|The revised paper has improved compared to the discussion paper. However, there are some uncertainties that should be emphasized in the revised paper.|
Line 186: Using 12 trajectories to represent one GOM or PBM measurement is a large source of uncertainty. Simply mentioning that other studies had done the same thing doesn’t address the uncertainties in this current study. The trajectory directions can change significantly over a 12 hr period, but this variation is not reflected in the GOM and PBM measurements. More discussion is needed on how these uncertainties would affect the PSCF results.
Lines 189-190: I suggest not using “successful” because the modeled sources haven’t been evaluated.
Line 211: Similar to the above comment, using 12 wind directions to represent one GOM and PBM concentration is a large source of uncertainty. Wind directions can change significantly over a 12 hr period but you wouldn’t know the impact to GOM or PBM because there is only one measurement. More discussion is needed on how these uncertainties would affect the CPF results.
Lines 295-297: The partial correlations from multiple linear regression should be reported. This gives the unique contribution of each independent variable to Kp. It’s not the same as performing two separate linear regressions for temperature and RH.
Line 303: What was the goodness of fit (R2) for this equation? Gas-particle partitioning models developed in other studies should be mentioned in the results as well. The studies below were able to obtain a higher R2 and validated the model with Kp data at other sites.
Rutter, A. P., & Schauer, J. J. (2007). The effect of temperature on the gas–particle partitioning of reactive mercury in atmospheric aerosols. Atmospheric Environment, 41(38), 8647-8657.
Cheng, I., Zhang, L., & Blanchard, P. (2014). Regression modeling of gas‐particle partitioning of atmospheric oxidized mercury from temperature data. Journal of Geophysical Research: Atmospheres, 119(20), 11864-11876.
Line 404: Uncertainties related to different aerosol composition and temperature with different wind directions needs to be mentioned here. Both of these factors can affect gas-particle partitioning, which in turn affects Kp or the GOM/PBM ratio.
Lines 486-488: These results could be very different using the median concentration for each cluster. If the mean and median concentration for each cluster varies significantly, the source contribution results would likely be very different. In this case, which measure is most representative of the concentration for each cluster and should be used to determine the source contributions?
Lines 466-476: The results here are very inconsistent. Line 466 states, “Compared to the other clusters, the source contributions of clusters 1 and 4, which represent regional transport, were relatively low for all Hg species (Table 4). Cluster 5 contributed more significantly, especially for GOM and PBM, indicating the importance of Korean sources.” Later on line 475 states, “Clusters 1 and 5 were used to represent the effect of sources outside of Korea and the cluster 4 was used to indicate the effect of sources in Korea.”