Implications from palaeoseismological investigations at the Markgrafneusiedl Fault (Vienna Basin, Austria) for seismic hazard assessment

Hintersberger, Esther; Decker, Kurt; Lomax, Johanna; Lüthgens, Christopher

doi:https://doi.org/10.5194/nhess-18-531-2018

Articles | Volume 18, issue 2

https://doi.org/10.5194/nhess-18-531-2018

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

Special issue:

Linking faults to seismic hazard assessment in Europe

https://doi.org/10.5194/nhess-18-531-2018

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

Articles | Volume 18, issue 2

Research article

|

21 Feb 2018

Research article |

| 21 Feb 2018

Implications from palaeoseismological investigations at the Markgrafneusiedl Fault (Vienna Basin, Austria) for seismic hazard assessment

Esther Hintersberger, Kurt Decker, Johanna Lomax, and Christopher Lüthgens

Download

Final revised paper (published on 21 Feb 2018)
Supplement to the final revised paper
Preprint (discussion started on 15 May 2017)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of “Implications from paleoseismological investigation at the Markgrafneusiedl Fault (Vienna Basin, Austria) for seismic hazard assessment “ by Hintersberger et al. submitted to Natural Hazards and Earth System Sciences.', Ryan Gold, 05 Jun 2017
- AC1: 'Response to RC1', Esther Hintersberger, 25 Aug 2017
RC2: 'nhess-2017-126 Reviewer report (CLARK)', Dan Clark, 10 Jul 2017
- AC3: 'Response to RC2', Esther Hintersberger, 25 Aug 2017
RC3: 'Annotated pdf version', Dan Clark, 10 Jul 2017
RC4: 'Revision of paper- Interactive comment', Maria Ortuño, 13 Jul 2017
- AC2: 'Response to RC4', Esther Hintersberger, 25 Aug 2017
RC5: 'Annotated version', Maria Ortuño, 13 Jul 2017

Peer-review completion

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

ED: Reconsider after major revisions (further review by Editor and Referees) (30 Aug 2017) by Oona Scotti

AR by Esther Hintersberger on behalf of the Authors (13 Oct 2017) Manuscript

ED: Referee Nomination & Report Request started (20 Oct 2017) by Oona Scotti

RR by Angela Landgraf (06 Nov 2017)

Suggestions for revision or reasons for rejection

Review of revised version of Hintersberger et al. 2017 “Implications from
palaeoseismological investigations at the Markgrafneusiedl Fault (Vienna Basin, Austria) for seismic hazard assessment”

I read the three reviews and subsequent major revision of the manuscript, including the response to the reviewers and the revised paper and supplement. I think that the authors have addressed the issues raised by the three referees in a comprehensive manner and documented these changes accordingly.

Below, I will briefly comment only on the main arguments and follow up with a few minor suggestions to be addressed before publication. To do so, I summarize the reviewer’s statements in black and my evaluation of their treatment by the authors in blue font color.

Three referees who are all experts in paleoseismology and familiar with trenching, earthquake histories and associated geochronology have seen the original paper. They all agree that this is an important and appropriate study to be published in Natural Hazards and Earth System Sciences, that it provides new paleoseismic results and will be an important contribution to the neotectonic and seismic hazard communities, including emergency managers.
To push the article to it’s full potential, all reviewers basically agree that the following issues need to be addressed:

Data & Discussion

(1) The event history and the event correlation between the sites, which is based on limited age control of bracketing units, needs some reworking to clearly outline the events by their confining units and to provide a common framework for their timing including uncertainties. To do so, two of the reviewers suggest using a probabilistic analysis with OxCal - that is good practice in many paleoseismic studies worldwide.

The authors have adapted their model using the suggested OxCal model. The modeled results are comparable to the previous “hands-on” calculation, which seems to me a good indication of a robust interpretation. The new approach and results are incorporated in the associated tables, method section, and figure and the code is added to the supplement. → Point seems fully addressed in this new version

(2) The anticipated fault behavior needs some redefinition or clarification. While the authors aim to differentiate between potential characteristic and super-cycle earthquake occurrence, the discussion and review revealed that periodic vs. clustered scenarios would be better end member scenarios to evaluate. Two of the referees additionally suggest to use to the coefficient of variation to statistically discriminate between the two.

The new manuscript has clarified terms and used only quasi-periodic vs. clustered behavior. → Point seems fully addressed in this new version

(3) All reviewers wish to see a discussion (or data) on a potential linkage between the MF and the more active VBTF with the effect of such a scenario on earthquake timing and maximum magnitude for the Vienna basin.

Although only briefly, this possibility is now discussed and the consequences of potentially larger magnitudes for combined ruptures mentioned. → Point is briefly addressed in this new version; authors give good arguments in the rebuttal letter that the data so far don’t allow for a more detailed discussion of this issue.

(4) Reviewers would like to see an improved discussion on the preservation potential of the earthquake evidence, which also includes alternative interpretation (or discussion) for origin and/or overprint of colluvial wedges. This is an important issue in low-strain areas, especially under influence of geomorphic consequences of glacial-interglacial cycles.

This issue got a new paragraph in the discussion chapter → fully addressed there and referred to also when discussing smaller ruptures in between clusters.

Presentation & Style

(1) The reviewers suggest that a methods section before the results will help to build a framework around the study without interrupting the text flow within the results by such information.

The new manuscript has a method chapter in a comprehensive and well-structured format → point fully addressed.

(2) A detailed site description (in figure) needs to supplement the trenching information to provide the reader with the geomorphic understanding of the area.

The new manuscript has a new figure (Figure 4) to account for these issues. Here, however, I feel that the points are not fully taken, see also my additional comments below. → needs revision

(3) The original (uninterpreted) photomosaics should be provided (at least as supplement).

→ Point fully addressed by incorporating the photomosaics into the supplemment

(4) The reviewers suggest a relabeling/restructuring of events to avoid misunderstandings about the number/sequence of earthquakes at or between the sites.

The event-labeling was changed accordingly. → Point fully addressed.

Additional comments

(1) Response to RC4 (Ice loading). While this may not necessarily need to find its way into the paper, I would like to emphasize that there is some discussion in the literature suggesting significant rebound of the Alps after LGM deglaciation (e.g., Norton and Hampel, 2010; Mey et al., 2016). Without knowing the exact location of the Alpine extent in Austria, the wavelength of isostatic compensation might be larger and includes the previous forebulge area, where vertical adjustment might have a different effect on normal faults than known for reactivated thrust faults that have been located below the Ice. In Switzerland, indications for extended glaciations also exist for MIS 6, so a time period just pre-dating the time interval of the earthquakes dated in this study.
(1) Please add references for the statement in the first paragraph of the introduction (i.e., the Sumatra, Tohoku, Haiti and Christchurch EQ-cases).
(2) Same paragraph, pg. 2, lines 5-8: There is much doubling, I suggest merging these sentences.
(3) It might be worth citing the introductory paper of the book Landgraf, A., Kuebler, S., Hintersberger, E. & Stein, S. (eds) Seismicity, Fault Rupture and Earthquake Hazards in Slowly Deforming Regions. Geological Society, London, Special Publications, 432, http://doi.org/10.1144/SP432.13 that addresses many of the mentioned issues (E. Hintersberger is one of the editors).
(4) Section 3.3: Did you follow the approaches of DuRoss et al., or Personius et al., as suggested by the reviewers? They might be cited here.
(5) Figure 4: I’m not sure, if this selection of field photos provided here adds much to the understanding (i.e., 4 C, D, F). 4 E is good, but could benefit from some annotation (e.g., fault scarp) – same for B. Also, perhaps a hillshade, instead of the DEM (in 4 A, but perhaps could be even considered for Fig. 3)) would bring out the geomorphology better. As a geomorphic map, it would also benefit from some annotation (perhaps having (A) as hillshade and (B) next to it as interpreted geomorphic map within the same extent (see Meriaux et al., 2014/2015 for nice examples of geomorphic mapping). Focus might be the fault scarp(s), but also incision. As the reviewer was asking for topographic profiles, such complementary profiles, derived from the DEM next to the trenches and annotated to highlight the trench dimensions and vertical separation at each site, would help the reader to better acknowledge the geomorphology of the sites. Additional comments on Fig. 4: Please provide orientation at field photos; consider moving the scale bar in A to the bottom so you can leave the coordinates in place. Indicate position and viewing direction of field photos on the overview map.
(6) “collapsed free-face” vs. “colluvial wedge”: I understand the difference from the description, but find this terminology confusing. An earthquake-driven colluvial wedge also at least partly derives from collapse of the free face (until the angle of repose is established and the diffusive processes take over). Maybe, you can call it “nontectonic wedge” or similar?
(7) Page 17, Line 1: Typo! Slip should be 1.8 m instead of 10.8 m – also, I did not quite follow the calculation. E1 –slip (based on B1 event) was measured as dip-slip between offset markers, so I guess, you assumed a vertical fault dip to translate the 10 cm of vertical slip for E1 that you added to the 0.8 m of B2 wedge height? Then, this amount (0.1 m for E1) should not be doubled for the second scenario, adding to 1.7 m only instead of 1.8? – Please clarify.
(8) In general, did you account for the listric shape and fault-dips of about 70° in places when calculating magnitudes from slip? I see that you consider vertical slip only, so the actual dip-slip might be slightly higher.
(9) Figures 1,3, and 4: give data source for the DEM

Referee Report: PDF

Hide

RR by Stéphane Baize (24 Nov 2017)

Suggestions for revision or reasons for rejection

Review on « Implications from palaeoseismological investigations at the Markgrafneusiedl Fault (Vienna Basin, Austria) for seismic hazard assessment » by Esther Hintersberger et al.

Stéphane Baize
Stephane.baize@irsn.fr

The authors present a new paleoseismological study on a normal fault, a splay in the releasing bend of the major Vienna Basin Transfer Fault System, Austria.
From morphotectonic and geological maps, detailed analysis of two trenches, one outcrop and a series of luminescence dating, they conclude with the occurrence of 5-6 paleoearthquakes in the last 120 ka. They deduce from those data two models of earthquake distribution in time, one with periodic events and the other with clustered events. Based on regional tectonics and seismic lines information, they also derive some information about the maximum credible earthquakes of the region.
The study is of particular importance for improving the seismic hazard evaluation of the region, especially when considered the short distance of this fault to the Austria capital.
The data are of excellent quality, the analyses reliable and the interpretation relevant. For those reasons, the paper is suited to NHESS and could be published with few minor corrections (see attached pdf).

I however underline three points that should be discussed/solved before publication:

Dealing with the section dedicated to discussion on the maximum credible earthquake.
The contribution of the authors is actually significant; “Mmax” is a mandatory input parameter of seismic hazard assessments. To avoid any misusage of such an M7 the authors should warn end-users that estimated M7 is most likely associated with a very long return period which will need to be correctly estimated if used in PSHA. On the other hand, if the authors suggest to consider M7 in deterministic seismic hazard analysis they should stress the high degree of uncertainty attached to this M7 estimate.

“The possible correlations of paleoearthquakes between the trenches allow for two different interpretations to assess the recurrence intervals of earthquakes”, leading to two alternative models of time distribution of events (periodic and clustered). I would suggest to moderate the related conclusions. The proposition of two models is to enhance the gathered data value for PSHA purposes, but I think deriving clustered or periodic models of a fault (system) from two trenches -with IRSL datings and with a not-so-finely bedded stratigraphy- is somewhat pushed a bit far. Applying time-dependent earthquake rupture forecast models in seismic hazard analyses requires many data to constrain the various parameters.

A more concerning point is the following. The interpretation of event B4 in trench SDF3 based on the W4 layer needs to be justified. From your logs and description, this gravel layer with its overlying silts look like a graded fluvial sedimentary pack, with homogenous thickness all along the trench. It does not look like a wedge associated with the degradation of the fault free face just after the earthquake. This discussion is important because it could question the number of events in one of the two trenches and potentially impact the discussion of the time distribution of earthquakes.

Referee Report: PDF

Hide

ED: Publish subject to minor revisions (review by editor) (27 Nov 2017) by Oona Scotti

Dear Esther,

I read the 5 reviews and the revised manuscript. I think that you have addressed most of the issues raised in a comprehensive manner and documented these changes accordingly. However, there are still minor comments left unanswered from the first round of review:
comments left unanswered from previous reviews
- “I encourage the authors to calculate the coefficient of variation (COV) for their two earthquake models (“event lines 1 and 2”), as a way to statistically distinguish/represent periodic vs clustered earthquake behaviour”
- “Do the authors think that the MF is an independent seismic source, or might earthquakes on the MF be correlated to larger strike-slip faulting events on the VBTF”
- “I wonder if the seismogenic events seen here are related with the unloading after the Glaciers retreated in repeated glaciations, leading to “pulses” of enhanced activity
and additional minor comments from the two new reviews that I believe you can handle following minor revisions.

So please, revise accordingly.

I have an additional technical question: As an end-user of paleoseismological data I would like to know where do you measure the "maximum" height of a colluvial wedge? Could you add it on the figures?

Editorial comments:
Acknowledgments: Please refer to the reviewers with their names; they have all declared their identity
Figure 3C should arrive after Figure 3A
Figure 6B should arrive after Figure 6A
There are two Table 1
“The trench SDF1 exposed about 30 m of Gaenserndorf terrace deposits in the footwall and ca. 10 m of mostly fine-graded sandy to silty sediments the hanging wall,”Insert: The trench SDF1 exposed about 30 m of Gaenserndorf terrace deposits in the footwall and ca. 10 m of mostly fine-graded sandy to silty sediments in the hanging wall,
Please add: “Detailed code information is given in the supplementary material”
Please replace: Photomosaique with Photomosaic
Please revise: “ See supplementary” material” for high-resolution photomosaique

Hide

AR by Esther Hintersberger on behalf of the Authors (24 Dec 2017) Author's response Manuscript

ED: Publish subject to technical corrections (04 Jan 2018) by Oona Scotti

AR by Esther Hintersberger on behalf of the Authors (13 Jan 2018) Manuscript

Download

Article (44531 KB)
Full-text XML

Short summary

The Vienna Basin is a low seismicity area, where historical data do not identify all potential earthquake sources. Despite observed Quaternary offset, there are no earthquakes along the Markgrafneusiedl Fault (MF). Results from 3 palaeoseismic trenches show evidence for 5–6 earthquakes with magnitudes up to M = 6.8 during the last 120 kyr. Therefore the MF should be considered as a seismic source, together with similar faults in the Vienna Basin, increasing the seismic potential close to Vienna.