Decadal change of summertime reactive nitrogen species and surface ozone over the Southeast United States

This article examines observations and modeling for two years with extensive field campaigns and examines decadal changes between the years. The article uses a somewhat coarse resolution (when applied to a region) and evaluates NOy species. The evaluation is mostly qualitative and the explanations for mean biases (the quantitative metric) are somewhat speculative. Overall the manuscript provides interesting insights into the decadal changes despite using short snapshots no inter-annual variation.

"Besides the base case that only includes ISOPNB for heterogeneous loss (Jacobs et al., 2014), we include two additional sensitivity tests to evaluate the potential impact of organic nitrate hydrolysis.One is "hydro_full" case including heterogeneous loss of a C5 dihydroxy dinitrate (DHDN) and monoterpene nitrates from OH oxidation (TERPN1), and the other one is "no_hydro" case assuming no heterogeneous loss for any organic nitrates." Obs:R=0.58,Slope=57 * Emissions are reported for the CONUS and average annual rates (1/mo), but spatial allocation and temporal allocation may be important to understand how they affect the region/time being reported.

Response 5
We show in the following Figure 2  Comment 12 402 -The low bias may be good for SEAC 4 S, but it would also be bad for ICARTT.This sentence reads as though there is a tidy explanation.

Response 12
We have modified the text in lines 371-373 as: "This low bias can be partially due to neglecting small alkyl nitrates, which could contribute 20 -30 ppt to ANs (less than 10% near the surface) during SEAC 4 RS (Fisher et al., 2016).Including small alkyl nitrates will increase modeled ANs a bit in ICARTT as well." Comment 13 446 -The discussion of implemented chemistry seems to belong in section 2.

Response 13
We have shortened the discussion in Section 2 as the reviewer suggested, particularly on heterogeneous chemistry.Comment 14 459 -R5 is not the exclusive fate of NO2.This should be more clear and consistent in the narrative.Perhaps using Ox would simplify and correct the narrative.

Response 14
We have explained Ox to connect R5 to the narrative in lines 432-433 as

Response 15
This is simulated abundances.We have stated in line 508-509 of the original manuscript as "Figure 5 shows the mean vertical profiles of modeled monoterpene nitrates (MNs) and isoprene nitrates (INs) during ICARTT and SEAC 4 RS." Comment 16 542 -"prompt production" is often used as a technical term in chemical mechanisms to mean implemented without the intermediate, perhaps rate limiting, reaction.If this is implemented as "prompt production", then it seems inappropriate to say that you "In our model, we see prompt production."please clarify.

Response 16
The text in lines 505-506 has been revised as: "In our model, we see a rapid increase of PROPNN after sunrise in the boundary layer (Figure S7), …" Comment 17 562 -This sentence and what precisely it references is unclear.I'm assuming 561,562 is observations.The differences from the model are calculable from 6, but not immediately available.Clarify and or add numbers to the text.

Response 17
This sentence has been explained as "Based on model estimates in Figure 6, most RON are reduced proportionally ..." in lines 530-531 of the revised manuscript.
Comment 18 702,705 and elsewhere -The Pollack study is compared to this study several times.The differences.I am not sure why this is important to the conclusions.

Response 18
The trend of RON, specifically PAN, is not only dependent on changes of NOx but also on that of VOC precursors.Pollack study exhibited a different conclusion from our study on PAN trend, likely due to different VOC precursors in the two studied regions.

Comment 4 *
DAM3 introduced in Figure 4 is an unclear nomenclature.If I understand it correctly, DAM3 is replaying the Y-axis with a subset of ANs.It is not a separate model.Why prepend the D to DAM3 and DObs?Response 4 Figure 4 has been revised as following:

Figure 2 .
Figure 2. Anthropogenic NOx emission rate during July-August 2013 of (a) NEI11v1 inventory and (b) RCP8.5.Comment 6 * The "discrepancy" introduced on lines 413-415 and explained in Figure S5 seems like a major point.Consider moving Figure S5 into the main text.Even if the figure stays in the supplement, describe the "discrepancy" and make a citation to clarify.

"
We show that the model can roughly reproduce the correlation of Ox (= O3 + NO2) vs. ANs during both ICARTT and SEAC 4 RS (Figure 4), ...." Comment 15 508,510 -Not clear here if you are referencing simulated or observed abundances.