{"id":"ffd5c1c0-4902-44c4-8826-b3aab1761856","arxiv_id":"2502.02839","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"The expected number of contaminating supernova-like transients inside LISA localization volumes only drops to about one for low-redshift mergers, and only when redshifts are available for every detected transient.","lead":"This paper calculates how many unrelated cosmic explosions, such as supernovae, will appear inside the error volumes that LISA will assign to massive black hole mergers. It finds that without redshift information for those explosions, follow-up searches would see roughly one hundred times more false candidates and could never reduce the list to a single likely counterpart.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Quoted z thresholds (z~0.8 pre-merger, z~1.5 post-merger) ride on unvalidated Fisher-matrix localization volumes from Mangiagli et al. (2020); a factor-of-few volume bias shifts the unity contours and would change the headline numbers.","rationale":"The paper's calculation is transparent and the qualitative message—that redshift information reduces contaminants by roughly two orders of magnitude—is robust to plausible rate uncertainties, since core-collapse supernovae dominate and their volumetric rate is known to factor-of-two precision. The genuinely novel mission-planning numbers are the unity thresholds at specific redshifts, and those depend almost linearly on the localization volume. The reader identified the same unvalidated external volume model; my reading adds a concrete mechanism: Eq. 1's separable volume approximation plus the post-merger extrapolation of fits that were derived from pre-merger Fisher information. This is not grounds for rejection—it is a standard conditional-acceptance request for validation of the inherited volume model. I therefore keep the reader's CONDITIONAL verdict and specify one check that would settle whether the concern actually lands.","tokens_in":19488,"tokens_out":4072,"duration_ms":44511,"concrete_test":"Recompute Vloc for a subset of the 1600-event grid (e.g., Mtot = 1e5, 1e6, and 1e7 Msun at z = 0.3, 0.8, 1.5, and 2.5, at tc = 24 h, 1 h, and after merger) using the full 15-parameter Fisher covariance matrix with the same waveform and LISA response as Mangiagli et al. (2020), integrating the actual 3D credible ellipsoid rather than Eq. 1's product approximation. For a few points, run a full Bayesian parameter estimation as a cross-check. If the Mangiagli polynomial volumes agree within a factor of ~2, the headline thresholds stand; if they differ by more than a factor of ~3, recompute the Figure 5 Ntrans maps and quote revised z thresholds, explicitly stating whether the after-merger fits are even defined.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim—that the number of contaminants reaches unity at z~0.8 one hour before coalescence and at z~1.5 after coalescence—is controlled by Vloc, not by the transient rates. Vloc is adopted without validation from the parametric Fisher-matrix fits of Mangiagli et al. (2020) in Section 2.2 via Eq. 1. Those fits average over mass ratio, spin, and other waveform parameters, and Eq. 1 approximates the 3D localization volume as a separable product of sky area and a radial luminosity-distance shell, discarding parameter correlations that are typically strong for LISA massive-black-hole signals. The 'After Merger' panels in Figures 4–5 require evaluating the fits at or beyond the end of the inspiral signal, a regime in which the Fisher predictions need explicit validation; Section 4.2 concedes that the adopted method ignores mass ratio and that rerunning the Mangiagli et al. experiment would be worthwhile, but no such test is provided. Because Ntrans is approximately linear in Vloc, a factor-of-few bias in Vloc translates directly into a shift of the z thresholds at which Ntrans=1. This is the weakest load-bearing assumption in the paper.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper estimates the number of unrelated electromagnetic transients expected to fall inside LISA gravitational-wave localization volumes for massive black hole mergers. The calculation combines localization volumes from the parametric Fisher-matrix fits of Mangiagli et al. (2020) with literature volumetric rates for several transient classes (SN Ia, SN Iax, CCSN, short and long GRBs, TDEs, FRBs) via the steady-state integral in Eq. (8). Two scenarios are compared: one in which redshifts are available for all detected transients, so only transients within the localization volume contribute, and one in which redshifts are unavailable, so all transients in the sky patch out to z=3 are counted. Results are presented as average contamination numbers for a merger catalog following the NewHorizon redshift distribution and as a grid in total mass and redshift at several times before and after coalescence. The headline findings are that with redshift information the expected number of contaminants drops to unity around z~0.8 one hour before coalescence and at z~1.5 after coalescence, while without redshifts the number increases by an average factor of ~100 and never drops below unity.","tokens_in":19728,"tokens_out":6612,"duration_ms":66797,"significance":"If the quantitative thresholds hold, the paper provides useful guidance for LISA follow-up strategy, particularly the need for transient redshifts. The calculation is transparent and internally consistent: Eq. (8) is a standard steady-state estimate, all inputs come from external catalogues and literature rate measurements, and no parameter is fitted to the target contamination counts. The qualitative conclusion that redshift information is critical is likely robust to moderate variations in the inputs. However, the exact z values at which the contamination count crosses unity rest on the adopted localization volumes and on point-value rate estimates, and the current manuscript does not quantify how those uncertainties propagate into the headline numbers.","major_comments":[{"comment":"The localization volumes that control Ntrans through Eq. (8) are taken exclusively from the parametric Fisher-matrix fits of Mangiagli et al. (2020), with no validation for the parameter space or time ranges used in this paper. Because Ntrans is approximately proportional to Vloc, a factor-of-few bias in Vloc translates directly into a shift of the headline thresholds, e.g., the z~0.8 pre-merger and z~1.5 post-merger unity crossing quoted in the abstract and Section 3.2. The 'After Merger' panels require evaluating the fits at or beyond the calibration regime, and the separable form of Eq. (1) discards correlations among luminosity distance, sky position, and other waveform parameters that are typically strong for LISA massive-black-hole signals. Section 4.2 acknowledges that the adopted method ignores mass ratio and that rerunning the Mangiagli et al. experiment would be worthwhile, but no quantitative test is supplied. I request a validation against full parameter estimation for a few representative (Mtot, z, tc) points, or at minimum a sensitivity study showing how the Ntrans=1 contours shift when Vloc is rescaled by a factor of 2-3.","section":"Section 2.2, Eq. (1); Figures 4-5"},{"comment":"The volumetric rates R(z) are used as point values without propagating their uncertainties. The shaded regions in Figure 3 show the 1-sigma scatter across simulated mergers, not the uncertainty in the adopted rates, so they understate the uncertainty in Ntrans. Since Ntrans is linear in each R(z), published rate uncertainties (typically factors of ~1.5-3 for supernovae, and larger for TDEs and FRBs, including the FRB beaming-fraction assumption) can move the z thresholds at which Ntrans crosses unity by a comparable amount. Please propagate rate uncertainties, for example by Monte Carlo sampling the rate normalizations and redshift-evolution parameters, and display the resulting uncertainty bands on the Ntrans=1 contours in Figures 4 and 5.","section":"Section 2.3-2.4, Eq. (8); Figure 3"},{"comment":"The 'without redshift' scenario integrates Eq. (8) from z=0 to zmax=3, and the abstract's factor-of-~100 increase and the statement that the count 'never drops below unity' rest on this choice. The cutoff is justified only by the qualitative statement that beyond z=3 transients are too faint for the largest telescopes; no detection or limiting-magnitude model is applied, so the integrated count is an upper limit whose numerical value depends on zmax. Because the factor ~100 is a headline result, please show the sensitivity of the factor and of the unity crossing to zmax (e.g., zmax=2 and zmax=4) and to a simple flux-limited detectability model.","section":"Section 2.4 and Section 3.2 (no-redshift scenario)"}],"minor_comments":[{"comment":"There is a typo in 'Howeber' in the first sentence of the third paragraph.","section":"Section 4.2"},{"comment":"In the Harris et al. reference, the journal name is written as 'Natur'; it should be 'Nature'.","section":"References"},{"comment":"The color bar labels and contour annotations in Figures 4 and 5 are dense and difficult to read at the current size; consider enlarging the panels or adding explicit contour values.","section":"Figures 4-5"},{"comment":"The first investigation uses 141,834 simulated mergers while the NewHorizon simulation predicts roughly five LISA-observable mergers over the mission; please clarify how the catalog is sampled from the rate distribution and why the large sample is appropriate for computing the average.","section":"Section 2.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and addresses a timely question for LISA follow-up planning. The central calculation is simple and reproducible, and the qualitative conclusion is likely robust. My main concern is that the precise z thresholds and the factor-~100 claim are presented with more confidence than the inputs warrant, and the manuscript's own caveats in Section 4.2 essentially concede the largest issue without quantifying it. I would be willing to reconsider after the requested sensitivity and validation tests are added."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a useful and overdue calculation—it counts the unrelated supernovae, TDEs, GRBs, and FRBs that will sit inside LISA localization volumes during massive black hole merger follow-up. The headline numbers are the right order of magnitude, and the qualitative conclusion (redshifts for candidate transients are not optional) is secure. But the exact thresholds—z~0.8 one hour before merger, z~1.5 after—ride on the localization volume fits from Mangiagli et al. (2020), and the paper does not stress-test them.\n\nWhat is genuinely new: the contamination count as a function of Mtot, z, and time to coalescence is a straightforward but sensible application of volumetric rate integration to LISA volumes. The authors use independent literature rates for six transient classes, integrate over the redshift shell, and compare idealized with- and without-redshift scenarios. The result that the no-redshift case is ~100 times worse and never drops below unity is striking and robust to the details. Credit also for Section 3.3 showing that core-collapse supernovae dominate; that helps planners know what they are fighting. The paper is transparent about its idealized observing assumptions and states the main caveat in Section 4.2.\n\nSoft spots, in proportion. The dominant input is Vloc, and it is adopted as a black box. The Fisher-matrix parametric fits average over mass ratio and spin, and the after-merger regime is exactly where those fits need independent validation. The authors concede this is worthwhile future work, but for a paper whose abstract quotes unity thresholds to one decimal place, a sensitivity test on Vloc is not optional. A factor-of-two or three error in Vloc shifts the z thresholds by maybe ±0.2–0.3; the 'never drops below unity' conclusion survives, but the precise contours in Figures 4–5 would not. Next, the transient rates are used as point values with no uncertainty propagation. The ~100 factor is so large that this is not fatal, but it makes the claimed precision in the abstract feel higher than the inputs justify. Finally, no code or data is shipped; for a calculation this simple, a table of Vloc fits or a script would make the thresholds independently checkable.\n\nWho is this for? LISA follow-up working groups, survey time-allocation committees, and anyone writing a strategy paper for multi-messenger astronomy. It is a solid first paper in what will be a series, and it deserves a serious referee. My recommendation: send it to review, but require the authors to add a sensitivity analysis on Vloc and to caveat the quoted thresholds accordingly.","headline":"Useful first estimate of LISA follow-up contamination; the factor-of-100 redshift penalty is robust, but the quoted z thresholds depend on Fisher-matrix volumes that the paper never stress-tests.","tokens_in":20294,"tokens_out":3715,"would_cite":true,"duration_ms":33835,"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 quantifies how many unrelated electromagnetic transients will contaminate LISA localization volumes and shows that redshift information is the decisive requirement for identifying true counterparts.","keywords":["gravitational waves","LISA","massive black hole mergers","electromagnetic counterparts","transient astronomy","supernovae","tidal disruption events","localization volumes"],"falsifier":"Run full Bayesian LISA parameter estimation on a grid of massive-black-hole merger waveforms spanning $10^5$–$10^7\\,M_\\odot$ and $z=0.2$–$3.2$, and compare the resulting localization volumes with the parametric fits; if the volumes differ by more than a factor of two, the reported unity-contaminant thresholds at $z\\simeq0.8$ pre-merger and $z\\simeq1.5$ post-merger are not reliable.","tokens_in":19293,"feed_emoji":"🔭","tokens_out":8709,"duration_ms":74739,"temperature":0.7,"pith_summary":"LISA will detect massive black hole mergers in advance, and telescopes will search those sky regions for an electromagnetic flash from the merger itself. This paper asks a prior question: how many unrelated transients—supernovae, tidal disruption events, gamma-ray bursts, fast radio bursts—will happen to lie in the same localization volume and masquerade as the counterpart. Using simulated merger populations and localization volumes, it finds that when every detected transient can be assigned a redshift, the expected contaminant count falls to about one or fewer for mergers at $z \\lesssim 0.8$ in the final hour before coalescence, and after coalescence the one-contaminant region extends to $z \\lesssim 1.5$. Without redshifts, the count jumps by an average factor of roughly 100 and never drops below unity, so the paper's practical conclusion is that follow-up strategy should treat redshift acquisition for candidate transients as a primary requirement.","feed_headline":"Redshifts cut LISA false-transient count one hundredfold","feed_subtitle":"Without redshifts, unrelated supernovae outnumber the true EM counterpart for every LISA merger.","key_machinery":"The counting machinery is a volume integral over the LISA localization volume: $N_{\\rm trans} = (\\Delta\\Omega/4\\pi) \\int_{z_{\\min}}^{z_{\\max}} R(z)\\, (dV_c/dz)\\,\\Delta t\\, dz$, where $\\Delta\\Omega$ and $\\Delta t$ are the sky localization region and transient lifetime, and $R(z)$ is the volumetric rate of each transient class. The localization volumes themselves come from parametric Fisher-matrix fits that give the fractional distance error and sky-area error as functions of merger total mass, redshift, and time to coalescence; these volumes shrink as coalescence approaches and grow with mass and redshift. Available redshift information enters by restricting the integration to the redshift shell spanned by the localization volume, whereas unavailable redshifts force integration over $z\\in[0,3]$, which is what produces the factor-of-100 increase.","core_discovery":"The central discovery is that the expected number of unrelated electromagnetic transients inside a LISA localization volume is controlled mainly by the size of that volume, not by the redshift dependence of transient rates, and that redshift information is what separates a tractable search from an intractable one. In the idealized scenario where a redshift is available for every transient detected in the LISA sky region, the expected contaminant count drops to unity at $z \\lesssim 0.8$ for mergers one hour before coalescence, and after coalescence it stays below unity for total masses $10^{5.5}\\,M_\\odot \\lesssim M_{\\rm tot} \\lesssim 10^{6.5}\\,M_\\odot$ out to $z \\lesssim 1.5$. In the opposite scenario, counting everything from $z=0$ to $z=3$ in the same sky region, the expected count rises by an average factor of about 100 and never falls below one, even for the closest mergers. Core-collapse supernovae make up roughly 80 percent of the contaminants at all redshifts considered.","pith_inferences":["Beyond the paper, the same counting integral can be applied to other future gravitational-wave detectors and to stellar-mass sources; the qualitative conclusion that redshift filtering dominates over localization-volume size should transfer, though the specific thresholds will change.","A partial-redshift scenario, not treated in detail here, would place the true contaminant count between the two extremes; even coarse photometric redshifts might remove most of the factor of 100, relaxing the need for spectroscopy of every candidate.","The paper's assumption that transient rates are known from $z=0$ to $3$ could be tested with the first years of wide-field time-domain surveys, which will measure the supernova rate at the redshifts LISA will probe."],"forward_implications":["Before coalescence, the contaminant count for low-redshift mergers ($z \\lesssim 0.2$) reaches unity as early as six hours before merger when redshifts are known, so pre-merger electromagnetic monitoring with redshift filtering is feasible only for nearby events.","After coalescence, mergers with total mass near $10^6\\,M_\\odot$ at $z \\lesssim 1.5$ have at most one contaminant, making them the cleanest targets for counterpart identification.","Without redshifts, every LISA follow-up field will contain roughly 100 unrelated transients, so 'find the new source in the error box' strategies cannot work by themselves.","Because roughly 80% of contaminants are core-collapse supernovae, real-time transient classification can remove most false candidates, but classification alone does not remove the need for redshifts.","The paper's three recommendations—host-galaxy redshift catalogs, real-time classification, and coordinated follow-up spectroscopy—follow directly from the factor-of-100 gap between the two scenarios."],"supporting_citations":[{"why":"Supplies the Fisher-matrix localization-volume fits used in Eqs. (1) and (8) to compute volumes for every simulated merger.","marker":"Mangiagli et al. (2020)"},{"why":"Provides the cosmological merger redshift distribution used to build the first catalog of simulated MBH mergers.","marker":"Volonteri et al. (2020)"},{"why":"Supplies the Type Ia supernova volumetric rate adopted at $z<1$.","marker":"Dilday et al. (2008)"},{"why":"Supplies the Type Ia supernova volumetric rate adopted at $z>1$.","marker":"Hounsell et al. (2018)"},{"why":"Supplies the core-collapse supernova volumetric rate, the dominant contaminant class.","marker":"Strolger et al. (2015)"},{"why":"Supplies the Type Iax supernova volumetric rate at $z=0$.","marker":"Foley et al. (2013)"},{"why":"Supplies the median transient lifetimes used as $\\Delta t$ for supernovae and gamma-ray bursts.","marker":"Villar et al. (2017)"},{"why":"Supplies the fast radio burst volumetric rate with an assumed beaming fraction.","marker":"Law et al. (2017)"}],"fun_headline_variants":["No-redshift LISA searches get 100x more false transients","Without redshifts, LISA false alarms never drop below one","Core-collapse supernovae dominate LISA contaminant population","Redshift info determines if LISA follow-up can succeed","LISA false transient counts hinge on redshift availability"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation inherits the size of every LISA localization volume from approximate fitting formulas for LISA's distance and sky-position measurement errors, averaged over parameters like mass ratio and spin; if those volumes are off by even a factor of a few, all contaminant counts and redshift thresholds shift by a comparable factor.","fun_headline_variants_meta":{"raw":{"variants":["No-redshift LISA searches get 100x more false transients","Without redshifts, LISA false alarms never drop below one","Core-collapse supernovae dominate LISA contaminant population","Redshift info determines if LISA follow-up can succeed","LISA false transient counts hinge on redshift availability"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000327,"raw_usage":{"total_tokens":1898,"prompt_tokens":1081,"completion_tokens":817,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":697,"completion_tokens_details":{"reasoning_tokens":734}},"tokens_in":697,"tokens_out":817,"duration_ms":8106,"temperature":1.0,"reasoning_tokens":734,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T10:54:59.303804+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run full Bayesian LISA parameter estimation on a grid of massive-black-hole merger waveforms spanning $10^5$–$10^7\\,M_\\odot$ and $z=0.2$–$3.2$, and compare the resulting localization volumes with the parametric fits; if the volumes differ by more than a factor of two, the reported unity-contaminant thresholds at $z\\simeq0.8$ pre-merger and $z\\simeq1.5$ post-merger are not reliable.","supporting_citations":[],"review_version":1}