Galaxies-Intergalactic Medium Interaction Calculation --I. Galaxy formation as a function of large-scale environment (0906.4350v1)

Abstract: [Abridged] We present the first results of hydrodynamical simulations that follow the formation of galaxies to z=0 in spherical regions of radius ~20 Mpc/h drawn from the Millennium Simulation. The regions have overdensities that deviate by (-2, -1, 0, +1, +2)sigma from the cosmic mean, where sigma is the rms mass fluctuation on a scale of ~20Mpc/h at z=1.5. The simulations have mass resolution of up to 10^6 Msun/h, cover the entire range of large-scale environments and allow extrapolation of statistics to the entire 500 (Mpc/h)^3 Millennium volume. They include gas cooling, photoheating from an ionising background, SNe feedback and winds, but no AGN. We find that the specific SFR density at z <~ 10 varies systematically from region to region by up to an order of magnitude, but the global value, averaged over all volumes, reproduces observational data. Massive, compact galaxies, similar to those observed in the GOODS fields, form in the overdense regions as early as z=6, but do not appear in the underdense regions until z~3. These environmental variations are not caused by a dependence of the star formation properties on environment, but rather by a strong variation of the halo mass function from one environment to another, with more massive haloes forming preferentially in the denser regions. At all epochs, stars form most efficiently in haloes of circular velocity ~ 250 km/s. However, the star formation history exhibits a form of "downsizing" (even in the absence of AGN): the stars comprising massive galaxies at z=0 have mostly formed by z=1-2, whilst those comprising smaller galaxies typically form at later times. However, additional feedback is required to limit star formation in massive galaxies at late times.

• The paper demonstrates that the global star formation rate density peaks around z≈2–3 with nearly an order-of-magnitude variation across different cosmic overdensities.
• The paper finds that specific star formation rates in halos are primarily driven by halo mass, especially notable around a circular velocity of 250 km/s.
• The paper links variations in the halo mass function to environmental density, indicating significant impacts on galaxy downsizing despite the absence of AGN feedback.

Overview of Galaxies-Intergalactic Medium Interaction Calculation: Galaxy Formation as a Function of Large-Scale Environment

This paper describes a set of hydrodynamical simulations, the Galaxies-Intergalactic Medium Interaction Calculation (GIMIC), hosted within the framework of the larger Millennium Simulation, aiming to explore galaxy formation and evolution as influenced by large-scale cosmic environments. This research investigates the dynamics within and between galaxies and the intergalactic medium (IGM), focusing on the formation and characteristics of galaxies across varied overdensity environments, ranging from rarefied cosmic voids to dense galaxy clusters.

Methodology

The paper utilizes the GADGET-3 simulation code, which is an evolution of the well-known GADGET-2, to capture the baryonic physics incorporated into the modeling of galaxy formation. The simulations incorporate sophisticated physics modules for star formation, stellar evolution, feedback mechanisms, radiative cooling, and chemodynamics. Key processes such as gas cooling, photoheating, and energetic feedback from supernovae and galactic winds are modeled without incorporating Active Galactic Nuclei (AGN) feedback, a potentially significant omission for impacting star formation, particularly in massive galaxies.

The simulations are conducted in five spherical regions, each approximately 20 Mpc/h in radius, extracted from the Millennium Simulation. These regions represent a spread of environmental overdensities of (-2, -1, 0, +1, +2)σ deviations from the cosmic mean at redshift z=1.5. This modeling approach allows researchers to capture and examine galaxy evolution dynamics in a realistic cosmic web context.

Key Findings

• Star Formation Rate Density (SFRD): The GIMIC simulations reveal that the global star formation rate density reaches its peak around z≈2-3, followed by a decline towards the present epoch. Notably, the SFRD shows significant dependence on overdensity environments, with variation between the most extreme regions by an order of magnitude across epochs.
• Specific Star Formation Rates (sSFR): The specific star formation rates in individual halos appear primarily determined by halo mass rather than the large-scale environment. Efficient star formation is most pronounced in halos with circular velocities around 250 km/s, a threshold where the impact of galactic winds diminishes.
• Environmental Influence and Halo Formation: Variations in the halo mass function across different environments account for the discrepancies in star formation efficiency. More massive halos, and hence galaxies, preferentially form in denser regions, influencing the overall SFRD for each environment.
• Galaxy Formation and Downsizing: The paper finds evidence of downsizing, where more massive galaxies form their stars earlier and quench their star formation sooner compared to lower mass galaxies. Despite the absence of AGN feedback, this downsizing effect emerges, suggesting a complex interplay of cooling, halo growth, and stellar feedback mechanisms.
• High Redshift Star Formation: At high redshift, the paper identifies massive, high-metallicity and compact galaxies forming first in the overdense regions—a phenomenon consistent with observations of early-universe massive galaxies.

Implications and Future Directions

The GIMIC simulations underscore the importance of environmental context in galaxy formation studies. The dependency of star formation and galaxy assembly on the local and extended cosmic environment highlights the need for simulations that encapsulate a wide range of overdensities to faithfully represent cosmic structure growth.

The paper indicates that while current simulations successfully capture many aspects of galaxy evolution, further refinement—particularly in the modeling of feedback processes such as AGN activity—is required to reconcile discrepancies in low-redshift galaxy properties. These findings pave the way for future work, exploring more sophisticated models of feedback, as seen in subsequent projects such as the Overwhelmingly Large Simulations (OWLS).

Overall, GIMIC provides an expansive and nuanced understanding of galaxy formation dynamics, linking small-scale physics with large-scale environmental factors, and sets a foundation for refined modeling approaches in cosmological simulations.