Correlation between tectonic stress regimes and methane seepage on the western Svalbard margin

Plaza-Faverola, Andreia; Keiding, Marie

doi:https://doi.org/10.5194/se-10-79-2019

Articles | Volume 10, issue 1

https://doi.org/10.5194/se-10-79-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/se-10-79-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 10, issue 1

Research article

|

15 Jan 2019

Research article |

| 15 Jan 2019

Correlation between tectonic stress regimes and methane seepage on the western Svalbard margin

Andreia Plaza-Faverola and Marie Keiding

Download

Final revised paper (published on 15 Jan 2019)
Supplement to the final revised paper
Preprint (discussion started on 18 Apr 2018)
Supplement to the preprint

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'methane seepage and north atlantic ridge activiy', Anonymous Referee #1, 10 May 2018
- SC1: 'Reply to anonymous referee #1', Marie Keiding, 15 May 2018
  - RC2: 'Correlation between tectonic stress regimes and methane seepage on the west of west-Svalbard margin', Anonymous Referee #1, 16 May 2018
    - AC3: 'Actions based on comments by referee 1', Andreia Plaza-Faverola, 15 Jun 2018
EC1: 'TE comment', Federico Rossetti, 16 May 2018
- AC1: 'Description of how we intend to implement the comments by referee 1 to improve the manuscript', Andreia Plaza-Faverola, 22 May 2018
RC3: 'Comments se-2018-11', Anonymous Referee #2, 23 May 2018
- AC2: 'Implementation of comments by referee 2', Andreia Plaza-Faverola, 28 May 2018

Peer-review completion

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

AR by Andreia Plaza-Faverola on behalf of the Authors (02 Jul 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (03 Jul 2018) by Federico Rossetti

RR by Anonymous Referee #1 (11 Jul 2018)

Suggestions for revision or reasons for rejection

As pointed out in my review of the original version of this paper, authors rightfully point out that the local stress field, where seepage occurs, may results from four different loading processes:
A: gravitational effects through topography effects and density contrasts;
B: Flexural effects due to sediment erosion and deposition;
C: Glacially flexural effects;
D: tectonic stresses resulting from the Vestnesa Ridge.
Authors claim that by investigating solely the ridge push effect they can demonstrate that it is the most significant effect that controls methane production. In order to undertake this demonstration they propose to use Okada ‘s elastic solution that addresses stresses caused by a dislocation in an homogeneous infinite plate.

I pointed out in my review that, in elastic theory, when various loading processes are applied on the same volume of an elastic material, the resulting stress fields is the sum of that caused by each of the various loading processes applied separately. In the present case, authors must demonstrate that effects of A+B+C are negligible as compared to that of D, their push ridge model, if they wish to conclude that only D is of import. But they fail to do so.
Interestingly, authors mention in their introduction that Wallman et al. (2018) conclude that hydrate dissociation off Svalbard is induced by isostatic glacial rebound. So clearly not everybody agrees with their proposition that glacial rebound is negligible.

Before addressing the stress analysis proposed by authors for the ridge push effect, I will point out that topography effects at a depth of 1200 m is of particularly import given the bathymetry shown on figure 1 , which indicates a slope extending from -1 km to -3km below sea level. Clearly the stress field close to this free surface depends strongly on topography and its effect must be analyzed properly before claiming it is negligible.
In Appendix A, Okada’s dislocation is embedded in the ductile part of the crust. This is precisely contradictory with the elastic hypothesis of Okada’s solution and renders the stress analysis erroneous. In addition, it is stated, p 4 - second paragraph, that the total sedimentary thickness is larger 5 km. Elastic characteristics of sediments differ strongly from that of basement crystalline rocks so that the homogeneous semi-infinite space analysis does not apply for the elastic part of the crust.
In conclusion, first the stress field analysis of the ridge push effect is erroneous, second the proposition that topography effects are negligible has not been demonstrated, third results from Wallman et al. show that glacial rebound is an important factor controlling methane production off Svalbard.
My final recommendation is not to publish this paper.

Hide

RR by Anonymous Referee #2 (12 Jul 2018)

ED: Reconsider after major revisions (13 Jul 2018) by Federico Rossetti

AR by Andreia Plaza-Faverola on behalf of the Authors (10 Aug 2018)

ED: Referee Nomination & Report Request started (14 Aug 2018) by Federico Rossetti

RR by Anonymous Referee #3 (28 Aug 2018)

Suggestions for revision or reasons for rejection

In this contribution the authors are trying to bring together tectonic modeling (stress/strain) and observations on fluid seepage at Vestensa, off Svalbard. Their model suggests a region of tensile stresses that mostly coincides with a region of actively degassing pockmarks on the Vestnesa Ridge, juxtaposed to zones of strike-slip stress and inactive pockmarks.
I am reviewing this contribution after a first initial set of reviews by different reviewers was completed. I am judging my response on (a) the scientific material presented in the revised manuscript and (b) the earlier comments made by one reviewer on the applicability of the model used.

I find the approach presented in this manuscript intriguing, results rather interesting, but remain skeptical on the conclusions drawn. The paper needs some modifications, especially more discussions and at least one other set of tests if the zone of tensile stress is indeed robust in its location by modifying plate boundary geometries (which can be drawn differently than what authors show).
A few general remarks that I think would help in getting this paper published:
(a) Pockmarks on the southern drift. The authors show faults mapped south of the MTF, where seismic and bathymetric data show abundant chains of pockmarks. Those pockmarks are inactive in the terminology of the authors (i.e. inactive= no degassing). I suggest the authors augment their discussion on how to explain those southern pockmark chains as well. The orientation of principle stresses is strikingly different south and north of the MTF, relative to mapped fault-orientations (parallel, perpendicular) and earthquake focal mechanisms. Why is this not important as well?
(b) Why only use degassing as sign of “activity”? Very recent heat-flow surveying (>100 measurements) taken across the pockmarks show that pockmarks may be still “active” even without gas discharge, as they can show a strong and localized heat-flow anomaly to the pockmark depressions themselves (and that occurs north and south of the MTF and the zone of tensile stress regime). The authors are aware of this study and data as well as preliminary results from the heat-flow study are available through the cruise report of expedition MSM57 and can be downloaded from the University of Bremen website.
(c) Degassing could be completely unrelated to the deep-rooted stresses – microbial activity rates are changing; there is coupling between ocean phenomena and the sediment with free gas phases and gas hydrates (e.g. seen in tidal modulated gas fluxes or decadal oscillations). I think, using degassing alone is “too short” an argument to investigate the relationship between stress and the fluid flow regimes.
(d) Sediment drift bodies such as the Vestnesa Ridge are heavily dynamic and evolve quickly. Topography is steep across the ridge and vertical stresses thus change fast. While the authors do say and have indicated in the revised version of their manuscript that they do not reject other causes (and associated stresses) of the difference in vent activity, but I find the outcome or limitation of discussion only on the zone of tensile stress rather weak overall, given this very dynamic and regionally fast-changing zone of study.
(e) The geometry of the plate boundaries, especially the MTF, seems to be fixed in all the discussion, but I could easily draw the MTF at a different angle (e.g. rotated by 5 degrees) by just looking at the bathymetry. Would this not affect the zone of tensile stress regime? The authors did look at parameter variations and assess effects in the actual zones of stresses, but they never changed the geometry. Maybe test quickly, if this is important or not?
(f) Coupling between crustal deformation and sediment load seems to have been assumed as 100%. Is that acceptable? Could those regimes be decoupled, i.e. the ridge-push is doing its own thing and sediment drape from contourites deposited during glacial/deglacial phases react to other forces?

Overall, the manuscript has its contribution that is interesting and intriguing, fully worth of getting published at some point; but as of yet, it has too little result and discussion to be fully acceptable. After the initial revision, the authors modified and restated many points in their discussion and conclusion but need to go even further. I suggest a more diverse discussion taking some of my points above into account and possibly assess the “error” in the stress-regime mapping a step further.

Then, what are the larger implications of such process? Could you maybe name other regions where similar processes occur? What other studies would help assess this phenomenon further? Borehole stress measurements? Are there maybe wire-line image log (from resistivity) data from the old ODP boreholes to look at the borehole stress by mapping breakouts or fractures? Could you use the shear-wave splitting technique to look at the stress components by doing a polarization analysis of the first arrivals (e.g. Tonegawa et al., 2016)? Are those changing across Vestnesa and to the Svyatogor Ridge south of the MTF?

Tonegawa, T., Obana, K., Yamamoto, Y., Kodaira, S., Wang, K., Riedel, M., Kao, H., Spence, G.D., 2016. Fracture alignments in marine sediments off Vancouver Island from Ps splitting analysis, Bulletin of the Seismological Society of America, 10.1785/0120160090.

Hide

RR by Anonymous Referee #4 (14 Sep 2018)

Suggestions for revision or reasons for rejection

Dear Authors,

Understanding flow localisation patterns and location is a challenging entreprise. The submarine features leading to the expulsion of methane and other gases are widely reported, but currently the mechanisms remain still very unclear and discussed. In this context, this study tries to address a challenging point, constraining the flow localisation in Western Svalbard based on a analytical mechanical model of the complex lithospheric regional layout.

Although the model used seems to be correctly applied, the major simplifications and hypotheses the authors make in order to use it significantly limit the conclusion that can be drawn. This point was also discussed and criticised in the past round of revisions.

I will propose some broad suggestions that may enhance clarity and readability of the article, then address some specific amendments.

> Since the aim of this study is a modelling exercise, it would be nice to have a more rigorous model description. The model configuration is not very clear. It would be very helpful for the reader to see on the figure the initial and boundary conditions. On Fig. 4 as well, it would be very helpful to have these information present, as well as the highlight of the Vastness ridge.

It would be very helpful for the reader to have a more structured result dissection and discussion. Maybe the authors could expose the full set of hypotheses and more quantitative facts in a first step, and then discuss and present their interpretation regarding the various aspects.

Also, I would ask the authors to clearly distinguish between tectonic stress configuration optimal for enhanced fluid motion (extension regime) and actual mechanisms that may lead to flow focussing within chimneys and pockmarks. Fault planes will fail to produce pipe like features from a mechanical point of view.

> The more specific comments are following:
L. 60-63: elastic stresses resulting from strain are instantaneous, thus I would be very careful using them to explain long term evolution and variations. They may be related to viscous creep rather than to purely elastic medium.

L. 73: please precise which fluids in the fluid dynamics you refer to.

L. 105-110: it is difficult to get the importance of the regional geology in the way it is brought to the reader. It would be very interesting to read about how these different geological formation impact the model, and the results.

L. 125-126: The last sentence is difficult to get, please rephrase.

L. 130: the GHSZ does not seem to be reported in Fig. 2, but rather in Fig. 3 ?

L. 134-138: What is the importance of those facts with regards to the proposed work ?

Section 4: Modelling approach. This section is about the aim of the paper, modelling. Thus, it would greatly benefit from enhanced details and a more rigorous model description, with regards to initial and boundary conditions especially.

L. 173: agreeing on the usage of a very simplified model, orientation of the stress field may be valid to interpret from the model, but I would be very careful regarding the magnitude of stresses, even if relative.

L. 210-213: Long and unclear sentence

L. 214: please precise style of principal stress

L. 218: please consider to rephrase avoiding the usage of “thing”

L. 219: please consider rephrasing as following “… provides a correct first order prediction of the stress field ORIENTATION in the upper crust…”

L. 221: consider the preferred usage of section number rather than below / above …

L. 236: consider rephrasing “thicker sediment thickness” to make it more clear

L. 240: what is seismic definition ?

L. 268: Gravitationnel stress would not only induce the suggested stress, but a complex pattern.

L. 292 what unit is mm/a ?

L. 234-235: Please make it clear to distinguish between tectonic stress that may enhance fluid circulation in specific location from the actually mechanism that will result in fluid localisation within pipes or chimneys. Faults and fractures will not be able to produce tubular like features, based on mechanics.

L. 335: consider replacing above by the appropriate section number.

Fig. 2: missing BSR in legend ?

Fig. 4: add boundary conditions, highlight the Vestnesa Ridge

Based on the aforementioned arguments, I would suggest to publish this article but asking minor revision.

Hide

ED: Reconsider after major revisions (15 Sep 2018) by Federico Rossetti

AR by Andreia Plaza-Faverola on behalf of the Authors (30 Oct 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (01 Nov 2018) by Federico Rossetti

RR by Michael Riedel (10 Nov 2018)

RR by Anonymous Referee #5 (02 Dec 2018)

Suggestions for revision or reasons for rejection

I congratulate with the authors for the work done. This is a first important step to assess the relevance of tectonic stress propagation in the sedimentary cover of continental margins in controlling fluid migration in the long term. I acknowledge the work done to improve the manuscript according to the reviews received in the first round. I add below a few overall minor remarks and suggestions addressing some aspects not covered before.

Line 219. Earthquakes focal mechanisms. In the introduction (line 94), you state that earthquakes focal mechanisms provide poorly constrained stress vectors. However, in following analysis the focal mechanisms are used as a robust evidence of the stress field, so that they can be used to validate the results of the model. I wander why the rather clear evidence from focal mechanisms was not properly used. Perhaps the initial statement in the introduction should be smoothed?

Line 232 and following text. One important aspect of your analysis, somehow not properly valued in the paper (although you address it in the discussion), is the role of compaction disequilibrium in the sedimentary column as a factor of weakening the strength of the sediments and therefore allowing the rupture to propagate upwards. I have a few arguments here that could be incorporated in the text with the purpose to give even more emphasis to the role of undercompaction:
You state rather vaguely here that the sedimentary column below the Vestnesa Ridge is not expected to be highly consolidated. I would not use the term ‘highly consolidated’. Sediments are either normal- or over- or under- consolidated. Over consolidation results from the maximum stress experienced by the sediment being higher than the present stress. This occurs only in the case of subaqueous erosion (tectonic or sedimentary), or exhumation of a marine sedimentary sequence, which is not your case. In a sedimentary sequence experiencing continuous sedimentation, like the Vestnesa sediment drift, you either have normal consolidation (if excess pore water induced by sedimentation can dissipate without building overpressure) or underconsolidation (in the case you have a rapidly building total stress in consequence of rapid sedimentation, or increase in the volume of the pore fluids by, in your case, volumetric expansion of methane). In this case your pore pressure may increase above hydrostatic, with decreasing effective stress (decreasing strength), to the point of reaching the total stress. Above this limit you may have hydrofracturing and sediment inflation. (I think this is the process you describe citing “ the minimum effective stress is negative in Line 428. So hydrofracturing in the sedimentary column may be induced independently, without the crustal stress contribution, by a combination of increasing sedimentary load and increasing pore fluid volume. Of course in your case you are adding another factor that is the propagation of the crustal stress into the sedimentary column. In your case this contribution of tensile stress is towards a decrease of the total stress on the grains in favor of underconsolidation. I see you refer to this in Line 425. Later in the discussion you report a more important details, like the sedimentation rate. It would be important to know also the lithology (sampled or inferred). This is because it is generally assumed that in low permeability sediments (clay-rich formations), a sedimentation rate of 1 mm a-1 is enough to produce a situation of underconsolidation alone (Rubey & Hubbert 1959; Fertl 1976). You are not too far from this value below the crest of the ridge. I would rephrase your sentence in Lines 320-322 not I the sense of a stress build up, but rather a strength decrease due to compaction disequilibrium.
I am mentioning this because I think you should make even more clear in your discussion, the role of the consolidation state in the Vestnesa Ridge sedimentary sequence, saying that pore fluid overpressure (decreasing strength) may results from rapid sedimentation rate and fluid volume increase (gas expansion), below the crest of the ridge (the depocentre). The decreased effective stress in this area is a good explanation of the reason why you find the gas expulsion features and the faults here.

RUBEY, W.W. & HUBBERT, M.K. 1959. Role of fluid pressure in mechanics of overthrust faulting: II. Overthrust belt in geosynclinal area of western Wyoming in light of fluid-pressure hypothesis. Geological Society of America Bulletin, 70, 167–206
FERTL, W.H. 1976. Abnormal Formation Pressures. Implications to Exploration, Drilling, and Production of Oil and Gas Resources. Developments in Petroleum Science, 2. Elsevier, Amsterdam.

Line 419. The role of gas hydrates in making the faults and fractures less permeable to fluid migration is very important. I suggest to cite and use the results of the paper:
Madrussani et al., 2009. Gas hydrates, free gas distribution and fault pattern on the west
Svalbard continental margin. GJI. doi: 10.1111/j.1365-246X.2009.04425.x
Where the authors provide geophysical evidence (Vp, Vs and attenuation) of the role of gas hydrates in decreasing the permeability of the faults and driving lateral gas migration below the base of the GHSZ.

Line 498. The meaning of bathymetry as a source of stress is, at least to me, a bit unclear. I understand you mean gravitational forces along a slope. Is it so? Is stress induced by bathymetry coded?

Minors:
Line 243 replace “gentle” with “small”
Line 288. Replace “understood” with “considered”
Line 312. replace “nearby” with “focusing around”
Line 355. I would delete the sentence “All the conditions are given for sustaining seepage along the entire ridge” as I find it unnecessary.

Hide

ED: Publish subject to minor revisions (review by editor) (02 Dec 2018) by Federico Rossetti

AR by Andreia Plaza-Faverola on behalf of the Authors (07 Dec 2018) Author's response Manuscript

ED: Publish as is (13 Dec 2018) by Federico Rossetti

ED: Publish as is (18 Dec 2018) by J. Huw Davies (Executive editor)

AR by Andreia Plaza-Faverola on behalf of the Authors (18 Dec 2018) Manuscript

Short summary

Vast amounts of methane are released to the oceans at continental margins (seepage). The mechanisms controlling when and how much methane is released are not fully understood. In the Fram Strait seepage may be affected by complex tectonic processes. We modelled the stress generated on the sediments exclusively due to the opening of the mid-ocean ridges and found that changes in the stress field may be controlling when and where seepage occurs, which has implications for seepage reconstruction.