Elucidation of Spatiotemporal structures from high-resolution blowing snow observations

Nishimura, Kouichi; Nemoto, Masaki; Ito, Yoichi; Omiya, Satoru; Shimoyama, Kou; Niiya, Hirofumi

doi:https://doi.org/10.5194/egusphere-2023-1845

Preprints

https://doi.org/10.5194/egusphere-2023-1845

Preprints

26 Sep 2023

| 26 Sep 2023

Elucidation of Spatiotemporal structures from high-resolution blowing snow observations

Kouichi Nishimura, Masaki Nemoto, Yoichi Ito, Satoru Omiya, Kou Shimoyama, and Hirofumi Niiya

Abstract. Systematic observations were conducted to investigate the spatio-temporal structures of blowing snow. Along a line perpendicular to the dominant wind direction on the leeside of a flat field, fifteen Snow Particle Counters (SPCs) and Ultra Sonic Anemometers (USAs) were placed 1.5 m apart. Data were recorded at high frequencies of 100 kHz for SPCs and 1 kHz for USAs. The horizontal mass flux distributions, representing the spatio-temporal variability of blowing snow, exhibited non-uniformity in both time and space and manifested periodic changes akin to snow waves. Additionally, the presence of 'snow snakes,' meandering near the snow surface, was observed. Quadrant analysis revealed predominant snow fluxes in quadrants Q1 (u'>0, w'>0) and Q4 (u'>0, w'<0). However, a more detailed parametric curve analysis indicated the existence of ejection events Q2 (u'<0, w'>0) before snow waves and in front of snow snakes, shifting to Q1 and Q4 afterward, implying the consideration of both top-down and bottom-up mechanisms for burst sweep events.

Received: 13 Aug 2023 – Discussion started: 26 Sep 2023

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 2410 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (2410 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

22 Oct 2024

Elucidation of spatiotemporal structures from high-resolution blowing-snow observations

Kouichi Nishimura, Masaki Nemoto, Yoichi Ito, Satoru Omiya, Kou Shimoyama, and Hirofumi Niiya

The Cryosphere, 18, 4775–4786, https://doi.org/10.5194/tc-18-4775-2024,https://doi.org/10.5194/tc-18-4775-2024, 2024

Short summary

Kouichi Nishimura, Masaki Nemoto, Yoichi Ito, Satoru Omiya, Kou Shimoyama, and Hirofumi Niiya

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2023-1845', Nikolas Aksamit, 11 Dec 2023

The authors present data from a very exciting and one-of-a-kind experiment on natural variations in blowing snow transport. By utilizing an unprecedented number of blowing snow particle counters and ultra-sonic anemometers, they have been able to create 3D representations of traveling blowing snow clouds. This group of researchers is likely the only collection in the world that is capable of performing such a field experiment, and I am very happy to see these data get into the public eye. I think a lot of interesting analysis may follow from this initial analysis and I am excited to see what follow up studies may happen.

General Comments:
As it stands, the authors have a done a good job highlighting a couple interesting cases where they were able to relate blowing snow behavior to specific signatures in adjacent wind measurements. It is not clear, however, how common these features were in general.
Some additional information on the features being presented could be very informative. For example, can you provide a measure of the average streamwise and spanwise spacing of the features? This could be calculated from autocorrelations in the two directions. As well, can you provide the characteristic aspect ratios for the wave and snake features you describe? This would be very helpful to relate to the responsible atmospheric structures. How much do these characteristics vary across your dataset?
As well, for the cryospheric science community, I believe a bit more of an explanation of the novelty of this approach could be beneficial. It is not common to study blowing snow from a time-resolved coherent structure standpoint and I think it could be insightful to introduce the quadrant analysis methodology a bit more systematically, explain what sweeps and ejections are commonly thought to represent in the turbulent boundary layer literature, and how that connects to the more common friction velocity-based descriptions of blowing snow flux.
Lastly, I would appreciate a reference, or an explanation, that differentiates between snow waves and snow snakes. These two features appear like they deserve to be in distinct classes, but I am not aware of a previous discussion on them being described as unique. That could be very interesting as they are likely caused by different atmospheric dynamics.

Specific Comments:
L26: I think you mean modeling of, not study of
L32-33: Can you provide an explanation of why you think turbulent sweeps and ejections, specifically, are crucial to discuss? Why not a different kind of coherent turbulent feature? Can we not generalize to say that we just need to take a structure-based view of snow transport given its intermittent dynamics and non-linear dependence on wind speed?
L36-37: Can you provide a citation here?
L46: Conducting is a different tense.
L72: You have two sections titled “Results and analysis”
L73: Are these average or instantaneous values you are plotting?
L75: Westerly?
L79-80: Are you referring to conditions at the AWS? Please specify.
L88: Can you show us that the particles are generally slower than the wind?
L91: Can you comment on what the range of diameters that is measurable by the SPC? Is this the upper end of that range?
L93: It would be very beneficial if in addition to time, you included an average “advected distance” on your x-axis. This could be something like you have mentioned with frozen turbulence: time x avg wind speed. That would really help us see the dimensions of the features you are describing. For example, is Figure 4 approximately 500 meters in the x-direction?
L96: “Precipitation observed with DFIR was”
L97: Can you please rephrase the sentence beginning with “They show the maps…”
L99: Westward wind direction or westerly wind?
L100: What was the period of the periodic changes?
L103: I believe your reference to Figure 5 is out of order. However, it would be beneficial to include a comparison with the photo, including an indicator of lengths.
L101-102: Here or later, can you please comment on what you think the snow waves are caused by?
L112-115: This is a bit surprising and somewhat counter-intuitive. In the atmospheric boundary layer, we often assume that the size of the eddies governing the flow increase as we move away from the surface, but your presentation here suggests the opposite. Can you please comment on what you think is happening and why this is different from the conventional view?
L116: Can you report what the dominant wind frequency is versus the dominant snow frequency? As well, at the higher end of the frequency spectrum, did you see any influence of stochastic particle-surface interactions that weren’t present in the wind?
L119-120: Can you provide a reference for quadrant analysis, such as Wallace’s review from 2016.
L125: What quadrant hole size did you use?
Figure 5: It would be helpful to note for those unfamiliar that the snow waves are organized in a lateral or spanwise orientation and the snow streamers are quasi-parallel to the streamwise direction.
Figure 7: Can you include the heights of the measurements in the labels.
Figure 8: This is a very nice figure and data that has never been available before. You appear to be using multiple fonts. As well, it could improve the presentation to write “Sweep or Ejection” to maintain symmetry with the adjacent colorbars.
L127: We can’t actually see that the snow fluxes were predominantly observed in Quadrant 1 in the figure because we have no idea the density of the dots.
Figure 9: There are some curious features here. For example, this looks like an 1/x relationship between Reynolds stress magnitude and snow flux for all quadrants. Do you have any comment on why the highest snow fluxes appear to be when the magnitude of u’w’ is on the lower end, and why snow fluxes taper off with larger u’w’? For those that use u’w’ to calculate friction velocity (u*) in blowing snow models, this should be a very disturbing finding.
Given your comments about Q1 and Q4 being dominant, can you create a plot comparable to Figure 9 that show the relationship between windspeed magnitude and snow flux? Is there still a 1/x scaling? If there is a stronger correlation, perhaps a “gust-factor” type approach would make sense to revisit (rather than u* or u’w’) for future blowing snow model development?
L130: Why are you specifying these features as “Ejection-like” and “sweep-like” here?
L156-159: This is very nice!
L180-189: Your results and Bauer’s results appear to be in contradiction as they reported Q2 and Q4 dominance, but you report Q1 and Q4, but you state that they align. Can you help me understand?
L183: I believe these studies were on sand, and not snow transport
L204-206: Please rephrase this sentence. It appears incomplete.
Figure 10: Do you have any idea why the massflux appears constant at y=0 in your top panel? It appears in your sweep/ejection panel that the ejections are preceding the start of transport. Does this support the idea of a bottom-up mechanism driving snow transport initiation? That is, the idea that eddies pull away from the surface (causing an ejection) and an in-rush of air fills the space (causing a sweep)?
Figure 11/12: It would be very helpful if you could mark the onset of snow transport in your parametric curves.

-Nikolas Aksamit

Citation: https://doi.org/10.5194/egusphere-2023-1845-RC1
- AC1: 'Reply on RC1', Kouichi Nishimura, 30 Dec 2023
  
  Dear Reviewer,
  Thank you very much for your careful review of our manuscript and for providing positive evaluations. Initially, please understand that in this manuscript, we aim to present a concise overview of our unique observations and the obtained results. It appears that emphasizing this aspect is crucial, and I will ensure to highlight it more prominently in the manuscript. Notably, the dataset resulting from our extensive observations is substantial, affording various avenues for deriving concrete conclusions that elucidate the spatiotemporal structures of blowing snow. More comprehensive analyses will be deferred to subsequent manuscripts, currently in progress.
  We deeply appreciate all of your insightful comments and suggestions, which have proven to be both informative and educational. In line with your recommendations, we are undertaking substantial revisions to the manuscript in collaboration with our coauthors.
  Specifically, we are in the process of calculating autocorrelations to ascertain the average streamwise and spanwise spacing of the observed features. Additionally, we plan to incorporate information on the characteristic aspect ratios of the wave and snake features. The introduction of the quadrant analysis methodology will be refined, elucidating why turbulent sweeps and ejections are pivotal in generating these characteristic structures.
  While the literature review has been extensive, our search for references explicitly detailing the distinctions between snow waves and snow snakes is ongoing. In the figures illustrating horizontal mass flux distributions, we will include "advective distance" to provide clarity regarding the dimensions of the observed structures. Moreover, we will delve into a more detailed explanation and discussion of the features presented in Figure 9.
  Once again, we appreciate your valuable input, and we are committed to enhancing the manuscript based on your feedback.
  Best regards,
  Kouichi Nishimura
  
  Citation: https://doi.org/10.5194/egusphere-2023-1845-AC1
RC2:
'Comment on egusphere-2023-1845', Anonymous Referee #2, 19 Dec 2023

This manuscript reports results of measuring wind and snow transport across a transverse cross-section of 15 stations, spaced at 1.5 metre intervals.
The field experiment and the data are undoubtedly interesting and have the potential to generate new insights into the relationship between boundary layer flow and the transport of snow. The analysis presented in the manuscript, however, is very limited and preliminary, and I feel that the data should be analysed in much more extent, detail, and rigour. The paper furthermore lacks quite a lot of key details about the experimental set-up and the data that were collected, and the initial interpretation of results with respect to streamers doesn’t seem suitable. I will briefly expand on these key issues below. I believe the study requires far more data analysis to make it relevant and up-to-date with the current particle transport literature and for this reason I believe the manuscript should be returned to the authors for later resubmission as a new manuscript.
Data analysis: figure 3 shows three periods of measurement, some of which span several days, and yet the results that are shown are only small snap-shots of data, often only 60 or 120 seconds. In some cases more data is evidently used, but there is no clarity on the transport and surface dynamics within these periods: the spectra in fig.7, for example, are based on longer time-series but the paper never really explains in detail what happens with the relative height of the instruments as snow accumulates underneath the sensors, except where results relate to 2 different heights on Feb.24. In other instances, results are shown from just one station along the transect (and again, only for a minute’s worth of data or so), as in fig.3 and fig.7. For fig.9 it’s not clear whether results are for one station or for all 15, and the same is not clear for table 1. In all these cases, if results are just from one station that clearly leaves a lot of data untouched, or if the results are based on all 15 stations, then it clearly leaves all internal variability unexplored. Fig.11 and 12 present descriptive samples of analysis, but there is no further quantification across the whole dataset. The results presented thus really seem only just an initial exploration and the study lacks a comprehensive investigation of all available data.
Details: linked to the above limitation is the lack of detail on the data collection periods and the experimental set-up. Relative sensor heights are continuously changing, for example, but there is no overview of this. (for that matter, the snow depth was evidently monitored throughout, as shown in fig.2e, but the text lacks any details about how it was measured). Figure 2b shows that UAs are positioned some distance away off the side of the SPCs and exact positioning details need to be shown. Wind directions vary throughout the measurement periods and this will have impacted on the planform mappings (even if direction varies by 10-15 degrees, this will impact on the apparent lateral spacing of time-series).
Analysis: my biggest concern is about the interpretation of patterns shown in fig.6b&c as streamers. First, streamers are generally only 10-20 cm wide (as is also clear from the photo in fig.5b), and yet the bands of high flux in figure 6 are clearly two metres wide or more. These are not streamers in terms of our general understanding. The 1.5 metre spacing between measurement points already precludes mapping these types of small-scale flux structures anyway. Second, the perfectly straight bands in that figure are unnatural and are almost certainly an artifact of differential sensor sensitivity or saturation. The same banding pattern is also visible in the maps of fig.4 and 8a. Real flux patterns would vary laterally over a time-scale of 10 seconds and this should be visible in the maps. I have seen similar artificial banding in transport maps myself and it was related to differential sensor response. I’m not familiar with the SPC used in the study here, but in sand transport studies that use laser counter sensors (‘Wenglors’) we know that the optics can get ‘smudged’ or dusty or affected by condensation on the outside and this leads to degraded performance. A first method of removing the artificial banding is to normalise the individual SPC time-series by a longer-term average. Because the fowling accumulates over time, a Reynolds decomposition using a window of 10-100 seconds may be useful here.
On a final note, I believe the power spectra in fig.7 do not show any significant peaks (as claimed in L112-117), only a typical Kolmogorov-type spectrum slope. The spectrum below frequencies of 10^-1 is far too coarse and unstable to assign meaning to what are essentially noise variations. The suggestion on L116, that coherent flow structures would increase in size between the suspension layer (higher above the surface) and the saltation layer (nearer the surface) runs completely counter to everything we know about turbulence and eddies (decreasing in size nearer to the surface), including the Hunt & Morrison model cited in the study here.
The general observation in this paper so-far that flux tends to relate better to u’>0 events or periods, is also, by the way, a finding that was reported in the sand transport literature by Weaver and Wiggs (2012, GRL).

Citation: https://doi.org/10.5194/egusphere-2023-1845-RC2
- AC2: 'Reply on RC2', Kouichi Nishimura, 30 Dec 2023
  
  Dear Reviewer,
  Thank you very much for carefully reviewing our manuscript and providing insightful comments, especially regarding the standpoint of sand particle transport. From the outset, please understand that, in this manuscript, our intention is to provide a brief overview of this unique observation and the obtained results. As pointed out by the reviewer, our analysis is still quite limited, and we are presenting preliminary results. It is probably essential to emphasize this aspect more prominently in the manuscript. The dataset obtained from the series of observations is substantial, offering various approaches to derive concrete conclusions explaining the spatiotemporal structures of blowing snow. More detailed analyses will be presented in subsequent manuscripts, which are currently in progress.
  All of your comments and suggestions are highly informative and educational. Following your recommendations, we are planning to revise the manuscript extensively through discussions with coauthors. For instance, we are eager to incorporate more detailed information about our experimental setup.
  In practice, the 1.5-meter spacing in our measurements may not be fine enough to discern the precise structures described here, particularly the streamers. However, by improving the color bar in the figures, we hope to indicate more precise and reasonable aspects. At this stage, the transport flux of structures has reached its maximum, hence represented as a simple bold red bar.
  The Snow Particle Counter (SPC) is a well-established blowing snow sensor that outputs both particle size and numbers, representing transport flux every second. As far as I know, it has also been utilized in sand transport research as a "Sand Particle Counter." Additionally, accurate calibrations of all sensors are carried out beforehand, allowing us to reasonably exclude the effect of sensitivity differences.
  In line with the suggestions, we are prepared to reconsider the power spectrum issues that seem to contradict common concepts previously revealed.
  Best regards,
  Kouichi
  
  Citation: https://doi.org/10.5194/egusphere-2023-1845-AC2

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2023-1845', Nikolas Aksamit, 11 Dec 2023

The authors present data from a very exciting and one-of-a-kind experiment on natural variations in blowing snow transport. By utilizing an unprecedented number of blowing snow particle counters and ultra-sonic anemometers, they have been able to create 3D representations of traveling blowing snow clouds. This group of researchers is likely the only collection in the world that is capable of performing such a field experiment, and I am very happy to see these data get into the public eye. I think a lot of interesting analysis may follow from this initial analysis and I am excited to see what follow up studies may happen.

General Comments:
As it stands, the authors have a done a good job highlighting a couple interesting cases where they were able to relate blowing snow behavior to specific signatures in adjacent wind measurements. It is not clear, however, how common these features were in general.
Some additional information on the features being presented could be very informative. For example, can you provide a measure of the average streamwise and spanwise spacing of the features? This could be calculated from autocorrelations in the two directions. As well, can you provide the characteristic aspect ratios for the wave and snake features you describe? This would be very helpful to relate to the responsible atmospheric structures. How much do these characteristics vary across your dataset?
As well, for the cryospheric science community, I believe a bit more of an explanation of the novelty of this approach could be beneficial. It is not common to study blowing snow from a time-resolved coherent structure standpoint and I think it could be insightful to introduce the quadrant analysis methodology a bit more systematically, explain what sweeps and ejections are commonly thought to represent in the turbulent boundary layer literature, and how that connects to the more common friction velocity-based descriptions of blowing snow flux.
Lastly, I would appreciate a reference, or an explanation, that differentiates between snow waves and snow snakes. These two features appear like they deserve to be in distinct classes, but I am not aware of a previous discussion on them being described as unique. That could be very interesting as they are likely caused by different atmospheric dynamics.

Specific Comments:
L26: I think you mean modeling of, not study of
L32-33: Can you provide an explanation of why you think turbulent sweeps and ejections, specifically, are crucial to discuss? Why not a different kind of coherent turbulent feature? Can we not generalize to say that we just need to take a structure-based view of snow transport given its intermittent dynamics and non-linear dependence on wind speed?
L36-37: Can you provide a citation here?
L46: Conducting is a different tense.
L72: You have two sections titled “Results and analysis”
L73: Are these average or instantaneous values you are plotting?
L75: Westerly?
L79-80: Are you referring to conditions at the AWS? Please specify.
L88: Can you show us that the particles are generally slower than the wind?
L91: Can you comment on what the range of diameters that is measurable by the SPC? Is this the upper end of that range?
L93: It would be very beneficial if in addition to time, you included an average “advected distance” on your x-axis. This could be something like you have mentioned with frozen turbulence: time x avg wind speed. That would really help us see the dimensions of the features you are describing. For example, is Figure 4 approximately 500 meters in the x-direction?
L96: “Precipitation observed with DFIR was”
L97: Can you please rephrase the sentence beginning with “They show the maps…”
L99: Westward wind direction or westerly wind?
L100: What was the period of the periodic changes?
L103: I believe your reference to Figure 5 is out of order. However, it would be beneficial to include a comparison with the photo, including an indicator of lengths.
L101-102: Here or later, can you please comment on what you think the snow waves are caused by?
L112-115: This is a bit surprising and somewhat counter-intuitive. In the atmospheric boundary layer, we often assume that the size of the eddies governing the flow increase as we move away from the surface, but your presentation here suggests the opposite. Can you please comment on what you think is happening and why this is different from the conventional view?
L116: Can you report what the dominant wind frequency is versus the dominant snow frequency? As well, at the higher end of the frequency spectrum, did you see any influence of stochastic particle-surface interactions that weren’t present in the wind?
L119-120: Can you provide a reference for quadrant analysis, such as Wallace’s review from 2016.
L125: What quadrant hole size did you use?
Figure 5: It would be helpful to note for those unfamiliar that the snow waves are organized in a lateral or spanwise orientation and the snow streamers are quasi-parallel to the streamwise direction.
Figure 7: Can you include the heights of the measurements in the labels.
Figure 8: This is a very nice figure and data that has never been available before. You appear to be using multiple fonts. As well, it could improve the presentation to write “Sweep or Ejection” to maintain symmetry with the adjacent colorbars.
L127: We can’t actually see that the snow fluxes were predominantly observed in Quadrant 1 in the figure because we have no idea the density of the dots.
Figure 9: There are some curious features here. For example, this looks like an 1/x relationship between Reynolds stress magnitude and snow flux for all quadrants. Do you have any comment on why the highest snow fluxes appear to be when the magnitude of u’w’ is on the lower end, and why snow fluxes taper off with larger u’w’? For those that use u’w’ to calculate friction velocity (u*) in blowing snow models, this should be a very disturbing finding.
Given your comments about Q1 and Q4 being dominant, can you create a plot comparable to Figure 9 that show the relationship between windspeed magnitude and snow flux? Is there still a 1/x scaling? If there is a stronger correlation, perhaps a “gust-factor” type approach would make sense to revisit (rather than u* or u’w’) for future blowing snow model development?
L130: Why are you specifying these features as “Ejection-like” and “sweep-like” here?
L156-159: This is very nice!
L180-189: Your results and Bauer’s results appear to be in contradiction as they reported Q2 and Q4 dominance, but you report Q1 and Q4, but you state that they align. Can you help me understand?
L183: I believe these studies were on sand, and not snow transport
L204-206: Please rephrase this sentence. It appears incomplete.
Figure 10: Do you have any idea why the massflux appears constant at y=0 in your top panel? It appears in your sweep/ejection panel that the ejections are preceding the start of transport. Does this support the idea of a bottom-up mechanism driving snow transport initiation? That is, the idea that eddies pull away from the surface (causing an ejection) and an in-rush of air fills the space (causing a sweep)?
Figure 11/12: It would be very helpful if you could mark the onset of snow transport in your parametric curves.

-Nikolas Aksamit

Citation: https://doi.org/10.5194/egusphere-2023-1845-RC1
- AC1: 'Reply on RC1', Kouichi Nishimura, 30 Dec 2023
  
  Dear Reviewer,
  Thank you very much for your careful review of our manuscript and for providing positive evaluations. Initially, please understand that in this manuscript, we aim to present a concise overview of our unique observations and the obtained results. It appears that emphasizing this aspect is crucial, and I will ensure to highlight it more prominently in the manuscript. Notably, the dataset resulting from our extensive observations is substantial, affording various avenues for deriving concrete conclusions that elucidate the spatiotemporal structures of blowing snow. More comprehensive analyses will be deferred to subsequent manuscripts, currently in progress.
  We deeply appreciate all of your insightful comments and suggestions, which have proven to be both informative and educational. In line with your recommendations, we are undertaking substantial revisions to the manuscript in collaboration with our coauthors.
  Specifically, we are in the process of calculating autocorrelations to ascertain the average streamwise and spanwise spacing of the observed features. Additionally, we plan to incorporate information on the characteristic aspect ratios of the wave and snake features. The introduction of the quadrant analysis methodology will be refined, elucidating why turbulent sweeps and ejections are pivotal in generating these characteristic structures.
  While the literature review has been extensive, our search for references explicitly detailing the distinctions between snow waves and snow snakes is ongoing. In the figures illustrating horizontal mass flux distributions, we will include "advective distance" to provide clarity regarding the dimensions of the observed structures. Moreover, we will delve into a more detailed explanation and discussion of the features presented in Figure 9.
  Once again, we appreciate your valuable input, and we are committed to enhancing the manuscript based on your feedback.
  Best regards,
  Kouichi Nishimura
  
  Citation: https://doi.org/10.5194/egusphere-2023-1845-AC1
RC2:
'Comment on egusphere-2023-1845', Anonymous Referee #2, 19 Dec 2023

This manuscript reports results of measuring wind and snow transport across a transverse cross-section of 15 stations, spaced at 1.5 metre intervals.
The field experiment and the data are undoubtedly interesting and have the potential to generate new insights into the relationship between boundary layer flow and the transport of snow. The analysis presented in the manuscript, however, is very limited and preliminary, and I feel that the data should be analysed in much more extent, detail, and rigour. The paper furthermore lacks quite a lot of key details about the experimental set-up and the data that were collected, and the initial interpretation of results with respect to streamers doesn’t seem suitable. I will briefly expand on these key issues below. I believe the study requires far more data analysis to make it relevant and up-to-date with the current particle transport literature and for this reason I believe the manuscript should be returned to the authors for later resubmission as a new manuscript.
Data analysis: figure 3 shows three periods of measurement, some of which span several days, and yet the results that are shown are only small snap-shots of data, often only 60 or 120 seconds. In some cases more data is evidently used, but there is no clarity on the transport and surface dynamics within these periods: the spectra in fig.7, for example, are based on longer time-series but the paper never really explains in detail what happens with the relative height of the instruments as snow accumulates underneath the sensors, except where results relate to 2 different heights on Feb.24. In other instances, results are shown from just one station along the transect (and again, only for a minute’s worth of data or so), as in fig.3 and fig.7. For fig.9 it’s not clear whether results are for one station or for all 15, and the same is not clear for table 1. In all these cases, if results are just from one station that clearly leaves a lot of data untouched, or if the results are based on all 15 stations, then it clearly leaves all internal variability unexplored. Fig.11 and 12 present descriptive samples of analysis, but there is no further quantification across the whole dataset. The results presented thus really seem only just an initial exploration and the study lacks a comprehensive investigation of all available data.
Details: linked to the above limitation is the lack of detail on the data collection periods and the experimental set-up. Relative sensor heights are continuously changing, for example, but there is no overview of this. (for that matter, the snow depth was evidently monitored throughout, as shown in fig.2e, but the text lacks any details about how it was measured). Figure 2b shows that UAs are positioned some distance away off the side of the SPCs and exact positioning details need to be shown. Wind directions vary throughout the measurement periods and this will have impacted on the planform mappings (even if direction varies by 10-15 degrees, this will impact on the apparent lateral spacing of time-series).
Analysis: my biggest concern is about the interpretation of patterns shown in fig.6b&c as streamers. First, streamers are generally only 10-20 cm wide (as is also clear from the photo in fig.5b), and yet the bands of high flux in figure 6 are clearly two metres wide or more. These are not streamers in terms of our general understanding. The 1.5 metre spacing between measurement points already precludes mapping these types of small-scale flux structures anyway. Second, the perfectly straight bands in that figure are unnatural and are almost certainly an artifact of differential sensor sensitivity or saturation. The same banding pattern is also visible in the maps of fig.4 and 8a. Real flux patterns would vary laterally over a time-scale of 10 seconds and this should be visible in the maps. I have seen similar artificial banding in transport maps myself and it was related to differential sensor response. I’m not familiar with the SPC used in the study here, but in sand transport studies that use laser counter sensors (‘Wenglors’) we know that the optics can get ‘smudged’ or dusty or affected by condensation on the outside and this leads to degraded performance. A first method of removing the artificial banding is to normalise the individual SPC time-series by a longer-term average. Because the fowling accumulates over time, a Reynolds decomposition using a window of 10-100 seconds may be useful here.
On a final note, I believe the power spectra in fig.7 do not show any significant peaks (as claimed in L112-117), only a typical Kolmogorov-type spectrum slope. The spectrum below frequencies of 10^-1 is far too coarse and unstable to assign meaning to what are essentially noise variations. The suggestion on L116, that coherent flow structures would increase in size between the suspension layer (higher above the surface) and the saltation layer (nearer the surface) runs completely counter to everything we know about turbulence and eddies (decreasing in size nearer to the surface), including the Hunt & Morrison model cited in the study here.
The general observation in this paper so-far that flux tends to relate better to u’>0 events or periods, is also, by the way, a finding that was reported in the sand transport literature by Weaver and Wiggs (2012, GRL).

Citation: https://doi.org/10.5194/egusphere-2023-1845-RC2
- AC2: 'Reply on RC2', Kouichi Nishimura, 30 Dec 2023
  
  Dear Reviewer,
  Thank you very much for carefully reviewing our manuscript and providing insightful comments, especially regarding the standpoint of sand particle transport. From the outset, please understand that, in this manuscript, our intention is to provide a brief overview of this unique observation and the obtained results. As pointed out by the reviewer, our analysis is still quite limited, and we are presenting preliminary results. It is probably essential to emphasize this aspect more prominently in the manuscript. The dataset obtained from the series of observations is substantial, offering various approaches to derive concrete conclusions explaining the spatiotemporal structures of blowing snow. More detailed analyses will be presented in subsequent manuscripts, which are currently in progress.
  All of your comments and suggestions are highly informative and educational. Following your recommendations, we are planning to revise the manuscript extensively through discussions with coauthors. For instance, we are eager to incorporate more detailed information about our experimental setup.
  In practice, the 1.5-meter spacing in our measurements may not be fine enough to discern the precise structures described here, particularly the streamers. However, by improving the color bar in the figures, we hope to indicate more precise and reasonable aspects. At this stage, the transport flux of structures has reached its maximum, hence represented as a simple bold red bar.
  The Snow Particle Counter (SPC) is a well-established blowing snow sensor that outputs both particle size and numbers, representing transport flux every second. As far as I know, it has also been utilized in sand transport research as a "Sand Particle Counter." Additionally, accurate calibrations of all sensors are carried out beforehand, allowing us to reasonably exclude the effect of sensitivity differences.
  In line with the suggestions, we are prepared to reconsider the power spectrum issues that seem to contradict common concepts previously revealed.
  Best regards,
  Kouichi
  
  Citation: https://doi.org/10.5194/egusphere-2023-1845-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Reconsider after major revisions (further review by editor and referees) (05 Feb 2024) by Guillaume Chambon

AR by Kouichi Nishimura on behalf of the Authors (17 Mar 2024) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (20 Mar 2024) by Guillaume Chambon

RR by Nikolas Aksamit (05 Apr 2024)

RR by Anonymous Referee #2 (18 Jun 2024)

Suggestions for revision or reasons for rejection

I have struggled to engage with the revised manuscript because the author-response to my concerns and suggestions is rather disappointing.

First, the authors stress many times that this manuscript only presents “preliminary results”, “a brief overview”, and that “analysis is still quite limited”. This is a very serious problem because scientific research publications should fundamentally not be preliminary, brief, or limited. Preliminary outputs may be presented at a conference, but they cannot be expected to become part of the established scientific literature, since any such preliminary work is by definition potentially still subject to change and modification. On this basis alone I believe this manuscript must be rejected because it is explicitly unable to contribute definitive scientific findings.

Second, my central concerns about the SPC data artefacts have been ignored. The authors stress that their SPC sensors are well calibrated and they give additional technical details, but they entirely fail to engage with my comments on the clearly artificial nature of the unrealistically elevated/depressed signals for specific sensors over long periods of time, as visually evident still in the revised snow flux ‘maps’ of figures 4c, 7, 10a, and 13a. In figure 13a, for example, it is abundantly clear that the elevated signal at Y=3 over a period of 8 seconds is not a genuine feature of variable snow flux, and further scrutiny of the snow flux maps shows that most individual sensors produce signals that are consistently at different relative intensities. The authors have not addressed this issue, nor have they recognised the likelihood that sensor optics are fouled by the challenging field conditions (i.e. independent of calibrations) and the artefacts continue to affect their interpretations.

Third, the author-response and the revisions in the manuscript still lack any recognition of the very significant limitations imposed by the 1.5 m lateral spacing of the instruments, i.e. that the spatial resolution is fundamentally limited to this scale and the data cannot, by definition, reveal any structures or trends that are smaller than this. The author-response claims that “…by adjusting the range of the color bar in the figures, the width of the streamers became reasonably thinner,…” but this is simply a visualization gimmick (further exacerbated by the relative sensor distortions); streamers are generally on the order of 10-20 cm wide, and the instrumentation array is fundamentally incapable of resolving the sort of spatial patterning (streamers) visible in figure 6b. Text in the conclusion (L214-216) claiming to have observed streamer families and nested streamers is simply not sustainable. It also means that the (new) 2D auto-correlation map of Figure 8 has a fundamental lateral (spatial) resolution of 1.5 metres and hence nearly all of the autocorrelation structure is simply an interpolation feature.

Fourth, although the author response notes that “analysis was conducted with all the sensor data,” this is only true for the spatio-temporal maps of snow flux and wind forcing, but instead these are severely limited because only very short snippets of time (on the order of 10s of seconds) are shown and discussed. Other results, such as the auto-correlation, power spectra, and quadrant analysis, are severely limited because they are derived only from single instruments or from just one specific measurement run. There is insufficient consistent analysis of all available data as well as its internal variability to produce any reliable findings.

Partly due to the above problems, the conclusions are still essentially trivial; the only finding that can be somewhat substantiated is the notion that snow flux correlates better with u’>0, but this is also already well-known from the sand transport literature (as is recognised here too). Even this is still quite limited though because the relevant results were only obtained from one measurement run. Authors claim that Table 1 also supports this finding, but this is not correct as that table only reports quadrant percentages, not links with snow flux.

In summary, the manuscript has too many weaknesses, the results are too preliminary and selective, and there are no robust findings that would make a useful contribution to the scientific literature.

Hide

ED: Publish subject to revisions (further review by editor and referees) (24 Jun 2024) by Guillaume Chambon

AR by Kouichi Nishimura on behalf of the Authors (02 Aug 2024) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (19 Aug 2024) by Guillaume Chambon

AR by Kouichi Nishimura on behalf of the Authors (23 Aug 2024) Author's response Manuscript

Post-review adjustments

AA: Author's adjustment | EA: Editor approval

AA by Kouichi Nishimura on behalf of the Authors (12 Oct 2024) Author's adjustment Manuscript

EA: Adjustments approved (16 Oct 2024) by Guillaume Chambon

Journal article(s) based on this preprint

22 Oct 2024

Elucidation of spatiotemporal structures from high-resolution blowing-snow observations

Kouichi Nishimura, Masaki Nemoto, Yoichi Ito, Satoru Omiya, Kou Shimoyama, and Hirofumi Niiya

The Cryosphere, 18, 4775–4786, https://doi.org/10.5194/tc-18-4775-2024,https://doi.org/10.5194/tc-18-4775-2024, 2024

Short summary

Kouichi Nishimura, Masaki Nemoto, Yoichi Ito, Satoru Omiya, Kou Shimoyama, and Hirofumi Niiya

Viewed

Total article views: 446 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
300	108	38	446	21	26

HTML: 300
PDF: 108
XML: 38
Total: 446
BibTeX: 21
EndNote: 26

Views and downloads (calculated since 26 Sep 2023)

Month	HTML	PDF	XML	Total
Sep 2023	25	16	3	44
Oct 2023	49	23	6	78
Nov 2023	22	2	1	25
Dec 2023	47	14	9	70
Jan 2024	18	6	2	26
Feb 2024	28	9	3	40
Mar 2024	20	11	1	32
Apr 2024	14	5	4	23
May 2024	14	6	1	21
Jun 2024	12	6	4	22
Jul 2024	15	4	2	21
Aug 2024	13	3	1	17
Sep 2024	17	2	1	20
Oct 2024	6	1	0	7

Cumulative views and downloads (calculated since 26 Sep 2023)

Month	HTML	PDF	XML	Total
Sep 2023	25	16	3	44
Oct 2023	49	23	6	78
Nov 2023	22	2	1	25
Dec 2023	47	14	9	70
Jan 2024	18	6	2	26
Feb 2024	28	9	3	40
Mar 2024	20	11	1	32
Apr 2024	14	5	4	23
May 2024	14	6	1	21
Jun 2024	12	6	4	22
Jul 2024	15	4	2	21
Aug 2024	13	3	1	17
Sep 2024	17	2	1	20
Oct 2024	6	1	0	7

Viewed (geographical distribution)

Total article views: 472 (including HTML, PDF, and XML) Thereof 472 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 22 Oct 2024

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (2410 KB)
Metadata XML

Short summary

It is crucial to consider organized structures such as turbulence sweeps and ejections when discussing the onset and development of snow transport. This study aims to systematically measure blowing and drifting snow to investigate their spatiotemporal structures. To achieve this goal, we have deployed fifteen Snow Particle Counters (SPCs) in designated test areas and are conducting measurements using an equal number of ultrasonic anemometers, providing high temporal resolution data.


Total:	0
HTML:	0
PDF:	0
XML:	0