This study aims to address the underestimation of volatile organic compounds (VOCs) in emission inventories by adjusting scaling factors, optimizing photochemical reactions, and revising the schemes for generating additional polycyclic aromatic hydrocarbons (PAHs). They utilized both ground-based and airborne observation data to evaluate the extent of the impact of improved VOCs emissions on the Seoul Metropolitan Area. This contributes to the enhancement of the model's ability to simulate urban air quality. This paper is exceptionally well-crafted, presenting intriguing findings. Consequently, I suggest it should be published following appropriate revisions.

1. To improve the alignment between observed VOC speciation and model predictions, the chemical mechanism has been revised and underestimated VOCs species has been increased. These revisions are designed to improve the understanding of VOC species in the atmosphere and to enhance the simulation capability of the model. I would like to know if these improvements are suitable for other regions than the SMA and the period other than May 1 to June 10, 2016.

2. Line 158, Are you referring to the process of individually adjusting the scaling factors for VOCs species and comparing them with observed values to determine the optimal scaling factors? And the determination of the best scaling factor typically involves assessing how well the adjusted model outputs align with actual measurements. What metrics were used in this study to quantify the differences between predicted and observed values? 

3. Line 219, The observations clearly show a shift from increasing to decreasing POx with increasing NOx at approximately 6 ppb. However, the model did not capture this feature at all. What could be the reason for this? In addition, many discrepancies between simulations and observations in this study have been explained by insufficient model resolution. Could this be further addressed through the use of high-resolution models?

4. After scaling VOCs, the model simulations of NOx concentrations tend to be underestimated during the day and overestimated at night in Fig. 4b. Could you provide a more detailed explanation for this phenomenon?

5. The model does not fully capture the vertical profiles of PNs in the observation in Fig. 6. Is this due to an underestimation in the VOC emissions?

6. Line 370, Please verify the absence of Figure 2d in the article.

7. Line 230 The word "tha" appears to be a spelling error. 

8. Figure 1 lacks a serial number indicating its order.

Travis et al. examine the impact of modified representation of VOC emissions and production of NOx reservoirs on modeled urban ozone production. The modifications are largely informed by the interpretation of KORUS-AQ observation campaign. Their findings reveal that significant augmentation of anthropogenic emissions is necessary to align with the observed VOC species and calculated OH reactivity. The enhanced VOCs also partly reduced the underestimation of PAN, but the underestimation of PNs is not resolved. The results emphasize the need for improved measurement and research into OVOCs emissions and chemistry to enhance the model's capacity to replicate PNs and ANs.

The study is surely significant and aligns with the scope of ACP. Although I do not have specific criticisms regarding the data analysis and interpretation, I do have two primary concerns that I would like the authors to address before I can recommend the manuscript for publication.

First, the modification to emission inventory and “improved” chemistry scheme is largely relies on the KORUS-AQ campaign. In the abstract, the authors state that the study is motivated the fact that “during KORUS-AQ, air quality models underestimated ozone, formaldehyde, and peroxyacetyl nitrate (PAN) indicating an underestimate of VOCs in the emissions inventory”. However, the GEOS-Chem model has been extensively applied to diverse regions globally, each with distinct chemical environments. For example, numerous studies have demonstrated that the standard version of the GEOS-Chem model (particularly versions after 12.0.0) tends to overestimate surface ozone concentrations in other polluted regions, such as China. This implies that the new chemistry may not fit other regions. My primary concern is whether the new chemical schemes introduced in this study, which might be integrated into the standard GEOS-Chem scheme but heavily rely on the findings from the KORUS-AQ campaign, will have universality for other regions. This issue needs to be evaluated before the scheme can be considered for adoption in GEOS-Chem. Otherwise, these improvements may appear to be tailored solely for the region constrained by KORUS-AQ studies.

Figure 8 and related discussions are a nice way to present the global impact of the added PNs chemistry on global chemistry. Can you also show the overall impact on surface ozone due to the modification of chemistry？

Secondly, the scaling of anthropogenic VOCs emissions appears to be somewhat arbitrary (Lines 158-166). While I acknowledge that the most of the model bias towards VOCs should be attributed to the emission inventory, the magnitude of the scaling factor will impact the interpretation of the effectiveness of the modified chemistry. I look forward to a more rigorous consideration of the emission scaling factor.

Minor comments.

Line 29: It is also important to note the version of GEOS-Chem model here.

Figure 5: suggest add the description to the title of the figure.