the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Ground solar absorption observations of total column CO, CO2, CH4, and aerosol optical depth from California’s Sequoia Lightning Complex Fire: Emission factors and modified combustion efficiency at large scales
Sajjan Heerah
Aaron G. Meyer
Harrison A. Parker
Manvendra Dubey
Francesca M. Hopkins
Abstract. With global wildfires becoming more widespread and severe, tracking their emissions of greenhouse gases and air pollutants is becoming increasingly important. Wildfire emissions have primarily been characterized by in situ laboratory, and field observations at fine scales. While this approach captures the mechanisms relating emissions to combustion phase and fuel properties, their evaluation on large scale plumes has been limited. In this study, we report remote observations of total column trace gases and aerosols in the 2020 wildfire season of smoke plumes from the Sierra Nevada of California with an EM27/SUN solar Fourier transform infrared (FTIR) spectrometer. We derive total column aerosol optical depth (AOD), emission factors (EF) and modified combustion efficiency (MCE) for these fires, and evaluate relationships between them based on combustion phase at large scales. We demonstrate that the EM27/SUN effectively detects changes of CO, CO2 and CH4 in the atmospheric column at ~10 km scales that are attributed to wildfire emissions. These observations are used to derive total column EFCO of 120.5 ± 12.2 and EFCH4 of 4.3 ± 0.8 for a large smoke plume event in mixed combustion phases. These values are consistent with in situ relationships measured in similar temperate coniferous forest wildfires. FTIR derived AOD was compared to a nearby AERONET station and observed ratios of AOD to XCO were consistent with those previously observed from satellites. We also show that co-located XCO observations from the TROPOMI satellite-based instrument are 9.7 % higher than our EM27/SUN observations during the wildfire period. Finally, we put wildfire CH4 emissions in context of the California state CH4 budget and estimate that 213.7 ± 49.8 Gg CH4 were emitted by large wildfires in California during 2020, about 13.6 % of the total state CH4 emissions in 2019. Our novel application of an EM27/SUN solar spectrometer to quantify wildfire emission ratios at large scales follows predictive relationships that are consistent with in situ studies, offering promise for extensive monitoring from ground networks and satellite remote sensing.
- Preprint
(1659 KB) -
Supplement
(206 KB) - BibTeX
- EndNote
Isis Frausto-Vicencio et al.
Status: closed
-
RC1: 'Comment on acp-2022-671', Anonymous Referee #1, 09 Nov 2022
In this work, the authors use FTIR (EM27/SUN) measurements of XCO, XCO2, XCH4, and AOD to estimate fire emission factors and combustion phases during the 2020 wildfire seasons in California. Focusing on two major fires, the Creek and the Sequoia Lightning fires, they report emission factors (EFCO/CO2 and EFCH4/CO2) and modified combustion efficiency (MCE) and compare them to values reported in the literature. In addition, XCO derived from the EM27/SUN is compared to TROPOMI observations in wildfires, suggesting an overestimation of about 10% of the satellite. Finally, the authors estimate emission of CH4 from wildfire and discuss this value with regards to the California CH4 state budget.
General comments
While the scientific asset of using such an instrument to study wildfire is explicit, discussion about the limitation of this study is clearly missing.
The paper is well written and easy to follow. Some parts are however very repetitive and the manuscript would need some restructuring. There are some repetitions in the paper and authors should reorganize some parts to make it more concise. For instance, some similar sentences are shown in different parts throughout the paper: lines 78-80 similar to lines 240-244, line 557-559 is quasi similar to lines 94-96, and 26-28.
The discussion section is interesting and well written. The introduction section should be more concise and emphasize the research state of art. More appropriate references are needed. I would recommend not using references in the conclusion section and focusing on summarizing the main results of this study. A paragraph in the conclusion is missing to highlight the limitations and the perspectives of this work.
Although authors are very thorough in the sensitivity tests, the main concerns are error estimations and background measurements.
The error estimations are missing in this study. There are no estimation of the measurement uncertainties of Xgas and AOD which should be incorporated and propagated in the calculation of ER, EF and MCE. The slopes of the linear fits should reflect errors propagation and be mentioned with an error bar (+-).
The background values are very important in the ER computation. How can you ensure that the 2nd percentile of the daily measured mixing ratios represent the background at this location? Why comparing the background SJV values to the very remote location of Mauna Loa? Could you find appropriate background values located at closer sites (Caltech, Mont Wilson, Dryden, other)? The ER values greatly depend on background concentrations and measurement precision. Error bars should be added to these estimates.
Some figures could be improved for clarity purpose. Ex: the time series does not display well intraday variations.
The word “large scale” is recurrent in the manuscript and does not seem to be appropriated. Is regional more appropriated? The title should be modified since “large scales” is vague.
Overall, this paper addresses relevant scientific questions within the scope of ACP journal, which is why I recommend its publication.
Specific comments
Please clarify the differences between Emission Ratios and Enhancement Ratios.
Abstract line 19: please define at “10km scales”. Is it vertical or horizontal scale?
Line 57: “fire conditions”, please explain what conditions.
Figure 3b: the error bars are the standard deviation. Errors on both TROPOMI and EM27/SUN measurements should be included in the linear fit.
Section 3.3: authors state that FTIR and AERONET AOD are in agreement. What is the R value? How can you prove it?
Figure 4: reduce point size or find a solution to better display intraday variations.
Figure 5: What are the measurement errors? Could you propagate the errors to obtain slopes values with all uncertainties?
Figure 9: what is the error bar on the Top 20 CA wildfires emissions?
Please verify the order of the references in the parenthesis throughout the manuscript. It should follow the ACP journal recommendations: https://www.atmospheric-chemistry-and physics.net/submission.html#references (ex: line 72; lines 110-111; …)
Title of section 3.5 should be more specific. SJV GHG’s sources are only dairy farms?
Technical corrections
Line 34-35: Rephrase this sentence and define particulate matter 2.5.
Line 56: Is reference “CARB 2018” appropriate?
Line 51: Is IPCC 2014 correct? Can you refer to a more recent report?
Line 65-66, 69, 70, 71 and more: add references.
Figure 5, 8: Change dots color or size to clearly display all the points.
Figure 3b, 6, D1: R2 should be R2
Citation: https://doi.org/10.5194/acp-2022-671-RC1 - AC1: 'Reply on RC1', Isis Frausto-Vicencio, 01 Feb 2023
-
RC2: 'Comment on acp-2022-671', Anonymous Referee #2, 10 Nov 2022
GENERAL COMMENTS
This paper analyzes measurements from an EM27/SUN FTIR during the 2020 wildfire season in California. The EM27/SUN measurements are in generally good agreement when compared against TROPOMI satellite CO columns and AOD from a nearby AERONET station. The EM27/SUN measurements are also used to estimate emission factors (CO, and CH4) and modified combustion efficiency. The paper then estimates that 2020 wildfires made up 13.6% of CH4 emissions in California, based on emission ratios of CH4/CO2 and a reported estimate of CO2 emissions from wildfires.
Overall, the paper is well-written, and presents new data and interpretation that contributes to knowledge on wildfire emissions. I have provided specific comments below, which describe where the manuscript could be improved, for example by addressing areas where content is a bit confusing, fully describing all figures and tables, using uncertainties throughout the paper, and making the discussion more complete. I recommend publication of the paper, with minor revisions.
SPECIFIC COMMENTS
Line 25: When are 2020 CH4 emissions expected to be available? (Consider updating to 2020, if the estimates are available before publication.)Lines 26, 94, 557: “a novel application” – this statement is vague and, and should specify what is novel about the application compared with previous studies.
Line 62-64: “observations… focus on aerosol burden from smoke plumes with limited attention to trace gases…” This isn’t entirely true - the paper should indicate that there have been a number of studies that have looked at trace gases emitted from fires using satellite data, including ratios of species and estimation of emissions for CO, NOx, NH3 (see for example, Griffin et al., AMT 2021; Adams et al. ACP 2019; Whitburn et al., Atmos. Env., 2015 and references therein).
Line 65 : “.. present a new technique…” Is ground-based FTIR is a particularly new technique?
Line 91: “… satellite greenhouse gas observations…” perhaps replace with “… observations of CO…”
Line 97: I found this section a little bit difficult to follow. Recommend giving an overview of the various instruments involved (e.g., describing of Fig 1 to a high level) and the description of the fire to the top of the section (before Sect. 2.1), and then using the subsections to give the technical details. I would recommend that you describe/name the fires here and then use consistent naming throughout the document. (For example, the name “Shotgun” fire is used in some places – is this the same as the “North Complex” fire in Table E1?)
Line 150: Should panels b, c of the figure be referenced/described here?
Line 163: “… for a novel evaluation…” – is this evaluation novel? Has TROPOMI CO not been evaluated under smoky conditions before?
Line 180: Should Sect 2.4 be moved into an appendix, since it mostly references methods described elsewhere?
Line 295: For the comparisons against TROPOMI should uncertainties be included in the fits, etc?
Line 307 (and similar statements lines 500, 546): “These results suggest an overestimation of 9.7% X_CO from TROPOMI observations of wildfires.” Is there evidence that the difference is due to overestimation from TROPOMI? Or could this be due to differences in sampling or biases in the EM27/SUN data? Is there an uncertainty attached to the 9.7%? Also, what is the reported uncertainty in TROPOMI measurements and in the EM27/SUN measurements? Is 9.7% within the range of uncertainties? Is this bias consistent with previous studies?
Line 332 (Figure 4): Can figure 4 be added to figure 2 as a panel? Could be helpful to see all the timeseries together and would reduce the number of figures needed.
Line 339: “McMillan et al. (2008) found values…” should the word “values” be replaced with “slopes”
Line 340: “… 40 to 74…” should units be provided for this?
Lines 339-343: I find the discussion in these sentences a bit confusing and could be organized a bit more clearly. For example, are the values from McMillan et al of 44-74 the same as the values that are for clean region described further down the paragraph?
Line 348 (Figure 5 caption): There is no “teal line” on the figure – does this mean the “teal markers”?
Line 356: “… a steady MCE as X_CO, X_CH4, and AOD increased, indicating influence of smoldering combustion” Please elaborate – why does this indicate smoldering?
Lines 365-371: The discussion on EFs should be merged with the discussion on lines 458-464. I also find parts of the discussion a bit hard to follow. Which studies were included in the table and why? Which values are most relevant for comparison against the present study? How does Lueker et al., 2001 compare to the results in this study?
Line 372 (Table 1 caption): Should mention the present study in the caption, e.g., “Summary of past airborne studies, and the present study…”
Line 381 (Figure 6): Do panels a-e add value to the paper? Is the same/similar information captured in the broader timeseries in Fig 2?
Line 388 (Figure 7): If including this figure in the paper, should describe its relevance in the text.
Line 389 (Sect. 3.5 title): This section title is vague. Perhaps split Sect. 3.5 into two sections (one section about ratios for livestock vs wildfire emissions and another section about estimating total methane emissions from wildfires in California?)
Lines 401-4110: I found the discussion of the different ratios difficult to follow, and could use a rewrite for clarity. For example on line 407, “Similar ratios… were found in Hanford…” – I assume this means ratios similar to the non-wildfire ratios?
Line 414-415: “… dairy farm operations are the dominant source of CH4 during fire and non-fire periods.” This seems to contradict the next sentence, which says that during the strong smoke influence period, CH4 from the smoke is dominant. Clarify.
Line 416: “The immense scale…” recommend starting a new section (or at least a new paragraph) here.
Lines 430-435 and 635-640: Please provide a bit more information about how Table E1 was filled in. Why is the ER from the study in table E1 (0.0084) different from the ER given in line 362 (0.0073)? How were the ERs derived from the EFs in the literature? How were uncertainties propagated? Which values in Table E1 correspond to Pritchard and which correspond to Xu? Were there any cases where both Xu and Pritchard had different ER values for the same vegetation, and if so how did you choose which to use? (Also, check that everything is consistent between the text and the appendix: Line 434 references Xu 2020, but line 634 table caption references Prichard 2020 and Xu 2022).
Lines 454-457: Repetitive - delete summary of the work and save this for the conclusion/abstract?
Lines 471-487: This discussion feels a biased toward FTIR measurements. What are the drawbacks to FTIR measurements compared to the other methods? Is there information that can be provided by aircraft that can’t be provided by FTIR? Do all of these methods have similar uncertainties in, e.g., emission factors? Are the FTIRs more difficult to operate than say continuous ground-based in-situ analyzers? Is there potential for satellite (especially the next generation of satellites), modelling, or other methods to add to knowledge on fire emission factors as well?
Line 491: “with great resolution” – subjective (remove or replace this).
Line 520-529: Can more context be provided on the discussion of wildfire emissions of CH4? Are there other estimates of CH4 emissions from wildfires in the California or is a data gap? When discussing possible climate feedbacks, is it expected that a lot of CH4 is emitted from fires globally? Are CH4 emissions from fires considered, for example, in IPCC reports?
TECHNICAL CORRECTIONS
Line 490: “12. Smoke event” (delete the period after 12)
Citation: https://doi.org/10.5194/acp-2022-671-RC2 - AC2: 'Reply on RC2', Isis Frausto-Vicencio, 01 Feb 2023
Status: closed
-
RC1: 'Comment on acp-2022-671', Anonymous Referee #1, 09 Nov 2022
In this work, the authors use FTIR (EM27/SUN) measurements of XCO, XCO2, XCH4, and AOD to estimate fire emission factors and combustion phases during the 2020 wildfire seasons in California. Focusing on two major fires, the Creek and the Sequoia Lightning fires, they report emission factors (EFCO/CO2 and EFCH4/CO2) and modified combustion efficiency (MCE) and compare them to values reported in the literature. In addition, XCO derived from the EM27/SUN is compared to TROPOMI observations in wildfires, suggesting an overestimation of about 10% of the satellite. Finally, the authors estimate emission of CH4 from wildfire and discuss this value with regards to the California CH4 state budget.
General comments
While the scientific asset of using such an instrument to study wildfire is explicit, discussion about the limitation of this study is clearly missing.
The paper is well written and easy to follow. Some parts are however very repetitive and the manuscript would need some restructuring. There are some repetitions in the paper and authors should reorganize some parts to make it more concise. For instance, some similar sentences are shown in different parts throughout the paper: lines 78-80 similar to lines 240-244, line 557-559 is quasi similar to lines 94-96, and 26-28.
The discussion section is interesting and well written. The introduction section should be more concise and emphasize the research state of art. More appropriate references are needed. I would recommend not using references in the conclusion section and focusing on summarizing the main results of this study. A paragraph in the conclusion is missing to highlight the limitations and the perspectives of this work.
Although authors are very thorough in the sensitivity tests, the main concerns are error estimations and background measurements.
The error estimations are missing in this study. There are no estimation of the measurement uncertainties of Xgas and AOD which should be incorporated and propagated in the calculation of ER, EF and MCE. The slopes of the linear fits should reflect errors propagation and be mentioned with an error bar (+-).
The background values are very important in the ER computation. How can you ensure that the 2nd percentile of the daily measured mixing ratios represent the background at this location? Why comparing the background SJV values to the very remote location of Mauna Loa? Could you find appropriate background values located at closer sites (Caltech, Mont Wilson, Dryden, other)? The ER values greatly depend on background concentrations and measurement precision. Error bars should be added to these estimates.
Some figures could be improved for clarity purpose. Ex: the time series does not display well intraday variations.
The word “large scale” is recurrent in the manuscript and does not seem to be appropriated. Is regional more appropriated? The title should be modified since “large scales” is vague.
Overall, this paper addresses relevant scientific questions within the scope of ACP journal, which is why I recommend its publication.
Specific comments
Please clarify the differences between Emission Ratios and Enhancement Ratios.
Abstract line 19: please define at “10km scales”. Is it vertical or horizontal scale?
Line 57: “fire conditions”, please explain what conditions.
Figure 3b: the error bars are the standard deviation. Errors on both TROPOMI and EM27/SUN measurements should be included in the linear fit.
Section 3.3: authors state that FTIR and AERONET AOD are in agreement. What is the R value? How can you prove it?
Figure 4: reduce point size or find a solution to better display intraday variations.
Figure 5: What are the measurement errors? Could you propagate the errors to obtain slopes values with all uncertainties?
Figure 9: what is the error bar on the Top 20 CA wildfires emissions?
Please verify the order of the references in the parenthesis throughout the manuscript. It should follow the ACP journal recommendations: https://www.atmospheric-chemistry-and physics.net/submission.html#references (ex: line 72; lines 110-111; …)
Title of section 3.5 should be more specific. SJV GHG’s sources are only dairy farms?
Technical corrections
Line 34-35: Rephrase this sentence and define particulate matter 2.5.
Line 56: Is reference “CARB 2018” appropriate?
Line 51: Is IPCC 2014 correct? Can you refer to a more recent report?
Line 65-66, 69, 70, 71 and more: add references.
Figure 5, 8: Change dots color or size to clearly display all the points.
Figure 3b, 6, D1: R2 should be R2
Citation: https://doi.org/10.5194/acp-2022-671-RC1 - AC1: 'Reply on RC1', Isis Frausto-Vicencio, 01 Feb 2023
-
RC2: 'Comment on acp-2022-671', Anonymous Referee #2, 10 Nov 2022
GENERAL COMMENTS
This paper analyzes measurements from an EM27/SUN FTIR during the 2020 wildfire season in California. The EM27/SUN measurements are in generally good agreement when compared against TROPOMI satellite CO columns and AOD from a nearby AERONET station. The EM27/SUN measurements are also used to estimate emission factors (CO, and CH4) and modified combustion efficiency. The paper then estimates that 2020 wildfires made up 13.6% of CH4 emissions in California, based on emission ratios of CH4/CO2 and a reported estimate of CO2 emissions from wildfires.
Overall, the paper is well-written, and presents new data and interpretation that contributes to knowledge on wildfire emissions. I have provided specific comments below, which describe where the manuscript could be improved, for example by addressing areas where content is a bit confusing, fully describing all figures and tables, using uncertainties throughout the paper, and making the discussion more complete. I recommend publication of the paper, with minor revisions.
SPECIFIC COMMENTS
Line 25: When are 2020 CH4 emissions expected to be available? (Consider updating to 2020, if the estimates are available before publication.)Lines 26, 94, 557: “a novel application” – this statement is vague and, and should specify what is novel about the application compared with previous studies.
Line 62-64: “observations… focus on aerosol burden from smoke plumes with limited attention to trace gases…” This isn’t entirely true - the paper should indicate that there have been a number of studies that have looked at trace gases emitted from fires using satellite data, including ratios of species and estimation of emissions for CO, NOx, NH3 (see for example, Griffin et al., AMT 2021; Adams et al. ACP 2019; Whitburn et al., Atmos. Env., 2015 and references therein).
Line 65 : “.. present a new technique…” Is ground-based FTIR is a particularly new technique?
Line 91: “… satellite greenhouse gas observations…” perhaps replace with “… observations of CO…”
Line 97: I found this section a little bit difficult to follow. Recommend giving an overview of the various instruments involved (e.g., describing of Fig 1 to a high level) and the description of the fire to the top of the section (before Sect. 2.1), and then using the subsections to give the technical details. I would recommend that you describe/name the fires here and then use consistent naming throughout the document. (For example, the name “Shotgun” fire is used in some places – is this the same as the “North Complex” fire in Table E1?)
Line 150: Should panels b, c of the figure be referenced/described here?
Line 163: “… for a novel evaluation…” – is this evaluation novel? Has TROPOMI CO not been evaluated under smoky conditions before?
Line 180: Should Sect 2.4 be moved into an appendix, since it mostly references methods described elsewhere?
Line 295: For the comparisons against TROPOMI should uncertainties be included in the fits, etc?
Line 307 (and similar statements lines 500, 546): “These results suggest an overestimation of 9.7% X_CO from TROPOMI observations of wildfires.” Is there evidence that the difference is due to overestimation from TROPOMI? Or could this be due to differences in sampling or biases in the EM27/SUN data? Is there an uncertainty attached to the 9.7%? Also, what is the reported uncertainty in TROPOMI measurements and in the EM27/SUN measurements? Is 9.7% within the range of uncertainties? Is this bias consistent with previous studies?
Line 332 (Figure 4): Can figure 4 be added to figure 2 as a panel? Could be helpful to see all the timeseries together and would reduce the number of figures needed.
Line 339: “McMillan et al. (2008) found values…” should the word “values” be replaced with “slopes”
Line 340: “… 40 to 74…” should units be provided for this?
Lines 339-343: I find the discussion in these sentences a bit confusing and could be organized a bit more clearly. For example, are the values from McMillan et al of 44-74 the same as the values that are for clean region described further down the paragraph?
Line 348 (Figure 5 caption): There is no “teal line” on the figure – does this mean the “teal markers”?
Line 356: “… a steady MCE as X_CO, X_CH4, and AOD increased, indicating influence of smoldering combustion” Please elaborate – why does this indicate smoldering?
Lines 365-371: The discussion on EFs should be merged with the discussion on lines 458-464. I also find parts of the discussion a bit hard to follow. Which studies were included in the table and why? Which values are most relevant for comparison against the present study? How does Lueker et al., 2001 compare to the results in this study?
Line 372 (Table 1 caption): Should mention the present study in the caption, e.g., “Summary of past airborne studies, and the present study…”
Line 381 (Figure 6): Do panels a-e add value to the paper? Is the same/similar information captured in the broader timeseries in Fig 2?
Line 388 (Figure 7): If including this figure in the paper, should describe its relevance in the text.
Line 389 (Sect. 3.5 title): This section title is vague. Perhaps split Sect. 3.5 into two sections (one section about ratios for livestock vs wildfire emissions and another section about estimating total methane emissions from wildfires in California?)
Lines 401-4110: I found the discussion of the different ratios difficult to follow, and could use a rewrite for clarity. For example on line 407, “Similar ratios… were found in Hanford…” – I assume this means ratios similar to the non-wildfire ratios?
Line 414-415: “… dairy farm operations are the dominant source of CH4 during fire and non-fire periods.” This seems to contradict the next sentence, which says that during the strong smoke influence period, CH4 from the smoke is dominant. Clarify.
Line 416: “The immense scale…” recommend starting a new section (or at least a new paragraph) here.
Lines 430-435 and 635-640: Please provide a bit more information about how Table E1 was filled in. Why is the ER from the study in table E1 (0.0084) different from the ER given in line 362 (0.0073)? How were the ERs derived from the EFs in the literature? How were uncertainties propagated? Which values in Table E1 correspond to Pritchard and which correspond to Xu? Were there any cases where both Xu and Pritchard had different ER values for the same vegetation, and if so how did you choose which to use? (Also, check that everything is consistent between the text and the appendix: Line 434 references Xu 2020, but line 634 table caption references Prichard 2020 and Xu 2022).
Lines 454-457: Repetitive - delete summary of the work and save this for the conclusion/abstract?
Lines 471-487: This discussion feels a biased toward FTIR measurements. What are the drawbacks to FTIR measurements compared to the other methods? Is there information that can be provided by aircraft that can’t be provided by FTIR? Do all of these methods have similar uncertainties in, e.g., emission factors? Are the FTIRs more difficult to operate than say continuous ground-based in-situ analyzers? Is there potential for satellite (especially the next generation of satellites), modelling, or other methods to add to knowledge on fire emission factors as well?
Line 491: “with great resolution” – subjective (remove or replace this).
Line 520-529: Can more context be provided on the discussion of wildfire emissions of CH4? Are there other estimates of CH4 emissions from wildfires in the California or is a data gap? When discussing possible climate feedbacks, is it expected that a lot of CH4 is emitted from fires globally? Are CH4 emissions from fires considered, for example, in IPCC reports?
TECHNICAL CORRECTIONS
Line 490: “12. Smoke event” (delete the period after 12)
Citation: https://doi.org/10.5194/acp-2022-671-RC2 - AC2: 'Reply on RC2', Isis Frausto-Vicencio, 01 Feb 2023
Isis Frausto-Vicencio et al.
Isis Frausto-Vicencio et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 300 | 110 | 12 | 422 | 30 | 2 | 4 |
- HTML: 300
- PDF: 110
- XML: 12
- Total: 422
- Supplement: 30
- BibTeX: 2
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1