simulations is a characteristic feature of small-scale DM power. A harder ionizing spectrum leads to more heating (E), but the small-scale features of the temperature map are similar to the other CDM models.