the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Dec 2023
Long-term observations of black carbon and carbon monoxide in the Poker Flat Research Range, central Alaska, with a focus on forest wildfire emissions
Abstract. Forest wildfires in interior Alaska represent an important black carbon (BC) source for the Arctic and sub-Arctic. However, BC observations in interior Alaska have not been sufficient to constrain the range of existing emissions. Here, we show our observations of BC mass concentrations and carbon monoxide (CO) mixing ratios in the Poker Flat Research Range (65.12° N, 147.43° W), located in central Alaska, since April 2016. The medians of the hourly BC mass concentration and CO mixing ratio throughout the observation period were 13 ng m-3 and 124.7 ppb, respectively. Significant peaks in the BC mass concentration and CO mixing ratio were observed at the same time, indicating influences from common sources. These BC peaks coincided with peaks at other comparative sites in Alaska, indicating large BC emissions in interior Alaska. Source estimation by FLEXPART-WRF confirmed a contribution of forest wildfires in Alaska when high BC mass concentrations were observed. For these cases, we found a positive correlation (r = 0.44) between the observed BC/∆CO ratio and fire radiative power (FRP) observed in Alaska and Canada. This finding indicates that the BC and CO emission ratio is controlled by the intensity and time progress of forest wildfires and suggests the BC emission factor or/and inventory could be potentially improved by FRP. We recommend that FRP be integrated into future bottom-up emission inventories to achieve a better understanding of the dynamics of pollutants from frequently occurred forest wildfires under the rapidly changing climate in the Arctic.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1535 KB)
-
Supplement
(838 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1535 KB) - Metadata XML
-
Supplement
(838 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
CC1: 'Comment on egusphere-2023-2764', Jian Liu, 11 Jan 2024
The authors have conducted a lot of analysis work in this paper, demonstrating a new way to help people understand the emissions over Arctic. There are some issues : 1) There are several models mentioned in this paper, as we know, models are impacted from the model itself, parameters, and input, such as emissions, so how did you find a balance between the models and their uncertainties, any assumptions included? If yes, more details should be given in the supplement. 2) FRP has widely been utilized in the top-down emission inventories, but for bottom-up emission researchers, the burned area and emission factors are used to estimate the biomass burning emission. Of course, the FRP should have some correlations with wildfires and these are acknowledged as two different technique routes for emission researchers. So the question here is how to help these researchers use the FRP or help each other, the authors should give accurate details in this part. If not, the conclusion will be weak. The authors should also make it clear that this might be useful to the bottom-up emissions in the results section. 3) Following Question 2), based on this recently published paper (https://doi.org/10.5194/acp-24-367-2024), why did the authors choose GFED only as the emission inventory in the FLEXPART-WRF, how about using an FRP-based emission inventory and checking if there was some relation between BC/△CO and the differences for two different routes' emission inventories? 4) In Fig. S2, there are some very high values from ground-based observations but not shown by NOAA hourly-average observations, does that mean the authors has a different data curation method from NOAA, if yes, why? 5) Some minor revisions should be corrected, e.g. Line 81, "section 3.2" should be "Section 3.2" and so on. Lines 405, and 416, the references should include the DOI link, and other similar issues in the reference part.
Citation: https://doi.org/10.5194/egusphere-2023-2764-CC1 -
AC1: 'Reply on CC1', Takeshi Kinase, 29 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2764/egusphere-2023-2764-AC1-supplement.pdf
-
CC2: 'Reply on AC1', Jian Liu, 23 Apr 2024
I appreciate your replies, which help me understand deeply your work. Wish you a success in the following submission.
Citation: https://doi.org/10.5194/egusphere-2023-2764-CC2
-
CC2: 'Reply on AC1', Jian Liu, 23 Apr 2024
-
AC1: 'Reply on CC1', Takeshi Kinase, 29 Mar 2024
-
RC1: 'Comment on egusphere-2023-2764', Anonymous Referee #1, 01 May 2024
The manuscript by Kinase et al. (Long-term observations of black carbon and carbon monoxide in the Poker Flat Research Range, central Alaska, with a focus on forest wildfire emissions) reports ~ 4 years of continuous monitoring data of black carbon (BC) and carbon monoxide (CO). The observation data were simulated by the FLEXPART-WRF model. The data were analyzed by focusing on high-concentration events that were likely induced by biomass burning. The authors suggest that the emission ratio of BC may depend on fire radiation power (FRP). The topic is within the scope of the journal. The organization and writing quality of the manuscript can be improved. I suggest all the authors of the manuscript carefully read through it again to improve both the general organization and expressions of individual contents. I have the following comments.
Major comment
I believe that Figure 6 shows the most important message of the manuscript. Although there might be a relationship between BC/delta_CO and FRP, the data are highly scattered (r = 0.44). At the current moment, it is very difficult for me to evaluate if such a trend exists. Comparison with some other data may help better understand the data. For instance, BC/delta_CO data from laboratory experiments (smoldering, flaming) can be compared with the present dataset. If the existing dataset from the laboratory supports the idea, the conclusion can be strengthened.
Other comments
Line 17
The observation period needs to be clearly stated, including the end of the observation.
Line 18
It is better to show some statistics (e.g., 10th and 90th tile values), rather than only showing the median.
Introduction
There are too many usages of 'significant.' Reserve the usage of such an adjective only for really meaningful cases (e.g., demonstrating statistical significance).
Line 88
The reviewer is not familiar with the observation site. Based on the description in section 2.1, it seems that the observation site is located in a forest. Can the influence of forest canopy on particle deposition be ignored when the sampling height is 5.5 m above the ground level?
Line 89
Are there any reasons why the authors used the PM1.0 cyclone? Most BC data are collected using PM2.5 inlets. The usage of PM1.0 cyclone makes comparison with literature data to be difficult.
Line 113-115
How well are these parameters constrained by previous studies? It would be helpful to have some references so that readers can find the original papers for these numbers.
Line 130
What kind of back trajectory model was employed? Are back trajectory calculations for ground surface trustable for the duration of 20 days?
Line 166
Concentrations of most aerosol species are different from one place to another place. The description is not very informative. It would be better to describe 1) how the concentration level of the present data is compared with previous studies, and 2) what kind of geographical differences might have induced the difference.
Line 172
The comparison of the CO data should belong to the method section.
Line 186
Figure S3 is not well organized. It may be better to separate the panel.
Line 202
Why was the observation suitable for obtaining the data that were influenced by wildfires?
Line 254
Are there any typical criteria to separate smoldering and flaming by FRP?
L290
Back trajectories only suggest the origin of air masses. It does not indicate anything about the occurrence of wildfires.
L298
I am not sure if the correlation can be considered as being robust.
L303
The selection of the window is somewhat arbitrary. How does the original data (without classification by the window) look like? Why do the authors think that such a window is needed? These points will need to be clearly described.
Citation: https://doi.org/10.5194/egusphere-2023-2764-RC1 - AC2: 'Reply on RC1', Takeshi Kinase, 25 Aug 2024
-
RC2: 'Comment on egusphere-2023-2764', Anonymous Referee #2, 22 Jul 2024
Kinase et al., 2023 "Long-term observations of black carbon and carbon monoxide in the Poker Flat Research Range, central Alaska, with a focus on forest wildfire emissions" presents long-term observations of BC mass concentrations and CO mixing ratios in the Poker Flat Research Range, which is located in Central Alaska. The authors use the FLEXPART-WRF model to estimate contributions from black carbon source regions and sectors and the HYSPLIT model to trace airmasses that originate from forest wildfires. This study shows a good correlation between the BC/ΔCO ratio and the fire radiative power (FRP) and concludes that the FRP should be used to improve emission inventories. This work provides very useful datasets and novel ideas. The methods and analysis support well the conclusion of this study. The text is well-written although in some parts the structure needs to be carefully revised.
I have some minor comments for the authors to consider, presented below:
Abstract: Although the measurements of this study were taken in Alaska, the high BC mass concentrations were originated from wildfires in both Alaska and Western Canada (section 3.5). Therefore, I would suggest discussing in the abstract about wildfires at high latitudes rather than just in interior Alaska. I would also suggest to mention already in the abstract the whole period of sampling at the PFRR.
Section 2.2: Some additional information about COSMOS is needed here. At least it should be added that this is a filter-based absorption technique.
Lines 163-164: Unclear sentence; please, rephrase.
Lines 201-202: Why are the black carbon emissions better captured in central Alaska? Is Alaska the only source region of the measured black carbon? A discussion on the topography of the sites and meteorological conditions is needed to support this statement.
Lines 219-223: Figure 2 shows also that emissions in Canada contribute to BC concentrations at the PFRR during the warm season. How important is this contribution? I would suggest discussing this in this section since you will link later the high BC concentrations with the forest wildfires in Canada.
Lines 251-256: I would suggest moving this text somewhere in the next section because it is not relevant to section 3.4.
Line 259: Can you support the statement here with the back trajectory analysis?
Line 261: East Asia is a separate category in Fig. 5. The text should be corrected.
Figure 4b. Typo in the label (x-axis).
Figure 5: I would suggest the x-axis to be the same as in Figure 2 for easier comparison.
Lines 291-292: Unclear sentence; please, rephrase.
Line 311: There is a typo.
Citation: https://doi.org/10.5194/egusphere-2023-2764-RC2 - AC3: 'Reply on RC2', Takeshi Kinase, 25 Aug 2024
Interactive discussion
Status: closed
-
CC1: 'Comment on egusphere-2023-2764', Jian Liu, 11 Jan 2024
The authors have conducted a lot of analysis work in this paper, demonstrating a new way to help people understand the emissions over Arctic. There are some issues : 1) There are several models mentioned in this paper, as we know, models are impacted from the model itself, parameters, and input, such as emissions, so how did you find a balance between the models and their uncertainties, any assumptions included? If yes, more details should be given in the supplement. 2) FRP has widely been utilized in the top-down emission inventories, but for bottom-up emission researchers, the burned area and emission factors are used to estimate the biomass burning emission. Of course, the FRP should have some correlations with wildfires and these are acknowledged as two different technique routes for emission researchers. So the question here is how to help these researchers use the FRP or help each other, the authors should give accurate details in this part. If not, the conclusion will be weak. The authors should also make it clear that this might be useful to the bottom-up emissions in the results section. 3) Following Question 2), based on this recently published paper (https://doi.org/10.5194/acp-24-367-2024), why did the authors choose GFED only as the emission inventory in the FLEXPART-WRF, how about using an FRP-based emission inventory and checking if there was some relation between BC/△CO and the differences for two different routes' emission inventories? 4) In Fig. S2, there are some very high values from ground-based observations but not shown by NOAA hourly-average observations, does that mean the authors has a different data curation method from NOAA, if yes, why? 5) Some minor revisions should be corrected, e.g. Line 81, "section 3.2" should be "Section 3.2" and so on. Lines 405, and 416, the references should include the DOI link, and other similar issues in the reference part.
Citation: https://doi.org/10.5194/egusphere-2023-2764-CC1 -
AC1: 'Reply on CC1', Takeshi Kinase, 29 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2764/egusphere-2023-2764-AC1-supplement.pdf
-
CC2: 'Reply on AC1', Jian Liu, 23 Apr 2024
I appreciate your replies, which help me understand deeply your work. Wish you a success in the following submission.
Citation: https://doi.org/10.5194/egusphere-2023-2764-CC2
-
CC2: 'Reply on AC1', Jian Liu, 23 Apr 2024
-
AC1: 'Reply on CC1', Takeshi Kinase, 29 Mar 2024
-
RC1: 'Comment on egusphere-2023-2764', Anonymous Referee #1, 01 May 2024
The manuscript by Kinase et al. (Long-term observations of black carbon and carbon monoxide in the Poker Flat Research Range, central Alaska, with a focus on forest wildfire emissions) reports ~ 4 years of continuous monitoring data of black carbon (BC) and carbon monoxide (CO). The observation data were simulated by the FLEXPART-WRF model. The data were analyzed by focusing on high-concentration events that were likely induced by biomass burning. The authors suggest that the emission ratio of BC may depend on fire radiation power (FRP). The topic is within the scope of the journal. The organization and writing quality of the manuscript can be improved. I suggest all the authors of the manuscript carefully read through it again to improve both the general organization and expressions of individual contents. I have the following comments.
Major comment
I believe that Figure 6 shows the most important message of the manuscript. Although there might be a relationship between BC/delta_CO and FRP, the data are highly scattered (r = 0.44). At the current moment, it is very difficult for me to evaluate if such a trend exists. Comparison with some other data may help better understand the data. For instance, BC/delta_CO data from laboratory experiments (smoldering, flaming) can be compared with the present dataset. If the existing dataset from the laboratory supports the idea, the conclusion can be strengthened.
Other comments
Line 17
The observation period needs to be clearly stated, including the end of the observation.
Line 18
It is better to show some statistics (e.g., 10th and 90th tile values), rather than only showing the median.
Introduction
There are too many usages of 'significant.' Reserve the usage of such an adjective only for really meaningful cases (e.g., demonstrating statistical significance).
Line 88
The reviewer is not familiar with the observation site. Based on the description in section 2.1, it seems that the observation site is located in a forest. Can the influence of forest canopy on particle deposition be ignored when the sampling height is 5.5 m above the ground level?
Line 89
Are there any reasons why the authors used the PM1.0 cyclone? Most BC data are collected using PM2.5 inlets. The usage of PM1.0 cyclone makes comparison with literature data to be difficult.
Line 113-115
How well are these parameters constrained by previous studies? It would be helpful to have some references so that readers can find the original papers for these numbers.
Line 130
What kind of back trajectory model was employed? Are back trajectory calculations for ground surface trustable for the duration of 20 days?
Line 166
Concentrations of most aerosol species are different from one place to another place. The description is not very informative. It would be better to describe 1) how the concentration level of the present data is compared with previous studies, and 2) what kind of geographical differences might have induced the difference.
Line 172
The comparison of the CO data should belong to the method section.
Line 186
Figure S3 is not well organized. It may be better to separate the panel.
Line 202
Why was the observation suitable for obtaining the data that were influenced by wildfires?
Line 254
Are there any typical criteria to separate smoldering and flaming by FRP?
L290
Back trajectories only suggest the origin of air masses. It does not indicate anything about the occurrence of wildfires.
L298
I am not sure if the correlation can be considered as being robust.
L303
The selection of the window is somewhat arbitrary. How does the original data (without classification by the window) look like? Why do the authors think that such a window is needed? These points will need to be clearly described.
Citation: https://doi.org/10.5194/egusphere-2023-2764-RC1 - AC2: 'Reply on RC1', Takeshi Kinase, 25 Aug 2024
-
RC2: 'Comment on egusphere-2023-2764', Anonymous Referee #2, 22 Jul 2024
Kinase et al., 2023 "Long-term observations of black carbon and carbon monoxide in the Poker Flat Research Range, central Alaska, with a focus on forest wildfire emissions" presents long-term observations of BC mass concentrations and CO mixing ratios in the Poker Flat Research Range, which is located in Central Alaska. The authors use the FLEXPART-WRF model to estimate contributions from black carbon source regions and sectors and the HYSPLIT model to trace airmasses that originate from forest wildfires. This study shows a good correlation between the BC/ΔCO ratio and the fire radiative power (FRP) and concludes that the FRP should be used to improve emission inventories. This work provides very useful datasets and novel ideas. The methods and analysis support well the conclusion of this study. The text is well-written although in some parts the structure needs to be carefully revised.
I have some minor comments for the authors to consider, presented below:
Abstract: Although the measurements of this study were taken in Alaska, the high BC mass concentrations were originated from wildfires in both Alaska and Western Canada (section 3.5). Therefore, I would suggest discussing in the abstract about wildfires at high latitudes rather than just in interior Alaska. I would also suggest to mention already in the abstract the whole period of sampling at the PFRR.
Section 2.2: Some additional information about COSMOS is needed here. At least it should be added that this is a filter-based absorption technique.
Lines 163-164: Unclear sentence; please, rephrase.
Lines 201-202: Why are the black carbon emissions better captured in central Alaska? Is Alaska the only source region of the measured black carbon? A discussion on the topography of the sites and meteorological conditions is needed to support this statement.
Lines 219-223: Figure 2 shows also that emissions in Canada contribute to BC concentrations at the PFRR during the warm season. How important is this contribution? I would suggest discussing this in this section since you will link later the high BC concentrations with the forest wildfires in Canada.
Lines 251-256: I would suggest moving this text somewhere in the next section because it is not relevant to section 3.4.
Line 259: Can you support the statement here with the back trajectory analysis?
Line 261: East Asia is a separate category in Fig. 5. The text should be corrected.
Figure 4b. Typo in the label (x-axis).
Figure 5: I would suggest the x-axis to be the same as in Figure 2 for easier comparison.
Lines 291-292: Unclear sentence; please, rephrase.
Line 311: There is a typo.
Citation: https://doi.org/10.5194/egusphere-2023-2764-RC2 - AC3: 'Reply on RC2', Takeshi Kinase, 25 Aug 2024
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 555 | 175 | 38 | 768 | 63 | 29 | 25 |
- HTML: 555
- PDF: 175
- XML: 38
- Total: 768
- Supplement: 63
- BibTeX: 29
- EndNote: 25
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Fumikazu Taketani
Masayuki Takigawa
Chunmao Zhu
Yongwon Kim
Petr Mordovskoi
Yugo Kanaya
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1535 KB) - Metadata XML
-
Supplement
(838 KB) - BibTeX
- EndNote
- Final revised paper