the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Oct 2024
Drivers of change in Peak Season Surface Ozone Concentrations and Impacts on Human Health over the Historical Period (1850–2014)
Abstract. Elevated concentrations of ozone at the surface can lead to poor air quality and increased risks to human health. There have been large increases in surface ozone over the historical period associated with socio-economic development. Here the change in peak season ozone (OSDMA8) is estimated for the first time using hourly surface ozone output from 3 CMIP6 models over the 1850 to 2014 period. Additional results are obtained from one model to quantify the impact from different drivers of ozone formation, including anthropogenic emissions of ozone and aerosol precursors, stratospheric ozone and climate change. The peak season ozone concentrations are used to calculate the risk to human health, in terms of the attributable fraction metric (the percentage of deaths from COPD associated with long-term exposure to elevated ozone concentrations). OSDMA8 concentrations are simulated to more than double across northern mid-latitude regions over the historical period, mainly driven by increases in anthropogenic emissions of NOX and global CH4 concentrations. Small contributions are made from changes in other anthropogenic precursor emissions (CO and non-CH4 VOCs), aerosols, stratospheric ozone and climate change. The proportion of the global population exposed to OSDMA8 concentrations above the theoretical minimum risk exposure level (32.4 ppb), increased from <20 % in 1855 to >90 % in 2010. This has also increased the risk to human health mortality due to COPD from long-term ozone exposure by up to 20 % across Northern Hemisphere regions in the present day. Like for OSDMA8 concentrations, the drivers of the increase in the ozone health risks are attributed mainly to changes in NOX and global CH4. Fixing anthropogenic NOX emissions at 1850 values can eliminate the risk to human health from long-term ozone exposure in the near present-day period. Understanding the historical drivers of ozone concentrations and their risk to human health can help to inform the dvelopment of future pathways that reduce this risk.
- Preprint
(2082 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 20 Dec 2024)
-
RC1: 'Comment on egusphere-2024-2732', Anonymous Referee #1, 21 Nov 2024
reply
A review of egusphere-2024-2732 by Steven Turnock et al.
General Comment:
This paper argues that increases in NOx emissions and CH4 concentrations have played a major role in the historical changes in surface ozone concentrations. Although many previous studies have already shown that these factors have a significant impact on the historical changes in surface ozone, there have been few studies that have analyzed the latest CMIP6 (AerChemMIP) data from this perspective, so to some extent this result is novel. In addition, the authors analyze the health effects of surface ozone changes. Although these results appear to be novel in the same sense, the interpretation of the results is very simplistic and further analysis and discussion of the results are desirable for publication in the journal. I would like the authors to refer to the following comments for revisions.
Major Comment:
The contribution of each driver to historical changes in OSDMA8 is quantitatively estimated in Section 3.2.1. Is it possible to quantitatively estimate the contribution of each driver to health impact in a similar way? With the current descriptions in the manuscript, it is difficult to consistently compare the contribution of each driver to OSDMA8 and its contribution to health impact. If such a comparison could be made, the difference between the impact of each driver on ozone concentration changes and the health impact would be visualized, the factors behind these differences could be discussed further, and suggestions for future countermeasures would become even more meaningful.
Specific Comments:
- Table1: It would be better to more clearly explain what is fixed in each sensitivity scenario. Methane is well described in the caption already, but the other drivers are not clearly explained in the table caption.
- L128: Do all UKESM1 sensitivity experiments use the same initial data?
- L129: heath -> health
- L145-147: The bias correction method for the baseline period should be explained in more detail here. Which was used for the correction: the difference or ratio between RAMP and each model? Did the method correct the 1-hour values and then calculate OSDMA8 or correct the OSDMA8 values directly, etc.?
- L175-177: A brief description is desirable here on the ozone response derived with the method of Wild et al. (2012) using the difference between the equilibrium and prescribed CH4 concentration. Whether the ozone increase or decrease? Are there any regional or temporal characteristics?
- L227: concentration -> remove
- L240-L242: It would be an overstatement that the CMIP6 models have an ability to simulate long-term change in surface ozone based on a comparison at only five remote locations.
- L292: The AF value in Greenland exceeds 10%. Why does it happen where the AF value is quite low at other high latitude regions in the Northern Hemisphere?
- Figure3: Since the average of the three models is mainly discussed in the manuscript, so the average value should also be included in the figure.
- L299: The number in brackets (e.g. 37%) needs an explanation.
- L312-L313: In this experiment, the CH4 concentration is set uniformly within the model domain, but the actual CH4 concentration has a relatively clear difference between the Northern and Southern Hemispheres. How do you think setting the uniform CH4 concentration affects the change in ground-level ozone concentration?
- L348: including -> remove
- L352: Do all models include online calculation of BVOC emissions?
- L354-L355: The UKESM1-0-LL has a smaller ozone sensitivity per unit temperature change (ppb/K) than other models (Zanis et al. 2022), so I guess it is possible that the impact of climate change on OSDMA8 is underestimated in UKESM1-0-LL model. Further discussion on it is desirable here.
- L363: According to the manuscript the sum of individual driver impact on historical OSDMA8 change is 20.4 ppb (8.6+1.5+5.9+0.8+0.8+2.8), and the historical change in OSDMA8 in histSST experiment is 12 ppb (described in L236). I couldn't understand how this number (20% larger) was calculated from these values.
- L377: concs?
- Figure A1: Southern Sub-Saharan Africa is in the wrong colour on the map.
Citation: https://doi.org/10.5194/egusphere-2024-2732-RC1 -
RC2: 'Comment on egusphere-2024-2732', Anonymous Referee #2, 02 Dec 2024
reply
General Comment:
This study provides a valuable analysis of historical surface ozone trends and their health impacts using CMIP6 data, which is particularly significant given the limited number of studies attributing global ozone trends to various drivers using this dataset. Notably, the results align well with findings from previous studies, supporting the potential applicability of CMIP6 models for ozone trend attribution. This consistency reinforces confidence in using CMIP6 data for understanding long-term changes in ozone and their underlying drivers.
Overall, the manuscript topic is highly relevant for ACP, and the results provide novel and insightful contributions. However, the paper could benefit from more in-depth analysis and discussion in many areas. A more detailed exploration of the methodological uncertainties and limitations would strengthen the robustness of the conclusions. Additionally, the discussion could expand on how these findings compare quantitatively to previous studies and how CMIP6 models might improve upon or differ from prior modeling frameworks.
Specific Comments:
- L147: Figure 1 provides a comprehensive visualization of the model-simulated OSDMA8 values and their differences from the RAMP dataset. However, I would suggest including an additional subfigure to explicitly illustrate the spatial distribution of the RAMP dataset. This would provide a clearer and more intuitive comparison between the model results and the observed dataset, helping readers better understand the spatial discrepancies.
- L155: As mentioned, four different bias correction techniques are available: mean bias, relative bias, delta correction, and quantile mapping. The study chose to apply only the delta correction method. It would strengthen the paper to clarify why delta correction was preferred. Were the other three methods unsuitable for this study’s context? Did the correction results from the other methods perform less effectively compared to delta correction? Clarifying this methodological choice would improve the transparency and robustness of the study.
- L163: After applying bias correction to the histSST experiment, how were the results quantitatively compared with other sensitivity experiments to isolate and measure the specific contributions of individual drivers? A more detailed explanation of the comparison process would be needed.
- L164: UKESM1-0-LL processes CH4 inputs differently from the other models. A more detailed comparison of the differences among these models would greatly enhance the analysis.
- L187: The stated global increase factors for NOx (>11 times), NMVOCs, CO, and CH4 (>2) since 1850 appear inconsistent with the trends shown in Figure 2. Further clarification is required to reconcile this discrepancy.
- L235: Were changes in OSDMA8 concentrations calculated as the absolute difference between 2010 and 1855, or as a trend over 1855–2010 (e.g., 12 ± 2.6 ppb per decade (50% per decade))?
- L261: Further details are needed on the differences between the models, particularly in areas where these differences significantly impact ozone formation.
- L304: Many studies have reported significant ozone changes in these areas due to anthropogenic emissions (Wang et al., 2022; Zhang et al., 2016;…). It would be valuable to include a comparison with these existing findings.
- L310: CH4 is also classified as a VOC, but its longer chemical lifetime allows it to contribute more significantly to ozone formation over larger spatial and temporal scales compared to other VOCs. Could you clarify why it leads to a stronger ozone enhancement effect than other VOCs?
- L321: After the signing of the Montreal Protocol, indicating a leveling off of stratospheric ozone decline, Figure 2 shows a significant increase in total column ozone since 2000. How was it determined that stratospheric ozone continued to decline during 2005–2014?
- L345/L332: Please specify whether the percentage change is positive (+) or negative (-).
- L357: Why are only the attribution results from UKESM1-0-LL presented here? Are the results from the other two models consistent? Please provide an explanation. Additionally, since these models are all online-coupled, this approach does not eliminate feedback effects between various factors and meteorology. How should the influence of these feedback effects be accounted for in the interpretation of the results?
References
Wang, H., Lu, X., Jacob, D. J., Cooper, O. R., Chang, K.-L., Li, K., Gao, M., Liu, Y., Sheng, B., Wu, K., Wu, T., Zhang, J., Sauvage, B., Nédélec, P., Blot, R., and Fan, S.: Global tropospheric ozone trends, attributions, and radiative impacts in 1995–2017: an integrated analysis using aircraft (IAGOS) observations, ozonesonde, and multi-decadal chemical model simulations, Atmos. Chem. Phys., 22, 13753–13782, https://doi.org/10.5194/acp-22-13753-2022, 2022.
Zhang, Y., Cooper, O. R., Gaudel, A., Thompson, A. M., Nédélec, P., Ogino, S.-Y., and West, J. J.: Tropospheric ozone change from 1980 to 2010 dominated by equatorward redistribution of emissions, Nat. Geosci., 9, 875–879, https://doi.org/10.1038/ngeo2827, 2016.
Citation: https://doi.org/10.5194/egusphere-2024-2732-RC2 -
CC1: 'Comment on egusphere-2024-2732', Owen Cooper, 19 Dec 2024
reply
This comment can be found in the attached pdf
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 211 | 67 | 20 | 298 | 2 | 3 |
- HTML: 211
- PDF: 67
- XML: 20
- Total: 298
- BibTeX: 2
- EndNote: 3
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1