This manuscript describes the development of a new module which realizes online consideration of influences of meteorological conditions on emissions used in air quality simulations. An application of this module on vehicle emissions and air quality simulations is also described.

I can understand importance of considering influences of meteorological conditions on emissions. However, the descriptions are very confusing for me. I have difficulties in understanding what was done in this study and what is a scientific significance of the module developed in this study.

MOVES is a model to estimate vehicle emissions as described in the first paragraph in Section 2.3. Its computational requirements are prohibitive in real-time air quality forecasting applications. Therefore, SMOKE-MOVES tool was developed to overcome the issues. It runs SMOKE to estimate air quality model-ready emissions using the MOVES EF LUTs with hourly meteorological inputs as described in the second paragraph in Section 2.3. However, MOVES EF LUT files require significant computational resources, such as memory and storage spaces as described in the third paragraph.

According to Figure 2, MOVES EF LUTs are the starting points in SMOKE-MetEmis and CMAQ-MetEmis. Therefore, I thought that SMOKE-MOVES is inevitable even if MOVES EF LUT files require significant computational resources.

However, the fourth paragraph of Section 3.1 says that the SMOKE-MetEmis can generate a single MetEmis_TBL emissions input file that can represent the 334 MOVES LUTs files and its size is significantly smaller than the size of the 334 MOVES LUTs files.

Does it mean the new module can generate a single MetEmis_TBL emissions input file which can be used instead of 334 MOVES LUTs files? But I cannot find how the new module can generate a single MetEmis_TBL emissions input file without using SMOKE-MOVES and MOVES LUTs files.

The third paragraph of Section 3.1 says that CMAQ-ready gridded daily emissions require approximately 1.9 hours per day while the CMAQ-MetEmis “inline” approach does not cause much computational processing time. Is the emission used in “Base” the one generated by SMOKE-MOVES taking 1.9 hours, and is the emission used in “MetEmis” the one generated by CMAQ-MetEmis taking only 1 minute? Does it mean the emission used in “Base” is less accurate and more time consuming than the emission used in “MetEmis”?

It is important to consider influences of meteorological conditions on emissions. However, if their influences are poorly represented, the model performance could be also poorer when they are considered. Therefore, accuracies of influences of meteorological conditions are critical in term of this study. However, as described in Line 171-175, the dependency of mobile emissions on local meteorology can vary by vehicle types, fuel types, road types, processes, vehicle speed for onroad vehicles, hour of day for off-network vehicles, as well as by pollutants. Uncertainties in them could be quite large. Nevertheless, this manuscript just believed the dependence of vehicle emissions on meteorological conditions represented in MOVES. Their uncertainties should be discussed.

Figure 4 (b) indicates that TOG emissions estimated in “Base” is mostly higher than “MetEmis”. This is not due to an issue of “online” or “offline”. The profile used in “Base” may not be just representative. The performance could become higher if a more representative meteorological profile is provided.

Then, improvement in model performance with the new module is quite small as shown in Table 3. Differences between “Base” and “MetEmis” are within uncertainties in concentrations simulated by chemical transport models.

Large differences can be seen only in the selected episodes. In case of the episode (July 24th, 2019) over San Jose, CA, the maximum VOC concentration is 1263 ppbC as shown in Table 6. This unrealistic high value could be originated in errors in the meteorological profiles used in “Base”.

Due to these issues, I cannot recommend accepting this manuscript in the current form.

This manuscript describes the CMAQ-MetEmis module, which allows to dynamically model the meteorological-induced MOVES onroad mobile emissions inline within the CMAQ air quality modelling system. The CMAQ-MetEmis module addresses the shortcomings (computational time and memory requirements) of the current “offline” approach, which is based on the SMOKE model. The strength of CMAQ-MetEmis is in its ability to take into account the impact of local meteorology on mobile emissions within a reasonable time frame to meet the air quality forecasting time constraints. The manuscript shows how the spatiotemporal enhancements of on-road mobile emissions predicted by “MetEmis” benefit the performance of an air quality modelling system, indicating the importance of dynamically estimate weather-aware mobile emissions. The paper is well written and structured, and its quality is good, which makes it a very good contribution to GMD. I therefore recommend to accept this manuscript for publication once the following comments have been addressed.

1. The meteorological-dependent expressions considered in this study are key to understand the emission and air quality results that are later presented. From my point of view these expressions should be included and briefly described in the manuscript, including also a discussion on the potential uncertainty associated to them.

2. Despite reporting interesting results, the contents of sections 3.3 and 3.4 are sometimes a bit difficult to follow (and not always in line with the proposed titles). I would recommend to restructure this part of the manuscript by: 1) creating a general section entitled e.g., “Effects of Weather-Aware Mobile Emissions on modeling performance”, and then including several subsection describing the differences between the base and MetEmis cases 1) on a general level, 2) at the city level and 3) for specific pollution episodes. Despite reporting a general improvement, authors should emphasize a bit more that these tend to only be relevant for NO2, and highlight that improving statistics of secondary pollutants (O3) or pollutants that mainly consist on secondary species (PM2.5) is more challenging due to the important role of other emission sources and processes (e.g., chemical reactions).

Other comments:

1. Development of CMAQ-MetEmis Coupler (line 215)  Should not this be header of a subsection?

2. Line 250 : “respectively for the winter (January) and winter (July)” (should not be summer?)

3. Legends in Figure 9.b and Figure 12.a are difficult to read and incomplete (what are the dots representing?)