General comments:

This paper presents a study to quantify the variability, source and transport of Ethane (C2H6) using solar absorption measurements from a ground-based high resolution Fourier transform infrared spectrometer performed over Hefei, eastern China. The measurements of five years (2015 - 2020) have been used to evaluate the GEOS-Chem model simulations, as well as to analyze dependencies on meteorological and emission factors using generalized additive models. Finally, the authors highlighted the sensitivities of model results to quantify relative contributions of various source categories and regions to the observed C2H6 abundances.

I have some concerns, which are mentioned below in the major and minor comments section. I recommend the publication of the manuscript after these points are addressed.

Major comments:

P7 L23: It is hard to see the difference in the C2H6 emission distribution from the plots of Fig. A1. Perhaps, showing the relative difference w.r.t. the annual mean would be a good way to highlight the seasonal change.

P10 L4: the overestimation of 17.4% in December is not so evident from the figure. Please give more information on how this is calculated?

P11 L40: the values do not match the figure, e.g., biogenic, please verify the absolute and relative contribution for other components as well.

P11 L41: these numbers should change based on the corrections done for the above comment.

Figure 3: Please provide some explanation on why the measurement uncertainties are lower during the summer time and vice-a-versa.

Minor comments:

P1 L32: together with atmospheric modelling

P1 L36: no brackets needed

P3 L4: ... namely the infrared working group (IRWG) of the Network for the Detection of ...

P4 L19: please provide the spectral range of the NIR and MIR observations.

P5 L30: shouldn't this be table 1?

P6 L7: zero level uncertainty reported is different from the value in table 1

P14 L36: and -3.93%, respectively

This study by Sun et al., for the first time, presents and quantifies the variability, source, and transport of C₂H₆ over densely populated and industrialized eastern China by using ground-based high resolution FTIR observation, GEOS-Chem model simulation, and the analysis of meteorological fields. The dependencies of C₂H₆ on meteorological factors and co-emitted gases are also analyzed by using generalized additive models (GAMs). The ground-based FTIR C₂H₆ time series are applied to evaluate the GEOS-Chem model for the specifics of C₂H₆ simulation over eastern China. The authors further run a series of GEOS-Chem sensitivity simulations to quantify relative contributions of various source categories and regions to the observed C₂H₆ abundance. They also conclude that there is a decreasing change rate in C₂H₆ due to the decrease in local and transported C₂H₆ emissions, which points to air quality improvement in China in recent years. The three-dimensional (3D) transport inflow and outflow pathways of C₂H₆ over the observation site are finally determined by the GEOS-Chem sensitivity simulations and the analysis of meteorological fields. Overall, this manuscript is well written, structured, and its topic fits well in the scope of ACP. I believe that the results of this study could be of interest to the general atmospheric science community and should be in the literature. However, a couple of minor points that should be corrected/clarified before publication.

Specific comments

1. GEOS-Chem is a powerful tool for source attribution of atmospheric composition; however, I feel that the way the authors implemented the model raises misleading and should be clarified. In Section 2.2, the GEOS-Chem model setup is described. On L25-26, the authors state a 1hr time step for surface variables and boundary layer heights. I am not sure what surface variables are in this case or is the boundary layer time step? I am guessing these are the emissions and boundary layer mixing time steps? Additionally, given the importance of the boundary layer in this studies, the authors should state which boundary layer mixing scheme was used. The authors also state a 3hr time step of all other variables. Is this referring to transport and chemistry time steps? If so, this seems exceedingly long, especially for the full-chemistry simulation. All these should be clearly described or clarified.
2. I noticed in Figure 3 that there is a mismatch in terms of time coverage between GEOS-Chem and FTIR observations. The time series of GEOS-Chem is about one year less than the FTIR observations. Is it possible to extend the time series of GEOS-Chem to match the FTIR observations?

Technical comments

1. Please provide correlation and error budget figures for the validations of GEOS-Chem model and GAMs model. I suggest it can be added to the supplement.
2. Figure 1: Please add (a), (b) and (c) to 3 subplots, and explained it in the caption.
3. Figure 3: short data gaps of up to a few months have occurred between 2016 and 2017. Please explain the reason. Is this due to data quality control?
4. The atmospheric circulation pattern technique mentioned in this study is actually the analysis of the meteorological fields. So for clarity, please replace all atmospheric circulation pattern technique terms with the analysis of the meteorological fields.
5. With respect to language, the text is in my impression occasionally penetrated with incorrect/awkward phrases. For example, from my perspective, “Conclusion” rather than “Summary and conclusion” is sufficient for the title of section 6. I am not a native speaker, therefore I did not attempt to correct all these flaws throughout the whole paper. Instead, I would recommend a linguistic revision of the whole text: I assume that either one of the coauthors with a good command of the English language or ACP can provide support for this task.