the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Natural and anthropogenic influence on tropospheric ozone variability over the Tropical Atlantic unveiled by satellite and in situ observations
Abstract. Tropospheric ozone over the South and Tropical Atlantic plays an important role in the photochemistry and energy budget of the atmosphere. In this remote region, tropospheric ozone estimates from reanalysis datasets show the largest discrepancies. The present study characterises the vertical and horizontal distribution of tropospheric ozone over the South and Tropical Atlantic during February 2017 using a multispectral satellite approach called IASI+GOME2 and in situ airborne measurements from the Atmospheric Tomography Mission (ATom). These observations are compared with three global chemistry reanalysis products: the Copernicus Atmosphere Monitoring Service reanalysis (CAMS reanalysis), the Tropospheric Chemistry Reanalysis version 2 (TCR-2), and the second Modern-Era Retrospective Analysis for Research and Applications (MERRA-2). The CO-enriched air masses from Western and Central Africa are lifted into the middle and upper troposphere over the ocean by strong upward motions. In the descending branches of the Hadley cells over the Southern Atlantic, stratospheric intrusions are observed. Air masses in the Southern Hemisphere are influenced by biomass burning sources from Central and Eastern Africa and lightning, as well as downdrafts from the stratosphere. According to in situ measurements of chemical tracers, tropospheric ozone attributed to biomass burning emissions of ozone precursors is approximately 13 ppb (~17 %) over 7 km (25° S–5° N) and approximately 38 ppb (~50 %) over 3 km (25° S–15° S). The intercomparison suggests a significant overestimation of three chemistry reanalysis products of lowermost troposphere ozone over the Atlantic in the Northern Hemisphere because of the overestimations of ozone precursors from anthropogenic sources from North America.
- Preprint
(2649 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-3758', Anonymous Referee #1, 15 Jan 2025
In “Natural and anthropogenic influence on tropospheric ozone variability over the Tropical Atlantic unveiled by satellite and in situ observations” Okamoto et al. discuss ozone levels over the Atlantic Ocean based on satellite observations, reanalysis products and in situ observations from the ATom 2 campaign in February 2017. The authors show that ozone can be attributed to various different sources in the studied region including biomass burning, urban influences and stratospheric intrusion. They report an overprediction of ozone in lower altitudes in all investigated reanalysis products.
This is an interesting study. However, I have several questions and comments, which need to be addressed before I can recommend this paper for publication. The manuscript is very long and contains too many figures. I recommend creating a Supplement and moving some figures and explanations there. Overall, the manuscript would benefit from being more concise. I further suggest including the other three ATom campaigns in this study. The authors investigate seasonality and as the ATom deployments took place in all seasons and all cover the studied region, the in situ data will be a valuable addition. While the authors present the differences in the reanalysis products for only two days in February 2017, the comparison with further in situ data could help to draw general conclusions on the performance of these products. Please find my detailed comments in the following.
Lines 35 f.: Stratospheric intrusion only contributes a small fraction to the overall tropospheric ozone (Lelieveld & Dentener, 2000, doi: 10.1029/1999JD901011)
Line 41 f.: biogenic sources? – There seems to be a word missing. I further recommend briefly discussing the formation mechanism of ozone from these precursors here (or in the previous paragraph).
Lines 45: MOZAIC?
Line 48: Several studies have shown that lightning is the dominant source of ozone in the upper troposphere, e.g. Schumann & Huntrieser, 2007 (doi: 10.5194/acp-7-3823-2007) Nussbaumer et al., 2023 (doi:10.5194/acp-23-12651-2023). Please consider citing this literature.
Line 54: Over the tropics, 9-13km is still tropospheric. The UTLS region only applies to Southern or Northern Extratropical Latitudes.
Line 57: What do the authors mean by “the observational gap of air pollution” and how does it relate to the tropical Atlantic?
Line 69: What “major outbreaks” are the authors referring to?
Lines 181 f.: Please briefly describe the measurements / instruments including uncertainties and detection limits of the trace gases used in this analysis.
Lines 187 f.: “(…) they use a pair of HCN biomass of a burning tracer and C2Cl4 of an urban tracer.” This sentence is difficult to read, I recommend rephrasing.
Line 190: The conditions for polluted and aged air are the same. I assume the first “<” should be “>” instead? Is well-mixed and aged air equivalent to clean / unpolluted air? What’s the lifetime of HCN and C2Cl4 in the troposphere?
Line 195: Are the lifetimes of X and CO similar? If not, does this introduce errors in this method due to different transport ranges?
Line 198: I recommend adding the definition of the emission ratio here as well.
Line 202: Is this equation used for all measurements and if yes, why were the air masses classified into the four categories before? Or is it only used to distinguish the sources of ozone in air masses defined as “mixed pollution air”?
Figure 2: This manuscript has many Figures. I recommend choosing one or two months (for example February and August) and showing the lightning intensity and the fire radiation power in one Figure. The remaining panels can be shown in a Supplement. This makes the comparison of the location of maximum lightning and maximum fire activity easier and reduces the number of Figures.
Line 291: Do the coordinates describe a box and therefore also include continental regions? If yes, does it maybe make more sense to only consider the maritime regions? It could help to add an outline of the discussed area in the Supplement or as a subpanel in Figure 3.
Figure 3: Is the SD by itself really meaningful in regard to the uncertainty of the reanalysis products / satellite observations? Maybe it would make more sense to look at the SD relative to the O3 measurement (SD/O3value*100%). I also recommend adding the mean monthly O3 in the Figure.
Line 332: How was this altitude range chosen? The chemical composition at 6km vs 12km is very different. While upper tropospheric ozone is strongly impacted by deep convective updraft as well as lightning activity, the free troposphere is rather decoupled from the surface and is not impacted by lightning. I recommend looking at the free troposphere, e.g. 3 - 10km, and the upper troposphere, e.g. >10 km, separately.
Section 3.2: This section makes me wonder why the authors limit themselves to one of the ATom deployments. The four deployments cover the same area during all seasons and could provide some more insights into the discrepancies with the reanalysis products.
Figure 6 + 7: These Figures (including their description in the text) are a repetition of Figures 4 and 5 and I recommend deleting them or moving them to the Supplement.
Lines 400 ff.: It is difficult to see the differences and similarities between the in situ observations and the satellite / reanalysis products just by the color. Maybe instead (or additionally) a vertical profile with averaged ozone could be used (Altitude as y-axis and Ozone mixing ratio as x-axis). Or an additional column could be added to the Figure showing the difference between the ATom2 data and the other products in each panel.
Line 407: “in the” is repeated.
Table 2: Why was only the latitude band between 10°N and 30°N investigated?
Figure 9: Similar to O3, it is difficult to rank the agreement between the ATom2 data and the other products just by color. I recommend using a more quantitative method.
Line 466: “not depicted by any”?
Figure 10: See comments above.
Lines 482: In the upper troposphere, lightning NOx is the most important source of O3. Could it make more sense to define lightning as an individual category?
Line 503: Can you briefly explain how “the probability of boundary layer influences” is obtained?
Lines 523: The results for CAMS and MERRA-2 could be added to the Supplement.
Figure 14: How was this plot generated? Some of the colored points were previously categorized as e.g. urban influence and not as mixed pollution. Please clarify this. What are the gray data points? Maybe instead of the absolute values, the relative contribution to the overall O3 could be shown.
Figure 15: I recommend moving this Figure to the Supplement.
Line 637: This hypothesis could be tested by calculating ozone loss terms along the ATom2 flight track.
Lines 657 f.: This sounds like the reanalysis products only show overestimations for urban air masses? I understood earlier that the downward motions in the Hadley cell are overestimated in the reanalysis products leading to an overestimation in ozone. Could you please clarify this?
Citation: https://doi.org/10.5194/egusphere-2024-3758-RC1 - CC1: 'Comment on egusphere-2024-3758', Owen Cooper, 04 Feb 2025
- CC2: 'Comment on egusphere-2024-3758', Debra E. Kollonige, 06 Feb 2025
-
CC3: 'Comment on egusphere-2024-3758', Kenneth Pickering, 06 Feb 2025
Comments on Okamoto et al. focusing on the models used in data assimilation systems and treatment of lightning NOx production:
Major comments:
I do not think that the authors should have included MERRA-2 in this analysis. There is no detailed tropospheric ozone photochemistry included in the model -- only a simple month-dependent O3 parameterization (Stanjer et al., 2008, JGR) derived from a 2-D chemistry model that was originally developed for the stratosphere. Monthly mean O3 production and loss terms from the 2-D model are utilized in the GEOS-5 model, and these terms are not really useful below the upper troposphere. As a result, O3 in the lower and middle troposphere is not well represented in the assimilation, especially with regard to longitudinal variations in O3. The MERRA-2 developers do not recommend use of the O3 values below 500 hPa.
Section 2.2 on the Atmospheric Chemistry Reanalyses: The descriptions of the three reanalyses do not say how lightning flashes are predicted in the models or say how lightning NOx (LNOx) emissions are treated (NOx production per flash and how those emissions are distributed vertically in the model). Therefore, the reader does not know how realistic the distribution of LNOx emissions is represented in the model. This horizontal and vertical distribution will have a large impact on the source attribution results for O3. LNOx is not included at all in MERRA-2, significantly affecting the longitudinal distribution of O3 especially in the upper troposphere. Assimilation of OMI and MLS O3 will partially correct for this deficiency, but not totally. Lack of LNOx leads to a small low bias in UT O3 in MERRA-2.
Specific comments:
Abstract line 22-23: most of the lift occurs over the continents, not the ocean
Figure 1: shows 2017 maps of the World-Wide Lightning Detection Network (WWLLN) Global Lightning Climatology (WGLC) from Kaplan and Lau (2021, Earth Sys. Sci. Data). Much greater maximum amounts of lightning are shown over Central and South America than over Africa. The satellite (OTD/LIS) climatology (Cecil et al., 2014, Atmos. Res.) is much different, showing the strongest maximum over Central Africa. WWLLN detection efficiency over Africa is very low. I'm wondering if enough attention has been given to this when the WGLC was created. Uncertainty in use of this climatology should be mentioned in the paper.
Figures 4, 6, and others: MERRA-2 O3 plots for 0-3 km should not be shown here – O3 is obviously wrong in the lower troposphere
Figures 5 and 7: MERRA-2 O3 plots for a specific day should not be trusted and therefore not shown. The chemistry in the model is based on monthly mean O3 production and loss rates.
Citation: https://doi.org/10.5194/egusphere-2024-3758-CC3 -
CC4: 'Comment on egusphere-2024-3758', Anne Thompson, 07 Feb 2025
6 Feb. 2025, Comments on Okamoto et al Paper
The paper as a whole follows a reasonable outline and is well-organized but it begins with serious omissions that must be remedied. It is easy to forget the pioneering satellite work or major experiments that paved the way for ATom or HIPPO (Wofsy: https://royalsocietypublishing.org/doi/10.1098/rsta.2010.0313 ), a similar experiment that preceded the ATom Atlantic and Pacific transects. The SAFARI-92/TRACE-A sampling over the Atlantic and similar DC-8 campaigns over the Pacific provided baseline data and the insights into tropical chemistry and dynamics that remain relevant.
Lines 40 ff. Context with prior published work. The south and tropical Atlantic is NOT one of the regions “with the sparest coverage of in-situ observations!” On the contrary, south Atlantic ozone structure is well-characterized and has been studied in detail since the early 1990s. Before that a classic paper by Jack Fishman on satellite data from TOMS (https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/JD092iD06p06627) called attention to the high tropospheric ozone over the south Atlantic. That was the beginning of “Tropical Atlantic” studies on ozone. A number of satellite-based products have focused on the tropical south Atlantic, e.g., Fishman et al., JGR, 1996; Thompson et al., GRL, 2000; Hudson et al., JGR, 1995; Leventidou et al., ACP, 2017.
Just as important, pioneering international aircraft campaigns led by the US, France and Germany probed causes of the “tropical south Atlantic ozone maximum” with flights from Brazil, western and southern Africa and from Brazil. The early aircraft campaigns were:
- TROPOZ II – led by A. Marenco from Toulouse!! For example: https://www.sciencedirect.com/science/article/abs/pii/S1352231099005087 by Gouget. Also - https://www.semanticscholar.org/paper/Study-of-ozone-formation-and-transatlantic-from-the-Jonqui%C3%A8res-Marenco/fcdb685886daa73cd211fb1e3e740d5138d2c02a
- Early IGAC “Projects” on biomass burning and south tropical Atlantic (BIBEX, STARE). In September-October 1992 a massive umbrella experiment “STARE” was conducted during the southern African burning season: IGAC/STARE/SAFARI-92/TRACE A (International Global Atmospheric Chemistry/South Tropical Atlantic Regional Experiment/Southern African Fire Atmospheric Research Initiative/Transport and Atmospheric Chemistry near the Equator-Atlantic) with multiple aircraft, including NASA’s DC-8 with comprehensive lidar ozone measurements, in-situ profiling of ozone precursors like NO/NO2, CO, hydrocarbons; SAFARI-92 operated a DC-3 over southern Africa. The aircraft measurements were augmented by balloon ozonesonde profiles from Brazil, Ascension Island, Brazzaville, Congo; Irene, South Africa. There are ~65 papers from the 1992 campaigns in Journal of Geophysical Research, published 1 Oct. 1996!
An outgrowth of the SAFARI-92/TRACE-A balloon data was the initiation of the SHADOZ (Southern Hemisphere ADditional Ozonesondes) network of ozone sounding stations that has operated from 1998 to the present day (Thompson et al., 2003; Thompson et al., 2017). SHADOZ provides ozone profiles 2-4 times/month from Ascension Island (central south tropical Atlantic, 8S, 15W), Natal, Brazil (6S, 35W) and Paramaribo (5N, 55W). The ozone profiles from these 3 stations since 1998 number more than 2000! Comment on Figs 9 and 10. Comparisons of satellite and model ozone with April 2017 SHADOZ sondes should be carried out for comparison to the ATOM comparisons.
The earlier work must be cited. One way to revise the paragraph starting on Line 40 …
The South and Tropical Atlantic has been a region of intense interest in the ozone community since a regional maximum in tropospheric ozone derived from satellite measurements was identified by Fishman and Larson (1987). This discovery was the motivation for a large-scale ground and aircraft study in the southern biomass burning season in September and October 1992: IGAC/STARE/SAFARI-92/TRACE A (International Global Atmospheric Chemistry/South Tropical Atlantic Regional Experiment/Southern African Fire Atmospheric Research Initiative/ Transport and Atmospheric Chemistry near the Equator-Atlantic). SAFARI-92/TRACE A confirmed the regional ozone feature with aircraft profiling and lidar plus ozonesondes deployed over Brazil, Ascension Island and 3 sites in subSaharan Africa. In addition, analyses of the comprehensive SAFARI-92/TRACE-A data confirmed links of the Atlantic maximum to fire activity over Africa (Fishman et al., 1996; Thompson et al., 1996) and to ozone formed from a combination of fires, deep convection and lightning activity over South America (Pickering et al., 1996). Based on the ozonesonde profiles, it was estimated that the relative contributions to the Atlantic ozone were approximately 2/3 from African sources and 1/3 from South America (Thompson et al., 1996). However, dynamical influences were required for the ozone feature to form. Krishnamurti et al. (1996) demonstrated that recirculation within the south Atlantic gyre allowed the ozone to accumulate so that the highest ozone amounts were over the ocean rather than the continents.
Shipboard ozone sampling over the tropical Atlantic provided additional insights into south tropical Atlantic ozone (Weller et al., 1996; Thompson et al., 2000). The ozone maximum occurred at all seasons, not only during the peak of southern hemisphere burning but also when African fire activity was at its greatest north of the ITCZ. This so-called “Atlantic ozone paradox” was associated with upper tropospheric-stratospheric subsidence and lightning in addition to fires (Thompson et al., 2000). These contributions were evaluated in an early model study (Moxim and Levy, 2000). The SAFARI-92/TRACE-A experiments were instrumental in assembling the SHADOZ (Southern Hemisphere Additional Ozonesondes; https://tropogsfc.nasa.gov/shadoz) network of stations that has operated from 1998 to the present day (Thompson et al. 2017). With coordinated launches of ozonesondes from more than 10 stations across the tropics, the Atlantic maximum is a strong feature with the south tropical Atlantic always exhibiting more tropospheric column ozone (5-15 Dobson Units,1 DU = 2.69x1016 cm-2). When looking at tropospheric ozone structure across the entire tropical band, the Atlantic ozone feature leads to a zonal wave-one pattern (Thompson et al., 2003).
More recently, the role of biomass burning (van der Werf et al., 2017), biogenic (Sindelarova et al., 2022) and lightning (Schumann and Huntreiser, 2007) contributions to Atlantic, African and South America ozone has been investigated. In-Service Aircraft….
REFERENCES TO ADD:
Fishman, J. and Larsen, J. (1987) Distribution of total ozone and stratospheric ozone in the tropics: Implications for the distribution of tropospheric ozone, J. Geophys. Res., 92 (D6), 6627-6634
Fishman, J., Watson, C. E., Larsen, J. C., Logan, J. A. (1990) Distribution of tropospheric ozone determined from satellite data, J. Geophys. Res, 95 (D4), 3599-3617
Fishman, J., Hoell, J. M., Jr., Bendura, R. D., McNeal, R. J., Kirchhoff, V. W. J. H. (1996) NASA GTE TRACE A experiment (September–October 1992): Overview, J. Geophys. Res., https://doi.org/10.1029/96JD00123
Hudson, R. D., Kim, J., Thompson, A. M. (1995) On the derivation of tropospheric column ozone from radiances measured by the total ozone mapping spectrometer, J. Geophys. Res. 100, 11137-11145;
https://doi.org/10.1029/94JD02435
Krishnamurthi, T. N., Sinha, M. C., Kanamitsu, M., Oosterhof, D., Fuelberg, H., Chatfield, R., Jacob, D. J., & Logan, J. (1996). Passive tracer transport relevant to the TRACE-A experiment. Journal of Geophysical Research-Atmospheres, 101(D19), 23889-23907
Leventidou, E. Weber, M., Eichmann, K-U., Burrows, J. P., Heue, K-P., Thompson, A. M., Johnson, B. (2017) Harmonisation and trends of 20-years tropical tropospheric ozone data, Atmos. Chem. Phys., https://doic.org/10.5194/acp-2017-815.
Moxim, W. J., and Levy II, H. (2000) A model analysis of the tropical South Atlantic Ocean tropospheric ozone maximum: The interaction of transport and chemistry J. Geophys. Res., https://doi.org/10.1029/2000JD900175
Pickering, K. E., Thompson, A. M., Pickering, K. E., Wang, Y., Tao, W-K., McNamara, D. P., Kirchhoff, V. W. J. H., Heikes, B. G., Sachse, G. W., Bradshaw, J. W. D., Gregory, G. L, Blake, D. R. (1996) Convective transport of biomass burning emissions over Brazil during TRACE-A, J. Geophys. Res., https://doi.org/10.1029/96JD00346
Thompson, A. M., Pickering, K. E., McNamara, D. P., Schoeberl, M. R., Hudson, R. D., Kim J. W., Browell, E. V., Kirchhoff, V. W. J. H., Nganga, D. (1996) Where did tropospheric ozone over southern Africa and the tropical Atlantic come from in October 1992? Insights from TOMS, GTE/TRACE-A and SAFARI-92, J. Geophys. Res., 101, 24,251-24,278; https://doi.org/10.1029/96JD01463
Thompson, A. M., Doddridge, B. G., Witte, J. C., Hudson, R. D., Luke, W. T., Johnson, J. E., Johnson, B. J., Oltmans, S. J., R. Weller, R. (2000), A tropical Atlantic paradox: Shipboard and satellite views of a tropospheric ozone maximum and wave-one in January-February 1999, Geophys. Res. Lett., https://doi.org/10.1029/1999GL011273
Thompson, A. M., J. C. Witte, S. J. Oltmans, F. J. Schmidlin, J. A. Logan, M. Fujiwara, V. W. J. H. Kirchhoff, F. Posny, G. J. R. Coetzee, B. Hoegger, S. Kawakami, T. Ogawa, J. P. F. Fortuin, H. M. Kelder (2003) Southern Hemisphere ADditional Ozonesondes (SHADOZ) 1998-2000 tropical ozone climatology. 2. Tropospheric Variability and the Zonal Wave-One, J. Geophys. Res., 108, 8241, doi: 10.1029/2002JD002241
Thompson, A. M., Witte, J. C., Sterling, C., Jordan, A. Johnson, B. J., Oltmans, S. J., Fujiwara, M., Vömel, M., Allaart, M., Piters, A., Coetzee, J. G. R., Posny, F., Corrales, E., Andres Diaz, J., Félix, C., Komala, N., Lai, Maata, M., Mani, F., Zainal, Z., Ogino, S-Y., Paredes, F., Luiz Bezerra Penha, T., . Raimundo da Silva, F., Sallons-Mitro, S., Selkirk, H. B., Schmidlin, F. J., Stuebi, R., Thiongo, K. (2017) First reprocessing of Southern Hemisphere Additional Ozonesondes (SHADOZ) Ozone Profiles (1998-2016). 2. Comparisons with satellites and ground-based instruments, J. Geophys. Res., 122, 13000-13025, doi: 10.1002/2017JD027406
Weller, R., Lilischkis, R., Schrems, O., Neuber, R., Wessel, S. (1996) Vertical ozone distribution in the marine atmosphere over the central Atlantic Ocean (56°S-50°N), J. Geophys. Res., 101, 1387–1399; https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/95JD02838
Citation: https://doi.org/10.5194/egusphere-2024-3758-CC4 -
EC1: 'Comment on egusphere-2024-3758', Farahnaz Khosrawi, 14 Feb 2025
I would like to thank referee 1 and the four community members - Owen Cooper, Debra Kollonige, Kenneth Pickering, and Anne Thompson - for posting their comments in the discussion. Since the community members are all experts in the field and provided valuable feedback and suggestions for improvement on the manuscript by Okamoto et al. I, after consulting with an executive editor, decided that these four comments together can be counted as a substitute for a second referee comment. Thus, the discussion can be closed now without waiting for additional referee reports.
Citation: https://doi.org/10.5194/egusphere-2024-3758-EC1
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
255 | 59 | 16 | 330 | 6 | 7 |
- HTML: 255
- PDF: 59
- XML: 16
- Total: 330
- BibTeX: 6
- EndNote: 7
Viewed (geographical distribution)
Country | # | Views | % |
---|---|---|---|
United States of America | 1 | 123 | 43 |
China | 2 | 28 | 9 |
France | 3 | 25 | 8 |
Germany | 4 | 19 | 6 |
Japan | 5 | 14 | 4 |
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1
- 123