Reviewer 1 This is an interesting and useful combination of field work observation, sampling and lab experimentation to examine modern plant communities and charcoalized plant parts at different roasting temperatures for the purpose of improving the development and interpretation of palaeoenvironmental records of charcoal (a proxy for fire) in boreal forests. The results should be of interest to a wide variety of researchers in EGU fields and interested in biogeosciences related to fire at local to large spatial scales, palaeovegetation research, and archaeology.

OVERVIEW: Overall I suggest some changes to clarify the intent of the author for certain words like “burned”, ‘impact’, ‘intensity’ etc - detailed below.. And some expansion on the discussion with respect to additional charcoal morphology literature in paleoenvironmental studies and archaeological studies. There should be some refinement on how explicit the ‘decomposition’ technique of the vegetation fuels was and thus, the use of the word ‘’burned’’. Is it known if the material flamed in the oven? Was it roasted? combustion? Pyrolysis? The oxygen and time variables are largely ignored and this should be written as a caveat in the experimental design to be explored elsewhere or in future studies. Similarly, the words charcoal, charred, and (roasted, unused term in the study) may need to be defined early. ‘Ashed’ is also used but not fully discussed - I suppose it meant a more complete combustion generating white ash that then crumbled apart into soot and flyash? Was there an explicit purpose statement? Something like, characterising the diversity of charcoal morphologies produced by boreal forest vegetation fuels at X study site?

R: Many thanks for the positive response that this paper was interesting and a useful contribution to the community, and for the useful and thoughtful comments to further improve it. In revising this manuscript, I am working to 1) Clarify some of the terms used and expanded on the caveats of experimental design; 2) Expand on the discussion concerning additional charcoal morphology literature in paleoenvironmental studies and archaeological studies. However, a comprehensive comparison with archaeological studies is beyond the scope of this study. The same is valid for an extensive comparison with studies focusing on transport and sorting of morphologies, which was the not subject of the current study.

SPECIFIC COMMENTS:

L36 - This might be semantic, or a question of (spatial and time) scale and thus the need for clarity.  The word impact is a bit ambiguous without further clarity on the context and use of this term. What is meant here ‘[a] fire impacts boreal forest’’. Over the long duree, is it the changing attributes of fires and the fire regime that impact the boreal forests? Does this mean one fire is impactful? Boreal forests have a lot of spatial heterogeneity in vegetation structure that is in part caused by fire and in part also influences fire itself. A changing fire regime has significant outcomes on the land cover. But if fire is a process in boreal forests itself, it seems more of a feature than of the biome rather than something that just impacts it. Throughout the paper the framing of the disturbance regimes needs to be balanced with how these disturbance regimes (mostly fire explored here) are a part of the system, and not something that just happens to the boreal forest and changes (impacts) it.

R: Thank you. I have rephrased this to acknowledge the role of disturbance by fire in the functioning of boreal forests, but also the concerns on the impact of changes in fire frequency or intensity. It reads: ‘’Disturbance by wildfires is among the most common disturbance types in boreal forests, triggering gap dynamics or stand-scale forest replacement depending on intensity and frequency (Goldammer, 2015). Ongoing and anticipating increase in the frequency and severity of wildfire in boreal forests is raising concerns on its impact on the composition of these forests as well as climate (Jones et al., 2020).

L40 - Fire intensities in nature have been shown to be able to reach much higher temperatures, even flame temperatures can be higher than the range explored here. Is this really the gamut of temperatures in hot boreal fires? This needs to be framed as a subset (or modal?) temperatures of fires (maybe this can be estimated from MODIS intensities? i.e. energy output detected by satellite, or if there are some published field-based measurements.).

R: Thank you for pointing this out. Indeed, a wildfire burns at much higher temperatures than 500 degrees, but with this sentence, I highlight the temperature range that leads to the charcoal formation. In the revised paper I introduce an additional sentence clarify the range of fire temperatures and those resulting in charcoal formation (Rein, 2014).

L49 - worth stating somewhere that the Courtney-Mustaphi and Pisaric, 2014 study discussed potential for not just focusing on known-fuel morphotypes for charcoal analysis but for categorising all morphologies found in a local-scale study to examine the variability; as this would be useful to explore relationships to not just the known-fuel-sources of charcoal but taphonomic processes and possibly fire types (or another variable).

R: Added: ‘’Courtney-Mustaphi and Pisaric (2014) also discussed the potential for categorising of charcoal morphologies to explore relationships to taphonomic processes and possibly fire types.

L60 - It would be useful to make distinctions between studies using ovens, flames, and other pyrolysis and combustion methods.

R: I am working to distinguish between results originating from studies that have used muffle over, open flame, or calorimetric methods.

L68 - spp? Or taxa? What was the minimal taxonomic resolution?

R: Genus is the minimum taxonomic resolution used here, however, most plants have been identified at the species level.

L69 - was there any testing in this study? It appears to be mostly a characterization study, which has merit. The purpose, objective, aims are not congruent with the content

R: I have replaced the word test with evaluating. The three main objectives presented in the introduction are then detailed in the Results /Discussion.

L79 - complete dryness. Was this checked? Before burning in the oven were the samples dried? Often one would dry at 105°C for 24 hours to drive off most moisture. Of course this may only influence the combustion to a limited extent in this study - but worth documenting for future comparison studies.

R: Thank you, I removed the word completely dryness. The material was air-dried for several weeks, but the moisture content was not measured. The plant appearance was dry and brittle in all instances.

L87 - what was the rationale for limiting oxygen? Were there any comparisons with oxygen not-limited burning and open flame burning?

R:  To slow the burning processes in order to better replicate the natural conditions where fuels are more abundant and layered, thus the oxygen is more limited. I have not attempted experiments under non-limiting oxygen nor under open flame conditions.

L198 - is Ericaceae ever investigated in this study?

R: Ledum, Vaccinium, Camadaphne, Oxycoccus are all part of the Ericaceae family, this is what I ment here, instead of listing their names I have used their family name, see also Table 1.

L231 - ‘’rounder’’, was this intended to mean circular? (as in 2-Dimensional), or roundedness as in the degree angles are eroded or not produced? Can these terms be written more explicitly for the reader. Note that both how circular something is, and roundedness can be quantified, semi-quantified or categorised. Was this done? Discussing if this may or may not be useful in future studies would be useful for readers and future analysts (Note Vanniere et al 2003 Journal of Archaeological Science, 30(10), pp.1283-1299, with reference to eroded charcoal in agricultural soils).

R: Rounder, intended as circular, as in 2-Dimensional scale. I have not quantified the degrees of angels. I am working to point out clearly the differences and the potential usefulness of additional measurements on roundedness.

L264 - Add a caveat about the need to do detailed comparative studies on graminoid versus conifer needle fuels and subsequent charcoal. And perhaps among Graminoid growth forms themselves: Poaceae subfamilies, Cyperaceae, and others.

R: Added. Comparative studies on graminoid charcoal originating from Poaceae (grass) versus Cyperaceae (sedge) family will further improve the identification of fuel types given the ecological differences of the two groups i.e., Cyperaceae growing on wetlands, and Poaceae often on dry habitats. Misidentification of charcoal morphotypes from graminoid and conifer needle and subsequently of fuel types is another caveat that could be improved through detailed comparative studies of these fuel types.

L279 - a useful document for comparing mosses etc for readers to compare in Quaternary and temperature ecosystems is Levesque et al 1988. Lévesque, P. E. M., Dinel, H., Larouche, A. 1988. Guide to the identification of plant macrofossils in Canadian peatlands. Land Resource Research Centre, Research Branch, Agriculture Canada, Ottawa.

R:  I am trying to get access to it.

L315 - intensity, as in heat/energy given off by fire?

R: The sentence adjusted to denote intensity as high-energy release

General comments: Introduction in general:

I think there needs to be a distinction between flame combustion, roasting by hot air (ovens), pyrolysis. This needs to come out more obvious to the read beginning in the abstract, methods, and discussions. It needs to be stated that dry roasting in an oven is a proxy for one type of heating of vegetation in a natural fire, different to flame burning, etc. This is evident in the statement by the authors on L124 that ‘All plant tissue was reduced to ash at 450 °C (Fig. S1).’ In natural fires, flame and air temperatures do reach higher. I think the main items that need to be acknowledged is that the oven approximates some aspects of the heating conditions of natural fuels and that a crucial variable that is not explored is time at a (burning) temperature.With roasting in an oven the influence of flame dynamics and turbulent air flow is missing to the same degree as fires outdoors. This needs to be acknowledged as part of the experimental design and open the need for additional research.

R: Many thanks for pointing out the inconsistent use of burning types i.e, roasting by hot air (ovens), and pyrolysis, which is in the current form of the paper were sometimes used interchangeably.  In the revised version I will make it clear that all combustions were in the oven thus roasting by hot air. I will additionally acknowledge the limitation of the experimental design to some aspects of the heating conditions of natural fuels, especially on the relation between time and the burning temperature.

It would be useful throughout and within this paper (if anything was combusted in a different method) to add the categorical naming of how the material was ‘burned’. See Table1 in the following publication:  https://doi.org/10.1016/S0031-0182(00)00174-7

R: All plant material used for burning experiments was dried before and burning conditions (muffle oven, preheating) were the same for all measurements. The only difference is the use of different temperatures. This information was added in the caption of Table 1. List of plant materials burned. All plants are from Siberia, Russia, except Picea abies, originating from Taunus, Germany. All plant material was dried before the combustion in the muffle oven.

I have some broad suggestions on how certain details are communicated.-The plant anatomy of bryophytes is treated rather colloquially and requires refinement. - Are the species names known for the bryophytes? Many burn differently at low temperature because they hold water droplets differently, making some taxa more difficult to ignite even under the same fire weather conditions.

R: I have used Sphagnum spp. (likely S. medium/S. divinum) and brown moss (Polytrichum commune). Their names are visible in all graphs and Table 1. I acknowledged that moss burned differently but have not discussed the reasons for this. I am working to find anatomical differences between them, however, the fact that different moss species hold water droplets differently, making some taxa more difficult to ignite under the same fire weather conditions, holds in nature but not in laboratory burning experiments, where all plants were dried prior combustion.

-The use of the word ‘twig’ needs some level of description here as twigs are different in deciduous, coniferous, herbaceous? and colloquial terms. Can this be more explicit throughout the paper as it may vary by plant types.

R: The term twig was used only for woody species i.e, deciduous and coniferous trees as well as deciduous shrubs, and denotes small branches close to where the leaves are attached to the plant. I will add this specification in Table 1.

-There is a lot of comparison with Mustaphi and Pisaric 2014; could this be expanded to many of the other morphology papers. A table on charcoal morphometric technique studies and the usefulness could help link with the editors comment on this study not presenting a tangible application of the study in its current form.

R: To accommodate this comment (and from Rev 1) I am working towards adding an extra table comparing morphometrics results from this study to those from other publications.

FIGURES:

-Perhaps the black font text would be best placed outside the photograph because of the overlap and poor contrast between the letters and charcoal fragments.

R: Thank you, I will revise the figures according to this suggestion

FIGURE4 - are some of these not charcoal? Again the Levesque et al 1988 publication might be worth comparing. 

R: All pictures come from laboratory-produced charcoal.

FIGURE5 - ‘chacoal production’ spelling in bold (bottom left). Can you quantify the aspect ratios?

R: Corrected to charcoal production. Aspect ratio equals L:W ratio.

TABLE1 - can you add growth forms of plants? (sort of in the plant type column) and the anatomical parts investigated in this study? For instance, Does ‘’leaves’’ include the Petiole? The veins? Does twig also just mean wood? Or something else? Soft young wood? High water content?

R: Where possible or needed I have included the growth form of plants.

ADDITIONAL PUBLICATIONS

Some important morphology studies are not discussed in the context of this study.  It would be appropriate to discuss these studies in a comparative manner and to build the case for the overall usefulness of morphological metrics.

R: Many thanks for providing such an extended list of literature on charcoal. I am working towards incorporating the most relevant ones into the revised form of the paper.

Additional literature on charcoal and lab burning:

Reviewer 2 General Comments:

Feurdean presents a dataset of experimental charcoal produced from 17 species endemic to boreal Siberian. Using these experimentally produced charcoal, Feurdean makes insights into the reliability and applicability of charcoal morphologies as a proxy of fuel type. Additionally, the author shows how charcoal mass is retained as a function of combustion temperature for these samples.

The manuscript is very interesting and presents a promising dataset for the paleofire field. Its efforts towards proxy calibration of charcoal morphology and morphometry represent a key advance. Its experimental characterization of charcoal produced from several new fuel types and taxa, as well as replication of previous experimental productions of charcoal, make it a useful contribution to the field. Mass retention during charcoal production represents a key gap in our understanding of the source to sink controls of sedimentary charcoal, and this manuscript helps to bridge this gap. Lastly, it improves on some of the methodological approaches of earlier experimental productions of charcoal particles.

However, the manuscript falls short in several ways. Although the bulk of the manuscript is in good order, and the study itself is robust and scientifically sound, the discussion and conclusions are, in my opinion, woefully underdeveloped, especially in light of the novelty of the dataset and approach. In a broader sense, the manuscript fails to fully deliver on the potential conclusions and insights that could be gained from a dataset which is truly brimming with potential. I recommend moderate revision prior to publication and have outlined, in my opinion, the manuscript’s primary shortcomings and areas needing improvement below.

R: I want to thank the reviewer for the useful and thoughtful comments, as well as the overall positive response in feeling that this paper was interesting and a useful contribution to the community.

Firstly, it lacks in-depth comparison to prior work, which will surely undermine its impact. For example, several of the taxa tested by the author have also been directly tested in previous experimental studies (e.g., Eriophorum vaginatum in Pereboom et al. (2020), Pinus sylvestris in Crawford and Belcher (2014)), yet there is no discussion of the similarities and differences of the morphometrics of the charcoal produced from these taxa. Similarly, previous experimental studies have used a variety of techniques (and temperatures) to produce charcoal, but limited comparison is made with these studies and their conclusions. How and why do the values differ and compare between this and other experimental charcoal studies? Although the Discussion focuses on the findings of this study, it does not sufficiently contextualize these findings within those of the published literature. For example, section 4.2 refers to several other studies, but only vaguely compares findings of charcoal particle elongation between these studies. The published aspect ratio data of these experimental studies should be more thoroughly discussed and explored if this is this to have a veritable impact on the field.

R: I am working to extend the comparison of results from this study with others from published literature. The comparisons are also readily visible in the added new Table.  As most of the published literature is on the L:W ratio, I have contextualized these findings better.

Secondly, the manuscript does not make actionable conclusions for the paleofire field. Besides a somewhat unconvincing (see specific comments below) description of the potential utility to use charcoal aspect ratios to distinguish fire and fuel types, the manuscript does not provide explicit descriptions of the morphometric values that be used to constrain interpretations of sedimentary charcoal. What is the cut-off for elongation indicative of graminoids?  What are the ranges of morphometric values that can be indicative of fuel types? What are the mean values of the aspect ratios of the fuel types that can be distinguished (wood, graminoids, and leaves, as indicated in the Conclusions section)? What is the quantitative relationship between temperature and charcoal mass retention? More specific and worthwhile conclusions need to be made from the dataset. At present, the manuscript is intriguing but does not provide explicit values and tools that can be applied to actual sediment samples, and in turn, inform paleofire interpretations.

R: These are all pertinent questions. I will update the Discussion of the manuscript to further explore and extend on: a) the cut-off for elongation indicative of graminoids?  B) the ranges of morphometric values that can be indicative of fuel types? C) the mean values of the aspect ratios of the fuel types that can be distinguished (wood, graminoids, and leaves), and D) the quantitative relationship between temperature and charcoal mass retention. On the other hand, as already stated in the Conclusion,  in many cases the overlapping aspect ratios between fuel types make it difficult to come with a clearly defined cut-off value or range values indicative of a specific fuel type.

Lastly, the author should provide the morphometric values derived from the experimental charcoal in Table 1 (or in the supplement) to enable others to more directly compare with this dataset. As it stands, future work will have to estimate values from the figures. To my knowledge, provision of explicit data value ranges is the norm in these types of studies. I strongly suggest the author provide these values to facilitate use of the insights provided by the manuscript.

R: I have prepared a table showing mean values of L:W, Length, and Area in the main paper, and one showing the full range of individual morphometric measurements across all temperatures in the Supplementary Material.

Specific Comments:

L103-104: What was the rationale for this sieve size, given the wide range of sieve sizes used in the paleofire field?

1. To get rid of smaller particles but this can anyway be done by automatic selection of grain size. I have added a sentence stating this: The sieving step could be skipped if the size of charcoal particles is detected automatically.

L203-206: Shouldn’t this be irrelevant given that the sampled were uniformly dried before combustion?

R: Perhaps. However, with this sentence I wanted to point out that while all other material collected was from alive plants,  the trunk wood comes from a dead tree.

L243: Neither of these citations are provided in the references list. The Clark (Clark 1988) and Higuera (Peters and Higuera 2007, Higuera et al. 2007) models of charcoal dispersal do not actually incorporate particle shape. To my knowledge, the model of Vachula and Richter (2018) is the only one to directly test the effect of charcoal particle shape on dispersal distance.

R: Thank you for pointing out this. Indeed, most models assume similar, round particles when simulating the particles' transports. Clark and Peters (2007) made a difference between longer versus rounded particles. They have derived a velocity index, which determined that elongated particles have a lower velocity index and have consequently a longer time of residency into the atmosphere than the rounder one.  I have extended this chapter that now reads: ‘’ The shape and density of charcoal particles affect their transport (Clark and Hussey, 1996; Higuera et al., 2014). Models and empirical data generally indicate that the amount of charcoal particle is greatest near the fire source (Clark et al., 1998; Clark and Royall, 1995; Higuera et al., 2007; Peters and Higuera, 2007; Tinner et al., 2006). However, models assume a uniform spherical particle shape and density, when in reality charcoal particles are of different shapes, sizes, and densities. Indeed, recent modelling results show that non-spherical particles have lower settling velocities than spherical particles and produce a spatially more extensive and heterogeneous particle-size distribution pattern (Vacula and Richter, 2018). Specifically, median dispersal distances for spherical and aspherical particles (>150 µm) could be up to 20 km apart (Vacula and Richter 2018). Similarly, Clark and Hussey (1996), derived a velocity index from sedimentary charcoal particles also found that non-spherical particles have lower setting velocities and higher residence time into the atmosphere than the elongated particles.

L248: Aleman et al. did not experimentally produce charcoal particles. The values referred to here were derived from environmental samples. This comparison is not appropriate, in my opinion.

R: Removed from here but added in the new Table dealing with comparisons with published literature.

L297-311: This conflation of fuel type and burn temperature is not convincing. Although the data presented in this paper clearly show the ability to differentiate fuel types, it seems a stretch to suggest that burn temperature might be inferred from fuel type assemblages and charred mass. How would this work for an environmental sample? How could charred mass be differentiated from total fire activity? Why isn’t Figure 2H referred to and discussed in this section? It should be useful in making these conclusions.

R: The differentiation of fuel types is not straightforward and cannot be made based on charred mass alone, but by looking at the association between charcoal morphologies and fuel type. For example, if we know (based on charred mass) that Sphagnum preserves as charcoal only after low-temperature burnings,  and if we find abundant Sphagnum charcoal morphotypes in the sediments, this could potentially mean that fire temperatures were low, otherwise Sphagnum would not have been preserved and be found as charcoal. Comparison with pollen and plant macrofossils would improve such interpretation. I’m therefore working to extend this aspect in the revised manuscript. Fig2H shows the aspect ratio for fuel mix at a single temperature (300 degrees) and has been used to determine the accuracy of aspect ratio in mix fuel types, as normally occurs in nature.

L359: Where are the morphometric measurement data? How can future researchers actually use these data to better inform their interpretations if they are not provided?

R: In the revised manuscript I will provide a) a new Table with median and SD values for all individual species and all temperatures; b) a new Table providing comparative results from this study to those from literature, and c) and a new Table SI; with full individual measurements at all temperatures.

Figures 2 and 3: The figure captions indicate that the boxplots summarize the median aspect ratios, lengths, and areas of particles for each taxa (i.e., each box plot depicts the median, standard deviation, and range of the median values of the measurements). If I understand correctly, though, these boxplots are actually summarizing the individual measurements. The medians are just one component of the boxplots? This ought to be clarified.

R: I have added this information, the new caption reads: The median aspect ratios of charred particles from (a–d) the individual measurements taxa burned at 250, 300, 350, and 400 °C, respectively, and (e–g) fuel types at burning temperatures.

Technical and typographical corrections:

L14-15: “Graminoids, Sphagnum, and wood”

R: Corrected. This sentence reads: Graminoids, Sphagnum, and wood (trunk) lose the most mass at low burn temperatures, whereas heathland shrub leaves, brown moss, and ferns retain the most mass at high burn temperatures.

L38: End parenthesis is missing from citation.

R: Parenthesis included.

L48: Consider rephrasing. No fires are ‘cool’. Consider using more specific fire regime characteristics (e.g., intensity, severity).

R: Rephrased. This sentence reads: This analytical limitation restricts the reconstruction of fuel sources, a crucial factor in determining fire type, i.e., the burning of surface fuels in low or high-intensity fires, or distinguishing between surface and crown fires, which requires greater distinction of fuel types (Courtney-Mustaphi and Pisaric, 2014; Feurdean et al., 2017; Hawthorne et al., 2018).

L71: “paper” singular

R: Done. The paper also discusses the advantages and limitations of laboratory-based burning studies for palaeofire reconstruction.

L85: “tests”

T: Done: This paper builds on previous morphological classifications of charcoal particles and tests the effects of increasing ...

L248: “Mustaphi”

R: Mustaphi and Pisaric (2014) also observed that burning monocotyledons from…

L287-288: “influence”

Done. Fuel arrangements also strongly influence fire propagation.

L292: “cool, surface fueled”

R: Done. Siberian forests burn alternatively as cool, surface fueled by graminoids…

L293: “Anderson”

R: Done, Anderson, 1982

L297-300: Please consider revising this sentence. Its present wording is difficult to decipher.

R: I have removed the sentence This has a practical importance for determining the ranges of fire severity and frequency that Siberian boreal tree taxa have tolerated in the past, and helps to evaluate their potential to adapt to new fire regimes in the future, so that the transition from the theoretic part to the practical application is directly visible. It reads: ‘’Siberian forests burn alternatively as cool, surface fuelled by graminoids, forbs, ferns, mosses or as hot, high-intensity surface (burning shrubs) and crown fires (Anderson, 1982). Results from this study suggest that the combined use of morphometric and morphological features and charred mass can help distinguish the predominant fuel source. Knowledge of the fuel source may in turn provide clues on fire type, i.e., the combination of fire intensity (temperature) and severity (effect on vegetation).  For example, charcoal particles with a higher aspect ratio, typical for graminoids, and dominantly graminoid morphologies that tend to preserve only at a lower temperature, likely indicate a graminoid fuel source, and therefore a cooler, lower-intensity fire (Fig. 5).’’

L310: “reveal”

R: Done: The burning experiments also reveal that charred particles …

References cited in this review:

Clark, J. S. (1988). Particle motion and the theory of charcoal analysis: source area, transport, deposition, and sampling. Quaternary research, 30(1), 67-80.

Crawford, A. J., & Belcher, C. M. (2014). Charcoal morphometry for paleoecological analysis: The effects of fuel type and transportation on morphological parameters. Applications in plant sciences, 2(8), 1400004.

Higuera, P. E., Peters, M. E., Brubaker, L. B., & Gavin, D. G. (2007). Understanding the origin and analysis of sediment-charcoal records with a simulation model. Quaternary Science Reviews, 26(13-14), 1790-1809.

Peters, M. E., & Higuera, P. E. (2007). Quantifying the source area of macroscopic charcoal with a particle dispersal model. Quaternary Research, 67(2), 304-310.

Vachula, R. S., & Richter, N. (2018). Informing sedimentary charcoal-based fire reconstructions with a kinematic transport model. The Holocene, 28(1), 173-178.

Citation: https://doi.org/10.5194/bg-2021-1-RC2

General comments

This manuscript provides new and additional information to the increasing body of knowledge of using sedimentary charcoal for more robust reconstructions of past fire regime including fire severity, fuel sources and fire type. The study combines different features analysed from modern burned plant materials and based on that provide assessment of possible methodology for defining fire type and fuel source also from sedimentary records. It would be interesting to see a comparison of the features from plant material burned in laboratory with features from actual sedimentary charcoal record from the same study area. However, I understand that this could be further work in addition to this manuscript. In general, this manuscript is well organized and written in clear language. Introduction is informative and the aims of the study are clearly stated. Methods used in the study are relevant and justified and the approach provides novel information and methodology for using sedimentary charcoal in reconstructing past fire regime in boreal landscape. The results are interesting and I would have hoped to see more comparison of the results from this study to previous research also using laboratory techniques or sedimentary charcoal to analyze charcoal morphology in regards to the combustion temperature and fuel source. I was also partly missing clear numerical values when describing the morphological features. For example, I would have hoped more defined information of what is the size range of smaller and larger fragments, when morphometrically categorizing fuel types. More exact numerical information of size, ratios, mass retention etc. would make these results more comparable with later studies utilizing the methodology presented here. Overall, I think this work brings important addition to the literature and methodology of using combination of different charcoal features for more robust reconstructions of past fire regime. I havemade some minor suggestions for the author in the detailed comments. In my opinion, this paper would be in wide interest of the readers of Biogeosciences and I recommend this paper to be accepted to publication after suggested minor revisions.

R: Many thanks for the positive response that this paper was interesting and a useful contribution to the community, and for the useful and thoughtful comments to further improve this paper. I am currently working to 1) Expand the Discussion on how the results from this study compared to previous research using laboratory techniques or sedimentary charcoal to analyse charcoal morphology in regards to the combustion temperature and fuel source, including a table showing numerical information on aspect ratio, length, and area; 2) Clarify the numerical information that distinguishes the size range of smaller and larger fragments; and 3) I am working on a few selected fossil sample to compare the features from plant material burned in the laboratory with features from the actual sedimentary charcoal record in the study area.

Detailed comments:

L 79: I would suggest adding right in the beginning that these identified plant materials were the actual samples that were later burned in the laboratory.

R: Added: ‘‘Plant materials used for laboratory burning experiments were identified in the field, stored in plastic bags for transportation, and air-dried.

L 87-88: So, nothing was actually used to initiated flame, but the burning was due to high temperatures? How well does this mimic the natural conditions for fire and does it have an effect on the charcoal features compared to the ones in sedimentary record?

R: Right, nothing was done to initiate the flame. Laboratory burning conditions are not identically to the natural wildfires (see the response to Rev1 for limitation of laboratory burning experiments). Results from charcoal morphometrics obtained under open flame to those in muffle oven compare well (Umbanhowar and McGrath, 1998; Orvis et al., 2005) and charcoal features obtain in the laboratory with those from sediment wildfire (Mustaphi and Pisaric, 2014) .

L 92-95: I would suggest to mark the mixed samples a bit more clearly. Now it takes some time to figure out that which proportion go to which sample.

R: R: Thank you, I agree that this is not that easy to read.  In the current version I present these samples along a gradient of transition from predominant surface fuel fire that burn with lower intensity (graminoid and moss), to surface fuel that burns with intermediate- to high-intensity (shrub), and finally to crown fuel that burns at high-intensity crown fires (wood and tree leaf). I am working to make the labels clearer.

L 126-128: Here is reference to the Fig. 1. However, there isn’t as many types given in the figure as here in the text. This is a bit confusing and I recommend to fix this or adding some explanation for leaving some features out form the figure.

R: Many thanks. The description of Fig. 1 in the text follows the groping of similar fuel types at the bottom of this figure.  In the revised manuscript I have better matched the names of these groups in Fig. 1 with those in the text. It reads: ‘’The percentage of charred mass retained at 300 °C (an intermediate temperature) was as follows, in decreasing order: moss and fern > wood (shrub twig) > leaf (shrub) > leaf (forb)> > wood (tree twig) > leaf (needles tree) > graminoid > Sphagnum > wood (trunk) (Fig. 1).’’

L 197-199: Here average results across all temperatures are referred to Fig. 1, but as far as I understand the figure presents results from different burning temperatures rather than averaged over all temperatures. I would suggest this to be clarified.

R: Thank you, I have revised the text to better illustrate Fig 1. It now reads: ‘’Comparison of results across temperatures suggest that graminoid, Sphagnum, and trunk wood produce the lowest amounts of charcoal per unit biomass and lost their mass more rapidly with increasing burning temperature (Fig. 1). Contrastingly, leaves of shrubs (Ericaceae), forbs, and ferns (Polypodiaceae), as well as fern stems (Equisetum), produced the highest amounts of charcoal per unit biomass and retained the most charred at higher temperatures (Fig. 1).

L 235-237: Here it is stated that larger fragments are more reliable to categorise fuel types. It would be useful to clearly state that what is the size ranges for what is considered larger and smaller fragments.

R: Indeed, in the original manuscript I have stated ‘’Because smaller charred fractions tend to be rounder (lower aspect ratio) than larger fractions, it suggests that larger charcoal fragments can be more confidently used to morphometrically categorise fuel types’’. In fact, results from this study show that the smaller the particles, the smaller the aspect ratio was. From here I have extrapolated that larger charcoal fragments can be more confidently used to morphometrically categorise fuel types. However, it is difficult to find a clear cut of the actual size that could more reliably be used for aspect ratio, but I will explore this aspect further.