{"id":"ca6afb74-5d6d-41ce-9f7a-6e84fe6ca9a8","arxiv_id":"2507.18728","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Many protostars with dynamical mass estimates have luminosities close to their photospheric emission, implying low current accretion rates for most, with a minority of luminous rapid accretors.","lead":"This paper compares measured masses and brightnesses of protostars to an empirical birthline curve, and finds many are too dim to be accreting much mass right now. It suggests star formation may occur in bursts for some objects, but the sample is small and biased.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The low-accretion conclusion depends on a by-eye birthline drawn above the ONC; if that zero-accretion L*(M) is 0.2-0.3 dex too high, many sources shift from photosphere-dominated to accretion-dominated, and the 'tracking' is never quantified.","rationale":"Reader's weakest_assumption is the same as the load-bearing step I identify: the unquantified, by-eye placement of the empirical birthline. I agree. The paper is explicitly preliminary and hedges in the Discussion, so the CONDITIONAL verdict is appropriate. My concrete check would quantify whether the 'many track' claim survives a less aggressive zero-accretion fiducial. If it does, the result is strengthened; if not, the central claim is an artifact of the chosen reference curve. The strongest_claim's key clause 'accretion luminosities may be at most comparable' is exactly the clause most sensitive to this choice, so the concern is central rather than peripheral. I do not see an internal inconsistency or a more severe flaw requiring REJECT; the data compilation and the explicit acknowledgment of biases are real assets. The main risk is that the conclusion is underdetermined by the current sample and the by-eye reference, which is precisely a CONDITIONAL rather than ACCEPT situation.","tokens_in":9753,"tokens_out":11241,"duration_ms":129530,"concrete_test":"Recompute the comparison using a reproducible birthline instead of the by-eye curve: bin the Da Rio et al. (2010) ONC sample in log Teff and take the 10th, 50th, and 90th percentiles of log L; convert each percentile envelope to L*(M) and R*(M) with the same Siess (2000) tracks. Then for every Table 1 source compute Δ = log L_bol − log L*(M) under (i) the paper's curve, (ii) the 90th-percentile envelope, and (iii) a curve shifted down by 0.2 dex (the approximate SPOTS/age difference). Count the fraction of sources with Δ < 0.2 (photosphere-dominated). If this fraction remains above half under (ii) and (iii), the low-accretion claim survives; if it falls below half, the by-eye placement is the proximate cause of the conclusion.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central inference of Sections 3 and 4—that many Table 1 protostars have L_bol close to the zero-accretion birthline and therefore L_acc ≲ L*—rests on the Section 2 construction of that birthline. It is 'drawn by eye to lie above almost all' of the Da Rio et al. (2010) ONC stars (Figure 1, footnote 1) and then mapped through Siess (2000) tracks into L*(M) and R*(M) (Figures 2-3). Three properties make this load-bearing. First, the placement is an upper envelope, not a central estimate; a zero-accretion fiducial chosen this high minimizes the inferred L_acc = L_bol − L* by construction. Second, 'track reasonably closely' is never made quantitative: no tolerance is defined, no residual statistic is computed, and points below the line are primarily attributed to anisotropic envelope radiation, so the comparison cannot be falsified. Third, plausible alternative calibrations shift the fiducial by ~0.2 dex: the SPOTS tracks used by the authors themselves differ by up to this amount, and the adopted 0.3-0.4 Myr isochrone is younger than the likely age of many embedded Class I sources. If the true L*(M) is 0.2-0.3 dex lower, objects sitting on the adopted birthline have L_acc comparable to or larger than L*, contradicting the abstract's 'at most comparable' claim. The paper's Section 4 hedge that Mdot ~1e-7 Msun/yr even with overestimated L* softens the evolutionary-rate conclusion, but it does not rescue the headline statement.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript assembles a sample of roughly thirty protostars with dynamical mass measurements and compares their bolometric luminosities with an empirical 'birthline' relation L*(M) and R*(M), constructed by placing a hand-drawn curve above the Orion Nebula Cluster stars in the HR diagram and calibrating it with Siess (2000) and SPOTS (Somers et al. 2020) evolutionary tracks. The authors find that most sources lie near this birthline and conclude that their accretion luminosities are at most comparable to photospheric luminosities, implying mass accretion rates far too low to build the final masses in typical protostellar lifetimes, with a minority of high-luminosity objects consistent with significant accretion. They interpret the luminosity distribution as a tentative hint of bimodality, while emphasizing the small sample and selection biases.","tokens_in":10101,"tokens_out":7757,"duration_ms":80772,"significance":"If the central inference is correct, it would challenge the view that typical protostars gain most of their mass through steady disk accretion over roughly 0.5 Myr, and it would add support to models in which episodic accretion dominates mass buildup. The paper's strengths are the compilation of a well-referenced observational sample in Table 1, the explicit comparison of Siess and SPOTS track calibrations, and an unusually candid acknowledgment of the method's limitations. The main weakness is that the zero-accretion comparison line is not derived quantitatively, so the conclusion is currently a suggestion rather than a robust measurement. The paper also offers a clear falsifiable prediction—luminosity distributions and accretion rates should cluster near the birthline for a well-defined sample—which future ALMA samples can test.","major_comments":[{"comment":"The empirical birthline is drawn by eye to lie above almost all ONC stars, making it an upper envelope of the HR diagram rather than a central estimate of the zero-accretion luminosity. Because L_acc in Eq. (2) is defined as L_bol - L*, this choice systematically minimizes the inferred accretion contribution. The manuscript explicitly notes in footnote 1 that no quantitative placement method was used, yet the central claim of the abstract depends on this placement. I request a quantitative sensitivity analysis: for example, recompute L_acc for a birthline shifted down by 0.2-0.3 dex (the difference between the Siess and SPOTS tracks in Fig. 2) and report the fraction of the sample for which L_acc exceeds L*.","section":"Section 2, Figure 1 and footnote 1"},{"comment":"The alternative SPOTS calibration differs from the fiducial by up to roughly 0.2 dex in luminosity, and the adopted 0.3-0.4 Myr isochrone is at the young end of plausible protostellar ages; older ages give lower L*(M). A 0.2-0.3 dex downward shift is therefore not a small perturbation for this analysis: it moves a substantial number of Table 1 sources from below or on the birthline to clearly above it, converting 'at most comparable' into 'accretion-dominated'. The statement that these differences are 'not significant for our purposes' should be backed by an explicit count or residual statistic.","section":"Section 2, Figures 2-3"},{"comment":"The comparison is not falsifiable as presented. Table 1 lists only a single bolometric luminosity per source with no published uncertainty, and many entries have no mass error bars. Points lying below the birthline are attributed to anisotropic envelope radiation and outflow cavities, with vertical bars applied for only three systems, without a systematic model of the inclination distribution. I request that the authors report residuals in log L at fixed M for the sample with available uncertainties, and that they either apply inclination corrections consistently to all sources or show that the qualitative conclusion is unchanged when all below-line points are treated as upper limits on L_bol.","section":"Section 3, Table 1 and Figure 4"},{"comment":"The hedge 'even if our birthlines somewhat overestimate photospheric contributions, resulting in higher accretion luminosities' is internally inconsistent: an overestimated L* lowers the inferred L_acc = L_bol - L*, so it would actually suppress the inferred accretion rates, not raise them. The robustness argument should be made with the correct sign, since this is the only quantitative caveat offered against the birthline-placement concern.","section":"Section 4"}],"minor_comments":[{"comment":"Section 5 repeats verbatim the 'Other differences from Tobin and Sheehan' paragraph that already appears in the Table 1 footnote; this duplication should be removed or the section merged with Section 3.","section":"Section 5"},{"comment":"The reference list entry for Reynolds et al. (2021) is garbled as '1k' instead of '17', and the T_bol column header is inconsistently typeset in the preprint.","section":"Table 1"},{"comment":"The term 'empirical birthline' may overstate the empirical content, because the curve is calibrated in mass and radius using Siess (2000) theoretical tracks; I suggest 'empirically anchored birthline' or a similar clarifying phrase.","section":"Section 2"},{"comment":"Because the core result is a comparison against birthline curves, I recommend providing the birthline tabulations L*(M) and R*(M) as machine-readable supplementary material rather than 'upon reasonable request'.","section":"Data availability"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Lee, here is my read of Hartmann et al. on the protostellar mass-luminosity relation.\n\nThe genuinely new thing here is the empirical birthline: instead of relying on theoretical birthlines from spherical collapse, they take the upper envelope of the Orion Nebula Cluster HR diagram, map it through Siess tracks into L*(M) and R*(M), and then compare that zero-accretion relation against every protostar with a dynamical mass they could find. The data compilation in Table 1, drawn mostly from Tobin & Sheehan 2024 plus a few updates, is useful in itself. The paper is refreshingly cautious: they flag selection biases, unresolved binaries, anisotropic radiation escape, and they explicitly say the bimodality is a hint, not a result.\n\nThe weak point is exactly where the stress-test puts it. The birthline is 'drawn by eye to lie above almost all' of the ONC stars. That makes it an upper envelope to the zero-accretion luminosity, not a central estimate. And 'track reasonably closely' is never quantified: no tolerance, no residual statistic, no error bars on the birthline. If the true L*(M) is 0.2 dex lower, which is within the quoted SPOTS vs. Siess difference, many of the points that look photosphere-dominated would actually have accretion luminosities comparable to or larger than L*. So the abstract's phrase 'at most comparable to their photospheric radiation' is not supported by a quantitative test.\n\nThat said, the paper's bigger conclusion—that many of these objects have mass accretion rates well below what is needed to build their masses in 0.5 Myr—is more robust. Even if you drop the birthline by 0.2 dex, the implied rates go up by roughly a factor of two, but they stay near 1e-7 Msun/yr, still far below M/0.5Myr ~ 2e-6 Msun/yr. The authors themselves make this hedge in Section 4. So the evolutionary point survives the calibration uncertainty; the fine-grained claim about photospheric vs. accretion dominance does not.\n\nThe remaining weaknesses are standard for this kind of compilation: small N, inhomogeneous selection, missing error bars on many entries, and only a handful of the edge-on systems get corrected luminosities. None of these are fatal given the preliminary framing. I would not cite this for a firm rate distribution, but I would cite it for the empirical birthline construction and for posing the bimodality question cleanly.\n\nWho should read it: observers and theorists working on protostellar accretion and star formation lifetimes. It deserves a serious referee—the question matters, and the paper is honest enough that a good referee can push for a quantitative birthline and a residual analysis. If I were the editor, I would send it out rather than desk reject.","headline":"A short, honest observational paper that builds a new empirical birthline from Orion and compiles existing dynamical masses; the central claim of low accretion rates is plausible but rests on a hand-drawn curve, so the 'at most comparable' phrasing is stronger than the evidence.","tokens_in":10653,"tokens_out":3431,"would_cite":true,"duration_ms":34231,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Many protostars with disk-derived masses sit near the no-accretion birthline, implying low mass-growth rates.","keywords":["protostars","mass-luminosity relation","birthline","accretion luminosity","dynamical masses","protostellar disks","star formation","mass accretion rates"],"falsifier":"A decisive check would be direct measurements of accretion luminosity, for instance spectral veiling or hydrogen recombination-line excesses, for the same protostars that have dynamical masses; if objects sitting on the birthline show accretion luminosities comparable to their bolometric luminosities, the low-accretion-rate conclusion would be contradicted.","tokens_in":2029,"feed_emoji":"⭐","tokens_out":2405,"duration_ms":121339,"temperature":0.7,"pith_summary":"This paper asks how much of a protostar's luminosity is starlight and how much is the glow of infalling gas, using objects whose masses were measured from the rotation of their surrounding disks. It constructs an empirical \"birthline,\" an estimate of the luminosity and radius a protostar would have if it were not accreting, by drawing a curve above almost all Orion Nebula Cluster stars and calibrating it with pre-main-sequence evolutionary tracks. Comparing observed protostars to this birthline, the paper finds that many lie close to it, so their accretion luminosities are at most comparable to their photospheric radiation. Through the standard relation between accretion luminosity, mass, radius, and accretion rate, that places typical accretion rates below what would be needed to build up the final stellar mass in a 0.5 Myr protostellar lifetime. A minority of objects are much brighter and consistent with rapid, evolutionarily important accretion, hinting at a bimodal distribution that the authors caution is not yet definitive.","feed_headline":"Protostars shine mostly by starlight, not infall","feed_subtitle":"Most studied protostars hug the zero-accretion birthline, so their accretion rates are too low to build mass.","key_machinery":"The central object is the empirical birthline: an assumed $L_*(M_*)$ and $R_*(M_*)$ relation for a protostar radiating only from its photosphere. It is constructed by drawing a curve by eye through the upper envelope of the Orion Nebula Cluster in the Hertzsprung–Russell diagram and calibrating it in mass with the Siess et al. (2000) pre-main-sequence tracks, with an alternative version built from starspot-included SPOTS models. The argument is carried by the accretion-luminosity equation $L_{\\rm bol} = L_* + 0.8 G M_* \\dot{M}/R_*$: with observed $L_{\\rm bol}$ and birthline $L_*$ and $R_*$ fixed, each object's vertical offset from the birthline is converted directly into an allowed accretion rate, so the clustering near the birthline forces the low inferred rates.","core_discovery":"The paper's discovery claim is that most protostars with dynamical mass estimates cluster near an empirical birthline, a hand-drawn locus in the luminosity–temperature diagram placed just above nearly all Orion Nebula Cluster stars and calibrated in mass with the Siess et al. (2000) tracks, alongside an alternative starspot-inclusive locus. Because the birthline represents the expected photospheric luminosity $L_*$ and radius $R_*$ at zero accretion, objects near it must have accretion luminosity $L_{\\rm acc} \\lesssim L_*$. Using $L_{\\rm bol} = L_* + 0.8 G M_* \\dot{M}/R_*$, this translates to accretion rates around $10^{-7}\\,M_\\odot\\,{\\rm yr}^{-1}$ or lower for most sources, rates too small to build up the final stellar mass within typical estimated protostellar lifetimes. A small subset lies roughly an order of magnitude above the birthline and is interpreted as undergoing evolutionarily significant mass accretion. The resulting bimodal appearance is presented as a hint, explicitly limited by small-number statistics and probable selection biases.","pith_inferences":["A quantitative extension the paper leaves implicit is a duty-cycle estimate: if low-luminosity protostars represent a long quiescent phase and high-luminosity sources represent short bursts that build most of the mass, the observed fraction of bright objects constrains the burst duty cycle to a few percent.","Future unbiased samples with many more dynamical masses could test the bimodality directly by looking for a two-peaked luminosity distribution at fixed mass, a signature that should be visible once selection effects are controlled.","If most embedded protostars truly accrete at roughly $10^{-7}\\,M_\\odot\\,{\\rm yr}^{-1}$, the total accretion luminosity in a forming cluster would be much smaller than commonly assumed, which would shift estimates of feedback and outflow energetics downward."],"forward_implications":["If most protostars with measured masses sit near the birthline, their current accretion rates are roughly $10^{-7}\\,M_\\odot\\,{\\rm yr}^{-1}$ or less, too small to increase the stellar mass significantly over a typical 0.5 Myr protostellar lifetime.","The small subset with luminosities an order of magnitude above the birthline must be accreting at rates near $M_*/0.1$ Myr, making those objects the ones capable of adding most of the final stellar mass.","Since substantial accretion luminosity would push an object well above the photospheric locus, the tight clustering near the birthline sets an upper limit to accretion luminosity for most of the sample.","The mass-measured sample is not obviously biased toward older evolutionary stages, since its bolometric-temperature distribution resembles the broader protostar population, so the low-luminosity objects are not simply aged protostars that have finished accreting.","If the apparent bimodality between low- and high-luminosity protostars is real, it would favor episodic or burst-dominated mass accretion, although the authors caution that the current sample is too small and biased to decide."],"supporting_citations":[{"why":"Supplies the pre-main-sequence evolutionary tracks used to calibrate the birthline's luminosity and radius as functions of mass.","marker":"Siess et al. (2000)"},{"why":"Provides the Orion Nebula Cluster luminosity–temperature diagram over which the empirical birthline is drawn.","marker":"Da Rio et al. (2010)"},{"why":"Gives the 0.8 coefficient in the accretion-luminosity equation that converts offsets from the birthline into accretion rates.","marker":"Gullbring et al. (1998)"},{"why":"Compiles the protostellar masses and bolometric luminosities that form the observed sample.","marker":"Tobin & Sheehan (2024)"},{"why":"Provides the SPOTS starspot evolutionary models used to construct the alternative birthline.","marker":"Somers et al. (2020)"},{"why":"Shows the spot-coverage models used to set the alternative SPOTS birthline.","marker":"Cao et al. (2022)"},{"why":"Supplies the typical protostellar lifetime estimate used to judge whether inferred accretion rates are high enough to build the stellar mass.","marker":"Evans et al. (2009)"},{"why":"Provides bolometric luminosities and SED modeling, including corrected luminosities for edge-on systems such as IRAS 04302 and L1489.","marker":"Furlan et al. (2008)"}],"fun_headline_variants":["Many protostars accrete too slowly to grow","Protostar sample reveals bimodal accretion hint","Low accretion prevails among protostars with measured masses","Few protostars accrete at rates that build mass","Protostellar accretion rates mostly below mass-building level"],"cache_read_input_tokens":12672,"weakest_assumption_plain":"The load-bearing premise is that the hand-drawn birthline, placed above nearly all Orion Nebula Cluster stars, faithfully represents the zero-accretion luminosity and radius of protostars as a function of mass; if that curve sits too high, the inferred accretion luminosities and rates are systematically too low, and the central conclusion would shift.","fun_headline_variants_meta":{"raw":{"variants":["Many protostars accrete too slowly to grow","Protostar sample reveals bimodal accretion hint","Low accretion prevails among protostars with measured masses","Few protostars accrete at rates that build mass","Protostellar accretion rates mostly below mass-building level"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000698,"raw_usage":{"total_tokens":3154,"prompt_tokens":947,"completion_tokens":2207,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":563,"completion_tokens_details":{"reasoning_tokens":2129}},"tokens_in":563,"tokens_out":2207,"duration_ms":16520,"temperature":1.0,"reasoning_tokens":2129,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T18:08:31.238951+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive check would be direct measurements of accretion luminosity, for instance spectral veiling or hydrogen recombination-line excesses, for the same protostars that have dynamical masses; if objects sitting on the birthline show accretion luminosities comparable to their bolometric luminosities, the low-accretion-rate conclusion would be contradicted.","supporting_citations":[],"review_version":2}