This paper provides a good overview of a data set consisting of important properties of the lower atmosphere during the MOSAiC expedition, combining measurements from radiosondes, a surface micro-meteorological tower, solar and IR radiometers, and a laser ceilometer.

Synthesis products of this sort provide a valuable framework for the analysis of other measurements such as surface turbulent exchange and the surface energy budget, aerosol properties, boundary layer clouds, gas phase chemistry, etc. For a large, multidisciplinary project, such as MOSAiC, it is particularly useful for the many different studies that may require one or more the properties documented in this data set to all use the same data or definitions and avoid the potential inconsistencies that would result from multiple groups all calculating various parameters independently. The authors thus provide a valuable service to the MOSAiC community with this data set.

The paper is well written, and clearly documents the procedures used to define the various quantities, and the quality control applied. I am happy to recommend publication with only minor editorial revisions.

Detailed comments:

L32 – recent estimates of the rate of Arctic warming are even higher than this, up to 4 times global mean (Rantanen, M.; Karpechko, A.Y.; Lipponen, A.; Nordling, K.; Hyvärinen, O.; Ruosteenoja, K.; Vihma, T.; Laaksonen, A. The Arctic has warmed nearly four times faster than the globe since 1979. Commun. Earth Environ. 2022, 3, 168)

L69: “in hopes that the dataset will be useful…” – perhaps “in the expectation that…” would be more appropriate phrasing

L110: re: flow distortion around ships, Achtert et al. (2015) includes CFD model estimates of the impact of flow distortion on wind profiles immediately above a research vessel. Berry et al (2001), while a pioneering study, focus on the, more extreme, impact on wind measurements on the ship itelf.

Achtert, P., I. M. Brooks, B. J. Brooks, B. I. Moat, J. Prytherch, P. O. G. Persson, and M. Tjernström. 2015: Measurement of wind profiles over the Arctic Ocean from ship-borne Doppler lidar. Atmos. Meas. Tech. 8, 4993-5007, doi:10.5194/amt-8-4993-2015

L176, delete “, for which an example was given in Sect. 2.2”, it’s unnecessary repetition.

L188-190: delete “, averaged between 5 minutes before and 5 minutes after the time of launch, following the same time interval format as the met tower data, for which an example was given in Sect. 2.2” – again, unnecessary – you already stated that this was done in “the same manner as for the met tower”, further repetition of details isn’t needed.

L216: on the issue of classifying inversion layers as distinct or a single inversion, Tjernstrom and Graverson (2009) is also relevant (https://doi.org/10.1002/qj.380)

Section 3.1 – The discussion of identifying temperature inversions is fine, and the inversions are of importance in their own right. It is perhaps worth noting, however, that from the perspective of stability (and linking to the Richardson number as an indicator of stability), it is not the presence of an inversion - a positive temperature gradient (regardless of threshold used in classifying it as such) – but the gradient relative the adiabatic lapse rate that is important…or the gradient in (virtual/equivalent) potential temperature.

L270: “Wind speed is provided in components so the user may calculate total wind speed as well as wind direction” – this is a bit arbitrary. If the user is interested simply in the wind (speed and direction, or its components) profile (values at specific heights and times) then it makes no difference whether you provide the components or speed and direction – they have the same storage requirements, and it’s easy to convert either to the other. If any averaging is required (in altitude or time) then it is much easier to work with components – avoiding the problem of averaging direction across the 0/360 wrap.

L279: “Rib is the ratio between buoyantly (from thermals) and mechanically (from wind shear) produced turbulence” – this is a little misleading. Rib is the ratio between buoyant and mechanical forcing, rather than turbulent production…for stable conditions there is no buoyant production, there is still a (negative) forcing. I suppose one could argue that this is a negative production of turbulence, but the word ‘production’ implies a positive value.

L324: “Stability regime from the…” -> “The stability regime from the…”

This manuscript nicely introduces and describes an aggregated dataset produced from some of the meteorological observations at the year-long MOSAiC drifting experiment. Thermodynamic measurements from frequent radiosondes and from a meteorological tower installed on the sea ice were combined with ceilometer retrievals and surface broadband radiation. These quantities were further used to calculate the bulk Richardson number, atmospheric stability, to locate temperature inversions, and to quantify the presence of low-level jets.

The combination of key measured variables together with derived quantities provides a solid tool, useful to characterize the Arctic atmospheric boundary layer over RV Polarstern during the whole MOSAiC expedition, and provides a valuable common ground that reduces the risk of inconsistencies.

The manuscript is well written and describes in detail the methods that were used to process the data, providing also useful examples. I recommend it for publication with minor revisions.

Specific comments:

60 – I suggest changing to something like “... including the atmospheric boundary layer (ABL) height and stability ...” as I find that the original formulation is a bit vague, especially since other ABL features are listed afterward.

Table 1 - The True "10m" height of the wind seemingly has a typo in the ranges (9.9 - 1.1m) since the central value is 10.34m.

143-146 - It seems that the temperature and humidity setup listed in Table 2 do not match the text in these lines. The instruments in the text (HMT330 and PTU300) differ slightly from the table (HMT337 and PTU307). Additionally, the 2 m temperature is associated with the PTU in the text and with the HMT in the table.

270 - I tend to think that the final user could benefit more from the speed and direction rather than the u and v components, especially in the case of quicklooks to be compared with the LLJ parameters, which have a speed and direction format (Table 3). However, I also see the value of u and v for other purposes.

295 - radiosonde profile