{"id":"b990d7d3-5601-4d84-ab04-5960d797650e","arxiv_id":"2512.13828","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"For LEO downlinks, aperture averaging meaningfully suppresses turbulence-induced power fluctuations, while for MEO links it gives little benefit and quantum state tomography requires much larger apertures or photon budgets.","lead":"This paper models how much light from low- and medium-Earth-orbit satellites reaches a ground telescope, including losses from atmospheric absorption, beam spreading, and turbulence. It then simulates how well a ground station could reconstruct quantum states transmitted from orbit, showing that low-orbit links are far more forgiving than medium-orbit links.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Aperture-averaging model uses total satellite distance instead of effective turbulent-layer distance, artificially suppressing MEO averaging; core LEO/MEO contrast likely an artifact of Eq. (18).","rationale":"The reader's weakest assumption identifies the aperture-averaging formulas as a source of uncertainty, but not the specific mis-specification of the propagation distance L. My concern is more pointed: Eq. (18) uses the total satellite-to-ground distance, whereas for a downlink the intensity correlation scale at the receiver is governed by the distance to the turbulent layer (typically the first ~10–20 km of the atmosphere). The paper itself acknowledges Giggenbach's alternative (Eqs. 19–21) and states that it shows strong agreement with measurements and outperforms more general approximations, but then chooses Andrews' formula solely for simplicity. This choice directly produces the paper's central result that aperture averaging is nearly ineffective for MEO. A correct treatment using L' ≈ 12 km at zenith yields strong aperture averaging for MEO as well, eliminating the claimed LEO/MEO contrast. Because this concerns the paper's main physical conclusion and not just a quantitative correction, the current version's central claim is not supported. The QST framework and link-budget methodology may remain useful, but the headline aperture-averaging result would require major revision, hence REJECT rather than CONDITIONAL. I have not based this on external consensus but on the internal inconsistency between the presented alternative model and the one chosen for simulations, plus the standard physics of Fresnel zones in downlink scintillation.","tokens_in":21074,"tokens_out":8240,"duration_ms":434573,"concrete_test":"Recompute Figs. 2 and 4 using the Giggenbach aperture-averaging factor Eq. (20) with L'(θ) from Eq. (21), using the same λ = 1550 nm, H-V profile, and telescope diameters D = 0.25/0.5/1.0/1.5 m. Also compute the PSI via Yura's path-integrated Eq. (23) with h_s from Eq. (24) over the H-V C_n² profile. If the MEO PSI at D = 1 m, ζ = 0° drops below 0.1 (or becomes comparable to the LEO value) instead of ≈0.98, the claimed LEO/MEO aperture-averaging contrast collapses. Check whether the MEO QST photon budgets in Fig. 6 change by more than an order of magnitude.","verdict_should_be":"REJECT","load_bearing_attack":"The central LEO/MEO contrast in Section III.A rests on Eq. (18), where the Andrews aperture-averaging factor Av_A(D) = [1 + 1.062(kD²/4L)]^(-7/6) is evaluated with L equal to the total slant range (420 km for LEO, 20,200 km for MEO). This is a horizontal-path formula: for a satellite downlink, the relevant Fresnel zone size at the receiver is set by the distance to the dominant turbulent layer near the ground, not by the satellite distance. Using the full slant range makes the Fresnel zone enormous (√(λL) ≈ 5.6 m for MEO at 1550 nm), so any practical aperture looks 'small' and averaging appears negligible. The paper itself introduces the Giggenbach model (Eqs. 19–21) with L' ≈ 12 km at zenith and states that it 'outperforms more general all-regime approximation functions,' yet discards it for simplicity. Replacing L by L' would make MEO aperture averaging comparable to LEO: for D = 1 m, ζ = 0, Av_G ≈ 0.01–0.02 instead of Av_A ≈ 0.98. The conclusion that aperture averaging is 'nearly ineffective for MEO' is thus an artifact of the chosen propagation distance, not a physical result. This is load-bearing because the paper's headline claim—that aperture averaging substantially suppresses LEO but not MEO fluctuations—and the associated design guidance for MEO QKD/QST would be reversed or strongly weakened under the more appropriate model.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a multiplicative transmittance model for satellite-to-ground optical links, combining internal receiver efficiency, atmospheric absorption/scattering along a slant path, diffraction loss, and turbulence-induced intensity fluctuations. Using the Hufnagel–Valley C_n^2 profile, the Rytov scintillation index, and the Andrews aperture-averaging factor, it simulates photon loss versus zenith angle for a LEO (420 km) and a MEO (20,200 km) satellite. It then feeds the resulting mean link transmittance into a quantum-state-tomography (QST) simulation with SIC-POVMs and Poisson shot noise, reporting average reconstruction fidelities for different apertures and photon budgets. The central claim is that aperture averaging is effective for LEO downlinks but nearly ineffective for MEO downlinks, so that MEO QST requires orders-of-magnitude larger photon budgets (about 10^7–10^9 photons) than LEO (about 10^4–10^6).","tokens_in":21510,"tokens_out":8563,"duration_ms":79665,"significance":"If correct, the numerical estimates would provide useful design guidance for satellite QKD/QST demonstrations, particularly for choosing receiver aperture and photon budgets as a function of orbital altitude and zenith angle. The paper's assembly of standard propagation formulas is mostly transparent, the parameter set in Table I is explicit, and the effective-pass-time analysis in Fig. 1 is a useful practical addition. The QST use case is also a sensible diagnostic framing. However, the central LEO-vs-MEO aperture-averaging contrast rests on a propagation-distance choice that the paper itself shows is questionable, and the QST simulations as written do not appear to include turbulent transmittance fluctuations. These issues are load-bearing and currently prevent the quantitative conclusions from being accepted.","major_comments":[{"comment":"The central LEO-vs-MEO contrast is an artifact of using the total slant range L in the Andrews aperture-averaging factor. Eq. (18) is a horizontal-path expression in which the Fresnel zone is evaluated at the total distance L; for a satellite downlink the intensity-correlation scale at the receiver is set by the dominant turbulent layers near the ground/tropopause, not by the satellite distance. The paper itself introduces the Giggenbach model in Eqs. (19)–(21), with L' ≈ 12 km at zenith, and states that it 'outperforms more general all-regime approximation functions', yet discards it for simplicity. Replacing L by L' changes the results dramatically: for D = 1 m, λ = 1550 nm and ζ = 0, Eq. (20) gives Av_G ≈ [1 + 1.062 × (4.05×10^6 m^{-1} × 1 m^2)/(9 × 1.2×10^4 m)]^{-7/6} ≈ 0.014, whereas Eq. (18) with L = 20,200 km gives Av_A ≈ 0.94. Thus a 1 m aperture would average MEO fluctuations al","section":"Section III.A, Eq. (18)"},{"comment":"The QST simulations do not appear to include turbulence-induced intensity fluctuations. The effective photon number is set to eN = ηN with η from Eq. (1), which includes the stochastic factor I, but the simulation then samples m_k ~ Pois(n_k) with n_k determined by a single mean transmittance; no draw from the log-normal distribution in Eq. (36) is described. As written, Figs. 5 and 6 quantify only shot noise under a deterministic mean loss, not the effect of atmospheric turbulence on QST fidelity. Statements such as 'turbulence amplifies the poor SNR' in the MEO case are therefore unsubstantiated by the simulation. The authors should either sample the transmittance I (with σ_j^2 = σ_I^2 or σ_P^2) and rerun the QST, or explicitly restrict their claims to tomography under mean-link losses and remove the turbulence-related interpretation.","section":"Section IV.A, Eqs. (37)–(38)"}],"minor_comments":[{"comment":"The text states h0 = 6600 km, but the numerical value exp(−α0 h0) ≈ 0.9675 requires h0 ≈ 6.6 km. The 'km' unit is a typo; if h0 = 6600 km were used in Eq. (5), atmospheric extinction would be negligible. Please correct and ensure Eq. (5) uses the same scale height.","section":"Section II.B, Eq. (4)"},{"comment":"The attenuation coefficient α0 = 5×10^-6 m^-1 is quoted for λ = 800 nm, but all simulations use λ = 1550 nm. Since γ(h) is wavelength-dependent, please either state the 1550 nm attenuation coefficient or justify using the 800 nm value in the link-budget calculations.","section":"Section II.B / Table I"},{"comment":"The variable z is used both as the slant-path coordinate and, implicitly, as altitude in C_n^2(z). Please clarify the integration variable and its relation to height above sea level to avoid ambiguity in the path integral.","section":"Section II.D.2, Eq. (13)"},{"comment":"The notation 'eN = ηN' is used without specifying whether η is the random transmittance or its mean. Given the earlier definition of I, please define whether eN is the expected received photon number or an instantaneous one; this also relates to the major comment above.","section":"Section IV.A, Eq. (38)"},{"comment":"The conflict-of-interest statement reads 'The author declares...' but the paper has three authors. Please change to 'The authors declare...'.","section":"Conflict of Interest"},{"comment":"References [55]–[57] appear not to be cited in the text; please add citations or remove them from the bibliography.","section":"References"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read: the paper is a competent, tutorial-style link-budget study with a quantum tomography use case that makes for a useful reference scan; the headline claim — aperture averaging suppresses LEO turbulence noise but is nearly useless for MEO — is very likely an artifact of applying the Andrews aperture-averaging formula with the total slant range instead of the distance to the dominant turbulent layer.\n\nWhat the paper does well: it carefully assembles the standard Beer-Lambert, Gaussian-diffraction, Rytov-scintillation, and log-normal framework, normalizes the turbulence factor correctly, and gives a rare clear comparison of the three aperture-averaging models (Andrews, Giggenbach, Yura). The parameterized LEO/MEO scan is genuinely handy for anyone planning a demonstrator, and the QST-as-diagnostic wrapper produces concrete photon budgets — roughly 10^4–10^6 photons for LEO, 10^7–10^9 for MEO. Those budgets are driven by geometric loss and are approximately right, so the QST feasibility conclusions mostly survive even if the turbulence part is wrong.\n\nThe soft spot is load-bearing. Eq. (18), Av_A(D) = [1 + 1.062(kD^2/4L)]^(-7/6), is a horizontal-path result, but the paper evaluates it with L equal to the full satellite range. For a downlink, the intensity correlation scale at the receiver is set by the distance to the turbulent layer, a few kilometers, not 20,200 km. The authors themselves introduce Giggenbach's L' ≈ 12 km (Eqs. 19–21), note it outperforms all-regime approximations, and then drop it for Andrews. Plugging L' into the same expression gives, for a 1 m aperture at MEO, Av ≈ 0.02 rather than ≈ 0.98 — i.e., strong aperture averaging, comparable to LEO. This also sits at odds with measured GEO downlink scintillation, which shows strong averaging on sub-meter apertures. So the central LEO/MEO contrast in Section III.A, and the design advice that follows, needs to be redone with the Giggenbach or Yura model the paper already cites.\n\nMinor issues, in proportion: h0 is printed as 6600 km where the arithmetic needs 6.6 km; alpha0 is quoted for 800 nm while the simulations run at 1550 nm; the H_OGS-dependent term added to the Hufnagel-Valley profile is underived and shifts ground-level C_n^2 by roughly a factor of 1.7; and the QST section never states whether eta is sampled per measurement from the log-normal distribution or used as its mean. The overlap with the authors' own [25] and [11] is substantial — the new material is the scan itself, the model comparison, and the QST wrapper.\n\nRecommendation: send it to a serious referee. It's a useful reference paper, but as it stands the aperture-averaging guidance is unsound, and that should be caught before anyone builds a budget on it. Major revision asked for: redo Section III.A with a downlink-appropriate formula, fix the typos, clarify the simulation, and moderate the novelty claims.","headline":"A competent, tutorial-style link-budget scan with a useful QST diagnostic, but the headline LEO/MEO aperture-averaging contrast is very likely an artifact of applying Andrews' formula with total slant range instead of the distance to the dominant turbulent layer.","tokens_in":22039,"tokens_out":9267,"would_cite":false,"duration_ms":87712,"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 establishes a multiplicative link-budget model for satellite-to-ground optical downlinks in which aperture averaging strongly suppresses turbulence-induced photon-loss fluctuations for LEO missions but barely helps MEO missions,","keywords":["free-space optics","optical downlink","link budget","atmospheric turbulence","aperture averaging","scintillation","quantum state tomography","satellite quantum communication"],"falsifier":"A direct falsifier is an experimental satellite-to-ground downlink at LEO and MEO altitudes that measures the power scintillation index and the variance of received photon counts for a range of aperture diameters; if for a MEO link a 1.0 m aperture reduces the scintillation variance by a factor close to the Andrews prediction (which for 20,200 km gives Av_A ≈ 1 even for 1.5 m), the model holds, but if strong turbulence or non-Kolmogorov effects yield a different dependence on D/L, the specific predictions of Eq. (18) would fail.","tokens_in":20912,"feed_emoji":"🛰️","tokens_out":1623,"duration_ms":18042,"temperature":0.7,"pith_summary":"The paper is trying to establish a practical, quantitative method for predicting photon loss in free-space optical downlinks from LEO and MEO satellites, with turbulence treated as a multiplicative random factor whose fluctuations can be mitigated by aperture averaging. It applies the model to a quantum state tomography use case to show how many received photons are needed to reconstruct quantum states with high fidelity. The central quantitative claim is that aperture averaging is highly effective for LEO (420 km) but nearly ineffective for MEO (20,200 km), so that LEO tomography needs roughly 10^4–10^6 photons while MEO tomography requires 10^7–10^9. A sympathetic reader would care because these numbers provide concrete design targets for satellite QKD, quantum networking, and optical communication demonstrators.","feed_headline":"Aperture averaging tames LEO turbulence but barely helps MEO","feed_subtitle":"A new link-budget model maps photon loss to quantum tomography fidelity, setting photon budgets of 10^4-10^6 for LEO vs 10^7-10^9 for MEO.","key_machinery":"The central object is the multiplicative transmittance model of Eq. (1), where the turbulence-induced random factor I is described by a log-normal distribution whose variance is the intensity scintillation index (ISI) σ²_I, computed from the Rytov variance using the Hufnagel–Valley C_n² profile, and where aperture averaging is introduced through the power scintillation index (PSI) σ²_P = Av_A(D) σ²_I with Av_A(D) = [1 + 1.062 (kD²/4L)]^(−7/6). This aperture-averaging factor carries the argument: it depends only on the ratio D²/L, which is what makes it effective for LEO (short L) and ineffective for MEO (long L). The QST use case then connects this machinery to quantum performance by replaci","core_discovery":"The central claim is that the total downlink transmittance factorizes into independent contributions, η = η_int η_atm(ζ) η_d(z) I, where η_int is internal loss, η_atm is Beer–Lambert absorption/scattering, η_d is diffraction-limited geometric collection, and I is a log-normal turbulence factor with mean unity. Using the Hufnagel–Valley turbulence profile and the Andrews aperture-averaging factor Av_A(D) = [1 + 1.062 (kD²/4L)]^(−7/6), the paper finds that for LEO downlinks the power scintillation index (PSI) is strongly reduced by aperture averaging, smoothing the photon-loss curves and stabilizing received power, whereas for MEO downlinks the same apertures yield only marginal averaging beca","pith_inferences":["The paper's separation of losses into independent multiplicative factors suggests a modular design methodology: one could validate each term (atmosphere, diffraction, turbulence) separately with dedicated experiments before combining them, which is not stated but follows from the factorization.","The near-ineffectiveness of aperture averaging for MEO implies that for MEO quantum links, the turbulence term remains nearly at the ISI level, meaning that time-correlated intensity fluctuations could introduce correlated errors in quantum protocols — an extension the paper does not explicitly explore.","The QST use case could be extended to entanglement-based protocols: since the fidelity of a single-photon reconstruction degrades with photon loss, entangled-state distribution over MEO would likely face even stricter constraints, as the paper only addresses single-photon polarization states.","The Hufnagel–Valley profile with fixed parameters (v_rms = 26.25 m/s, C0 = 1.7×10^-14) is a single representative snapshot; a testable extension would be to run the same model under a range of C_n² profiles (e.g., different seasons or site altitudes) to map the spread in required photon budgets."],"forward_implications":["If the model is correct, LEO optical downlink systems can stabilize received power and enable high-fidelity quantum state tomography with moderate telescopes (50–100 cm) and photon budgets around 10^4–10^6, while MEO systems need much larger apertures or photon budgets of 10^7–10^9.","The results imply that aperture averaging should be a primary design lever for LEO links but is insufficient for MEO links, where adaptive optics or advanced signal processing would be needed to mitigate turbulence-induced fluctuations.","The fidelity curves as a function of zenith angle provide a practical scheduling constraint: quantum communication and tomography windows should be limited to zenith angles where the effective photon number stays above the required threshold.","The model offers a benchmark for comparing theoretical lower bounds of photon loss with experimental values, which tend to be higher, helping to identify missing loss factors in real satellite missions.","The distinction between ISI and PSI quantifies how much turbulence-induced variance can be suppressed by aperture size, offering a direct design trade-off between telescope diameter and transmitted photon number."],"fun_headline_variants":["For LEO, apertures beat turbulence; MEO sees little gain","Link budget model ties turbulence to quantum tomography","LEO gains from aperture averaging, MEO barely benefits","Quantum downlink budgets: 10^4-10^6 photons for LEO, 10^7-10^9 for MEO","Turbulence factorized: LEO smoothed, MEO not by apertures"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The quantitative results rest on the assumed validity of the Hufnagel–Valley C_n² profile and the weak-turbulence Rytov/aperture-averaging formulas applied at zenith angles up to 80° and for paths up to 20,200 km; if the turbulence profile is unrepresentative or strong-fluctuation effects set in, the computed scintillation indices, the LEO/MEO contrast, and the QST fidelity curves all shift.","fun_headline_variants_meta":{"raw":{"variants":["For LEO, apertures beat turbulence; MEO sees little gain","Link budget model ties turbulence to quantum tomography","LEO gains from aperture averaging, MEO barely benefits","Quantum downlink budgets: 10^4-10^6 photons for LEO, 10^7-10^9 for MEO","Turbulence factorized: LEO smoothed, MEO not by apertures"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000134,"raw_usage":{"total_tokens":1006,"prompt_tokens":801,"completion_tokens":205,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":545,"completion_tokens_details":{"reasoning_tokens":115}},"tokens_in":545,"tokens_out":205,"duration_ms":3439,"temperature":1.0,"reasoning_tokens":115,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T06:29:45.350809+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct falsifier is an experimental satellite-to-ground downlink at LEO and MEO altitudes that measures the power scintillation index and the variance of received photon counts for a range of aperture diameters; if for a MEO link a 1.0 m aperture reduces the scintillation variance by a factor close to the Andrews prediction (which for 20,200 km gives Av_A ≈ 1 even for 1.5 m), the model holds, but if strong turbulence or non-Kolmogorov effects yield a different dependence on D/L, the specific predictions of Eq. (18) would fail.","supporting_citations":[],"review_version":1}