This is an interesting and well written paper, discussing water vapour exchange between the boundary layer and free troposphere over China, using budgeting techniques.

My main concern is whether the results and conclusions presented are significant enough and of sufficient general interest. In a revised version of the manuscript I would encourage the authors to really focus on ensuring the abstract and conclusions highlight what they feel are the key new findings presented by their work, and what the general interest of these are (i.e. a wider ranging interest than just local meteorology over China). This may require further work on some aspects of the paper that were tantalising but lacked depth. For example:

• Sect 3.3 - do the authors have a suggestion why the water vapour exchange changes in the way it does due to ENSO - what is the mechanism for this relationship? i.e. what components and inputs of the water budget are altered, and why? I think this is one of the key general interest parts of the paper, but it requires more in-depth discussion than merely presenting correlations, given you've gone to the effort of devising the budgeting tools that allow you to answer the questions of why the correlations exist.
• L507 - I accept that a complete analysis here is beyond the scope of this study, but I think the paper would be strengthened by at least providing some discussion of the mechanisms responsible for the water vapour transport results presented.

Minor comments:

• L60 - should be cancelling (not canceling)
• L80 - Boutle et al (2011, QJ, doi:10.1002/qj.783) is possibly a better reference here
• L125 - would be better to refer to "grid length" than resolution, since that is what is quoted (the resolution is ~5 times the grid-length!)
• L225 - "variables were extracted from the vertical level closest to the top of the ABL" - why not interpolate the variables to the ABL height, using the levels either side. I'm slightly worried that choosing the closest level may give significant differences if that level is above or below the BL top, i.e. are you always choosing free tropospheric values or always choosing ABL values, or a random mix of the two? And does this matter?

The manuscript discusses the water vapor exchange between the atmospheric boundary layer (ABL) and free troposphere (FT). The water vapor exchange between the ABL and FT is an important phenomenon related to e.g. precipitation, clouds, tropical cyclone formation etc. Therefore, it is quite important to improve the understanding of the water vapor exchange. The authors are using WRF simulations for seven years to study the phenomena. They model and the parameterization schemes used are well evaluated against meteorological observations. The manuscript is very well written and structured and, in my opinion, it is quite easy for the reader to follow. The structure and results are already quite good, and the methods used are well evaluated, the results are discussed and compared well to existing literature and also the uncertainties of the results are discussed well. Therefore, I have only minor suggestions before I can suggest the publishing of the paper. Her are my detailed comments:

1. Figure 2: Do you have an explanation why the model seems to be underestimating the ABLH in winter months, but overestimating during the summer months?
2. Figure 2 (+others): It would be good for the reader to point out in the caption, that the winter panel and the summer panels have different scales in y-axes.
3. P8 L239: Please give proper citation for the ECMWF data used in the study.
4. P9 L273: Do you mean Sect 3.2 in stead of Sect. 3b?
5. Figure 8: Is the map showing only statistically significant grids? If not what percentage of the grids were significant? Were there any spatial variation of the significancy?
6. P16 L458–459: Which section do you refer to with Sect. 3a?
7. Summary: Even if it is good to have some sort of summary of the results, I would prefer (also or instead of summary) a short conclusions section that would also point out the most important findings of this study. In addition, it should be also clearly pointed out in the abstract.