arXiv subjects

Michele Ginolfi

Publications and source records attributed to Michele Ginolfi.

At least 19 recordsLinked to original sources

arXiv2026

Tracing Lyman alpha escape in the CRISTAL-02 galaxy at z~5.3

We investigate the mechanisms regulating Lyman-alpha (Ly$α$) escape in the star-forming galaxy CRISTAL-02 at z~5.3. The galaxy has clumpy morphology, suggestive that it may be interacting with other system(s). Two clumps (A and B, hereafter) are suggested as a site for intense star-formation or potential AGN candidates. We investigate how the local gas, dust, and feedback shape the escape of Ly$α$ photons around these clumps. Using VLT/MUSE and JWST/NIRSpec IFU observations, complemented by NIRCam UV imaging, we constructed spatially matched emission-line maps. We derived flux, line-ratio, and extinction maps, together with spatially resolved Ly$α$ escape fractions and ionizing photon production efficiencies. We find that Ly$α$ is significantly more extended than H$α$ and UV, reaching ~33 kpc and preferentially extending along the cold molecular gas outflow traced by [C II] emission. Clumps A and B show contrasting Ly$α$ properties: Clump A has lower dust attenuation and enhanced Ly$α$/H$α$ ratios and escape fraction, whereas Clump B is brighter in H$α$ and UV but has suppressed Ly-$α$ despite a higher ionizing photon production efficiency. These results indicate that Ly$α$ escape is strongly influenced by the local H I geometry, dust, and outflows and cannot be explained by ionizing photon production alone. The observed Ly$α$ morphology and zELDA radiative transfer modeling favor an outflow-driven escape scenario, while the available data cannot uniquely distinguish between AGN and star formation-driven feedback.

astro-ph.GA↗

arXiv2026

Unsupervised selection and characterisation of Little Red Dots in JWST surveys with manifold learning

Little Red Dots (LRDs) are compact, red sources discovered at high redshift by JWST whose physical nature and selection function remain debated. We investigate whether an unsupervised machine-learning approach applied to multi-band photometry can identify LRD-like objects, and other populations, without relying on predefined colour cuts. Using UMAP, a manifold-learning (dimensionality-reduction) method, we place ~242,000 isolated, well-measured sources from the ASTRODEEP-JWST catalogue on a two-dimensional map, where objects with similar broadband colours, morphology, and photometric redshift lie close together. We then use spectroscopically confirmed LRDs to identify where LRD-like objects lie within this map, compare the resulting areas with published colour cuts, and validate our data-driven selection against archival NIRSpec spectra from the DJA. We find that the spectroscopically selected LRDs concentrate in two well-defined regions with no colour cut imposed, tracing populations that differ mainly in redshift, a difference imprinted in their broadband colours. The main region reaches a purity of ~0.78 at ~0.82 completeness on the spectroscopically classified subset, competitive with, or cleaner than, literature colour cuts, and yields ~100 additional candidates. We also test the method as a general tool for population discovery: the manifold recovers the locations of brown dwarfs and broad-line AGN with no explicit criterion, and isolates rare pathological outliers. Overall, unsupervised manifolds, anchored by sparse high-confidence spectroscopic labels, provide an efficient, assumption-light framework for characterising populations, comparing selection methods on a common basis, and discovering rare objects in large photometric datasets.

astro-ph.GA↗

arXiv2026

Beyond the Largest Lyapunov Exponent: Entropy-Based Diagnostics of Chaos in Henon-Heiles and N-Body Dynamics

The largest Lyapunov exponent is widely used to diagnose chaos in gravitational dynamics, but in mixed phase spaces and finite-N systems it does not always provide a complete description of orbital complexity and phase-space transport. Entropy-based diagnostics may offer a complementary perspective. We investigate whether trajectory-based information entropy can provide a useful diagnostic of chaos in gravitational systems and how it relates to the largest Lyapunov exponent as a function of orbital energy and of the number of degrees of freedom. We computed the largest Lyapunov exponent and a coarse-grained Shannon entropy for ensembles of trajectories in the Henon-Heiles potential and for test-particle orbits in live N-body realizations of a Plummer model. We then compared the dependence of both quantities on orbital energy and, for the N-body case, on particle number. In the Henon-Heiles system, the Shannon entropy follows the transition from weak to widespread chaos and exhibits an energy dependence that closely mirrors that of the largest Lyapunov exponent. For test-particle orbits in live N-body potentials, both diagnostics indicate stronger chaos for more tightly bound trajectories. However, their dependence on N differs: the largest Lyapunov exponent remains nearly constant over the explored range of particle numbers, whereas the Shannon entropy decreases monotonically as N increases. These results show that the information entropy can complement the largest Lyapunov exponent and may better capture changes in global phase-space mixing, especially in systems where the leading Lyapunov exponent alone is not sufficiently informative. It therefore provides a promising alternative for diagnosing chaos when tangent-space dynamics is unavailable or computationally expensive, and it is naturally suited to systems with densely sampled trajectories, such as minor bodies in the Solar System.

astro-ph.EP↗

arXiv2026

Adversarial robustness of a U-Net-based model observer for CT protocol optimization

Artificial intelligence is increasingly used in medical imaging, yet its robustness to input perturbations remains a critical concern for a wide clinical adoption. To this end, we used adversarial examples to systematically probe vulnerabilities of a U-Net-based model observer for computed tomography protocol optimization, performing detection and localization of low-contrast objects in a phantom dataset. Adversarial attacks were generated using both gradient-based and optimization-based white-box methods. Fast gradient perturbations produced high misclassification rates, reaching up to 75% at intermediate perturbation levels while remaining visually imperceptible. Localization was more robust, with success rates of about 25% for small perturbations and 42% at moderate levels. In contrast, optimization-based attack achieved success rates close to 50% for both tasks. To mitigate these vulnerabilities, dynamic adversarial training was implemented. This reduced the success rate of optimization-based attacks to 7% for classification and 13% when including localization-specific training, demonstrating a substantial robustness improvement without compromising task performances, confirmed by localization receiver operating characteristic analysis. To further interpret model behavior, radiomic texture analysis was performed on original and adversarial images. While most global image statistics remain stable, specific texture-related features exhibit consistent changes in successful attacks, highlighting the model's sensitivity to subtle local intensity patterns. Overall, adversarial training improves robustness without degrading performance, while radiomic analysis reveals interpretable links between texture alterations and prediction failures, supporting more reliable and explainable AI systems for medical imaging.

physics.med-ph↗

arXiv2026

The ALPINE-CRISTAL-JWST Survey: The Fast Metal Enrichment of Massive Galaxies at z~5

We present the stellar mass-metallicity relation (MZR) and mass-metallicity-star formation relation ("fundamental metallicity relation"; FMR) of 18 massive (log(M/Msun) = 9.5 - 11) main-sequence galaxies at z~5 from the ALPINE-CRISTAL-JWST sample. This sample complements recent studies by JWST at up to two orders of magnitude lower stellar masses. The metallicities are derived using strong optical lines, and verified by temperature-based oxygen abundance measurements for five galaxies for which faint auroral lines are detected. We find that the metal abundance evolves, on average, from 40% to 60% solar between z~5 and cosmic noon (z~2) at the massive end of the MZR, suggesting already significant metal enrichment at early times. The FMR at z=5 exhibits a 5x larger scatter (preferentially to lower metallicities) compared the local FMR relation. This scatter can be explained by a bursty star formation and the direct build-up of metals in early galaxies as well as differences in age and outflow efficiencies. Capitalizing on all available samples, we find that the observed MZR and FMR over three orders of stellar mass is generally in good agreement with results from cosmological simulation, although some underestimate the metal enrichment at low stellar masses. This may be due to too efficient metal-rich outflows. We show that the ALPINE-CRISTAL-JWST galaxies likely joined the current FMR at z~10 and will evolve into massive (log(M/Msun)~11.4) galaxies with super-solar metallicities by z=0.

astro-ph.GA↗

arXiv2026

The ALPINE-CRISTAL-JWST Survey: Gas-phase abundance gradients of main sequence star-forming galaxies and their kinematics at $4 < z < 6$

We present gas-phase radial metallicity profiles for 20 main-sequence galaxies at $4 0.05{\rm dexkpc^{-1}}$ at $1σ$, and none have a significant negative gradient. We investigated the correlation between the metallicity gradients and the intrinsic gas velocity dispersion $σ_0$ as well as the $V_{\rm rot}/σ_0$ ratio of the disks. Combining our sample with mass-matched literature samples at $3<z<7$, we found a negative shallow correlation between $V_{\rm rot}/σ_0$ and the metallicity gradients, but no strong relationships with $σ_0$. As $V_{\rm rot}/σ_0$ increases towards later cosmic times, the observed negative trend with $V_{\rm rot}/σ_0$ is consistent with the overall cosmic evolution of metallicity gradients from high to low redshifts. This suggests that disk maturity plays a crucial role in shaping the radial metallicity gradients. [Abridged abstract]

astro-ph.GA↗

arXiv2026

Milky-Way-like stars in a galaxy core 8 billion years ago revealed by gravitational lensing

The assembly of stellar-dominated cores in elliptical galaxies is key to understanding how cosmic structures evolved. Gravitational lensing offers unique insights into the nature of their stars. We report the discovery of the smallest known quadruply lensed quasar (radius ~0.2"), whose lensing galaxy at redshift 1.055 (5.5 billion years after the Big Bang) features a lensing mass of only ~2x10^10 M_sun. A Bayesian analysis, based on the system's exceptional properties and standard scaling relations, allowed us to sample the central galactic initial mass function with unmatched accuracy and in a previously uncharted regime in terms of mass and redshift. We found it consistent with the Milky Way one, while excluding bottom-heavy functions. This suggests that the core either grew slowly or underwent early disruptive events altering its stellar build-up, in contrast with the classical view that bulges form rapidly and remain unchanged by later interactions.

astro-ph.GA↗

arXiv2026

The ALPINE-CRISTAL-JWST Survey: Chemical Abundance Comparison Between the ISM and CGM of Main-Sequence Galaxies at z=4-6

Gaseous halos around galaxies play an important role in galaxy evolution. The exchange of metals from the interstellar medium (ISM) to the circumgalactic medium (CGM) are caused by the formation, feedback, and/or merging history of galaxies. We study the variation in chemical composition between the ISM ($\lesssim3\,$kpc) and CGM ($\sim5-10\,$kpc) for a sample of $M_{\star}>10^{9.5}\,M_{\odot}$ main-sequence galaxies at $4<z<6$ with both JWST and ALMA observations. Using JWST/NIRSpec integral field spectroscopy, we derive the optical line ratios from the ISM and the CGM for our sample focusing on the typical optical lines used for metallicity studies. Our comparison shows that the ISM and the CGM have similar chemical abundances. This indicates that the CGM of these typical $4<z<6$ galaxies is enriched to the level of their ISM in the early universe. Using statistical tests, we find that some of the line ratios show marginal differences between the ISM and CGM. Combined with \texttt{Cloudy} modeling, our results suggest that a difference in ionization level (higher for the ISM) is the dominant reason for the observed ratio difference of oxygen line ratios. There is also indication of a deficit in the nitrogen abundance with respect to oxygen in the CGM, which suggests a delay in redistribution of secondary nitrogen. Finally, an enhanced $F_{[\rm CII]\rm 158μm}/F_{\rm Hα}$ ratio is observed in the CGM, suggesting that feedback and/or mergers play a key role in metal mixing.

astro-ph.GA↗

arXiv2026

Direct pathway to the Early Supermassive Black Holes: A Red Super-Eddington Quasar in a Massive Starburst Host at $z=7.2$

We present a panchromatic optical-mm characterization of GNz7q, a recently identified X-ray weak, rapidly growing red quasar embedded within a dusty starburst galaxy at $z=7.1899$, using the full suite of JWST/NIRCam, NIRSpec, MIRI, and archival NOEMA observations. Our deep NIRSpec/G395M spectroscopy reveals unambiguous broad Balmer emission (FWHM $=2221\pm20$kms$^{-1}$), confirming a super-Eddington accreting black hole ($λ_{\rm Edd}=2.7\pm0.4$) with a mass of $\log(M_{\rm BH}/M_{\odot})=7.55\pm0.34$, using accretion-rate corrected BH mass estimators. After subtracting the point source, we robustly detect stellar emission from the host galaxy across multiple NIRCam and MIRI filters. Out joint morphological-spectral analysis yields a stellar mass of $\log (M_*/M_\odot)=10.5\pm0.4$ and an intense star formation rate of ${\rm SFR}=330\pm97\,M_\odot\,\rm yr^{-1}$, confirming the host as a massive, dusty starburst galaxy. We find that GNz7q lies on the local $M_{\rm BH}$-$M_*$ relation ($M_{\rm BH}/M_*\simeq 0.001$) and is well positioned to evolve into the locus of massive SDSS quasars with $\log (M_{\rm BH}/M_\odot)\approx 9$ and $M_*\approx 10^{11}\,M_\odot$ at $z\sim 6$, owing to its remarkably rapid growth in both the black hole and its host galaxy. This stands in stark contrast to many recently reported JWST AGN populations at similar redshifts, including the little red dots (LRDs), whose weak or undetected star formation makes it difficult for them to grow into the massive galaxies hosting SDSS-like quasars. These results suggest that GNz7q marks as a rare, pivotal phase of early BH-galaxy co-eolution, plausibly providing a crucial direct pathway to the supermassive black hole systems within the first billion years of the Universe.

astro-ph.GA↗

arXiv2026

One cloud is not enough: extreme conditions bias chemical abundances in high-redshift galaxies

Since its launch, JWST has opened an unprecedented opportunity to characterise the ionised ISM of high-redshift galaxies using well-established rest-frame UV/optical diagnostics from the local Universe. At the same time, these observations challenge the validity of such classical methods when applied to the extreme environments typical at high redshift. We present an in-depth analysis of the ISM in three representative case studies at $z=2 - 6$ (MARTA 4327, the Sunburst Arc and RXCJ2248-ID) conducted within a multi-cloud photoionisation modelling framework (HOMERUN). We show that even a small fraction of unresolved high-density clumps can contribute more than half of the observed flux of auroral lines, while only negligibly to standard optical density tracers. As a result, $T_{\mathrm{e}}$-method metallicities can be underestimated by $\sim 0.15 - 0.3$ dex, as for MARTA 4327. By modelling rest-frame UV and optical data, we demonstrate that discrepancies between abundances obtained from diagnostics tracing different zones do not necessarily imply chemical inhomogeneities. In RXCJ2248-ID, the disagreement between UV and optical N/O may naturally arise from ionisation and density structure alone. In contrast, we find evidence for genuine chemical stratification in the Sunburst Arc, where a component enriched in nitrogen coexists with a chemically normal one. Finally, we argue that very-high-ionisation lines may be explained within a pure star-formation scenario invoking matter-bounded regions. However, in the case of RXCJ2248-ID, we cannot rule out a minor contribution from an AGN based solely on the observed fluxes. These results indicate that classical diagnostics can be significantly biased in high-redshift galaxies and that self-consistent, physically motivated tools are therefore essential to properly interpret the complex ISM conditions and chemical enrichment in the early Universe.

astro-ph.GA↗

arXiv2026

GA-NIFS: AGN activity in a Lyα emitter within a triple-AGN system anchored by a passive galaxy at z=3

Massive quenched galaxies at z>3 challenge models of early galaxy evolution, as their rapid formation and abrupt quenching require efficient feedback, often linked to active galactic nuclei (AGN). The quiescent galaxy GS10578 at z=3.1 is a key example of this population. Previous JWST/NIRSpec IFU data revealed an AGN outflow and uncovered a compact pair of AGN separated by ~5 kpc. In addition, VLT/MUSE spectroscopy has identified a third AGN candidate at a projected distance of ~30 kpc, associated with a luminous Lya emitter (LAE2) characterised by high-ionisation UV lines, although rest-frame optical diagnostics were not previously available. We aim to confirm the nature of LAE2 using rest-frame optical diagnostics enabled by new JWST data, and to characterise the physical and ionisation properties of both LAE2 and a distinct nearby Lya emitter (LAE1) that lacks any detectable continuum counterpart. We analyse new NIRSpec IFU observations targeting the optical nebular lines of LAE1 and LAE2, complemented with MUSE data, as part of the GA-NIFS project. We extract integrated and spatially resolved spectra, construct emission-line maps, and use standard diagnostic diagrams to determine ionisation sources and kinematics. LAE2 exhibits line ratios fully consistent with an embedded AGN. Its optical lines display a clumpy morphology and irregular kinematics on sub-kpc scales. Except for Lya, LAE1 remains undetected in all nebular lines and in JWST imaging. The similarity of the LAE1 and LAE2 Lya profiles in both velocity and flux suggests that LAE1 traces resonantly scattered emission rather than in-situ star formation. Our analysis reveals that the environment of GS10578 contains both multi-black-hole activity and gas structures on tens-of-kpc scales, offering new insights into how feedback and satellite interactions influence the late evolutionary stages of quenched massive galaxies.

astro-ph.GA↗

arXiv2025

A search for cold molecular outflows in cosmic noon galaxies

The flow of baryons in and out of galaxies is the primary driver for galaxy evolution. In addition to depleting the gas reservoir of galaxies, outflows also enrich their circumgalactic medium (CGM) with processed gas -- which can further impact the next stages of gas accretion, resulting in the presence of molecular gas beyond the stellar component of galaxies, out to CGM scales. Here, we aim to search for cold molecular gas (MH2) in the CGM of typical main-sequence (MS) star-forming galaxies (SFGs) at cosmic noon (zmed=1.3), where we expect outflows to be particularly prominent. Using Band 3 CO(2-1) data from the Atacama Large Millimeter and submillimeter Array (ALMA), we study the spatial extent of the MH2 of a sample of 26 SFGs, via stacking techniques. We compare this extent to that of the stacked stellar emission of our sample traced by UltraVISTA Ks band data. We also search for broad wings in the stacked spectrum which can be indicative of ongoing outflows. Within the noise level of the observations, we find that the total intrinsic MH2 of our sample spatially extends to scales of ~12 kpc, similarly to the stellar emission (~13 kpc). We do not find broad wings in the stacked spectrum that could hint at ongoing molecular outflows, but we find a tentative minor excess of CO(2-1) emission at negative velocities that might be indicative of outflows, where the redshifted gas is optically thick. The absence of high-velocity molecular gas suggests that molecular outflows traced by CO(2-1) emission are weak in MS SFGs at cosmic noon. These weak outflows thus fail to expel a significant amount of molecular gas to CGM scales, as indicated by the absence of molecular emission extending beyond the stellar emission region. This lack of CO emission at large radii could also imply that the molecular gas does not survive at such distances.

astro-ph.GA↗

arXiv2025

The ALPINE-CRISTAL-JWST Survey: Stellar and nebular dust attenuation of main-sequence galaxies at z~4-6

Characterizing dust attenuation is crucial for revealing the intrinsic physical properties of galaxies. We present an analysis of dust attenuation in 18 spectroscopically confirmed star-forming main-sequence galaxies at $z = 4.4-5.7$ observed with JWST/NIRSpec IFU and NIRCam, selected from the ALPINE and CRISTAL ALMA large programs. We fit the emission line fluxes from NIRSpec and the broad-band photometry from NIRCam with Prospector, using both spatially integrated emission and $\sim0.6$ kpc pixel-by-pixel measurements. We derive the stellar-to-nebular dust attenuation ratio ($f=E(B-V)_{\mathrm{star}}/E(B-V)_{\mathrm{neb}}$) from the SED fits and the Balmer decrement with H$α$ and H$β$. Although individual galaxies show large scatter, the best-fit value is $f = 0.51^{+0.04}_{-0.03}$, slightly higher than that measured for local starburst galaxies. We find weak correlations of $f$ with galaxy properties, increasing with higher specific star-formation rates, younger stellar ages, and more recent star-formation. For the range of $E(B-V)_{\mathrm{star}} = 0.009-0.15$ mag for in our sample, assuming $f = 1$ (often adopted in high-redshift studies) instead of $f = 0.51$ underestimate line luminosities and ionizing photon production efficiency $ξ_\text{ion}$ by $\sim3-36\%$ and $\sim4-46\%$, respectively. We also find that the total stellar masses estimated from spatially-integrated SED fits with a delayed-$τ$ star-formation histories are systematically smaller than the sum of pixel-by-pixel SED fits, with a median offset of $\sim 0.26$ dex, likely because the integrated fits are biased toward luminous young stellar populations.

astro-ph.GA↗

arXiv2025

The ALPINE-CRISTAL-JWST Survey: NIRSpec IFU Data Processing and Spatially-resolved Views of Chemical Enrichment in Normal Galaxies at z=4-6

We present a statistical study of spatially resolved chemical enrichment in 18 main-sequence galaxies at $z=4$--6, observed with \jwst/NIRSpec IFU as part of the ALPINE-CRISTAL-\jwst\ survey. Performing an optimized reduction and calibration procedure, including local background subtraction, light-leakage masking, stripe removal, and astrometry refinement, we achieve robust emission-line mapping on kiloparsec scales. Although line-ratio distributions vary across galaxies in our sample, we generally find mild central enhancements in [O\,\textsc{iii}]/H$β$, [O\,\textsc{ii}]/[O\,\textsc{iii}], [S\,\textsc{ii}]$_{6732}$/[S\,\textsc{ii}]$_{6718}$, H$α$/H$β$, and $L_{\rm Hα}/L_{\rm UV}$, consistent with elevated electron density, dust obscuration, and bursty star formation accompanied by reduced metallicity and ionization parameter. These features point to inside-out growth fueled by recent inflows of pristine gas. Nevertheless, the median metallicity gradient is nearly flat over a few kpc scale, $Δ\log({\rm O/H}) = 0.02 \pm 0.01$ dex kpc$^{-1}$, implying efficient chemical mixing through inflows, outflows, and mergers. From pixel-by-pixel stellar and emission-line characterizations, we further investigate the resolved Fundamental Metallicity Relation (rFMR). Metallicity is described by a fundamental plane with stellar mass and SFR surface densities, but with a stronger dependence on $Σ_{\rm SFR}$ than seen in local galaxies. Our results indicate that the regulatory processes linking star formation, gas flows, and metal enrichment were already vigorous $\sim$1 Gyr after the Big Bang, producing the nearly flat metallicity gradient and a stronger coupling between star formation and metallicity than observed in evolved systems in the local universe.

astro-ph.GA↗

arXiv2025

No [CII] or dust detection in two Little Red Dots at z$_{\rm spec}$ > 7

Little Red Dots (LRDs) are compact, point-like sources characterized by their red color and broad Balmer lines, which have been debated to be either dominated by active galactic nuclei (AGN) or dusty star-forming galaxies (DSFGs). Here we report two LRDs (ID9094 and ID2756) at z$_{\rm spec}$>7, recently discovered in the JWST FRESCO GOODS-North field. Both satisfy the "v-shape" colors and compactness criteria for LRDs and are identified as Type-I AGN candidates based on their broad H$β$ emission lines (full width at half maximum: 2280$\pm$490 km/s for ID9094 and 1070$\pm$240 km/s for ID2756) and narrow [OI] lines ($\sim$ 300-400 km/s). To investigate their nature, we conduct deep NOEMA follow-up observations targeting the [CII] 158${\rm μm}$ emission line and the 1.3 mm dust continuum. We do not detect [CII] or 1.3 mm continuum emission for either source. Notably, in the scenario that the two LRDs were DSFGs, we would expect significant detections: $>16σ$ for [CII] and $>3σ$ for the 1.3 mm continuum of ID9094, and $>5σ$ for [CII] of ID2756. Using the 3$σ$ upper limits of [CII] and 1.3 mm, we perform two analyses: (1) UV-to-FIR spectral energy distribution (SED) fitting with and without AGN components, and (2) comparison of their properties with the L$_{[CII]}$-SFR$_{tot}$ empirical relation. Both analyses are consistent with a scenario where AGN activity may contribute to the observed properties, though a dusty star-forming origin cannot be fully ruled out. Our results highlight the importance of far-infrared observations for studying LRDs, a regime that remains largely unexplored.

astro-ph.GA↗

arXiv2025

AstroLLaVA: towards the unification of astronomical data and natural language

We present AstroLLaVA, a vision language model for astronomy that enables interaction with astronomical imagery through natural dialogue. By fine-tuning the LLaVA model on a diverse dataset of $\sim$30k images with captions and question-answer pairs sourced from NASA's `Astronomy Picture of the Day', the European Southern Observatory, and the NASA/ESA Hubble Space Telescope, we create a model capable of answering open-ended questions about astronomical concepts depicted visually. Our two-stage fine-tuning process adapts the model to both image captioning and visual question answering in the astronomy domain. We demonstrate AstroLLaVA's performance on an astronomical visual question answering benchmark and release the model weights, code, and training set to encourage further open source work in this space. Finally, we suggest a roadmap towards general astronomical data alignment with pre-trained language models, and provide an open space for collaboration towards this end for interested researchers.

astro-ph.IM↗

arXiv2025

Classifying spectra of emission-line regions with neural networks -- An application to integral field spectroscopic data of M33

Emission-line regions are key to understanding the properties of galaxies, as they trace the exchange of matter and energy between stars and the interstellar medium (ISM). In nearby galaxies, individual nebulae can be identified as HII regions, planetary nebulae (PNe), supernova remnants (SNR), and diffuse ionised gas (DIG) with criteria on single or multiple emission-line ratios. However, these methods are limited by rigid classification boundaries, the narrow scope of information they are based upon, and the inability to account for line-of-sight nebular superpositions. In this work, we use artificial neural networks to classify these regions using their optical spectra. Our training set consists of simulated spectra, obtained from photoionisation and shock models, and processed to match observations obtained with MUSE. We evaluate the performance of the network on simulated spectra for a range of signal-to-noise (S/N) levels and dust extinction, and the superposition of different nebulae along the line of sight. At infinite S/N the network achieves perfect predictive performance, while as the S/N decreases, the classification accuracy declines, reaching an average of ~80% at S/N(H$α$)=20. We apply our model to real spectra from MUSE observations of the galaxy M33, where it provides a robust classification of individual spaxels, even at low S/N, identifying HII regions and PNe and distinguishing them from SNRs and diffuse ionized gas, while identifying overlapping nebulae. We then compare the network's classification with traditional diagnostics and find satisfactory agreement. Using activation maximisation maps, we find that at high S/N the model mainly relies on weak lines (e.g. auroral lines of metal ions and He recombination lines), while at the S/N level typical of our dataset the model effectively emulates traditional diagnostic methods by leveraging strong nebular lines.

astro-ph.GA↗

arXiv2025

Unveiling the Fast Acceleration of AGN-Driven Winds at Kiloparsec Scales

Supermassive black holes at the centre of galaxies gain mass through accretion disks. Models predict that quasi-spherical winds, expelled by the black hole during active accretion phases, have a key role in shaping galaxy evolution by regulating star formation, the distribution of metals over kiloparsec scales, and by sweeping ambient gas to the outskirts and beyond of galaxies. Nonetheless, the mechanism driving these outflows and the amount of energy exchanged between the wind and the galaxy's interstellar medium remain unclear. Here, we present a detailed analysis of the kinematical properties of winds in a sample of nearby active galaxies using the novel kinematic tool MOKA3D, which takes into account the clumpy nature of the ISM. We find remarkable similarities among the properties of the outflows in all the galaxies examined. In particular, we provide the first evidence that outflows exhibit a regular trend in radial velocity, initially constant or slightly decreasing, followed by rapid acceleration starting at approximately 1 kpc from the nucleus, despite the seemingly complex kinematics observed. The observed behavior aligns with our current theoretical understanding of Active Galactic Nuclei outflows, where a momentum-driven phase transitions to an energy-conserving phase just beyond approximately 1 kpc. The constant velocity of the momentum-driven wind is then rapidly accelerated following the inefficient Compton cooling of post-shock material and the transition to energy conservation. The measured radial terminal velocities of the outflows are always larger than the escape velocities from the host galaxies, confirming the key role of outflows in shaping the galaxy properties and evolution, as a manifestation of AGN feedback. Our results, only made possible by our novel kinematic analysis tool, are crucial to understand the origin and the powering mechanism of these winds.

astro-ph.GA↗