Pith. sign in

REVIEW 3 major objections 6 minor 130 references

Evolved stars return only a quarter of the gas M33 burns to form stars, so the galaxy will deplete its gas within a billion years unless fresh gas falls in.

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

2026-08-01 11:06 UTC pith:I7CQHKGA

load-bearing objection Solid full-disc mass-loss census for M33, but the headline 3σ ISM-depletion significance is inflated once the paper's own factor-two error on ζ is propagated. the 3 major comments →

arxiv 2607.19989 v1 pith:I7CQHKGA submitted 2026-07-22 astro-ph.GA astro-ph.SR

The UK Infrared Telescope M33 monitoring project. VI. Feedback from dusty stellar winds across the galactic disc

Atefeh Javadi , Jacco Th. van Loon , Mina Alizadeh , Seyed Azim Hashemi , Elham Saremi This is my paper
classification astro-ph.GA astro-ph.SR
keywords stars: mass-lossAGB starsred supergiantsgalaxies: M33interstellar mediumstar formationinfrared variabilitydusty winds
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.

This paper measures the winds from pulsating evolved stars across the entire disc of the galaxy M33, using near-infrared and mid-infrared monitoring to identify long-period variables and to convert their dust emission into gas mass-loss rates. The central result is a total mass-return rate of about 0.1 solar masses per year, which is 4.5 ± 1 times lower than the star formation rate of 0.45 ± 0.10 solar masses per year. If these numbers are right, M33's interstellar medium will be exhausted in about a gigayear unless gas is accreted from its surroundings. This matters because it is a direct, galaxy-wide measure of the feedback loop coupling dying stars to future star formation, and it shows that spiral galaxies do not necessarily recycle enough gas internally to keep forming stars indefinitely.

Core claim

The paper claims that the total rate at which dusty winds from AGB stars and red supergiants return mass to the ISM of M33 is ζ ≈ 0.1 M⊙/yr (within a factor of two), about a factor 4.5 ± 1 below the star formation rate. It also finds that mass-loss rate is approximately proportional to luminosity (birth mass), with weaker dependences on pulsation period and amplitude; super-AGB stars exceed 10^-4 M⊙/yr, and red supergiants show three mass-loss modes (below, around, above the nuclear burning rate). The dominant mass-loss phase lasts only 6e4-2e5 yr. The authors conclude external gas supply is required.

What carries the argument

The self-similar scaling relation for radiation-driven dusty winds, (τ L^(3/4)) / (ψ^(1/2) Ṁ) ≈ constant, connects dust optical depth, luminosity, gas-to-dust ratio, and mass-loss rate. The authors calibrate this by fitting SEDs of 294 stars with a dust radiative transfer model, then build empirical relations between near-IR colour and optical depth/bolometric correction, applying them to the full variable-star sample. A radially varying gas-to-dust ratio accounts for the metallicity gradient.

Load-bearing premise

Every mass-loss rate relies on the assumption that the wind is a steady radiation-pressure-driven outflow with a specific density structure; if pulsation shocks, binarity, or other driving mechanisms are important, all rates and the total feedback are systematically off.

What would settle it

Measure the outflow kinematics of dusty AGB stars in M33 (e.g., with ALMA observations of molecular lines) and compare the density structure to the radiation-driven wind model; a mismatch would invalidate the derived mass-loss rates and the 0.1 M⊙/yr total.

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

If this is right

  • M33 will exhaust its ISM in about 1 Gyr unless it accretes gas at roughly 0.35 M⊙/yr.
  • Mass-loss rate scales linearly with luminosity, so more massive evolved stars return proportionally more mass; super-AGB stars reach extreme rates.
  • The dominant mass-loss phase is a brief final episode (< 2e5 yr), much shorter than the TP-AGB or RSG phase.
  • Carbon stars do not dominate dust return; silicates dominate, with carbonaceous dust contributing ~1/4 in the outer disc and ~1/7 in the centre.

Where Pith is reading between the lines

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

  • If the feedback deficit is real, M33 should currently be accreting gas from its surroundings at a detectable rate; 21-cm observations of inflowing halo gas can test this.
  • The near-linear Ṁ-L relation suggests the integrated mass return is nearly independent of short-term star-formation history, providing a robust population-wide 'recycling rate' for other galaxies.
  • A targeted search for the most extreme, heavily obscured carbon stars would test whether the 0.1 M⊙/yr total is an underestimate; the authors note they may have missed a few such stars.

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

3 major / 6 minor

Summary. This paper presents the sixth instalment of the UKIRT M33 monitoring project. The authors combine WFCAM JHKs time-series photometry with Spitzer 3.6/4.5/8 micron imaging to identify long-period variables across the M33 disc, model the SEDs of 294 evolved stars with the dusty radiative transfer code, calibrate optical-depth and bolometric-correction relations in terms of near-IR colours, and apply these to the full variable and non-variable samples. They derive approximate proportionality between mass-loss rate and luminosity, estimate that the dominant dust-producing mass-loss phase lasts ~0.6-2e5 yr, and obtain a total mass-return rate from dusty stellar winds of zeta ~ 0.1 Msun/yr ("give or take a factor two"). Comparing with the star formation rate xi = 0.45 +/- 0.10 Msun/yr, they conclude that M33's ISM will be depleted within ~1 Gyr unless external gas is accreted.

Significance. If correct, this is one of the few galaxy-wide, empirical censuses of mass return from AGB stars and RSGs, and it provides a valuable benchmark for stellar evolution and ISM feedback models. The paper's strengths are its large, multi-epoch, near- and mid-IR dataset; the direct SED modelling of 294 sources; the careful rejection of foreground stars, YSOs and other contaminants; and the systematic cross-checks against spectroscopically confirmed carbon stars, RSGs and symbiotic stars, as well as against relations from the LMC, SMC and Milky Way. The central mass-return estimate is defensible at the factor-of-two level. However, as discussed below, the headline significance of the ISM-shortfall claim and the luminosity-scaling claim are not supported by the paper's own error budget and method.

major comments (3)
  1. [Sec. 5.2] The claim that the mass-return rate zeta = 0.1 Msun/yr falls short of the SFR xi = 0.45 +/- 0.10 Msun/yr "at more than 3 sigma significance" is not supported by the paper's own uncertainty. Section 4.3 states zeta is "give or take a factor two". The quoted ratio xi/zeta = 4.5 +/- 1 propagates only the SFR uncertainty, treating zeta as exact. With zeta in [0.05, 0.2] Msun/yr and xi = 0.45 +/- 0.10, the ratio spans roughly 1.8 to 9. The deficit is therefore not established at 3-sigma. Please propagate the full uncertainty on zeta, or present the comparison as a factor-of-two-limited estimate, and soften the abstract and conclusion accordingly.
  2. [Sec. 4.2.1, Eq. (1); Sec. 3.2] The reported near-proportionality between mass-loss rate and luminosity, log Mdot = (0.98 +/- 0.18) log L - 9.35, is partly a construction artifact. In Sec. 3.2, Mdot is obtained from the self-similar radiatively driven wind scaling (tau L^{3/4})/(psi^{1/2} Mdot) ~ constant, i.e. Mdot is proportional to tau L^{3/4} psi^{-1/2}. Thus even a sample with tau independent of L would show a slope of 0.75 in log Mdot versus log L. The fitted slope 0.98 +/- 0.18 differs from 0.75 by only ~1.3 sigma, so the data do not independently establish a slope of unity; they indicate only a weak positive correlation of tau with L on top of the assumed wind scaling. Please report the fit relative to this built-in baseline and avoid presenting Eq. (1) as an empirical discovery of Mdot proportional to L.
  3. [Sec. 5.1.2, Eq. (7), Fig. 19] The inference that the dominant mass-loss phase lasts 0.6-2e5 yr, shorter than the TP-AGB/RSG phase, is model-dependent in a way that is not fully acknowledged. The ratio eta in Eq. (7) compares integrated mass loss (Mdot_i times delta t_i) with birth mass. A discrepancy of factors 2.5-5 could be resolved either by shortening delta t, as the paper assumes, or by a systematic overestimate of Mdot (e.g. from the wind scaling or dust opacity assumptions) or by an underestimate of birth mass. The paper gives no independent calibration of the LPV duration; it assumes a 2e5 yr duration for carbon stars and then derives the other durations. Please state this degeneracy explicitly and present the timescale as conditional on the adopted Mdot calibration and model lifetimes.
minor comments (6)
  1. [Sec. 1] "Messier 1771" appears to be a typo; the intended object is Messier 33.
  2. [References and Appendix A] Corrupted names appear in the text: "Miko/suppress lajewska" and "tjhemsleves". Please proofread the bibliographic and appendix text.
  3. [Sec. 5.2, Fig. 16] The text refers to "Figure 20" when discussing the radial variation of the replenishment rate; the relevant radial profile appears to be Fig. 16. Please check all cross-references.
  4. [Sec. 3.1] The statement that "in some cases no acceptable match" led to changing T_star or T_d should be quantified: how many stars, and by how much were the temperatures changed? This affects reproducibility of the SED fits.
  5. [Sec. 3, Sec. 4.3] The carbon-star classification threshold of 1.5-4 Msun is model-based. The comparison with spectroscopically confirmed stars is reassuring, but a sensitivity test using e.g. a 3.5 Msun upper limit would help quantify the effect on the total carbon-star mass-return rate.
  6. [Sec. 2.2, Sec. 4.3] The survey completeness factor of 0.3-0.5 is derived from the Spitzer-variable comparison. Please clarify whether this is a detection completeness for LPVs or a classification completeness, since the two could affect the mass-return correction differently.

Circularity Check

1 steps flagged

Secondary Mdot-L proportionality is partly inherited from the assumed radiatively-driven-wind scaling; central feedback budget remains independently anchored.

specific steps
  1. self definitional [§3.1–3.2 scaling relations; §4.2.1 Eq. (1); abstract]
    "On the basis of self-similarity of radiatively-driven winds there are scaling relations ... such that the combination of (τL^{3/4})/(ψ^{1/2} Mdot) is approximately constant. ... The mass-loss rate then follows from the self-similar description of radiation-driven winds ... determined by the values for τ, L and ψ. ... log(Mdot/Msun/yr) = (0.98 ± 0.18) × log(L/Lsun) − 9.35 ± 0.05. This suggests proportionality between mass-loss rate and luminosity."

    Eq. (1) is fitted to Mdot values that were not measured independently but computed from the adopted scaling relation, which already contains log Mdot ∝ 0.75 log L + log τ − 0.5 log ψ. The fitted luminosity exponent (0.98 ± 0.18) therefore equals the built-in 0.75 plus the τ(L) relation from the same SED models; it is not an independent empirical determination. Since 0.75 is within about 1.3σ of the fitted slope, the abstract's claim that mass loss is 'approximately proportional to luminosity' largely restates the assumed wind scaling rather than testing it. The total mass-return budget does not rely on Eq. (1), so the circularity is partial and secondary.

full rationale

The central result — the total mass-return rate ζ ≈ 0.1 M⊙/yr and its factor-four shortfall relative to the star-formation rate — is an empirical census. Individual Mdot values are obtained from SED fits (τ, L, ψ) through the dusty radiative-transfer code, summed over the LPV population, corrected for completeness, and compared with independent external calibrations (Goldman et al. 2017; van Loon et al. 2005a; Groenewegen & Sloan 2018; the initial–final mass relation). That chain is not circular: the radiatively-driven-wind scaling relation is an external, explicitly stated assumption, and the derived budget is anchored to photometry. However, one secondary conclusion — that Mdot is 'approximately proportional to luminosity' (Eq. 1 and the abstract) — is partly circular, because every Mdot in the fit was computed from the same assumed scaling relation, which already contains L^{3/4}. The fitted slope 0.98±0.18 is within ~1.3σ of the built-in 0.75, so the proportionality claim mostly restates the input scaling rather than independently measuring a luminosity law. This does not invalidate the total feedback estimate, which does not depend on Eq. (1). The statistical overstatement of the ξ/ζ significance noted by the skeptic is a correctness/error-propagation issue, not a circularity, and is therefore not scored here.

Axiom & Free-Parameter Ledger

7 free parameters · 8 axioms · 0 invented entities

The ledger lists the radial gas-to-dust ratios, fixed temperatures, classification thresholds, an assumed LPV duration, and the fitted colour–optical-depth/BC scaling relations, plus the completeness correction. The central claim also rests on the self-similar wind scaling and Padova/Williams stellar-evolution benchmarks.

free parameters (7)
  • Radial gas-to-dust mass ratio ψ = 200 (r<2.5 kpc), 250 (2.5–3.5), 300 (3.5–4), 350 (4–4.5), 400 (>4.5 kpc)
    Assumed from M33 metallicity gradient (Gratier et al. 2017); Mdot ∝ ψ^{1/2}, so the radial variation changes Mdot by ~40% across the disc.
  • Inner dust temperature T_d = 900 K (sometimes changed)
    Fixed in dusty SED fits; directly affects derived optical depth and luminosity for the 294 modelled stars.
  • Stellar temperature T_* = 3000 K (sometimes changed)
    Fixed in dusty SED fits; affects the photospheric SED shape and hence the derived luminosity and optical depth.
  • Carbon-star classification mass range = 1.5–4 M_sun
    Stars with birth mass in this range are classed as carbon stars; misclassification changes Mdot by up to an order of magnitude.
  • Assumed LPV phase duration for carbon stars = 2 × 10^5 yr
    Used to calibrate the pseudo-evolutionary tracks; the derived timescales for other mass ranges scale with this assumed value.
  • τ–colour and BC–colour scaling coefficients (Tables 2 & 3) = Listed in Tables 2 and 3
    Fit to the 294 SED-modelled stars and then applied to all other stars; assumes the same relations hold across the whole disc.
  • Survey completeness correction = factor ~2–3.3 (1/0.3 to 1/0.5)
    Multiplied onto the variable-star mass return to obtain the headline 0.1 M_sun/yr; based on internal completeness estimates rather than external calibration.
axioms (8)
  • domain assumption Self-similar radiation-driven wind scaling: (τ L^{3/4})/(ψ^{1/2} Mdot) ≈ const (Ivezić & Elitzur 1999)
    Invoked in §3.2 to convert SED-fit optical depths and luminosities into mass-loss rates for all stars.
  • domain assumption The dusty radiative-transfer code correctly models dusty circumstellar envelopes
    Used in §3.1; the SED fits (by visual inspection) and all derived τ and L rely on this.
  • domain assumption All stars lie at the adopted distance modulus μ = 24.9 mag
    Adopted from Bonanos et al. (2006); luminosity and Mdot scale with distance squared and L^{3/4}.
  • domain assumption Padova evolutionary models (Marigo et al. 2008, 2017) give correct birth masses, luminosities and TP-AGB lifetimes
    Used to convert K-band brightness to birth mass and to classify carbon stars; also used to estimate pulsation durations.
  • domain assumption Initial–final mass relation of Williams et al. (2009) is correct
    Used in §5.1.2 as the benchmark for calibrating the mass-loss timescales.
  • domain assumption Constant SFR over the past 10 Gyr and a Salpeter IMF for the pseudo-evolutionary tracks
    Explicitly assumed in §5.1.2 to construct mass-loss vs. time tracks; the assumed carbon-star duration (2e5 yr) sets the absolute scale.
  • domain assumption Interstellar reddening to M33 stars is negligible compared to circumstellar reddening at near-IR wavelengths
    Justified in §2 (E(B−V)~0.10–0.16 mag); for non-variable stars, however, the authors note the application of scaling relations is an upper limit.
  • domain assumption The near-IR variability detection selects the stars undergoing the heaviest mass loss
    Used throughout; the final mass-return census uses variable stars, corrected for incompleteness.

pith-pipeline@v1.3.0-alltime-deepseek · 36263 in / 22948 out tokens · 211046 ms · 2026-08-01T11:06:50.069613+00:00 · methodology

0 comments
read the original abstract

We have conducted a near-infrared monitoring campaign at the UK InfraRed Telescope (UKIRT), of the Local Group spiral galaxy M33 (Triangulum). In this sixth paper of the series, we measure the dust and gas mass-loss rates by the pulsating Asymptotic Giant Branch (AGB) stars and red supergiants (RSGs) across the stellar disc of M33. We combined our time-averaged near-IR photometry with the multi-epoch mid-IR photometry obtained with the Spitzer Space Telescope, and employed a combination of spectral energy distribution modelling and scaling relations. We found that the mass-loss rate is approximately proportional to luminosity (birth mass), with additional weaker dependence on pulsation period and/or amplitude (reflecting stellar evolution). As a population, AGB stars contribute most to the mass return into the interstellar medium (ISM). Super-AGB stars also reach very high mass-loss rates, in excess of $10^{-4}$ M$_\odot$ yr$^{-1}$. The mass loss of RSGs appears to be subject to different modes, with rates well below, around, and well above the nuclear burning timescale. The timescale for the dominant mass loss phase is $\sim0.6$-$2\times10^5$ yr, shorter than the thermal-pulsing AGB or RSG phases. The rate at which stars return mass to the ISM, $\sim0.1$ M$_\odot$ yr$^{-1}$ is about four times lower than the star formation rate, which would deplete the current ISM mass within about a Gyr, thus requiring additional, external gas supplies to sustain the long-term future of star formation in M33.

Figures

Figures reproduced from arXiv: 2607.19989 by Atefeh Javadi, Elham Saremi, Jacco Th. van Loon, Mina Alizadeh, Seyed Azim Hashemi.

Figure 1
Figure 1. Figure 1: Mosaic image of M 33 with overlain in magenta those WFCAM variables that are also identified with Spitzer. (J − Ks) < 0.08 mag. Indeed, comparison between the ob￾served colours and theoretical isochrones (e.g., papers III–V) suggests this is the case for the bulk of the stars. At mid-IR wavelengths the attenuation diminishes further still (see van Loon et al. 2003). Compared to the effects of circumstellar… view at source ↗
Figure 2
Figure 2. Figure 2: shows the mid-IR colour–magnitude diagrams (CMDs) for the stars that are identified in both surveys, overlain with Padova isochrones (Marigo et al. 2017). There is a generally excellent agreement between data and models, except for very bright sources with Ks < 13 mag, and the brightest 8-µm sources. As discussed in paper IV, stars with Ks < 11 mag and (often) redder H–Ks than J–Ks are fore￾ground stars, a… view at source ↗
Figure 3
Figure 3. Figure 3: Example SEDs of presumed carbon stars. The hori￾zontal ”errorbars” on the data represent the width of the photo￾metric passbands. The best matching SEDs modelled with dusty are shown with solid lines. For comparison, the best matching fits using silicates are shown with dotted lines. 3.2 Scaling relations On the basis of self-similarity of radiatively-driven winds there are scaling relations (Ivezi´c & Eli… view at source ↗
Figure 5
Figure 5. Figure 5: Combinations of the optical depth (τ), luminosity (L), gas-to-dust mass ratio (ψ) and mass-loss rate (M˙ ); the horizontal lines show the average values (solid) and ± standard deviations (dotted). The open red squares show the results if the carbon stars are presumed to be oxygen-rich instead [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Relations between optical depth (τ) and near-IR colours (top: J–Ks; bottom: H–Ks) for carbon stars (red squares) and M-type stars (blue triangles). The open red squares show the results if we adopt oxygenous dust for the presumed carbon stars. The red and blue lines are the fitted relations for carbon and M-type stars, respectively [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
Figure 8
Figure 8. Figure 8: Mass-loss rate vs. luminosity, for spectroscopically con￾firmed carbon stars from Block et al. (2007; red squares), 24- µm sources from Montiel et al. (2015; magenta pentagons) lumi￾nous stars from Martin & Humphreys (2017; blue crosses), RSGs from Drout et al. (2012; blue triangles), η Carinæ analogues from Khan et al. (2013; red stars) and symbiotic binary stars from Miko lajewska et al. (2017; green poi… view at source ↗
Figure 7
Figure 7. Figure 7: Relations between bolometric correction to the Ks band (BCKs) and near-IR colours (top: J–Ks; bottom: H–Ks) for carbon stars (red squares) and M-type stars (blue triangles). The open red squares show the results if we adopt oxygenous dust for the presumed carbon stars. The red and blue lines are the fitted relations for carbon and M-type stars, respectively. parameters, and present an assessment of the mas… view at source ↗
Figure 9
Figure 9. Figure 9: Mass-loss rate vs. luminosity, modelled with dusty for low-mass AGB stars (green triangles), intermediate-mass carbon stars (red squares) and massive AGB stars and RSGs (blue tri￾angles). The open red squares show the results if the presumed carbon stars are presumed to be oxygen-rich instead. The lines have the same meaning as in figure 8. The magenta line traces the fit to the mass-loss rate vs. luminosi… view at source ↗
Figure 10
Figure 10. Figure 10: Left: mass-loss vs. luminosity for all stars including non-variable stars. The blue and green triangles represent massive luminous M-type stars and low-mass stars (at lower luminosities), respectively. The red squares represent presumed carbon stars. Large yellow symbols identify the stars modelled with dusty; other UKIRT (WFCAM) variables are identified by black squares. The bottom panel shows the result… view at source ↗
Figure 11
Figure 11. Figure 11: Mass-loss rate vs. Ks-band amplitude, for stars mod￾elled with dusty – the open red squares show the results if the carbon stars are presumed to be oxygen-rich [PITH_FULL_IMAGE:figures/full_fig_p011_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Mass-loss rate vs. J–Ks colour, for stars modelled with dusty – the open red squares show the results if the carbon stars are presumed to be oxygen-rich. pulsation periods give the shocked atmosphere more time to grow grains, onto which the radiation field can impart momentum to drive a wind – deemed important for carbon stars (e.g., Winters et al. 2000; Wachter et al. 2002; Nowotny et al. 2010, 2011) as … view at source ↗
Figure 14
Figure 14. Figure 14: Luminosity vs. pulsation period. The symbols are as in figure 13; the red and blue lines represent the Period– Luminosity relationships derived by Menzies et al. (2008) and Guandalini & Busso (2008), respectively. relation: low-mass stars probably experiencing the effects of the reduced mass as a result of their sustained heavy mass loss (Wood 2000), whilst the most massive AGB stars may be more luminous … view at source ↗
Figure 15
Figure 15. Figure 15: Mass-loss rate vs. luminosity (top left), age (bottom left) and birth mass (top right) for the WFCAM variable stars, and vs. birth mass for all stars including non-variable stars (bottom right). The top panels show the cumulative distribution (integrated from the right); the bottom panels show binned contributions. Distinction is made between likely carbon stars (red) and M-type stars (blue). for a temper… view at source ↗
Figure 16
Figure 16. Figure 16: Top: radial profile of mass-return-rate surface density across a large part of M 33 deprojected onto the galactic plane; bottom: the same, but for the central square kiloparsec of M 33. though we note that the Goldman et al. formula is valid only for oxygen-rich stars. However, while we do not find any de￾pendence of the ratio for either M-type or carbon stars on luminosity or pulsation period ( [PITH_FU… view at source ↗
Figure 18
Figure 18. Figure 18: Mass-loss rate vs. luminosity amplitude. The symbols have the same meaning as in figure 10. sor & Davies work and ours are not at all in contradiction with one another, but rather complement eachother. Correlations with pulsation amplitude can be rather muddled by the fact that the magnitude scale is a relative scale, and more luminous stars naturally tend to pulsate with smaller amplitudes expressed this… view at source ↗
Figure 20
Figure 20. Figure 20: Ratio of pulsation duration and age, as tabulated in the Marigo et al. (2008) isochrones (black) and Marigo et al. (2017) isochrones (magenta), vs. stellar birth mass [PITH_FULL_IMAGE:figures/full_fig_p017_20.png] view at source ↗
Figure 19
Figure 19. Figure 19: Top: ratio of the mass lost during the LPV phase, and the birth mass. The horizontal dotted lines indicate the situation where the star expels all of its mass. For comparison we also plot the same fractional mass loss derived from the initial–final mass relation determined by Williams et al. (2009). Bottom: ratio of our fractional mass loss, and that of Williams et al. Clearly, the stars appear to lose mo… view at source ↗
Figure 22
Figure 22. Figure 22: Pseudo-evolutionary tracks of the mass-loss rate and accumulated mass loss, for populations of low-mass AGB stars, intermediate-mass (carbon star) AGB stars, massive AGB stars and RSGs. This is based on a constant SFR over the past 10 Gyr, Salpeter IMF and monotonically increasing mass-loss rate. We confirm the three local enhancements surrounding the remarkably quiescent nucleus found already in Paper II… view at source ↗
Figure 23
Figure 23. Figure 23: Left: map of mass-return-rate surface density over the disc of M 33; Right: ratio of the cold dust mass and dust-mass return-rate surface density. more problematic that stars appeared to lose more mass than they were born with. The total size of the evolved star population also sets a firm limit on the possible in￾completeness level of our survey; we recall that we detected more than half of the heavily r… view at source ↗
Figure 24
Figure 24. Figure 24: Ratio of total mass returned by variable stars, and the recent SFR (black), and the mass fraction of carbonaceous dust returned to the ISM (red), both plotted vs. radial distance to the centre of M 33. local, recent star formation rate. This ratio approaches (but does not quite reach) unity around r = 3–4 kpc, whereas the mass return rate falls short of sustaining star formation by a few times in the outs… view at source ↗
Figure 25
Figure 25. Figure 25: Top: map of mass-return-rate surface density for low￾mass stars (1–1.5 M⊙); Bottom: for carbon stars (1.5–4 M⊙). dust – see for comparison the SMC (Boyer et al. 2012) and outer disc of the Milky Way (Ishihara et al. 2011). 6 SUMMARY OF CONCLUSIONS This paper is the culmination of our long-term monitoring programme of infrared variable star in M 33, presenting the measurement of the mass-loss rates and the… view at source ↗

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

130 extracted references · 3 linked inside Pith

  1. [1]

    Barsukova E.\ A., Govanskij V.\ P., Hornoch K., Fabrika S., Pietsch W., Sholukhova O., Valeev A.\ F., 2011, MNRAS, 413, 1797

    2011

  2. [2]

    Beasor E.\ R., Davies B., 2016, MNRAS, 463, 1269

    2016

  3. [3]

    Beasor E.\ R., Davies B., 2018, MNRAS, 475, 55

    2018

  4. [4]

    Benjamin R.\ A.\ et al., 2005, ApJ, 630, L149

    2005

  5. [5]

    Bladh S., H\"ofner S., Aringer B., Eriksson K., 2015, A&A, 575, A105

    2015

  6. [6]

    Block D.\ L.\ et al.\ , 2007, A&, 471, 467

    2007

  7. [7]

    Bolatto A.\ D., 2007, ApJ, 655, 212

    2007

  8. [8]

    Bonanos A.\ Z.\ et al., 2006, ApJ, 652, 313

    2006

  9. [9]

    Bonanos A.\ C.\ et al., 2010, AJ, 140, 416

    2010

  10. [10]

    Boothroyd A.\ I., Sackmann I.-J., 1992, ApJ, 393, L21

    1992

  11. [11]

    Boyer M.\ L.\ et al., 2012, ApJ, 748, 40

    2012

  12. [12]

    Buckalew B.\ A.\ et al., 2006, ApJS, 162, 329

    2006

  13. [13]

    Burggraf B.\ et al., 2015, A&A, 581, A12

    2015

  14. [14]

    Cioni M.-R.\ L.\ et al., 2008, A&A, 487, 131

    2008

  15. [15]

    Clark J.\ S.\ et al., 2012, A&A, 541, A146

    2012

  16. [16]

    Contreras Pe\ na C.\ et al., 2017, MNRAS, 465, 3011

    2017

  17. [17]

    Davidge T., 2000, AJ, 119, 748

    2000

  18. [18]

    Davidge T., 2018, ApJ, 856, 129

    2018

  19. [19]

    Deul E.\ R., van der Hulst J.\ M., 1987, A&AS, 67, 509

    1987

  20. [20]

    Draine B.\ T., Lee H.\ M., 1984, ApJ, 285, 89

    1984

  21. [21]

    Drout M., Massey P., Meynet G., 2012, ApJ, 750, 97

    2012

  22. [22]

    Dwek E., ApJ, 501, 643

  23. [23]

    Ekstr\"om S.\ et al., 2012, A&A, 537, A146

    2012

  24. [24]

    Freedman W.\ L., Wilson C.\ D., Madore B.\ F., 1991, ApJ, 372, 455

    1991

  25. [25]

    Georgy C., 2012, A&A, 538, L8

    2012

  26. [26]

    Georgy C., Ekstr\"om S., Meynet G., Massey P., Levesque E.\ M., Hirschi R., Eggenberger P., Maeder A., 2012, A&A, 542, A29

    2012

  27. [27]

    Goldman S.\ R.\ et al., 2017, MNRAS, 465, 403

    2017

  28. [28]

    Goldman S.\ R.\ et al., 2018, MNRAS, 473, 3835

    2018

  29. [29]

    Gonz\'alez-L\'opezlira R.\ A., 2018, 856, 170

    2018

  30. [30]

    Gratier P.\ et al., 2017, A&A, 600, A27

    2017

  31. [31]

    Greenhill L.\ J., Moran J.\ M., Reid M.\ J., Gwinn C.\ R., Menten K.\ M., Eckart A., Hirabayashi H., 1990, ApJ, 364, 513

    1990

  32. [32]

    Groenewegen M.\ A.\ T., Sloan G.\ C., Soszy\'nski I., Petersen E.\ A., 2009, A&A, 506, 1277

    2009

  33. [33]

    Groenewegen M.\ A.\ T., Sloan G.\ C., 2018, 609, A114

    2018

  34. [34]

    Gruendl R.\ A., Chu Y.-H., 2009, ApJS, 184, 172

    2009

  35. [35]

    Gruendl R.\ A., Chu Y.-H., Seale J.\ P., Matsuura M., Speck A.\ K., Sloan G.\ C., Looney L.\ W., 2008, ApJ, 688, L9

    2008

  36. [36]

    Guandalini R., Busso M., 2008, A&A, 488, 675

    2008

  37. [37]

    Gullieuszik M.\ et al., 2012, A&A, 537A, 105

    2012

  38. [38]

    Hanner M.\ S., 1988, NASA Conf.\ Pub.\ 3004, 22

    1988

  39. [39]

    Hodierna G.\ B., 1654, De Systemate Orbis Cometici, Deque Admirandis Coeli Characteribus (About the systematics of the cometary orbit, and about the admirable objects of the sky), Palermo

  40. [40]

    H\"ofner S., 2008, A&A, 491, L1

    2008

  41. [41]

    H\"ofner S., Olofsson H., 2018, A&ARv, 26, 1

    2018

  42. [42]

    Hubble E., Sandage A., 1953, ApJ, 118, 353

    1953

  43. [43]

    Humphreys R.\ M., Leitherer C., Stahl O., Wolf B., Zickgraf F.-J., 1988, A&A, 203, 306

    1988

  44. [44]

    Iben I.\ Jr., Renzini A., 1983, ARA&A, 21, 271

    1983

  45. [45]

    Ishihara D., Kaneda H., Onaka T., Ita Y., Matsuura M., Matsunaga N., 2011, A&A, 534, A79

    2011

  46. [46]

    Ivezi\'c Z , Elitzur M., 1995, ApJ, 445, 415

    1995

  47. [47]

    Ivezi\'c Z , Elitzur M., 1997, MNRAS, 287, 799

    1997

  48. [48]

    Ivezi\'c Z , Nenkova M., Elitzur M., 1999, dusty User Manual (University of Kentucky)

    1999

  49. [49]

    Javadi A., van Loon J.\ Th., Mirtorabi M.\ T., 2011a, MNRAS, 411, 263 (Paper I)

  50. [50]

    Javadi A., van Loon J.\ Th., Mirtorabi M.\ T., 2011b, MNRAS, 414, 3394 (Paper II)

  51. [51]

    Javadi A., van Loon J.\ Th., Mirtorabi M.\ T., 2011c, in: Why Galaxies Care About AGB Stars II, eds.\ F.\ Kerschbaum, T.\ Lebzelter & R.\ F.\ Wing, ASPC, 445, 497

  52. [52]

    Javadi A., van Loon J.\ Th., Khosroshahi H., Mirtorabi M.\ T., 2013, MNRAS, 432, 2824 (Paper III)

    2013

  53. [53]

    Javadi A., Saberi M., van Loon J.\ Th., Khosroshahi H., Golabatooni N., Mirtorabi M.\ T., 2015, MNRAS, 447, 3973 (Paper IV)

    2015

  54. [54]

    Javadi A., van Loon J.\ Th., Khosroshahi H., Tabatabaei F., Hamedani Golshan R., Rashidi M., 2017, MNRAS, 464, 2103 (Paper V)

    2017

  55. [55]

    Jones A.\ P.\ et al., 2014, Faraday Discussions, 168, 313

    2014

  56. [56]

    Jones O.\ C.\ et al., 2017, MNRAS, 470, 3250

    2017

  57. [57]

    Jura M., Kleinmann S.\ G., 1989, ApJ, 341, 359

    1989

  58. [58]

    Jura M., Kleinmann S.\ G., 1990, ApJS, 73, 769

    1990

  59. [59]

    Kang X., Chang R., Yin J., Hou J., Zhang F., Zhang Y., Han Z., 2012, MNRAS, 426, 1455

    2012

  60. [60]

    Khan R., Stanek K.\ Z., Kochanek C.\ S., Bonanos A.\ Z., 2011, ApJ, 732, 43

    2011

  61. [61]

    Khan R., Stanek K.\ Z., Kochanek C.\ S., 2013, ApJ, 767, 52

    2013

  62. [62]

    Khan R., Kochanek C.\ S., Stanek K.\ Z., Gerke J., 2015, ApJ, 799, 187

    2015

  63. [63]

    Koch E., Rosolowsky E., Johnson M.\ C., Kepley A.\ A., Leroy A., 2018, preprint (arXiv:1801.10568)

    Pith/arXiv arXiv 2018

  64. [64]

    Laki\'cevi\'c M.\ et al., 2015, ApJ, 799, 50

    2015

  65. [65]

    Marigo P., Girardi L., Bressan A., Groenewegen M.\ A.\ T., Silva L., Granato G.\ L., 2008, A&A, 482, 883

    2008

  66. [66]

    Marshall J.\ R., van Loon J.\ Th., Matsuura M., Wood P.\ R., Zijlstra A.\ A., Whitelock P.\ A., 2004, MNRAS, 355, 1348

    2004

  67. [67]

    Massey P., Armandroff T.\ E., Pyke R., Patel K., Wilson C.\ D., 1995, AJ, 110, 2715

    1995

  68. [68]

    Matsuura M., Woods P.\ M., Owen P.\ J., 2013, MNRAS, 429, 2527

    2013

  69. [69]

    Matsuura M.\ et al., 2009, MNRAS, 396, 918

    2009

  70. [70]

    Mattsson L., H\"ofner S., Herwig F., 2007, A&A, 470, 339

    2007

  71. [71]

    Mauron N., Josselin E., 2011, A&A, 526, A156

    2011

  72. [72]

    McDonald I., Zijlstra A.\ A., Boyer M.\ L., 2012, MNRAS, 427, 343

    2012

  73. [73]

    McDonald I., van Loon J.\ Th., Decin L., Boyer M.\ L., Dupree A.\ K., Evans A., Gehrz R.\ D., Woodward C.\ E., 2009, MNRAS, 394, 831

    2009

  74. [74]

    McQuinn K.\ B.\ W.\ et al., 2007, ApJ, 664, 850

    2007

  75. [75]

    Meixner M.\ et al., 2006, AJ, 132, 2268

    2006

  76. [76]

    Menzies J., Feast M, Whitelock P., Olivier E., Matsunaga N., da Costa G., 2008, MNRAS, 385, 1045

    2008

  77. [77]

    Messier C., 1771, Mem.\ Acad., p448

  78. [78]

    Miko ajewska J., Shara M.\ M., Caldwell N., I kiewicz K., Zurek D., 2017, 465, 1699

    2017

  79. [79]

    Montiel E.\ J., Srinivasan S., Clayton G.\ C., Engelbracht C.\ W., Johnson C.\ B., 2015, AJ, 149, 57

    2015

  80. [80]

    Nelson E.\ J.\ et al., 2018, preprint (arXiv:1801.02647)

    Pith/arXiv arXiv 2018

Showing first 80 references.