Qualitative modeling of seismite formation

C. PSHENICHNY; S. GUZMÁN; R. CARNIEL; B. LONGHINOS

Tromsø

Conferencia; Energies of the high North – Arctic Conference 2012; 2012

Artic Frontiers

Seismically-induced deformations in rocks and sediments, being an interesting geological phenomenon per se, are also indicative of past seismic events of big magnitude. Deformed sediments, or seismites, occur widely in the Arctic and yet wider southward, in Fennoscandia and adjacent areas of north-western part of Russian Plane. Based on the theory of extended glacial sheet development in the Pleistocene, strong earthquakes there were caused in early to middle Holocene by relaxation of the earthcrust after retreat of the last glacial sheet. If so, alike environment may be developing now in some islands of the Arctic Ocean due to well-documented melting of ice caps in the course of ongoing general warming. Taking into the account that, for the same very reason of warming, economic activity in the Arctic is expected to flourish in the near decades, the issue of potential seismic response to glacier retreat becomes unignorable and may have geotechnical and infrastructural implications. Moreover, melting permafrost imposed on tectonic movements also creates redistribution of pressure and water in the sediments and may provide an array of structures very similar to seismitic. Yet more interesting, and possibly dangerous for human activity and lives, could be interference of melting and seismic events. Meanwhile, existing classifications of seismitic and related structures are cumbersome, wooly and incomplete, as stated by their authors. They usually fail to incorporate paragenetically related phenomena like slope processes (importantly for the Arctic, including those that require negligible slope angle and can develop on virtually horizontal surface) and in general cannot account for the diversity of scenarios of earthquake and post-earthquake processes in the subsurface and on the surface. However, exactly this issue is of the greatest practical relevance. Obviously, an approach other than bare classification should be tried for conceptualization of the formation of seismites. A promising option is application of an artificial intelligence tool developed specially for the geosciences, the event bush method. The virtue of this approach is not to register each variety of seismically-induced deformation and put in a drawer but, rather, offer a way to qualitatively infer as complex deformation scenario as supposed by field data, including the related slope and environmental processes, from a limited set of theoretically plausible and practically reasonable postulates. This approach will be demonstrated in the presentation, from its methodological foundations to application examples from throughout the world.