Light-absorption of dust and elemental carbon in snow in the Indian Himalayas and the Finnish Arctic

Svensson, Jonas; Ström, Johan; Kivekäs, Niku; Dkhar, Nathaniel B.; Tayal, Shresth; Sharma, Ved P.; Jutila, Arttu; Backman, John; Virkkula, Aki; Ruppel, Meri; Hyvärinen, Antti; Kontu, Anna; Hannula, Henna-Reetta; Leppäranta, Matti; Hooda, Rakesh K.; Korhola, Atte; Asmi, Eija; Lihavainen, Heikki

doi:https://doi.org/10.5194/amt-11-1403-2018

Articles | Volume 11, issue 3

https://doi.org/10.5194/amt-11-1403-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/amt-11-1403-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 11, issue 3

Research article

|

12 Mar 2018

Research article |

| 12 Mar 2018

Light-absorption of dust and elemental carbon in snow in the Indian Himalayas and the Finnish Arctic

Jonas Svensson, Johan Ström, Niku Kivekäs, Nathaniel B. Dkhar, Shresth Tayal, Ved P. Sharma, Arttu Jutila, John Backman, Aki Virkkula, Meri Ruppel, Antti Hyvärinen, Anna Kontu, Henna-Reetta Hannula, Matti Leppäranta, Rakesh K. Hooda, Atte Korhola, Eija Asmi, and Heikki Lihavainen

Download

Final revised paper (published on 12 Mar 2018)
Preprint (discussion started on 12 Jul 2017)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Large and un-explored uncertainties undermine conclusions', Anonymous Referee #1, 08 Aug 2017
- AC1: 'Author's reply to comments', J. Svensson, 05 Oct 2017
RC2: 'Review', Anonymous Referee #3, 09 Aug 2017
- AC2: 'Author's reply to comments', J. Svensson, 05 Oct 2017

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Jonas Svensson on behalf of the Authors (06 Oct 2017)

ED: Referee Nomination & Report Request started (07 Oct 2017) by Mingjin Tang

RR by Anonymous Referee #3 (13 Oct 2017)

Suggestions for revision or reasons for rejection

Second review of “ Light-absorption of dust and elemental carbon in snow from the Indian Himalaya and the Finnish Arctic “ by Svensson et al.

The authors have addressed most of the comment satisfactorily, however there are several points that need further elaboration before I recommend publishing.

1. I repeat my original point below because there is no answer to the question why SiC is the best proxy for dust in two different regions that are subject to very different dust sources. I see the point that laboratory assays are important and they are valid, but the question is whether they are relevant. Why is SiC not a random choice?
p. 10, l. 16/22: Information on light absorbing constituents in SiC and stone crush are needed. The light absorption of mineral dust is strongly influenced by e.g., hematite, goethite etc. There is no point in determining a MAC of a “random” dust sample in the laboratory to apply it to the ambient samples with unknown contents of light absorbing constituents as the authors state themselves on p. 20, l. 13. It is not clear, why dust samples are tested in the laboratory that have no relevance for the ambient dust samples. Also, in Figure 7, the discrepancy between the gravimetrically determined SiC mass on filters and the estimated based on the MAC when values are > 7 g / m2 are just discarded without any further discussion. Is it possible, that the method does not work for high dust loadings? I suggest that the dust related aspects of the laboratory assays are drastically shortened to the information relevant for the ambient samples.

2. Even though there is no data pointing towards local sources, this aspect needs to be discussed.
l. 13 – 27: What about local dust sources? There are most likely bare rock or mountain walls from which mineral dust can be deposited on the glacier. In general dust sources and in particular local dust sources can be highly variable so that any kind of interpretation in terms of trends is difficult, especially if no information about the origin is available. If the authors had mineral dust size distributions at least a statement on potential role of local sources if very large particles are present could be made.

3. My suggestion how to test the importance of LAI loading on the filter as described below has been ignored.
l. 28 – 34: The interpretation that the MAC is decreasing towards the top of the snow pit is not convincing and an explanation why this might be the case is missing completely. There are not enough data points to conclude a trend in the MAC. The profile rather shows that LAI deposition is highly variable. Towards the bottom of the snow pit the ratio is also lower and if the point at 10 cm depth were not as low as observed, probably no trend would be inferred. The authors introduced on p. 20 the hypothesis that potentially large loadings on filters play a role for the observed variability of the MAC (see comment further above). To test this and to develop a potential explanation why the MAC changes, I recommend plotting in Figure 13 the ratio of the optical and TOM EC versus the ratio of Dust/EC and OC/EC to check if there is a relation with the overall LAI content. Also, indicate in the figure which ratio is meant. In the conclusion, the potential reasons for the observation should be given as well.

4. I do not agree with the authors’ answer that there is no point in removing the outlier. While it is true that it shows the scatter in the data, and therefore it is important to include it in the figure, it also drives the correlation. It is located in a ‘convenient’ spot, because it moves the slope closer to 1. As the authors say if the outlier were not there the agreement would be poorer. This needs to be discussed.
Figure 8: Is the outlier at Sunderhunga taken into account for the linear regression? If so, what would the slope look like without? It would be even greater than 19 % and that is a relatively large deviation between the optical measurement of EC with TOM and the estimated EC based on the MAC and the PSAP data. What does this discrepancy mean for the validity of the methods?

Minor point, in my opinion the title should read: “Light-absorption of dust and elemental carbon in snow in the Indian Himalaya and the Finnish Arctic”

Hide

RR by Anonymous Referee #1 (18 Oct 2017)

Suggestions for revision or reasons for rejection

Response response for Reviewer #1 of the first iteration of “Contribution of dust and elemental carbon to the reduction of snow albedo in the Indian Himalaya and the Finnish Arctic.” by Svensson et al., for ACPD.

The authors have addressed several of my main comments: they have added a relevant estimate of filter under catch, and satisfactorily addressed most minor comments.

Main comments “leftover”:
1) Carbonates. The authors have added clarification about the decomposition protocol they use, indicating that inspection of thermographs is sufficient to constrain artifacts of carbonates showing up as EC. However, I still recommend that assessments of uncertainty be improved/included.
a) Indeed, I had not understood the specific benefits of the EUSAAR_2 protocol previously; it certainly offers improved rejection of carbonates compared to IMPROVE. However, as pointed out before, Cavelli et al. 2010 deals with atmospheric aerosols that likely have much lower mineral dust loads than the samples here. Examination of their Figure 3 indicates a small response within the higher temperature times of their He/O2 devolution from natural calcite. For mineral dust/EC ratios as in the snow, the question of potentially significant artifacts even with the EUSAAR_2 protocol still requre assessment. Further the authors themselves suggest that ambiguous separation of EC/OC with their technique introduces scatter in their results; to quote from their response:
“The statement that the “good agreement” in fig.5 between the two optical measurements indicate that most of the scatter between TOM and PSAP is related to where the OCEC instrument determines to place the split point between OC and EC, still holds “
I do not understand how this could be reconciled with insensitivity to carbonates.

b) Zhang et al. also provide extra information about this issue. On the one hand, since they use IMPROVE_A, their estimates likely significantly overestimate the artifacts using EUSAAR_2. However, citing their 20% estimate is likely not appropriate beyond their sample ensemble: their snow was significantly higher in EC content than that here (on average ~1500 ng/g, their Figure 4), and it is not known how the mineral content of their samples relates, in amount and composition to those here. I expect that carbonate artifacts scale only with carbonate concentration, not relative to EC load. Another way to present their results could be via examination of their figure 4, where 5 samples were acidified to test carbonate contamination, and on average, shifted downwards in EC content by approximately 200ng/g each.

2) On the strength/relevance of constraints on ambient observations of the laboratory test. I am concerned that the authors did not understand some portions my original major comment 2. To paraphrase: measurements of eBC (i.e. interpretation of light transmission through a complex and uncharacterized filter) with different instruments (PSAP and OC/EC instrument) were carried out. Comparison of these two determinations does not give information about the actual BC or eBC on the filter, but only on the precision and relationship of the light transmission measurement. This is why it is not a clear constraint on the evolved EC measurement, and provides no information about the validity of the separate thermally evolved EC measurement or the relationship to actual eBC. This is why my suggestion of analyzing the results in the context of relative loadings (known from known dilutions) might help: the scaling of results with particulate load for the two different techniques (light attenuation or thermal evolution) could reveal influences of minerals on the BC load determination.

3) On the discussion of the trend of ratio with depth (Figure 13). I don’t think adding the sentence “This observation is rather hypothetical, and it needs to be further explored.” to this section sufficiently addresses my comment that there is no statistically relevant evidence for a trend.

4) It is very unlikely that the R^2 value given for 5 is correct

A separate question I am curious about is the potential impact of carbonate thermal decomposition in the OC stage of the EUSAAR_2 on the optical OC charring constraints; has this been considered?

Hide

ED: Reconsider after major revisions (24 Oct 2017) by Mingjin Tang

AR by Jonas Svensson on behalf of the Authors (09 Jan 2018)

ED: Referee Nomination & Report Request started (15 Jan 2018) by Mingjin Tang

RR by Anonymous Referee #3 (25 Jan 2018)

RR by Anonymous Referee #2 (09 Feb 2018)

ED: Publish as is (13 Feb 2018) by Mingjin Tang

AR by Jonas Svensson on behalf of the Authors (14 Feb 2018)

Short summary

Receding glaciers in the Himalayas are of concern. Here we present measurements of light-absorbing impurities, known to contribute to the ongoing glacier decrease, in snow from Indian Himalayas and compare them to snow samples from the Finnish Arctic. The soot particles in the snow are shown to have lower light absorbing efficiency, possibly affecting their radiative forcing potential in the snow. Further, dust influences the snow in the Himalayas to a much greater extent than in Finland.