General Comments

In their manuscript “Local-Scale uncertainty of seasonal mean and extreme values of in-situ snow depth and snow fall measurements” Buchmann et al. present a detailed analysis of local bias in snow depth measurements and snow climate indicators from a set of 30 station pairs. They provide a comprehensive overview on differences between snow measurements for different winter periods to half year season of an up to 77 year long time series. They discuss the results in relation to climate trends and local impact of station surroundings on snow accumulation. The work gives a good basis for estimating the uncertainty in snow related parameters induced by local station setting. In general the manuscript is concisely written and the results are appropriately discussed. Considering the supplementary material the presentation of the results in figures and tables is substantial. Just the discussion and conclusion section should be revised for better comprehensibility (see specific and detailed comment).

Specific Comments

(1) I would suggest to change the title to “Local bias of...” and to avoid the term “uncertainty” here and in the discussion. “Local bias” is used most in the manuscript, and is in terms of locally specific characteristics in snow accumulation at the station just a part of the overall uncertainty in local snow measurements in addition to e.g. measurement errors. A complete analysis of the uncertainty in terms of variability of the local bias might include not only the mean values, but also max/min/range/sigma of the seasonal values (see detailed comments/ L98). “Local variance” or “Local variation” might by another options more presenting the local effects of station setting as mentioned in the manuscript (e.g. soil, buildings, …).

(2) The presented analysis is an update of the work published by Buchmann et al. (2021) with an increased number of station pairs for a longer time period. Nevertheless, basic methods are the same, and results do not differ fundamentally. I recommend the authors to accordingly present the new findings and differences between both studies.

(3) In the current version of the manuscript, results and discussion are presented right together. I suggest to separate both and to align the discussion linking the different results (some biases are related to the same source of error). Further, the conclusion are rather to long and can be shortened by several sentences (see specific comments) to be more condensed. The conclusion might present aspects on the order of local biases, that might be negligible and what the local station bias means for the interpretation of climatologies (trends) based on single (mountain) stations.

Detailed Comments (L = line)

L10: Remove daily scale and change to: “Daily measurements of snow depth...”

L22: Here and throughout the manuscript: for me it was difficult to follow with the shortening with the brackets. At least it stopped the flow in reading. Please consider to present in two separate sentences.

L47: Here and throughout the manuscript: there appears on redundant pair of brackets in the citation using e.g. L48: Please add “half year” in front of period

L56ff: the research questions are announced but not formulated as such. In addition methods are presented here (point 2). Please rephrase this part to list the analysed parameters or formulate the RQs

L72: remove the brackets and write: “...within a distance of 3 km horizontally and 150 m vertically...”

L72/73: 1 km: Please fix the space between number and unit such that it is not seperated at the line break (here and in the entire manuscript)

L74: 1770m: This might be a high elevation for station pairs, but it is not in context of mountain stations and snow measurements. Would be nice to see a discussion if and how the results are expected to be scaleable to higher located mountain stations.

L75/76: Shift the dates of the 77 year time period (1943 to 2020 ;

L76) directly to this number (L75). End the sentence after “station-pairs” in L76.

L86: since all measurements are made in the past, please present more detail here. L98: since time periods are presented, the analysis of the distribution of seasonal/monthly values within the time span would be of interest, too. Please consider to give those numbers.

L104: To my knowledge the standard hydrological year of mid Europe is defined by 1.10. to 31.09.. Why did you use the period from 1.9. Please give at least a citation.

L121: Consider to use an abbreviation of “local bias” for further use in the manuscript

L138: The finding that all outliers are produced by the same station pairs should be discussed in terms, if such station-pairs are representative at all.

Figure 1: The RPD for all HN parameters is between 10 and 20%. Please present this in the results. It would be interesting to see the statistics on the variability of the parameters when moving a 30-year period over the entire time series of the three station pairs with most parallel years.

L152: Replace “entire” ba “half-year”

L207: Replace “uncertainty” by “local bias”

L208: Figure A2: Please consider to put the Figure in the manuscript.

L219: I did not get the sentence starting with “Among...”. Please rephrase.

Figure 4: Here and in the text: Consider to use Dend instead of Dstop.

L238-244: This is an good example to open a separate discussion

Figure 5: What does the error-bars show? Please present in the caption.

L259: “Analysis....” This is a result and has not to be mentioned in the conclusions again. Please revise the conclusion section and avoid to present redundant results and discussions.

L264: Which estimation uncertainty is meant here?

L291: What is the threshold to neglect local biases? This might be an interesting discussion for the discussion sector, too.

References: Buchmann, Moritz, Aschauer, Johannes, Begert, Michael, and Marty, Christoph: Snow climate indicators derived from parallel manuel snow measurements, https://doi.org/10.16904/envidat.218, 2021.

The submitted paper by Buchmann et al. describes a methodology and results of a study to assess the impact that local-scale variability in snow depth and snowfall measurements have on determining seasonal mean and extreme indices. The study uses 30 co-located (within 3 km horizontally and 150 m vertically) pairs of manual measurements throughout Switzerland where the pairs of stations are operated by independent institutions (SLF and MeteoSwiss). The authors show the differences in seasonal mean snow depth, sum (accumulation) of new snow, maximum 3-day new snow sum, days with snow on the ground and days with snowfall when they are calculated with the co-located measurements. In general, the variability varies by site and the period used to calculate the indices. The authors attempted to examine some potential causes of the differences, such as sunshine hours during ablation, but concluded that the metadata records are inadequate for any detailed or conclusive results.

The paper provides some insight into the potential variability in local-area climate indices that users can expect due to measurement station location and local-scale variability in snow cover properties, and the potential pitfalls of extrapolating point-measurement-derived indices to the regional or landscape scale. The paper is interesting and relatively well written. I do have a couple of concerns that should be addressed before this paper can be published in TC. My major concerns are as follows:

1. I have concerns about the use of the term “uncertainty”, largely in the title, abstract, and conclusions. I don’t consider myself a metrology expert, but to me, “uncertainty” is a metric attached to a measurement to inform the user of the range of values to be expected when the measurement is made with respect to what the true value actually is. Therefore, each manual measurement presented in this paper would have an attached uncertainty, and that uncertainty would contribute to the overall uncertainty in the calculation of seasonal climate indices. However, the more appropriate terminology for what is actually being assessed here is “variability”, or specifically, the impact of spatial variability on the indices. This suggested revision doesn’t impact the interpretation of the results (in fact, the term “uncertainty” is really only used in the title, abstract, and conclusion, and not in the results) so updating these sections with more appropriate terminology should be a relatively easy revision.
2. The paper presents some insight into the impact of local-scale variability of snow cover measurements on seasonal climate-related indices and offers some explanation as to why snow cover measurements can be quite variable in space. I believe that it is implied, but both the authors and the readers need to understand that it is highly unlikely that two measurements can adequately assess local-scale variability. I suggest that this point be clearly made (with references where appropriate) so as not to accidentally mislead the reader.

More minor and specific issues and suggested revisions are embedded in the attached annotated document.

Buchmann et al. are presenting an in-depth analysis of a large set of closely located station pairs with parallel snow measurement in Switzerland. Most of the sites have several decades of parallel measurements which gives valuable insight in possible differences, caused by the impact of the station environment in contrast to the climate/region the stations may represent.

The achieved results of the presented statistical analysis give an estimated magnitude of the added uncertainties which can be expected due to (the often unknown) variety in the less-than-ideal siting of the station. This information is valuable for any analysis of long-term snow measurement, especially in case of single located stations, as metadata describing the location for such kind of measurements are often lacking the necessary accuracy or may even be missing entirely. 

The paper is written in an understandable language and the analysis method is described adequately. I do recommend the publication of the study, but do have some concerns and suggestions that should be addressed before publication in TC:

1. The authors introduce a lot of abbreviations for their measured and calculated parameters, station names, time periods (seasons) and analysis methods. While some of those are well-known and their use well-established and justified, are others solely defined in this paper. The shear amount is severely hampering the reading process, as the reader is required to either memorize the abbreviations or searching the meaning of the abbreviations somewhere in the text. This comment is of technical nature and should be easy to address. Please refer to the attached file with specific comments for possible solutions.
2. The description of the results sometimes lacks clarity and accuracy. Obvious exceptions are just very briefly and sometimes not at all mentioned, neither are possible reasons for those exceptions discussed. Please find some concrete examples in the attached file with specific comments.
3. The authors were not able to conclude on the exact causes for the observed differences due to the lack of metadata for their sites. That fact is well described and creating awareness of this kind of problem that may likely exist for other long-term snow measurements is one of the main messages of the paper. I do think, however, that it would be appropriate to recommend a list of necessary metadata to record for ongoing snow measurements and if possible, also give advice on which sites to avoid entirely.
4. Further, as also one of the other reviewers mentions, it is necessary to clearly state that also for the case of parallel measurements of two ideal and neighbored sites, it is likely that significant differences in the described variables may still be observed. I can imagine that some of the still existing neighbored sites (such that metadata could be achieved and described for the more recent measurements) in this study could be used to illustrate, discuss, and possibly quantify these differences and how they compare to the differences found for the long-term analysis presented in this study.

Please find more minor and specific comments and suggestions in the attached file.