The great divide between warm and cold gas kinematics in the early universe

Abstract

Galaxies with high redshift (z) are galaxies in the early universe, and they differ from local ones. They often show more irregular morphologies and significantly different kinematic properties, as turbulent motion dominates. These early galaxies have higher gas velocity dispersion (σ) and lower rotation-to-dispersion (vrot/σ) values. Recent findings from both the JWST and the Atacama Large Millimeter/submillimeter Array (ALMA) have challenged these observed trends. Recent surveys from JWST have found a higher number of disc galaxies than previous Hubble observations at the same redshifts. Furthermore, both JWST and ALMA have found an increasing number of dynamically cold (vrot/σ > 4) disc galaxies in the early universe, as well as (σ) at high redshifts being lower than expected from extrapolating from local surveys.

These differences between observations could arise for several reasons. Characterizing the kinematics of local galaxies is typically achieved through observations of emission lines from cold gas tracers. This is not possible for high-redshift galaxies as a very long integration time would be needed. Instead emission lines from warm gas tracers are used. ALMA has used observations of the [C II] 158 µm emission line to find dynamically cold discs in the early universe. This raises the question of whether the discrepancies between observations of cold and warm gas tracers are caused by the use of different.

To investigate this, we use the cosmological simulation MEGATRON. MEGATRON is based on the RAMSESRTZ code, uniquely it can follow the non-equilibrium chemistry of metals fully coupled to the radiative transfer. We analyze data in the range of 11 ≤ z ≤ 4 for 479 halos across 9 timesteps. We calculate gas rotation curves, velocity dispersions, and rotation-to-dispersion ratios, both weighted by flux densities, density, and the mass of the gas. For the flux densities, we use five different emission lines of both warm and cold gas tracers. By comparing these results, one can determine whether they behave systematically differently in the high-redshift universe and if this is what causes the different kinematic results observed between cold and warm gas tracers.

The analysis reveals that the choice of tracer has a significant impact on both the velocity dispersion and the rotation-to-dispersion ratio. For the velocity dispersion, warm gas tracers such as [O III] can be up to two times larger than that of the cold gas tracer CO, and have a rotation-to-dispersion ratio that is half as large. We also found a strong correlation between mass and dispersion values, with high-mass galaxies exhibiting higher dispersions. For individual galaxies, the kinematic values can change significantly over redshift. Warm gas tracers have a much higher variability in their dispersion values, while for rotation-to-dispersion ratios CO, [O II], and [O III] were the most variable ones. Factors such as mergers influenced these changes.

We concluded that cold gas tracers and warm gas tracers can yield significantly different values for both velocity dispersions and rotation-to-dispersion ratios, implying that observational effects do not solely account for the difference between high-redshift observations of cold and warm tracers, but also reflect a real systematic difference. Values for singular galaxies can change significantly over time, and different emission lines will also change by varying amounts, due to warm gas tracers reacting more strongly to factors such as stellar feedback. We concluded that cold gas tracers, such as CO, are better suited for kinematic observations than warm gas tracers, such as [O III]

Popular Abstract

Do we all live in a simulation? I can’t answer that question for you and me, but the galaxies I have analyzed certainly do. Galaxy formation and evolution is an important field in astrophysics, but also a complicated one. While we have many observations of galaxies, it would be helpful to compare this data with theories of how galaxies work. This is where cosmological simulations come into play. With simulations, we can program how we think all the different aspects of galaxy formation should work, let the simulation run, see how things change over time, and then compare it to observations. In this way, we can test different ideas and see what works best.

From the first galaxies to the present day, there has been a significant transformation in their nature. Since the universe is expanding, in the early universe, galaxies were closer together and things were very different! Galaxies would collide with each other more often, and this would affect many of their properties, such as the shape and how quickly stars form. The behavior of the gas in the galaxy would also change due to this, which is what we will look into in this thesis.

The gas velocity dispersion (σ) is a way of measuring how large the spread of velocities of the gas is; higher values mean a larger spread. Studies found that the σ values for galaxies decrease with time. Another impor- tant value is the rotational support (v_rot/σ), which shows whether the gas is supported more by rotation or gas pressure. Higher values mean the galaxy is dynamically cold and supported more by rotation. In contrast to velocity dispersions, the values for v_rot/σ increase with time. Recent observations from the ALMA and JWST have given different results. These telescopes have observed galaxies in the early universe to have lower velocity dispersion values than would be expected from local studies of galaxy kinematics observed through warm gas tracers, and have also observed an increasing number of galaxies with high v_rot/σ values in the early universe. These findings raise the question of whether the formation of disc galaxies in the early universe works differently from what we think and happens earlier than previously thought.

Different elements are used for different observations, with some tracing warmer gas and some colder gas. Cold gas tracers are more suitable for local observations, while warm gas tracers are needed to study galaxies at greater distances. However, this raises the question of whether the different behavior between more local galaxies and galaxies far away is real or caused by the fact that we use different gas tracers? This is the essence of what this thesis has explored.

We have used the cosmological simulation MEGATRON to explore this. We will use the data from the simulation to calculate both the velocity dispersions and rotational support, and examine the values obtained for galaxies of different masses and how they change over time.

We found that warm gas tracers have higher velocity dispersion values than the cold gas tracers, and this difference increases at higher masses. For v_rot/σ the cold gas tracers have higher values, there is however not as clear of an increase with higher masses. The warm gas tracers tend to change their velocity dispersions more dramatically over time compared to the cold gas tracers, though there wasn’t as clear a distinction for the v_rot/σ values.

• author: Larsson, Eddie
• supervisor: Corentin Cadiou; Oscar Agertz
• organization: Astrophysics; Department of Physics
• course: ASTM32 20251
• year: 2025
• type: H2 - Master's Degree (Two Years)
• subject: Physics and Astronomy
• keywords: galaxies: kinematics and dynamics, galaxies: high-redshift, galaxies: evolution
• report number: 2025-EXA249
• language: English
• id: 9208942
• date added to LUP: 2025-08-06 08:53:36
• date last changed: 2025-08-06 08:53:36