Eight new feedback tests, from Mg II quenching to AGN bubbles and the Lyα forest

Quick scan

Observed CGM and outflows

Cool Mg II absorption declines through the green valley

• Authors/date: Simon Xinlin Wu et al.; submitted August 24, 2026; v1; submitted to ApJ.
• Study design: The authors combine Mg II absorption measurements for 716 galaxies at $0.07<z<2.7$, including 169 new and 547 archival galaxies. The sample is compared with a virial-radius-normalized radial profile. A star-formation offset metric, $\sigma_{\mathrm{SFO}}$, places galaxies on a common scale from the star-forming main sequence across roughly 10 billion years of cosmic time. 1
• Result: Mg II absorption declines with projected distance from the host. The residual around the mean profile is highest for star-forming galaxies, intermediate for green-valley galaxies, and lowest for quiescent galaxies. The same ordering appears in the inner-CGM covering fraction for . Star-forming galaxies also show enhanced absorption along both polar and disk directions, a pattern consistent with bipolar outflows and co-planar accretion. The absorption kinematics are consistent with a predominantly bound cool CGM. 1
• Why read: The paper supplies a large, redshift-spanning observational test of how the cool CGM changes as star formation fades. The green-valley population makes the transition gradual in the data rather than a binary star-forming-versus-quenched split.
• Caveat: Mg II is a cool-gas tracer, and the polar excess is consistent with outflows rather than a unique measurement of their mass or energy. The sample establishes a relation between star-formation state and cool CGM structure; it does not by itself isolate AGN feedback from accretion history or other causes of quenching.

NGC 253 traces a wind through cosmic-ray transport

• Authors/date: Shengtao Wang et al.; submitted August 24, 2026; v1; accepted for publication in A&A.
• Study design: ASKAP 943 MHz and MWA 216 MHz total-intensity images provide resolutions of 13 and 45 arcsec, with rms noise levels of 16 $\mu$Jy beam and 1 mJy beam. After thermal-emission subtraction, the authors fit vertical synchrotron-intensity and spectral-index profiles with one-dimensional advection and diffusion models. 2
• Result: The northwest radio spur reaches about 9 kpc above the disk and the southeast spur about 8 kpc. The central region is better described by advection, while diffusion becomes more important farther out. The fitted advection speed reaches the estimated escape speed at about 5.5 kpc. Below that height, the combined thermal, magnetic, cosmic-ray, and ram pressures exceed the estimated gravitational pressure. 2
• Why read: The study connects a phase-sensitive radio observable to the transport of a magnetized outflow over a resolved vertical scale. The pressure comparison gives simulations a concrete height range in which wind acceleration is plausible.
• Caveat: The one-dimensional fits describe the observed transport profiles; they do not uniquely separate cosmic-ray pressure from the other pressure terms or determine a total mass-loading rate. NGC 253 is one edge-on starburst, so the fitted transition should not be treated as a universal wind law.

NGC 4438 shows why jet scaling depends on radius

• Authors/date: Luan Luan et al.; submitted August 26, 2026 and updated August 27, 2026; v2; accepted by ApJ.
• Study design: NGC 4438 contains two nuclear bubbles on roughly 200 pc scales and a lopsided outflow extending to about 10 kpc. The authors combine radio and X-ray modeling with a spatially resolved inventory of bulk kinetic, thermal, cosmic-ray, magnetic, and radiative energy. 3
• Result: Bulk kinetic energy dominates the current budget of the nuclear bubbles, while roughly half of the injected energy has been converted into thermal, cosmic-ray, and magnetic energy or lost radiatively. The thermal and magnetic pressures remain consistent within uncertainties across the bubble sizes examined. A radio-luminosity/jet-power relation calibrated on kiloparsec-scale bubbles matches the galaxy-scale outflow but substantially overestimates the power of the 200 pc nuclear bubbles. 3
• Why read: The paper puts the coupling-location question on a single object with multiple spatial scales. A jet-power proxy calibrated at kiloparsec scales cannot be moved inward without checking how the energy has been partitioned.
• Caveat: The nuclear and galaxy-scale structures are linked to the same AGN only plausibly in the abstract's description. The energy inventory depends on joint radio-X-ray modeling, and the unresolved central engine leaves the highest-energy particle acceleration mechanism open.

AGN coupling and long-timescale quenching

A jet prescription puts energy where the bubbles put it

• Authors/date: Ross J. Turner, Andrew Sullivan, and William R. Q. Gaffney; submitted August 26, 2026; v1; accepted in MNRAS.
• Study design: The authors extend the Radio AGN in Semi-analytic Environments (RAiSE) dynamical model into an analytic prescription for the radial and polar-angle distribution of AGN feedback energy. The calculation follows buoyantly rising bubbles subject to ablation and the gravitational collapse of swept-up gas in shocked shells across 10 representative cluster environments. 4
• Result: Buoyant bubbles deposit energy preferentially near the core radius where the density gradient is steep, while collapsing shocked shells heat flatter cluster cores. Outbursts with W and Myr confine their energy to the inner 10 kpc. For Myr, less than 1% of the injected energy remains within 30 kpc. Repeated outbursts offset radiative cooling in all but the densest cores within 100 kpc for duty cycles . 4
• Why read: This paper turns anisotropic jet evolution into a prescription that can be inserted into larger simulations. It gives a direct way to test whether a feedback model deposits energy at the radius where an observable requires it.
• Caveat: The prescription is analytic and calibrated against representative environments rather than a single resolved cluster sample. The cooling-offset result is conditional on the assumed bubble and shocked-shell evolution, duty cycle, and cluster-core density.

MACER separates AGN luminosity from cumulative gas removal

• Authors/date: Yuxuan Zou et al.; submitted August 26, 2026; v1; to be submitted to ApJ.
• Study design: High-resolution MACER hydrodynamic simulations include multiphase gas, star formation, stellar feedback, cosmological gas inflow, and self-consistent radiative and mechanical AGN feedback. The simulations follow one disk galaxy across to 10 and compare instantaneous luminosity with cold-gas content and star formation. 5
• Result: The simulated host retains a cold-gas fraction with little dependence on instantaneous AGN luminosity, while it still undergoes strong gas depletion and quenching. During the quenching phase, the cold-gas mass falls by nearly three orders of magnitude. AGN luminosity varies over roughly – years, whereas repeated outflows deplete the galaxy-scale reservoir over about 1 Gyr. 5
• Why read: The simulation gives a timescale-based explanation for why gas-rich quasar hosts and effective long-term ejective feedback can coexist. Cold-gas mass tracks the star-formation rate more closely than the instantaneous AGN luminosity in this setup.
• Caveat: The result comes from one high-resolution disk-galaxy framework and its chosen inflow and feedback prescriptions. The simulation reproduces a weak luminosity-gas connection; it does not show that every gas-rich AGN host has undergone the same cumulative history.

The metallicity relation reverses in massive galaxies

• Authors/date: Laura Carnevale et al.; submitted August 25, 2026; v1.
• Study design: The authors compare EAGLE, SIMBA, Illustris, and IllustrisTNG simulations with Sloan Digital Sky Survey observations. The analysis tests the relation among stellar mass, gas-phase metallicity, and star-formation rate, with special attention to the metallicity versus specific-star-formation-rate trend. 8
• Result: In EAGLE, IllustrisTNG, and SDSS, the usual anti-correlation between metallicity and sSFR reverses above . The positive correlation appears across all four simulations and SDSS when lower-star-forming galaxies are included. The reversal persists to about in the simulations. The authors associate the pattern with strong nuclear outflows that can quench star formation while expelling enriched gas. 8
• Why read: This paper offers a cross-simulation and observational test for models in which outflows change both star formation and the composition of the gas left behind. The mass threshold gives a concrete regime in which a simple gas-regulator interpretation becomes harder to apply.
• Caveat: The nuclear-outflow explanation is presented as an interpretation, not as a uniquely identified engine. The observed reversal depends on the metallicity diagnostic, and the simulation comparison does not directly measure the CGM gas expelled by the putative outflows.

IGM accounting and measurement systematics

PIFFLE assigns an FRB excess to foreground halos

• Authors/date: Qi Guo et al.; submitted August 28, 2026; v1.
• Study design: The paper analyzes FRB20230907D, localized to a host at with observed dispersion measure . Subaru/PFS and SDSS spectroscopy, published group catalogs, Rubin/LSST imaging, and eROSITA X-ray data are combined to identify foreground structures. A friends-of-friends search and a modified-NFW halo-gas model then estimate the dispersion-measure contribution of each system. 6
• Result: The analysis identifies a foreground system at with and another group at . The two systems contribute estimated observer-frame dispersion measures of and pc cm, respectively, for a combined central value of 230 pc cm. The foreground structures, together with Milky Way, diffuse IGM, Virgo, M49, and host contributions, account for the excess within the model uncertainties. 6
• Why read: The paper shows how halo gas across four decades in halo mass can enter an IGM-style observable. For feedback studies, the practical lesson is that a high dispersion measure needs a foreground inventory before it can be used to infer diffuse baryons or thermal history.
• Caveat: The halo contribution is model-dependent because the calculation assumes a modified-NFW gas profile. The paper explains one sightline; the result is a foreground decomposition rather than a direct measurement of feedback energy.

Ly$\alpha$ forest power needs numerical convergence

• Authors/date: Tomáš Šoltinský et al.; submitted August 25, 2026; v1.
• Study design: The authors use the Sherwood and Sherwood-Relics cosmological hydrodynamic simulations, with box sizes up to 160 cMpc, to calculate the three-dimensional Ly$\alpha$-forest flux power spectrum over . The analysis varies simulation volume, mass resolution, optical-depth grid resolution, reionization timing, and photoheating rate, then applies a Zel'dovich control-variate correction. 7
• Result: Box size has a modest effect over most scales and orientations after the correction. Lower mass resolution changes the spectrum by up to about 13%, while an insufficient optical-depth grid can increase small-scale power by up to about 35%. Reionization timing leaves a percent-level imprint at , a factor-of-two change in photoheating changes large-scale power by about 4–8%, and spatially inhomogeneous H I reionization boosts large-scale power by up to 70% at . 7
• Why read: The paper identifies the numerical effects that a future Ly$\alpha$-forest feedback or thermal-history measurement must control before interpreting a small power-spectrum difference as astrophysics.
• Caveat: The result calibrates simulation requirements for an observable; it does not identify a particular feedback source. The relative sizes of the effects depend on the simulation suites, redshift range, and power-spectrum estimator used here.

What can be compared across the set