{"id":"be2eb719-802a-4658-812e-aeefddfd4ab3","arxiv_id":"1908.01316","paper_version":3,"verdict":"UNVERDICTED","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"The authors propose a space-based mid-infrared nulling interferometer and estimate it could detect 423 nearby exoplanets, including about 30 in the habitable zone, enabling atmospheric biosignature searches.","lead":"This white paper argues that a large space-based mid-infrared interferometer is the best way to study the atmospheres of rocky exoplanets and search for signs of life. It presents simulations suggesting such a mission could detect over 400 nearby exoplanets and characterize the atmospheres of dozens of habitable-zone planets.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Mission concept rests on an end-to-end nulling performance at 3–20 µm that is extrapolated from narrowband lab tests; the required null floor and in-space formation jitter are not jointly demonstrated.","rationale":"The reader’s weakest assumption correctly identifies the nulling interferometer’s feasibility as the load-bearing element of the paper’s central claim. The paper’s own Section 4.2 acknowledges that cryogenic nulling, spatial filters, deformable mirrors, and space-qualified formation-flying metrology still require development, so the central yield and characterization claims are conditional on an integrated capability that has not yet been demonstrated. I considered alternative potential concerns—the transfer of Kepler occurrence rates to the solar neighborhood, the absence of a detailed characterization-phase exposure-time calculation, and the use of an Earth-twin retrieval template—but these are secondary and, at least at white-paper level, reasonably supported by cited literature and internal consistency. The strongest point of failure is therefore the end-to-end null-depth performance: the cited lab results are narrowband and not co-tested with the flight-like formation and cryogenic environment, and the paper provides no sensitivity analysis showing how yield degrades if the null floor worsens. This does not change the reader’s verdict: the document remains an advocacy white paper whose scientific case is plausible but unverified, so UNVERDICTED is still appropriate. The proposed concrete test would move the verdict toward CONDITIONAL if the end-to-end simulation fails, or toward ACCEPT if it succeeds at realistic error levels.","tokens_in":20210,"tokens_out":10369,"duration_ms":119855,"concrete_test":"End-to-end simulation: take the complex transmission and phase error budget measured on the Planet Detection Testbed (Martin et al. 2012), add representative PROBA-3-level formation jitter (mm-level separation noise, arcsec pointing noise) and cryogenic wavefront errors for a 170 m baseline, and propagate through the same post-processing pipeline. Require the recovered residual null to stay below 1e-7 of the stellar flux at the key biosignature wavelengths (e.g., O3 near 9.6 µm, CH4 near 7.7 µm, N2O near 7.8 µm, CO2 near 15 µm). If this cannot be met even in simulation, the mission concept as specified lacks a demonstrated path to its core contrast requirement.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that a large MIR nulling interferometer uniquely enables Q1–Q3 depends on reaching a post-processed stellar leakage of order 1e-7–1e-8 of the host star flux across 3–20 µm with ~170 m baselines and ~2.5 m collectors (§2.2). The cited laboratory evidence is narrower than that requirement: Martin et al. (2012) report a null depth of 8e-6 at 10 µm with 10% bandwidth, and 1e-8 total suppression only after post-processing; the Adaptive Nuller reached 1e-5 at 34% bandwidth (§4.1). Neither demonstrates the required chromatic null stability from 3–20 µm, cryogenic operation, or behavior under the metrology and pointing errors of a free-flying, multi-spacecraft 170 m baseline. PROBA-3’s mm-level and arcsec-level formation precision controls spacecraft position, but the actual science requirement is optical-path-difference stability to a small fraction of 10 µm, maintained by internal metrology and delay lines that have not been tested together with MIR nulling in space. The paper itself lists these as required future developments (§4.2), so the yield and characterization numbers assume an integrated performance currently extrapolated, not measured. If the in-flight null floor is only a few times worse than the lab value, the Earth–Sun contrast near 1e-7 is no longer detectable, and the central 'dozens of terrestrial exoplanets' scenario lacks a demonstrated instrumental basis.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This white paper argues that a large space-based mid-infrared nulling interferometer is the only currently conceived approach that can detect and spectroscopically characterize dozens of terrestrial exoplanets, addressing three questions: the prevalence of atmospheric biosignatures (Q1), the fraction of planets with conditions suitable for liquid water (Q2), and the diversity of terrestrial exoplanet atmospheres (Q3). It presents a spectral retrieval simulation of an Earth-twin observed at R=100 and SNR=20 over 3-30 micron; a statistical analysis of the power of a null result as a function of the number of characterized habitable-zone planets; an updated Monte Carlo yield simulation based on Kepler occurrence rates and a 320-star sample within 20 pc, predicting about 423 detectable planets and about 30 habitable-zone terrestrial planets; and a review of technology readiness and required developments. The central claim is that the MIR approach surpasses reflected-light alternatives and should be pursued as an ESA L-class mission.","tokens_in":20542,"tokens_out":8067,"duration_ms":78331,"significance":"If the quantitative yields and retrieval performance hold up, the proposed mission would provide the first statistically meaningful sample of terrestrial exoplanet atmospheres, enabling direct searches for biosignature gases (O3, CH4, N2O) and measurements of surface conditions. The paper is timely for the Voyage 2050 planning cycle and provides a useful synthesis of occurrence-rate statistics, retrieval methods, and interferometry technology status. Its strengths include the explicit use of external empirical benchmarks for planet occurrence and exozodi levels, the transparent acknowledgment that several key technologies are not yet space-qualified (Section 4.2), and a quantitative comparison with a reflected-light retrieval study. The principal weakness is the dependence of the central claims on extrapolated nulling performance and simplified statistical and retrieval models, which the paper does not yet subject to sensitivity analysis.","major_comments":[{"comment":"The simulated yield of about 423 detected planets and about 30 habitable-zone terrestrial planets (Figure 6) assumes a nulling interferometer with about 170 m baselines and about 2.5 m collectors that achieves a post-processed stellar leakage sufficient to detect an Earth-Sun contrast of about 1e-7 at 10 micron. The cited laboratory demonstrations do not cover this parameter range: Martin et al. (2012) report a null depth of 8e-6 at 10 micron with 10% bandwidth (with 1e-8 total suppression after post-processing), and the Adaptive Nuller reached 1e-5 at 34% bandwidth. Neither demonstrates the required chromatic null stability from 3-20 micron, cryogenic operation, or performance under the metrology and pointing errors of a free-flying multi-spacecraft configuration. The paper itself identifies cryogenic validation as future work (Section 4.2). Consequently the yield numbers, which are central to the paper's argument for a large mission, are extrapolations. To make the claim load-bearing, the authors should add a sensitivity analysis showing how the yield and the number of habitable-zone planets vary with nulling floor (e.g., 1e-6 vs 1e-5), instrument throughput, and exozodi level, or explicitly state that the yield estimates are upper bounds pending technology demonstration.","section":"Sections 1.2.3, 2.2, and 4.1"},{"comment":"The retrieval study used to demonstrate the scientific potential of MIR spectra for Earth-twin characterization is not described in sufficient detail to be reproducible or to substantiate the claimed precision. Neither the atmospheric forward model, the retrieval algorithm, the noise model, nor the prior ranges for the free parameters (including the pressure-temperature profile) are specified. The conclusion that 'all parameters are well constrained' rests on a single simulation at R=100 and SNR=20 per resolution element over 3-30 micron; no tests of robustness to clouds, haze, or model uncertainty are presented. The comparison with the reflected-light retrievals of Feng et al. (2018) is also imperfect because the spectral coverage, resolution, and SNR definitions differ between the two studies. Since this comparison underpins the claim in the Abstract that the MIR approach surpasses other approaches, the retrieval analysis needs either additional methodological detail and robustness checks, or the claim should be softened to reflect the illustrative character of the simulation.","section":"Section 1.2.1 and Figures 2-3"},{"comment":"The statistical power analysis for a possible null result assumes that the 30 or 50 observed habitable-zone planets are an unbiased, independent sample from the population described by H0 and that the outcome 'provides conditions for liquid water' can be determined unambiguously for each planet. In practice, the detectability of a planet in the search phase depends on its radius, insolation, and distance, and the characterization of surface conditions from a MIR spectrum will carry systematic uncertainties. These selection effects and measurement errors would degrade the statistical power relative to the idealized calculation, so the statement that '30 planets is the minimum number needed' (Section 1.2.3) is an optimistic lower bound. Please add a caveat to that effect and, if possible, a simple simulation that incorporates selection effects.","section":"Section 1.2.2 and Figure 4"}],"minor_comments":[{"comment":"The figure caption and text mention a spectral range of 3-30 micron, while the title and abstract emphasize 3-20 micron; please unify the wavelength range or explicitly state why the difference is negligible.","section":"Section 1.2.1 and Figure 2"},{"comment":"The left panel's color bar and the numbers inside the bins are not explained in the caption; please add a legend or note that numbers indicate expected detections per bin.","section":"Figure 6"},{"comment":"The description of PRISMA and PROBA-3 could be clearer: the text states PRISMA demonstrated sub-cm positioning and PROBA-3 will achieve mm-level precision, but it does not explain whether the PROBA-3 requirement is for inter-satellite separation or line-of-sight jitter; this distinction matters for the nulling application.","section":"Section 4.1"},{"comment":"There are several distinct Defrère et al. (2018) entries in the reference list; please disambiguate them in the text (e.g., Defrère et al. 2018a,b,c) and ensure each citation points to the correct paper.","section":"References"},{"comment":"The citation 'Ertel et al. (in prep); Ertel et al., 2018' is unusual for a peer-reviewed paper; perhaps cite only Ertel et al. (2018) and mention that the HOSTS results include a future publication.","section":"Section 4.3"}],"recommendation":"major_revision","confidential_remarks":"This is a programmatic white paper rather than a standard research article, and should be judged accordingly. The central argument is plausible, but the quantitative yield and retrieval claims need to be explicitly framed as extrapolations. The paper is not circular: its key inputs (Kepler occurrence rates, HOSTS exozodi, laboratory nulling) are external benchmarks, and the yield simulation is updated from the authors' own prior work in a transparent way. The main risk is the lack of a sensitivity analysis for the assumed nulling performance; this should be addressed in revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things about this paper. First, it is a white paper for ESA's Voyage 2050 process, not a research preprint, so the verdict should be about whether the advocacy is well-argued and fairly grounded. Second, the central claim is that a free-flying MIR nulling interferometer is the only near-term path to characterize dozens of terrestrial exoplanet atmospheres and search for biosignatures. On the science case, I think the paper is persuasive. The retrieval comparison with Feng et al. in reflected light is the most useful piece: it quantifies what MIR emission adds (CH4, N2O, CO, radius, surface pressure and temperature) and what it gives up (O2, red edge). That is a real contribution to the design conversation, even if each ingredient is adapted from prior work.\n\nThe yield simulation is plausible but softer. The 423-planet and ~30 habitable-zone numbers come from a Monte Carlo that updates Kammerer & Quanz (2018) with JWST/MIRI sensitivity and a 3.5x throughput penalty, on a sample of 320 FGKM stars. The inputs are external (Kepler SAG13 rates, Dressing & Charbonneau), so there is no circularity problem. But the paper does not provide enough detail on the instrument model, the noise budget, or the sensitivity of the yield to assumptions like occurrence rate, exozodi, and null floor. The HOSTS median exozodi of 4.5x solar is mentioned but not propagated into the yield error bars. For a planning document that is acceptable; for a refereed claim it would need more.\n\nThe stress-test concern is the load-bearing one. The lab evidence is real but narrower than the requirement: the Planet Detection Testbed achieved 8e-6 null depth at 10 µm with 10% bandwidth, and 1e-8 total suppression only after post-processing, while the mission needs ~1e-7 to 1e-6 nulling across 3–20 µm under cryogenic, free-flying conditions. PROBA-3 demonstrates mm-level formation flying, not the optical path difference stability to a fraction of 10 µm with internal metrology and delay lines. The paper is honest about this—Section 4.2 explicitly lists cryogenic nulling validation and integrated optics as required future work—so the authors are not hiding the gap. But the yield numbers in Section 1.2.3 assume the required performance is available. That gap should be stated more prominently.\n\nWho is this for? Someone writing ESA or NASA roadmap documents, or a colleague who wants a compact, well-referenced case for MIR nulling and a fair comparison to reflected light. It deserves a serious referee, though the referee's job should be to pressure the engineering assumptions and ask for sensitivity analysis, not to reject the argument.\n\nMy recommendation: engage with it as a planning document. If you cite it, cite the retrieval comparison and the yield estimate with the caveat that both assume an undemonstrated end-to-end nulling performance.","headline":"A well-argued Voyage 2050 white paper for a MIR nulling interferometer: the science case is strong and the retrieval comparison is genuinely useful, but the yield numbers rest on an end-to-end nulling performance that has not been demonstrated.","tokens_in":21279,"tokens_out":1507,"would_cite":false,"duration_ms":14103,"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":"A large space-based mid-infrared interferometer is the only approach able to survey dozens of terrestrial exoplanet atmospheres for signs of life and test how common Earth-like conditions are.","keywords":["mid-infrared interferometry","terrestrial exoplanets","atmospheric characterization","biosignatures","habitable zone","nulling interferometer","spectral retrieval","space mission concept"],"falsifier":"The decisive test is engineering: an integrated cryogenic demonstration in which the nulling system reaches a post-processing starlight suppression near $10^{-7}$ at 10 μm while the collector spacecraft simultaneously maintain millimeter-level formation over ~150 m baselines - if either capability falls short at the required scale, the mission concept and its yield estimates collapse. On the science side, a radial-velocity census of the proposed target stars that found temperate small planets to be much rarer than the Kepler-based occurrence rates assumed in the yield simulations would settle the yield claim directly, by showing that the projected ~30 habitable-zone planets were an overestimate.","tokens_in":2553,"feed_emoji":"🔭","tokens_out":7868,"duration_ms":224743,"temperature":0.7,"pith_summary":"This paper argues that the only way to find out whether any other world harbors signs of life, and how common Earth-like surface conditions are, is a single kind of instrument that has not yet been built: a large space-based mid-infrared interferometer measuring the heat emitted by terrestrial planets around nearby stars. The authors claim that no currently adopted mission or ground facility, including JWST and the coming extremely large ground telescopes, can obtain thermal emission spectra of temperate rocky planets in the numbers needed, namely dozens. Three quantitative results carry the argument: a spectral retrieval showing that a simulated Earth-twin's mid-infrared spectrum pins down atmospheric abundances, surface pressure, surface temperature, and radius; a statistical calculation showing that roughly 30-50 characterized planets suffice for a null result to bound how rare Earth-like conditions are; and Monte Carlo yield simulations predicting more than 400 detectable planets, including about 30 terrestrial planets in the empirical habitable zone, around 320 stars within 20 parsecs. The paper presents this mission concept as the timely next step for a European-led long-term space science program.","feed_headline":"One space mission could detect 400 nearby exoplanets","feed_subtitle":"Thermal spectra of rocky worlds would reveal biosignatures and surface conditions no planned telescope can reach.","key_machinery":"The central object is the mid-infrared thermal emission spectrum of a terrestrial planet, whose absorption bands - ozone near 9.6 μm, methane near 7.7 μm, nitrous oxide near 7.8 μm, water near 6.2 μm, carbon dioxide near 15 μm, and carbon monoxide near 4.67 μm - carry the information about composition, temperature structure, surface pressure, and surface temperature. The instrument proposed to obtain these spectra is a nulling interferometer, a formation of collector telescopes of roughly 2.5 m aperture spaced over baselines up to about 170 m whose beams are combined so the host star's light interferes destructively, reaching the contrast of $10^{-7}$ to $10^{-6}$ needed to see a planet some 0.1 arcseconds from a Sun-like star at 10 parsecs. Around these two anchors sit three supporting mechanisms: an atmospheric retrieval framework that inverts the simulated Earth-twin spectrum into posterior distributions for key parameters; a Poisson-statistics calculation that sizes the sample a null result would need; and a Monte Carlo simulator that converts assumed planet occurrence rates, random orbits, and instrument sensitivities into the predicted 423 detections.","core_discovery":"The central claim, stated on the paper's own terms, is that spatially resolved observations of thermal emission from terrestrial exoplanets in the mid-infrared — a search phase of broad-band imaging near 5.6, 10, and 15 μm followed by low-resolution spectroscopy from roughly 3 to 20 μm — are the most powerful and the only currently viable route to the three driving science goals: detecting atmospheric biosignatures, assessing whether surface conditions allow liquid water, and mapping the diversity of terrestrial-planet atmospheres. The retrieval study shows that a single mid-infrared spectrum of an Earth-twin at spectral resolution $R=100$ and signal-to-noise 20 yields molecular abundances to about 0.5 dex, surface pressure to 0.1 dex, surface temperature to better than 5 K, and radius to a few percent, while recovering the pressure-temperature profile; methane and nitrous oxide are accessible only in the thermal infrared, ozone serves as a proxy for molecular oxygen, and carbon monoxide appears as a contextual anti-biosignature. The yield analysis predicts that a search phase of under one year of on-source time across 320 nearby FGK and M stars detects over 400 planets spanning $0.5$-$6.0\\,R_\\oplus$ and $0.1$-$1000$ times Earth's insolation, with about 30 small planets in the empirical habitable zone - the minimum sample for a statistically meaningful null result about habitability. The paper concludes that none of the currently adopted projects, from the ground or in space, can deliver this dataset, and that the technology for a mid-infrared nulling interferometer has matured to the point where a large space mission is the logical next step.","pith_inferences":["An implication the authors leave implicit is that the two observation windows are complementary rather than rivals: oxygen is seen only in reflected light while methane and nitrous oxide appear only in the mid-infrared, so the strongest roadmap may be a staged sequence in which a reflected-light survey identifies candidate targets and the mid-infrared mission delivers the definitive biosignature s","The claimed retrieval accuracy for an Earth-twin (surface temperature within 5 K, pressure within 0.1 dex) could be validated before any mission flies by running the same retrieval pipeline on Earth itself, observed as a distant point source by a mid-infrared spectrograph.","The paper notes that most of its ~320 target stars have unmeasured exozodiacal dust levels and that southern-sky targets are not covered by current surveys; a ground-based survey of the southern stars could test the assumed dust background, since higher dust levels would eat into the predicted ~30 habitable-zone detections.","The sample-sizing logic for the null result transfers to any future technique that delivers tens of temperate terrestrial planets, meaning the statistical framework stands even if the specific interferometer architecture is superseded."],"forward_implications":["A single mid-infrared spectrum of an Earth-twin at $R=100$ and $\\mathrm{SNR}=20$ lets one retrieve molecular abundances to about 0.5 dex, surface pressure to 0.1 dex, surface temperature to better than 5 K, and radius to a few percent, from the emission spectrum alone.","If none of roughly 30-50 characterized habitable-zone planets shows conditions that allow liquid water, that null result rejects the hypothesis that 50% of such planets are habitable at about $5\\sigma$ (with 30 planets) and can bound the true fraction down to 10-20% (with 50 planets).","A search phase of under three years on 320 stars within 20 pc would detect more than 400 exoplanets across $0.5$-$6\\,R_\\oplus$ and $0.1$-$1000$ times Earth's insolation, including about 30 terrestrial planets in the empirical habitable zone, enough for the null-result test.","Methane, nitrous oxide, and carbon monoxide are only detectable in the thermal infrared, and ozone provides a mid-infrared proxy for oxygen, so reflected-light surveys would miss key biosignature and anti-biosignature gases even for the few planets they can reach.","No currently selected mission or planned ground facility can assemble the required sample, so if the paper is right, answering whether Earth is unique requires building this mission."],"supporting_citations":[{"why":"Laboratory demonstration of a four-beam nulling interferometer reaching null depth $8\\times10^{-6}$ at 10 μm with 10% bandwidth and $10^{-8}$ total starlight suppression after post-processing; it is the feasibility anchor for the required planet-to-star contrast.","marker":"Martin et al. 2012"},{"why":"Reflected-light retrieval study of an Earth-twin at 0.4-1.0 μm; it supplies the direct comparison showing that mid-infrared thermal emission recovers abundances plus radius, surface temperature, and pressure that reflected light cannot.","marker":"Feng et al. 2018"},{"why":"Earlier Monte Carlo yield simulation that this paper updates; it supplies the framework and technical assumptions behind the predicted 423 exoplanet detections.","marker":"Kammerer & Quanz 2018"},{"why":"M-dwarf planet occurrence rates and radius distributions used as the input statistics for the M-star component of the 320-star yield simulation.","marker":"Dressing & Charbonneau 2015"},{"why":"Defines the empirical habitable zone used throughout the paper, via the 'Recent Venus' inner and 'Early Mars' outer insolation limits of 0.35-1.75 times Earth's insolation.","marker":"Kopparapu et al. 2013"},{"why":"Empirical mass-radius relation used to set the rocky-to-gas-planet transition and thus the radius and mass boundaries of what counts as terrestrial.","marker":"Chen & Kipping 2017"},{"why":"Survey results giving the median exozodiacal dust level around Sun-like stars (about 4.5 times the Solar System's zodiacal light) that the astrophysical feasibility of the detections hinges on.","marker":"Ertel et al. 2018"},{"why":"Recent review of interferometric exoplanet mission concepts and technology readiness; it documents the nulling demonstrations, TRL levels, and formation-flying precision requirements.","marker":"Defrère et al. 2018"},{"why":"Earth-twin thermal emission spectrum modeling on which the synthetic observation used in the retrieval study is based.","marker":"von Paris et al. 2013"},{"why":"Atlas of mid-infrared biosignature absorption bands that underpins the paper's argument that ozone, methane, and nitrous oxide are uniquely accessible in the thermal infrared.","marker":"Schwieterman et al. 2018"}],"fun_headline_variants":["One mid-IR mission could unmask 400 exoplanet atmospheres","Thermal spectra of 400 exoplanets could reveal biosignatures","Mid-IR eyes on 400 exoplanets to probe habitability and life","Rocky worlds' thermal fingerprints: 400 planets in one mission","How unique is Earth? Mid-IR mission could tell"],"cache_read_input_tokens":23168,"weakest_assumption_plain":"The load-bearing premise is an engineering one: a fleet of free-flying telescope spacecraft can hold formation to millimeter precision over baselines of roughly 150 meters while suppressing the host star's light to about one part in a million after post-processing, even though the deep-nulling and the formation-flying capabilities have so far only been demonstrated separately, not as one working system in space.","fun_headline_variants_meta":{"raw":{"variants":["One mid-IR mission could unmask 400 exoplanet atmospheres","Thermal spectra of 400 exoplanets could reveal biosignatures","Mid-IR eyes on 400 exoplanets to probe habitability and life","Rocky worlds' thermal fingerprints: 400 planets in one mission","How unique is Earth? Mid-IR mission could tell"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000761,"raw_usage":{"total_tokens":3494,"prompt_tokens":1175,"completion_tokens":2319,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":791,"completion_tokens_details":{"reasoning_tokens":2227}},"tokens_in":791,"tokens_out":2319,"duration_ms":16539,"temperature":1.0,"reasoning_tokens":2227,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:16:11.080663+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The decisive test is engineering: an integrated cryogenic demonstration in which the nulling system reaches a post-processing starlight suppression near $10^{-7}$ at 10 μm while the collector spacecraft simultaneously maintain millimeter-level formation over ~150 m baselines - if either capability falls short at the required scale, the mission concept and its yield estimates collapse. On the science side, a radial-velocity census of the proposed target stars that found temperate small planets to be much rarer than the Kepler-based occurrence rates assumed in the yield simulations would settle the yield claim directly, by showing that the projected ~30 habitable-zone planets were an overestimate.","supporting_citations":[{"cited_title":"2012, Appl","cited_arxiv_id":null,"evidence_quote":"Laboratory demonstration of a four-beam nulling interferometer reaching null depth $8\\times10^{-6}$ at 10 μm with 10% bandwidth and $10^{-8}$ total starlight suppression after post-processing; it is the feasibility anchor for the required planet-to-star contrast."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Earlier Monte Carlo yield simulation that this paper updates; it supplies the framework and technical assumptions behind the predicted 423 exoplanet detections."},{"cited_title":"2017, The Astrophysical Journal, 834, 17","cited_arxiv_id":null,"evidence_quote":"Empirical mass-radius relation used to set the rocky-to-gas-planet transition and thus the radius and mass boundaries of what counts as terrestrial."},{"cited_title":"2018, AJ, 155, 194 Event Horizon Telescope Collaboration, Akiyama, K., Alberdi, A., et al","cited_arxiv_id":null,"evidence_quote":"Survey results giving the median exozodiacal dust level around Sun-like stars (about 4.5 times the Solar System's zodiacal light) that the astrophysical feasibility of the detections hinges on."}],"review_version":1}