CEE 437: Geophysics

Geophysical Methods -- measure a physical, electrical, or chemical property of soil, rock and pore fluids • Magnetic • Electrical Resistance • Temperature • Gravity • Ground Penetrating Radar • Seismic Wave Propagation

o Reflection o Refraction o manmade and natural source

Seismic Refraction commonly used to find depth to water table or bedrock

For a dipping layer, the dip, δ, can be calculated by reversing profile.

Multiple Layers

Seismic Reflection primarily used by oil companies to explore depths greater than 100 m

Seattle Fault Geophysics Seismic Reflection – to identify bedrock depths and sediment layers Gravity – to identify basin and uplift depths and positions Aeromagnetics – to identify basaltic conglomerate marker-bed Micro-seismicity – to identify faults and seismic slip

170 Geological Society of America Bulletin, February 2002

BLAKELY et al.

Figure 1. Regional setting. Large-scale map shows isostatic residual gravity over the Se-attle basin and surrounding areas, where gravity lows reflect thick sections of basin-fillingdeposits. Faults generalized from Yount and Gower (1991) and Johnson et al. (1996).Crosshatch pattern indicates urbanized areas. S—Seattle, T—Tacoma, E—Everett, B—Bremerton. Inset: Dotted rectangle shows area of large-scale map of Figure 1; dashedrectangle indicates area of Figures 2, 3, 4, and 6.

Figure 2. (A) Aeromagnetic anomalies overthe Seattle fault zone and Seattle uplift.Letters A, B, and C indicate anomalies dis-cussed in text. Dotted line shows location ofmagnetic profile (Fig. 5). (B) Aeromagneticanomalies filtered in order to emphasizeshallow magnetic sources. Data from Awere continued upward 50 m, then sub-tracted from the original data.

M

strands have remained uncertain along most ofits length.

The frontal fault of the Seattle fault zonewas the likely source of a M 7 earthquake thatoccurred ;1100 yr ago (A.D. 900–930), caus-ing tectonic uplift (Bucknam et al., 1992),landslides (Jacoby et al., 1992), and a localtsunami (Atwater and Moore, 1992). Upliftpatterns from that earthquake are consistentwith the south-side-up model for the fault.Field evidence from Bainbridge Island (Buck-nam et al., 1992) indicates that the pre-upliftshoreline at Restoration Point, ;1.5 km southof Eagle Harbor, was 7 m lower than it is to-day, whereas the pre-uplift shoreline at amarsh near Winslow, on the north side of Ea-gle Harbor, was 1.5 m higher (R.C. Bucknam,2001, written communication). Thus, near Ea-gle Harbor, an active strand of the Seattle faultzone must lie within narrow spatial limits nearthe topographic surface. A similar pattern isseen on the east side of Puget Sound: At AlkiPoint, south of the frontal fault, the pre-upliftshoreline was 6 m lower than it is today (R.C.Bucknam, 2001, written commun.), whereasat West Point north of the frontal fault, thepre-uplift shoreline was 3 m higher (Atwaterand Moore, 1992). Considering that sea levelhas risen ;1 m since the uplift, net motionwas dominantly south-side up both west andeast of Puget Sound.

This relatively simple thrust-fault model iscomplicated by the recent discovery of aneast-striking scarp on Bainbridge Island(Bucknam et al., 1999), referred to as the ToeJam Hill scarp. Contrary to the long-term his-tory on the Seattle fault zone, the topographicexpression along the Toe Jam Hill scarp isconsistent with a north-side-up fault, and re-cent geologic field evidence confirms this in-terpretation (Nelson et al., 1999). Moreover, aM 4.9 earthquake that occurred near Bremer-ton in 1997 had a focal mechanism also con-sistent with north-side-up movement (Weaveret al., 1999; T.M. Van Wagoner, R.S. Crosson,K.C. Creager, G. Medema, and L. Preston,2001, personal commun.), and other relocatedearthquake hypocenters throughout the area ofthe Seattle fault zone have components ofnorth-side-up motion (T.M. Van Wagoner, R.S.Crosson, K.C. Creager, G. Medema, and L.Preston, 2001, personal commun.). The impli-cations of the Toe Jam Hill fault and recentearthquakes are discussed subsequently.

The location of the deformation front andseveral thrusts in the Seattle fault zone are rea-sonably well determined in Puget Sound andother waterways by marine seismic-reflectionstudies (Pratt et al., 1997; Johnson et al., 1994,1999). Geologic mapping (Yount and Gower,1991) shows that the Tertiary strata in thehanging wall of the fault are dipping steeply

to the north, and sparse outcrops can be tracedeastward along strike for .50 km. However,the cover of young glacial deposits, water, andvegetation makes it difficult to map the pre-cise location and configuration of the Seattlefault zone between the widely spaced seismic-reflection crossings, particularly beneath thehighly developed regions of Seattle, Bremer-ton, and Bellevue. For these reasons, the Se-attle area is an excellent candidate for high-resolution potential-field studies.

AEROMAGNETIC INTERPRETATION

The Seattle uplift (Fig. 1) is underlain atshallow depth by a complex package of Eo-cene and younger volcanic and sedimentaryrocks. The contrasting magnetic properties ofthese rocks are ideal for aeromagnetic map-ping of structures in the middle and uppercrust. Along the Seattle fault zone, a distinc-tive pattern of magnetic anomalies follows theeastward trend of bedrock in the upthrownblock and reliably reflects the underlying,steeply dipping stratigraphy.

Strands of the Seattle Fault Zone

Aeromagnetic data over the southern mar-gin of the Seattle basin (Fig. 2) display a pack-age of three east-trending magnetic anomalies.From north to south, they consist of an elon-gate, narrow magnetic high, a broad magneticlow, and a complex magnetic high; their east-west extent is .50 km. The northern anomaly(anomaly A in Fig. 2) is remarkably linear andnarrow; it trends east from Dyes Inlet to PugetSound and from Lake Washington to 10 kmeast of Lake Sammamish. On Bainbridge Is-land, this anomaly directly overlies a basaltconglomerate within the Miocene fluvial de-posits of the Blakely Harbor Formation (Ful-mer, 1975), which strikes east and dips 728–808N. East of Lake Washington, a similaranomaly also is caused by a steeply dippingMiocene volcanic conglomerate correlativewith the Blakely Harbor Formation. Theserocks were presumably deposited in the Se-

Geological

Societyof

Am

ericaB

ulletin,February

2002171

SEA

TT

LE

FAU

LT

ZO

NE

,W

ASH

ING

TO

N

Geological Society of America Bulletin, February 2002 175

SEATTLE FAULT ZONE, WASHINGTON

Figure 5. Magnetic model of the Seattlefault zone. (A) Calculated and observedmagnetic profiles. See Figure 2A for profilelocation. Profile extends north-south alongthe east shore of Bainbridge Island (see Fig.4 for location). A, B, and C indicate anom-alies discussed in text. (B) Magnetic model.

Sources are assumed to be infinitely extended in the east and west directions. Dip of frontalfault (508S) determined from seismic-reflection data (ten Brink et al., 1999; Johnson etal., 1999). Dip (748N) of Miocene conglomerate (labeled by its magnetization M 5 0.8 A/m) from geologic mapping (Yount and Gower, 1991). The Crescent Formation is the unitwith a magnetization of 1.7 A/m. (C) Interpretation favoring a structural connection be-tween Bainbridge Island earthquakes and Holocene scarp. OPF—Orchard Point fault,BHF—Blakely Harbor fault, FF—frontal fault, S—Holocene scarp seen in lidar topo-graphic data. (D) Interpretation favoring the Holocene scarp as a back thrust to the mainthrust sheet. First-motion solution and earthquake locations from Weaver et al. (1999).

found that nearly 60% of hypocenters fall be-tween 15 and 25 km, with a mean depth of 17.6km, placing most crustal earthquakes in the Se-attle region within Crescent Formation base-ment. Within the Seattle fault zone, hypocen-ters depart significantly from this depth range,forming a near-vertical zone that extends wellabove the 15 km depth to near the topographicsurface. These shallow earthquakes do not liealong the thrust fault predicted from our mag-netic model, nor along thrust faults describedin earlier studies (e.g., Johnson et al., 1994;Pratt et al., 1997; Brocher et al., 2001). Thisrecent seismic activity may reflect one or morebasin-loading faults stressed by the advancingwedge of the thrust sheet, or deformation with-in an upper plate that moves along a nearlyhorizontal plane of slip. In any case, the patternof recent seismicity appears to be a departurefrom the long-term deformational history of theSeattle fault zone.

SUMMARY

New, high-resolution aeromagnetic dataprovide constraints on the location, length,and geometry of the Seattle fault zone. Thecorrelation of aeromagnetic anomalies withtilted upthrown-block stratigraphy defines thelocation of the Seattle fault zone within nar-row limits over a distance of 50 km. Thus de-termined, the fault zone on Bainbridge Islandcoincides with recent seismicity, a postglacialfault scarp, and the M 7 earthquake that oc-curred ;1100 yr ago (A.D. 900–930) on theSeattle fault zone. The details of the relation-ship between earthquake sources in the foot-wall and those in the hanging wall and the roleof the frontal fault at depth remains to be re-solved, as well as the kinematic relationshipsamong current earthquakes, paleoseismic evi-dence, and fault geometry.

APPENDIX

The aeromagnetic survey (Blakely et al., 1999)was flown along north-south lines spaced 400 mapart and along east-west control lines spaced 8 kmapart. Flight altitude was 250 m above terrain, or aslow as permitted by safety considerations. A theo-retical flight surface, based on a digital topographicmodel, was computed in advance of the survey, andreal-time, differentially corrected Global Position-ing System (GPS) navigation was used during flightto maintain the desired flight surface. Two ground-based magnetometers were used to monitor and cor-rect for time-varying magnetic fields. Total-fieldanomalies were computed based on the Internation-al Geomagnetic Reference Field updated to the dateof the survey.

ACKNOWLEDGMENTS

We are grateful to Ralph Haugerud, Tom Pratt,Brian Sherrod, Derek Booth, Kathy Troost, Tom

172 Geological Society of America Bulletin, February 2002

BLAKELY et al.

Figure 3. Geologic compilation of the area including the Seattle fault zone and Seattle uplift, simplified from Yount and Gower (1991),Frizzell et al. (1984), and Tabor et al. (1993). Seismic-reflection interpretation from Johnson et al. (1999). Uplift data from Atwater andMoore (1992) and R.C. Bucknam (2001, written commun.); values reported as elevation of pre-uplift (or pre-subsidence) shorelinerelative to present mean highest high water (MHHW). White dots indicate magnetic contacts interpreted from aeromagnetic data anddiscussed in text. A—Alki Point, D—Duwamish River, B—Beacon Hill, W—West Seattle, M—Mercer Island, E—Elliott Bay, R—Restoration Point, EH—Eagle Harbor, DI—Dyes Inlet.

172 Geological Society of America Bulletin, February 2002

BLAKELY et al.

Figure 3. Geologic compilation of the area including the Seattle fault zone and Seattle uplift, simplified from Yount and Gower (1991),Frizzell et al. (1984), and Tabor et al. (1993). Seismic-reflection interpretation from Johnson et al. (1999). Uplift data from Atwater andMoore (1992) and R.C. Bucknam (2001, written commun.); values reported as elevation of pre-uplift (or pre-subsidence) shorelinerelative to present mean highest high water (MHHW). White dots indicate magnetic contacts interpreted from aeromagnetic data anddiscussed in text. A—Alki Point, D—Duwamish River, B—Beacon Hill, W—West Seattle, M—Mercer Island, E—Elliott Bay, R—Restoration Point, EH—Eagle Harbor, DI—Dyes Inlet.

176 Geological Society of America Bulletin, February 2002

BLAKELY et al.

Figure 6. Upper-plateearthquakes of the Seat-tle fault zone. (A) Epicen-ters of M 1.5 and largerearthquakes occurringsince January 1980 areshown by circles, wheresize of circle indicates rel-ative magnitude. Dottedpurple line shows pre-dicted traces of the Seat-tle fault zone, taken fromFigure 4. (B) Cross sec-tion across the Seattlefault zone. Earthquakeswithin pink box of Figure6A projected parallel tofault zone. S—Holocenescarp seen in lidar topo-graphic data, FF—fron-tal fault (see also Fig. 5,C and D, for relationshipto other faults). Thedashed line at 15 km em-phasizes the fact thatmost hypocenters are be-low this depth.

Brocher, and Uri ten Brink for discussions thathelped formulate many of the ideas in this paper.Early reviews by Bob Jachens and Tom Brocher andlater reviews by Bob Crosson, Silvio Pezzopane,Peter La Femina, and Chuck Connor were particu-larly helpful.

REFERENCES CITED

Atwater, B.F., and Moore, A.L., 1992, A tsunami ;1000yr ago in Puget Sound, Washington: Science, v. 258,p. 1614–1617.

Blakely, R.J., and Simpson, R.W., 1986, Approximating

edges of source bodies from magnetic and gravityanomalies: Geophysics, v. 51, p. 1494–1498.

Blakely, R.J., Wells, R.E., Yelin, T.S., Madin, I.P., and Bee-son, M.H., 1995, Tectonic setting of the Portland-Van-couver area, Oregon and Washington: Constraintsfrom low-altitude aeromagnetic data: Geological So-ciety of America Bulletin, v. 107, p. 1051–1062.