Recent Advances in Wildland Fire Smoke Dynamics Research in the United States

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a well-structured review that synthesizes a wide range of recent literature on Wildland Fire Smoke Dynamics Research in the United Statues. The manuscript is timely, comprehensive, and written in a clear and engaging style. It effectively identifies key advances, research gaps, and future directions, making it a valuable resource for both new and experienced researchers in the field. I have only a few minor comments.

• Figure 2. Negative R² does not make sense. I suggest removing this index.
• There appear to be two sections labeled "2.3". Please correct this duplication and update all subsequent section numbers accordingly.
• Please check the captions of all figures. The formatting of figure captions is inconsistent throughout the manuscript. For example, both “Fig. 3” and “Figure 3” are used, and the font style alternates between bold and regular. Please ensure consistency in figure referencing and formatting according to the journal’s guidelines.
• Line 385: Please change ‘US.’ To ‘US’.
• Lines 385-387: The finding that mean plume heights are higher for wildfires in the central U.S. than in the western U.S. seems counterintuitive, given the typically larger fire sizes during recent decades in the West. Please clarify this result and provide potential explanations or supporting evidence.
• Lines 554-561: The font type is strange here.
• Reference 27. Please remove the ’27.’ Here.

Author Response

Please see attached file for responses.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript explores wildland fire smoke dynamics research as a review paper. The coverage is very broad, from field campaigns and satellite remote sensing, to coupled scientific modeling and machine learning applications. Overall, the review manuscript is impressive, and the extensive author list representing key institutions in the field is truly great. Nevertheless, there are several technical points and areas for potential improvements, which can further enhance the manuscript's clarity, impact, and completeness.

(1) The paper is now structured into distinct sections for "Advances in Measurements" and "Advances in Modeling." However, the manuscript could be strengthened by more explicitly weaving a narrative that connects the findings from one section to the challenges in the next. In particular, Section 2.3 outlines new knowledge about smoke particle evolution from campaigns like FIREX-AQ, while Section 3.1 points out that models still struggle to reproduce this evolution. The paper would be more impactful if these links were made more direct. Thus, the authors should explicitly state how a specific measurement advancement has directly challenged or confirmed a specific modeling assumption, or causes changes in modern research. The authors should also review the weaknesses and technical shortcoming that exist in current chemical transport models.

(2) The manuscript mentioned "WUI" in Section 4.3: "Gaps and Research Needs", but this may understate the magnitude of this emerging challenge. WUI fires are not simply wildland fires that happen near buildings, but represent a fundamentally different combustion regime. The fuel source changes from biomass to a complex mixture of building materials, plastics, vehicles, and household chemicals, which results in a completely different emission profile, and may also lead to totally different chemical pathways apart from pure biomass smoke. Thus, the authors can add a new sub-section titled "the unique challenge of WUI smoke dynamics", which discusses different dispersion patterns in the built environment, but also outlines the critical gap in our knowledge of WUI-specific emission factors, chemical composition (e.g., dioxins, furans, heavy metals), and the subsequent health implications, which are likely far more severe than traditional wildland smoke.

(3) The authors could make the distinction between fires and wildfires more explicitly, especially the variation of dynamics in specific cities of US.  This could lead to a potential research gap in relevant discipline, so please state the research gap clearly in the Conclusion as well. Also, when discussing the discovery and scientific finding based on a prescribed fire campaign, the authors should also state clearly the potential limitations of extrapolating the existing / current finding to a high-intensity wildfire scenario.

(4) In Section 4.1, the gap in measuring deep plumes exist because it is a feature of large wildfires, which are difficult to be measured or quantified, whereas most instrumented burns are prescribed fires with less intense plume rise. Emphasizing this point can also add the significance of the identified research gaps in relevant disciplines.

(5) Section 3.6 provides an overview of Machine Learning (ML) applications. However, the treatment is largely descriptive. A critical review should be added to address the pros and cons, the inherent limitations of these techniques from physical science perspectives and from scientific point of view. ML models are powerful interpolators, but can be poor extrapolators at the same time, especially when one wishes to project future fire scenarios, especially nowadays there is a changing climatic system. Thus, the use of ML techniques has to be very cautious, because ML is usually like a black box, and when noisy datasets are available, the ML results could be totally wrong. Some of the technical challenges have to be discussed when ML is applied into large scale retrieval studies, or into climatic studies. Furthermore, in Conclusion, the authors should also highlight the importance of developing "physics-informed" or "explainable AI" (XAI) models that respect physical laws, for example, conservation laws, as ML techniques applicable in wildfire simulation is not only staying in data fitting, but have to consist of science contents. The authors should address this point seriously by referring to the following references as well in your main text, and should incorporate the key ideas of the references below into relevant paragraphs:

https://cdnsciencepub.com/doi/10.1139/er-2020-0019

https://fireecology.springeropen.com/articles/10.1186/s42408-024-00335-2

https://www.nature.com/articles/s41598-024-52821-x

https://link.springer.com/article/10.1007/s11676-024-01783-x

(6) The manuscript explores how remote sensing and large-scale data analytics are useful in climatic prediction and wildfire forecasting. However, they didn't provide full explanations of the roles of satellite informatics and remote sensing in relevant assessments, as well as the changing importance of these data analytic techniques in fire monitoring and wildfire assessment, as well as assessing the climatic variables that are related to wildfire occurrence. Modeling limitations are not mentioned as well. Please acknowledge the following references and incorporate some of the ideas here into your revised manuscript:

https://apctt.org/sites/default/files/attachment/2025-05/06_Application%20of%20satellite%20informatics.pdf

https://www.tandfonline.com/doi/full/10.1080/15481603.2024.2348863

https://www.sciencedirect.com/science/article/pii/S1574954125002298

(7) For "Gaps and Research Needs", a summary figure is important. The conceptual figure can include some ideas about the next-generation, fully-coupled research and forecasting / monitoring system related to wildfire detection and prediction. A new schematic diagram should be added to Section 4, which should show inputs (e.g., real time satellite datasets, dynamic fuels), the coupled model core (e.g., CFD, chemistry behind, meteorological concepts, fire behavior etc.), potential outputs (e.g., smoke concentrations, health impact, potential health consequences, rise of plumes etc.). Arrows should also be added into the diagram to indicate the flow of data, feedback loops that took place etc., as well as the missing connection between these important objects or concepts in current research etc.

(8) The introduction of this manuscript identifies "human health" as a primary motivator of this research, and the conclusion also mentions public health protection. However, the review focuses almost exclusively on the physical and chemical processes, but not discusses the connection with human health conditions and neighborhood environmental conditions. Thus, a more comprehensive review of how specific advancement in dynamics research can enable better health impact studies. The authors may also explore how wildfire occurrence could lead to air pollution and changing climatic conditions, which could induce devastating health impacts to citizens. Some specific health impacts due to unfavorable environmental conditions can also be mentioned and discussed, for example, the ones shown in the following references:

https://www.sciencedirect.com/science/article/pii/S2667278221001073

Table 1 of https://www.mdpi.com/1660-4601/18/12/6532 

https://www.sciencedirect.com/science/article/pii/S0021755724001499

(9) Please explain that more accurate plume rise models (as discussed in Section 3.4) are not just a modeling exercise, but are very useful and helpful in determining ground-level concentrations, which is the primary input for epidemiological exposure models, or health-impact studies.

(10) Improved modeling of secondary organic aerosol (SOA) formation is also important, and the authors can highlight more on this topic in the review manuscript, because these secondary particles have different sizes and chemical compositions, which will cause and bring out different health effects, as compared with primary emissions. This would reinforce the societal relevance of the technical advancements described in previous sections of the manuscript.

(11) In overall, please also provide more critical perspective and analyses of new technologies adopted in wildfire prediction and measurement of wildfire dynamics. Some sections or sub-sections should focus more on science, rather than social science perspectives.

(12) Figure 4 is too brief, please add more key terms and descriptions, or labelling within the figure.

(13) The different graphs based on different models in Figure 6 should be indicated using different colors. The current figure is a bit misleading and unclear.

(14) Figure 8(b) and (c) show R = 0.97 and R = 0.07 respectively. Why is there a significant difference / discrepancy? Please explain from scientific perspectives behind, not just from statistical point of view. Please provide some reasons based on understanding of physical science, or the setting up of the machine learning based model.

After addressing all aforementioned points or current weaknesses, I am sure this manuscript can become a statement on the state of the science in wildland fire smoke dynamics. I am looking forward to seeing the revised version of this review paper.

Author Response

Please see attached file for responses. 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have tried their very best to enhance the quality of the current manuscript, which is very nice. However, there are still some more further modifications needed:

(1) Regarding point 2:

"The research needs are discussed to improve emission estimation, obtain chemical data, evaluate human health impacts, and improve modeling tools" - This is very important in environmental studies and assessments, but how could all these be incorporated into relevant studies? The authors should provide proper examples with illustrations and highlight the importance of relevant aspects in your last part of the manuscript.

(2) Regarding point 3:

"research gap and need to understand the differences in measured smoke properties between prescribed fire and wildfire, develop integrated systems of statistical, physical and ML models, and conduct the simultaneous measurements of these components for wildfires and WUI fires" - the reviewer agree with the authors on this, but please discuss the existing progress in relevant discipline, for example, how could can the existing statistical, physical and ML models perform so far? What are the technical shortcoming induced?

(3) Regarding point 6:

We understand that the authors have added a small paragraph discussing how satellite datasets can be used to detect fire and smoke dynamics, as well as other environmental retrieval tasks, however please also acknowledge the contribution of these references below (in previous report):

Please acknowledge the following references and incorporate some of the ideas here into your revised manuscript:

https://apctt.org/sites/default/files/attachment/2025-05/06_Application%20of%20satellite%20informatics.pdf

https://www.tandfonline.com/doi/full/10.1080/15481603.2024.2348863

https://www.sciencedirect.com/science/article/pii/S1574954125002298

(4) Regarding point 10

The authors added that "multiple chamber experiments and one-dimensional Lagrangian chemical transport model to understand the importance of the intermediate VOC fragmentation" - The reviewers agree and think it is good, but is there other ways of detecting intermediate VOC fragmentation from chemistry point of view? Please try to add in more details of the relevant lab-work conducted in this area too.

After addressing aforementioned comments, I think the manuscript is serving as a good review paper on fire and smoke dynamics.

Author Response

The authors have tried their very best to enhance the quality of the current manuscript, which is very nice. However, there are still some more further modifications needed:

Thank you for continuously providing valuable and constructive comments for improving this manuscript. We followed these comments carefully and made modifications accordingly.

(1) Regarding point 2:

"The research needs are discussed to improve emission estimation, obtain chemical data, evaluate human health impacts, and improve modeling tools" - This is very important in environmental studies and assessments, but how could all these be incorporated into relevant studies? The authors should provide proper examples with illustrations and highlight the importance of relevant aspects in your last part of the manuscript.

Response: We adopt a diagram (Fig. 9) provided by the U.S. National Academies of Sciences, Engineering, and Medicine (ref. 171) as an example to illustrate incorporation of the studies on emission estimation, chemical data, human health impacts, and modeling. A short paragraph is added to explain this diagram (L922-928).

 (2) Regarding point 3:

"research gap and need to understand the differences in measured smoke properties between prescribed fire and wildfire, develop integrated systems of statistical, physical and ML models, and conduct the simultaneous measurements of these components for wildfires and WUI fires" - the reviewer agree with the authors on this, but please discuss the existing progress in relevant discipline, for example, how could can the existing statistical, physical and ML models perform so far? What are the technical shortcoming induced?

Response: A paragraph is added to discuss the need to develop integrated systems of statistical, physical and ML models. It uses fire behavior modeling as an example. Strengths and shortcomings of each of the three types of modeling tools are briefly described. It is suggested that one of the focuses of modeling study is to develop hybrid modelling approaches that integrate multiple methods to harness their combined strengths and in the meanwhile minimize their weaknesses. (L950-961).

(3) Regarding point 6:

We understand that the authors have added a small paragraph discussing how satellite datasets can be used to detect fire and smoke dynamics, as well as other environmental retrieval tasks, however please also acknowledge the contribution of these references below (in previous report):

Please acknowledge the following references and incorporate some of the ideas here into your revised manuscript:

https://apctt.org/sites/default/files/attachment/2025-05/06_Application%20of%20satellite%20informatics.pdf

https://www.tandfonline.com/doi/full/10.1080/15481603.2024.2348863

https://www.sciencedirect.com/science/article/pii/S1574954125002298

Response: Thanks for providing these valuable studies. We used the ideas provided in these studies in three places. (1) A paragraph is added to describe how the idea of applying satellite technology to deal with the climatic challenges at urban scale is used to deal with the challenge of WUI fire and smoke. (L289-299). (2) A paragraph is added to describe how the idea of a compilation pipeline to integrate ground- and satellite-based telemetry tracking wildlife datasets is used to develop integrated fire and smoke datasets. (L769-774). (3) The great value and capabilities of ML and DL in processing massive satellite fire and smoke satellite data is suggested, based on the idea about the potential of ML and DL to automate the detection and counting of wildlife individuals with higher detection rates than conventional surveys. (L946-949)

(4) Regarding point 10

The authors added that "multiple chamber experiments and one-dimensional Lagrangian chemical transport model to understand the importance of the intermediate VOC fragmentation" - The reviewers agree and think it is good, but is there other ways of detecting intermediate VOC fragmentation from chemistry point of view? Please try to add in more details of the relevant lab-work conducted in this area too.

Response: Two paragraphs are added to address this comment. One describes techniques used to conduct offline and online SOA analyses through field measurements (L327-334). The other describes some chamber measurements of short- and long-term photochemical evolution of biomass burning OA (L355-371).

After addressing aforementioned comments, I think the manuscript is serving as a good review paper on fire and smoke dynamics.

Response: Thanks again!