• KS
• Galaxy
• Feedback
• Datacube
• Cosmology
• BPT
• ALMA
• MUSE
• Outflow
• SFR

In the modern astrophysics, the most widely accepted cosmological model describing the evolution of the Universe is the ΛCMD cosmological paradigm, in which the geometry and expansion of the Universe depend on the constant Λ associated with the dark energy and the dark matter formed by non-relativistic particles (i.e., Cold Dark Matter). According to this model, galaxies would form when the gas cools down, and the first structures start forming by the gravitational collapse of gas in the potential wells of dark matter halos (White & Rees (1978)).

Due to the conservation of angular momentum during the collapse, the cool inward-flowing gas creates the rotating disk. Successively, stars form in the dense gas regions of this disk as a result of gravitational instabilities.

Although the proposed galaxy formation model based on the cooling process of gas in dark matter halos has provided a reasonable estimate of the typical mass scale of galaxies, the simple picture has a long-standing problem called “Overcooling Problem”. The cooling process is so efficient that the model predicts more galaxies with respect to those reported by current observing programs. Modern models and simulations of galaxy formation and evolution need to include ”feedback” processes to overcome the Overcooling Problem and reconcile observations to the theoretical expectations. Such feedback processes inject energy into the dark matter halos, increasing the temperature of the gas and preventing overcooling.

For galaxies in the low-mass regime, theoretical works expect that the feedback processes are caused by the radiation emitted by stars and shocks generated by the explosion of supernovae. As the number of stars and supernovae in a galaxy depends on the rate of formation of new stars, these feedback processes are usually dubbed “star formation feedback” (Dekel & Silk, 1986; Efstathiou, 2000). In massive galaxies star formation feedback is not sufficient to overcome the gravitational potential and regulate the cooling process. In this case models expect that the radiation emitted by the accretion disk of a super-massive black hole, also called the active galactic nucleus (AGN), plays a crucial role in solving the overcooling problem. In this case we refer to this process as “AGN feedback” (Benson et al.,2003).

In the last twenty years, several observing programs have reported evidence for both star formation and AGN feedback identifying radiation emitted by gas moving away from the galaxy at high velocity (up to 1000 km/s). Indeed radiation pressure from stars and AGN and supernovae explosion accelerates gas up to high velocity allowing it to escape from the gravitational potential of the galaxy and heat gas on a large scale through shocks. These disruptive events could inhibit the formation of new stars and explain the discrepancy between the prediction of galaxy formation models and the number of galaxies observed in the Universe. However, despite such evidence the efficiency of such feedback process in solving the overcooling problem is still debated and an open question of modern astrophysics. Therefore, a fundamental step to decipher the galactic evolution is to understand the competing roles of gas inflow and outflow (Veilleux et al., 2005).

This thesis aims to study the properties of the nearby galaxy NGC6810, located at ∼27 Mpc from us (z = 0.006541), and investigate the impact of the galactic outflow on the star formation activity. To reach our goal, we have exploited rest-frame optical and millimetre observations carried out with the spectrograph Multi Unit Spectroscopic Explorer (MUSE) mounted at Very Large Telescope (VLT) and interferometer telescope Atacama Large Millimetre Array (ALMA), respectively. In particular, we used MUSE data to analyze the optical transitions of atomic hydrogen (Hα at 6564.61 Å and Hβ at 4862.68 Å), oxygen ([OIII] at 4960.295 Å and 5008.240 Å), and nitrogen ([N II] at 6549.86 Å at 6585.27 Å) that trace the warm (T ∼ 10 4 K) ionised gas in galaxies. We have analyzed the rotational transition of the carbon monoxide (CO(2-1) at 2.6 mm) with ALMA to map the cold (T = 10 − 20 K) molecular gas in the galaxy. We stress that this is the first time that star-formation feedback has been studied in a local galaxy by using multi-wavelength facilities and combining the analyzing of ionized and molecular gas.

The identification of the galactic outflows and its impact on the host galaxy has been performed through the kinematic analysis of molecular and ionised gas traced by the aforementioned emission lines. The final calibrated data obtained with MUSE and ALMA facilities are a 3D array, called datacube, that has 2 spatial dimensions along the axis x and y, and 1 spectral dimension along the z-axis. In summary, a datacube stores spectral data associated with individual portions of the sky, usually called spatial pixels. The size of ALMA datacube of NGC6810 is (660 × 660) px × 240 spectral channels, corresponding to about 435000 spectra, and the size of MUSE datacube is (320 × 318) px × 1635 spectral channels corresponding to about 90000 spectra.

Initially, we have developed an algorithm to automatically perform the Gaussian fitting on the emission line data for all spectra of ALMA and MUSE data. The developed algorithm has enabled us to identify and fit the emission line over the field of view and take into account fake signals due to noise fluctuations in those spectra where the signal-to-noise ratio of the emission lines is comparable to the noise level.

The results of the fitting procedure on the data are the kinematics maps: flux, velocity, and velocity dispersion maps. The flux map gives us information about the bulk of the emission of the targeted line, and it corresponds to the integrated flux of the Gaussian profile in each spatial pixel. The velocity map reports the centroid of the line in each pixels and can be used to study the motion along the line of sight due to Doppler effect. The velocity dispersion map corresponds to the width of the emission line and provides information on the random motions of the gas in the galaxy. Here we summarise the results from the kinematics analysis of the ALMA and MUSE observations:
• The gradient of the velocity maps indicates a galaxy in which the gas in the South of the galaxy is receding along the line of sight with a velocity of ∼300 km/s while the gas in North side is approaching with a velocity of ∼ −300 km/s. The velocity gradient pattern is consistent with what is expected from a classic rotating disk. However, we have identified regions of the galaxy where both velocity and velocity dispersion are not in agreement with the rotating disk kinematic. Such discrepancies in terms of kinematics indicate the presence of galactic outflows in the galaxy.
• The disk inclination of the galaxy, which is fundamental to determine the dynamics of the rotating disk, was recovered by applying the public code kinemetry algorithm on the velocity and dispersion map of molecular gas. We have estimated a disk inclination i ≃ 60 ◦ (and position angle P A ≃ 5 ◦ ).
• The average velocity dispersion (i.e. random motion intensity) of the cold molecular gas is 20.8±0.1 km/s , while the average velocity dispersion of the warm ionised gas is 75±11 km/s. In those regions where we have indetified the emission of the outflowing gas the velocity dispersion of the ionized gas is higher than 140 km/s, which is 2-3 times higher than the average value.

Once we have obtained the velocity map corrected for the inclination of the galaxy, we have estimated the average velocity and the velocity dispersion as a function of the distance from the galaxy center. The radial velocity profile results in a typical galaxy rotation curve, with a steep rise at small radii, a peak at ∼ 180 km/s, and a flat profile for larger radii. The radial velocity dispersion profile suggests that the kinematics of the central regions are more chaotic with a velocity dispersion of 38 ± 8 km/s (from r = 0 kpc to r = 1 kpc), while the intensity of random motion decreases at larger radii with a velocity dispersion of 17 ± 5 km/s (from r = 1 kpc to r = 3.1 kpc).

The ratio between the velocity curve and the velocity dispersion is a useful diagnostic to discriminate if the gravitational stability of the galaxy is supported by the random motions (v/σ < 2) or by the rotation (v/σ > 2). In average NGC6810 appears to be rotational-dominated with a v/σ = 8.7±0.5 that is consistent with the typical values observed in local spiral galaxies (v/σ ∼ 10; Swinbank et al. 2017; Harrison et al. 2017; Di Teodoro et al. 2016).

Another useful diagnostic to analyze the impact of the outflow into the galaxy environment is the Toomre parameter (Q), which is a criterion related to the gravitational stability of the gas and valid for disk galaxies, such as NGC6810. The Toomre stability parameter is proportional to two terms, the epicyclic frequency and velocity dispersion of the gas, that are related to the intensity perturbation with respect to a normal orbital motion, and it is inversely proportional to the gas mass density. Q ≥ 1 means that the gas is stable against gravitational collapse (the disk is stabilized by differential rotation on large scales and by pressure gradients at smaller scales), whilst Q ≤ 1 means that the rotating disk becomes unstable for axisymmetric perturbations because the pressure due to the velocity dispersion and the differential rotation cannot stabilize the gravity perturbation. We have thus computed the Toomre parameter converting the CO(2-1) flux map into a gas mass density map of NGC6810. We have found that Q < 1 in most of the galaxy, suggesting ongoing star-formation activity in NGC6810. Only a few central regions have revealed Q > 1 probably due to the impact of the galactic outflow.

Another possible method to investigate the impact of feedback in galaxies is to study the star- formation activity in galaxies.
Although star formation is a complex process, several observing programs have claimed that the formation of new stars is mainly driven by the gravitational collapse of the gas and thus of the gas mass surface density. Schmidt 1959 and Kennicutt 1989 found for the first time that the star-formation rate (i.e. number of stars formed per year; SFR) surface density depends on the gas surface density. Therefore the last part of the thesis aims at estimating the number of stars produced per year in NGC6810 and verifying if this is consistent with what is expected from the empirical relation called Kennicutt-Schmidt law or not.

The intensity of Balmer transitions are good tracers of star-formation activity if the energy absorbed by the atomic hydrogen is due to the radiation emitted by young stars (Diamond-Stanic & Rieke, 2012). The Balmer line cannot be used as SFR tracer when there is an AGN in the galaxy and its radiation intensity is larger than that of young stars. To distinguish the two cases, we have used the so-called BPT diagram (Baldwin et al. (1981)) that exploits the [NII]6584/Hα and [OIII]5007/Hβ line ratio to distinguish between the ionization mechanism of gas. Our analysis has shown no signature of AGN ionization, and the radiation absorbed by the gas is mainly associated with young stars.

We have therefore used the Hα emission line to determine the SFR in the galaxy. As the emission of rest-frame optical lines might be altered by the presence of dust along the line of sight, we have initially determined the dust attenuation by using the theoretical Hα/Hβ flux ratio in the absence of dust. After the correction of Hα emission for dust attenuation, we have determined the SFR surface density in every pixel of the field of view and compared it with the surface gas mass density determined by CO(2-1) line. Despite the luminosity-weighted SFR and gas mass density being in agreement with what is expected from the Kennicutt-Schmidt law, we have found that some regions of the galaxy deviate from this relation. Some regions are located above the empirical relation and this indicates that the SFR is enhanced. On the other hand, about 80% lie below the relation, probably due to the feedback mechanism.

In conclusion, we have found that the galactic outflow does not have a large-scale impact on NGC6810 galaxy as both gas kinematics and star-formation activity are little affected by feedback mechanism. However, this spatial resolved study has revealed that the star-formation activity in some regions of the galaxy is regulated by the outflows, and the formation rate of stars in these regions is smaller than what is expected from empirical relations. Finally, this study shows that the kinematics analysis of nearby galaxies, such as NGC6810, using different instruments to map both the warm and cold gas phases yields a better understanding of the impact of outflows in galaxies.