REVIEW 3 major objections 5 minor 5 references
A positioning system for Baikal-GVD
T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read The paper claims that the Baikal-GVD hydroacoustic positioning system locates optical modules with an average accuracy of 12 ± 6 cm, equivalent to sub-nanosecond timing for Cherenkov light.
desk verdict Useful Baikal-GVD status report, but the 12±6 cm accuracy is an interpolation consistency residual, not an absolute positioning calibration. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing mechanism is the chain from acoustic distance measurements to absolute coordinates: stationary nodes at the string anchors define a reference frame; acoustic modems, arranged as beacons along the strings and nodes at the bottom, measure distances; beacon coordinates are trilaterated from those distances; and optical-module coordinates are obtained by linear interpolation along a piece-wise linear string model. The error study itself relies on an extra set of beacons whose trilaterated coordinates are compared with their interpolated coordinates, making the interpolation assumption the testable link between beacon-level and optical-module-level accuracy.
What would settle it
Place an acoustic beacon at a location whose absolute coordinates are known independently—for example, by a tightly moored reference frame surveyed from the surface with GPS and a high-precision depth sensor—and compare its trilaterated acoustic coordinates with the surveyed ones over a season; discrepancies consistently above about 20 cm would invalidate the claimed sub-nanosecond timing equivalence.
Extended reading notes
Core claim
The central claim is that the hydroacoustic positioning system of Baikal-GVD is accurate enough to turn an unknown string drift of tens of meters into a coordinate uncertainty that is equivalent to a subnanosecond time calibration. Functionally, every cluster uses stationary acoustic nodes anchored to the lake floor plus acoustic beacons mounted along the strings; beacon coordinates are reconstructed by trilaterating measured distances to the nodes, and optical modules between beacons are then positioned by assuming the string is piece-wise linear. To quantify the error, two extra beacons were installed between standard beacons on two strings, and for a season their trilaterated coordinates were compared with the positions obtained by interpolation from the surrounding beacons. The resulting mean error, 12 ± 6 cm, is presented as the positioning accuracy of an individual optical module and as comparable to other large-scale neutrino telescopes.
Load-bearing premise
The estimate presumes that the acoustic trilateration of the test beacons is effectively the true position, so any error in the stationary node coordinates, the distance measurements, or the assumed sound-speed profile is invisible in the 12 ± 6 cm number.
Editorial extensions
If this is right
- A real-time position service can feed reconstructed beacon coordinates to the event pipeline every minute or so, so time calibration can track seasonal drift instead of assuming fixed geometry.
- The 12 ± 6 cm accuracy is presented as equivalent to a sub-nanosecond timing error, which is the level needed for Cherenkov direction reconstruction.
- The measured drift of up to 50 m and average speed of 0.5 cm/s means static deployment coordinates are insufficient for a full-size detector; the positioning system is necessary for the completed cubic-kilometre array.
- The demonstrated correlation of beacon motion across strings and clusters implies that global string-drift patterns could be modeled and partly predicted, reducing the required polling rate.
Reading between the lines
- A natural test the paper does not perform is an independent absolute check, e.g. a beacon with a co-located high-precision pressure sensor, since the 12 cm figure measures agreement between two acoustic/interpolation estimates rather than agreement with an external reference.
- If the node coordinates or the assumed sound-speed profile are biased beyond a few centimetres, the reported accuracy would shift by that bias, so the headline number should be read as internal precision until an absolute external calibration is done.
- The same distributed acoustic network could double as a current-meter: correlated beacon drift over time is a water-motion signal that could be inverted to estimate flow velocities across the detector volume.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports the status and performance of the hydroacoustic positioning system (APS) of the Baikal-GVD neutrino telescope. The system uses 171 acoustic modems (22 stationary nodes and 149 beacons) to trilaterate beacon coordinates on flexible strings, from which optical-module (OM) coordinates are obtained by piece-wise linear interpolation. The authors present beacon drift measurements (up to ~50 m, average speed 0.5 cm/s), correlation patterns across strings and clusters, and an error estimate for OM positioning: comparing trilaterated coordinates of two extra beacons with their interpolated coordinates yields a season-long mean error of 12 ± 6 cm. The abstract and conclusions state that this accuracy is sufficient for subnanosecond time calibration of the detector.
Significance. If the 12 ± 6 cm figure were a true absolute positioning accuracy, the paper would demonstrate that Baikal-GVD's acoustic system meets the telescope's timing-calibration requirements despite string drift. The drift statistics and correlation measurements are useful and consistent with the detector's design constraints, and the paper gives a concrete status update on a commissioned system. The explicit comparison with other large-scale neutrino telescopes is informative. However, the central accuracy claim is not established by the presented analysis, as detailed below; the strength of the paper is therefore the system description and the drift phenomenology rather than the claimed absolute accuracy.
major comments (3)
- [Section 3, Figure 5] The error estimate of 12 ± 6 cm is an internal-consistency residual, not an absolute positioning accuracy. The comparison is between two outputs of the same acoustic system: the trilaterated coordinate of each extra beacon and the coordinate obtained by interpolating neighboring beacons on the same string. These two quantities share the same node coordinates (from surface GPS trilateration), the same acoustic ranging hardware, and the same sound-speed/time model. Any bias common to those measurements—for instance, a systematic offset in node coordinates, an error in the assumed sound-speed profile, or a shared modem timing offset—cancels in the difference. Consequently, the quoted figure does not bound the absolute error of OM coordinates, and the conclusion in Section 4 that the system 'allows positioning optical modules of the telescope with an average accuracy of 12 ± 6 cm, which is equivalent to a subnanosecond time calibration' is not supported by the presented data. The authors should either reframe the claim as an internal consistency check or provide an independent ground-truth comparison (e.g., distances to a known calibration source or an external positioning method) and a sensitivity analysis of the residual to node-coordinate and sound-speed perturbations.
- [Section 3, Figure 5] The manuscript provides no statistical details for the quoted mean and spread. It is not stated how many measurements contribute to the 12 ± 6 cm figure, whether the ±6 cm is a standard deviation of the distribution or a standard error of the mean, how the residual varies in time (Figure 5 is not described in the text), or what the per-beacon results are. A seasonal average over 'April 2018 to February 2019' with no sample size or error budget for the three listed factors (AM measurement accuracy, interpolation distance, and beacon mobility) makes the central numerical claim impossible to evaluate. The authors should report the distribution of residuals, the number of points, the per-beacon statistics, and a breakdown of systematic uncertainties from node coordinates and sound-speed assumptions.
- [Section 3, interpolation model] The piece-wise linear string model used to interpolate beacon coordinates onto OM positions is assumed without validation. The positioning error from interpolation is stated to depend on the distance between the OM and the beacons, but no test is presented of how well a straight-line segment between beacons represents the actual string shape, which can be curved under water currents. This is a load-bearing premise for the claimed OM accuracy. The authors should justify the model with, for example, bending-angle measurements, mechanical string simulations, or an explicit error term that accounts for deviations from linearity.
minor comments (5)
- [References] Reference [1] is listed as 'Status, these proceedings' without a title or author list; this should be completed for the proceedings version.
- [Throughout] The spelling 'Cerenkov' is inconsistent with the standard 'Cherenkov'; please unify.
- [Figure 5] The caption of Figure 5 should state what the plotted curve represents (e.g., daily means, raw residuals, or cumulative statistics), and the axes should be labeled with units.
- [Section 1] The sentence 'An AM is submerged at the depth of ∼ 1 meter at several sites on the surface' is unclear; presumably the AM is lowered to about 1 m below the surface, not 'submerged at the depth of ∼ 1 meter at several sites on the surface'.
- [Section 2] The text says 'Figure 3 also shows, that the coordinates...' and 'Figure 4 illustrates correlation' but the panels in Figure 4 are not described; a short explanation of the plotted quantities (which coordinate, which time window) would improve readability.
Circularity Check
12±6 cm 'accuracy' is an internal-consistency residual: both compared quantities share the same acoustic ranging, node coordinates, and sound-speed model.
-
self definitional
[Section 3, 'Positioning precision' (Figure 5); Section 4, 'Conclusions']
"The error estimate is acquired by comparing trilaterated coordinates of these beacons with the coordinates acquired via interpolation. ... The mean positioning error for both beacons over the season is 12 ± 6 cm ... It allows positioning optical modules of the telescope with an average accuracy of 12 ± 6 cm, which is equivalent to a subnanosecond time calibration."
The two quantities being compared are not independent ground-truth positions. The trilaterated beacon coordinates and the interpolated coordinates are both computed from the same acoustic distances to the same stationary nodes, using the same node coordinates (from surface GPS trilateration) and the same sound-speed/time model. A systematic error in any of these shared inputs—e.g., a uniform sound-speed-profile bias, a node-coordinate offset, or a common modem timing delay—shifts both estimates coherently and cancels in the difference. Thus the 12±6 cm residual measures only the internal consistency between trilateration and the piece-wise-linear interpolation model, not the absolute OM coordinate error.
full rationale
The central derivation chain is short: beacon coordinates are trilaterated from acoustic distances to stationary nodes; OM coordinates are obtained by interpolating beacon coordinates under a piece-wise linear string model; the quoted positioning error is the difference between a trilaterated extra beacon and an interpolated position on the same string. This is a self-comparison within one measurement chain. The residual is a legitimate test of the string-interpolation model and of random ranging consistency, but it is not a measurement of absolute positioning accuracy. Any bias common to the acoustic ranges—sound-speed offset, node-coordinate error, modem timing bias—enters both compared terms and cancels. Therefore the 'average accuracy of 12 ± 6 cm' is, by construction, an internal consistency residual rather than a bound on absolute coordinate error, and the subnanosecond time-calibration claim is only as strong as the unverified node-coordinate and sound-speed assumptions. I do not charge reference [2] as circular: it is a separate prior experimental report by the collaboration, and the current residual is computed from current data; the self-citation supports the error budget but does not by itself define the target result. The piece-wise linear assumption is a modeling premise, not an input-output tautology. The paper would need an independent external benchmark—such as mechanically surveyed target positions or a different ranging technology—to convert the internal residual into an absolute accuracy claim. Overall the central claim is partially circular, so the score is 4.
Assumptions & free parameters
assumptions (4)
- domain assumption Acoustic distance measurements by the EvoLogics S2C R42/65 modems have a precision of a few centimeters, as demonstrated in [2].
- domain assumption Node coordinates, determined by trilateration from surface GPS sites shortly after installation, are stationary and accurate.
- domain assumption A string can be modeled as piece-wise linear between beacons.
- domain assumption The sound speed profile in Lake Baikal is either uniform or accounted for in the acoustic ranging.
Cite this review
Pith. "Pith review of A positioning system for Baikal-GVD." pith.science (2026). https://pith.science/paper/WYP6XUWI
@misc{pith2026190805529,
author = {Pith},
title = {Pith review of: A positioning system for Baikal-GVD},
year = {2026},
howpublished = {\url{https://pith.science/paper/WYP6XUWI}},
note = {Machine review of arXiv:1908.05529}
}
read the original abstract
A cubic kilometer scale neutrino telescope Baikal-GVD is currently under construction in Lake Baikal. Baikal-GVD is designed to detect Cerenkov radiation from products of astrophysical neutrino interactions with Baikal water by a lattice of photodetectors submerged between the depths of 1275 and 730 m. The detector components are mounted on flexible strings and can drift from their initial positions upwards to tens of meters. This introduces positioning uncertainty which translates into a timing error for Cerenkov signal registration. A spatial positioning system has been developed to resolve this issue. In this contribution, we present the status of this system, results of acoustic measurements and an estimate of positioning error for an individual component.
Figures
Figures from the paper (1 more)
Reference graph
Works this paper leans on
-
[1]
Avrorin A. D. et al., Status, these proceedings
-
[2]
Avrorin, A. V . et al. (2013). A hydroacoustic positioning system for the experimental cluster of the cubic-kilometer-scale neutrino telescope at Lake Baikal. Instruments and Experimental Techniques, 56(4), 449-458
work page 2013
-
[3]
Kebkal, O. et al. (2011, June). D-MAC: Media access control architecture for underwater acoustic sensor networks. In OCEANS 2011 IEEE-Spain (pp. 1-8). IEEE
work page 2011
-
[4]
Riccobene, G. (2019). The Positioning system for KM3NeT. In EPJ Web Conf. (V ol. 207, p. 07005)
work page 2019
-
[5]
Ardid, M. (2009). Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 602(1), 174-176. 4
work page 2009
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.