the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 06 Oct 2025
Low-level jets in the southern North Sea: implications for wind turbine performance using Doppler lidar observations
Abstract. Accurate knowledge of wind conditions experienced by wind turbines is essential to assess their performance. Among these conditions, low-level jets (LLJs) are important to consider since they have a direct impact on wind turbines, as their cores are frequently located within the rotor layer. In this context, this study investigates the characteristics of LLJs and their formation mechanisms using 3.3 years of long-range Doppler lidar measurements obtained at Dunkerque, a coastal city on the southern North Sea. In addition, data from an ultrasonic anemometer and from the ERA5 reanalyses of the European Centre for Medium-Range Weather Forecasts (ECMWF) were used to determine the conditions favoring LLJ occurrence. The analysis revealed that LLJs were present in 15.6 % of the 117,411 measured wind profiles. The average jet core speed was 8.4 m s−1 , with a mean core height of 267 m. LLJs were more frequent during nighttime, especially in spring and summer. These characteristics were consistent with those obtained at other sites in the North Sea region, with some differences attributable to the location of Dunkerque on the coast near the Dover Strait. This position introduced additional formation mechanisms for LLJs, including land–sea thermal gradients and wind channeling in the English Channel. The impact of LLJs on wind turbines of varying dimensions was then assessed for both energy production and structural loads. For conventional turbines, with a hub height around 100 m, LLJs counter-intuitively tend to decrease power production at high wind speeds. Conversely, for more recent and future wind turbines, LLJs will improve power production in all conditions. The increase in turbine size will also greatly reduce their exposure to detrimental wind shear conditions, both in terms of speed and direction shear.
- Preprint
(5511 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on wes-2025-183', Anonymous Referee #1, 28 Oct 2025
-
RC2: 'Referee Comment on wes-2020-113 "Low-level jets in the southern North Sea: implications for wind turbine performance using Doppler lidar observations.”', Anonymous Referee #2, 16 Nov 2025
Summary:
The research described in the manuscript investigates low-level jets (LLJs) at Dunkerque, a coastal city on the southern North Sea. The main data were obtained from lidar observations up to an altitude of 1500 m above the ground level. The researchers calculated the atmospheric stability conditions using measurements from a sonic anemometer. They also measured temperatures at two sea points to calculate the land-sea gradient.
Results were compared to several studies in the area, including a similar study previously performed at the same location (Dieudonné et al., 2023). One of the main contributions of the study is that they extended the observational range up to an altitude of 1500 m, indeed multiplying by five the observational range in the previous study (up to 300 m). Consequently, they were able to obtain more accurate results, as they could account for high altitude LLJs previously missing. The authors also used innovative means to compare the impacts on wind turbines of LLJ and non-LLJ winds.
The manuscript presents a topic of high interest and potential practical use, especially to guide the location of future wind energy projects and the sizing of wind turbines. This is particularly important for the region, where energy plays a strategic role. The article is promising; however, there are several concerns that merit the authors attention. I would recommend the authors to address/respond the following comments:
Major comments:
- The way the authors define and reference the continuity criteria used to select LLJs is not well organized. The authors use two continuity criteria, one in the time scale and another one in the space (altitude) scale, but their uses lack standardization throughout the document.
Line 109: “… a continuity criterion was introduced …”.
The authors should consider specifying that this is a ‘time continuity criterion’, to differentiate it from the ‘height continuity criterion’ loosely described at the end of the same paragraph. If we follow the convention shown in the 1st column of Table 3, the durations of ‘1.5 hours’, ‘30 minutes’, and ‘1 hour’ are not different criteria, but the possible values used for the time continuity criterion.
Lines 113-114: “… LLJ core heights were required not to vary by more than 50 m between two consecutive profiles.”
Since the 2nd column of Table 3 shows not just ’50 m’ but also ‘100 m’ and ‘N/A’, this sentence is loosely defining, without naming it, the ‘height continuity criterion’. Again, and following the convention shown in the 2nd column of Table 3, the height differences ’50 m’ and ‘100 m’ are not different criteria, but the possible values used for the height continuity criterion.
Since subsection 2.2 is where both continuity criteria are mentioned for the first time in the document, this subsection should explicitly define their names and describe all their possible values.
Lines 172-173: “With the same objective of excluding isolated jets, a continuity criterion in height between two consecutive profiles was introduced.”This definition of the height continuity criterion should probably belong to subsection 2.2 (part of the Methodology section), not in the Results section. Here, it will probably suffice to say that the results soon-after correspond to the different values of the height continuity criterion.
Line 186: “… when using the same continuity criteria (Dieudonné et al., 2023).”
Dieudonné et al. use both continuity criteria (time and height) and even a direction continuity criterion. I assume that the sentence means using the same time continuity criterion of 1.5 hours and the same height continuity criterion of 50 m. Am I assuming right? If that is the case, it would be great to explicitly specify it.
Lines 260-261: “… same continuity criterion …”
It’s inferable from the context, but explicitly saying ‘… same time continuity criteria …” will make the meaning instantly clear.
Line 389: “… the use of different continuity criteria.”
The context leads to infer that this is attributable to different threshold of the time continuity criterion. Am I assuming right?
- The authors are not consistent describing the minimum selection criterion.
Line 176: “Initially, a criterion on the wind decrease below the jet was also applied …”.
Since wind speed always decreases below the jet peak, I infer that the authors mean “Initially, a minimum selection criterion was also applied, as seen in Figure 2.” Am I inferring correctly? If that is the case, the sentence should be corrected accordingly. Additionally, the authors should consider describing the criterion here, as they did later in the caption of Figure 2.
Lines 176-177: “… (Andreas et al., 2000; Karipot et al., 2006; Tuononen et al., 2017).”
- Andreas et al. and Tuononen et al. mirror the main criterion (that the local maximum is at least 2 m/s stronger and at least 25% stronger than the local minima above the core) but applied also to the local minima below the core. That is slightly different from what is shown in Figure 2 where the threshold is just 1 m/s. If the criterion description from both references fits better what the authors applied as ‘minimum selection criterion’, they should explain accordingly in the manuscript.
- Karipot et al. analyzed an intermittent LLJ on “the night of June 25–26, 2004”, but they don’t mention the criteria used for their selection (or at least I can’t find them). The authors should consider if the citation is relevant in this place.
Line 181: “… no criterion was imposed on the wind decrease below the jet.”
The assertion is not consistent with what is shown in Table 3, 3rd row. Maybe the authors mean that the minimum selection criterion was not used for the final conclusions. Please, confirm.
Table 3 (general caption and 3rd column header): “… and criterion on the wind decrease below the jet.”; “Wind decrease below the core”.
Since wind speed always decreases below the jet peak, I infer that the authors mean the ‘minimum selection criterion’. Am I inferring correctly? If that is the case, the authors should update accordingly.
Minor comments:
- Lines 22-23: “… LLJs complicate both forecasting and energy production due to the fluctuations in wind speed and wind direction that they cause.”
The word “fluctuation” (i.e., change with time) appears imprecise here. It leads to infer that wind speeds and directions exhibit more turbulent behavior during LLJs than during unstable regimes. LLJs tend to be quite coherent structures, interrupted by intermittent burst of instabilities. From the reference: “… during the nighttime (i.e., the hours when the LLJs are present), simulated TKE is fairly intermittent and characterized by a turbulent bursting type nature (Nunalee and Basu, 2014).” The authors should clarify if by ‘fluctuation’ they mean this intermittent behavior. - Lines 102-103: “… a wind speed fall-off of at least 2 ms−1 and 20 % compared to the minimum above.”
This is understandably a rephrasing of the the-facto criterion ‘a wind speed maximum that is at least 2ms-1 and 25% faster than the next minimum above.” The criterion is defined relative to the maximum strength as follows: (max-min)/max=0.2 (the 20 % criterion), which is the same as saying max=min*1.25 (the 25 % criterion). The rephrasing shouldn’t give the impression that there is a fall-off from the minimum value. My personal suggestion would be, for example, “… a wind speed fall-off of at least 2 ms−1 and 20 % from the LLJ maximum to the minimum above.” - Line 127: “ERA5 spatial resolution is 0.25◦…”.
They have several products with different resolutions. Apparently, 0.25◦ corresponds to the high-resolution deterministic reanalysis, grided from the native resolution of 1◦. That is fine, but would it be possible to clarify this? - Lines 137-139: “Three wind turbine models (Table 2) … (i) the most common offshore turbines in the North Sea, (ii) the most advanced turbine model currently available, and (iii) the turbine model projected for future uses (Global Wind Energy Council, 2024).”
Since the information on Table 2 includes rotor diameters and hub heights, I assume that the authors are referring to the figure in page 53 of the report cited. I also assume that the case (i) corresponds to the year 2020, the case (ii) to the year 2023, the case (iii) to the year 2030, and that the hub height was calculated as the tip height minus half of the rotor diameter. May the authors confirm? If that is the case, they may consider adding the page number to the citation, to help readers to understand how numbers in Table 2 were obtained. - Following on the previous comment, the figure in the reference shows trends worldwide, not just in the North Sea. Therefore, I suppose that the authors have additional information that the most common offshore turbines in the North Sea are the ones that were trending worldwide in 2020. May they confirm?
- Line 162: “… the shear was divided into two contributions: from the upper blade tip to the hub and from the lower blade tip to the hub.”
Following on that sentence and noting in Figure 14 that positive shears are infrequent for the blade in the lower position (in both LLJ and non-LLJ situations), I assume that the term ‘wind shear contribution’ is the wind shear as ‘felt’ by an observer moving from the blade root to the blade tip. for example, for the blade in the lower position the average contributed wind shear on the blade would be cWShear = (windspeed@lowertip – windspeed@hub) / (height@lowertip – height@hub). Is my inference correct? Wind shears exert distributed loads (bending moments) that contribute mechanical shears on the blades; authors should note, however, that compounded interactions are more complex in a 3-blades wind turbine. - Lines 189-191: “Cases presenting two simultaneous jets at different altitudes were very rare … 4.1 % of the jet profiles and 0.64 % of time.”
“Very rare” is maybe a too strong conclusion for a 4.1 % occurrence. That percentage, although small, is significant. More cautious phrasing would be “infrequently” or “ quite uncommon”. - Table 3, 4th column: “Jet detection height”.
The caption can confuse some readers as they may infer that those are the heights where LLJs were detected. Maybe “Maximum scanning height” is more precise. - Figure 4 caption: “The black line represents the coastline direction, with the sea located on the upper part and the land on the lower part.”
There are two black lines in each subfigure, one labeled ‘Coastline’ and the other one with no label. The labeled black line appears to be parallel to the coastline seen at the right in Figure 1b. What is the meaning of the unlabeled black line? - Line 236: “… northeasterly and easterly LLJs followed the global distribution pattern … “.
While that appears to be true for NE LLJs (the yellow dashed line), Figure 7a apparently show a different pattern for E LLJs (the green line). The global distribution peaks on the months of June-July, while the E LLJs distribution peaks on the month of September. The description merits a reconsideration. - Line 238: “… they occurred over a shorter period …”
Since the number of NW LLJs (the blue line) outside the period is not negligible, a milder claim may be more precise, something like “… they tend to occur over a shorter period …” - Line 242: “High-speed LLJs occurred year-round, but their peak occurrence was observed in March and October …”
I infer that the authors mean WS> 15 m/s (the yellow areas). Figure 7b appears to show that those jets are mostly observed during the months of October, April, and March (in that order). Consequently, April is more prone to high-speed LLJs than March. This observation doesn’t contradict the authors description that high-speed LLJs tend to peak during the periods where S LLJs are more frequent, especially considering that the frontier between April and May is artificial, but omitting April seems a little forced. - Figure 8: “… which closely corresponds to the local solar time.”
‘closely’ is imprecise. More informative is to indicate that the local time in Dunkerque is UTC+1 in normal time and UTC+2 during daylight saving time. - Lines 261-262: “… with LLJs less than 3 hours apart being considered as a single event … The longest event lasted 23.6 h, and the average duration was 5.4 h.”
LLJ events can last much more than 3 hours, that is a fact. Therefore, those results are completely reasonable. The problem is the 3-hour method that is not sufficiently described. In a 23.6-hour event there are many profiles that are separated by more than 3 hours and nevertheless were considered (correctly) as part of a single event. I suppose that the method worked like a rolling window in which LLJ profiles separated by 3 hours or less were considered belonging to the same event, even if there were intermediate non-LLJ profiles. Am I assuming correctly? - Figure 10: “Distribution of the Pasquill atmospheric stability classes …”
It is interesting to see LLJs developing in extreme unstable conditions. Since the Monin-Obukhov stability parameter (used to determine the Pasquill stability classes, subsection 2.3) was calculated using measurements from the ultrasonic anemometer located at the low altitude z=15 m (line 96), maybe there is a residual layer above where LLJs develop. Or this specific LLJ type is explained by a mechanism other than Blackadar alone (for example, sea breezes). - Lines 388-389: “… an average duration of 3.1 h, which is shorter than the 5.4 h obtained in the present study. This difference may be attributed to the use of different continuity criteria.”
That explanation is reasonable, as well as the explanation given soon after. On the other hand, since the longest event was also greater in the new study (23.6 h versus 20 h 50 min in Dieudonné et al.), would it be possible also that high-altitude LLJs tend to last longer? - Lines 406-407: “… these jets would correspond to purely local sea breeze cases, without contribution of the wind channeling into the Dover Strait.”
As stated soon after, there is probably a second mechanism; therefore, the phrase “would correspond to purely local sea breeze cases” to encompass all NW LLJs may result inaccurate. Since Figure 11 shows that NW LLJs occurs through a wide range of temperature gradients (both negative and positive), would it be possible that NW LLJs observed during positive temperature gradients might be due to purely diurnal sea breezes, and NW LLJs observed during negative temperature gradients might be due to purely nocturnal inertial restoration (i.e., Blackadar mechanism)? - Appendix A. Data availability.
Figure A1 and Table A1 are positioned before the section heading.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 206 | 38 | 19 | 263 | 20 | 23 |
- HTML: 206
- PDF: 38
- XML: 19
- Total: 263
- BibTeX: 20
- EndNote: 23
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Please see attached pdf