The present manuscript investigates the formation of pyroconvection using the Australian New Year's event as an example. The results of sub-kilometer atmospheric simulations are described and compared with satellite data. For a high-frequency representation of fire emission fluxes, the GFAS data product, which is only available for daily mean values, is extended to include a description of a daily cycle. In sensitivity experiments, the effects of heat and water vapor fluxes are specifically investigated and compared with a reference case without corresponding fluxes. The study uses realistic atmospheric simulations to demonstrate the complexity of the formation of pyroconvection and emphasizes the special role of atmospheric stability for the convective transport of smoke aerosol into the upper troposphere.

The manuscript summarizes a carefully conducted scientific study. The methods used are appropriate for the research question and the presentation of the results is clear and logically structured. The figures are easy to understand and of high quality. I recommend publication in ACP, provided that the comments below have been considered and appropriate changes incorporated into the manuscript.

Main comments:

1. The present work proposes a scaling of the flux of sensible heat and water vapor emission from fires that has a pronounced maximum in the early afternoon. This scaling, which has been derived in the existing literature as typical behavior for pyroconvection, does not at all match the observed behavior during the ANY event, for which strong pyroconvective activity was documented during the nighttime hours. In other words, the present study suffers from the logical inconsistency that a generally valid parameterization for pyroconvective fire emissions is applied to an extreme case that deviates from the rule and does not follow this typical course. The authors address this aspect in their discussion section, but they need important parts of the presentation of the results to discuss deviations in the temporal course of the pyroconvection, which probably come about due to the insufficient assumptions in the parameterization. The general inconsistency will not be resolved with the existing material, but a preceding presentation of the temporal course of the Australian fire in comparison to the typical pyroconvective evolution may help to better understand the assumptions made and their influence.

2. The presentation of the results is very descriptive and emphasizes unimportant details. I would have liked to see more focus on the convective processes involved and an exploration of the mechanisms for pyroconvective transport based on the case study. Furthermore, I would like to emphasize that the discussion section needs to be revised. Here, the limitations of the study and the resulting implications need to be addressed in more detail. An additional summary of the results without reflection is not needed here.

Minor Comments:

- l.13 – 15: Please shorten and rephrase the sentence. Be clearer about what you want to communicate.

- l.20: The Luderer reference is actually supporting the argument about diurnal changes.

- l. 26: Balch actually states for the Australian event that “Night-time MODIS active fire detections represented 76% […] of total detections” – not supporting your argument here.

- l.27 (and l. 400) Fromm reference is wrong. I guess it should be the untold story paper

- paragraph starting at l. 36: Here you jump from climate change, to interactive fire simulations to factors at impact pyroconvective transport. Please improve the line of argument here and use paragraphs as structural elements to separate between topics.

- l. 60: “plume production”: Please change wording.

- l. 76: Start a new paragraph with “In this study…”

- l. 76-77: “intense fire-atmosphere interaction…”: Please rephrase! This is misleading. You don’t conduct studies on an interactively coupled fire-atmosphere system.

- Sect. 2.2: The description of aerosol emission fluxes is missing here. Do you also scale them with d? Please mention ICON and ART versions and refer, if possible, to a specific version tag.

- Sect 2.3: Please add information on the domain size (horizontal extent) and typical vertical layer spacing to the description.

- l. 128: “complex microphysical proceses” → Unclear what is meant here.

- end of Sect 2.3: Please add the values of peak sensible heat and peak fire-induced water vapor flux to the description. Please also provide the equivalent latent heating potentially released by condensing the additional water vapor.

- l. 194: “Besides …” This is too detailed and insignificant. Please delete it!

- l. 202: “However …” Again irrelevant information here.

- Table 1: Please use positive CIN value following the standard definitions e.g. of AMS glossary. Please only print significant digits in the table! Please support your simulated values with observations. Please provide a comparison of your simulated profiles to radiosonde data in your response.

- l. 220: CAPE: Please clarify which CAPE definition is used – surface-based, most-unstable, mixed-layer, etc…

- l. 234-236: Plume definition should be moved to method section. The 100 largest altitudes definition introduces dependence on the grid which is sub-optimal.

- l. 235: “the sum of LWC+IWC […] must be larger…” This is not clear! Do you sum up values of an intensive quantity?

- paragraph starting with l. 257: There are too many methodological descriptions mixed in with the presentation of results. It is especially unclear how pyroCb is defined for REF (l. 260).

- l. 265: How do you define “lifespan”?

- l. 270: “Our analysis shows …” This is not clear! How and where do you show observational uncertainties? Please be more precise!

- l. 293 – 307: The paragraphs appear to be a summary of the results presented before. To my opinion, this does not fit into the discussion section where the focus should be on limitations, implications and relation to existing knowledge.

- Open science: I recommend making analysis and plotting scripts openly available to improve the reproducibility of your results.

The present manuscript focusing on the 2019/2020 Australian New Year's bushfire event. This study analyzes the role of wildfire-induced heat and moisture fluxes in the genesis and development of pyroconvective clouds. The goal is to refine the understanding and numerical representation of these processes using the ICON-ART model for simulations, which includes fire effects like sensible heat and moisture release, and fire data obtained from the Global Fire Assimilation System (GFAS), based on MODIS satellite observations.

Three simulation setups were tested, REF, SH and SHLH. In addition, NASA’s 3D Wind Algorithm was used to retrieve plume and cloud heights from satellite data.

The introduction is well written and effectively contextualizes the challenges involved in simulating pyroconvective clouds, highlighting the complexity of this type of atmospheric phenomenon.

In the methods and materials section, the model, its parameterizations, and configurations are clearly and objectively described, allowing for a good understanding of the adopted approach.

The presentation of the results is quite detailed. However, the excessive use of acronyms compromises the flow of reading, as the reader must constantly refer to the legend or glossary to understand the terms. This can be particularly challenging for readers who are not fully familiar with the field.

In the discussion section, the authors appropriately address the uncertainty factors that may have contributed to the differences between the model results and the observational data. However, the way these uncertainties were estimated or calculated is not sufficiently clear. For example, in line 270, the authors state: “Our analysis shows that the height retrievals have an error range of ±200 to 300 m,” but it is not specified how this error range was obtained — it would be important to clarify whether this estimate was derived from comparisons with observational data, statistical analysis, or another method.

Additionally, in line 129, the authors state: “However, this study does not consider aerosol-cloud and aerosol-radiation interaction.” It would be relevant to further discuss what types of uncertainties this omission may introduce into the model’s performance, especially considering that such interactions can significantly impact the development and evolution of convective clouds.

I also suggest including a more in-depth discussion on the possible uncertainties associated with the plume height values obtained from the NASA 3D wind retrieval algorithm, as well as a comparison between the vertical resolution of these height estimates and the vertical resolution of the model used in this study. Are they consistent? Such an analysis would help better assess the compatibility between the data and the model results.

Would it be possible for the authors to consider incorporating data from other satellites to reduce the uncertainties related to the plume height estimates obtained by the model? For example, data from active remote sensing instruments?

Overall, the article makes several important scientific contributions to the understanding and modeling of pyro-convective cloud dynamics, especially in the context of extreme wildfire events. 

The study introduces the parameterization scheme for fire-induced sensible heat and moisture release using satellite data (GFAS).

The manuscript provides clear evidence that sensible heat is the dominant factor in driving plume rise and cloud formation, and demonstrates how wildfires can inject aerosols and trace gases into the upper troposphere and lower stratosphere, potentially affecting global atmospheric composition and radiation balance. By improving our ability to simulate how wildfires impact the atmosphere, the study helps bridge gaps between fire science, meteorology, and climate science. In this regard, I recommend the publication of the manuscript, provided that the reviewers' suggestions are adequately addressed.