The manuscript entitled " Exploring the variations in ambient BTEX in urban Europe and its environmental health implications" aimed to offer essential insights into BTEX pollution in European urban environments. However, No more new idea and deep insights about the pollution characteristics, as well as the environmental risk of BTEX pollution were put forth in the manuscript. In my opinion, the current manuscript can't be recommended for publication in Atmospheric Chemistry and Physics.

Specific comments:

1. The first is that the method is not innovative (only analyzing spatio-temporal changes, characteristic ratios, and health risks).
2. The second is that the conclusion is also not innovative (the main source is not quantitative, many of which are the results of previous research).
3. Thirdly, the results were relatively arbitrary and did not take into account the influence of various factors on the external field. For example, in the section of Line 182-187, it is not very appropriate to conclude that transportation and industrial activities are the main sources of BTEX pollutants in the study area based solely on the different concentrations of different types of pollutants, as the meteorological conditions in these locations may vary greatly.

The manuscript, while providing and overview over several BTEX datasets in one place to me appears to be lacking depth in the analysis and truly novel results. To me it appears that the paper is more of a measurement report paper. Hence while I am not convinced about novel findings related to atmospheric chemistry, I do believe it can be considered as a measurement report.

However, I have several concerns regarding the statistical analysis presented:

It is clear from the box plots in Figures in the supplement and main text that the data does not follow a normal distribution at any of the sites. The mean is always significantly higher than the median. It is very likely that all these distributions will fail a normality test when subjected to one hence statistical parameters used to represent the data must be robust parameters.

My statistics and data analysis related concerns are as follows:

1)Average and standard deviation are not the correct statistical parameters to report when data fails the normality test. Instead, the median would be more appropriate than the mean. Sigma can be approximated in a robust manner from the median median deviation MAD (1 sigma = 1.48*MAD). Authors can also report the median and the range of values observed instead.

2) The difference of median between seasons must be assessed with a robust test such as Mann-Whitney test since the data is non-normal. The robust substitute of ANOVA would be a Kruskal-Wallis test for equal median although a Mann-Whitney pairwise comparison between season may turn out to be more revealing and interesting.

3) All the fits in Figure 5 appear to be ordinary least square regressions despite the fact that both x and y axis contain measured data that carries measurement uncertainties. OLR is only suitable when the quantity on the X-Axis is absolute and error free. Please use major axis regression (MA) also known as orthogonal regression to fit your data. Also: You need to give the error of the slope and intercepts. Figure 5B urban panel the line fit for summer doesn’t appear to have much to do with the data at least I can’t see the corresponding circles anywhere near the line. The slope may be driven by some extreme values outside the x and y axis boundaries. Please check your data

My other concerns are es follows

1) Referring to sites with the site code instead of the name of the city makes the manuscript hard to follow. It would be nicer to refer to them with the name of the city and nature of the site even if it costs a few more words.

2)Please make all comparison qualitative: E.g. a statement like

“ Nonetheless, compared to the period before 2000, the levels of benzene and other BTEX compounds have shown a decreasing trend due to the successful implementation of air quality measures in Greece, such as the extension of metro lines and the use of catalytic converters in cars” Implies a time series trend analysis was done which I can’t find in the results section. So it appears to be a comparison between values reported in this study and a previous study. Then why not state clearly what the was then, and is now.

3)Please highlight and explain interesting data instead of only focusing on the data that meats the expectations of BTEX=traffic & Industry. Why is the B/T rations in urban Helsinki and suburban Paris so high?

4)Why are B/T ratio’s higher at industrial sites. While toluene is known to be an industrial solvent significant toluene emission should actually lower the ratio. Industrial benzene use has supposedly been phased out due to the carcinogenic nature. Can the authors comment on the industrial benzene source? Alternatively, is this actually representative of industrial sites in Europe in general since 2/3 industrial sites are in Lyon. This could be related to local emissions from a specific industry.

Minor corrections

Line 109-113 BTEX became TEX please correct