[Next Section] [Previous Section] [Contents]
Coseismic subsidence from a great (>8 moment magnitude) subduction zone earthquake in the Cascadia margin would produce post-earthquake hazards of coastal flooding that could persist for decades. Great subduction zone earthquakes in Alaska and Chile resulted in coastal coseismic subsidence of one to three meters (Plafker, 1972; Heaton and Hartzell, 1986). In this thesis evidence of coseismic subsidence is identified from submergence events recorded in abruptly buried late-Holocene wetland deposits, located in bays of British Columbia, Washington, Oregon, and northern California. The abrupt burial of wetlands by bay mud is indicative of coastal submergence events that have been interpreted to represent coseismic subsidence from great earthquakes (Atwater, 1987). Existing and new core and cutbank data for the most recently buried wetland deposit is compiled to establish predicted rise of relative sea-level following a Cascadia earthquake. Two bays, Grays Harbor, Washington, and the Necanicum estuary, Oregon, are analyzed for increased flooding potential in the years following an elastic response to a great earthquake.
In the Cascadia subduction zone (Figure 1), the most recent, and regionally recognized, buried wetland deposits are dated at ~300 radiocarbon years before present (RCYBP). Older and less uniformly distributed abrupt burial deposits are reported back to 5,000 years, with an average recurrence interval of 500 years (Atwater and Hemphill-Haley, 1996; Darienzo and Peterson, 1995). Submergence events are recognized by deposits with vertical successions of upper-intertidal (tidal flat-low marsh) or supratidal (high marsh-forest edge) wetlands that are buried by lower or mid-intertidal mud. The peat-mud sequences are often separated by abrupt contacts (~0.5 cm), implying coseismic subsidence from great earthquakes in the CSZ (Atwater, 1987). The most recent burial has been linked with paleoliquefaction in the lower Columbia River valley (Obermeier, 1995). Some peat-mud couplets are separated by sand deposits interpreted to have been deposited by near-field tsunami inundation (Atwater, 1987; Darienzo and Peterson, 1990; Clague and Bobrowsky, 1994).
Knowledge of the regional variation in the amount of coseismic subsidence is needed to evaluate potential hazards of post-earthquake flooding. Estimates of regional coastal subsidence are also needed to test megathrust fault models and to constrain tsunami excitation models (Savage, 1983; Hyndman and Wang, 1995; Baptista and others 1996). Studies that focus on the magnitude and frequency of flooding have been completed for most of the central Cascadia margin (Army Corp of Engineers, 1971; Harris and others, 1979; Soil Conservation Service, 1979; FEMA, 1990; Phipps, 1990). However, flooding studies within the Cascadia margin have not addressed the combined affect of storm-surge flooding following coseismic subsidence.
In addition to the compilations of existing subsidence data for an Oregon database (Darienzo, 1991; Briggs, 1994 ), new evaluations of paleosubsidence in southwestern Washington, Grays Harbor estuary, and in northwestern Oregon, the Necanicum Estuary, Tillamook Bay, and Siletz Bay are completed. These bays provide a variety of hydrographic conditions and different responses to coastal flooding (Percy and others, 1974; Peterson and others, 1984). The objectives of this study are to: (1) show the regional extent of buried wetland deposits in Oregon by compiling data from published and unpublished studies and; (2) test the internal consistency of paleotidal indicators of coseismic subsidence; (3) develop a regional model of estimated magnitude of subsidence based on the results of objective 2 using the most recently buried wetland deposit observed in Oregon and Washington; and (4) establish the potential of post seismic coastal flooding for 10 and 100 year flood events in two representative coastal settings.
Figure 1. Diagram of the Cascadia margin. Solid line with triangles represents the trench, i.e., megathrust fault exposure between subsducting Juan de Fuca plate and overriding North American plate. Longitude 130° W-126° W correspond to UTM-E zone 9. Longitude 126° W-122° W correspond to UTM E zone 10. Modified from Kulm and others, (1984).
Samples from buried wetland deposits document changes in plant macrofossils, organic content, and microfossil assemblages across the most recent (circa 1700 AD) earthquake subsidence contact. Estimates of paleo-subsidence applied to flood elevations serve as a first order analysis of what impact predicted regional subsidence has on coastal flooding. The study results indicate that prehistoric CSZ coastal subsidence can be estimated from: (1) changes in organic content; (2) changes in proportions of fresh, diatom species; and (3) observed plant macrofossil evidence across the peat-mud couplet. Regional trends of the amount of subsidence are generally related to the coastal distance from the subduction zone trench. Finally, the relative impact of coastal subsidence on potential flooding depends on the amount of subsidence and the hydrography of the tidal basin.