Dear authors of essd-2020-209,

by now I got around to looking at the second round of review reports. Both referees are very happy with the improvements to the paper. However, they have some remaining concerns regarding the accessibility and description of the dataset itself.

Please consider the very detailed suggestions of the referees and revise the data publication (and related parts in the manuscript) accordingly, so that the dataset can reach its full potential.

Best regards,
Sibylle Hassler

Dear authors of essd-2020-209,

after going through the referee comments and your revised manuscript, and after consultation with the chief and managing editors, we see one issue remaining before we can consider publishing your manuscript:

It is not ESSD data policy to include datasets that can't be readily accessed, and the precipitation data would qualify as such. Also, from the manuscript (and checking back with one of the referees) it is not clear for which parts of the calculations you actually use this data. Is it part of the upscaling? Is it part of the trend analysis in 3.2? There's a mention of "quantifying the coupling strength between precipitation 35 and SM at 10-year scale" in the abstract and conclusions. Did you do this or is this an outlook?

It would be great if you could clarify where exactly you used this data. And if it is necessary. Maybe we can find solutions on how to solve this issue of inaccessability. For example, in previous cases, aggregated values of data from the CMA could be saved in the repository. At the moment, simply stating that the data is (basically) not available, won't be sufficient.

I hope this helps and we can solve this issue together to publish your dataset.

Best regards,
Sibylle Hassler

Dear authors of essd-2020-209,

thank you for addressing the issue with the precipitation dataset from the CMA. However, I still have some concerns regarding this data.

- The description of the hourly data still lacks metadata, additionally it is not clear which data exactly was used for the analysis. Could you please update the User guide in your repository and describe the data acquisition procedure in similar detail as you did for the soil moisture products (how to choose the station you used, which time frame, etc...).

- Manuscript section 3.2: You wrote in your response that the trend calculation for the precipitation data was done similarly to the soil moisture data. However, in the manuscript, there is no mention whatsoever about precipitation data in section 3.2. Please revise this so that it is clear that you did the trend analysis for both soil moisture and precipitation. An obvious way would be to describe the equations in general terms (x instead of SM as a variable) and then describe for which variables you did the analysis.

- Monthly precipitation data: It is a good idea to include a slightly aggregated data product if you can't supply the original data that you used in the analysis. However, it is not clear if monthly data was used at all and where, or if it is simply provided as an informative dataset because the daily values are not accessible. Please make this very clear in the text. Furthermore, there is only very little metadata about this dataset except the coordinates. What kind of instruments were used (maybe there's a standard in the CMA, but it would be good to have more information, similar to the level of detail you provide for soil moisture)? Additionally, it is not clear how you aggregated the data to monthly values. It seems like they are averages of each month (because the numbers are too low compared to your annual precipitation amounts in section 2.1), which would be very unusual for precipitation data. In any case, you should explain why you have done monthly values, give sufficient metadata (comparable to the soil moisture metadata), describe what kind of aggregation you used and why, and where you use this data.

Also, while skimming through the manuscript again, I noticed a number of typos and phrasing issues. Please consult an English native speaker to check for those so that the manuscript can be read easily.

I hope my comments are helpful to revise the manuscript and data product accordingly.

Best regards,
Sibylle Hassler

Dear authors of the Tibetan Plateau observatory soil moisture manuscript,

thank you very much for thoroughly addressing my previous comments about the precipitation data. I think it is much clearer now how and where the data was used.

I have two minor comments that you might address in the next step:
- Please remove the "SM" in line 244, as by now the sentence refers to both SM and precip time series.
- It would be good to also include the mean annual precipitation for the Maqu network in section 2.1.1 as you do it for the other two networks and also provide the monthly data for Maqu now. Could you please add this value?

Best regards,
Sibylle Hassler

Dear authors,

thanks for also addressing these last points.

Best regards,
Sibylle Hassler