the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Black carbon concentrations and modeled smoke deposition fluxes to the bare ice dark zone of the Greenland Ice Sheet
Peng Xian
Joshua Schwarz
Abstract. Ice-albedo feedbacks in the ablation region of the Greenland Ice Sheet (GrIS) are difficult to constrain and model due in part to our limited understanding of the seasonal evolution of the bare-ice region. To help fill observational gaps, 13 surface samples were collected on the GrIS across the 2014 summer melt season from patches of snow that were visibly light, medium, and dark colored. These samples were analyzed for their refractory black carbon (rBC) concentrations and size distributions with a Single Particle Soot Photometer coupled to a characterized nebulizer. We present a size distribution of rBC in fresh snow on the GrIS, as well as from surface hoar in the bare ice dark zone of the GrIS. The size distributions from the surface hoar samples appear unimodal, and were overall smaller than the fresh snow sample, with a peak around 0.3 µm. The fresh snow sample contained very large rBC particles that had a pronounced bimodality in peak size distributions, with peaks around 0.2 µm and 2 µm. rBC concentrations ranged from a minimum of 3 µg-rBC/L-H2O in light-colored patches at the beginning and end of the melt season, to a maximum of 32 µg-rBC/L-H2O in a dark patch in early August. On average, rBC concentrations were higher (20 µg-rBC/L-H2O ± 10 µg-rBC/L-H2O) in patches that were visibly dark compared to medium patches (7 µg-rBC/L-H2O ± 2 µg-rBC/L-H2O) and light patches (4 µg-rBC/L-H2O ± 1 µg-rBC/L-H2O), suggesting BC aggregation contributed to snow aging on the GrIS, and vice versa. Additionally, concentrations peaked in light and dark patches in early August, which is likely due to smoke transport from wildfires in Northern Canada and Alaska as supported by the Navy Aerosol Analysis and Prediction System (NAAPS) reanalysis model. According to model output, 26 mg/m3 of biomass burning derived smoke was deposited between April 1st and August 30th, of which 85 % came from wet deposition and 67 % was deposited during our sample collection timeframe. The increase in rBC concentration and size distributions immediately after modelled smoke deposition fluxes suggest biomass burning smoke is a source of BC to the dark zone of the GRIS. Thus, role of BC in the seasonal evolution of the ice-albedo feedback should continue to be investigated in the bare-ice zone of the GrIS.
Alia Lauren Khan et al.
Status: closed
-
RC1: 'Comment on tc-2022-258', Anonymous Referee #1, 10 Mar 2023
Comment on tc-2022-258 - Black carbon concentrations and modeled smoke deposition fluxes to the bare ice dark zone of the Greenland Ice Sheet
This manuscript analyses 13 surface snow samples collected in the Greenland Ice Sheet for refractory black carbon (rBC), measuring their concentrations and size distributions on a Single Particle Soot Photometer (SP2) + a customized but extensively tested nebulizer. Concentrations of rBC in the samples were compared with the Navy Aerosol Analysis and Prediction System (NAAPS) reanalysis model to establish rBC dry and wet deposition fluxes, and were linked to biomass burning smoke from wildfires in Northern Canada and Alaska. The sampling occurred in the 2014 summer melt season, from three different types of snow patches, visually classified as light, medium and dark colored, which presented increasing rBC concentrations, respectively. The aim of this work was to provide observational data to models that analyze ice-albedo feedbacks in the ablation region of the Greenland Ice Sheet. The authors’ comparison of rBC measurements in-situ with NAAPS model showed that global aerosol models may be overestimating BC deposition, and recommend further investigation of this issue.
General comments:
This manuscript presents relevant research for the scope of this journal, addressing the role of black carbon (here as rBC) in the surface snow of a region of Greenland with restricted data. Considering that BC effects in ice and snow are still highly uncertain, field measurements are important and necessary to provide data for models involving this particle.
The title clearly reflect the contents of the paper, and the abstract provide a concise and complete summary of the research carried out. The methods are well detailed and their usage is backed by previous works cited in the manuscript. The language is adequate and clear, as well as the manuscript structure. The authors do not extrapolate their conclusions more than their results go, and acknowledge their dataset is not big enough to observe patterns in the rBC size distribution on their samples. There are interesting comparisons of the ground observations and the NAAPS model, showing the model to be off for one specific snow sample – and thus justifying the importance of evaluating the model with a larger sample size of rBC ground observations across the Arctic – but also validating the model results with rBC measurements in dark patches of snow, as those show an increase in concentrations just after specific smoke events. This work also presents more evidence on post-deposition aggregation of rBC particles (seen in the increase of the modal peak of rBC particle size in light patches over the duration of the season).
My understanding is that this research has merit: it presents new data, is scientifically sound and, based on results, raises an important question about models overestimating BC deposition in Greenland snow. It is the first work presenting rBC size distributions in snow and surface hoar of the bare ice zone of the Greenland Ice Sheet. Thus, I recommend publication after minor revision.
I have only one specific comment and very few minor corrections. I hope my review was useful to this journal and to the authors, and would be glad to review this manuscript a second time should the editors decide for it. Also hope to see the questions raised in this manuscript to be further investigated in future works.
Cheers,
The reviewer.
Specific comments:
Ln 108-122: Is it possible to add the volume of samples in this section, somewhere?
Data availability: would the rBC data be made available with the manuscript?
Minor corrections:
Line 63: “… and was determined…” - shouldn’t it be “were”?
Figure 1: The source of the images is mentioned in duplicity (both in the bottom of the image and in the caption). In my opinion this is unnecessary, and the authors could remove one of them.
Lines 317-318: “… especially during episodically…” - seems odd to me, please check and if wrong, rephrase.
Citation: https://doi.org/10.5194/tc-2022-258-RC1 -
AC1: 'Reply on RC1', Alia Khan, 12 Mar 2023
The authors appreciate the constructive comments from Reviewer 1. We have revised the manuscript based on the suggestions from Reviewer 1. Our responses are below to the specific comments and minor corrections.
Specific comments:
Ln 108-122: Is it possible to add the volume of samples in this section, somewhere?
Author Response: The following sentence has been amended to, ”…were collected in 150 mL pre-cleaned and combusted amber glass bottles.”
Data availability: would the rBC data be made available with the manuscript?
Author Response: Yes, the rBC concentrations are listed in Table 1. A Data Availability statement has been added for clarity, “The rBC and NAAPS modeled deposition data are included in Table 1.”
Minor corrections:
Line 63: “… and was determined…” - shouldn’t it be “were”?
Author Response: This sentence has been updated for clarity, “The mean concentration of the samples collected was 2.6 ng/g and the mean peak was 15 ng/g. Based on these results, it was determined that EC/OC do not influence the snow albedo in the NW sector of the GrIS dry zone (Polashenski et al., 2015a).”
Figure 1: The source of the images is mentioned in duplicity (both in the bottom of the image and in the caption). In my opinion this is unnecessary, and the authors could remove one of them.
Author Response: The authors agree and the redundant information has been removed from the Figure caption.
Lines 317-318: “… especially during episodically…” - seems odd to me, please check and if wrong, rephrase.
Author Response: Thanks, ‘during’ has been deleted from the sentence.
Citation: https://doi.org/10.5194/tc-2022-258-AC1 -
RC2: 'Reply on AC1', Anonymous Referee #1, 13 Mar 2023
Dear authors,
Thank you for the prompt reply and clarifications. I am glad with the manuscript as it is now.
Cheers,
The reviewer
Citation: https://doi.org/10.5194/tc-2022-258-RC2 -
AC2: 'Reply on RC2', Alia Khan, 18 Apr 2023
The authors appreciate the constructive comments from Reviewer 2. We have revised the manuscript based on the suggestions from Reviewer 2. Our responses are below to the specific comments and minor corrections.
Specific Reviewer Comments:
The manuscript mentions that the study site is near the S6 automatic weather station. I think providing data from the weather station over the course of the sampling period would help to provide context for the measurements. In particular, the paper discusses that flushing or melt might have caused the reduction in rBC values before the August 11 measurements. Would it be possible to determine if major rain events occurred or if particularly high temperatures or large deviations in surface energy balance might have resulted in this flushing/melting between August 2 and August 11? I do not know what data are available from this AWS, so perhaps this is not possible.
Author Response: This is a great idea but unfortunately it doesn’t look like precipitation is measured here: https://www.projects.science.uu.nl/iceclimate/aws/files_oper/oper_01309
Work by Lewis et al. (2021) in Geophysical Research Letters also presents black carbon concentration measurements from the percolation zone of the Greenland Ice Sheet made using an SP2. I believe all of their measurements were further from the margin of the ice sheet and resulted in BC concentrations less than 1.5 ng/g, but it still seems worth including in the review of BC measurements in Greenland.
Author Response: The following sentence has been added to the introduction, “Similarly, rBC concentrations from the percolation zone of the GrIS have been shown to be relatively low, less than 1.5 ng/g (Lewis et al., 2021).”
It may be worth discussing what the magnitude of the impact of these levels of rBC concentration would be on the overall broadband albedo. I understand that assumptions would need to be made about grain size, other LAIs, solar conditions, etc., but using the SNICAR model (as I note is done in the Khan et al. 2017 JGR: Atmospheres paper) could provide a range of the potential effect given these BC concentrations. Because it is clear that there are a number of other LAIs (particularly in the surface hoar samples), this could be more difficult if there are too many assumptions that must be made, so I would leave it to the authors’ discretion if they think this is worthwhile.
Author Response: We appreciate the suggestion and actually had a similar thought during the manuscript preparation. Prior to submission, we chatted with the developer of the SNICAR model, Mark Flanner. We decided that there would be too many assumptions needed to be made to explore albedo modeling and radiative forcing with this dataset, however, we absolutely agree that this is of utmost interest as a next step in this work.
Line 184: It is assumed that the rBC ratio to total mass biomass burning smoke is 7%. Please add in a couple of sentences indicating what factors affect this percentage and given that this is a median value, perhaps how large of a range might be expected.
Author Response: The following sentences have been amended/added to this section: “Estimating atmospheric properties related to biomass burning is highly complex and is influenced by wide variety of factors such as the type of fuel, combustion temperature, and atmospheric conditions. Also, the chemical, optical and physical properties of biomass burning aerosols can change during atmospheric transport and dispersion. The mass ratio of rBC to total mass in biomass burning smoke particles is estimated to be 5–10% black carbon in the NAAPS-RA model based on field studies (see a summary in Reid et al., 2005) and here we chose 7% as a median value.”
Line 217: It may be useful to show each size distribution for the individual samples, perhaps in a supplement if not in the main paper, in order to demonstrate that there is no pattern.
Author Response: A third panel has been added to Figure 2, “…and C) the size distribution of each surface hoar sample categorized as light, medium and dark.”
Figures 3 and 5: Consider marking the sample days with a vertical line so it is easy for the reader to see how these model results line up with the sampling.
Author Response: The five sampling dates are marked in orange on Figure 3.
Figure 5: Consider separating out the total precipitation into a separate figure stacked below because it can be difficult to parse the overlapping lines.
Author Response: Total precipitation has been removed and added as a second panel in Figure 5.
Line 260: Where it states the modeled smoke deposition is 300 µg/L, what time period is that over?
Author Response: This sentence has been amended to, “The modeled smoke deposition flux is 3000 µg/m2/day or 300 µg/L over 24 hours.”
Lines 262-264: If the measurement is already in units of µg-rBC/L-H2O, doesn’t that already account for the amount of snow water equivalent? Perhaps I am misunderstanding, but I’m not sure how multiplying by 10% would make sense, so please clarify. I do think that the assumption that 7% of smoke is BC would be a source of uncertainty worth discussing more (see comment above about line 184).
Author Response: Thanks for catching this. The authors agree and the following sentence has been deleted, “If we assume the snow water equivalent is 10%, then the rBC-snow concentration (i.e., the concentration of rBC in the fresh wet snow being deposited) would be 2.1 µg-rBC/L-H2O.”
Technical Corrections:
Line 37: “Thus, role” – “Thus, the role”
Author Response: This update has been made.
Line 60: “higher concentrations” – “higher concentration”
Author Response: This update has been made.
Line 169: “Moderate Imaging” – “Moderate Resolution Imaging”
Author Response: This update has been made.
Line 210: “sizes” – “size”
Author Response: This update has been made.
Line 244: “based NAAPS-RA” – “based on NAAPS-RA”
Author Response: This update has been made.
Line 245: should it be AOT instead of AOD?
Author Response: This update has been made.
Line 260: should that be 10 L/m^2?
Author Response: This update has been made.
Line 266: “NAAPs” – “NAAPS”
Author Response: This update has been made.
Line 283: add in DOY so that it is easier for reader to reference the figure
Author Response: This update has been made.
Figure 5: I think the right y-axis should be red instead of blue, if I understand correctly.
Author Response: Total precipitation has been removed and added as a second panel in Figure 5, per the suggestion above.
Line 318: remove word “during” or otherwise rephrase
Author Response: This sentence has been amended to, “However, NAAPS model results also indicate the increase is likely related to accumulation of episodically deposited wildfire-derived smoke. For example, the smoke event in early August, which brought smoke from the western Northern Hemisphere.”
Lines 379 + 382: references missing article titles
Author Response: This update has been made.
References:
Lewis, G., Osterberg, E., Hawley, R., Marshall, H. P., Meehan, T., Graeter, K., et al. (2021). Atmospheric blocking drives recent albedo change across the western Greenland ice sheet percolation zone. Geophysical Research Letters, 48, e2021GL092814. https://doi.org/10.1029/2021GL092814
Author Response: The reference has been added to the manuscript.
Citation: https://doi.org/10.5194/tc-2022-258-AC2 -
AC4: 'Reply on AC2', Alia Khan, 18 Apr 2023
Apologies, this text was meant to be in response to Reviewer 2's comments below.
Citation: https://doi.org/10.5194/tc-2022-258-AC4
-
AC4: 'Reply on AC2', Alia Khan, 18 Apr 2023
-
AC2: 'Reply on RC2', Alia Khan, 18 Apr 2023
-
RC2: 'Reply on AC1', Anonymous Referee #1, 13 Mar 2023
-
AC1: 'Reply on RC1', Alia Khan, 12 Mar 2023
-
RC3: 'Comment on tc-2022-258', Anonymous Referee #2, 27 Mar 2023
General Comments:
The paper presents novel results on the concentration and size distribution of refractory black carbon (rBC) in the dark snow zone of the Greenland Ice Sheet. Using the NAAPS model, the paper compares the in-situ measurements with modeled smoke deposition, likely sourced from wildfires in North America. Using this model, the paper also covers potential differences in wet and dry deposition of black carbon, identifying wet deposition as a large source of black carbon, at least in this particular year. This dataset is novel—the first to present rBC measurements of this kind in this region of Greenland. It offers some new insights into rBC concentrations and the aggregation of rBC with other light absorbing impurities (LAIs). This is significant because these are the first measurements of their kind, and they provide initial insight into the behavior of black carbon in the dark snow zone of the Greenland Ice Sheet. The potential albedo feedback in this region is critically important. The purpose of the work is clear, and while the number of samples is limited, the authors acknowledge this point and are careful not to draw conclusions beyond what their data indicate. The methods are sound and well described. Overall, the paper is clearly written and well-structured. I have a few specific comments that I think might improve clarity and completeness, but generally, I think this paper is an important contribution and will be of interest to the readers of The Cryosphere!
Specific Comments:
The manuscript mentions that the study site is near the S6 automatic weather station. I think providing data from the weather station over the course of the sampling period would help to provide context for the measurements. In particular, the paper discusses that flushing or melt might have caused the reduction in rBC values before the August 11 measurements. Would it be possible to determine if major rain events occurred or if particularly high temperatures or large deviations in surface energy balance might have resulted in this flushing/melting between August 2 and August 11? I do not know what data are available from this AWS, so perhaps this is not possible.
Work by Lewis et al. (2021) in Geophysical Research Letters also presents black carbon concentration measurements from the percolation zone of the Greenland Ice Sheet made using an SP2. I believe all of their measurements were further from the margin of the ice sheet and resulted in BC concentrations less than 1.5 ng/g, but it still seems worth including in the review of BC measurements in Greenland.
It may be worth discussing what the magnitude of the impact of these levels of rBC concentration would be on the overall broadband albedo. I understand that assumptions would need to be made about grain size, other LAIs, solar conditions, etc., but using the SNICAR model (as I note is done in the Khan et al. 2017 JGR: Atmospheres paper) could provide a range of the potential effect given these BC concentrations. Because it is clear that there are a number of other LAIs (particularly in the surface hoar samples), this could be more difficult if there are too many assumptions that must be made, so I would leave it to the authors’ discretion if they think this is worthwhile.
Line 184: It is assumed that the rBC ratio to total mass biomass burning smoke is 7%. Please add in a couple of sentences indicating what factors affect this percentage and given that this is a median value, perhaps how large of a range might be expected.
Line 217: It may be useful to show each size distribution for the individual samples, perhaps in a supplement if not in the main paper, in order to demonstrate that there is no pattern.
Figures 3 and 5: Consider marking the sample days with a vertical line so it is easy for the reader to see how these model results line up with the sampling.
Figure 5: Consider separating out the total precipitation into a separate figure stacked below because it can be difficult to parse the overlapping lines.
Line 260: Where it states the modeled smoke deposition is 300 µg/L, what time period is that over?
Lines 262-264: If the measurement is already in units of µg-rBC/L-H2O, doesn’t that already account for the amount of snow water equivalent? Perhaps I am misunderstanding, but I’m not sure how multiplying by 10% would make sense, so please clarify. I do think that the assumption that 7% of smoke is BC would be a source of uncertainty worth discussing more (see comment above about line 184).
Technical Corrections:
Line 37: “Thus, role” – “Thus, the role”
Line 60: “higher concentrations” – “higher concentration”
Line 169: “Moderate Imaging” – “Moderate Resolution Imaging”
Line 210: “sizes” – “size”
Line 244: “based NAAPS-RA” – “based on NAAPS-RA”
Line 245: should it be AOT instead of AOD?
Line 260: should that be 10 L/m^2?
Line 266: “NAAPs” – “NAAPS”
Line 283: add in DOY so that it is easier for reader to reference the figure
Figure 5: I think the right y-axis should be red instead of blue, if I understand correctly.
Line 318: remove word “during” or otherwise rephrase
Lines 379 + 382: references missing article titles
References:
Lewis, G., Osterberg, E., Hawley, R., Marshall, H. P., Meehan, T., Graeter, K., et al. (2021). Atmospheric blocking drives recent albedo change across the western Greenland ice sheet percolation zone. Geophysical Research Letters, 48, e2021GL092814. https://doi.org/10.1029/2021GL092814
Citation: https://doi.org/10.5194/tc-2022-258-RC3 -
AC3: 'Reply on RC3', Alia Khan, 18 Apr 2023
The authors appreciate the constructive comments from Reviewer 2. We have revised the manuscript based on the suggestions from Reviewer 2. Our responses are below to the specific comments and minor corrections.
Specific Comments:
The manuscript mentions that the study site is near the S6 automatic weather station. I think providing data from the weather station over the course of the sampling period would help to provide context for the measurements. In particular, the paper discusses that flushing or melt might have caused the reduction in rBC values before the August 11 measurements. Would it be possible to determine if major rain events occurred or if particularly high temperatures or large deviations in surface energy balance might have resulted in this flushing/melting between August 2 and August 11? I do not know what data are available from this AWS, so perhaps this is not possible.
Author Response: This is a great idea but unfortunately it doesn’t look like precipitation is measured here: https://www.projects.science.uu.nl/iceclimate/aws/files_oper/oper_01309
Work by Lewis et al. (2021) in Geophysical Research Letters also presents black carbon concentration measurements from the percolation zone of the Greenland Ice Sheet made using an SP2. I believe all of their measurements were further from the margin of the ice sheet and resulted in BC concentrations less than 1.5 ng/g, but it still seems worth including in the review of BC measurements in Greenland.
Author Response: The following sentence has been added to the introduction, “Similarly, rBC concentrations from the percolation zone of the GrIS have been shown to be relatively low, less than 1.5 ng/g (Lewis et al., 2021).”
It may be worth discussing what the magnitude of the impact of these levels of rBC concentration would be on the overall broadband albedo. I understand that assumptions would need to be made about grain size, other LAIs, solar conditions, etc., but using the SNICAR model (as I note is done in the Khan et al. 2017 JGR: Atmospheres paper) could provide a range of the potential effect given these BC concentrations. Because it is clear that there are a number of other LAIs (particularly in the surface hoar samples), this could be more difficult if there are too many assumptions that must be made, so I would leave it to the authors’ discretion if they think this is worthwhile.
Author Response: We appreciate the suggestion and actually had a similar thought during the manuscript preparation. Prior to submission, we chatted with the developer of the SNICAR model, Mark Flanner. We decided that there would be too many assumptions needed to be made to explore albedo modeling and radiative forcing with this dataset, however, we absolutely agree that this is of utmost interest as a next step in this work.
Line 184: It is assumed that the rBC ratio to total mass biomass burning smoke is 7%. Please add in a couple of sentences indicating what factors affect this percentage and given that this is a median value, perhaps how large of a range might be expected.
Author Response: The following sentences have been amended/added to this section: “Estimating atmospheric properties related to biomass burning is highly complex and is influenced by wide variety of factors such as the type of fuel, combustion temperature, and atmospheric conditions. Also, the chemical, optical and physical properties of biomass burning aerosols can change during atmospheric transport and dispersion. The mass ratio of rBC to total mass in biomass burning smoke particles is estimated to be 5–10% black carbon in the NAAPS-RA model based on field studies (see a summary in Reid et al., 2005) and here we chose 7% as a median value.”
Line 217: It may be useful to show each size distribution for the individual samples, perhaps in a supplement if not in the main paper, in order to demonstrate that there is no pattern.
Author Response: A third panel has been added to Figure 2, “…and C) the size distribution of each surface hoar sample categorized as light, medium and dark.”
Figures 3 and 5: Consider marking the sample days with a vertical line so it is easy for the reader to see how these model results line up with the sampling.
Author Response: The five sampling dates are marked in orange on Figure 3.
Figure 5: Consider separating out the total precipitation into a separate figure stacked below because it can be difficult to parse the overlapping lines.
Author Response: Total precipitation has been removed and added as a second panel in Figure 5.
Line 260: Where it states the modeled smoke deposition is 300 µg/L, what time period is that over?
Author Response: This sentence has been amended to, “The modeled smoke deposition flux is 3000 µg/m2/day or 300 µg/L over 24 hours.”
Lines 262-264: If the measurement is already in units of µg-rBC/L-H2O, doesn’t that already account for the amount of snow water equivalent? Perhaps I am misunderstanding, but I’m not sure how multiplying by 10% would make sense, so please clarify. I do think that the assumption that 7% of smoke is BC would be a source of uncertainty worth discussing more (see comment above about line 184).
Author Response: Thanks for catching this. The authors agree and the following sentence has been deleted, “If we assume the snow water equivalent is 10%, then the rBC-snow concentration (i.e., the concentration of rBC in the fresh wet snow being deposited) would be 2.1 µg-rBC/L-H2O.”
Technical Corrections:
Line 37: “Thus, role” – “Thus, the role”
Author Response: This update has been made.
Line 60: “higher concentrations” – “higher concentration”
Author Response: This update has been made.
Line 169: “Moderate Imaging” – “Moderate Resolution Imaging”
Author Response: This update has been made.
Line 210: “sizes” – “size”
Author Response: This update has been made.
Line 244: “based NAAPS-RA” – “based on NAAPS-RA”
Author Response: This update has been made.
Line 245: should it be AOT instead of AOD?
Author Response: This update has been made.
Line 260: should that be 10 L/m^2?
Author Response: This update has been made.
Line 266: “NAAPs” – “NAAPS”
Author Response: This update has been made.
Line 283: add in DOY so that it is easier for reader to reference the figure
Author Response: This update has been made.
Figure 5: I think the right y-axis should be red instead of blue, if I understand correctly.
Author Response: Total precipitation has been removed and added as a second panel in Figure 5, per the suggestion above.
Line 318: remove word “during” or otherwise rephrase
Author Response: This sentence has been amended to, “However, NAAPS model results also indicate the increase is likely related to accumulation of episodically deposited wildfire-derived smoke. For example, the smoke event in early August, which brought smoke from the western Northern Hemisphere.”
Lines 379 + 382: references missing article titles
Author Response: This update has been made.
References:
Lewis, G., Osterberg, E., Hawley, R., Marshall, H. P., Meehan, T., Graeter, K., et al. (2021). Atmospheric blocking drives recent albedo change across the western Greenland ice sheet percolation zone. Geophysical Research Letters, 48, e2021GL092814. https://doi.org/10.1029/2021GL092814
Author Response: The reference has been added to the manuscript.
Citation: https://doi.org/10.5194/tc-2022-258-AC3
-
AC3: 'Reply on RC3', Alia Khan, 18 Apr 2023
Status: closed
-
RC1: 'Comment on tc-2022-258', Anonymous Referee #1, 10 Mar 2023
Comment on tc-2022-258 - Black carbon concentrations and modeled smoke deposition fluxes to the bare ice dark zone of the Greenland Ice Sheet
This manuscript analyses 13 surface snow samples collected in the Greenland Ice Sheet for refractory black carbon (rBC), measuring their concentrations and size distributions on a Single Particle Soot Photometer (SP2) + a customized but extensively tested nebulizer. Concentrations of rBC in the samples were compared with the Navy Aerosol Analysis and Prediction System (NAAPS) reanalysis model to establish rBC dry and wet deposition fluxes, and were linked to biomass burning smoke from wildfires in Northern Canada and Alaska. The sampling occurred in the 2014 summer melt season, from three different types of snow patches, visually classified as light, medium and dark colored, which presented increasing rBC concentrations, respectively. The aim of this work was to provide observational data to models that analyze ice-albedo feedbacks in the ablation region of the Greenland Ice Sheet. The authors’ comparison of rBC measurements in-situ with NAAPS model showed that global aerosol models may be overestimating BC deposition, and recommend further investigation of this issue.
General comments:
This manuscript presents relevant research for the scope of this journal, addressing the role of black carbon (here as rBC) in the surface snow of a region of Greenland with restricted data. Considering that BC effects in ice and snow are still highly uncertain, field measurements are important and necessary to provide data for models involving this particle.
The title clearly reflect the contents of the paper, and the abstract provide a concise and complete summary of the research carried out. The methods are well detailed and their usage is backed by previous works cited in the manuscript. The language is adequate and clear, as well as the manuscript structure. The authors do not extrapolate their conclusions more than their results go, and acknowledge their dataset is not big enough to observe patterns in the rBC size distribution on their samples. There are interesting comparisons of the ground observations and the NAAPS model, showing the model to be off for one specific snow sample – and thus justifying the importance of evaluating the model with a larger sample size of rBC ground observations across the Arctic – but also validating the model results with rBC measurements in dark patches of snow, as those show an increase in concentrations just after specific smoke events. This work also presents more evidence on post-deposition aggregation of rBC particles (seen in the increase of the modal peak of rBC particle size in light patches over the duration of the season).
My understanding is that this research has merit: it presents new data, is scientifically sound and, based on results, raises an important question about models overestimating BC deposition in Greenland snow. It is the first work presenting rBC size distributions in snow and surface hoar of the bare ice zone of the Greenland Ice Sheet. Thus, I recommend publication after minor revision.
I have only one specific comment and very few minor corrections. I hope my review was useful to this journal and to the authors, and would be glad to review this manuscript a second time should the editors decide for it. Also hope to see the questions raised in this manuscript to be further investigated in future works.
Cheers,
The reviewer.
Specific comments:
Ln 108-122: Is it possible to add the volume of samples in this section, somewhere?
Data availability: would the rBC data be made available with the manuscript?
Minor corrections:
Line 63: “… and was determined…” - shouldn’t it be “were”?
Figure 1: The source of the images is mentioned in duplicity (both in the bottom of the image and in the caption). In my opinion this is unnecessary, and the authors could remove one of them.
Lines 317-318: “… especially during episodically…” - seems odd to me, please check and if wrong, rephrase.
Citation: https://doi.org/10.5194/tc-2022-258-RC1 -
AC1: 'Reply on RC1', Alia Khan, 12 Mar 2023
The authors appreciate the constructive comments from Reviewer 1. We have revised the manuscript based on the suggestions from Reviewer 1. Our responses are below to the specific comments and minor corrections.
Specific comments:
Ln 108-122: Is it possible to add the volume of samples in this section, somewhere?
Author Response: The following sentence has been amended to, ”…were collected in 150 mL pre-cleaned and combusted amber glass bottles.”
Data availability: would the rBC data be made available with the manuscript?
Author Response: Yes, the rBC concentrations are listed in Table 1. A Data Availability statement has been added for clarity, “The rBC and NAAPS modeled deposition data are included in Table 1.”
Minor corrections:
Line 63: “… and was determined…” - shouldn’t it be “were”?
Author Response: This sentence has been updated for clarity, “The mean concentration of the samples collected was 2.6 ng/g and the mean peak was 15 ng/g. Based on these results, it was determined that EC/OC do not influence the snow albedo in the NW sector of the GrIS dry zone (Polashenski et al., 2015a).”
Figure 1: The source of the images is mentioned in duplicity (both in the bottom of the image and in the caption). In my opinion this is unnecessary, and the authors could remove one of them.
Author Response: The authors agree and the redundant information has been removed from the Figure caption.
Lines 317-318: “… especially during episodically…” - seems odd to me, please check and if wrong, rephrase.
Author Response: Thanks, ‘during’ has been deleted from the sentence.
Citation: https://doi.org/10.5194/tc-2022-258-AC1 -
RC2: 'Reply on AC1', Anonymous Referee #1, 13 Mar 2023
Dear authors,
Thank you for the prompt reply and clarifications. I am glad with the manuscript as it is now.
Cheers,
The reviewer
Citation: https://doi.org/10.5194/tc-2022-258-RC2 -
AC2: 'Reply on RC2', Alia Khan, 18 Apr 2023
The authors appreciate the constructive comments from Reviewer 2. We have revised the manuscript based on the suggestions from Reviewer 2. Our responses are below to the specific comments and minor corrections.
Specific Reviewer Comments:
The manuscript mentions that the study site is near the S6 automatic weather station. I think providing data from the weather station over the course of the sampling period would help to provide context for the measurements. In particular, the paper discusses that flushing or melt might have caused the reduction in rBC values before the August 11 measurements. Would it be possible to determine if major rain events occurred or if particularly high temperatures or large deviations in surface energy balance might have resulted in this flushing/melting between August 2 and August 11? I do not know what data are available from this AWS, so perhaps this is not possible.
Author Response: This is a great idea but unfortunately it doesn’t look like precipitation is measured here: https://www.projects.science.uu.nl/iceclimate/aws/files_oper/oper_01309
Work by Lewis et al. (2021) in Geophysical Research Letters also presents black carbon concentration measurements from the percolation zone of the Greenland Ice Sheet made using an SP2. I believe all of their measurements were further from the margin of the ice sheet and resulted in BC concentrations less than 1.5 ng/g, but it still seems worth including in the review of BC measurements in Greenland.
Author Response: The following sentence has been added to the introduction, “Similarly, rBC concentrations from the percolation zone of the GrIS have been shown to be relatively low, less than 1.5 ng/g (Lewis et al., 2021).”
It may be worth discussing what the magnitude of the impact of these levels of rBC concentration would be on the overall broadband albedo. I understand that assumptions would need to be made about grain size, other LAIs, solar conditions, etc., but using the SNICAR model (as I note is done in the Khan et al. 2017 JGR: Atmospheres paper) could provide a range of the potential effect given these BC concentrations. Because it is clear that there are a number of other LAIs (particularly in the surface hoar samples), this could be more difficult if there are too many assumptions that must be made, so I would leave it to the authors’ discretion if they think this is worthwhile.
Author Response: We appreciate the suggestion and actually had a similar thought during the manuscript preparation. Prior to submission, we chatted with the developer of the SNICAR model, Mark Flanner. We decided that there would be too many assumptions needed to be made to explore albedo modeling and radiative forcing with this dataset, however, we absolutely agree that this is of utmost interest as a next step in this work.
Line 184: It is assumed that the rBC ratio to total mass biomass burning smoke is 7%. Please add in a couple of sentences indicating what factors affect this percentage and given that this is a median value, perhaps how large of a range might be expected.
Author Response: The following sentences have been amended/added to this section: “Estimating atmospheric properties related to biomass burning is highly complex and is influenced by wide variety of factors such as the type of fuel, combustion temperature, and atmospheric conditions. Also, the chemical, optical and physical properties of biomass burning aerosols can change during atmospheric transport and dispersion. The mass ratio of rBC to total mass in biomass burning smoke particles is estimated to be 5–10% black carbon in the NAAPS-RA model based on field studies (see a summary in Reid et al., 2005) and here we chose 7% as a median value.”
Line 217: It may be useful to show each size distribution for the individual samples, perhaps in a supplement if not in the main paper, in order to demonstrate that there is no pattern.
Author Response: A third panel has been added to Figure 2, “…and C) the size distribution of each surface hoar sample categorized as light, medium and dark.”
Figures 3 and 5: Consider marking the sample days with a vertical line so it is easy for the reader to see how these model results line up with the sampling.
Author Response: The five sampling dates are marked in orange on Figure 3.
Figure 5: Consider separating out the total precipitation into a separate figure stacked below because it can be difficult to parse the overlapping lines.
Author Response: Total precipitation has been removed and added as a second panel in Figure 5.
Line 260: Where it states the modeled smoke deposition is 300 µg/L, what time period is that over?
Author Response: This sentence has been amended to, “The modeled smoke deposition flux is 3000 µg/m2/day or 300 µg/L over 24 hours.”
Lines 262-264: If the measurement is already in units of µg-rBC/L-H2O, doesn’t that already account for the amount of snow water equivalent? Perhaps I am misunderstanding, but I’m not sure how multiplying by 10% would make sense, so please clarify. I do think that the assumption that 7% of smoke is BC would be a source of uncertainty worth discussing more (see comment above about line 184).
Author Response: Thanks for catching this. The authors agree and the following sentence has been deleted, “If we assume the snow water equivalent is 10%, then the rBC-snow concentration (i.e., the concentration of rBC in the fresh wet snow being deposited) would be 2.1 µg-rBC/L-H2O.”
Technical Corrections:
Line 37: “Thus, role” – “Thus, the role”
Author Response: This update has been made.
Line 60: “higher concentrations” – “higher concentration”
Author Response: This update has been made.
Line 169: “Moderate Imaging” – “Moderate Resolution Imaging”
Author Response: This update has been made.
Line 210: “sizes” – “size”
Author Response: This update has been made.
Line 244: “based NAAPS-RA” – “based on NAAPS-RA”
Author Response: This update has been made.
Line 245: should it be AOT instead of AOD?
Author Response: This update has been made.
Line 260: should that be 10 L/m^2?
Author Response: This update has been made.
Line 266: “NAAPs” – “NAAPS”
Author Response: This update has been made.
Line 283: add in DOY so that it is easier for reader to reference the figure
Author Response: This update has been made.
Figure 5: I think the right y-axis should be red instead of blue, if I understand correctly.
Author Response: Total precipitation has been removed and added as a second panel in Figure 5, per the suggestion above.
Line 318: remove word “during” or otherwise rephrase
Author Response: This sentence has been amended to, “However, NAAPS model results also indicate the increase is likely related to accumulation of episodically deposited wildfire-derived smoke. For example, the smoke event in early August, which brought smoke from the western Northern Hemisphere.”
Lines 379 + 382: references missing article titles
Author Response: This update has been made.
References:
Lewis, G., Osterberg, E., Hawley, R., Marshall, H. P., Meehan, T., Graeter, K., et al. (2021). Atmospheric blocking drives recent albedo change across the western Greenland ice sheet percolation zone. Geophysical Research Letters, 48, e2021GL092814. https://doi.org/10.1029/2021GL092814
Author Response: The reference has been added to the manuscript.
Citation: https://doi.org/10.5194/tc-2022-258-AC2 -
AC4: 'Reply on AC2', Alia Khan, 18 Apr 2023
Apologies, this text was meant to be in response to Reviewer 2's comments below.
Citation: https://doi.org/10.5194/tc-2022-258-AC4
-
AC4: 'Reply on AC2', Alia Khan, 18 Apr 2023
-
AC2: 'Reply on RC2', Alia Khan, 18 Apr 2023
-
RC2: 'Reply on AC1', Anonymous Referee #1, 13 Mar 2023
-
AC1: 'Reply on RC1', Alia Khan, 12 Mar 2023
-
RC3: 'Comment on tc-2022-258', Anonymous Referee #2, 27 Mar 2023
General Comments:
The paper presents novel results on the concentration and size distribution of refractory black carbon (rBC) in the dark snow zone of the Greenland Ice Sheet. Using the NAAPS model, the paper compares the in-situ measurements with modeled smoke deposition, likely sourced from wildfires in North America. Using this model, the paper also covers potential differences in wet and dry deposition of black carbon, identifying wet deposition as a large source of black carbon, at least in this particular year. This dataset is novel—the first to present rBC measurements of this kind in this region of Greenland. It offers some new insights into rBC concentrations and the aggregation of rBC with other light absorbing impurities (LAIs). This is significant because these are the first measurements of their kind, and they provide initial insight into the behavior of black carbon in the dark snow zone of the Greenland Ice Sheet. The potential albedo feedback in this region is critically important. The purpose of the work is clear, and while the number of samples is limited, the authors acknowledge this point and are careful not to draw conclusions beyond what their data indicate. The methods are sound and well described. Overall, the paper is clearly written and well-structured. I have a few specific comments that I think might improve clarity and completeness, but generally, I think this paper is an important contribution and will be of interest to the readers of The Cryosphere!
Specific Comments:
The manuscript mentions that the study site is near the S6 automatic weather station. I think providing data from the weather station over the course of the sampling period would help to provide context for the measurements. In particular, the paper discusses that flushing or melt might have caused the reduction in rBC values before the August 11 measurements. Would it be possible to determine if major rain events occurred or if particularly high temperatures or large deviations in surface energy balance might have resulted in this flushing/melting between August 2 and August 11? I do not know what data are available from this AWS, so perhaps this is not possible.
Work by Lewis et al. (2021) in Geophysical Research Letters also presents black carbon concentration measurements from the percolation zone of the Greenland Ice Sheet made using an SP2. I believe all of their measurements were further from the margin of the ice sheet and resulted in BC concentrations less than 1.5 ng/g, but it still seems worth including in the review of BC measurements in Greenland.
It may be worth discussing what the magnitude of the impact of these levels of rBC concentration would be on the overall broadband albedo. I understand that assumptions would need to be made about grain size, other LAIs, solar conditions, etc., but using the SNICAR model (as I note is done in the Khan et al. 2017 JGR: Atmospheres paper) could provide a range of the potential effect given these BC concentrations. Because it is clear that there are a number of other LAIs (particularly in the surface hoar samples), this could be more difficult if there are too many assumptions that must be made, so I would leave it to the authors’ discretion if they think this is worthwhile.
Line 184: It is assumed that the rBC ratio to total mass biomass burning smoke is 7%. Please add in a couple of sentences indicating what factors affect this percentage and given that this is a median value, perhaps how large of a range might be expected.
Line 217: It may be useful to show each size distribution for the individual samples, perhaps in a supplement if not in the main paper, in order to demonstrate that there is no pattern.
Figures 3 and 5: Consider marking the sample days with a vertical line so it is easy for the reader to see how these model results line up with the sampling.
Figure 5: Consider separating out the total precipitation into a separate figure stacked below because it can be difficult to parse the overlapping lines.
Line 260: Where it states the modeled smoke deposition is 300 µg/L, what time period is that over?
Lines 262-264: If the measurement is already in units of µg-rBC/L-H2O, doesn’t that already account for the amount of snow water equivalent? Perhaps I am misunderstanding, but I’m not sure how multiplying by 10% would make sense, so please clarify. I do think that the assumption that 7% of smoke is BC would be a source of uncertainty worth discussing more (see comment above about line 184).
Technical Corrections:
Line 37: “Thus, role” – “Thus, the role”
Line 60: “higher concentrations” – “higher concentration”
Line 169: “Moderate Imaging” – “Moderate Resolution Imaging”
Line 210: “sizes” – “size”
Line 244: “based NAAPS-RA” – “based on NAAPS-RA”
Line 245: should it be AOT instead of AOD?
Line 260: should that be 10 L/m^2?
Line 266: “NAAPs” – “NAAPS”
Line 283: add in DOY so that it is easier for reader to reference the figure
Figure 5: I think the right y-axis should be red instead of blue, if I understand correctly.
Line 318: remove word “during” or otherwise rephrase
Lines 379 + 382: references missing article titles
References:
Lewis, G., Osterberg, E., Hawley, R., Marshall, H. P., Meehan, T., Graeter, K., et al. (2021). Atmospheric blocking drives recent albedo change across the western Greenland ice sheet percolation zone. Geophysical Research Letters, 48, e2021GL092814. https://doi.org/10.1029/2021GL092814
Citation: https://doi.org/10.5194/tc-2022-258-RC3 -
AC3: 'Reply on RC3', Alia Khan, 18 Apr 2023
The authors appreciate the constructive comments from Reviewer 2. We have revised the manuscript based on the suggestions from Reviewer 2. Our responses are below to the specific comments and minor corrections.
Specific Comments:
The manuscript mentions that the study site is near the S6 automatic weather station. I think providing data from the weather station over the course of the sampling period would help to provide context for the measurements. In particular, the paper discusses that flushing or melt might have caused the reduction in rBC values before the August 11 measurements. Would it be possible to determine if major rain events occurred or if particularly high temperatures or large deviations in surface energy balance might have resulted in this flushing/melting between August 2 and August 11? I do not know what data are available from this AWS, so perhaps this is not possible.
Author Response: This is a great idea but unfortunately it doesn’t look like precipitation is measured here: https://www.projects.science.uu.nl/iceclimate/aws/files_oper/oper_01309
Work by Lewis et al. (2021) in Geophysical Research Letters also presents black carbon concentration measurements from the percolation zone of the Greenland Ice Sheet made using an SP2. I believe all of their measurements were further from the margin of the ice sheet and resulted in BC concentrations less than 1.5 ng/g, but it still seems worth including in the review of BC measurements in Greenland.
Author Response: The following sentence has been added to the introduction, “Similarly, rBC concentrations from the percolation zone of the GrIS have been shown to be relatively low, less than 1.5 ng/g (Lewis et al., 2021).”
It may be worth discussing what the magnitude of the impact of these levels of rBC concentration would be on the overall broadband albedo. I understand that assumptions would need to be made about grain size, other LAIs, solar conditions, etc., but using the SNICAR model (as I note is done in the Khan et al. 2017 JGR: Atmospheres paper) could provide a range of the potential effect given these BC concentrations. Because it is clear that there are a number of other LAIs (particularly in the surface hoar samples), this could be more difficult if there are too many assumptions that must be made, so I would leave it to the authors’ discretion if they think this is worthwhile.
Author Response: We appreciate the suggestion and actually had a similar thought during the manuscript preparation. Prior to submission, we chatted with the developer of the SNICAR model, Mark Flanner. We decided that there would be too many assumptions needed to be made to explore albedo modeling and radiative forcing with this dataset, however, we absolutely agree that this is of utmost interest as a next step in this work.
Line 184: It is assumed that the rBC ratio to total mass biomass burning smoke is 7%. Please add in a couple of sentences indicating what factors affect this percentage and given that this is a median value, perhaps how large of a range might be expected.
Author Response: The following sentences have been amended/added to this section: “Estimating atmospheric properties related to biomass burning is highly complex and is influenced by wide variety of factors such as the type of fuel, combustion temperature, and atmospheric conditions. Also, the chemical, optical and physical properties of biomass burning aerosols can change during atmospheric transport and dispersion. The mass ratio of rBC to total mass in biomass burning smoke particles is estimated to be 5–10% black carbon in the NAAPS-RA model based on field studies (see a summary in Reid et al., 2005) and here we chose 7% as a median value.”
Line 217: It may be useful to show each size distribution for the individual samples, perhaps in a supplement if not in the main paper, in order to demonstrate that there is no pattern.
Author Response: A third panel has been added to Figure 2, “…and C) the size distribution of each surface hoar sample categorized as light, medium and dark.”
Figures 3 and 5: Consider marking the sample days with a vertical line so it is easy for the reader to see how these model results line up with the sampling.
Author Response: The five sampling dates are marked in orange on Figure 3.
Figure 5: Consider separating out the total precipitation into a separate figure stacked below because it can be difficult to parse the overlapping lines.
Author Response: Total precipitation has been removed and added as a second panel in Figure 5.
Line 260: Where it states the modeled smoke deposition is 300 µg/L, what time period is that over?
Author Response: This sentence has been amended to, “The modeled smoke deposition flux is 3000 µg/m2/day or 300 µg/L over 24 hours.”
Lines 262-264: If the measurement is already in units of µg-rBC/L-H2O, doesn’t that already account for the amount of snow water equivalent? Perhaps I am misunderstanding, but I’m not sure how multiplying by 10% would make sense, so please clarify. I do think that the assumption that 7% of smoke is BC would be a source of uncertainty worth discussing more (see comment above about line 184).
Author Response: Thanks for catching this. The authors agree and the following sentence has been deleted, “If we assume the snow water equivalent is 10%, then the rBC-snow concentration (i.e., the concentration of rBC in the fresh wet snow being deposited) would be 2.1 µg-rBC/L-H2O.”
Technical Corrections:
Line 37: “Thus, role” – “Thus, the role”
Author Response: This update has been made.
Line 60: “higher concentrations” – “higher concentration”
Author Response: This update has been made.
Line 169: “Moderate Imaging” – “Moderate Resolution Imaging”
Author Response: This update has been made.
Line 210: “sizes” – “size”
Author Response: This update has been made.
Line 244: “based NAAPS-RA” – “based on NAAPS-RA”
Author Response: This update has been made.
Line 245: should it be AOT instead of AOD?
Author Response: This update has been made.
Line 260: should that be 10 L/m^2?
Author Response: This update has been made.
Line 266: “NAAPs” – “NAAPS”
Author Response: This update has been made.
Line 283: add in DOY so that it is easier for reader to reference the figure
Author Response: This update has been made.
Figure 5: I think the right y-axis should be red instead of blue, if I understand correctly.
Author Response: Total precipitation has been removed and added as a second panel in Figure 5, per the suggestion above.
Line 318: remove word “during” or otherwise rephrase
Author Response: This sentence has been amended to, “However, NAAPS model results also indicate the increase is likely related to accumulation of episodically deposited wildfire-derived smoke. For example, the smoke event in early August, which brought smoke from the western Northern Hemisphere.”
Lines 379 + 382: references missing article titles
Author Response: This update has been made.
References:
Lewis, G., Osterberg, E., Hawley, R., Marshall, H. P., Meehan, T., Graeter, K., et al. (2021). Atmospheric blocking drives recent albedo change across the western Greenland ice sheet percolation zone. Geophysical Research Letters, 48, e2021GL092814. https://doi.org/10.1029/2021GL092814
Author Response: The reference has been added to the manuscript.
Citation: https://doi.org/10.5194/tc-2022-258-AC3
-
AC3: 'Reply on RC3', Alia Khan, 18 Apr 2023
Alia Lauren Khan et al.
Alia Lauren Khan et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 265 | 97 | 20 | 382 | 3 | 1 |
- HTML: 265
- PDF: 97
- XML: 20
- Total: 382
- BibTeX: 3
- EndNote: 1
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1