General Comments:

This rich hydrologic dataset will be useful for future analyses of long-term changes in hydrologic functioning in the DMC catchment. The fact that it captured a significant drought period makes it especially unique. The data could also be used for inter-site comparisons to better understand how landscape characteristics influence hydrologic behaviour in response to disturbance (e.g., climatic extremes).

The hydrologic data are useable in the current format and size, although would require significant data wrangling to prepare for specific analyses. The GIS data are useable in the current format and size. I provide some specific comments below to improve the accessibility and useability of the hydrologic data.

The article is well structured and clear. The data collection and analyses methods are described in sufficient detail for the most part. All references to other data sets or papers seem appropriate. I have provided minor technical comments below that could help to clarify certain aspects of the data collection and analyses methods. The figures and tables are high quality and I only have a few minor suggestions for improving them.

Despite my suggestions for improvement, by reading the article and downloading the dataset I am confident that I could use the data.

Specific Comments:

1. The majority of the digital object identifiers provided in the manuscript take the reader to the Demnitz_Isotope_Hydrological_Data page (number 623) on the IGP Freshwater Research and Environmental Database website. There a description of the available data is given; however, it is not possible to download the data here. In the ‘Data files’ section, it says that a user needs to make a request to the three contacts to access the file. However, in section 7 of the manuscript, the reader is directed to data package 622, where the data can be downloaded. I recommend that the ‘Data files’ – Actions column on package 623 be updated with a link to 622. Or to make the process even more streamlined, just eliminate 623 altogether. It’s not clear why there are two data packages.
2. There are no apparent inconsistencies in the data and no unusual data formats, however, I think there could be some improvements in the way the data is organized that could reduce the pre-processing time for a user. My recommendations for improving the data organization are to:
   1. Make sure that all columns have a header (e.g., on a few sheets the year, date, or date_time header is missing, colP header on ‘Forest_A_Isotopes’ is missing, etc.).
   2. Make sure to standardize the column headers across sheets and I recommend using lower case with an underscore separator between words to keep the headers simple (e.g., don’t use different types of brackets around the units).
   3. Consider integrating different levels of column headers (e.g., instead of ‘Average Moisture [m3/m3]’ and ‘20cm’, use ‘avgsm_m3_m3_20cm’… or something like that). On the readme page the authors could create a data dictionary that translates these abbreviated headers for the user.
   4. Move all the ‘readme’ textboxes off of the data sheets and onto a single readme sheet (use the ‘General Readme’ sheet).
   5. Use ISO 8601 date format consistently on all sheets.
   6. Make sure that blank cells in data columns are replaced with ‘NaN’.

Technical Comments:

38: Remove one of the ‘long-term’.

43: Consider ‘… to investigate links between…’.

46: Consider ‘… importance of links between…’.

50-1: I think the quotation marks should go around “blue water” and “green water”, not just around the colours.

53: What do the authors mean by robust data sets? I am familiar with the concept of robust statistics, but unsure what constitutes a robust data set.

81: NGP only used three times – consider just spelling it out.

93: Spell out northeast.

93: Consider removing ‘its integrative characteristics;’.

103: I recommend explaining here what the characteristics of a drought sensitive region are. The authors could also consider mentioning whether or not the data can potentially be used to understand the hydrologic functioning of other drought sensitive regions beyond northeast Germany.

Fig. 1: Instead of one symbol for all measurements, I recommend using different symbols to represent meteorological, soil, sap flow, groundwater, and isotope measurements.

134: Consider replacing ‘… historic evolution…’ with ‘… history…’.

141: ‘… high (~90 %) proportions of evapotranspiration…’ is a bit confusing – consider ‘… high water losses due to evapotranspiration (~ 90 % of total precipitation)…’.

141-3: Here, the authors explain some of the characteristics of a drought sensitive region. I recommend these be mentioned earlier when the concept is first introduced.

150: Was it net radiation that was measured?

157: Were standard rain gauges used? If so, replace ‘simple’ with ‘standard’.

179-80: Consider reorganizing the sentence, e.g., Water samples were also collected at Bruchmill using an autosampler (ISCO…) and analyzed for stable water isotopes.

197-8: Are these nested sites marked on the map? If so, it needs to be clear which sites they are.

200-1: I recommend putting this information on groundwater levels in the previous section.

202: Were any twigs sampled or did they have to have certain characteristics?

207-8: What was the average and max storage time?

209: Either introduce these abbreviations when the water source is first mentioned in the isotope section or add the water source in front of each abbreviation in the brackets.

211: Replace ‘CDRS’ with ‘CRDS’.

212: Consider revising ‘… to direct liquid water equilibrium method for soil water’ to ‘… soil water extracted via the direct liquid water equilibrium method’.

Table 2: Instead of repeating the abbreviations in brackets after dates, consider creating sub-rows from the data type column to the far-right column of the table, i.e., a sub-row for each data type.

Table 2: I think for some of the spatial resolution entries the direction (vertical or horizontal) should be given.

Figure 2: In the caption consider replacing ‘Spatial data availability…’ with ‘Measurement period for each parameter at each site…’.

228: Consider ‘data’ instead of ‘amount’.

241: Consider ‘hydrologic’ instead of ‘response’.

243-4: Consider replacing ‘Variability was also more sensitive under forested land cover…’ with ‘There was higher variability in volumetric soil moisture under forested land cover…’.

Figure 4: It looks like the winter 2020 peak groundwater level at both sites is slightly higher than 2019. Could this suggest recovery?

Figure 4: I’m not sure if this really matters, but the panels are introduced out of order in the caption.

Figure 5: The panels could be stretched out a bit more vertically. This would especially help with seeing some of the detail in panel (c). The titles above each panel could be repositioned beside the a), b), c), etc. to achieve this.

312: There was not a lot of information given in the methods on data quality control. I recommend including the accuracy of all the measurements in the dataset.

Tetzlaff et al. have posted a strong dataset with hydrologic and isotopic data from several compartments representing blue water and green water from a mixed land use catchment in northeast Germany. The dataset is strong and useful – I could not benefit fully due to problems I had accessing the metadata. While the data quality appeared to be strong they could be better organized and made to be more intuitive to follow with better column labels. Even if explanations are present in the metadata, more intuitive headings could avoid making the search. Reviewer 1 did a thorough job of noting these instances, so some of my comments may overlap. I’ve divided my comments onto editorial comments on the text followed by comments pertaining to the data.

Editorial suggestions:

General: This is described as a 2-year dataset but there were references to groundwater levels and stream discharge from 20 and even 40 years ago. The text made no mention that the older data would be posted, but they are there on the website. It would be good to mention up front that you focused on the richer, newer data, but these “historical perspective” data are also posted, and particularly with the unusual drought during these two years, the older hydrologic data provide nice context.

Line 22. Suggest to delete “though.”

Line 23. Parallel construction: “we” should consistently be within or outside of each numbered clause

Line 82. Drop “s” from “lowlands.”

Line 92. Fix “data of two years of data….”

Line 155. Suggest either ‘Measurements were made’ or ‘Throughfall was collected.’

Line 156. 10-m

Line 160. “18 sensors”

Line 170-171. Suggest “Water level measurements were established….”

Line 210. Spell check: “being”

Line 226. Delete comma after “including”

Line 236. Change “is” to “in”

Fig. 2b. Start of record, July 2018 – are these good data? Throughfall >>rainfall only for this short period?

Line 243-244. more sensitive = greater?

Line 253. Consider simplifying sentence by deleting “despite” and “this”

Line 278. Variation damped?  (not samples)

Line 301. data are

Line 310. Spell check: “complementary”

Data:

Like Reviewer 1, I had some confusion over the data versions. My comments were originally on the file in 623, but then I saw the author response to Reviewer 1 that site 622 had the revised dataset. The two versions looked the same to me, but if my comments are off it could be a version confusion issue.

Unlike Reviewer 1 I could not download any metadata.  Both the .eml and.xml file downloads were blocked.

The General “readme” page at Site 622 says all sample times are CET, but makes no mention of accommodations (or not) for the summer time change. This may be clarified elsewhere but should be stated here.

In the Hasenfelde Met_data tab, each parameter has an average, max, and min value for each hour (nice!), but the wind speed and direction are not specified. Are these averages or instantaneous values on the hour?

The precipitation isotope tab states (in readme text in upper left) that aside from the daily Hasenfelde site, other sites might be event or weekly – is there any indication of which samples are weekly? This applies to Throughfall as well, its tab states throughfall was collected on the same schedule as precip.

On the Discharge tab, for Denmitz Mill, using the rating curve in cell F5, I can’t reproduce the flows from the water levels.

In the Discharge_Isotope tab, the readme upper left says sample type is specified in site name but it is actually in the Type: field in the header for each site. And in this field (Row2), Grab” in grab sample is misspelled.

Groundwater tab: “Div/0 error in cells for GW4, March 2019. I don’t understand readme note in upper left corner about only showing data for 2018, as all sites go to 2020. Duplicate identical data columns could be removed.

Forest A and Grass A SMC tabs – These tabs are hard to decipher – headers are minimal, sites are overly abbreviated where there is ample space to expand, and there are multiple columns with identical labels. Even if these are described in metadata elsewhere there should be more information directly in the tab to help decipher what is being presented. Is there a difference between Soil Moisture and Average Soil Moisture?

Forest A and Grass A Isotope tabs – Reporting of units in column headings is uneven, for instance columns J and K, grav_soil_moist (no units given) and water content g (just g? Not per unit volume?)

Vegetation_isotopes tab, row 21 has data beyond where headers stop.