the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 Nov 2025
Divergent iron dissolution pathways controlled by sulfuric and nitric acids from the ground-level to the upper mixing layer
Abstract. Iron (Fe) plays a crucial role in the global biogeochemical cycle, marine ecosystems, and human health. Despite extensive research on Fe dissolution, the understanding of the mechanism of the Fe acidification process remains highly controversial. Here, we revealed significant differences in Fe acid dissolution between the upper mixing layer and the ground-level of a megacity. The results showed that air masses with elevated n[SO42−]/n[NO3−] ratios (5.4 ± 3.7) yielded more enhanced iron solubility (%FeS, 8.7 ± 2.4 %) in the upper mixing layer after atmospheric aging compared to those (1.6 ± 0.7 and 3.3 ± 0.4 %, respectively) at the ground-level near source regions of acidic gases. Further analysis suggested that Fe dissolution is primarily driven by sulfuric acid in the upper mixing layer different from nitric acid at the ground-level, attributing to the aging processes of acidic species during long-range transport. %FeS also exhibits a clear size dependence: sulfuric-acid dominates in submicron aerosols (Dp<1 μm), leading to elevated %FeS (3.4 ± 3.8 %), whereas alkaline mineral dust in supermicron particles (Dp>1 μm) neutralizes nitric acid and suppresses Fe dissolution (1.7 ± 2.2 %). Our finding highlighted that sulfuric acid dominates Fe acidification process in the upper layer and fine particles, but the contribution of nitric acid to Fe dissolution at the ground-level is equally important. Our study provides new data sets for testing model’s capability to simulate dissolved Fe concentration and deposition and will help to improve the accuracy of Fe solubility predictions.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1417 KB)
-
Supplement
(5330 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1417 KB) - Metadata XML
-
Supplement
(5330 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-5423', Anonymous Referee #1, 06 Dec 2025
-
AC2: 'Reply on RC1', Weijun Li, 14 Jan 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5423/egusphere-2025-5423-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Weijun Li, 14 Jan 2026
-
RC2: 'Comment on egusphere-2025-5423', Anonymous Referee #2, 16 Dec 2025
This manuscript discusses the different dissolution pathways of Fe by nitrate and sulfate at two distinct locations. It is very interesting that sulfate dominates Fe dissolution in the upper mixing layer, while nitrate contributes to it at the ground level. The discussion integrates various insights, and I believe this paper provides important knowledge on atmospheric iron chemistry.
My most important concern is as follows. The manuscript mainly attributes the differences between the two sites to altitude. However, it may also be possible that the observed differences reflect differences in the sampling periods rather than altitude alone, and this point requires further discussion. In Figure 5, NOx emissions appear to be higher in September, when the HZ sampling was conducted than in July. Could the higher nitrate concentrations observed at HZ be influenced by such seasonal differences? I suggest that the authors demonstrate, using previous studies and/or available data, the seasonal variability of iron, TSP, nitrate, and sulfate at Damingshan and/or Hangzhou, and clarify that the observed site-to-site differences cannot be explained solely by seasonal variation.
In addition, I found it somewhat difficult to identify the fundamental factor that determines whether nitrate or sulfate contributes more strongly to iron solubility. Is it simply the difference in relative concentrations? Or could differences in their size distributions between the two sites play an important role? It would significantly strengthen the manuscript if the authors could provide a clearer discussion on what drives this difference.
Specific Comments
- L.68: The manuscript states that proton concentrations for proton-promoted dissolution are estimated only by the ratios of sulfate to calcite. Is this approach universally adopted in previous studies? For example, do models such as GEOS-Chem or IMPACT use different methods?
- L.74: Since a main objective of this study is the comparison between the ground level and the upper mixing layer, I suggest briefly explaining why the upper mixing layer is particularly important and why it is the focus of this study.
- L.76: The abbreviation “DMS” is used for Damingshan. In atmospheric chemistry, this could be confused with dimethyl sulfide. Consider using the site name without abbreviation.
- L.90: Please briefly describe how the mixing layer height (MLH) was determined.
- L.97: Quartz filters may have relatively high blank levels for trace elements compared with other materials (e.g. PTFE). Why did you choose quartz filters?
- L.115: Non-sea-salt sulfate does not appear in the results and discussion. Please check whether this description is necessary or not.
- L.140: It would be helpful to include filter blank values normalized by filter area (ng cm2-) in addition to ng m⁻³. Please also clarify how the air volume (m³) was assumed in calculating ng m⁻³, and indicate the contribution of filter blank to the samples.
- L.146: the soluble Fe and total Fe ‘concentration’, respectively
- L.157: Is the NH₃ data in 2025 from HZ or DMS?
- L.164: How was TSP measured? Please add a brief description of the method.
- Figure 2: Does the deviation represent variability among the three sample sets? If so, I recommend adding a brief explanation.
- Figure S3: Panels (a) and (b) are missing from the figure.
- L.210: In “DMS” TSP samples…
- L.223–226: The author discusses differences between IMPACT model results and observations at different locations. It is difficult to know if it is because of the model bias or location-dependent effects. Ideally, both sites should be compared using observations. At least, it would be helpful to demonstrate that the IMPACT model reasonably reproduces the size distributions in Figure 4, in addition to nitrate and sulfate concentrations (in TSP at HZ), before making this comparison.
- Figure S6: Please use either “sulfate” or “sulphate” consistently.
- L.240: In Fig. S7, the correlations between SO₄²⁻ and Feₛ or %Feₛ appear similarly strong for both supermicron and submicron particles. Please check this carefully. If the correlation is indeed weaker for supermicron particles, it would be helpful to show this quantitatively in the text.
- L.245: Why is the negative correlation between pH and Feₛ or %Feₛ not observed for submicron particles? Could mechanisms other than proton-promoted dissolution be involved?
- L.247: Compared with what?
- Section 4.1: Based on Figure 5, DMS does not appear to experience longer-range transport than HZ. I suggest expanding the map range so that the full trajectories are visible and comparing transport distances explicitly. In addition, both trajectories are shown at 500 m altitude; for a mountain site, trajectories at 1000 m or 1500 m may better represent air masses reaching the upper mixing layer. These may also show longer transport pathways. I also recommend reconsidering whether the differences in trajectory direction between the two sites reflect seasonal effects, site location, or altitude, by comparing trajectories under different conditions.
- L.266: “In” general
- Figure S9: Please unify the color-bar scale for the clarity.
- L.288 (Text S2): I agree with the conclusion that organic acids likely do not contribute significantly to iron dissolution. However, if oxalate concentrations are estimated from sulfate and the lack of correlation between oxalate/FeT and %Feₛ is used as evidence, this would also imply no correlation between sulfate/FeT and %Feₛ (including at DMS), which seems inconsistent with the main text’s argument that sulfate contributes to iron dissolution. If oxalate was not directly measured, it may be preferable not to estimate it from sulfate. I think discussion in the third paragraph (“While organic acids can …”) is enough.
- Figure 6b: For DMS, should the p-value also be reported as p > 0.05?
- Figure 6c: I’m also curious about the DMS results; please consider including them.
- Figure 7a: Rather than low FeT combined with high (n[SO₄²⁻]+n[NO₃⁻])/[FeT], could the issue be that Feₛ was too low to allow %Feₛ to be determined? Please clarify.
- Figure 7b: The x-axis is shown on a logarithmic scale, but was the correlation analysis performed in linear space as in Fig. 6b? Please confirm that the correlation methods are applied consistently.
Citation: https://doi.org/10.5194/egusphere-2025-5423-RC2 -
AC1: 'Reply on RC2', Weijun Li, 14 Jan 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5423/egusphere-2025-5423-AC1-supplement.pdf
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-5423', Anonymous Referee #1, 06 Dec 2025
This paper investigated the mechanisms by which sulfuric and nitric acids influence iron (Fe) dissolution at different altitudes through a comparative analysis of aerosol samples collected at ground level in Hangzhou and in the upper mixing layer at Mountain Daming. It reported that there were significant vertical differences in Fe solubility and related aging processes of acidic species. The results are interesting and will help improve our understanding on the biogeochemical cycling of atmospheric Fe. Generally, this manuscript is well structured and written, I will be happy to recommend this manuscript for publication after a minor revision.
Major comments
- Line 100-105: Only 7 TSP samples per site and 3 sets of MOUDI samples (DMS only, none for HZ) are mentioned. The limitations of this sample size and its potential impact on statistical significance should be more explicitly addressed in the discussion.
- Brief summary on air pollution conditions during the sampling period should be supplemented.
- Line 156-161: For aerosol pH calculation, only size-resolved aerosols were considered, what about the TSP samples? In addition, 2025 NH₃ data was used to estimate 2021 levels at DMS, while acknowledged, may introduce uncertainty. I suggest the author implement a sensitivity analysis or further discussion on the potential impact of this assumption on pH calculation.
- Figure 2e: In consideration of the deviation, single peak pattern may be more suitable for size distributions of total Fe rather than bimodal pattern.
- Line 287-289. The concentrations of oxalate were estimated by empirical relationship with sulfate, resulting in associated uncertainty. Therefore, the result“organic acids played a limited role in Fe dissolution” should be more conservative. In addition, what about the potential synergistic effect between organic and inorganic acids on Fe dissolution？
Minor comments
- The relevant work of Tang et al. (2025, AMT) and Chen et al. (2024, EST) might be considered for citation in this manuscript.
- Line 185 “…or 10-18…”, “or” should be replaced with “and”.
Citation: https://doi.org/10.5194/egusphere-2025-5423-RC1 -
AC2: 'Reply on RC1', Weijun Li, 14 Jan 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5423/egusphere-2025-5423-AC2-supplement.pdf
-
RC2: 'Comment on egusphere-2025-5423', Anonymous Referee #2, 16 Dec 2025
This manuscript discusses the different dissolution pathways of Fe by nitrate and sulfate at two distinct locations. It is very interesting that sulfate dominates Fe dissolution in the upper mixing layer, while nitrate contributes to it at the ground level. The discussion integrates various insights, and I believe this paper provides important knowledge on atmospheric iron chemistry.
My most important concern is as follows. The manuscript mainly attributes the differences between the two sites to altitude. However, it may also be possible that the observed differences reflect differences in the sampling periods rather than altitude alone, and this point requires further discussion. In Figure 5, NOx emissions appear to be higher in September, when the HZ sampling was conducted than in July. Could the higher nitrate concentrations observed at HZ be influenced by such seasonal differences? I suggest that the authors demonstrate, using previous studies and/or available data, the seasonal variability of iron, TSP, nitrate, and sulfate at Damingshan and/or Hangzhou, and clarify that the observed site-to-site differences cannot be explained solely by seasonal variation.
In addition, I found it somewhat difficult to identify the fundamental factor that determines whether nitrate or sulfate contributes more strongly to iron solubility. Is it simply the difference in relative concentrations? Or could differences in their size distributions between the two sites play an important role? It would significantly strengthen the manuscript if the authors could provide a clearer discussion on what drives this difference.
Specific Comments
- L.68: The manuscript states that proton concentrations for proton-promoted dissolution are estimated only by the ratios of sulfate to calcite. Is this approach universally adopted in previous studies? For example, do models such as GEOS-Chem or IMPACT use different methods?
- L.74: Since a main objective of this study is the comparison between the ground level and the upper mixing layer, I suggest briefly explaining why the upper mixing layer is particularly important and why it is the focus of this study.
- L.76: The abbreviation “DMS” is used for Damingshan. In atmospheric chemistry, this could be confused with dimethyl sulfide. Consider using the site name without abbreviation.
- L.90: Please briefly describe how the mixing layer height (MLH) was determined.
- L.97: Quartz filters may have relatively high blank levels for trace elements compared with other materials (e.g. PTFE). Why did you choose quartz filters?
- L.115: Non-sea-salt sulfate does not appear in the results and discussion. Please check whether this description is necessary or not.
- L.140: It would be helpful to include filter blank values normalized by filter area (ng cm2-) in addition to ng m⁻³. Please also clarify how the air volume (m³) was assumed in calculating ng m⁻³, and indicate the contribution of filter blank to the samples.
- L.146: the soluble Fe and total Fe ‘concentration’, respectively
- L.157: Is the NH₃ data in 2025 from HZ or DMS?
- L.164: How was TSP measured? Please add a brief description of the method.
- Figure 2: Does the deviation represent variability among the three sample sets? If so, I recommend adding a brief explanation.
- Figure S3: Panels (a) and (b) are missing from the figure.
- L.210: In “DMS” TSP samples…
- L.223–226: The author discusses differences between IMPACT model results and observations at different locations. It is difficult to know if it is because of the model bias or location-dependent effects. Ideally, both sites should be compared using observations. At least, it would be helpful to demonstrate that the IMPACT model reasonably reproduces the size distributions in Figure 4, in addition to nitrate and sulfate concentrations (in TSP at HZ), before making this comparison.
- Figure S6: Please use either “sulfate” or “sulphate” consistently.
- L.240: In Fig. S7, the correlations between SO₄²⁻ and Feₛ or %Feₛ appear similarly strong for both supermicron and submicron particles. Please check this carefully. If the correlation is indeed weaker for supermicron particles, it would be helpful to show this quantitatively in the text.
- L.245: Why is the negative correlation between pH and Feₛ or %Feₛ not observed for submicron particles? Could mechanisms other than proton-promoted dissolution be involved?
- L.247: Compared with what?
- Section 4.1: Based on Figure 5, DMS does not appear to experience longer-range transport than HZ. I suggest expanding the map range so that the full trajectories are visible and comparing transport distances explicitly. In addition, both trajectories are shown at 500 m altitude; for a mountain site, trajectories at 1000 m or 1500 m may better represent air masses reaching the upper mixing layer. These may also show longer transport pathways. I also recommend reconsidering whether the differences in trajectory direction between the two sites reflect seasonal effects, site location, or altitude, by comparing trajectories under different conditions.
- L.266: “In” general
- Figure S9: Please unify the color-bar scale for the clarity.
- L.288 (Text S2): I agree with the conclusion that organic acids likely do not contribute significantly to iron dissolution. However, if oxalate concentrations are estimated from sulfate and the lack of correlation between oxalate/FeT and %Feₛ is used as evidence, this would also imply no correlation between sulfate/FeT and %Feₛ (including at DMS), which seems inconsistent with the main text’s argument that sulfate contributes to iron dissolution. If oxalate was not directly measured, it may be preferable not to estimate it from sulfate. I think discussion in the third paragraph (“While organic acids can …”) is enough.
- Figure 6b: For DMS, should the p-value also be reported as p > 0.05?
- Figure 6c: I’m also curious about the DMS results; please consider including them.
- Figure 7a: Rather than low FeT combined with high (n[SO₄²⁻]+n[NO₃⁻])/[FeT], could the issue be that Feₛ was too low to allow %Feₛ to be determined? Please clarify.
- Figure 7b: The x-axis is shown on a logarithmic scale, but was the correlation analysis performed in linear space as in Fig. 6b? Please confirm that the correlation methods are applied consistently.
Citation: https://doi.org/10.5194/egusphere-2025-5423-RC2 -
AC1: 'Reply on RC2', Weijun Li, 14 Jan 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5423/egusphere-2025-5423-AC1-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 347 | 96 | 37 | 480 | 41 | 29 | 26 |
- HTML: 347
- PDF: 96
- XML: 37
- Total: 480
- Supplement: 41
- BibTeX: 29
- EndNote: 26
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Guochen Wang
Xuedong Cui
Bingye Xu
Can Wu
Minkang Zhi
Keliang Li
Liang Xu
Qi Yuan
Yuntao Wang
Zongbo Shi
Akinori Ito
Shixian Zhai
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1417 KB) - Metadata XML
-
Supplement
(5330 KB) - BibTeX
- EndNote
- Final revised paper
This paper investigated the mechanisms by which sulfuric and nitric acids influence iron (Fe) dissolution at different altitudes through a comparative analysis of aerosol samples collected at ground level in Hangzhou and in the upper mixing layer at Mountain Daming. It reported that there were significant vertical differences in Fe solubility and related aging processes of acidic species. The results are interesting and will help improve our understanding on the biogeochemical cycling of atmospheric Fe. Generally, this manuscript is well structured and written, I will be happy to recommend this manuscript for publication after a minor revision.
Major comments
Minor comments