No abstract

International audienceKamchatsky Peninsula lies within a complex meeting place of tectonic plates, in particular, the orthogonal interaction of the west-moving Komandorsky Island block with mainland Kamchatka. Examining the Holocene history of vertical deformation of marine wave-built terraces along the peninsular coast, we differentiated tectonic blocks undergoing uplift and tilting separated by zones of stable or subsided shorelines. We analyzed ∼200 excavations along >30 coastal profiles and quantified vertical deformation on single profiles as well as along the coast using paleoshorelines dated with marker tephras. For the past ∼2000 yr, the average rates of vertical deformation range from about -1 to +7 mm/yr. Uplift patterns are similar to those detected from historical leveling and from mapping of the stage 5e Quaternary marine terrace (ca. 120 ka). Average vertical deformation in the Holocene is highest for the shortest studied time period, from ca. A.D. 250 to 600, and it is several times faster than rates for marine oxygen isotope stage (MIS) 5e terraces. Vertical displacements observed along the coast are most likely coseismic and probably have included subsidence as well as uplift events. Because subsidence is generally associated with erosion, almost surely more prehistoric large earthquakes occurred than are recorded as topographic steps in these terraces. We suggest that the distribution of coastal uplift and subsidence observed along the Kamchatsky Peninsula coastline is qualitatively explained by the squeezing of the Kamchatsky Peninsula block between the Bering and Okhotsk plates, and the Komandorsky Island block

Abstract-The near-field expression of the tsunami produced by the 15 November 2006 Kuril earthquake (M w 8.1-8.4) in the middle Kuril Islands, Russia, including runup of up to 20 m, remained unknown until we conducted a post-tsunami survey in the summer of 2007. Because the earthquake occurred between summer field expeditions in 2006 and 2007, we have observations, topographic profiles, and photographs from three months before and nine months after the tsunami. We thoroughly surveyed portions of the islands of Simushir and Matua, and also did surveys on parts of Ketoi, Yankicha, Ryponkicha, and Rasshua. Tsunami runup in the nearfield of the middle Kuril Islands, over a distance of about 200 km, averaged 10 m over 130 locations surveyed and was typically between 5 and 15 m. Local topography strongly affected inundation and somewhat affected runup. Higher runup generally occurred along steep, protruding headlands, whereas longer inundation distances occurred on lower, flatter coastal plains. Sediment transport was ubiquitous where sediment was availabledeposit grain size was typically sand, but ranged from mud to large boulders. Wherever there were sandy beaches, a more or less continuous sand sheet was present on the coastal plain. Erosion was extensive, often more extensive than deposition in both space and volume, especially in areas with runup of more than 10 m. The tsunami eroded the beach landward, stripped vegetation, created scours and trim lines, cut through ridges, and plucked rocks out of the coastal plain.

Paleotsunami records from the central Kuril Island segment of the Japan-Kuril-Kamchatka subduction zone indicate that the region has been frequently inundated by tsunamis. As many as 20e22 tsunami deposits are recognized on Matua Island for the past 3300 yr with an average tsunami recurrence interval of ~150 yr, and 34e36 tsunami deposits are evident on Simushir Island for the past 2350 yr with an average recurrence of ~65 yr. These intervals are short, but comparable to other segments of the Japan-Kuril-Kamchatka subduction zone. Results from all survey locations reveal shortening recurrence intervals toward the present, especially for the last 600 yr, indicating a possible preservation bias. On Simushir, tsunamis at least 11 m higher than the modern tsunamis in 2006 and 2007 occurred every ~300 yr on average. On Matua, tsunamis with slightly farther inundation than the 2006 and 2007 tsunamis occurred every ~215 yr while those with at least 100 m farther inland inundation occur every ~750 yr. Our paleotsunami record almost certainly includes tsunamis that are not from great subduction zone earthquakes in the central Kuril segment: we expect the Matua record includes volcanic tsunamis and the Simushir record includes tsunamis from the southern Kuril segment.