Measurements and Modelling of Offshore Wind Profiles in a Semi-Enclosed Sea

Round  1

Reviewer 1 Report

The manuscript deals with wind and turbulence profiles at Ôstergarnsholm. It is a well written manuscript. Especially the part on LLJ is very interesting and merits publication in itself.

A 30° cone continuous scanning lidar is used in this study. This type of lidars scans in the vertical by focusing the lidar beam at prescribed measuring heights. As a consequence the probe length (contrary to pulse lidars) becomes a strong function of height. The probe length is small for low heights and increases very much as function of height. This large probe length at large heights hampers the estimation of fluxes ( wind profiles are not so sensitive) from lidars of this type. Furthermore also the horizontal averaging increases  with height. In consequence the huge difference between the lidar and WRF estimated results illustrated in Fig. 9 (a  + b) to a high degree might be caused by way the lidar operates by smooting the signals in both horizontal and vertical directions. This also explains to good comparision between lidar and tower measurements of the momentum flux in Fig. 9b.

It is noted that the estimation of TI is not explained, which suggest that the estimates are taken from the output from the lidar and thus based on a proprietary and unpublished routine by the factory of the lidar. This point is not clarified in the manuscript however, and the authors needs to consider if they want (and dare) to use results from a black box routine.

Fig. 7 shows a comparison between lidar and WRF wind speed estimates. The comparison has a correlation coefficient of 0.65, which is surprising low and much lower than found in similar comparisons of measurements off-shore. It is hinted in the manuscript that the roughness at Gotland is too low in WRF – this raises the question why publish results from WRF runs that you already now consider are not correct. Additionally some of poor behaviour in Fig. 7 and 9 might be caused by the roughness problem in WRF.

Line 249:  Looking at table 1, the bias between 300 and 100 m is smaller in 6 cases and larger in 6 cases,  contradicting what is said on line 247.

Conclusion:

I am very happy with the part of the manuscript that deals with LLJ, which merits publication. However the parts dealing with TI and momentum fluxes should not  be published because of the inherent problems estimating fluxes from lidar continuous wave wind lidars. Furthermore there seems to be problems in the WRF simulations as well, which to an unknown degree might influence the turbulence parameters derived from WRF, both can contribute to the poor comparison shown in Fig. 7.

My final recommendation is to leave out the presentation and discussion on TI and momentum transfer, the exiting analysis of the LLJ is more than enough to justify the publication of the manuscript.

Author Response

We agree with the reviewer that there are some uncertainties in the LIDAR turbulence data that were not properly discussed. We removed all but one of the figures dealing with turbulence intensity, but we kept the parts about momentum flux since we feel that we have a better understanding of the errors in this parameter, and we also added more discussion on the errors. The answers to each of the reviewers questions are found in the attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

This work presents an elaborated study about LLJ monitoring both with a remote sensing method and WRF model. At first sight, this paper seems to be a very accurate scientific report [with advantages and disadvantages]. The novelty [or motivation] of the study must be improved, specially that a many study in the specific field are already reported. In this order, I kindly ask the authors for an improvement of the introduction part [for a better motivate their study].

About the LIDAR system, it is able for up to 300 m measurements? Please cite. On the zxlidar website [https://www.zxlidars.com/wp-content/uploads/2018/09/ZX_Lidars_Product_Guide_2018-ZX-300M-Summary.pdf] in not specified that the range up to 300 m lidar measurements.

Line 97: “The measurement heights in the present study are chosen to 28, 39, 50, 100, 150, 200, 250 and 300 m above ground surface”

-        Please clarify if it is 28 or 29! It looks like a value of 29 m [for LIDAR measurements] that is used throughout the study;

-        The values of 39 m and 50 m were used only for time series distribution [fig 14]? If yes, please specify that.

In figure 13, if I well understand, the first day from June 2017, is represented on the both panels. If so, the data doesn’t correspond.

The figures must be rearranged. It is quite difficult for readers to follow the text and figures in the present format.

I recommend to the authors for a short update on the conclusions chapter, for example how all these results can improve the WRF model….by presenting some perspectives will make this study, more interesting. As example, an offshore wind measurements lidars network can improve the model accuracy and both the turbulence coefficients and wind parameters can be used as input parameters for an artificial neural network. It's up to the authors if they will consider this recommendation.

Author Response

We have worked to improve the introduction and conclusion sections in order to motivate the study more carefully and make it more coherent. The answer to each of the comments are included in the attached file.

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper uses meteorological tower, lidar and WRF model data to examine the marine boundary layer at Gotland Island.  The study examines wind, turbulence and momentum flux profiles, and looks at low level jet occurrence, and in particular the representation of these jets in the WRF model.  The study concludes the WRF model representation of low level jets needs improvement.

Overall it is a neat introductory study on a years worth of data.  As data collection continues this will hopefully form the basis for larger data set studies, perhaps comparing other models and parameterisations, other instruments, and potentially look at climatological changes in wind speed and direction, and jet occurrence.

Major corrections

Page 1 line 24 – not sure “adjustment time” is the most appropriate term.  Similarly on line 25 “non-standard”.  By “adjustment time” I think you might mean “response time”, and “non-standard” I think perhaps “wind profiles with inflection points at one or several heights…….” would be better wording?

Page 2 line 79 onwards – it might be helpful to include a plot of which wind directions are included and which aren’t.  I ended up sketching one myself as I found it difficult to visualise from numbers what was what.  This could potentially be overlayed on the map in figure 1 right.

Page 3 lines 90 to 97 – you have chosen certain lidar levels for comparison to other instruments, but what height levels are available?  I.e. what is the height resolution of the lidar?

Page 5 line 165 – “speed above” is somewhat ambiguous.  Does this refer to the range gates of the lidar?  And how many range gates above the height in question are considered?  The same question applies to classification 2 on line 168, and classification 3 on line 170.  If the jet occurred at 300 m, such that there was no lidar data available “above” this height, would this jet not be recorded?

Page 5 line 170 – why does this classification permit a wind speed difference of 2 m/s, where classifications 1 and 2 only 1 m/s?

Page 6 line 205 – what does the “it is” refer to in “It is especially prominent…..”?  Do you mean the incorrect winds in the lidar?

Page 7 figure 3(a) – it is generally best to make the x and y axes scales match, so the line of best fit runs through the middle of the plot, as has been done in (b).

Page 7 figure 3(b) – wouldn’t it be possible to correct the direction wrap around effect at +/- 180 by shifting the lidar data +/- 180 to make it align with the tower?  This seems to have been done for Figure 7?

Page 9 figure 7 – you mention there are 12 levels of WRF data below 100m, can you comment on why you chose 100 m for the comparison in this plot?  What does the data look like at other levels?  In particular what does the comparison look like around 29 m where you can also compare to the tower data?  You go on to discuss 300 m height, but did you compare all available heights?  You say on page 9 line 248 the wind speed difference increases with height, is this strictly true, or just in comparison of 100 m to 300 m?  Could it be the 300 m data is influenced by another factor?  Perhaps make this point a bit clearer to the reader.

Page 10 line 273 – you note the magnitude of the momentum flux profiles are higher in WRF, can you provide a comment as to why this might be?

Page 11 figure 10 – can you provide a comment on changes in magnitude with different conditions?  Similarly why does the lidar data show a reduction in magnitude above 100 m in very stable conditions?

Page 13 figure 12 – I assume these data are from the entire year?  Or are they from April-July?  Please make this clear.  Also, would it be possible to either plot a standard deviation or similar, something to indicate the spread in wind speeds averaged?

Page 14 figure 13 – the filled horizontal areas are labelled as lowermost, middle and uppermost, where they could be referred to by their colours.  I’m also not entirely clear on the point you are trying to make including the data at 100 m?

Minor corrections

Page 3 line 90 – repeated “the”

Page 3 line 104 – “not included” would perhaps read better than “taken away”

Page 4 line 127 – “….is determined and by fitting a Gaussian…..” remove “and”

Page 4 line 131 – insert spaces on either side of the dash in “2016-7”

Page 4 line 133 – figure 1 is labelled as left and right, there is no a

Page 4 line 137 – insert “a” in “….shown that it takes A long time….”

Page 4 line 138 – insert “been” in “….which has BEEN shown to give….”

Page 4 line 139 – remove “th” from the 6

Page 4 line 140 – add an “s” to hour

Page 4 line 143 – missing comma after Dudhia shortwave radiation scheme

Page 5 line 156 – insert “the” before “sea surface”

Page 5 line 158 – add an “s” to “height”

Page 5 line 160 – insert “the” before “sea surface”

Page 5 line 161 – insert “the” before “colder sea surface”

Page 5 line 169 – these criteria are first called “classifications” on line 164, but are then changed to “criterion”.  It is a minor point but consistency is good.

Page 5 line 178 – replace “is” with “are” in “….from WRF IS calculated….”

Page 7 figure 3 – remove the capital T from “….1:1 ratio, The equation…….”

Page 8 figure 4 – the caption states “lighter colors”, but it all appears to be in one colour?

Page 9 line 204 – “island Gotland” should read “Gotland island” or “island of Gotland” throughout

Page 10 figure 9(b) – it is hard to tell the difference between the lidar dashed line and the WRF solid line.  Perhaps using a different colour or a more broken line would help?  This is not so much of a problem in (a), and is clearer on screen than on a print out.

Page 10 line 258 – remove “the” and add “s” in “….divided into THE sea and land sectorS.”

Page 10 line 269 – move “also” to between can and clearly so that it reads “….behaviour can ALSO clearly be seen….”

Page 12 line 309 – insert “the” in “….largest during THE nighttime….”, and similarly for line 310

Page 12 line 311 – sentence beginning “This indicates that LLJs are commonly…….” Needs to be re-worded.  The meaning is not clear.

Page 12 line 317 – insert “a” in “….there are only A few cases….”

Page 12 line 318 – replace “by” with “in” in “….hard to capture IN the model.”

Page 14 line 353 – please state which day around 15 LST the LLJ forms

Page 16 line 385 – move “also” to between “can” and “be” so that it reads “….height can ALSO be seen in….”

Page 17 line 416 – insert “the” between “to” and “wind” so it reads “….of importance TO the wind industry….”

Page 17 line 425 – replace “point to” with “suggest” so it reads “observations SUGGEST that….”

Page 17 line 432 – remove the “s” from “addition”

Page 17 line 442 – either reword or remove “compared to now”

Page 17 line 452 – replace “difficulty” with “difficulties”

Author Response

We thank the reviewer for a thorough review, in which we improved the manuscript by complying to most of the remarks. We include the answers to each of the comment in the attached file.

Author Response File: Author Response.pdf

Round  2

Reviewer 1 Report

na

Reviewer 2 Report

The authors have solved all the problems raised by me.The article may be published.

Reviewer 3 Report