{"id":"e55ba2c4-5c62-4bf8-a245-304b35362109","arxiv_id":"1908.06705","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"low","formal_verification":"none","parameter_count":1,"one_line_summary":"H.E.S.S. developed scheduling and alert tools for gravitational wave follow-up and used them to observe GW170814, GW170817, and S190512at, finding no significant very-high-energy gamma-ray emission.","lead":"This paper describes how the H.E.S.S. gamma-ray telescopes follow up gravitational wave alerts and reports the observations made during the LIGO/Virgo runs O2 and O3. No very-high-energy gamma-ray counterpart was detected, so the value is in the prompt-response strategy and the upper limits it produced.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quantitative upper limits in Section 4 are not self-contained: they assume an E^-2 spectrum without stating a confidence level, so the VHE constraints cannot be independently assessed; the prompt-response claim itself is credible and unchanged.","rationale":"The reader's verdict is CONDITIONAL, and I agree with that assessment. The strongest claim of a working prompt-response pipeline is supported by the reported observation latency (5 minutes after the updated localization for GW170817) and the coverage claim, which are consistent with the cited detailed analyses. The weakest point is the spectral assumption for the upper limits, which the reader also identified; in addition, the proceedings do not state the confidence level of the limits. These are completeness issues that do not invalidate the central claim but do justify a conditional acceptance. No new concern moves the verdict in either direction, so UNCHANGED is appropriate.","tokens_in":5257,"tokens_out":4465,"duration_ms":39533,"concrete_test":"Recompute the integral upper limit for GW170814 from the reported sensitivity: 11 runs of 28 minutes with 5-sigma sensitivity of 20% Crab flux, assuming an E^-2 spectrum over 250 GeV to 10 TeV. Check whether the derived limit matches the quoted value in Figure 3 and whether it is at the usual 95% confidence level; if the confidence level is unspecified or the limit changes by more than 20% under a spectral index of -2.5, the stated constraints are incomplete.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claims are the integral upper limits on VHE emission from GW170814 and GW170817 (Section 4.1/4.2, Figures 3/4). These limits are derived assuming an E^-2 power-law spectrum, but the text does not state the confidence level (e.g., 95% CR) at which they are set, nor does it present the systematic uncertainties in the IACT analysis. Without these, the quoted constraints are not fully interpretable as standalone results. This is a completeness gap of a conference proceedings, not an internal inconsistency, and it is the same assumption the reader flagged. The scheduling and coverage claims are supported by the reported observation times and the ~90% coverage statement; the prompt-response capability is credible because the underlying analyses are cross-checked in the cited papers. The load-bearing concern is therefore that the quantitative VHE constraints are under-specified, not that the pipeline claim is false.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper describes the very-high-energy (VHE) gravitational-wave follow-up strategy developed for the H.E.S.S. instrument, focusing on the optimized scheduling algorithms and their implementation in the VoAlerter alert system. It reports on the H.E.S.S. observations of the binary black hole merger GW170814 and the binary neutron star merger GW170817 during LIGO/Virgo O2, and of the binary black hole event S190512at during O3. For GW170814, the text states that no significant gamma-ray emission was found and that the observations covered about 90% of the final localization, with integral flux upper limits derived for 250 GeV < E < 10 TeV assuming an E^-2 spectrum. For GW170817, H.E.S.S. began observations five minutes after receiving the updated localization, subsequently monitored the remnant over several nights, and again found no significant VHE emission, with upper limits between 270 GeV and 8.55 TeV assuming an E^-2 spectrum. The S190512at observations are presented as a technical trial of the alert chain. The central claim is that the optimized follow-up strategy was proven successful during O2, demonstrating prompt response and VHE constraints on compact binary mergers.","tokens_in":5418,"tokens_out":7918,"duration_ms":71931,"significance":"If the reported results are taken at face value, the paper demonstrates two things of interest to the multi-messenger community: H.E.S.S. can react to a GW alert within minutes and tile a meaningful fraction of the localization region, and the resulting observations place the first constraints on VHE emission from two O2 mergers. The paper's strengths are that the significance maps and upper limits come from established H.E.S.S. analysis chains (the Model analysis with ImPACT as an independent cross-check), that the quantitative response metrics (five-minute latency, roughly 90% coverage, 11 and 3 observation runs) are concrete and falsifiable, and that the paper explicitly refers to the externally published collaboration analyses [10] and [13] for the underlying results. The scheduling algorithms themselves are not new conceptually but their integration into an operational alert system is a useful practical contribution. The main weakness, as detailed below, is that the quantitative upper limits are under-specified as presented, so the constraints cannot be fully interpreted from this text alone.","major_comments":[{"comment":"The integral upper limits quoted for GW170814 (250 GeV < E < 10 TeV) and for GW170817 (270 GeV < E < 8.55 TeV) are presented without stating the confidence level at which they are set and without quoting the systematic uncertainties of the IACT analysis. The text says the limits are derived assuming an E^-2 power-law spectrum, but a reader cannot tell whether these are 95% or 99% credible limits, and Figure 4's caption repeats the spectral assumption without adding the confidence level. Because these limits are the paper's central quantitative constraints on VHE emission, their missing confidence level makes them not independently interpretable; the authors should state the confidence level explicitly and either give the systematic uncertainties or explicitly defer to the full analyses in [10] and [13].","section":"Section 4.1 and Figure 4 (Section 4.2)"},{"comment":"The statement that 'The H.E.S.S. observations cover around 90% of the localisation of the final reconstruction' is not defined or demonstrated quantitatively. It is unclear whether 'coverage' means the fraction of the 90% credible region's area, the fraction of its probability mass, or the fraction of the sky map within the instrument FoVs, and Figure 3 does not show a coverage map or a quantitative overlay. Since this figure is central to the claim that the scheduling strategy was successful, the authors should define the coverage metric and provide the corresponding map or cumulative-probability curve.","section":"Section 4.1"}],"minor_comments":[{"comment":"The word 'Unverse' in the abstract is a typo for 'Universe'.","section":"Abstract"},{"comment":"The phrase 'from ∼ 30 GeVs to about 100 TeV' should read 'from ∼ 30 GeV to about 100 TeV'.","section":"Section 2"},{"comment":"The phrase 'including neither sun light nor moon light, and reduced values of light pollution' is ungrammatical; consider 'including no sunlight or moonlight and low levels of light pollution'.","section":"Section 3.1, visibility constraints"},{"comment":"The caption contains the garbled phrase 'orange corresponds illustrates a the minimum angular distance'; it should be rewritten for clarity.","section":"Figure 1 caption"},{"comment":"The sentence 'Each of the observations reached a sensitivity of about 20% of the flux from the Crab nebula at 5 sigma' should specify the energy range and the analysis configuration for which this sensitivity is quoted.","section":"Section 4.1"},{"comment":"The sentence 'More information on the H.E.S.S. results of this follow-up observations' should be 'More information on the H.E.S.S. results of these follow-up observations'.","section":"Section 4.2"},{"comment":"The false alarm rate is typeset as '1/16.6 years −1'; it should be '1/16.6 yr^-1' or similar.","section":"Section 5"}],"recommendation":"major_revision","confidential_remarks":"This is a proceedings-style contribution whose main value is a concise record of the H.E.S.S. prompt-response capability. The missing confidence level is a standard completeness issue for such papers and is fixable by referencing the companion analyses. No concerns about novelty disclosure or citation behavior arise."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a conference proceedings, not a new result paper. The O2 follow-ups of GW170814 and GW170817 are already published in [10] and [13]; the only genuinely new item is the S190512at technical trial, which is explicitly presented as a chain check with no derived science. If you are looking for a new measurement, this isn't it. If you want a clear description of how an IACT array organizes GW follow-up, it is a useful, honest status report.\n\nWhat it does well: the scheduling section is concrete. The 2D and 3D algorithms are described clearly, and the VoAlerter workflow in Figure 2 explains the decision path. The headline operational claim—that H.E.S.S. can start observing within minutes of a GW localization update—is supported by the GW170817 timeline (first obs 5 min after the BAYESTAR update, 17:59 UTC). The ~90% coverage claim for GW170814 is also stated and is consistent with the referenced analysis. The paper is appropriately modest: the O3 event is called a technical trial, and the upper limits are flagged as assuming an E^-2 spectrum.\n\nSoft spots: the quantitative limits in Section 4 are under-specified in the text alone. No confidence level is stated (95%? 99%?), and the systematic uncertainty from the IACT analysis is not given. A reader who only reads this proceedings cannot fully interpret the flux limits. That is a real completeness gap, but minor: the limits are taken from published papers, where the details presumably appear. The paper would be improved by a sentence pointing to the CL and systematics, or by restating them. The E^-2 spectral assumption is a model input, attributed to the GRB literature; not a flaw, but the reader should know the limits are not model-independent. No data or code are provided, which is expected for a proceedings but limits independent checking.\n\nCitation pattern: the reliance on the collaboration's own [10] and [13] is legitimate because those are separate, externally published analyses; this paper is a summary, not a re-derivation. No circularity problem.\n\nWho it's for: people working on multi-messenger follow-up, especially IACT scheduling, and anyone wanting a compact record of H.E.S.S.'s O2/O3 GW program. It will be a useful reference for the collaboration's capabilities, but it won't change any astrophysical conclusions.\n\nRecommendation: worth a serious referee if submitted as a proceedings paper; the referee should ask for confidence levels and systematics in the text or an explicit pointer to the papers that give them. As a standalone journal article it would be too thin because the main results are elsewhere. I would accept it for the proceedings with minor revision, and I'd cite it only as a program description, not as a source of new constraints.","headline":"A well-written conference proceedings that clearly describes H.E.S.S.'s GW follow-up system, but the only new element is an O3 technical trial with no scientific result; the quantitative limits are under-specified in the text alone.","tokens_in":6023,"tokens_out":2314,"would_cite":false,"duration_ms":23508,"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":"The H.E.S.S. gamma-ray array shows a working gravitational-wave follow-up chain, covering 90% of GW170814's localization and responding within five minutes.","keywords":["gravitational waves","very-high-energy gamma rays","H.E.S.S.","IACT","multi-messenger astronomy","GW170817","GW170814","counterpart search"],"falsifier":"Replay the archived H.E.S.S. pointing and slewing logs for GW170814 or GW170817 against the published gravitational-wave localization maps: if the first pointings do not track the highest-probability regions, or if a future compact-merger counterpart emits a hard-spectrum very-high-energy signal that the scheduler deprioritizes, the strategy's optimality claim would be refuted.","tokens_in":5080,"feed_emoji":"🔭","tokens_out":8942,"duration_ms":87160,"temperature":0.7,"pith_summary":"The paper reports on the gravitational-wave follow-up program of the H.E.S.S. imaging atmospheric Cherenkov telescope array and argues that the array can act as a prompt multi-messenger responder. It describes scheduling algorithms that rank telescope pointings by the probability content of a gravitational-wave localization and an alert system that puts those pointings into action within minutes. The demonstration cases are two events from the O2 observing run: the binary black hole merger GW170814, whose final localization was about 90% covered by H.E.S.S. observations, and the neutron star merger GW170817, where observations began five minutes after an updated localization. With no significant very-high-energy gamma-ray source detected, the paper reports upper limits in the roughly 250 GeV to 10 TeV band, and it presents a later black hole merger follow-up as a successful full-chain trial. These results matter because they show that ground-based gamma-ray telescopes can contribute directly to gravitational-wave counterpart searches and can constrain non-thermal emission from compact mergers.","feed_headline":"Gamma-ray array covered 90% of GW170814's likely sky region","feed_subtitle":"Prompt gamma-ray follow-up began five minutes after GW170817's updated alert and set flux limits.","key_machinery":"The carrying mechanism is the scheduling engine inside the H.E.S.S. online alert system, named VoAlerter. It converts a gravitational-wave localization into an ordered list of pointings: a 2D algorithm uses the sky probability map directly, while a 3D algorithm combines the distance estimate with the positions of known galaxies. The engine favors low zenith angles so that observations reach the lowest possible energy threshold, a choice justified by the expectation that neutron-star merger counterparts are gamma-ray bursts with soft spectra. For the results quoted here, the analysis uses monoscopic data from the 28-meter telescope to concentrate on low energies, and the flux limits are derived assuming a power-law spectrum $E^{-2}$.","core_discovery":"On its own terms, the paper claims that H.E.S.S. now has a working gravitational-wave follow-up chain: the alert system receives a localization, selects a scheduling algorithm, and re-points the telescopes within minutes. The evidence is the O2 campaign, where H.E.S.S. observed about 90% of the final localization region of the binary black hole merger GW170814 over three nights and began observing the neutron star merger GW170817 five minutes after the updated localization was issued. No significant very-high-energy gamma-ray emission was found in either follow-up. The paper therefore presents integral upper limits on the flux between roughly 250 GeV and 10 TeV for GW170814 and between 270 GeV and 8.55 TeV for GW170817, both computed under an assumed $E^{-2}$ power-law spectrum, and it treats a successful O3 trial follow-up of the black hole merger S190512at as confirmation that the alert chain operates under current conditions.","pith_inferences":["Editorial inference: the scheduler's optimality claim is testable in hindsight; replaying H.E.S.S. pointing logs against the published localization maps would quantify how much probability is covered per minute compared with unoptimized pointing.","Editorial inference: because the quoted limits assume $E^{-2}$, the same dataset can in principle be recast for other spectral indices, so the constraint is broader than the single quoted number.","Editorial inference: the low-zenith priority depends on the soft-spectrum gamma-ray burst picture; a future hard-spectrum counterpart detection would motivate a scheduler that optimizes over multiple spectral hypotheses.","Editorial inference: with the smaller localization regions expected from future detector networks, the probability-ordered pipeline becomes relatively more efficient, so the demonstrated capability is likely to become more valuable."],"forward_implications":["Ground-based gamma-ray observatories can participate in gravitational-wave alerts as prompt responders, with reaction times of minutes rather than hours.","A single IACT array can cover the dominant part of a gravitational-wave localization; H.E.S.S. reports about 90% coverage of GW170814's final 90% credible region.","Null detections from well-timed observations turn into quantitative flux upper limits across roughly 250 GeV to 10 TeV, constraining non-thermal emission models of compact mergers.","The O3 trial on a distant binary black hole merger shows that the same chain functions under current alert conditions, including false-alarm-rate screening.","Extended monitoring of the GW170817 remnant between 120 and 250 days after merger probes the late-time behavior of the source, beyond the prompt phase."],"supporting_citations":[{"why":"Detection paper for GW170814, the binary black hole trigger that defined the follow-up target.","marker":"[9]"},{"why":"Companion H.E.S.S. paper supplying the significance map and upper limits for GW170814.","marker":"[10]"},{"why":"Detection of GW170817, the neutron-star merger event that set the prompt-response goal.","marker":"[11]"},{"why":"BAYESTAR localization algorithm that produced the updated sky map used to schedule the prompt H.E.S.S. pointing.","marker":"[12]"},{"why":"H.E.S.S. paper with the GW170817 follow-up results, including the 270 GeV to 8.55 TeV upper-limit map.","marker":"[13]"},{"why":"Source of the expectation that neutron-star merger counterparts appear as gamma-ray bursts, justifying the low-energy priority.","marker":"[5]"},{"why":"Gamma-ray burst spectral measurements showing soft spectra, the empirical basis for favoring low zenith angles.","marker":"[6]"},{"why":"LALInference reconstruction used for the GW170814 localization overlay.","marker":"[14]"}],"fun_headline_variants":["H.E.S.S. chased GW170814 across 90% of its sky map","H.E.S.S. responds to GW alerts in minutes, finds no VHE","No gamma-ray counterpart in H.E.S.S. GW follow-up campaigns","GW170814 follow-up: H.E.S.S. sets limits, no gamma-ray burst"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The quantitative limits and the decision to favor low-energy, low-zenith observations both assume that any very-high-energy counterpart to a compact merger has a soft spectrum like $E^{-2}$; if the true spectrum is different, the quoted constraints and pointing priorities may not represent the source.","fun_headline_variants_meta":{"raw":{"variants":["H.E.S.S. chased GW170814 across 90% of its sky map","H.E.S.S. responds to GW alerts in minutes, finds no VHE","No gamma-ray counterpart in H.E.S.S. GW follow-up campaigns","GW170814 follow-up: H.E.S.S. sets limits, no gamma-ray burst"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.002015,"raw_usage":{"total_tokens":7858,"prompt_tokens":949,"completion_tokens":6909,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":565,"completion_tokens_details":{"reasoning_tokens":6821}},"tokens_in":565,"tokens_out":6909,"duration_ms":49064,"temperature":1.0,"reasoning_tokens":6821,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:35:56.489401+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Replay the archived H.E.S.S. pointing and slewing logs for GW170814 or GW170817 against the published gravitational-wave localization maps: if the first pointings do not track the highest-probability regions, or if a future compact-merger counterpart emits a hard-spectrum very-high-energy signal that the scheduler deprioritizes, the strategy's optimality claim would be refuted.","supporting_citations":[{"cited_title":"PRL 119.14 (2017): 141101","cited_arxiv_id":null,"evidence_quote":"Detection paper for GW170814, the binary black hole trigger that defined the follow-up target."},{"cited_title":"Searches for TeV gamma-ray counterparts to Gravitational Wave events with H.E.S.S","cited_arxiv_id":"1906.10426","evidence_quote":"Companion H.E.S.S. paper supplying the significance map and upper limits for GW170814."},{"cited_title":"PRL 119.16 (2017): 161101","cited_arxiv_id":null,"evidence_quote":"Detection of GW170817, the neutron-star merger event that set the prompt-response goal."},{"cited_title":"ApJ, 2017, vol","cited_arxiv_id":null,"evidence_quote":"H.E.S.S. paper with the GW170817 follow-up results, including the 270 GeV to 8.55 TeV upper-limit map."},{"cited_title":"ApJ , 2019, vol","cited_arxiv_id":null,"evidence_quote":"Gamma-ray burst spectral measurements showing soft spectra, the empirical basis for favoring low zenith angles."}],"review_version":1}