Freshwater (paleo)tsunamis – a review

Abstract

In freshwater systems (rivers and lakes), historical and recent tsunamis have been documented and their traces have been found in the geological record, but studies of paleotsunamis (prehistorical tsunamis) in such environments are still underrepresented. This contribution reviews paleotsunami studies with a focus on the post −2011 period and uses historical events to highlight some areas of research that have received little attention. In the past decade, the number of paleotsunami studies has increased and this includes those carried out over freshwater settings. However, studies of lacustrine paleotsunamis compared to studies on marine paleotsunamis are still rare and those for rivers are to our knowledge non-existent. Similarly, studies of historical tsunamis generated by meteorological disturbances have been carried out but there have been none for their paleotsunami counterpart. Thus, within this review, to cover all different aspects of tsunami generation processes in freshwater systems, we have used several historical examples, although there is a notable focus on lacustrine paleotsunamis. This review shows that future studies of freshwater paleotsunamis are necessary in order to better understand their causes, frequencies and hazard potential.

Introduction

Tsunami is a Japanese term for “harbor wave” (Darbyshire and Ishiguro, 1957; Goff et al., 2016). The current definition of the term ‘tsunami’ describes a series of propagating waves of extremely long wavelength and period, usually generated by sudden disturbances of the water column associated with earthquakes occurring below or near the ocean floor. Additional generating mechanisms include volcanic eruptions, subaerial and submarine mass-movements, and bolide or other impacts upon the ocean surface (Tsunami Glossary, 2019). Tsunamis are invariably considered to be associated with marine settings and the earthquakes that cause them, a perception that has been reinforced by recent events such as the 2004 Indian Ocean and the 2011 Tohoku-oki events. While most marine tsunamis are most likely generated by plate displacements along sea-floor ruptures during megathrust earthquakes, recent events such as the 2018 Anak Krakatau tsunami (lateral collapse of Anak Krakatau volcano, Grilli et al., 2019; Takagi et al., 2019) indicate that this is by no means always the case.

Rapid displacement of large water masses can occur in any aqueous system. Worldwide historical documents and eyewitness reports have shown that tsunamis do not only occur in open oceans but also in confined fjords (e.g. 1958 Lituya Bay impulse wave in Alaska, US: Miller, 1960; Fritz et al., 2009) and in freshwater systems such as rivers and lakes (e.g. Schnellmann et al., 2002; Fritsche et al., 2012; Kremer et al., 2012; Clark et al., 2015; Donaldson et al., 2019; Hu et al., 2020). These historical events allow us to document the existence, causes and consequences of such tsunamis. Unlike historical tsunamis, we only know about the occurrence of paleotsunamis (prehistorical tsunamis) through the traces that they have left behind in the geological record. We distinguish between direct traces that are the deposits of the tsunami itself on lake shores or backwash deposits on the lake floor (Dirksen et al., 2011; Freundt et al., 2006; Moore et al., 2006; Moore et al., 2014) and indirect traces of paleotsunamis that are reflected in the geological record of the causal mechanism (e.g. Schnellmann et al., 2002; Kremer et al., 2012; Bozzano et al., 2009). In the latter case, numerical modelling is used to support the hypothesis that a freshwater paleotsunami occurred and to assess the magnitude of the inferred event (e.g. modelling the tsunamigenic effects of large mass-movements in lakes; Kremer et al., 2014). In the former case, research on the geological traces of freshwater paleotsunamis is rare.

When searching for “paleotsunamis in lakes” (and its synonyms), around 13 to 1000 results are found on the “web of knowledge” and on “google scholar”, respectively. However, the majority of these results refer to marine tsunamis that are recorded in coastal lakes (e.g. Kempf et al., 2017). A review of the literature indicates that there are few publications with a specific focus on paleotsunamis generated in lakes (De Lange and Moon, 2016; Dirksen et al., 2011; Freundt et al., 2006; Kremer et al., 2014; Kremer et al., 2015; Leithold et al., 2019; Moore et al., 2006; Mountjoy et al., 2019; Nigg et al., 2021; Schnellmann et al., 2002; Strupler et al., 2018). There appear to be no publications referring specifically to river paleotsunamis or to paleo-meteotsunamis.

In this study, we use the following definition of a tsunami: “A series of waves that are formed by a sudden displacement of the water, caused in or adjacent to a freshwater system (lake and river) by subaerial and subaqueous mass-movements, volcanic activity, co-seismic fault displacement and meteorological effects” (Fig. 1). The preposition “paleo” refers to the prehistoric period where historical (written) documentation is absent. As the historical period varies between countries and cultures, we consider the definition used in the original publications (e.g. In Switzerland, historical documents describe natural hazards already in the 6th century (Gisler et al., 2007) while in New Zealand the first written records are dated around 1840 CE (Clark et al., 2015)). Historical events related to human activity, e.g. mass-movements triggered by construction and quarry works close to the shore, were not considered. In addition, historical wave events generated by e.g. ice avalanches in moraine-dammed proglacial lakes (e.g. Clague and Evans, 2000) were also not included. Many of these lakes have formed due to glacier retreat after the Little Ice Age and are therefore considered only short-term structures in geological terms.

The main objective of this study is to review the literature on paleotsunamis in freshwater systems (rivers, lakes). We focus mainly on the period since the devastating 2011 Tohoku-oki tsunami. In particular, we emphasize how increased tsunami awareness has led to a series of follow-up studies that also investigated the lacustrine realm. However, since the number of studies of freshwater paleotsunamis is limited, we also use historical case studies and pre-2011 literature to complete a review of the full range of processes that can generate these events. In the following, we briefly review the historical and prehistorical freshwater tsunami dataset with a focus on the paleotsunamis. We, then, discuss the advances made in paleotsunami research since 2011, the state of current research and propose future research ideas.