The paper presents a novel rocket-borne instrument MESS (MEteoric Smoke Sampler) for sampling of Meteoric Smoke Particles (MSP) in mesosphere and justifying its aerodynamic properties. The authors present results of aerodynamic simulations for the instrument flying at supersonic velocities and for MSP-flow through this environment into the instrument.

The paper is well-structured, and the results are clearly present.  

However, from the beginning and until the end of reading it is clear that the paper needs a proper proofreading (there are many typos and unclear sentences). Furthermore, physical quantities and their units must be separated by a whitespace (e.g., 85 km). I marked some of unclear sentences in the supplemented pdf.

Regarding science/technical questions:

1. I, personally, would be interested in the parameter “angle of attack”, which is mentioned but not addressed. For example, which angle (value) is a critical angle, so, that at larger angles no particle sampling is possible?

2. Which speed of particles inside the instrument is acceptable for a “successful collection”, i.e., for the particle to stick to the grid?

The paper investigates an instrument design to collect meteor smoke particles on TEM samples during a sounding rocket flight for post-flight analyses in a laboratory. It focuses on how these particles are transported from the environment, through the geometry of the instrument, onto the TEM samples, during the conditions of a super sonic flow. To track the particles a public model and an in-house model are combined and applied to different altitudes and payload speeds, taking into account that MSP can be immersed in ice particles.

A study on how particles interact with such an instrument is very important in order to optimize the outcome of such measurements.

General comments:

The manuscript could improve on language and precision from title to summary.

The manuscript is sometimes imprecise on whether ice particles or dust particles or large dust particles are at focus.

Specific comments:

Abstract:

Line 7: The dimensions given in the abstract are different from the drawing.

Line 14: Last sentence difficult to understand.

Section 1

Line 28+ : The PMSE / radar physics are not sufficiently (e.g. turbulence, bragg scale structures) or imprecise described (“ wavelength of scattered radars”), particle size might not play a direct role here.

The source of the assumed particle size distribution should be given more explicitly.

Line 36: It is mentioned that built-in dust probes are the only way for in-situ composition measurements of these particles, however an honorable mention of mass spectrometers is justified given the fact that mass spectrometers are flown into this atmospheric region since many decades.

Line 46 - 51: The goal for the paper should be described more precise, large dust, large MSP, ice particles, MSP in ice particles? IP abbreviation not introduced.

Section 2

Figure 1: The simplified drawing for the instrument in the simulation omits the lid completely. It seems as if the lid could have a significant impact on the flow and potential shadowing for certain angles of attack, especially for low apogee (low speed) mission profiles where the angle of attack is higher.

Line 74+: The mentioned pressure valve function is unclear: if the instrument opens after nose cone ejection and closes at apogee, how does it maintain the pressure at nose cone ejection in the instrument? Why is it needed at all?

Further if the pressure valve is venting the instrument during flight, the “air cushion” formed by the trapped air inside the funnel is considerable smaller as simulated in section 4.1 ?

Section 3:

Line 80:  grammar

Line 91: Since it is very unlikely that dust particles, sublimate through background gas collisions, the author means ice particles?

Line 133: The Knudsen number cannot characterize the mesosphere, as it is a relation between the environment and the length of a characteristic structure (payload or instrument).

Figure 2: little difference between the 2 panels in the figure, maybe possible to improve.

Section 3.3:  If the section only treats ice particles, this should be made more clear.

Section 4:

Line 156: the unit is “ms^-1“ not “m.s^-1”  throughout the paper

Line 168: the current atmospheric model is called NRLMSISE-00. The given reference is outdated and the name is insufficient but the value seem legit.

Figure 3 – 6 : While one figure nicely shows how the shock is formed around the instrument it is quite difficult to assess the simulation results for their numbers, e.g. the number density inside the funnel. A summary plot of a point in the funnel or at point of interest would probably be of more value and reduce the number of figures with rather similar appearance.

Figure 7: The figure is supposed to show particle trajectories from the simulations. As it appears the particle size only gives very trivial cases of a ray-like behavior and does not show how e.g. a particle is trapped or subjected to background collisions.

Line 224: How is the given conclusion be drawn from Figure 7?

Line 232: If an initial ice particle splits several times to finally create remnants all below 0.8 nm and thus ultimately is removed from the simulation, would that not remove quite a lot of potential MSP? i.e. underestimate the total collected amount?

Line 211 period after “Figure 7a”

Line 225 grammar around “being taken away”

Line 230 radii

In section 4.6 it would be nice to see the assumed particle densities, at least by pointing to a certain figure or maybe even from a reprint of the underlying densities, to evaluate the numbers.

Maybe include the ideal case if any particle entering the funnel would be collected, even if the results have a large uncertainty.

Line 384: The tilt under an angle of attack can be easily simulated with DS2V using 2D flows, why would you need 3D?