HOLOCENE SEA-LEVEL CHANGES, EARTHQUAKES AND TSUNAMI AROUND BERING GLACIER

Bering Glacier lies in a dynamic yet poorly understood region in terms of relative sea-level change. In addition to the interaction of global, eustatic sea level change with crustal uplift or subsidence resulting from glacio-hydrostatic adjustment processes, the area lies at the convergent plate margin between the Pacific and North American plates.

A rock exposure on the east shore of Tashalich Arm contains fossil shells of Penitella pennita, a shallow subtidal, rock-boring species. 14C ages indicate relative sea level above present ~9200-5500 BP, with the coastline much further to the north. Despite numerous investigations of terrace-like features on the hills around Bering Glacier we have found no evidence to demonstrate that they are raised shoreline features, so we have no direct evidence of the elevation of relative sea level between the time of ice retreat and the oldest ages from the Taschlich Arm exposure.

Diatom analysis of sediments close to the mouth of Seal River record sea level below present from ~2800 to at least 300 BP, ending with a rapid fall of sea level. This may indicate land uplift during an earthquake, but we need to corroborate this by finding further locations with comparable evidence.

Cores across the marsh at Cape Suckling reveal multiple mud-peat couplets. A sharp boundary at the base of a peat horizon may indicate land uplift during an earthquake, raising intertidal mudflat above the level of high tides. We are currently undertaking diatom analyses to test this hypothesis. 14C ages on the base of peat horizons indicate possible correlation of one ~300 years ago with the sequence at Seal River and another with the 1899 Cape Yakatak earthquake. The oldest mud-peat couplet, 930-740 BP, shows greatest relative land uplift together with a sand layer that we interpret as a tsunami deposit caused by vertical displacement of the sea floor during a great earthquake.

Bering Glacier lies between the major area of uplift during the great 1964 earthquake and the area uplifted in 1899. Our evidence suggests that it potentially records co-seismic land movements from great earthquakes generated from movement on two segments of the subduction zone: in AD1964 and ~900 years ago the segment to the west, and possibly AD1899 and ~300 years ago, the segment to the east, although the evidence for these two is less convincing.