General comments

This manuscript presents the diurnal isotopic variability of precipitation observed at Rio Carlo in the central-southern part of Brazil, between 2019 -- 2021. The authors collected 90 precipitation samples for isotopic analysis at 5-10 minute intervals and then divided the samples into daytime and nighttime. They found there is a clear diurnal isotopic variation, with lower oxygen isotope ratio and higher d-excess in night than in daytime precipitation in summer. They speculated that the local effects may be a major factor in this diurnal feature, as the HYSPLIT model showed similar trajectories of air masses that brought precipitation during the day and at night. And, with the fact that nighttime precipitation is characterized by higher RH, lower LCL, and higher BT, they concluded that the post-condensation process, such as re-evaporation of raindrops and isotopic exchange, may be responsible for the diurnal isotopic differences.

There is no doubt that the new observations are valuable to the isotope hydrologist and paleoclimatologist. And, personally speaking, their found diurnal isotopic differences of precipitation is very interesting. I agree that the post condensation effect “may” affect the diurnal isotopic variations in summer, but I think there are other factors that contribute to the diurnal variation. I don’t think that just a few hundred meters drop in LCL and an increase in RH from 86 to 97 is sufficient to explain the observed isotopic changes. As noted in this manuscript, isotopes reflect the precipitation history from the upstream. So, similar trajectories between day and night do not guarantee that the precipitation history is identical. Precipitation history can vary with the size and organization of convective clouds, and intense rainfall in night-time suggests that convective activity is more active than during the day. To verify a hypothesis, it must be tested quantitatively, not qualitatively.

Unfortunately, most of the arguments are based on speculation and I can’t find some key explanations. For example, 5-10 min sampling interval means that isotopic data include both Intra-event and Inter-event samples. To detect the “true” diurnal isotopic variations, the number of precipitation events that occur during the night and day is very important. However, I can’t find a description of how many precipitation events are used to calculate the daily and nightly averages. I suspect that the authors do not differentiate between inter-event and intra-event samples. This is very critical of this study. I therefore conclude that the manuscript requires major revision before it is of suitable quality for the publication in ACP. Since it will take longer time for revision and the content of the manuscript will largely change, my evaluation of this manuscript is “reject”.

Other comments

• I can’t understand how many trajectories are used to identify the moisture origin for each precipitation event. What time did you launch the trajectory analysis? How many trajectories do you compute? A single trajectory analysis does not provide reliable moisture source for each precipitation event. And, it is necessary to consider the replenished moisture from the surface during the transport to discuss the moisture sources.
• Introduction of a conceptual model is not a “Results”. Results just show your observed results. Your speculation and conceptual model should be noted in Discussion section. The manuscript should be reorganization before resubmission.
• The same reference was cited with different names. For example, Aemisegger et al. 2015a and Aemisegger et al. 2015b refer to the same paper. Similar mistakes is found for Kurita et al. 2013a and Kurita et al. 2013b. Check the reference carefully.