{"id":"cb34cefb-c19c-4fd3-8e8a-e6b223b7bf61","arxiv_id":"1908.08579","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The authors release a public MySQL database of thousands of MAPPINGS V radiative shock models, adding low-metallicity and age-truncated grids to the existing 3MdB photoionization resource.","lead":"This paper presents 3MdBs, an online SQL database of radiative shock models computed with the MAPPINGS V code, including standard, low-metallicity, and young shock grids. It is a resource paper meant to give the astrophysics community easy, queryable access to thousands of precomputed shock model line intensities.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The database's low-velocity shock grid rests on an unvalidated time-dependent preionization solver; Section 4.2 shows ~0.3 dex unexplained differences from Allen08, so the stored 100-200 km/s intensities are the least secure part of the central claim.","rationale":"The paper is a useful and clearly written resource: it documents the database schema, provides SQL and Python access examples, shows comparison plots, and honestly discloses that dust is neglected and that MAPPINGS V results differ from MAPPINGS III. The existence of the database and the described grids is plausible, and the authors do not hide the differences between their grid and Allen et al. (2008). However, the scientific value of the stored line intensities depends centrally on the new time-dependent preionization treatment introduced in Section 2.1. This treatment is claimed to be essential below about 200 km/s, yet the paper gives no independent check of it and even states that determining the cause of the observed differences with Allen08 is beyond the authors' expertise. That leaves the low-velocity part of the grids, and all precursor-inclusive spectra, as the least secure element of the central claim. The reader's weakest_assumption identified exactly this point, and I agree. A conditional acceptance remains appropriate: the database can be accepted as a resource, but the paper should either provide the proposed consistency check or retain an explicit, prominent caveat that the 100-200 km/s models and precursor-inclusive spectra have not been validated against observations or an independent time-dependent calculation.","tokens_in":14811,"tokens_out":9207,"duration_ms":101656,"concrete_test":"Retrieve the stored preshock temperature and H I/He I ionization fractions for the 250, 300, 500, and 1000 km/s, n0 = 1 cm^-3, B = 1 µG solar-abundance models from the Allen08 replica grid. For the same parameters, run convergent iterative-equilibrium preionization models using the radiation field from the shocked gas, as in Allen et al. (2008), and compare the two sets. Agreement to within a few percent at these velocities, where the equilibrium assumption is valid, would support the new solver; significant disagreement would show that the Section 4.2 line-ratio differences are not explained by atomic data alone and that the low-velocity 100-200 km/s grid, which depends on the new solver, is not credible.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2.1 states that the Allen et al. (2008) iterative preionization method fails below about 200 km/s and that MAPPINGS V 'for the first time' treats preshock ionization and thermal structure in a fully time-dependent, self-consistent manner. Every model in the database from 100 to 200 km/s, plus its precursor emission, depends on this new solver. The paper provides no convergence test, no comparison to an independent time-dependent photoionization calculation, and no observational validation of this regime. Section 4.2 reports differences of up to ~0.3 dex in [O III]/Hβ and other ratios relative to the Allen08 grid and explicitly declines to determine whether the differences come from code improvements, atomic data updates, or the new preionization treatment. Because the database is presented as a replacement for Allen08, users cannot tell whether the low-velocity line ratios stored in it are more correct or merely different. This is a correctness risk, not a style issue: the same solver is used for the precursor emission in all shock+precursor sums above 100 km/s, so any flaw propagates through the combined spectra as well.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents 3MdBs, a MySQL-accessible database of radiative shock models computed with MAPPINGS V. Three grids are described: a re-computation of the Allen et al. (2008) shock grids with MAPPINGS V, a low-metallicity extension using Gutkin et al. (2016) abundances, and grids of age-truncated (young) shocks. The database stores line intensities from 660 Å to 609 µm for the shock, precursor, and combined regions, together with ionic fractions, temperatures, column densities, and other plasma properties. The paper includes BPT-type comparisons with Allen et al. (2008), documents differences of up to ~0.3 dex in some line ratios, and gives SQL and Python examples for querying the database.","tokens_in":15007,"tokens_out":8622,"duration_ms":82932,"significance":"If the stored models are correct, this is a valuable community resource for interpreting shock excitation in H II regions, LINERs, and high-redshift galaxies. The paper has clear strengths: the grids are precisely parameterized, the absence of dust is explicitly disclosed, the comparison to the previous Allen et al. (2008) grids is shown, and the SQL/MySQL distribution with a web interface is a practical step beyond scattered flat files. The forward-model nature of the work makes circularity a non-issue. However, the scientific value depends critically on the new time-dependent preionization treatment introduced in Section 2.1, and that treatment is not validated against independent calculations or observations; the 100-200 km/s regime and all precursor spectra above 100 km/s therefore carry the main correctness risk. The low-metallicity and truncated grids are likewise introduced without external benchmarks.","major_comments":[{"comment":"The new time-dependent preionization treatment is the physical basis for every model at 100-200 km/s and for every precursor spectrum above 100 km/s, yet the paper provides no convergence test, no comparison with an independent time-dependent photoionization calculation, and no observational validation in this regime. Since the paper states that the earlier iterative method used by Allen et al. (2008) fails below about 200 km/s, there is no external benchmark for these models. The database documentation should carry an explicit caveat, and the authors should add at least one convergence test and one independent comparison for the low-velocity and precursor models.","section":"Section 2.1 (and Section 2.2)"},{"comment":"The paper reports differences of up to ~0.3 dex in [O III]/Hβ and other line ratios relative to Allen et al. (2008) and then states that it is beyond its scope to determine whether these arise from code improvements, atomic data updates, or the new preionization treatment. Because the grid is presented as an exact replica and potential replacement of the Allen et al. (2008) grid, users cannot tell whether the stored intensities are more correct or merely different. This is a load-bearing gap: without identifying the origin of the offset, the claim that the database supersedes Allen et al. (2008) is not supported. Please analyze the origin of the main differences or explicitly present the grid as a candidate update pending validation.","section":"Section 4.2, Figures 3-5"},{"comment":"The grid definition is internally inconsistent. The text says '36 individual shock velocities (from 100 up to 1000 km s−1 in steps of 25 km s−1)', but this range contains 37 values; the stated total of 1440 models in the first sub-grid implies 5 abundances × 8 magnetic fields × 36 velocities, not 37. The authors should correct the count, state the exact velocity list, give total model counts for all grids, and provide a checksum or export so that the 'exact replica' claim is independently verifiable.","section":"Section 4.1, Tables 2-3"},{"comment":"The truncated/young shock grids are a central new contribution, but the paper does not state precisely how each model was terminated beyond the field 'cut off temp', and it offers no comparison with the published young-shock models cited in the text (Raymond et al. 1980; Contini & Shaviv 1982; Binette et al. 1985). The 'cut' distinction is therefore not well enough defined for users to interpret the stored spectra. Please specify the stopping criterion (temperature threshold, age, or column density) and provide at least one comparison to existing young-shock calculations.","section":"Section 4.4"}],"minor_comments":[{"comment":"The abstract and body contain several typos and unclear phrases: 'in other to explore', 'publicized', 'visualized', 'the data in distributed', and 'book case' should be corrected.","section":"Abstract and general text"},{"comment":"The caption states 'GRID SAMPLE OF THE LOW METALLICITY GRID DESCRIBED IN SECTION 3.2', but the low-metallicity grid is actually described in Section 4.3; also, the caption should clarify whether all combinations of the listed parameter values were computed.","section":"Table 3 caption"},{"comment":"The description for the ZINC field says 'Abundance of helium in log(Zn/H)'; it should say 'Abundance of zinc in log(Zn/H)'.","section":"Table 7"},{"comment":"The caption says the figure is 'identical' to Fig. 20 of Allen et al. (2008) even though the models were computed with a different code; please use a phrase such as 'comparable' or 'equivalent' instead.","section":"Figure 2 caption"},{"comment":"The paper states that 4132 emission lines are stored but does not provide the line list in the text; a link, appendix, or sample line list would help readers assess the wavelength and species coverage.","section":"Section 3.1"},{"comment":"The database is referenced only by a URL; a persistent identifier or DOI would make the resource more robust for long-term citation and archival access.","section":"Section 3.2"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Chris — this is a resource paper, and for that genre it's a decent one. What's actually new: a low-metallicity shock grid built on Gutkin et al. abundances, an age-truncated (young shock) grid, and the first systematic comparison of MAPPINGS V against the Allen et al. (2008) MAPPINGS III grids at fixed parameters. The database itself is the product: 4132 lines per model from 660 Å to 609 µm, shock/precursor/shock+precursor columns, ion fractions and column densities, all queryable through MySQL and with a working web explorer. The SQL example in Table 1 demonstrates the interface concretely. If this holds up, it reduces real friction for people who need ready-made shock grids.\n\nThe paper is honest where it counts. The authors admit they can't say whether the ~0.3 dex line differences from Allen08 come from code changes or atomic data, and they state plainly that dust is excluded. That's the right way to present a re-computation.\n\nNow the soft spots, in order. The first is Section 2.1. The paper says the Allen08 iterative preionization fails below ~200 km/s and that MAPPINGS V 'for the first time' treats preshock photoionization and thermal structure time-dependently. Every 100–200 km/s model and all the precursor emission above 100 km/s depends on this solver, and there is no convergence test, no independent calculation to compare against, and no observational check of that regime. The stress-test note lands: at low velocities the stored line ratios may be different without being better. This is the least secure part of the central claim. It's not fatal—the 200–1000 km/s grid sits on a method with a long track record—but it's a genuine caveat that a referee should ask to have validated or explicitly flagged at the point of use.\n\nSecond, reproducibility is conditional. There's no versioned snapshot of the grids, no released scripts; the live MySQL server can change over time. The main grid contents are plausible but not independently verifiable from the text. That's the main reason I'd condition acceptance on a persistent archive of the presented grids.\n\nThe citation pattern is fine; the self-citation to Morisset et al. (2015) is the infrastructure paper this extends. No circularity problem—these are forward models from a published code, not fitted to anything.\n\nBottom line: this deserves a serious referee and will see real use if the snapshot and low-velocity validation are handled. I'd send it to review, with those two conditions.","headline":"A genuinely useful shock-model resource with honest caveats; the unvalidated 100–200 km/s preionization regime and the lack of a versioned snapshot are the real conditions on acceptance.","tokens_in":15541,"tokens_out":5736,"would_cite":true,"duration_ms":54088,"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":"3MdBs is a public SQL-queryable database of fully radiative MAPPINGS V shock models: an exact replica of the Allen et al. (2008) grids, a low-metallicity extension, and grids of young, age-truncated shocks.","keywords":["astronomical databases","radiative shock models","MAPPINGS V","3MdBs","preionization","low-metallicity shocks","young shocks","BPT diagnostic diagram"],"falsifier":"Run an independent MAPPINGS V calculation of a stored model, for example a 400 km s$^{-1}$ shock with $n_0 = 1$ cm$^{-3}$, $B = 3.23\\,\\mu$G and solar abundances, and compare every stored line intensity; then compare the stored precursor line fluxes against UV/optical spectra of a well-studied radiative shock such as a Cygnus Loop filament, since a systematic mismatch with observed precursor spectra or an irreproducible stored intensity would show the database values are biased.","tokens_in":14597,"feed_emoji":"💾","tokens_out":19546,"duration_ms":160868,"temperature":0.7,"pith_summary":"The paper establishes 3MdBs, the shock-model extension of the Mexican Million Models database, in which thousands of fully radiative shock models computed with MAPPINGS V are stored and made queryable online through the standard MySQL/SQL protocol. For each model the database keeps the intensities of 4,132 emission lines from 660 Å to 609 µm for the shocked gas, the photoionized precursor, and the sum of both, together with ionic fractions, temperatures, and column densities. Three grids are available at publication: an exact parameter replica of the Allen et al. (2008) shock grids, a low-metallicity extension built on the Gutkin et al. (2016) abundances, and grids of young, age-truncated shocks. A reader should care because shock-excitation diagnostics such as BPT and LINER diagrams have until now had to be read from scattered tables and files, whereas this database makes them queryable in seconds, and the paper's first comparisons already show that recomputing the classic grids with MAPPINGS V moves [O III] 5007 Å/Hβ upward by up to about 0.3 dex.","feed_headline":"Thousands of shock-model spectra go live in a public database","feed_subtitle":"The archive recomputes the Allen 2008 grids and adds low-metallicity and young shocks.","key_machinery":"The carrying mechanism is the database architecture itself: a modular set of twelve MySQL tables, including shock parameters, abundances, five emission-line tables split by wavelength interval (660–938 Å, 939–1527 Å, 1528–2999 Å, 3000–7499 Å, and 7500 Å–609 µm), and tables for ionic fractions, ionic temperatures, and ionic column densities, linked by model and abundance identifiers so that any parameter combination or line ratio can be retrieved with a single SQL query. The scientific content is produced by the MAPPINGS V shock and photoionization code (version 5.1.13), and in particular by its iterated, fully time-dependent treatment of preionization: the code solves the photoionization, recombination, photoelectric heating, and line cooling of the gas as it approaches the shock front and repeats until the precursor state converges, replacing the older iterative method that assumed ionization equilibrium and is valid only above about 200 km s$^{-1}$.","core_discovery":"On the authors' terms, the central claim is the creation of a shared resource: a live MySQL database, named 3MdBs, that delivers fully radiative shock models in which each model is identified by shock velocity, preshock density, transverse magnetic field, abundance set, and (for the young grid) shock age, with line intensities stored for the shocked gas, the precursor, and their sum. The flagship grid reproduces Allen et al. (2008) exactly in parameter space, with five abundance sets, velocities from 100 to 1,000 km s$^{-1}$, magnetic fields from $10^{-4}$ to 10 $\\mu$G, and densities from 0.01 to 1,000 cm$^{-3}$, recomputed with MAPPINGS V, whose new time-dependent treatment of preshock ionization replaces the iterative equilibrium method that fails below about 200 km s$^{-1}$. Recomputing changes the diagnostics: [O III] 5007 Å/Hβ comes out up to about 0.3 dex stronger than in the MAPPINGS III models in parts of the BPT diagram, while [C II] 2327 Å and [Ne III] 15.5 µm come out weaker. The extension grids add two cautionary results: at very low metallicity the metal lines almost vanish while hydrogen and helium lines persist, and young, age-truncated shocks at low metallicity can fall below the Kewley et al. (2001) shock classification curve.","pith_inferences":["Editorial inference: the age-truncated grid stores cumulative line fluxes as a function of time, so the same data could be used to estimate the age of a young supernova remnant or Herbig-Haro object by matching observed line ratios against the stored time series, a use the paper does not develop.","Editorial inference: all stored models exclude dust, so the optical line intensities represent a dust-free case; a dust-inclusive extension of the same schema would test how much the diagnostic diagrams shift in dusty environments.","Editorial inference: the reported shift of up to 0.3 dex in [O III]/Hβ between code versions implies that published AGN and LINER classification boundaries calibrated on the older MAPPINGS III grid may need recalibration, a conclusion the authors do not draw.","Editorial inference: the time-dependent preionization treatment makes a specific, testable prediction, that below roughly 200 km/s equilibrium preionization fails and the precursor ionization structure should differ systematically, which could be checked against UV spectra of slow radiative shocks."],"forward_implications":["Anyone with a SQL client can reproduce the Allen et al. (2008) shock grids and build diagnostic diagrams such as BPT or LINER plots in seconds, instead of reading model tables out of published papers.","The MAPPINGS V recomputation shifts some line ratios relative to the MAPPINGS III grid, with [O III] 5007 Å/Hβ up to about 0.3 dex stronger in parts of the diagram, so shock classifications and parameter estimates based on the older grid carry a systematic offset.","At very low metallicity the metal collisionally excited lines become negligible while hydrogen and helium lines remain, so a shock component superimposed on a photoionized region could make standard abundance methods underestimate quantities such as O/H.","Young, age-truncated shocks at low metallicity can fall below the Kewley et al. (2001) classification curve, so low [S II]/Hα or [N II]/Hα ratios do not by themselves exclude a shock origin.","Because the database is live, new grids, including future MAPPINGS V versions or community-requested parameter sets, become immediately accessible to every user the moment they are added."],"supporting_citations":[{"why":"Supplies the grid being replicated: five abundance sets, the velocity, magnetic-field, and density ranges, and the MAPPINGS III baseline against which the new models are compared.","marker":"Allen et al. (2008)"},{"why":"Provides the MAPPINGS V improvements, above all the iterated time-dependent preionization treatment that generates every stored model.","marker":"Sutherland & Dopita (2017)"},{"why":"Describes the original 3MdB photoionization database whose SQL logic and access model 3MdBs extends to shock models.","marker":"Morisset et al. (2015)"},{"why":"Defines the abundance sets and the element-by-element method used to build the low-metallicity grid.","marker":"Gutkin et al. (2016)"},{"why":"Supplies two of the five abundance sets (solar and twice solar) used in the Allen 2008 replica sub-grid.","marker":"Dopita & Sutherland (1996)"},{"why":"Provides the LINER diagnostic diagrams against which the new MAPPINGS V grids are compared and where the difference of up to 0.3 dex appears.","marker":"Molina et al. (2018)"},{"why":"Defines the classification curve that the young, low-metallicity shock models are shown to fall below.","marker":"Kewley et al. (2001)"}],"fun_headline_variants":["Shock model archive goes live with MAPPINGS V grids","Public shock database offers young and metal-poor models","3MdBs: live MySQL shock models, hundreds of grids","MAPPINGS V shock grids now online, from IR to X-ray","Radiative shock database expands Allen 2008 with age"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The database's scientific value rests on one untested premise: that the new time-dependent preionization treatment in MAPPINGS V correctly computes the ionization and thermal state of the gas entering the shock, since the paper states this treatment is new, notes that the older method fails below about 200 km/s, and does not compare the predicted line intensities with observations.","fun_headline_variants_meta":{"raw":{"variants":["Shock model archive goes live with MAPPINGS V grids","Public shock database offers young and metal-poor models","3MdBs: live MySQL shock models, hundreds of grids","MAPPINGS V shock grids now online, from IR to X-ray","Radiative shock database expands Allen 2008 with age"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000279,"raw_usage":{"total_tokens":1647,"prompt_tokens":923,"completion_tokens":724,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":539,"completion_tokens_details":{"reasoning_tokens":638}},"tokens_in":539,"tokens_out":724,"duration_ms":6933,"temperature":1.0,"reasoning_tokens":638,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:35:04.638334+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run an independent MAPPINGS V calculation of a stored model, for example a 400 km s$^{-1}$ shock with $n_0 = 1$ cm$^{-3}$, $B = 3.23\\,\\mu$G and solar abundances, and compare every stored line intensity; then compare the stored precursor line fluxes against UV/optical spectra of a well-studied radiative shock such as a Cygnus Loop filament, since a systematic mismatch with observed precursor spectra or an irreproducible stored intensity would show the database values are biased.","supporting_citations":[{"cited_title":"2015, rmxaa, 51, 103","cited_arxiv_id":null,"evidence_quote":"Describes the original 3MdB photoionization database whose SQL logic and access model 3MdBs extends to shock models."},{"cited_title":"J., Maoz , D., Runnoe , J","cited_arxiv_id":null,"evidence_quote":"Provides the LINER diagnostic diagrams against which the new MAPPINGS V grids are compared and where the difference of up to 0.3 dex appears."}],"review_version":1}