This paper presents the design and construction of a unique system, RoLi, for the tower-based profile studies. Exemplary high-resolution profile data are provided, demonstrating the robust performance and successful application at the ATTO tower with a payload of ~80kg. I can see the future application of this system measuring an expanded number of compounds, maybe even VOCs. The onboard system and rails provide strong support for the scientific instruments in terms of safety control, data acquisition, and time synchronization. While several artefacts were noticed and appropriately addressed by the authors, recommendations are provided to further improve. The capability of this system for the vertical study of fog, clouds, and accumulated particles across the rain forest canopy is clearly demonstrated. Overall, I believe this manuscript suits well with the scope of AMT and is well-written, warranting publication after addressing two concerns and minor technical questions.

1. The aluminum rail in this RoLi system. I don’t see a detailed description and evaluation of the strength of this commercial rail system. The Authors stated that “…is ideal for RoLi” but according to Figures 1 and 2, the thickness of the aluminum rail is notably smaller than the supporting frame. Considering the pulling forces from the ~80kg instrument payload, it would be more convincing if the authors could add some evaluation results about the strength. I am sure some words from the manufacturer's specifications would strengthen this claim.
2. Potential contamination of the RoLi system. In this paper, the authors demonstrated the application of RoLi system in the research area related to aerosol profiles. I am sure with such a payload, this system can support measurement of other targets, even VOCs. It would be great if the authors could clarify whether lubricants, solvents, or maintenance materials could introduce interference. Such words would enhance the credibility for multi-purpose future applications.

Below are some more specific comments:

1. Line 105: Insulation is crucial for the tower-based measurements. I am thinking the 100 kΩ resistance threshold for power supply reactivation described by the authors might lead to frequent restarts when the unstable resistance fluctuates around 100 kΩ. Maybe a higher restart resistance threshold would be better to prevent unnecessary power cycling and reduce hardware stress.
2. Line 155: My experience with Raspberry Pi won’t allow me to trust such a small processor. The connection between the RoLi system and the ground station is essential since the loss of connection would trigger the alarm. I don’t know how frequently the connection check was scheduled (which should be mentioned), but I have double that this Raspberry Pi can handle such a burden of work.
3. Line 164: I understand that the selection of Cyberbajt YAGI 24-16-2.4 was well discussed by authors and technicians, but some descriptions would be great to show the capability of the effective range and signal strength, etc.
4. Line 221-226: When there is heavy wind with wind speed exceeding 15 m/s, the contingency plan is to send the RoLi system “go to home”, as stated by the authors. My point of view is that staying where it is would be safer. During situations with heavy wind or lost connection, either moving upward or downward is not safe, especially when the connection alarm is triggered, which would send the system home, but if there is something wrong with the rail lower than the current system location, the system is out of control to stop, leading to a dangerous situation. Please correct me if there are unmentioned advantages of the current plan.
5. Line 242: Overheating inside the box is always painful. My recommendation is to use aluminum-coated foam wrapping around the box. It works great based on my experience.
6. Line 299: It would be more informative if the authors could provide the list of instruments or the compounds being measured permanently at the top of the ATTO tower.
7. Line 381-382: I am sure the authors had done that, but are there other parameters measured at the top of the ATTO tower that can help with the analysis?
8. Line 407-409: A small K-type thermocouple works great in the described situation. Lightweight and adaptable. So thin that it won’t be bent by heavy wind. You can set it away from the box and the frame to avoid interference.

Dear authors, congratulations on an interesting paper about this relevant measurement technology. In general, the article is well presented, and the topic is interesting. Although I certainly appreciate articles focused on the development and engineering of measurement systems, it must be stated that they often precede the scientific articles using data collected by these systems. Therefore, the system description articles are often a little thin on the scientific results, focusing their science mostly on the validation and trustworthiness of the newly developed system. Therefore, my review is based on these premises.

Specific Comments:

Abstract Line 6 (and article lines 224, 226, and 298): In the abstract, you describe the RoLi's profiling speed using the more traditional SI unit of m/s. However, in lines 224, 226, and 298, you use mm/s, including a negative sign in line 224. I recommend making these units follow a standard, ideally m/s.

Line 26: This sentence has a strange wording. Is the tower "AT" the core of the project?

Line 34: The wording here may need to be improved because it is confusing. Although you offer a reference, I still fail to follow the logic of the argument. How does a large "footprint" enable the integration of regional processes? Would not a network of small flux towers better represent mesoscale and regional processes? 

Line 47: The four-dimensional argument here is overstated. Even though the RoLi is increasing resolution in the vertical dimension, it is still attached to a single tower (i.e., a single point in the region) and not a network of tall towers. This needs to be reworded not to overstate the impact of this system.

Line 57: The argument regarding UAS flight time is incorrect. In your scale of profiles, every 30 minutes with 0.5 m/s ascent speeds, there are several commercially available UAS that can easily achieve those flight performances. I believe the real argument against UAS use here is the pilot/or human oversight requirements and the current limitations on availability of battery recharging/swapping technologies.

Obs: The best argument for your system is around line 50 - 55 (the short inlet and the measurement accuracy)!

Line 326: Here, you mention the nocturnal Low-level jet as "clearly visible"; Well, it is not. In part because the x-axis only has date and not time, so I can at best only estimate day and night, but also because with data for four days, the stronger winds could be associated with other features, such as the rains described. Therefore, I suggest rewording this explanation and improving the figure to mark sunrise and sunset for each day.

Line 334 - 336: Similar to the comment for line 326, this daily cycle is also not "clearly highlighted".

Line 346 - 358: The wording for this paragraph is a bit strange. You start describing the behavior of particles, then transition to relating it to wind shear, only to return to addressing the particles, making it confusing. This is particularly true in line 352 when you start a sentence with "The origin of these particles" after talking about winds. Therefore, it is unclear to the reader which particles are referred to by "these".

Lines 359 - 367: I particularly liked that you addressed system limitations and included a section for it later. Good Job!

Figures 4 and 5: Improve the x-axis to include time and add markers for sunrise and sunset to make day-night cycles clearer.

Section 3.4 is very good. Good job!

Line 444: You conclude that "the short inlet line of RoLi avoids

445 the effects and unavoidable losses associated with long inlet lines". Which is a confusing sentence because you are saying it avoids the unavoidable (even though the "unavoidable" adjective is attached to the long lines). Additionally, you make this conclusion without having shown a clear comparison between the short and long line systems. Therefore, I do not believe this is a conclusion that the data shown in the article supports. You have to either add the supporting data or remove the conclusion.

Conclusion: Is this system open-source? Can others use it? If so, be sure to highlight it as it would greatly increase the value of this article to the community.

General Recommendation:

The system is interesting, and the article is good. However, I believe it could be improved by including more intercomparison data to validate the performance of the RoLi system. As is, the data presented only indicates a minimal agreement with the expected atmospheric patterns of the region for that time of year, without a validation of the instances in which it does not.

Although I understand that a 330 m tower does not fit in a controlled chamber, and there are not many other towers around, I believe the authors could have demonstrated the performance of the measurements against the point measurements of the tower, the inferred profile from the tower point measurements, and a NWP simulation.

Finally, it is unclear if others in the community can benefit from this development. Can I also build a RoLi, or is it proprietary? Can I access and use RoLi data? If not, why should I care about it? Please make these answers clear to the reader. I believe doing so will elevate the impact of this article in the community.