the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Apr 2022
Use of fluorescent sand to assess plot-scale hydrological connectivity and sediment transport on young moraines in the Swiss Alps
Abstract. The surface of the earth is constantly changing, especially in mountain areas, where glacial retreat uncovers large bodies of unconsolidated sediment. However, our knowledge on overland flow (OF) generation and related sediment transport is still limited, partly due to a lack of methods to trace the pathways of water and sediment on the surface. To investigate how different surface characteristics affect hydrological- and sediment connectivity on natural hillslopes, we studied five plots on two moraines of different ages in a proglacial area of the Swiss Alps. On all plots, we performed sprinkling experiments to determine OF characteristics, i.e., total OF, peak OF flow rate, peak turbidity and sediment concentrations, and total sediment yield. Here we test if a fluorescent sand tracer, together with UV lamps and a high-resolution camera, can be used to visualize the pathways of OF and to determine the typical sediment transport distances. The results highlight the ability of the setup to detect sand movement, even for individual fluorescent sand particles (300–500 µm grain size), and to distinguish between the two main mechanisms of sediment transport: OF-driven erosion and splash erosion. The experiments also revealed that the higher rock cover on the younger moraine enhanced surface hydrological connectivity and resulted in larger sediment transport distances. In contrast, the higher vegetation cover on the older moraine promoted infiltration and reduced the length of the surface flow pathways and erosion. The study, thus, demonstrates the potential of the new method to improve the observation of surface hydrological connectivity and sediment transport. These observations can help to improve our understanding of OF and sediment transport in complex natural settings.
- Preprint
(2696 KB) - Metadata XML
-
Supplement
(4371 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2022-165', Anonymous Referee #1, 01 Jun 2022
This work assesses the feasibility of using glow-in-the-dark colored sand to study sediment transport in five 4 m × 6 m plots on two Swiss moraines. To this end, sprinkling experiments are executed on the plots, overland flow (OF) is measured with a Bernoulli tube at the downslope end of the plots, and high-resolution photographs are taken before and after sprinkling experiments. Such measurements are used to determine OF characteristics as well as the maximum spatial extent traveled by the sand.
These are the main results:
- If the soil particles of the plot match the dimensions of the colored sand, this approach is successful at showing the redistribution of sediments onto the soil surface. This can be executed even if no OF is measured at the downslope end of the plot and without any pictures/videos being taken during the sprinkling experiments.
- Areas of the plot surface where the colored sand was redistributed tend to be consistent with some of the areas where OF was observed through dye staining experiments.
- Total OF and sediment yield could be predicted by the combination of the plot rock cover, vegetation cover, and rainfall intensity. However, such parameters were not strongly related to the changes in the spatial extent traveled by sand.
- Microtopography and vegetation cover can explain sand redistribution on the plots.
The paper proposes and investigates an experimental method to study the changes of the Earth surface where unconsolidated sediments are uncovered, such as in glacial moraines. This is within the scope of HESS to study the “spatial characteristics of the global water resources and related budgets, in all compartments of the Earth system” as well as “the role of physical processes in the cycling of continental water in all its phases and at all scales”.
While the paper presents novel data, I am not particularly convinced it presents novel concepts, ideas, or tools. The idea of using sediment color to quantify sediment sources and transport has a long tradition in Hydrology, and has been addressed with multiple techniques spanning from diffuse reflectance spectrometry to fluorescent tracers. Additionally, OF timing and surface hydrological connectivity have already been investigated with liquid and particle tracers as well as a vast array of sensing techniques, including image analysis.
I agree with the Authors that the main conclusion of the article is the convenience of using colored sand to partially (since some of the sand disappeared) show the redistribution of sediments in small scale plots without taking pictures/videos continuously during the experiments. This piece of information may be useful for conducting research in environments like those illustrated in the paper.
Additional conclusions entail the facts that sand redistribution most likely occurred along OF paths, and vegetation cover was largely responsible for the observed response in total OF and sediment yield. These conclusions support well-known results available in the literature.
Regarding the scientific methods and assumptions as well as result traceability, I think that further parameters should have been considered and should be included in the paper to better understand how the experiments were executed (see later).
In my opinion, one major issue is that results are not sufficient to support some of the interpretations and conclusions. I do not agree with the Authors that colored sand illustrated OF pathways. OF pathways are routes taken by water and may not necessarily all coincide with paths taken by heavy sand particles. I agree that sediments redistributed where OF occurred, but water may have also followed a lot more paths that were not necessarily taken by sand. OF pathways can be studied by using neutrally buoyant (which do not sink) particles of more varied granulometry than those used by the Authors. Similarly, I do not think that much can be concluded on the surface hydrological connectivity of the plots based on the reported results. For example, even if a detailed description is not provided, it looks like sand was deployed on top of the vegetation in the 1860 plot. After the sprinkling experiments, such sand disappears since it is probably masked by the vegetation top cover. On the other hand, water did not necessarily follow the same path (that is, it did not remain trapped below the vegetation) and, therefore, nothing can be said on the relative OF paths in that area. In parallel, I suggest the title of the article is opportunely edited to include the sole assessment of sediment transport.
To properly assess the consistency of sediment and OF pathways, the Authors should include an experimental validation section where the brilliant blue dye tracer solution is deployed during all sprinkling experiments at locations consistent with the colored sand. By continuously (and automatically) capturing dye pathways, reliable information could be inferred on the hydrological connectivity and overall OF pathways in the plots. For instance, data on the average dimensions and length of the rills/OF pathways would be particularly useful to fully understand the mechanism of sediment transport. A more careful analysis of this aspect would highly enrich the work.
The manuscript did not always give proper credit to related work: several investigations on experimental studies for sediment redistribution are not included among the references (see for instance the work by Martínez-Carreras et al., 2010). Likewise, the objectives of several cited works are not properly stated. Here a few examples.
In the Introduction, it is stated that climate change is also expected to increase the frequency, intensity, and amount of heavy precipitation. This is a very general statement and cited works are not related to glacier covers nor to environments like those studied in the paper. Moreover, reference Maruffi et al., 2022 is not even reported in the Bibliography.
In several instances, the Authors refer to the continuous acquisitions of photos/videos as an experimental disadvantage. However, continuous pictures could provide reliable information on the mechanisms underlying the formation of OF pathways, which, instead, cannot be properly justified in this work (see the conclusions on rain splash detachment and transport that sound rather arbitrary).
The methodology should also undergo an extensive revision based on the specific comments reported below. Presentation quality is generally good apart from minor points highlighted in the following.
Specific comments:
Abstract: I think this part should be re-written by better stating the actual results of the paper and reducing deductions on the hydrological connectivity that cannot be properly supported.
Introduction: the first introductory part on climate change as well as Hortonian and saturation-excess overland flow is too general and should be more cautiously related to the literature.
Regarding the research questions, I think those should be reformulated since fluorescent sand particles cannot be properly used to trace water.
Sprinkling experiments: I think a major flaw exists in the selection of the average rainfall intensities adopted in the experiments. In Table 2, rainfall intensities are not consistent at all among the plots. How can we compare results for the 1860L plot – HI experiment (81 mm/hr) to the 1990L plot – HI experiment (48 mm/hr)? Conversely, the experiment on the 1860L plot – MI (43 mm/hr) replicates a rainfall like that on the 1990H plot – HI. Rainfall durations are also inconsistent. Thus, given such diverse “meteorological forcing”, it is unclear what we can infer from these experiments since plot characteristics, cover, and “complexity” are also diverse.
Image pre-processing and analysis: It is not explained how image spatial resolution was adjusted to make each pixel refer to an area of 1 mm2.
A main assumption of the image analysis method is that “the main part of the sand ribbon did not move during an experiment”. I think this is a rather strong assumption and should at least be supported with photographic material (i.e., a time-lapse during the experiment).
The following methodological details should be considered:
- The time between sand deployment onto the plot surface and the beginning of the sprinkling experiment should be included.
- Likewise, average wind values should be integrated since the wind may have influenced the redistribution of sand particles.
- Further details on how the sand was deployed should also be stated. The presence of vegetation implies that some of the sand particles remained on the top of the vegetation cover while some other reached the soil surface. Such “depth” effects cannot be observed in images but may have most likely influenced the sand redistribution.
- The order at which experiments were executed also plays a role in the results. Sand particles in the final (HI) experiments may have moved through already existing rills. So, are these experiments (and the related results) truly independent on each other?
Discussion: Lines 461-468. I do not think this comment is relevant since some of the cited works aimed at identifying OF pathways/timing. This work instead can only deduce OF paths from sediment redistribution.
Some details on the sand particles should be included as well:
- Were the particles charged? The occurrence of aggregates suggests so and this may have influenced their motion and redistribution onto the soil surface.
- Did photoluminescent material lost from the sand particles coat/bind to other soil particles? This may have created “fake” colored particles.
- Did the sand particles exhibit photo-quenching effects or changes in their color following exposure to sunlight/soil pH? These are typical photoluminescence effects that may have also influenced results. A preliminary laboratory characterization of the sand material should have been executed before experiments outdoors.
Code and data availability
The reported statement is not in agreement with HESS policy (https://www.hydrology-and-earth-system-sciences.net/policies/data_policy.html)
Minor points
Line 289: “plot containing” is unclear
Line 228: moraine is misspelled
In conclusion, I support consideration of this work upon a major revision of its objectives, better pondering of the existing literature, and inclusion of validation experiments that may help in broadening the overall scope of the work. Further, justification of the methodological flaws (see the comments on the sprinkling experiments and the Discussion) should be provided by supporting laboratory tests on the characterization of the sand material.
Citation: https://doi.org/10.5194/egusphere-2022-165-RC1 - AC1: 'Reply on RC1', Fabian Maier, 18 Jul 2022
-
RC2: 'Comment on egusphere-2022-165', Anonymous Referee #2, 29 Sep 2022
This study employs night-time sprinkling experiments, dyed water and fluorescent sand distributed on steep moraine slopes to investigate runoff generation and sediment transport at the plot scale. Drone-based orthophotos, digital surface models and field measurements are used to acquire and analyse surface characteristics, for example slope, vegetation and rock cover, roughness and soil properties. While the movement of fluorescent sand particles was monitored, overland flow and sediment yield from the erosion plots were measured at the downslope end of the erosion plots.
Overall, the paper is well written, and the described method (the innovativeness of which I cannot properly judge) appears as an interesting means to detect and quantify surface runoff, sediment mobilisation and transport (distances) across such steep slopes. The paper also contains interesting insights, for example regarding the influence of stone cover on hydrological connectivity (and consequently also sediment transport), which generally makes it a good contribution for both the hydrological and geomorphological community. I found references both to methodological and scientific papers to be ample, and the aims clearly stated. While generally the conclusions appear to be supported by the results, I think that the “connectivity” part of the objectives could be better elaborated and/or discussed. While “connectivity” is in the title of this study, I feel that the applicability of the fluorescent sand method to investigate connectivity at a more relevant (i.e. larger) spatial scale, for example hillslope-channel coupling should be discussed more explicitly. This is related to the question as to how much connectivity on such small spatial scales matters; if surface runoff re-infiltrates downslope of the plot, then the connectivity observed ON the plot would have limited relevance regarding hillslope-channel coupling. I do acknowledge though that the authors are aware of this, as the title judiciously contains the restriction “plot scale”, but considering the relevance of connectivity at (an even slightly) larger spatial scales (i.e. the hillslope scale) for catchment-scale sediment dynamics, it would be really interesting to learn what the authors think about the potential application to sediment transport on and delivery from hillslopes or hillslope sections larger than their plots. For example how large do they think is the area that can be monitored by their method?
I have a number of concerns that I feel need to be addressed before a revised version could be accepted for publication.
- The relation of shutter speed, ISO setting and motion blur on resulting images is not properly described as “high shutter speed” means shorter exposure times and hence smaller risk of motion blur.
- I would question whether hillslopes whose average (!) gradient differs by as much as 15° are in fact “comparable”. Slope is a factor, and of course it is very difficult to find every combination of inclination, vegetation cover etc on a lateral moraine of limited size. Furthermore, the plots have a different relative position – though generally consisting of the same parent material, footslopes could have different properties than top- or midslope locations due to decades of soil redistribution, for example. So maybe the “comparable” needs to be supported by other evidence or be tuned down a bit by acknowledging the variability that makes strong, unambigious (cor)relations with single controlling factors somewhat unlikely. By the way, the correlations should be reported with the correlation coefficients AND the p value, not only with the latter that only shows the significance, but not the strength (and hence something like relevance) of the correlation.
- A major concern for me is the fact that a digital model derived via SfM is always a digital surface model (DSM) in the presence of vegetation, not an elevation (=bare ground) model. The authors chose to measure slope and roughness in the field (although slope could have been measured on the DSM by measuring points on the slope, their distance and elevation difference) and correctly justify this with vegetation being contained in the DSM. On the other hand, they start the paragraph with the statement that the “DTM was used … to determine the potential direction of OF”. Where vegetation is present, it would be impossible to correctly identify potential flowpaths, and moreover I imagine that the vegetation covers the surface on which the sand particles travel, which might render them invisible, so that the modelled potential flowpaths are not realistic, and can moreover not be compared to the real-world flow paths if/where the latter are not visible.
- Lines 162-169: I did not understand why high-resolution images taken by a different drone were taken for visualisation purposes (“background”) and for determination of rock cover, while you used a different drone (with apparently poorer resolution as I understood the text) to produce the DSM with “coarse” (4 cm is not too coarse anyway) resolution. The authors could explain better why they used which system, and what are “georeferenced photos” – on a steep slope, I can hardly imagine how georeferencing (without orthorectification) could produce images as a suitable mapping basis.
I have added similar, and some more comments (also typo corrections etc) in the annotated version of the manuscript.
-
AC2: 'Reply on RC2', Fabian Maier, 12 Oct 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-165/egusphere-2022-165-AC2-supplement.pdf
Status: closed
-
RC1: 'Comment on egusphere-2022-165', Anonymous Referee #1, 01 Jun 2022
This work assesses the feasibility of using glow-in-the-dark colored sand to study sediment transport in five 4 m × 6 m plots on two Swiss moraines. To this end, sprinkling experiments are executed on the plots, overland flow (OF) is measured with a Bernoulli tube at the downslope end of the plots, and high-resolution photographs are taken before and after sprinkling experiments. Such measurements are used to determine OF characteristics as well as the maximum spatial extent traveled by the sand.
These are the main results:
- If the soil particles of the plot match the dimensions of the colored sand, this approach is successful at showing the redistribution of sediments onto the soil surface. This can be executed even if no OF is measured at the downslope end of the plot and without any pictures/videos being taken during the sprinkling experiments.
- Areas of the plot surface where the colored sand was redistributed tend to be consistent with some of the areas where OF was observed through dye staining experiments.
- Total OF and sediment yield could be predicted by the combination of the plot rock cover, vegetation cover, and rainfall intensity. However, such parameters were not strongly related to the changes in the spatial extent traveled by sand.
- Microtopography and vegetation cover can explain sand redistribution on the plots.
The paper proposes and investigates an experimental method to study the changes of the Earth surface where unconsolidated sediments are uncovered, such as in glacial moraines. This is within the scope of HESS to study the “spatial characteristics of the global water resources and related budgets, in all compartments of the Earth system” as well as “the role of physical processes in the cycling of continental water in all its phases and at all scales”.
While the paper presents novel data, I am not particularly convinced it presents novel concepts, ideas, or tools. The idea of using sediment color to quantify sediment sources and transport has a long tradition in Hydrology, and has been addressed with multiple techniques spanning from diffuse reflectance spectrometry to fluorescent tracers. Additionally, OF timing and surface hydrological connectivity have already been investigated with liquid and particle tracers as well as a vast array of sensing techniques, including image analysis.
I agree with the Authors that the main conclusion of the article is the convenience of using colored sand to partially (since some of the sand disappeared) show the redistribution of sediments in small scale plots without taking pictures/videos continuously during the experiments. This piece of information may be useful for conducting research in environments like those illustrated in the paper.
Additional conclusions entail the facts that sand redistribution most likely occurred along OF paths, and vegetation cover was largely responsible for the observed response in total OF and sediment yield. These conclusions support well-known results available in the literature.
Regarding the scientific methods and assumptions as well as result traceability, I think that further parameters should have been considered and should be included in the paper to better understand how the experiments were executed (see later).
In my opinion, one major issue is that results are not sufficient to support some of the interpretations and conclusions. I do not agree with the Authors that colored sand illustrated OF pathways. OF pathways are routes taken by water and may not necessarily all coincide with paths taken by heavy sand particles. I agree that sediments redistributed where OF occurred, but water may have also followed a lot more paths that were not necessarily taken by sand. OF pathways can be studied by using neutrally buoyant (which do not sink) particles of more varied granulometry than those used by the Authors. Similarly, I do not think that much can be concluded on the surface hydrological connectivity of the plots based on the reported results. For example, even if a detailed description is not provided, it looks like sand was deployed on top of the vegetation in the 1860 plot. After the sprinkling experiments, such sand disappears since it is probably masked by the vegetation top cover. On the other hand, water did not necessarily follow the same path (that is, it did not remain trapped below the vegetation) and, therefore, nothing can be said on the relative OF paths in that area. In parallel, I suggest the title of the article is opportunely edited to include the sole assessment of sediment transport.
To properly assess the consistency of sediment and OF pathways, the Authors should include an experimental validation section where the brilliant blue dye tracer solution is deployed during all sprinkling experiments at locations consistent with the colored sand. By continuously (and automatically) capturing dye pathways, reliable information could be inferred on the hydrological connectivity and overall OF pathways in the plots. For instance, data on the average dimensions and length of the rills/OF pathways would be particularly useful to fully understand the mechanism of sediment transport. A more careful analysis of this aspect would highly enrich the work.
The manuscript did not always give proper credit to related work: several investigations on experimental studies for sediment redistribution are not included among the references (see for instance the work by Martínez-Carreras et al., 2010). Likewise, the objectives of several cited works are not properly stated. Here a few examples.
In the Introduction, it is stated that climate change is also expected to increase the frequency, intensity, and amount of heavy precipitation. This is a very general statement and cited works are not related to glacier covers nor to environments like those studied in the paper. Moreover, reference Maruffi et al., 2022 is not even reported in the Bibliography.
In several instances, the Authors refer to the continuous acquisitions of photos/videos as an experimental disadvantage. However, continuous pictures could provide reliable information on the mechanisms underlying the formation of OF pathways, which, instead, cannot be properly justified in this work (see the conclusions on rain splash detachment and transport that sound rather arbitrary).
The methodology should also undergo an extensive revision based on the specific comments reported below. Presentation quality is generally good apart from minor points highlighted in the following.
Specific comments:
Abstract: I think this part should be re-written by better stating the actual results of the paper and reducing deductions on the hydrological connectivity that cannot be properly supported.
Introduction: the first introductory part on climate change as well as Hortonian and saturation-excess overland flow is too general and should be more cautiously related to the literature.
Regarding the research questions, I think those should be reformulated since fluorescent sand particles cannot be properly used to trace water.
Sprinkling experiments: I think a major flaw exists in the selection of the average rainfall intensities adopted in the experiments. In Table 2, rainfall intensities are not consistent at all among the plots. How can we compare results for the 1860L plot – HI experiment (81 mm/hr) to the 1990L plot – HI experiment (48 mm/hr)? Conversely, the experiment on the 1860L plot – MI (43 mm/hr) replicates a rainfall like that on the 1990H plot – HI. Rainfall durations are also inconsistent. Thus, given such diverse “meteorological forcing”, it is unclear what we can infer from these experiments since plot characteristics, cover, and “complexity” are also diverse.
Image pre-processing and analysis: It is not explained how image spatial resolution was adjusted to make each pixel refer to an area of 1 mm2.
A main assumption of the image analysis method is that “the main part of the sand ribbon did not move during an experiment”. I think this is a rather strong assumption and should at least be supported with photographic material (i.e., a time-lapse during the experiment).
The following methodological details should be considered:
- The time between sand deployment onto the plot surface and the beginning of the sprinkling experiment should be included.
- Likewise, average wind values should be integrated since the wind may have influenced the redistribution of sand particles.
- Further details on how the sand was deployed should also be stated. The presence of vegetation implies that some of the sand particles remained on the top of the vegetation cover while some other reached the soil surface. Such “depth” effects cannot be observed in images but may have most likely influenced the sand redistribution.
- The order at which experiments were executed also plays a role in the results. Sand particles in the final (HI) experiments may have moved through already existing rills. So, are these experiments (and the related results) truly independent on each other?
Discussion: Lines 461-468. I do not think this comment is relevant since some of the cited works aimed at identifying OF pathways/timing. This work instead can only deduce OF paths from sediment redistribution.
Some details on the sand particles should be included as well:
- Were the particles charged? The occurrence of aggregates suggests so and this may have influenced their motion and redistribution onto the soil surface.
- Did photoluminescent material lost from the sand particles coat/bind to other soil particles? This may have created “fake” colored particles.
- Did the sand particles exhibit photo-quenching effects or changes in their color following exposure to sunlight/soil pH? These are typical photoluminescence effects that may have also influenced results. A preliminary laboratory characterization of the sand material should have been executed before experiments outdoors.
Code and data availability
The reported statement is not in agreement with HESS policy (https://www.hydrology-and-earth-system-sciences.net/policies/data_policy.html)
Minor points
Line 289: “plot containing” is unclear
Line 228: moraine is misspelled
In conclusion, I support consideration of this work upon a major revision of its objectives, better pondering of the existing literature, and inclusion of validation experiments that may help in broadening the overall scope of the work. Further, justification of the methodological flaws (see the comments on the sprinkling experiments and the Discussion) should be provided by supporting laboratory tests on the characterization of the sand material.
Citation: https://doi.org/10.5194/egusphere-2022-165-RC1 - AC1: 'Reply on RC1', Fabian Maier, 18 Jul 2022
-
RC2: 'Comment on egusphere-2022-165', Anonymous Referee #2, 29 Sep 2022
This study employs night-time sprinkling experiments, dyed water and fluorescent sand distributed on steep moraine slopes to investigate runoff generation and sediment transport at the plot scale. Drone-based orthophotos, digital surface models and field measurements are used to acquire and analyse surface characteristics, for example slope, vegetation and rock cover, roughness and soil properties. While the movement of fluorescent sand particles was monitored, overland flow and sediment yield from the erosion plots were measured at the downslope end of the erosion plots.
Overall, the paper is well written, and the described method (the innovativeness of which I cannot properly judge) appears as an interesting means to detect and quantify surface runoff, sediment mobilisation and transport (distances) across such steep slopes. The paper also contains interesting insights, for example regarding the influence of stone cover on hydrological connectivity (and consequently also sediment transport), which generally makes it a good contribution for both the hydrological and geomorphological community. I found references both to methodological and scientific papers to be ample, and the aims clearly stated. While generally the conclusions appear to be supported by the results, I think that the “connectivity” part of the objectives could be better elaborated and/or discussed. While “connectivity” is in the title of this study, I feel that the applicability of the fluorescent sand method to investigate connectivity at a more relevant (i.e. larger) spatial scale, for example hillslope-channel coupling should be discussed more explicitly. This is related to the question as to how much connectivity on such small spatial scales matters; if surface runoff re-infiltrates downslope of the plot, then the connectivity observed ON the plot would have limited relevance regarding hillslope-channel coupling. I do acknowledge though that the authors are aware of this, as the title judiciously contains the restriction “plot scale”, but considering the relevance of connectivity at (an even slightly) larger spatial scales (i.e. the hillslope scale) for catchment-scale sediment dynamics, it would be really interesting to learn what the authors think about the potential application to sediment transport on and delivery from hillslopes or hillslope sections larger than their plots. For example how large do they think is the area that can be monitored by their method?
I have a number of concerns that I feel need to be addressed before a revised version could be accepted for publication.
- The relation of shutter speed, ISO setting and motion blur on resulting images is not properly described as “high shutter speed” means shorter exposure times and hence smaller risk of motion blur.
- I would question whether hillslopes whose average (!) gradient differs by as much as 15° are in fact “comparable”. Slope is a factor, and of course it is very difficult to find every combination of inclination, vegetation cover etc on a lateral moraine of limited size. Furthermore, the plots have a different relative position – though generally consisting of the same parent material, footslopes could have different properties than top- or midslope locations due to decades of soil redistribution, for example. So maybe the “comparable” needs to be supported by other evidence or be tuned down a bit by acknowledging the variability that makes strong, unambigious (cor)relations with single controlling factors somewhat unlikely. By the way, the correlations should be reported with the correlation coefficients AND the p value, not only with the latter that only shows the significance, but not the strength (and hence something like relevance) of the correlation.
- A major concern for me is the fact that a digital model derived via SfM is always a digital surface model (DSM) in the presence of vegetation, not an elevation (=bare ground) model. The authors chose to measure slope and roughness in the field (although slope could have been measured on the DSM by measuring points on the slope, their distance and elevation difference) and correctly justify this with vegetation being contained in the DSM. On the other hand, they start the paragraph with the statement that the “DTM was used … to determine the potential direction of OF”. Where vegetation is present, it would be impossible to correctly identify potential flowpaths, and moreover I imagine that the vegetation covers the surface on which the sand particles travel, which might render them invisible, so that the modelled potential flowpaths are not realistic, and can moreover not be compared to the real-world flow paths if/where the latter are not visible.
- Lines 162-169: I did not understand why high-resolution images taken by a different drone were taken for visualisation purposes (“background”) and for determination of rock cover, while you used a different drone (with apparently poorer resolution as I understood the text) to produce the DSM with “coarse” (4 cm is not too coarse anyway) resolution. The authors could explain better why they used which system, and what are “georeferenced photos” – on a steep slope, I can hardly imagine how georeferencing (without orthorectification) could produce images as a suitable mapping basis.
I have added similar, and some more comments (also typo corrections etc) in the annotated version of the manuscript.
-
AC2: 'Reply on RC2', Fabian Maier, 12 Oct 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-165/egusphere-2022-165-AC2-supplement.pdf
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 665 | 256 | 48 | 969 | 90 | 29 | 33 |
- HTML: 665
- PDF: 256
- XML: 48
- Total: 969
- Supplement: 90
- BibTeX: 29
- EndNote: 33
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 314 | 33 |
| Switzerland | 2 | 112 | 12 |
| Germany | 3 | 93 | 10 |
| China | 4 | 66 | 7 |
| Russia | 5 | 43 | 4 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 314