the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 May 2025
Fire activity in the northern Arctic tundra now exceeds late Holocene levels, driven by increasing dryness and shrub expansion
Abstract. Tundra ecosystems are characterized by small, rare and infrequent fires due to cold, often waterlogged conditions, and limited biomass. However, ongoing climate warming and drying in northern soils and peatlands contribute to increasingly frequent and extensive wildfires. To place recent fire regimes in the context of long-term variability and to better understand interactions between moisture, vegetation, and fire, we reconstructed wildfire history over the past 3000 years using a network of charcoal records in combination with vegetation and hydrological datasets and satellite-derived fire datasets from northern Arctic Alaska. Composite charcoal records show minimal fire activity from ~1000 BCE to 500 CE, followed by a slight increase from 500 CE onwards. This long-term pattern shifted abruptly around 1880 CE, when fire activity exceeded any levels observed in the preceding three millennia. Individual charcoal records show a more heterogeneous fire pattern before 1880 CE and a more homogeneous one thereafter. Our findings suggest that deepening of water tables and peatland drying associated with permafrost thaw have facilitated woody encroachment, especially by more flammable Ericaceous shrubs. These vegetation changes have increased fuel availability and flammability, ultimately driving the recent surge in wildfire activity. This study highlights the importance of moisture–vegetation–fire feedback in shaping tundra fire regimes and the vulnerability of Arctic ecosystems to fire. This is particularly evident in areas experiencing pronounced drying and the expansion of flammable shrub taxa. We also found that the charcoal source area of our tundra fire encompasses broader landscapes over tens of kilometre.
- Preprint
(3029 KB) - Metadata XML
-
Supplement
(25 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-2318', Ramesh Glückler, 19 Jun 2025
General comments:
In their manuscript “Fire activity in the northern Arctic tundra now exceeds late Holocene levels, driven by increasing dryness and shrub expansion”, the authors explore timely relationships between past wildfire dynamics, vegetation, and hydrology in the Alaskan tundra. For that they use a combination of new and previously published paleoecological data on past wildfire activity, vegetation composition, and the peatland water table, and combine this paleoecological data with more recent fire observations from remote sensing. The main result, summarized in the manuscript’s title, is based on increased reconstructed fire activity after 1880 CE, coinciding with drying and shrub encroachment. Overall, the manuscript presents a valuable perspective on past wildfire activity, especially by its approach to compile data from multiple sites and to include a reconstruction of the peatland water table. The comparison of charcoal-based fire reconstructions with more recent fire observations is intriguing and contributes to ongoing efforts of determining regional charcoal source area. The manuscript content fits well within the scope of Biogeosciences, and would be an interesting contribution to the journal for researchers across various disciplines, such as paleoecology, peatland ecology, fire ecology, hydrology, or tundra vegetation dynamics. However, in its current state the manuscript would in my opinion benefit from some further polishing and some clarifications regarding the proposed drivers behind the reconstructed fire activity. Before publication I would therefore recommend some moderate editing, including clarifications regarding the applied methods and interpretations, a more comprehensive description of the results, and some changes to the structure of the discussion to guide the reader more smoothly. My suggestions therefore do not concern the great underlying data and analysis, but rather potential improvements in the way these are presented and discussed, so the manuscript can stand at its full potential. From my perspective, a correspondingly revised version should be very much considered for publication in Biogeosciences. Please find more detailed comments below:
Specific comments:
Manuscript structure and clarifications
From my perspective, some changes to the manuscript’s structure would improve reading flow and potentially clarify some remaining questions. First, it was not always completely clear to me which data were already published before (and exactly where), and which were newly created, and for some of the data, I was missing a clear description. I think all of this may be mostly related to the short results section, which only includes sub-sections for the fire-related data (charcoal and remote sensing). However, in the methods section, also new chronological data, data on vegetation (from pollen and plant macrofossils), and data for the water table (testate amoebae) is mentioned. I think even if previously published, since they were also processed as part of this manuscript, they should also be featured and briefly described in the results section, where they are currently missing. Expanding the manuscript in this regard for methods would be appreciated as well, since it is currently not always clear which methods were applied due to the short nature of the description (e.g., in L122 a composite of shrub data (and later water table data) is mentioned, but lacking details on how exactly it was composited, and in L126 the abbreviation “DTW” is introduced, but not clearly described). If the testate amoebae data was published before, what were the main outcomes? How would the newly composited water table curve be described? Introducing the data in a bit more detail would help the reader follow along in the discussion. Speaking of the discussion, I would personally not start with the methodological interpretation of charcoal source area, but rather with the most important aspects of the manuscript mentioned in the title – although this is definitely a subjective preference. However, also here some aspects should be expanded upon. For example, in L195 historical fire observations were mentioned, but not further described. Which kind of fire activity was recorded between 1880 to 1920? Does it fit to the reconstructed charcoal-based signal? Finally, instead of the last sentence of the discussion, I suggest to include an actual conclusion section that briefly summarizes the findings and provides an additional outlook. I feel that would provide a smoother end to the manuscript.Interpretation – Fire before the drought?
I generally agree with the main conclusion of relating moisture conditions, fuel types, and fire activity, as this would be expected, and it would certainly be great to see this relationship clearly in paleoecological data. However, some aspects of the suggested relationships, as seen in the actual data, may need a bit more explanation. In L209, it is stated that the driest conditions of the Late Holocene (during the past few decades) coincided with the most intense/severe fire period. However, looking at Fig. 3a it seems rather that severe fires occurred between c. 1900 to 1970 CE, whereas in Fig. 3c it seems that drier-than-average conditions only occurred since ca. 2000 CE, when the levels of biomass burned were very low. A similar discrepancy is evident in the timing of increased shrub counts in Fig. 3b when compared to the reconstructed biomass burning, even though in L225 it says that woody biomass would have promoted fires. Somehow it seems to me as though the level of biomass burning increased first, and the shrubs and drying followed – but what would then be the driver behind the fire activity? How can these temporal differences be understood in light of the proposed relationships, and does proxy taphonomy play a role as well? In case I did not misinterpret the plots, I think that these aspects should be included in more detail in the discussion.Technical corrections:
L34: Suggest to re-phrase sentences, e.g. “[…] charcoal records in combination with data on vegetation, hydrology, and satellite-derived fire observations from […]”. In the following sentence: “A regional composite of charcoal records shows […]”
L44: Maybe instead of ending a bit abruptly after this result regarding source area, a sentence briefly summarizing the importance of the study, or providing some outlook, would be fitting here?
L46: Suggest to re-phrase to e.g. “[…] increase in the number, size, and intensity of individual fires, and in the length and duration of fire seasons […]”
L52: “Sayedi” et al., 2024
L53: Suggest to re-phrase here, e.g. “[…] contributed to a deviation from the previously low wildfire activity”
L67: “peatlands”
L66, 70: No reference for “Vachula, 2020” or “Vachula et al., 2020” is listed in the reference list. Do you refer to Vachula (2021) in L70 (see references below)? Maybe just double-check that all citations are mentioned in the reference list and vice-versa.
L73: “tussock sedges-moss-dwarf shrubs” seems like a bit of a clunky term, maybe there is another way to refer to this vegetative zone? Also note additional space before “dwarf”
L74: not sure if an “en”-dash is needed in “testate amoebae-based” – maybe double-check journal guidelines.
L85: “Annual precipitation averages 25 mm […]” – is that the average annual precipitation sum across multiple years? For that it seems quite low. Or do you refer to the average across all months of a year? In that case, I think that the first would provide better context. Maybe it is good to just clarify which average this refers to.
L92: I am not sure I can follow the description of how cores were retrieved with a shovel, maybe you can provide a bit more detail?
L101: Sites “Ga I and GA II” or sometimes written with a space, sometimes without (e.g., L126)
L107: Suggest to include here: “[…] intervals across cores from all sites”
L107: Which bleaching agent was used in the procedure? This should be stated in any case, but especially since previous analyses found that certain bleaching agents may dissolve charcoal particles from low-intensity fires (Constantine and Mooney, 2021), which (I suppose) are the kind of fires one may expect at these sites
L112: Sometimes “charcoal influx” is used, sometimes “charcoal accumulation rate (CHAR)”. I suggest to standardize these terms across the manuscript, as they seem to be used interchangeably.
L112: The unit description here is missing some “minus” and numbers – I think it should be “particles cm-2 yr-1”. This also goes for other instances throughout the manuscript (e.g., right afterwards: “particles cm-3” and “year cm-1”, or in L146) – Also note that in some figure captions, the unit is written as “#/cm2/yr”, I’d recommend to standardize one type across the manuscript.
L121: “smoothed”?
L136: Add space: “70 km” and standardize across the manuscript
L137: “[…] the distance from the nearest […]”
L145: Add space: “mean = 0.04 […]”
L148: In this case for ranges I think that “en”-dashes should indeed be used. Maybe consult the journal guidelines for a revision (same in L165, L170)
L152: Suggest to replace “the” with “an”
L154: “macro charcoal” – double check that this term is always used in correct and equal spelling across the manuscript (suggest to use either “macro-charcoal”, “macrocharcoal”, or “macroscopic charcoal”)
L158: Suggest to start with “Our satellite […]” and use the term “buffer” instead of “radius”?
L157: The “extremely large fire” is 103.896 ha or 103,896 ha? It may be good to double-check that the use of comma/dot for decimals is always correct and according to the international norm, including in the corresponding figure caption etc. In L160, a space is missing before “2 ha”, and later a “=” is missing in “median 115 ha”.
L168: Before, no spaces were used when mentioning e.g. “Fig. 4c”. Recommend to standardize across the manuscript.
L171: “[…] while the satellite data during 2001–2017” is not a full sentence, maybe this part could be re-phrased?
L176: Delete space at beginning of the line
L180: I would have expected some citations already within this sentence, to make it clear which study is related to the 2 km source area or the wider area. In the next sentence “Vachula et al., 2020” is mentioned, which may again may need to be corrected.
L186: Missing comma in citation
L191: I am not sure data on fire intensity was presented here? As it’s a similar theme, I would also refrain from mentioning an “intense fire period” (L209) and rather use the term “severe”, to reduce potential confusion about fire regime terminology.
L211: “Smaller-amplitude deepening of the water table […]”?
L220: Suggest to add commas: “[…] fuel type and availability, and fire, particularly […]”
Figure 2: I am not sure if I understand this sentence in the caption: “The total charcoal influx represents the unidentified charcoal.” – could that be re-phrased and clarified? Also, note the CHAR unit being written differently here than in the main text. In L432, standardize the use of spaces and capitals when defining “Common Era” (including other instances across the manuscript). In the last sentence of the caption, it says that the smaller plots on the right side cover the last two centuries, yet their x-axes seem only to reach back to 1960 CE?
Figure 3: In caption in L442: Is it correct to write “[…] as counts. (c) Composite record […]”? Regarding the figure itself, would there be a way to separate the inserted plots more clearly from the main plots, so the numbers don’t overlap with the background plot? For example, they could all be decreased in size a little bit and inserted on the left side of the respective main plot, where there is more space. Two of the inserted plots also currently miss x-axes. The y-axis description of plot a) is not completely clear to me – it should probably say that higher values mean more biomass burning, but from just the one arrow/symbol it is not fully clear. Maybe it would be better just to use two arrows and a description, such as in plot c)? Also, it seems that the x-axis line in plot c) has a variable thickness, whereas it is not visible at all in some of the other inserted plots.
Figure 4: In caption, I suggest to list the cut peaks as “a) 2001 (341 ha), 2007 (659 ha); b) 2007 (103.896 ha)” – note the b) 2007 value again here, is it really supposed to be 103.9 ha? In the last sentence, standardize use of spaces around “=”. In the figure itself, I am wondering if a) and b) could not be combined into a single plot, with the bars of different width in the back and CHAR in the front, maybe in a different color? Also, is it really “Cumulative charcoal influx” in a)? The compilation method by Blarquez et al. (2014) does not just sum the different records, so it’s likely not cumulative – if that was done here, it should be stated in the methods section. In plot c), there are variable distances between the legend entries, and the x-axis line is interrupted.
Table 1: “Coordinates”
Appendix C: Suggest writing either just “particles >500 µm” or “particles larger than 500 µm”.
Data availability: Ideally, the data could be uploaded during the revision so that the final DOIs can be included in the manuscript, which would make finding the data a lot easier in the future. In any case, I’d recommend to fully reference Neotoma, e.g.: “[…] deposited to the Neotoma Paleoecology Database (www.neotomadb.org; Williams et al., 2018)”
Supplement S1: “Satellite-based”
Author contribution: “conceived the study”
Affiliations: Suggest to double-check spelling of institutions, order of institutional levels, and the inclusion of the full address
References mentioned in this review:
Blarquez, O., Vannière, B., Marlon, J. R., Daniau, A.-L., Power, M. J., Brewer, S., & Bartlein, P. J. (2014). paleofire: An R package to analyse sedimentary charcoal records from the Global Charcoal Database to reconstruct past biomass burning. Computers & Geosciences, 72, 255–261. https://doi.org/10.1016/j.cageo.2014.07.020
Constantine, M., & Mooney, S. (2021). Widely used charcoal analysis method in paleo studies involving NaOCl results in loss of charcoal formed below 400°C. The Holocene, 09596836211041740. https://doi.org/10.1177/09596836211041740
Vachula, R. S. (2021). A meta-analytical approach to understanding the charcoal source area problem. Palaeogeography, Palaeoclimatology, Palaeoecology, 562, 110111. https://doi.org/10.1016/j.palaeo.2020.110111
Williams, J. W., Grimm, E. C., Blois, J. L., Charles, D. F., Davis, E. B., Goring, S. J., Graham, R. W., Smith, A. J., Anderson, M., Arroyo-Cabrales, J., Ashworth, A. C., Betancourt, J. L., Bills, B. W., Booth, R. K., Buckland, P. I., Curry, B. B., Giesecke, T., Jackson, S. T., Latorre, C., … Takahara, H. (2018). The Neotoma Paleoecology Database, a multiproxy, international, community-curated data resource. Quaternary Research, 89(1), 156–177. https://doi.org/10.1017/qua.2017.105
Citation: https://doi.org/10.5194/egusphere-2025-2318-RC1 -
AC1: 'Reply on RC 1 and RC 2', Angelica Feurdean, 29 Jul 2025
29 July, 2025, Frankfurt am Main
We thank both reviewers for providing encouraging feedback and valuable comments on the manuscript, which will help enhance its accuracy and readability. We agree that some of the methods and interpretations, as well as parts of the discussions, were brief. In the revised manuscript, we expanded these parts based on these suggestions. For more details on the main comments and suggestions, see our response to specific comments below. The complete response to all comments will follow together with the revised manuscript.
Ramesh Glückler General comments:
In their manuscript “Fire activity in the northern Arctic tundra now exceeds late Holocene levels, driven by increasing dryness and shrub expansion”, the authors explore timely relationships between past wildfire dynamics, vegetation, and hydrology in the Alaskan tundra. For that they use a combination of new and previously published paleoecological data on past wildfire activity, vegetation composition, and the peatland water table, and combine this paleoecological data with more recent fire observations from remote sensing. The main result, summarized in the manuscript’s title, is based on increased reconstructed fire activity after 1880 CE, coinciding with drying and shrub encroachment. Overall, the manuscript presents a valuable perspective on past wildfire activity, especially by its approach to compile data from multiple sites and to include a reconstruction of the peatland water table. The comparison of charcoal-based fire reconstructions with more recent fire observations is intriguing and contributes to ongoing efforts of determining regional charcoal source area. The manuscript content fits well within the scope of Biogeosciences, and would be an interesting contribution to the journal for researchers across various disciplines, such as paleoecology, peatland ecology, fire ecology, hydrology, or tundra vegetation dynamics. However, in its current state the manuscript would in my opinion benefit from some further polishing and some clarifications regarding the proposed drivers behind the reconstructed fire activity. Before publication I would therefore recommend some moderate editing, including clarifications regarding the applied methods and interpretations, a more comprehensive description of the results, and some changes to the structure of the discussion to guide the reader more smoothly. My suggestions therefore do not concern the great underlying data and analysis, but rather potential improvements in the way these are presented and discussed, so the manuscript can stand at its full potential. From my perspective, a correspondingly revised version should be very much considered for publication in Biogeosciences. Please find more detailed comments below:
Specific comments: Manuscript structure and clarifications
1. From my perspective, some changes to the manuscript’s structure would improve reading flow and potentially clarify some remaining questions. First, it was not always completely clear to me which data were already published before (and exactly where), and which were newly created, and for some of the data, I was missing a clear description. I think all of this may be mostly related to the short results section, which only includes sub-sections for the fire-related data (charcoal and remote sensing). However, in the methods section, also new chronological data, data on vegetation (from pollen and plant macrofossils), and data for the water table (testate amoebae) is mentioned. I think even if previously published, since they were also processed as part of this manuscript, they should also be featured and briefly described in the results section, where they are currently missing. Expanding the manuscript in this regard for methods would be appreciated as well, since it is currently not always clear which methods were applied due to the short nature of the description (e.g., in L122 a composite of shrub data (and later water table data) is mentioned, but lacking details on how exactly it was composited, and in L126 the abbreviation “DTW” is introduced, but not clearly described). If the testate amoebae data was published before, what were the main outcomes? How would the newly composited water table curve be described? Introducing the data in a bit more detail would help the reader follow along in the discussion.R: Methods. In revising this manuscript, we have expanded the Methods section in the main text (3.1 Chronology) and the Appendices to include the chronology of all sites, not just the three unpublished sites (SG, RH, and DL). Additionally, we created separate subchapters to expand the description of testate amoebae water table DWT reconstruction (3.3 Testate amoebae reconstruction of hydrological changes) and pollen and plant macrofossils-based vegetation reconstructions (3.4 Plant macrofossils and pollen-based reconstruction of vegetation dynamics), including the creation of their respective composite records.
Results. Following the same logic as for methods, we have created separate subchapters that include brief descriptions of results from pollen and plant macrofossil-based vegetation reconstructions, as well as testate amoebae water table DWT reconstructions.
- Speaking of the discussion, I would personally not start with the methodological interpretation of charcoal source area, but rather with the most important aspects of the manuscript mentioned in the title – although this is definitely a subjective preference. However, also here some aspects should be expanded upon. For example, in L195 historical fire observations were mentioned, but not further described. Which kind of fire activity was recorded between 1880 to 1920? Does it fit to the reconstructed charcoal-based signal? Finally, instead of the last sentence of the discussion, I suggest to include an actual conclusion section that briefly summarizes the findings and provides an additional outlook. I feel that would provide a smoother end to the manuscript.
R: Thank you for the suggestions regarding the manuscript's structure. The reason for starting the Discussion with the methodological Interpretation of charcoal source areas is that readers have an overview of what charcoal records represent in terms of distance from the fires and biomass burnt. Once these aspects are established, readers will be able to understand the long-term charcoal record more effectively. For this reason, we would like to maintain the manuscript's structure, starting with the Interpretation of charcoal source areas. However, we fully agree with the reviewer's suggestion to include a separate conclusion section, in which information on the main findings from this study, along with further outlook, is presented. Regarding historical fire observations (L195), they refer to evidence coming from aerial photographs published in Miller et al. (2023). We will clarify these aspects in the revised manuscript.
Interpretation – Fire before the drought?
I generally agree with the main conclusion of relating moisture conditions, fuel types, and fire activity, as this would be expected, and it would certainly be great to see this relationship clearly in paleoecological data. However, some aspects of the suggested relationships, as seen in the actual data, may need a bit more explanation. In L209, it is stated that the driest conditions of the Late Holocene (during the past few decades) coincided with the most intense/severe fire period. However, looking at Fig. 3a it seems rather that severe fires occurred between c. 1900 to 1970 CE, whereas in Fig. 3c it seems that drier-than-average conditions only occurred since ca. 2000 CE, when the levels of biomass burned were very low. A similar discrepancy is evident in the timing of increased shrub counts in Fig. 3b when compared to the reconstructed biomass burning, even though in L225 it says that woody biomass would have promoted fires. Somehow it seems to me as though the level of biomass burning increased first, and the shrubs and drying followed – but what would then be the driver behind the fire activity? How can these temporal differences be understood in light of the proposed relationships, and does proxy taphonomy play a role as well? In case I did not misinterpret the plots, I think that these aspects should be included in more detail in the discussion.R: Following the reviewer's two suggestions, we have additionally calculated the charcoal accumulation rate (CHAR) on charcoal values interpolated at the median temporal resolution of each record to account for marked changes in the sedimentation rate in records in the uppermost sediments compared to the rest of the core. This interpolated composite CHAR record aligns temporally more closely with composite records in DTW and vegetation. However, some temporary asynchrony changes in fire-vegetation-moisture persist, and we believe that taphonomic processes connected to the type of biomass burned (see our response to reviewer 2) and the spatial resolution of different proxies. In the revised version, we will expand the discussion to provide a more precise explanation of these processes.
Technical corrections:.
R: thank you for technical and language corrections, we will carefully incorporate them in the revision of the manuscript. To most of those who asked a question and not referring to rephrasing, we responded below.
L34: Suggest to re-phrase sentences, e.g. “[…] charcoal records in combination with data on vegetation, hydrology, and satellite-derived fire observations from […]”. In the following sentence: “A regional composite of charcoal records shows […]”
L44: Maybe instead of ending a bit abruptly after this result regarding source area, a sentence briefly summarizing the importance of the study, or providing some outlook, would be fitting here?
L46: Suggest to re-phrase to e.g. “[…] increase in the number, size, and intensity of individual fires, and in the length and duration of fire seasons […]”
L52: “Sayedi” et al., 2024
L53: Suggest to re-phrase here, e.g. “[…] contributed to a deviation from the previously low wildfire activity”
L67: “peatlands”
L66, 70: No reference for “Vachula, 2020” or “Vachula et al., 2020” is listed in the reference list. Do you refer to Vachula (2021) in L70 (see references below)? Maybe just double-check that all citations are mentioned in the reference list and vice-versa.
R: We refer to Vachula RS, Sae-Lim J, Russell JM. Sedimentary charcoal proxy records of fire in Alaskan tundra ecosystems. Palaeogeography, Palaeoclimatology, Palaeoecology. 1;541:109564.
L73: “tussock sedges-moss-dwarf shrubs” seems like a bit of a clunky term, maybe there is another way to refer to this vegetative zone? Also note additional space before “dwarf”
R: We will simplify the term as tussock and shrubby tussock tundra.
L74: not sure if an “en”-dash is needed in “testate amoebae-based” – maybe double-check journal guidelines.
L85: “Annual precipitation averages 25 mm […]” – is that the average annual precipitation sum across multiple years? For that it seems quite low. Or do you refer to the average across all months of a year? In that case, I think that the first would provide better context. Maybe it is good to just clarify which average this refers to.
R: Thank you for spotting the mistake, a zero was missing! Annual precipitation averages 250 mm.
L92: I am not sure I can follow the description of how cores were retrieved with a shovel, maybe you can provide a bit more detail?
L101: Sites “Ga I and GA II” or sometimes written with a space, sometimes without (e.g., L126)
L107: Suggest to include here: “[…] intervals across cores from all sites”
L107: Which bleaching agent was used in the procedure? This should be stated in any case, but especially since previous analyses found that certain bleaching agents may dissolve charcoal particles from low-intensity fires (Constantine and Mooney, 2021), which (I suppose) are the kind of fires one may expect at these sites
R: We used sodium hypochlorite (household bleach).
L112: Sometimes “charcoal influx” is used, sometimes “charcoal accumulation rate (CHAR)”. I suggest to standardize these terms across the manuscript, as they seem to be used interchangeably.
R: Thank you, for consistence we have decided to sue charcoal accumulation rate (CHAR).
L112: The unit description here is missing some “minus” and numbers – I think it should be “particles cm-2 yr-1”. This also goes for other instances throughout the manuscript (e.g., right afterwards: “particles cm-3” and “year cm-1”, or in L146) – Also note that in some figure captions, the unit is written as “#/cm2/yr”, I’d recommend to standardize one type across the manuscript.
R: Thank you, we will uniformised both the name and units measure through the manuscript.
L121: “smoothed”?
L136: Add space: “70 km” and standardize across the manuscript
L137: “[…] the distance from the nearest […]”
L145: Add space: “mean = 0.04 […]”
L148: In this case for ranges I think that “en”-dashes should indeed be used. Maybe consult the journal guidelines for a revision (same in L165, L170)
L152: Suggest to replace “the” with “an”
L154: “macro charcoal” – double check that this term is always used in correct and equal spelling across the manuscript (suggest to use either “macro-charcoal”, “macrocharcoal”, or “macroscopic charcoal”)
R: We will standardise the term to the term macrocharcoal thought the manuscript.
L158: Suggest to start with “Our satellite […]” and use the term “buffer” instead of “radius”?
L157: The “extremely large fire” is 103.896 ha or 103,896 ha? It may be good to double-check that the use of comma/dot for decimals is always correct and according to the international norm, including in the corresponding figure caption etc. In L160, a space is missing before “2 ha”, and later a “=” is missing in “median 115 ha”.
R: It is a very large fire of over 103 kiloha.
L168: Before, no spaces were used when mentioning e.g. “Fig. 4c”. Recommend to standardize across the manuscript.
L171: “[…] while the satellite data during 2001–2017” is not a full sentence, maybe this part could be re-phrased?
L176: Delete space at beginning of the line
L180: I would have expected some citations already within this sentence, to make it clear which study is related to the 2 km source area or the wider area. In the next sentence “Vachula et al., 2020” is mentioned, which may again may need to be corrected.
L186: Missing comma in citation
L191: I am not sure data on fire intensity was presented here? As it’s a similar theme, I would also refrain from mentioning an “intense fire period” (L209) and rather use the term “severe”, to reduce potential confusion about fire regime terminology.
R: We will use fire frequency and size.
L211: “Smaller-amplitude deepening of the water table […]”?
L220: Suggest to add commas: “[…] fuel type and availability, and fire, particularly […]”
Figure 2: I am not sure if I understand this sentence in the caption: “The total charcoal influx represents the unidentified charcoal.” – could that be re-phrased and clarified? Also, note the CHAR unit being written differently here than in the main text. In L432, standardize the use of spaces and capitals when defining “Common Era” (including other instances across the manuscript). In the last sentence of the caption, it says that the smaller plots on the right side cover the last two centuries, yet their x-axes seem only to reach back to 1960 CE?
R: Apologies for the clumpy sentence and inconsistency. Total charcoal influx refers to the total amount of charcoal at each site, as opposed to the charcoal that was differentiated into morphological types.
Figure 3: In caption in L442: Is it correct to write “[…] as counts. (c) Composite record […]”? Regarding the figure itself, would there be a way to separate the inserted plots more clearly from the main plots, so the numbers don’t overlap with the background plot? For example, they could all be decreased in size a little bit and inserted on the left side of the respective main plot, where there is more space. Two of the inserted plots also currently miss x-axes. The y-axis description of plot a) is not completely clear to me – it should probably say that higher values mean more biomass burning, but from just the one arrow/symbol it is not fully clear. Maybe it would be better just to use two arrows and a description, such as in plot c)? Also, it seems that the x-axis line in plot c) has a variable thickness, whereas it is not visible at all in some of the other inserted plots.
Figure 4: In caption, I suggest to list the cut peaks as “a) 2001 (341 ha), 2007 (659 ha); b) 2007 (103.896 ha)” – note the b) 2007 value again here, is it really supposed to be 103.9 ha? In the last sentence, standardize use of spaces around “=”. In the figure itself, I am wondering if a) and b) could not be combined into a single plot, with the bars of different width in the back and CHAR in the front, maybe in a different color? Also, is it really “Cumulative charcoal influx” in a)? The compilation method by Blarquez et al. (2014) does not just sum the different records, so it’s likely not cumulative – if that was done here, it should be stated in the methods section. In plot c), there are variable distances between the legend entries, and the x-axis line is interrupted.
R: Thank you for your suggestions to correct the figures and figure captions. We will carefully incorporate your suggestions in the revised version of the manuscript. Basically, on figure four is the charcoal accumulation rate at all sites (not composite),
Table 1: “Coordinates”
Appendix C: Suggest writing either just “particles >500 µm” or “particles larger than 500 µm”.
Data availability: Ideally, the data could be uploaded during the revision so that the final DOIs can be included in the manuscript, which would make finding the data a lot easier in the future. In any case, I’d recommend to fully reference Neotoma, e.g.: “[…] deposited to the Neotoma Paleoecology Database (www.neotomadb.org; Williams et al., 2018)”
R: Thank you. We will add all datasets used in this manuscript to the SI and send this to Neotoma
Supplement S1: “Satellite-based”
Author contribution: “conceived the study”
Affiliations: Suggest to double-check spelling of institutions, order of institutional levels, and the inclusion of the full address
References mentioned in this review:
Blarquez, O., Vannière, B., Marlon, J. R., Daniau, A.-L., Power, M. J., Brewer, S., & Bartlein, P. J. (2014). paleofire: An R package to analyse sedimentary charcoal records from the Global Charcoal Database to reconstruct past biomass burning. Computers & Geosciences, 72, 255–261. https://doi.org/10.1016/j.cageo.2014.07.020
Constantine, M., & Mooney, S. (2021). Widely used charcoal analysis method in paleo studies involving NaOCl results in loss of charcoal formed below 400°C. The Holocene, 09596836211041740. https://doi.org/10.1177/09596836211041740
Vachula, R. S. (2021). A meta-analytical approach to understanding the charcoal source area problem. Palaeogeography, Palaeoclimatology, Palaeoecology, 562, 110111. https://doi.org/10.1016/j.palaeo.2020.110111
Williams, J. W., Grimm, E. C., Blois, J. L., Charles, D. F., Davis, E. B., Goring, S. J., Graham, R. W., Smith, A. J., Anderson, M., Arroyo-Cabrales, J., Ashworth, A. C., Betancourt, J. L., Bills, B. W., Booth, R. K., Buckland, P. I., Curry, B. B., Giesecke, T., Jackson, S. T., Latorre, C., … Takahara, H. (2018). The Neotoma Paleoecology Database, a multiproxy, international, community-curated data resource. Quaternary Research, 89(1), 156–177. https://doi.org/10.1017/qua.2017.105
Review 2 General comments:
This manuscript contributes to our understanding of fire-regime variability on millennial to centennial timescales on the Alaskan North Slope along the Dalton Highway corridor. The authors use macroscopic charcoal counts from peatland soils to reconstruct charcoal influx associated with past fires. They also reconstruct vegetation compositional changes at local and regional scales using a combination of macrofossil and pollen identification from their sites, respectively. These data, along with previous reconstructions for water table changes from testate amoebae at a subset of sites provides information for qualitatively evaluating feedbacks between hydrology, vegetation, and landscape flammability. The authors also use spatially explicit fire datasets to constrain the source area of the macroscopic charcoal counts. Overall, this is a well-developed dataset and robust contribution that expands our understanding of the palaeoecological history of this rapidly changing region. I appreciate that the findings are generally not overstated, with the exception of the manuscript title that makes a very bold claim which may only tenuously be supported by the data (please see specific comments). Overall, I think that some revisions are needed to clarify the interpretations of the datasets and the overarching conclusions. Specifically, greater attention to chronological control and constraints, as well as a more nuanced explanation of the some of the temporal patterns, would improve the manuscript. I hope the specific comments I provide below are helpful to the authors.
Specific comments:
- I would really like to see some background information in the introduction as to the sources of charcoal in peatlands and the the pros/cons of using that setting for fire history reconstructions. This does not have to be overly long, but a bit of general background on the methods of extracting fire reconstructions from peatlands specifically would be quite useful to the reader.
R: Thank you. In the revised version of the manuscript, we have clarified that: 1) compared to lakes, peatlands provide local-scale reconstructions of past fire regimes, and that 2) this fine-scale resolution reflects the pronounced heterogeneity in local peat moisture, vegetation composition, and structure, and the relatively small catchment area.
- Chronological control is probably the most important part of any paleo study. I would highly recommend adding the age-depth models for all nine sites to the appendix so that readers can easily examine the fidelity of all models. A master table with all AMS and 210Pb ages per depth for all sites would also be useful to add the appendix.
R: Thank you for this observation. Following the reviewer 1 suggestions, in the revised version of the paper, we will provide a table including the AMS14C and Pb210 dates, as well as figures for all sites.
- One of the major claims in this paper is that fire activity since 1880 CE was higher than the previous millennia. At first glance, the data presented in Figure 2 is compelling (and exciting!). However, it is unclear to me if the charcoal influx data for each site has been interpolated to the median or average sampling resolution for each record. The reason I am concerned about this is that without a consistent temporal resolution, trends in influx could solely be a function of changes in the sedimentation rate. Specifically, the sedimentation rate changes dramatically in all records presented in the appendix in the uppermost sediments compared to the rest of the core. Equal time sampling will be important to compare past to recent trends, and should be explained carefully in the text if the take home is indeed that fire regimes have changed.
R: Thank you for this observation regarding the effect of changing sediment accumulation rates on trends in charcoal accumulation rates (CHAR). We have subsampled all records continuously at 1 cm resolution. In the original version of the paper, the charcoal accumulation rates have not been interpolated to median or average sampling resolution before the construction of the composite CHAR record (Fig 3), nor the CHAR plots at individual sites (Fig. 2). Indeed, the sedimentation rate increased markedly in the upper part of the record compared to the rest of the record. We have recalculated the CHAR i.e, interpolated the charcoal data to a median resolution for each record. We also used interpolated charcoal accumulation rates to construct the composite CHAR record. This interpolation appears to have slightly altered the pattern composite CHAR compared to the untransformed data. However, it preserves the increasing trend in recent times, implying that the key takeaway message — that recent fire activity is the highest compared to the late Holocene level — remains valid. The recent increase in composite CHAR values, based on interpolated values, appears to align more closely with those of plant macros and DTW. We are incorporating the new results in the revised version of the manuscript.
- Another point in regard to chronological control is line 37-38 in the abstract. Is it possible that the heterogenous fire patterns observed in the past compared to more recent sediments is a function of both 1) lower temporal resolution in the deeper parts of the peat cores and 2) the much larger age error of the deeper sediments? I think these types of issues should be discussed explicitly in the text to provide a more conservative and nuanced interpretation of the data.
R: Even though the temporal trends in CHAR are slightly shifted compared to the original, untransformed record, this interpolation does not affect the original observed pattern; that fire patterns were more heterogeneous among sites in the past compared to more recent times.
- In terms of the analysis comparing the timing of recent fires and the distance of fires from the depositional sites (summarized in figure 4) --- I am not sure I understand this analysis and it would be helpful to see some clarification. From my understanding, peaks in the charcoal influx data seem to capture deposition of charcoal from long distances, and yet do not capture the larger fires (specifically the Anaktuvuk River Fire that burned ~1000 km2)? I think the disconnect here is that I do not quite understand how the timing of the fire peaks were determined, and/or what the age model error around those timings actually is. From what I can tell, peak analysis to identify which peaks are most likely to represent fire events versus redeposition processes were not done on these cores. These types of peak analyses that are commonly used in lake sediment studies (with minimum count screening, signal-to-noise analyses, etc) may not be appropriate for peat cores (see comment #1), which is fine. However, I think it is important to carefully clarify how these influx peaks can capture fires from long distances only at certain time periods. I think the text from lines 180-187 is not substantial enough to really explain this disconnect. What I would recommend is to 1) carefully explain how the timing of fire peaks was determined in the paleorecords and the temporal span of each peak, 2) clearly explain how many peaks seem to correspond to a nearby fires and how many do not, and then 3) where they do or do not, explain the processes that may explain the pattern (i.e., are peaks more related to what actually burned in the fire, or the distance from the fires to the site, or both). I think articulating this carefully in this way will greatly help the manuscript.
R: Regarding data type: In Figure 4, the peak in charcoal refers to elevated values in the charcoal accumulation rate in several consecutive samples. We have not examined charcoal peaks as in Higuera et al. (2009), which implies statistical separation of charcoal peaks from charcoal background, peak evaluation (i.e., minimum count screening, signal-to-noise analyses, etc.). The main reason we have not conducted, or rather not presented peak analysis, is the generally very low charcoal counts and numerous zero samples, and at some sites, low median sedimentation rate relative to the entire record, which argues against using CHAR analysis (Higuera et al., 2009). Our preliminary peak analysis test using CHAR analysis confirmed that charcoal peaks could not be satisfactorily separated from the background at many sites. However, as the reviewer observed, the interpolated CHAR values appear to match temporally better periods of high fire activity in satellite data.
R: Regarding the correlations between the charcoal peaks and fire sources, Figure 4 and Table S1 show that most documented fires occurred at considerable distances (3–29 km) from our sampling sites, and the fire source areas at our sites overlapped. Although the reviewer's suggestions provide an elegant way to address this issue, it remains challenging to determine whether the elevated CHAR values primarily reflect fire size, fire proximity, or a combination of both. Nevertheless, in the revised version, we have expanded the potential connection between charcoal values and the source of the fire.
- In lines 184-187 – the authors mention that gramminoids have lower charcoal rendition rate per unit biomass because they may be completely consumed in fires. Would this potential preservation bias based on vegetation type partially explain why more charcoal seems to be present in the records when shrubs are more common the landscape? I think this in an important point to address in the interpretation. While it makes intuitive sense that more flammable species and higher biomass on the landscape would result in larger and/or more frequent burning (and/or create positive feedbacks), preservation biases from graminoid burning versus shrub burning could also be important to address.
R Thank you for this observation. We have slightly expanded on the effect that fuel type and amount, as well as burning temperature may have had on charcoal amount.
Modern fire data - please clarify in the methods how the area burned for the point data was estimated. Was this based on field surveys by the AICC? For the polygon data, was estimated area burned based soley on the size of the polygon, or were non-vegetated surfaces removed from the area burned estimate?
- The final comment I have is regarding the water table analyses. While I think it is awesome to have that type of information in the study (especially considering how important this could be to future fire regime changes), it was unclear to me how spatially consistent water table changes actually are in this region. For example, I am thinking about how that area has large ice wedges that can cause subsidence and intersection with the water table as they thaw, and how fires themselves may cause surface subsidence through these thaw processes. Thus…how sure are the authors that water table changes from these reconstructions actually apply to the whole area and/or that these records are appropriate to combine into a composite? I think a bit more text explaining the logic here would go a long way to improving the interpretation.
R: Our study area does not lie in an ice wedge region. In all instances, we cored from the hummock area of each peatland. No fires have been documented at the coring sites between 1969 and the present based on satellite images. Furthermore, given the very low charcoal amount (larger than 150 microns) and the absence of larger charcoal particles (mm) in the entire record, this likely indicates that our coring sites have likely not burned during the entire study period (present to a maximum of 3000 BCE). Therefore, it is unlikely that fire has caused the thaw of permafrost and surface subsidence.
Technical corrections:
- Figure 1 and Table 1: Please add the site codes to the map OR the site numbers to the table. It is not possible to easily figure out which site is which on the map.
- Table 1: I would recommend adding original citations for previously cored sites to the data table. Also, it is not clear to me if the charcoal, pollen, and macrofossil analyses are new for all sites, or only for a subset of sites. Clarification of new versus pre-existing analyses in the methods section would be very helpful.
- Figure 1: I can only see two 2007 fires in the polygon data on this map. I am assuming that the other 7 fires occurring in 1983, 1990, 2001, 2009, and 2017 fires are so small that they are obscured by the point data? In any case, I think the solution here would be to use a single color for the fire polygons and label the years of these polygons directly on the map so that the reader can see them (or some other workable solution that allows the fire polygons to be seen by the reader).
- Line 53 – “low fire patterns” is a confusing phrase. Please reword and clarify the meaning.
- Line 63 – feedback should be plural
- Figure 2 caption – parentheses bracket missing in the first line
R: We agree with the reviewer comments and, in agreement with suggestions from reviewer 1, we will provide more in the methods chapter and tables.
Kind regards
Angelica Feurdean on behalf of the other authors
Citation: https://doi.org/10.5194/egusphere-2025-2318-AC1
-
AC1: 'Reply on RC 1 and RC 2', Angelica Feurdean, 29 Jul 2025
-
RC2: 'Comment on egusphere-2025-2318', Anonymous Referee #2, 14 Jul 2025
General comments:
This manuscript contributes to our understanding of fire-regime variability on millennial to centennial timescales on the Alaskan North Slope along the Dalton Highway corridor. The authors use macroscopic charcoal counts from peatland soils to reconstruct charcoal influx associated with past fires. They also reconstruct vegetation compositional changes at local and regional scales using a combination of macrofossil and pollen identification from their sites, respectively. These data, along with previous reconstructions for water table changes from testate amoebae at a subset of sites provides information for qualitatively evaluating feedbacks between hydrology, vegetation, and landscape flammability. The authors also use spatially explicit fire datasets to constrain the source area of the macroscopic charcoal counts. Overall, this is a well-developed dataset and robust contribution that expands our understanding of the palaeoecological history of this rapidly changing region. I appreciate that the findings are generally not overstated, with the exception of the manuscript title that makes a very bold claim which may only tenuously be supported by the data (please see specific comments). Overall, I think that some revisions are needed to clarify the interpretations of the datasets and the overarching conclusions. Specifically, greater attention to chronological control and constraints, as well as a more nuanced explanation of the some of the temporal patterns, would improve the manuscript. I hope the specific comments I provide below are helpful to the authors.
Specific comments:
- I would really like to see some background information in the introduction as to the sources of charcoal in peatlands and the the pros/cons of using that setting for fire history reconstructions. This does not have to be overly long, but a bit of general background on the methods of extracting fire reconstructions from peatlands specifically would be quite useful to the reader.
- Chronological control is probably the most important part of any paleo study. I would highly recommend adding the age-depth models for all nine sites to the appendix so that readers can easily examine the fidelity of all models. A master table with all AMS and 210Pb ages per depth for all sites would also be useful to add the appendix.
- One of the major claims in this paper is that fire activity since 1880 CE was higher than the previous millennia. At first glance, the data presented in Figure 2 is compelling (and exciting!). However, it is unclear to me if the charcoal influx data for each site has been interpolated to the median or average sampling resolution for each record. The reason I am concerned about this is that without a consistent temporal resolution, trends in influx could solely be a function of changes in the sedimentation rate. Specifically, the sedimentation rate changes dramatically in all records presented in the appendix in the uppermost sediments compared to the rest of the core. Equal time sampling will be important to compare past to recent trends, and should be explained carefully in the text if the takehome is indeed that fire regimes have changed.
- Another point in regards to chronological control is line 37-38 in the abstract. Is it possible that the heterogenous fire patterns observed in the past compared to more recent sediments is a function of both 1) lower temporal resolution in the deeper parts of the peat cores and 2) the much larger age error of the deeper sediments? I think these types of issues should be discussed explicitly in the text to provide a more conservative and nuanced interpretation of the data.
- In terms of the analysis comparing the timing of recent fires and the distance of fires from the depositional sites (summarized in figure 4) --- I am not sure I understand this analysis and it would be helpful to see some clarification. From my understanding, peaks in the charcoal influx data seem to capture deposition of charcoal from long distances, and yet do not capture the larger fires (specifically the Anaktuvuk River Fire that burned ~1000 km2)? I think the disconnect here is that I do not quite understand how the timing of the fire peaks were determined, and/or what the age model error around those timings actually is. From what I can tell, peak analysis to identify which peaks are most likely to represent fire events versus redeposition processes were not done on these cores. These types of peak analyses that are commonly used in lake sediment studies (with minimum count screening, signal-to-noise analyses, etc) may not be appropriate for peat cores (see comment #1), which is fine. However, I think it is important to carefully clarify how these influx peaks can capture fires from long distances only at certain time periods. I think the text from lines 180-187 is not substantial enough to really explain this disconnect. What I would recommend is to 1) carefully explain how the timing of fire peaks was determined in the paleorecords and the temporal span of each peak, 2) clearly explain how many peaks seem to correspond to a nearby fires and how many do not, and then 3) where they do or do not, explain the processes that may explain the pattern (i.e., are peaks more related to what actually burned in the fire, or the distance from the fires to the site, or both). I think articulating this carefully in this way will greatly help the manuscript.
- In lines 184-187 – the authors mention that gramminoids have lower charcoal rendition rate per unit biomass because they may be completely consumed in fires. Would this potential preservation bias based on vegetation type partially explain why more charcoal seems to be present in the records when shrubs are more common the landscape? I think this in an important point to address in the interpretation. While it makes intuitive sense that more flammable species and higher biomass on the landscape would result in larger and/or more frequent burning (and/or create positive feedbacks), preservation biases from graminoid burning versus shrub burning could also be important to address.
- Modern fire data - please clarify in the methods how the area burned for the point data was estimated. Was this based on field surveys by the AICC? For the polygon data, was estimated area burned based soley on the size of the polygon, or were non-vegetated surfaces removed from the area burned estimate?
- The final comment I have is in regards to the water table analyses. While I think it is awesome to have that type of information in the study (especially considering how important this could be to future fire regime changes), it was unclear to me how spatially consistent water table changes actually are in this region. For example, I am thinking about how that area has large ice wedges that can cause subsidence and intersection with the water table as they thaw, and how fires themselves may cause surface subsidence through these thaw processes. Thus…how sure are the authors that water table changes from these reconstructions actually apply to the whole area and/or that these records are appropriate to combine into a composite? I think a bit more text explaining the logic here would go a long way to improving the interpretation.
Technical corrections:
- Figure 1 and Table 1: Please add the site codes to the map OR the site numbers to the table. It is not possible to easily figure out which site is which on the map.
- Table 1: I would recommend adding original citations for previously cored sites to the data table. Also, it is not clear to me if the charcoal, pollen, and macrofossil analyses are new for all sites, or only for a subset of sites. Clarification of new versus pre-existing analyses in the methods section would be very helpful.
- Figure 1: I can only see two 2007 fires in the polygon data on this map. I am assuming that the other 7 fires occurring in 1983, 1990, 2001, 2009, and 2017 fires are so small that they are obscured by the point data? In any case, I think the solution here would be to use a single color for the fire polygons and label the years of these polygons directly on the map so that the reader can see them (or some other workable solution that allows the fire polygons to be seen by the reader).
- Line 53 – “low fire patterns” is a confusing phrase. Please reword and clarify the meaning.
- Line 63 – feedback should be plural
- Figure 2 caption – parentheses bracket missing in the first line
Citation: https://doi.org/10.5194/egusphere-2025-2318-RC2 -
AC1: 'Reply on RC 1 and RC 2', Angelica Feurdean, 29 Jul 2025
29 July, 2025, Frankfurt am Main
We thank both reviewers for providing encouraging feedback and valuable comments on the manuscript, which will help enhance its accuracy and readability. We agree that some of the methods and interpretations, as well as parts of the discussions, were brief. In the revised manuscript, we expanded these parts based on these suggestions. For more details on the main comments and suggestions, see our response to specific comments below. The complete response to all comments will follow together with the revised manuscript.
Ramesh Glückler General comments:
In their manuscript “Fire activity in the northern Arctic tundra now exceeds late Holocene levels, driven by increasing dryness and shrub expansion”, the authors explore timely relationships between past wildfire dynamics, vegetation, and hydrology in the Alaskan tundra. For that they use a combination of new and previously published paleoecological data on past wildfire activity, vegetation composition, and the peatland water table, and combine this paleoecological data with more recent fire observations from remote sensing. The main result, summarized in the manuscript’s title, is based on increased reconstructed fire activity after 1880 CE, coinciding with drying and shrub encroachment. Overall, the manuscript presents a valuable perspective on past wildfire activity, especially by its approach to compile data from multiple sites and to include a reconstruction of the peatland water table. The comparison of charcoal-based fire reconstructions with more recent fire observations is intriguing and contributes to ongoing efforts of determining regional charcoal source area. The manuscript content fits well within the scope of Biogeosciences, and would be an interesting contribution to the journal for researchers across various disciplines, such as paleoecology, peatland ecology, fire ecology, hydrology, or tundra vegetation dynamics. However, in its current state the manuscript would in my opinion benefit from some further polishing and some clarifications regarding the proposed drivers behind the reconstructed fire activity. Before publication I would therefore recommend some moderate editing, including clarifications regarding the applied methods and interpretations, a more comprehensive description of the results, and some changes to the structure of the discussion to guide the reader more smoothly. My suggestions therefore do not concern the great underlying data and analysis, but rather potential improvements in the way these are presented and discussed, so the manuscript can stand at its full potential. From my perspective, a correspondingly revised version should be very much considered for publication in Biogeosciences. Please find more detailed comments below:
Specific comments: Manuscript structure and clarifications
1. From my perspective, some changes to the manuscript’s structure would improve reading flow and potentially clarify some remaining questions. First, it was not always completely clear to me which data were already published before (and exactly where), and which were newly created, and for some of the data, I was missing a clear description. I think all of this may be mostly related to the short results section, which only includes sub-sections for the fire-related data (charcoal and remote sensing). However, in the methods section, also new chronological data, data on vegetation (from pollen and plant macrofossils), and data for the water table (testate amoebae) is mentioned. I think even if previously published, since they were also processed as part of this manuscript, they should also be featured and briefly described in the results section, where they are currently missing. Expanding the manuscript in this regard for methods would be appreciated as well, since it is currently not always clear which methods were applied due to the short nature of the description (e.g., in L122 a composite of shrub data (and later water table data) is mentioned, but lacking details on how exactly it was composited, and in L126 the abbreviation “DTW” is introduced, but not clearly described). If the testate amoebae data was published before, what were the main outcomes? How would the newly composited water table curve be described? Introducing the data in a bit more detail would help the reader follow along in the discussion.R: Methods. In revising this manuscript, we have expanded the Methods section in the main text (3.1 Chronology) and the Appendices to include the chronology of all sites, not just the three unpublished sites (SG, RH, and DL). Additionally, we created separate subchapters to expand the description of testate amoebae water table DWT reconstruction (3.3 Testate amoebae reconstruction of hydrological changes) and pollen and plant macrofossils-based vegetation reconstructions (3.4 Plant macrofossils and pollen-based reconstruction of vegetation dynamics), including the creation of their respective composite records.
Results. Following the same logic as for methods, we have created separate subchapters that include brief descriptions of results from pollen and plant macrofossil-based vegetation reconstructions, as well as testate amoebae water table DWT reconstructions.
- Speaking of the discussion, I would personally not start with the methodological interpretation of charcoal source area, but rather with the most important aspects of the manuscript mentioned in the title – although this is definitely a subjective preference. However, also here some aspects should be expanded upon. For example, in L195 historical fire observations were mentioned, but not further described. Which kind of fire activity was recorded between 1880 to 1920? Does it fit to the reconstructed charcoal-based signal? Finally, instead of the last sentence of the discussion, I suggest to include an actual conclusion section that briefly summarizes the findings and provides an additional outlook. I feel that would provide a smoother end to the manuscript.
R: Thank you for the suggestions regarding the manuscript's structure. The reason for starting the Discussion with the methodological Interpretation of charcoal source areas is that readers have an overview of what charcoal records represent in terms of distance from the fires and biomass burnt. Once these aspects are established, readers will be able to understand the long-term charcoal record more effectively. For this reason, we would like to maintain the manuscript's structure, starting with the Interpretation of charcoal source areas. However, we fully agree with the reviewer's suggestion to include a separate conclusion section, in which information on the main findings from this study, along with further outlook, is presented. Regarding historical fire observations (L195), they refer to evidence coming from aerial photographs published in Miller et al. (2023). We will clarify these aspects in the revised manuscript.
Interpretation – Fire before the drought?
I generally agree with the main conclusion of relating moisture conditions, fuel types, and fire activity, as this would be expected, and it would certainly be great to see this relationship clearly in paleoecological data. However, some aspects of the suggested relationships, as seen in the actual data, may need a bit more explanation. In L209, it is stated that the driest conditions of the Late Holocene (during the past few decades) coincided with the most intense/severe fire period. However, looking at Fig. 3a it seems rather that severe fires occurred between c. 1900 to 1970 CE, whereas in Fig. 3c it seems that drier-than-average conditions only occurred since ca. 2000 CE, when the levels of biomass burned were very low. A similar discrepancy is evident in the timing of increased shrub counts in Fig. 3b when compared to the reconstructed biomass burning, even though in L225 it says that woody biomass would have promoted fires. Somehow it seems to me as though the level of biomass burning increased first, and the shrubs and drying followed – but what would then be the driver behind the fire activity? How can these temporal differences be understood in light of the proposed relationships, and does proxy taphonomy play a role as well? In case I did not misinterpret the plots, I think that these aspects should be included in more detail in the discussion.R: Following the reviewer's two suggestions, we have additionally calculated the charcoal accumulation rate (CHAR) on charcoal values interpolated at the median temporal resolution of each record to account for marked changes in the sedimentation rate in records in the uppermost sediments compared to the rest of the core. This interpolated composite CHAR record aligns temporally more closely with composite records in DTW and vegetation. However, some temporary asynchrony changes in fire-vegetation-moisture persist, and we believe that taphonomic processes connected to the type of biomass burned (see our response to reviewer 2) and the spatial resolution of different proxies. In the revised version, we will expand the discussion to provide a more precise explanation of these processes.
Technical corrections:.
R: thank you for technical and language corrections, we will carefully incorporate them in the revision of the manuscript. To most of those who asked a question and not referring to rephrasing, we responded below.
L34: Suggest to re-phrase sentences, e.g. “[…] charcoal records in combination with data on vegetation, hydrology, and satellite-derived fire observations from […]”. In the following sentence: “A regional composite of charcoal records shows […]”
L44: Maybe instead of ending a bit abruptly after this result regarding source area, a sentence briefly summarizing the importance of the study, or providing some outlook, would be fitting here?
L46: Suggest to re-phrase to e.g. “[…] increase in the number, size, and intensity of individual fires, and in the length and duration of fire seasons […]”
L52: “Sayedi” et al., 2024
L53: Suggest to re-phrase here, e.g. “[…] contributed to a deviation from the previously low wildfire activity”
L67: “peatlands”
L66, 70: No reference for “Vachula, 2020” or “Vachula et al., 2020” is listed in the reference list. Do you refer to Vachula (2021) in L70 (see references below)? Maybe just double-check that all citations are mentioned in the reference list and vice-versa.
R: We refer to Vachula RS, Sae-Lim J, Russell JM. Sedimentary charcoal proxy records of fire in Alaskan tundra ecosystems. Palaeogeography, Palaeoclimatology, Palaeoecology. 1;541:109564.
L73: “tussock sedges-moss-dwarf shrubs” seems like a bit of a clunky term, maybe there is another way to refer to this vegetative zone? Also note additional space before “dwarf”
R: We will simplify the term as tussock and shrubby tussock tundra.
L74: not sure if an “en”-dash is needed in “testate amoebae-based” – maybe double-check journal guidelines.
L85: “Annual precipitation averages 25 mm […]” – is that the average annual precipitation sum across multiple years? For that it seems quite low. Or do you refer to the average across all months of a year? In that case, I think that the first would provide better context. Maybe it is good to just clarify which average this refers to.
R: Thank you for spotting the mistake, a zero was missing! Annual precipitation averages 250 mm.
L92: I am not sure I can follow the description of how cores were retrieved with a shovel, maybe you can provide a bit more detail?
L101: Sites “Ga I and GA II” or sometimes written with a space, sometimes without (e.g., L126)
L107: Suggest to include here: “[…] intervals across cores from all sites”
L107: Which bleaching agent was used in the procedure? This should be stated in any case, but especially since previous analyses found that certain bleaching agents may dissolve charcoal particles from low-intensity fires (Constantine and Mooney, 2021), which (I suppose) are the kind of fires one may expect at these sites
R: We used sodium hypochlorite (household bleach).
L112: Sometimes “charcoal influx” is used, sometimes “charcoal accumulation rate (CHAR)”. I suggest to standardize these terms across the manuscript, as they seem to be used interchangeably.
R: Thank you, for consistence we have decided to sue charcoal accumulation rate (CHAR).
L112: The unit description here is missing some “minus” and numbers – I think it should be “particles cm-2 yr-1”. This also goes for other instances throughout the manuscript (e.g., right afterwards: “particles cm-3” and “year cm-1”, or in L146) – Also note that in some figure captions, the unit is written as “#/cm2/yr”, I’d recommend to standardize one type across the manuscript.
R: Thank you, we will uniformised both the name and units measure through the manuscript.
L121: “smoothed”?
L136: Add space: “70 km” and standardize across the manuscript
L137: “[…] the distance from the nearest […]”
L145: Add space: “mean = 0.04 […]”
L148: In this case for ranges I think that “en”-dashes should indeed be used. Maybe consult the journal guidelines for a revision (same in L165, L170)
L152: Suggest to replace “the” with “an”
L154: “macro charcoal” – double check that this term is always used in correct and equal spelling across the manuscript (suggest to use either “macro-charcoal”, “macrocharcoal”, or “macroscopic charcoal”)
R: We will standardise the term to the term macrocharcoal thought the manuscript.
L158: Suggest to start with “Our satellite […]” and use the term “buffer” instead of “radius”?
L157: The “extremely large fire” is 103.896 ha or 103,896 ha? It may be good to double-check that the use of comma/dot for decimals is always correct and according to the international norm, including in the corresponding figure caption etc. In L160, a space is missing before “2 ha”, and later a “=” is missing in “median 115 ha”.
R: It is a very large fire of over 103 kiloha.
L168: Before, no spaces were used when mentioning e.g. “Fig. 4c”. Recommend to standardize across the manuscript.
L171: “[…] while the satellite data during 2001–2017” is not a full sentence, maybe this part could be re-phrased?
L176: Delete space at beginning of the line
L180: I would have expected some citations already within this sentence, to make it clear which study is related to the 2 km source area or the wider area. In the next sentence “Vachula et al., 2020” is mentioned, which may again may need to be corrected.
L186: Missing comma in citation
L191: I am not sure data on fire intensity was presented here? As it’s a similar theme, I would also refrain from mentioning an “intense fire period” (L209) and rather use the term “severe”, to reduce potential confusion about fire regime terminology.
R: We will use fire frequency and size.
L211: “Smaller-amplitude deepening of the water table […]”?
L220: Suggest to add commas: “[…] fuel type and availability, and fire, particularly […]”
Figure 2: I am not sure if I understand this sentence in the caption: “The total charcoal influx represents the unidentified charcoal.” – could that be re-phrased and clarified? Also, note the CHAR unit being written differently here than in the main text. In L432, standardize the use of spaces and capitals when defining “Common Era” (including other instances across the manuscript). In the last sentence of the caption, it says that the smaller plots on the right side cover the last two centuries, yet their x-axes seem only to reach back to 1960 CE?
R: Apologies for the clumpy sentence and inconsistency. Total charcoal influx refers to the total amount of charcoal at each site, as opposed to the charcoal that was differentiated into morphological types.
Figure 3: In caption in L442: Is it correct to write “[…] as counts. (c) Composite record […]”? Regarding the figure itself, would there be a way to separate the inserted plots more clearly from the main plots, so the numbers don’t overlap with the background plot? For example, they could all be decreased in size a little bit and inserted on the left side of the respective main plot, where there is more space. Two of the inserted plots also currently miss x-axes. The y-axis description of plot a) is not completely clear to me – it should probably say that higher values mean more biomass burning, but from just the one arrow/symbol it is not fully clear. Maybe it would be better just to use two arrows and a description, such as in plot c)? Also, it seems that the x-axis line in plot c) has a variable thickness, whereas it is not visible at all in some of the other inserted plots.
Figure 4: In caption, I suggest to list the cut peaks as “a) 2001 (341 ha), 2007 (659 ha); b) 2007 (103.896 ha)” – note the b) 2007 value again here, is it really supposed to be 103.9 ha? In the last sentence, standardize use of spaces around “=”. In the figure itself, I am wondering if a) and b) could not be combined into a single plot, with the bars of different width in the back and CHAR in the front, maybe in a different color? Also, is it really “Cumulative charcoal influx” in a)? The compilation method by Blarquez et al. (2014) does not just sum the different records, so it’s likely not cumulative – if that was done here, it should be stated in the methods section. In plot c), there are variable distances between the legend entries, and the x-axis line is interrupted.
R: Thank you for your suggestions to correct the figures and figure captions. We will carefully incorporate your suggestions in the revised version of the manuscript. Basically, on figure four is the charcoal accumulation rate at all sites (not composite),
Table 1: “Coordinates”
Appendix C: Suggest writing either just “particles >500 µm” or “particles larger than 500 µm”.
Data availability: Ideally, the data could be uploaded during the revision so that the final DOIs can be included in the manuscript, which would make finding the data a lot easier in the future. In any case, I’d recommend to fully reference Neotoma, e.g.: “[…] deposited to the Neotoma Paleoecology Database (www.neotomadb.org; Williams et al., 2018)”
R: Thank you. We will add all datasets used in this manuscript to the SI and send this to Neotoma
Supplement S1: “Satellite-based”
Author contribution: “conceived the study”
Affiliations: Suggest to double-check spelling of institutions, order of institutional levels, and the inclusion of the full address
References mentioned in this review:
Blarquez, O., Vannière, B., Marlon, J. R., Daniau, A.-L., Power, M. J., Brewer, S., & Bartlein, P. J. (2014). paleofire: An R package to analyse sedimentary charcoal records from the Global Charcoal Database to reconstruct past biomass burning. Computers & Geosciences, 72, 255–261. https://doi.org/10.1016/j.cageo.2014.07.020
Constantine, M., & Mooney, S. (2021). Widely used charcoal analysis method in paleo studies involving NaOCl results in loss of charcoal formed below 400°C. The Holocene, 09596836211041740. https://doi.org/10.1177/09596836211041740
Vachula, R. S. (2021). A meta-analytical approach to understanding the charcoal source area problem. Palaeogeography, Palaeoclimatology, Palaeoecology, 562, 110111. https://doi.org/10.1016/j.palaeo.2020.110111
Williams, J. W., Grimm, E. C., Blois, J. L., Charles, D. F., Davis, E. B., Goring, S. J., Graham, R. W., Smith, A. J., Anderson, M., Arroyo-Cabrales, J., Ashworth, A. C., Betancourt, J. L., Bills, B. W., Booth, R. K., Buckland, P. I., Curry, B. B., Giesecke, T., Jackson, S. T., Latorre, C., … Takahara, H. (2018). The Neotoma Paleoecology Database, a multiproxy, international, community-curated data resource. Quaternary Research, 89(1), 156–177. https://doi.org/10.1017/qua.2017.105
Review 2 General comments:
This manuscript contributes to our understanding of fire-regime variability on millennial to centennial timescales on the Alaskan North Slope along the Dalton Highway corridor. The authors use macroscopic charcoal counts from peatland soils to reconstruct charcoal influx associated with past fires. They also reconstruct vegetation compositional changes at local and regional scales using a combination of macrofossil and pollen identification from their sites, respectively. These data, along with previous reconstructions for water table changes from testate amoebae at a subset of sites provides information for qualitatively evaluating feedbacks between hydrology, vegetation, and landscape flammability. The authors also use spatially explicit fire datasets to constrain the source area of the macroscopic charcoal counts. Overall, this is a well-developed dataset and robust contribution that expands our understanding of the palaeoecological history of this rapidly changing region. I appreciate that the findings are generally not overstated, with the exception of the manuscript title that makes a very bold claim which may only tenuously be supported by the data (please see specific comments). Overall, I think that some revisions are needed to clarify the interpretations of the datasets and the overarching conclusions. Specifically, greater attention to chronological control and constraints, as well as a more nuanced explanation of the some of the temporal patterns, would improve the manuscript. I hope the specific comments I provide below are helpful to the authors.
Specific comments:
- I would really like to see some background information in the introduction as to the sources of charcoal in peatlands and the the pros/cons of using that setting for fire history reconstructions. This does not have to be overly long, but a bit of general background on the methods of extracting fire reconstructions from peatlands specifically would be quite useful to the reader.
R: Thank you. In the revised version of the manuscript, we have clarified that: 1) compared to lakes, peatlands provide local-scale reconstructions of past fire regimes, and that 2) this fine-scale resolution reflects the pronounced heterogeneity in local peat moisture, vegetation composition, and structure, and the relatively small catchment area.
- Chronological control is probably the most important part of any paleo study. I would highly recommend adding the age-depth models for all nine sites to the appendix so that readers can easily examine the fidelity of all models. A master table with all AMS and 210Pb ages per depth for all sites would also be useful to add the appendix.
R: Thank you for this observation. Following the reviewer 1 suggestions, in the revised version of the paper, we will provide a table including the AMS14C and Pb210 dates, as well as figures for all sites.
- One of the major claims in this paper is that fire activity since 1880 CE was higher than the previous millennia. At first glance, the data presented in Figure 2 is compelling (and exciting!). However, it is unclear to me if the charcoal influx data for each site has been interpolated to the median or average sampling resolution for each record. The reason I am concerned about this is that without a consistent temporal resolution, trends in influx could solely be a function of changes in the sedimentation rate. Specifically, the sedimentation rate changes dramatically in all records presented in the appendix in the uppermost sediments compared to the rest of the core. Equal time sampling will be important to compare past to recent trends, and should be explained carefully in the text if the take home is indeed that fire regimes have changed.
R: Thank you for this observation regarding the effect of changing sediment accumulation rates on trends in charcoal accumulation rates (CHAR). We have subsampled all records continuously at 1 cm resolution. In the original version of the paper, the charcoal accumulation rates have not been interpolated to median or average sampling resolution before the construction of the composite CHAR record (Fig 3), nor the CHAR plots at individual sites (Fig. 2). Indeed, the sedimentation rate increased markedly in the upper part of the record compared to the rest of the record. We have recalculated the CHAR i.e, interpolated the charcoal data to a median resolution for each record. We also used interpolated charcoal accumulation rates to construct the composite CHAR record. This interpolation appears to have slightly altered the pattern composite CHAR compared to the untransformed data. However, it preserves the increasing trend in recent times, implying that the key takeaway message — that recent fire activity is the highest compared to the late Holocene level — remains valid. The recent increase in composite CHAR values, based on interpolated values, appears to align more closely with those of plant macros and DTW. We are incorporating the new results in the revised version of the manuscript.
- Another point in regard to chronological control is line 37-38 in the abstract. Is it possible that the heterogenous fire patterns observed in the past compared to more recent sediments is a function of both 1) lower temporal resolution in the deeper parts of the peat cores and 2) the much larger age error of the deeper sediments? I think these types of issues should be discussed explicitly in the text to provide a more conservative and nuanced interpretation of the data.
R: Even though the temporal trends in CHAR are slightly shifted compared to the original, untransformed record, this interpolation does not affect the original observed pattern; that fire patterns were more heterogeneous among sites in the past compared to more recent times.
- In terms of the analysis comparing the timing of recent fires and the distance of fires from the depositional sites (summarized in figure 4) --- I am not sure I understand this analysis and it would be helpful to see some clarification. From my understanding, peaks in the charcoal influx data seem to capture deposition of charcoal from long distances, and yet do not capture the larger fires (specifically the Anaktuvuk River Fire that burned ~1000 km2)? I think the disconnect here is that I do not quite understand how the timing of the fire peaks were determined, and/or what the age model error around those timings actually is. From what I can tell, peak analysis to identify which peaks are most likely to represent fire events versus redeposition processes were not done on these cores. These types of peak analyses that are commonly used in lake sediment studies (with minimum count screening, signal-to-noise analyses, etc) may not be appropriate for peat cores (see comment #1), which is fine. However, I think it is important to carefully clarify how these influx peaks can capture fires from long distances only at certain time periods. I think the text from lines 180-187 is not substantial enough to really explain this disconnect. What I would recommend is to 1) carefully explain how the timing of fire peaks was determined in the paleorecords and the temporal span of each peak, 2) clearly explain how many peaks seem to correspond to a nearby fires and how many do not, and then 3) where they do or do not, explain the processes that may explain the pattern (i.e., are peaks more related to what actually burned in the fire, or the distance from the fires to the site, or both). I think articulating this carefully in this way will greatly help the manuscript.
R: Regarding data type: In Figure 4, the peak in charcoal refers to elevated values in the charcoal accumulation rate in several consecutive samples. We have not examined charcoal peaks as in Higuera et al. (2009), which implies statistical separation of charcoal peaks from charcoal background, peak evaluation (i.e., minimum count screening, signal-to-noise analyses, etc.). The main reason we have not conducted, or rather not presented peak analysis, is the generally very low charcoal counts and numerous zero samples, and at some sites, low median sedimentation rate relative to the entire record, which argues against using CHAR analysis (Higuera et al., 2009). Our preliminary peak analysis test using CHAR analysis confirmed that charcoal peaks could not be satisfactorily separated from the background at many sites. However, as the reviewer observed, the interpolated CHAR values appear to match temporally better periods of high fire activity in satellite data.
R: Regarding the correlations between the charcoal peaks and fire sources, Figure 4 and Table S1 show that most documented fires occurred at considerable distances (3–29 km) from our sampling sites, and the fire source areas at our sites overlapped. Although the reviewer's suggestions provide an elegant way to address this issue, it remains challenging to determine whether the elevated CHAR values primarily reflect fire size, fire proximity, or a combination of both. Nevertheless, in the revised version, we have expanded the potential connection between charcoal values and the source of the fire.
- In lines 184-187 – the authors mention that gramminoids have lower charcoal rendition rate per unit biomass because they may be completely consumed in fires. Would this potential preservation bias based on vegetation type partially explain why more charcoal seems to be present in the records when shrubs are more common the landscape? I think this in an important point to address in the interpretation. While it makes intuitive sense that more flammable species and higher biomass on the landscape would result in larger and/or more frequent burning (and/or create positive feedbacks), preservation biases from graminoid burning versus shrub burning could also be important to address.
R Thank you for this observation. We have slightly expanded on the effect that fuel type and amount, as well as burning temperature may have had on charcoal amount.
Modern fire data - please clarify in the methods how the area burned for the point data was estimated. Was this based on field surveys by the AICC? For the polygon data, was estimated area burned based soley on the size of the polygon, or were non-vegetated surfaces removed from the area burned estimate?
- The final comment I have is regarding the water table analyses. While I think it is awesome to have that type of information in the study (especially considering how important this could be to future fire regime changes), it was unclear to me how spatially consistent water table changes actually are in this region. For example, I am thinking about how that area has large ice wedges that can cause subsidence and intersection with the water table as they thaw, and how fires themselves may cause surface subsidence through these thaw processes. Thus…how sure are the authors that water table changes from these reconstructions actually apply to the whole area and/or that these records are appropriate to combine into a composite? I think a bit more text explaining the logic here would go a long way to improving the interpretation.
R: Our study area does not lie in an ice wedge region. In all instances, we cored from the hummock area of each peatland. No fires have been documented at the coring sites between 1969 and the present based on satellite images. Furthermore, given the very low charcoal amount (larger than 150 microns) and the absence of larger charcoal particles (mm) in the entire record, this likely indicates that our coring sites have likely not burned during the entire study period (present to a maximum of 3000 BCE). Therefore, it is unlikely that fire has caused the thaw of permafrost and surface subsidence.
Technical corrections:
- Figure 1 and Table 1: Please add the site codes to the map OR the site numbers to the table. It is not possible to easily figure out which site is which on the map.
- Table 1: I would recommend adding original citations for previously cored sites to the data table. Also, it is not clear to me if the charcoal, pollen, and macrofossil analyses are new for all sites, or only for a subset of sites. Clarification of new versus pre-existing analyses in the methods section would be very helpful.
- Figure 1: I can only see two 2007 fires in the polygon data on this map. I am assuming that the other 7 fires occurring in 1983, 1990, 2001, 2009, and 2017 fires are so small that they are obscured by the point data? In any case, I think the solution here would be to use a single color for the fire polygons and label the years of these polygons directly on the map so that the reader can see them (or some other workable solution that allows the fire polygons to be seen by the reader).
- Line 53 – “low fire patterns” is a confusing phrase. Please reword and clarify the meaning.
- Line 63 – feedback should be plural
- Figure 2 caption – parentheses bracket missing in the first line
R: We agree with the reviewer comments and, in agreement with suggestions from reviewer 1, we will provide more in the methods chapter and tables.
Kind regards
Angelica Feurdean on behalf of the other authors
Citation: https://doi.org/10.5194/egusphere-2025-2318-AC1
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 424 | 42 | 15 | 481 | 17 | 10 | 24 |
- HTML: 424
- PDF: 42
- XML: 15
- Total: 481
- Supplement: 17
- BibTeX: 10
- EndNote: 24
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1