Review in the attached pdf.

Review of the manuscript “Velocity of magnetic holes in the solar wind from cluster multipoint Measurements” by Henriette Trollvik, et al.

This manuscript considers the dynamics and propagation velocity of magnetic holes observed in the solar wind. Magnetic holes are now common structures in space and heliospheric plasmas. They have been observed within the magnetosheath, magnetotail, and in the solar wind upstream of Earth and other planets. They have also been detected to accompany switchback events at small heliocentric distances. As such, these structures have high scientific importance for plasma heating and the dynamics of the solar wind plasma. This manuscript is well-written and has taken an interesting approach to perform a statistical study of these structures through using data from the Cluster mission collected during an early phase of the mission.

However, there are aspects in this work that need further consideration and improvement, and the paper has the potential to make important contributions after additional work and resubmission. These points are listed below:

The authors track the minimum magnetic field point to estimate the velocity of the structures. Tsurutani et al. (2011, cited in the manuscript) suggested that magnetic depressions (or magnetic holes) can change in size as they propagate. Particularly, rotational holes that are created as a result of reconnection in the solar wind can be very dynamic. Once the reconnection begins, it can continue to evolve for long periods and internal structures of the reconnection process can continuously change the boundary and the pressure balance. Yao et al. (2020, not cited) also showed that the magnetic dips can be expanding or contracting. Have the authors considered the possibility of contracting and/or expanding structures?

Yao, S. T., Hamrin, M., Shi, Q. Q., Yao, Z. H., Degeling, A. W., Zong, Q.‐G., et al. [2020]. Propagating and dynamic properties of magnetic dips in the dayside magnetosheath: MMS observations. Journal of Geophysical Research: Space Physics, 125, e2019JA026736. https://doi.org/10.1029/2019JA026736

Line 24: “… These structures have no velocity in the plasma frame but are convected with…”

How does this statement fits with the goal of this paper? It looks like to be the answer to the open question identified in line 37: “An important open question regarding magnetic holes is determination of their velocities in the solar wind frame.” Perhaps you can add in line 24 that “some studies” have shown that MH are stationary in the plasma rest frame.

Line 70: I suggest moving the discussion for magnetic hole event selection before discussing Figure 1. As it currently reads, it seems that the MH events are identified visually.

Line 77: The solar wind is a quasi-neutral flow. In the pristine solar wind, any real physical differences between electron and ion densities are immediately restored. As authors indicated the difference in measured densities are instrumental. However, the same plasma density should be used to calculate derived parameters, (i.e. Alfven speed, etc.). Can you comment on why you introduce these different instruments, and which one is the ultimate source of the plasma density in the study?  If the two measurements are complementary to give a better time resolution that should be stated in the text.

Also Line 77: “the latter being more reliable.”

This should probably be double checked. In Fig. 1, the electron density (black line) seems to be flat within the MHs, while ion densities increase, a typical signature in MHs.

Line 82-87:  Was it also a requirement for all 4 s/c to show a similar magnetic depression levels? It is also possible that different s/c might cross different parts of the solenoid/cylindrical structure of the MH. Can you comment on that?

Line 96-97: “The plane perpendicular to ….” This sentence does not make sense as written. A plane cannot represent the field strength. Please rephrase.

Line 107:  Is t_alpha,beta influenced by the size of the averaging (sliding) window discussed in line 82?

Line 107: What were the conditions for cross-correlation? Please comment of you applied a certain threshold for the correlation. It would be interesting to see how this cross-correlation limits the number of events and/or the level of error in determining the velocity. A paragraph describing this would be a god addition to the paper and helps to justify the importance of your conclusions.

Line 135: The number of events seems very small to do a statistical analysis. Is it possible to extent the period of study? At a later phase of the mission, the s/c trajectories moved to cover the solar wind, and all 4 s/c were still able to measure the magnetic field. If lack of plasma measurements is an issue, one possibility is to use the shifted omni data, for instance, to determine the Alfven speed.

Line 163: What is the significance of the mean velocity in the timing frame? It depends on the solar wind velocity and normal direction. Related to this, can the authors comment whether based on this study, these structures are still to be considered pressure balanced? If there is a perceived velocity in the plasma rest frame, meaning that the structure either is lagging or pushing forward, should this also cause a sort of asymmetry between the leading and trailing boundaries?