Effect of soil property uncertainties on permafrost thaw projections: a calibration-constrained analysis

Harp, D. R.; Atchley, A. L.; Painter, S. L.; Coon, E. T.; Wilson, C. J.; Romanovsky, V. E.; Rowland, J. C.

doi:https://doi.org/10.5194/tc-10-341-2016

Articles | Volume 10, issue 1

https://doi.org/10.5194/tc-10-341-2016

© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/tc-10-341-2016

© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 10, issue 1

Research article

|

11 Feb 2016

Research article |

| 11 Feb 2016

Effect of soil property uncertainties on permafrost thaw projections: a calibration-constrained analysis

D. R. Harp, A. L. Atchley, S. L. Painter, E. T. Coon, C. J. Wilson, V. E. Romanovsky, and J. C. Rowland

Download

Final revised paper (published on 11 Feb 2016)
Supplement to the final revised paper
Preprint (discussion started on 29 Jun 2015)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC C1331: 'review for Harp et al.', Anonymous Referee #1, 28 Jul 2015
- AC C1397: 'Responses to anonymous referree #1', D.R. Harp, 10 Aug 2015
RC C1605: 'Review', Anonymous Referee #2, 09 Sep 2015
- AC C1661: 'Response to referee #2', D.R. Harp, 16 Sep 2015

Peer-review completion

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

AR by D.R. Harp on behalf of the Authors (23 Sep 2015) Author's response Manuscript

ED: Reconsider after major revisions (30 Sep 2015) by Julia Boike

AR by D.R. Harp on behalf of the Authors (11 Nov 2015) Author's response Manuscript

ED: Referee Nomination & Report Request started (11 Nov 2015) by Julia Boike

RR by Anonymous Referee #2 (01 Dec 2015)

Suggestions for revision or reasons for rejection

The paper has been greatly improved in the revised version. The structure is clearer, it is easier to understand what the authors are doing, and the number of figures in the paper has been reduced to an acceptable amount. The authors have addressed many of the concerns of the reviewers and the associated editor.

However, there are some remaining concerns, which prevent me from recommending the paper to be accepted as is. Therefore I rather recommend the paper to be reconsidered after another round of major revisions.

The most serious point is, that it is obvious from figure 9, that the authors use a grid resolution of 10 cm below a depth of about 90 cm. While this might be appropriate for the original 2006 climate, where the active layer thickness is only about 30 cm, it might cause serious discretisation errors at the end of the simulated period, when the active layer depth is around 100 cm. The gradients especially in water potential are very high at the freezing front and require a proper resolution. The consequence of a too coarse resolution would be smoothed gradients, which might partially explain the very low Stefan numbers at the end of the simulation. The authors therefore have to conduct grid refinement studies to check if grid convergence has been obtained. If this has been done already, it should be mentioned in the text and should be documented in the supplement. As this is a major point, I have to propose to reconsider the paper after major revisions.

There are some minor points as well:
- Equation (4): phi is here used for the first time, but not explained.
- Line 359-365: For me this paragraph can be misunderstood. For me a "fit" can be the adaptation of a model to measured data (which was performed for 2013), whereas for 2014 one can see the agreement between the model and the data (as there was no calibration for 2014, as far as I understood). You should make this clearer.
- You should also add the time series at different depth for the three locations for 2014 to the supplement, as it is done for 2013 in figure S2-S4.
- Line 427-432: The statement, that the partitioning between methane and carbon dioxide will not change is very bold, based on the available data. The active layer depth and the annual thaw depth duration as well as the annual mean liquid saturation are very aggregated parameters, whereas the production of methane or carbon dioxide depends on local conditions at certain times. For example methane is even produced in soils under generally aerobic conditions (which is explained by anaerobic microsites). Especially due to the high gas saturations down to 1 metre (as shown in figure 7) I would rather expect a change in the biochemical processes.
- Line 444-446: The changes in the underlying permafrost excites in me the question, what the lower boundary condition of the model is. If I understand line 155-156 correctly, you are simulating a domain down to 50m? Are you then using no-flux at the lower boundary?
- Figure 6, Line 448-462: To me it is not evident, what is shown in this figure. I reckon that it is temperature profiles at a certain time (the time of maximal thaw depth) for different parameter scenarios for 2100 with the colour specifying the day, when this maximum is reached. However, what exactly is plotted for 2006? Is this the median of the curves for different scenarios also at the day of maximum thaw depth or is it averaged over days 246-260? Please make this clearer.
- Figure 7: please comment in the text especially on the higher gas saturation, which will lead to a continuous gas phase and thus aerobic conditions down to at least 80 cm. Also it might be mentioned that the high liquid saturation down to 80 cm might result in increased subsurface run-off (depending on the topography, which I do not know). I assume that this is not included in the model.
- Line 529-544: "Some strong correlations are apparent..." this is rather sloppy language. There are only four strong correlations, and three of them are with the same variable, the porosity of the peat soil. Then there is a moderate correlation between the residual water content of both peat and mineral soil with the annual mean liquid saturation, which is to be expected. I am not sure, if this justifies four large figures. If you are confident, that the regression in table 2 has some justification, it would rather be interesting, to visually assess the agreement between ALT, D and Stefan number predicted only based on the porosity of the mineral soil and your simulation results.

Hide

ED: Reconsider after major revisions (03 Dec 2015) by Julia Boike

AR by D.R. Harp on behalf of the Authors (15 Jan 2016) Author's response Manuscript

ED: Publish subject to technical corrections (19 Jan 2016) by Julia Boike

AR by D.R. Harp on behalf of the Authors (19 Jan 2016) Manuscript

Short summary

This paper investigates the uncertainty associated with permafrost thaw projections at an intensively monitored site. Permafrost thaw projections are simulated using a thermal hydrology model forced by a worst-case carbon emission scenario. The uncertainties associated with active layer depth, saturation state, thermal regime, and thaw duration are quantified and compared with the effects of climate model uncertainty on permafrost thaw projections.