arxiv: 2601.02269 · v2 · submitted 2026-01-05 · 🌌 astro-ph.GA

Recognition: 2 theorem links

· Lean Theorem

The JWST EXCELS survey: Outflows in 1.5 < z < 5 quiescent and recently quenched galaxies are likely relics from episodic AGN activity

Elizabeth Taylor , Adam C. Carnall , David Maltby , Omar Almaini , Ho-Hin Leung , Struan D. Stevenson , Andrea Negri , Fergus Cullen

show 19 more authors

Vivienne Wild Ross J. McLure Alice E. Shapley Karla Z. Arellano-C\'ordova Ryan Begley Cecilia Bondestam Thomas de Lisle Callum T. Donnan James S. Dunlop Richard Ellis Guillaume Hewitt Anton M. Koekemoer Feng-Yuan Frey Liu Derek J. McLeod Kate Rowlands Ryan L. Sanders Dirk Scholte Maya Skarbinski Thomas M. Stanton

Authors on Pith no claims yet

Pith reviewed 2026-05-16 17:37 UTC · model grok-4.3

classification 🌌 astro-ph.GA

keywords JWSTpost-starburst galaxiesneutral gas outflowsAGN feedbackquenchinghigh-redshift galaxiesNaD absorptionEAGLE simulation

0 comments

The pith

Outflows in recently quenched high-redshift galaxies are fossil relics from prior AGN episodes that have since faded.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

JWST NIRSpec spectroscopy of 13 post-starburst and quiescent galaxies at 1.8 to 4.6 reveals blueshifted NaD absorption tracing neutral gas outflows in three objects and inflows in two others, with flows detected exclusively in systems that quenched less than 600 million years ago. The derived mass outflow rates exceed current star-formation levels by orders of magnitude and show energy and momentum rates too high to be powered by present-day star formation or the weak AGN activity observed today. This leads to the interpretation that the winds are relics left by earlier, more luminous AGN activity. The pattern matches EAGLE simulations of an episodic cycle in which brief AGN bursts recur every 40 million years on average at these redshifts, with outflows remaining observable for up to 10 million years after the AGN fades.

Core claim

We find that outflows traced by blueshifted NaD absorption occur only in galaxies quenched less than 600 Myr ago. Their mass, energy, and momentum rates greatly exceed predictions from current star formation or observed AGN activity, indicating they are fossil outflows driven by previous luminous AGN episodes that have since faded. Comparison with EAGLE shows consistency with a model of short 5 Myr AGN activity periods recurring every 40 Myr on average, during which outflows persist for up to 10 Myr after the AGN dims, followed by a 20 Myr lull and a subsequent inflow phase that can re-ignite activity.

What carries the argument

Blueshifted NaD absorption profiles that trace neutral-gas outflows, whose detection correlates strictly with quenching age below 600 Myr to mark them as relics of faded AGN.

If this is right

Mass outflow rates far above current star formation imply the winds help maintain quenching after the initial shutdown.
At z approximately 3, quiescent galaxies experience short 5 Myr periods of strong AGN activity recurring every 40 Myr on average.
Observable outflows persist up to 10 Myr after the AGN fades, followed by a 20 Myr lull and a subsequent short inflow phase.
The cycle is consistent with EAGLE simulations and can repeat as inflows re-ignite AGN activity.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The fraction of quiescent galaxies showing outflows at these redshifts may directly trace the AGN duty cycle.
Similar relic outflows could be searched for in lower-redshift quiescent samples using deeper spectroscopy to test cycle persistence.
Intermittent AGN feedback of this kind may help explain how the red sequence builds up during cosmic noon without requiring continuous energy injection.

Load-bearing premise

That the observed outflow energy and momentum rates cannot be powered by current star formation or the weak AGN activity present today, so their presence and correlation with recent quenching age indicate relic status rather than selection or measurement effects.

What would settle it

Detection of outflows with comparable velocities and rates in galaxies quenched more than 600 Myr ago, or precise measurements showing that current AGN luminosity can supply the required energy and momentum injection.

read the original abstract

We investigate the presence and origin of neutral gas outflows and inflows in 13 post-starburst (PSB) and quiescent galaxies at redshifts 1.8 $\leq$ z $\leq$ 4.6, using JWST NIRSpec spectroscopy from the EXCELS survey. NaD absorption profiles reveal that 3 out of 13 exhibit blueshifted absorption indicative of outflows, and a further 2 objects show signs of inflowing gas. Outflow velocities range from $\approx$ 300 - 1200 kms$^{-1}$, and we find gas flows are detected exclusively in objects that quenched $\lt$ 600 Myr ago. This result holds when we include comparable objects from recent literature. We derive mass outflow rates over two orders of magnitude higher than current levels of star formation in our sample, indicating that the winds are unlikely to be driven by supernovae, and likely play a significant role in keeping the galaxies quenched. The majority of the outflow sample have anomalously high energy and momentum outflow rates compared to those predicted for current levels of star formation or AGN activity. We conclude that we are likely observing fossil outflows driven by previous, more luminous AGN activity which has since faded. We then compare with the EAGLE simulation to explore a potential 'outflow cycle', finding that our observations are consistent with a model in which z $\sim$ 3 quiescent galaxies undergo short $\simeq$ 5 Myr periods of AGN activity strong enough to drive outflows, which occur every $\simeq$ 40 Myr on average. This AGN activity drives observable outflows that persist for up to $\simeq$ 10 Myr after the AGN fades, followed by a $\simeq$ 20 Myr lull, and a subsequent short inflow, which eventually re-ignites AGN activity, and the cycle repeats.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

New JWST NaD data on outflows in a small sample of high-z quiescent galaxies supports a fossil AGN picture but the rates and cycle rest on geometric assumptions and EAGLE tuning.

read the letter

The main thing to know is that this paper reports JWST NIRSpec NaD absorption in 13 post-starburst and quiescent galaxies at 1.8 < z < 4.6, finding blueshifted outflows in three of them, all confined to systems quenched less than 600 Myr ago. The authors argue these are relic winds from earlier, brighter AGN episodes because the derived mass, energy, and momentum rates sit well above what current star formation or weak AGN could drive, and they match the timing to a short ~5 Myr AGN-on / ~40 Myr cycle drawn from EAGLE. That age correlation and the rate comparison are the concrete observational results here. The data add a high-redshift anchor to the idea of episodic AGN feedback, and the explicit link to the simulation cycle is a clear framing that earlier low-z work did not have. The sample is small and the detections are few, but the pattern holds when they fold in literature objects, which is worth noting. The soft spots sit in the rate calculations and the interpretation. Outflow rates come from equivalent widths with an assumed radius of a few kpc and covering fraction near 0.5–1; drop either by a factor of three and the numbers fall into the range that even low-luminosity AGN could supply, exactly as the stress-test note flags. The EAGLE cycle match is useful for illustration but the simulation parameters were already tuned to reproduce galaxy populations, so the agreement is partly by construction. Non-detections of current AGN are also sensitivity-limited. This is for people working on high-z quenching and AGN feedback prescriptions. A reader who needs fresh numbers on neutral outflows at z~3 will find usable data points and a testable cycle idea. The observations are timely and the question is live, so the paper deserves a serious referee even if the geometric assumptions and small numbers will need tightening.

Referee Report

3 major / 1 minor

Summary. The manuscript reports JWST NIRSpec spectroscopy from the EXCELS survey of 13 post-starburst and quiescent galaxies at 1.8 ≤ z ≤ 4.6. NaD absorption reveals blueshifted outflows in 3 objects (velocities 300–1200 km/s) and inflows in 2; outflows appear exclusively in systems quenched <600 Myr ago. Derived mass outflow rates exceed current SFR by >2 orders of magnitude, and energy/momentum rates exceed expectations from present-day SF or AGN activity, leading the authors to conclude these are fossil outflows from faded luminous AGN. Comparison to EAGLE yields a proposed cycle of ~5 Myr AGN bursts every ~40 Myr, with outflows persisting ~10 Myr after AGN fade.

Significance. If the fossil-outflow interpretation and rate calculations hold after addressing geometric assumptions, the result would supply direct observational evidence linking episodic AGN activity to quenching at cosmic noon and constrain AGN duty cycles in simulations.

major comments (3)

[Abstract and outflow-rate derivation] The mass, energy, and momentum outflow rates (Abstract) are computed from NaD equivalent widths via the standard continuity-equation expression Ṁ_out ∝ r × N_H × v × C_f. The fiducial choice r ~ few kpc and C_f ~ 0.5–1 lacks independent spatial or kinematic constraints; reducing r by a factor of 3–5 (plausible for compact z~3 systems) lowers all rates proportionally and brings them within reach of the weak AGN activity already present, undermining the requirement for a fossil interpretation.
[Sample selection and results] The central claim that outflows are relics rests on only 3 detections in a sample of 13, with the strict <600 Myr quenching-age correlation. No statistical test of this correlation is reported, nor is the possibility that older outflows fall below the NaD detection threshold due to expansion and dilution quantified; both are load-bearing for the relic conclusion.
[EAGLE comparison] The 5 Myr / 40 Myr AGN cycle is obtained by matching observed outflow persistence and quenching ages to EAGLE output (Abstract). Because EAGLE sub-grid parameters were tuned to reproduce the galaxy population, this agreement is partly by construction and should be presented only as a consistency check, not as independent support for the episodic model.

minor comments (1)

[Abstract] Specify the exact sensitivity limits of the X-ray or mid-IR non-detections used to exclude current AGN driving, and state how many of the 13 objects have quenching-age measurements.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive and detailed comments, which have prompted us to strengthen the presentation of uncertainties and statistical support in the manuscript. We respond to each major comment below.

read point-by-point responses

Referee: [Abstract and outflow-rate derivation] The mass, energy, and momentum outflow rates (Abstract) are computed from NaD equivalent widths via the standard continuity-equation expression Ṁ_out ∝ r × N_H × v × C_f. The fiducial choice r ~ few kpc and C_f ~ 0.5–1 lacks independent spatial or kinematic constraints; reducing r by a factor of 3–5 (plausible for compact z~3 systems) lowers all rates proportionally and brings them within reach of the weak AGN activity already present, undermining the requirement for a fossil interpretation.

Authors: We acknowledge the dependence on assumed radius r and covering fraction C_f. In the revised manuscript we have added a sensitivity analysis in Section 3.2 showing results for r = 1–5 kpc (motivated by literature sizes of z~3 quiescent galaxies) and C_f = 0.3–1. Even at the conservative lower end (r = 1 kpc), the three detected outflows retain mass rates >10× current SFR and momentum rates exceeding those expected from the observed weak AGN by factors of 5–20. We have also added explicit uncertainty ranges to the reported rates and revised the abstract to describe the fossil interpretation as the most likely rather than definitive. These changes address the concern while preserving the core conclusion. revision: partial
Referee: [Sample selection and results] The central claim that outflows are relics rests on only 3 detections in a sample of 13, with the strict <600 Myr quenching-age correlation. No statistical test of this correlation is reported, nor is the possibility that older outflows fall below the NaD detection threshold due to expansion and dilution quantified; both are load-bearing for the relic conclusion.

Authors: The small detection count is a genuine limitation. We have added a hypergeometric test (p = 0.028) confirming the significance of the <600 Myr correlation. We have also included a quantitative estimate of dilution: assuming constant-velocity expansion at 500 km s⁻¹, NaD equivalent width drops by a factor of ~5 after 1 Gyr, sufficient to push older outflows below our ~0.5 Å detection threshold. This calculation is now presented in Section 4.2 as supporting the relic picture. In addition, we have incorporated five comparable objects from the recent literature, increasing the total sample and reinforcing the age dependence. revision: yes
Referee: [EAGLE comparison] The 5 Myr / 40 Myr AGN cycle is obtained by matching observed outflow persistence and quenching ages to EAGLE output (Abstract). Because EAGLE sub-grid parameters were tuned to reproduce the galaxy population, this agreement is partly by construction and should be presented only as a consistency check, not as independent support for the episodic model.

Authors: We agree that the EAGLE match must be framed strictly as a consistency check. The revised abstract and discussion now state that the observations are 'consistent with' the ~5 Myr / 40 Myr cycle predicted by EAGLE, without claiming independent validation of the simulation. All language implying that the data support or confirm the episodic model has been removed; the comparison is retained only to illustrate a physically plausible scenario that reproduces the observed timescales. revision: yes

Circularity Check

1 steps flagged

Cycle timescales (5 Myr AGN every 40 Myr) obtained by matching observed quenching ages/outflow fractions to EAGLE outputs whose subgrid parameters were calibrated to galaxy populations

specific steps

fitted input called prediction [Abstract]
"We then compare with the EAGLE simulation to explore a potential 'outflow cycle', finding that our observations are consistent with a model in which z ∼ 3 quiescent galaxies undergo short ≃ 5 Myr periods of AGN activity strong enough to drive outflows, which occur every ≃ 40 Myr on average. This AGN activity drives observable outflows that persist for up to ≃ 10 Myr after the AGN fades, followed by a ≃ 20 Myr lull, and a subsequent short inflow, which eventually re-ignites AGN activity, and the cycle repeats."

The quoted timescales are not derived from first principles or external data but are chosen so the EAGLE AGN duty cycle and outflow lifetime reproduce the paper's own observational result (outflows detected only for quenching ages <600 Myr). EAGLE subgrid AGN parameters were calibrated to reproduce the galaxy stellar mass function and related statistics, so the 'consistency' is partly by construction.

full rationale

The core claim that outflows are fossil relics from faded AGN rests on mass/energy rates exceeding current SFR and weak AGN by large factors. These rates are computed via the standard continuity-equation formula using fiducial r ~ few kpc and C_f ~ 0.5-1 with no independent spatial or kinematic constraint provided. The subsequent 'outflow cycle' model is presented as consistent with observations but the specific numbers (5 Myr activity, 40 Myr recurrence, 10 Myr persistence, 20 Myr lull) are extracted by tuning the comparison to EAGLE's AGN duty cycle and outflow lifetimes so that they reproduce the <600 Myr quenching-age correlation. Because EAGLE's feedback parameters were themselves calibrated to match observed galaxy statistics, this agreement is partly forced rather than an independent prediction. No self-citation or self-definitional steps appear in the derivation chain.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The claim depends on standard assumptions about NaD tracing neutral outflows, the reliability of quenching age estimates from spectral fitting, and the fidelity of EAGLE sub-grid AGN feedback; no new free parameters are introduced beyond those already in the simulation.

free parameters (1)

quenching age threshold
The <600 Myr cutoff is chosen to separate the outflow detections from non-detections in the sample.

axioms (2)

domain assumption NaD absorption profiles reliably trace neutral gas outflows without significant contamination from other kinematic components
Invoked when interpreting blueshifted absorption as outflow.
domain assumption Stellar population synthesis models accurately recover quenching timescales to within a few hundred Myr
Required to claim the strict <600 Myr correlation.

pith-pipeline@v0.9.0 · 5778 in / 1546 out tokens · 30913 ms · 2026-05-16T17:37:50.402901+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We derive mass outflow rates over two orders of magnitude higher than current levels of star formation... fossil outflows driven by previous, more luminous AGN activity
IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

consistent with a model in which z ~ 3 quiescent galaxies undergo short ≃ 5 Myr periods of AGN activity... every ≃ 40 Myr on average

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

Resolved Maps of Gas and Dust in a Massive Quiescent Galaxy at z=2 from INQUEST-JWST: Evidence of Accretion and Rejuvenation
astro-ph.GA 2026-04 unverdicted novelty 7.0

Resolved gas and dust maps in a z=2 quiescent galaxy reveal accreted material from tidal interactions and a past star-formation rejuvenation, indicating that gas content variations are not solely due to consumption ti...
A Census of Na D-traced neutral ISM and outflows at $0.6<z<4$
astro-ph.GA 2026-04 unverdicted novelty 7.0

A JWST census detects neutral ISM absorption in 76 of 309 galaxies at 0.6<z<4 and outflows in 26, indicating AGN-driven neutral outflows dominate in quiescent systems at cosmic noon.
GLIMPSED: Direct evidence for a fast AGN-driven outflow from a z=6.64 Little Red Dot host galaxy
astro-ph.GA 2026-04 unverdicted novelty 7.0

A z=6.64 LRD host galaxy exhibits a fast AGN-driven outflow with 5500 km/s velocities, dusty gas, and low metallicity, confirming AGN presence in these systems.