Interactive comment on “ 222 Rn calibrated mercury fluxes from terrestrial surface of southern Africa derived from observations at Cape Point , South Africa ”

Overview: The manuscript entitled “222Rn calibrated mercury fluxes from terrestrial surface of southern Africa derived from observations at Cape Point, South Africa”, by F. Slemr et al. describes observationally constrained estimates of terrestrial Hg fluxes from southern Africa. The analysis draws upon nearly 5 years (2007–2011) of measurements of gaseous elemental mercury (GEM) and 222Rn made at the Global Atmospheric Watch (GAW) station Cape Point. Estimated terrestrial 222Rn fluxes are combined with correlations between GEM and 222Rn during periods with elevated 222Rn to derive terrestrial GEM fluxes. The derived fluxes are corrected for 222Rn decay using an assumed transport time. As the authors point out, terrestrial Hg fluxes are poorly constrained by existing observations, particularly outside of northern mid-

Overview: The manuscript entitled "222Rn calibrated mercury fluxes from terrestrial surface of southern Africa derived from observations at Cape Point, South Africa", by F. Slemr et al. describes observationally constrained estimates of terrestrial Hg fluxes from southern Africa.The analysis draws upon nearly 5 years (2007-2011) of measurements of gaseous elemental mercury (GEM) and 222Rn made at the Global Atmospheric Watch (GAW) station Cape Point.Estimated terrestrial 222Rn fluxes are combined with correlations between GEM and 222Rn during periods with elevated 222Rn to derive terrestrial GEM fluxes.The derived fluxes are corrected for 222Rn decay using an assumed transport time.As the authors point out, terrestrial Hg fluxes are poorly constrained by existing observations, particularly outside of northern mid-C1112 latitude regions.This study therefore makes an important contribution to understanding regional terrestrial fluxes.
Overall the paper is well written and provides a good review of relevant past work.My main criticisms, described under "Specific Comments", are: (1) some important analytical and methodological details are missing; (2) the true mean Hg flux may not be properly described without weighting each 222Rn "event" by its duration; (3) excluding GEM depletion events may bias the results.Minor issues are identified under "Technical Corrections".I have also included under "Technical Corrections" some suggested grammatical corrections.
Although I have identified some important issues with the manuscript content/organization and with the analysis methodology, I expect my concerns can be addressed without major revisions.I recommend that the paper be published in ACP after the authors address the issues identified below.

Specific Comments
1. Tekran calibrations: The authors make no mention of calibration of the Hg measurements against a Tekran 2505 saturated Hg0 vapor source (or similar primary standard).This information also appears not to have been provided in the preceding work by Brunke et al. (2010).I am assuming the authors did in fact employ a primary Hg calibration standard.Therefore, for completeness, it would be helpful if the authors could include some quantitative data on their primary calibrations.This information is necessary for assessing the overall uncertainty in the measurements.Similarly, further detail on error associated with the 222Rn measurements would be helpful (e.g., quoted from Brunke et al., 2002).As for Hg, only the detection limit is quoted, thereby providing a measure of precision error alone.Here, it is at least clear to the reader that another publication (Brunke et al., 2002) can be consulted for further details on measurement error.
2. Description of data analysis protocol: The paper could be improved by inclusion of more specific details on how the data were treated.The authors mention that they defined "events" as times periods when 222Rn was >1000 mBq m−3, and that they filtered short duration Hg depletion and pollution events.However, it would be helpful if the authors could define more quantitatively what filter criteria were used (e.g., threshold concentrations of GEM, CO, etc.).It is mentioned that the 15 min integrated GEM concentrations were reduced to 30 min averages prior to merging with other measurements, but the time resolution of the 222Rn measurements is not given.This information also does not appear to be provided in Brunke et al. (2002), while Whittlestone and Zahorowski (1998) indicate a time resolution of 45 min.Some further details need to be included for clarity.
3. Wet deposition measurements: On page 8220, the authors write "the occurrence of precipitation was investigated for 7 of the events with the highest emission and 5 events associated with the highest deposition".Prior to this statement, no mention of deposition measurements at the Cape Point station is made.Only later do the authors reference Gichuki and Mason (2013) when discussing precipitation measurements on page 8221.These measurements should be described, along with any additional relevant meteorological measurements, in the Experimental section.

Error in mean values:
In some instances it is unclear what is represented by the error values cited for mean GEM/222Rn slopes and intercepts.Standard errors are explicitly noted in some places, but standard deviations are implied in others.For simplicity and clarity it would help if the authors could choose a single metric (e.g., standard error) and describe in the experimental section which metric is to be used, or more clearly identify each quoted error as standard deviation or standard error.5. Time-weighting of GEM/222Rn correlation slopes during 222Rn events.It seems that a better estimate of the average GEM flux could be obtained by time-weighting each 222Rn event (i.e., weighting the event by duration).If the authors choose not to do this, they should at least provide some mention of this potential source of bias in their discussion of error (perhaps along with discussion of 222Rn decay on page C1114 8220).Point are "numerous".Presumably these events are associated with enhanced RGM production and deposition.It seems it would be helpful if the authors could test the sensitivity of their results to incorporation of the GEM depletion events into their 222Rn event analysis.I understand this may not be feasible, but if so it would strengthen the analysis.
7. Correction for 222Rn decay: It would be helpful if the authors could provide a brief description to justify the assumed 2 day transport time that was used to correct the derived GEM fluxes for 222Rn decay.

Technical Corrections
1. Title: The authors might want to consider shortening their title.To do so, they could eliminate "derived from observations at Cape Point, South Africa" or "South Africa".I also suggest inserting "the" between "from" and "terrestrial" 2. Abstract; "emission ratio" vs. "flux ratio": The authors refer to the GEM/222Rn "emission ratio" in the Abstract, but use the term "flux ratio" elsewhere.Through their discussion, they demonstrate that the 222Rn flux is always positive.However, since the same is not true for GEM (i.e., the flux is bi-directional), it seems that, in the context of the present analysis, "flux ratio" is more appropriate than "emission ratio".Thus, only trial 222Rn flux for southern Africa is somewhat confusing as the Griffiths et al. paper considers Australia, not South Africa.Some clarification would be helpful here.
24. Conclusions, page 8222: Shouldn't the end date be December 2011, not "December 2009"?25.Conclusions, page 8222: I suggest rewriting "was with −0.01 ± 0.34 ng m−2 h−1 (standard error with n = 191, after correcting for 222Rn decay) not. .." as "was −0.01 ± 0.34 ng m−2 h−1 (standard error with n = 191, after correcting for 222Rn decay), which is not. .." 26. Figure 1: It would help to include the heading "Month" below the x-axis.Also, it seems that the caption should refer to "monthly" event frequency, not "seasonal" frequency.I suggest rewriting "all 222Rn events" as "222Rn events ("all")" and inserting "("significant")" after "correlations".I also suggest adding to the caption a brief description of the event criteria (e.g., 222Rn > 1000 mBq/m3).27. Figure 2: As it is impossible to achieve a significant slope of 0, the bin for "significant" slopes centered at 0.00 pg mBq−1 in this plot is somewhat confusing.I suggest including the range of slopes included in each bin (or at least in the bin centered at 0).Also, I suggest changing "Signif" to "significant" in the legend, to be consistent with Figure 1.

6 .
Contribution of RGM deposition to total Hg flux: On page 8220 the authors cite the measurements of Sorensen et al. (2010) as demonstrating that reactive gaseous mercury (RGM) concentrations in the marine boundary layer around southern Africa are small.Considering that the Sorensen et al. dataset is limited to a short time period in the context of the present dataset, I don't think the Sorensen et al. measurements are fully adequate to support the authors' conclusion that RGM makes a minimal contribution to Hg flux in the region.It seems that the Brunke et al. (2010) analysis is far more relevant to estimating the potential influence of RGM deposition on the overall Hg flux as observed at Cape Point.Brunke et al. indicate that GEM depletion events at Cape