the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Jul 2022
Contribution of marine biological emissions to gaseous methylamines in the atmosphere: an emission inventory based on satellite data
Abstract. Methylamines can readily react with acidic gases in the atmosphere, which consequently has an important impact on the atmospheric environment. It is difficult to measure amines in field studies due to their high reactivity, and therefore, numerical modelling is an effective tool to study ambient amines. However, the contribution of marine biological emissions (MBE), an important source of methylamines (MA), has been insufficiently investigated in the current emission inventory. Therefore, this study utilized satellite data such as Sea Surface Temperature (SST), Chlorophyll-a (Chla), Sea Surface Salinity (SSS) and model simulation data (Wind Speed, WS) to establish a more reasonable MBE inventory of amines. Spatial and temporal distribution of methylamine emissions indicates that MBE fluxes of monomethylamine (MMA) and trimethylamines (TMA) can be comparable with or even higher than that of terrestrial anthropogenic emissions (AE), while for dimethylamines (DMA), the ocean acts as a sink. The method used in this study can better reflect the exchange direction of amines between ocean and atmosphere, and reflect the emission characteristics of different amines. From WRF-Chem simulation results, the concentration of amines in the coastline was found to increase significantly due to the contribution of MBE. Wind and Chla were potentially the most important factors affecting MBE fluxes. WS is directly used in the calculation of ocean-atmosphere exchange coefficient Kg, and the direction of the prevailing winds in different seasons affects the area of influence of the MBE. Chla indirectly influences the calculation results of exchange flux by affecting the calculation of pH. In addition, the emission fluxes and spatial distribution of AE and wet deposition also affect the simulation of amines.
- Preprint
(2681 KB) - Metadata XML
-
Supplement
(2454 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on acp-2022-394', Anonymous Referee #1, 29 Jul 2022
In the current study the authors provide a combined satellite and model study to estimate oceanic emissions of methylamine (MMA), dimethylamine (DMA) and trimethylamine (TMA) along the Chinese coastlines. Therefore, sea surface temperature (SST), chlorophyll-a (chl-a), sea surface salinity (SSS) and wind speed data were used. Recent investigations show that satellite data are a useful tool to simulate and understand emissions from the ocean. The study reveals that amine emissions from the ocean can have significant contribution to gas-phase TMA and MMA concentrations, but not to DMA concentrations. Through sensitivity studies wind speed and chl-a concentrations were found to be important drivers of amine emissions. The modeled gas-phase concentrations of MMA, DMA and TMA are compared with measurements in that region and found good matches. Regarding the importance of amines for new particle formation and current limitations the paper addresses relevant scientific questions in the field of atmospheric chemistry.
Nevertheless, emissions from the oceans towards the atmosphere require good established concentration measurements within the sea water. These are not given for the Chinese coastline and thus the authors used an average value derived from different measurements. Here, high uncertainties can exist. Furthermore, the authors do not use established measured physical and chemical parameters for these amines. Besides, it seems that there is a bug in the calculation of the pKa value. Therefore, the simulations have to be reperformed.
The paper needs major revision before publication.
Main Comments
Parameters such as pKa and Henry’s Law coefficient are important to calculate the amine flux into the atmosphere. The authors use Henry’s Law coefficients obtained for NH3 as it was done in recent studies. This approach is feasible if such values are not given in literature. However, these values are determined.
Why are the authors not using the Henry’s Law coefficients for MMA, DMA and TMA as provided in Sander (2015) and Leng et al. (2015)?
The pKa values used in table 4 from Gibb et al. (1999) are valid for 25°C not 20°C. Why are the temperature dependent pKa values given in the review of Ge et al. (2011b) not used?
The simulations have to be reperformed by using the values for MMA, DMA, and TMA.The authors use observed values of MMA, DMA and TMA dissolved in sea water from other sea areas and state that “all sites are located in densely populated areas”. However, values from Hawaii or the Arabian Sea are used which are obviously not as densely populated areas as the Chinese coastline. Why do the authors use only these values, but neglect other measured values from Yang et al. (1994), Gibb et al. (1999), and van Pinxteren et al. (2019)?
Formula 10
In the publication of Khoo et al. (1977) the formula is different. It is
pKa = pKwa + (0.1552 – 0.0003142T)*I instead of the applied
pKa = pKwa + (0.1552 – 0.003142T)*I
Furthermore, there was a correction of this prediction method by Bell et al. (2007; 2008). This has to be checked and the simulations have to be reperformed.The authors state that chl-a influences the emission of amines into the atmosphere, but concentrate only on the pH effect. In environments rich of biological activity such as the sea-surface microlayer DMA concentrations can be up to one order of magnitude larger than in the bulk (van Pinxteren et al, 2019). A sensitivity study dealing the possible effect of higher chl-a on dissolved amine concentrations is missing and has to be done.
For anthropogenic emissions the authors use the Amines-to-Ammonia mass emission ratio derived for the Yangtze River Delta region. This ratio will result into lower TMA emissions compared to MMA and DMA. However, in the review of Ge et al. (2011a) animal husbandry TMA emissions dominate MMA and DMA that is not reflected in the current study by the lower agricultural emission ratio. Thus, the high ocean contribution might be coincidence because of the underestimation. The ratio for agricultural emission from Mao et al. (2018) is further 0.00043 instead of the applied 0.0004.
Besides, recently a new source of C2 and C3 amines was detected in China (Chang et al., 2021) that might be a potent TMA source. This is not represented, yet.Why do the authors not treat uptake on aerosols that was determined to be important for the lifetime of amines by Yu and Luo (2014)?
The authors do refer to a mass ratio of amines with ammonia from Zheng et al. (2015) for the chemical boundary conditions of the model, but there a ratio between amines as well as ammonia with NOx is given. From the values of Zheng et al. (2015) it is hard to recalculate the ratio presented in this study.
There a two recent measurements campaigns of marine MMA, DMA and TMA in the gas phase at the study area. However, it seems the results are not well discussed. For example, TMA measured is around one order of magnitude higher in Gao et al. (2022), but a discussion is missing. Furthermore, in table 6 the values are sometimes in ng or µg m-3 instead of pptv making a comparison difficult. I suggest that the authors include their modeled average values together with pptv also in ng m-3 for better comparison.
Overall, from the most recent studies it seems that during winter the model overpredict DMA, but underpredict TMA.Minor Comments
Line 61
Does this refer towards NH3? Should be better specified.Line 93
Delete “And”Line 104
“Multiple”Lines 108-109
The authors focus only on a very small part of the North Pacific. The sentence should be rephrased.Line 142
Please provide a reference for the reaction rate coefficients.Line 145
How long was the spin-up time for the model?Line 255
From the figure it seems that it is April > October > July > January.Line 257
“to the at the ocean” rephraseLine 302
Is this the average of the full area? Has to be mentioned.Line 303
Is the last part really needed?Line 311
The referring to MBE is missing for the TMA comparison.Line 334
Provide a reference for the degradation in sediments.Line 342
With decreasing SST the emission increases.Section 3.2
For the reader it would be better when the discussion is structured into (i) MMA, (ii) DMA and (iii) TMA as it is for the figures. Separate figures showing the percentage changes over the ocean will help to better understand and follow the discussion.Line 364
Table 7 instead of table 6Line 395
How is the difference of HONO + OH between the simulation with and without MBE?Line 405
Add that the low agricultural emissions of DMA are related to the contribution.Line 419
In figure S2 and S3 the distribution of the residential emissions is not shown.Section 3.3.2
A discussion of DMS when the WS is reduced is missing. Why are there so strong changes in July?Section 3.3.3
When the Henry’s Law coefficient is changed also the emission flux is changed. This discussion is missing.Line 491
Because of the missing sea water data and the missing uptake on aerosol particles, it could be questionable if the last part of the sentence (“which is more consistent with the reality”) is really true. It might be true on global scale, but not necessarily on regional scale.Table 6
Please add the year of the measurement. Provide also average values.Figure 4
The figure caption describes MBE and AE contribution different as mentioned in the figure legend.Figure S2-S6
Give the name of month instead of a number and symbol.References
Bell et al., Ammonia/ammonium dissociation coefficient in seawater: A significant numerical correction, Environ. Chem., 4, 183–186, https://doi.org/10.1071/EN07032, 2007.
Bell, et al., Corrigendum to: Ammonia/ammonium dissociation coefficient in seawater: A significant numerical correction. Environ. Chem., 5, 258. https://doi.org/10.1071/en07032_co, 2008
Chang et al., Discovery of a Potent Source of Gaseous Amines in Urban China, Environ. Sci. Technol. Lett., 8, 725-731, https://doi.org/10.1021/acs.estlett.1c00229, 2021.
Ge et al., Atmospheric amines - Part I. A review, Atmos. Environ., 45, 524-546, https://doi.org/10.1016/j.atmosenv.2010.10.012, 2011a.
Ge et al., Atmospheric amines e Part II. Thermodynamic properties and gas/particle partitioning, Atmos. Environ., 45, 561-577, https://doi.org/10.1016/j.atmosenv.2010.10.013, 2011b.
Gibb et al.: Distributions and biogeochemistries of methylamines and ammonium in the Arabian Sea, Deep-Sea Res. Part II, 46, 593–615, https://doi.org/10.1016/S0967-0645(98)00119-2, 1999.
Leng et al., Temperature-Dependent Henry’s Law Constants of Atmospheric Amines, J. Phys. Chem. A, 119, 8884-8891, https://doi.org/10.1021/acs.jpca.5b05174, 2015.
Sander, R.: Compilation of Henry's law constants (version 4.0) for water as solvent, Atmos. Chem. Phys., 15, 4399–4981, https://doi.org/10.5194/acp-15-4399-2015, 2015.
van Pinxteren et al.: Aliphatic amines at the Cape Verde Atmospheric Observatory: Abundance, origins and sea-air fluxes, Atmos. Environ., 203, 183-195, https://doi.org/10.1016/j.atmosenv.2019.02.011, 2019
Yang et al.: Seasonal variation in concentration and microbial uptake of methylamines in estuarine waters, Mar. Ecol. Prog. Ser., 108, 303-312, https://doi.org/10.3354/meps108303, 1994.
Citation: https://doi.org/10.5194/acp-2022-394-RC1 -
AC1: 'Reply on RC1', Xuemei Wang, 10 Jan 2023
The comment was uploaded in the form of a supplement: https://acp.copernicus.org/preprints/acp-2022-394/acp-2022-394-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Xuemei Wang, 10 Jan 2023
-
RC2: 'Comment on acp-2022-394', Anonymous Referee #2, 05 Aug 2022
Overall impression
The study aims to estimate the gas phase concentration of three different amines using WRF-Chem. In addition to previous similar attempts using anthropogenic amine emissions only, the authors apply an online marine biogenic amine emission scheme. This new method seems sound and valid. However, the manuscript lacks of a thorough validation. (Some) Measurement results of different studies are presented in a table but never discussed in context with the simulated concentrations. Furthermore, the conclusions and discussion is not clearly formulated. For example, it is often unclear what type of aggregation / averages are referred to when presenting relative differences. This made it hard to follow the discussion and the results and likely lead to misunderstandings. Finally, potential shortcomings of the current method (e.g. loss to atmospheric aerosols is not considered) is not discussed anywhere in the paper.
Despite the fact that the manuscript could be improved in terms of language and clearer formulations at many points, the manuscript also needs major revision in the interpretation of results, discussion and conclusion. Since measured concentrations were presented, an evaluation section that compares these measurements, in particular the ones of the same year, with the modelled concentrations.
Major comments
l. 138-139: Why is loss to aerosols not considered? Since it was correctly described in the introduction that amines play a role in aerosol formation, there should be a considerable loss to aerosols. This should be discussed here and in the conclusions.
l. 145-146: It is stated that 2015 was chosen since it had ‚more consecutive days with field observations’, however, these field observations were not used in the manuscript for validation purposes. Furthermore, the authors cite e.g. Zheng et al., 2015, in this context, who performed measurements in 2012, so this study probably does not help to explain why 2015 was chosen.
l. 148-149: The mentioned verification discussion is missing in the manuscript.
l. 349-350: Weren't the simulation periods chosen to compare with observations? What were the available observations in the simulated time periods of 2015? Above it was mentioned, that daily measurements are available in certain periods. These periods were simulated. So why don’t the authors take the grid cell where the measurement took place and produce a daily mean from the simulation and compare this to the observation?
l. 351-352: Since the range of observed concentrations spans ~an order of magnitude for each of the three investigated amines, it's hard to say that simulation results are 'close'. What you could generally infer is that simulation and available observations agree in terms of order of magnitude. The increase due to MBE does not seem to be very strong seen the numbers in Table 6 to justify the statement 'inclusion of marine biogenic emissions can make up for the shortcomings of the previous models' underestimation of amines'
l. 355: The mentioned verification part is missing. In the text, there are no comparisons between measured and simulated values and no discussion of the evaluation of the simulated values against the observations shown in Table 6.
l. 364-365: Which region correspond these numbers to? I get different relative changes when using the numbers given in Table 6.
Later I see, that these numbers seem to refer to only regions with an increase shown in Table 7, right? Why do the authors present these numbers and not the domain average in the text (MA in Table 7 if I understood correctly)? In any case, it need to be made much clearer in the whole discussion sections what averages the given numbers in the text refer to. At present it is impossible to follow the discussion of results in a fair amount of time.
Conclusion section: It has to be stated that the drawn conclusions are only valid for the given months in the year 2015. What is with potential limitations since loss to atmospheric aerosol is not considered? This should be discussed. Overall none of the results of the sensitivity simulations in sections 3.3 is summarized in the conclusion. Why not? And a comparison to the available measurements is missing.
l. 484-486: The authors should summarize what they actually did. This numerical modelling to simulate gas phase amine concentration. In addition to anthropogenic emission MBE of amines was treated in an online manner.
Minor comments
Please check the manuscript with a native English speaker if possible. At many instances, some formulations were misleading or seemed at least very unfamiliar.
l. 20: ‚has been insufficiently investigated in the current emission inventory.‘ What do the author’s want to express? An emission inventory cannot investigate.
l. 60-62: Stronger and more easily oxidized than what? Likely ammonia, then please state it like this.
l. 63-65: Is the lower concentration of amines compared to ammonia really a result of faster loss processes only or at least partly due to the lower emission rates?
l. 76-80: It is referred to ‚increased concentrations’, an ‚increase by 1-2 orders of magnitude‘ and a ‚significant increase‘, but it is not mentioned compared to what concentrations have increased?
l. 89-90: Not sure if China’s location alone can be called unique. Densely populated coastal areas can be found in many places in the world. The combination of location and pollution is rather what could be called unique.
l. 95: ‚amines’ role in atmospheric chemistry‘ sounds strange. Better: ‚the role of amines in …’?
l. 101: Is ‚reasonable‘ the right word? ‚Complete‘ or ‚detailed‘ might fit better.
l. 102-103: First, is the emission mass ratio in the Mao et al., 2018, AE emission inventory really ‚arbitrary‘? Second, since this scheme is used as it is, and just an online MBE scheme was added, your study does not ‚overcome the arbitrariness‘ in the AE emission inventory. Please reformulate these sentences in a more clearer way.
l. 104: Multiple -> multiple
l. 121: This is confusing here. Which is the ‚previous method’ you are referring to?
l. 127-129: Is an ammonia / amine ratio, e.g. similar to traffic, also applied for ammonia emissions from ships? If not, why not?
l. 150: What do the authors mean with 'representative'? Do these months represent a specific season? If so, is this also the case for these months in the year 2015? As the authors state later, at least SST might be non-typical for parts of the 2015, which indicates it is not representative in general. So, please clarify what is meant with ‚representative’ in the context of this paper. I doubt that 10 days of one month can be called ‚representative‘ for the same month in a climate view or in other years.
l. 154: ‚chemical boundary‘ -> ‚chemical boundary conditions‘
l. 154-155: Did the authors run CAM-CHEM themselves to generate boundary conditions? This is what I understand from reading. However, the webpage given a few sentences later suggest that driving data for boundary conditions can be download. Please clarify.
l. 168: Why the reference to these measurements of concentrations in air and in water by Gibb et al., 1999? How exactly were these used?
l. 188: solvated -> dissolved?
l. 189-190: ‚this study selected the average of the observed values of other sea areas (Table 3)’. I don’t understand what the authors want to express here. Which numbers of Table 3 are used in the emission algorithm and how exactly? The description of your method might need more detail or clarification.
l. 198-199: Perhaps related to the previous comment… It would be helpful if the authors can give their calculated average values in Table 3. This helps the reader to much easier follow the method.
l. 208: The unit is probably is 45 g kg-1. The term ‚within‘ suggest a range but only one number is given.
l. 226-228: Was this done by the authors? If yes, please describe in more detail how the SST data was combined with the other satellite data sets mentioned here. According to the description before, the NESDIS SST data is already on 5 km horizontal resolution.
l. 235: ‚SST presents the seasonality‘ sounds unfamiliar. Please check for proper English.
l. 257: Obsolete "at the".
l. 275: ion intensity -> ionic strength?
section 2.5.4: Is time-resolved (e.g. hourly) wind speed utilized or monthly mean?
l. 293: Meaning of FNL is missing.
l. 294-295: Well, there is considerable bias for RH. This might not be important for your study, but in any case should not be stated suggestion good agreement. Most importantly is that model and observations of wind speed agree reasonably well.
l. 302: Are the numbers presented in Table 5 for one grid cell or domain mean? Or only over ocean?
l. 303-307: Why not also in table 5 change the signs, which is then consistent with the rest of the figures and the paper?
l. 308, Fig. 3: From table 5 I understand that the ocean is a source for MMA and TMA, but a sink for DMA. Why aren't therefore negative emission fluxes seen for DMA over ocean in Fig. 3?
l. 309, Fig. 4: Is that domain mean?
l. 309: ‚MBE emission‘ -> emission to much, since the E in MBE already refers to emission.
l. 310, 311: TMA -> ‚MBE flux of TMA‘ or similar.
l. 309-314: Why no discussion of DMA emission flux, but only for MMA and TMA?
l. 315-335: Just as a suggestion: These paragraphs might be more useful directly after section 2.5, before section 3.1.
l. 315, Fig. 5: Which values were used for the other three input variables, while only one is varied? I understand that the four plots do not show the contribution, but the variation of the emission flux due to the variation of one variable while the others were held constant. It is important for the interpretation and the magnitude of the contribution to provide a clearer description here.
Please also clarify which area or aggregation Fig. 5 refers to. Is that temporal domain mean or ocean mean or one grid cell?
l. 316: ‚Kg, which is directly proportional to Kg’ -> Probably the latter Kg is meant to be something else.
l. 325, 338, 344, 346 and later occurrences: overflow -> emission, release or similar. Check with native speaker.
l. 347: 'affecting the variation of the exchange fluxes' would be more precise in my view.
l. 349, 354: What is the temporal and spatial resolution of the model? This should be mentioned in section 2.3 where the model is described.
l. 358-360: Future tense ('will') does not sound correct.
l. 360-361: No significant change in autumn and winter seems to be the case for MMA and DMA only, according to Table 6.
l. 381 and other occurrences of ‚offshore‘: Shouldn't ‚offshore‘ be on the open ocean, at least farer away from the coast on the ocean? Why is land mass with 500m above sea level considered as ‚offshore’?
l. 393: Formulation sounds incorrect. Perhaps better 'was lowered / decreases by less than 10%'.
l. 396-398: Can that be shown with the model results, e.g. NO, OH concentration maps?
l. 400: What do the authors mean with ‚model mechanism‘?
l. 419-420: ‚they have relatively high fluxes’: What is the relative contribution of residential emissions in the mentioned areas? It seems that this is rather uniform and hence the relative change is more uniform (hence the relative change not as 'obvious' as you say).
l. 420-421: ‚the difference‘ -> If you refer to Fig. S2 and S3, the authors probably mean ‚the relative difference‘.
l. 423: What is high value and low value area?
l. 445-446: Shouldn't the change of pH due to Chla change directly affect the sea air exchange of DMA as well? What do you mean with 'the change of MBE of DMA is not considered'?
l. 447: 'range of DMA' -> 'range of DMA change'
l. 449-450: ‚therefore, DMA reacts with ·OH faster than the other two amines in the upwind direction, and its concentration is less affected.‘ Can the authors please explain in more detail what they mean here?
l. 466-467: ‚going below 50 mm in October’: What region does that refer to? Also in July many places in the domain have precipitation amounts < 50 mm.
l. 475: ‚DMA is only affected by terrestrial AE‘: Didn't it have negative emission fluxes over sea water, i.e. is lost into the ocean surface (Fig. 4)?
l. 476: variation range of its concentration -> variation range of its concentration change
l. 490: It's not the satellite data itself, but its application in an online emission scheme, that might reflect the emission situation.
l. 494-495: Is that 500km away from the coast on land or 500km away over the ocean?
l. 495-496: These are the maximum increases for TMA due to MBE, aren't they? Why don’t authors instead mention the average increases in the text as was done for MMA. I believe that more readers would be interested in averages than in the extremes? Anyhow, it always must be clear what the given numbers are. This is missing throughout the manuscript in many places and makes it rather poor.
l. 499: ‚WS and Chla were found to be the dominant factors affecting MBE fluxes’: This is the case for all amines, right? Then please state it like this.
l. 507: ‚and the ocean also transforms from a source of amines to a sink’: Was that observed in the simulations, since such result was not presented in section 3.3.2? And was it the case for all amines or only some? Since the Chla concentrations, according to Fig. 2, were mostly < 10 mg m-3, 50% increase should only make it a sink for DMA, and very slightly MMA perhaps.
Table 6: Some observation results are presented in µg m-3 and ng m-3. These should be transformed to pptv. Why are the Gao et al. 2022 observations of TMA much larger than for the other studies?
Fig. 2: The numbers in the plots: Black for SST, Chla, SSS and white color for WS would probably give better readability.
Fig. 4: Labels are wrong. In the figure, dashed is MBE and solid lines are AE.
Comments to the author response to the quick initial review comments
In the response, changes in lines 362, 418, 464 are mentioned (with ‚descriptions and data sources’), but I can’t find these changes.
The comparison to measurements can of course be done only in a qualitative manner, but it should at least be discussed in the manuscript and not only presented in a table (Table 6).
Citation: https://doi.org/10.5194/acp-2022-394-RC2 -
AC2: 'Reply on RC2', Xuemei Wang, 10 Jan 2023
The comment was uploaded in the form of a supplement: https://acp.copernicus.org/preprints/acp-2022-394/acp-2022-394-AC2-supplement.pdf
-
AC3: 'Reply on AC2', Xuemei Wang, 10 Jan 2023
<strong>Publisher’s note: this comment is a copy of AC2 and its content was therefore removed.</strong>
Citation: https://doi.org/10.5194/acp-2022-394-AC3 -
AC4: 'Reply on AC2', Xuemei Wang, 10 Jan 2023
<strong>Publisher’s note: this comment is a copy of AC2 and its content was therefore removed.</strong>
Citation: https://doi.org/10.5194/acp-2022-394-AC4
-
AC3: 'Reply on AC2', Xuemei Wang, 10 Jan 2023
-
AC2: 'Reply on RC2', Xuemei Wang, 10 Jan 2023
Status: closed
-
RC1: 'Comment on acp-2022-394', Anonymous Referee #1, 29 Jul 2022
In the current study the authors provide a combined satellite and model study to estimate oceanic emissions of methylamine (MMA), dimethylamine (DMA) and trimethylamine (TMA) along the Chinese coastlines. Therefore, sea surface temperature (SST), chlorophyll-a (chl-a), sea surface salinity (SSS) and wind speed data were used. Recent investigations show that satellite data are a useful tool to simulate and understand emissions from the ocean. The study reveals that amine emissions from the ocean can have significant contribution to gas-phase TMA and MMA concentrations, but not to DMA concentrations. Through sensitivity studies wind speed and chl-a concentrations were found to be important drivers of amine emissions. The modeled gas-phase concentrations of MMA, DMA and TMA are compared with measurements in that region and found good matches. Regarding the importance of amines for new particle formation and current limitations the paper addresses relevant scientific questions in the field of atmospheric chemistry.
Nevertheless, emissions from the oceans towards the atmosphere require good established concentration measurements within the sea water. These are not given for the Chinese coastline and thus the authors used an average value derived from different measurements. Here, high uncertainties can exist. Furthermore, the authors do not use established measured physical and chemical parameters for these amines. Besides, it seems that there is a bug in the calculation of the pKa value. Therefore, the simulations have to be reperformed.
The paper needs major revision before publication.
Main Comments
Parameters such as pKa and Henry’s Law coefficient are important to calculate the amine flux into the atmosphere. The authors use Henry’s Law coefficients obtained for NH3 as it was done in recent studies. This approach is feasible if such values are not given in literature. However, these values are determined.
Why are the authors not using the Henry’s Law coefficients for MMA, DMA and TMA as provided in Sander (2015) and Leng et al. (2015)?
The pKa values used in table 4 from Gibb et al. (1999) are valid for 25°C not 20°C. Why are the temperature dependent pKa values given in the review of Ge et al. (2011b) not used?
The simulations have to be reperformed by using the values for MMA, DMA, and TMA.The authors use observed values of MMA, DMA and TMA dissolved in sea water from other sea areas and state that “all sites are located in densely populated areas”. However, values from Hawaii or the Arabian Sea are used which are obviously not as densely populated areas as the Chinese coastline. Why do the authors use only these values, but neglect other measured values from Yang et al. (1994), Gibb et al. (1999), and van Pinxteren et al. (2019)?
Formula 10
In the publication of Khoo et al. (1977) the formula is different. It is
pKa = pKwa + (0.1552 – 0.0003142T)*I instead of the applied
pKa = pKwa + (0.1552 – 0.003142T)*I
Furthermore, there was a correction of this prediction method by Bell et al. (2007; 2008). This has to be checked and the simulations have to be reperformed.The authors state that chl-a influences the emission of amines into the atmosphere, but concentrate only on the pH effect. In environments rich of biological activity such as the sea-surface microlayer DMA concentrations can be up to one order of magnitude larger than in the bulk (van Pinxteren et al, 2019). A sensitivity study dealing the possible effect of higher chl-a on dissolved amine concentrations is missing and has to be done.
For anthropogenic emissions the authors use the Amines-to-Ammonia mass emission ratio derived for the Yangtze River Delta region. This ratio will result into lower TMA emissions compared to MMA and DMA. However, in the review of Ge et al. (2011a) animal husbandry TMA emissions dominate MMA and DMA that is not reflected in the current study by the lower agricultural emission ratio. Thus, the high ocean contribution might be coincidence because of the underestimation. The ratio for agricultural emission from Mao et al. (2018) is further 0.00043 instead of the applied 0.0004.
Besides, recently a new source of C2 and C3 amines was detected in China (Chang et al., 2021) that might be a potent TMA source. This is not represented, yet.Why do the authors not treat uptake on aerosols that was determined to be important for the lifetime of amines by Yu and Luo (2014)?
The authors do refer to a mass ratio of amines with ammonia from Zheng et al. (2015) for the chemical boundary conditions of the model, but there a ratio between amines as well as ammonia with NOx is given. From the values of Zheng et al. (2015) it is hard to recalculate the ratio presented in this study.
There a two recent measurements campaigns of marine MMA, DMA and TMA in the gas phase at the study area. However, it seems the results are not well discussed. For example, TMA measured is around one order of magnitude higher in Gao et al. (2022), but a discussion is missing. Furthermore, in table 6 the values are sometimes in ng or µg m-3 instead of pptv making a comparison difficult. I suggest that the authors include their modeled average values together with pptv also in ng m-3 for better comparison.
Overall, from the most recent studies it seems that during winter the model overpredict DMA, but underpredict TMA.Minor Comments
Line 61
Does this refer towards NH3? Should be better specified.Line 93
Delete “And”Line 104
“Multiple”Lines 108-109
The authors focus only on a very small part of the North Pacific. The sentence should be rephrased.Line 142
Please provide a reference for the reaction rate coefficients.Line 145
How long was the spin-up time for the model?Line 255
From the figure it seems that it is April > October > July > January.Line 257
“to the at the ocean” rephraseLine 302
Is this the average of the full area? Has to be mentioned.Line 303
Is the last part really needed?Line 311
The referring to MBE is missing for the TMA comparison.Line 334
Provide a reference for the degradation in sediments.Line 342
With decreasing SST the emission increases.Section 3.2
For the reader it would be better when the discussion is structured into (i) MMA, (ii) DMA and (iii) TMA as it is for the figures. Separate figures showing the percentage changes over the ocean will help to better understand and follow the discussion.Line 364
Table 7 instead of table 6Line 395
How is the difference of HONO + OH between the simulation with and without MBE?Line 405
Add that the low agricultural emissions of DMA are related to the contribution.Line 419
In figure S2 and S3 the distribution of the residential emissions is not shown.Section 3.3.2
A discussion of DMS when the WS is reduced is missing. Why are there so strong changes in July?Section 3.3.3
When the Henry’s Law coefficient is changed also the emission flux is changed. This discussion is missing.Line 491
Because of the missing sea water data and the missing uptake on aerosol particles, it could be questionable if the last part of the sentence (“which is more consistent with the reality”) is really true. It might be true on global scale, but not necessarily on regional scale.Table 6
Please add the year of the measurement. Provide also average values.Figure 4
The figure caption describes MBE and AE contribution different as mentioned in the figure legend.Figure S2-S6
Give the name of month instead of a number and symbol.References
Bell et al., Ammonia/ammonium dissociation coefficient in seawater: A significant numerical correction, Environ. Chem., 4, 183–186, https://doi.org/10.1071/EN07032, 2007.
Bell, et al., Corrigendum to: Ammonia/ammonium dissociation coefficient in seawater: A significant numerical correction. Environ. Chem., 5, 258. https://doi.org/10.1071/en07032_co, 2008
Chang et al., Discovery of a Potent Source of Gaseous Amines in Urban China, Environ. Sci. Technol. Lett., 8, 725-731, https://doi.org/10.1021/acs.estlett.1c00229, 2021.
Ge et al., Atmospheric amines - Part I. A review, Atmos. Environ., 45, 524-546, https://doi.org/10.1016/j.atmosenv.2010.10.012, 2011a.
Ge et al., Atmospheric amines e Part II. Thermodynamic properties and gas/particle partitioning, Atmos. Environ., 45, 561-577, https://doi.org/10.1016/j.atmosenv.2010.10.013, 2011b.
Gibb et al.: Distributions and biogeochemistries of methylamines and ammonium in the Arabian Sea, Deep-Sea Res. Part II, 46, 593–615, https://doi.org/10.1016/S0967-0645(98)00119-2, 1999.
Leng et al., Temperature-Dependent Henry’s Law Constants of Atmospheric Amines, J. Phys. Chem. A, 119, 8884-8891, https://doi.org/10.1021/acs.jpca.5b05174, 2015.
Sander, R.: Compilation of Henry's law constants (version 4.0) for water as solvent, Atmos. Chem. Phys., 15, 4399–4981, https://doi.org/10.5194/acp-15-4399-2015, 2015.
van Pinxteren et al.: Aliphatic amines at the Cape Verde Atmospheric Observatory: Abundance, origins and sea-air fluxes, Atmos. Environ., 203, 183-195, https://doi.org/10.1016/j.atmosenv.2019.02.011, 2019
Yang et al.: Seasonal variation in concentration and microbial uptake of methylamines in estuarine waters, Mar. Ecol. Prog. Ser., 108, 303-312, https://doi.org/10.3354/meps108303, 1994.
Citation: https://doi.org/10.5194/acp-2022-394-RC1 -
AC1: 'Reply on RC1', Xuemei Wang, 10 Jan 2023
The comment was uploaded in the form of a supplement: https://acp.copernicus.org/preprints/acp-2022-394/acp-2022-394-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Xuemei Wang, 10 Jan 2023
-
RC2: 'Comment on acp-2022-394', Anonymous Referee #2, 05 Aug 2022
Overall impression
The study aims to estimate the gas phase concentration of three different amines using WRF-Chem. In addition to previous similar attempts using anthropogenic amine emissions only, the authors apply an online marine biogenic amine emission scheme. This new method seems sound and valid. However, the manuscript lacks of a thorough validation. (Some) Measurement results of different studies are presented in a table but never discussed in context with the simulated concentrations. Furthermore, the conclusions and discussion is not clearly formulated. For example, it is often unclear what type of aggregation / averages are referred to when presenting relative differences. This made it hard to follow the discussion and the results and likely lead to misunderstandings. Finally, potential shortcomings of the current method (e.g. loss to atmospheric aerosols is not considered) is not discussed anywhere in the paper.
Despite the fact that the manuscript could be improved in terms of language and clearer formulations at many points, the manuscript also needs major revision in the interpretation of results, discussion and conclusion. Since measured concentrations were presented, an evaluation section that compares these measurements, in particular the ones of the same year, with the modelled concentrations.
Major comments
l. 138-139: Why is loss to aerosols not considered? Since it was correctly described in the introduction that amines play a role in aerosol formation, there should be a considerable loss to aerosols. This should be discussed here and in the conclusions.
l. 145-146: It is stated that 2015 was chosen since it had ‚more consecutive days with field observations’, however, these field observations were not used in the manuscript for validation purposes. Furthermore, the authors cite e.g. Zheng et al., 2015, in this context, who performed measurements in 2012, so this study probably does not help to explain why 2015 was chosen.
l. 148-149: The mentioned verification discussion is missing in the manuscript.
l. 349-350: Weren't the simulation periods chosen to compare with observations? What were the available observations in the simulated time periods of 2015? Above it was mentioned, that daily measurements are available in certain periods. These periods were simulated. So why don’t the authors take the grid cell where the measurement took place and produce a daily mean from the simulation and compare this to the observation?
l. 351-352: Since the range of observed concentrations spans ~an order of magnitude for each of the three investigated amines, it's hard to say that simulation results are 'close'. What you could generally infer is that simulation and available observations agree in terms of order of magnitude. The increase due to MBE does not seem to be very strong seen the numbers in Table 6 to justify the statement 'inclusion of marine biogenic emissions can make up for the shortcomings of the previous models' underestimation of amines'
l. 355: The mentioned verification part is missing. In the text, there are no comparisons between measured and simulated values and no discussion of the evaluation of the simulated values against the observations shown in Table 6.
l. 364-365: Which region correspond these numbers to? I get different relative changes when using the numbers given in Table 6.
Later I see, that these numbers seem to refer to only regions with an increase shown in Table 7, right? Why do the authors present these numbers and not the domain average in the text (MA in Table 7 if I understood correctly)? In any case, it need to be made much clearer in the whole discussion sections what averages the given numbers in the text refer to. At present it is impossible to follow the discussion of results in a fair amount of time.
Conclusion section: It has to be stated that the drawn conclusions are only valid for the given months in the year 2015. What is with potential limitations since loss to atmospheric aerosol is not considered? This should be discussed. Overall none of the results of the sensitivity simulations in sections 3.3 is summarized in the conclusion. Why not? And a comparison to the available measurements is missing.
l. 484-486: The authors should summarize what they actually did. This numerical modelling to simulate gas phase amine concentration. In addition to anthropogenic emission MBE of amines was treated in an online manner.
Minor comments
Please check the manuscript with a native English speaker if possible. At many instances, some formulations were misleading or seemed at least very unfamiliar.
l. 20: ‚has been insufficiently investigated in the current emission inventory.‘ What do the author’s want to express? An emission inventory cannot investigate.
l. 60-62: Stronger and more easily oxidized than what? Likely ammonia, then please state it like this.
l. 63-65: Is the lower concentration of amines compared to ammonia really a result of faster loss processes only or at least partly due to the lower emission rates?
l. 76-80: It is referred to ‚increased concentrations’, an ‚increase by 1-2 orders of magnitude‘ and a ‚significant increase‘, but it is not mentioned compared to what concentrations have increased?
l. 89-90: Not sure if China’s location alone can be called unique. Densely populated coastal areas can be found in many places in the world. The combination of location and pollution is rather what could be called unique.
l. 95: ‚amines’ role in atmospheric chemistry‘ sounds strange. Better: ‚the role of amines in …’?
l. 101: Is ‚reasonable‘ the right word? ‚Complete‘ or ‚detailed‘ might fit better.
l. 102-103: First, is the emission mass ratio in the Mao et al., 2018, AE emission inventory really ‚arbitrary‘? Second, since this scheme is used as it is, and just an online MBE scheme was added, your study does not ‚overcome the arbitrariness‘ in the AE emission inventory. Please reformulate these sentences in a more clearer way.
l. 104: Multiple -> multiple
l. 121: This is confusing here. Which is the ‚previous method’ you are referring to?
l. 127-129: Is an ammonia / amine ratio, e.g. similar to traffic, also applied for ammonia emissions from ships? If not, why not?
l. 150: What do the authors mean with 'representative'? Do these months represent a specific season? If so, is this also the case for these months in the year 2015? As the authors state later, at least SST might be non-typical for parts of the 2015, which indicates it is not representative in general. So, please clarify what is meant with ‚representative’ in the context of this paper. I doubt that 10 days of one month can be called ‚representative‘ for the same month in a climate view or in other years.
l. 154: ‚chemical boundary‘ -> ‚chemical boundary conditions‘
l. 154-155: Did the authors run CAM-CHEM themselves to generate boundary conditions? This is what I understand from reading. However, the webpage given a few sentences later suggest that driving data for boundary conditions can be download. Please clarify.
l. 168: Why the reference to these measurements of concentrations in air and in water by Gibb et al., 1999? How exactly were these used?
l. 188: solvated -> dissolved?
l. 189-190: ‚this study selected the average of the observed values of other sea areas (Table 3)’. I don’t understand what the authors want to express here. Which numbers of Table 3 are used in the emission algorithm and how exactly? The description of your method might need more detail or clarification.
l. 198-199: Perhaps related to the previous comment… It would be helpful if the authors can give their calculated average values in Table 3. This helps the reader to much easier follow the method.
l. 208: The unit is probably is 45 g kg-1. The term ‚within‘ suggest a range but only one number is given.
l. 226-228: Was this done by the authors? If yes, please describe in more detail how the SST data was combined with the other satellite data sets mentioned here. According to the description before, the NESDIS SST data is already on 5 km horizontal resolution.
l. 235: ‚SST presents the seasonality‘ sounds unfamiliar. Please check for proper English.
l. 257: Obsolete "at the".
l. 275: ion intensity -> ionic strength?
section 2.5.4: Is time-resolved (e.g. hourly) wind speed utilized or monthly mean?
l. 293: Meaning of FNL is missing.
l. 294-295: Well, there is considerable bias for RH. This might not be important for your study, but in any case should not be stated suggestion good agreement. Most importantly is that model and observations of wind speed agree reasonably well.
l. 302: Are the numbers presented in Table 5 for one grid cell or domain mean? Or only over ocean?
l. 303-307: Why not also in table 5 change the signs, which is then consistent with the rest of the figures and the paper?
l. 308, Fig. 3: From table 5 I understand that the ocean is a source for MMA and TMA, but a sink for DMA. Why aren't therefore negative emission fluxes seen for DMA over ocean in Fig. 3?
l. 309, Fig. 4: Is that domain mean?
l. 309: ‚MBE emission‘ -> emission to much, since the E in MBE already refers to emission.
l. 310, 311: TMA -> ‚MBE flux of TMA‘ or similar.
l. 309-314: Why no discussion of DMA emission flux, but only for MMA and TMA?
l. 315-335: Just as a suggestion: These paragraphs might be more useful directly after section 2.5, before section 3.1.
l. 315, Fig. 5: Which values were used for the other three input variables, while only one is varied? I understand that the four plots do not show the contribution, but the variation of the emission flux due to the variation of one variable while the others were held constant. It is important for the interpretation and the magnitude of the contribution to provide a clearer description here.
Please also clarify which area or aggregation Fig. 5 refers to. Is that temporal domain mean or ocean mean or one grid cell?
l. 316: ‚Kg, which is directly proportional to Kg’ -> Probably the latter Kg is meant to be something else.
l. 325, 338, 344, 346 and later occurrences: overflow -> emission, release or similar. Check with native speaker.
l. 347: 'affecting the variation of the exchange fluxes' would be more precise in my view.
l. 349, 354: What is the temporal and spatial resolution of the model? This should be mentioned in section 2.3 where the model is described.
l. 358-360: Future tense ('will') does not sound correct.
l. 360-361: No significant change in autumn and winter seems to be the case for MMA and DMA only, according to Table 6.
l. 381 and other occurrences of ‚offshore‘: Shouldn't ‚offshore‘ be on the open ocean, at least farer away from the coast on the ocean? Why is land mass with 500m above sea level considered as ‚offshore’?
l. 393: Formulation sounds incorrect. Perhaps better 'was lowered / decreases by less than 10%'.
l. 396-398: Can that be shown with the model results, e.g. NO, OH concentration maps?
l. 400: What do the authors mean with ‚model mechanism‘?
l. 419-420: ‚they have relatively high fluxes’: What is the relative contribution of residential emissions in the mentioned areas? It seems that this is rather uniform and hence the relative change is more uniform (hence the relative change not as 'obvious' as you say).
l. 420-421: ‚the difference‘ -> If you refer to Fig. S2 and S3, the authors probably mean ‚the relative difference‘.
l. 423: What is high value and low value area?
l. 445-446: Shouldn't the change of pH due to Chla change directly affect the sea air exchange of DMA as well? What do you mean with 'the change of MBE of DMA is not considered'?
l. 447: 'range of DMA' -> 'range of DMA change'
l. 449-450: ‚therefore, DMA reacts with ·OH faster than the other two amines in the upwind direction, and its concentration is less affected.‘ Can the authors please explain in more detail what they mean here?
l. 466-467: ‚going below 50 mm in October’: What region does that refer to? Also in July many places in the domain have precipitation amounts < 50 mm.
l. 475: ‚DMA is only affected by terrestrial AE‘: Didn't it have negative emission fluxes over sea water, i.e. is lost into the ocean surface (Fig. 4)?
l. 476: variation range of its concentration -> variation range of its concentration change
l. 490: It's not the satellite data itself, but its application in an online emission scheme, that might reflect the emission situation.
l. 494-495: Is that 500km away from the coast on land or 500km away over the ocean?
l. 495-496: These are the maximum increases for TMA due to MBE, aren't they? Why don’t authors instead mention the average increases in the text as was done for MMA. I believe that more readers would be interested in averages than in the extremes? Anyhow, it always must be clear what the given numbers are. This is missing throughout the manuscript in many places and makes it rather poor.
l. 499: ‚WS and Chla were found to be the dominant factors affecting MBE fluxes’: This is the case for all amines, right? Then please state it like this.
l. 507: ‚and the ocean also transforms from a source of amines to a sink’: Was that observed in the simulations, since such result was not presented in section 3.3.2? And was it the case for all amines or only some? Since the Chla concentrations, according to Fig. 2, were mostly < 10 mg m-3, 50% increase should only make it a sink for DMA, and very slightly MMA perhaps.
Table 6: Some observation results are presented in µg m-3 and ng m-3. These should be transformed to pptv. Why are the Gao et al. 2022 observations of TMA much larger than for the other studies?
Fig. 2: The numbers in the plots: Black for SST, Chla, SSS and white color for WS would probably give better readability.
Fig. 4: Labels are wrong. In the figure, dashed is MBE and solid lines are AE.
Comments to the author response to the quick initial review comments
In the response, changes in lines 362, 418, 464 are mentioned (with ‚descriptions and data sources’), but I can’t find these changes.
The comparison to measurements can of course be done only in a qualitative manner, but it should at least be discussed in the manuscript and not only presented in a table (Table 6).
Citation: https://doi.org/10.5194/acp-2022-394-RC2 -
AC2: 'Reply on RC2', Xuemei Wang, 10 Jan 2023
The comment was uploaded in the form of a supplement: https://acp.copernicus.org/preprints/acp-2022-394/acp-2022-394-AC2-supplement.pdf
-
AC3: 'Reply on AC2', Xuemei Wang, 10 Jan 2023
<strong>Publisher’s note: this comment is a copy of AC2 and its content was therefore removed.</strong>
Citation: https://doi.org/10.5194/acp-2022-394-AC3 -
AC4: 'Reply on AC2', Xuemei Wang, 10 Jan 2023
<strong>Publisher’s note: this comment is a copy of AC2 and its content was therefore removed.</strong>
Citation: https://doi.org/10.5194/acp-2022-394-AC4
-
AC3: 'Reply on AC2', Xuemei Wang, 10 Jan 2023
-
AC2: 'Reply on RC2', Xuemei Wang, 10 Jan 2023
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 691 | 214 | 33 | 938 | 76 | 19 | 22 |
- HTML: 691
- PDF: 214
- XML: 33
- Total: 938
- Supplement: 76
- BibTeX: 19
- EndNote: 22
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1