Dear authors,

Thank you for your careful consideration of both the suggestions raised by the reviewers, and their concerns with respect to the scope of Climate of the Past. Your contribution to the understanding of processes controlling the variability of precipitation isotopic composition in Japan on intra to inter-annual timescales is fully relevant for the "proxy calibration" scope of the journal, and I appreciate your effort to stress the implications of your results for the documentation of past changes in the East Asian monsoon.

I am therefore pleased to accept the publication of your manuscript in Climate of the Past, with however the request for minor modifications which are listed in the attached document.

You need to revise the abstract which is not sufficiently informative about the tools and methods developed in your manuscript. Please mention the new observation data (precipitation at the event scale, vapour), the existing data (GNIP monthly precipitation isotopic composition data) and their combined use with an analysis of air mass trajectories and upstream precipitation.

I also recommend to change the writing of the verbs in the manuscript, so that all your work is reported in the present tense, while the description of single past events is reported in the past tense. At the moment, it is too incoherent.

I have one question about the authors' conclusions on the use of surface vapour isotopic data. It seems to me that they conclude that the vapour signal is (i) strongly affected by precipitation reevaporation, and (ii) suggests a lack of strong signal associated with different oceanic sources and transport paths. However, most of the detailed interpretation is based on the precipitation data. This could deserve a word of conclusion, in the results, or in the main conclusions.

I also provide a suggestion for the last sentence of the conclusions.

In Figure 1 and Figure 10 captions, please change "counter interval" to "contour interval". In Figure 8 caption, please replace "precipitation" by "Precipitation". In Figure 9 caption, please change "Anomalous plot" to "Anomalies of". In the panels of Figure 9, the number of significant digits for correlation coefficients should be reduced to 2 digits (0.48 instead of 0.483).

With these improvements, please provide a revised manuscript which clearly shows the changes which have been implemented, and, after a brief verification, it will be accepted for publication.

Dear authors,

Thank you for your final revisions. I am pleased to accept your manuscript for publication in Climate of the Past. Sincerely,

Valérie.