{"id":"182675cc-24af-4aa2-9e46-f11050af1583","arxiv_id":"1909.02663","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The paper describes the design of UCIRC2, a dual-band 10 and 12 micron infrared cloud monitor for the EUSO-SPB2 balloon, with an actively temperature-stabilized camera stage.","lead":"UCIRC2 is a two-band infrared camera system designed to monitor clouds beneath the EUSO-SPB2 balloon telescope. This paper describes the engineering design and planned calibration, but not yet any flight data.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Cloud-height retrieval rests on Eq. 2.1 assumptions that are unvalidated; two-band inversion cannot separate cloud emissivity, surface state, and multilayer structure.","rationale":"The reader's weakest assumption correctly identifies the single-layer, thermal-equilibrium, known-surface assumption in Eq. 2.1 as the load-bearing element of the cloud-height claim. The paper is an engineering status report; the design details are plausible and the calibration plan is sensible, but the physics of the retrieval is the place where the central claim could fail. The two-band scheme fundamentally cannot distinguish between a cold, low-emissivity cloud and a warmer, higher-emissivity cloud when surface and emissivity are uncertain, and multilayer scenes are common in the atmosphere. The concern is not an external disagreement with consensus; it is an internal gap between the stated retrieval equation and the general claim in the abstract. Because the authors explicitly describe the assumption and the instrument has not yet flown, the appropriate verdict is the same conditional acceptance the reader proposed. A synthetic retrieval study is a concrete, low-cost test that would either validate the claim under realistic conditions or force the claim to be restricted. I therefore agree with the reader's assessment and see no reason to change the verdict.","tokens_in":4682,"tokens_out":3178,"duration_ms":41010,"concrete_test":"Run a synthetic end-to-end retrieval test: generate two-band top-of-atmosphere radiances with a line-by-line or MODTRAN-like radiative transfer model for (a) a single opaque cloud with known surface, (b) a two-layer cloud (high thin cirrus over lower cloud), (c) varying surface emissivity and temperature, and (d) wavelength-dependent cloud emissivity. Apply the Eq. 2.1 inversion (or the actual algorithm from reference [7]) and compute CTH bias and uncertainty versus truth for each scene over the plausible atmospheric range. If the retrieved CTH error exceeds about 1 km in any multilayer or unknown-surface case, the abstract's general claim should be qualified to the single-layer, known-surface configuration.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that minute-cadence 10/12 micron images determine cloud height is supported only by Eq. 2.1, Ptot = epsilon*Pc + (1-epsilon)*PE. This equation assumes a single, opaque-in-emission cloud layer with a single wavelength-independent emissivity epsilon, in thermal equilibrium, over a surface of known temperature and emissivity. Real scenes violate each of these: cloud emissivity is spectrally variable and often low; semi-transparent cirrus over lower clouds produces a mixed radiance; surface temperature/emissivity (land, sea ice, broken clouds) is not known to the required accuracy. With only two spectral channels, one cannot simultaneously solve for Tc, epsilon(lambda1), epsilon(lambda2), and surface state, so the retrieval is biased. Additionally, CTH is inferred from Tc via an assumed atmospheric lapse rate, but no atmospheric profile or actual retrieval algorithm is described. Since the purpose is exposure correction for UHECR fluorescence, a biased low CTH (warm effective emission level) directly overestimates the observable volume. The paper itself flags the single-layer assumption but provides no accuracy estimate or validation plan beyond thermal calibration; hence 'allow determination' is a design promise rather than a demonstrated capability.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript describes the design, construction, and planned calibration of UCIRC2, a two-band infrared cloud monitor for the EUSO-SPB2 balloon mission. UCIRC2 uses two uncooled microbolometer cameras centered at approximately 10 and 12 micrometers to image the cloud field below the balloon once per minute. The authors argue that brightness temperatures in these two bands can be combined, through Eq. (2.1), to estimate cloud temperature and hence cloud-top height, which is needed to compute the effective exposure of the fluorescence and Cherenkov detectors. The paper presents the mechanical and thermal design, the electronics and software architecture, the data volume budget, and the planned thermovac and blackbody calibration procedures, but it reports no measured data from the completed instrument.","tokens_in":4947,"tokens_out":3816,"duration_ms":46036,"significance":"If the instrument performs as designed, UCIRC2 would fill an important operational gap for EUSO-SPB2: continuous cloud-top height and coverage information within the detector field of view, which is necessary for accurate exposure correction in ultra-high-energy cosmic ray fluorescence observations. The paper's strengths are in the engineering domain: the thermal control concept is well motivated by the balloon environment, the wide field of view is matched to the detector needs, the data volume estimate is concrete and modest, and the calibration plan addresses pixel-by-pixel gain and offset in a temperature-stabilized configuration. These parts are plausible and follow standard practice. The central scientific claim, however, is not supported by any measured data or by a complete retrieval algorithm; the paper reports a design and a plan, not a validated capability. The significance of the manuscript therefore hinges on whether the retrieval assumptions can be made explicit and tested, and the claims are high.","major_comments":[{"comment":"Equation (2.1) does not by itself break the degeneracy between emissivity and temperature with only two spectral channels. In each band, the cloud emissivity epsilon_lambda is an independent unknown, and it is well known that water and ice cloud emissivity varies significantly between 10 and 12 micrometers; the equation as written uses a single epsilon, silently assuming wavelength-independent emissivity or imposing an implicit closure relation. With two bands one has two measurements but, in general, more unknowns (Tc, epsilon_10, epsilon_12, and the surface state). The paper should state the exact retrieval algorithm, including the assumed spectral emissivity model or regularization, and should quantify the sensitivity of the retrieved cloud-top height to emissivity mismatch and to a non-opaque, multi-layer cloud scene.","section":"Section 2, Eq. (2.1)"},{"comment":"The abstract claims that minute-cadence IR images 'allow the determination' of cloud height and coverage, but no test, calibration, or validation data are presented. Section 4 describes the thermovac and blackbody calibration program entirely in the future tense ('will be tested', 'will be performed'), and Figure 5 shows a preliminary, not yet implemented, calibration setup. The claim is therefore a design promise rather than a demonstrated capability. The authors should either present laboratory or field validation (for example, retrieval tests against known cloud scenes, lidar ceilometer data, or UCIRC1 flight data) or add a quantified error budget and revise the abstract and introduction to state that the capability is expected but not yet verified.","section":"Section 4 and Abstract"},{"comment":"The retrieval chain from cloud temperature to cloud-top height is incomplete. The paper does not specify the atmospheric thermal profile or lapse-rate model used to convert Tc to CTH, and it does not account for line-of-sight atmospheric emission and absorption between the cloud and the balloon, or for a heterogeneous surface (land, sea ice, broken clouds) rather than a uniform ocean of known temperature. These effects are not negligible in the 10-12 micrometer window and directly bias the effective cloud emission level. Because a small error in cloud temperature translates into a substantial altitude error in the upper troposphere, the authors should provide a complete forward model, state all assumptions, and give at least an order-of-magnitude bias analysis for realistic scenes.","section":"Section 2, Eq. (2.1) and Section 4"}],"minor_comments":[{"comment":"The wording 'two IR cameras observe at wavelengths of 10 µm and one at 12 µm' is grammatically awkward, and Figure 4 labels one camera as '9µ Camera' while the text says the cameras are centered at 10 and 12 micrometers; the labels should be made consistent.","section":"Section 3.1"},{"comment":"There is a typo in 'reconstrction method' and 'Blackbody Power Ratio CTH reconstrction method'; also, the sentence about the filter choice says the bands facilitate both the 'Blackbody Power Ratio' and 'Radiative Transfer Equation' methods, but only reference [7] is cited and no details of these methods are given, so the reader cannot judge whether the chosen bands are adequate.","section":"Section 3.1"},{"comment":"Figure 2 shows uncalibrated UCIRC1 images, not UCIRC2 images; the caption and text should state this clearly and explain that the image only demonstrates qualitative cloud coverage, not the quantitative cloud-height retrieval claimed for UCIRC2.","section":"Figure 2 and Section 2"},{"comment":"The calibration description would be more useful if it specified the calibration target's emissivity, the planned temperature range and number of set points, and how the pixel-by-pixel gain and offset will be derived from the acquired images; this would allow a reader to assess whether the planned calibration can actually support the temperature accuracy needed for the retrieval.","section":"Section 4"}],"recommendation":"major_revision","confidential_remarks":"This is essentially a conference proceedings paper describing hardware design, with the retrieval method presented at a high level and no measured performance. For a journal publication, the central claim about cloud-height determination needs either a complete retrieval algorithm with error analysis or explicit validation data; otherwise the paper should be framed strictly as a design write-up with prospective capabilities. If the venue is an instrument-design proceedings archive, a softened abstract and an added limitations paragraph might suffice, but for the present scope I recommend requiring the retrieval treatment to be completed or the claims to be substantially tempered."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nQuick take: this is a well-written design report for a balloon-borne infrared cloud monitor. The engineering is thoughtful and the calibration plan is sensible, but the central retrieval claim—that 10/12 micron images allow determination of cloud top height—is unvalidated and the retrieval physics is oversimplified. Read it as a status report, not a science result.\n\nWhat's new: the specific implementation of dual-band thermal IR imaging on a high-altitude balloon, using two uncooled microbolometer cameras with filters at 10 and 12 microns. The active temperature control system (Peltier coolers, heat pipes, resistive heater, PID controller) is described in enough detail to be credible and reproducible. The idea to power-cycle the cameras via GPIO and to take summed image bursts for SNR is practical. The pixel-by-pixel calibration over a range of camera temperatures is good practice. The paper is clearly written and appropriately references prior work (UCIRC1, JEM-EUSO, [7] for retrieval methods).\n\nThe soft spots are real but not fatal. Equation 2.1 assumes a single, opaque cloud layer with a single emissivity, over a surface of known temperature. Real scenes violate these assumptions; with only two bands you can't separate emissivity, surface state, and multilayer structure. The paper does not describe the actual retrieval algorithm, nor the atmospheric lapse rate used to convert Tc to CTH, nor any error estimate. The stress-test note is on target: the abstract's \"allow the determination\" should be read as \"designed to allow\" pending calibration and validation. I'd like to see a limitations paragraph that says this explicitly. The missing test data are not a flaw for a conference status report—the instrument wasn't built yet—but the wording should not over-promise.\n\nBottom line: this is a useful engineering contribution for the EUSO-SPB2 effort and for future space-based UHECR missions. It deserves peer review in its venue (ICRC), but the authors should qualify the retrieval claims. I'd accept conditionally.","headline":"A solid engineering description of a balloon IR cloud monitor, but the cloud-height retrieval is a design promise, not a demonstrated capability.","tokens_in":5467,"tokens_out":3094,"would_cite":false,"duration_ms":34547,"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":"Two infrared cameras read cloud height below a balloon telescope.","keywords":["infrared cloud monitor","cloud top height","two-band radiometry","microbolometer camera","balloon payload","cosmic ray exposure","instrument calibration"],"falsifier":"Compare UCIRC2's retrieved cloud-top heights with simultaneous ground-based lidar or radiosonde measurements during a flight over a region with broken or multi-layer clouds; if the two-band heights diverge from the measured cloud heights whenever more than one cloud layer is present, the single-layer assumption is the failure point.","tokens_in":4538,"feed_emoji":"☁️","tokens_out":8544,"duration_ms":88472,"temperature":0.7,"pith_summary":"This paper presents the design, construction, calibration, and testing plan for UCIRC2, an infrared cloud monitor that will fly on the EUSO-SPB2 balloon payload. The central claim is that a pair of calibrated infrared images, one centered at 10 μm and one at 12 μm, taken every minute from a downward-looking balloon, is enough to determine both cloud coverage and cloud-top height within the field of view of the cosmic-ray detectors. That matters because high clouds can hide the brightest part of an ultra-high-energy cosmic-ray air shower, so the effective detector volume depends on knowing where clouds are. The paper argues that UCIRC2's wide field of view and minute-cadence image pairs provide this information continuously during night observation.","feed_headline":"Balloon cloud camera maps cloud height every minute","feed_subtitle":"UCIRC2's paired 10 and 12 micron images reveal which clouds block the view of cosmic-ray showers.","key_machinery":"The load-bearing mechanism is the two-band brightness-temperature measurement. UCIRC2 pairs two 640×480 uncooled IR cameras with different bandpass filters, one near 10 μm and one near 12 μm, so the same scene is seen in two bands spanning the thermal emission peak. The retrieval uses the relation $P_\\mathrm{tot} = \\epsilon P_c + (1-\\epsilon) P_E$, which expresses the detector power as a mixture of cloud emission and surface emission through the unknown cloud emissivity $\\epsilon$; two bands break the degeneracy between $\\epsilon$ and cloud temperature, and the temperature then gives cloud-top height. The argument depends on a pixel-by-pixel calibration of each camera against a temperature-controlled blackbody target, with the camera stage held at a fixed set point by Peltier coolers, heat pipes, and a resistive heater, so the measured brightness temperatures remain reliable across the roughly 40 °C to −40 °C flight temperature range.","core_discovery":"The paper argues that two-band brightness-temperature imaging from above the atmosphere can break the degeneracy between cloud emissivity and cloud temperature, giving cloud-top height without assuming a fixed emissivity. UCIRC2 pairs two uncooled microbolometer cameras, one filtered to transmit roughly 9.6–11.6 μm and the other 11.5–12.9 μm, and captures a pair of images every 60 seconds while the payload observes at night. The retrieval models the power on the detector as $P_\\mathrm{tot} = \\epsilon P_c + (1-\\epsilon) P_E$, where $\\epsilon$ is cloud emissivity, $P_c$ is the cloud's blackbody power, and $P_E$ is the surface's power; with two bands, both $\\epsilon$ and the cloud temperature $T_c$ can be estimated, and $T_c$ maps to cloud-top height. The paper also details the thermal-control and calibration design that makes these radiometric measurements trustworthy in the balloon environment.","pith_inferences":["A natural extension is a three-band or hyperspectral cloud monitor, which would allow the single-layer assumption to be tested in flight by comparing estimates from different band pairs.","The same pixel-by-pixel calibration and thermal-control scheme could be applied to any uncooled microbolometer camera on a high-altitude balloon or small satellite, where ambient temperature swings otherwise distort the radiometric response.","Combining the minute-cadence cloud maps with the fluorescence telescope's triggered events would allow each cosmic-ray event to be flagged for whether its line of sight was cloud-free, a flag that could be used as a statistical weight in the energy spectrum."],"forward_implications":["During EUSO-SPB2 night observation, UCIRC2 will produce a cloud coverage and cloud-top-height map every minute across a field of view wider than the fluorescence telescope's, so the cosmic-ray exposure calculation can be corrected for clouds.","Because the IR field of view is wider than the photodetector field of view, cloud conditions in the detector's swept volume between images can be inferred from the IR images.","The two-band approach removes the need to fix cloud emissivity in advance, since the two measured brightness temperatures solve for both emissivity and cloud temperature.","The temperature-stabilized enclosure and vacuum-compatible construction allow uncooled microbolometer cameras to take quantitative radiometric data from a balloon at high altitude.","The burst-capture summation and bzip2 compression keep the data volume near 0.5 MB per minute, which is small enough to store and transmit for the long flight."],"supporting_citations":[{"why":"Defines the EUSO-SPB2 mission and its fluorescence and Cherenkov detection modes, which set the requirement for continuous cloud information in the detector field of view.","marker":"[3]"},{"why":"Describes the first-generation UCIRC cloud monitor and its flight, the design baseline UCIRC2 improves on.","marker":"[4]"},{"why":"Presents the cloud-top-height retrieval methods that motivate UCIRC2's two-band brightness-temperature approach.","marker":"[7]"}],"fun_headline_variants":["Two infrared bands unmask cloud height for cosmic-ray watch","Cloud tops revealed by twin thermal eyes on a balloon","Two-band IR camera measures cloud height every minute from balloon","Infrared duo on balloon breaks cloud emissivity-temperature ambiguity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The cloud-height retrieval assumes the scene contains a single cloud layer in thermal equilibrium with its surroundings above a surface of known temperature and emissivity; if the real field of view holds multiple layers, non-equilibrium clouds, or unknown surface conditions, the inferred cloud-top heights can be biased.","fun_headline_variants_meta":{"raw":{"variants":["Two infrared bands unmask cloud height for cosmic-ray watch","Cloud tops revealed by twin thermal eyes on a balloon","Two-band IR camera measures cloud height every minute from balloon","Infrared duo on balloon breaks cloud emissivity-temperature ambiguity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000582,"raw_usage":{"total_tokens":2717,"prompt_tokens":903,"completion_tokens":1814,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":519,"completion_tokens_details":{"reasoning_tokens":1748}},"tokens_in":519,"tokens_out":1814,"duration_ms":14121,"temperature":1.0,"reasoning_tokens":1748,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T04:43:00.203638+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compare UCIRC2's retrieved cloud-top heights with simultaneous ground-based lidar or radiosonde measurements during a flight over a region with broken or multi-layer clouds; if the two-band heights diverge from the measured cloud heights whenever more than one cloud layer is present, the single-layer assumption is the failure point.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the EUSO-SPB2 mission and its fluorescence and Cherenkov detection modes, which set the requirement for continuous cloud information in the detector field of view."},{"cited_title":"Allen, M","cited_arxiv_id":null,"evidence_quote":"Describes the first-generation UCIRC cloud monitor and its flight, the design baseline UCIRC2 improves on."},{"cited_title":"Anzalone, M","cited_arxiv_id":null,"evidence_quote":"Presents the cloud-top-height retrieval methods that motivate UCIRC2's two-band brightness-temperature approach."}],"review_version":1}