General Comments:

The study analyzes 85 Lagrangian surface drifter trajectories available for the eastern-central North Sea and the Skagerrak. The drifters were released during 17 deployment campaigns, and cover the period from 2017 to 2021; they are uniformly structured in standard NetCDF format.

First, tidal energy spectra and tidal ellipses are derived, demonstrating significant differences between the shallow areas of the eastern North Sea and the Skagerrak. Second, the data are employed to derive gridded surface current distributions in a Eulerian framework.

Altogether, the paper is written in a clear and concise manner. Also, the English language is used appropriately. Moreover, the surface drifter data are definitely worth to be published, since the provide a complementary view of the ocean surface circulation, when compared with data from mooring stations or from model simulations.

However, at present the paper shows a severe weakness, which should be accounted for, before it is suitable for publication in ESSD. 

The benefit of the gridded data is extremely questionable. Hence, its calculation and presentation make no sense, and therefore, it should be removed from the manuscript. There are three major reasons for this criticism:

1)    Mesoscale dynamics occur in a spatio-temporal space, and hence, a temporal average over several months    or even years, as done in this study, smooths out most of the mesoscale-structures. Looking at the 15 km resolution gridded data in Figure 8, this problem becomes evident. In fact, only the very coarse cyclonic Skagerrak Gyre circulation is visible. This averaging problem becomes even more obvious, when looking at the finer resolution results. It is apparent, that the 7.5 km, 5 km, and 3 km resolution results do not provide any significant additional information. When looking at the higher resolution results, only the impact of the interpolation scheme can be observed.

2)    The argument that compared to model simulations, these gridded data with a 0.125 degrees resolution can provide an improved understanding with respect to the mesoscale dynamics in the North Sea is not acceptable. First, the severe problem mentioned above regarding the temporal averaging of different mesoscale patterns does not occur in model simulations. If required, models do provide results with a temporal resolution even on the subtidal timescale. Moreover nowadays, the spatial resolution of most North Sea models is of the order of 3 km (see e.g. Paetsch et al., 2017), which is nearly one order of magnitude better than the standard resolution of the gridded data, which are presented in this study.

3)    When converting Lagrangian data to a Eulerian framework you always face the theoretical problem, of how to deal with the Stokes drift, which is inherently included when averaging Lagrangian data over the wind wave scale and/or the tidal scale; both is actually done in the current study. Hence, a discussion about the treatment of the Stokes Drift, when converting drifter data to a Eulerian gird, is definitely necessary. However, at present, this issue has been totally ignored by the authors.

Having in mind these severe problems mentioned above, it is clear that all aspects related to the gridding of the drifter trajectories must be deleted in the manuscript. In contrast, the authors should focus on the real advantages of their very attractive drifter data set. Firstly, the data are an excellent source for model validation. Single drifter trajectories are ideal subjects to be compared with model tracer trajectories in a one-to-one comparison study, using the same starting point and the same time period in retrospective tracer simulations. Secondly, drifter data can nicely be used to derive dispersion properties, as for example performed in Ricker et al. (2022). Since, ideally in hydrodynamical simulations the dispersion rate has to be calibrated for each specific model area, this kind of independent dispersion information can be very helpful for numerous model investigations.

Review of „Lagrangian surface drifter observations in the North Sea: An overview on high resolution tidal dynamics and submesoscale surface currents“ by Lisa Deyle, Thomas H. Badewien, Oliver Wurl and Jens Meyerjürgens.

Anonymous reviewer #2

13.12. 2023

In the manuscript, a data set derived from Lagrangian surface drifter observations is presented. The drifters were deployed in the south-eastern North Sea, namely the German Bight, and along a section across the Skagerrak, from 2017 to 2021. They will be publicly available in standard NetCDF format at the Pangaea data base.

The study consists of a tidal analysis for the data set, deriving tidal energy spectra and tidal ellipses. Here, distinct differences for the shallow shelf areas in the German Bight and along the Danish east coast versus the deep Skagerrak are displayed. To present the mean residual circulation in the research area, the residual velocities were derived along trajectories and transferred into an Eulerian representation by calculating box-averages on a regular grid. For the Skagerrak, the grid resolution was reduced step-wise from 15 km to 3 km to get insights into the meso- and submesoscale components of the surface circulation.

General comments:

The data set presented here is definitely valuable for a detailed insight into the surface currents on different temporal and spatial scales and should be published. However, the manuscript itself and analysis presented therein have a number of weaknesses. Thus, altogether I suggest major revisions of the manuscript.

1. The English language, in some parts, is not satisfying. I recommend to work through the whole text with a native speaker.

1. The presentation of the used instrument and data set is too scarce. Of course, in Meyerjürgens et al., 2019, a detailed description of the development of the instrument, the measurements and the sampling is given. Yet, Meyerjürgens et al., 2019 is only based on a subset of the data presented here and the reviewed text should stand on its own, as far as the basic information is concerned.

For example:

Determining the position of the drifter and transmitting the position data ashore are key components of the measurements. However, these two components are not correctly described in the text (see lines 129 to 130 and for comparison Meyerjürgens et al., 2019 “Positioning and Telemetry”.

The picture of the instrument in this manuscript (Figure 2) is less helpful than the pictures of the instrument in Meyerjürgens et al., 2019 (Figure 2).

A reflection on the seasonal coverage of the data set is missing. On top of that, other statistical information such as the number of individual measurement from which the box-averages (Figure 5) were calculated is missing.

The manuscript refers to Lilly and Perez-Brunius, ESSD 2021, which is a good example for how to present a data set of this kind.

1. When dealing with different scales of motion, I expect at first some reflections on the time and length scales of the different processes in the research area and considerations if and how these processes can be observed with the data set at hand. See for example Cushman-Roisin and Beckers, Introduction to geophysical fluid dynamics, Elsevier 2011, or search for “baroclinic Rossby radius”.

It is definitely not possible to sort out different scales by only decreasing the grid-cell length!

I will not go into detail for this part of the manuscript but also go along with the comments from reviewer #1.

1. Figure 5 a/b: Although a and b are meant to show the zonal/meridional component of the velocities, the arrows are similar in both pictures. I recommend having only one picture and show in colors the magnitude of the absolute value of the velocity and the direction; red – eastward, blue – westward.

Individual comments:

English language:

Line 45: I suggest “..consisted of mooring arrays with…” instead.

Line 65/66: I don’t understand the logical implication of “.. resulting in..” in this sentence. Please rewrite.

Line 68: I don’t understand the logical implication of  “..due to..” in this sentence. Please rewrite.

Line 92: I don’t understand the logical implication of “In addition” in this sentence. Please rewrite.

Line 92: “These are partly due to nonlinear tidal interactions.” I don’t understand this statement. Please explain and rewrite.

Line 122: I suggest “..and measured for up to several month”.

Line 180: “…a grid is created that is composed of several cells.” I have no idea of what else a grid can consist of. This part of the sentence give the same information two times. Please rewrite or rather shorten.

Content:

Line 31: The North Sea is also influenced by river-runoff from Forth, Humber, Thames, Seine, Meuse, Glomma. See for example “North Sea Region Climate Change Assessment”, Editors Markus Quante and Franciscus Colijn, Springer 2016. DOI 10.1007/978-3-319-39745-0.

Line 36/37: The residual currents are driven by density gradients and wind.

Line 51-53: This is basically correct, but whether or not the data set is able to resolve small scales of motion depends on the time increment of the measurements. And this also holds for Eulerian measurements. This part is not really describing the difference between Lagrangian and Eulerian measurements. Please rewrite.

Line 75. Please write southwest instead of south.

Line 81-83: Please add a reference.

Line 87: I suggest giving all numbers in m/s and not changing the units within the paper.

Line 93-94: Interaction needs different components: please specify.

Line 103-104: “… the residual circulations are assumed to be smaller compared to tidal currents and wind-induced currents”. Please rewrite this sentence, because wind-induced currents are part of the residual currents.

Line 114: Please specify why this drifters can be deployed in shallow water and others cannot.

Table 1: Is the number of trajectories similar to the number of instruments?

Line 134: I don’t understand the terminus “typical circulation trend” – do you mean “mean residual circulation”? Please explain.

Line 135: It should be: “moved westward”.

Line 173: I have never heard about “the complex-valued velocity”. The vector of the surface velocity consists of a zonal and a meridional component. But why is it complex? Please explain.

Line 188/89: How many data gaps (percentage) in the grid have been filled by this method?

Line 211 and 220: Please explain why you need a time series of “at about 14.2 days” and “at least 10.4 days”?

Line 243/244: I think “which results in” is the wrong logical implication.

Additionally, you can have “long residence times” in a grid cell, if velocities are small. Thus, you have to explain this statement or show it is proven.

Line 248-252: The naming “eastern inflow of the North Atlantic" is not correct, especially if you are dealing with surface currents.

I miss some information about variability of the features “inflow of the North Atlantic”, “Norwegian Coastal Current” and “outflow from the Baltic Sea”. They are highly variable and thus averaging about all data in a grid cell, independent of the time resolution/coverage can produce artefacts (more information about this topic/problem can be found at  Lilly and Perez-Brunius, ESSD 2021 and Lilly and Perez-Brunius, Nonlin. Processes Geophys, 2021 from your references).

I don’t see any “small eddies” in the map (Figure 5).

…

Line 260/261: “..which means that the north-south motion increases”. This is a change in direction of only 6%. I suggest to write “slightly increases”.

Line 260: see Line 87.

Line 270/271: Please explain or rewrite: “Other irregularities are difficult to distinguish from the rest of the noise”. What do you mean with “the rest of the noise”?

5 Discussion, first paragraph: It is difficult to distinguish between general statements, taken from the literature, and results from the analyses of the presented data set. Please clarify. From this study we do not “gain an understanding of the transition processes between the mesoscale ocean circulation and” smaller scales.

I leave out comments on the rest of the manuscript.