The authors have constructed comprehensive emission data of organic compounds from mobile sources by utilizing conventional emission data of nonmethane organic gas (NMOG) and particulate organic matter (OM). The developed emission data include information about volatility and detailed chemical composition of organic compounds, and are important basis for accurate simulation of atmospheric PM and O3. This manuscript is well written, and clearly organized.

I recommend this manuscript for publication after the following minor comments are addressed.

Line 168: I could not follow the procedure for converting OM emissions to CROC emissions. I guess that gas-particle partitioning during filter sampling is critical in this conversion and that sampling temperature and OM concentrations are key parameters. However, you calculate EF_CROC as a function of EF_OM, and you did not consider sampling temperature and OM concentrations in this conversion. Did you assume that EF_OM is proportional to OM concentrations at sources, and that effect of sampling temperature is negligible? I would like this point to be clearly stated.

Line 220: daily-averaged measurements?

Numbering of Figures and Tables should be carefully checked. (e.g., CROC/OM ratio in Line 284 is in Table S6 (not Table S5). Fig. S41 (Line 351) and Fig. S42 (Section S10) are not shown in the Supplementary Material.

This article aims at detailing the speciation of reactive organic compounds in OM and NMOG emissions. This article provides many details on the specification, which may be very useful for other modellers. However, many papers already estimate reactive organic compound specifications from OM and NMOG emissions, and the introduction does not sufficiently explain what is done in the other papers. Also, for clarity, the measurement biases that this article aims to correct should be introduced and explained in a paragraph in the introduction.

Minor comments :

• Line 60 : Is IVOC more impacted than SVOC for filter artefacts ? Please explain why.
• In the introduction line 64 and thereafter, please detail the main assumption currently used in the litterature to estimate IVOC and SVOC. Although IVOC may be specified from NMVOC or GROC, it may also be directly speciated in the measurements (see for example Sarica et al. Env. Pollution, doi:10.1016/j.envpol.2023.121955)
• What is the advantage to define GROC ? In the methodology defined here, GROC/NMOG ratio and IVOC/GROC ratio with speciation for each of them need to be specified. Why is it better than what is usually done, i.e. simply define a speciation for VOC and determe a IVOC/NMVOC ratio ?
• Line 278 : How are estimated the source-specific adjustment factors ? Where are they detailed ? Table S5 details the volatility profiles (which are key properties for SOA modelling). Where do those come from ? What are the incertainties associated to those profiles ?