{"id":"0a61ee8a-f8d3-4bec-a84b-e6ac9f86f809","arxiv_id":"2505.00781","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Atom gradiometers, especially a space-based AEDGE+-like design, could probe a 10% dark matter clump subcomponent in the 10^6 to 10^10 kg window and order-10 ultralight dark matter overdensities through purely gravitational phase shifts.","lead":"This paper calculates how proposed atom gradiometers could detect dark matter through gravity alone, from heavy compact clumps to ultralight fields. It finds that a space-based AEDGE+-like detector could reach an unexplored clump mass window and be far more sensitive than laser interferometers to fast dark matter density oscillations.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Projected AEDGE+ reach hinges on the shot-noise-limited assumption; unresolved binary foregrounds sit directly in the signal band and are not included in the SNR integrals.","rationale":"The paper is internally consistent and the derivations are transparent, so I find no fatal flaw in the physics. The central claim, however, is quantitative: the AEDGE+-like reach for f_DM ~ 0.1 clumps and f_DM ~ 10 ULDM overdensities assumes atom shot-noise-limited operation across a band where the authors themselves show astrophysical foregrounds dominate. The reader's weakest assumption identifies exactly this point, and the paper explicitly defers a detailed noise analysis. My stress-test confirms this is the most load-bearing assumption: it is not a question of gauge choice or a missing factor in the signal calculation, but of whether the detector noise budget used in Eq. (25) is realistic. The parametric enhancement of atom gradiometers over laser interferometers for oscillating metric perturbations is a conceptual result that does not depend on this noise assumption, so the conditional verdict remains appropriate rather than a rejection.","tokens_in":33890,"tokens_out":31981,"duration_ms":345712,"concrete_test":"Recompute the projected upper limits from Eq. (25) for the 'space-based (outside)' configuration with the noise PSD replaced by S_n(omega) + S_fg(omega), where S_fg is the unresolved binary foreground strain converted to gradiometer phase noise using the curves in Fig. 5 (or published models such as Karnesis et al. 2021 and LVK BBH). Evaluate f_DM at m = 1e-17 eV and at clump masses 1e6 to 1e10 kg. If S_fg exceeds S_n in the signal band by more than an order of magnitude, the headline f_DM ~ 10 and f_DM ~ 0.1 curves shift by the corresponding factor; if a mitigation strategy reduces S_fg below shot noise, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline detection claims for an AEDGE+-like detector rest on the Table I assumption that the 'space-based (outside)' gradiometer is atom shot-noise limited with sqrt(S_n) = 1e-4 / sqrt(Hz) between omega_min/2pi = 1e-3 Hz (benchmark) and 10 Hz. Appendix A and Fig. 5 show that unresolved galactic and extragalactic binary foregrounds exceed the shot-noise strain curve below about 1e-2 Hz, which directly overlaps the ULDM signal at m ~ 1e-17 eV (2m ~ 4.8e-3 Hz) and the clump events with impact parameters b ~ 5e6 to 5e9 m discussed in Sec. III C. The SNR calculation in Eq. (25) uses only the shot-noise PSD S_n(omega); the astrophysical foreground is not added to the noise budget. The authors explicitly defer this in Secs. III C, IV C, and V. Therefore the projected O(0.1) clump fraction and O(10) ULDM overdensity are conditional on an unvalidated noise assumption. The parametric Einstein-term enhancement over laser interferometers is independent of this and remains supported, but the quantitative reach curves are not yet robust predictions.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript studies the purely gravitational signatures of dark matter in proposed long-baseline atom gradiometers. Using the phase-shift framework of Ref. [68], the authors decompose the gradiometer signal into Doppler, Einstein, and Shapiro contributions, and apply it to two dark matter scenarios: compact dark clumps with masses between about 10^6 and 10^10 kg, and ultralight dark matter with masses below about 10^-17 eV. For clumps, they find that the Doppler term dominates and project that an AEDGE+-like space-based gradiometer could probe a clump subcomponent at the 10% level of the local dark matter density. For ultralight dark matter, they find that the Einstein gravitational-redshift term, which is not velocity suppressed, dominates and yields a parametric enhancement over laser interferometers; they project sensitivity to overdensity factors around 10 for masses below 10^-17 eV. The calculations are presented for terrestrial (AION-km, MAGIS-km) and space-based (MAGIS-space, AEDGE, AEDGE+) concepts, with Monte Carlo and analytic estimates for the clump reach and power-averaged sensitivity curves for the ultralight case.","tokens_in":34048,"tokens_out":10488,"duration_ms":116005,"significance":"If the quantitative projections hold, this paper identifies an otherwise unexplored mass window for purely gravitational dark matter searches and makes a credible parametric argument that atom gradiometers are better suited than laser interferometers to fast-oscillating metric perturbations sourced by ultralight dark matter. The strength of the paper is that the signal derivations are internally consistent, the heuristic estimates agree with the full calculation, and the authors are unusually transparent about the assumptions behind their reach projections. The central parametric claim does not depend on the unvalidated noise assumptions; however, the headline numerical reaches for AEDGE+ do depend on them, and the paper itself repeatedly defers the needed noise analysis. The projected sensitivity curves should therefore be treated as conditional until foregrounds are included or explicitly shown to be subdominant.","major_comments":[{"comment":"The step from the Fourier-domain expressions in Eq. (36) to the amplitude formula Eq. (37) is not shown. The normalization of Eq. (37) depends on the treatment of the Rayleigh-distributed field amplitude phi_0, the Fourier convention for the delta-function peak at omega = 2m, and the combination K_+(omega) + K_-(omega). Since Eq. (37) sets the normalization of the projected reach curves in Fig. 4, please provide the intermediate derivation or explicitly state the averaging convention used for sqrt(<|Delta phi|^2>).","section":"Section IV B, Eq. (37)"}],"minor_comments":[{"comment":"There is a typo in the sentence 'Phi does not interact affect the free evolution of photons'; it should read 'does not affect'.","section":"Section IV B"},{"comment":"The table reports sqrt(S_n) in units of 1/sqrt(Hz), while Eq. (13) and related expressions use S_n in units of Hz^-1. The notation is understandable but should be made uniform to avoid confusion.","section":"Table I and Section II A"},{"comment":"The caption of Fig. 2 states SNR_t ~ 2, while the text states SNR_t^2 ~ 4 with a chi-squared threshold. This is consistent, but the caption could clarify that the threshold is on SNR^2.","section":"Section III C"},{"comment":"The abstract says 'depending on astrophysical backgrounds' while Section V says 'under optimistic noise projections'. Using one consistent qualifier would make the conditional nature of the headline claims clearer.","section":"Abstract and Section V"},{"comment":"The factor min(1, 2L/b) in Eq. (7) is stated without derivation; clarifying that it comes from the relative acceleration of two separated accelerometers in a gradient field would improve readability.","section":"Section III A, Eq. (7)"}],"recommendation":"major_revision","confidential_remarks":"The paper's main quantitative claims are conditional on a noise assumption that the authors themselves identify as unresolved. This is not a fatal flaw, and the parametric enhancement argument is well supported, but the headline reach numbers are likely to be over-quoted if the foreground issue is not addressed in the published version. I recommend requiring either an inclusion of foregrounds in the noise budget or a prominently displayed degraded-reach estimate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read it. It's a serious, carefully derived sensitivity forecast for purely gravitational DM searches with atom gradiometers. The new physics is real: a space-based AEDGE+-like detector could open a clump mass window (10^6–10^10 kg) that no other gravitational probe touches, and the paper identifies a parametric advantage—the Einstein time-delay term—that makes AGs more sensitive than laser interferometers to fast ULDM density oscillations. That last point is the strongest part; it's an argument about the measurement itself, not about noise curves, so it survives even if the projected reaches move.\n\nThe derivations in Secs. IIIB and IVB are internally consistent and match the heuristics; the velocity suppression of the Doppler term and the Shapiro cancellation in Newtonian gauge are shown explicitly. They use their own published framework (Ref. [68]), which is a general derivation, not fitted to these claims, so the circularity concern is low. The paper is also unusually transparent about what it doesn't know.\n\nThe real soft spot is the noise budget. The headline AEDGE+ reach assumes shot-noise-limited performance between 10^-6 and 10^-3 Hz, but Appendix A's Fig. 5 shows unresolved galactic and extragalactic binary foregrounds exceed that curve below about 10^-2 Hz—exactly where the ULDM signal at m~10^-17 eV and clump events with b ~ 5x10^6–10^9 m live. The SNR integral in Eq. (25) only includes Sn, and the authors explicitly leave noise analysis for future work. The stress-test concern lands. That said, they flag it in the text, including a passage admitting the constraints would weaken to f_DM~1 for clumps around 10^6 kg and to f_DM>10^3 for ULDM if foregrounds dominate. The projected O(0.1) clump fraction and O(10) ULDM overdensity are conditional, not predictions. A reader should not take those numbers as robust.\n\nThis paper deserves a serious referee. The physics derivations and the parametric enhancement argument are strong, the noise caveat is real but identified, and the mass window is genuinely new territory. I'd send it to review with the expectation that the authors strengthen the foreground discussion, possibly by adding foregrounds to the noise budget or at least making the conditional nature of the headline claims impossible to miss. For my own work, I'd cite it for the ULDM gravitational detection argument. For a reading group, yes—the two derivations and the sensitivity comparison are worth reading carefully.","headline":"Careful, transparent sensitivity forecast with a genuinely new parametric advantage for atom gradiometers; the headline reach numbers depend on an unvalidated shot-noise assumption that the paper itself flags.","tokens_in":34638,"tokens_out":1863,"would_cite":true,"duration_ms":19938,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["95.35.+d","04.80.Nn"],"model":"deepseek-v4-flash","headline":"A space-based atom gradiometer could detect dark matter clumps at about 10 percent of the local density and ultralight overdensities of ten times local, purely through gravity.","keywords":["atom gradiometer","dark matter clumps","ultralight dark matter","gravitational redshift","atom interferometry","gravitational wave detectors","matched filtering","dark matter gravitational signatures"],"falsifier":"Measure the acceleration-noise spectral density of a space-based gradiometer in the $10^{-6}$–$10^{-3}$ Hz band; if it exceeds $\\sqrt{S_a}\\sim3\\times10^{-18}\\,\\mathrm{m\\,s^{-2}/Hz^{1/2}}$, the value the projections assume, the claimed $f_{\\rm DM}$ reach weakens. A one-year run that sees no transient event above threshold for $f_{\\rm DM}\\approx0.1$ with clumps in the $10^6$–$10^{10}$ kg range would strongly disfavor the predicted 90 percent probability of at least one detectable encounter.","tokens_in":33649,"feed_emoji":"🛰️","tokens_out":10721,"duration_ms":98238,"temperature":0.7,"pith_summary":"This paper asks whether proposed long-baseline atom gradiometers—devices that compare the phases of two atom interferometers separated by a large baseline—can detect dark matter using gravity alone. It argues that a space-based gradiometer with a $4.4\\times10^7$ m baseline and a $150$ s interrogation time could see a dark matter subcomponent of about 10 percent of the local density if that component is made of compact clumps of mass $10^6$–$10^{10}$ kg. For ultralight dark matter, the paper claims the same detector could probe an overdensity of roughly ten times the local density for masses below $10^{-17}$ eV, using the oscillating gravitational field produced by density and pressure fluctuations. The key reason is that the gradiometer's observable includes the relative gravitational redshift between separated atomic clouds, a signal that is not velocity suppressed, unlike the tidal (Doppler) effect that dominates laser interferometers. If the paper is right, this opens a mass window and a detection channel—pure gravity—that no other planned probe covers.","feed_headline":"Atom gradiometers could detect dark matter through gravity alone","feed_subtitle":"A space-based design could reach clump masses of 10^6-10^10 kg and probe ultralight overdensities of about ten times local.","key_machinery":"The load-bearing object is the coordinate-invariant decomposition of the gradiometer phase shift into Doppler (tidal displacement), Einstein (gravitational redshift), and Shapiro (photon time delay) terms, with detector response kernels $K_\\pm(\\omega)$ fixed by the baseline $L$, interrogation time $T$, and number of large-momentum-transfer kicks $n$ (Eqs. (1)–(3)). For clump signals the Doppler term dominates and is computed from the Newtonian potential of a passing point mass. For ultralight dark matter the Einstein term dominates and is driven by the $h_{00}$ component of the metric, which the field's density and pressure fluctuations produce at frequency $2m$; because this term is not velocity suppressed, it survives where the Doppler response is suppressed by the dark matter velocity. The reach estimates combine these phase-shift formulas with matched filtering against a shot-noise-limited phase-noise spectrum, using the 90th-percentile minimum impact parameter for clump encounters and a coherence-time-dependent signal-to-noise treatment for ultralight dark matter.","core_discovery":"The paper's central claim is that the gravitational phase shift measured by an atom gradiometer has three physically distinct components—Doppler, Einstein, and Shapiro—and that this decomposition changes the reach for dark matter. For compact dark clumps the Doppler term, the tidal displacement of atoms along the baseline, dominates; the projected acceleration sensitivity of a space-based gradiometer then yields a 90 percent upper limit of $f_{\\rm DM}\\sim\\mathcal{O}(0.1)$ on the clump fraction for masses $10^6\\,{\\rm kg}\\lesssim M\\lesssim10^{10}\\,{\\rm kg}$. For ultralight dark matter the fast-oscillating metric perturbations from density and pressure fluctuations enter chiefly through the Einstein term, the relative gravitational redshift between the two atom clouds; unlike the Doppler term this is not suppressed by the dark matter velocity, so the gradiometer is parametrically more sensitive than a laser interferometer with comparable strain sensitivity. The paper projects that such a detector could probe an ultralight dark matter overdensity of $\\mathcal{O}(10)$ times the local density for $m\\lesssim10^{-17}$ eV. The same calculation shows that terrestrial km-scale gradiometers would need much larger overdensities to see either class of signal.","pith_inferences":["The same Einstein-term enhancement should apply to any clock-comparison measurement, not only atom gradiometers; a network of two or more space-based gradiometers could cross-correlate to extract the stochastic ultralight signal from colored noise, a direction the paper leaves open.","Because the gravitational signal is independent of the atomic species, comparing two gradiometers that use different atomic transitions could serve as a built-in check against non-gravitational systematic shifts.","If unresolved binary foregrounds are as strong as current models suggest, the headline $\\mathcal{O}(0.1)$ clump fraction and $\\mathcal{O}(10)$ ultralight overdensity claims could degrade by orders of magnitude; a direct measurement of that foreground in the $10^{-3}$–$10^{-2}$ Hz band would sharpen or overturn these projections.","Treating clumps as point-like masses sets a lower bound on the signal; extending the calculation to finite-size objects such as axion stars or dark-photon stars would change the signal when the impact parameter is comparable to the object's radius, most likely at the low-mass end of the reach."],"forward_implications":["A space-based atom gradiometer would become the first purely gravitational probe of dark clumps in the $10^6$–$10^{10}$ kg window, reaching a 10 percent subcomponent of the local dark matter density.","For ultralight dark matter below about $10^{-17}$ eV, the same detector could explore overdensities of roughly ten times the local density, a regime that planetary ephemeris and laser-ranging measurements can only reach at much larger densities.","Atom gradiometers are parametrically more sensitive than laser interferometers with comparable strain sensitivity to fast-oscillating metric perturbations, so the instrument would double as a dark matter detector and a mid-frequency gravitational wave detector.","Terrestrial km-scale designs are unlikely to reach interesting dark matter fractions for either clumps or ultralight fluctuations, so the space-based configuration carries the claimed reach.","If unresolved galactic and extragalactic binary foregrounds contaminate the $10^{-6}$–$10^{-2}$ Hz band, the projected reach degrades, though the best space-based design should still outperform a design with atoms confined inside the satellites."],"supporting_citations":[{"why":"Supplies the coordinate-invariant Doppler/Einstein/Shapiro phase-shift decomposition and the detector response kernels used for both signal calculations.","marker":"[68]"},{"why":"Provides the laser-interferometer Monte Carlo projections for compact clumps that the atom gradiometer reach is compared against.","marker":"[13]"},{"why":"Establishes the laser-interferometer clump reach that the space-based gradiometer improves on, a case the paper notes was not covered there.","marker":"[11]"},{"why":"Derives the laser-interferometer response to ultralight dark matter density fluctuations that the parametric comparison extends.","marker":"[37]"},{"why":"Supplies the minimum-impact-parameter statistics for clump encounters used in the sensitivity analysis.","marker":"[16]"},{"why":"Provides the large-momentum-transfer response kernels and transient-gravitational-wave phase-shift formulas that the signal calculations map onto.","marker":"[64]"},{"why":"Sets the space-based gradiometer parameters and the shot-noise-limited sensitivity assumptions underlying the projections.","marker":"[50]"}],"fun_headline_variants":["Atom gradiometers hunt dark matter via gravity alone","Space atom gradiometer could probe dark clumps and ultralight fields","Atom gradiometers gain edge over lasers for dark matter gravity","Gravitational dark matter detection with atom gradiometers","Atom gradiometers reveal dark matter through gravitational time delay"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The headline reach assumes the space-based gradiometer is atom shot-noise limited with a phase-noise floor of $10^{-4}/\\sqrt{\\mathrm{Hz}}$ between about $10^{-6}$ and $10^{-3}$ Hz, and that unresolved astrophysical binary signals and asteroid gravity-gradient noise do not fill that band.","fun_headline_variants_meta":{"raw":{"variants":["Atom gradiometers hunt dark matter via gravity alone","Space atom gradiometer could probe dark clumps and ultralight fields","Atom gradiometers gain edge over lasers for dark matter gravity","Gravitational dark matter detection with atom gradiometers","Atom gradiometers reveal dark matter through gravitational time delay"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00025,"raw_usage":{"total_tokens":1617,"prompt_tokens":1072,"completion_tokens":545,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":688,"completion_tokens_details":{"reasoning_tokens":460}},"tokens_in":688,"tokens_out":545,"duration_ms":5169,"temperature":1.0,"reasoning_tokens":460,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T04:35:53.679393+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the acceleration-noise spectral density of a space-based gradiometer in the $10^{-6}$–$10^{-3}$ Hz band; if it exceeds $\\sqrt{S_a}\\sim3\\times10^{-18}\\,\\mathrm{m\\,s^{-2}/Hz^{1/2}}$, the value the projections assume, the claimed $f_{\\rm DM}$ reach weakens. A one-year run that sees no transient event above threshold for $f_{\\rm DM}\\approx0.1$ with clumps in the $10^6$–$10^{10}$ kg range would strongly disfavor the predicted 90 percent probability of at least one detectable encounter.","supporting_citations":[{"cited_title":"Gravitational Wave Measurement in the Mid-Band with Atom Interferometers","cited_arxiv_id":"2309.07952","evidence_quote":"Supplies the coordinate-invariant Doppler/Einstein/Shapiro phase-shift decomposition and the detector response kernels used for both signal calculations."}],"review_version":1}