General comments

The preprint from Johansson et al. presents NH3 measurement from the GLORIA instrument abord HALO and compares the data to CAMS model simulations (reanalysis and forecast) for two distinct situations in the upper troposphere. First, NH3 in the Asian monsoon and second biomass burning event from South America over the Southern Atlantic. While the model represent well NH3 in the Asian monsoon, it overestimates NH3 in biomass burning events into the upper troposphere over the Southern Atlantic. In contrast to NH3, PAN is simulated well with the model. Johansson et al. suggest some hints and reasons, why the model has difficulties in NH3 representation for some NH3 sources, but a deeper analysis on the reasons is out of the scope of the paper.

The subject of NH3 in the upper troposphere, especially with its potential indirect contribution to the Earth’s radiation budget and the difficulties in the representation in models fits well in the scope of ACP.

Overall the paper is easy to read and the results are presented in a logically structured manner. It has only minor issues (presented below) and I recommend publishing the paper.

Specific comments

Title: I suggest to either write “ammonia” or “NH3” in the title, but not both.

Figure 4-6: I suggest to add the information of the take off and landing base for each flight in the Figure description. That helps to better understand the flight direction and thus the latitudinal location as it is hard to differentiate the three flight in Figure 1b.

Figure 4: I can only identify one PV line. Is it the 2 PVU or 4 PVU line? The same is valid for Fig 5 and Fig 6. I suggest either to add a label in the plot or clarify in the caption, which line is visible.

L145-147: How is the horizonal distribution of NH3 over the Indian subcontinent? Could different sources on the ground with different NH3 emissions have lead to two distinct air masses characteristics east and west of the flight tracks?

L160-162: The presents of clouds explains the reduced quality of the NH3 retrieval, but could they also be a hint of NH3 loss and transformation processes (e.g. particle formation), leading to reduced NH3 mixing ratios? How are clouds represented in CAMS and could it be a hint of missing scavenging effects for NH3??? Are there any information on the ammonium (NH4+), thus the particle phase available, either for Gloria or model data, which could give further ideas on the deviation between model and measurements.

L214: You state that NH3 over the South Atlantic is likely from biomass burning. Could there also be a contribution from livestock and fertilizers?

L224: NH3 is very variable in biomass burning, depending on a lot of factors, e.g. fire size, soil composition, fuel composition, fire weather etc. I suggest to mention some of the reasons, why NH3 is hard to simulate, especially from BB sources, at least at some point, see also L243.

L225-231: As CAMS seems to reproduce the large scale Asian Monsoon transport of NH3 well in comparison to biomass burning. Could that be a size problem of CAMS, as the resolution for small scale phenomena is not sufficient enough?

L237: You state strong biomass burning in South America and Africa. I could not find any other statement concerning Africa as a source region in the draft for the flight during SouthTrac. Is Africa a potential source region, which need to be taken into account?

L243: I agree that clarifying the reasons for NH3 overestimation is beyond the scope of the presented analysis. Nevertheless, I suggest to mention some more difficulties of NH3 in simulations. Some suggestions are: NH3 is very variable in biomass burning, depending on a lot of factors, e.g. fire size, soil composition, fuel composition, fire weather etc. The chemical aging of NH3 in BB plumes can be fast. BB emit moisture and particles, thus accelerating transformation and scavenging processes, leading to NH3 loss. Consequently, it is not further transported in comparison to PAN from South America to the Southern Atlantic upper troposphere. What about NH4 in model and Gloria data (see also L160-162)

L247: I strongly agree that NH3 is undersampled.

Technical corrections

L12: “pollution gases, such as peroxyacetyl nitrate (PAN) show the ability” -> “pollution gases, such as peroxyacetyl nitrate (PAN), show the ability”

L13: Suggestion “CAMS models to reproduce the biomass burning plumes above the South Atlantic in general.” -> ““CAMS models to generally reproduce the biomass burning plumes above the South Atlantic.”

L15: Suggestion “… different accuracy for biomass burning and agricultural sources in the Asian Monsoon” -> “… different accuracy for biomass burning in comparison to agricultural sources in the Asian Monsoon”

L28: “.. local air quality, and for climate through …” -> “.. local air quality and for climate through …” also L42, L54

L38: “or AIRS (Atmospheric Infrared Sounder) global…” -> “or AIRS (Atmospheric Infrared Sounder), global…”

L47: “In the present work we utilize…” -> “In the present work, we utilize…”

L64: “… sampling up to 0.0625 cm−1, and compensation of aircraft movements and active targeted…” -> “… sampling up to 0.0625 cm−1, compensation of aircraft movements, and active targeted…”

L79, L81: “… 1.3 km, and estimated…” -> “… 1.3 km and estimated…” similar L92, L94, L100, L106, L110 (2x), L174, L184

L85: “…, and second, the CAMS ..” -> “… and second, the CAMS ..” similar L86

L90/91: “2003 and 2021, and uses 60 vertical levels between 0.1 and 1000 hPa, and a horizontal…” -> “2003 and 2021, uses 60 vertical levels between 0.1 and 1000 hPa and a horizontal…”

L149: Suggestion “and supplementary figures 1-2).” -> “and supplementary Fig. S1 and Fig. S2).”

L177: Suggestion “NH3 as measured by GLORIA (Fig. 5a) instead, does not…” -> “NH3 as measured by GLORIA (Fig. 5a), does not…”

L192: “(Figs6b)” -> “(Fig. 6b)”

L195: “… simulated plumes, compared to…” -> “… simulated plumes compared to…”

L200: “enhancements of ammonia.” -> “enhancements of NH₃.” Also L234, L253

This study presents valuable model-measurements comparisons of NH₃ in the upper troposphere and lower stratosphere, focused on the biomass burning event and the Asian summer monsoon over two specific locations. Large enhancement of NH₃ was observed in the upper troposphere near the Asian monsoon region but not over the biomass burning plumes. The CAMS model simulation shows enhancement of NH₃ in both the regions. Overall, the results are well organized and presented with sufficient scientific evaluation. However, there are some weaknesses that could be improved to make this work more impactful.

General Comments:

1. The goal of this study should be mentioned clearly. Is it to evaluate the CAMS model? Or to understand the differences in CAMS reanalysis and real time products? Is the end goal to improve the model performances?
2. Both NH₃ and PAN are both emitted from biomass burning and discussed in the manuscript together. I am not sure how to they are related to each other. Can we expect the same enhancement in them due to biomass burning? What are their lifetimes? In all the figures, GLORIA measurements of NH₃ and PAN are compared with the CAMS NH₃. What does PAN simulated from CAMS look like for these cases?
3. The CAMS model performance is evaluated for the two specific events, Asian monsoon and biomass burning. It would be valuable to include some information about how CAMS performs on climatological perspectives. Can satellite measurements of NH₃ be compared with the CAMS model outputs?
4. Both the CAMS reanalysis and forecast products were compared in this study. Is either one expected to be better than the other?
5. For all the figures, the timing of enhancement in the measurements and the model seems to be different. Therefore, it is hard to draw conclusions based on these figures. Do we expect the timing to be closer?
6. For the model-measurements differences, only some speculations were made without any evidence. Are there any supporting material to show, for example, that the loss process of HN₃ is underestimated?

Specific Comments:

L5 (Abstract) – within the Asian monsoon -> either Asian monsoon anticyclone or Asian monsoon region?

L7 (Abstract) – We compare… model. -> We compare…model to evaluate its performances.

L8 (Abstract) – both reanalysis and forecast model were used here. Why?

L15 (Abstract) – What does ‘different accuracy’ mean? Is it higher or lower? Or better?

L24 (page 2) – Is this due to warming climate?

L33 (page 2) – Here ‘quasi non’ might mean ‘close to zero’. I would like to know if it exists or not.

L47 (page 2) – This sentence is important in stating the goal of this work. Is it to analyze the model performance or describe the distribution? Or it could be to evaluate the model under two different conditions?

L66 (page 3) – What does ‘mostly interferograms…used’ mean?

L79 (page 4) – ‘…adjusted slightly’ can be explained explicitly.

Section 2.2.1 (Page 4) – Here, the CAMS performance is stated for PAN. Is there a direct relationship between PAN and NH₃ other than both are biomass burning tracers? Are there any studies focused on NH₃?

Figure 2 (page 6) – The color bar is saturated at 1000 pptv. However, the values in Fig. 2a & 2b could be much higher than 1000 pptv.

L145 (page 7) – ‘This is remarkable…changed’. -> I am not sure what this sentence means.

L162 (page 9) – Were there cloud or aerosols? Fig. 4a (also 5a) looks likes very noisy to me.

L243 (page 12) – ‘There are..hints…’ -> Can this be shown or explained further?