The Active Galactic Nuclei (AGN) population can be represented by two main categories. In the first category, Quasars and Seyfert galaxies, sources of high bolometric luminosities, capable of generating winds through radiation pressure due to its accretion rate close to Eddington, usually found in wide angle outflows, are commonly distinguished as radiative-mode (or sometimes also called as quasar-mode) AGN. In the second category, the so called radio-mode AGN (or jet-mode, or kinetic-mode), the central engine launches powerful collimated jets of relativistic particles accelerated in the inner regions of the accretion disk due to its intense magnetic fields. The origin of the difference between these two categories is believed to happen within the accretion disk structure and internal thermodynamics, and the resulting mass accretion rates [1] (and references therein).

Radio emission is one of the most distinctive tracers of nuclear activity in galaxies. The radio jets can be extremely powerful, extending up to Mpc scales and producing strong feedback in the surrounding medium of early-type galaxy hosts in the center of galaxy clusters ^[2][3][4][5][2,3,4,5]. But do radio jets produce feedback in its host galaxies? This is the central topic of this short review of the interplay between radio-emission from AGN and its host galaxies. Fast (≳1000 km s−1

) H i 21 cm absorption outflows observed using WSRT ^[6][7][6,7] and X-ray detections related to shocks signatures due to the jet-gas interactions, e.g., observed with Chandra ^[8][9][8,9] illustrate how multi-wavelength analysis is fundamental to properly characterize the role of radio feedback.

With the emergence of Integral Field Spectroscopy (hereafter IFS) optical instruments in the past years, such as the Gemini GMOS [10], VLT MUSE [11] and GTC MEGARA [12], along with deep observations in other wavelength instruments such as ALMA [13] and most recently JWST, recent studies of outflows in local AGN have been able to characterize and constrains the outflow properties, resolving their kinematics and determining its extent within the host galaxies. High resolution spectra allow to extract information in the ionized and molecular gas phases, such as velocity dispersion and emission line ratios, helping understand how gas excitation works. This has been proven useful in radio-loud AGN, characterized by shock-driven outflows due to jet-gas interactions.

The relativistic jets present in radio-loud AGN, when coupled with the circumnuclear gas either in the narrow-line region (within the radius of influence of the AGN ionizing radiation) or further out in the ISM (within the inner few kpcs), can be responsible for the heating and acceleration of this gas resulting in outflows. These outflows are usually detected in the turbulent ionized emitting gas by measuring its kinematics and isolating the kinematic components linked to the AGN feedback (e.g., ^[14][15][16][60,61,62]). The superposition of one or more kinematic components to that originating in gas rotating in the galaxy potential can result in a complex ionized gas spectrum, with many components, making the decomposition process hard to constrain ^[17][63]. With the advent of better spatial and spectral resolutions IFUs, the number of resolvable kinematic components has increased, and the interpretation of these components and their origin has also increased in complexity.

The case of the Seyfert 2 galaxy NGC 7130 is a clear example of such complexity, as discussed in Comerón et al. ^[18][17], using MUSE narrow-field adaptive optics observations. Even though the radio jet in this Seyfert galaxy is not powerful, one or maybe two outflowing ionized gas components seem to be interacting with the radio jet, as revealed by the very detailed emission-line decomposition performed by the authors, which comprises a total of nine components, with six being connected to outflows. It is important to note that the fitting of several components should be statistically justified so that the decomposition does not introduce artificial components into the emission-line fit.

Usually a rotation component is observed in the gas velocity field of radio-loud AGN, but alongside more complex kinematics, as illustrated in the study performed in the MURALES survey ^[19][20][64,65], in which a sample of 37 radio galaxies from the Third Cambridge Catalog (3C) were observed with the MUSE IFS. The ionized gas velocity maps of these radio galaxies appear to show some rotation, but present a much more disturbed pattern than the expected “web diagram” characteristic of the isovelocity curves of ordered rotation. Although these complex kinematics may not be the case for all radio-loud AGN, they are commonly observed, as several other studies of individual galaxies have also shown, such as in 4C +29.30 ^[21][16], Cygnus A ^[22][66] and UGC 05771 ^[23][67], and are usually caused by the interaction between the radio jet and the circumnuclear gas.

An alternative method to track different kinematic components is to probe the velocity field along channel maps, which is possible when using IFS observations. Arp 102B is one case for which the channel maps can aid in the interpretation of how the radio jet interacts with the surrounding emitting gas.Hα channel maps, indicate that a spiral arm-like structure correlates spatially with the radio jet, and the emitting gas is observed both in blueshifted and redshifted velocities. Researchers  have interpreted that these velocities trace the outflowing gas pushed by the radio jet oriented very close to the plane of the sky. The channel maps shows emission from the “walls” surrounding the outflow being pushed aside, seen in both blueshifted and redshifted velocities.

When the jet collimation is lost due to interaction with dense gas or jet precession, sometimes a gas bubble is formed, and the inflation of the bubble by the outflow also impacts the ISM. A nuclear starburst leading to numerous supernovae explosions can also produce outflowing bubbles from galaxy centers. The bubble serves as a shock front between the outflow and the ISM gas, and when it erupts several galaxy properties can be altered, including its chemical composition. Perhaps the most famous case is of the own galaxy. Bipolar bubbles at the Galaxy center have been observed in several wavelengths, including in the radio at 1.3 GHz with MeerKAT ^[24][71], and it is still debatable whether these bubbles originate from AGN or star formation feedback. A similar case where there is ambiguity in the feedback origin is NGC 3079 ^[25][72]. While studying the AGN-starburst composite galaxy NGC 6764, Hota and Saikia ^[26][73] also found an inconclusive origin for the feedback, and compared it to 10 other sources. These are all associated with AGN activity, but the radio and optical emission could be also affected by central starbursts. Pure starburst sources do not seem to present the same structures, indicating that AGN must be present to produce the large-scale bubbles.

Besides the effect on the ionized gas phase, that trace the hot and turbulent outflows, molecular gas can also show signatures of jet-gas interaction. As the ionized gas represents only a fraction of the total gas mass, which is dominated by the molecular gas in the galaxy inner regions, the outflow gas mass should also be dominated by the molecular gas, specially in low AGN bolometric luminosities ^[27][74]. Mass outflow rates observed in molecular gas are about 2–3 orders of magnitude higher than those traced in the ionized gas phase (∼1000M⊙ yr−1 as compared to ∼1–10 M⊙ yr−1, respectively) in AGN with bolometric luminosity ∼1046 erg s−1 (e.g., ^[28][75]). As a comparison, star-formation rates of 10–100 M⊙ yr−1(see Figure 3 in Fiore et al. ^[28][75]) are estimated for AGN in the same luminosity range, displaying that these outflows must indeed originate from AGN activity.

As shown in a study of the galaxy IC5063 by Dasyra et al. ^[29][76], by estimating the internal and external pressures of the molecular clouds one can infer the impact of the radio jet in the star formation processes within the galaxy, where both suppression and enhancement can simultaneously happen. In NGC 613, disturbed gas can be traced in the nucleus due to molecular outflows mainly boosted by the radio jet ^[30][77]. While modeling the molecular gas outflows in the young radio galaxy 4C 31.04, Zovaro et al. ^[31][78] could reproduce the observed kinematics by assuming that the gas is being pushed and expanded by the radio jet in an energy bubble while generating shocks within this bubble, originating the observed H2 emission. As observed (and discussed above) for the ionized gas phase, rotation is also usually observed in the molecular gas phase, but with distortions commonly found in the inner few hundred pc, as analyzed by Ruffa et al. ^[32][79] in a sample of six low excitation radio galaxies using ALMA observations. However, in the case of this sample, the authors interpreted that these non-rotating components are related to inflowing gas, since they seem to be correlated with structures such as spirals or bars, known for being mechanisms causing the gas to lose angular momentum to reach the nucleus and feed the SMBH. This illustrates that other signatures besides the gas velocities, should be used in the search of outflows, such as velocity dispersion and line ratios and their relation to the observed kinematics. Other cases of galaxies presenting jet-gas interaction signatures through the analysis of their molecular gas kinematics include NGC 1377 ^[33][80], ESO 420-G13 ^[34][81] and NGC 7319 ^[35][82], among others.

Detailed 3D hydro-dynamical simulations indicate that indeed the gas can be perturbed by the radio jet within the host galaxy. Not only is the jet responsible for disturbing and shaping the emitting gas distribution, as discussed in Wagner and Bicknell ^[36][83], the path taken by the jet and its morphology is also affected by the inhomogeneity of the gas density. In this scenario, the jet collimation, its power, and how it spatially couples with the gas are important parameters to estimate the feedback energetics. One example of such simulations is displayed, where the gas density distribution is shown while the radio jet evolves with time: the jet carves its way through the gas, pushing it both aside and forward to larger distances while heating it and possibly creating shock ionization. How easy and straight is the path of the jet through the gas depends on its density and porosity distributions.

During the interaction between the relativistic jet and the surrounding gas, gas ionization through fast shocks can occur, with the excited gas emitting characteristic spectra that provide information about the physical parameters of the shock ionization. Observed line ratios in the narrow-line region (NLR) indicating the presence of shocks usually lie in the Low-Ionization Emission Line Region (LINER, ^[37][88]) part of optical diagnostic diagrams, such as the well known BPT diagrams ^[38][89], depending on parameters such as the shock velocity and gas density ^[39][40][90,91]. Although photoionization models have a considerable overlap with shock models in the BPT diagrams, high values of low-ionization line ratios such as [N ii]/Hα and [S ii]/Hα are usually interpreted as tracers of shocks when gas emission is observed due to jet influence.

The presence of shocked gas also commonly correlates with broader emission lines, which trace the increase of gas turbulence. This is the case of 3C 293 ^[16][62], where jet-driven outflows are detected along the radio emission and two kinematic components are needed to reproduce the ionized gas emission-line profiles. A clear increase in the [N ii]/Hα and [S ii]/Hα line ratios is observed in the regions where a broader component is required to reproduce the emission lines (see Figure 6 in ^[16][62]), indicating that the shock velocity increases in the regions where the highest ratios are observed.

Shock excitation has been found in several other objects presenting radio jets, such as M51 ^[14][60], Coma A ^[41][93], the Beetle galaxy ^[42][94], PKS B1934-63 ^[43][95], 3C 320 ^[44][96], 4C 31.04 ^[31][78], 3C 433 ^[45][97], J1220+3020 ^[46][98], Cygnus A ^[22][66], among others. Shock excitation has been found not only in the ionized gas phase, but also sometimes in molecular or neutral gas. High gas velocities of up to ∼1000 km s−1, or even higher, are commonly found in these sources, and an increase of velocity dispersion is also a good tracer of the regions presenting shocks.

After characterizing gas outflows due to the AGN jet feedback, it is useful to estimate its kinetic power in order to quantify its impact on the host galaxy and compare it with models and other sources of feedback. The kinetic power of the outflow attributed to the kinematic disturbance produced by the radio jet can be calculated via ^[47][99]:

The calculations above are not restricted to radio-loud AGN, but are general to any type of outflow. For the ionized gas, the gas density is usually estimated using the [S II]λ6717,31Åline ratio, and the filling factor can be obtained using a measured hydrogen emission-line luminosity, such as Hα ^[48][100]:

Scaling relations using the estimated mass outflow rates M˙ and outflow kinetic power E˙ of AGN-driven feedback have been the object of studies in the past few years. Fiore et al. ^[28][75] investigated the relation between the bolometric luminosity and both the mass outflow rate and outflow kinetic power for 94 AGN covering a large range of redshifts (from local Universe up to z∼6) compiling observational data from the literature and homogeneously calculating the mass outflow rates and powers. Ionized and molecular gas show clear correlations between Lbol and both M˙ and E˙, with somewhat higher M˙ and E˙ for the molecular gas, that seems to reach the coupling efficiency in the range 1–10%Lbol, as required in models (e.g., ^[49][50][101,102]) to have a significant impact on the host galaxy (e.g., by pushing the gas out of the galaxy and halting star-formation) while for the ionized gas only ≈30% of the galaxies reach this efficiency.

More recent studies (e.g., and references therein ^[51][52][103,104]) on kinematic feedback via ionized gas outflows have obtained higher gas densities and resulting lower powers than those estimated in Fiore et al. ^[28][75] for most AGN. In some cases, like in the radio-galaxy 4C +29.30 ^[21][16], the outflow power can reach a few percent of the AGN power LAGN, implying strong feedback, but, in most cases, this power is below 1%LAGN. The kinematic feedback is usually present, heating and disturbing the gas – a “maintenance mode feedback", but not high enough to push the gas out of the galaxy or immediately halt star formation. On the other hand, AGN feedback occurs not only via outflows, with recent model estimates suggesting that at most 20% of the AGN feedback is in kinetic form (e.g., ^[53][105]).

The relations between E˙ and LAGN and M˙ and LAGN discussed above seem to apply both to radio-loud and radio-quiet AGN. In the case of the latter, Villar Martin et al. ^[54][106] has shown that, for a sample of 13 nearby (z<0.2) radio quiet QSOs, most of which showing signatures of interactions, 10 had extended radio emission. In addition, they found this radio emission was correlated with the optical Hα emission, indicating jet-gas interaction. Thus radio-mode feedback is also present in radio-quiet AGN, when the jet is spatially coupled with the ISM gas.