the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Aug 2021
The statistics of blowing snow occurrences from multi-year autonomous snow flux measurements in the French Alps
Abstract. Wind-driven snow transport has important implications for the spatial-temporal heterogeneity of snow distribution and snowpack evolution in mountainous areas, such as the European Alps. The climatological and hydrological significance of this region have been extensively investigated using satellite and numerical models. However, knowledge of the spatiotemporal variability of blowing snow is in its infancy because of inaccuracies in satellite-based blowing snow algorithms and the absence of quantitative assessments. Here, we present the spatiotemporal variability and magnitude of blowing snow events, and explore the potential links with ambient meteorological conditions using near surface blowing snow observations from the ISAW outdoor environmental monitoring network. Results show frequent occurrence of blowing snow, and contrasting seasonal variability in the French Alps. On average, monthly blowing snow days range from 5.0 to 14.3 days when using the snow flux threshold of 0.1 g m−2 s−1. The minimum and maximum frequencies of blowing snow days are observed in September and January, respectively, accounting for between 16.7 % and 46.1 % of the month. However, the frequency of monthly blowing snow days varies widely between stations, and this variability is more pronounced at lower threshold levels. Blowing snow events with relatively high magnitudes of snow mass flux (1.0 g m−2 s−1) occur more frequently than low-intensity events (snow mass flux ranges from 0.1 to 0.5 g m−2 s−1). By imposing a minimum duration of 4 h, the monthly cumulative hours with blowing snow occurrences can be up to 255 hours, but show significant seasonal and spatial variability. The considerable variability observed across this region can be explained by contrasting local climate (particularly wind speed and air temperature), snowpack properties, topography and vegetation. The snow-mass transported during relatively high magnitude blowing snow events accounts for about 90 % of all the transported snow mass, highlighting the importance of major events. Blowing snow events that occur with concurrent snowfall are generally associated with intense snow transport. Transport of wet snow and dry snow is mostly concentrated in the range of 0.1 to 0.5 g m−2 s−1 and 0.5 to 1.0 g m−2 s−1, respectively. Understanding the spatiotemporal variability of blowing snow occurrences and the potential links with ambient meteorological conditions is critical for constructing effective avalanche disaster warning systems, and for promoting quantitative evaluation and development of satellite retrieval algorithms and blowing snow models.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(4616 KB)
Interactive discussion
Status: closed
-
RC1: 'Comment on tc-2021-260', Anonymous Referee #1, 06 Sep 2021
This study presented here aims to present the spatiotemporal variations in blowing snow occurrence in the French Alps and to explore the potential links with ambient meteorological conditions using long-term meteorological data and FlowCapt measured snow mass flux. This analysis might help to improve model simulations of blowing snow, as local measurements and satellite-based retrievals are linked. Nevertheless, the study is not placed in a broader context, which does not show me the applicability of this study in other studies. I have three main concerns about the study.
First, the authors say the regional characteristics of blowing snow occurrence are investigated, but the blowing snow measurements are not connected with the local topography and the available other observations such as snow depth and precipitation. For example, it doesn’t make sense to state that the minimum frequencies of blowing snow days are observed in September, without connecting it to the terrain. It is very likely that most of the stations barely have snow in September in the French Alps, especially at altitudes <2000 m, which is supported with Fig. 10. If the authors are willing to improve this point it is needed that they show other metadata of the stations, such as the type of vegetation, exact altitude and the topographical features such as sheltered or exposed areas and relate these to the observed snow flux. Also, Fmor is often highlighted as the station with the maximum values, but this should be explained with the features of the station.
Second, an extensive review of the FlowCapt sensor is missing. The research in this study should be related to previous work with such a sensor. More information is needed on reliability and applications of the sensor. Furthermore, I am also not sure about the reliability of the FlowCapt sensor if it got partly covered by snow. This study also is contradictory at this point. L124 states that the sensor can monitor snow drift as long as the sensor is partially exposed, but in L146/147 it is said that the measurements suffer from changes in this exposed length. The reliability of the sensor had already been discussed in other literature. For example, Lehning and Fierz (2008) stated that the FlowCapt measurements have to be regarded as an index measurement rather than a precise flux measurement. It should therefore be doubted whether it is useful to distinguish between 0.1, 0.5 and 1 g m-2s-1.
Third, it is needed to distinguish between a discussion and conclusion. The study lacks of a discussion and the study is not placed in a broader context. This makes the study not useful for other studies, as numbers are presented that apply to nine stations at the French Alps, without relating it to local topography. Already from observations during fieldwork, we know that blowing snow occurs very frequently, and the quantitative use of FlowCapt sensors, as already said, is questionable.
Detailed comments:
L39: palys must be plays
L41: which effects of snow cover on the ground thermal regime are meant?
L47/48: can you please add some examples of climatological and hydrological effects on changes in snow distribution?
L49/50: please add references to these two statements.
L53: Li and Pomeroy, not Pomeriy.
L60: aclanche must be avalanches
L62: please consider replacing heat fluxes by energy fluxes.
L69: 2) physically based snow models
L70: please add some more recent references, such as Liston et al (2007) or even more recent physically based snow models (e.g. SNOWPACK).
L73: despite sharing a similar overall spatial pattern
L82: observations instead of observation
L83: spatiotemporal patterns
L86: please provide examples of where the measurements are done, even though they are sparse
L87: and uneven distribution. And do you mean the FlowCapt observations with near-surface measurements?
L88: accurate information is not a good statement here. In other studies, it was already stated that the FlowCapt sensor are an indicator of snow flux and not of precise measurements. Do you maybe mean other accurate information? Which?
L92: transport, with (the comma is placed wrong)
L101: in France and please elaborate on what these ISAW stations are. Where are they used for and what kind of data do they provide? Since when are measurements available?
L104/105: please mention or show which years are used in this study. A Gantt chart with available data per station would be a valuable addition. Also please state over how many years the data is averaged.
L116: compoled? Do you mean resampled to an hourly resolution?
L117: please consider using abbreviations for the meteorological variables, as these variables are repeated here from L106-108.
L123: what is meant with large climatic range?
L124: I want to know more about the functionality of the sensor if it is partially exposed. In general and as stated above, more information is needed about the FlowCapt sensors.
Fig. 1: please provide the elevations at an higher resolution and/or a table with the exact elevation of the station and its expositions.
L134: the air temperature (space missing)
L135: what is meant by ranges? Please also consider using abbreviations for wind speed and direction or otherwise remove the second time ‘wind’.
L136/137: how often does it occur that the wind speed and direction remain unchanged? Is this a measurement error?
L139: a SR50 is a snow depth sensor. Please mention it. The paper would also benefit from a figure that compares precipitation, snow depth and snow drift observations, as these are strongly related.
L141: the months without snow are not the same from year to year. Please compare and display the snow flux observations with the snow depth and only discard months without snow. Show the statistics of the amount of months with snow. Therefore it is also needed to show which years measurements were available (see L104/105).
L143: outlier observations. Please use observations instead of measurements throughout the manuscript.
L145: why do you discard periods with positive air temperatures?
L148: please elaborate more on these inaccurate snow depth observations. How inaccurate are the snow depth observations compared to the uncertainty of the FlowCapt sensor?
L159/160: which values were used by Amory (2020) and how much do they differ from what is used in the French Alps? Do these thresholds make sense, if we keep in mind that the FlowCapt sensor only indicates snow drift and is not very accurate in its quantitative assessment?
Fig. 2+3: the timescale of these figures (and the other figures showing the months) is very confusing. Please add a broken axis between May and September and use a scatter plot instead of a line.
Sect. 3.1: the mentioning of the minimum and maximum for specific stations is only interesting if it is related to another parameter, such as the local topography or extreme wind values.
L193: please refer to which Fig. XX after Fche
L197: what is found in the study of Guyomarc'h et al?
Fig. 3: how are events counted that overlap two months? E.g. a blowing snow event start 31 January and lasts until 2 February. Where is this assigned to? Fig. 3 can be removed in my opinion, as it does not provide a lot of additional information if Fig. 2 and Fig. 4 are displayed.
L200: BSD frequency is very vague. Please decide if you want to show the amount of blowing snow events (including its duration) or the amount of days that blowing snow occurs.
L202: please refer to Fig. 4 after the numbers.
L203: please relate the highest and lowest frequencies to the local conditions at the stations.
L207: please prove and show why/how the significant spatial variability was related to the local ambient conditions.
L211: Fmor and Fcmb stations
L211: how can an event be a fraction of a number? Is it averaged? Please provide information over how many years this average is calculated.
L213: only small variations
Fig. 4: please provide information on how the frequency of BSD is calculated. Is it BSD/#days? Also add the broken line to the x-axis.
Fig. 5: add the broken line to the x-axis. Add title above the Fig. (0.1, 0.5 and 1 g m-2s-1) and provide more information on 5ghi in the caption.
L233: 76.99% (significancy) and blowing instead of blwoing
L234: this sentence is very unclear. High or low magnitude? Accounting 88.4% of the time? What do you mean with both magnitudes accounting?
L241: fluxes
L247: Seasonal variability (capital letter)
Sect. 3.2: to which months do you assign the events that overlap two months? See comment at Fig. 3.
Fig. 6: this is a very unclear figure. The grey dots are related to the months. Again, please do not connect these points as the summer months are missing. Furthermore, as I’ve already mentioned my concern about the reliability of the FlowCapt sensor and the thresholds, I would remove this figure. If you are really willing to keep the figure, please add it as scatterplot to Fig. 5.
L271: the low value in May at Fber is probably related to very small amounts of snow in May. Please validate this with snow depth and precipitation data. Otherwise, mentioning this station as station with the largest variations is not adding much information to your manuscript.
Fig. 7: caption: increasing mean wind speed (WS) and 0.1 and 1.0 g m-2s-1, respectively. In the labels m/s must be m s-1 and also show wind speeds of 0 m s-1. If this is not possible, elaborate more on the chosen threshold of 3 m s-1.
L294: wind speeds in Figure 7 (space is missing)
L295: these high wind speeds also lead to blowing snow. Even though they are rare, please add them to the study and elaborate the magnitude of these events.
Fig. 8: this should not be a line plot, as the summer months are missing. Please consider making a scatter plot such as Fig. 5.
L304: high wind speeds (space was missing). How high are relatively high wind speeds?
L307: relatively strong inter-particle bonding. This is an interesting statement, but please prove it.
L308: how do you know about snow compaction? Please prove this with literature or observations.
L309: low and (space is missing)
L310: how do you know this is ‘likely caused’? Please prove this statement.
L312: is this snow wetness measured? How do you know this? I’d rather say high water content of the snow.
L316: this statement does not fit in the context of your results and it should be supported by other literature.
Fig. 9: please elaborate in the text further about the differences between NoSF and SF. It is not entirely clear in both text and figure.
Fig. 10: hard to understand. Please represent the data in a graphical way. Anyways, this plot also shows why blowing snow in September is the least frequent, as we often do not have snow. Caption: unavailable and rarely occur.
L343: the frequency of blowing snow with concurrent snowfall was significantly lower. But how are the measurements handled during snowfall? Again, more information is needed about how the sensor works. I can imagine it won’t work during snowfall and thus the frequency seems to be lower. Second, what is significant in this context?
L373: closely related to large-scale atmospheric circulation and local topography. Please add references to this statement.
Fig. 11: Do these wind roses display the frequencies? The scale of the axis is different (between 30 and 50) for the different stations. Please make this equal.
Sect. 3.4 about wind: it is concluded that wind direction is not an effective factor to estimate the occurrence of blowing snow. But this is concluded while no comparison has been made with the local topography, because that surely influences the blowing snow.
Sect. 4: this section is called summary, but the study would strongly benefit from a discussion to place the results in a broader context and to be critical about its own results. A proper conclusion should follow the discussion.
L406: of course September has the least amount of blowing snow days, as there is barely snow in September (Fig. 10). This is not a proper conclusion or a new finding where the community can benefit from.
L418: snow event in terms of
L431: please provide a direct link to the data set
References:
Lehning, M., & Fierz, C. (2008). Assessment of snow transport in avalanche terrain. Cold Regions Science and Technology, 51(2-3), 240-252. https://doi.org/10.1016/j.coldregions.2007.05.012
Liston, G., Haehnel, R., Sturm, M., Hiemstra, C., Berezovskaya, S., & Tabler, R. (2007). Simulating complex snow distributions in windy environments using SnowTran-3D. Journal of Glaciology, 53(181), 241-256. doi:10.3189/172756507782202865
Citation: https://doi.org/10.5194/tc-2021-260-RC1 -
RC2: 'Comment on tc-2021-260', Anonymous Referee #2, 02 Nov 2021
The statistics of blowing snow occurrences from multi-year autonomous snow flux measurements in the French Alps
The authors present a statistical analysis of blowing snow transport at a number of meteorological stations in the French Alps. Relying on blowing snow flux information from FlowCapt instruments, the authors analyze the spatiotemporal variability of blowing snow in one region of one mountain range. Even in such a relatively small domain, this is a significant open question, largely because of the strong presence of spatiotemporal variability in snow cover characteristics and turbulent winds in complex topography. While the authors provide a detailed breakdown of FlowCapt time series, they do little to describe the inherent differences between meteorological sites, and they do nothing to address well known problems with FlowCapt data for quantitative studies.
Major and minor comments are detailed below. If the editor finds the manuscript fit for resubmission, there are grammatical and spelling issues that need to be addressed as well.
Major Comments:
I don’t think the FlowCapt data can be accurately used in the ways that the authors are intending. There is a very strong critique of the product (i.e. Cierco et al., 2007) that highlights fundamental issues with the data, specifically for the kind of work that is being submitted here. However, the work of Cierco et al., (2007), nor the issues detailed by them are nowhere mentioned in the present work.
To quote a few key conclusions from Cierco:
“Particle velocity has been shown to be a very important parameter in the measurement process. Flowcapt still appears to be unable to consider it properly because of erroneous assumptions at the basis of the calibration process. As a result, the proposed calibration of the sensor produces large overestimation of the snow flux. Moreover, this trend appears to drastically increase with height because of the properties of the atmospheric boundary layer...
If we consider a 1-m long tube, the variation in wind speed between the top and the bottom of the sensor produces an error in the estimation of a mean speed for particles...
It should be stressed that with the best correction found we do not expect major improvements in terms of measurement accuracy. Indeed, because of the power law that relates snow flux to the signal, substantial changes in snow flux only led to a small range of variations in the recorded signal...
To conclude, sensor saturation must be considered when extreme wind occurs, even in natural flows, and vibration de-coupling at low temperatures needs to be improved...
The recordings from Flowcapt may be efficiently corrected so as to constitute database containing extensive amounts of field data, but a great deal of work is still needed to determine the data reliability...
If Flowcapt is still used efficiently for operational purposes (to accurately determine the time step of blowing snow occurrences and to obtain a qualitative estimation of snow fluxes), its usefulness in research works is quite limited : the device requires additional calibrations and its output signal needs further corrections. “
In light of these conclusions, I do not think the numerical flux measurements can be used in this quantitative way, especially as the data are treated as is. Furthermore, I am skeptical that using the three thresholds as qualitative ranges (e.g. low, medium, high) can be performed as we do not know if 0.1 gm^-2s^-1 means the same thing from one storm (or gust) to the next. This significantly undercuts the analysis in this manuscript, and I do not see a clear path to publication given these large uncertainties as the conclusions heavily rely on these quantitative flux values.
Given the uncertainty in FlowCapt data, there is room for the manuscript to calculate a binary blowing snow presence from the FlowCapt data, but this would likely result in a significant change to the manuscript and its conclusions to maintain scientific interest.
Minor Comments:
L21 Define ISAW
L41 “plays”
L53 Pomeroy
L60 avalanche
L83 spatiotemporal
L87 uneven
L92 move comma
L116 compiled
Figure 1. We need significantly more information about the meteorological stations, including aspect, elevation, lat/lon, and a site description (treeline, clearing, alpine, glacier, ridgeline, valley bottom, moraine, etc.)
L134 air not theair
L136 If you are dealing with wind speed, you only have non-negative scalar values. Therefore a mean windspeed of zero, means a constant windspeed of zero, and you automatically know what the maximum windspeed is.
L148-149, This is troubling. What do you mean “only relative values of snow flux”? How can one measurement be compared to another measurement, when they may both contain different errors?
L153-154 How did you calculate whether snow was melted? A zero-degree threshold is not an indicator of snow surface water content. This method needs to be elaborated on, or possibly refined.
L158-159 In light of the accuracy study of Cierco, it needs to be clarified that these numerical thresholds have little to do with the physical fluxes that they represent. The authors have essentially studied different numerical output from the FlowCapt sensors, which may or may not be caused by different blowing snow conditions, and may or may not be related to the reality in the field.
L410-412: How does this correspond with the wind characteristics at the site? This would seem to imply that either the wind is more often strong than weak, snow scours only during strong storms and then transport would ceases (but there is still a large reservoir of available snow for transport), or that the FlowCapt instrument is biased towards high fluxes.
Reference:
Cierco, F.-X., Naaim-Bouvet, F. and Bellot, H.: Acoustic sensors for snowdrift measurements: How should they be used for research purposes?, Cold Reg. Sci. Technol., 49(1), 74–87, doi:10.1016/j.coldregions.2007.01.002, 2007.
Citation: https://doi.org/10.5194/tc-2021-260-RC2
Interactive discussion
Status: closed
-
RC1: 'Comment on tc-2021-260', Anonymous Referee #1, 06 Sep 2021
This study presented here aims to present the spatiotemporal variations in blowing snow occurrence in the French Alps and to explore the potential links with ambient meteorological conditions using long-term meteorological data and FlowCapt measured snow mass flux. This analysis might help to improve model simulations of blowing snow, as local measurements and satellite-based retrievals are linked. Nevertheless, the study is not placed in a broader context, which does not show me the applicability of this study in other studies. I have three main concerns about the study.
First, the authors say the regional characteristics of blowing snow occurrence are investigated, but the blowing snow measurements are not connected with the local topography and the available other observations such as snow depth and precipitation. For example, it doesn’t make sense to state that the minimum frequencies of blowing snow days are observed in September, without connecting it to the terrain. It is very likely that most of the stations barely have snow in September in the French Alps, especially at altitudes <2000 m, which is supported with Fig. 10. If the authors are willing to improve this point it is needed that they show other metadata of the stations, such as the type of vegetation, exact altitude and the topographical features such as sheltered or exposed areas and relate these to the observed snow flux. Also, Fmor is often highlighted as the station with the maximum values, but this should be explained with the features of the station.
Second, an extensive review of the FlowCapt sensor is missing. The research in this study should be related to previous work with such a sensor. More information is needed on reliability and applications of the sensor. Furthermore, I am also not sure about the reliability of the FlowCapt sensor if it got partly covered by snow. This study also is contradictory at this point. L124 states that the sensor can monitor snow drift as long as the sensor is partially exposed, but in L146/147 it is said that the measurements suffer from changes in this exposed length. The reliability of the sensor had already been discussed in other literature. For example, Lehning and Fierz (2008) stated that the FlowCapt measurements have to be regarded as an index measurement rather than a precise flux measurement. It should therefore be doubted whether it is useful to distinguish between 0.1, 0.5 and 1 g m-2s-1.
Third, it is needed to distinguish between a discussion and conclusion. The study lacks of a discussion and the study is not placed in a broader context. This makes the study not useful for other studies, as numbers are presented that apply to nine stations at the French Alps, without relating it to local topography. Already from observations during fieldwork, we know that blowing snow occurs very frequently, and the quantitative use of FlowCapt sensors, as already said, is questionable.
Detailed comments:
L39: palys must be plays
L41: which effects of snow cover on the ground thermal regime are meant?
L47/48: can you please add some examples of climatological and hydrological effects on changes in snow distribution?
L49/50: please add references to these two statements.
L53: Li and Pomeroy, not Pomeriy.
L60: aclanche must be avalanches
L62: please consider replacing heat fluxes by energy fluxes.
L69: 2) physically based snow models
L70: please add some more recent references, such as Liston et al (2007) or even more recent physically based snow models (e.g. SNOWPACK).
L73: despite sharing a similar overall spatial pattern
L82: observations instead of observation
L83: spatiotemporal patterns
L86: please provide examples of where the measurements are done, even though they are sparse
L87: and uneven distribution. And do you mean the FlowCapt observations with near-surface measurements?
L88: accurate information is not a good statement here. In other studies, it was already stated that the FlowCapt sensor are an indicator of snow flux and not of precise measurements. Do you maybe mean other accurate information? Which?
L92: transport, with (the comma is placed wrong)
L101: in France and please elaborate on what these ISAW stations are. Where are they used for and what kind of data do they provide? Since when are measurements available?
L104/105: please mention or show which years are used in this study. A Gantt chart with available data per station would be a valuable addition. Also please state over how many years the data is averaged.
L116: compoled? Do you mean resampled to an hourly resolution?
L117: please consider using abbreviations for the meteorological variables, as these variables are repeated here from L106-108.
L123: what is meant with large climatic range?
L124: I want to know more about the functionality of the sensor if it is partially exposed. In general and as stated above, more information is needed about the FlowCapt sensors.
Fig. 1: please provide the elevations at an higher resolution and/or a table with the exact elevation of the station and its expositions.
L134: the air temperature (space missing)
L135: what is meant by ranges? Please also consider using abbreviations for wind speed and direction or otherwise remove the second time ‘wind’.
L136/137: how often does it occur that the wind speed and direction remain unchanged? Is this a measurement error?
L139: a SR50 is a snow depth sensor. Please mention it. The paper would also benefit from a figure that compares precipitation, snow depth and snow drift observations, as these are strongly related.
L141: the months without snow are not the same from year to year. Please compare and display the snow flux observations with the snow depth and only discard months without snow. Show the statistics of the amount of months with snow. Therefore it is also needed to show which years measurements were available (see L104/105).
L143: outlier observations. Please use observations instead of measurements throughout the manuscript.
L145: why do you discard periods with positive air temperatures?
L148: please elaborate more on these inaccurate snow depth observations. How inaccurate are the snow depth observations compared to the uncertainty of the FlowCapt sensor?
L159/160: which values were used by Amory (2020) and how much do they differ from what is used in the French Alps? Do these thresholds make sense, if we keep in mind that the FlowCapt sensor only indicates snow drift and is not very accurate in its quantitative assessment?
Fig. 2+3: the timescale of these figures (and the other figures showing the months) is very confusing. Please add a broken axis between May and September and use a scatter plot instead of a line.
Sect. 3.1: the mentioning of the minimum and maximum for specific stations is only interesting if it is related to another parameter, such as the local topography or extreme wind values.
L193: please refer to which Fig. XX after Fche
L197: what is found in the study of Guyomarc'h et al?
Fig. 3: how are events counted that overlap two months? E.g. a blowing snow event start 31 January and lasts until 2 February. Where is this assigned to? Fig. 3 can be removed in my opinion, as it does not provide a lot of additional information if Fig. 2 and Fig. 4 are displayed.
L200: BSD frequency is very vague. Please decide if you want to show the amount of blowing snow events (including its duration) or the amount of days that blowing snow occurs.
L202: please refer to Fig. 4 after the numbers.
L203: please relate the highest and lowest frequencies to the local conditions at the stations.
L207: please prove and show why/how the significant spatial variability was related to the local ambient conditions.
L211: Fmor and Fcmb stations
L211: how can an event be a fraction of a number? Is it averaged? Please provide information over how many years this average is calculated.
L213: only small variations
Fig. 4: please provide information on how the frequency of BSD is calculated. Is it BSD/#days? Also add the broken line to the x-axis.
Fig. 5: add the broken line to the x-axis. Add title above the Fig. (0.1, 0.5 and 1 g m-2s-1) and provide more information on 5ghi in the caption.
L233: 76.99% (significancy) and blowing instead of blwoing
L234: this sentence is very unclear. High or low magnitude? Accounting 88.4% of the time? What do you mean with both magnitudes accounting?
L241: fluxes
L247: Seasonal variability (capital letter)
Sect. 3.2: to which months do you assign the events that overlap two months? See comment at Fig. 3.
Fig. 6: this is a very unclear figure. The grey dots are related to the months. Again, please do not connect these points as the summer months are missing. Furthermore, as I’ve already mentioned my concern about the reliability of the FlowCapt sensor and the thresholds, I would remove this figure. If you are really willing to keep the figure, please add it as scatterplot to Fig. 5.
L271: the low value in May at Fber is probably related to very small amounts of snow in May. Please validate this with snow depth and precipitation data. Otherwise, mentioning this station as station with the largest variations is not adding much information to your manuscript.
Fig. 7: caption: increasing mean wind speed (WS) and 0.1 and 1.0 g m-2s-1, respectively. In the labels m/s must be m s-1 and also show wind speeds of 0 m s-1. If this is not possible, elaborate more on the chosen threshold of 3 m s-1.
L294: wind speeds in Figure 7 (space is missing)
L295: these high wind speeds also lead to blowing snow. Even though they are rare, please add them to the study and elaborate the magnitude of these events.
Fig. 8: this should not be a line plot, as the summer months are missing. Please consider making a scatter plot such as Fig. 5.
L304: high wind speeds (space was missing). How high are relatively high wind speeds?
L307: relatively strong inter-particle bonding. This is an interesting statement, but please prove it.
L308: how do you know about snow compaction? Please prove this with literature or observations.
L309: low and (space is missing)
L310: how do you know this is ‘likely caused’? Please prove this statement.
L312: is this snow wetness measured? How do you know this? I’d rather say high water content of the snow.
L316: this statement does not fit in the context of your results and it should be supported by other literature.
Fig. 9: please elaborate in the text further about the differences between NoSF and SF. It is not entirely clear in both text and figure.
Fig. 10: hard to understand. Please represent the data in a graphical way. Anyways, this plot also shows why blowing snow in September is the least frequent, as we often do not have snow. Caption: unavailable and rarely occur.
L343: the frequency of blowing snow with concurrent snowfall was significantly lower. But how are the measurements handled during snowfall? Again, more information is needed about how the sensor works. I can imagine it won’t work during snowfall and thus the frequency seems to be lower. Second, what is significant in this context?
L373: closely related to large-scale atmospheric circulation and local topography. Please add references to this statement.
Fig. 11: Do these wind roses display the frequencies? The scale of the axis is different (between 30 and 50) for the different stations. Please make this equal.
Sect. 3.4 about wind: it is concluded that wind direction is not an effective factor to estimate the occurrence of blowing snow. But this is concluded while no comparison has been made with the local topography, because that surely influences the blowing snow.
Sect. 4: this section is called summary, but the study would strongly benefit from a discussion to place the results in a broader context and to be critical about its own results. A proper conclusion should follow the discussion.
L406: of course September has the least amount of blowing snow days, as there is barely snow in September (Fig. 10). This is not a proper conclusion or a new finding where the community can benefit from.
L418: snow event in terms of
L431: please provide a direct link to the data set
References:
Lehning, M., & Fierz, C. (2008). Assessment of snow transport in avalanche terrain. Cold Regions Science and Technology, 51(2-3), 240-252. https://doi.org/10.1016/j.coldregions.2007.05.012
Liston, G., Haehnel, R., Sturm, M., Hiemstra, C., Berezovskaya, S., & Tabler, R. (2007). Simulating complex snow distributions in windy environments using SnowTran-3D. Journal of Glaciology, 53(181), 241-256. doi:10.3189/172756507782202865
Citation: https://doi.org/10.5194/tc-2021-260-RC1 -
RC2: 'Comment on tc-2021-260', Anonymous Referee #2, 02 Nov 2021
The statistics of blowing snow occurrences from multi-year autonomous snow flux measurements in the French Alps
The authors present a statistical analysis of blowing snow transport at a number of meteorological stations in the French Alps. Relying on blowing snow flux information from FlowCapt instruments, the authors analyze the spatiotemporal variability of blowing snow in one region of one mountain range. Even in such a relatively small domain, this is a significant open question, largely because of the strong presence of spatiotemporal variability in snow cover characteristics and turbulent winds in complex topography. While the authors provide a detailed breakdown of FlowCapt time series, they do little to describe the inherent differences between meteorological sites, and they do nothing to address well known problems with FlowCapt data for quantitative studies.
Major and minor comments are detailed below. If the editor finds the manuscript fit for resubmission, there are grammatical and spelling issues that need to be addressed as well.
Major Comments:
I don’t think the FlowCapt data can be accurately used in the ways that the authors are intending. There is a very strong critique of the product (i.e. Cierco et al., 2007) that highlights fundamental issues with the data, specifically for the kind of work that is being submitted here. However, the work of Cierco et al., (2007), nor the issues detailed by them are nowhere mentioned in the present work.
To quote a few key conclusions from Cierco:
“Particle velocity has been shown to be a very important parameter in the measurement process. Flowcapt still appears to be unable to consider it properly because of erroneous assumptions at the basis of the calibration process. As a result, the proposed calibration of the sensor produces large overestimation of the snow flux. Moreover, this trend appears to drastically increase with height because of the properties of the atmospheric boundary layer...
If we consider a 1-m long tube, the variation in wind speed between the top and the bottom of the sensor produces an error in the estimation of a mean speed for particles...
It should be stressed that with the best correction found we do not expect major improvements in terms of measurement accuracy. Indeed, because of the power law that relates snow flux to the signal, substantial changes in snow flux only led to a small range of variations in the recorded signal...
To conclude, sensor saturation must be considered when extreme wind occurs, even in natural flows, and vibration de-coupling at low temperatures needs to be improved...
The recordings from Flowcapt may be efficiently corrected so as to constitute database containing extensive amounts of field data, but a great deal of work is still needed to determine the data reliability...
If Flowcapt is still used efficiently for operational purposes (to accurately determine the time step of blowing snow occurrences and to obtain a qualitative estimation of snow fluxes), its usefulness in research works is quite limited : the device requires additional calibrations and its output signal needs further corrections. “
In light of these conclusions, I do not think the numerical flux measurements can be used in this quantitative way, especially as the data are treated as is. Furthermore, I am skeptical that using the three thresholds as qualitative ranges (e.g. low, medium, high) can be performed as we do not know if 0.1 gm^-2s^-1 means the same thing from one storm (or gust) to the next. This significantly undercuts the analysis in this manuscript, and I do not see a clear path to publication given these large uncertainties as the conclusions heavily rely on these quantitative flux values.
Given the uncertainty in FlowCapt data, there is room for the manuscript to calculate a binary blowing snow presence from the FlowCapt data, but this would likely result in a significant change to the manuscript and its conclusions to maintain scientific interest.
Minor Comments:
L21 Define ISAW
L41 “plays”
L53 Pomeroy
L60 avalanche
L83 spatiotemporal
L87 uneven
L92 move comma
L116 compiled
Figure 1. We need significantly more information about the meteorological stations, including aspect, elevation, lat/lon, and a site description (treeline, clearing, alpine, glacier, ridgeline, valley bottom, moraine, etc.)
L134 air not theair
L136 If you are dealing with wind speed, you only have non-negative scalar values. Therefore a mean windspeed of zero, means a constant windspeed of zero, and you automatically know what the maximum windspeed is.
L148-149, This is troubling. What do you mean “only relative values of snow flux”? How can one measurement be compared to another measurement, when they may both contain different errors?
L153-154 How did you calculate whether snow was melted? A zero-degree threshold is not an indicator of snow surface water content. This method needs to be elaborated on, or possibly refined.
L158-159 In light of the accuracy study of Cierco, it needs to be clarified that these numerical thresholds have little to do with the physical fluxes that they represent. The authors have essentially studied different numerical output from the FlowCapt sensors, which may or may not be caused by different blowing snow conditions, and may or may not be related to the reality in the field.
L410-412: How does this correspond with the wind characteristics at the site? This would seem to imply that either the wind is more often strong than weak, snow scours only during strong storms and then transport would ceases (but there is still a large reservoir of available snow for transport), or that the FlowCapt instrument is biased towards high fluxes.
Reference:
Cierco, F.-X., Naaim-Bouvet, F. and Bellot, H.: Acoustic sensors for snowdrift measurements: How should they be used for research purposes?, Cold Reg. Sci. Technol., 49(1), 74–87, doi:10.1016/j.coldregions.2007.01.002, 2007.
Citation: https://doi.org/10.5194/tc-2021-260-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 832 | 238 | 40 | 1,110 | 35 | 34 |
- HTML: 832
- PDF: 238
- XML: 40
- Total: 1,110
- BibTeX: 35
- EndNote: 34
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
