Tsunami recurrence in the eastern Alaska-Aleutian arc: A Holocene stratigraphic record from Chirikof Island, Alaska

Abstract

Despite the role of the Alaska-Aleutian megathrust as the source of some of the largest earthquakes and tsunamis, the history of its pre–twentieth century tsunamis is largely unknown west of the rupture zone of the great (magnitude, M 9.2) 1964 earthquake. Stratigraphy in core transects at two boggy lowland sites on Chirikof Island’s southwest coast preserves tsunami deposits dating from the postglacial to the twentieth century. In a 500-m-long basin 13–15 m above sea level and 400 m from the sea, 4 of 10 sandy to silty beds in a 3–5-m-thick sequence of freshwater peat were probably deposited by tsunamis. The freshwater peat sequence beneath a gently sloping alluvial fan 2 km to the east, 5–15 m above sea level and 550 m from the sea, contains 20 sandy to silty beds deposited since 3.5 ka; at least 13 were probably deposited by tsunamis. Although most of the sandy beds have consistent thicknesses (over distances of 10–265 m), sharp lower contacts, good sorting, and/or upward fining typical of tsunami deposits, the beds contain abundant freshwater diatoms, very few brackish-water diatoms, and no marine diatoms. Apparently, tsunamis traveling inland over low dunes and boggy lowland entrained largely freshwater diatoms. Abundant fragmented diatoms, and lake species in some sandy beds not found in host peat, were probably transported by tsunamis to elevations of >10 m at the eastern site. Single-aliquot regeneration optically stimulated luminescence dating of the third youngest bed is consistent with its having been deposited by the tsunami recorded at Russian hunting outposts in 1788, and with the second youngest bed being deposited by a tsunami during an upper plate earthquake in 1880. We infer from stratigraphy, 14 C-dated peat deposition rates, and unpublished analyses of the island’s history that the 1938 tsunami may locally have reached an elevation of >10 m. As this is the first record of Aleutian tsunamis extending throughout the Holocene, we cannot estimate source earthquake locations or magnitudes for most tsunami-deposited beds. We infer that no more than 3 of the 23 possible tsunamis beds at both sites were deposited following upper plate faulting or submarine landslides independent of megathrust earthquakes. If so, the Semidi segment of the Alaska-Aleutian megathrust near Chirikof Island probably sent high tsunamis southward every 180–270 yr for at least the past 3500 yr.