Vorträge, Seminare, Ereignisse

Abstract
TBD

Abstract
Globular clusters (GCs) are old, densely-packed groups of thousands to millions of stars and have long been hypothesized to be the building blocks of the Milky Way’s stellar halo due to their high star formation efficiency. Excitingly, early results from JWST suggest that clustered star formation may also be dominant in other galaxies at high redshifts. Although the Milky Way is a perfect laboratory in which to study GCs due to our ability to resolve individual stars, the clusters we observe today are likely to be only the most massive survivors of what was once a much larger population of Galactic GCs. In this talk, I will summarize how we can trace the contribution of the Milky’s Way’s disrupted GCs to the Galaxy by identifying their most unusual stars, those with such high nitrogen-to-oxygen ratios ([N/O]) that they can reliably be identified as having originated in a GC even long after escaping. I will then summarize our recent work to identify which of these anomalous GC-origin stars in the field may have been born in surviving Milky Way GCs by associating stars and clusters via their integrals of motion. Although in bulk it appears that the high-[N/O] stars originated in inner Galaxy GCs, specific star-to-cluster tagging is complicated by the fact that integrals of motion are likely not conserved due to the effects of dynamical friction and the Galactic bar. These results have interesting implications for the feasibility of chemodynamical studies of the Milky Way halo.

Abstract
To arrange a visit with the speaker during the visit, please contact their host: Anna de Graaff

Abstract
Jellyfish galaxies provide a striking view of the interaction between galaxies and their surrounding environment. As galaxies move through the intracluster medium, ram-pressure stripping can remove their interstellar gas and produce extended, asymmetric tails. These structures offer a unique opportunity to study how environmental processes shape the evolution of galaxies, yet the physical connection between the properties of the host galaxy and those of its stripped material remains poorly understood. In this work, we present a new framework to identify and characterize ram-pressure-stripped gas in galaxies from the IllustrisTNG50 cosmological simulation. We identify Jellyfish galaxies using the MARK–log param- eter space, which provides a physically motivated characterization of galaxies dominated by ram pressure. We then extend the phase-space approach of Min et al. (2019) by incorporating the local gas density, allowing the stripped material to be identified on a particle-by-particle basis and its physical properties to be quantified. This approach enables us to move beyond a simple Jellyfish/non-Jellyfish classification and investigate the physical properties of the tails themselves. In particular, we explore the correlations between the properties of the host galaxies, their interaction with the intracluster medium, and the resulting tail properties, including their extent and gas content. By connecting the conditions experienced by a galaxy with the structure produced by ram-pressure stripping, this framework provides a new way to investigate the physical mechanisms governing the formation and evolution of Jellyfish tails.

Abstract
The WST project aims to design and construct an innovative 10-metre class wide-field spectroscopic survey telescope (WST) in the southern hemisphere. It will feature the parallel operation of two cutting-edge instruments: a high-multiplex (32,000), large field-of-view (3 square degrees) multi-object spectrograph (MOS) operating in both low- and high-resolution modes, and a giant panoramic integral field spectrograph (IFS). WST’s ambitious top-level requirements place it well ahead of all existing and planned facilities. In its first five years of operation, the MOS is expected to observe 300 million galaxies, 25 million stars at low resolution, and 2 million stars at high resolution, while the IFS will deliver 4 billion spectra. These capabilities will enable transformative science across a wide range of astrophysical domains. A key strength of WST lies in the synergy between its MOS and IFS instruments, offering highly complementary spectroscopic survey modes. This dual approach is central to the project’s scientific ambition. WST is envisioned as the next major facility for ESO following the ELT, and a proposal will be submitted in response to the upcoming ESO Call for Ideas in 2027. The project is backed by a large consortium of leading institutes from European countries and Australia, with deep expertise in astronomical instrumentation and survey science. To arrange a visit with the speaker during the visit, please contact their host: Dylan Nelson

Abstract
The Vera C. Rubin Observatory has reached an important milestone by starting regular operations as the Legacy Survey of Space and Time (LSST). However, turning the alert stream into scientific discoveries still requires methods beyond pure machine learning, and the stream of alerts relayed via broker teams still needs to be vetted by observers who are less robotic and more human. In this context, we report on the first year of operations of one important aspect of our in-kind contribution: our LSST TVSSC Galactic Science Observing Priority Manager. We provide insights into the reasoning behind combining multiple classifier probabilities, catalogue queries, and time series photometry. We summarize how we have used the first identified targets for immediate follow-up observations with the Las Cumbres Observatory global network of telescopes, and we discuss which functionalities will be developed in future. From a scientific perspective, we will discuss the relevance of microlensing observations in the bulge fields of the Nancy Grace Roman Space Telescope and beyond.

Abstract
The PHANGS-HST and JWST Treasury Programs have produced the most extensive inventories to date of optically visible and dust-obscured star clusters across nearby galaxies. A coherent picture is emerging of the earliest phases of star formation, built from complementary studies that consistently point to short dust-clearing timescales (~3 Myr), in agreement with earlier optical studies. Together, these results underscore the central role of early, pre-supernova feedback in regulating the star formation cycle. Drawing on the statistical power of tens of thousands of clusters spanning dozens of galaxies, I will synthesize what PHANGS and related programs have revealed about cluster evolution, feedback, and the use of clusters as “clocks” for timing key phases of star formation. Looking ahead, I will highlight work on the Habitable Worlds Observatory, NASA’s next flagship astrophysics mission after Roman, which will operate as a true “Super-Hubble,” with UV–optical capabilities to transform our understanding of star formation and fields across astrophysics. To arrange a visit with the speaker during the visit, please contact their host: Ralf Klessen

Abstract
Little Red Dots (LRDs) are compact sources at z > 5 discovered with JWST, characterized by broad Balmer emission lines, narrow absorption features, and strong Balmer breaks. These signatures point to dense, optically thick hydrogen gas surrounding rapidly growing black holes, but their interpretation depends critically on how Balmer photons propagate through this gas. In this talk, I will present the formation of Balmer lines in LRDs via scattering processes using 3D Monte Carlo radiative transfer simulations. I will first discuss how various scattering processes reshape hydrogen-line profiles and ratios, demonstrating that broad and asymmetric Balmer lines do not necessarily trace gas kinematics alone. I will then focus on modeling the Halpha and Hbeta profiles of LRDs observed with JWST/NIRSpec and constraining the n=2 hydrogen column density from the Balmer break. The observations reveal a fundamental puzzle: Hbeta absorption is systematically deeper than Halpha, despite the substantially larger scattering cross-section of Halpha. A homogeneous medium cannot reproduce this ordering or simultaneously satisfy the column-density constraints from Halpha, Hbeta, and the Balmer break. We show that these tensions are resolved by a multiphase medium consisting of dense, optically thick clumps embedded in diffuse hot gas. Our results demonstrate that Balmer spectra encode the unresolved phase structure of gas in LRDs.

Abstract
The common envelope interaction happens between a growing or giant star of any mass and a more compact companion. The compact objects inspirals in the outer layers of the large star fully ejecting the envelope or merging with the core. The CE interaction is at the origin of a number of single and binary objects such as the progenitors of Type Ia supernovae, cataclysmic variables or the hypothesised Thorne-Zytkow objects. While 3D hydro simulations have come a long way in the last decade, observations of this fast phase have lagged, until recently. Time-resolved surveys have recently observed a number of objects that today are grouped into a single (albeit somewhat diverse) class called Luminous Red Novae. These have been identified as common envelope mergers and marrying these observations with simulations has become a priority to explain the observations and to (finally!) constrain the simulations. I will present the first simulations that target directly LRN light curves also including the effects of dust nucleation, essential to reproduce light and colours of these objects. To arrange a visit with the speaker during the visit, please contact their host: Fritz Röpke

Abstract
Active galactic nuclei (AGNs) are among the most powerful engines in the Universe. Understanding how accretion, obscuration, and feedback are connected is therefore essential for developing a complete picture of black-hole and galaxy evolution. In particular, observations of individual quasars have suggested a possible connection between powerful ionised outflows and unusually weak X-ray emission relative to established mid-infrared–X-ray relations. Testing this connection statistically requires large, homogeneous, multiwavelength samples. In this talk, I will introduce the development of the SDSS-V Black Hole Mapper quasar catalogue and discuss the analysis presented in Demke et al. (2026). The sample combines optical spectroscopy from SDSS-V with X-ray information from eROSITA and mid-infrared photometry from WISE. I will then present the extension of this analysis carried out as part of my master’s thesis. By increasing the sample size by about a factor of three and adding X-ray spectral information, this dataset enables a more detailed investigation of the relationships between X-ray emission, mid-infrared luminosity, black-hole mass, accretion rate, and the kinematics of ionised gas traced by the [O III] emission line.

Abstract
Strongly lensed supernovae (SNe) are emerging as a new probe of cosmology and SNe. When a SN is strongly lensed by a foreground galaxy, multiple images of the SN will appear around the foreground galaxy at different times. Current surveys including Euclid and the Rubin Observatory Legacy Survey of Space and Time will capture hundreds of strongly lensed SNe, expanding the existing sample by two orders of magnitude. These events offer excellent opportunities to: (1) measure the Hubble constant via lensing time delays, shedding light on the Hubble tension, and (2) obtain unprecedented constraints on SN progenitors through early-phase spectra. I will give an overview of the first discoveries of lensed SNe and their cosmological implications. To arrange a visit with the speaker during the visit, please contact their host: Philipp Girichidis

Abstract
ABSTRACT 1: AGN feedback plays an important role in the evolution of their host galaxy, yet the lifecycle of such an AGN event is still not fully understood. First observations and new simulations suggest the presence of AGN driven outflows even after the galaxy nucleus is no longer active. Using resolved spectroscopic data obtained with integral field units, these so-called fossil outflows can be studied on a larger scale. Predictions suggest that up to 60% of all observed outflows might be of fossil nature. In this talk, I will present a selection method for candidate fossil outflow regions in the SDSS-IV MaNGA galaxy sample comprising about 10,000 galaxies at low redshift. The method is based on kinematic elevation of the ionized gas, traced by broad [OIII]5007 emission line widths. By comparing the gas kinematics with the stellar velocity dispersion, gas motions dominated by the gravitational potential can be distinguished from outflow-driven motions. We identified a fossil outflow candidate sample of 269 non-AGN galaxies showing they are common in low redshift galaxies. I discuss some promising candidates in more detail as well as an analysis of the outflow properties of the outflow candidates compared to currently active galaxies. Simple outflow age calculations suggest outflow detections up to about 8 Myr after the AGN shutdown. ABSTRACT 2: One of the prime mechanisms suspected of driving the decline of star formation in massive galaxies is the so-called Active Galactic Nucleus (AGN) feedback, which still remains to this day as one of the least understood concepts in galaxy evolution. AGN feedback acts via many different, often combined, modes and when trying to decipher these processes it is crucial to tackle the different forms of feedback one by one to dissect their multi-faceted nature and understand which effects result from which causes. In this talk I will present a case-study conducted as part of my Bachelor Thesis by my group on one AGN host galaxy, dubbed the Cone Source, aimed at investigating how AGN feedback in the form of biconical outflows can locally affect the molecular gas reservoir as well as star formation. For this purpose, we combined and analyzed spatially resolved 12CO(1-0) observations taken by the Northern Extended Millimeter Array (NOEMA) with optical-IFU data from the MaNGA survey. In this data we find correlations of stellar properties with the multi-phase outflow strength as well as signs of a recent starburst phase within our galaxy, highlighting the complex evolution and interplay between AGN and their host galaxies.

Abstract
Galaxies do not grow in isolation, but within a cosmic web of intergalactic gas. The evolution of the galaxies and this intergalactic medium are closely connected. As the first galaxies formed, their light heated the gas surrounding them in a process known as reionization. Measuring the progress of reionization can give us an indirect view of how and when the first galaxies formed, complementary to direct observations of the galaxies themselves with telescopes like JWST. In this talk, I will discuss how we can constrain the timing of reionization with different observational probes, such as absorption in the spectra of distant luminous objects. I will present my work interpreting these observations with large cosmological simulations of the intergalactic medium. I will discuss what is known about the history of reionization, and what questions are still to be answered. To arrange a visit with the speaker during the visit, please contact their host: Sarah Bosman

Abstract
Quenching mechanisms have been extensively studied over the past decade, yet identifying which processes shut down star formation, their relative importance, and on what timescales, remains a central problem in galaxy evolution. In this talk, I will present the first systematic census of quenching mechanisms using kinematic asymmetries in a large sample (> 6700 galaxies) of spatially resolved galaxies, providing a unified view of what halts star formation in the local Universe (z < 0.2). Our analysis reveals that the most effective quenching pathways leave minimal kinematic imprints by the time galaxies are fully quenched, pointing to slow-acting processes (>3 Gyr) such as starvation and maintenance feedback. To compare these observations with theoretical predictions, we use mock galaxies from IllustrisTNG to investigate satellite quenching pathways, focusing on the role of the circumgalactic medium (CGM). Using a merger-tree-based algorithm, we reconstruct the baryonic, dark-matter, structural, and chemical histories of about 7300 spatially resolved galaxies, using time since infall as the physical axis along which quenching unfolds. For the first time, we present the coupled evolution of the hot and cool gas reservoirs after infall. Satellites lose about 90% of their hot CGM within about 4 Gyr, with the depletion increasing with residence time and remaining largely independent of stellar mass. The hot gas mass correlates strongly with SFR, establishing the CGM as a long-term fuel reservoir. Despite losing little stellar mass, satellites experience gas depletion and tidal stripping of their dark matter and metal-poor stellar outskirts, becoming more compact and metal-rich than centrals at fixed mass. Our results suggest that the gradual erosion of the hot CGM is a key link between infall and the slow shutdown of star formation.

Abstract
Theories of galaxy formation, structure, and evolution increasingly include the effects of cosmic rays on gas dynamics and energy balance. This is usually done by treating the cosmic rays as a fluid that is coupled to the thermal gas by magnetic fields on global scales and electromagnetic fluctuations on kinetic scales. I will discuss the foundations and assumptions of the fluid model, assess whether it is correct and complete, where it can be improved, and how it can be tested. To arrange a visit with the speaker during the visit, please contact their host: Max Gronke

Abstract
ESO is constructing the Extremely Large Telescope (ELT), which will become the world’s largest optical and near-infrared telescope and a flagship facility for ground-based astronomy in the coming decades. With its 39-metre primary mirror and unprecedented sensitivity and spatial resolution, the ELT will open new parameter space across a wide range of astrophysical topics, from exoplanets and star formation to galaxy evolution and cosmology. Realising this scientific potential depends critically on a new generation of highly sophisticated instruments developed by international consortia together with ESO. In this talk, I will provide an overview of the ELT and its instrumentation, with particular emphasis on the two instruments that are most advanced in development: MICADO, the high-resolution near-infrared imager, and METIS, the mid-infrared imager and spectrograph. I will discuss their scientific capabilities, technical challenges, and the transformational opportunities they will provide for future astronomy.

Abstract
To arrange a visit with the speaker during the visit, please contact their host: Max Gronke

Abstract
In this talk, I will present the SOFIA Legacy Project FEEDBACK, which has provided the first large-scale, velocity-resolved view of stellar feedback in galactic star-forming regions. This was achieved by utilising the [CII] 158 micron and [OI] 63 micron lines, which are key tracers of photodissociation regions (PDRs) at the interface between ionised and molecular gas. By probing feedback across a wide range of physical scales and evolutionary stages, FEEDBACK provides a comprehensive observational perspective on the formation, evolution and eventual dispersal of molecular clouds. One of the project’s major findings concerns cloud formation and dispersal. In Cygnus X, widespread high- velocity [CII] emission predominantly tracing neutral gas reveals the formation of molecular clouds via converging atomic flows (Schneider et al. 2023). In the same region, the Diamond Ring exemplifies a distinct feedback geometry: it arises as the terminal phase of a stellar wind bubble expanding into a flat molecular cloud, where bipolar breakout along the low-density directions above and below the cloud plane produces a true ring - not a sphere - as the observable signature. This ring morphology is detected in [CII] and confirmed by dedicated HD simulations following the evolution of a bubble in a flat molecular sheet (Dannhauer et al. 2025). Cloud dispersal is further evidenced across several FEEDBACK regions (RCW79; Bonne et al. 2023; Bonne et al. submitted), where [CII] emission reaches velocities of up to 25 km/s, organised partly in fragmented shells and partly in flows escaping through low-density channels. Derived mass-loss rates of order 0.01 Msun/yr imply short cloud erosion timescales of less than 5 Myr, providing direct observational evidence that stellar feedback can rapidly disperse molecular clouds and regulate star formation efficiency. Across the full FEEDBACK sample expanding bubbles are a common feature. While classical expanding shells are observed, the [CII] data often reveal fragmented, incomplete or asymmetrical bubbles, frequently accompanied by high-velocity wings and multiple kinematic components. Expansion velocities exceeding 10 km/s are detected in most regions, but their dynamics often require additional energy input from stellar winds beyond pure thermal pressure. Overall, FEEDBACK has provided a revised model of feedback-driven cloud evolution across scales: (i) molecular clouds form and disperse efficiently on short timescales; (ii) the geometry of feedback is fundamentally anisotropic and is governed by the turbulent, porous structure of the interstellar medium (ISM); (iii) [CII] can be used to trace both expanding shells and escaping gas flows; and (iv) individual embedded systems, such as S106 and the Diamond Ring, demonstrate that feedback can be detected from protostellar to cloud scales. Alongside other observations presented in this talk it highlights the power of FIR observations. Building on these results, the next step will be taken by the GEco survey on the CCAT/FYST telescope, with expected first light autumn this year. GEco will extend the FEEDBACK studies by mapping the [CII] and mid-J CO lines across entire molecular cloud complexes and the broader Galactic ecosystem on a large scale. Together with FEEDBACK, GEco will also probe the carbon cycle in the galactic interstellar medium.

Abstract
In recent years, cosmic rays (CRs) have emerged as a key factor shaping the interstellar medium (ISM) properties and influencing galaxy evolution. Yet, despite significant advances, modelling CR transport on galactic scales remains challenging, largely due to the complex microphysical interactions between CRs and thermal gas. As a result, the true extent of CR impact on ISM structure and evolution is still uncertain. In this talk, I will present our latest efforts to model CR transport within the TIGRESS MHD simulations of the multiphase ISM. I will discuss how CRs affect the dynamical and thermal state of the ISM in Milky Way-like galaxies, showing that while they have little effect on the star formation rate in galactic disks, they significantly accelerate warm outflowing gas with mass loading factors higher than in simulations without CRs. These results hold across different galactic environments. Beyond these theoretical insights, some simulations include a spectral treatment of CR protons and electrons, allowing us to construct synthetic observables - such as CR spectra and non-thermal emission - for direct comparison with observations. I will present our initial results comparing synthetic CR spectra with measurements in the solar neighbourhood, highlighting a very good agreement, and discuss the implications for CR transport. To arrange a visit with the speaker during the visit, please contact their host: Philipp Girichidis

Abstract
Gaia Data Release 4, to be published on 2 December 2026 at 12:00 CET, is the first release based on Gaia’s complete nominal mission, covering roughly 5.5 years of observations, almost doubling the 34-month observation baseline underlying Gaia DR3. Because proper-motion precision improves with the observation time span to the power of 3/2 - much faster than the square-root scaling of parallax precision - this extended baseline yields a disproportionate leap in astrometric accuracy for anything involving stellar motion: binary and multiple-star orbits, cluster dynamics, and subtle accelerations caused by unseen companions. The full release comprises astrometric, photometric, and spectroscopic catalogues for close to two billion sources, together with substantially expanded non-single-star solutions, variability, and astrophysical parameters. This talk previews the scientific scope of the release through the topics addressed by the DR4 Performance Verification Papers, spanning the full breadth of Gaia science: dormant black holes in binary systems; exoplanets revealed through subtle astrometric and spectroscopic signatures; the structure and dynamics of open and globular star clusters; asteroids, asteroid families, and comets in the Solar System; the orbits and dynamics of the nearest stars to the Sun; the coolest and least massive stars at the bottom of the main sequence; classical pulsating stars tracing the spiral structure of the Milky Way; time delays in gravitationally lensed quasars; and the first three-dimensional cartography of a galaxy beyond the Milky Way, the Large Magellanic Cloud. Together, these studies illustrate how Gaia DR4 will transform our view of the Milky Way and its neighbourhood, from the Solar System to the structure of the Local Group and beyond. As the detailed scientific results remain under embargo until the release date, this talk will present the concrete findings in full after 2 December 2026, 12:00 CET.

Abstract
To arrange a visit with the speaker during the visit, please contact their host: Michela Mapelli

Abstract
Very massive stars (VMS) are born with masses above 100 Msun and represent the most extreme, albeit rare, class of stellar objects. They shine with luminosities exceeding a million times that of the Sun and drive powerful line-driven winds that inject momentum and chemically processed material into the interstellar medium. VMS are invoked to explain distinctive spectral features in unresolved stellar populations, yet the stars themselves remain poorly understood. At the highest masses, the scarcity of observational constraints limits our ability to test models of VMS evolution, while at high metallicity, uncertainties in mass loss remain a major challenge. In this talk, I will motivate the detailed investigation of O/WNh stars as key laboratories for probing VMS evolution. In the Milky Way, only nine such stars are currently known, and none has previously been quantitatively characterized using detailed stellar atmosphere modeling. For the first time, we have computed models reproducing the observed optical spectrum of WR 31-1, the most luminous Galactic O/WNh star. I will present the results of this pilot study and discuss how they motivate future targeted observational campaigns.

Abstract
Shock waves are the universe's most prolific particle accelerators. The same mechanism, diffusive shock acceleration, is invoked to explain energetic particles in the solar wind, cosmic rays from supernova remnants, and relativistic electrons at merger shocks in galaxy clusters. In this picture, particles gain energy by repeatedly scattering back and forth across the shock front, aided by turbulent magnetic fields. Turbulence, however, does not have to produce ordinary diffusion. If particles occasionally make very long, uninterrupted excursions, they spread much faster than a random walk allows. This behaviour, known as a Lévy walk or superdiffusion, appears in systems ranging from wild fires to particles transport in the heliosphere. Radio relics are arcs of synchrotron emission that can extend over megaparsecs, tracing shocks driven by colliding galaxy clusters. They provide a natural laboratory for testing how particles are transported on large scales. We discovered that several relics show faint radio emission reaching up to 100 kpc ahead of the shock, where standard theory predicts almost nothing. I will show that superdiffusive electron transport explains this precursor naturally. I will then discuss how the spectral index profile ahead of the shock, measurable with LOFAR, MeerKAT and soon the SKA, can distinguish true superdiffusion from simply faster normal diffusion. In the final part, I will show what other curiosities recent radio observations have revealed about cosmic-ray transport in the intra-cluster and intra-group medium. To arrange a visit with the speaker during the visit, please contact their host: Max Gronke

Abstract
The evolution of galaxy clusters is strongly shaped by the physics of the intracluster medium (ICM), where processes such as turbulence, mixing, and galaxy-ICM interactions drive much of the cluster’s dynamical state. One key ingredient is viscosity, which can suppress small-scale gas motions and modify density fluctuations and magnetic fields. Yet, both the strength and nature of viscosity in the ICM remain uncertain. In this talk, I will present results from high-resolution cosmological simulations that explicitly include viscosity, with a particular focus on the Coma cluster. Recent XRISM observations of Coma have revealed unexpectedly low small-scale gas motions, challenging our current understanding of ICM dynamics. I will explore whether viscosity can help explain these observations, comparing inviscid simulations with models including both isotropic and anisotropic viscosity. I will show how these different treatments affect cluster morphology, turbulence, and ICM fluctuations, and discuss how simulations and observations together can constrain the microphysics of the ICM.

Abstract
TBD

Abstract
The Dark Energy Spectroscopic Instrument (DESI) Data Release 1, published in March 2025, provides a lot of spectra of the ionized interstellar medium in HII-regions within supposed dwarf galaxies. Using the strong line calibration method, we derive their gas-phase oxygen abundance (12+log(O/H)), as a metallicity indicator. Comparing these metallicities with stellar-mass estimates (CIGALE, FASTSPECFIT), provided by the DESI VACs, we find that a "significant percentage" of the dwarf galaxies deviate significantly from the mass-metallicity relation. We find that this is due to deblending issues of extended galaxies in Legacy Survey (LS) data releases - including DR9, on which DESI DR1 photometry is based - leading to faulty luminosities across all color bands in DESI DR1. This subsequently disturbs mass estimates. Using segmentation maps applied to model images, which we reconstruct from LS DR9 Tractor catalog data, we estimate the luminosities in different bands for over 3 million galaxies at z < 0.2. We find that for a "significant percentage" of the galaxies our derived magnitudes are more than 1 mag brighter than DESI DR1 results. With this corrected photometry, genuine outliers from the mass-metallicity relation can now be observed.

Abstract
TBD

Abstract
Studying the impact of Active Galactic Nuclei (AGN) on the interstellar medium at high redshift is crucial for understanding how galaxies evolve. In this talk, I will first present results from 4C +03.24, a powerful radio galaxy at z ~ 3.5, using integral-field spectroscopy obtained with JWST/NIRSpec. The joint analysis of our observations with archival VLT/MUSE and HST data reveals extended continuum emission (~14 kpc), possibly arising from recent star formation and associated with nearby companions. The presence of a gap in the [CII] emission within the bipolar region dominated by AGN photo-ionization may provide direct evidence of negative AGN feedback. We further find that the outflow power associated with the kinematically disturbed ionized gas peaks ~ 2 kpc from the nucleus, with a low kinetic coupling efficiency of ~0.0063%. In the second part, I will discuss new results on AGN across different epochs (1 < z < 6), obtained from archival JWST data.