{"id":"054e5ebf-b2c2-41f2-b6c9-1ff1d4329376","arxiv_id":"2506.01733","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"First Crab Nebula observations with the SST-1M stereoscopic system validate Monte Carlo performance predictions at 510 m altitude, giving energy thresholds near 1 TeV and angular resolution down to 0.10 degrees.","lead":"Two prototype gamma-ray telescopes in the Czech Republic measured the Crab Nebula at very high energies from a low-altitude site. The observation matches detailed simulations and shows the stereoscopic system reaches the multi-TeV sensitivity expected from its design.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The transferred daytime V_AOD from Košetice is the most load-bearing assumption; an unrepresentative aerosol column shifts the energy scale by ~10% per 0.1 V_AOD and directly scales the Crab flux normalization, so the MC-validation claim rests on it.","rationale":"The paper is a careful commissioning study with strong internal consistency: the data/MC comparisons in Figs. 15, 18, 19, and 20 are shown without normalization, the Crab SED agrees with external instruments, and the systematic budget is unusually thorough. The residual near-threshold discrepancies and the background-statistics limitations are acknowledged and affect the SED only weakly, since the spectral fit starts at 2.5 TeV and the background systematics are statistically inconclusive but small. The most load-bearing element is the atmospheric aerosol model: the entire energy scale, and therefore the flux normalization of the central validation, is tied to a single V_AOD value taken from a remote daytime site. The paper itself quantifies the sensitivity (10% energy shift per 0.1 V_AOD) and lists this as the dominant energy-scale systematic, but it does not demonstrate that the transferred value is representative of the nights used. This is a testable assumption, and the test above uses data already shown in the paper. The reader's weakest_assumption identifies the same point, and I agree; the verdict remains CONDITIONAL pending that test, so no change is recommended.","tokens_in":31680,"tokens_out":6673,"duration_ms":77286,"concrete_test":"Compute, for each night in the Crab sample, the airmass-corrected rate of events with intensity >200 p.e. (the quantity shown in Fig. 14) and compare it with the same rate predicted by sim_telarray MC generated at V_AOD = 0.0, 0.05, 0.1, and 0.2, interpolating to find the best-fit V_AOD per night. If the best-fit values do not cluster around 0.05 and do not track the Košetice daytime V_AOD, the remote daytime transfer is invalid and the 5% energy-scale systematics are underestimated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the MC model reliably describes the low-altitude response depends on the atmospheric model used to generate the IRFs. Section 2.4 fixes V_AOD = 0.05 for the entire campaign based on daytime sun-photometer data from Košetice, 45 km away, with no on-site nighttime validation. Section 8.1 states that a 0.1 change in V_AOD shifts the light/energy scale by about 10%; Table 2 converts the assumed 0.05 uncertainty into 5% energy-scale systematics and 18-19% flux-normalization systematics. If the transferred column is wrong by 0.1, or if the day-to-night transfer is unrepresentative, the measured Crab SED shifts by an amount comparable to the total quoted systematic budget (22-23%), and the claimed good agreement with other observatories would no longer independently validate the MC. The muon-based optical efficiency tuning and the event-rate comparison in Fig. 15 can partially absorb a wrong V_AOD and therefore do not by themselves confirm the aerosol model. The paper acknowledges this limitation but provides no quantitative test of the transfer assumption.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents the first Crab Nebula observations with the two SST-1M telescopes at Ondřejov, 510 m a.s.l., using 46/52 h of mono data and 33 h of stereo data collected between September 2023 and March 2024. The authors calibrate the SiPM cameras from dark runs and pedestals, tune the optical efficiency with muon rings, adopt a fixed V_AOD = 0.05 atmosphere from a daytime sun-photometer 45 km away, and produce a large CORSIKA + sim_telarray MC production with per-telescope NSB levels. Random-forest regressors and classifiers are used for energy, direction, and gamma/hadron separation; IRFs are derived and applied in gammapy to fit a power-law Crab spectrum over 2.5-50 TeV and to create a significance skymap. The claimed results are analysis-level energy thresholds of 1 TeV (mono) and 1.3 TeV (stereo), energy resolutions of about 20% and 10%, angular resolutions of 0.18° and 0.10°, a factor ~2 stereo sensitivity improvement, and a Crab SED in good agreement with previous instruments within the quoted systematics. The systematic budget is reported as <10% in energy scale, 22-23% in flux normalization, and 3-6% in spectral index.","tokens_in":31900,"tokens_out":11084,"duration_ms":117369,"significance":"The paper has real strengths: an unusually detailed calibration chain (dark-run photoelectron spectra, NSB-induced voltage-drop model, muon-ring optical-efficiency monitoring), a very large MC production, an open-source analysis pipeline (sst1mpipe) with GADF-compliant outputs, MC-data comparisons at several analysis levels, and a dedicated off-axis performance study. If the performance claims hold, the paper is a valuable demonstration that a compact, low-altitude, SiPM-based stereo IACT can operate in the multi-TeV band with controlled systematics. The main reservation is that the external validation and the energy scale rest on the untested transfer of daytime V_AOD from Košetice, 45 km away; the paper's own sensitivity analysis shows this term dominates the flux-normalization budget. The claim of agreement with other observatories is also qualitative and needs a quantitative compatibility measure.","major_comments":[{"comment":"The fixed V_AOD = 0.05 used for the entire campaign is derived from daytime Sun-photometer measurements at Košetice, 45 km from Ondřejov, with no on-site nighttime validation. Section 8.1 states that a 0.1 change in V_AOD shifts the light/energy scale by about 10%, and Table 2 converts the assumed 0.05 V_AOD uncertainty into 5% energy-scale and 18-19% flux-normalization systematics. An unquantified transfer error of order 0.1 would therefore move the Crab flux normalization by an amount comparable to the total 22-23% systematic budget and could invalidate the claimed agreement with other observatories. The limitation is acknowledged in the text, but no quantitative test of the transfer assumption is provided. I ask the authors to add a concrete cross-check, for example comparing the run-to-run event-rate stability (Figure 14) or a Cherenkov-transparency metric with the Košetice V_AOD time series, or repeating the Crab spectral fit with IRFs generated at V_AOD = 0.10 and 0.15 and reporting the resulting flux shift.","section":"Section 2.4, Section 8.1, Table 2"},{"comment":"The abstract and Section 9 conclude that the measured Crab SED is in good agreement with other observatories, but the evidence is a visual overlap of bands and points. Because the total flux-normalization uncertainty is 22-23% and the statistical errors are sizeable, the agreement could be trivially satisfied. Please provide a quantitative measure (e.g., chi2/ndof or pulls of the stereo flux points relative to a reference Crab spectrum, propagated with the systematic covariances) to support the claim. This is load-bearing because the SED comparison is the main external validation of the MC model.","section":"Section 6.2, Figure 16"},{"comment":"The MC-data validation is partly a consistency check: the MC is tuned using dark-run spectra (Figure 1), pedestal NSB distributions (Figure 5), muon-ring light yield (Figure 3), and the intensity threshold is chosen from the data/MC rate comparison (Figure 15). Consequently those comparisons are internal consistency checks rather than independent predictions. The only external benchmarks are the Crab excess rates and the SED, whose interpretation depends on the V_AOD energy scale. Please state explicitly which comparisons are predictive rather than tuned, and give the statistical precision with which the Crab excess rate constrains the gamma-ray effective area and the energy scale.","section":"Section 7, Figures 15, 18-20"}],"minor_comments":[{"comment":"The phrase 'thus are copped for in the analysis' should read 'thus are accounted for in the analysis'.","section":"Section 1"},{"comment":"The sentence 'as it was observed after operation of SST-1M-1 that lowering R_bias would decrease the impact of their order to lower the effects of NSB' is garbled and should be rewritten.","section":"Section 2.1.2"},{"comment":"Equation (4) is missing a square root and has unbalanced parentheses; it should read ΔΓ = 2 sqrt((5%/SBR_LE)^2 + (5%/SBR_HE)^2) / log(E_max/E_min), or the equivalent with the intended parentheses.","section":"Section 8.3, Eq. (4)"},{"comment":"The line giving the best-fit coordinates has unbalanced parentheses: '((α2000 = 83.62°, δ2000 = 21.99°) ± 0.02°' needs an extra closing parenthesis.","section":"Section 6.3"},{"comment":"Duplicate entries appear for Aleksić et al. (2016), Bose et al. (2022), and Heller et al. (2017); these should be consolidated.","section":"References"},{"comment":"Please clarify the exact livetimes used in the spectral analysis: 46/52 h raw, 33/27 h mono, and 25 h stereo are quoted, but the text should state explicitly why the mono livetimes differ between telescopes and which livetime enters the flux normalization.","section":"Section 6.1/6.2"},{"comment":"The abbreviation 'C.U.' is used for Crab units; please define it at first use.","section":"Section 5.4"}],"recommendation":"major_revision","confidential_remarks":"This is a solid commissioning and performance paper that fits the scope of A&A. The main blocker is the V_AOD transfer assumption; the authors can likely address it with an additional cross-check and a quantitative SED comparison. I would not reject the paper, but the central 'good agreement' claim needs more support than a visual comparison."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a careful, honest commissioning paper, and the Crab detection is the real thing. But the conclusions about \"validating\" MC performance should be read as instrument self-consistency rather than an independent check: the MC was tuned to the same dark runs, pedestals, and muon rings used for comparison, and the atmosphere is a fixed V_AOD taken from a daytime photometer 45 km away.\n\nWhat's actually new: first stereo Crab observations with SST-1M, first full sim_telarray-to-data comparison for this telescope pair, and a genuinely thorough systematics budget. The SED agrees with MAGIC, VERITAS, HESS, LHAASO, and others within quoted errors, and the skymap significance distribution is clean. The muon-ring optical efficiency tracking and the NSB voltage-drop modeling are well done. For the SST-1M/CTA path this is a milestone, even if the physics target is a standard candle.\n\nSoft spots, in order of real weight:\n- The V_AOD concern is real. Section 2.4 takes a daytime, 45-km-away measurement, averages it to 0.05, and applies it to all nights. Section 8.1 says a 0.1 change in V_AOD shifts the energy scale by ~10% and the flux normalization by 18-19%. The total flux-normalization systematics are 22-23%, so an unrepresentative aerosol column would eat the entire budget. They acknowledge this and name on-site monitoring as mitigation, but they do not test the transfer with data. This limits the strength of the \"MC validated\" claim.\n- Several headline numbers (thresholds, resolutions, sensitivity) are MC-computed. The data validation shows good agreement above threshold, but there are acknowledged residual discrepancies near threshold that are left for future work.\n- The background systematics are stated as inconclusive because of limited statistics. That is honest, but it means the 5% background systematic is more an assumption than a measurement.\n\nNone of this is fatal. The central claim—SST-1M stereo at 510 m detects the Crab, and the tuned MC describes the response above ~1 TeV—holds up. The paper is candid about its limitations, and sst1mpipe is on GitHub, so the pipeline is at least inspectable.\n\nWho it's for: IACT instrumentalists, especially people in the CTA/SST path or planning low-altitude arrays. I'd send it to peer review with a request for a quantitative test of the V_AOD transfer or a more cautious wording about MC validation. It deserves serious referee time.","headline":"Solid commissioning paper: real Crab detection and unusually thorough systematics, but the 'MC validation' is partly a consistency check because the MC is tuned to the data and the aerosol model is borrowed from a site 45 km away.","tokens_in":32897,"tokens_out":2120,"would_cite":true,"duration_ms":24399,"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":"The two SST-1M telescopes, at 510 m altitude, detect the Crab Nebula and reproduce its TeV spectrum, validating the full Monte Carlo model of the low-altitude array.","keywords":["SST-1M","imaging atmospheric Cherenkov telescopes","Crab Nebula","silicon photomultipliers","stereoscopic observations","Monte Carlo validation","multi-TeV gamma-ray astronomy","low-altitude observatory"],"falsifier":"Put an on-site lidar or sun/moon photometer at Ondrejov during a repeat Crab campaign and measure the vertical aerosol optical depth on the same nights the telescopes observe. If the true $V_{\\rm AOD}$ differs from 0.05 by 0.1, the reconstructed flux normalization would shift by about 18–19%, comparable to or larger than the quoted total systematics; a repeat measurement that refuses to reproduce the Crab SED under the assumed atmosphere would falsify the energy-scale calibration, while agreement under measured $V_{\\rm AOD}$ would confirm it.","tokens_in":31469,"feed_emoji":"🔭","tokens_out":8971,"duration_ms":87539,"temperature":0.7,"pith_summary":"Two 4-meter Imaging Atmospheric Cherenkov Telescopes with silicon-photomultiplier cameras, installed at the low altitude of 510 meters above sea level, have detected the Crab Nebula in both single-telescope and stereoscopic modes. The paper's central claim is that the measured very-high-energy spectrum of the Crab matches the results of established TeV observatories within uncertainties, and that a carefully tuned Monte Carlo model of the atmosphere, optics, camera, and electronics describes the real data from trigger level up to reconstructed gamma-ray images. If true, the SST-1M array works as a multi-TeV gamma-ray instrument despite the extra atmospheric attenuation at low altitude, with an analysis-level energy threshold of 1 TeV in mono and 1.3 TeV in stereo, stereo energy resolution around 10%, and an acceptance that stays nearly flat out to 2.5 degrees off axis. The Crab observation is used as a standard-candle validation: agreement between data and simulation is shown in event rates, shower-image parameters, angular distributions, and the derived spectral energy distribution.","feed_headline":"Two low-altitude telescopes catch the Crab Nebula at TeV energies","feed_subtitle":"A 33-hour stereo campaign reproduces the standard-candle spectrum, backing the array's Monte Carlo performance model.","key_machinery":"The load-bearing mechanism is a full end-to-end Monte Carlo chain, calibrated to the two specific telescopes: CORSIKA generates the air showers, sim_telarray ray-traces the Cherenkov light through a Davies-Cotton 4-meter mirror and the SiPM camera, and the atmospheric model combines ERA5 molecular density profiles with MODTRAN aerosol transmission using a vertical aerosol optical depth of 0.05 taken from a daytime sun photometer 45 km away. The camera model is tuned against dark-run photoelectron spectra (Borel-distributed crosstalk), muon-ring images for optical efficiency, and a measured voltage-drop model for SiPM behavior under night-sky background. The tuned simulations produce instrument response functions and train random-forest (machine-learning) regressors and a gamma-hadron classifier; the Crab Nebula data then serve as the independent check that the whole chain holds.","core_discovery":"The discovery claimed is that a stereoscopic pair of small-size IACTs operating at 510 m above sea level can observe the Crab Nebula and reconstruct its multi-TeV spectrum, and that the instrument's complete simulation—including the SiPM camera response, Davies-Cotton optics, atmospheric transmission, and machine-learning energy and direction reconstruction—reproduces the measured Crab signal without renormalization. Using 33 hours of stereo data (25 hours after quality cuts) and mono data from both telescopes, the fit of a power law $\\mathrm{d}\\phi/\\mathrm{d}E = \\phi_0 (E/E_0)^{-\\Gamma}$ over 2.5–50 TeV gives flux normalizations at $E_0 = 7$ TeV of $(1.76–2.02)\\times 10^{-13}\\,\\mathrm{cm}^{-2}\\mathrm{s}^{-1}\\mathrm{TeV}^{-1}$ and spectral indices $\\Gamma = 2.68–2.78$ for the three datasets, consistent within uncertainties with published results from major TeV observatories. The same comparison validates the energy and angular resolutions, the background model, and the off-axis acceptance predicted by Monte Carlo.","pith_inferences":["Beyond the paper: a dedicated on-site nocturnal aerosol monitor (lidar or photometer) would directly test the strongest assumption; if it confirms $V_{\\rm AOD} \\approx 0.05$, the energy-scale systematics would be frozen, and if it does not, the 18–19% flux-normalization uncertainty would need to be revised upward.","Beyond the paper: the same dark-run and muon-ring tuning methodology could be transferred to other SiPM-based IACT cameras, potentially providing a uniform calibration strategy for next-generation arrays.","Beyond the paper: the flat off-axis acceptance suggests the SST-1M pair could serve as a survey instrument for extended Galactic sources, but a dedicated pointed campaign would be needed to demonstrate that capability."],"forward_implications":["If the Monte Carlo model is right, the quoted instrument response functions can be trusted for future science: mono energy threshold 1 TeV, stereo 1.3 TeV, energy resolution ~20% mono and ~10% stereo, and angular resolution 0.18° mono and 0.10° stereo.","Stereo observation improves flux sensitivity by about a factor of two over mono, reaching roughly 7% of the Crab flux in 50 hours above the energy threshold.","The acceptance stays within about 10% of flat out to 2.5° off axis, so extended sources and poorly localized transients can be observed without a strong loss of performance.","The validated pipeline can produce spectra and sky maps of other sources in the multi-TeV range from this low-altitude site, using the MC-derived response functions directly.","The current systematic budget—under 10% in energy scale, about 22–23% in flux normalization, and 3–6% in spectral index—sets the precision of the instrument, with the dominant terms being the aerosol assumption and night-sky-background variability."],"supporting_citations":[{"why":"Describes the SST-1M design, optics, PSF measurements, and software array trigger; supplies the instrument model that the Monte Carlo must reproduce.","marker":"Alispach et al. 2025"},{"why":"CORSIKA generates the air-shower and Cherenkov-light simulations from which all instrument response functions are derived.","marker":"Heck et al. 1998"},{"why":"sim_telarray simulates ray tracing, telescope geometry, camera response, and atmospheric attenuation; the detector half of the Monte Carlo chain.","marker":"Bernlöhr 2008"},{"why":"Measured SiPM behavior under night-sky background and provided the voltage-drop model used to correct image intensities.","marker":"Nagai et al. 2019"},{"why":"Supplies the Crab Nebula reference spectrum used for weighting and comparison, and the stereo direction-reconstruction approach.","marker":"Aleksi´c et al. 2016"},{"why":"Provides the Crab Nebula coordinates adopted for the signal region in the spectral and sky-map analysis.","marker":"H. E. S. S. Collaboration 2020"},{"why":"Reflected-region background estimation used in the spectral analysis.","marker":"Berge et al. 2007"},{"why":"Significance formula used for sensitivity estimates and the significance sky map.","marker":"Li & Ma 1983"},{"why":"Ring background method used to compute excess and significance maps.","marker":"Pühlhofer et al. 2003"}],"fun_headline_variants":["Low-altitude telescope pair sees Crab Nebula in TeV gamma rays","Crab Nebula detected by low-altitude SST-1M stereo system","First Crab observations from low-altitude stereoscopic IACTs","TeV Crab Nebula observed by low-altitude telescope pair","SST-1M stereoscopic system catches Crab Nebula at TeV"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the daytime aerosol measurement taken at a site 45 km away, averaged to a single value of 0.05 for the whole campaign, faithfully represents the vertical aerosol profile over Ondrejov on the actual observation nights; if that column or its height distribution is wrong, the energy scale shifts by about 10% per 0.1 of optical depth and the entire Monte Carlo validation moves with it.","fun_headline_variants_meta":{"raw":{"variants":["Low-altitude telescope pair sees Crab Nebula in TeV gamma rays","Crab Nebula detected by low-altitude SST-1M stereo system","First Crab observations from low-altitude stereoscopic IACTs","TeV Crab Nebula observed by low-altitude telescope pair","SST-1M stereoscopic system catches Crab Nebula at TeV"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000762,"raw_usage":{"total_tokens":3494,"prompt_tokens":1167,"completion_tokens":2327,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":783,"completion_tokens_details":{"reasoning_tokens":2233}},"tokens_in":783,"tokens_out":2327,"duration_ms":17346,"temperature":1.0,"reasoning_tokens":2233,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T11:35:11.786293+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Put an on-site lidar or sun/moon photometer at Ondrejov during a repeat Crab campaign and measure the vertical aerosol optical depth on the same nights the telescopes observe. If the true $V_{\\rm AOD}$ differs from 0.05 by 0.1, the reconstructed flux normalization would shift by about 18–19%, comparable to or larger than the quoted total systematics; a repeat measurement that refuses to reproduce the Crab SED under the assumed atmosphere would falsify the energy-scale calibration, while agreement under measured $V_{\\rm AOD}$ would confirm it.","supporting_citations":[{"cited_title":"2025, Journal of Cosmology and Astroparticle Physics, 2025, 047","cited_arxiv_id":null,"evidence_quote":"Describes the SST-1M design, optics, PSF measurements, and software array trigger; supplies the instrument model that the Monte Carlo must reproduce."},{"cited_title":"N., Schatz, G., & Thouw, T","cited_arxiv_id":null,"evidence_quote":"CORSIKA generates the air-shower and Cherenkov-light simulations from which all instrument response functions are derived."},{"cited_title":"2007, A&A, 466, 1219","cited_arxiv_id":null,"evidence_quote":"Reflected-region background estimation used in the spectral analysis."}],"review_version":1}