Protoclusters are the unvirialized progenitors of present-day galaxy clusters, and represent one of the most powerful laboratories for studying how the large-scale environment shapes galaxy evolution. A key question in protocluster research is whether the dense environment directly boosts or suppresses star formation in member galaxies.
These overdense regions are studied through star-forming galaxies such as Lyman Break Galaxies (LBGs) and Lyman-alpha Emitters (LAEs). Also, submillimeter galaxies (SMGs) are considered tracers of protocluster regions.
Using different hydrodynamical cosmological simulations, we have been studying the galaxy population in protoclusters across cosmic time. Here are some highlights of our latest results.
Galaxies in Protoclusters (Andrews et al. 2025)
In a recent paper by Andrews et al. (2025), we use TNG100 and TNG300 to study the physical properties of galaxies residing in protocluster environments. We select a set of the most massive clusters at z=0 and track the regions where they are placed back in time. In our analysis, we include Lyman-α emitters (LAEs), which are key tracers of large-scale structure and protocluster environments.
Our results show that protocluster galaxies follow the same star formation rate - stellar mass scaling relation as average field galaxies, but a larger fraction appear to have undergone major mergers in recent history, resulting in an enhanced star formation rate at approximately 60% above field levels, and a flatter distribution in both SFR and stellar mass.
Furthermore, we show that protocluster galaxies (including LAEs) begin to quench much earlier (z ~ 0.8–1.6) than their field counterparts (z ~ 0.5–0.9), highlighting the profound influence of large-scale environment on the overall formation history of galaxies. Our work provides key theoretical predictions for ongoing wide-field observational campaigns.
Submillimeter galaxies in cosmological simulation (Kumar et al. 2025, 2026)
SMGs are among the most intensely star-forming galaxies in the Universe, with star formation rates often exceeding hundreds to thousands of solar masses per year. They are predominantly observed at z ~ 2–3 and are thought to be the progenitors of today's massive elliptical galaxies. Despite their importance, reproducing their observed properties in cosmological simulations has remained a major challenge.
Source number counts of submillimeter galaxies in the FLAMINGO (solid blue), TNG300 (dashed orange), TNG100 (dashdotted green), and EAGLE (dotted pink) simulations. Shaded regions show 1σ errors for 100 realizations of the modeled submillimeter galaxy population
In Kumar et al. (2025), we test parametric models for SMGs across three state-of-the-art cosmological hydrodynamical simulations, EAGLE, IllustrisTNG, and FLAMINGO, and produce forecasts for the upcoming surveys. Our results show that FLAMINGO simulation successfully reproduces the observed redshift distribution and source number counts of SMGs without requiring a top-heavy initial mass function. We find that SMGs with flux densities above 1 mJy contribute up to ~27% of the cosmic star formation rate density at z ~ 2.6, consistent with recent observations, and that their abundance rises from z = 6 to z = 2.5, followed by a sharp decline toward lower redshifts.
Left column: Contribution of submillimeter galaxies with S_850 > 1 mJy to the cosmic SFRD in the FLAMINGO (solid blue), TNG300 (dashed orange), and TNG100 (dash-dotted green) simulations. Dotted curves represent the total SFRD in corresponding colors. The dotted black color shows the Madau & Dickinson (2014) fit to observations. The top panel shows predictions using the L21 relation, while the bottom panel using the H13 relation. The shaded regions around the SFRD of submillimeter galaxies represent the 1σ uncertainty from 100 realizations. For comparison, we show submillimeter observations. Right column: Ratio of submillimeter to total SFRD corresponding to the left column. We use the Madau & Dickinson (2014) cosmic SFRD to estimate the observational ratio. At peak activity, the submillimeter contribution in the FLAMINGO simulation is about 27% for the H13 Model.
Top: The percentage of SMGs and non-SMGs in different environments. Middle: Redshift evolution of the SMG and non-SMG fractions across cosmic environments, normalized by the total galaxy population in each environment. Bottom: Redshift evolution of the SMG-to-non-SMG ratio in different cosmic environments. SMGs and non-SMGs are modeled for galaxies with M∗ ≥ 10^9 M⊙.
Where do the most intensely star-forming galaxies in the Universe live, and how does their environment shape their evolution? In Kumar et al. (2026), we leverage the large-volume FLAMINGO cosmological simulation to map SMGs across the full cosmic web, from dense cluster cores to cosmic filaments and voids, across redshifts z = 1 to 4. Using the DisPerSE cosmic web finder, we define five distinct environments (inner cluster-halo, outer cluster-halo, inner filament, outer filament, and void/wall) at each redshift, tracking how both the SMG population and the large-scale structure evolve over cosmic time. Our results reveal that the fraction of SMGs in the inner cluster-halo environment declines from ~30% at z = 4 to ~3% by z = 1, as star formation is progressively suppressed in massive halos. Furthermore, the brightest SMGs (S₈₅₀ > 10 mJy) are found exclusively in inner cluster-halos, highlighting a strong connection between SMG luminosity and environmental density. In inner cluster halos, SMGs form two distinct populations: central SMGs dominate the high-mass end, and satellite SMGs contribute at lower masses. SMGs dominate star formation in dense environments, contributing up to 80% of the total star formation rate in inner cluster-halos at z = 4, but less than 50% in low-density regions, demonstrating that the dusty starburst phase is not a random event, but is intimately connected to the densest regions of the cosmic web.