Emission factors and evolution of SO<sub>2</sub> measured from biomass burning in wild and agricultural fires

Rickly, Pamela; Guo, Hongyu; Campuzano-Jost, Pedro; Jimenez, Jose L.; Wolfe, Glenn M.; Bennett, Ryan; Bourgeois, Ilann; Crounse, John D.; Dibb, Jack E.; DiGangi, Joshua P.; Diskin, Glenn S.; Dollner, Maximilian; Gargulinski, Emily M.; Hall, Samuel R.; Halliday, Hannah S.; Hanisco, Thomas F.; Hannun, Reem A.; Liao, Jin; Moore, Richard; Nault, Benjamin A.; Nowak, John B.; Robinson, Claire E.; Ryerson, Thomas; Sanchez, Kevin J.; Schöberl, Manuel; Soja, Amber J.; St. Clair, Jason M.; Thornhill, Kenneth L.; Ullmann, Kirk; Wennberg, Paul O.; Weinzierl, Bernadett; Wiggins, Elizabeth B.; Winstead, Edward L.; Rollins, Andrew W.

doi:https://doi.org/10.5194/acp-2022-309

Preprints

https://doi.org/10.5194/acp-2022-309

Preprints

13 May 2022

Status: this preprint is currently under review for the journal ACP.

Emission factors and evolution of SO₂ measured from biomass burning in wild and agricultural fires

Pamela Rickly^1,2, Hongyu Guo^1,3, Pedro Campuzano-Jost^1,3, Jose L. Jimenez^1,3, Glenn M. Wolfe⁴, Ryan Bennett⁵, Ilann Bourgeois^1,2, John D. Crounse⁶, Jack E. Dibb⁷, Joshua P. DiGangi⁸, Glenn S. Diskin⁸, Maximilian Dollner⁹, Emily M. Gargulinski¹⁶, Samuel R. Hall¹⁰, Hannah S. Halliday¹¹, Thomas F. Hanisco⁴, Reem A. Hannun^4,12, Jin Liao^4,13, Richard Moore⁸, Benjamin A. Nault¹⁴, John B. Nowak⁸, Claire E. Robinson^8,15, Thomas Ryerson^2,a, Kevin J. Sanchez⁸, Manuel Schöberl⁹, Amber J. Soja^8,16, Jason M. St. Clair^4,12, Kenneth L. Thornhill⁸, Kirk Ullmann¹⁰, Paul O. Wennberg^6,17, Bernadett Weinzierl⁹, Elizabeth B. Wiggins⁸, Edward L. Winstead⁸, and Andrew W. Rollins² Pamela Rickly et al.

Show author details

¹Cooperative Institute for Research in Environmental Science, University of Colorado, Boulder, CO, USA
²Chemical Sciences Laboratory, NOAA, Boulder, CO, USA
³Department of Chemistry, University of Colorado, Boulder, CO, USA
⁴Atmospheric Chemistry and Dynamics Lab, NASA Goddard Space Flight Center, Greenbelt, MD, USA
⁵Bay Area Environmental Research Institute, NASA Ames Research Center, Moffett Field, CA
⁶Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA, USA
⁷Earth System Research Center, University of New Hampshire, Durham, NH, USA
⁸NASA Langley Research Center, Hampton, VA, USA
⁹Faculty of Physics, Aerosol Physics and Environmental Physics, University of Vienna, 1090 Vienna, Austria
¹⁰Atmospheric Chemistry Observations and Modeling Laboratory, National Center for Atmospheric Research, Boulder, CO, USA
¹¹Environmental Protection Agency, Research Triangle, NC, USA
¹²Joint Center for Earth Systems Technology, University of Maryland Baltimore County, Baltimore, MD 21250, USA
¹³Goddard Earth Science Technology and Research (GESTAR) II, University of Maryland Baltimore County, Baltimore, MD, USA
¹⁴CACC, Aerodyne Research, Inc.
¹⁵Science Systems and Applications, Inc., Hampton, VA, USA
¹⁶National Institute of Aerospace, Resident at NASA Langley Research Center, Hampton, VA, USA
¹⁷Division of Engineering and Applied Science, California Institute of Technology, Pasadena, CA, USA
^anow at: Scientific Aviation, Boulder, CO, USA

¹Cooperative Institute for Research in Environmental Science, University of Colorado, Boulder, CO, USA
²Chemical Sciences Laboratory, NOAA, Boulder, CO, USA
³Department of Chemistry, University of Colorado, Boulder, CO, USA
⁴Atmospheric Chemistry and Dynamics Lab, NASA Goddard Space Flight Center, Greenbelt, MD, USA
⁵Bay Area Environmental Research Institute, NASA Ames Research Center, Moffett Field, CA
⁶Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA, USA
⁷Earth System Research Center, University of New Hampshire, Durham, NH, USA
⁸NASA Langley Research Center, Hampton, VA, USA
⁹Faculty of Physics, Aerosol Physics and Environmental Physics, University of Vienna, 1090 Vienna, Austria
¹⁰Atmospheric Chemistry Observations and Modeling Laboratory, National Center for Atmospheric Research, Boulder, CO, USA
¹¹Environmental Protection Agency, Research Triangle, NC, USA
¹²Joint Center for Earth Systems Technology, University of Maryland Baltimore County, Baltimore, MD 21250, USA
¹³Goddard Earth Science Technology and Research (GESTAR) II, University of Maryland Baltimore County, Baltimore, MD, USA
¹⁴CACC, Aerodyne Research, Inc.
¹⁵Science Systems and Applications, Inc., Hampton, VA, USA
¹⁶National Institute of Aerospace, Resident at NASA Langley Research Center, Hampton, VA, USA
¹⁷Division of Engineering and Applied Science, California Institute of Technology, Pasadena, CA, USA
^anow at: Scientific Aviation, Boulder, CO, USA

Hide author details

Received: 26 Apr 2022 – Discussion started: 13 May 2022

Abstract. Fires emit sufficient sulfur to affect local and regional air quality and climate. This study analyzes SO₂ emission factors and variability in smoke plumes from US wild and agricultural fires, and their relationship to sulfate and hydroxymethanesulfonate (HMS) formation. Observed SO₂ emission factors for various fuel types show good agreement with the latest reviews of biomass burning emission factors, producing an emission factor range of 0.47–1.2 g SO₂ kg^-1 C in the emissions. These emission factors vary with geographic location in a way that suggests that deposition of coal burning emissions and application of sulfur-containing fertilizers likely play a role in the larger observed values, which are primarily associated with agricultural burning. A 0-D box model generally reproduces the observed trends of SO₂ and total sulfate (inorganic + organic) in aging wildfire plumes. In many cases, modeled HMS is consistent with the observed organosulfur concentrations. However, a comparison of observed organosulfur and modeled HMS suggests that multiple organosulfur compounds are likely responsible for the observations, but that the chemistry of these compounds yield similar production and loss rates to that of HMS, resulting in good agreement with the modeled results. We provide suggestions for constraining the organosulfur compounds observed during these flights and we show that the chemistry of HMS can allow for organosulfur to act as a S(IV) reservoir under conditions of increased pH (>6) and liquid water content (>10^-7 g m^-3). This can facilitate long-range transport of sulfur emissions resulting in increased SO₂ and eventually sulfate in transported smoke.

How to cite. Rickly, P., Guo, H., Campuzano-Jost, P., Jimenez, J. L., Wolfe, G. M., Bennett, R., Bourgeois, I., Crounse, J. D., Dibb, J. E., DiGangi, J. P., Diskin, G. S., Dollner, M., Gargulinski, E. M., Hall, S. R., Halliday, H. S., Hanisco, T. F., Hannun, R. A., Liao, J., Moore, R., Nault, B. A., Nowak, J. B., Robinson, C. E., Ryerson, T., Sanchez, K. J., Schöberl, M., Soja, A. J., St. Clair, J. M., Thornhill, K. L., Ullmann, K., Wennberg, P. O., Weinzierl, B., Wiggins, E. B., Winstead, E. L., and Rollins, A. W.: Emission factors and evolution of SO₂ measured from biomass burning in wild and agricultural fires, Atmos. Chem. Phys. Discuss. [preprint], https://doi.org/10.5194/acp-2022-309, in review, 2022.

Download & links

Preprint (PDF, 2810 KB)

Supplement (1951 KB)

Pamela Rickly et al.

Status: final response (author comments only)

RC1:
'Comment on acp-2022-309 Please check rate constant', Anonymous Referee #1, 19 May 2022

This paper reports the surprising result that HMS can act as a sulfur reservoir by conversion of HMS back to bisulfite and/or sulfite on relatively short time scales. The reaction describing this conversion is given in Table S1 and the work of Song et al. (2021) (and references therein) is referenced. In this paper's Table S1 the pre-exponential factor for the rate of HMS loss by this reaction is 6.2 x 10 +8 , however, in Song et al the factor is k = 6.2 x 10 -8, a difference of 16 orders of magnitude. Maybe this is a typo, or maybe it explains the unexpected short lifetime of HMS? It would be optimal for the authors to address this issue prior to a full review of the paper since it impacts a large portion of the manuscript.

Citation: https://doi.org/10.5194/acp-2022-309-RC1
- AC1: 'Reply on RC1', Pamela Rickly, 19 May 2022
  
  Thank you for pointing this out. This is a typo in Table S1, but the reaction is correctly represented in the model as k = 6.2 x 10 -8. The SI will be corrected accordingly.
  
  Citation: https://doi.org/10.5194/acp-2022-309-AC1
RC2:
'Comment on acp-2022-309', Anonymous Referee #2, 19 May 2022
The manuscript investigates the sulfur dioxide (SO2) emissions, and sulfate and hydroxymethanesulfonate (HMS) formation from US wild and agricultural fires. The study includes a combination of field data, collected during the course of two flights (one in Boise, ID and the other in Salina, KS) and modeling, using a 0-D box model. The authors provide SO2 fuel emission factors in agreement with previously literature reported values and provide evidence that HMS acts as S(IV) reservoir under higher pH and liquid water content conditions. This new role of HMS provides valuable insight on the role of sulfur-containing species under biomass burning conditions. I believe that the authors have conducted an analytical work that is of interest to the readers of ACP, however there are some points that need to be addressed. Please find below some comments that I believe need to be addressed in order to clarify specific results and the model mechanism used in this work.

The authors provide an HMS loss rate constant of: 6.2 × 10^8 × exp(−11400 × (1/T − 1/298)) +4.8 × 103 × (Kw/ H + ) × exp(−4700 × (1/T − 1/298), while Song et al. (2021, ACP), which is used as the reference of the rate, provide a rate constant of: 6.2 × 10^−8 × exp(−11400 × (1/T − 1/298)) +4.8 × 103 × (Kw/ H + ) × exp(−4700 × (1/T − 1/298). I assume that this is a typo, but please clarify.

Assuming that the HMS loss rate has a typo, additional comments are:

The role of HMS as S(IV) reservoir is very interesting, especially since this result is mainly under conditions of pH>6 and high LWC, in which HMS has been shown to be unstable and prone to additional reactions. In the model, the formation and decomposition of HMS is included (Table S1), however its reactions with OH and H2O2, which has been shown to occur at pH>6 (Kok et al., 1986. J. Geoph. Res.; Martin et al., 1989, Atmos. Environ.; Chapman et al., 1990, Atmos. Environ.) are not included. How are the results affected upon inclusion of these reactions?

The model represents efficiently the field data of August 3^rd, however it does not capture the trend of all the field data for the case of August 7^th. Since both days are within the same campaign in Boise, ID, what was the main differences between these days? It would be interesting to provide a brief explanation on why the two days differ, as provided for the two passes of the 7^th of August.

Field data are provided for mainly August 3^rd and 7^th, which correspond to the Boise flights. It would be beneficial to provide field data and the model performance for the Saline flights. Are the main results the same for both flights? This is not very clear.

It is stated in the manuscript that HMS can be over-predicted and that additional organosulfur species can be “the result of further reactions of HMS suggesting that the model is correctly reproducing the HMS formation chemistry, but indicating that the model aqueous phase chemistry is incomplete” (lines 639-641). The inclusion of HMS oxidation via OH and H2O2 might improve the HMS prediction for the cases that pH>6, however for more acidic conditions there is another pathway that can lead to sulfate formation but also add to the HCHO loading and potentially affect the HMS chemistry. HCHO can react directly with H2O2 forming hydroxymethyl hydroperoxide, which can then decompose to reform HCHO and H2O2 (Dovrou et al., 2022, PNAS). Since H2O2 and HCHO are observed via the flight measurements, could this pathway be useful for the model representation of these species as well as the organosulfur chemistry (as it provides further information regarding HCHO (source of HMS))?
Citation: https://doi.org/10.5194/acp-2022-309-RC2
RC3: 'Comment on acp-2022-309', Anonymous Referee #1, 06 Jun 2022

This paper investigates the emissions and chemistry of sulfur in wildfire smoke plumes measured during the FIREX study. The emissions estimates provide addition data to the existing knowledge; the results are consistent with past/published studies. An estimate on the overall contribution of wildfires to atmospheric SO2/sulfate would be nice to put this in perspective; are these fires an important or only minor source in the regions of the fires and the US as a whole.

The second, and much more speculative part of this study is the investigation, of SO2 oxidation pathways and fate of the products. The analysis of organo-sulfates, specifically HMS is novel, but highly uncertain since there is actually no HMS data presented so only generalized comparisons to observations are possible, ie, predicted HMS in a flight with highest evidence of organo-sulfates. The HMS analysis is also highly uncertain. As noted, (line 686), pH and LWC have the largest effect on HSM, and the dependance of HMS on pH and LWC is highly non-linear. Furthermore, pH and LWC themselves are uncertain and predicted by a model and no assessment is really shown to assess the pH and LWC predictions. Based on what little data is provided, it appears that the pH can vary substantially within even one smoke plume, making the predicted HMS highly uncertain. Overall, the authors should provide more details on the ISORROPIA predictions of LWC and pH since these are the key variables in the model. Actual HMS data is critical to assess the conclusions of this paper. It is noted that the SAGA MC instrument can detect S(IV) species, but interferences from SO2 make interpretation difficult. However, SAGA also measured particle ionic composition on filters, which should not have the SO2 artifact issue (or should be much less). Was that data checked to see if there was an evidence of HMS, or lack of it, on the contrasting flights discussed? It is curious why this is not discussed. Overall, the first part of the paper on emissions seems solid, the secondar part provides some interesting results, but they are really not verifiable and so it is hard to assess their validity and value. Can the authors reference any papers that show evidence that HMS may be present the atmosphere in general or more specifically in regions of fire influence?

Specific Comments.

Lines 85 to 92, what is missing here is some idea on the relative contribution of biomass burning to overall sulfate in various regions. Eg, since this paper is about fires in the US, can references be sited or estimates made on the relative contribution of fires to total S near and long distances from the fires.

Line 160 and on: This measurement technique allows for the speciation of submicron non-refractory particulate mass and the direct separation of inorganic and organic species having the same nominal mass to charge ratio (DeCarlo et al., 2006; Canagaratna et al., 2007). How is separation of inorganic and organic species possible if have same mass to charge?

Line 174 and on. What is the SAGA instrument (from Fig S8 it is the MC), which should be specified in the main text. Also state if both instruments sample over similar particle size ranges or not, and if not, what is the possible effect? What about comparing sulfate from the SAGA MC and Filters to see if there is substantial sulfate at higher particle sizes compared to those measured by the AMS and MC.

Regarding uptake of gases, such as SO2, HCHO …. I assume (3.4) & (3.6) is uptake to the dry particles since the LAS reported dry distributions? If so, it is not clear how uptake is to ambient wet particles and equilibrium between gas-LWC is handled. For species that may react slowly in the particle or LWC phase, is equilibrium established between the gas and particles by this model? What are the time scales for equilibrium for the various species and what are the time scales for which concentrations are changing in the plume being modeled? Same applies for calculation of pH where an equilibrium model is used (ISORROPIA). In essence, is equilibrium assumed, and if so, is it that reasonable (ie, provide justifications).

Regarding the pH prediction (lines 290 and on), it would be useful to show the particle phase ammonium predicted and measured (and same for gas phase ammonia), this would give both an idea of the model prediction and the amount of data thrown out due to the 40% criteria. Cations are mentioned, but not in much detail, specifically, what about K+, which could be high and seems odd not to be in the pH calculation.

Lines 347 to 362: Maybe the lack of correlation with MCE is that the MCE dynamic range is small? One might try looking at BC/OA ratios, just out of curiosity.

Line 465 and Fig 6a. There are very few data points for T<265K. Please comment on why this is (eg, what is the aircraft altitude) and how this limits conclusions drawn from this plot. One may wish to make a plot of altitude vs SO2/total S, since alt and T are related. Could this be explained by instrument sampling artifacts as a function of altitude? If this trend is driven by liquid reactions, then test with looking at predicted liquid water vs T or altitude.

Define vertical axis in Fig 7a fig caption.

Line 521, what is the uncertainty here, if uncertainty in pH and LWC are considered? Eg, 30% is stated for the various measurements and chemistry (I assume) for a given pH and LWC, but how sensitive is this to pH and LWC? (This is a common issue with the whole modeling section). The average pH for this smoke plume was 5.3 (range -2 to .8 ), but the LWC data is not given (at least I did not see it), please provide. It would be interesting to note the ratio of LWC to dry aerosol mass, which could be estimated based on the AMS data. This would give a sense of how wet the particles were.

Why not make plots of pH and LWC vs plume age for all these modeled plumes, given that these are key variables?

Also in Fig 7, what is the relative change in SO2 and sulfate in the 6 hrs? It appears to be fairly small. Fit the SO2 and sulfate data vs time with a line and test if the slope is actually not zero. Fig 7b, kind of gives a false sense of the importance of SO2 oxidation since it shows a huge time dependance, but in reality, there is very little change. It would be more realistic to plot the vertical axis as rate of oxidation divided by total sulfate, or something similar.

Regarding Section 4.2.2. Similar questions apply as noted above regarding sensitively of predictions to pH and LWC. What species is driving the pH so high (7.2)? Is this a realistic particle pH? The LWC is 10^-7 g/m3; what fraction of dry particle mass is that? Say the particle mass is 100 ug/m3, that is a ratio of 0.1, a small amount of LWC.

Line 595, agree within 40% for what pH and LWC?

In the second pass, how did the pH change relative to the first pass? Was it even higher due to higher LWC (x10)?

The sensitivity analysis is unclear (section 4.2.3.). Figure 11 shows that no HMS will be produced since loss is much larger than production, but the conditions of this flight (Aug 3) showed little organo-sulfate, so I guess there is consistency. Since highest concentrations of organo-sulfate was observed on the second leg of the 7 Aug flight, why not do a sensitivity analysis for this data. In fact, I would focus on this data, showing the predicted pH and LWC in detail and possibly also show how well ISORROPIA does by providing a comparison of the partitioning of ammonia, ammonium, nitric acid, and nitrate for this data. This could provide much more insight than the current analysis.

Citation: https://doi.org/10.5194/acp-2022-309-RC3

Pamela Rickly et al.

Supplement

https://doi.org/10.5194/acp-2022-309-supplement

Pamela Rickly et al.

Viewed

Total article views: 520 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
399	108	13	520	42	2	7

HTML: 399
PDF: 108
XML: 13
Total: 520
Supplement: 42
BibTeX: 2
EndNote: 7

Views and downloads (calculated since 13 May 2022)

Month	HTML	PDF	XML	Total
May 2022	303	78	10	391
Jun 2022	96	30	3	129

Cumulative views and downloads (calculated since 13 May 2022)

Month	HTML	PDF	XML	Total
May 2022	303	78	10	391
Jun 2022	96	30	3	129

Viewed (geographical distribution)

Total article views: 540 (including HTML, PDF, and XML) Thereof 540 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 25 Jun 2022

Short summary

Biomass burning sulfur dioxide (SO₂) emission factors range from 0.27–1.1 g kg^-1 C. Biomass burning SO₂ can quickly form sulfate and organosulfur, but these pathways are dependent on liquid water content and pH. Hydroxymethanesulfonate (HMS) appears to be directly emitted from some fire sources, but is not the sole contributor to the organosulfur signal. It is shown that HMS and organosulfur chemistry may be an important S(IV) reservoir with the fate dependent on the surrounding conditions.


Total:	0
HTML:	0
PDF:	0
XML:	0

Emission factors and evolution of SO2 measured from biomass burning in wild and agricultural fires

Supplement

Viewed

Viewed (geographical distribution)

Emission factors and evolution of SO₂ measured from biomass burning in wild and agricultural fires