Cannon Beach Tsunami Hazard Research Project
The Cannon Beach tsunami hazard assessment began in 2005 with support from the National Tsunami Hazard Mitigation Program of the National Oceanic and Atmospheric Administration (NOAA). The project is managed by DOGAMI. Cooperating investigators are Y. Joseph Zhang and Antonio Baptista of the Center for Coastal Margin Observation & Prediction of Oregon Health & Science University, Chris Goldfinger of Oregon State University, and Kelin Wang of the Pacific Geoscience Centre, Sydney, B.C. A final report summarizing findings is in preparation and will be released as a DOGAMI Special Paper. The following are some illustrations and a draft abstract from the Special Paper.
Location of the Cannon Beach study area (right panel) relative to major offshore tectonic plates and plate boundaries (left panel). Contour lines in red and blue in left panel are simulated coseismic deformation from a Cascadia earthquake source scenario used in the investigation for tsunami modeling (Large 9) . Contours are in 0.5-m intervals and depict the north-south extent of scenario tsunami sources used in this study; red to orange lines = uplift; blue lines = subsidence; FZ = Fracture Zone; SZ = subduction zone.
Photographs showing the impact of the 1964 Alaska tsunami at Cannon Beach (courtesy of the Cannon Beach Historical Society). (A) Oblique aerial photograph of the lower Elk Creek valley (now Ecola Creek) that flows through downtown Cannon Beach. Decking from the old Elk Creek bridge was torn from its abutments and transported 300 m upstream. A foundation in the lower part of the photo marks the original postition of a house that was carried 400 m upstream and deposited between the bridge sections. (B) View to the northwest across Elk Creek showing bridge pilings and piers, all that remain after the tsunami destroyed the Elk Creek bridge in 1964. Bell Harbor Motel can be seen in the distance across the creek. (C) The Bell Harbor Motel suffered considerable damage from flooding during the tsunami, including broken windows, water damage and destruction caused by drift logs. The roof of a different building was left in the front yard of the Motel after being carried several hundred yards by the waves. (D) View to the southeast looking across Elk Creek at the bridge remains and house transported hundreds of meter inland by the tsunami. Figure is from Witter (2008).
Four panels showing preliminary rupture modes identified using turbidite correlation of Goldfinger et al. (2008) for a ~3,000-year period; additional events for the last ~10,000 years are described by Goldfinger et al. (in preparation). (A) Full rupture, represented at all sites by seven events. (B) Mid-southern rupture, represented at all sites as far south as 40.7° N by five events. (C) Southern rupture from central Oregon southward to at least 41° N, represented by one event. (D) Southern Oregon/northern California events, represented by one event. Rupture terminations appear to occur at three forearc structural uplifts: Nehalem Bank (NB), Heceta Bank (HB), and Coquille Bank (CB ). Latitudinal boundaries of episodic tremor and slip (ETS) events proposed for the downdip subduction interface (Brudzinski et al., 2007) are shown by white dashed lines. Figure is taken from Goldfinger et al. (2008).
Abstract (October 6, 2008 draft)
To update the tsunami hazard assessment method for Oregon, we (1) evaluate geologically reasonable variability of the earthquake rupture process on the Cascadia megathrust, (2) compare those scenarios to geological and geophysical evidence for plate locking, (3) specify 25 deterministic earthquake sources, and (4) use the resulting vertical coseismic deformations as initial conditions for simulation of Cascadia tsunami inundation at Cannon Beach, Oregon. Because of the Cannon Beach focus, the north-south extent of source scenarios is limited to Neah Bay, Washington to Florence, Oregon. We use the marine paleoseismic record to establish recurrence bins from the 10,000 year event record and select representative coseismic slips from these data. Assumed slips on the megathrust are 8.4 m (290 yrs of convergence), 15.2 m (525 years of convergence), 21.6 m (748 years of convergence), and 37.5 m (1298 years of convergence) which, if the sources were extended to the entire Cascadia margin, give Mw varying from approximately 8.3 to 9.3. Additional parameters explored by these scenarios characterize ruptures with a buried megathrust versus splay faulting, local versus regional slip patches, and seaward skewed versus symmetrical slip distribution. By assigning variable weights to the 25 source scenarios using a logic tree approach, we derived percentile inundation lines that express the confidence level (percentage) that a Cascadia tsunami will NOT exceed the line. Lines of 50, 70, 90, and 99 percent confidence correspond to maximum runup of 8.9, 10.5, 13.2, and 28.4 m (NAVD88). The tsunami source with highest logic tree weight (preferred scenario) involved rupture of a splay fault with 15.2 m slip that produced tsunami inundation near the 70 percent confidence line. Minimum inundation consistent with the inland extent of three Cascadia tsunami sand layers deposited east of Cannon Beach within the last 1000 years suggests a minimum of 15.2 m slip on buried megathrust ruptures. The largest tsunami run-up at the 99 percent isoline was from 37.5 m slip partitioned to a splay fault. This type of extreme event is considered to be very rare, perhaps once in 10,000 years based on offshore paleoseismic evidence, but it can produce waves rivaling the 2004 Indian Ocean tsunami. Cascadia coseismic deformation most similar to the Indian Ocean earthquake produced generally smaller tsunamis than at the Indian Ocean due mostly to the 1 km shallower water depth on the Cascadia margin. Inundation from distant tsunami sources was assessed by simulation of only two Mw 9.2 earthquakes in the Gulf of Alaska, a hypothetical maximum considered event developed by the Tsunami Pilot Study Working Group (2006) and the historical maximum, the tsunami from the 1964 Prince William Sound Earthquake; maximum runups were, respectively, 12.4 m and 7.5 m.
Tsunami runup on the order of 30-35 m near the epicenter of the great Sumatra-Andaman Islands earthquake of 26 December 2004 (Tsuji, et al., 2005) and geologic similarities between the Cascadia subduction zone and the Sunda-Andaman subduction zone (Dengler, 2006, Goldfinger and McNeill, 2006) prompted the question: Could local tsunami of this size be generated by the Cascadia subduction zone? An ancillary question is how large is the hazard posed by distant tsunami? This paper addresses the first question through comprehensive earthquake source characterization of the north-central Cascadia subduction zone and numerical simulations of tsunami flooding at Cannon Beach, Oregon. Cannon Beach was chosen for its relatively small size, variety of topography, and known Cascadia tsunami deposits. Inland reach of paleo-tsunami deposits places a minimum constraint on Cascadia inundation. The second question was answered by simulation of historical inundation by the 1964 Alaska tsunami and a hypothetical worst-case tsunami also originating from the Gulf of Alaska. The 1964 simulation provided a benchmark test of the tsunami simulation model, since a rich record of historical observations was available. The final objective was development of a cost effective approach for tsunami hazard assessment of the entire Oregon coast.
Goldfinger, C., Grijalva, K., Buergmann, R., Morey, A. E., Johnson, J. E., Nelson, C. H., Gutierrez-Pastor, J., Ericsson, A., Karabanov, E., Chaytor, J. D., Patton, J. A., and Gracia, E., 2008, Late Holocene rupture of the northern San Andreas Fault and possible stress linkage to the Cascadia Subduction Zone: Bulletin of the Seismological Society of America, v. 98, no. 2, p. 861-889.
Goldfinger, C., Nelson, C. H., Morey, A., Johnson, J.E., Gutierrez-Pastor, J., Eriksson, A. T., Karabanov, E., Patton, J., Gracia, E., Enkin, R., Dallimore, A., Dunhill, G., in preparation, Turbidite Event History: Methods and Implications for Holocene Paleoseismicity of the Cascadia Subduction Zone, USGS Professional Paper 1661-F 130p, 56 Figures.
Witter, R.C., 2008, Prehistoric Cascadia tsunami inundation and runup at Cannon Beach, Clatsop County, Oregon: Oregon Department of Geology and Mineral Industries technical report to the Cannon Beach Rural Fire Protection District: Oregon Department of Geology and Mineral Industries Open-File Report, in preparation.
Oregon Department of Geology and Mineral Industries
800 NE Oregon Street, Suite 965, Portland, OR 97232-2162
(971) 673-1555, FAX (971) 673-1562
email us at DOGAMI
Your best source for outdoor recreation and natural resource information, plus the largest selection of maps in the Northwest.