the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Jun 2026
How accurate is the beam position of scanning lidars? An intercomparison
Abstract. Scanning wind lidar provide a flexible measurement solution to accurately quantify near-shore and offshore wind resource and site conditions to help reduce risk and costs for prospective offshore wind developments. Typically these devices are set up using a near-horizontal beam position (elevation angle) in order to obtain wind speed measurements at distances of 3-10 km at heights in the atmospheric boundary layer relevant to wind turbines. Consequently, precise knowledge of the beam position and any associated positioning errors of the lidar scan head is required to minimise the uncertainty in the measurement height and thus the resulting wind speed measurements. A number of Hard Target Testing (HTT) methods are typically used to determine and correct for beam position errors, and include the use of static hard targets (SHT; e.g. buildings), dynamic hard targets (DHT; i.e. drone) and the sea surface itself (Sea Surface Levelling; SSL). However, to-date, these methods have not been consistently compared. This paper therefore systematically compares the performance of the SHT, DHT and SSL methods in determining the static elevation angle error (elevation offset) for scanning lidar wind measurement campaigns. A comparison experiment is conducted using two independent pulsed scanning lidar devices (type Vaisala WindCube 400S) at a near-shore wind measurement test site in the UK, using an offshore meteorological mast as the static reference hard target across all methodologies. Differences in the mean elevation offset relative to the reference target range from -0.03◦ to -0.20◦ for SHT, -0.002◦ for DHT, and from -0.05◦ to +0.07◦ for SSL. The uncertainties found from the methods range from approximately ±0.03◦ for DHT to up to ±0.20◦ for SSL. Overall, the results show that in the absence of offshore static hard targets, the DHT method can reliably determine the elevation offset with lowest uncertainty. While practically easier to execute, the SSL method showed more varying results between the two devices tested and exhibited larger uncertainties due to a number of factors including the determination of the sea surface distance, sea state and device tilt. When performed thoroughly, the elevation offset can be obtained from all methods to within the specified angular positioning error (±0.1◦) of the scanning lidar model. While the SSL method exhibits larger uncertainties, it offers an alternative approach when resources (e.g. suitable SHT candidates) are limited and can be implemented regularly to quantify time-dependent angular drifts for both near-shore and offshore applications.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Wind Energy Science.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(15412 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
EC1: 'Comment on wes-2026-94', Etienne Cheynet, 15 Jun 2026
-
AC3: 'Reply on EC1', Julia Gottschall, 12 Oct 2026
We would like to thank the editor and two referees for their time and effort in reviewing our paper, providing constructive feedback and insightful comments that have helped us improve the quality of the manuscript.
We have provided a separate set of line-by-line responses to each reviewer. The points raised with the EC are mostly already covered by these, which is why we are not responding to this separately.
Citation: https://doi.org/10.5194/wes-2026-94-AC3
-
AC3: 'Reply on EC1', Julia Gottschall, 12 Oct 2026
-
RC1: 'Comment on wes-2026-94', Anonymous Referee #1, 09 Jul 2026
Understanding the accuracy of scanning LiDAR pointing is very important for a range of applications. The present paper by M. Young and colleagues addresses this in a highly technical, well structured and extensive report. As a LiDAR user, I was happy to see several personal concerns addressed. Relying on ‘common’ approaches such as static hard-targetting and sea surface leveling, or more recent approaches such as dynamic hard-targetting, the authors estimate absolute azimuthal accuracies by computing azimuthal offsets and error. These angular uncertainties can be translated into vertical (height) uncertainties for a given sample range. This work can help future users to better anticipate risks, and adapt their experimental configurations as a function of their needs.
Whilst recommending this study for publication, I invite the authors to consider my comments and questions, which I hope will help clarify the scope and contant of the study.
Key comments :
My main concern is the definition of what accuracy is, and perhaps my comments will encourage you to better define it :
- Whilst you produce absolute uncertainties, its relative importance will depend on the local vertical gradient in horizontal wind speed (dependent at leading order on both the height above the surface, and the wind speed and associated wind shear), and whether a user wishes to focusing on specific structures (wind turbine wakes) and length scales (e.g. turbulent intensity).
- An important but briefly mentionned result is that for ‘temporally consecutive scans of the same static hard target, φlidar is constant’ (L322). Are these scans of the same type as those presented in Figure 4 and 5, and if so, please refer to these figures here. If possible, I believe these results (showing that φlidar is constant) should be shown, at least in the appendix, to convince the reader that there is no error here other than the pixel picking confidence level (cf. Table 5). Furthermore, I invite you to detail the duration of which each scan is performed. Indeed, the hard target gives you an average target position over the duration of the scan, which helps ensure that φlidar is constant by smoothing possible higher-frequency fluctuations. What is your take on the possible higher-frequency modulations that cannot be captured here ? Could elevation angle accuracy be sensitive to gusts, turbulence, that could modulate the LiDAR attitude through pressure fluctuations and possible vibrations and resonance ? I have a personal interest in your take on beam pointing accuracy at high-frequencies, and believe this should be discussed/presented as a limitation (up to a certain extent) of your accuracy estimates.
Minor comments (line by line) :
- L38 : You highlight the importance of accurately knowing lidar positionning errors here. Saying this here may lead readers to think that this only applies to LiDAR coastal deployments on land. It may be worth reformulating. You could introduce coastal deployments in the previous paragraph, and focus this one on the general need to accurately know lidar positionning errors.
- L47 : I find that the Δφ annotation is ambiguous here, and it may be clearer to directly refer to psi offsets or error. Talking about the variations of Δφ is ambiguous as I do not know at this point if you are talking about an offset, an error, or fluctuations around and offset to which you would need to attach information about the frequency of these variations, which you do not thoroughly detail this further in the manuscript.
- L97 : You write ‘additional uncertainties’, but they are not additional or new. They are always there. I would suggest presenting the paragraph by first presenting all identified sources of error (or making a reference to your Table 5). You can then say that offshore motion adds further complexity. Furthermore, your mention of the ‘weight of the lidar head’ gives the impression that this is the only source of SLiDAR error.
- L133 : I would complete this sentence by writing that vertical variations are expected to be larger in the lower part of the boundary layer, or in and around turbine wakes.
- L164 : It may be useful to already mention here that the possible tidal phase lag in your study area due to the distance of the tidal gauge is accounted for in the error estimates, and make a reference to the information you present in Table 5.
- Figure 3 : This figure needs a spatial scale, and a marker for the tidal gauge, or at least an arrow pointing in the direction of where the tidal gauge is.
- L237 : From what database, with what frequency and over what duration, was your reference mean sea level computed ? There is a possible ambiguity here between the hydrographic/historic water level reference, and the tidal gauge reference ?
- L290 : Whilst I am aware of the litterature on the topic of backlash, I recommend that you define the term here.
- L452 : I note the conservative nature of your filtering here. Considering this, could you clarify here why you have surface detections beyond 0° in elevation in your figure 10i ? Do you expect a bias on your results due to the fact that you are eleminiating so much data, and therefore the effective area over which you keep the intercepts with the data ?
- L453 : I understand here that you are left with 30-50 % and 50-90 % of data to work with after filtering. Is this correct ? I believe this sentence is worth clarifying to remove any ambiguity.
- L466 : Belt or gear systems seems to have a lot of importance on the errors you quantify for either RHI or PPI scans, and I am happy to see that this is being addressed in your work. Would it not be more convenient to evaluate PPI and RHI accuracy as a function of the system version using a hard target solution instead ? Has this been done elsewhere ? Have you tried hard-targetting a drone by alternating between RHIs and PPIs ? Whilst I agree that the SSL technique is straight-forward to test this, the high uncertainties associated with the SSL technique may mean that it is not be the best candidate for this comparison.
Typos :
- L4 : Consider adding a comma after ‘3-10 km’ for the clarity of the sentence, and also consider also providing vertical scales relevant to wind turbines.
- L70 : I suggest deleting half to sentence to have it begin at ‘Its variability’
- L227 : space missing in ‘levelingmethodologies’
Best regards,
W. Bruch
Citation: https://doi.org/10.5194/wes-2026-94-RC1 -
AC1: 'Reply on RC1', Julia Gottschall, 12 Oct 2026
We would like to thank the editor and two referees for their time and effort in reviewing our paper, providing constructive feedback and insightful comments that have helped us improve the quality of the manuscript.
We have provided a separate set of line-by-line responses to each reviewer (see attached documents).
While working on the revisions, we also discovered and corrected a small translation error in the reference measurements (azimuth and range/distance) for some of the long range static hard target candidates in Table 6 (now Table 5). This error has been corrected for, and does not alter any of the main results and conclusions of the manuscript.
-
AC1: 'Reply on RC1', Julia Gottschall, 12 Oct 2026
-
RC2: 'Comment on wes-2026-94', Anonymous Referee #2, 21 Aug 2026
Dear authors,
please find our comments in the attached document.
-
AC2: 'Reply on RC2', Julia Gottschall, 12 Oct 2026
We would like to thank the editor and two referees for their time and effort in reviewing our paper, providing constructive feedback and insightful comments that have helped us improve the quality of the manuscript.
We have provided a separate set of line-by-line responses to each reviewer (see attached documents).
While working on the revisions, we also discovered and corrected a small translation error in the reference measurements (azimuth and range/distance) for some of the long range static hard target candidates in Table 6 (now Table 5). This error has been corrected for, and does not alter any of the main results and conclusions of the manuscript.
-
AC2: 'Reply on RC2', Julia Gottschall, 12 Oct 2026
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 374 | 215 | 105 | 694 | 130 | 98 |
- HTML: 374
- PDF: 215
- XML: 105
- Total: 694
- BibTeX: 130
- EndNote: 98
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript addresses an interesting topic of clear relevance to Wind Energy Science. The discussion provides useful information on wind-speed errors and on the practical advantages and limitations of the different hard-target test methodologies. However, I find that the wind-energy implications of the required angular accuracy could be discussed more broadly.
At present, the discussion mainly considers the effect of angular errors on wind speed. This is relevant, but the acceptable angular error depends strongly on the intended application. For example, an error of 0.2° may be acceptable for estimating one-point turbulence statistics at a range of 1 km, but too large for coherence studies based on two lidars with nearly parallel beams, as highlighted during the COTUR experiment by Cheynet et al. (2021). I therefore recommend that the authors clarify which level of angular accuracy, calibration procedure, or hard-target test is appropriate for different applications, including mean wind-speed estimation, one-point turbulence statistics, and two-point turbulence statistics. This broader discussion would help strengthen the manuscript’s relevance to Wind Energy Science, rather than framing the work mainly as a technical contribution to Doppler wind lidar technology.
The literature review should also be completed and better contextualised with respect to previous and recent work on long-range scanning lidars for wind-energy applications. Vasiljević et al. (2016) is mentioned only briefly, although it contains relevant quantitative information on static pointing accuracy, hard-target mapping, backlash, and repeatability of long-range WindScanner systems. I recommend that the authors compare their results with those reported by Vasiljević et al. (2016). This would help readers assess whether the present results are consistent with earlier long-range scanning lidar systems and, if not, whether the differences may be due to differences in hardware or calibration procedure.
In addition, Mann et al. (2026), which includes some of the present manuscript’s co-authors, reports a related offshore lidar campaign focused on turbulence measurements in the marine boundary layer and novel hard target tests. This recent study could be used to provide more context on the wind-energy applications and practical motivation for accurate beam positioning.
References:
Cheynet, E., Flügge, M., Reuder, J., Jakobsen, J. B., Heggelund, Y., Svardal, B., ... & Godvik, M. (2021). The COTUR project: remote sensing of offshore turbulence for wind energy application. Atmospheric Measurement Techniques, 14(9), 6137–6157.
Mann, J., Patel, A., Sjöholm, M., Thorsen, G. R., Simon, E. I., Hung, L. Y., & Gottschall, J. (2026). An experimental campaign to measure turbulence in the marine boundary layer. Wind Energy Science Discussions, 2026, 1–27.
Vasiljević, N., Lea, G., Courtney, M., Cariou, J. P., Mann, J., & Mikkelsen, T. (2016). Long-range WindScanner system. Remote Sensing, 8(11), 896.