{"id":"796e2220-db17-4877-ac93-167d173464c7","arxiv_id":"2505.21929","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"M87's giant radio lobes are continuously refilled AGN outflows with an age of about 30-50 Myr, requiring (0.2-2) x 10^44 erg/s for the diffuse component and (1-11) x 10^44 erg/s for the whole source.","lead":"New images from the MWA and VLA trace M87's giant diffuse radio lobes and give their spectra from 60 MHz to 10.55 GHz. The lobes appear to be continuously refilled AGN-driven bubbles, about 30 to 50 million years old, with outflow powers near 10^44 erg per second.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 30–50 Myr age and outflow powers depend on interpreting the 1.72 GHz spectral break as CI/JP synchrotron aging; if the break instead reflects lost short-spacing high-frequency flux or a multi-population spectrum, the continuous-injection claim and age-specific powers fail.","rationale":"The reader's weakest assumption and my concern overlap, but I sharpen it with a concrete systematic: frequency-dependent short-spacing loss in the retained S-band images. The paper explicitly excludes the six highest S-band images for this reason but does not demonstrate that the retained images are unaffected, and the 10.55 GHz anchor is a single-dish point whose recovery of the extended diffuse lobes is not quantified. The manuscript's own caution about the JP-model correction factor and the Beq dependence (Section 3.4) reinforces that the age is model-dependent. I do not escalate to REJECT because the jet can supply the required power also via the independent Pkin–L151 and radio-luminosity estimates, and the stellar-wind disfavoring is robust. However, the 'continuously injected' framing and the precise P_out ranges are conditional on the aging interpretation, so the reader's CONDITIONAL verdict is appropriate, and my proposed test would settle the load-bearing issue.","tokens_in":33118,"tokens_out":14033,"duration_ms":152811,"concrete_test":"Re-image the VLA S-band data over the full 2–4 GHz band with a short-spacing correction, e.g., by joint deconvolution with the Effelsberg 10.55 GHz total-power map or by adding zero-spacing flux, and re-extract the diffuse-region spectrum. Refit the CI and JP models to the corrected 0.06–10.55 GHz data. If the recovered high-frequency flux is consistent with a power law through 4 GHz (νb shifts to >4 GHz), the 1.72 GHz break and the 30–50 Myr age are artifacts of missing short spacings; if the break persists after correction, the aging interpretation is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim is the continuously injected outflow power. In §3.3.2 the six highest-frequency S-band images are excluded because their total flux drops from missing short baselines, and the retained S-band points are imaged with a uv-range whose shortest spacing grows with frequency. If the reconstructed high-frequency diffuse flux is progressively underestimated, the fitted CI break at νb = 1.72 GHz in §3.3.2 is an artifact rather than an aging break. Eq. (3) then yields a spuriously long 35 Myr lifetime, and §4.1's excavation-time powers via Eq. (4) are not valid. The JP fits to R1–R3 (νb = 6–13 GHz) do not resolve this: the two highest breaks lie near or above the highest retained frequency, and the factor 2–3 JP-to-dynamical-age correction is taken from the literature, not measured here. The sound-crossing time (~54 Myr) is consistent only after that correction, so it does not independently confirm the CI age. The qualitative conclusion that a jet can supply the energy is more robust because it also follows from the Pkin–L151 and 1% radio-efficiency estimates, but the 'continuously injected' interpretation and the wind-versus-jet discrimination rest on the aging interpretation.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents MWA (70-230 MHz) and VLA (1-4 GHz) observations of the ~46 kpc diffuse radio lobes of M87, supplemented by LOFAR, 325 MHz VLA, and 10.55 GHz Effelsberg data. After flux scaling to the RCB scale, the authors construct 60 MHz-10.55 GHz spectra of the lobes' diffuse region and three subregions, fit a continuous-injection model with alpha_inj = -0.86 and nu_b = 1.72 GHz to the diffuse region and JP models with nu_b = 6-13 GHz to the subregions, derive B_eq ~ 10 uG from equipartition, and convert the break frequencies into synchrotron ages of 30-50 Myr (including a literature-based correction for the JP ages). Combining these ages with pressure/volume energy estimates yields outflow powers of ~(0.2-2) x 10^44 erg/s for the diffuse lobes and ~(1-11) x 10^44 erg/s for the whole source. The paper argues that stellar winds cannot supply this power, that the AGN jet can, and that the current AGN wind is insufficient unless average AGN activity was ~100 times higher over the past 30-50 Myr.","tokens_in":33418,"tokens_out":10892,"duration_ms":104425,"significance":"The observational core is a useful contribution: the imaging and flux-scaling procedures are standard and transparent, the wideband spectral coverage is significantly better than earlier work, and the spectral index maps and flux tables will be of lasting value. If the synchrotron-aging interpretation is correct, the paper provides one of the more complete energy budgets for M87's large-scale lobes and a concrete constraint on AGN feedback. The main caveat is that the quantitative age and power claims are conditional on identifying the 1.72 GHz break as single-population CI/JP radiative aging; this assumption, together with the short-spacing systematics at high frequency, needs to be tested explicitly before the headline numbers can be regarded as robust.","major_comments":[{"comment":"The six highest-frequency S-band images are excluded because their total flux drops when short baselines are absent, but the retained S-band images are not shown to be immune to the same effect. Table 1 gives the S-band uvmin growing from 0.21 to 0.38 kλ, and the footnote in §3.3.2 states that M87's ~9' lobes require baselines shorter than ~0.38 kλ. If the high-frequency diffuse flux is progressively underestimated across the retained points, the fitted CI break at nu_b = 1.72 GHz would be an imaging artifact rather than a synchrotron-aging break, which would invalidate the age from Eq. (3) and the excavation-time powers from Eq. (4). I request a quantitative stability test: refit the CI model with the highest retained S-band points removed, compare the reconstructed diffuse flux against independent single-dish/Effelsberg measurements at comparable frequencies, and vary the lower uv cutoff to show that nu_b is stable.","section":"§3.3.2, Table 1, Fig. 7"},{"comment":"The central 'continuously injected outflow' conclusion identifies the 1.72 GHz spectral break with radiative aging of a single CI electron population, but curvature of this kind can also result from a superposition of multiple outbursts, re-acceleration, adiabatic losses, or a non-power-law injection spectrum. The JP fits do not remove this degeneracy: the fitted break frequencies for R1–R3 are 5.7–12.7 GHz, all above the highest retained frequency (~3.5 GHz), so the JP ages are model extrapolations rather than direct measurements. I ask for a formal comparison (e.g., CI versus two-population or interrupted-CI fits to the same 48-point spectrum, with an information criterion) or, failing that, an explicit statement that the age and power numbers are conditional on the single-population aging interpretation.","section":"§3.3.2–§3.4, Table 5"},{"comment":"The sound-crossing time does not independently confirm the CI age. The raw JP lifetimes are 11–15 Myr, and the agreement with t_s ~ 54 Myr is obtained only after multiplying by a literature-based factor of 2–3 (Turner et al. 2018a; Mahatma et al. 2019). As written, the 'confirmation' is built into the adopted correction. Please show the uncorrected comparison explicitly and justify the applicability of the 2–3 correction to these specific lobe regions, or soften the claim that continuous injection is confirmed by the sound-crossing time.","section":"§4.1 and end of §3.4"},{"comment":"There are two inconsistent estimates of the current AGN-wind power in this section. The first, P_w = 0.5 Mdot_w v_w^2 with Mdot_w ~ 0.1–0.2 M_sun/yr and v_w ~ 0.2c, gives P_w up to 2.3 x 10^44 erg/s, which is sufficient to produce the lobes' diffuse components; the second, Eq. (5), gives ~10^41–10^42 erg/s. The conclusion that the current wind cannot power the lobes rests entirely on the second estimate and on the adopted launching radius R_launch ~ 10^2–10^4 R_s. The authors should reconcile these estimates or explicitly identify the assumption that rules out the crude upper limit, and should propagate the resulting uncertainty into the 'few percent' statement.","section":"§4.2.2"}],"minor_comments":[{"comment":"A typical supernova releases 10^51 erg, not 10^51 erg s^-1; the units should be corrected.","section":"§4.2.1"},{"comment":"The telescope name is written inconsistently as 'MW A' and 'MWA'; use one form consistently.","section":"Throughout"},{"comment":"The model labels 'CI: b = 1.72 = 0.86' and 'JP: b = 12.71 = 0.86' are missing the symbol for the injection spectral index; the labels should read, e.g., 'alpha_inj = -0.86'.","section":"Figs. 7–8"},{"comment":"Equation (2) appears to have unbalanced parentheses; please check the typeset form of the Beck & Krause (2005) expression.","section":"§3.4, Eq. (2)"},{"comment":"The radio luminosity is computed assuming the spectrum extends from 10 MHz to 100 GHz, but the integration limits and the spectral model used outside the observed band are not specified; a brief statement would help reproducibility.","section":"§3.5"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid observational contribution and the data reduction looks careful, but the headline age and power claims are more model-dependent than the abstract implies. I would like to see the requested short-spacing stability tests and a reconciliation of the two AGN-wind power estimates before publication. No concerns about citation practice or journal scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a useful paper. It provides the best-sampled wideband spectra of M87's diffuse lobes to date, using MWA, VLA, LOFAR and Effelsberg data from 60 MHz to 10.55 GHz, and it ties those spectra to an energy budget for the kpc-scale lobes. The imaging, flux scaling and spectral fitting are standard and transparent, and the paper is honest about many of its assumptions.\n\nThe genuinely new items are the five-band MWA images, the VLA L- and S-band images, and the tighter CI break frequency, νb = 1.72 ± 0.28 GHz, along with new flux densities for the diffuse components and a more detailed power budget than previous work. The qualitative conclusion that the jet can energetically power the lobes and that galactic stellar winds are negligible is robust: it also follows from the Pkin–L151 relation and the 1% radio efficiency argument, independent of the aging analysis.\n\nThe soft spot is the quantitative claim of continuous injection with an age of 30–50 Myr and the associated outflow powers. This rests on treating the 1.72 GHz break as synchrotron aging of a single CI electron population. The stress-test concern about the excluded S-band points is legitimate and not fully addressed. The six highest-frequency S-band images were dropped because their total flux drops, and the shortest baseline grows with frequency, so the retained S-band points may progressively resolve out diffuse lobe flux. If so, the break is an artifact of missing short-spacing flux, and the age and excavation-time powers are not valid. The JP fits to R1–R3 do not rescue the interpretation because their breaks at 6–13 GHz lie near or above the highest retained frequency, and the factor 2–3 JP-to-dynamical-age correction is taken from the literature. The sound-crossing time is consistent only after that correction, so it is not independent confirmation.\n\nThe paper would be strengthened by a test of the missing-flux hypothesis (for example, using a single-dish measurement at S-band or modelling the uv coverage) and by softening the 'confirmation' language in Section 4.1. The reduced data products are not deposited, only the raw data links; that is a minor issue.\n\nThis deserves a serious referee. It is a careful single-object study with improved data products, and the main conclusions are probably right even if the quantitative ages and powers carry more uncertainty than the text suggests. I would send it to peer review with a request for the missing-flux test and language revision.","headline":"Solid wideband spectral study of M87's lobes with a load-bearing but addressable concern about missing short-spacing flux driving the break frequency.","tokens_in":34057,"tokens_out":3809,"would_cite":true,"duration_ms":38350,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Wideband radio spectra of M87's 46-kiloparsec diffuse lobes show they were inflated by a continuously injected outflow with power of order $10^{44}$ erg/s, pointing to AGN activity rather than stellar winds.","keywords":["galaxies: active","radio continuum: galaxies","galaxies: individual: M 87","techniques: interferometric","radio lobes","synchrotron ageing","continuous injection model","AGN feedback"],"falsifier":"Measure the hard X-ray inverse-Compton emission from the diffuse lobes: because the CI model fixes the electron population, the predicted IC flux at a given field strength is specific, and an observed field far from $B_{\\rm eq}\\simeq10\\,\\mu$G would change the synchrotron age and outflow power enough to re-open the wind-versus-jet question.","tokens_in":32895,"feed_emoji":"📡","tokens_out":9596,"duration_ms":82266,"temperature":0.7,"pith_summary":"The paper assembles a well-sampled radio spectrum of the two giant diffuse lobes of M87, spanning 60 MHz to 10.55 GHz, and asks what kind of outflow built them. It argues that the lobes were filled by a continuously injected outflow with power $\\sim(0.2-2)\\times10^{44}$ erg s$^{-1}$ for the diffuse emission and $\\sim(1-11)\\times10^{44}$ erg s$^{-1}$ for the whole radio source, sustained over roughly 30–50 Myr. On that energy budget, the paper concludes that supernova-driven galactic winds are negligible, that the AGN jet can supply the required energy, and that the AGN wind seen today is too weak unless the nucleus was on average about $10^2$ times more active in the past. This matters because M87 is close enough that these lobes can be used as a detailed test case for which feedback mechanism actually inflates radio lobes.","feed_headline":"Radio data pin M87's giant lobes to AGN outflows, not stars","feed_subtitle":"Well-sampled 60 MHz–10.55 GHz spectra date the lobes at 30–50 Myr and rule out stellar winds as the power source.","key_machinery":"The load-bearing machinery is the continuous-injection (CI) synchrotron-ageing model: the observed radio spectrum of a lobe is assumed to come from electrons injected continuously with a power-law spectrum that then cool by synchrotron and inverse-Compton losses, producing a break at $\\nu_{\\rm b}$. The break frequency, combined with the equipartition magnetic field through the synchrotron-lifetime formula, gives a radiative age; the age is then equated to the excavation time $t_{\\rm exc}=E/P_{\\rm out}$ to convert stored lobe energy into an outflow power. A second check is the JP impulsive-injection model applied to three steeper-spectrum subregions, whose age is corrected upward by a factor of 2–3 because the JP model underestimates dynamical ages for active lobes.","core_discovery":"Using MWA and VLA images together with LOFAR and Effelsberg data, the authors reconstruct 100-arcsecond-resolution spectra of the lobes' diffuse region and fit them with a continuous-injection (CI) synchrotron model, obtaining an injection spectral index $\\alpha_{\\rm inj}\\simeq-0.86$ and a break frequency $\\nu_{\\rm b}\\simeq1.72$ GHz. Equipartition analysis gives $B_{\\rm eq}\\simeq10\\,\\mu$G and a minimum pressure of $\\simeq9\\times10^{-12}$ dyn cm$^{-2}$. Comparing the synchrotron lifetime with the sound crossing time of the lobes yields an age of about 30–50 Myr, and equating that age to the excavation time gives outflow powers of $\\sim(0.2-2)\\times10^{44}$ erg s$^{-1}$ for the diffuse lobes and $\\sim(1-11)\\times10^{44}$ erg s$^{-1}$ for the whole source. From these numbers the paper concludes that galactic stellar winds cannot account for the lobes, the jet can, and the current AGN wind would need an average factor $\\sim10^2$ enhancement over the past 30–50 Myr to be the driver.","pith_inferences":["If the CI interpretation holds, the same spectral-age machinery could be applied to the lobes of other nearby low-luminosity AGN to map AGN duty cycles from the ratio of required past-to-present outflow power.","The factor-of-five gap between the lobe minimum pressure ($\\simeq9\\times10^{-12}$ dyn cm$^{-2}$) and the surrounding thermal pressure suggests that the lobes must carry substantial magnetic pressure or are not in pressure balance, which would change the excavation time and power estimates.","The misalignment between the lobe axis and the parsec-scale jet hints that the jet direction may have changed over tens of Myr; a decade-long proper-motion program on the lobe edges could look for the roughly 830 km s$^{-1}$ expansion that the sound-crossing argument predicts."],"forward_implications":["If the lobes are continuously inflated over 30–50 Myr, the present-day AGN wind power of roughly $10^{41-42}$ erg s$^{-1}$ cannot be the sole driver; the nucleus must have been on average about $10^2$ times more active in the past.","Galactic stellar winds are ruled out as the main driver: the star formation rate required, roughly 600–4700 $M_\\odot$ yr$^{-1}$ over the past 30–50 Myr, is orders of magnitude above M87's observed upper limit of $<0.08\\,M_\\odot$ yr$^{-1}$.","Jet power estimates assembled from different scales fall in the range $0.1\\times10^{44}$ to $10\\times10^{44}$ erg s$^{-1}$, bracketing the required outflow power of about $10^{44}$ erg s$^{-1}$.","The spectral uniformity and sharp edges of the lobes imply a turbulent, externally confined plasma, supporting the picture of a continuously injected, pressure-balanced outflow.","The agreement between the sound crossing time (about 54 Myr) and the synchrotron-based age supports the adopted 30° viewing angle and the continuous-injection scenario over an interrupted single outburst."],"supporting_citations":[{"why":"Supplies the RCB flux-scale method, the earlier CI fit to M87's halo with a ~40 Myr synchrotron age, and the LOFAR images at 60 and 140 MHz.","marker":"de Gasperin et al. (2012)"},{"why":"Provides the revised equipartition formulae used to derive $B_{\\rm eq}$ and minimum pressure from the 227 MHz intensity and spectral index maps.","marker":"Beck & Krause (2005)"},{"why":"Provides the synchrotron-lifetime formula (Eq. 3) converting break frequency and magnetic field into a radiative age.","marker":"Murgia et al. (2011)"},{"why":"Defines the impulsive-injection JP spectral model used for the three older-particle regions in the lobes.","marker":"Jaffe & Perola (1973)"},{"why":"Defines the continuous-injection spectral model and supplies the assumed 10 MHz–100 GHz radio band for luminosity estimates.","marker":"Pacholczyk (1970)"},{"why":"Provides the kinetic jet power–151 MHz luminosity relation used as an independent outflow-power estimate.","marker":"Godfrey & Shabala (2013)"},{"why":"Supplies the X-ray temperature and density profiles used for the sound speed, sound crossing time, and thermal pressure at the lobe radius.","marker":"Matsushita et al. (2002)"},{"why":"Supports the factor 2–3 upward correction of JP-derived dynamical ages, which underlies the 30–45 Myr source age.","marker":"Turner et al. (2018a)"},{"why":"Constrains the lobe viewing angle to <35°, used to deproject lobe size and compute volumes and energies.","marker":"Werner et al. (2010)"}],"fun_headline_variants":["M87's giant radio lobes: AGN-driven, not stellar","M87 lobes are AGN outflows, not stellar wind","M87's 46-kpc lobes powered by AGN, not stars","Radio spectra link M87's giant lobes to AGN jets","M87's lobes: AGN beats stellar wind"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The age and power estimates assume that the measured spectral break is synchrotron ageing of a single continuously injected electron population, so the break frequency can be converted into a radiative lifetime; if the break instead comes from multiple outbursts, re-acceleration, adiabatic losses, or a non-power-law injection spectrum, the 30–50 Myr age, the outflow powers, and the wind-versus-jet discrimination do not follow.","fun_headline_variants_meta":{"raw":{"variants":["M87's giant radio lobes: AGN-driven, not stellar","M87 lobes are AGN outflows, not stellar wind","M87's 46-kpc lobes powered by AGN, not stars","Radio spectra link M87's giant lobes to AGN jets","M87's lobes: AGN beats stellar wind"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001314,"raw_usage":{"total_tokens":5479,"prompt_tokens":1195,"completion_tokens":4284,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":811,"completion_tokens_details":{"reasoning_tokens":4196}},"tokens_in":811,"tokens_out":4284,"duration_ms":30198,"temperature":1.0,"reasoning_tokens":4196,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T13:19:44.956086+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the hard X-ray inverse-Compton emission from the diffuse lobes: because the CI model fixes the electron population, the predicted IC flux at a given field strength is specific, and an observed field far from $B_{\\rm eq}\\simeq10\\,\\mu$G would change the synchrotron age and outflow power enough to re-open the wind-versus-jet question.","supporting_citations":[],"review_version":1}