the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 06 Oct 2026
Constraining speciated mercury dry deposition models with urban observations in eastern China
Abstract. Atmospheric dry deposition constitutes a major pathway of terrestrial mercury (Hg) loading globally, yet estimates of its fluxes remain highly uncertain, particularly in regions with elevated particulate matter (PM) concentrations. This study employed long-term surrogate surface measurements to characterize speciated atmospheric Hg deposition at an urban site in eastern China. Over the period of July 2020 to June 2021, measured total atmospheric mercury (TAM) dry and wet deposition fluxes amounted to 29.19 and 7.29 μg m−2 yr−1, respectively, corresponding to a dry-to-wet ratio of 4.0. The observed gaseous oxidized mercury (GOM) dry deposition flux over the same period accounted for 17 % of the total dry deposition. These refined flux measurements provided observational constraints for speciated Hg dry deposition modeling. Applying a site-specific correction factor of 2.21 to denuder-based GOM concentrations avoided a 46 % underestimation of the mean GOM dry deposition flux. The explicit inclusion of coarse particulate-bound mercury (PBM) increased the deposition velocity by a factor of 4.5, and revealed that coarse particles accounted for 80 % of total PBM dry deposition. Replacing the reference leaf area index (LAI) with site-specific values reflecting local vegetation phenology better captured bidirectional gaseous elemental mercury (GEM) exchange, including net emission during certain summer periods. The refined simulations further showed that PBM and GEM co-dominated Hg dry deposition at this site, unlike many less PM-polluted regions where GEM plays a more dominant role. This observationally constrained framework improves speciated atmospheric Hg dry deposition modelling and reduces uncertainty in global Hg budget estimates.
- Preprint
(1403 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 17 Nov 2026)
- RC1: 'Comment on egusphere-2026-5395', Anonymous Referee #1, 07 Oct 2026 reply
-
RC2: 'Comment on egusphere-2026-5395', Anonymous Referee #2, 25 Oct 2026
reply
This manuscript combines surrogate-surface observations and modeling to examine speciated mercury dry deposition at an urban site in eastern China. The observational dataset is valuable, and the analysis of GOM concentration corrections, PBM particle-size distributions, and local vegetation parameters addresses important uncertainties in mercury deposition estimates. However, several methodological details require clarification, particularly the model inputs, sampling statistics, and comparability between modeled fluxes and surrogate-surface measurements. The evidence supporting improved model performance also needs further evaluation. I therefore recommend major revision. My specific comments are provided below.
1.The model includes vegetation GEM exchange processes that are not fully represented by artificial-turf measurements. Although the authors acknowledge this limitation and conduct sensitivity tests, please explain how this difference affects the evaluation of the local LAI scheme and clarify the extent to which the comparison supports the conclusions about its performance.
2.Please clarify how samples are counted in Sections 2.2.1 and 2.2.5. Distinguish sampling cycles, replicate samples, and valid samples retained after quality control, and specify the corresponding time periods. In particular, please reconcile the 297 precipitation samples collected with the 309 valid samples, and clarify whether “73 turf samples” actually refers to73 sampling cycles.
3.Lines 123-136: Please check the use of “ ; ” and the formatting of units. Please also provide units for all dimensional variables and parameters in the equations throughout the manuscript. For example, the gravitational settling velocities reported in lines 241-243 lack units, please add them.
4.Line 129: The definition of an “equivalent volume of ambient precipitation” is unclear. Please explicitly define the calculation of Mrain.
5.Lines 143-144: Please replace “bi-weekly” with “every two weeks” or “twice a week,” as appropriate, to clarify the replacement frequency.
6.Section 2.2.4 and line 95: RM is defined as GOM + PBM, but the subsection entitled “Membrane-based offline methods for RM” describes only size-resolved PBM measurements. Please revise the title and associated descriptions.
7Lines 174-176: The interlaboratory comparison and the reproducibility RSD of 9 ± 4% are stated twice in succession. Please remove the duplicate statement.
8.Lines 248-259: In Equation (16), χcappears both on the left-hand side and in the numerator of the first term on the right side. Please verify whether the latter should instead be the ambient concentration , consistent with the subsequent definition. The units also require attention: χc is given in ng m-3, whereas , χa, χst and χg are described in μg m-3. Please use consistent units or explicitly state the conversions used in the calculations.
9.The methods do not specify the sources, temporal resolution, or corresponding vegetation types of the local and reference LAI, nor whether deposition observations were used to adjust them. Please add a dedicated subsection describing both LAI datasets and the key model inputs.
10.Lines 515–517: The authors suggest that PBM may interfere with Aerohead measurements of GOM dry deposition under high particle-loading conditions, but no experimental evidence or direct literature support is provided. Please explain the basis for this statement and reconcile it with the description in Section 2.2.1 that the sampler design prevents contamination by particle deposition. Although GOM contributes relatively little to TAM dry deposition, the interference remains unquantified. Please clarify which conclusions are meant by the statement that it is “unlikely to alter the main conclusion.”.
11.Lines 327-329: The seasonal results support an important influence of precipitation amount on wet deposition, but the evidence that atmospheric RM abundance plays a secondary role remains insufficient. Please provide an analysis of temporally matched RM and precipitation Hg concentrations, or qualify the proposed ranking of their relative importance.
12.Figures 3, 5, and 6: The text identifies the reported statistics as uncentered R², whereas the figures label them only as r². Please make the terminology consistent, define the statistic in the methods, and report the number of paired observations.
13.Figure 5 does not clearly show winter validation data. If no winter observations are available, claims of improved seasonal performance throughout the year should be qualified.
14.Please use a consistent spelling convention throughout. Given the current mixture of “modeling/modelling” and “modeled/modelled,” either convention would be acceptable, but it should be applied consistently.
15.Lines 24-25: The statement in the abstract that the framework “reduces uncertainty in global Hg budget estimates” exceeds what is directly evaluated in this single-site study.
16.Line 217: The heading “Dynamic particle-size-regarded model for PBM dry deposition” is awkward.
Citation: https://doi.org/10.5194/egusphere-2026-5395-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 199 | 47 | 23 | 269 | 26 | 22 |
- HTML: 199
- PDF: 47
- XML: 23
- Total: 269
- BibTeX: 26
- EndNote: 22
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This study measured the dry and wet deposition fluxes of Hg at an urban site in mainland China and compared them with empirical deposition models. The results provide valuable insights into atmospheric Hg dry and wet deposition fluxes in urban environments and are useful for constraining Hg dry deposition in regional and global models. I therefore consider the study suitable for publication in ACP. The sampling and chemical analysis were well conducted, and the manuscript is well written. I broadly agree with the interpretations of the observations. One of my main concerns points to the uncertainty of the method, particularly the extent to which the measurements can represent Hg deposition fluxes under natural conditions. This is also important for comparison with modelling studies and for constraining speciated Hg dry deposition modeling. This concern arises because the surrogate surfaces used here differ from natural Earth surfaces, which may lead to uncertainty in the estimate of the "real" Hg dry deposition fluxes. I agree that the methods used here may currently be the most applicable approaches. However, the uncertainties associated with these methods should be discussed and noted. Some minor suggestions are listed below.
Line 13: a one-year of observations is not suitable to be described as a long-term study. I would suggest to remove ‘long-term’ here.
Line 39: better to specify that it is within the planetary boundary layer.
Line 54-55: I agree CEMs might be currently the most promising method for the measurements of GOM concentrations. While it remains inaccurate to state that it is the most promising method for the quantification of GOM dry deposition fluxes.
Line 152: RM should be PBM? It is unclear to me why the using of a size fraction sampling of PBM would enable more accurate and refine RM concentrations. Is the PTFE membranes would also effectively collect GOM?
Line 192: the Vd may vary with the Earth’s surface types. The authors should specify which types of surfaces are relevant to this study.
Equation (10): 11 distinct particle sizes PBM were measured in this study. why not to estimate PBM dry deposition fluxes using the 11 particle sizes?
Line 263 and 268, and many other places throughout the manuscript: I would suggest to use one deposition flux unit. μg m-2 yr-1 or ng m-1 yr-1.
Table 1: better to specify the particle size of PBM.
Line 312: what are the deposition fluxes simulated for the typical urban areas in East Asia?
Line 333: Hg wet deposition at Mt. Damei and Mt. Waliguan was not reported by Eom et al., 2021.
Line 364-365: different empirical settings of GOM Vd should be also the cause for the discrepancy between observations and modelling simulation.
Line 384: can this fraction be supported by previous studies or the measurements in this study? It appears that you have measured the size-fraction PBM concentrations at the sampling site.
Table 4: the GOM and PBM dry deposition velocities estimated there seems to be lower and higher than typical models. I would suggest to add a comparison between this study and modeling studies.
Line 471-475: are there any previous studies have measured the litterfall Hg deposition in urban areas of mainland China? if yes, please add a comparison with previous observations.