I think that the presentation quality requires substantial improvements in three aspects, as follows.

1) The VOC observed datasets are poorly presented. This manuscript did not thoroughly discuss the trend of speciated VOCs but lumped the species into functional groups. It is impossible to evaluate the model evaluation processes without VOC-specific information. In addition, FID is not a common analytical tool to quantify oxygenated VOCs, which requires a thorough description of the analytical characteristics.

2) The emission inventory's speciation information is not thorough enough. All inventory information is presented in a lumped fashion except for alkane. It is vague how this information was integrated into the box model framework.

3) The sensitivity tests of the box model need to be conducted and thoroughly discussed. The oxidation product accumulation in the box model frame must be verified by comparing it with the observed value. This is particularly important as this study concludes that OVOCs play an important role in ozone production. Therefore, it is important to present a quantitative discussion of how much of OVOCs in the studied area is coming from direct emission vs photochemical production. 

The manuscript entitled " The impact multi-decadal of changes in VOCs speciation on urban ozone chemistry: A case study in Birmingham, United Kingdom" aimed to quantify the impacts of the real-world changes in VOCs sources on urban O3 production rate, and also evaluate the relative importance of different VOCs functional group classes on the O₃ production. The manuscript provides some valuable advice for the pollution control of O₃ in the area with relative clean air quality. I therefore suggest a necessary revision of this manuscript before final publication in Atmospheric Chemistry and Physics.

1. The innovation of the paper needs further unearthing, as there are already many similar literatures;
2. In this study only 38 VOCs species were detected, which was much less than the photochemical species requirements of PAMS and also not conducive to the operation of the OBM model. It is also necessary for the author to explain the quality control of the online VOCs monitoring instrument.
3. Is the research based on case study representative in evaluating photochemical pollution in a certain region?
4. The research result showed that road transport played the most important VOC role in the case study ozone photochemistry despite it only contributing 3.3% of national VOCs emissions in 2019. Since the proportion of emissions from road transportation is so limited, how to further control?