The current version of the manuscript ‘Operational and experimental snow observation systems in the upper Rofental: data from 2017 to 2023’ describes a dataset collected at three sites in the Rofental in the Austrian Alps. The dataset contains standard measurements as well as some more experimental sensor setups for snow measurements, e.g., to record blowing snow events at one site.

The manuscript extends the ESSD paper ‘The Rofental: a high Alpine research basin (1890–3770ma.s.l.) in the Oetztal Alps (Austria) with over 150 years of hydrometeorological and glaciological observations’ by Strasser et al. 2018. Moreover, it was already submitted in a previous version to TDC in 2021 with a dataset until 2020, which was, however, at this time too short. This dataset was now extended with three more years, which makes the dataset in my opinion sufficiently long for publication in ESSD.

In general, the Rofental research catchment is a great site for high-alpine research. As such freely available datasets are still very scarse, this dataset can be very helpful for snow monitoring and the development and testing of snow models in high-alpine regions. The dataset is easily accessible and is fully described in the manuscript; data gaps are mentioned and discussed.

The manuscript in its current version has been significantly improved by the reviewing process in 2021 compared to its previous version (not published).

I have only a few minor points:

p.7, l.160f: ‘The station comprises a large set of operational snow cover sensors’. In this context, the word ‘large’ seems to be a bit too much. à ‘The station comprises several operational snow cover sensors’

p.10, Figure 4: Please indicate a line at the year 2017 as the dataset for this paper starts in this year.

p.11, Figure 13 (and further figures): It is good that you keep the same colour for individual stations throughout the manuscript. However, yellow might be a too light colour.

p.12, l. 262: Mention that the season 2019/2020 is an example in sub-section 7.2. Why do you not show further seasons?

p.17, Figure 11: I would choose other colours here than the colours, you chose for representing data of specific stations.

p.18, l.369: ‘…over six winter seasons’. This is not the case for all sensors. Please add this information also in the conclusions.

This paper documents an excellent and important dataset for analyzing mountain snow processes.  I highly recommend the publication of this paper and have only minor comments below.  The only detractions I have in the score above on originality stem from the fact that some of this is an extension of a previous paper (albeit over a long enough time period and with enough new data that it is well worth publishing), and on presentation quality there are a few english grammar changes that could be updated in the proof stage. 

1) In the context of alpine catchment hydrology, please document the vegetation cover (and lack there of) in most of this basin, there are trees and grasses at lower elevations, if not where these data were collected. 

2) Is there a flag provided in the data to note times when instruments are covered? 

3) Can you document how well the point observations of snow depth represent the surrounding snow depth at all sites?  Clearly at the exposed/sheltered site pair, there is a lot of variability. Is that true at all sites? 

4) Are the 10 minute data an average of higher frequency measurements over the prior 10 minutes? The surrounding +/- 5 min period? or are they instantaneous? 

5) When replacing sensors, was any cross-validation/comparison made between old and new sensors? 

6) What is done as part of the "thorough check for obvious errors"?  Is this primarily thresholds and change thresholds? Are the "corrected" values flagged as such? 

I enjoyed the evaluation of more experimental measurements of snow flux, snow density, and drift occurence.  The authors might want to look at a similar dataset that was recently published and included snow particle flux as well as ~5 minute repeat scans with terrestrial scanning lidar during many such blowing snow events (Lundquist et al 2024 https://doi.org/10.1175/BAMS-D-23-0191.1)

This manuscript presents a comprehensive data set for three research sites in the Upper Rofental. The time series presented comprise data from 2017 to 2023 and come from three meteorological and snow-hydrological stations. The manuscript extends a ESSD paper published in 2018. Since 2017, the observation network has been extended with a new automatic weather station and has been complemented by sensors continuously recording snow cover properties. The manuscript documents these extensions and presents the new data sets that have been recorded between 2017 and 2023.

The Rofental research catchment is a very well-instrumented high Alpine environmental research basin, combining glaciological, hydrological and meteorological observations. The dataset presented here is very valuable, especially concerning new sensor technologies measuring snow properties and certainly merits publication in ESSD. The dataset is easy to access and well documented.

The manuscript presents interesting exemplary use cases of data analysis but I think several points must be clarified:

Line 174: ‘temperature correction for longwave radiation…’, specify sensor temperature to correct for the longwave emission of the sensor (I suppose)

Section 6 ‘Meteorological data’

Is the wind speed following seasonal changes?

Specify how wind gusts are measured, what does ‘wind gust’ mean?

Rather than discussing outgoing solar radiation, present and discuss albedo values.

The precipitation gauge under-catch is a serious measurement problem and may explain the maximal amounts of precipitation recorded in summer. Further analysis is required. The under-catch can be quantified (at least approximately) by comparing the measurements from different types of rain gauges and the snow scale in winter.

Lines 248-249: No explanation for the differences in HS measurements a few meters apart from January to May 2022 (panel (e))?

Section 7.2 and Figure 8 are not clear.

The mismatch between SWE and HS measurements is worrying and should be analyzed further. As the sensors used are not sufficiently specified (see other comments), it is difficult to understand if the measured variables are at the same site and the interpretation of the data remains unclear. For instance, the legend of Figure 8 mentions ‘SWE melt out date is two weeks later than HS’, but we don’t not if these variables are measured at the same location. Furthermore, temperature measurements in panel (b) should be used to estimate the melt out date (before HS melt out date?). The different temperatures in panel (c) are not visible (use different colors).

Section 7.3

The two configurations of the flat bands (in diagonal or horizontally) should be described before (in Section 4.3).

The comparison of snow density values derived from S1, S2 and the snow scale is unclear (lines 314-321). Should the density derived from the snow scale (total SWE and HS) only be compared with the S1 diagonal flat band measurements, since the S2 measurements concern snow density at the base of the snowpack? The interpretation of density measurements needs to be clarified.

Section 7.4

Lines 326-328: ‘The measurement principle… with different results’. As the uncertainty of acoustic snow drift sensor seems quite high, could the authors be more specific? Give an estimation of the uncertainty range, the main measurement problems… The following sentence states that ‘it is still the only way to continuously measure and detect drifting snow events with a certain reliability’. What about optical snow particle counters?

The analysis of a snow drift event based on different measurement instruments is interesting but I see two main problems:

- the SWE is measured with different types of sensors in the exposed and sheltered sites (snow pillow and snow scale). As SWE measurements by different sensors can be quite different (for instance Figure 7), the analyze of SWE differences between the exposed and sheltered sites require a better comparison between snow pillow and snow scale measurements (a comparison at the same location for instance).

- The blowing snow flux measured by acoustic sensor can be perturbed by snowfall. Thus, with this sensor (and due to the difficulty to measure the snowfall rate in strong wind conditions), it is difficult to quantify the blowing snow flux during a precipitation event and to relate it to a wind speed threshold for snow erosion. Thus, the interpretation of snow particle fluxes and changes in SWE is problematic. It would be more convincing to analyze a snow drift event without precipitation.

In panel (a), snow depth in the sheltered site shows little deposition compared with the large deposition recorded in SWE (panel (a)). This shows a discrepancy between SWE and HS measurements at the sheltered site during the period of the main blowing snow event?

Figures:

The text in the figures is often too small and difficult to read (Figures 4, 5, 6, 7, 8, 9 and 11, axe titles in particular). The legend should clearly state from which sensor is derived each variable (for instance from which sensor is derived precipitation in Figure 5 or 11?). This is an important point.

The map in figure 1 should clearly highlight the three stations discussed in the paper.

Figure 3 is very useful but should clearly highlight the new instruments (compared to the 2018 publication).

Legend of Figure 4 ‘Narrow bars indicate a second sensor for a variable’: not clear to me.

Figure 6: SWE in mm w.e. The scale of HS should go to 200 cm in (c) to be coherent with (b) and (e). In (e): ‘USH-9’ is not clear.

Figure 7 is interesting but not clear. Specify from which the sensors are derived HS and SWE. The yellow line is not sufficiently visible (chose another color).

Figure 9: explain S1 and S2 in the legend. Panel (c): SWE in mm w.e.

Tables 1, 2 and 3: better highlight the new sensors installed since 2018