the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Oct 2023
Brief communication: Influence of snow cover on albedo reduction by snow algae
Abstract. Snow algae contribute to snowmelt by darkening the surface, reducing its albedo. However, the potential consequences of algae under the surface (such as after a fresh snowfall) on albedo reduction is not known. In this study, we examined the impact of sub-surface snow algae on surface energy absorption. The results indicate energy absorption across all analysed wavelength ranges when snow algae are snow-covered, an effect that was correlated with both cell densities and chlorophyll-a concentrations. These findings suggest that snow algae lower albedo and thus increase snow melt even when snow-covered.
- Preprint
(699 KB) - Metadata XML
-
Supplement
(71 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2023-2379', Andrew Gray, 30 Nov 2023
-
AC1: 'Reply on RC1', Pablo Almela, 21 Dec 2023
RC1: 'Comment on egusphere-2023-2379', Andrew Gray, 30 Nov 2023
GENERAL
Reviewer: A nice study with a clear potential for impact.
Main concern centers around the concept of albedo used throughout. I wasn't clear that what was being measured was albedo because measurements didn't seem to have been made with respect to incoming solar irracience. The results are still valid, but the termanology needs addressing so as not to confuse the reader and perpetuate the idea that HDRF measurements are the same as Albedo. This could cleared up some in the methods section also. More info needed, for example on controls taken and how the experimental design influenced snow reflectance.
Other minor comments are in attached pdf.
Response: We thank the reviewer for the time dedicated reviewing this manuscript, as well as for the positive remarks about the significance of the presented results. As suggested, the data presented are indeed based on reflectance measurements and not albedo per se. We have carefully revised the manuscript to correct and clarify this concept. We have also addressed minor comments, which together have helped to improve the manuscript. Detailed responses can be found below.
MINOR COMMENTS
Reviewer: Ln 43. Probably better to reference original work here, Lutz, Takeuchi, Yallop etc
Response: We thank the reviewer for this suggestion. A new reference has been included here. Please see line 43.
Reviewer: Lns 76-83. Did you use a spectralon panel as an irradiance measure? And if so did you place it under the PVC pipe?
Response: Yes, a spectralon panel was used to determine the blank. This was done directly on the reference and not through the PVC pipe. More details have been included in the manuscript to clarify this matter. Please see lines 80-81.
Reviewer: Ln 86. It isn't the albedo because its a directional measurement.
Response: We apologize to the reviewer for the misunderstanding. The manuscript has been modified according to the suggestion.
Reviewer: Ln 91. I know its a different system, be we observe much more deeply buried blooms in Antarctica. Ind have measured reflectance impacts under c. 7cm of snow.
Response: We appreciate the reviewer's comment about the presence of snow algae below 2 cm from the surface. In our sampling, we've observed algae below 2 cm as well, but they mostly show up on the surface. We've taken into account the Onuma et al., 2021 model, which suggests that a snow cover over 20mm marks the 'disappearance' of a bloom (sets the initial cell concentration to 0). This reference wasn't included in the previous version but has been included in the current one for better clarity. Due to these considerations, and aligning with Cook et al., 2017, we found it fitting to set the upper limit of our study at 2 cm.
Reviewer: Lns 93-104. Did you run controls to see what reflectance impact your experiment had because of changing the snow structure?
Response: We didn’t run controls to measure the impact on reflectance. We used snow with similar physical characteristics (subsurface snow) to measure the maximum reflectance in our snow field and perform the experiment. We presume that the influence in wavelengths where pigments primarily absorb is attributed mainly to their presence rather than alterations in the snow's physical properties, although these alterations might have some effect, albeit to a varying extent. Hence, while these controls would have been helpful in comprehending the alterations in the snow following manipulation, we don't view them as indispensable.
Reviewer: Ln 107. So the algal cell density measurements are based on a 2cm thick puck of snow?
Response: Yes, the algal measurement was done on a 2 cm deep core corresponding to the same surface where albedo measurements were conducted. On this bloom, the algae was present on the very surface of the snow (as seen in Fig.1) so we decided not to keep digging for counting cells.
Reviewer: Ln 116. It would be nice to know how soon after sampling microscopy was conducted. And if a period greater than a few hours, what preservation methods were used
Response: Microscopy was conducted on the same day of sampling. Some snow was still present in the samples upon arrival at the lab so we assumed that there was no change with respect to what was sampled. More details have been included in the manuscript to clarify this matter. Please see lines 119-120.
Reviewer: Ln 150. Do all of these references actually measure primary production? I don't think Kahn 2021 does
Response: Some of the cited references lack primary production measurements, yet they depict snow algae as the main primary producers within this ecosystem. However, to avoid misunderstandings we have removed the reference from this sentence as suggested by the reviewer.
Reviewer: Ln 153. Would be nice to include a macroinvertebrate paper here too
Response: We thank the reviewer for this suggestion. We agree that incorporating a reference to macroinvertebrates in snow algae blooms would enhance our understanding and improve the insight into the impact of these communities on the ecosystem. This addition provides greater context in this article for the potential reader, and the reference has been included. Please see line 156.
Reviewer: Lns 153-155 Rephrase: Snow algae absorb light energy primarily in the ranges where their specific pigments absorb light most effectively.
Response: The manuscript has been modified accordingly. Please see lines 157-159.
Reviewer: Ln 155. The ASD FS4 measures 3500 to 2500nm
Response: Although the ASD FS4 measures 350 to 2500nm, we have limited our study to the range of 350-1150 because this is where the algae exerts its most significant influence, making it more compelling to showcase.
Reviewer: Ln 175. Again, you're measuring HDRF, not albedo, and albedo should be measured across the solar spectrum instead of the VNIR range
Response: We appreciate the reviewer’s comment. We have modified 'albedo' by ‘HDRF’ or 'reflectance' in those sentences where we considered it necessary. However, we want to highlight that we specify ‘spectral albedo’ and not ‘albedo’, so we would be making it clear that we are referring to a specific wavelength region, in this case between 350 nm and 1150 nm.
Reviewer: Ln 179. Report stats here.
Response: We thank the reviewer for this suggestion. Stats have been reported. Please see line 186.
Reviewer: Lns 181-183. the effect is on albedo/reflectance, not on the algal bloom itself. Reword
Response: The sentence has been reworded. Please see line 189.
Reviewer: Lns 184-185. reword to make a bit clearer
Response: The sentence has been reworded to make it clearer. Please see lines 190-192
Reviewer: Lns 186-188. Or that the snow pack physical structure has been modified in your experiment
Response: Although physical alterations of the snow after manipulation may affect its reflectance, we assume that the magnitude of the algae's impact at the wavelengths where the pigments predominantly absorb is much greater. We also like to emphasize that the results presented here are relative and are not intended to establish standard reflectance values that can be extrapolated as a function of biomass and snow cover, but rather to describe for the first time the energy absorption capacity of these biological communities even when they are covered by snow.
Reviewer: Lns 191-193. This data would be good to show. Did you calculate radiative forcing under the snow?
Response: We do not have this data. For this reason, we have modified the sentence, to make it clear that this has not been measured directly in our study. Please see line 199.
Reviewer: Ln 199. clarify what this means
Response: What we aim to emphasize is that the snow cover over snow algae reduces the pigment absorption (in contrast to a surface bloom), potentially causing their effects to evade detection in visible range scans. Furthermore, because the algae remain unseen by the naked eye, their presence beneath the surface might be misinterpreted as background noise rather than a true impact on the albedo, as this effect hadn't been determined until now. The manuscript has been modified accordingly to clarify these ideas. Please see lines 206-207.
Reviewer: Ln 209. Satellites don't necessarily require this. In fact your results show that chlorophyll absorbance is still evident under surface snow. Which would therefore be detected by a remote sensing approach. What it would do is underestimate the cell density as the chlorophyll signal would be weaker.
Response: We appreciate the reviewer's suggestion. The manuscript has been modified accordingly, including the idea of the ‘biomass underestimation’ suggested. Please see lines 226-229.
Reviewer: Lns 212-214. I don't think you can make this conclusion about metabolic activity without measuring it. And if you have a field reference or irradiance, you could actually measure the radiative forcing with your data, which would be really useful.
Response: We agree that we cannot ensure that metabolic activity exists when algae occur below the surface. This is something we need to study, and we are already planning an experiment for the next melting season. Consequently, we have removed any mention of metabolic activity from the manuscript.
Reviewer: Figure 2
I may be misinterpreting your methods, but I don't think these show what the graph says they do. Firstly, they're not albedo, but I've discussed that. Secondly, if the subsequent measurements are taken relative to a clean snow surface (as a reference), then it is a measure or HDRF relative to clean snow. Not an intrinsic property of the bloom itself.
Response: We thank the reviewer for this suggestion. The figure has been modified by replacing ‘albedo’ with ‘HDRF’. However, it's important to highlight that the measurements depicted in this figure are not taken relative to a clean snow surface for reference but relative to the bloom, and that is why the reflectance increases as we add more layers).
Reviewer: Another point, were the clean snow references made with the PVC pipe too? If not then a lot of the lower HDRF could be accounted for from the pipe. The light you measure reflecting from snow typically enters the snow pack a few cm away from the point where you take the measurement, especially if you are working with a 2cm FOV.
Response: We thank the reviewer's concern for the methodological approach used in this work. The clean snow reference was not made through the PVC pipe. We agree that this could impact the HDRF values obtained, but it would mainly occur at wavelengths below 350 nm (e.g., Hasan et al., 2015: 10.1039/C4RA16043F) so it should not invalidate our results.
Citation: https://doi.org/10.5194/egusphere-2023-2379-AC1 -
AC2: 'Reply on RC1 (UPDATED VERSION)', Pablo Almela, 20 Jan 2024
Dear Reviewer 1,
Please consider the updated document where we have edited the responses, taking into account the suggestions from reviewers 2 and 3. Simultaneously, we have made adjustments to the line numbers associated with each response.
Thanks again for taking the time and effort to review the manuscript.
—--------------------------------------------------------------------------------------------
RC1: 'Comment on egusphere-2023-2379', Andrew Gray, 30 Nov 2023
GENERAL
Reviewer: A nice study with a clear potential for impact.
Main concern centers around the concept of albedo used throughout. I wasn't clear that what was being measured was albedo because measurements didn't seem to have been made with respect to incoming solar irracience. The results are still valid, but the termanology needs addressing so as not to confuse the reader and perpetuate the idea that HDRF measurements are the same as Albedo. This could cleared up some in the methods section also. More info needed, for example on controls taken and how the experimental design influenced snow reflectance.
Other minor comments are in attached pdf.
Response: We thank the reviewer for the time dedicated reviewing this manuscript, as well as for the positive remarks about the significance of the presented results. As suggested, the data presented are indeed based on reflectance measurements and not albedo per se. We have carefully revised the manuscript to correct and clarify this concept.
Other significant modifications, as per the recommendations of the other reviewers, comprise:
-We have reanalyzed our HDRF data and conducted comparisons by calculating the area under the curve (AUC).
-Figure 3A now depicts variations in reflectance rather than percentages, and the same applies to Figure 3B (second column).
We have also addressed minor comments, which together have helped to improve the manuscript. Detailed responses can be found below.
MINOR COMMENTS
Reviewer: Ln 43. Probably better to reference original work here, Lutz, Takeuchi, Yallop etc
Response: We thank the reviewer for this suggestion. A new reference has been included here. Please see line 45.
Reviewer: Lns 76-83. Did you use a spectralon panel as an irradiance measure? And if so did you place it under the PVC pipe?
Response: Yes, a spectralon panel was used to determine the blank. This was done directly on the reference and not through the PVC pipe. More details have been included in the manuscript to clarify this matter. Please see lines 87-88.
Reviewer: Ln 86. It isn't the albedo because its a directional measurement.
Response: We apologize to the reviewer for the misunderstanding. The manuscript has been modified according to the suggestion.
Reviewer: Ln 91. I know its a different system, be we observe much more deeply buried blooms in Antarctica. Ind have measured reflectance impacts under c. 7cm of snow.
Response: We appreciate the reviewer's comment about the presence of snow algae below 2 cm from the surface. In our experience sampling across the mountain west region of the US, we've also observed algae below 2 cm. However, based on our observations in real-time in the field at this snowfield, 2 cm seemed like a reasonable depth which is also consistent with the Onuma et al., 2021 model. This model suggests that a snow cover over 20mm marks the 'disappearance' of a bloom (sets the initial cell concentration to 0). This reference wasn't included in the previous version and has been included in the current one for better clarity. There was misunderstanding with Cook et al., 2017, since it was referring to ice and not snow, so this reference has not been included in the current version. Please see lines 101-103.
Reviewer: Lns 93-104. Did you run controls to see what reflectance impact your experiment had because of changing the snow structure?
Response: We didn’t run controls to measure the impact on reflectance. We used snow with similar physical characteristics (subsurface snow) to measure the maximum reflectance in our snowfield and perform the experiment. We presume that the influence in wavelengths where pigments primarily absorb is attributed mainly to their presence rather than alterations in the snow's physical properties, although these alterations might have some effect, albeit to a varying extent. Hence, while these controls would have been helpful in comprehending the alterations in the snow following manipulation, we don't view them as indispensable.
Reviewer: Ln 107. So the algal cell density measurements are based on a 2cm thick puck of snow?
Response: Yes, the algal measurement was done on a 2 cm deep core corresponding to the same surface where albedo measurements were conducted. More details have been included in the manuscript (lines 118-119). On this bloom, the algae was present on the very surface of the snow (as seen in Fig.1) so we decided not to keep digging for counting cells.
Reviewer: Ln 116. It would be nice to know how soon after sampling microscopy was conducted. And if a period greater than a few hours, what preservation methods were used
Response: Microscopy was conducted on the same day of sampling. Some snow was still present in the samples upon arrival at the lab so we assumed that there was no change with respect to what was sampled. More details have been included in the manuscript to clarify this matter. Please see line 127.
Reviewer: Ln 150. Do all of these references actually measure primary production? I don't think Kahn 2021 does
Response: Some of the cited references lack primary production measurements, yet they depict snow algae as the main primary producers within this ecosystem. However, to avoid misunderstandings we have removed the reference from this sentence as suggested by the reviewer. Lines 162-163.
Reviewer: Ln 153. Would be nice to include a macroinvertebrate paper here too
Response: We thank the reviewer for this suggestion. We agree that incorporating a reference to macroinvertebrates in snow algae blooms would enhance our understanding and improve the insight into the impact of these communities on the ecosystem. This addition provides greater context in this article for the potential reader, and the reference has been included. Please see line 165.
Reviewer: Lns 153-155 Rephrase: Snow algae absorb light energy primarily in the ranges where their specific pigments absorb light most effectively.
Response: The manuscript has been modified accordingly. Please see line 166.
Reviewer: Ln 155. The ASD FS4 measures 3500 to 2500nm
Response: Although the ASD FS4 measures 350 to 2500nm, we have limited our study to the range of 350-1150 because this is where the algae exerts its most significant influence, making it more compelling to showcase. More details have been included in the methodology to clarify this point (lines 83-86).
Reviewer: Ln 175. Again, you're measuring HDRF, not albedo, and albedo should be measured across the solar spectrum instead of the VNIR range
Response: We appreciate the reviewer’s comment. We have changed 'albedo' to either ‘HDRF’ or 'reflectance' in those cases where we refer to our data (e.g., lines 80, 94, 104, 106, 117, 161, 168, 180, 181, 197, 200)
Reviewer: Ln 179. Report stats here.
Response: We thank the reviewer for this suggestion. Stats have been reported. Please see line 198.
Reviewer: Lns 181-183. the effect is on albedo/reflectance, not on the algal bloom itself. Reword
Response: The sentence has been reworded. Please see line 199.
Reviewer: Lns 184-185. reword to make a bit clearer
Response: The sentence has been reworded to make it clearer. Please see lines 201.
Reviewer: Lns 186-188. Or that the snow pack physical structure has been modified in your experiment
Response: Although physical alterations of the snow after manipulation may affect its reflectance, we assume that the magnitude of the algae's impact at the wavelengths where the pigments predominantly absorb is much greater. We also like to emphasize that the results presented here are relative and are not intended to establish standard reflectance values that can be extrapolated as a function of biomass and snow cover, but rather to describe for the first time the energy absorption capacity of these biological communities even when they are covered by snow.
Reviewer: Lns 191-193. This data would be good to show. Did you calculate radiative forcing under the snow?
Response: We do not have this data. For this reason, we have modified the sentence, to make it clear that this has not been measured directly in our study. Please see line 212.
Reviewer: Ln 199. clarify what this means
Response: The manuscript has been modified accordingly to clarify these ideas. Please see lines 220-224.
Reviewer: Ln 209. Satellites don't necessarily require this. In fact your results show that chlorophyll absorbance is still evident under surface snow. Which would therefore be detected by a remote sensing approach. What it would do is underestimate the cell density as the chlorophyll signal would be weaker.
Response: We appreciate the reviewer's suggestion. The manuscript has been modified accordingly, including the idea of the ‘biomass underestimation’ suggested. Please see lines 235-238.
Reviewer: Lns 212-214. I don't think you can make this conclusion about metabolic activity without measuring it. And if you have a field reference or irradiance, you could actually measure the radiative forcing with your data, which would be really useful.
Response: We agree that we cannot ensure that metabolic activity exists when algae occur below the surface. This is something we need to study, and we are already planning an experiment for the next melting season. Consequently, we have removed any mention of metabolic activity from the manuscript.
Reviewer: Figure 2
I may be misinterpreting your methods, but I don't think these show what the graph says they do. Firstly, they're not albedo, but I've discussed that. Secondly, if the subsequent measurements are taken relative to a clean snow surface (as a reference), then it is a measure or HDRF relative to clean snow. Not an intrinsic property of the bloom itself.
Response: We thank the reviewer for this suggestion. The figure has been edited accordingly.
Reviewer: Another point, were the clean snow references made with the PVC pipe too? If not then a lot of the lower HDRF could be accounted for from the pipe. The light you measure reflecting from snow typically enters the snow pack a few cm away from the point where you take the measurement, especially if you are working with a 2cm FOV.
Response: We thank the reviewer's concern for the methodological approach used in this work. The clean snow reference was not made through the PVC pipe. We agree that this could impact the HDRF values obtained, but it would mainly occur at wavelengths below 350 nm (e.g., Hasan et al., 2015: 10.1039/C4RA16043F) so it should not invalidate our results.
Citation: https://doi.org/10.5194/egusphere-2023-2379-AC2
-
AC1: 'Reply on RC1', Pablo Almela, 21 Dec 2023
-
RC2: 'Comment on egusphere-2023-2379', Alexander Thomson, 02 Jan 2024
A neat, compact study on a pressing subject with clear ramifications to the quantification of BAR from snow algae.
There are a couple of issues with the final plot (3B) and regression analysis which require addressing. I am also uncertain about the use of "% albedo increase" as the central unit in the study. I think this would be worthwhile re-assessing, perhaps using relative albedo values calculated against a standard (panel or clean snow standard). But this is not my field of expertise, and so I am happy to be corrected on this.
The results are impactful with wide ranging implications for snow algae research, including remote sensing approaches, quantification of BAR, and ecology. That there are implications from the work is pointed out in in the text. It would great to see these implications, as well as some of the direct implications of the results (e.g. variation in response of different absorption wavelengths to snow depth), explored further in the discussion. But I understand that this is a brief communication piece and may be limited for words and space.
Full summary comments and in document comments attached below.
-
AC3: 'Reply on RC2', Pablo Almela, 20 Jan 2024
RC2: 'Comment on egusphere-2023-2379', Alexander Thomson, 02 Jan 2024
GENERAL
Reviewer: A nice study with a clear potential for impact.
General comments:
A neat, compact study on a pressing subject with clear ramifications to the quantification of BAR from snow algae.
My main query revolves around the use of “% albedo increase” as the reporting unit. I am not totally familiar with spectral and albedo research, however I think this has the potential to confuse the discussion, and to misrepresent some relationships (e.g. albedo vs snow depth). Perhaps it would be worthwhile to replot the data against a relative albedo value instead (either relative to a control panel measurement, or a snow control measurement)?
I am also unsure about the plots and linear regression approaches in figure 3B. The figure
legends and labelling of points was fairly unclear. The values for 0 cm snow depth appear to be missing from plot 3 B, and the corresponding regression analysis. I am also not sure linear regression values are useful on values from sequential experimental results like those in the cell density plot, or values that have not been normalised to cell density.
The results are impactful with wide ranging implications for snow algae research, including
remote sensing approaches, quantification of BAR, and ecology. That there are implications from the work is pointed out in the text, but it would great to see these implications explored further in the results and discussion.
Response: We thank the reviewer for dedicating time to review our manuscript. We sincerely appreciate all your valuable comments and suggestions, which helped us in improving the quality of the manuscript. We have carefully revised the manuscript to correct and clarify the concept of reflectance (HDRF), since the initial version included a misconception regarding the measured parameter (not albedo). Consequently, we have reworked Fig3, and the data now illustrates 'HDRF' and 'increase in HDRF'. We have reanalyzed our data and conducted comparisons by calculating the area under the curve (AUC), which we acknowledge as a more accurate approximation, as suggested by rev 3. We opted for a linear regression test due to the independence of observations and the expected linear relationship between the tested variables. The correlations are more robust as we present reflectance vs snow cover as a dependent variable and the increase in reflectance vs cell density. We have also addressed minor comments, which together have helped to improve the manuscript. Detailed responses can be found below.
Comments in text (line number indicated):
32 - More no? See Hotaling review table in below ref.
Response: Albedo reduction has been updated. Please see line 34.
33 - More so than ice algae?
Response: We thank the reviewer for the suggestion. Ice algae make a greater contribution to BAR reduction (i.e., Hotaling et al. (2021). The ms has been modified accordingly. Please see line 35.
38 - There is much greater diversity than this. Perhaps rephrase along the lines of "dominant taxa observed in snow algae bloom formation". As a US counter example see Van Hees et al. 2023 on Chlainomonas spp. Blooms
Response: We appreciate the reviewer's suggestion, and the manuscript has been edited accordingly. Please see line 35. However, Chlainomonas spp. has not been included as a dominant taxon in the manuscript, given that it does not typically represent a majority group, except in the specific region where the cited study was conducted. We will show these results in a paper soon.
38 - Orange blooms are common in the Antarctic (Remias et al 2013)
Response: We have modified the manuscript, stating that blooms can acquire a reddish color (which would include orange) and we included the reference. Please see line 40.
43 - Maybe find a primary reference showing cell abundance vs IRF/melt?
Response: We thank the reviewer for this suggestion. A new reference has been included here. Please see line 45.
46 - rephrase sentence
Response: We thank the reviewer for this suggestion. The manuscript has been revised and modified. Please see lines 47-48.
92 - Personally Ive observed snow algae in dense layers much deeper than 2 cm (5-15 cm). Whether these blooms were actively growing is beside the point to this study which is assessing impact on reflectance, not biology. I would have perhaps tested to deeper snow depths.
Response: We appreciate the reviewer's comment about the presence of snow algae below 2 cm from the surface. In our samplings during the summer, we've observed algae below 2 cm as well, but they mostly show up on the surface. We've taken into account the Onuma et al., 2021 model, which suggests that a snow cover over 20mm marks the 'disappearance' of a bloom (sets the initial cell concentration to 0). This reference wasn't included in the previous version but has been included in the current one for better clarity. Please see line103. We found it fitting to set the upper limit of our study at 2 cm, particularly based on our observations in real-time in the field. Future studies could expand on this and explore blooms with obvious abundant pigmented biomass deeper in the snowpack.
95 - My understanding of albedo and spectral data is limited, however would it not be better to measure the reduction in albedo from a set standard (a control panel? or the control snow patch?) ? I am unsure about the wording of albedo "increasing" in this way, even if the increase is a proportional increase (%). Ive commented below, more specifically on the results, too.
Response: We thank the reviewer for the suggestion. In the new version of the manuscript we have specified that a spectralon was used for the white reference (please see lines 87-88). We have also considered it appropriate to remove the phrase you mention about proportional increase, since we now present different data in Fig 3.
98 - was the PVC tube used for the reference as well?
Response: The clean snow reference was not made through the PVC pipe. This could impact the HDRF values obtained, but it would mainly occur at wavelengths below 350 nm (e.g., Hasan et al., 2015: 10.1039/C4RA16043F) so it should not invalidate our results. In this version we have also specified that the PVC cylinder was not in the field of view when measuring reflectance. Please see line 96-97.
100 - I think assuming generally consistent snow structure across your study snow patch and samples is reasonable. However it would be useful to report some of those characteristics (grain size, density) for future comparison with other sites/snow packs. My understanding of spectral/snow physics are poor, but I assume that the snow structure will make a significant difference to how much snow cover affects BAR?
Response: We agree with the reviewer that the physical properties of snow play a key role in the BAR effects. However, we do not have this information since at the time of sampling our SLF Snow Sensor was not working. Therefore we regret that we cannot include that information in the manuscript.
108 - Any evidence of subsurface cells? What was the depth of the observed bloom?
Response: This algae bloom was concentrated in the surface, with no evident presence of cells below the initial centimeter. As a precautionary measure, we sampled a core of snow extending 2 centimeters deep (refer to lines 117-119 for details).
118 - was the composition of cells/species consistent across the 6 measurement plots?
Response: While it would be interesting to know the composition of the algal communities in the six plots, we didn't extract DNA from these samples. Our main objective did not involve studying the community and, lacking experience as taxonomists in snow algae, we cannot confirm if there were differences in the compositions.
137 - those?
Response: We thank the reviewer for the suggestion. The manuscript has been edited accordingly.
138 - not sure the last section of this sentence adds to the discussion. Too many variables to make such statements.
Response: This sentence has been removed from the manuscript.
171 - was there a difference in the effect of snow depth on reflectance in different absorption ranges? Does green and red absorption get reduced at the same rate with snow depth increase? Would be interesting to discuss
Response: We thank the reviewer for the suggestion and agree that it would be interesting to include this comparison in the discussion. The manuscript has been edited accordingly. Please see lines 185-193.
176 - see comment above, and on Figure 3B
179 - so absorption in these wavelengths is potentially affected more by snow cover?
Worth discussing?
Response: The data in question has been omitted from the manuscript and is consequently not available for discussion. However, in response to your earlier suggestion, we have included a very interesting comparison of absorption within the chlorophyll and carotenoid ranges (see line 185-193).
180 - See comments on figure 3B - the cell density plot is not all that clear, but is it suggesting that snow cover has a greater effect on higher biomass blooms? Again, i think using "increase in albedo" maybe confuses this point. Plotting relative albedo vs cell density instead might show the relationship more clearly? And more intuitively show that high cell count results in lower albedo, and that the increase in reflectance with snow cover is steeper for high biomass blooms (which i think your figure 2 plot 6 shows as well)?
Response: We thank the reviewer for the suggestion. We have modified the data shown in Figure 3. We now present ‘HDRF’ (instead of ‘albedo’) and 'increase in HDRF' (instead of 'increase in albedo (%)'). Now it is clearly shown how a greater cell density is associated with a greater increase in reflectance when snow is added on top of the snow algae.
183 - See comments on graph, but could you not normalise for cell density before performing linear regression between reflectance and snow depth? This is conflating two factors (cell density and snow depth) on reflectance. As discussed in comments above, that cell density interacts with the effect on albedo is interesting in itself, and could be explored, but not sure this linear regression is the best way to do so.
Response: We thank the reviewer for the suggestion. As mentioned above, we have modified the data shown in Figure 3. We now have HDRF (instead of albedo) and 'increase in HDRF' (instead of 'increase in albedo (%)'). The latter refers to the changes in reflectance with respect to the initial measurements of the snow algae on the surface, but it is no longer a percentage but the difference between both measurements. This allows us to relativize the measurements in each plot and thus compare their effect between them. Although we considered normalizing the data (log cell density vs reflectance), it did not improve clarity visually or in regression analyses. Consequently, we have opted not to modify this section. We hope all these changes will enhance the clarity of the data presented in the manuscript.
191 - indicate is a bit strong. perhaps "suggest". Unless you can include melt data or calculations?
Response: The manuscript has been edited accordingly. Please see line 212.
198 - Agree in general with this, and a crucial data point in calculating IRF across regions from remote sensing methods. But also agree with previous reviewer - your data suggests spectral approaches might well detect some chlorophyll absorption through shallow snow cover! Maybe worth expanding this comment/discussion?
Response: We agree with the reviewer. The manuscript has been edited accordingly to rev1 and rev2. Please see lines 220-222.
203 - Was this shown in this analysis? Any evidence for this?
Response: We thank the reviewer for the comment. Our data did not provide sufficient evidence to conclusively support this claim. Therefore, the statement has been removed from the manuscript.
206 - Just suggestions, but could the ramifications of this study be explored further?
- could this understanding help constrain the boundaries of detection of covered snow algae? or allow more refined detection of subsurface blooms?
- what is the relative impact of snow cover on BAR at a landscape scale? Do covered snow algae blooms contribute a missed proportion of BAR using current methods? If so is that missed proportion significant? A rough back of the envelope simulation/model of a bloom/snowfield might be interesting to test this?
- any ecological significance? green vs red snow blooms? PS efficiency at depth vs surface?
Response: We appreciate the reviewer's suggestion. In the conclusions, we have incorporated a comment regarding the potential consequences of underestimating snowmelt rates (please see lines 241-244). But because this is a brief communication, we are limited in elaborating extensively on these details. We really hope that the manuscript will lead to further studies considering subsurface blooms and their potential impact.
213 - Maybe needs rephrasing - the findings suggest subsurface snow habitats might still be suitable for PS activity in snow algae, facilitated by absorption and localised melting effects
Response: We thank the reviewer for this suggestion. We have decided to remove this sentence from the conclusions following the comments of reviewer 1. We do not have data that proves PS below the surface, but we are already working on it to test it as soon as we have the opportunity.
216 - could these be explored?
Response: We thank the reviewer for this suggestion. The manuscript has been modified accordingly, as mentioned above. Please see lines 241-244.
Fig 1D - Any scale?
Response: We have incorporated the scale in the revised version. Please see line 255.
Fig 3B - should this not be normalised for cell counts? weak correlation is due to the variation in cell count, not the weak relationship between snow depth and reflectance
Fig 3B - Is this showing that reflectance increases more steeply when you have a higher biomass? The cell count plots and figure legend are very unclear - are the points spectral measurement replicates? Again, perhaps worth trying to plot these using relative albedo (vs a control value), rather than increase in absorption? Also, similarly to plots next door, is an r2 value appropriate if you are plotting points from the same plot at different snow depths?
Fig 3B - Should the values for 0 cm snow cover not be included? And the regression not extend to this? This seems to me to be of central importance to the study, as reflectance seems to increase more steeply between 0 cm and 0.5 cm depth? Though again, may be an artefact of using "% increase in albedo"
Response: We thank the reviewer for all suggestions regarding Figure 3. Figure 3B has been modified, as discussed above. This new version resolves the weak correlations shown previously, clarifies the data shown (the figure has been redesigned) and also presents the data for 0 cm snow cover as suggested.
Citation: https://doi.org/10.5194/egusphere-2023-2379-AC3
-
AC3: 'Reply on RC2', Pablo Almela, 20 Jan 2024
-
RC3: 'Comment on egusphere-2023-2379', S. McKenzie Skiles, 09 Jan 2024
Snow is bright natural surface in the visible wavelengths where ice is transparent and snow reflectance is dominated by frequent scattering opportunities at ice/air interfaces. It is well understood that, in the visible wavelengths, light penetrates tens of centimeters into the snowpack and thereby, light absorbing particles present in the near surface will lower the snow albedo. The authors assess this impact for very near surface snow algae using field experiments whereby they used an ASD field spectrometer to measure the spectral reflectance of snow with algae at the surface, and then incrementally covered the surface with visibly cleaner subsurface snow, measuring the spectral reflectance with each increment. They additionally collected snow sampled and analyzed characteristics of snow algae (cell counts and chlorophyll) and related those results to reflectance measurements.
This is a nicely focused field study, the scope of which is suitable for a brief communication. It will be of interest to the snow algae/LAP communities following revisions to clarify methods and results before publication. My feedback is organized first by major feedback and then line-by-line revisions.
Major:
The authors should be consistent in referring to their ASD measurements by the right terminology. The measurements are the hemispherical conical reflectance factor (HCRF), not albedo. Even though the authors do recognize that light is not reflected by snow evenly in all directions and that they made a directional measurement, they go on to indicate that reflectance and albedo are equivalent (not the case). Albedo, the ratio of incoming to outgoing solar radiation over the hemisphere, cannot be measured with a field spectrometers bare optical fiber and requires a remote cosine receptor (RCR) foreoptic. The authors cite Cook et al., 2017 in the paper, if they refer back to this paper they may find figures 5 and 6 informative, as well as the discussion of best practices in field spectroscopy. Another useful paper is Schaepman-Strub, 2006 which discusses in detail the difference between measurable remote sensing quantities. The measurements in this study should only be referred to as reflectance, and ideally HCRF (commonly simplified to HDRF), including updating axis labels on Figure 2 and Figure 3.
Cook et al., 2017 - https://tc.copernicus.org/articles/11/2611/2017/tc-11-2611-2017.pdf
Related, and could be very useful, but behind paywall: Cook et al., 2019 Bio-optical properties of Terrestrial Snow and Ice https://link.springer.com/chapter/10.1007/978-3-030-20587-4_3
Schaepman-Strub et al., 2006 - https://doi.org/10.1016/j.rse.2006.03.002
Relatedly, the methods section and Figure 1 need to be updated to make the ASD measurement methods more clear. For example – the use of a white reference panel is not mentioned despite it being a necessary part of reflectance measurements. Also, discussion is needed as to why they chose to measure at an angle in the backscatter direction, and to calculate the instruments field of view. Additionally, please clarify that the PVC pipe was not in the field of view or used in the reflectance measurement, as may be inferred from Figure 1 (which should also be updated). Finally, please describe how the broadband values, used compare to algae properties, were calculated (i.e. integration with respect to spectral irradiance).
I think it would also be useful to better motivate the experiment and contextualize the results to include, even briefly as would be suitable for a brief communication, more background on the current state of the science. Specifically in terms spectral controls on snow albedo (LAPs in visible, ice absorption/grain size in NIR), how those controls determine the depth to which light penetrates (deeper in the visible then NIR wavelengths), and why that is relevant for algae and BAR. For example, multiple studies have previously researched 1) light penetration at different depths for algae photosynthesis/photo-inhibition, and 2) the impact of below surface LAPs on snow albedo. It is not clear in the study, as currently written, that the current state of the science is such that we understand how far light penetrates into snow and that the impact that algae (and other LAPs) has on albedo, when present beneath cleaner snow layers, can be modeled. For example, both the SNICAR-ADv3 and the PBSAM model can simulate absorption by algae in subsurface layers. I’ll use one of my studies as an example (but only because it is best known to me, the authors are not under any pressure to include these specific citations in their study): using a gravimetrics measuring board Skiles and Painter, 2017 collected 3 cm incremental samples across the top 30 cm of the snowpack and related dust and BC concentration to albedo over a daily spring time series, finding that dust in the top 6-8 centimeters exhibits the strongest relationship with albedo decline. Skiles and Painter, 2018 then used those measurements to run the multilayer version of the SNICAR model, representing dust and BC in multiple layers, including beneath cleaner snow. Onuma and Takeuchi have done similar studies in Greenland for algae, dust, and BC. A strong motivation for this work would be that, although we can model it, there are few in situ reflectance studies (such as this one) in midlatitude environments of algae laden snow with cleaner snow on top.
References that could be helpful for more background (not exhaustive)-
Snow optical properties, light penetration into snow: Warren, 1982, Libois et al, 2013, Warren et al., 2006, Zhou ett al. 2003, Warren et al. 2013
Snow/algae: Curl et al., 1972, Gorton et al., 2001, Stibal et al., 2007, Cook et al., 2017 (already cited, but in a limited fashion)
SNICAR/PBSAM: Flanner et al., 2021, Onuma et al., 2020
Impact of mineral dust, at different depths below the surface, on albedo:
Skiles and Painter, 2017, Skiles and Painter, 2018
BC at different depths: Doherty et al., 2013
Line-by-line
15-16: The optical properties fresh snow and older aged snow, like that used in this study to cover algae, are quite different. Fresh snow is much brighter and smaller grained, I would remove the parenthetical aside (such as after fresh snowfall) because that would not be an outcome of the reflectance measurements from this specific experiment.
18: This portion of the abstract would be better used to describe the quantitative results from the experiment– what were the relationships between reflectance, depth below surface, and cell densities/chlorophyll.
Line 71: Can you clarify what you mean by snow algae ‘could be clearly seen in the distance’ – is this to indicate it was visibly widespread surrounding the site of the experiment?
Line 77: Include spectral range and resolution of instrument.
Line 82: Reiterate point made above, because it is not the typical approach, please provide some justification as to why measurements were made at an angle (and in the backscatter direction?). For that angle and distance above target, please provide the approximate area being measured for context and comparison to sample size being collected.
Line 84: Reiterating point made above that this is not an albedo measurement – and to please clarify the role of the PVC cylinder.
Line 87: The reflectance measurements would indicate that the snow is not free of LAPs (the visible reflectance would be higher), and almost certainly they are there, given the time of the experiment in late summer and that widespread algae growth corresponds with deposition that provides nutrients. It could be useful to clarify that relative to algae laden snow at the surface, the subsurface snow was visually brighter and therefore used as a proxy for clean snow. It would also be useful to include in results/discussion that analysis was not undertaken to quantify other LAP concentrations, and therefore would be unknown and an uncertainty in the presented results.
Line 91/92: : There would not be such a distinct/hard limit – it would depend on many factors including snow properties, surface roughness, and liquid water content. I also don’t not see this statement in Cook et al., 2017 – 2 cm is mentioned as a practical limit on sample collection in ice - but algae in Greenland ice would also be different then algae on a snowfield so another citation/justification for this depth would be more suitable.
Line 106 - Is there a reason to use pigment instead of chlorophyll in the section header?
Line 121 – It is more typical to report snow density in kg m-3, consider including in place of or in addition to the current units. In August the snow on a snowfield would be significantly more dense than 200 kg m-3, probably more in the range of 500-600 kg m-3, unless it had very recently snowed (from this pictures this does not appear to be the case). Please justify the use of this value.
Lines 131-134 – Belongs in methods
Line 137 – There are more studies now that have studied algae content, it would be useful to have more studies included for reference (especially ones that are not at very high latitudes such as Greenland/the Arctic). Comparing midlatitude sites to arctic sites for algae is not apples to apples, it would expected that cell counts would be lower in the arctic relative to midlatitude study locations given constraints on algae growth (solar, water, nutrients).
Line 139 – It would be useful to have pictures of the six plots, which could be suitably included in any of the figures. If you chose them based on the visual intensity of their color variation in concentration would be a descriptor of the expected outcome of the experimental design, it could be useful to phrase it in this way.
Line 155 – An ASD Fieldspec 4 measures 350 – 2500 nm, please clarify why the spectral range indicated here is not the full range and provide a justification (in methods) about why the full range was not used.
Line 160 – Please clarify (in methods) how these comparisons are made - by integrating over the carotenoid and chlorophyll absorption features? Or by using average reflectance values (which would not be appropriate given reflectance magnitude is sensitive to measurement and solar zenith angles).
Line 167 – In the near infrared absorption is dominated by ice, and reflectance signatures would be indicative of effective grain radius/ice path length (for reference see Fig. 1 in Cook et al., 2017). Please clarify.
Line 178 (and broader results reporting) – Working from spectral reflectance to a single broadband albedo value is not trivial, and from the described methods it was not done in this paper. Even going from spectral reflectance to broadband reflectance requires integrating with respect to irradiance since incoming solar radiation is not evenly distributed across the solar spectrum. More details are needed on how the results presented in Figure 3 were calculated. Please report differences in the actual values as opposed to a percent difference, given the broadband values themselves are not reported (i.e 0.4 increasing to 0.7, delta0.3).
Line 199 -200 – Please clarify the point being made here - you should be able to detect algae beneath the surface from visible/spectral measurements given that it is absorbing and lowers the albedo – this is the outcome of the study (and is the basis of LAP radiative forcing retrievals from remote sensing, e.g. Painter et al., 2012). Visible to the eye and visible to remote sensing are two different things, and remote sensing has many different spectral and spatial scales making it hard to blanket statements about remote sensing can and cannot do. Also, if it is present at a depth where it doesn’t lower the albedo, then it would be undetectable, but it also wouldn’t be relevant for BAR?
Painter et al, 2012: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2012GL052457
202 – As noted in the study algae primarily reduces albedo/reflectance in the visible wavelengths, and NIR is indicative of ice/grain size absorption - so this statement needs clarification or more justification.
(In the reflectance plots, the NIR reflectance is highest for the cleaner snow - it was subsurface and cleaner, so it makes sense that the grain size would be smaller due to lower energy inputs to drive snow metamorphism - therefore would have higher NIR reflectance. The NIR is lower for the snow algae at surface because it's absorbing, and the grains would be much larger. In the plots the NIR reflectance is essentially the same for the 'snow added experiments' that indicates the grain size of the added snow is consistent. Light penetrates ~1-2 cm into snow in the NIR wavelengths, so it would make sense that your ‘snow added’ spectra in the NIR would be (spectrally) a combination between the cleaner snow spectra and the algae laden spectra.)
Line 206- I would suggest taking this out or adding more context and specifics on how these results would be integrated into watershed melt models, which is quite complicated.
Line 213 – Since this study did not calculate melt rates or do a time series study to measure them, it would be more suitable to stick to relevant conclusions drawn from the field experiment or provide more context for this statement.
Line 214 – To reiterate a point made above, if it lowers the albedo then it is detectable by remote sensing – but I might be missing the point being made – please clarify.
Citation: https://doi.org/10.5194/egusphere-2023-2379-RC3 -
AC4: 'Reply on RC3', Pablo Almela, 20 Jan 2024
RC3: 'Comment on egusphere-2023-2379', S. McKenzie Skiles, 09 Jan 2024
GENERAL
Reviewer: Snow is bright natural surface in the visible wavelengths where ice is transparent and snow reflectance is dominated by frequent scattering opportunities at ice/air interfaces. It is well understood that, in the visible wavelengths, light penetrates tens of centimeters into the snowpack and thereby, light absorbing particles present in the near surface will lower the snow albedo. The authors assess this impact for very near surface snow algae using field experiments whereby they used an ASD field spectrometer to measure the spectral reflectance of snow with algae at the surface, and then incrementally covered the surface with visibly cleaner subsurface snow, measuring the spectral reflectance with each increment. They additionally collected snow sampled and analyzed characteristics of snow algae (cell counts and chlorophyll) and related those results to reflectance measurements.
This is a nicely focused field study, the scope of which is suitable for a brief communication. It will be of interest to the snow algae/LAP communities following revisions to clarify methods and results before publication. My feedback is organized first by major feedback and then line-by-line revisions.
Response: We appreciate the reviewer for dedicating time to review our manuscript and for the positive comments on the significance of the presented results.The manuscript has been thoroughly revised to clarify what is referred to as major. Additionally, we have addressed minor comments, contributing to overall enhancements in the manuscript. Please find detailed responses below.
Major:
The authors should be consistent in referring to their ASD measurements by the right terminology. The measurements are the hemispherical conical reflectance factor (HCRF), not albedo. Even though the authors do recognize that light is not reflected by snow evenly in all directions and that they made a directional measurement, they go on to indicate that reflectance and albedo are equivalent (not the case). Albedo, the ratio of incoming to outgoing solar radiation over the hemisphere, cannot be measured with a field spectrometers bare optical fiber and requires a remote cosine receptor (RCR) foreoptic. The authors cite Cook et al., 2017 in the paper, if they refer back to this paper they may find figures 5 and 6 informative, as well as the discussion of best practices in field spectroscopy. Another useful paper is Schaepman-Strub, 2006 which discusses in detail the difference between measurable remote sensing quantities. The measurements in this study should only be referred to as reflectance, and ideally HCRF (commonly simplified to HDRF), including updating axis labels on Figure 2 and Figure 3.
Cook et al., 2017 - https://tc.copernicus.org/articles/11/2611/2017/tc-11-2611-2017.pdf
Related, and could be very useful, but behind paywall: Cook et al., 2019 Bio-optical properties of Terrestrial Snow and Ice https://link.springer.com/chapter/10.1007/978-3-030-20587-4_3
Schaepman-Strub et al., 2006 - https://doi.org/10.1016/j.rse.2006.03.002
Relatedly, the methods section and Figure 1 need to be updated to make the ASD measurement methods more clear. For example – the use of a white reference panel is not mentioned despite it being a necessary part of reflectance measurements. Also, discussion is needed as to why they chose to measure at an angle in the backscatter direction, and to calculate the instruments field of view. Additionally, please clarify that the PVC pipe was not in the field of view or used in the reflectance measurement, as may be inferred from Figure 1 (which should also be updated). Finally, please describe how the broadband values, used compare to algae properties, were calculated (i.e. integration with respect to spectral irradiance).
I think it would also be useful to better motivate the experiment and contextualize the results to include, even briefly as would be suitable for a brief communication, more background on the current state of the science. Specifically in terms spectral controls on snow albedo (LAPs in visible, ice absorption/grain size in NIR), how those controls determine the depth to which light penetrates (deeper in the visible then NIR wavelengths), and why that is relevant for algae and BAR. For example, multiple studies have previously researched 1) light penetration at different depths for algae photosynthesis/photo-inhibition, and 2) the impact of below surface LAPs on snow albedo. It is not clear in the study, as currently written, that the current state of the science is such that we understand how far light penetrates into snow and that the impact that algae (and other LAPs) has on albedo, when present beneath cleaner snow layers, can be modeled. For example, both the SNICAR-ADv3 and the PBSAM model can simulate absorption by algae in subsurface layers. I’ll use one of my studies as an example (but only because it is best known to me, the authors are not under any pressure to include these specific citations in their study): using a gravimetrics measuring board Skiles and Painter, 2017 collected 3 cm incremental samples across the top 30 cm of the snowpack and related dust and BC concentration to albedo over a daily spring time series, finding that dust in the top 6-8 centimeters exhibits the strongest relationship with albedo decline. Skiles and Painter, 2018 then used those measurements to run the multilayer version of the SNICAR model, representing dust and BC in multiple layers, including beneath cleaner snow. Onuma and Takeuchi have done similar studies in Greenland for algae, dust, and BC. A strong motivation for this work would be that, although we can model it, there are few in situ reflectance studies (such as this one) in midlatitude environments of algae laden snow with cleaner snow on top.
References that could be helpful for more background (not exhaustive)-
Snow optical properties, light penetration into snow: Warren, 1982, Libois et al, 2013, Warren et al., 2006, Zhou ett al. 2003, Warren et al. 2013
Snow/algae: Curl et al., 1972, Gorton et al., 2001, Stibal et al., 2007, Cook et al., 2017 (already cited, but in a limited fashion)
SNICAR/PBSAM: Flanner et al., 2021, Onuma et al., 2020
Impact of mineral dust, at different depths below the surface, on albedo:
Skiles and Painter, 2017, Skiles and Painter, 2018
BC at different depths: Doherty et al., 2013
Response: We thank the reviewer for the valuable suggestions. We apologize for the confusion in the terminology used in the previous version. The measurements in this study are now only referred to reflectance (HDRF). We have also edited the current manuscript by contextualizing the impacts of other light absorbing impurities (LAIs), and specifying several studies that assessed the vertical effect on the albedo reduction of these contaminants. However, we are limited in the number of references we can include and by the word count in the short communication format. The methodology has been clarified. We hope that the edits and added information allows the potential reader to contextualize our study a little better.
Line-by-line
15-16: The optical properties fresh snow and older aged snow, like that used in this study to cover algae, are quite different. Fresh snow is much brighter and smaller grained, I would remove the parenthetical aside (such as after fresh snowfall) because that would not be an outcome of the reflectance measurements from this specific experiment.
Response: We thank the reviewer for its comment on the different properties of fresh and old snow. In this sentence we indicate that snow algae can be buried after blooming by a snowfall, to give the reader context about our study, but not as something that is directly addressed. To provide further context, we've included an additional example—avalanches—illustrating how algae might occur beneath the surface without involving fresh snow. Please see line 16.
18: This portion of the abstract would be better used to describe the quantitative results from the experiment– what were the relationships between reflectance, depth below surface, and cell densities/chlorophyll.
Response: We appreciate the reviewer's suggestion and we agree that incorporating qualitative results into the abstract would be valuable. Unfortunately, the abstract is restricted to 100 words and thus we chose the current format to succinctly summarize our results.
Line 71: Can you clarify what you mean by snow algae ‘could be clearly seen in the distance’ – is this to indicate it was visibly widespread surrounding the site of the experiment?
Response: We find it important to mention that the bloom exhibited a high cell density, as its presence was evident from a distance. We have rephrased this to make it clearer. Please see line 75
Line 77: Include spectral range and resolution of instrument.
Response: The spectral range and resolution of the instrument have been incorporated in the revised version. Please see lines 80-86.
Line 82: Reiterate point made above, because it is not the typical approach, please provide some justification as to why measurements were made at an angle (and in the backscatter direction?). For that angle and distance above target, please provide the approximate area being measured for context and comparison to sample size being collected.
Response: An angle of 60 degrees was used to be sure that we were measuring reflectance at an angle other than the direction of sunlight at the time of the experiment. The approximate measured area has been included in the manuscript. Please see line 92.
Line 84: Reiterating point made above that this is not an albedo measurement – and to please clarify the role of the PVC cylinder.
Response: We thank the reviewer for the suggestion. In the manuscript, the term 'albedo' has been substituted with either 'reflectance' or 'HDRF' consistently. The role of the PVC was to control the thickness of the snow layers added. This part of the manuscript has been edited accordingly. Please see lines 95-96.
Line 87: The reflectance measurements would indicate that the snow is not free of LAPs (the visible reflectance would be higher), and almost certainly they are there, given the time of the experiment in late summer and that widespread algae growth corresponds with deposition that provides nutrients. It could be useful to clarify that relative to algae laden snow at the surface, the subsurface snow was visually brighter and therefore used as a proxy for clean snow. It would also be useful to include in results/discussion that analysis was not undertaken to quantify other LAP concentrations, and therefore would be unknown and an uncertainty in the presented results.
Response: In the current version of the manuscript, we have specified that the subsurface snow was visually brighter (please see line 98), and also clarify that LAP concentrations other than algae were not quantified (please see lines 205-209).
Line 91/92: : There would not be such a distinct/hard limit – it would depend on many factors including snow properties, surface roughness, and liquid water content. I also don’t not see this statement in Cook et al., 2017 – 2 cm is mentioned as a practical limit on sample collection in ice - but algae in Greenland ice would also be different then algae on a snowfield so another citation/justification for this depth would be more suitable.
Response: We apologize to the reviewer for the misunderstanding with Cook et al., 2017. In our experience sampling across the mountain west region of the US, we've also observed algae below 2 cm. However, based on our observations in real-time in the field at this snowfield, 2 cm seemed like a reasonable depth which is also consistent with the Onuma et al., 2021 model. This model suggests that a snow cover over 20mm marks the 'disappearance' of a bloom (sets the initial cell concentration to 0). This reference wasn't included in the previous version and has been included in the current one for better clarity. Please see lines 101-103.
Line 106 - Is there a reason to use pigment instead of chlorophyll in the section header?
Response: There is no reason, and specifying which pigment has been analyzed would be more convenient. Therefore, the manuscript has been modified accordingly. Please see line 116.
Line 121 – It is more typical to report snow density in kg m-3, consider including in place of or in addition to the current units. In August the snow on a snowfield would be significantly more dense than 200 kg m-3, probably more in the range of 500-600 kg m-3, unless it had very recently snowed (from this pictures this does not appear to be the case). Please justify the use of this value.
Response: There was no special rationale behind the density of 200 kg/m3 beyond being a general value used for it. Following the reviewer's recommendation, we have changed this value to 500 kg/m3 and updated the related data. See line 132.
Lines 131-134 – Belongs in methods
Response: We thank the reviewer for this suggestion. The sentence has been relocated to the methodology section (refer to line 121-122), and the results and discussion paragraph has been revised accordingly (see line 147-148).
Line 137 – There are more studies now that have studied algae content, it would be useful to have more studies included for reference (especially ones that are not at very high latitudes such as Greenland/the Arctic). Comparing midlatitude sites to arctic sites for algae is not apples to apples, it would expected that cell counts would be lower in the arctic relative to midlatitude study locations given constraints on algae growth (solar, water, nutrients).
Response: We thank the reviewer for this suggestion. Unfortunately we are limited with the number of references we can include in a brief communication. However, the intent behind this statement is to show the global variability in cell densities within snow algae blooms, providing readers with a comprehensive contextual understanding. Given that cell density is related with the quantity of pigments/carotenoids (as shown in this study) and pigments/carotenoids are related with the impact on the BAR reduction, we believe it is important to present this range in cell densities occurrence.
Line 139 – It would be useful to have pictures of the six plots, which could be suitably included in any of the figures. If you chose them based on the visual intensity of their color variation in concentration would be a descriptor of the expected outcome of the experimental design, it could be useful to phrase it in this way.
Response: We agree with the reviewer that including the pictures of the plots could be interesting for a better description of the samples. Unfortunately, we do not have the pictures of the six plots so we cannot show them in the manuscript.
Line 155 – An ASD Fieldspec 4 measures 350 – 2500 nm, please clarify why the spectral range indicated here is not the full range and provide a justification (in methods) about why the full range was not used.
Response: We chose this spectral range for biological relevance, as biological processes and structural features relevant to organisms are prominent in this interval. Also to simplify the data and graphs presented in this study. The manuscript has been modified accordingly. Please see lines 83-86.
Line 160 – Please clarify (in methods) how these comparisons are made - by integrating over the carotenoid and chlorophyll absorption features? Or by using average reflectance values (which would not be appropriate given reflectance magnitude is sensitive to measurement and solar zenith angles).
Response: We thank the reviewer for this comment. We have reanalyzed our data and conducted comparisons by calculating the area under the curve (AUC), which we acknowledge as a more accurate approximation, as suggested. As a result, we have made corresponding edits to the methodology (lines 139-141), results (lines 168-187), and Figure 3.
Line 167 – In the near infrared absorption is dominated by ice, and reflectance signatures would be indicative of effective grain radius/ice path length (for reference see Fig. 1 in Cook et al., 2017). Please clarify.
Response: We thank the reviewer for this suggestion. Since we want to focus on the biological impact, this statement has been removed from the sentence to avoid any misunderstanding.
Line 178 and broader results reporting) – Working from spectral reflectance to a single broadband albedo value is not trivial, and from the described methods it was not done in this paper. Even going from spectral reflectance to broadband reflectance requires integrating with respect to irradiance since incoming solar radiation is not evenly distributed across the solar spectrum. More details are needed on how the results presented in Figure 3 were calculated. Please report differences in the actual values as opposed to a percent difference, given the broadband values themselves are not reported (i.e 0.4 increasing to 0.7, delta0.3).(
Response: We thank the reviewer for this suggestion. In the current version of the manuscript, we have not use the term '% albedo increase' as it was found to be confusing and may not be the most fitting. Instead, our reporting now revolves around 'HDRF' and 'increase in HDRF,' reflecting alterations in reflectance across snow layers. We trust that this edit contributes to a clearer interpretation of the results outlined in the manuscript.
Line 199 -200 – Please clarify the point being made here - you should be able to detect algae beneath the surface from visible/spectral measurements given that it is absorbing and lowers the albedo – this is the outcome of the study (and is the basis of LAP radiative forcing retrievals from remote sensing, e.g. Painter et al., 2012). Visible to the eye and visible to remote sensing are two different things, and remote sensing has many different spectral and spatial scales making it hard to blanket statements about remote sensing can and cannot do. Also, if it is present at a depth where it doesn’t lower the albedo, then it would be undetectable, but it also wouldn’t be relevant for BAR?
Painter et al, 2012: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2012GL052457
Response: We thank the reviewer for the suggestion. This sentence has been edited following reviewer 1's suggestions. However, isn't it possible that the snow cover prevents the algae from being visible to the naked eye? In figure 1 it can be seen how, after adding 2 cm of snow, there is no apparent trace of the bloom. But indeed, this should not prevent them from being detected by remote sensing. We hope that the edits have clarified this point in the manuscript. Please see lines 220-222.
202 – As noted in the study algae primarily reduces albedo/reflectance in the visible wavelengths, and NIR is indicative of ice/grain size absorption - so this statement needs clarification or more justification.
Response: We thank the reviewer for the suggestion. This sentence has been removed following reviewer 1's suggestions.
(In the reflectance plots, the NIR reflectance is highest for the cleaner snow - it was subsurface and cleaner, so it makes sense that the grain size would be smaller due to lower energy inputs to drive snow metamorphism - therefore would have higher NIR reflectance. The NIR is lower for the snow algae at surface because it's absorbing, and the grains would be much larger. In the plots the NIR reflectance is essentially the same for the 'snow added experiments' that indicates the grain size of the added snow is consistent. Light penetrates ~1-2 cm into snow in the NIR wavelengths, so it would make sense that your ‘snow added’ spectra in the NIR would be (spectrally) a combination between the cleaner snow spectra and the algae laden spectra.)
Response: We really thank the reviewer for this clarification about the spectra in the NIR for the snow layers added to the plots. This helps us better understand the physical properties of the environment where the algae is blooming, and therefore understand the results.
Line 206- I would suggest taking this out or adding more context and specifics on how these results would be integrated into watershed melt models, which is quite complicated.
Response: We thank the reviewer for the suggestion. This sentence has been removed in the current version.
Line 213 – Since this study did not calculate melt rates or do a time series study to measure them, it would be more suitable to stick to relevant conclusions drawn from the field experiment or provide more context for this statement.
Response: We have modified this part in alignment with the feedback provided by the other reviewers. Nevertheless, we consider it important to mention the potential increase in melting rates, as it stands as a pivotal potential conclusion from this study (highlighted merely as a suggestion). Given our demonstrated evidence of energy absorption by the algae under the surface, there is a logical expectation that this absorbed energy would be transmitted as heat to the snow surrounding the cells, consistent with what happens on the surface.
Line 214 – To reiterate a point made above, if it lowers the albedo then it is detectable by remote sensing – but I might be missing the point being made – please clarify.
Response: We apologize to the reviewer for the misunderstanding in the first version of the manuscript. Now we have clarified this part.
Citation: https://doi.org/10.5194/egusphere-2023-2379-AC4
-
AC4: 'Reply on RC3', Pablo Almela, 20 Jan 2024
Status: closed
-
RC1: 'Comment on egusphere-2023-2379', Andrew Gray, 30 Nov 2023
A nice study with a clear potential for impact.
Main concern centers around the concept of albedo used throughout. I wasn't clear that what was being measured was albedo because measurements didn't seem to have been made with respect to incoming solar irracience. The results are still valid, but the termanology needs addressing so as not to confuse the reader and perpetuate the idea that HDRF measurements are the same as Albedo. This could cleared up some in the methods section also. More info needed, for example on controls taken and how the experimental design influenced snow reflectance.
Other minor comments are in attached pdf.
-
AC1: 'Reply on RC1', Pablo Almela, 21 Dec 2023
RC1: 'Comment on egusphere-2023-2379', Andrew Gray, 30 Nov 2023
GENERAL
Reviewer: A nice study with a clear potential for impact.
Main concern centers around the concept of albedo used throughout. I wasn't clear that what was being measured was albedo because measurements didn't seem to have been made with respect to incoming solar irracience. The results are still valid, but the termanology needs addressing so as not to confuse the reader and perpetuate the idea that HDRF measurements are the same as Albedo. This could cleared up some in the methods section also. More info needed, for example on controls taken and how the experimental design influenced snow reflectance.
Other minor comments are in attached pdf.
Response: We thank the reviewer for the time dedicated reviewing this manuscript, as well as for the positive remarks about the significance of the presented results. As suggested, the data presented are indeed based on reflectance measurements and not albedo per se. We have carefully revised the manuscript to correct and clarify this concept. We have also addressed minor comments, which together have helped to improve the manuscript. Detailed responses can be found below.
MINOR COMMENTS
Reviewer: Ln 43. Probably better to reference original work here, Lutz, Takeuchi, Yallop etc
Response: We thank the reviewer for this suggestion. A new reference has been included here. Please see line 43.
Reviewer: Lns 76-83. Did you use a spectralon panel as an irradiance measure? And if so did you place it under the PVC pipe?
Response: Yes, a spectralon panel was used to determine the blank. This was done directly on the reference and not through the PVC pipe. More details have been included in the manuscript to clarify this matter. Please see lines 80-81.
Reviewer: Ln 86. It isn't the albedo because its a directional measurement.
Response: We apologize to the reviewer for the misunderstanding. The manuscript has been modified according to the suggestion.
Reviewer: Ln 91. I know its a different system, be we observe much more deeply buried blooms in Antarctica. Ind have measured reflectance impacts under c. 7cm of snow.
Response: We appreciate the reviewer's comment about the presence of snow algae below 2 cm from the surface. In our sampling, we've observed algae below 2 cm as well, but they mostly show up on the surface. We've taken into account the Onuma et al., 2021 model, which suggests that a snow cover over 20mm marks the 'disappearance' of a bloom (sets the initial cell concentration to 0). This reference wasn't included in the previous version but has been included in the current one for better clarity. Due to these considerations, and aligning with Cook et al., 2017, we found it fitting to set the upper limit of our study at 2 cm.
Reviewer: Lns 93-104. Did you run controls to see what reflectance impact your experiment had because of changing the snow structure?
Response: We didn’t run controls to measure the impact on reflectance. We used snow with similar physical characteristics (subsurface snow) to measure the maximum reflectance in our snow field and perform the experiment. We presume that the influence in wavelengths where pigments primarily absorb is attributed mainly to their presence rather than alterations in the snow's physical properties, although these alterations might have some effect, albeit to a varying extent. Hence, while these controls would have been helpful in comprehending the alterations in the snow following manipulation, we don't view them as indispensable.
Reviewer: Ln 107. So the algal cell density measurements are based on a 2cm thick puck of snow?
Response: Yes, the algal measurement was done on a 2 cm deep core corresponding to the same surface where albedo measurements were conducted. On this bloom, the algae was present on the very surface of the snow (as seen in Fig.1) so we decided not to keep digging for counting cells.
Reviewer: Ln 116. It would be nice to know how soon after sampling microscopy was conducted. And if a period greater than a few hours, what preservation methods were used
Response: Microscopy was conducted on the same day of sampling. Some snow was still present in the samples upon arrival at the lab so we assumed that there was no change with respect to what was sampled. More details have been included in the manuscript to clarify this matter. Please see lines 119-120.
Reviewer: Ln 150. Do all of these references actually measure primary production? I don't think Kahn 2021 does
Response: Some of the cited references lack primary production measurements, yet they depict snow algae as the main primary producers within this ecosystem. However, to avoid misunderstandings we have removed the reference from this sentence as suggested by the reviewer.
Reviewer: Ln 153. Would be nice to include a macroinvertebrate paper here too
Response: We thank the reviewer for this suggestion. We agree that incorporating a reference to macroinvertebrates in snow algae blooms would enhance our understanding and improve the insight into the impact of these communities on the ecosystem. This addition provides greater context in this article for the potential reader, and the reference has been included. Please see line 156.
Reviewer: Lns 153-155 Rephrase: Snow algae absorb light energy primarily in the ranges where their specific pigments absorb light most effectively.
Response: The manuscript has been modified accordingly. Please see lines 157-159.
Reviewer: Ln 155. The ASD FS4 measures 3500 to 2500nm
Response: Although the ASD FS4 measures 350 to 2500nm, we have limited our study to the range of 350-1150 because this is where the algae exerts its most significant influence, making it more compelling to showcase.
Reviewer: Ln 175. Again, you're measuring HDRF, not albedo, and albedo should be measured across the solar spectrum instead of the VNIR range
Response: We appreciate the reviewer’s comment. We have modified 'albedo' by ‘HDRF’ or 'reflectance' in those sentences where we considered it necessary. However, we want to highlight that we specify ‘spectral albedo’ and not ‘albedo’, so we would be making it clear that we are referring to a specific wavelength region, in this case between 350 nm and 1150 nm.
Reviewer: Ln 179. Report stats here.
Response: We thank the reviewer for this suggestion. Stats have been reported. Please see line 186.
Reviewer: Lns 181-183. the effect is on albedo/reflectance, not on the algal bloom itself. Reword
Response: The sentence has been reworded. Please see line 189.
Reviewer: Lns 184-185. reword to make a bit clearer
Response: The sentence has been reworded to make it clearer. Please see lines 190-192
Reviewer: Lns 186-188. Or that the snow pack physical structure has been modified in your experiment
Response: Although physical alterations of the snow after manipulation may affect its reflectance, we assume that the magnitude of the algae's impact at the wavelengths where the pigments predominantly absorb is much greater. We also like to emphasize that the results presented here are relative and are not intended to establish standard reflectance values that can be extrapolated as a function of biomass and snow cover, but rather to describe for the first time the energy absorption capacity of these biological communities even when they are covered by snow.
Reviewer: Lns 191-193. This data would be good to show. Did you calculate radiative forcing under the snow?
Response: We do not have this data. For this reason, we have modified the sentence, to make it clear that this has not been measured directly in our study. Please see line 199.
Reviewer: Ln 199. clarify what this means
Response: What we aim to emphasize is that the snow cover over snow algae reduces the pigment absorption (in contrast to a surface bloom), potentially causing their effects to evade detection in visible range scans. Furthermore, because the algae remain unseen by the naked eye, their presence beneath the surface might be misinterpreted as background noise rather than a true impact on the albedo, as this effect hadn't been determined until now. The manuscript has been modified accordingly to clarify these ideas. Please see lines 206-207.
Reviewer: Ln 209. Satellites don't necessarily require this. In fact your results show that chlorophyll absorbance is still evident under surface snow. Which would therefore be detected by a remote sensing approach. What it would do is underestimate the cell density as the chlorophyll signal would be weaker.
Response: We appreciate the reviewer's suggestion. The manuscript has been modified accordingly, including the idea of the ‘biomass underestimation’ suggested. Please see lines 226-229.
Reviewer: Lns 212-214. I don't think you can make this conclusion about metabolic activity without measuring it. And if you have a field reference or irradiance, you could actually measure the radiative forcing with your data, which would be really useful.
Response: We agree that we cannot ensure that metabolic activity exists when algae occur below the surface. This is something we need to study, and we are already planning an experiment for the next melting season. Consequently, we have removed any mention of metabolic activity from the manuscript.
Reviewer: Figure 2
I may be misinterpreting your methods, but I don't think these show what the graph says they do. Firstly, they're not albedo, but I've discussed that. Secondly, if the subsequent measurements are taken relative to a clean snow surface (as a reference), then it is a measure or HDRF relative to clean snow. Not an intrinsic property of the bloom itself.
Response: We thank the reviewer for this suggestion. The figure has been modified by replacing ‘albedo’ with ‘HDRF’. However, it's important to highlight that the measurements depicted in this figure are not taken relative to a clean snow surface for reference but relative to the bloom, and that is why the reflectance increases as we add more layers).
Reviewer: Another point, were the clean snow references made with the PVC pipe too? If not then a lot of the lower HDRF could be accounted for from the pipe. The light you measure reflecting from snow typically enters the snow pack a few cm away from the point where you take the measurement, especially if you are working with a 2cm FOV.
Response: We thank the reviewer's concern for the methodological approach used in this work. The clean snow reference was not made through the PVC pipe. We agree that this could impact the HDRF values obtained, but it would mainly occur at wavelengths below 350 nm (e.g., Hasan et al., 2015: 10.1039/C4RA16043F) so it should not invalidate our results.
Citation: https://doi.org/10.5194/egusphere-2023-2379-AC1 -
AC2: 'Reply on RC1 (UPDATED VERSION)', Pablo Almela, 20 Jan 2024
Dear Reviewer 1,
Please consider the updated document where we have edited the responses, taking into account the suggestions from reviewers 2 and 3. Simultaneously, we have made adjustments to the line numbers associated with each response.
Thanks again for taking the time and effort to review the manuscript.
—--------------------------------------------------------------------------------------------
RC1: 'Comment on egusphere-2023-2379', Andrew Gray, 30 Nov 2023
GENERAL
Reviewer: A nice study with a clear potential for impact.
Main concern centers around the concept of albedo used throughout. I wasn't clear that what was being measured was albedo because measurements didn't seem to have been made with respect to incoming solar irracience. The results are still valid, but the termanology needs addressing so as not to confuse the reader and perpetuate the idea that HDRF measurements are the same as Albedo. This could cleared up some in the methods section also. More info needed, for example on controls taken and how the experimental design influenced snow reflectance.
Other minor comments are in attached pdf.
Response: We thank the reviewer for the time dedicated reviewing this manuscript, as well as for the positive remarks about the significance of the presented results. As suggested, the data presented are indeed based on reflectance measurements and not albedo per se. We have carefully revised the manuscript to correct and clarify this concept.
Other significant modifications, as per the recommendations of the other reviewers, comprise:
-We have reanalyzed our HDRF data and conducted comparisons by calculating the area under the curve (AUC).
-Figure 3A now depicts variations in reflectance rather than percentages, and the same applies to Figure 3B (second column).
We have also addressed minor comments, which together have helped to improve the manuscript. Detailed responses can be found below.
MINOR COMMENTS
Reviewer: Ln 43. Probably better to reference original work here, Lutz, Takeuchi, Yallop etc
Response: We thank the reviewer for this suggestion. A new reference has been included here. Please see line 45.
Reviewer: Lns 76-83. Did you use a spectralon panel as an irradiance measure? And if so did you place it under the PVC pipe?
Response: Yes, a spectralon panel was used to determine the blank. This was done directly on the reference and not through the PVC pipe. More details have been included in the manuscript to clarify this matter. Please see lines 87-88.
Reviewer: Ln 86. It isn't the albedo because its a directional measurement.
Response: We apologize to the reviewer for the misunderstanding. The manuscript has been modified according to the suggestion.
Reviewer: Ln 91. I know its a different system, be we observe much more deeply buried blooms in Antarctica. Ind have measured reflectance impacts under c. 7cm of snow.
Response: We appreciate the reviewer's comment about the presence of snow algae below 2 cm from the surface. In our experience sampling across the mountain west region of the US, we've also observed algae below 2 cm. However, based on our observations in real-time in the field at this snowfield, 2 cm seemed like a reasonable depth which is also consistent with the Onuma et al., 2021 model. This model suggests that a snow cover over 20mm marks the 'disappearance' of a bloom (sets the initial cell concentration to 0). This reference wasn't included in the previous version and has been included in the current one for better clarity. There was misunderstanding with Cook et al., 2017, since it was referring to ice and not snow, so this reference has not been included in the current version. Please see lines 101-103.
Reviewer: Lns 93-104. Did you run controls to see what reflectance impact your experiment had because of changing the snow structure?
Response: We didn’t run controls to measure the impact on reflectance. We used snow with similar physical characteristics (subsurface snow) to measure the maximum reflectance in our snowfield and perform the experiment. We presume that the influence in wavelengths where pigments primarily absorb is attributed mainly to their presence rather than alterations in the snow's physical properties, although these alterations might have some effect, albeit to a varying extent. Hence, while these controls would have been helpful in comprehending the alterations in the snow following manipulation, we don't view them as indispensable.
Reviewer: Ln 107. So the algal cell density measurements are based on a 2cm thick puck of snow?
Response: Yes, the algal measurement was done on a 2 cm deep core corresponding to the same surface where albedo measurements were conducted. More details have been included in the manuscript (lines 118-119). On this bloom, the algae was present on the very surface of the snow (as seen in Fig.1) so we decided not to keep digging for counting cells.
Reviewer: Ln 116. It would be nice to know how soon after sampling microscopy was conducted. And if a period greater than a few hours, what preservation methods were used
Response: Microscopy was conducted on the same day of sampling. Some snow was still present in the samples upon arrival at the lab so we assumed that there was no change with respect to what was sampled. More details have been included in the manuscript to clarify this matter. Please see line 127.
Reviewer: Ln 150. Do all of these references actually measure primary production? I don't think Kahn 2021 does
Response: Some of the cited references lack primary production measurements, yet they depict snow algae as the main primary producers within this ecosystem. However, to avoid misunderstandings we have removed the reference from this sentence as suggested by the reviewer. Lines 162-163.
Reviewer: Ln 153. Would be nice to include a macroinvertebrate paper here too
Response: We thank the reviewer for this suggestion. We agree that incorporating a reference to macroinvertebrates in snow algae blooms would enhance our understanding and improve the insight into the impact of these communities on the ecosystem. This addition provides greater context in this article for the potential reader, and the reference has been included. Please see line 165.
Reviewer: Lns 153-155 Rephrase: Snow algae absorb light energy primarily in the ranges where their specific pigments absorb light most effectively.
Response: The manuscript has been modified accordingly. Please see line 166.
Reviewer: Ln 155. The ASD FS4 measures 3500 to 2500nm
Response: Although the ASD FS4 measures 350 to 2500nm, we have limited our study to the range of 350-1150 because this is where the algae exerts its most significant influence, making it more compelling to showcase. More details have been included in the methodology to clarify this point (lines 83-86).
Reviewer: Ln 175. Again, you're measuring HDRF, not albedo, and albedo should be measured across the solar spectrum instead of the VNIR range
Response: We appreciate the reviewer’s comment. We have changed 'albedo' to either ‘HDRF’ or 'reflectance' in those cases where we refer to our data (e.g., lines 80, 94, 104, 106, 117, 161, 168, 180, 181, 197, 200)
Reviewer: Ln 179. Report stats here.
Response: We thank the reviewer for this suggestion. Stats have been reported. Please see line 198.
Reviewer: Lns 181-183. the effect is on albedo/reflectance, not on the algal bloom itself. Reword
Response: The sentence has been reworded. Please see line 199.
Reviewer: Lns 184-185. reword to make a bit clearer
Response: The sentence has been reworded to make it clearer. Please see lines 201.
Reviewer: Lns 186-188. Or that the snow pack physical structure has been modified in your experiment
Response: Although physical alterations of the snow after manipulation may affect its reflectance, we assume that the magnitude of the algae's impact at the wavelengths where the pigments predominantly absorb is much greater. We also like to emphasize that the results presented here are relative and are not intended to establish standard reflectance values that can be extrapolated as a function of biomass and snow cover, but rather to describe for the first time the energy absorption capacity of these biological communities even when they are covered by snow.
Reviewer: Lns 191-193. This data would be good to show. Did you calculate radiative forcing under the snow?
Response: We do not have this data. For this reason, we have modified the sentence, to make it clear that this has not been measured directly in our study. Please see line 212.
Reviewer: Ln 199. clarify what this means
Response: The manuscript has been modified accordingly to clarify these ideas. Please see lines 220-224.
Reviewer: Ln 209. Satellites don't necessarily require this. In fact your results show that chlorophyll absorbance is still evident under surface snow. Which would therefore be detected by a remote sensing approach. What it would do is underestimate the cell density as the chlorophyll signal would be weaker.
Response: We appreciate the reviewer's suggestion. The manuscript has been modified accordingly, including the idea of the ‘biomass underestimation’ suggested. Please see lines 235-238.
Reviewer: Lns 212-214. I don't think you can make this conclusion about metabolic activity without measuring it. And if you have a field reference or irradiance, you could actually measure the radiative forcing with your data, which would be really useful.
Response: We agree that we cannot ensure that metabolic activity exists when algae occur below the surface. This is something we need to study, and we are already planning an experiment for the next melting season. Consequently, we have removed any mention of metabolic activity from the manuscript.
Reviewer: Figure 2
I may be misinterpreting your methods, but I don't think these show what the graph says they do. Firstly, they're not albedo, but I've discussed that. Secondly, if the subsequent measurements are taken relative to a clean snow surface (as a reference), then it is a measure or HDRF relative to clean snow. Not an intrinsic property of the bloom itself.
Response: We thank the reviewer for this suggestion. The figure has been edited accordingly.
Reviewer: Another point, were the clean snow references made with the PVC pipe too? If not then a lot of the lower HDRF could be accounted for from the pipe. The light you measure reflecting from snow typically enters the snow pack a few cm away from the point where you take the measurement, especially if you are working with a 2cm FOV.
Response: We thank the reviewer's concern for the methodological approach used in this work. The clean snow reference was not made through the PVC pipe. We agree that this could impact the HDRF values obtained, but it would mainly occur at wavelengths below 350 nm (e.g., Hasan et al., 2015: 10.1039/C4RA16043F) so it should not invalidate our results.
Citation: https://doi.org/10.5194/egusphere-2023-2379-AC2
-
AC1: 'Reply on RC1', Pablo Almela, 21 Dec 2023
-
RC2: 'Comment on egusphere-2023-2379', Alexander Thomson, 02 Jan 2024
A neat, compact study on a pressing subject with clear ramifications to the quantification of BAR from snow algae.
There are a couple of issues with the final plot (3B) and regression analysis which require addressing. I am also uncertain about the use of "% albedo increase" as the central unit in the study. I think this would be worthwhile re-assessing, perhaps using relative albedo values calculated against a standard (panel or clean snow standard). But this is not my field of expertise, and so I am happy to be corrected on this.
The results are impactful with wide ranging implications for snow algae research, including remote sensing approaches, quantification of BAR, and ecology. That there are implications from the work is pointed out in in the text. It would great to see these implications, as well as some of the direct implications of the results (e.g. variation in response of different absorption wavelengths to snow depth), explored further in the discussion. But I understand that this is a brief communication piece and may be limited for words and space.
Full summary comments and in document comments attached below.
-
AC3: 'Reply on RC2', Pablo Almela, 20 Jan 2024
RC2: 'Comment on egusphere-2023-2379', Alexander Thomson, 02 Jan 2024
GENERAL
Reviewer: A nice study with a clear potential for impact.
General comments:
A neat, compact study on a pressing subject with clear ramifications to the quantification of BAR from snow algae.
My main query revolves around the use of “% albedo increase” as the reporting unit. I am not totally familiar with spectral and albedo research, however I think this has the potential to confuse the discussion, and to misrepresent some relationships (e.g. albedo vs snow depth). Perhaps it would be worthwhile to replot the data against a relative albedo value instead (either relative to a control panel measurement, or a snow control measurement)?
I am also unsure about the plots and linear regression approaches in figure 3B. The figure
legends and labelling of points was fairly unclear. The values for 0 cm snow depth appear to be missing from plot 3 B, and the corresponding regression analysis. I am also not sure linear regression values are useful on values from sequential experimental results like those in the cell density plot, or values that have not been normalised to cell density.
The results are impactful with wide ranging implications for snow algae research, including
remote sensing approaches, quantification of BAR, and ecology. That there are implications from the work is pointed out in the text, but it would great to see these implications explored further in the results and discussion.
Response: We thank the reviewer for dedicating time to review our manuscript. We sincerely appreciate all your valuable comments and suggestions, which helped us in improving the quality of the manuscript. We have carefully revised the manuscript to correct and clarify the concept of reflectance (HDRF), since the initial version included a misconception regarding the measured parameter (not albedo). Consequently, we have reworked Fig3, and the data now illustrates 'HDRF' and 'increase in HDRF'. We have reanalyzed our data and conducted comparisons by calculating the area under the curve (AUC), which we acknowledge as a more accurate approximation, as suggested by rev 3. We opted for a linear regression test due to the independence of observations and the expected linear relationship between the tested variables. The correlations are more robust as we present reflectance vs snow cover as a dependent variable and the increase in reflectance vs cell density. We have also addressed minor comments, which together have helped to improve the manuscript. Detailed responses can be found below.
Comments in text (line number indicated):
32 - More no? See Hotaling review table in below ref.
Response: Albedo reduction has been updated. Please see line 34.
33 - More so than ice algae?
Response: We thank the reviewer for the suggestion. Ice algae make a greater contribution to BAR reduction (i.e., Hotaling et al. (2021). The ms has been modified accordingly. Please see line 35.
38 - There is much greater diversity than this. Perhaps rephrase along the lines of "dominant taxa observed in snow algae bloom formation". As a US counter example see Van Hees et al. 2023 on Chlainomonas spp. Blooms
Response: We appreciate the reviewer's suggestion, and the manuscript has been edited accordingly. Please see line 35. However, Chlainomonas spp. has not been included as a dominant taxon in the manuscript, given that it does not typically represent a majority group, except in the specific region where the cited study was conducted. We will show these results in a paper soon.
38 - Orange blooms are common in the Antarctic (Remias et al 2013)
Response: We have modified the manuscript, stating that blooms can acquire a reddish color (which would include orange) and we included the reference. Please see line 40.
43 - Maybe find a primary reference showing cell abundance vs IRF/melt?
Response: We thank the reviewer for this suggestion. A new reference has been included here. Please see line 45.
46 - rephrase sentence
Response: We thank the reviewer for this suggestion. The manuscript has been revised and modified. Please see lines 47-48.
92 - Personally Ive observed snow algae in dense layers much deeper than 2 cm (5-15 cm). Whether these blooms were actively growing is beside the point to this study which is assessing impact on reflectance, not biology. I would have perhaps tested to deeper snow depths.
Response: We appreciate the reviewer's comment about the presence of snow algae below 2 cm from the surface. In our samplings during the summer, we've observed algae below 2 cm as well, but they mostly show up on the surface. We've taken into account the Onuma et al., 2021 model, which suggests that a snow cover over 20mm marks the 'disappearance' of a bloom (sets the initial cell concentration to 0). This reference wasn't included in the previous version but has been included in the current one for better clarity. Please see line103. We found it fitting to set the upper limit of our study at 2 cm, particularly based on our observations in real-time in the field. Future studies could expand on this and explore blooms with obvious abundant pigmented biomass deeper in the snowpack.
95 - My understanding of albedo and spectral data is limited, however would it not be better to measure the reduction in albedo from a set standard (a control panel? or the control snow patch?) ? I am unsure about the wording of albedo "increasing" in this way, even if the increase is a proportional increase (%). Ive commented below, more specifically on the results, too.
Response: We thank the reviewer for the suggestion. In the new version of the manuscript we have specified that a spectralon was used for the white reference (please see lines 87-88). We have also considered it appropriate to remove the phrase you mention about proportional increase, since we now present different data in Fig 3.
98 - was the PVC tube used for the reference as well?
Response: The clean snow reference was not made through the PVC pipe. This could impact the HDRF values obtained, but it would mainly occur at wavelengths below 350 nm (e.g., Hasan et al., 2015: 10.1039/C4RA16043F) so it should not invalidate our results. In this version we have also specified that the PVC cylinder was not in the field of view when measuring reflectance. Please see line 96-97.
100 - I think assuming generally consistent snow structure across your study snow patch and samples is reasonable. However it would be useful to report some of those characteristics (grain size, density) for future comparison with other sites/snow packs. My understanding of spectral/snow physics are poor, but I assume that the snow structure will make a significant difference to how much snow cover affects BAR?
Response: We agree with the reviewer that the physical properties of snow play a key role in the BAR effects. However, we do not have this information since at the time of sampling our SLF Snow Sensor was not working. Therefore we regret that we cannot include that information in the manuscript.
108 - Any evidence of subsurface cells? What was the depth of the observed bloom?
Response: This algae bloom was concentrated in the surface, with no evident presence of cells below the initial centimeter. As a precautionary measure, we sampled a core of snow extending 2 centimeters deep (refer to lines 117-119 for details).
118 - was the composition of cells/species consistent across the 6 measurement plots?
Response: While it would be interesting to know the composition of the algal communities in the six plots, we didn't extract DNA from these samples. Our main objective did not involve studying the community and, lacking experience as taxonomists in snow algae, we cannot confirm if there were differences in the compositions.
137 - those?
Response: We thank the reviewer for the suggestion. The manuscript has been edited accordingly.
138 - not sure the last section of this sentence adds to the discussion. Too many variables to make such statements.
Response: This sentence has been removed from the manuscript.
171 - was there a difference in the effect of snow depth on reflectance in different absorption ranges? Does green and red absorption get reduced at the same rate with snow depth increase? Would be interesting to discuss
Response: We thank the reviewer for the suggestion and agree that it would be interesting to include this comparison in the discussion. The manuscript has been edited accordingly. Please see lines 185-193.
176 - see comment above, and on Figure 3B
179 - so absorption in these wavelengths is potentially affected more by snow cover?
Worth discussing?
Response: The data in question has been omitted from the manuscript and is consequently not available for discussion. However, in response to your earlier suggestion, we have included a very interesting comparison of absorption within the chlorophyll and carotenoid ranges (see line 185-193).
180 - See comments on figure 3B - the cell density plot is not all that clear, but is it suggesting that snow cover has a greater effect on higher biomass blooms? Again, i think using "increase in albedo" maybe confuses this point. Plotting relative albedo vs cell density instead might show the relationship more clearly? And more intuitively show that high cell count results in lower albedo, and that the increase in reflectance with snow cover is steeper for high biomass blooms (which i think your figure 2 plot 6 shows as well)?
Response: We thank the reviewer for the suggestion. We have modified the data shown in Figure 3. We now present ‘HDRF’ (instead of ‘albedo’) and 'increase in HDRF' (instead of 'increase in albedo (%)'). Now it is clearly shown how a greater cell density is associated with a greater increase in reflectance when snow is added on top of the snow algae.
183 - See comments on graph, but could you not normalise for cell density before performing linear regression between reflectance and snow depth? This is conflating two factors (cell density and snow depth) on reflectance. As discussed in comments above, that cell density interacts with the effect on albedo is interesting in itself, and could be explored, but not sure this linear regression is the best way to do so.
Response: We thank the reviewer for the suggestion. As mentioned above, we have modified the data shown in Figure 3. We now have HDRF (instead of albedo) and 'increase in HDRF' (instead of 'increase in albedo (%)'). The latter refers to the changes in reflectance with respect to the initial measurements of the snow algae on the surface, but it is no longer a percentage but the difference between both measurements. This allows us to relativize the measurements in each plot and thus compare their effect between them. Although we considered normalizing the data (log cell density vs reflectance), it did not improve clarity visually or in regression analyses. Consequently, we have opted not to modify this section. We hope all these changes will enhance the clarity of the data presented in the manuscript.
191 - indicate is a bit strong. perhaps "suggest". Unless you can include melt data or calculations?
Response: The manuscript has been edited accordingly. Please see line 212.
198 - Agree in general with this, and a crucial data point in calculating IRF across regions from remote sensing methods. But also agree with previous reviewer - your data suggests spectral approaches might well detect some chlorophyll absorption through shallow snow cover! Maybe worth expanding this comment/discussion?
Response: We agree with the reviewer. The manuscript has been edited accordingly to rev1 and rev2. Please see lines 220-222.
203 - Was this shown in this analysis? Any evidence for this?
Response: We thank the reviewer for the comment. Our data did not provide sufficient evidence to conclusively support this claim. Therefore, the statement has been removed from the manuscript.
206 - Just suggestions, but could the ramifications of this study be explored further?
- could this understanding help constrain the boundaries of detection of covered snow algae? or allow more refined detection of subsurface blooms?
- what is the relative impact of snow cover on BAR at a landscape scale? Do covered snow algae blooms contribute a missed proportion of BAR using current methods? If so is that missed proportion significant? A rough back of the envelope simulation/model of a bloom/snowfield might be interesting to test this?
- any ecological significance? green vs red snow blooms? PS efficiency at depth vs surface?
Response: We appreciate the reviewer's suggestion. In the conclusions, we have incorporated a comment regarding the potential consequences of underestimating snowmelt rates (please see lines 241-244). But because this is a brief communication, we are limited in elaborating extensively on these details. We really hope that the manuscript will lead to further studies considering subsurface blooms and their potential impact.
213 - Maybe needs rephrasing - the findings suggest subsurface snow habitats might still be suitable for PS activity in snow algae, facilitated by absorption and localised melting effects
Response: We thank the reviewer for this suggestion. We have decided to remove this sentence from the conclusions following the comments of reviewer 1. We do not have data that proves PS below the surface, but we are already working on it to test it as soon as we have the opportunity.
216 - could these be explored?
Response: We thank the reviewer for this suggestion. The manuscript has been modified accordingly, as mentioned above. Please see lines 241-244.
Fig 1D - Any scale?
Response: We have incorporated the scale in the revised version. Please see line 255.
Fig 3B - should this not be normalised for cell counts? weak correlation is due to the variation in cell count, not the weak relationship between snow depth and reflectance
Fig 3B - Is this showing that reflectance increases more steeply when you have a higher biomass? The cell count plots and figure legend are very unclear - are the points spectral measurement replicates? Again, perhaps worth trying to plot these using relative albedo (vs a control value), rather than increase in absorption? Also, similarly to plots next door, is an r2 value appropriate if you are plotting points from the same plot at different snow depths?
Fig 3B - Should the values for 0 cm snow cover not be included? And the regression not extend to this? This seems to me to be of central importance to the study, as reflectance seems to increase more steeply between 0 cm and 0.5 cm depth? Though again, may be an artefact of using "% increase in albedo"
Response: We thank the reviewer for all suggestions regarding Figure 3. Figure 3B has been modified, as discussed above. This new version resolves the weak correlations shown previously, clarifies the data shown (the figure has been redesigned) and also presents the data for 0 cm snow cover as suggested.
Citation: https://doi.org/10.5194/egusphere-2023-2379-AC3
-
AC3: 'Reply on RC2', Pablo Almela, 20 Jan 2024
-
RC3: 'Comment on egusphere-2023-2379', S. McKenzie Skiles, 09 Jan 2024
Snow is bright natural surface in the visible wavelengths where ice is transparent and snow reflectance is dominated by frequent scattering opportunities at ice/air interfaces. It is well understood that, in the visible wavelengths, light penetrates tens of centimeters into the snowpack and thereby, light absorbing particles present in the near surface will lower the snow albedo. The authors assess this impact for very near surface snow algae using field experiments whereby they used an ASD field spectrometer to measure the spectral reflectance of snow with algae at the surface, and then incrementally covered the surface with visibly cleaner subsurface snow, measuring the spectral reflectance with each increment. They additionally collected snow sampled and analyzed characteristics of snow algae (cell counts and chlorophyll) and related those results to reflectance measurements.
This is a nicely focused field study, the scope of which is suitable for a brief communication. It will be of interest to the snow algae/LAP communities following revisions to clarify methods and results before publication. My feedback is organized first by major feedback and then line-by-line revisions.
Major:
The authors should be consistent in referring to their ASD measurements by the right terminology. The measurements are the hemispherical conical reflectance factor (HCRF), not albedo. Even though the authors do recognize that light is not reflected by snow evenly in all directions and that they made a directional measurement, they go on to indicate that reflectance and albedo are equivalent (not the case). Albedo, the ratio of incoming to outgoing solar radiation over the hemisphere, cannot be measured with a field spectrometers bare optical fiber and requires a remote cosine receptor (RCR) foreoptic. The authors cite Cook et al., 2017 in the paper, if they refer back to this paper they may find figures 5 and 6 informative, as well as the discussion of best practices in field spectroscopy. Another useful paper is Schaepman-Strub, 2006 which discusses in detail the difference between measurable remote sensing quantities. The measurements in this study should only be referred to as reflectance, and ideally HCRF (commonly simplified to HDRF), including updating axis labels on Figure 2 and Figure 3.
Cook et al., 2017 - https://tc.copernicus.org/articles/11/2611/2017/tc-11-2611-2017.pdf
Related, and could be very useful, but behind paywall: Cook et al., 2019 Bio-optical properties of Terrestrial Snow and Ice https://link.springer.com/chapter/10.1007/978-3-030-20587-4_3
Schaepman-Strub et al., 2006 - https://doi.org/10.1016/j.rse.2006.03.002
Relatedly, the methods section and Figure 1 need to be updated to make the ASD measurement methods more clear. For example – the use of a white reference panel is not mentioned despite it being a necessary part of reflectance measurements. Also, discussion is needed as to why they chose to measure at an angle in the backscatter direction, and to calculate the instruments field of view. Additionally, please clarify that the PVC pipe was not in the field of view or used in the reflectance measurement, as may be inferred from Figure 1 (which should also be updated). Finally, please describe how the broadband values, used compare to algae properties, were calculated (i.e. integration with respect to spectral irradiance).
I think it would also be useful to better motivate the experiment and contextualize the results to include, even briefly as would be suitable for a brief communication, more background on the current state of the science. Specifically in terms spectral controls on snow albedo (LAPs in visible, ice absorption/grain size in NIR), how those controls determine the depth to which light penetrates (deeper in the visible then NIR wavelengths), and why that is relevant for algae and BAR. For example, multiple studies have previously researched 1) light penetration at different depths for algae photosynthesis/photo-inhibition, and 2) the impact of below surface LAPs on snow albedo. It is not clear in the study, as currently written, that the current state of the science is such that we understand how far light penetrates into snow and that the impact that algae (and other LAPs) has on albedo, when present beneath cleaner snow layers, can be modeled. For example, both the SNICAR-ADv3 and the PBSAM model can simulate absorption by algae in subsurface layers. I’ll use one of my studies as an example (but only because it is best known to me, the authors are not under any pressure to include these specific citations in their study): using a gravimetrics measuring board Skiles and Painter, 2017 collected 3 cm incremental samples across the top 30 cm of the snowpack and related dust and BC concentration to albedo over a daily spring time series, finding that dust in the top 6-8 centimeters exhibits the strongest relationship with albedo decline. Skiles and Painter, 2018 then used those measurements to run the multilayer version of the SNICAR model, representing dust and BC in multiple layers, including beneath cleaner snow. Onuma and Takeuchi have done similar studies in Greenland for algae, dust, and BC. A strong motivation for this work would be that, although we can model it, there are few in situ reflectance studies (such as this one) in midlatitude environments of algae laden snow with cleaner snow on top.
References that could be helpful for more background (not exhaustive)-
Snow optical properties, light penetration into snow: Warren, 1982, Libois et al, 2013, Warren et al., 2006, Zhou ett al. 2003, Warren et al. 2013
Snow/algae: Curl et al., 1972, Gorton et al., 2001, Stibal et al., 2007, Cook et al., 2017 (already cited, but in a limited fashion)
SNICAR/PBSAM: Flanner et al., 2021, Onuma et al., 2020
Impact of mineral dust, at different depths below the surface, on albedo:
Skiles and Painter, 2017, Skiles and Painter, 2018
BC at different depths: Doherty et al., 2013
Line-by-line
15-16: The optical properties fresh snow and older aged snow, like that used in this study to cover algae, are quite different. Fresh snow is much brighter and smaller grained, I would remove the parenthetical aside (such as after fresh snowfall) because that would not be an outcome of the reflectance measurements from this specific experiment.
18: This portion of the abstract would be better used to describe the quantitative results from the experiment– what were the relationships between reflectance, depth below surface, and cell densities/chlorophyll.
Line 71: Can you clarify what you mean by snow algae ‘could be clearly seen in the distance’ – is this to indicate it was visibly widespread surrounding the site of the experiment?
Line 77: Include spectral range and resolution of instrument.
Line 82: Reiterate point made above, because it is not the typical approach, please provide some justification as to why measurements were made at an angle (and in the backscatter direction?). For that angle and distance above target, please provide the approximate area being measured for context and comparison to sample size being collected.
Line 84: Reiterating point made above that this is not an albedo measurement – and to please clarify the role of the PVC cylinder.
Line 87: The reflectance measurements would indicate that the snow is not free of LAPs (the visible reflectance would be higher), and almost certainly they are there, given the time of the experiment in late summer and that widespread algae growth corresponds with deposition that provides nutrients. It could be useful to clarify that relative to algae laden snow at the surface, the subsurface snow was visually brighter and therefore used as a proxy for clean snow. It would also be useful to include in results/discussion that analysis was not undertaken to quantify other LAP concentrations, and therefore would be unknown and an uncertainty in the presented results.
Line 91/92: : There would not be such a distinct/hard limit – it would depend on many factors including snow properties, surface roughness, and liquid water content. I also don’t not see this statement in Cook et al., 2017 – 2 cm is mentioned as a practical limit on sample collection in ice - but algae in Greenland ice would also be different then algae on a snowfield so another citation/justification for this depth would be more suitable.
Line 106 - Is there a reason to use pigment instead of chlorophyll in the section header?
Line 121 – It is more typical to report snow density in kg m-3, consider including in place of or in addition to the current units. In August the snow on a snowfield would be significantly more dense than 200 kg m-3, probably more in the range of 500-600 kg m-3, unless it had very recently snowed (from this pictures this does not appear to be the case). Please justify the use of this value.
Lines 131-134 – Belongs in methods
Line 137 – There are more studies now that have studied algae content, it would be useful to have more studies included for reference (especially ones that are not at very high latitudes such as Greenland/the Arctic). Comparing midlatitude sites to arctic sites for algae is not apples to apples, it would expected that cell counts would be lower in the arctic relative to midlatitude study locations given constraints on algae growth (solar, water, nutrients).
Line 139 – It would be useful to have pictures of the six plots, which could be suitably included in any of the figures. If you chose them based on the visual intensity of their color variation in concentration would be a descriptor of the expected outcome of the experimental design, it could be useful to phrase it in this way.
Line 155 – An ASD Fieldspec 4 measures 350 – 2500 nm, please clarify why the spectral range indicated here is not the full range and provide a justification (in methods) about why the full range was not used.
Line 160 – Please clarify (in methods) how these comparisons are made - by integrating over the carotenoid and chlorophyll absorption features? Or by using average reflectance values (which would not be appropriate given reflectance magnitude is sensitive to measurement and solar zenith angles).
Line 167 – In the near infrared absorption is dominated by ice, and reflectance signatures would be indicative of effective grain radius/ice path length (for reference see Fig. 1 in Cook et al., 2017). Please clarify.
Line 178 (and broader results reporting) – Working from spectral reflectance to a single broadband albedo value is not trivial, and from the described methods it was not done in this paper. Even going from spectral reflectance to broadband reflectance requires integrating with respect to irradiance since incoming solar radiation is not evenly distributed across the solar spectrum. More details are needed on how the results presented in Figure 3 were calculated. Please report differences in the actual values as opposed to a percent difference, given the broadband values themselves are not reported (i.e 0.4 increasing to 0.7, delta0.3).
Line 199 -200 – Please clarify the point being made here - you should be able to detect algae beneath the surface from visible/spectral measurements given that it is absorbing and lowers the albedo – this is the outcome of the study (and is the basis of LAP radiative forcing retrievals from remote sensing, e.g. Painter et al., 2012). Visible to the eye and visible to remote sensing are two different things, and remote sensing has many different spectral and spatial scales making it hard to blanket statements about remote sensing can and cannot do. Also, if it is present at a depth where it doesn’t lower the albedo, then it would be undetectable, but it also wouldn’t be relevant for BAR?
Painter et al, 2012: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2012GL052457
202 – As noted in the study algae primarily reduces albedo/reflectance in the visible wavelengths, and NIR is indicative of ice/grain size absorption - so this statement needs clarification or more justification.
(In the reflectance plots, the NIR reflectance is highest for the cleaner snow - it was subsurface and cleaner, so it makes sense that the grain size would be smaller due to lower energy inputs to drive snow metamorphism - therefore would have higher NIR reflectance. The NIR is lower for the snow algae at surface because it's absorbing, and the grains would be much larger. In the plots the NIR reflectance is essentially the same for the 'snow added experiments' that indicates the grain size of the added snow is consistent. Light penetrates ~1-2 cm into snow in the NIR wavelengths, so it would make sense that your ‘snow added’ spectra in the NIR would be (spectrally) a combination between the cleaner snow spectra and the algae laden spectra.)
Line 206- I would suggest taking this out or adding more context and specifics on how these results would be integrated into watershed melt models, which is quite complicated.
Line 213 – Since this study did not calculate melt rates or do a time series study to measure them, it would be more suitable to stick to relevant conclusions drawn from the field experiment or provide more context for this statement.
Line 214 – To reiterate a point made above, if it lowers the albedo then it is detectable by remote sensing – but I might be missing the point being made – please clarify.
Citation: https://doi.org/10.5194/egusphere-2023-2379-RC3 -
AC4: 'Reply on RC3', Pablo Almela, 20 Jan 2024
RC3: 'Comment on egusphere-2023-2379', S. McKenzie Skiles, 09 Jan 2024
GENERAL
Reviewer: Snow is bright natural surface in the visible wavelengths where ice is transparent and snow reflectance is dominated by frequent scattering opportunities at ice/air interfaces. It is well understood that, in the visible wavelengths, light penetrates tens of centimeters into the snowpack and thereby, light absorbing particles present in the near surface will lower the snow albedo. The authors assess this impact for very near surface snow algae using field experiments whereby they used an ASD field spectrometer to measure the spectral reflectance of snow with algae at the surface, and then incrementally covered the surface with visibly cleaner subsurface snow, measuring the spectral reflectance with each increment. They additionally collected snow sampled and analyzed characteristics of snow algae (cell counts and chlorophyll) and related those results to reflectance measurements.
This is a nicely focused field study, the scope of which is suitable for a brief communication. It will be of interest to the snow algae/LAP communities following revisions to clarify methods and results before publication. My feedback is organized first by major feedback and then line-by-line revisions.
Response: We appreciate the reviewer for dedicating time to review our manuscript and for the positive comments on the significance of the presented results.The manuscript has been thoroughly revised to clarify what is referred to as major. Additionally, we have addressed minor comments, contributing to overall enhancements in the manuscript. Please find detailed responses below.
Major:
The authors should be consistent in referring to their ASD measurements by the right terminology. The measurements are the hemispherical conical reflectance factor (HCRF), not albedo. Even though the authors do recognize that light is not reflected by snow evenly in all directions and that they made a directional measurement, they go on to indicate that reflectance and albedo are equivalent (not the case). Albedo, the ratio of incoming to outgoing solar radiation over the hemisphere, cannot be measured with a field spectrometers bare optical fiber and requires a remote cosine receptor (RCR) foreoptic. The authors cite Cook et al., 2017 in the paper, if they refer back to this paper they may find figures 5 and 6 informative, as well as the discussion of best practices in field spectroscopy. Another useful paper is Schaepman-Strub, 2006 which discusses in detail the difference between measurable remote sensing quantities. The measurements in this study should only be referred to as reflectance, and ideally HCRF (commonly simplified to HDRF), including updating axis labels on Figure 2 and Figure 3.
Cook et al., 2017 - https://tc.copernicus.org/articles/11/2611/2017/tc-11-2611-2017.pdf
Related, and could be very useful, but behind paywall: Cook et al., 2019 Bio-optical properties of Terrestrial Snow and Ice https://link.springer.com/chapter/10.1007/978-3-030-20587-4_3
Schaepman-Strub et al., 2006 - https://doi.org/10.1016/j.rse.2006.03.002
Relatedly, the methods section and Figure 1 need to be updated to make the ASD measurement methods more clear. For example – the use of a white reference panel is not mentioned despite it being a necessary part of reflectance measurements. Also, discussion is needed as to why they chose to measure at an angle in the backscatter direction, and to calculate the instruments field of view. Additionally, please clarify that the PVC pipe was not in the field of view or used in the reflectance measurement, as may be inferred from Figure 1 (which should also be updated). Finally, please describe how the broadband values, used compare to algae properties, were calculated (i.e. integration with respect to spectral irradiance).
I think it would also be useful to better motivate the experiment and contextualize the results to include, even briefly as would be suitable for a brief communication, more background on the current state of the science. Specifically in terms spectral controls on snow albedo (LAPs in visible, ice absorption/grain size in NIR), how those controls determine the depth to which light penetrates (deeper in the visible then NIR wavelengths), and why that is relevant for algae and BAR. For example, multiple studies have previously researched 1) light penetration at different depths for algae photosynthesis/photo-inhibition, and 2) the impact of below surface LAPs on snow albedo. It is not clear in the study, as currently written, that the current state of the science is such that we understand how far light penetrates into snow and that the impact that algae (and other LAPs) has on albedo, when present beneath cleaner snow layers, can be modeled. For example, both the SNICAR-ADv3 and the PBSAM model can simulate absorption by algae in subsurface layers. I’ll use one of my studies as an example (but only because it is best known to me, the authors are not under any pressure to include these specific citations in their study): using a gravimetrics measuring board Skiles and Painter, 2017 collected 3 cm incremental samples across the top 30 cm of the snowpack and related dust and BC concentration to albedo over a daily spring time series, finding that dust in the top 6-8 centimeters exhibits the strongest relationship with albedo decline. Skiles and Painter, 2018 then used those measurements to run the multilayer version of the SNICAR model, representing dust and BC in multiple layers, including beneath cleaner snow. Onuma and Takeuchi have done similar studies in Greenland for algae, dust, and BC. A strong motivation for this work would be that, although we can model it, there are few in situ reflectance studies (such as this one) in midlatitude environments of algae laden snow with cleaner snow on top.
References that could be helpful for more background (not exhaustive)-
Snow optical properties, light penetration into snow: Warren, 1982, Libois et al, 2013, Warren et al., 2006, Zhou ett al. 2003, Warren et al. 2013
Snow/algae: Curl et al., 1972, Gorton et al., 2001, Stibal et al., 2007, Cook et al., 2017 (already cited, but in a limited fashion)
SNICAR/PBSAM: Flanner et al., 2021, Onuma et al., 2020
Impact of mineral dust, at different depths below the surface, on albedo:
Skiles and Painter, 2017, Skiles and Painter, 2018
BC at different depths: Doherty et al., 2013
Response: We thank the reviewer for the valuable suggestions. We apologize for the confusion in the terminology used in the previous version. The measurements in this study are now only referred to reflectance (HDRF). We have also edited the current manuscript by contextualizing the impacts of other light absorbing impurities (LAIs), and specifying several studies that assessed the vertical effect on the albedo reduction of these contaminants. However, we are limited in the number of references we can include and by the word count in the short communication format. The methodology has been clarified. We hope that the edits and added information allows the potential reader to contextualize our study a little better.
Line-by-line
15-16: The optical properties fresh snow and older aged snow, like that used in this study to cover algae, are quite different. Fresh snow is much brighter and smaller grained, I would remove the parenthetical aside (such as after fresh snowfall) because that would not be an outcome of the reflectance measurements from this specific experiment.
Response: We thank the reviewer for its comment on the different properties of fresh and old snow. In this sentence we indicate that snow algae can be buried after blooming by a snowfall, to give the reader context about our study, but not as something that is directly addressed. To provide further context, we've included an additional example—avalanches—illustrating how algae might occur beneath the surface without involving fresh snow. Please see line 16.
18: This portion of the abstract would be better used to describe the quantitative results from the experiment– what were the relationships between reflectance, depth below surface, and cell densities/chlorophyll.
Response: We appreciate the reviewer's suggestion and we agree that incorporating qualitative results into the abstract would be valuable. Unfortunately, the abstract is restricted to 100 words and thus we chose the current format to succinctly summarize our results.
Line 71: Can you clarify what you mean by snow algae ‘could be clearly seen in the distance’ – is this to indicate it was visibly widespread surrounding the site of the experiment?
Response: We find it important to mention that the bloom exhibited a high cell density, as its presence was evident from a distance. We have rephrased this to make it clearer. Please see line 75
Line 77: Include spectral range and resolution of instrument.
Response: The spectral range and resolution of the instrument have been incorporated in the revised version. Please see lines 80-86.
Line 82: Reiterate point made above, because it is not the typical approach, please provide some justification as to why measurements were made at an angle (and in the backscatter direction?). For that angle and distance above target, please provide the approximate area being measured for context and comparison to sample size being collected.
Response: An angle of 60 degrees was used to be sure that we were measuring reflectance at an angle other than the direction of sunlight at the time of the experiment. The approximate measured area has been included in the manuscript. Please see line 92.
Line 84: Reiterating point made above that this is not an albedo measurement – and to please clarify the role of the PVC cylinder.
Response: We thank the reviewer for the suggestion. In the manuscript, the term 'albedo' has been substituted with either 'reflectance' or 'HDRF' consistently. The role of the PVC was to control the thickness of the snow layers added. This part of the manuscript has been edited accordingly. Please see lines 95-96.
Line 87: The reflectance measurements would indicate that the snow is not free of LAPs (the visible reflectance would be higher), and almost certainly they are there, given the time of the experiment in late summer and that widespread algae growth corresponds with deposition that provides nutrients. It could be useful to clarify that relative to algae laden snow at the surface, the subsurface snow was visually brighter and therefore used as a proxy for clean snow. It would also be useful to include in results/discussion that analysis was not undertaken to quantify other LAP concentrations, and therefore would be unknown and an uncertainty in the presented results.
Response: In the current version of the manuscript, we have specified that the subsurface snow was visually brighter (please see line 98), and also clarify that LAP concentrations other than algae were not quantified (please see lines 205-209).
Line 91/92: : There would not be such a distinct/hard limit – it would depend on many factors including snow properties, surface roughness, and liquid water content. I also don’t not see this statement in Cook et al., 2017 – 2 cm is mentioned as a practical limit on sample collection in ice - but algae in Greenland ice would also be different then algae on a snowfield so another citation/justification for this depth would be more suitable.
Response: We apologize to the reviewer for the misunderstanding with Cook et al., 2017. In our experience sampling across the mountain west region of the US, we've also observed algae below 2 cm. However, based on our observations in real-time in the field at this snowfield, 2 cm seemed like a reasonable depth which is also consistent with the Onuma et al., 2021 model. This model suggests that a snow cover over 20mm marks the 'disappearance' of a bloom (sets the initial cell concentration to 0). This reference wasn't included in the previous version and has been included in the current one for better clarity. Please see lines 101-103.
Line 106 - Is there a reason to use pigment instead of chlorophyll in the section header?
Response: There is no reason, and specifying which pigment has been analyzed would be more convenient. Therefore, the manuscript has been modified accordingly. Please see line 116.
Line 121 – It is more typical to report snow density in kg m-3, consider including in place of or in addition to the current units. In August the snow on a snowfield would be significantly more dense than 200 kg m-3, probably more in the range of 500-600 kg m-3, unless it had very recently snowed (from this pictures this does not appear to be the case). Please justify the use of this value.
Response: There was no special rationale behind the density of 200 kg/m3 beyond being a general value used for it. Following the reviewer's recommendation, we have changed this value to 500 kg/m3 and updated the related data. See line 132.
Lines 131-134 – Belongs in methods
Response: We thank the reviewer for this suggestion. The sentence has been relocated to the methodology section (refer to line 121-122), and the results and discussion paragraph has been revised accordingly (see line 147-148).
Line 137 – There are more studies now that have studied algae content, it would be useful to have more studies included for reference (especially ones that are not at very high latitudes such as Greenland/the Arctic). Comparing midlatitude sites to arctic sites for algae is not apples to apples, it would expected that cell counts would be lower in the arctic relative to midlatitude study locations given constraints on algae growth (solar, water, nutrients).
Response: We thank the reviewer for this suggestion. Unfortunately we are limited with the number of references we can include in a brief communication. However, the intent behind this statement is to show the global variability in cell densities within snow algae blooms, providing readers with a comprehensive contextual understanding. Given that cell density is related with the quantity of pigments/carotenoids (as shown in this study) and pigments/carotenoids are related with the impact on the BAR reduction, we believe it is important to present this range in cell densities occurrence.
Line 139 – It would be useful to have pictures of the six plots, which could be suitably included in any of the figures. If you chose them based on the visual intensity of their color variation in concentration would be a descriptor of the expected outcome of the experimental design, it could be useful to phrase it in this way.
Response: We agree with the reviewer that including the pictures of the plots could be interesting for a better description of the samples. Unfortunately, we do not have the pictures of the six plots so we cannot show them in the manuscript.
Line 155 – An ASD Fieldspec 4 measures 350 – 2500 nm, please clarify why the spectral range indicated here is not the full range and provide a justification (in methods) about why the full range was not used.
Response: We chose this spectral range for biological relevance, as biological processes and structural features relevant to organisms are prominent in this interval. Also to simplify the data and graphs presented in this study. The manuscript has been modified accordingly. Please see lines 83-86.
Line 160 – Please clarify (in methods) how these comparisons are made - by integrating over the carotenoid and chlorophyll absorption features? Or by using average reflectance values (which would not be appropriate given reflectance magnitude is sensitive to measurement and solar zenith angles).
Response: We thank the reviewer for this comment. We have reanalyzed our data and conducted comparisons by calculating the area under the curve (AUC), which we acknowledge as a more accurate approximation, as suggested. As a result, we have made corresponding edits to the methodology (lines 139-141), results (lines 168-187), and Figure 3.
Line 167 – In the near infrared absorption is dominated by ice, and reflectance signatures would be indicative of effective grain radius/ice path length (for reference see Fig. 1 in Cook et al., 2017). Please clarify.
Response: We thank the reviewer for this suggestion. Since we want to focus on the biological impact, this statement has been removed from the sentence to avoid any misunderstanding.
Line 178 and broader results reporting) – Working from spectral reflectance to a single broadband albedo value is not trivial, and from the described methods it was not done in this paper. Even going from spectral reflectance to broadband reflectance requires integrating with respect to irradiance since incoming solar radiation is not evenly distributed across the solar spectrum. More details are needed on how the results presented in Figure 3 were calculated. Please report differences in the actual values as opposed to a percent difference, given the broadband values themselves are not reported (i.e 0.4 increasing to 0.7, delta0.3).(
Response: We thank the reviewer for this suggestion. In the current version of the manuscript, we have not use the term '% albedo increase' as it was found to be confusing and may not be the most fitting. Instead, our reporting now revolves around 'HDRF' and 'increase in HDRF,' reflecting alterations in reflectance across snow layers. We trust that this edit contributes to a clearer interpretation of the results outlined in the manuscript.
Line 199 -200 – Please clarify the point being made here - you should be able to detect algae beneath the surface from visible/spectral measurements given that it is absorbing and lowers the albedo – this is the outcome of the study (and is the basis of LAP radiative forcing retrievals from remote sensing, e.g. Painter et al., 2012). Visible to the eye and visible to remote sensing are two different things, and remote sensing has many different spectral and spatial scales making it hard to blanket statements about remote sensing can and cannot do. Also, if it is present at a depth where it doesn’t lower the albedo, then it would be undetectable, but it also wouldn’t be relevant for BAR?
Painter et al, 2012: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2012GL052457
Response: We thank the reviewer for the suggestion. This sentence has been edited following reviewer 1's suggestions. However, isn't it possible that the snow cover prevents the algae from being visible to the naked eye? In figure 1 it can be seen how, after adding 2 cm of snow, there is no apparent trace of the bloom. But indeed, this should not prevent them from being detected by remote sensing. We hope that the edits have clarified this point in the manuscript. Please see lines 220-222.
202 – As noted in the study algae primarily reduces albedo/reflectance in the visible wavelengths, and NIR is indicative of ice/grain size absorption - so this statement needs clarification or more justification.
Response: We thank the reviewer for the suggestion. This sentence has been removed following reviewer 1's suggestions.
(In the reflectance plots, the NIR reflectance is highest for the cleaner snow - it was subsurface and cleaner, so it makes sense that the grain size would be smaller due to lower energy inputs to drive snow metamorphism - therefore would have higher NIR reflectance. The NIR is lower for the snow algae at surface because it's absorbing, and the grains would be much larger. In the plots the NIR reflectance is essentially the same for the 'snow added experiments' that indicates the grain size of the added snow is consistent. Light penetrates ~1-2 cm into snow in the NIR wavelengths, so it would make sense that your ‘snow added’ spectra in the NIR would be (spectrally) a combination between the cleaner snow spectra and the algae laden spectra.)
Response: We really thank the reviewer for this clarification about the spectra in the NIR for the snow layers added to the plots. This helps us better understand the physical properties of the environment where the algae is blooming, and therefore understand the results.
Line 206- I would suggest taking this out or adding more context and specifics on how these results would be integrated into watershed melt models, which is quite complicated.
Response: We thank the reviewer for the suggestion. This sentence has been removed in the current version.
Line 213 – Since this study did not calculate melt rates or do a time series study to measure them, it would be more suitable to stick to relevant conclusions drawn from the field experiment or provide more context for this statement.
Response: We have modified this part in alignment with the feedback provided by the other reviewers. Nevertheless, we consider it important to mention the potential increase in melting rates, as it stands as a pivotal potential conclusion from this study (highlighted merely as a suggestion). Given our demonstrated evidence of energy absorption by the algae under the surface, there is a logical expectation that this absorbed energy would be transmitted as heat to the snow surrounding the cells, consistent with what happens on the surface.
Line 214 – To reiterate a point made above, if it lowers the albedo then it is detectable by remote sensing – but I might be missing the point being made – please clarify.
Response: We apologize to the reviewer for the misunderstanding in the first version of the manuscript. Now we have clarified this part.
Citation: https://doi.org/10.5194/egusphere-2023-2379-AC4
-
AC4: 'Reply on RC3', Pablo Almela, 20 Jan 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 512 | 321 | 52 | 885 | 70 | 34 | 56 |
- HTML: 512
- PDF: 321
- XML: 52
- Total: 885
- Supplement: 70
- BibTeX: 34
- EndNote: 56
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
A nice study with a clear potential for impact.
Main concern centers around the concept of albedo used throughout. I wasn't clear that what was being measured was albedo because measurements didn't seem to have been made with respect to incoming solar irracience. The results are still valid, but the termanology needs addressing so as not to confuse the reader and perpetuate the idea that HDRF measurements are the same as Albedo. This could cleared up some in the methods section also. More info needed, for example on controls taken and how the experimental design influenced snow reflectance.
Other minor comments are in attached pdf.