{"id":"60b008f5-36b3-45de-8312-2b3c224bfbd9","arxiv_id":"2505.02818","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Prior astrometric orbits from CHES would boost HWO's habitable-planet detection completeness by about 10% and detection efficiency by 2-30x, adding roughly 5-10 expected detections.","lead":"This paper simulates how a Chinese astrometry mission (CHES) could help a future NASA direct-imaging observatory (HWO) find Earth-like planets. It finds that knowing planet orbits beforehand raises detection completeness by about 10% and makes observing time two to thirty times more efficient.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Synergy gains assume every injected planet is already detected by CHES; the paper never applies a CHES detection gate, so the ~10% completeness gain and 5–10 added yields are upper bounds until that gate is modeled.","rationale":"The reader's weakest assumption and this stress-test concern coincide: the synergy scenario assumes perfect prior knowledge of planetary orbits. I checked whether §4.2 provides independent support for that assumption; it simulates CHES noise and MCMC retrieval, but it is a fitting exercise rather than a detection-completeness calculation. The full §5.2 population is injected and then assumed known, so the prior set is not filtered by CHES sensitivity. The 0.3 µas signal versus 1 µas total noise comparison makes this materially important: at 10 pc the per-epoch S/N is order unity, so detection requires many epochs and an explicit detection-threshold argument that the paper does not supply. The bias direction is clear—excluding undetected planets can only reduce the synergy gains—so this is a load-bearing upper-bound concern rather than a fatal flaw. Other limitations, such as HWO instrument design, occurrence-rate uncertainty, exozodi levels, and binary companions, are explicitly acknowledged in §7 and are standard planning uncertainties; they do not undermine the internal logic. The paper remains a useful conditional estimate, but the headline numbers should be reported as conditional on CHES detection completeness, which is exactly the reader's CONDITIONAL verdict. Therefore no verdict change is needed.","tokens_in":29416,"tokens_out":7734,"duration_ms":91613,"concrete_test":"Recompute the §5.2 prior-knowledge scenario with an explicit CHES detection gate: for each injected planet, including two-planet systems, simulate the CHES astrometric time series using the §4.2 noise model with σ_total ≈ 1 µas, the common-target schedules from Tan et al. 2024, and the 40-epoch uniform schedule for HWO-only targets; fit each system with the Nii-C MCMC code; and admit a planet to the known-orbit prior set only if the fit recovers its period within 20% and its predicted maximum-contrast epoch within the half-width of the contrast peak, with a false-alarm probability below 1%. Then recompute Figure 6 completeness, the benefit-to-cost ratios, and Table 3 yields.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claims are computed in §5.2 under 'assuming their orbits are already known,' but the preceding noise model in §4.2 gives CHES a total noise of roughly 1 µas per epoch, while an Earth-mass planet at 1 au around a solar-type star at 10 pc produces a 0.3 µas astrometric signal (§2.1). The orbital-retrieval exercise in §4.2 and Figure 4 fits injected planets and reports prediction quality; it does not apply a detection threshold, so it cannot establish how many of the injected planets CHES would actually find. For the full completeness and yield population, which includes planets with radii down to 0.5 R⊕ and planets at larger distances, the astrometric signal is often weaker, yet every planet is still admitted to the prior set. Consequently, the ~10% completeness gain and the 5–10 added planets in Table 3 are upper bounds conditional on perfect CHES detection, not expected values of the proposed synergy. The paper does not flag this assumption as a limitation in §7; it only notes that stellar characteristics are not fully modeled in orbital retrieval.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper simulates the synergy between the proposed CHES astrometry mission and the Habitable World Observatory (HWO) direct-imaging mission for detecting Earth-like planets around nearby stars. The authors model Keplerian orbits, Lambert-phase reflected-light contrast, a Hybrid Lyot Coronagraph response, and signal-to-noise criteria to compute detection completeness and expected planet yields for the 164-star HWO catalog. They find that prior CHES astrometry, assumed to provide known orbits, increases imaging completeness by about 10%, improves the benefit-to-cost ratio by factors of two to thirty, and adds roughly five to ten expected detections under the Bryson et al. (2021) occurrence-rate model. The paper also provides a per-star priority ranking.","tokens_in":29643,"tokens_out":5948,"duration_ms":65024,"significance":"If the stated gains are robust, the paper provides a quantitative, mission-relevant argument for coordinating CHES astrometry with HWO imaging and offers a concrete target-priority list. Its strengths include the use of standard physical models (Lambert phase function, Kepler orbits, HLC curves, Bryson occurrence rates), an end-to-end simulation pipeline based on public tools (EXOSIMS, RCETC, Nii-C, synphot), and a clear detection criterion (S/N > 7 with separation and magnitude limits). The main result, however, is conditional on the assumption that CHES has already detected every planet whose orbit is used to schedule HWO observations; the paper does not apply a CHES detection threshold to the injected population, so the headline completeness and yield gains are upper bounds rather than expected values. The orbital-retrieval exercise in Section 4.2 demonstrates phase-prediction accuracy only for Earth-mass planets at the center of the habitable zone and does not assess whether CHES would detect the fainter, smaller, or more distant planets in the completeness sample. This is a load-bearing caveat that needs to be either modeled or explicitly stated as a limitation.","major_comments":[{"comment":"The synergy scenario assumes that the orbits of all injected planets are already known ('assuming their orbits are already known'), but the preceding orbital-retrieval analysis in §4.2 is performed only for Earth-mass planets at the center of the habitable zone and does not apply a detection threshold. Given the stated CHES total noise of approximately 1 µas per epoch and an astrometric signal of 0.3 µas for an Earth-mass planet at 1 au around a solar-type star at 10 pc, a substantial fraction of the injected population (radii down to 0.5 R⊕, distances beyond 10 pc, and eccentric orbits) would likely not be detected by CHES. Consequently, the approximately 10% completeness gain and the five-to-ten added planets in Table 3 are conditional upper bounds, not expected values. The authors should either apply a realistic CHES detection gate in the Monte Carlo completeness and yield calculations, or explicitly reframe all synergy gains as upper limits and acknowledge this in Section 7.","section":"§5.2, Fig. 6, Table 3"},{"comment":"The abstract states that the synergy yields 'approximately 37 and 47 planets' in the conservative and optimistic habitable zones under the low-occurrence model, but Table 3 reports 41.78 and 54.33 for imaging plus astrometry, and 37.11 and 46.80 for imaging alone. The numbers quoted in the abstract do not match the table; the sentence appears to mix the imaging-only conservative value with an incorrect optimistic value. This inconsistency should be corrected so the abstract faithfully represents Table 3.","section":"Abstract vs. Table 3"},{"comment":"The benefit-to-cost comparison mixes different observing strategies: the imaging-only scenario uses optimized multi-epoch revisits (Nv up to 6), while the astrometry-assisted scenario uses a single observation at the predicted maximum-contrast phase. The reported efficiency gain of two to thirty therefore conflates the value of orbit knowledge with the reduction in visit count and does not share a common time budget between scenarios. This is not necessarily an error, but it should be stated explicitly so the efficiency factor is not over-interpreted as a pure sensitivity gain.","section":"§5.2, Eq. (18)"}],"minor_comments":[{"comment":"The sentence 'The detection completeness Cdet is defined as which is the fraction of possible planets that will be detected' is grammatically incomplete and should be rephrased.","section":"§5.1"},{"comment":"The text says 'standard derivations of 0.36 and 0.74 µas' but should read 'standard deviations'; also the noise components are described as Gaussian, which is reasonable but should be stated as an assumption.","section":"§4.2"},{"comment":"In the sentence 'we construct girds over Rp and a' the word 'girds' should be 'grids'; similar typos appear elsewhere (e.g., 'the parameters combination', 'such system are').","section":"§6"},{"comment":"The Kepler equation block is garbled in the text: the line 'E:' and the expression for tan(f/2) are missing their proper equation formatting and the connecting relation between E and the mean anomaly is unclear.","section":"Eq. (2)"},{"comment":"The paper states that the Bryson et al. (2021) parameters are taken from the Poisson likelihood fit, but it does not list the adopted values of F0, C0, α, β, γ, or the form of g(Teff). Providing these values (or an equation reference with the specific numbers) would improve reproducibility.","section":"§5.3"}],"recommendation":"major_revision","confidential_remarks":"The paper is a mission-synergy feasibility study whose central quantitative claims rest on the known-orbit assumption. The CHES detection-gate issue is fixable within the manuscript's scope: either add a detection filter to the injection pipeline or reframe the gains as upper bounds. The abstract/Table 3 mismatch is a straightforward correction. The manuscript also appears not fully polished (garbled equations, typos), which suggests it is a draft; the authors should be asked to clean it up. The heavy reliance on the authors' own CHES papers for noise and scheduling models is reasonable for a mission-specific study, but the reviewers may wish to see the retrieval-quality results benchmarked against an independent astrometric retrieval code."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a solid, clearly-scoped simulation study. It shows that if CHES provides accurate orbits in advance, HWO's completeness goes up by about 10%, detection efficiency by 2–30x, and habitable-planet yield by 5–10. The general point isn't new, but the CHES/HWO-specific numbers, per-target completeness maps, two-planet scheduling, and priority rankings are. I think the reader's conditional verdict is about right.\n\nWhat's good: the simulation is internally consistent and grounded in standard models—Lambert phase, Kepler orbits, HLC coronagraph curves, Bryson occurrence rates. The orbital retrieval in §4.2 demonstrates the contrast prediction concept. The benefit-to-cost definition and the priority list are genuinely useful for mission planning. The paper is honest about instrument uncertainties.\n\nWhere it's soft: the stress-test note is correct. Section 5.2 explicitly says \"assuming their orbits are already known,\" but the paper never applies a CHES detection gate. With CHES total noise ~1 µas and an Earth-mass planet at 10 pc giving only 0.3 µas, many injected planets—especially smaller ones and more distant targets—would not actually be detected by CHES. So the claimed completeness gain and yield increase are upper bounds, not expected values. This is a real caveat, but not fatal, because the assumption is stated. The abstract and conclusions should carry the caveat more prominently. Minor: no code released and no Monte Carlo error bars, which would help reproducibility.\n\nI disagree with the stress-test only in emphasis: this is a planning tool, not an empirical claim. The upper-bound nature is worth one paragraph in the discussion, not a blocker.\n\nWho should read it: anyone working on HWO target selection or astrometry-imaging synergy. It deserves a serious referee. I'd accept it with a request to model CHES detection completeness or at least quantify the impact of a detection threshold on the claimed yields.","headline":"A credible, clearly-scoped simulation showing CHES prior astrometry can modestly boost HWO completeness and yield—but the headline numbers assume CHES already detects every injected planet.","tokens_in":30194,"tokens_out":2607,"would_cite":false,"duration_ms":27395,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Astrometric foreknowledge can lift direct-imaging yields of habitable planets by five to ten.","keywords":["astrometry","direct imaging","habitable planets","exoplanet detection completeness","CHES mission","HWO mission","yield simulation","Kepler occurrence rates"],"falsifier":"An end-to-end simulation that passes the actual CHES astrometric time series, including its roughly 1 microarcsecond Gaussian and stellar-activity noise, through the orbital retrieval code, and only feeds successfully recovered orbits into the HWO scheduler, would settle whether the claimed completeness and yield gains hold; the present paper instead assumes those orbits are already known and does not report the recovery fraction for these specific targets.","tokens_in":29201,"feed_emoji":"🔭","tokens_out":10289,"duration_ms":102690,"temperature":0.7,"pith_summary":"This paper tries to establish that running a micro-arcsecond astrometry mission before a direct-imaging flagship can make the imaging of Earth-like planets substantially more efficient, and it quantifies the gain for two concrete planned missions, CHES and HWO. The authors simulate CHES measuring stellar wobbles to fix the orbits of injected planets, then use those orbits to schedule HWO observations at the moment each planet reflects the most starlight and is outside the coronagraph's inner working angle. Under their adopted detection limit, prior astrometric knowledge improves detection completeness by roughly 10 percent and detection efficiency by factors of two to thirty, and it raises the predicted number of detectable habitable planets by five to ten, to about 37 planets in the conservative habitable zone and 47 in the optimistic one under the low-occurrence Kepler model. If correct, the result argues that precursor astrometry is a cheap way to buy yield from a future imager without changing the imager itself. The paper closes by converting the simulations into a priority ranking of HWO target stars.","feed_headline":"Astrometry before direct imaging adds 5–10 habitable planets","feed_subtitle":"Simulated CHES astrometry lets HWO spot Earth-like planets with up to 30x less observing time.","key_machinery":"The machinery is the phase-dependent reflected-light contrast $C = A_g \\phi(\\beta) (R_p/r)^2$ with the Lambert phase function, combined with Keplerian orbit propagation. CHES astrometry constrains the orbital elements; from those elements the paper predicts the time $t_p$ at which a planet reaches maximum contrast while remaining outside the inner working angle, and HWO observes only at those times. Detection completeness is computed by injecting planet populations, binning radius versus semi-major axis, and applying three criteria — angular separation within the working angles, magnitude limit, and $S/N > 7$ — with a benefit-to-cost ratio $f = (S/N)\\,C_{\\mathrm{det}}/(N_v \\tau)$ that measures signal and completeness per unit observing time. For two-planet systems the scheduler compares simultaneous and sequential visits to maximize this ratio. Expected yields are obtained by multiplying per-cell completeness by the adopted Kepler-based occurrence-rate model.","core_discovery":"The central discovery is that known orbits are the dominant lever in direct-imaging detection of habitable planets. Treating CHES and HWO as prototypes, the paper injects one- and two-planet systems around 164 nearby solar-type stars, retrieves their orbits from simulated CHES astrometry with a realistic micro-arcsecond noise budget, and computes whether an HWO-like coronagraph would detect them under three criteria: angular separation between the inner and outer working angles, apparent magnitude brighter than the limit, and signal-to-noise ratio above 7. With prior astrometry, observations are scheduled at the phase of peak reflected-light contrast, and the paper finds completeness rises by about 10% for most targets, the benefit-to-cost ratio improves by a factor of two to thirty, and expected yields grow by 5 to 10 planets, from roughly 37 to 42 in the conservative habitable zone and from roughly 47 to 54 in the optimistic zone under the low-occurrence model. The gains are largest for planets near 1 au, where astrometric signals are strongest, and the efficiency gains are largest for the nearest common targets.","pith_inferences":["A testable extension of this logic is that the yield gain should scale monotonically with astrometric precision: re-running the pipeline with higher-noise astrometry would show the added planets shrinking toward zero, which would confirm that orbital knowledge, not the mere existence of a precursor, drives the effect.","The quoted gain of five to ten added planets should be read as an optimistic bound, because the paper credits CHES with perfect knowledge of every injected orbit; folding in the actual CHES detection and retrieval success rate would likely push the realized gain toward the lower end.","The same scheduling principle could be transferred to starshade imagers or to narrower-band photometry, where the phase dependence of reflected light is different; the target ranking would then shift, offering a direct way to prioritize which astrometric orbits are worth measuring first.","From a mission-planning perspective, these results suggest that a relatively small astrometry mission flown before a flagship imager may buy yield more cheaply than increasing the imager's aperture or contrast, since the gain here comes from scheduling rather than hardware."],"forward_implications":["If CHES observes before HWO, the imager can time each target to near-maximum reflected-light contrast, reducing required exposure time by factors of two to thirty for the same signal-to-noise.","Expected habitable-planet yield under the low-occurrence model rises by five to ten planets, from about 37 to 42 in the conservative habitable zone and from about 47 to 54 in the optimistic zone.","Astrometric priors add roughly 10% completeness, recovering some smaller and closer-in planets that imaging alone misses, with the largest completeness gains appearing for more distant stars.","The efficiency gains are largest for the nearest stars shared by both missions, so precursor astrometry changes the priority ordering of HWO targets, not just the total yield."],"supporting_citations":[{"why":"Defines the CHES mission concept, telescope, noise sources, and astrometric precision that the retrieval simulation relies on.","marker":"Ji et al. 2022"},{"why":"Supplies the CHES observation schedules, exposure times, and orbital retrieval method used to predict optimal imaging phases.","marker":"Tan et al. 2024"},{"why":"Provides the HWO reference design parameters, including aperture, contrast, and magnitude limit, used in the imaging model.","marker":"Mamajek & Stapelfeldt 2024"},{"why":"Provides the 164-star HWO target catalogue used for all completeness and yield simulations.","marker":"Harada et al. 2024"},{"why":"Supplies the completeness optimization framework with revisit counts, intervals, exposure times, and the benefit-to-cost ratio that the paper extends.","marker":"Stark et al. 2015"},{"why":"Defines the conservative and optimistic habitable zone boundaries used for planet injection and yield calculations.","marker":"Kopparapu et al. 2014"},{"why":"Supplies the Kepler-derived occurrence-rate model from which expected yields are computed.","marker":"Bryson et al. 2021"},{"why":"Provides the signal-to-noise model including residual starlight, zodiacal light, and exozodiacal light noise.","marker":"Stark et al. 2014"},{"why":"Gives the idealized perfect-precursor-knowledge scenario that this paper compares against and extends to multi-planet systems.","marker":"Plavchan et al. 2024"},{"why":"Provides the earlier result that prior radial-velocity knowledge raises imaging yields by about 30%, serving as the benchmark for the claimed improvement.","marker":"Morgan et al. 2021"}],"fun_headline_variants":["CHES astrometry makes HWO imaging 2-30x more efficient","Orbital data from CHES adds 5-10 habitable planets for HWO","Astrometry first: HWO finds Earth-like planets up to 30x faster","Prior CHES orbits boost HWO planet yield and cut observing time"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that every injected planet's orbit is already known from CHES astrometry before the imaging schedule is set; if CHES's roughly 1 microarcsecond total noise cannot actually detect and retrieve the orbits of the faintest Earth analogues, whose astrometric signal is only about 0.3 microarcseconds at 10 parsecs, then the claimed 10% completeness gain and the five to ten added yields would be smaller.","fun_headline_variants_meta":{"raw":{"variants":["CHES astrometry makes HWO imaging 2-30x more efficient","Orbital data from CHES adds 5-10 habitable planets for HWO","Astrometry first: HWO finds Earth-like planets up to 30x faster","Prior CHES orbits boost HWO planet yield and cut observing time"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000863,"raw_usage":{"total_tokens":3769,"prompt_tokens":999,"completion_tokens":2770,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":615,"completion_tokens_details":{"reasoning_tokens":2684}},"tokens_in":615,"tokens_out":2770,"duration_ms":21264,"temperature":1.0,"reasoning_tokens":2684,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:40:16.131117+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"An end-to-end simulation that passes the actual CHES astrometric time series, including its roughly 1 microarcsecond Gaussian and stellar-activity noise, through the orbital retrieval code, and only feeds successfully recovered orbits into the HWO scheduler, would settle whether the claimed completeness and yield gains hold; the present paper instead assumes those orbits are already known and does not report the recovery fraction for these specific targets.","supporting_citations":[],"review_version":1}