See comments in attached document.

Review of the manuscript „Brief communication: On the definition of the low-level jet“

By C. Hallgren, J.A. Aird, S. Ivanell, H. Körnich, R.J. Barthelmie, S.C. Pryor, and E. Sahlée

The article presents a suggestion for a definition of the low-level jet (LLJ) with the aim to harmonize  and make LLJ statistics of different sites easier to compare. The article compares statistics of the new methodology, which is based on wind shear below and above the jet core, with statistics based on the frequently used definition, which uses a threshold for the reduction of wind speed above the jet core. The results are compared for 3 offshore and 3 onshore sites of particular interest for wind energy.

The article addresses an important topic of making LLJ statistics comparable for different sites and methods. The new method is well justified and explained, and there are interesting results comparing the different sites. The paper is of large interest to the Wind Energy Community.

However, my main concern is that the new definition is only applied to simulation data. Such statistics is often determined from  observational data, e.g. based on wind lidar profiles. If this method should be used by a broad community in the future, it is of high importance to take into account measurement data and compare to the statistics based on former criteria. Measurement data are usually not as „smooth“ as simulation data, and there may be more variability of wind speed within one profile of wind speed. Also the treatment of lacking data from a certain height on, e.g. due to clouds, should be considered. If this provides similar results to the numerical data, and if the articles gives guidelines how to identify the LLJ form experimental data, e.g. in which height intervals, it will be much more convincing to the wind energy community.

The sites have been chosen at locations where observational data are available. So it requires large extra work, but should be possible, in principle. Therefore I would suggest to vote for „Major revision“ before the article is ready fo publication.

Some remarks are listed below.

1. 1: I would suggest to avoid using the term „non-ideal wind speed profiles“ in the abstract. It should first be defined what you mean with this. Maybe substitute with „logarithmic“
2. 17: please provide references for the different phenomena that are associated with LLJ
3. 31-34: The reason for the strong negative shear, low turbulence, entrainment and separation of layers is normally a temperature inversion. Maybe stat this explicitly?
4. 35-39: Please compare the „extreme examples“ to values of LLJ wind shear in the literature. E.g. for Emeis (2010) the wind shear values during LLJ events that were observed were only between around 0.03 and 0.095. So what you call „very strong shear“ is below the lowest shear of experimental data.
5. 55: your reults are based on ERA5. How well is ERA5 able to represent LLJ? Maybe there is a validation in the literature? If yes, please quote as a reference, if not, could you provide such an example yourself? Maybe at least exemplarily for one month for one of the sites?
6. 67: You suggest to make use of the height interval 10-500 m. For experimental data it would be good to have a suggestion how to calculate shear if this altitude range is not available. E.g. some lidars have an overlap issue and start measuring only from 40 m on, som have an upper range of 200m, etc.
7. 90 unclear, please rephrase: „On average the LLJ occur in weaker wind conditions than the climatology“

What about strong logarithmic increase of wind shear? Could this falsely be identified as a LLJ, e.g. in experimental data, when the wind speed minimum above the jet core is not covered by the data?