Pith. sign in

REVIEW 4 major objections 3 minor 87 references

Full radio SED fitting lowers inferred galaxy luminosities by ~3x and shifts the radio luminosity function ~1 dex at high z, explaining the persistent radio excess in cosmic star formation rates.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-02 06:40 UTC pith:7UIF75MV

load-bearing objection A useful SKA science-case chapter: the review is solid and the mock forecast is new, but the headline numbers are inherited from T-RECS and need to be presented as illustrative, not as measurements. the 4 major comments →

arxiv 2607.12073 v3 pith:7UIF75MV submitted 2026-07-13 astro-ph.GA

Tracing cosmic star formation history through radio continuum spectral energy distribution and non-thermal emission

Fangxia An , Fatemeh S. Tabatabaei , Nick Seymour , Maryam Khademi , Zunli Yuan , Wenjie Wang , Joe A. Grundy , Mark T. Sargent
show 6 more authors
Tao Wang Yuheng Zhang Yinghe Zhao Hiddo S. B. Algera Elias Brinks Mattia Vaccari
This is my paper
classification astro-ph.GA
keywords radio continuumspectral energy distributionstar formation historystar-forming galaxiesk-correctionradio luminosity functionSKA AA4cosmic star formation rate density
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper argues that estimating a galaxy's star formation rate from radio continuum requires the shape of its radio spectrum, not just the flux at one frequency. Using mock SKA AA4 surveys built from the T-RECS catalogue, the authors show that fitting the full 150 MHz–20 GHz radio SED—instead of assuming a fixed spectral index of 0.70—reduces the median rest-frame 1.4 GHz luminosity of star-forming galaxies from 5.16×10^22 to 1.57×10^22 W/Hz and tightens the scatter from 0.70 to 0.54 dex. This propagates into the radio luminosity function, shifting it by up to ~1 dex at high redshift, and offers a plausible explanation for why radio-derived cosmic star formation rate densities from the MeerKAT DEEP2 field sit systematically above UV–IR estimates. If true, the broad frequency coverage of SKA AA4 is not a luxury but a necessity for an unbiased census of star formation across cosmic time.

Core claim

The central claim is that the widespread single-power-law k-correction (alpha=0.70) overestimates rest-frame 1.4 GHz luminosities of distant star-forming galaxies because real radio SEDs flatten at low frequencies and curve over 150 MHz–20 GHz. Fitting the full SED with SKA AA4 data lowers the median L_1.4 from 5.16e22 to 1.57e22 W/Hz, reduces scatter from 0.70 to 0.54 dex, and moves the inferred evolving RLF by up to about a dex at high redshift. The paper connects this directly to the MeerKAT DEEP2 SFRD excess over UV–IR estimates, and shows the Ultra-deep Tier (0.05 uJy/beam at 1.4 GHz) will reach SFRs of ~1 Msun/yr at z~5 and ~10 Msun/yr at z~8, constraining the faint end of the RLF to z

What carries the argument

The load-bearing tool is broadband radio SED fitting for k-corrections: three assumptions are compared—fixed spectral index alpha=0.70 (single-band), a two-point index between SKA-Mid Bands 1 and 2, and a Bayesian fit to the full 150 MHz–20 GHz SED. The full-SED fit captures low-frequency flattening (free-free absorption) and spectral curvature (cosmic-ray electron ageing and inverse-Compton losses), which lowers inferred L_1.4 and shrinks scatter. Mock samples come from the T-RECS sfgsdeep catalogue; RLFs are derived with binned (Page–Carrera) and unbinned maximum-likelihood fits under the LADE evolution model g(z)=(1+z)^{p1+p2 ln(1+z)}, f(z)=(1+z)^{k1+k2 z}, selected by AIC/BIC.

Load-bearing premise

The forecast inherits its central result from the T-RECS mock catalogue: if real high-redshift star-forming galaxies do not show the low-frequency flattening and spectral curvature that T-RECS builds in, the factor-of-three luminosity drop and dex-level RLF shift would not occur; the paper itself labels these forecasts illustrative and notes the mock lacks flux uncertainties and cannot separate radio-quiet AGN.

What would settle it

Re-observe the same MeerKAT DEEP2 galaxies with SKA AA4 multi-band coverage, fit each source's full 150 MHz–20 GHz SED, and recompute rest-frame 1.4 GHz luminosities and the cosmic SFRD; if the full-SED SFRD no longer exceeds UV–IR estimates, the fixed-power-law bias hypothesis survives—if the excess persists, the explanation fails. A cheaper version: apply the same k-correction comparison to existing MIGHTEE-COSMOS SEDs and check whether the measured spectral-index distribution (median alpha Band1–2 ≈ 0.76 ± 0.08 in the mock) matches the real data.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Rest-frame 1.4 GHz luminosities derived with a full SED fit drop by a factor of ~3 in the median, and the scatter across the sample shrinks from 0.70 to 0.54 dex.
  • The inferred radio luminosity function shifts by up to ~1 dex at z~3–5, meaning fixed-alpha k-corrections bias source counts and evolution—not just normalisation.
  • The MeerKAT DEEP2 excess of radio-based cosmic SFRD over UV–IR estimates has a direct, non-exotic explanation: the fixed alpha=0.70 assumption overestimated rest-frame luminosities.
  • Ultra-deep SKA AA4 tiers detect SFGs with SFRs down to ~1 Msun/yr at z~5 and ~10 Msun/yr at z~8, placing the faint end of the RLF within reach.
  • Integrated mid-radio (1–10 GHz) luminosities correlate almost linearly with total-infrared SFR (slope 1.00±0.04) with smaller scatter than monochromatic radio bands, making MRC the preferred tracer once an LF framework for integrated luminosities is built.

Where Pith is reading between the lines

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

  • If the factor-of-three reduction holds for real galaxies, previously published radio SFRDs that assumed a single power law may need systematic downward revision, narrowing the gap with UV–IR without invoking extra obscured star formation.
  • The same bias should affect radio-quiet AGN demographics: sources in the T-RECS mock that are actually RQ AGN are treated as SFGs, so a joint SED plus multi-wavelength AGN classifier would test whether part of the 'excess' originates in AGN contamination rather than spectral shape.
  • A direct testable extension: apply full-SED k-corrections to existing MIGHTEE-COSMOS data (1.5<z<3.5) and check whether the reported redshift-dependent flattening of alpha_nt matches the T-RECS SED shapes well enough to reproduce the 3x luminosity drop.
  • A luminosity function built specifically for integrated MRC luminosities—flagged by the paper as missing—could replace monochromatic 1.4 GHz LFs as the standard for cosmic SFH, since MRC–SFR is closer to linear and less scattered.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 3 minor

Summary. This SKA AA4 science chapter reviews radio continuum emission as a star-formation tracer and its use in measuring the cosmic star formation history. Sections 2–4 synthesize recent observational evidence that radio SEDs of star-forming galaxies are curved and evolve with redshift, discuss empirical SFR calibrations, and summarize current radio luminosity function (RLF) measurements. The new element is in Section 5, where mock samples constructed from the T-RECS catalogue are used to forecast SKA AA4 performance. The paper claims that broad-frequency SED fitting reduces the median rest-frame 1.4 GHz luminosity estimate from 5.16×10^22 W Hz^-1 (fixed α=0.70) to 1.57×10^22 W Hz^-1 (full 150 MHz–20 GHz fit), lowers the scatter from 0.70 to 0.54 dex, shifts the high-redshift RLF by up to ~1 dex, and thus 'provides a plausible explanation' for the MeerKAT DEEP2 SFRD excess reported by Matthews et al. (2024). It also forecasts deep SKA AA4 constraints on the RLF out to z~5.

Significance. The review portions of the chapter are valuable: they bring together independent multi-frequency observations (LOFAR, uGMRT, MeerKAT, VLA) showing spectral curvature and redshift-dependent spectral indices, and they make a strong case that fixed-slope k-corrections introduce systematic biases in radio SFRD measurements. The authors also deserve credit for explicitly acknowledging several limitations of their mock forecasts in §5.1.2 and §5.3, including the absence of flux uncertainties, contamination by radio-quiet AGN, and the absence of a stable luminosity-function framework for integrated mid-radio luminosities. If the quantitative forecast were robust, it would strengthen the rationale for SKA AA4's broad-band design and would help reconcile radio- and UV/IR-based cosmic SFRD estimates. However, because the headline numbers are inherited from the SED model built into the T-RECS mock catalogue rather than calibrated against independent data, the quantitative reach of the forecast is currently more limited than the presentation suggests.

major comments (4)
  1. [§5.1.3–5.1.4] The central quantitative result—median L_1.4 dropping from 5.16×10^22 to 1.57×10^22 W Hz^-1, scatter 0.70→0.54 dex, and RLF offsets up to ~1 dex—is effectively inherited from the T-RECS mock catalogue. T-RECS was constructed with curved radio SEDs, including the low-frequency flattening that the full SED fit recovers. The direction of the suppression is independently supported by observed spectral curvature (An et al. 2021, 2024; Tabatabaei et al. 2025), but the factor-of-three median shift and the ~1 dex RLF shift are not calibrated against observed SED-shape distributions. The word 'illustrative' appears in §5.1.2, yet the abstract and §5.1.3 present these numbers as SKA AA4's expected performance. Please add robustness tests varying the T-RECS SED prescription (e.g., turnover frequency, spectral-index distribution, redshift evolution) and/or calibrate the SED-shape distribution to the
  2. [§5.1.2–5.1.3] The mock catalogue has no measurement uncertainties, but the analysis applies a Bayesian SED fit to the tabulated flux densities and quotes scatter values as if they were forecast uncertainties. With noise-free fluxes, the 'Bayesian' fit reduces to a deterministic inversion of the catalogue's SED model, and the quoted scatter reflects the spread of input SED shapes, not the measurement scatter SKA AA4 will deliver. To support the claim that broad-band SED fitting reduces scatter from 0.70 to 0.54 dex as an observational forecast, the authors should inject realistic per-band noise (including calibration and confusion) and then evaluate the recovered luminosities and their uncertainties.
  3. [§5.1.2, §5.3] The T-RECS sfgsdeepcatalogue includes radio-quiet AGN, but the analysis treats all sources as SFGs because the catalogue lacks the information needed to separate them. The paper notes this caveat, but it does not propagate the effect into the RLF and SFRD forecasts. Since the high-redshift samples are exactly where AGN contamination is hardest to identify (see Fig. 10), the contamination could be a load-bearing systematic for the claimed z~5 RLF constraints. Please quantify the fraction of RQ AGN in the selected mock samples and demonstrate that the conclusions survive either by removing these sources using the full T-RECS information or by applying a statistical correction.
  4. [§5.3 (final bullet)] The chapter explicitly states that LF parameterisations used for monochromatic radio luminosities do not yield stable results when applied to integrated MRC luminosities. This is an important limitation, because Eq. (3) identifies MRC as the preferred SFR tracer, yet the cosmic SFH forecasts in §5.2 are based on monochromatic L_1.4 RLFs. The chain from broad-band SED fitting to improved cosmic SFH is therefore incomplete: the better SFR tracer (MRC) is not used in the final SFH forecast. Please either develop a stable MRC-LF framework (even in a simplified form) or state explicitly that the SFH forecasts remain tied to L_1.4 and that an MRC-based SFH framework is required to capitalise on the SED-fitting advantage.
minor comments (3)
  1. [§5.1.2 / Table 2] The survey parameters are referenced to 'Table 2 and Figure 6 of Prandoni et al. (2026)' but are not reproduced. Since the forecast results depend directly on these assumed depths and areas, including the key numbers in this chapter would improve readability and reproducibility.
  2. [§2, Eq. (1)] The symbols S(0), SFR(0), and D(0) are not all defined in the text immediately surrounding Eq. (1). A short definition of D(0) as the luminosity distance of the z≈0 analogue would help.
  3. [Figure 8 caption] The legend entries 'log(L1.4) with α = 0.7 3σ' and 'log(L1.4) with αLow−Mid 3σ' are ambiguous; it would be clearer to state explicitly that the shaded bands represent 3σ confidence intervals of the maximum-likelihood fits.

Circularity Check

0 steps flagged

No significant circularity: the SKA AA4 forecasts are explicitly mock-based and illustrative, not empirical derivations, and the empirical anchors (low-frequency flattening, MRC-SFR calibration) are independent observational results.

full rationale

The quantitative claims in §5.1.3–§5.1.4 (median L_1.4 dropping from 5.16e22 to 1.57e22 W/Hz and ~1 dex RLF shifts) are obtained by applying k-correction methods to the T-RECS mock catalogue. This is a simulation-based forecast, not a first-principles derivation about the sky. The paper explicitly states that 'the mock-catalogue-based forecasts presented in this section should be interpreted as illustrative predictions... rather than as precise predictions of the final SKA AA4 constraints' (§5.1.2). No fitted parameter is renamed as a prediction: the fixed-α vs SED-fit comparison is a controlled experiment on mock data, designed to show how much the choice of spectral model matters. The direction of the effect is anchored to independent observational evidence for low-frequency flattening (An et al. 2021, 2024), and the MRC-SFR calibration rests on the externally falsifiable MIGHTEE-COSMOS analysis (Tabatabaei et al. 2025). Overlap of some co-authors with those cited papers is normal scientific self-citation and does not make the cited observational results circular. The acknowledged limitation that T-RECS may not represent the true high-z SED distribution is a model-dependence caveat, not circularity. No equation in the paper reduces to its own inputs; therefore the circularity score is 0.

Axiom & Free-Parameter Ledger

4 free parameters · 5 axioms · 0 invented entities

The chapter introduces no new physical entities. Its freedom is concentrated in (a) hand-chosen and MCMC-fitted parameters describing the luminosity function and k-corrections, and (b) simulation and instrument assumptions inherited from T-RECS and Prandoni et al. The most load-bearing element is the T-RECS mock's embedded SED model, which effectively encodes the paper's main forecast.

free parameters (4)
  • Fixed spectral index alpha=0.70 = 0.70
    Hand-chosen representative synchrotron index used for k-correction in the fixed-index scenario and for the Ultra-deep Tier (§5.1.2, §5.1.3). The paper's comparison of L_1.4 estimates depends on this choice.
  • LADE evolution parameters p1, p2, k1, k2 = not quoted
    Fitted to mock RLFs via MCMC in §5.1.4 (Eq. 7). The inferred RLF evolution and the resulting SFRD forecasts depend on these values, which are not reported in the text.
  • Local RLF parameters (Phi*, L*, beta, sigma) = not quoted
    MCMC-fitted alongside the LADE parameters when fitting the modified Schechter function (Eq. 4) in §5.1.4. These shape the RLF used for the cosmic SFRD integration.
  • MRC-SFR calibration (Eq. 3) = normalization 10^(0.38±0.10), slope 1.00±0.04
    Adopted from Tabatabaei et al. 2025 to translate radio luminosities into SFRs for the sensitivity forecasts (e.g., SFR down to ~1 Msun/yr at z~5). These are externally fitted coefficients, not derived in this paper.
axioms (5)
  • domain assumption Flat LCDM cosmology with H0=67.27 km/s/Mpc, Omega_m=0.32, Omega_Lambda=0.68 (Planck 2016)
    Adopted in the Introduction and used for all distance and volume calculations in the mock forecasts.
  • domain assumption Radio SEDs of SFGs are composed of synchrotron and free-free emission, with curvature from free-free absorption and spectral ageing
    This physical picture underpins the SED fitting and the interpretation of spectral flattening in §2 and §5.1.3. It is standard in the field but not derived in the paper.
  • domain assumption T-RECS mock catalogue faithfully represents the radio source population (SED shapes, counts, redshift evolution)
    All quantitative forecasts in §5 are based on T-RECS; the paper explicitly calls the forecasts 'illustrative' but the numbers still rely on this external simulation being realistic.
  • domain assumption LADE model (Eq. 5) with luminosity evolution f(z) and density evolution g(z) describes the RLF over 0<z<5
    Used as the framework for all RLF fits in §5.1.4. Although 25 functional forms are tested, the assumed LADE family excludes other physical parameterizations.
  • domain assumption SKA AA4 survey parameters from Prandoni et al. (2026) are achievable (sensitivities, areas, confusion limits)
    The Deep multi-band and Ultra-deep tier sensitivities and areas are taken as given from the companion chapter; the forecasts stand or fall with those instrument assumptions.

pith-pipeline@v1.3.0-alltime-deepseek · 23061 in / 11219 out tokens · 118591 ms · 2026-08-02T06:40:49.698631+00:00 · methodology

0 comments
read the original abstract

As a tracer of massive star formation unaffected by dust, the radio continuum emission provides a unique window into the formation of the first stars and galaxies in the Universe. Recent observations show that the integrated rest-frame mid-radio (~1-10 GHz) luminosity of galaxies serves as one of the most robust tracers of the star formation rate (SFR). These studies further demonstrate that the synchrotron spectral index and the shape of the radio spectral energy distribution (SED) evolves with redshift as a consequence of the cosmic evolution of star formation activity. These findings underscore the importance of deep multi-band radio continuum observations in calibrating the SFR of early galaxies and understanding the astrophysical processes governing their assembly and evolution over cosmic time. This chapter presents recent progress in radio SFR calibrations for star-forming galaxies (SFGs) and reviews radio-continuum studies of the cosmic star formation history (SFH). We highlight the transformative potential of SKA AA4, whose broad frequency coverage and high sensitivity will enable well-constrained radio SEDs for SFGs across a wide redshift range.

Figures

Figures reproduced from arXiv: 2607.12073 by Elias Brinks, FangXia An, Fatemeh S. Tabatabaei, Hiddo S. B. Algera, Joe A. Grundy, Mark T. Sargent, Maryam Khademi, Mattia Vaccari, Nick Seymour, Tao Wang, Wenjie Wang, Yinghe Zhao, Yuheng Zhang, Zunli Yuan.

Figure 1
Figure 1. Figure 1: Theoretical radio SED of a typical star-forming galaxy in the observed-frame at different cosmic times (𝑧 = 0.1 or local, 𝑧 = 3 or cosmic noon, and 𝑧 = 7 at the epoch of reionisation) assuming no evolution. Also shown are the SKA-LOW (pink) and SKA-MID (cyan) frequency coverage. The radio spectra comprise a free-free (FF) and synchrotron (Syn) components, both of which are suffering the same free-free abso… view at source ↗
Figure 2
Figure 2. Figure 2: Left- non-thermal spectral index 𝛼nt against redshift z for MIGHTEE-COMOS SFGs located at 1.5 < z < 3.5 (Tabatabaei et al., 2025). The curve show best fit to the mean values in redshift bins of 0.2 (red points) with Pearson correlation coefficient, 𝑟 𝑝. Gray-shaded area show the 99% confidence bound. Right- The fitted rest-frame SEDs (solid lines) of a M51-like galaxy at five different redshifts adapted fr… view at source ↗
Figure 3
Figure 3. Figure 3: Selection of the commonly used 1.4 GHz to star formation rate conversion factors (Bell, 2003; Mancuso et al., 2015; Matthews et al., 2021; Molnár et al., 2021; Cook et al., 2024; Seymour et al., 2026). If determined over a limited redshift range, this is indicated in the legend. If determined over a limited range of SFRs or radio luminosities, this is indicated by the truncated lines. synchrotron emission … view at source ↗
Figure 4
Figure 4. Figure 4: ):  SFRTIR M⊙ yr−1  = 10 (0.29 ± 0.13) ×  SFR1.3 M⊙ yr−1 (0.75 ± 0.04) , (2)  SFRTIR M⊙ yr−1  = 10(0.38±0.10)  SFRMRC M⊙ yr−1 (1.00 ± 0.04) . (3) This is expected because a more complete sampling of CREs with different energies is attained using the integrated radio SED than a single radio band. 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 log10[SFR1.3(M yr 1 )] 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 l o g 1 0[S F… view at source ↗
Figure 5
Figure 5. Figure 5: Cosmic star formation rate density (SFRD) as a function of redshift. The figure is reproduced from Cochrane et al. (2023), who used the high-sensitivity 150 MHz data from LoTSS to estimate the cosmic SFRD up to 𝑧 ≲ 4. Other recent radio continuum–based measurements are overlaid and listed in the rightmost column of the legend. Except for the results from Karim et al. (2011, VLA-COSMOS 1.4 GHz), Novak et al… view at source ↗
Figure 6
Figure 6. Figure 6: Rest-frame 1.4 GHz luminosity as a function of redshift for the mock SFG sample, assuming a fixed radio spectral index of 𝛼 = 0.70. The left and right panels show the predicted source distributions for the Deep multi-band Tier and Ultra-deep Tier survey strategies, respectively, following the survey parameters presented in [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Left: Distributions of rest-frame 1.4 GHz luminosities derived using three spectral-index as￾sumptions in the 𝑘-correction: a fixed 𝛼 = 0.70 (filled grey), a two-point spectral index measured between SKA-Mid Bands 1 and 2 (blue hatched), and a spectral index fitted using the T-RECS tabulated flux densities from 150 MHz to 20 GHz, covering most of the SKA-Low frequency range and the full SKA-Mid range consi… view at source ↗
Figure 8
Figure 8. Figure 8: Rest-frame 1.4 GHz RLFs in different redshift bins. The points with error bars show the binned RLFs estimated using the method of Page and Carrera (2000), while the smooth curves are derived from a full maximum-likelihood analysis of the underlying sample. Grey symbols and curves correspond to luminosities computed assuming a fixed spectral index of 𝛼 = 0.70, appropriate when only SKA-Mid Band 2 observatio… view at source ↗
Figure 9
Figure 9. Figure 9 [PITH_FULL_IMAGE:figures/full_fig_p021_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Predicted fraction of SKA-detected sources without optical counterparts from LSST. Blue squares, yellow circles and red hexagons denote respectively the fraction of SKA detected galaxies (𝑆1.4 GHz ≥ 2 𝜇Jy, corresponding to the 5𝜎 detection threshold of the Deep multi-band Tier) that are fainter than the limiting magnitudes of the LSST 𝑟 band (𝑟 = 27.5), the Euclid Wide and Deep surveys (𝐻 = 24.5 and 26.5,… view at source ↗

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

87 extracted references · 7 canonical work pages

  1. [5]

    doi: 10.1088/0004-637X/803/2/51. R. Beck.A&ARv, 24:4, Dec

    doi:10.1088/0004-637x/803/2/51

  2. [6]

    doi: 10.1007/s00159-015-0084-4. P. S. Behroozi, R. H. Wechsler, and C. Conroy.ApJ, 770(1):57, June

    doi:10.1007/s00159-015-0084-4

  3. [10]

    doi: 10.1051/0004-6361/202141286. R. Bouwens et al.ApJ, 902(2):112, Oct

    doi:10.1051/0004-6361/202141286

  4. [12]

    doi: 10.1088/0004-637X/803/1/34. D. Calzetti et al.ApJ, 714(2):1256–1279, May

    doi:10.1088/0004-637x/803/1/34

  5. [16]

    doi: 10.1111/j.1365-2966.2010.16534.x. R. K. Cochrane et al.MNRAS, 523(4):6082–6102, Aug

    arXiv 2010

  6. [17]

    doi: 10.1093/mnras/stad1602. J. J. Condon.ARA&A, 30:575–611, Jan

    doi:10.1093/mnras/stad1602

  7. [19]

    doi: 10.1093/mnras/stae1215. G. De Zotti et al.A&A, 689:A272, Sept

    doi:10.1093/mnras/stae1215

  8. [20]

    doi: 10.1051/0004-6361/202449313. J. Delhaize et al.A&A, 602:A4, June

    doi:10.1051/0004-6361/202449313

  9. [22]

    doi: 10.1051/0004-6361/201629367. I. Delvecchio et al.A&A, 647:A123, Mar

    doi:10.1051/0004-6361/201629367

  10. [23]

    doi: 10.1051/0004-6361/202039647. S. Dey et al.ApJ, 938(2):152, Oct

    doi:10.1051/0004-6361/202039647

  11. [26]

    doi: 10.1093/mnras/stw3088. A. Enia et al.ApJ, 927(2):204, Mar

    doi:10.1093/mnras/stw3088

  12. [27]

    doi: 10.3847/1538-4357/ac51ca. M. Fernández-Torreiro et al.MNRAS, 527(4):11945–11961, Feb

    doi:10.3847/1538-4357/ac51ca

  13. [28]

    doi: 10.1093/mnras/ stad3145. T. J. Galvin et al.MNRAS, 474:779–799, Feb

    doi:10.1093/mnras/

  14. [29]

    doi: 10.1093/mnras/stx2613. F. Gao, T. Wang, and Y. Wang.A&A, 694:A261, Feb

    doi:10.1093/mnras/stx2613

  15. [31]

    doi: 10.1093/mnras/stac1393. S. M. L. Greis et al.MNRAS, 470(1):489–499, Sept

    doi:10.1093/mnras/stac1393

  16. [32]

    doi: 10.1093/mnras/stx1252. J. A. Grundy et al.PASA, 42:e002, Jan

    doi:10.1093/mnras/stx1252

  17. [33]

    doi: 10.1017/pasa.2024.119. C. Gruppioni et al.MNRAS, 432(1):23–52, June

    doi:10.1017/pasa.2024.119 2024

  18. [35]

    25 SFH through SED & non-thermal emission An, Tabatabaei, Seymour et al

    doi: 10.1093/mnras/stae2528. 25 SFH through SED & non-thermal emission An, Tabatabaei, Seymour et al. M. J. Hardcastle et al.MNRAS, 462(2):1910–1936, Oct

    doi:10.1093/mnras/stae2528 1910

  19. [37]

    doi: 10.1093/mnras/stab3202. V. Heesen et al.AJ, 147(5):103, May

    doi:10.1093/mnras/stab3202

  20. [38]

    doi: 10.1088/0004-6256/147/5/103. V. Heesen et al.arXiv e-prints, art. arXiv:2208.11068, Aug

    Pith/arXiv arXiv

  21. [41]

    doi: 10.3847/1538-4357/aba044. F.-Y. Liu et al.MNRAS, 545(2):staf1961, Jan

    doi:10.3847/1538-4357/aba044

  22. [43]

    doi: 10.1051/0004-6361/201424937. C. Mancuso et al.ApJ, 810(1):72, Sept

    doi:10.1051/0004-6361/201424937

  23. [44]

    doi: 10.1088/0004-637X/810/1/72. J. Marvil, F. Owen, and J. Eilek.AJ, 149(1):32, Jan

    doi:10.1088/0004-637x/810/1/72

  24. [45]

    doi: 10.1088/0004-6256/149/1/32. A. M. Matthews, J. J. Condon, W. D. Cotton, and T. Mauch.ApJ, 914(2):126, June

    doi:10.1088/0004-6256/149/1/32

  25. [46]

    doi: 10.3847/1538-4357/abfaf6. A. M. Matthews et al.ApJ, 966(2):194, May

    doi:10.3847/1538-4357/abfaf6

  26. [47]

    F.Mazoochi,R.Karimi,M.H.ZhoolidehHaghighi,andF.S.Tabatabaei.MNRAS,548(4):stag627, June

    doi: 10.3847/1538-4357/ad3912. F.Mazoochi,R.Karimi,M.H.ZhoolidehHaghighi,andF.S.Tabatabaei.MNRAS,548(4):stag627, June

    work page doi:10.3847/1538-4357/ad3912

  27. [48]

    doi: 10.1093/mnras/stag627. D. C. Molnár et al.MNRAS, 504(1):118–145, June

    work page doi:10.1093/mnras/stag627

  28. [49]

    doi: 10.1093/mnras/stab746. E. J. Murphy.ApJ, 706(1):482–496, Nov

    doi:10.1093/mnras/stab746

  29. [51]

    doi: 10.1088/0004-637X/737/2/67. E. J. Murphy et al.ApJ, 839(1):35, Apr

    doi:10.1088/0004-637x/737/2/67

  30. [52]

    doi: 10.3847/1538-4357/aa62fd. M. R. Nasirzadeh et al.A&A, 691:A199, Nov

    doi:10.3847/1538-4357/aa62fd

  31. [53]

    doi: 10.1051/0004-6361/202451110. M. Novak et al.A&A, 602:A5, June

    doi:10.1051/0004-6361/202451110

  32. [54]

    doi: 10.1051/0004-6361/201629436. E. F. Ocran et al.MNRAS, 491(4):5911–5924, Feb

    doi:10.1051/0004-6361/201629436

  33. [55]

    M.J.PageandF.J.Carrera.MNRAS,311(2):433–440,Jan.2000

    doi: 10.1093/mnras/stz3401. M.J.PageandF.J.Carrera.MNRAS,311(2):433–440,Jan.2000. doi: 10.1046/j.1365-8711.2000. 03105.x. F. Panessa et al.Nature Astronomy, 3:387–396, Apr

    arXiv 2000

  34. [56]

    Planck Collaboration et al.A&A, 594:A13, Sept

    doi: 10.1038/s41550-019-0765-4. Planck Collaboration et al.A&A, 594:A13, Sept

    doi:10.1038/s41550-019-0765-4

  35. [57]

    26 SFH through SED & non-thermal emission An, Tabatabaei, Seymour et al

    doi: 10.1051/0004-6361/201525830. 26 SFH through SED & non-thermal emission An, Tabatabaei, Seymour et al. I. Prandoni et al. InAdvancing Astrophysics with the SKA – II (AASKAII)

    doi:10.1051/0004-6361/201525830

  36. [58]

    doi: 10.1093/mnras/stv3020. B. Robertson et al.ApJ, 970(1):31, July

    work page doi:10.1093/mnras/stv3020

  37. [59]

    doi: 10.3847/1538-4357/ad463d. J. Sabater et al.A&A, 622:A17, Feb

    doi:10.3847/1538-4357/ad463d

  38. [60]

    doi: 10.1051/0004-6361/201833883. W. Saunders et al.MNRAS, 242:318–337, Jan

    doi:10.1051/0004-6361/201833883

  39. [61]

    A.M.M.Scaifeetal.MNRAS,406(1):L45–L49,July2010

    doi: 10.1093/mnras/242.3.318. A.M.M.Scaifeetal.MNRAS,406(1):L45–L49,July2010. doi: 10.1111/j.1745-3933.2010.00878. x. M. Schmidt.ApJ, 151:393, Feb

    arXiv 2010

  40. [63]

    doi: 10.3847/2041-8213/ab3c63. N. Seymour et al.ApJ, 998(2):306, Feb

    doi:10.3847/2041-8213/ab3c63 2041

  41. [64]

    doi: 10.3847/1538-4357/ae3a7d. W. Silima et al.MNRAS, 544(1):799–814, Nov

    doi:10.3847/1538-4357/ae3a7d

  42. [65]

    doi: 10.1093/mnras/staf1698. F. Sun et al.ApJ, 980(1):12, Feb

    doi:10.1093/mnras/staf1698

  43. [66]

    doi: 10.3847/1538-4357/ad9d0e. F. Sweijen et al.A&A, 658:A3, Feb

    doi:10.3847/1538-4357/ad9d0e

  44. [67]

    doi: 10.1051/0004-6361/202039871. F. Tabatabaei et al.ApJ, 989(1):44, Aug

    doi:10.1051/0004-6361/202039871

  45. [68]

    doi: 10.3847/1538-4357/ade233. F. S. Tabatabaei and E. M. Berkhuijsen.A&A, 517:A77, July

    doi:10.3847/1538-4357/ade233

  46. [69]

    doi: 10.1051/0004-6361/ 200913593. F. S. Tabatabaei et al.A&A, 475(1):133–143, Nov

    doi:10.1051/0004-6361/

  47. [71]

    doi: 10.1051/0004-6361/201220249. F. S. Tabatabaei et al.ApJ, 836(2):185, Feb

    work page doi:10.1051/0004-6361/201220249

  48. [72]

    doi: 10.3847/1538-4357/836/2/185. F. S. Tabatabaei et al.MNRAS, 517(2):2990–3007, Dec

    doi:10.3847/1538-4357/836/2/185

  49. [73]

    doi: 10.1093/mnras/stac2514. M. Talia et al.ApJ, 909(1):23, Mar

    doi:10.1093/mnras/stac2514

  50. [74]

    doi: 10.3847/1538-4357/abd6e3. N. J. Thykkathu et al.MNRAS, 548(3):stag616, May

    doi:10.3847/1538-4357/abd6e3

  51. [75]

    doi: 10.1093/mnras/stag616. K. Tisanić et al.A&A, 621:A139, Jan

    work page doi:10.1093/mnras/stag616

  52. [76]

    doi: 10.1051/0004-6361/201834002. D. van der Vlugt et al.ApJ, 941(1):10, Dec

    doi:10.1051/0004-6361/201834002

  53. [77]

    doi: 10.3847/1538-4357/ac99db. S. Verley, E. Corbelli, C. Giovanardi, and L. K. Hunt.A&A, 493(2):453–466, Jan

    doi:10.3847/1538-4357/ac99db

  54. [78]

    doi: 10.1051/0004-6361:200810566. T. Wang et al.ApJ, 816(2):84, Jan

    doi:10.1051/0004-6361:200810566

  55. [79]

    doi: 10.3847/0004-637X/816/2/84. T. Wang et al.Nature, 572(7768):211–214, Aug

    doi:10.3847/0004-637x/816/2/84

  56. [80]

    doi: 10.1038/s41586-019-1452-4. T. Wang et al.ApJL, 988(1):L35, July

    doi:10.1038/s41586-019-1452-4

  57. [81]

    27 SFH through SED & non-thermal emission An, Tabatabaei, Seymour et al

    doi: 10.3847/2041-8213/adebe7. 27 SFH through SED & non-thermal emission An, Tabatabaei, Seymour et al. W. Wang, Z. Yuan, H. Yu, and J. Mao.A&A, 683:A174, Mar. 2024a. doi: 10.1051/0004-6361/ 202347746. W. Wang et al.ApJ, 997(2):176, Feb

    doi:10.3847/2041-8213/adebe7 2041

  58. [82]

    doi: 10.3847/1538-4357/ae2474. Y. Wang et al.A&A, 685:A79, May 2024b. doi: 10.1051/0004-6361/202347787. I. H. Whittam et al.MNRAS, 543(1):507–517, Oct

    doi:10.3847/1538-4357/ae2474

  59. [83]

    doi: 10.1093/mnras/staf1505. T. Yang et al.ApJL, 1000(2):L42, Apr

    doi:10.1093/mnras/staf1505

  60. [84]

    doi: 10.3847/2041-8213/ae4803. I. Yoon et al. InAdvancing Astrophysics with the SKA – II (AASKAII)

    doi:10.3847/2041-8213/ae4803 2041

  61. [85]

    doi: 10.3847/1538-4365/ab855b. Z. Yuan et al.ApJSS, 260(1):10, May

    doi:10.3847/1538-4365/ab855b

  62. [86]

    doi: 10.3847/1538-4365/ac596a. J. A. Zavala et al.ApJ, 909(2):165, Mar

    doi:10.3847/1538-4365/ac596a

  63. [87]

    doi: 10.3847/1538-4357/abdb27. 28

    doi:10.3847/1538-4357/abdb27

  64. [1968]

    doi: 10.1086/149446. B. Sebastian and O. Bait.ApJL, 882(2):L19, Sept

    doi:10.1086/149446

  65. [1990]

    doi: 10.1093/mnras/245.1.101. M. S. Clemens, A. Scaife, O. Vega, and A. Bressan.MNRAS, 405(2):887–897, June

    work page doi:10.1093/mnras/245.1.101

  66. [1992]

    doi: 10.1146/annurev.aa.30.090192.003043. R. H. W. Cook et al.MNRAS, 531(1):708–727, June

    arXiv

  67. [2003]

    doi: 10.1086/367829. M. L. Bernet et al.Nature, 454(7202):302–304, July

    doi:10.1086/367829

  68. [2006]

    doi: 10.1086/506610. A. Karim et al.ApJ, 730(2):61, Apr

    doi:10.1086/506610

  69. [2007]

    doi: 10.1051/0004-6361:20078174. F. S. Tabatabaei et al.A&A, 552:A19, Apr

    doi:10.1051/0004-6361:20078174

  70. [2008]

    doi: 10.1038/nature07105. A. Bonaldi et al.MNRAS, 482(1):2–19, Jan

    doi:10.1038/nature07105

  71. [2009]

    doi: 10.1088/0004-637X/706/1/482. E. J. Murphy et al.ApJ, 737(2):67, Aug

    doi:10.1088/0004-637x/706/1/482

  72. [2010]

    doi: 10.1088/0004-637X/714/2/1256. C. M. Casey, D. Narayanan, and A. Cooray.Phys. Rep., 541(2):45–161, Aug

    doi:10.1088/0004-637x/714/2/1256

  73. [2011]

    doi: 10.1088/0004-637X/730/2/61. R. C. Kennicutt and N. J. Evans.ARA&A, 50:531–608, Sept

    doi:10.1088/0004-637x/730/2/61

  74. [2013]

    doi: 10.1093/mnras/stt308. C. L. Hale et al.MNRAS, 536(3):2187–2211, Jan

    doi:10.1093/mnras/stt308

  75. [2014]

    doi: 10.1016/ j.physrep.2014.02.009. X. Chi and A. W. Wolfendale.MNRAS, 245:101, July

    2014

  76. [2015]

    doi: 10.22323/1.215.0071. H. S. B. Algera et al. InAdvancing Astrophysics with the SKA – II (AASKAII)

    doi:10.22323/1.215.0071

  77. [2016]

    doi: 10.1093/mnras/stw1763. H. Hassani et al.MNRAS, 510(1):11–31, Feb

    doi:10.1093/mnras/stw1763

  78. [2017]

    doi: 10.1051/0004-6361/201629430. I. Delvecchio et al.A&A, 602:A3, June

    doi:10.1051/0004-6361/201629430

  79. [2018]

    doi: 10.1093/mnras/stx2728. J. S. Dunlop et al.MNRAS, 466(1):861–883, Apr

    doi:10.1093/mnras/stx2728

  80. [2019]

    24 SFH through SED & non-thermal emission An, Tabatabaei, Seymour et al

    doi: 10.1093/mnras/sty2603. 24 SFH through SED & non-thermal emission An, Tabatabaei, Seymour et al. M. Bonato et al.A&A, 656:A48, Dec

    doi:10.1093/mnras/sty2603

Showing first 80 references.