|Most of my comments have been well answered. And more details and deep discussions have been added in the revised manuscript. However there are still some points which are not well considered: |
1) Although I asked the author to show the comparisons of a-priori profiles and profiles measured by the MAX-DOAS in the major comment 1 of the previous review, a proper replythe was not given by the author. Please reply for this point.
2) In the revised draft, the author applied the MAX-DOAS HCHO profile in the calculations of OMI AMFs. And the author attributed the overestimations of the corrected OMI HCHO VCDs compared to MAX-DOAS VCDs to “the MAX-DOAS measurements are not sensitive to HCHO at higher altitude”. This is a strong claim on the credibility of MAX-DOAS HCHO measurements. OMI tropospheric HCHO SCD retrievals might also play a role here since some post corrections might be applied. If the author will insist on the claim, a study on typical proportions of HCHO located at upper troposphere in total HCHO columns and a sensitivity study of MAX-DOAS HCHO profile retrievals need to be done. I suggest the author modify their statement.
3) Regarding my second major comment, the author might not understand my argument. My argument is whether or not the lifetime weighting factor is meaningful to improve the backward propagation method. Both the trajectory backward time and lifetime can impact the weighting factor based on your equation 1, but both are arbitrarily determined without a reasonable sensitivity study shown in the draft. Meanwhile if the NO2 life time of 6 hours is assumed, do the trajectories of less than 12 hours also play a role, e.g. 1, 2, 3… hours? And how is the mixing of air parcels with different backward time considered in your approach?
For HCHO, the author assumed a shorter backward time, however as we see in Fig. 8c, the HCHO map observed by OMI is much smoother than OMI NO2 map shown in Fig. 7. Although HCHO lifetime is shorter than NO2, secondary formation of HCHO from its VOC precursors along regional transports can contribute to your measurement site. In another word, HCHO which the MAX-DOAS instrument measured was not all from primary emissions, but also from secondary formations. This effect means a larger backward time than 1 hour should be applied. The author might reference the sensitivity test on the effect of trajectory backward time on reconstructed maps shown in the recent paper of Wang et al., 2019 (which is cited in your draft). They applied the similar approach and determined the backward time reasonably by comparing reconstructed maps with satellite maps.
4) Regarding effects of meteorology on pollutants around the Youth Olympic Games, the author wrote “as the meteorological conditions are very similar during the three periods”, but it is not true for temperature. Figure 10b indicates that the temperature in the pre-Olympic period is higher than those in the Olympic and post-Olympic periods. Temperature can impact secondary formation of HCHO. The effect might contribute to the higher HCHO concentrations in the pre-Olympic period than those in the other two periods. The author needs to consider the effect.
In addition, as I suggested in the major comment 3 of the previous review, “difference of transport conditions in the three periods should also be discussed”. The statistics of trajectories might be needed to show for the discussion. However a reply to this point was not given for this point.
5) Please add the information of the reply to my specific comment 8 into the draft.