the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Sep 2025
Identifying regions that can constrain anthropogenic Hg emissions uncertainties through modelling
Abstract. Anthropogenic mercury (Hg) emissions are a major contributor to global Hg pollution. However, limitations in emission inventories and modeling approaches impede accurate quantification of Hg emissions and Hg ecosystem inputs, complicating the evaluation of mitigation policies. This study investigates how uncertainties in anthropogenic emissions, compared to chemistry and meteorology modeling uncertainties, affect model performance in model-observation comparisons, and explores strategies to evaluate emission uncertainties. We performed modeling experiments that incorporated four global anthropogenic emission inventories, which differ in Hg emissions by up to 630 Mg in Asia, 259 Mg in South America, and 252 Mg in Africa. Additionally, we employed two different chemical schemes and two meteorological datasets. Inventory differences were the primary driver of significant differences across modeled total gaseous mercury (TGM) concentrations in the Northern Hemisphere, resulting in ranges of up to 0.47 ng m−3 in China and 0.32 ng m−3 in India. These differences influenced Root Mean Square Error scores in TGM model–observation comparisons, ranging from 0.03 to 0.19 in Asia, 0.12 to 0.25 in the Arctic, and 0.02 to 0.14 in the USA in an annual mean. A signal-to-noise ratio (SNR) analysis identified regions such as the eastern U.S., Greenland, and Arctic Russia as valuable for constraining anthropogenic emissions. The existing limited Southern Hemisphere network offers limited constraints on emissions but provides possible insights into Hg chemistry. These findings highlight the need for an expanded monitoring network and more refined emission inventories to reduce uncertainties and improve the accuracy of global Hg policy evaluation.
- Preprint
(13964 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-4018', Anonymous Referee #1, 08 Oct 2025
-
RC2: 'Comment on egusphere-2025-4018', Hélène Angot, 09 Oct 2025
This manuscript presents a modeling study that quantifies the influence of uncertainties in anthropogenic mercury (Hg) emissions, chemistry, and meteorology on simulated atmospheric Hg concentrations and wet deposition. The authors use a signal-to-noise ratio (SNR) approach to identify regions where model-observation comparisons are most suitable for evaluating anthropogenic emissions uncertainties. The topic is timely and relevant, the experimental setup is interesting, and the results are valuable for improving future Hg modeling and monitoring efforts. However, several issues need to be clarified and discussed in more depth before the paper can be considered for publication. My main concerns relate to (1) the comparison between modeled and observed Hg species, (2) the regional aggregation choices that affect the interpretation of SNR, and (3) the discussion of results and their practical implications for monitoring. Detailed comments are provided below.
Major comments
1. Consistency between GEM and TGMIn the Methods, the authors refer to “GEM” observations but later discuss “TGM” in the Results and Discussion. It is unclear what is actually being compared. Are the authors comparing observed GEM to modeled GEM + GOM + PBM?
Since atmospheric observations primarily represent GEM (as GOM and PBM are usually negligible and highly uncertain), the model–observation comparison should be limited to modeled GEM, not total gaseous mercury. This distinction must be clarified throughout the manuscript and in the figures.2. Need for a regional map and station overview
A map showing the boundaries of the defined regions (e.g., Arctic, Asia, Southern Hemisphere) and the location of the monitoring stations is essential. This would help readers understand which datasets are used and how the “regional” results are aggregated. For example, the observed Arctic signal appears unusual, with surprisingly high concentrations in winter and spring. Which stations were included? The authors should verify that the seasonal cycles used are consistent with established patterns (see e.g., Angot et al., (2016a)).
3. Regional aggregation and observed variability
The statement that “The discrepancy in modeled TGM in Asia cannot be easily constrained by model-observation comparisons using the current observation sites. The reason is that, in this case, the range of the model results falls within the range of TGM measurement variability” should be interpreted with caution. Part of this large observational variability may arise from spatial aggregation of heterogeneous sites across a vast region. Focusing on smaller, more homogeneous subregions could reduce the apparent observational noise and improve the ability to constrain emission-related uncertainties.
4. Southern Hemisphere (SH) emissions and SNR interpretation
The authors conclude that SH monitoring networks are not ideal for constraining anthropogenic emissions uncertainties. While this may be true, the low SNR in the SH could also stem from poorly constrained emission inventories in this region. The apparent weak emissions signal may reflect limitations in the input data, not just a true physical insensitivity. This should be explicitly discussed.
5. Real-world feasibility of proposed monitoring strategies
Some of the recommendations in the Discussion appear unrealistic given logistical constraints. For example:
- The suggestion that the Arctic’s high winter SNR makes it “an excellent location” for studying background Hg ignores practical challenges (cold, darkness, snow/ice cover).
- The proposal to enhance wet deposition monitoring in South Africa overlooks that it is a dry region, often under drought conditions.
- Similarly, strengthening monitoring in Greenland and Arctic Russia is operationally difficult especially in the current geopolitical context.
These limitations should be acknowledged to ensure recommendations remain grounded in real-world feasibility.
Minor Comments
- Line 42: Dastoor et al., 2024 should be Dastoor et al., 2025.
- Figure 1: The emission inventories cover different time windows. Which years are compared? Please clarify this to avoid comparing “apples to oranges.” Similarly, why did the authors use different emission years in their “inventories” simulations (e.g., 2015 for AMAP/GMA, 2012 for EDGAR, 2013–2015 for STREETS, and 2010 for WHET)?
- Methods: Note that the GEOS-Chem version used is outdated and does not include the most recent Hg chemistry (Shah et al., 2021). The EDGAR inventory is also not the latest version, and another recent global inventory (Qiu et al., 2025) could be mentioned. The authors need not rerun simulations, but these limitations should be acknowledged in the Discussion.
- Line 149: “GOM wet deposition flux”, replace with “wet deposition flux.”
- Lines 196–197: Please consider citing recent studies describing the summertime phenomenon more accurately (Angot et al., 2016a; Araujo et al., 2022; Huang et al., 2025; Yue et al., 2023). Ahmed et al. (2023) refers to springtime events, and the Dastoor papers do not specifically address this point.
- Lines 217–218: Please consider citing Angot et al., (2016a, c, b), which provide multi-year Antarctic Hg observations directly relevant to this discussion. I mention these papers not because I am the author, but because they contain the most comprehensive long-term Antarctic datasets available to date.
- Line 242: Typo: “MEteo.”
- Lines 329–330: The statement is valid only in winter; please clarify.
References
Ahmed, S., Thomas, J. L., Angot, H., Dommergue, A., Archer, S. D., Bariteau, L., Beck, I., Benavent, N., Blechschmidt, A.-M., Blomquist, B., Boyer, M., Christensen, J. H., Dahlke, S., Dastoor, A., Helmig, D., Howard, D., Jacobi, H.-W., Jokinen, T., Lapere, R., Laurila, T., Quéléver, L. L. J., Richter, A., Ryjkov, A., Mahajan, A. S., Marelle, L., Pfaffhuber, K. A., Posman, K., Rinke, A., Saiz-Lopez, A., Schmale, J., Skov, H., Steffen, A., Stupple, G., Stutz, J., Travnikov, O., and Zilker, B.: Modelling the coupled mercury-halogen-ozone cycle in the central Arctic during spring, Elem. Sci. Anthr., 11, 00129, https://doi.org/10.1525/elementa.2022.00129, 2023.
Angot, H., Dastoor, A., De Simone, F., Gårdfeldt, K., Gencarelli, C. N., Hedgecock, I. M., Langer, S., Magand, O., Mastromonaco, M. N., Nordstrøm, C., Pfaffhuber, K. A., Pirrone, N., Ryjkov, A., Selin, N. E., Skov, H., Song, S., Sprovieri, F., Steffen, A., Toyota, K., Travnikov, O., Yang, X., and Dommergue, A.: Chemical cycling and deposition of atmospheric mercury in polar regions:review of recent measurements and comparison with models, Atmospheric Chem. Phys., 16, 10735–10763, https://doi.org/10.5194/acp-16-10735-2016, 2016a.
Angot, H., Dion, I., Vogel, N., Legrand, M., Magand, O., and Dommergue, A.: Multi-year record of atmospheric mercury at Dumont d’Urville, East Antarctic coast: continental outflow and oceanic influences, Atmos Chem Phys, 16, 8265–8279, https://doi.org/10.5194/acp-16-8265-2016, 2016b.
Angot, H., Magand, O., Helmig, D., Ricaud, P., Quennehen, B., Gallée, H., Del Guasta, M., Sprovieri, F., Pirrone, N., Savarino, J., and Dommergue, A.: New insights into the atmospheric mercury cycling in central Antarctica and implications on a continental scale, Atmos Chem Phys, 16, 8249–8264, https://doi.org/10.5194/acp-16-8249-2016, 2016c.
Araujo, B. F., Osterwalder, S., Szponar, N., Lee, D., Petrova, M. V., Pernov, J. B., Ahmed, S., Heimbürger-Boavida, L.-E., Laffont, L., Teisserenc, R., Tananaev, N., Nordstrom, C., Magand, O., Stupple, G., Skov, H., Steffen, A., Bergquist, B., Pfaffhuber, K. A., Thomas, J. L., Scheper, S., Petäjä, T., Dommergue, A., and Sonke, J. E.: Mercury isotope evidence for Arctic summertime re-emission of mercury from the cryosphere, Nat. Commun., 13, 4956, https://doi.org/10.1038/s41467-022-32440-8, 2022.
Huang, S., Yuan, T., Song, Z., Chang, R., Peng, D., Zhang, P., Li, L., Wu, P., Zhou, G., Yue, F., Xie, Z., Wang, F., and Zhang, Y.: Oceanic evasion fuels Arctic summertime rebound of atmospheric mercury and drives transport to Arctic terrestrial ecosystems, Nat. Commun., 16, 903, https://doi.org/10.1038/s41467-025-56300-3, 2025.
Qiu, X., Liu, M., Zhang, Y., Zhang, Q., Lin, H., Cai, X., Li, J., Dai, R., Zheng, S., Wang, J., Zhu, Y., Shen, H., Shen, G., Wang, X., and Tao, S.: Declines in anthropogenic mercury emissions in the Global North and China offset by the Global South, Nat. Commun., 16, 1179, https://doi.org/10.1038/s41467-025-56274-2, 2025.
Shah, V., Jacob, D. J., Thackray, C. P., Wang, X., Sunderland, E. M., Dibble, T. S., Saiz-Lopez, A., Černušák, I., Kellö, V., Castro, P. J., Wu, R., and Wang, C.: Improved Mechanistic Model of the Atmospheric Redox Chemistry of Mercury, Environ. Sci. Technol., https://doi.org/10.1021/acs.est.1c03160, 2021.
Yue, F., Angot, H., Blomquist, B., Schmale, J., Hoppe, C. J. M., Lei, R., Shupe, M. D., Zhan, L., Ren, J., Liu, H., Beck, I., Howard, D., Jokinen, T., Laurila, T., Quéléver, L., Boyer, M., Petäjä, T., Archer, S., Bariteau, L., Helmig, D., Hueber, J., Jacobi, H.-W., Posman, K., and Xie, Z.: The Marginal Ice Zone as a dominant source region of atmospheric mercury during central Arctic summertime, Nat. Commun., 14, 1–13, https://doi.org/10.1038/s41467-023-40660-9, 2023.
Citation: https://doi.org/10.5194/egusphere-2025-4018-RC2
Data sets
Identifying regions that can constrain anthropogenic Hg emissions uncertainties through modelling C. Gournia et al. https://doi.org/10.7910/DVN/Z3FKWE
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 130 | 45 | 13 | 188 | 6 | 5 |
- HTML: 130
- PDF: 45
- XML: 13
- Total: 188
- BibTeX: 6
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Gournia et al. present a modeling study aimed at assessing the contribution of uncertainty in anthropogenic Hg emissions to model errors with respect to surface observations. The goal of the study is to help identify regions where additional observations would be most effective in constraining anthropogenic emissions. The authors conclude that uncertainties in emissions are strongly influence surface concentrations in the northern hemisphere in the model, while uncertainties in chemistry dominate in the souther hemisphere. They identify eastern US, Greenland, and Russian Arctic as regions where observations could effectively constrain anthropogenic emissions.
Understanding and reducing uncertainties in emission inventories is important for scientific applications of Hg models and for informing policy decisions. This study address a critical aspect of Hg modeling. However I have significant concerns about the study’s methods and conclusions, as outlined below:
(1) One of the study's main conclusions is that observations in Greenland & the Arctic would help constrain Hg emission inventories. I find this conclusion difficult to accept. As shown by the authors in Fig 1, most of the uncertainty in Hg emissions is in Asia and S. America. It follows that more observations in these regions (and immediately downwind) would be most effective in reducing the emissions uncertainty, not observations in remote regions like the Arctic. This incorrect conclusion likely arises from their use of the “SNR” metric, which seems unsuitable for this purpose. The SNR values depend more on the day-to-day variation ( "noise") in the modeled surface concentrations (Fig. 4), and less to the model uncertainties that the authors are tying to assess.
(2) The study overlooks an important source of uncertainty in Hg modeling: the exchange of mercury between the atmosphere and land, ocean, and the biosphere. These exchanges are a significant source of uncertainty in Hg modeling and must be considered when assessing the relative importance of emission inventories compared to other processes in the model.
(3) The Hg chemistry in the model used in the study is outdated and the uncertainties in chemistry considered in the study do not reflect our current understanding. See Saiz-Lopez et al. 2020 (10.1073/pnas.1922486117) and related work. This affects the study’s conclusion about the relative importance of emission uncertainty in comparison to uncertainty in chemistry.
Minor and technical comments:
1) Table 1: Please clarify which quantity the uncertainty ranges refer to. Do they represent global emissions or emissions at the grid-point level?
2) Fig 1(B): Which species’ emission range is depicted—TGM or GEM?
3) Fig 2: Please specify the years of the emission inventories being compared.
4) Table 2: The correct name of the meteorological dataset is MERRA-2, not MERRA.
5) Line 145: MERRA 2 & GEOS-FP are products of the same weather model (GEOS) and therefore do not represent meteorological uncertainty in any meaningful way.
6) Fig 4(A) and (B): Units for the color bars are missing.