Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland)

Quaternary International xxx (2017) 1e13

Contents lists available at ScienceDirect

Quaternary International journal homepage: www.elsevier.com/locate/quaint

Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland) S.C. Fabbri a, *, M. Herwegh a, H. Horstmeyer b, M. Hilbe c, C. Hübscher d, K. Merz b, F. Schlunegger a, C. Schmelzbach b, B. Weiss d, F.S. Anselmetti c a

Institute of Geological Sciences, Baltzerstrasse 1þ3, 3012, Bern, Switzerland Institute of Geophysics, Dept. of Earth Sciences, Sonneggstr. 5, ETH Zürich, CH-8092, Zürich, Switzerland Institute of Geological Sciences, Oeschger Centre of Climate Change Research, University of Bern, Baltzerstr. 1þ3, 3012, Bern, Switzerland d Institute of Geophysics, Center for Earth System Research and Sustainability, University of Hamburg, Bundesstr. 55, D-20146 Hamburg, Germany b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 31 March 2016 Received in revised form 13 January 2017 Accepted 25 January 2017 Available online xxx

In the vicinity of Lake Thun at the front of the Bernese Alps (Switzerland), we performed a multidisciplinary neotectonic study combining onshore and offshore geological data and geophysical measurements in order to identify potentially active fault structures. Paleoseismic reconstructions on the northern margin of the Alps have documented several strong earthquakes with moment magnitudes 6 during the Late Quaternary, which have long recurrence intervals of 1,000 to 2,000 years. Such earthquakes are expected to produce surface ruptures. In this light, we investigated the study area located near Lake Thun primarily for on-fault evidence, to date still a shortcoming in Switzerland. We detected several features indicating potential fault activity, such as aligned subaquatic morphological depressions, offset horizons observed in reflection seismic profiles of lake sediments and in ground-penetrating radar images, all delineating a fault trace. Observations of fluvial deposits in a nearby gravel pit in the prolongation of the inferred structure supports these findings. A narrow zone with rotated long axes of pebbles (inclining at ~60 ) is clearly distinguishable and crosscuts the original bedding with predominantly horizontal orientation of pebble axes. This zone further shows an apparent 1.1 m offset of oxidized horizons and is therefore considered as a potential fault plane in a normal faulting regime. A dated radiocarbon age of ~11,000 years BP of the gravel deposits hence suggests a younger fault activity during the Holocene. The Einigen Fault Zone (EFZ), proposed on the basis of these observations, is considered as a complex fault system with a combination of dextral strike-slip and normal faulting, as suggested by GPR images. Observations in the gravel pit and radar data independently show that it includes at least two fault strands. However, while five earthquakes with epicentral intensities I0  VI and numerous smaller seismic events are known within less than 30 km epicentral distance to Lake Thun over the past 400 years, none of these seem to coincide with the location of the EFZ. © 2017 Published by Elsevier Ltd.

Keywords: Seismicity Einigen Fault Zone On-fault evidence Paleoseismology Switzerland

1. Introduction Current probabilistic seismic hazard assessments in Switzerland are mainly based on historically documented and instrumentally recorded earthquakes (Wiemer et al., 2009). The €h et al., earthquake catalogue of Switzerland (e.g. ECOS-09, Fa 2011) builds such a basis for the hazard assessment, covering a

* Corresponding author. E-mail address: [email protected] (S.C. Fabbri).

time span of roughly 1,000 years with ~20,000 events (including eight historic earthquakes with moment magnitudes of 5.7e6.2 along the Swiss Alps). However, a compilation of paleoseismic data based mainly on earthquake-triggered mass movements in several lakes along the northern margin of the Central Swiss Alps shows that strong earthquakes, with local intensities I > VI - VII, occur on fairly regular recurrence periods of ~1,000e2,000 years (Strasser et al., 2013). The strongest of these seismic events have reached reconstructed moment magnitudes of 6.2e6.7 (Strasser et al., 2006, 2013). These recurrence intervals are beyond the

http://dx.doi.org/10.1016/j.quaint.2017.01.033 1040-6182/© 2017 Published by Elsevier Ltd.

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

2

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

time span of the earthquake catalogue of Switzerland (ECOS). To quantify the hazard of future earthquakes and their associated damage, in particular that of rare and strong events, a thorough analysis of geological evidence for past events is essential. This study aims to find such evidence related to recently active tectonic fault structures in a terrestrial and lacustrine setting in the vicinity of Lake Thun (Switzerland). Apart from its tectonic setting with strong contrasts from the north to the south of Lake Thun and several geological and morphological indications suggesting neotectonic activity, the area was exposed to five moderate earthquakes in the last three centuries (moment magnitudes 4.7e5.2). These characteristics make the area likely to have experienced even stronger earthquakes in the past, so that it is ideal to investigate recent deformation caused by neotectonics. Earthquake-triggered subaquatic mass movements, turbidites and small-scale in situ deformation features (e.g. liquefaction structures, microfaults, mushroom-like intrusions, etc.) in lake sediments are considered as significant off-fault paleoseismic evidence and have been used to document prehistoric earthquakes throughout the Holocene and Late Pleistocene period, reflecting recent tectonic activity in the Alpine region (Monecke et al., 2004; Schnellmann et al., 2006; Strasser et al., 2011). The strongest of these events (moment magnitude Mw > 6) are expected to produce surface ruptures due to the size and displacement of the slipping surface (Wells and Coppersmith, 1994; Stirling et al., 2002). Considering the present off-fault paleoseismic evidence in the Alpine region, primary on-fault evidence should be present as well. However, seismogenic fault structures with clear surface ruptures are scarcely found in Switzerland (Madritsch et al., 2010). To date, only few on-fault studies identified active fault structures and neotectonic features in Switzerland in order to improve our understanding of the hazard imposed by strong earthquakes. Maurer and Deichmann (1995) showed in their investigation of two earthquake clusters in the western Swiss Alps that presently active fault planes in the Helvetic nappes belong to the reactivation of Riedel shears of a large-scale dextral strike-slip structure (Pavoni, 1980). Ustaszewski and Pfiffner (2008) investigated neotectonic faulting in the western and central Swiss Alps. They identified two tectonic faults based on mapping lineaments on aerial photographs and subsequent field studies characterizing the structures using displaced post-glacial landforms or sedimentary infills. We also follow their definition of the time frame concerning ‘neotectonics’ that includes all fault activity since the Last Glacial Maximum (LGM) 23 ka ago (Wirsig et al., 2016, and references therein). A similar approach was chosen in the eastern Swiss Alps by Persaud and Pfiffner (2004), revealing the difficulty of assigning observed lineaments and active faults to its physical source, which can be of tectonic nature, surface uplift due to post-glacial rebound or a combination of the two. Meghraoui et al. (2001) identified an active normal fault near the eastern edge of the Upper Rheingraben based on geomorphological and geophysical analysis, supplemented by six trenches at two different sites, and they attribute two prehistoric events as well as the 1356 Basel earthquake (Mw ¼ 6.7, ECOS) to this fault. Ustaszewski et al. (2007) documented an active fault located in the western Swiss Alps. The fault was formed during Alpine nappe emplacement and cataclasite formation as well as numerous periods of fluid flow dated between 0.5 Ma and 2.5 Ma derived from combined 230Th/234U and 234U/238U ratios. They observed several stages of reactivation with its most recent activity in Late Holocene times indicated by displaced and OSL dated late-glacial loess and slope-wash deposits. For a hydrothermally active strike-slip fault at Grimsel pass, Belgrano et al.

(2016) suggest a close interplay between recent faulting, seismic activity and fluid flux of meteoric waters down to depths of at least 4e5 km. Here a complex geometric system consisting of major fault zones and associated fault linkages in Riedel orientations accommodates young strike-slip movements. In an instrumentally based approach for the identification of an active fault structure, Kastrup et al. (2007) related the temporal clustering of a series of seismic events in the vicinity of Fribourg, in the Molasse Basin of western Switzerland, to a large-scale, N-S striking fault. They concluded that the N-S striking epicenter alignment, similarity in deformational style and orientation of faults mapped from seismic reflection data around Fribourg fit well with the fault characteristics of the crystalline basement and its overlying Molasse sediments in the larger region. Using crosscorrelation based relocation of epicenters, seismic reflection data and magnetic data, they consider the Fribourg fault as being active. In France, de La Taille et al. (2015) recognized Riedel-like fault structures in Lake Le Bourget affecting Holocene sediments and lake-floor morphology. They interpreted the structures as imprints of two known strike-slip faults crossing the lake basin. Irrespective of these examples, the prevalent lack of on-fault evidence is attributed primarily to high erosion rates in the Alpine region and to pervasive anthropogenic landscape modification, as suggested by Ustaszewski et al. (2007). They note that the scarceness of precisely dateable Quaternary deposits imposes an additional challenge when it comes to the exact timing of fault movements. Furthermore, temporal and spatial clustering of seismic events leads to variable recurrence intervals, sometimes accompanied by long periods of quiescence, which complicates the recognition of mid- to long-term occurrence patterns (Michetti et al., 2005). This study presents a potentially active fault structure near Einigen on the southern shore of Lake Thun at the front of the Bernese Alps. Our multidisciplinary approach combines amphibious geomorphology including terrestrial and subaquatic topography data with subsurface geological and geophysical data. An existing digital elevation model derived from airborne laser scanning (swissALTI3D, swisstopo) and a newly acquired highresolution bathymetric data set of Lake Thun are combined in an effort to identify topographic features potentially associated with fault traces. Including the lake floor has the advantage that the diffusive impact of erosive processes on the landscape is much smaller, as underwater processes occur at rates that are much slower than on land (Sturm and Matter, 1972, 1978). In a second step, we complement the geomorphologic data with subsurface data using ground-penetrating radar (GPR) on land and seismic reflection data in the lake. In addition, 'trench-style' outcrops in a gravel pit are used to further refine and quantify a potential fault zone through mapping displaced layers and rotated clasts. The combination of these various data sets provides a series of evidence pointing towards a recently active fault and thus supplements the thin catalog of known Quaternary fault structures supported by onfault evidence. These findings increase the awareness for seismic hazards in the Bernese Alps and the approach builds a possible recipe to discover other potentially active fault traces at the forefront of the Swiss Alps. 2. Geologic and seismotectonic setting Perialpine Lake Thun (Fig. 1) is situated in the upper Aare valley between Interlaken and Bern, and is located at the northern front of the Alpine nappes. The overdeepened basin of Lake Thun is surrounded by the Penninic and Helvetic thrust nappes and the Subalpine Molasse. The basin is elongated orthogonal to the general strike direction of the Alpine front

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

3

Fig. 1. Hill-shade of digital elevation model (swissALTI3D from swisstopo) superimposed with colored and shaded bathymetric map. Red circles: earthquake locations taken from ECOS-09. Orange and green circles: earthquakes recorded in the period 01.01.2009 until 31.01.2016 taken from WebDC3 (Web Interface to SED Waveform and Event Archives). Green circles: earthquake swarm at Diemtigen with 86 events ML  1.0 shown of total 306 events. The strongest historically documented earthquake struck the area in 1729 CE with Mw ¼ 5.2 and I0 ¼ VI. Inset: Overview of Switzerland showing principal tectonic units (J: Jura mountains; M: Molasse basin, H: Helvetic nappes, P: Penninic nappes, C: Crystalline basement, A: Austroalpine nappes). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

(WSWeENE). The northeastern shore of the lake lies predominantly in the tectonic units of the Subalpine Molasse and the Helvetic nappes. This situation is in strong contrast to the southwestern shore, on which mostly structurally higher Penninic units are outcropping (Fig. 1). These obvious differences on the opposite lake shores suggest a major tectonic boundary along the lake axis and high tectonic activity during nappe emplacement, i.e. from Eocene times throughout the Late Miocene (Pfiffner, 2011). During the Quaternary, several glacial advances coupled with the erosive power of subglacial meltwater created overdeepenings and shaped the study area (e.g. Reber and Schlunegger, 2016). Several hundred meters of ice (top of ice at 900e1400 m a.s.l) covered the upper Aare valley between Bern and Interlaken ~23 ka ago during the LGM (Bini et al., 2009). While on the steep slopes to the north and northeast of Lake Thun, prominent depositional glacial features from the last glacial period are rare, many of them are present on the fairly flat landscape to the south and southwest. A number of relatively continuous moraine crests running parallel to the lake axis populate the southern shoreline near the confluence of the Simme and Kander Rivers. These could potentially serve as neotectonic markers. However, as the area of interest was characterized by the confluence of several glaciers (Aare-, Kander-, Simme-, Lütschine- as well as smaller local glaciers), depositing moraines at different stages (Hantke and Wagner, 2005), one single offset moraine crest can only serve as marker if a depositional or glaciotectonic origin

during the different stages of glacial advance and retreat can be ruled out. Focal mechanisms close to Fribourg, some 40 km to the northwest of Lake Thun, show that the Alpine foreland today is dominated by a strike-slip stress regime with a slight normal faulting component, while the Helvetic nappes at the front of the Alps are characterized by strike-slip faulting to thrusting with Paxes striking NW - SE (Kastrup et al., 2004; Singer et al., 2014). 2.1. Local seismicity During the last 300 years, the area around Lake Thun has experienced six earthquakes with epicentral intensities I0 ranging €h et al., 2011). They all occurred within from VI to VII (ECOS-09, Fa epicentral distance to Lake Thun of less than 30 km. The strongest historically known earthquake with an estimated moment magnitude of Mw ¼ 5.2 is documented to have hit the area of Frutigen in 1729 CE (Fig. 1). The seismogenic source occurred at a depth of 12 km with an epicentral intensity of VI (Mw error estimate  0.5; location error  50 km). A recently recorded earthquake swarm (‘Diemtigen swarm’), starting in April 2014 and reaching more than 270 events by the end of 2014, reveals details on the seismic activity of the region (Diehl et al., 2015). In total, 306 events were localized until the end of January 2016, and 86 events show local magnitudes of ML  1.0. Focal mechanisms of the strongest events (May 10th 2014, ML 2.7; October 15th, 2014, ML

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

4

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

3.2) indicate a combination of thrusting and sinistral strike-slip faulting with a ENE dipping fault plane, i.e. with a strike roughly parallel to the lake axis. Diehl et al. (2015) note that a relation between the Frutigen earthquake and the swarm activity at Diemtigen might be possible, considering the proximity of the events (~4 km) and the large location uncertainty of the historic earthquake. 3. Methods 3.1. Multibeam bathymetry High-resolution bathymetry data of Lake Thun (Fig. 1) were acquired during 14 days in September and October 2014 using a Kongsberg EM2040 multibeam echo sounder. The positioning system used for the bathymetric campaign was a Leica GX1230 GNSS receiver in combination with the swipos GIS/GEO real-time kinematic positioning service provided by swisstopo. Survey lines were oriented mostly parallel to the lake's long axis, while areas of shallow water depth (<15 m) were surveyed in a shoreparallel pattern. Bathymetric data cover most of the lake basin from the deepest areas up to 5 m water depth. The processed point cloud has been rasterized for geomorphologic analysis of the lake floor. The resulting digital bathymetric map has a cell size of 1 m with a vertical accuracy in the order of a few decimeters. Hilbe et al. (2011) and Hilbe and Anselmetti (2014) provide a more detailed review on the acquisition procedure, equipment setup and the use of swath bathymetry tools in lake research. The lake floor was screened for topographic offsets, linear features and morphologic depressions using mostly shaded relief maps and slope gradient maps. 3.2. Terrestrial geomorphology The digital elevation model (swissALTI3D, swisstopo) with a raster size of 2 m was scanned by eye for linear features indicative for recent fault activity, using shaded relief images along the Aare valley between Bern and Interlaken. Such conspicuous geomorphic and hydrological features can be for instance depressions aligned with adjacent linear mounds, anomalously deep hillside troughs with little or no contributing drainage area, closed depressions, linear swales, offsets of fluvial terrace risers or moraine crests, deflections of creeks and river courses, as well as springs and gas seeps (e.g. Hunter et al., 2011). 3.3. Lacustrine reflection seismic survey We conducted a multi-channel reflection seismic survey on Lake Thun. We used a Sercel two-chamber Mini GI airgun (15/15 in3) in combination with a Geometrics MicroEel streamer of 97 m length and 24 channels (4 m channel distance with 3 hydrophones per channel) as acquisition tools and recorded 42 lines with over 180 km total length. Previous seismic surveys on Lake Thun (e.g. Finckh et al., 1984) used airguns and a multi-channel setup as well, however, they did not achieve the resolution reached in this campaign or suffered from limited penetration depth as in a singlechannel pinger (3.5 kHz) seismic campaign carried out in 2007 (Wirth et al., 2011). The setup used in this survey overcomes the drawbacks of limited resolution and penetration depth, using stateof-the-art equipment and processing work flow. Shot time interval was 10 s at 6.6 km/h survey speed, resulting in an average shot spacing of 18 m and an assigned common-mid point (CMP) interval of 2 m, corresponding to half the receiver spacing. The frequency spectra show reasonable signal power up to 500 Hz, resulting in a vertical resolution of the topmost stratigraphic sequences of l/

4 ¼ 0.7 m. An average theoretical vertical resolution of l/4 ¼ 2.5 m is obtained, when using a velocity of 1500 m/s and a main frequency of 140 Hz (Widess, 1973; Chopra et al., 2006). The position of the survey vessel was tracked with a Garmin GPSmap76Cx GPS receiver recording one point every 2 s with 2e5 m accuracy. Since the survey velocity was kept constant during acquisition, the easting and northing positions were individually smoothed and interpolated to 1 s. The smoothing was done with a local regression using weighted linear least square and a first degree polynomial model. The following processing steps were applied to the seismic data: frequency bandpass filtering, muting, velocity-model creation based on normal-move-out (NMO) analysis and recorded soundvelocity profiles from bathymetric survey, multiple suppression using surface-related multiple elimination (SRME; Verschuur et al., 1992), NMO corrections and CMP stacking, post-stack FX-deconvolution, and post-stack Kirchhoff depth migration. The bedrock topography including potential offsets was of primary interest. Major challenges were encountered close to the shoreline, where water depth is only a few tens of meters and surface- related multiples were difficult to remove. In this context, the SRME algorithm revealed its limitations. Due to the rather short streamer length, the CMP-fold was sometimes as low as 2 or 3. A longer streamer with the same receiver spacing or a shorter shot interval would have led to a higher fold. 3.4. Ground-penetrating radar To clarify the existence and trace of a potential fault, we collected a series of 2D ground-penetrating radar lines close to Lake Thun in December 2015 (Fig. 2). We tested different antenna frequencies of 50, 100 and 250 MHz, with the 100 MHz antenna mounted on a sledge and continuous recording leading to the best trade-off between horizontal/vertical resolution and depth penetration. We used a 1 m offset between the two 100 MHz antennae. Most of the profiles were tracked with a Leica GX1230 GNSS receiver in real time, with the GPS antenna placed on the sledge. For the processing, we used a constant trace bin size of 20 cm and applied the following processing steps (Gross et al., 2004; Schmelzbach et al., 2011): ‘dewow’, removal of systematic horizontal noise from recording devices and cables using eigenvector filtering, amplitude scaling (removing the averaged amplitude envelope and increasing the amplitude of later arrivals with respect to earlier ones), frequency bandpass filtering, FX-deconvolution, and phase-shift migration. The migration velocities were based on a common-mid-point measurement with an initial offset of 1 m and total distance increase per point of 0.2 m. Calculated interval velocities range from 0.08 to 0.124 m ns1. Power lines running parallel to a nearby highway created strong reflection hyperbola, since the antennae were unshielded and rendered some of the profiles unusable due to spurious reflections. 3.5. Outcrop data Field studies on outcrops of mainly Quaternary deposits were focused on areas where conspicuous topographic features were recognized in the amphibious digital elevation model (combined terrestrial digital elevation model SwissALTI3D and bathymetric elevation model). Special attention was given to outcrops and gravel pits in the vicinity of nearby linear features. Two pits were investigated for the occurrence of typical fault indications such as rotated pebbles or offset horizons indicating fault traces. Shearing on a fault may be indicated by rotation of clast long axes parallel to the fault plane, often termed as “shear fabric” (McCalpin, 2009a). Especially in normal faulting regimes, clast's long axes may be

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

5

Fig. 2. Amphibious geomorphology showing main features of the Einigen Fault Zone (EFZ). Extent of map is given in Fig. 1. Illumination angle is: 315/45, vertical exaggeration is ¼ 2 for bathymetry. Bottom: Moraines and gypsum sinkholes are taken from swisstopo's GeoCover dataset (Swiss Federal Office of Topography, swisstopo, Wabern, Switzerland). The orange star marks the location of the main fault strand in the GPR profile. The white star denotes the main fault strand encountered in seismic line 40. Extent of Fig. 4C shows the location of the gravel pit Gesige. A: subaquatic channel, B: channel-levee, C: gentle valley, D: alignment of subaquatic depressions. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

aligned parallel to the fault dip, i.e. at 50 e90 inclination (e.g. Olig et al., 1994). However, special care is required in glacially overprinted deposits, since cryoturbation and subglacial deformation can lead to very similar expressions.

Radiocarbon dating of one wood fragment found in a gravel pit, was carried out at the Laboratory for the Analysis of Radiocarbon with AMS, at University of Bern, using a Mini Carbon Dating System (MICADAS), developed by ETH Zurich.

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

6

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

4. Results 4.1. Geomorphologic data: combined terrestrial and lacustrine digital elevation model Numerous features potentially indicative for neotectonic activity can be found in the area between Interlaken and Bern as single occurrences. The WNW-ESE alignment of gas releasing pockmarks was observed to the south of Einigen and close to the western shoreline of Lake Thun. Similar activities were noticed to the southwest of Interlaken at the shallow subaquatic platform, coinciding with the location of three closely-spaced small earthquakes within the lake (Fig. 1). Another example is the prominent deflection of a 150 years old abandoned river channel, as well as the occurrence of several lineaments and landslides on opposite sides of the Gürbe valley flanks. However, only one location near the village of Einigen shows multiple different features providing high likelihood for a neotectonic structure (Fig. 2). The bathymetric map shows an alignment of 6 circular morphologic depressions, with the largest being 45 m in diameter and 2e6.5 m in depth measured to the lowest and highest point of the rim, respectively (Figs. 2 and 3). Some of these depressions release water and gas (pers. comm. V. Maurer, ‘Amt für Wasser und Abfall’, Canton of Bern). Adjacent to the north of these depressions, prominent traces of a subaquatic slide can be identified, with a sharp, nearly semi-circular head scarp on the slope and a depositional lobe of mass-movement deposits with a hummocky surface morphology in the flat basin area below. The toe of slope, which runs approximately along a straight line in the area and is highlighted by a dashed black line in Fig. 2, is offset by ~150 m at the location of the mass movement. In Fig. 2, it can be further observed: A - (in the lake), a not very prominent, shallow channel extending from the toe of the slope (which is offset by 40 m) to the deeper areas of the basin. It is outlined by orange lines in Fig. 2 and 50e80 m wide and ~300 m long. The channel is located in the eastern continuation of B and C. B - (in the lake), a very prominent, deeply incised channel-levee structure that runs from the shore to the toe of slope. It starts on the shore near the mouth of a small creek and near the former water outlet of a hydropower plant. It is most probably related to the latter. C - (on land), a gentle valley cutting across the complex of shoreparallel moraine ridges, marked by linear swales and descending to the lakeshore. D - (in the lake), aligned depressions located in the eastern continuation of C.

However, the recognition of topographic offsets, linear features and morphologic depressions on the lake floor is complicated due to the artificial deviation of the River Kander from its original course directly into Lake Thun 300 years ago. The sudden large volumes of sediment input and its associated intense river-bed erosion (Wirth et al., 2011) have the potential to level out previously visible offsets. The impact of strong erosion on the river bed is recognizable in the very west of Fig. 2. The Kander River exhibits particular unstable slopes to the east and west of the river at about the same latitude as the gentle valley on land (C). 4.2. Gravel pit A gravel pit (gravel pit of ‘Gesige’), lies in the direct prolongation of the morphologically distinct valley crosscutting several moraine crests (Fig. 2). The gravel pit walls expose well-graded sandy gravel. Judging from its unconsolidated state and from the fact that no crushed pebbles with grain-to-grain contacts occur, the sediment has never been glacially overprinted. The facies of the deposits shows a typical fluvial character of a braided river system with the long axes of the pebbles oriented horizontal or nearly horizontal (Boggs, 2009). In distinct zones on the gravel pit walls (indicated by white arrows in Fig. 4A, B), a significant deviation from this overall flat-lying orientation with strongly inclined pebbles is observed. In these ~0.5 m-wide zones, the apparent dips of the longitudinal axes of clasts reveal a median absolute deviation of 61.5 ± 10.5 (50 counts). Outside this zone, apparent dips amount to 13.8 ± 5.8 (85 counts), which is in the range of expected values for a braided river system (10 e30 , rarely up to 40 ) with imbricated pebbles (Pettijohn, 1930; Boggs, 2009). The steeply dipping gravel clasts persist beyond the excavated surface and reach shallower deposits. Furthermore, two slightly oxidized, decimeter-thick sedimentary layers, recognizable by their reddish color (Fig. 4), can be traced along the gravel pit walls, but they are vertically offset by ~1.1 m across the zone where tilted gravels occur. The age of one of these oxidized layers is constrained by one 14C date (BE no. 3288.1.1, Einigen01) of a tree trunk to a 2-sigma range of 10,772e11,102 cal years BP. The geometry of this conspicuous zone was tracked with a differential RTK-GPS system (Leica GX1230 GNSS) from March 2015 till April 2016 as excavation of the gravel pit progressed. Exact 3Dpositions of zones with rotated pebbles, or with significantly higher erosion on the overall stable gravel pit walls were measured and mapped. These measurements were given variable “confidence values” on the basis of detection quality: a clearly visible series of inclined pebbles was rated with very high confidence, while a

Fig. 3. Subaquatic morphologic depressions (pockmarks) between 15 and 50 m water depth (see Fig. 2 for location). The largest pockmark is 6.5 m deep measured from the western rim to the center. Water and gas release within the depressions is well documented by divers. Small arrows indicate tiny pockmarks measuring just a few meters across. The escarpment marked with a ticked line is part of a 340 m-long headwall escarpment of a possible mass movement extending further south.

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

location with only slight back erosion without rotated clasts at all was assigned a low confidence. The inset in Fig. 4C shows 9 locations with intermediate to very high confidence. The points are color-coded according to their elevation. The local distribution of the findings indicates that there are at least two individual, roughly planar zones striking NW-SE. 4.3. Geophysical data 4.3.1. Reflection seismic data Two reflection seismic profiles cross the conspicuous zone close to the shoreline (Figs. 5 and 6; see Fig. 2 for location of seismic lines 35 and 40). A simplified seismic stratigraphy (Figs. 5 and 6) shows, from bottom to top, a succession of bedrock (seismic sequence S4; high-amplitude and laterally continuous reflections), glacial till (S3; high amplitudes, slightly chaotic facies with partially intact layering), glacio-lacustrine and lacustrine deposits, accumulated during the Late Glacial (S2; horizontal layering with low amplitudes) and Holocene sediments (S1; high-amplitude and laterally continuous reflections). Seismic line 35 (Fig. 5) crosses the subaquatic channel (outlined by orange lines in Fig. 2) in an area where the bedrock reflections show considerably lower amplitude and lose lateral continuity at 280e300 m depth below lake level. The bedrock

7

surface has a step-like character showing an offset of several meters at 300 m depth affecting glacial deposits (S3). Directly above, the lacustrine and Holocene sediments are offset as well by 1.5e2 m. Line 40 was recorded closer to the shoreline and hence encounters bedrock at significantly shallower depth (Fig. 6). Also, the sedimentary succession is thinner and it is more difficult to recognize all seismic sequences defined in line 35. Glacial- (S3) and glacio-lacustrine (higher-amplitude facies at the base of S2) deposits seem to be missing or too thin to be clearly resolved (Fig. 6). The top of the bedrock is well distinguishable and seems to be offset at several locations. The most prominent disruption (at ~800 m in Fig. 6) is close to the alignment of the morphologic subaquatic depressions and reveals an apparent offset of a few meters, which we associate with the suspicious Einigen zone. The bedrock surface in this area seems to have a pronounced relief, which appears to influence also the topography of the overlying Holocene sediments. 4.3.2. Ground- penetrating radar (GPR) The GPR survey covers an area of about 65,000 m2 on the northeastern margin of the alluvial plain of the former River Kander adjacent to the Gesige gravel pit (Fig. 2), imaging thus deposits similar to those encountered in the gravel pit. Sediment thicknesses above the bedrock, estimated from borehole data, range between ~25 m close to the gravel pit and ~5 m when approaching the gentle valley (C in Fig. 2) towards Lake Thun (‘Amt für Geoinformation’ and ‘Amt für Wasser und Abfall’, Canton of Bern). In the area with relatively thin sediment cover to the east, the bedrock surface can be imaged by GPR. A line crossing the conspicuous zone in proximity to the valley (Fig. 7; Fig. 2 for location) shows that the bedrock surface, clearly distinguishable with its high-amplitude reflection, has a pronounced topography, which is offset at several locations. Above this step-like bedrock surface, reflections of the sediment coverage and overlying deposits seem to be truncated as well. 5. Discussion 5.1. Evidence for neotectonic faulting

Fig. 4. A: Outcrop photographs from the gravel pit of Gesige. White arrows indicate EFZ and alignment of rotated pebbles observable in the front as well as in the background of the picture. B: close-up view of A with offset oxidized layer. Black horizon is displayed for orientation purposes. White arrows mark steeply dipping gravel clasts. A and B were taken on the 20.10.2014. C was taken on the 05.04.2015, after excavation progressed. The tree trunk in C to the left within the oxidized layer was 14C dated with a 2-sigma range of 10,772e11,102 cal years BP. The apparent offset amounts to 1.1 m. Inset in C: D-GPS points when dipping clasts were encountered. Error of location is about 100 times smaller than point size. The color gives relative elevation of measured points (yellow: 597.9 m a.s.l.; red: 608.0 m a.s.l.). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The combination of terrestrial and subaquatic geomorphology with subsurface geological and geophysical data provides evidence related to a potentially active fault structure in the vicinity of Lake Thun. In the following, all observed evidence is listed from west to east along the E-W striking conspicuous zone. In the Gesige gravel pit, the average inclination of long clast axes of ~62 in discrete, narrow zones and oxidized horizons offset across these zones indicate a vertical displacement that we interpret as a strong argument for a recently active fault structure. The tilted pebbles evidence the existence of at least two NW-SE striking fault strands, as indicated by the inferred geometry from GPS positions of tilted clasts (Fig. 4C inset). Two 1.1 m vertically shifted horizons depict a clear offset coinciding with the location of the tilted pebbles. The fault zone in the gravel pit is unstable, erodes faster and is more sensitive to weathering as indicated by the fresh cone of debris at the base of the cliff directly below the fault zone (Fig. 4C). Assuming that the inclination of the shear fabric seen in the rotated clasts marks the minimal dip angle of the fault plane, the dip of ~62 fits well into the expected range for normal faults of 50 e90 (McCalpin, 2009b). We cannot determine whether the 1.25 m slip on the presumed fault surface (corresponding to a 1.1 m vertical offset) was created by a single displacement or by multiple events. A single event producing such a displacement would require a magnitude of ~6.7 (Wells and Coppersmith, 1994). Moreover, the fault zone might be characterized by several strands as seen in GPR data, which,

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

8

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

together with the uncertainty of number of slip events, do not allow a paleomagnitude estimation. Ground-penetrating radar data reveal a disrupted bedrock topography that furthermore affects overlying sediments. A clearly offset bedrock horizon at several locations supports the interpretation of a complex fault zone, formed by several strands. The most

prominent strand marked with an orange star in Fig. 7 is also visible on adjacent lines. It matches the trend and location of the zone encountered in the gravel pit, running through the valley bottom (orange star in Fig. 2), and is interpreted as the main fault strand. An entire set of normal and reverse faults surround the main fault in a 150-m-wide area (Fig. 7). The main fault strand and a second

Fig. 5. Seismic reflection profile 35, 3 vertically exaggerated. Bottom: Seismic sequence stratigraphic interpretation: S0 ¼ water column, S1¼Holocene sediments, S2 ¼ lacustrine and glacio-lacustrine deposits, S3 ¼ glacial till, S4 ¼ bedrock. The location of the seismic sections is given in Fig. 2. The orange arrows depict the borders of the subaquatic valley constraint by orange lines in Fig. 2. The offset affecting Holocene sediments amounts to 1.5e2 m. Overall deformation is associated with a normal faulting regime. Local amplitude anomalies in S2 are interpreted as likely related to free gas and the presence of fault systems. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

prominent strand farther to the SSE reveal reverse faulting. To the NNW, one shorter strand with an apparent offset of 1.2 m shows normal faulting. The three offsets affecting the shallowest deposits at 1e2 m below the surface are dominated by apparent dips of ~20 . This low value can probably be attributed to the low angle between the GPR line and the fault. An alternative explanation for the offset horizons observed in the gravel pit and the GPR profiles are karstic processes, as sinkholes, which originate from dissolution of the gypsum in the Penninic nappes (Fig. 2), and are quite common in the area of Einigen. However, the average inclination angle of gypsum sinkhole walls in GPR images is ~82 (measured on 4 walls of 2 independent well recognizable sinkholes) and differs clearly from measured apparent ~62 dips of rotated pebbles in the gravel pit. In the GPR profile, the local sinkhole shows typical blanking underneath (Fig. 7). The inclinations of the northern and southern flanks amounts to 80.9 and 82.9 respectively, which is significantly steeper than the overall trend of the apparent dipping angles corresponding to the different fault strands (Fig. 7). These characteristics are observable in several lines, where sinkhole locations are mapped and intersect GPR profiles. Furthermore, the sinkholes have mostly circular shapes, whereas the observed features in the gravel pit line up along a linear trend with a hanging wall and a footwall block and do not show the circular shape of a subsiding karst feature. We thus

9

consider a sinkhole as unlikely to be the origin of the deformation structures in the gravel pit. The gravel pit is also located close to the bed of the river Kander. Nevertheless, we are confident that the observed fault strands in the gravel pit are not directly related to gravitational instabilities of the river bed. The faults in the pit strike NW-SE as indicated by GPS data and are hence orthogonal to the course of the river bed and potential listric fault planes. However, instable slopes in the Kander river bed may well be a consequence of fault-weakened lithologies susceptible to erosion and hence provide additional evidence outlining the path of the fault trace to the west. The gentle valley running across several moraine crests coincides with the deformations in the gravel pit and with the offsets in the GPR data, providing further evidence of a complex fault system along this zone. The course of the valley is curved and was most likely shaped by a combination of erosion by subglacial meltwater and gypsum dissolution, both of which are strongly favored by fault-weakened lithologies. During the Last Glacial Maximum, overpressured subglacial meltwater in the confluence zone between the Kander, Lütschine and Aare Glacier may have eroded locally the bedrock along the pre-weakened bedrock structure leading to the gentle valley. At that time, the area was covered by roughly 700 m of ice (Bini et al., 2009) producing a complex subglacial draining system (e.g. Bennett and Glasser,

Fig. 6. Seismic reflection profile 40, 3 vertically exaggerated. Bottom: Seismic sequence stratigraphic interpretation: S0 ¼ water column, S1¼Holocene sediments, S2 ¼ lacustrine deposits, S4 ¼ bedrock. The location of the seismic section is given in Fig. 2. The white star denotes the location of the main fault strand (refer to Fig. 2 for location). At 800 m distance along the profile, the bedrock shows a deformed and offset topography attributed to a normal fault.

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

10

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

Fig. 7. GPR line 8, 4.75 vertically exaggerated. The location of the profile is indicated as blue line in Fig. 2. Horizontal noise from recording devices could not be entirely removed by eigenvector filtering and is still present on the profile. Black lines: Interpreted of top bedrock. Red lines: Sinkhole from gypsum dissolution and its associated deformation features. Yellow: Sedimentary cover. Orange lines: supposed fault structures. The orange star marks the main EFZ strand (refer to Fig. 2 for location). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2009). Additionally, high flow rates can lead to substantial gypsum dissolution. A water flow of 1 m/s can remove up to 1.7 m/yr of a gypsum face along a river, as observed in the River Ure at Ripon Parks, North Yorkshire (James et al., 1981). Moreover, the Penninic nappes to the WNW and to the ESE close to Spiez dip towards the north and south, respectively (Beck and Gerber, 1925). The valley hence lies directly in the core of an antiform, that is, in combination with a fault, rather sensitive to erosion of any form. Further evidence suggesting a fault zone is given by the active gas and water seeps forming the subaquatic morphologic depressions on the adjacent lake floor, which are located on the same lineament. This implies that they are likely related and fed by one linear system, as for instance a fault. It is a common observation that faults enhance susceptibility for fluid flow in a karstic system €uselmann et al., and have a big impact on fluid flow in caves (Ha 1999). The linear arrangement of pockmarks along the strike of a fault is a well-known phenomenon and has been termed “faultstrike” pockmarks (Soter, 1999; Pilcher and Argent, 2007). The coincidence of pockmarks and subaquatic fluid expulsion is a common feature and has been observed at many sites (Hovland et al., 2002; Dimitrov and Woodside, 2003; Kuscu et al., 2005). These observations point towards enhanced fluid transport along fault planes, possibly originating from deep thermogenic sources (Brooks et al., 1974; Hasiotis et al., 1996; Etiope et al., 2010; Belgrano et al., 2016). In lakes, spatial overlap between the occurrence of

pockmarks and active fault systems has been reported (Cartwright et al., 2004; Wessels et al., 2010; Reusch et al., 2015). These interpretations are further supported by the lacustrine seismic reflection data that cover the area where the lineament continues towards the deep basin: The two reflection seismic profiles (Figs. 5 and 6) show well discernable bedrock topography, which is offset at several locations that we have interpreted to be various fault strands. Line 35 shows amplitude anomalies in seismic unit S2 that might be related to the presence of a fault system with a major fault strand that offsets the glacial deposits by several meters and reveals an apparent dip of 36 (Fig. 5; position 1400e1500 m). The likely linked strand right above has an apparent dip of 33 , offsets the Holocene deposits by 1.5e2 m and shows evidence for normal faulting as well. Line 40 is marked by disrupted bedrock topography likely due to several faults. One of the fault strands offsets Holocene sediments by 2 m, dips 24.2 and shows normal faulting (Fig. 6; position 700e850). These low inclination values can be mostly attributed to the low angle between the seismic sections and the fault trace as well and the true dips might be higher. 5.2. Timing The gentle valley towards Lake Thun is crossed by several moraine crests, none of them showing clear lateral displacement on the digital elevation model. The moraines were formed shortly after

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

the abandonment of the piedmont glaciers maximum extent and during the first major recessional phase leading to ice-free conditions in the northern Alpine foreland at ~18 ka (Reitner, 2007; IvyOchs et al., 2008; Wirsig et al., 2016). The vertical offset of all individual fault strands combined in the GPR profile amounts to ~7 m though, indicative for significant movements throughout the existence of the fault system, i.e. prior to the LGM. The valley has not only a surface expression, but also a significant imprint in the bedrock topography as documented by i) GPR profiles, ii) maps of bedrock topography based on drilling data (‘Amt für Wasser und Abfall’, Canton of Bern) and iii) by the fact that bedrock outcrops to the north and south of the valley. The fact that the fault strands in the gravel pit offset a sediment layer that has been 14C dated to ~11,000 yr BP shows that the most recent movement has occurred during the Holocene and can be interpret as a reactivation of the fault system. 5.3. Broader tectonic context of the fault Located close the northern edge of the Alpine nappe stack, the proposed ~ E-W trending fault zone, hereafter termed Einigen Fault Zone (EFZ), can be placed in a larger Alpine tectonic context: Pfiffner (2011) suggests a major fault along Lake Thun responsible for the sudden southwestern termination of the front of the Helvetic nappes. Considering the significant geological differences of the units occurring on both lake sides and the large displacement of the border of the North Alpine Front across Lake Thun, a fault that has been active during nappe emplacement might be the most plausible interpretation explaining the different observations on both sides of the lake. The sinistral movements inferred from the Diemtigen earthquake swarm show active fault planes with an average strike of ~350 with a P-axis oriented NW - SE (Diehl et al., 2015), and thus are parallel to the Lake Thun basin, as well as to the active sinistral N-S striking Fribourg fault axis some 30 km to the northwest of Diemtigen (Kastrup et al., 2007). The fault plane solutions from Diemtigen are in good agreement with the general stress regime along the northern front of the Alps (Kastrup et al., 2004) and indicate a NW-SE compressional stress field. This orientation of the largest principal stress axis today is, however, ideally oriented to reactivate the EFZ. In general, the EFZ strikes E - W, starting with normal faults in the west, delineated by the rotated pebbles in the gravel pit. Moving along the fault to the east, the course of the Einigen fault bends from NW - SE to ENE - WSW (Fig. 2). GPR profiles reveal predominantly reverse faults in the south of the fault and normal faulting in the north of the section, implying a pattern reminiscent of flower structures as it is typical for strike-slip regimes. Within the lake, normal faulting again dominates the seismic profiles. At the transition of shallow subaquatic areas to the deep basin, the toe of slope shows significant lateral offsets indicative for dextral strike-slip movement (Fig. 2). We therefore conclude that the EFZ is a combination of dextral strike-slip faulting with a normal faulting component. The superposition of two faulting modes is underpinned by the curved course of the fault trace, whereas a simple fault sense would create rather a straight rupture path. An angle of about 40 lies between the longitudinal axis of the lake basin and the EFZ. Assuming the existence of the hypothetical large-scale fault structure along Lake Thun, we can at this stage just speculate whether the EFZ has (antithetic) Riedel shear-like character (e.g. Sylvester, 1988; Christie-Blick and Biddle, 1985; Belgrano et al., 2016) with respect to this superior fault structure, or whether there is no link between the EFZ and the Lake Thun basin. However, the EFZ is orthogonally oriented with respect to the triggered fault plane during the Diemtigen sequence. In that perspective, the EFZ could either be an antithetic Riedel shear structure or simply a

11

transfer fault linking the Diemtigen fault with the Lake Thun fault. Further investigation of lacustrine reflection seismic data will shed new light on the relationship between the EFZ and a potential strike-slip fault along Lake Thun. 6. Summary and conclusions We present multiple evidence of Late Quaternary deformation along an ~ E-W trending zone in the area of Lake Thun. The most reliable explanation for all the observed features is the presence of a fault trace. The entire length of the proposed structure, termed “Einigen Fault Zone” (EFZ) amounts to at least ~4 km, but further continuation to the east and west is likely. Using geomorphological analyses, subsurface geophysical data as well as outcrop studies, a number of features are identified, ranging from small pockmarks in Lake Thun to a gentle valley of 1.5 km in length, cutting across several shore-parallel moraine ridges. GPR and seismic data both indicate the existence of significantly disrupted bedrock surfaces with displaced overlying sediments. Gravel pit outcrops and the geomorphological signatures in and around the lake (aligned pockmarks, subaquatic valley, erosive river slopes) further provide strong evidence, that the EFZ indeed has a causal link to recent activity and marks a neotectonic zone. Some of these features rely on fault-weakened lithologies and fluid flow along fault-controlled pathways, others display clear offset layers. The observations in the gravel pit with rotated clasts indicate that the fault was active in Holocene times especially as it can be traced beyond the dated layer of ~11,000 years BP. In general, the offsets in Holocene sediments are significantly smaller than what is observable in the bedrock, suggesting repeated and ongoing phases of fault activity through time. However, there is no seismic event recorded in the earthquake catalogue (ECOS) indicating current activity of the structure. The closest documented and as “certain” classified event is located 3 km away to the southeast (Mw 3.1, I0 ¼ IV, Faulensee-Spiez, location error  10 km, 02.02.1926). Although no direct local seismicity related to the Einigen fault was recorded to date, the capability of the fault causing significant rupturing and displacements, accompanied by local seismic shaking, cannot be denied. Combining amphibious geomorphologic, geophysical and geological subsurface data revealed a list of evidence supporting the existence of a so far unknown Einigen Fault Zone that might be indicative of an active neotectonic feature. We used geomorphological data to perform a first screening of paleoseismologic indicators. As evidence for a potential fault structure accumulated, we conducted a multi-channel reflection seismic survey and a ground-penetrating radar investigation in order to image the subsurface and improve our understanding of the potential fault trace. The combination of terrestrial and offshore paleoseismic evidence and its combined significance point towards a roughly 4 km long strike-slip fault with a normal faulting component that experienced its last movement during Holocene times. Despite the lack of nearby instrumentally recorded seismicity within the last 40 years, and being aware that single-evidence of paleoseismic indicators are not sufficient to claim active fault structures, we are nevertheless confident that the sum of all presented indicators reveal that the EFZ is a good candidate for an active fault near the northern front of the Alps. Acknowledgments The processing of the seismic reflection data and georadar data was performed with Kingdom Suite 2015 provided by IHS and SeisSpace/ProMAX provided through Halliburton/Landmark's University Grant Program. This work was partially financed by the

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

12

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13

building insurance of the Canton of Bern GVB and the Federal Office of Topography swisstopo. We are thankful for the collaboration with various people at Vigier Beton Einigen, with special thanks to Arnold Gertsch and Peter J. Haller and his team. We appreciated the technical support by Sven Winter during the seismic campaign. We would also like to thank the local landowners Christian Wolf and Peter Zeller who granted access to their property for GPR measurements. We wish to thank reviewer A. Pfiffner and an anonymous reviewer for their helpful comments that improved the quality and clarity of the manuscript.

References Beck, P., Gerber, E., 1925. Geologische Karte Thun-stockhorn. Kümmerly & Frey, Bern. Belgrano, T.M., Herwegh, M., Berger, A., 2016. Inherited structural controls on fault geometry, architecture and hydrothermal activity: an example from Grimsel Pass, Switzerland. Swiss J. Geosci. 1e20. Bennett, M.R., Glasser, N.F., 2009. In: Glacial Geology: Ice Sheet and Landforms, Second ed. Wiley-Blackwell. Bini, A., Buoncristiani, J.-F., Couterrand, S., Ellwanger, D., Felber, M., Florineth, D., Graf, H.R., Keller, O., Kelly, M., Schlüchter, C., Schoeneich, P., 2009. Switzerland during the Last Glacial Maximum 1: 500'000. Bundesamt für Landestopografie swisstopo. Boggs, S., 2009. Petrology of Sedimentary Rocks. Cambridge University Press. Brooks, J.M., Gormly, J.R., Sackett, W.M., 1974. Molecular and isotopic composition of two seep gases from the gulf of Mexico. Geophys Res. Lett. 1. Cartwright, J., Wattrus, N., Rausch, D., Bolton, A., 2004. Recognition of an early Holocene polygonal fault system in Lake Superior: implications for the compaction of fine-grained sediments. Geology 32, 253e256. Chopra, S., Castagna, J., Portniaguine, O., 2006. Seismic resolution and thin-bed reflectivity inversion. CSEG Rec. 31, 19e25. Christie-Blick, N., Biddle, K.T., 1985. Deformation and basin formation along strikeslip faults. SEPM Spec. Pub 1 (37), 1e34. de La Taille, C., Jouanne, F., Crouzet, C., Beck, C., Jomard, H., de Rycker, K., Van Daele, M., 2015. Impact of active faulting on the post LGM infill of Le Bourget Lake (western Alps, France). Tectonophysics 664, 31e49. Diehl, T., Deichmann, N., Clinton, J., Kastli, P., Cauzzi, C., Kraft, T., Behr, Y., Edwards, B., Guilhem, A., Korger, E., Hobiger, M., Haslinger, F., Fah, D., Wiemer, S., 2015. Earthquakes in Switzerland and surrounding regions during 2014. Swiss J. Geosci. 108, 425e443. Dimitrov, L., Woodside, J., 2003. Deep sea pockmark environments in the eastern Mediterranean. Mar. Geol. 195, 263e276. Etiope, G., Zwahlen, C., Anselmetti, F.S., Kipfer, R., Schubert, C.J., 2010. Origin and flux of a gas seep in the northern Alps (giswil, Switzerland). Geofluids 10, 476e485. €h, D., Giardini, D., Ka €stli, P., Deichmann, N., Gisler, M., Schwarz-Zanetti, G., Fa Alvarez-Rubio, S., Sellami, S., Edwards, B., Allmann, B., 2011. ECOS-09 Earthquake Catalogue of Switzerland Release 2011 Report and Database. Public catalogue, 17. 4. 2011. Swiss Seismological Service ETH Zurich. Report SED/RISK. Finckh, P., Kelts, K., Lambert, A., 1984. Seismic stratigraphy and bedrock forms in perialpine lakes. Geol. Soc. Am. Bull. 95, 1118e1128. Gross, R., Green, A.G., Horstmeyer, H., Begg, J.H., 2004. Location and geometry of the Wellington Fault (New Zealand) defined by detailed three-dimensional georadar data. J. Geophys Res-Sol Ea. 109. Hantke, R., Wagner, G., 2005. Eiszeitliche und nacheiszeitliche Gletscherst€ ande im € Berner Oberland: Erster Tei, Ostliches Oberland bis zur Kander. Mitteilungen der Naturforschenden Gesellschaft Bern 62, 107e134. Hasiotis, T., Papatheodorou, G., Kastanos, N., 1996. A pockmark field in the Patras Gulf (Greece) and its activation during the 14/7/93 seismic event. Mar. Geol. 130, 333e344. €uselmann, P., Jeannin, P.-Y., Bitterli, T., 1999. Relationships between karst and Ha tectonics: case-study of the cave system north of Lake Thun (Bern, Switzerland). Geodin. Acta 12, 377e387. Hilbe, M., Anselmetti, F.S., 2014. Signatures of slope failures and river-delta collapses in a perialpine lake (Lake Lucerne, Switzerland). Sedimentology 61, 1883e1907. Hilbe, M., Anselmetti, F.S., Eilertsen, R.S., Hansen, L., Wildi, W., 2011. Subaqueous morphology of Lake Lucerne (Central Switzerland): implications for mass movements and glacial history. Swiss J. Geosci. 104, 425e443. Hovland, M., Gardner, J.V., Judd, A.G., 2002. The significance of pockmarks to understanding fluid flow processes and geohazards. Geofluids 2, 127e136. Hunter, L.E., Howle, J.F., Rose, R.S., Bawden, G.W., 2011. LiDAR-assisted identification of an active fault near Truckee, California. B Seismol. Soc. Am. 101, 1162e1181. Ivy-Ochs, S., Kerschner, H., Reuther, A., Preusser, F., Heine, K., Maisch, M., Kubik, P.W., Schluchter, C., 2008. Chronology of the last glacial cycle in the European Alps. J. Quat. Sci. 23, 559e573. James, A.N., Cooper, A.H., Holliday, D.W., 1981. Solution of the gypsum cliff (Permian, Middle Marl) by the River Ure at Ripon Parks, North Yorkshire. Proc.

Yorks. Geol. Polytech. Soc. 43, 433e450. Kastrup, U., Deichmann, N., Frohlich, A., Giardini, D., 2007. Evidence for an active fault below the northwestern Alpine foreland of Switzerland. Geophys J. Int. 169, 1273e1288. Kastrup, U., Zoback, M.L., Deichmann, N., Evans, K.F., Giardini, D., Michael, A.J., 2004. Stress field variations in the Swiss Alps and the northern Alpine foreland derived from inversion of fault plane solutions. J. Geophys Res-Sol Ea. 109, 1e22. Kuscu, I., Okamura, M., Matsuoka, H., Gokasan, E., Awata, Y., Tur, H., Simsek, M., Kecer, M., 2005. Seafloor gas seeps and sediment failures triggered by the August 17, 1999 earthquake in the Eastern part of the Gulf of Izmit, Sea of Marmara, NW Turkey. Mar. Geol. 215, 193e214. Madritsch, H., Preusser, F., Fabbri, O., Bichet, V., Schlunegger, F., Schmid, S.M., 2010. Late Quaternary folding in the Jura Mountains: evidence from syn-erosional deformation of fluvial meanders. Terra Nova. 22, 147e154. Maurer, H., Deichmann, N., 1995. Microearthquake cluster detection based on waveform similarities, with an application to the western Swiss Alps. Geophys J. Int. 123, 588e600. McCalpin, J.P., 2009a. Field techniques in paleoseismology-terrestrial environments. Int. Geophys Ser. 95, 29e118. McCalpin, J.P., 2009b. Paleoseismology in Extensional tectonic environments. Int. Geophys Ser. 95, 171e269. Meghraoui, M., Delouis, B., Ferry, M., Giardini, D., Huggenberger, P., Spottke, I., Granet, M., 2001. Active normal faulting in the upper Rhine graben and paleoseismic identification of the 1356 Basel earthquake. Science 293, 2070e2073. Michetti, A.M., Audemard, F.A., Marco, S., 2005. Future trends in paleoseismology: integrated study of the seismic landscape as a vital tool in seismic hazard analyses. Tectonophysics 408, 3e21. Monecke, K., Anselmetti, F.S., Becker, A., Sturm, M., Giardini, D., 2004. The record of historic earthquakes in lake sediments of Central Switzerland. Tectonophysics 394, 21e40. Olig, S.S., Lund, W.R., Black, B.D., 1994. Large mid-Holocene and late Pleistocene earthquakes on the Oquirrh fault zone. Utah. Geomorphol. 10, 285e315. Pavoni, N., 1980. Comparison of focal mechanisms of earthquakes and faulting in the helvetic zone of the central Valais, Swiss Alps. Eclogae Geol. Helv. 73, 551e558. Persaud, M., Pfiffner, O.A., 2004. Active deformation in the eastern Swiss Alps: postglacial faults, seismicity and surface uplift. Tectonophysics 385, 59e84. Pettijohn, F.J., 1930. Imbricate arrangement of pebbles in a pre-Cambrian conglomerate. J. Geol. 38, 568e573. Pfiffner, A.O., 2011. Structural Map of the Helvetic Zone of the Swiss Alps, Including Vorarlberg (Austria) and Haute Savoie (France), 1:100 000. Geological Special Map 128. Federal Office of Topography swisstopo. Federal Office of Topography swisstopo, Wabem, p. 128. Pilcher, R., Argent, J., 2007. Mega-pockmarks and linear pockmark trains on the West African continental margin. Mar. Geol. 244, 15e32. Reber, R., Schlunegger, F., 2016. Unravelling the moisture sources of the Alpine glaciers using tunnel valleys as constraints. Terra Nova. 28, 202e211. Reitner, J.M., 2007. Glacial dynamics at the beginning of Termination I in the Eastern Alps and their stratigraphic implications. Quatern Int. 164e65, 64e84. Reusch, A., Loher, M., Bouffard, D., Moernaut, J., Hellmich, F., Anselmetti, F.S., Bernasconi, S.M., Hilbe, M., Kopf, A., Lilley, M.D., Meinecke, G., Strasser, M., 2015. Giant lacustrine pockmarks with subaqueous groundwater discharge and subsurface sediment mobilization. Geophys Res. Lett. 42, 3465e3473. Schmelzbach, C., Tronicke, J., Dietrich, P., 2011. Three-dimensional hydrostratigraphic models from ground-penetrating radar and direct-push data. J. Hydrol. 398, 235e245. Schnellmann, M., Anselmetti, F.S., Giardini, D., McKenzie, J.A., 2006. 15,000 Years of mass-movement history in Lake Lucerne: implications for seismic and tsunami hazards. Eclogae Geol. Helv. 99, 409e428. Singer, J., Diehl, T., Husen, S., Kissling, E., Duretz, T., 2014. Alpine lithosphere slab rollback causing lower crustal seismicity in northern foreland. Earth Planet S. C. Lett. 397, 42e56. Soter, S., 1999. Macroscopic seismic anomalies and submarine pockmarks in the Corinth-Patras rift, Greece. Tectonophysics 308, 275e290. Stirling, M., Rhoades, D., Berryman, K., 2002. Comparison of earthquake scaling relations derived from data of the instrumental and preinstrumental era. B Seismol. Soc. Am. 92, 812e830. Strasser, M., Anselmetti, F.S., Fah, D., Giardini, D., Schnellmann, M., 2006. Magnitudes and source areas of large prehistoric northern Alpine earthquakes revealed by slope failures in lakes. Geology 34, 1005e1008. Strasser, M., Hilbe, M., Anselmetti, F.S., 2011. Mapping basin-wide subaquatic slope failure susceptibility as a tool to assess regional seismic and tsunami hazards. Mar. Geophys Res. 32, 331e347. Strasser, M., Monecke, K., Schnellmann, M., Anselmetti, F.S., 2013. Lake sediments as natural seismographs: a compiled record of Late Quaternary earthquakes in Central Switzerland and its implication for Alpine deformation. Sedimentology 60, 319e341. €nge im Thunersee. Sturm, M., Matter, A., 1972. Sedimente und Sedimentationsvorga Eclogae Geol. Helv. 65, 563e590. Sturm, M., Matter, A., 1978. Turbidites and varves in Lake Brienz (Switzerland): deposition of clastic detritus by density currents. Spec. Publs. Int. Assoc. Sediment. 2, 147e168. Sylvester, A.G., 1988. Strike-slip faults. Geol. Soc. Am. Bull. 100, 1666e1703. Ustaszewski, M., Herwegh, M., McClymont, A.F., Pfiffner, O.A., Pickering, R.,

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033

S.C. Fabbri et al. / Quaternary International xxx (2017) 1e13 Preusser, F., 2007. Unravelling the evolution of an Alpine to post-glacially active fault in the Swiss Alps. J. Struct. Geol. 29, 1943e1959. Ustaszewski, M., Pfiffner, O.A., 2008. Neotectonic faulting, uplift and seismicity in the central and western Swiss Alps. Geol. Soc. Lond. Spec. Publ. 298, 231e249. Verschuur, D.J., Berkhout, A.J., Wapenaar, C.P.A., 1992. Adaptive surface-related multiple elimination. Geophysics 57, 1166e1177. Wells, D.L., Coppersmith, K.J., 1994. New Empirical relationships among magnitude, rupture length, rupture width, rupture area, and surface displacement. B Seismol. Soc. Am. 84, 974e1002. Wessels, M., Bussmann, I., Schloemer, S., Schluter, M., Boder, V., 2010. Distribution, morphology, and formation of pockmarks in Lake Constance, Germany. Limnol.

13

Oceanogr. 55, 2623e2633. Widess, M.B., 1973. How thin is a thin bed. Geophysics 38, 1176e1180. Wiemer, S., Giardini, D., Fah, D., Deichmann, N., Sellami, S., 2009. Probabilistic seismic hazard assessment of Switzerland: best estimates and uncertainties. J. Seismol. 13, 449e478. Wirsig, C., Zasadni, J., Ivy-Ochs, S., Christl, M., Kober, F., Schlüchter, C., 2016. A deglaciation model of the Oberhasli, Switzerland. J. Quat. Sci. 31, 46e59. Wirth, S.B., Girardclos, S., Rellstab, C., Anselmetti, F.S., 2011. The sedimentary response to a pioneer geo-engineering project: tracking the Kander River deviation in the sediments of Lake Thun (Switzerland). Sedimentology 58, 1737e1761.

Please cite this article in press as: Fabbri, S.C., et al., Combining amphibious geomorphology with subsurface geophysical and geological data: A neotectonic study at the front of the Alps (Bernese Alps, Switzerland), Quaternary International (2017), http://dx.doi.org/10.1016/ j.quaint.2017.01.033