About the project

Research Project Objectives (scientific problem aimed to be solved by the proposed project, project’s research hypotheses)

The overall objective of this project is to refine the palaeoenvironmental consequences of the – primarily southern – Baltic Sea Basin countries on the basis of reconstructions of glacio-isostatic rebound resulting from advance and retreat of the Scandinavian Ice Sheet. Isostatic rebound led to earthquakes that, if the conditions were suitable, left traces in the form of laterally extensive seismites characterised by a high concentration of soft-sediment deformation structures (SSDS). The recognition of such seismites as well as the palaeoenvironmental interpretation needed for their reconstructions will be based on sedimentological investigations in combination with methods that are less commonly applied in this context. These methods will include: (1) dating of sediments by (1a) OSL and (1b) 14 C, (2) numerical modelling of (2a) tectonic activity during ice advance and retreat in areas around (and particularly south of) the Baltic Sea Basin, of (2b) soft-sediment deformation structures development triggered by high-magnitude earthquakes, and of (2c) differences in the lithology of the earthquake-affected sediments, and (3) field studies of mesoscopic deformation structures.

Study area
The investigations will be carried out primarily in Poland, Germany, Latvia, Estonia and Lithuania as all these territories form the very part of Europe where the Pleistocene Scandinavian glaciations had their maximum extent (Fig. 1). We omit the Kaliningrad region intentionally due to the political situation. As the Scandinavian countries (Norway, Sweden, Finland and Denmark) also were – either or not partially – covered by the Scandinavian Ice Sheet, they also underwent similar situations regarding the ice front as did the more southern countries. It may therefore well be that some fieldwork in these more northern countries will also be required. Although it is obvious that a thick ice cap causes high pressure on the mineral substratum, resulting in crustal subsidence during rapid ice accretion and in uplift during melting of the ice mass – and thus sometimes in
earthquakes – there are other triggers of earthquakes as well. In fact, tectonic activity and volcanic activity are much more common causes of earthquakes and these may also leave traces in the form of seismites (particularly Italy has fine examples; Moretti, 2000; Moretti et al., 2011). It may therefore be useful to compare seismites caused by glacio-isostatic rebound with those triggered by tectonics and volcanism.

The study of seismites
Palaeoseismic studies depend on the recognition of seismites on the basis of their characteristics, including soft-sediment deformation structures formed due to liquefaction and fluidization, and based on the study of the geometry of the deformed beds and on lithofacies analysis. Seismically-induced liquefaction processes are commonly connected to seismic shocks with a magnitude of five or more (Ambraseys, 1988). Such seismic shocks form a major hazard in many areas of Europe (Fig. 2). The seismites caused by more than 90% of the historical seismic events are located within 40 km from the epicentre (Galli, 2000). Most of the soft-sediment deformation structures related to earthquakes with a magnitude of 5-7 occur even within a distance of 20 km from the epicentre (Papadopoulos and Lefkopoulos, 1993). Earthquakes with a lower magnitude may also trigger liquefaction, but only in an area close to the epicentre. Thus, the type and size of layers with soft-sediment deformation structures of a seismic origin are a function of the magnitude of the earthquakes (Guiraud and Plaziat, 1993). The spatial distribution and lateral changes in the type and size of these soft-sediment deformation structures can be used to locate main active faults (Alfaro et al., 2010) or the
epicentres of deep earthquakes.

Fig 1. Extent of the Scandinavian ice sheet.

Fig 2. Seismic hazard map of the Europe and northernmost Africa (from Giardini et al., 2003).

Our investigations will concentrate not only on the recognition of seismites and the analysis of their soft-sediment deformation structures, but also on tracing faults in the bedrock which may have been reactivated during glaciation and deglaciation phases. Many of these faults must be considered as ‘active faults’ because the great majority of seismites occur globally in tectonically active areas. However, the areas south of the Baltic Sea Basin are presently not affected by significant tectonic activity, so the faults in the vicinity of the Quaternary seismites in this area were most probably reactivated by the changing pressures exerted by the advancing and retreating ice fronts. It is likely that also differences in the thickness of the Quaternary sediments nearby these faults contributed to their reactivation.

Aims of numerical modelling of the glacio-isostatic rebound:
– estimation of the maximum earthquake magnitudes during deglaciations;
– defining the characteristics of the areas where re-activation was most frequent during retreat of the Scandinavian Ice Sheet;
– identification of the deep tectonic structures which were reactivated during glacio-isostatic rebound.

Isostatic rebound can leave traces in the form of earthquake-induced soft-sediment deformation structures. All studies of soft-sediment deformation structures show that the final morphology is largely related to the characteristics of the initial sediment, the driving force acting during deformation, and the duration of a deformable state. Furthermore, experiments producing seismically-induced soft-sediment deformation structures (Moretti et al., 1999) show that the final morphology is independent of the acceleration (and magnitude) of an earthquake but, instead, is related to the initial sediment properties. Loading, for instance, is a widespread process facilitated by reversed density gradients, but in the case of seismites a loadcast can develop only if liquefaction takes place in the layer under the parent layer of the loadcast or its ultimate form, a pseudonodule (Moretti and Ronchi, 2011). Liquefaction of a sediment requires not only a water- saturated state, but it also needs a trigger, i.e. – in the case of this project – a seismic shock wave. The thickness of the sediments deformed by seismically-induced liquefaction is not related to the magnitude of the seismic shock.

Which types of soft-sediment deformation structures are formed as a result of shocks has been investigated in several experiments (e.g., Owen, 1992; Moretti et al., 1999); these showed that particularly loadcasts and associated structures are easily formed. This is understandable, as the seismically-induced shock waves that are responsible for the development of soft-sediment deformation structures in water-saturated, unconsolidated sediments in the uppermost decimetres of the sedimentary succession are S-waves, which result in alternations of compression and tension within the sediment, thus allowing material to sink into the underlying layer, even if there is hardly any difference in density (see Rossetti, 1999). The final morphology of soft-sediment deformation structures resulting from liquefaction and fluidization (the processes that play the most important role during loadcasting) depend mainly on the initial sedimentary setting, the driving force and the duration of the deformable state, whereas the nature of the trigger mechanism seems to play a minor or negligible role (Owen and Moretti, 2011; Owen et al., 2011). In other words, soft-sediment deformation structures can have identical morphologies, independent of whether they were formed due to a seismic shock or by any other trigger mechanism.
There is still considerable debate about how to recognize seismites. Recently Owen and Moretti (2011) proposed, based on literature (e.g. Sims, 1975; Obermeier, 1996; Rossetti, 1999; Wheeler, 2002; Hilbert- Wolf et al., 2009) and field examples, the following criteria to recognize seismites:
(1) a large areal extent;
(2) lateral continuity of deformed sediment;
(3) vertical repetition;
(4) soft-sediment deformation structures with a morphology comparable with structures described from earthquake-affected layers;
(5) proximity to active faults;
(6) dependence of complexity or frequency with distance from the triggering fault.

The frequency, size and/or complexity of seismically-induced soft-sediment deformation structures should diminish with increasing distance from the epicentre (Rodríguez-Lopez et al., 2007). It has become clear since then, however, that there are many pitfalls and that the various criteria are valid only if the geological context is relatively simple (Moretti and Van Loon, 2014), so that experience of the researchers is required.

Seismites have been reported from almost all sedimentary environments, but particularly from lacustrine successions (Sims, 1975; Alfaro et al., 1997; Moretti and Ronchi, 2011), including glaciolacustrine ones (Gruszka and Van Loon, 2007). Our investigation will focus mainly on fluvial and lacustrine sediments (Fig. 3) in all countries where sediments occur that may help to meet the objectives of the project.

Fig. 3. Soft-sediment deformation structures triggered by earthquakes. A: Levels of two seismites in Western Pomerania – NE Poland (see Van Loon and Pisarska-Jamroży, 2014a); B: Seismite level in the Ujście Basin (NW Poland); C: Three of seven levels of seismites in a succession in NE Latvia; D: Seismites in SE Latvia; E: Succession with twelve seismites in SE Latvia

It is worth noting that soft-sediment deformation structures can be caused by a wide variety of triggers, including glaciotectonics, gravity-induced sliding or slumping, permafrost-induced processes, cryoturbation, and fluidization or liquefaction due to instabilities resulting from reversed density gradients to mention only the most common ones. Most soft-sediment deformation structures within Quaternary deposits in Poland, Germany, Latvia, Estonia and Lithuania have been interpreted as a result of periglacial processes. Indeed, many periglacial structures and seismically-induced soft-sediment deformation structures result from loading, but the trigger mechanisms are completely different (Pisarska-Jamroży et al. in press). A fundamental question therefore now arises with the recently largely increased insight into the formation and characteristics of seismites: are all Pleistocene layers in Poland, Germany, Latvia, Estonia and Lithuania that contain abundant soft-sediment deformation structures and that were ascribed earlier to periglacial processes indeed due to periglacial conditions, or were at least part of the deformations triggered by seismic shocks due to glacial rebound?

Based on preliminary research, we tend to conclude that part of the ‘periglacial structures’ were formed due to shock waves.

Many periglacial structures and seismically-induced soft-sediment deformation structures result from loading, but the trigger mechanisms are completely different. In the periglacial environment water in sediments during temperature changes around its freezing point can induce intense deformation (Van Loon, 2009). The freezing/thawing alternations during the Pleistocene glacials affected many sediments, resulting in numerous soft-sediment deformation structures. Some soft- sediment deformation structures are considered by various authors (Jahn, 1975; Clapperton, 1993; French, 2007) as characteristic of periglacial processes. Frost fissures, frost wedges and associated structures may, indeed, be the only group that needs no other processes as well. Cryoturbation (due to ice pressure and fluidisation of the sediment after ice melting) must, however, be considered as a result of typically periglacial processes in combination with much more common deformational processes such as loading and liquefaction (Benedict, 1976; Jary, 2009).

Aims of numerical modelling of soft-sediment deformation structures:
– identification of soft-sediment deformation structures, which are typical for glacial rebound;
– identification of soft-sediment deformation structures developed in lithologically different sediments.

The Scandinavian Ice Sheet advanced and retreated several times, and its behaviour can be regarded as a proxy of climate change. Approximately 10% of the Earth is currently covered by ice. Fluctuations in the extent of the ice sheet had and still have consequences for the occurrence of earthquakes, and thus also for natural hazards (Tian et al., 2015).

Project hypothesis
– Earthquakes triggered by rebound after retreat of an ice cap occurred in the vicinity of tectonic structures (faults) in the bedrock within the Baltic Sea Basin.
– The more rapidly deglaciation occurred, the larger the number of and the stronger the magnitudes of the earthquakes resulting from the glacial rebound.
– Shock-induced soft-sediment deformation structures due to glacial rebound are in most respects comparable with soft-sediment deformation structures in non-glacigenic deposits.