Modelling the physical multiphase interactions of HNO<sub>3</sub> between snow and air on the Antarctic Plateau (Dome C) and coast (Halley)

Chan, Hoi Ga; Frey, Markus M.; King, Martin D.

doi:https://doi.org/10.5194/acp-18-1507-2018

Articles | Volume 18, issue 3

https://doi.org/10.5194/acp-18-1507-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/acp-18-1507-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 18, issue 3

Research article

|

02 Feb 2018

Research article |

| 02 Feb 2018

Modelling the physical multiphase interactions of HNO₃ between snow and air on the Antarctic Plateau (Dome C) and coast (Halley)

Hoi Ga Chan, Markus M. Frey, and Martin D. King

Download

Final revised paper (published on 02 Feb 2018)
Preprint (discussion started on 15 Dec 2016)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Interactive comment on "Modeling the Physical Multi-Phase Interactions of HNO3 Between Snow and Air on the Antarctic Plateau (Dome C) and coast (Halley)" by Hoi Ga Chan et al.', Anonymous Referee #1, 13 Jan 2017
- AC1: 'Reply to Referee 1', Hoi Ga Chan, 20 Apr 2017
RC2: 'Review of Chan et al.', Anonymous Referee #2, 20 Feb 2017
- AC2: 'Reply to Referee 2', Hoi Ga Chan, 20 Apr 2017
EC1: 'Editor Comment', Thorsten Bartels-Rausch, 19 Dec 2017

Peer-review completion

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

AR by Hoi Ga Chan on behalf of the Authors (20 Apr 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (25 Apr 2017) by Thorsten Bartels-Rausch

RR by Anonymous Referee #2 (14 May 2017)

Suggestions for revision or reasons for rejection

[Summary]

The authors have addressed my concerns on the model formulation of ice surface and bulk concentrations of HNO3 by providing rationales based upon experimental findings reported in the literature. I can accept their rationales, but then I am inclined to agree with one of the previous comments by the other referee in that the present model approach is not really based upon the “first principles”. Also, I am still inclined to feel that one of the major scientific contents in this paper, namely, the rejection of the viability of the disordered interface (DI) and its interactions with bulk solid ice in the air-snow exchange of HNO3, should be reconsidered seriously before the final publication of this manuscript is recommended.

Readability of the manuscript has increased significantly from the previous version. I still have some editorial suggestions for the authors to consider for improving the readability furthermore.

After all, the proof of the viability/applicability of their “Model 2” (surface adsorption, co-condensation, solid-state diffusion and liquid micro-pockets) in the surface air-snow HNO3 exchange is probably what this paper is all about. I will support the final publication of this study once the authors can satisfactorily address my remaining concerns.

[Specific comments]

1. The authors now explicitly argue that “the solid-state solution of nitrate molecules can occur concurrently with surface adsorption” (L150-151). If I understand correctly, the present model formulation dictates that a thermodynamic limitation does not exist anymore once HNO3 is retained on/in the outermost layer of their hypothesized spherical ice grain. As such, HNO3 trapped on the ice surface via Langmuir adsorption is assumed to be transferred instantaneously across the depth of the order of several micrometers to the outermost layer of bulk ice and subject to solid-state diffusion further inside based on the diffusivity having been determined experimentally. I am inclined to disagree to the claim that this model formulation is entirely based upon the first principles.

2. It appears to me that the treatment of the surface to bulk mass transfer of HNO3 is even more problematic in “Model 1” where the authors assume that HNO3 dissolved in the DI is subject to solid-state diffusion into the bulk volume of ice without thermodynamic limitation. This assumption effectively asserts that the equilibrium solubility of HNO3 in the entire volume of solid ice is identical to that in the liquid-like DI. Does any of the earlier studies advocating the role of DI assume such an equivalence between the DI and the bulk ice? So I still feel that the present study does not really offer the refutation of the HNO3 uptake onto the DI as a viable process.

3. In section 5.1.1, the authors refer to their model results based upon “Kinetic approach” and “Equilibrium approach” suggested earlier by Bock et al. (2016). These models are not exactly the same as the standard suite of models, “Model 1” and “Model 2”, in the present study. Hence the authors are apparently nudged to the names of the models used in Bock et al. (2016). I was confused when I first read this paper and again when I read the revised manuscript, partly because the transfer of HNO3 into liquid micro-pockets is treated by equilibrium in “Model 2”. I would rather like the authors to streamline the naming convention so that the “Kinetic” and “Equilibrium” models referred to here are called by numbers. For example, the “Kinetic” model can be a sub-category of “Model 1”. It is of course still useful to refer to the fact that these models are based on the “Kinetic” and “Equilibrium” approaches by Bock et al. (2016). Also, I would have liked it if the authors had introduced and discussed Eq. (19) in section 3, even though it does not belong to their standard model formulation.

[Minor comments and technical suggestions]

L70-71, “Murray et al., 2015; 3)”: An apparent typo.

L77-78: Please double check if this sentence makes sense.

L86: dynamicS

L177: “a 2” -> “two”

L178: “one” -> “an”

L241: “… growth law described by Flanner and Zander (2006):”

L334: below THE freezing POINT

L354: “particulate nitrate” -> “total particulate and gaseous nitrate” (?)

L389: well-mixed SO that

L425: Cv(RMSE) = 0.73 (Table 4)

L508: “inf” -> “in”

L512-513: “surface, layer leading to” -> “surface layer, leading to”

L513: “find in a more stable” -> “found in the particulate”

L522: information IS AVAILABLE on

L525 & L550: “maybe” -> “may be”

L581: requireD

L599 & L670: “larger” -> “higher”

L665: “sensitivity” -> “sensitive”

Table 1: Mathematical expressions (second column) for “Interfacial mass transport to a liquid phase” and “Gas-phase diffusion to the surface of a spherical droplet” should be inverted.

Hide

RR by Anonymous Referee #3 (26 May 2017)

Suggestions for revision or reasons for rejection

The authors have performed an interesting modeling study of an important and valuable dataset. The technical work is a worthwhile contribution to the literature; some new approaches are introduced and reasonable agreement with field data is achieved. However, the results are not currently presented in a manner that is acceptable for publication. The authors disparage previous work in the text for requiring tuning parameters and unrealistic assumptions. This is a fair criticism of nearly all modeling studies in this field, since fundamental understanding lags significantly behind our needs for process modeling, and therefore assumptions are required. The problem is that the authors present their modeling approaches as being superior to previous work without acknowledging up front the great uncertainty surrounding several of the "first principals" assumptions made in this study. It is not my intention to call out these assumptions as erroneous, but rather to insist that the authors acknowledge them and discuss the uncertainties that they introduce. For example, in model 1: The range of the DI onset temperature, To, is based on experimental data, but it should be acknowledged that this is a type of tuning parameter. Another assumption must be made regarding the depth/volume of the DI, although I did not see this discussed in the text. Uptake of HNO3 to the DI is assumed (as it is in nearly all snowpack modeling studies, for lack of a better option) to follow Henry's Law for bulk aqueous solutions, where in reality there is no physical basis for this assumption. For model 2, eq. 4, which was developed assuming that the liquid solution is ideal and does not take into account partitioning of nitrate between the gas, liquid, and solid phases, is applied without discussion. The authors should also take care to acknowledge when aspects of their models are adopted from or are similar to previous modeling studies, in addition to highlighting their innovations.

Hide

ED: Reconsider after major revisions (29 May 2017) by Thorsten Bartels-Rausch

AR by Anna Mirena Feist-Polner on behalf of the Authors (28 Jul 2017) Author's response

ED: Referee Nomination & Report Request started (17 Aug 2017) by Thorsten Bartels-Rausch

ED: Reconsider after minor revisions (Editor review) (12 Sep 2017) by Thorsten Bartels-Rausch

Dear Hoi Ga Chan et al.

Thank you very much for going through the review process and addressing the referee comments so detailed. This is a very timely and detailed work on a highly controversial topic, so I’m very much looking forward to your work being published in ACP. Yet, after going through the reply to the referee, there remain to details that are not entirely clear. Despite the intensive review process, I therefore need to come back to you to clarify these two aspects.

One point is connected to the thickness of the DI and eq. 14 What means infinitesimal thickness mathematically precisely and how do you define a concentration in such a DI with a volume of or approaching 0? With this question, I also refer to the molecular budget, which I’m sure you have looked at, but which would be worth mentioning. Did I understand correctly, that the flux into the DI matches the flux from the DI into the bulk and that the concentration in the DI based on Henry is established? If so, I do not understand the statement on page 25 (825): “In general, the grain boundary concentration of nitrate defined by solvation into the DI is much larger than when it is defined by the combination of surface adsorption and co-condensation on ice.“ Does this imply that the fluxes are balanced?
The second aspect touches question 2 of report 1: You don not explicitly mention an upper limit of solubility in the bulk ice. That of the DI is given by Henry. That of the inner ice is given by the solid solution? If correct, I suggest to clearly state this (lines 216-230) to prevent the impression that the whole grain can become liquid-like as the DI holding such high concentrations of solutes.
Some minor suggestions:

Page 3, line 90: Please define Co-condensation here.
Page 4, line 124. I would not say that the idea of liquid co-existing with ice comes from the Domine paper. This is given by thermodynamics; Domine strongly argued for arrangement in pockets as you state later. I suggest to rather cite Cho or McNeill, ACP (2011)
Page 7, line 210: I suggest to mention that the pH in the liquid content of the ice is not equal to the pH of molten snow.
Page 9, line 275: The adsorption as shown by Ullerstam might be under saturated, but it is still in equilibrium. So, I can’t quite follow this argument.
Page 12, line 371:replace water with brine or solution.

With best regards,
looking forward to your clarification.

Thorsten Bartels-Rausch

Hide

AR by Hoi Ga Chan on behalf of the Authors (16 Oct 2017) Author's response Manuscript

ED: Publish subject to technical corrections (19 Dec 2017) by Thorsten Bartels-Rausch

Short summary

Emissions of reactive nitrogen from snowpacks influence remote air quality. Two physical air–snow models for nitrate were developed. One assumes that below a threshold temperature the air–snow grain interface is pure ice and above it a disordered interface emerges. The other assumes an air–ice interface below melting and that any liquid present is concentrated in micropockets. Only the latter matches observations at two Antarctic lcoations covering a wide range of environmental conditions.

Modelling the physical multiphase interactions of HNO3 between snow and air on the Antarctic Plateau (Dome C) and coast (Halley)

Download

Interactive discussion

Peer-review completion

Modelling the physical multiphase interactions of HNO₃ between snow and air on the Antarctic Plateau (Dome C) and coast (Halley)