Earthquakes in the past few thousand years have left signs of land-level change, tsunamis, and shaking along the Pacific coast at the Cascadia subduction zone. Sudden lowering of land accounts for many of the buried marsh and forest soils at estuaries between southern British Columbia and northern California. Sand layers on some of these soils imply that tsunamis were triggered by some of the events that lowered the land. Liquefaction features show that inland shaking accompanied sudden coastal subsidence at the Washington-Oregon border about 300 years ago. The combined evidence for subsidence, tsunamis, and shaking shows that earthquakes of magnitude 8 or larger have occurred on the boundary between the overriding North America plate and the downgoing Juan de Fuca and Gorda plates. Intervals between the earthquakes are poorly known because of uncertainties about the number and ages of the earthquakes. Current estimates for individual intervals at specific coastal sites range from a few centuries to about one thousand years.

Salt marsh subsurface deposits (0-4 rn depth) in Netarts Bay, a small coastal lagoon of northern Oregon, record six events of marsh burial in the last several thousand years. Five of the buried marsh surfaces show sharp, nonerosional upper contacts with either anomalous sand layers (tsunami deposits) or tidal fiat mud deposits. These sequences indicate episodic, abrupt subsidence of the marsh surfaces to low intertidal levels. In contrast, lower marsh contacts with underlying intertidal muds are gradafional, indicating gradual uplift and development of the marsh. Three in&pendent measures of deposit elevation relative to mean tidal level (percent organics, diatom assemblages, and percent eolian sand) have been used to estimate vertical displacements of marsh surfaces. Abrupt subsidence displacements of 1-1.5 rn alternate with gradual uplift displacements of the order of 0.5-1.0 m. The vertical tectonic movements are interpreted to reflect coseismic strain release (abrupt subsidence) following interseismic strain accumulation (gradual uplift), associated with interplate coupling between the Juan de Fuca Plate and the North American plate in the Cascadia subduction zone. Recurrence intervals between subsidence events range from possibly less than 300 years to at least 1000 years, with the last dated event likely occurring 300-400 radiocgbon years before present (RCYBP). Significant x'•C age overlaps of at least four subsidence events recorded at Netarts and reported for southern Washington and other northern

[1] The first probabilistic tsunami flooding maps have been developed. The methodology, called probabilistic tsunami hazard assessment (PTHA), integrates tsunami inundation modeling with methods of probabilistic seismic hazard assessment (PSHA). Application of the methodology to Seaside, Oregon, has yielded estimates of the spatial distribution of 100-and 500-year maximum tsunami amplitudes, i.e., amplitudes with 1% and 0.2% annual probability of exceedance. The 100-year tsunami is generated most frequently by far-field sources in the Alaska-Aleutian Subduction Zone and is characterized by maximum amplitudes that do not exceed 4m,with an inland extent of less than 500 m. In contrast, the 500-year tsunami is dominated by local sources in the Cascadia Subduction Zone and is characterized by maximum amplitudes in excess of 10 mand an inland extent of more than 1k m. The primary sources of uncertainty in these results include those associated with interevent time estimates, modeling of background sea level, and accounting for temporal changes in bathymetry and topography.N onetheless, PTHA represents an important contribution to tsunami hazard assessment techniques; viewed in the broader context of risk analysis, PTHA provides amethod for quantifying estimates of the likelihood and severity of the tsunami hazard, which can then be combined with vulnerability and exposure to yield estimates of tsunami risk.

A new approach to detect Holocene subduction-zone earthquakes combines the results from ground-penetrating radar (GPR), Vibracores, and accelerator mass spectrometry (AMS) dates from a barrier spit located west of Willapa Bay, southwest Washington. GPR data show a 10-m-thick facies of beach sand within which we identify, and Vibracores confirm, beach-parallel, wave-eroded, buried scarps mantled with multiple beds of magnetite. The eight GPR-detected buried scarps are interpreted to be eroded by minor transgressions caused by instantaneous barrier subsidence during earthquakes associated with the Juan de Fuca plate subducting under the North American plate. Of these scarps, four have been AMS dated at 300, 1110, 2540, and 4250 (radiocarbon) yr B.P. No datable material has yet been found for the other four radar-detected scarps, but we interpolate and extrapolate dates of 1800, 3400, 5000, and 5800 yr B.P.

Prehistoric great earthquakes (Mw 8-9) in the central Cascadia margin have produced coastal subsidence (0-2 m) that has resulted in chronic lowland flooding and catastrophic beach erosion. Geologic records of wetland burial in 13 bays from Washington and Oregon are used with simple calculations to estimate likely coastal flooding and beach retreat following a great Cascadia earthquake. Plant macrofossils and peat-to-mud ratios recorded in core logs discriminate between forest, marsh, colonizing marsh, and mud-flat tidal settings. Transitions between these settings, i.e., across abrupt burial contacts, demonstrate either 0+ or -0.5, 1+ or -0.5, or 2+ or -0.5 m of paleosubsidence. Paleosubsidence from the last Cascadia event (AD 1700) decreases from 2+ or -0.5 m in the eastern reaches of south-west Washington bays to 0+ or -0.5 m in the western reaches of central Oregon bays. First-order estimates of post-subsidence flooding hazards are based on the predicted regional subsidence added to current 10- and 100-year flood elevations. At least 525 km of bay shorelines are threatened by chronic flooding following coseismic subsidence. Catastrophic beach retreat is estimated from Bruun's Rule to range from 50 to 300 m depending on assumed depth of closure, measured beach-berm heights, and predicted coseismic subsidence along the margin. Shorelines that are susceptible to catastrophic beach retreat from coseismic subsidence total at least 250 km in longshore distance.