{"id":"64c4641d-57e0-46e8-bb3f-471a414a5f1b","arxiv_id":"2509.01988","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"PRIMA imaging and spectroscopy could characterize HST-dark galaxies in about 50 and 100 hours respectively, closing the JWST-ALMA gap and probing their dust, gas, and AGN content.","lead":"This paper argues that the planned far-infrared space telescope PRIMA is the essential next facility for studying HST-dark galaxies, a massive, dusty population invisible to Hubble but detected by ALMA. It estimates that PRIMA could observe 50 to 100 such galaxies in about 50 hours of imaging plus 100 hours of spectroscopy, filling the gap between JWST and ALMA.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"FIRESS line list includes [OIII] 88 μm at z=2.2, which lies outside the 24-235 μm band; line detectability also depends on unvalidated z~0 L_line-L_IR relations.","rationale":"The paper is a mission science case, so the central claim is feasibility of PRIMA observations. The reader already flagged the SED and line-ratio assumptions as the weakest point. My stress test finds a concrete internal inconsistency in the FIRESS line list that makes the feasibility estimate even less secure: the [O III] 88 μm line cannot be observed at z=2.2 because it falls outside FIRESS's spectral range. This is independent of any template uncertainty and suggests the line-detectability calculations were not fully checked against the instrument bandpass. The dependence on local line-luminosity relations is also real, as the paper's own comparison with Ref. 62 shows order-of-magnitude time variations. These issues do not invalidate the overall case for PRIMA — many other lines and diagnostics remain accessible — but they support a conditional verdict rather than full acceptance. The reader's verdict is already CONDITIONAL, and my concern reinforces it without changing the outcome.","tokens_in":14294,"tokens_out":7257,"duration_ms":78569,"concrete_test":"Run a bandpass check: compute λ_obs = λ_rest × (1+z) for the five claimed lines at z=2.2 and over the z=2-2.5 interval against FIRESS's 24-235 μm. Use the public PRIMA ETC to recompute 5σ exposure times for only the in-band lines using the Ref. 61 relations. If [O III] 88 is excluded and the total time still yields 20-25 galaxies in ~100 h, the corrected claim holds; if the time increases by >50%, the FIRESS component should be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 4.2 states that with ~2 h/source FIRESS will detect [Ne II] 12.8, [Si II] 34.5, [O I] 63, [O III] 88, and [O IV] 25.9 for a galaxy with L_IR = 10^12.8 Lsun at z=2.2. But FIRESS covers 24-235 μm (Sec. 2). At z=2.2, [O III] 88 μm is observed at 88×3.2 = 281.6 μm, outside the band. This is an internal inconsistency, not a modeling uncertainty. The same time estimates use local L_line-L_IR relations (Ref. 61) calibrated at z~0; applying them to z~2-2.5 HST-dark galaxies is unvalidated. The paper itself notes that using a different [Ne II] calibration changes the required time from ~2 h to ~10 min (Ref. 62), showing the sensitivity. Because the central claim includes a quantitative FIRESS detection capability in ~100 h, an impossible line in the list and strong scaling-relation sensitivity make that estimate unreliable as stated.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper argues that PRIMA/PRIMAger imaging (25–265 μm) and FIRESS spectroscopy (24–235 μm) are necessary to characterize HST-dark galaxies: ALMA-selected, optically invisible, massive dusty sources at z≳2 whose SEDs are currently constrained only by a few IRAC/ALMA photometric points. The authors present a science case based on the literature sample of roughly 90 ALMA-detected HST-dark galaxies, discuss their location relative to the main sequence and gas-depletion times, and provide quantitative forecasts: about 50 hours of PRIMAger pointed observations would detect 50–100 such galaxies at 5σ, improving L_IR uncertainties from ~0.5 dex to ~0.1–0.15 dex; about 100 hours of FIRESS low-resolution spectroscopy would detect fine-structure lines in 20–25 of the brightest, lower-redshift (z≈2–2.5) sources, enabling AGN/starburst diagnostics and ISM studies. The paper is explicitly an observing-strategy and feasibility study for PRIMA.","tokens_in":14651,"tokens_out":4478,"duration_ms":54566,"significance":"If the quantitative forecasts are reliable, the paper makes a strong and timely case for PRIMA's unique role in filling the wavelength gap between JWST and ALMA for the HST-dark population. The qualitative scientific argument is convincing: those galaxies are dusty, likely massive, and poorly constrained, and mid-/far-IR photometry plus line spectroscopy is the obvious route to their physical properties. The paper also usefully quantifies the expected gain in L_IR precision using published simulation work and identifies concrete AGN/SF line diagnostics. Its main weaknesses are that the most quantitative claims rest on an unpublished median SED, local scaling relations applied at high redshift without validation, and at least one internally inconsistent line detectability statement. These issues do not invalidate the qualitative science case but make the headline exposure-time estimates unreliable as stated.","major_comments":[{"comment":"The text states that for L_IR=10^12.8 Lsun at z=2.2, FIRESS with ~2 h/source will detect [Ne II] 12.8, [Si II] 34.5, [O I] 63, [O III] 88, and [O IV] 25.9. FIRESS covers 24–235 μm (Sec. 2). At z=2.2, [O III] 88 μm is observed at 88×(1+2.2)=281.6 μm, outside the FIRESS band. This is not a modeling uncertainty but an internal inconsistency directly affecting the claimed ~100 h / 20–25 galaxy spectroscopic capability. The estimate should be re-derived with an in-band line set, or the redshift/luminosity assumptions should be changed accordingly.","section":"§4.2, FIRESS line list"},{"comment":"The ~2 h/source FIRESS estimates are based on local (z≈0) L_line–L_IR relations from Ref. 61 applied to z≈2.2 galaxies. No validation of these relations at high redshift is provided, and the paper itself notes that adopting the Ref. 62 [Ne II] calibration changes the required exposure from ~2 h to ~10 min. A factor-of-12 sensitivity to the adopted scaling relation means that the central 'about 100 h for 20–25 galaxies' estimate is not robust. Please bracket the predictions with different calibrations and state which lines/quoted times would survive the full plausible range.","section":"§4.2, L_line–L_IR scaling relations"},{"comment":"The PRIMAger exposure-time estimates (10 h for faint/high-z, 1 h for the bulk, <0.1 h for bright sources) are obtained by rescaling factsheet sensitivities to a 'median SED' compiled in 'Gruppioni et al. in preparation.' Likewise, Fig. 2's density contours and Fig. 4's luminosity distribution are attributed to the same unpublished paper. The central quantitative claims of the manuscript are therefore not reproducible from the submitted text or publicly available data. Please include the median SED, the sample definitions, and the resulting SEDs/templates in an appendix or public repository, or provide an alternative public reference.","section":"§4.1 and Figs. 2, 4, 5"}],"minor_comments":[{"comment":"The PRIMA wavelength coverage is quoted variously as '25 to 265 μm', '24 to 265 μm', '25 to 250 μm', and the instrument section describes PHI1 as 24–45 μm, PHI2 as 45–84 μm, and PPI bands as 96–235 μm. Please harmonize these ranges and clarify whether there is a real gap between 84 and 96 μm and between 235 and 265 μm.","section":"§1/§2, wavelength ranges"},{"comment":"Typo: 'PRIMger' should be 'PRIMAger'.","section":"§4.1"},{"comment":"The sentence 'by considering for the latter the relation provided for a low AGN fraction, f_AGN < 40% the 5 to 40 μm range, valid also for no or negligible AGN' is grammatically unclear. Define f_AGN more precisely and state the wavelength range over which it is measured.","section":"§4.2"},{"comment":"Equation (1) appears in raw 'EQ-TARGET' format in the text; ensure it is rendered properly. Also define the units of ϵ explicitly when the numerical value 0.1 is introduced.","section":"Eq. (1)"},{"comment":"The legend of Fig. 6 is dense and the dashed/dotted line styles are hard to distinguish in grayscale. Consider labeling the curves directly or using a tabular summary of the L_IR precision values.","section":"Fig. 6"}],"recommendation":"major_revision","confidential_remarks":"This is an instrument-science paper from a team heavily involved in PRIMA, and several of the quantitative inputs (Refs. 59 and 61, and the 'in preparation' papers) come from the same group. That is not a disqualifier, but it raises the bar for independent verification of the exposure-time and precision claims. The internal inconsistency about [O III] 88 μm should be caught before publication; it is the kind of error that undermines confidence in the quantitative section even if the qualitative science case is sound."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper makes a clear, practical case for PRIMA to observe HST-dark galaxies, and it has real value as a mission-planning document. The quantitative estimates are new: roughly 50 hours of PRIMAger for 50–100 sources and 100 hours of FIRESS for 20–25 bright, low-z systems, plus an L_IR precision gain from 0.5 dex to 0.1–0.15 dex. These numbers are explicitly order-of-magnitude, and the authors are honest about that. The scientific motivation is well constructed, drawing together ALMA and JWST results and explaining why only far-IR can fill the gap. I'd want this on hand when planning PRIMA observations.\n\nThe soft spots are real but mostly not fatal. The stress-test note is correct: at z=2.2, [O III] 88 μm is observed at 281.6 μm, outside FIRESS's 24–235 μm range. That's an internal inconsistency in the line list, and it weakens the credibility of the FIRESS time estimates as stated. The authors likely meant a different line or a different redshift, but as published it's wrong.\n\nThe other concerns are weaker. The local L_line–L_IR relations from Ref. 61 (first author is the paper's first author) applied at z~2–2.5 are a legitimate worry, and the paper itself shows the sensitivity by noting an alternative [Ne II] calibration cuts the required time from 2 hours to 10 minutes. That range of uncertainty should be in the abstract, not buried in the text. The reliance on 'Gruppioni et al. in preparation' for the galaxy sample and median SED is also a real limitation—key figures can't be independently checked yet. And 'only facility' is overstated; JWST/MIRI and ALMA do cover parts of the relevant range, though PRIMA's full 25–265 μm coverage is unique.\n\nTaken as a whole, the central feasibility claim holds: PRIMA should detect and characterize these galaxies in reasonable exposure times. But the FIRESS line-specific numbers need correction and caveats before being used in proposals.\n\nThis paper is primarily for PRIMA-focused astronomers and those planning high-z obscured-galaxy surveys. It deserves a serious referee—the science is relevant and the estimates are mostly reproducible—but the referee should demand the line-list fix and independent validation of the scaling relations.","headline":"A useful PRIMA science case for HST-dark galaxies, but check the FIRESS line list: [O III] 88 μm at z=2.2 falls outside the instrument's 24–235 μm band.","tokens_in":697,"tokens_out":1598,"would_cite":true,"duration_ms":32898,"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":"This paper argues that PRIMA's 25-265 μm imaging and spectroscopy can reveal whether HST-dark galaxies are powered by stars or buried black holes and measure the physical state of their gas.","keywords":["HST-dark galaxies","dusty star-forming galaxies","far-infrared astronomy","mid-infrared spectroscopy","fine-structure lines","AGN diagnostics","interstellar medium","galaxy evolution"],"falsifier":"After PRIMA is in operation, a ~50 h PRIMAger campaign on roughly 100 ALMA-selected HST-dark galaxies would test the claim directly: if fewer than about 50 sources are detected at 5σ, or if the recovered L_IR scatter stays near 0.5 dex instead of shrinking to 0.1-0.15 dex, the predicted feasibility and precision gain fail.","tokens_in":14231,"feed_emoji":"🔭","tokens_out":8961,"duration_ms":91967,"temperature":0.7,"pith_summary":"Massive, dusty galaxies invisible in the deepest HST images—HST-dark galaxies—are turning up in ALMA surveys at redshifts 2.5 to 7 and may account for a large share of early star formation, but their spectral energy distributions between JWST and ALMA wavelengths remain essentially unmeasured. This paper argues that PRIMA is the facility that can close that gap: roughly 50 hours of PRIMAger imaging would detect 50-100 ALMA-selected HST-dark galaxies and shrink the uncertainty on their total infrared luminosities from about 0.5 dex to 0.1-0.15 dex. Another ~100 hours of FIRESS spectroscopy would detect key fine-structure lines in the 20-25 brightest systems at z=2-2.5, letting observers determine whether an AGN or star formation powers each galaxy and measure gas density, ionization, and temperature. If these observations deliver, HST-dark galaxies can finally be placed in the evolutionary sequence, testing whether they are the missing progenitors of today's massive ellipticals and why current simulations underpredict their numbers.","feed_headline":"150 hours of PRIMA could decode the universe's hidden galaxies","feed_subtitle":"Two instruments fill the JWST–ALMA gap, revealing dust, gas, and the engines of early massive galaxies.","key_machinery":"The load-bearing mechanism is wavelength coverage. PRIMAger's 12 photometric bands from 25 to 265 μm (PHI1, PHI2, and four polarimetric PPI bands) land exactly in the rest-frame mid/far-infrared region where HST-dark SEDs are currently unconstrained, where PAH features and hot AGN-heated dust emit and where the dust peak itself sits. FIRESS's low-resolution R=100 mode covers 24-235 μm simultaneously, catching the mid/far-infrared fine-structure lines—especially [Ne V] and [O IV] as AGN ionization diagnostics and [O III] 52/88 as a density diagnostic—whose ratios are dust-insensitive probes of the gas ionized by stars or an AGN. The feasibility calculation takes the ALMA-selected HST-dark sam","core_discovery":"The paper's central claim is that the PRIMA observatory can observationally open the otherwise inaccessible mid- and far-infrared window on HST-dark galaxies. With about 50 hours of pointed PRIMAger observations, 50-100 ALMA-selected HST-dark galaxies would be detected at 5σ across the full 25-265 μm range—roughly 1 hour per source for the bulk of the population at 3≲z<4 and L_IR≳10^12.5 L_sun, 10 hours for fainter higher-redshift sources, and less than 0.1 hours for the brightest low-redshift tail. Adding the PRIMAger bands to current IRAC+ALMA SEDs improves L_IR precision from a scatter of about 0.5 dex to 0.1-0.15 dex. Spectroscopically, about 100 hours with FIRESS in low-resolution mode","pith_inferences":["The same PRIMAger/FIRESS strategy could be applied to JWST-selected extremely red galaxies, extending the census of heavily obscured massive galaxies beyond the ALMA-selected population.","Combining PRIMAger 25-265 μm photometry with JWST rest-frame optical data may break the dust-temperature/redshift degeneracy, likely revising current photometric redshifts and stellar mass estimates for individual sources.","If [Ne V] is detected in a substantial fraction of the sample, it would imply a large population of Compton-thick AGN in the early universe, with consequences for black-hole seeding and for the AGN contribution to cosmic reionization.","FIRESS line ratios such as [O III] 52/88 could supply the first electron-density measurements in HST-dark galaxies, linking their short depletion times to the turbulent or clumpy state of their gas reservoirs."],"forward_implications":["A ~50 h PRIMAger program would replace single-point ALMA photometry with 12-band SEDs, giving the first constraints on dust temperature, dust mass, and photometric redshifts for HST-dark galaxies.","L_IR precision would improve from about 0.5 dex to 0.1-0.15 dex, turning current order-of-magnitude SFRs and gas depletion times into quantitative evolutionary constraints.","A ~100 h FIRESS program would detect mid/far-infrared fine-structure lines in 20-25 z=2-2.5 HST-dark galaxies, separating AGN- from star-formation-powered systems and measuring gas density, ionization, and chemical conditions.","If the AGN fraction exceeds 40%, [Ne V] and [O IV] become detectable in 0.4-2 h per source, providing direct black-hole accretion measurements for heavily obscured AGN that X-ray facilities would likely miss.","Constraining the abundance and physical properties of HST-dark galaxies would test galaxy formation simulations that currently underpredict the number density of massive dusty galaxies at z>2 by one to two orders of magnitude."],"supporting_citations":[{"why":"Supplies the H-dropout sample of optically invisible massive galaxies and the argument that they dominate the early stellar-mass density.","marker":"[2]"},{"why":"Provides GOODS-ALMA detections that define optically dark galaxies and the estimate that they make up to 20% of ALMA blind samples.","marker":"[17]"},{"why":"Supplies the ASAGAO near-IR-dark ALMA sample that contributes to the compiled HST-dark population and its median SED.","marker":"[19]"},{"why":"Supplies the ALPINE serendipitous HST-dark galaxies whose SEDs and IR luminosity distribution set the exposure-time brackets.","marker":"[20]"},{"why":"Adds the AS2UDS invisible submillimetre galaxies to the literature sample used for the physical-property analysis.","marker":"[21]"},{"why":"Gives the optically thin dust assumption that converts ALMA continuum into molecular gas masses and depletion times.","marker":"[45]"},{"why":"Provides the simulated PRIMAger SED fits that quantify the L_IR precision gain from about 0.5 to 0.1-0.15 dex.","marker":"[59]"},{"why":"Supplies the local L_line-L_IR calibration used to predict FIRESS line fluxes and the limiting SFR and black-hole accretion rate.","marker":"[61]"}],"fun_headline_variants":["PRIMA's 150 hours unveil the hidden dusty giants","HST-dark galaxies illuminated by PRIMA's dual instruments","Mapping the dusty past: PRIMA closes the far-IR gap","PRIMA sees through dust to reveal early massive galaxies"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The exposure-time and detectability estimates rest on the assumed median SED of HST-dark galaxies and on local L_line-L_IR scaling relations; if these galaxies are colder, fainter, or have weaker line-to-IR ratios at z≈2-3, the required observing times would be substantially longer.","fun_headline_variants_meta":{"raw":{"variants":["PRIMA's 150 hours unveil the hidden dusty giants","HST-dark galaxies illuminated by PRIMA's dual instruments","Mapping the dusty past: PRIMA closes the far-IR gap","PRIMA sees through dust to reveal early massive galaxies"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000678,"raw_usage":{"total_tokens":2967,"prompt_tokens":842,"completion_tokens":2125,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":586,"completion_tokens_details":{"reasoning_tokens":2057}},"tokens_in":586,"tokens_out":2125,"duration_ms":14529,"temperature":1.0,"reasoning_tokens":2057,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T11:58:47.030142+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"After PRIMA is in operation, a ~50 h PRIMAger campaign on roughly 100 ALMA-selected HST-dark galaxies would test the claim directly: if fewer than about 50 sources are detected at 5σ, or if the recovered L_IR scatter stays near 0.5 dex instead of shrinking to 0.1-0.15 dex, the predicted feasibility and precision gain fail.","supporting_citations":[{"cited_title":"A dominant population of optically invisible massive galaxies in the early Universe,","cited_arxiv_id":null,"evidence_quote":"Supplies the H-dropout sample of optically invisible massive galaxies and the argument that they dominate the early stellar-mass density."},{"cited_title":"GOODS-ALMA: 1.1 mm galaxy survey. I. Source catalog and optically dark galaxies,","cited_arxiv_id":null,"evidence_quote":"Provides GOODS-ALMA detections that define optically dark galaxies and the estimate that they make up to 20% of ALMA blind samples."},{"cited_title":"ALMA 26 arcmin² survey of GOODS-S at 1 mm (ASAGAO): near-infrared-dark faint ALMA sources,","cited_arxiv_id":null,"evidence_quote":"Supplies the ASAGAO near-IR-dark ALMA sample that contributes to the compiled HST-dark population and its median SED."},{"cited_title":"The ALPINE-ALMA [CII] survey: the nature, luminosity function, and star formation history of dusty galaxies up to z ∼ 6,","cited_arxiv_id":null,"evidence_quote":"Supplies the ALPINE serendipitous HST-dark galaxies whose SEDs and IR luminosity distribution set the exposure-time brackets."},{"cited_title":"An ALMA survey of the S2CLS UDS field: optically invisible submillimetre galaxies,","cited_arxiv_id":null,"evidence_quote":"Adds the AS2UDS invisible submillimetre galaxies to the literature sample used for the physical-property analysis."},{"cited_title":"ISM masses and the star formation law at z = 1 to 6: ALMA observations of dust con- tinuum in 145 galaxies in the COSMOS survey field,","cited_arxiv_id":null,"evidence_quote":"Gives the optically thin dust assumption that converts ALMA continuum into molecular gas masses and depletion times."},{"cited_title":"Disentangling the co-evolution of galaxies and supermassive black holes with PRIMA,","cited_arxiv_id":null,"evidence_quote":"Provides the simulated PRIMAger SED fits that quantify the L_IR precision gain from about 0.5 to 0.1-0.15 dex."},{"cited_title":"Tracing black hole accretion with SED decomposition and IR lines: from local galaxies to the high-z Universe,","cited_arxiv_id":null,"evidence_quote":"Supplies the local L_line-L_IR calibration used to predict FIRESS line fluxes and the limiting SFR and black-hole accretion rate."}],"review_version":1}