Active tectonics of the Seattle fault and central Puget Sound, Washington— Implications for earthquake hazards Samuel Y. Johnson* U.S. Geological Survey, M.S. 966, Box 25046, Denver Federal Center, Denver, Colorado 80225 Shawn V. Dadisman } Jonathan R. Childs U.S. Geological Survey, M.S. 999, 345 Middlefield Road, Menlo Park, California 94025 William D. Stanley U.S. Geological Survey, M.S. 966, Box 25046, Denver Federal Center, Denver, Colorado 80225 ABSTRACT segments. Regional seismic-hazard assess- SEATTLE FAULT ments must (1) incorporate new information We use an extensive network of marine on fault length, geometry, and displacement Danes et al. (1965) first suggested the presence high-resolution and conventional industry rates on the Seattle fault, and (2) consider of a significant west-trending fault in the Puget seismic-reflection data to constrain the loca- the hazard presented by the previously un- Lowland through Seattle on the basis of gravity tion, shallow structure, and displacement recognized, north-trending fault zone. data. They inferred a steeply north dipping zone rates of the Seattle fault zone and crosscut- consisting of two parallel normal faults with ting high-angle faults in the Puget Lowland INTRODUCTION about 11 km of vertical slip. Gower et al. (1985) of western Washington. Analysis of seismic briefly outlined geologic relationships across the profiles extending 50 km across the Puget The Seattle fault is a zone of thrust or reverse Seattle fault, and Yount and Holmes (1992) sug- Lowland from Lake Washington to Hood faults that strikes through downtown Seattle in gested that the fault dipped to the south and had Canal indicates that the west-trending Seat- the densely populated Puget Lowland of west- reverse displacement. Johnson et al. (1994) used tle fault comprises a broad (4–6 km) zone of ern Washington (Fig. 1). The fault coincides industry seismic-reflection data from Puget three or more south-dipping reverse faults. with large gravity and magnetic anomalies Sound to show that the Seattle fault is a broad Quaternary sediment has been folded and (Danes et al., 1965; Finn et al., 1991) and forms zone comprising south-dipping thrust or reverse faulted along all faults in the zone but is the boundary between an uplift of Tertiary faults. Johnson et al. also inferred that the Seattle clearly most pronounced along fault A, the rocks to the south and the Seattle basin to the fault has been active from about 40 Ma to the northernmost fault, which forms the bound- north (Johnson et al., 1994). The Seattle fault is present, and linked north-vergent thrust faulting ary between the Seattle uplift and Seattle considered to be active (e.g., Gower et al., to flexural subsidence in the adjacent Seattle basin. Analysis of growth strata deposited 1985; Bucknam et al., 1992); however, its pre- basin (Fig. 1). They suggested that the Seattle across fault A indicate minimum Quater- cise location, lateral geometry, displacement fault represents a restraining transfer zone be- nary slip rates of about 0.6 mm/yr. Slip rates history, and slip rates are poorly defined. We tween right-lateral shear zones near Hood Canal across the entire zone are estimated to be collected an extensive network of marine high- and the southwest Washington Cascade foothills 0.7–1.1 mm/yr. resolution seismic-reflection profiles across the (Fig. 1; Gower et al., 1985). Pratt et al. (1997) The Seattle fault is cut into two main seg- Seattle fault to better define these uncertainties proposed a complementary model in which the ments by an active, north-trending, high- and provide constraints for earthquake hazard Seattle fault is one structural component of a angle, strike-slip fault zone with cumulative assessments. Our purpose in this paper is to north-directed thrust sheet that underlies the cen- dextral displacement of about 2.4 km. Faults present the results of these marine geophysical tral Puget Lowland from the Black Hills on the in this zone truncate and warp reflections in surveys and local complementary onland inves- southwest to the southern Whidbey Island fault Tertiary and Quaternary strata and locally tigations. Previous interpretations of the Seattle (Johnson et al., 1996) on the north. coincide with bathymetric lineaments. Cu- fault were derived from conventional industry Gower et al. (1985) suggested that the large mulative slip rates on these faults may ex- seismic-reflection data (Johnson et al., 1994; thickness of Quaternary strata in the Seattle ceed 0.2 mm/yr. Assuming no other crosscut- Pratt et al., 1997) collected from Puget Sound, basin indicates possible large Quaternary offsets, ting faults, this north-trending fault zone and these data were also incorporated in our and noted an uplifted Holocene marine terrace divides the Seattle fault into 30–40-km-long analysis. Because much of the geologic frame- within the Seattle fault zone at Restoration Point western and eastern segments. Although this work of the Puget Lowland is obscured by (Fig. 2A, see insert). Bucknam et al. (1992) doc- geometry could limit the area ruptured in Quaternary deposits, dense vegetation, Puget umented as much as 7 m of uplift at Restoration some Seattle fault earthquakes, a large event Sound waterways, and urban sprawl, marine Point and inferred that it occurred during a large ca. A.D. 900 appears to have involved both seismic surveys provide critical information for (M > 7) earthquake on the Seattle fault ca. A.D. understanding the structure and evolution of 900. This earthquake was accompanied by a *E-mail: [email protected]. the region. tsunami in Puget Sound (Atwater and Moore, GSA Bulletin; July 1999; v. 111; no. 7; p. 1042–1053; 8 figures. 1042 Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/111/7/1042/3383196/i0016-7606-111-7-1042.pdf by guest on 27 September 2021 EARTHQUAKE HAZARDS, SEATTLE FAULT AND CENTRAL PUGET SOUND, WASHINGTON 124°W 122°W Vancouver Island ▼ MB ▼ SJF ▼ ▼ SJ VI SCF LRF ▼ DMF ▼ ▼ SWF DAF ▼ GP ° ▼ ▼ ▼ 48 N ▼ ▼ ▼ l l Olympic l ? l l l ▼ l l ▼ Mountains l ▼ Pacific Ocean l l ▼ l ▼ SB ? l SF l ▼ ▼ S ▼ ▼ ▼ HC V CBF ▼ T Cascade Range 50 km Puget Lowland BH O Surficial deposits ▼ (Quaternary) MR Sedimentary rocks (Paleogene to Neogene) Cascade igneous rocks Coast Range DF (Oligocene and younger) and Columbia river Basalt Group (Miocene) SHZ Crescent Formation and ▼ other Eocene volcanic rocks ▼ MSH MA Basement rocks (pre-Tertiary) Figure 1. Schematic geologic map of northwestern Washington showing the Puget Lowland and flanking Cascade Mountains, Coast Range, and Olympic Mountains. Abbreviations for cities: O—Olympia; S—Seattle; T—Tacoma; VI—Victoria. Abbreviations for faults (heavy lines), modern Cascade volcanoes (triangles), and other geologic features: BH—Black Hills; CBF—Coast Range Boundary fault; DAF—Darrington fault; DF—Doty fault; DMF—Devils Mountain fault: GP—Glacier Peak; HC—Hood Canal; LRF—Leech River fault; MA—Mount Adams; MB—Mount Baker; MR—Mount Rainier; MSH—Mount Saint Helens; SB—Seattle basin; SCF—Straight Creek fault; SF—Seattle fault; SHZ—Saint Helens zone; SJ—San Juan Islands; SJF—San Juan fault; SWF—southern Whidbey Island fault. Geology from maps and com- pilations of Tabor and Cady (1978), Washington Public Power Supply System (1981), Gower et al. (1985), Walsh et al. (1987), Whetten et al. (1988), Yount and Gower (1991), Tabor et al. (1993), and Tabor (1994). 1992), landslides in Lake Washington (Fig. 2, A along the Seattle fault in the past 16 k.y., which evant recurrence intervals could be shorter or and B; Jacoby et al., 1992; Karlin and Abella, suggests that most postglacial uplift occurred longer. After assuming a thrust-sheet model of 1992, 1996) and rock avalanches in the Olympic during the ca. A.D. 900 event and that recur- deformation, Pratt et al. (1997) used fault- Mountains (Fig. 1; Schuster et al., 1992). rence intervals for such large events must be on segment lengths and fold geometries to deduce Rates of displacement and earthquake recur- the order of several thousand years. However, an average slip rate of 0.25 mm/yr for the Seat- rence intervals for the Seattle fault are essen- Thorson (1996) also speculated that motion on tle fault over the past 40 m.y. However, other tially unknown. Thorson (1993) used elevations the Seattle fault over the past 15 k.y. may be models of Puget Sound structure are possible of glacial deltas to infer about 9 m of uplift anomalous because of deglaciation and that rel- and there is no basis for assuming that this in- Geological Society of America Bulletin, July 1999 1043 Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/111/7/1042/3383196/i0016-7606-111-7-1042.pdf by guest on 27 September 2021 JOHNSON ET AL. ferred long-term Tertiary rate applies to Quater- STRATIGRAPHY AND SEISMIC nary time. Pratt et al. also calculated the total STRATIGRAPHY Stratigraphic unit Age surface area of the Seattle fault from their model Vashon Till 15 ka and concluded that earthquakes of magnitude Tertiary Rocks 7.6–7.7 were possible. Esperance Sand Seismogenic depths below Puget Sound are Tertiary rock units imaged on seismic- Lawton Clay typically about 15–25 km (Ludwin et al., 1991). reflection data in the vicinity of the Seattle fault Drift Vashon Since 1970, when the regional seismic network include (1) mainly basaltic rocks of the lower Sediments of the became operational, the largest two earthquakes Eocene Crescent Formation, (2) unnamed nonglacial 22 ka associated with the Seattle fault include a M 5.0 Eocene marine strata, (3) turbidite sandstone and Olympia interval event that occurred at a depth of about 17 km siltstone of the upper Eocene to Oligocene Blake- beneath Point Robinson on 29 January 1995 ley Formation, and (4) nonmarine sedimentary Possession Drift 80 ka (Dewberry and Crosson, 1996), and a M 4.9 rocks of the Miocene Blakely Harbor Formation Whidbey Formation 100 ka event that occurred at a depth of 7 km beneath (Johnson et al., 1994).