Holocene Tectonics and Fault Reactivation in the Foothills of the North Cascade Mountains, Washington

Seeing the True Shape of Earth’s Surface themed issue

Holocene tectonics and fault reactivation in the foothills of the north Cascade Mountains, Washington

Brian L. Sherrod1, Elizabeth Barnett1, Elizabeth Schermer2, Harvey M. Kelsey3, Jonathan Hughes4, Franklin F. Foit, Jr.5, Craig S. Weaver1, Ralph Haugerud1, and Tim Hyatt6 1U.S. Geological Survey, Department of Earth and Space Sciences, University of Washington, Box 351310, Seattle, Washington 98195, USA 2Geology Department, Western Washington University, Bellingham, Washington 98225, USA 3Department of Geology, Humboldt State University, Arcata, California 95521, USA 4Department of Geography, University of the Fraser Valley, 33844 King Road, Abbotsford, British Columbia V2S 7M8, Canada 5School of Earth and Environmental Sciences, Washington State University, Pullman, Washington 99164-2812, USA 6Nooksack Tribe, 5016 Deming Road, PO Box 157, Deming, Washington 98244, USA

ABSTRACT Creek and Canyon Creek faults formed in McCaffrey et al., 2007). GPS rates are higher in the early to mid-Tertiary as normal faults the southern forearc (6–10 mm/yr) and decrease We use LiDAR imagery to identify two and likely lay dormant until reactivated as to the north (Fig. 1C). With geodetic rates falling fault scarps on latest Pleistocene glacial out- reverse faults in a new stress regime. The to near zero just north of the United States–Brit-

wash deposits along the North Fork Nook- most recent earthquakes—each likely Mw > ish Columbia border, questions remain regarding sack River in Whatcom County, Washington 6.3 and dating to ca. 8050–7250 calendar the activity of faults in the northern Puget Low- (United States). Mapping and paleoseismic years B.P. (cal yr B.P.), 3190–2980 cal. yr land (Washington). investigation of these previously unknown B.P., and 910–740 cal. yr B.P.—demonstrate Forearc basins and uplifts in western Wash- scarps provide constraints on the earth- that reverse faulting in the northern Puget ington—from south to north, the Tacoma, quake history and seismic hazard in the Lowland poses a hazard to urban areas Seattle , and Everett Basins—are deﬁ ned mainly northern Puget Lowland. The Kendall scarp between Seattle (Washington) and Vancou- on the basis of high-amplitude geophysical lies along the mapped trace of the Boul- ver, British Columbia (Canada). anomalies, with faults hypothesized where the der Creek fault, a south-dipping Tertiary gradients are strongest (Blakely et al., 2002; normal fault, and the Canyon Creek scarp INTRODUCTION Brocher et al., 2001; Danes et al., 1965). Many lies in close proximity to the south-dipping of the faults found between the basins and the Canyon Creek fault and the south-dipping Geologic and geodetic data show that near- adjacent uplifts are active and present substan- Glacier Extensional fault. Both scarps are surface faults in the Cascadia forearc accom- tial seismic hazard to the region (Fig. 1B). The south-side-up, opposite the sense of displace- modate north-south shortening. Models of the best-known of these basin-bounding faults are ment observed on the nearby bedrock faults. forearc (Fig. 1A) show a series of migrating, the Tacoma, Seattle, Southern Whidbey Island, Trenches excavated across these scarps clockwise-rotating forearc blocks (Wells and and Darrington–Devils Mountain faults (Buck- exposed folded and faulted late Quaternary Simpson, 2001; Wells et al., 1998). This clock- nam et al., 1992; Johnson et al., 2004a, 2004b, glacial outwash, locally dated between ca. wise rotation causes convergence in western 1994; Sherrod, 2001; Sherrod et al., 2008, 12 and 13 ka, and Holocene buried soils and Washington (United States) where the Oregon 2004). The Darrington–Devils Mountain fault scarp colluvium. Reverse and oblique fault- Coast Range block impinges on Tertiary vol- forms a broad boundary between the northern ing of the soils and colluvial deposits indi- canic rocks and sediments, compressing these edge of the Everett Basin and accreted Meso- cates at least two late Holocene earthquakes, Tertiary rocks against the southern edge of the zoic rocks and Tertiary sedimentary rocks in the while folding of the glacial outwash prior to British Columbia (Canada) Coast Mountains. adjacent uplift. North of this uplift lies the Bell- formation of the post-glacial soil suggests This compression results in a series of struc- ingham forearc basin, the northernmost forearc an earlier Holocene earthquake. Abrupt tural basins separated by uplifts in the northern basin. The Bellingham Basin preserves Eocene changes in bed thickness across faults in the Cascadia forearc. Geological models show that to Quaternary sedimentary rocks in its interior Canyon Creek excavation suggest a lateral blocks in the northern Cascadia forearc move and is bounded on the north by pre-Tertiary component of slip. Sediments in a wetland northward relative to the Coast Mountains at sedimentary and metamorphic rocks. The Bell- adjacent to the Kendall scarp record three rates of 7–9 mm/yr (Wells and Simpson, 2001; ingham Basin is anomalous compared to the pond-forming episodes during the Holocene— Wells et al., 1998). other forearc basins in western Washington we infer that surface ruptures on the Boul- Geodetic studies show that the region between because prior to our studies, no known active der Creek fault during past earthquakes 46.5°N and 49.5°N is undergoing north-south faults lined the basin margins, yet geologic and temporarily blocked the stream channel shortening averaging ~3 mm/yr to 4.4 mm/yr geodetic data suggest that the basin should be and created an ephemeral lake. The Boulder (Hyndman et al., 2003; Mazzotti et al., 2002; the locus of active faulting (Kelsey et al., 2012).

Geosphere; August 2013; v. 9; no. 4; p. 827–852; doi:10.1130/GES00880.1; 16 ﬁ gures; 5 tables; 1 supplemental ﬁ le. Received 26 October 2012 ♦ Revision received 7 May 2013 ♦ Accepted 17 June 2013 ♦ Published online 16 July 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

Figure 1. (A) Location map of 124°W 121°W B the Paciﬁ c Northwest, showing A the Cascadia subduction zone, GPS Profile (Fig. 1C) BoundaryBoundary ofof FFraserraser - PPugetuget LLowlandowland BRFB CBCB NORTH R mobile forearc blocks, and British F AMERICAN Columbia stable North America (WA— PLATE VancouverV Isl. V a Washington; OR—Oregon; 50°N n BFGFB Coast Mtns c FG Penticton o CRC F B.C.B.C. buttress. uv R OWL—Olympic-Wallowa lin- 49°N e FFFF C r a Is BBBB U.S.U.S. s l. eament). Thumbtack indicates c AreaArea ofof FigureFigure 11BB CLFCL a F B d i that the position of crust shown a 36 s WA

mm/yr b

d in gray relatively is ﬁ xed based SCFS u OWLOW C c L

t F

i on very-long-baseline inter- JUAN DE o regon DDMFD 45° O D n OR AreaArea ooff MF

FUCA PLATE z

ferometry (vlbi) results from o FigureFigure 2 LRFLRF SWIFS EEBB ne Cascade 48°N W Oregon I Penticton, British Columbia, volcanic arc F Coast PacificPacific Range OlympicOlympic which indicate no resolvable PACIFIC OceanOcean MtnsMtns PLATE SBSB NBNB change of position with respect Pacific 53 mm/yr SFSF Ocean Nevada TFTF MBMB to stable North America (Wells °

4 0 50 km et al., 1998). (B) Enlarged map 200 km Californiai OFOF lif TBTB showing geologic and tectonic 130°W 125° 120° features in northwestern Wash- ington and southwestern Brit- SN ish Columbia. (C) Residual C Geologic Units – 1A GPS shortening rates (after 10 Residual GPS removing elastic subduction Sierra Nevada block zone component from GPS Coast Range block 8 signal) from southern Oregon Basin and Range 6 to southern British Columbia Columbia River Basalt Group (redrawn following Mazzotti 4 Washington forearc block et al., 2002). V North equals northward-directed velocities; Basement rocks V north (mm/yr) 2 Extensional magmatism error bars show 95% conﬁ dence 0 level; gray shaded area shows Symbols – 1A and 1B 95% conﬁ dence level for aver- Forearc block motion 44 46 48 50 age velocity. Abbreviations: Lattitude (°N) B—Bellingham; BB—Belling- Relative motion (amount indicated) Geologic Units - 1B ham Basin; BRF—Beaufort North American plate “fixed” Range fault; BFGF—Benson Quaternary sediments Quaternary volcano fault–Ganges fault system; Quaternary volcanics CB—Comax Basin; CRFF— Structural basin (inferred from gravity anomalies) Cameron River–Fulford fault; Tertiary continental deposits Fault, black where Holocene CLF—Cowichan Lake fault; Tertiary marine deposits DDMF—Darrington–Devils movement is known or suspected Mountain fault; EB—Everett Concealed fault, black where Tertiary volcanic rocks (excl. Siletzia) Basin; LRF—Little River Holocene movement is known or Volcanic rocks of Siletzia suspected fault; MB—Muckleshoot Basement rocks (pre-Tertiary and intrusive rocks) Basin; NB—North Bend Basin; OF—Olympia fault; SCF— Straight Creek fault; SB—Seattle Basin; SF—Seattle fault; SWIF—Southern Whidbey Island fault; TB—Tacoma Basin; TF—Tacoma fault; V—Vancouver (Vancouver Island faults from England and Calon, 1991). Digital geologic data from Washington Division of Geology and Earth Resources (2005).

This paper demonstrates active faulting on a dence for multiple Holocene earthquakes in an A broad lowland repeatedly glaciated in the set of recently discovered fault scarps along the area of the forearc previously thought to be tec- Quaternary occupies the region (Booth, 1994; eastern edge of the Bellingham Basin in north- tonically quiescent. Easter brook, 1985). Marine waters of Puget western Washington (Fig. 2). These scarps lie Sound partially occupy large areas of this low- adjacent to and represent reactivation of previ- BACKGROUND GEOLOGY land, and Quaternary glacial deposits partially ously mapped bedrock normal faults (Fig. 2B). ﬁ ll the lowland. These glacial deposits uncon- Observations collected from LiDAR mapping, Forearc basins in western Washington lie formably drape an older set of Tertiary volcanic fault-scarp excavations, and sediment cores between the Olympic Mountains to the west and and sedimentary rocks preserved in the forearc from a wetland adjacent to one scarp show evi- the Cascade Mountains to the east (Fig. 2A). basins and adjacent uplifts.

828 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

123.0°W 122.0°W A C A

49.5°N S C A B D E

BRITISHBRITISH M COLUMBIACOLUMBIA O StraitS of GeGeorgia VaVa t ra U it o f G N e oorgia rg T i a CDCD A PRPR SSSS CanadaCanada AArearea ofof FigureFigure 3 I N BBBB U.S.U.S. KKendallendall BB GGEXEX S A TRTR A′ CCCFCF

d GIGGII BellinghamBellingham n BCFBCF

a l Mt.Mt. BBakeraker

s I PPDD

v u ′ o B

a VancouverV Island SJISJI WASHINGTONWASHINGTON ViVi 48.5°N DDDMFDMF 0 25 km

StraitStraiStraitt ofof JuanJu ofan Juandede FucaFu deca Fuca SYMBOLS GEOLOGIC UNITS Quaternary active fault Fault (dashed where inferred) Quaternary basin (geophysically defined) Late Quaternary continental deposits Quaternary volcano (mostly glacial) Exploration borehole Quaternary volcanic rocks and lahars Earthquakes M<2 M2–4 M>4 Neogene Boundary Bay Formation (Hopkins, 1968; only in subsurface on B A Chuckanut Fm. member names shown A′ NW Boulder Creek fault below are from Lapen (2000) SE cross section) 900 Padden 600 member Eocene – Oligocene sedimentary deposits llsls ultrul aal tr F . a e . r 300 m pl r bbr SSlliidede a a b m ffii mbrm n c MapleM e mmbr.br. Tertiary volcanic rocks 0 d Slliide

Elevation (m) d a Sl metavmetavoolcalcannicic PaddenP m ide membermmbr.br. rockrocks ? Sllideide ? mbr. Cretaceous sedimentary rocks 0 900 Glacier extensional fault Distance (m) Pre-Tertiary igneous rocks C Pre-Tertiary metamorphic rocks

NW SE CD TR Bay BBPR ′

Bellingham PD Burrard Inlet Fraser River Fraser River False Creek

SS BB Mtn. fault Vedder 0 Sumas Mtn. fault 2.5 5 2x vertical exaggeration Depth (km) 0 5 10 20 25 Distance (km)

Figure 2. (A) Geology of the Bellingham Basin region, showing earthquake locations for the upper 20 km of crust and known active faults (black lines). Gray lines are mapped faults but are not known as active. The purple dotted line shows the outline of the active Quaternary Bellingham Basin (Kelsey et al., 2012). (B) Cross section A–A′ of bedrock through the Boulder Creek fault on Sumas Mountain (see Fig. 3 for location of cross section) using geologic data of Dragovich et al. (1997a). Location of Glacier Extensional fault (GEX) is inferred from Tabor et al. (2003). (C) Cross section B–B′ through the Bellingham Basin using borehole data from Mustard and Rouse (1994). See Figure 2A for location of cross section. Abbreviations used: Vi—Victoria; Va—Van- couver; GI—Gulf Islands; SJI—San Juan Islands; PR—Richﬁ eld-Pure Point Roberts borehole; CD—Conoco-Dynamic Mud Bay borehole; SS—Richﬁ eld-Pure Sunnyside borehole; BB—AHEL Birch Bay borehole; TR—AHEL Terrell borehole; PD— Pelican Dome borehole; BCF—Boulder Creek fault; CCF—Canyon Creek fault; DDMF—Darrington–Devils Mountain fault.

Geosphere, August 2013 829

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

Three groups of rocks and sediments char- Miocene Boundary Bay Formation (Hopkins, Kendall, Washington, where LiDAR mapping acterize the bedrock geology of the study area 1968; Mustard and Rouse, 1994). Lastly, Qua- revealed evidence for surface faulting along the (Figs. 2 and 3). First, pre-Tertiary metamorphic ternary glacial and nonglacial deposits blanket Boulder Creek fault, a fault originally described and igneous rocks underlie much of the area most of the region below 350 m in elevation. as a basin-bounding normal fault by Miller and north of the Nooksack River, and in places, The Quaternary Bellingham Basin, a smaller Misch (1963). minor amounts of sedimentary deposits cover vestige of the larger Tertiary Georgia Basin, lies these rocks. Second, the primary structural low west of the Cascade Mountains between the Boulder Creek Fault in the region, the Bellingham Basin (Fig. 2C), San Juan Islands in Washington and the Coast preserves Tertiary sedimentary rocks, includ- Range Mountains north of Vancouver, Brit- The Boulder Creek fault (Figs. 2B and 3) was ing the Eocene Chuckanut Formation, Eocene ish Columbia (Fig. 2A). Here we focus on the formally deﬁ ned as a south-dipping normal fault and Oligocene Huntington Formation, and mid- northeastern part of the Bellingham Basin near that bounds the northern margin of the Chuckanut

122.3°W 122.2°W 122.1°W 122.0°W

QQgogo SumasSumas pTpT 49.0°N ttoo CCultusultus LakeLake (~7(~7 km)km) ult fa y e n l i l QaQa ta un a o V QlsQls M t r a l de i u b a ed f QgdmQgdm VedderV Mountain fault m k pTpT e u QgtQgt l re GEXG o pTpT C E CColumbia Valley r e X ld u QgoQgo QgoQgo o QgdQgd BoulderB Creek fault QgdQgd QgtQgt pTpT pTpT EcEc QafQaf KendallKendall QgdQgd A MMapleaple FallsFalls QlsQls CanyonC Creek fault a ny on C r

48.9°N e QQgogo ek f au QafQaf OOEhEh lt QlsQls NooksackN River o F o k C QgtQgt s SCFS QlsQls QgdQgd a EcEc c k R iv AArearea ofof FigureFigure 4 e pTpT r EcEc

A’A’ QQgogo QlsQls MCFMCF 20012001 M3.0M3.0 QgdmQgdm QlsQls pTpT 19901990 M5.0M5.0 DDemingeming

0 5 kmkm EEcc

QlsQls

LEGEND Strike and dip of bedding Qa Alluvium Fault Strike and dip of overturned bedding Qaf Alluvial fan deposits Concealed Fault Qls Landslide deposits OEh Huntington Fm. Scarp Strike and dip of foliation Qgo Sumas glacial outwash Ec Chuckanut Fm. Strike and dip of vertical foliation Syncline Qgdm Everson glaciomarine deposits pT Pre-Tertiary rocks Town Earthquake location Qgt Glacial till Anticline Earthquake focal mechanism Qgd Undifferentiated glacial deposits

Figure 3. Digital 1:100,000-scale geologic map data draped over a shaded-relief LiDAR image; digital geologic data from Washington Divi- sion of Geology and Earth Resources (2005). Earthquake locations shown are mainly from the 1990 Deming M5.1 earthquake sequence, and earthquake focal mechanisms are from the University of Washington focal mechanism catalog (http://www.pnsn.org). Cross section A–A′ is depicted in Figure 2B.

830 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

Formation and separates Chuckanut Formation one lobe fl owed west down the Strait of Juan easily interpreted images (Figs. 4–7). The most from pre-Tertiary rocks to the north (Miller and de Fuca and the second lobe fl owed south into prominent features observed on LiDAR images Misch, 1963). Miller and Misch (1963) showed the Puget Lowland (the Puget Lobe). Calibrated of the Kendall area are late Quaternary glacial the fault as a normal fault in cross sections radiocarbon ages of organic matter below and striations, etched bedding surfaces in eroded depicting the regional geology. Based on geo- above till of the Vashon stade of the Fraser bedrock, and planar outwash surfaces occupying logic mapping, Tabor et al. (2003) suggested it glaciation of Armstrong et al. (1965) provide valley fl oors (Fig. 4). is a NE-SW–trending normal fault that places limiting ages on the Puget Lobe’s advance and We measured elevation profi les on terraces north-dipping Eocene Chuckanut Formation retreat (Porter and Swanson, 1998). Follow- and across scarps using raster map profi ling sandstones to the south against pre-Tertiary ing retreat of the Cordilleran Ice Sheet, marine routines in Grass GIS version 6.4. We analyzed metamorphic rocks to the north. Dragovich water inundated the area, depositing glacio- data produced in the profi ling procedure using et al. (1997a) mapped the Smith Creek and marine drift (Armstrong and Brown, 1954). Matlab—a script loads the LiDAR profi le data McCauley Creek faults, in addition to the Boul- Wood in the glaciomarine drift yielded an age into Matlab and then we handpicked profi le seg- der Creek fault, near Kendall, and argue that of 13,960–13,490 cal. yr B.P. (Kovanen and ments corresponding to the original upper and the 1990 M5.0 Deming earthquake occurred on Easterbrook, 1998). After deposition of gla- lower surfaces and the scarp face and/or debris the McCauley Creek fault, a small thrust fault ciomarine deposits in the area and emergence slope. The script then fi ts lines to those segments, northwest of Deming (Fig. 3). The earthquake of the area, ice from the Coast Mountains re- determines the point on the scarp face with the was initially reported at a depth of 12.6 km occupied much of the Fraser Lowland (the maximum slope, and calculates the offset of (Advanced National Seismic System catalog, Sumas stade of Armstrong et al., 1965). Wood the original surfaces through that point. Where http://earthquake.usgs.gov/monitoring/anss/) collected from ice-contact deposits at the mar- slopes of the original surfaces are remarkably but later updated to a depth of 3–4 km based on gin of this re-advance near Cultus Lake, Brit- different, we measured scarp height using offset recorded aftershock distributions (Dragovich ish Columbia, yielded an age of 13,770–12,630 of the crest and toe of the scarp, regardless of et al., 1997b). To date, there are no observations cal. yr B.P. (Kovanen and Easterbrook, 2001). what angle the original surfaces are. (e.g., seismic refl ection profi les) bearing on the Outwash deposits downstream of this ice mar- We collected organic material from trenches subsurface geometry of the faults in the region gin and downstream of our study area contained and cores for radiocarbon analysis to constrain surrounding the Deming earthquake epicenter. charcoal layers intercalated in sandy gravels ages of stratigraphic horizons and past earth- Furthermore, Dragovich et al. (1997a) suggested beneath terraces above the Nooksack River; quakes (Table 1). Radiocarbon procedures used that the McCauley Creek fault is kinematically samples of these charcoal layers yielded ages of accelerator mass spectrometry (AMS) to ana- linked with other nearby faults, including the 12,680–12,400 and 12,870–12,570 cal. yr B.P. lyze detrital charcoal samples collected from Boulder Creek fault, as part of a larger thrust (Kovanen and Easterbrook, 2001). Evidence scarp excavations and plant material picked fault system. Map patterns alone suggest that all presented by Harrington and Clark (2011) and from sieved samples of wetland deposits. of these faults are normal faults separating the Osborn et al. (2012) argues against the sugges- We prefer delicate plant material over char- Chuckanut and pre-Tertiary rocks (Dragovich tion by Kovanen and Easterbrook (2001) that coal clasts for 14C analyses because recycling et al., 1997b; Lapen, 2000; Tabor et al., 2003). alpine glaciers emanating from Mount Baker of delicate plant material in the environ- and areas to the east occupied the three Nook- ment over time is less likely than for charcoal Canyon Creek Fault sack Valley forks at this time. (Gavin, 2001). We sieved samples of peat and gyttja (organic-rich mud) from gouge cores in Tabor et al. (2003) mapped several faults METHODS wetland deposits to separate organic detritus northwest of Mount Baker, including an inferred (seeds, leaf parts, and conifer needles) from sur- structure we informally refer to as the Canyon Our study relied on LiDAR data obtained rounding matrix. These sieved samples yielded Creek fault (Fig. 3). The Canyon Creek fault from several sources, including the Puget Sound delicate plant materials that could not survive trends north-south near Mount Baker and bends LiDAR Consortium, the Nooksack Tribe, and subaerial transport and exposure. Thus, ages northwesterly at its intersection with the Nook- the U.S. Geological Survey (USGS) (all of the of these delicate samples closely approximate sack River valley. The dip and kinematics of the LiDAR data available are from the Puget Sound the age of deposition, whereas detrital charcoal Canyon River fault are not known. Cross sec- LiDAR Consortium; http://pugetsoundlidar.ess samples may have had in-built (inherited) ages tions by Tabor et al. (2003) show the fault as a .washington.edu/index.html). We rely mostly on of hundreds of years at the time of deposition near-vertical or south-dipping reverse fault in a USGS LiDAR survey designed in accordance (Gavin, 2001). the hanging wall of the south-dipping Glacier with Federal Emergency Management Agency We list the laboratory-reported radiocarbon Extensional fault (Fig. 2). Detailed kine matics LiDAR data collection standards to provide pulse ages in 14C yr B.P., and used the computer pro- of the Canyon Creek fault are unknown but spacings no greater than 1.4 m (~0.5 pulse/m2), gram OxCal (Bronk Ramsey, 1995) and the mapped offset of pre-Tertiary rocks indicates horizontal accuracy of ≤1 m Root Mean Square INTCAL09 calibration data of Reimer et al. right-lateral and/or east-side-down offset. Error (RMSE), and vertical accuracy of (2009) to calibrate the reported ages as prob- ≤18.5 cm RMSE (37 cm in vegetated areas). We ability distributions. The 95% confi dence inter- Quaternary Deposits imported individual data tiles consisting of grids val of each distribution is reported as cal. yr B.P. with 6 ft2 (1.82 m2) cells into Grass GIS (ver- (before A.D. 1950), thus, a calibrated age of During the last glaciation, the Cordilleran sion 6.4, http://grass.osgeo.org) and mosaicked 1000 cal. yr B.P. is about A.D. 950. We round Ice Sheet expanded southward from the Coast the tiles into a single bare-earth digital elevation ages for interpreted earthquakes to the nearest Mountains and Fraser Lowland in southwest- model. We then used hillshaded-relief images decade to account for additional uncertainty in ern British Columbia into northwestern Wash- and slope maps to interpret geomorphic features relating sample ages to stratigraphic contacts ington. At about the latitude of the Olympic and surfaces. A histogram-equalized grayscale inferred as having a tectonic origin (Stuiver and Mountains, the ice sheet divided into two lobes; profi le applied to slope maps yielded the most Polach, 1977).

Geosphere, August 2013 831

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

A 122.15°W 122.10°W 122.05°W 122.00°W

1 y e e d l li l LLandslide ds FracturesFractures a a n V n a 48.95°N 2 LLandslide d a i s Pre-TertiaryP bedrock b l i re d - m Te e r u lt tia l u r o a y f b ColumbiaC Valley k ed e ro e ck r LLandslideandslide r CreekC fault LLandslidea ulde nd BoulderBo sli 3 MapleMaple FallsFalls de LLandslideandslide LLandslidea nd sl KendallKendall ide CanyonCanyon CreekCreek sscarpcarp (Figure(Figure 66)) LLandslideandslide 4 HHornetornet KendallKendall scarpscarp ((FigureFigure 55))

48.90°N r e iv 6 k R e c k c g id o a in l r s dd s d k e d e o BBedding n b o a 5 NooksackN LLandslide River LLandslide e a n n EoceneEocene bbedrockedrock e d c s o l g i EoceneE bedrock d LLandslideandslide n i e

d FFlowlow

BBedding 0 1 2 km

Geologic units Modern stream channel Columbia Valley terrace (CVT) Quaternary alluvial fan

Landslide Nooksack Valley terrace (NVT) Holocene alluvial fan

Scarp (points to end of scarp) Older alluvium (Quaternary) Lines Mapped fault Inferred or concealed fault Scarp observed 1 Terrace profile (WA DNR) (WA DNR) on LiDAR (profiles below)

State Plane Easting (m) B West East 404000 40700 410000 413000 416000 419000 270 6 240 2

210 3 NVTNVT 180 1 5 4 Elevation (m) 150 CVT 120 210000 213000 216000 219000 222000 225000 South North State Plane Northing (m) Figure 4. (A) Shaded slope map of LiDAR data from Whatcom County, Washington. Slope is calculated from a 6-ft-cell digital elevation model and shaded using a histogram-equalized shading algorithm (light areas have low slope angles, dark areas have higher slope angles). WA DNR—Washington State Department of Natural Resources. (B) Profi les of terrace elevations in the Columbia Valley and North Fork Nooksack River. Graph colors correspond to colors for mapped terraces in A and profi le locations are shown on A by dotted black lines (numbers indicate corresponding profi le below, colored dots indicate ends of profi le shown on both map and profi les). Hor- net indicates the location of the Hornet trench. Shaded area in B indicates Nooksack Valley terraces.

832 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

122.14°W 122.13°W 122.12°W 122.11°W 122.10°W 122.09°W A 48.92°N

1771 KKendallendall 7 7

1631 HornetHornet 6 AArearea ofof FigureFigure 1010 3

2 3 4 5

1371 3 7 131131 1871 87 1

48.91°N 1631 6 3

1 447 7

ack River 7 Nooks 2 Fork 1271 NorthNorth Fork Nooksack River 1291 29

Inset image - slope image with stronger contrast to show warp on terrace surface

Explanation 1 Location of scarp profile (profiles shown below; 0 1 km red dot at north end, blue dot at south end) Trench location Contour interval = 2 meters Scarp (points to end of scarp on image) North

NSNSNS 144 174 214 1 2 3

140 170 210

136 166 206

Elevation (m) 132 162 202 V.E. ~20 V.E. ~20 V.E. ~6.5 128 158 198 0 100 200 300 400 500 0 100 200 300 400 500 0 50 100 150 Distance (m) NSNS Pop-up of 4 V.E. ~25 5 198 198 Eocene bedrock

194 194

190 190 Elevation (m) 186 186 V.E. ~25 182 182 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Distance (m)

Figure 5. Shaded-relief image of the Kendall scarp. Image is a slope map calculated from LiDAR data and shaded using a histogram-equalized shading algorithm. Contour interval is 2 m. White circle indicates the location of the Hornet trench on each image. Black lines indicate the locations of scarp elevation proﬁ les 1–5 shown at bottom of ﬁ gure. V.E.—vertical exaggeration.

We constrain ages of deposition and deforma- is particularly powerful because it incorporates culate ages of past events (entered in the model tion using a Bayesian analysis of our radiocar- prior chronologic information, such as strati- as boundaries). Ages of samples from pre- bon ages in the OxCal radiocarbon calibration graphic order, laboratory uncertainty, ages of earthquake soils in the trenches and wetland program (Lienkaemper and Bronk Ramsey, known events, and historical constraints, to are entered as phases because their true strati- 2009). Bayesian analysis in paleoseismology reweigh the probability distributions and cal- graphic order is not known.

Geosphere, August 2013 833

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

122.02°W 122.01°W 122.00°W 121.99°W Explanation 0 500 meters AreaArea ofof FigureFigure 7 190190 35 Contour interval = 5 meters 1 2 3353 33533 3 5 Scarp profile; red=north end, 32532 1 SmugglerSmuggler 5 blue=south end

3153 Trench location 15 4 End of scarp

5 N. Fork Nooksack River 9

1951 North

AreaArea ofof FigureFigure 8 48.92°N

31531 3053 5 2002 2852 0 00 8 5 2752 5 7 5 2952 9 5

2552 5 2602 2652 5 2 6 6 770 0 5 5 0 2 ek 0 re 2252 2502 n C 3 5 nyo 2302 0 CanyonCa Creek

5 5 0 2052 3 2352

340 340 340 340 12V.E. ~3 V.E. ~3 3 V.E. ~3 4 V.E. ~3 320 320 320 320

300 300 300 300

280 280 280 280

Elevation (m) 260 260 260 260 240 240 240 240 0 100 200 300 400 0 100 200 300 400 0 100 200 300 400 0 100 200 300 400 Distance (m)

Figure 6. Shaded-relief image of the Canyon Creek scarp. Contour interval is 5 m. Canyon Creek, the scarp’s namesake, is in the lower right corner of the LiDAR image. Numbered lines indicate locations of scarp profi les (profi les shown at bottom of fi gure). V.E.—vertical exaggeration.

122.016°W 122.013°W 122.010°W 122.007°W 122.004°W 2702 2 2502 7 404 5 2602 0 2302 0 0 6 2202 3 0 34034 2 0 0 0 2102 LoggingL road 1 o g 0 LandslideLandslide g in 33033 g 0

ro 48.924°N a A d B 320320 ‘Pyrite’‘Py forest rite ’ fo SSmugglermuggler res 31031 t DisturbedDis area 0 turb ed area 3003 00

290290 1951 C 95 D

Nooksac 48.921°N 28028 k Riv 0 HighwayHig er 0 100 200 m hw ay 2702 N 70

Figure 7. Enlarged LiDAR image showing the west end of the Canyon Creek scarp. Contour interval is 1 m. Letters refer to: A—downhill-facing scarp; B—uphill-facing scarp; C and D—possibly right- laterally offset stream channels. Disturbed area shown on LiDAR image is from recent logging activities.

834 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

TABLE 1. RADIOCARBON AND TEPHRA AGES FROM EXCAVATIONS AND WETLAND CORES Field Lab 13C/12C AMS radiocarbon age Calibrated age Site Unit number number ratio (yr B.P.) (2σ, cal. yr B.P.) Material Wetland KST3-20-29 Beta-210792 –28.1 128.5 ± 0.6 pBC* Modern Plant material from gyttja in youngest couplet # 3 KST3-22-22 Beta-210789 † 880 ± 40 920–720 Seeds from gyttja in youngest couplet # 3 KST3-22-29 Beta-210790 –27.6 920 ± 40 920–740 Flower/seed parts from peat below gyttja in youngest couplet 3 KST3-22-44 Beta-210791 –26.6 680 ± 40§ 680–560 Seeds from peat below gyttja in youngest couplet # 2 KM-FG 53-55 Split A Beta-223978 † 3020 ± 40 3340–3080 Thuja plicata leaves and conifer needles from gyttja in middle couplet # 2 KM-FG 55-57 Split A Beta-223979 –27.8 2960 ± 40 3320–2990 Thuja plicata leaves and conifer needles from top of soil/peat in middle couplet 1 Tephra 7780–7480 Mazama ash (Zdanowicz et al., 1999) Hornet # 4aAb HT-008 Beta-210083 –28.3 1100 ± 40 1120–930 Charcoal from lower part of stratigraphically highest buried soil # 2c HT-002 Beta-210084 –21.8 2890 ± 40 3200–2890 Charcoal from top of stratigraphically lowest buried soil Smugglers # 4 BS003 Beta-248645 –26.3 11,730 ± 60 13,760–13,400 Charcoal 4 BS004 Beta-248646 –25.4 40 ± 40 260–0 Charcoal 7 BS005 Beta-248647 –27.9 3760 ± 40 4240–3990 Charcoal 4 BS006 Beta-248648 –25.5 3920 ± 40 4500–4240 Charcoal *Sample contained post–nuclear bomb carbon and was not included in the OxCal analysis. †The original sample was too small for a 13C/12C ratio measurement. However, a ratio including both natural and laboratory effects was measured during the 14C detection to derive an AMS (accelerator mass spectrometry) radiocarbon age suitable for applicable calendar calibration. §Sample was not in stratigraphic agreement with other samples from the same couplet so we did not include this sample in our OxCal analysis. It is possible that roots or falling trees and branches penetrated the upper couplet in places and disturbed the deposits. #Sample used in OxCal analysis (see Fig. 8).

RESULTS To test this hypothesis, we placed excavations consists of discontinuous beds of sand and across the two scarps to map the shallow sub- gravel that commonly exhibit parallel and cross- Scarps Observed on LiDAR Images surface stratigraphy and search for evidence of stratifi cation. Long axes of imbricated pebbles recent deformation. and cobbles are sub-parallel to bedding in the The most notable LiDAR features are two We placed one excavation across the Kendall outwash. In the middle of the trench, stratifi ca- scarps, one located near Kendall and the second scarp near Kendall (herein called the Hornet tion in the sand and gravel disappears, mainly located east of the town of Maple Falls, Wash- trench) and one excavation across the Canyon below the scarp slope and upper original sur- ington, near Canyon Creek (Figs. 5, 6, 7, and 8). Creek scarp (herein called the Smuggle trench) face. Kovanen and Easterbrook (2001) and A ~4.3-km-long, south-side-up scarp near Ken- east of Maple Falls (Fig. 3). Barnett (2007) and Lapen (2000) mapped similar deposits nearby as dall (herein called the Kendall scarp; Fig. 5) Seidlecki (2008) discussed other trenches across undifferentiated Sumas glacial outwash deposits ranges from 2 m to 4.1 m high, and deforms the Kendall scarp. Excavations across both and describe the sediments as loose, moderately both Pleistocene outwash surfaces shown on scarps observed on LiDAR images revealed to well-sorted sandy gravels with subrounded to Figure 4. A second scarp, located ~5.5 km east folded and faulted sequences of outwash grav- rounded clasts derived from the Coast Plutonic of Maple Falls near Canyon Creek (herein called els and till, overlain by Holocene soils and Complex in British Columbia and from local the Canyon Creek scarp; Figs. 6 and 7) is ~2 km scarp colluvium. Both trenches exposed sedi- sources in the Nooksack Valley. Relief on the long, is south-side-up, has a maximum height of ments containing organic materials (charcoal) upper surface of unit 1 roughly mirrors the scarp ~6.9 m, and lies northeast of the inferred trace suitable for radiocarbon dating (see below). topography. Charcoal from nearby correlative of the Canyon Creek fault. The scarp forms both Photo mosaics of each trench are included in the units yielded radiocarbon ages between 13,960 uphill- and downhill-facing escarpments across Supplemental File1. We also conducted a strati- and 12,400 cal. yr B.P. (Kovanen and Easter- a hillslope bordering the Nooksack River; the graphic study by coring a small wetland adja- brook, 2001, 1998). scarp is truncated by the modern fl oodplain of cent to the Kendall scarp in hopes of obtaining a A dark-colored sandy silt with pebbles lies on the river at its western end and dies out in a land- more refi ned record of earthquake timing. top of unit 1 (Fig. 9A). This silt (unit 2) exhibits a slide at its eastern end. The changes in facing silt loam to loamy sand texture and a moderately direction, and two apparently right-laterally off- Kendall Scarp Excavation developed, medium sub-blocky soil structure. set stream channels, suggest a signifi cant lateral Relief on the surface of unit 1 mimics the scarp component of slip (locations C and D on Fig. 7). This excavation exposed folded and faulted topography and confi nes unit 2 to an area below The Kendall and Canyon Creek scarps are stratifi ed sandy gravels overlain by forest soils the scarp slope. We see no evidence of unit 2 in roughly coincident with the mapped traces of and colluvial deposits (Figs. 5 and 9A; see the uppermost part of the trench (high side of the Boulder Creek and Canyon Creek faults. Table 2 for details of each stratigraphic unit). the scarp). Charcoal from the top of unit 2 has a The Canyon Creek scarp has the same sense of The oldest stratum exposed in the trench (unit 1) calibrated radiocarbon age of 3200–2890 cal. yr slip (southwest-side-up) as the Canyon Creek B.P. (lab analysis number Beta-210084, 2890 ± fault, but the Kendall scarp is south-side-up and 1Supplemental File. Photomosaics of Hornet 40 yr B.P.; Table 1). the Boulder Creek fault is south-side-down. Our Trench and Smuggler Trench. If you are viewing the Unit 3, a wedge-shaped set of gravelly sand PDF of this paper or reading it offl ine, please visit working hypothesis is that both scarps observed http://dx.doi.org/10.1130/GES00880.S1 or the full- strata, overlies unit 2 in the middle of the trench on LiDAR images are fault scarps resulting text article on www.gsapubs.org to view the Supple- below the toe of the scarp. Primary dips of the from recent surface-rupturing earthquakes. mental File. gravelly sand strata roughly mirror the north-

Geosphere, August 2013 835

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

North South North South 137 ABColumbia Valley scarp 136 168 135 166 d=4.1 134 133 164 132 d=2.5 Elevation (m) 131 162 Quarry Road 130 V.E. ~40 V.E. ~30 160 0 100 200 300 400 500 600 0 50 100 150 200 250 300 350 400 450

C D 172 192 190 170 d=3.4 188 168 186

Elevation (m) 166 184 d=3.4 164 V.E. ~25 Chickadee V.E. ~25 Hornet 182 0 50 100 150 200 250 300 350 400 450 0 100 200 300 400 500 600

200 E Note: Difference in 276 F 198 two height estimates 272 196 reflect possible Bedrock 194 folding/tilting of 268 264 192 terrace surface d=3.9 260 190 256

Elevation (m) 188 d=~3.2 252 186 d=2.0 248 Canyon 184 V.E. ~16 Stellar V.E. ~5 244 0 100 200 300 400 500 600 050100150200 Distance (m) Distance (m)

Figure 8. Profi les from selected sites along Kendall and Canyon Creek scarps used to estimate maximum scarp heights for use in calculating total post-glacial slip rates. (A) Columbia Valley corresponds approximately to profi le 1 on Figure 5. Orange dotted lines show minimum offset of scarp crest and toe. (B) Quarry Road. (C) Chickadee. Both B and C correspond to sites within 100 m of profi le 2 on Figure 5. (D) Hornet. (E) Stellar. D and E correspond approximately to profi les 4 and 5, respectively, on Figure 5. (F) Canyon Creek profi le corresponds to location 3 on Figure 6. Lowercase “d” indicates calculated scarp height based on lines fi tted to the scarp treads. V.E.—vertical exaggeration.

sloping scarp above—the underlying silt loam brated radiocarbon age of 1120–930 cal. yr B.P. and bottom of the scarp) and thinnest along the (unit 2) is almost horizontal, suggesting the (Beta-210083, 1100 ± 40 14C yr B.P.; Table 1). scarp slope. steeper dips in unit 3 are primary. The gravelly Unit 5 is a small V-shaped pendant of pebbly Several faults cut the stratigraphic sequence in sands in unit 3 pinch out at the toe of the scarp, sand and serves as a vertical boundary between the Hornet trench (Fig. 9A). Fault F1 reversely such that the deposits form a distinct wedge on unit 2 to the south and units 3 and 4 to the north. offsets the entire stratigraphic sequence up top of unit 2. Unlike the distinctly stratiﬁ ed unit 3 nearby, through the unit 4. Fault F2 splays off of F1, and A second dark-colored sandy silt with peb- unit 5 is massive and has a weak sub-blocky soil both faults form the sides of unit 5, a pendant- bles (unit 4) lies above unit 3 and forms a slight structure. Unit 5 forms a complex set of both hori- shaped gravelly sand deposit. Faults F3, F4, and angular unconformity with units 2 and 3. Unit zontal and vertical contacts that separate units 4 F5 reversely offset and rotate layers of sandy 4 has a silt loam to loam texture and exhibits and 5 from stratigraphically higher unit 6, a mas- gravels in unit 1, and development of modern a well-developed, very ﬁ ne to medium granu- sive sandy silt to silty sand with primary dips that soils (unit 7) destroyed evidence of the upper lar to sub-blocky soil structure. Like units 2 and roughly mirror the topography of the scarp. ends of the faults. Rotated clasts, some with 3, unit 4 lies beneath the toe of the scarp and A dark-colored sandy silt to silty sand (unit near-vertical long axes, indicate the location of dips gently to the north, mirroring the slope of 7) lies at the top of the stratigraphic sequence faults in unit 1. Dipping layers of unit 1 between the scarp above. A near-vertical contact sepa- throughout the trench. Unit 7 has a silt loam to the faults attest to folding either before or dur- rates the silt loam of unit 4 from pebbly sands loam texture and exhibits a moderately devel- ing faulting events. We found no indications of in unit 5, while a conformable contact separates oped, medium sub-blocky soil structure—similar lateral motion on the faults but that does not pre- unit 4 from overlying sand deposits (unit 6). in both respects to units 2 and 4 below. Unit 7 is clude the possibility that all of these faults have Charcoal from the lower part of unit 4 has a cali- thickest at both ends of the trench (i.e., the top reverse oblique motion.

836 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington S w c C S 1c 1 B d c b 7dC 7dC 7 6 7cBw 7cBw 7 A 1Ab 1Ab 1 e A 1Ae 1Ae 1 7 Boulder O a 7aO 7 ns of the d c A 1c 1c 1 1Ad 1Ad 1 Silty fine sand Pebbly cobbly sand Loamy sand Root cast Pebbly sand Sand to loamy sand ne. 8 e 4 5 8 6 7 rc c A A 1Ae 1Ae 1 b 1Ac 1Ac 1 Youngest buried soil Youngest Modern soils Youngest scarp/slope colluvium Youngest 1Ac 1b 1b 1 6 6 Inferred contact Primary stratigraphic contact Inferred fault Scarp freeface/edge of colluvial wedge Former ground surface buried by colluvium Primary fault (with motion indicated) Radiocarbon sample c rc r rc Symbols Faults Contacts and Bedding c d 1c 1c 1 A 1Ad 1Ad 1 p b c 1bp 1bp 1 1c 1c 1 c rc r rc B Loamy sand to Pebbly sand Pebbly sand with boulders Pebbly sand Cobbly med-crs sand A 1B 1B 1 F3 (110°, 75° NE) F3 (110°, w b c ne; med—medium; vfn—very ﬁ 7bA 7 7bA 1c 1c 1 B c 2 7cBw 7 7cBw c p 3C 3A 3B 3D 1c 1c 1 b C 1bp 1bp 1 d Oldest buried soil

6 6 Oldest scarp colluvium Oldest slope colluvium

7dC 7 7dC c

8 8 A

7 7 1Ac 1 1Ac

p b 1bp1 Silt loam to Silt loam to Forest litter Silt loam to Silt loam to d

F5 A 7dC 1Ad 1 7bA 7aO 7cBw 7cBC B c 1B1B Modern soils 1B 1 1c 1B1B 1c 1 F4 Sandy gravel Pebbly med-crs sand Sandy gravel Fine-med sand Sheared gravelly sand c

rc r rc c

A d

1Ac 1 1Ac

A b

1 1Ad B 1Ad F3 1c 1b 1a 1b 1 1b 1bp 1B 1 1B 1m Glacial outwash deposits

B b

1B1 1b 1 1b

b 1 1Ac

B 1Ac

1B 1 A 1B

1Ab 1 1Ab

d Pebbly sand Pebbly silt sand Sandy silt - silty sand Cobbly pebbly sand Fine - coarse sand Pebbly sand

A c

c 1 1Ad 1Ad c A 7 7 2c 2 1c 1c 1 1Ac 1Ac 1 6 6 C

B b 5

3b 6a 6B c

3a p C 7cBC 7cBC 7

b B F1 B

Oldest scarp colluvium 1bp 1 Youngest scarp colluvium Youngest

b 8 8 1B 1 1B

2bBCb 2 2bBCb F2 D B 3D 3D 3 3B 3B 3

5 5

b b

c 3 3b

3b a A

3 3c 3c 3a 3 3a a D 2aAb 2aAb 2 B 3D 3D 3 3B 3B 3 a c rc r rc 6a 6a 6 HT-002 (Beta-210084) HT-002 2890 +/– 40 C14 yrs BP 5 5 b

C b A F2 (085°, vertical) 3C 3C 3

1Ab 1 1Ab C BS006 (Beta-248648) 3920 +/– 40 C14 yrs BP

B B

Loamy sand Silt loam to Undifferentiated 6B 6 6B

Silt loam to Silt loam to b

F1 (334°, 39° SW)

2bBCb 2 2bBCb c c BS005 (Beta-248647) 3760 +/– 40 C14 yrs BP Dashed line of F2 indicates trace fault across bench in trench 2c 1c 1 1c A 2aAb 4aAb 1Ac 1Ac 1 c 4bBCb

2bBCb Youngest buried soil Youngest A Oldest buried soil 1c 1 1c 3A 3A 3 b a d 1a 1a 1 C

A B p 7 7

1Ad 1Ad 1

b b b b

1 1bp 4bBCb 1bp 4bBCb 4 A C a B 4aAb 4 b b 4bBCb 4 1b 1 1b 4 4 p excavations along the scarp. Abbreviations: cob—cobble; pbl—pebble; crs—coarse; fn—ﬁ Abbreviations: excavations along the scarp. Figure 9. (A) Excavation log of the Hornet trench. (B) Excavation log of the Smuggler’s trench. See Figures 5 and 6 for locatio 5 and 6 for See Figures trench. (B) Excavation log of the Smuggler’s 9. (A) Excavation log of the Hornet trench. Figure b 1bp 1 1bp 1 meter d BS003 (Beta-248645) +/– 60 C14 yrs BP 11730 Destratified cob peb sand Cobbly vfn-crs sand Pebbly fn-crs sand Cobbly pbl sand Pebbly crs sand Coarse sand A C 1Ad 1 B 2 2

c a 7cBC 7cBC 7 A A

1Aa1 BS004 (Beta-248646) 40 +/– C14 yrs BP b Smuggler’s trench 1B 1Ac 1Ab N 1Aa 1Ae 1Ad 7bA 7bA 7 Hornet trench b

HT-008 (Beta-210083) HT-008 +/– 40 C14 yrs BP 1100 Glacial outwash deposits b

A B A A 1Ab 1

1Ab 1 1Ab

Geosphere, August 2013 837

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al. m oil u i e e y v l k k u b l l a a a o u u thquake b c n q q i o h h r n t t i d r r p e a a , d p e e h e o s t t l p s s a e o l e e v w e t g g e v u n n d e o u u , l d l i o o , l a y y o l i i i s c o o t m m s a s l s t o o t g r r e s f f s r e e e o r r f n m m o o f f e u u f i i o c f f v v o o o n t u u l l l l s o n i n o o z o i e o l c c r z z i i o P r p p r r r h o o e a a t h or weathered glacial outwash (unit 1) from Sumas advance BC horizon colluvium or weathered glacial outwash (unit 1); 7dC is a weak C horizon below the from Sumas advance probably correlative to unit 3a from Sumas advance. probably correlative to unit 5 (undifferentiated) from Sumas advance from Sumas advance from Sumas advance h c C c a A O Late Pleistocene glacial outwash, probably S L Scarp colluvium from penultimate earthquake; B S Late Pleistocene glacial outwash, probably e n o N Tectonic deformationTectonic Genesis and unit interpretation e e e e e n n n n n movement on F1 and F2 faulting along F1, F3, F4, and F5 F1, F3, and F5 to the north (25°–35°) and cut by o o o o o N N N N N Stratally disrupted and displaced by Cut by F1 and possibly folded Scarp colluvium from penultimate earthquake Folded and faulted; steeply dipping †† oldest to youngest). Labels for subunits that correspond with A, B, or E horizons include oldest to youngest). Labels for subunits that correspond with k k b b s s k k , , e f f b b - - v s s f f i v v , , s e , , s n m m beds or planar beds graded bedding beds and planar beds 2 2 a o , - - , soil structures N 2 M 1 1 1-2, m, sbk Possibly rotated clockwise by 2 2, m, sbk Cut by movement on F1 Buried late Holocene forest soil Low angle cross Depositional and/or h - v s , f - f v h e e Organic , , material** n n f vf, sv-h 3, vf-f, sv vf-f, h m, sv sv-h 2 o o , - N 1-3, f, sv-h; 3, N 1 1 1, f-m, h vf-m, gr-sbk 1-2, Cut by F1 and possibly folded Buried BC horizon of late Holocene forest soil 1-2, vf-f, sv-h 2-3, vf-m, sbk Cut by F1 and possibly folded AB horizon of late Holocene forest soil Buried 1, f, sh 3, m-f, sbk Cut and folded by movement on F1AB horizon of late Holocene forest soil Buried # 2 4 3 / / / 2 5 4 R R Y Y Y 5 . 0 0 2 1 1 (10YR3/4 (10YR2/2) (10YR2/1) (10YR3/2) (10YR3/4) (7.5YR2.5/2 Matrix color § Cobbles § % % 2 1 <1% 10YR4/3 < < <1% 10YR5/4 ~2% 10YR5/4 TABLE 2. DESCRIPTIONS OF STRATIGRAPHIC UNITS OBSERVED IN THE HORNET TRENCH THE HORNET IN UNITS OBSERVED 2. DESCRIPTIONS OF STRATIGRAPHIC TABLE ~15% 2.5Y5/4 1, co, h;2-3, ~20% 2.5Y5/4 1, m, h;1, f, 2%–5% 10YR4/3 Pebbles 5%–10% 10YR4/3 30%–40% ~10% 2.5Y6/4 f, sv-v 1, Imbrication Cut by F3 and F4 Late Pleistocene glacial outwash, probably 60%–70% 2.5Y5/3 1, co, sv; 2, † ne ) r e t t i l ne to t s e d r n loam loam loam loam loam loam loam loam to coarse) coarse) o a F Silt loam to Silt loam to Silt loam to ( Silt loam to Sand 20%–30% ~1% 2.5Y5/3 1-2, vf-f, sv Massive Cut by F1 and possibly folded Scarp colluvium from penultimate earthquake; Sand 30%–40% <2% 2.5Y4/3 1, f-m, sh Massive Cut by movement on F1, F4, and F5 Late Pleistocene glacial outwash, probably S Sand 30%–40% 10YR4/3 None Low angle cross Silt loam to Sand <1% 2.5Y5/3 1, f-m, h Planar, Silt loam to Silt loam to Sand 30%–40% <2% 2.5Y5/3 1, f, sv Massive Stratally disrupted and cut by F1–F5 Pleistocene glacial outwash, probably Late s n i a m e r t n silty sand pebbles sand silty sand granules and sand sand sand silt sand silty sand silt silt sand a l P O Terms for soil horizon properties follow Natural Resources Conservation Service notation and description. Terms Texture terms follow Natural Resources Conservation Service notation and description. Texture Estimate of area covered by clasts using size charts. Clasts were mostly subrounded to rounded, and occasionally faceted. Primary color is dominant Munsell of matrix, taken dry (moist in parentheses, if taken). O † § # of organic material present. Root terms follow Natural Resources Conservation Service notation and description. **Type †† *Units shown on the trench log are designated by a unit code based on lithology, stratigraphic position, and inferred age (from *Units shown on the trench log are designated by a unit code based lithology, a 7aA A Sandy silt to Unit* Horizon Lithology Matrix texture 7aBC/7dC BC Sandy silt to 7 1Aa -- Coarse sand, 6a -- Pebbly silty 5 -- Pebbly sand Sand 15%–25% 2.5Y5/4 1, co, h;1, m, h; 4aAb Ab Pebbly sandy 6b -- Sandy silt to 4bBCb3a BCb Pebbly sandy -- Cobbly pebbly 2bBCb BCb Pebbly sand Loamy sand ~20% 10YR5/3 1, f, sv None Cut and folded by movement on F1 Buried BC horizon of late Holocene forest s the appropriate soil horizon designation. 3b2aAb -- Fine to coarse Ab Sandy silt Silt loam to 1Ab -- Pebbly coarse 1B -- Cobbly pebbly 3c -- Pebbly sand Sand ~15%–20% 10YR5/4 1-2, f-vf, sh-v Massive Cut by F1 and possibly folded Scarp colluvium from penultimate ear 2c -- Pebbly sandy 1Ad -- Pebbly sand Sand (ﬁ 1Ae -- Cobbly sand Sand (very ﬁ 1Ac -- Cobbly pebbly

838 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

In summary, we correlate stratifi ed sandy turbance (Beta-248645, 11,730 ± 60 14C yr B.P.; the loamy sands. We discount a radiocarbon age gravels to similar deposits nearby mapped as Beta-248646, 40 ± 40 14C yr B.P.; Table 1). Char- of 40 ± 40 14C yr B.P. as probable contamination late Pleistocene glacial outwash deposits dated coal from near the top of unit 4 has a calibrated from root stirring. We interpret these dark- between 13,960–12,400 cal. yrs old. Overly- radiocarbon age of 4500–4240 cal. yr B.P. (Beta- colored deposits as buried soils within boulder- ing these outwash deposits is a sequence of two 248648, 3920 ± 40 14C yr B.P.; Table 1). rich, gravelly colluvium. Two faults cut through dark-colored, loamy deposits containing char- A set of thin, gravelly sands—units 5 and the entire section; one fault reversely or obliquely coal with radiocarbon ages between 3200 and 6—straddles the scarp near the top of the strati- offsets the glacial outwash and younger soils, 930 cal. yr B.P.; both loamy deposits intercalate graphic sequence in the trench. One of these and the second fault has apparent lateral offset. in a wedge-shaped deposit of stratifi ed sandy layers, unit 5, occupies the northern third of the gravel. We interpret these dark-colored deposits trench and overlies units 3 and 4. The second Stratigraphy of the Kendall Scarp Wetland as buried soils between layers of gravelly col- of these layers, unit 6, occupies the southern luvium. A series of faults cut through the entire two-thirds of the trench exposure. The con- Small ponds and wetlands adjacent to fault section and reversely or obliquely offset both tact between units 5 and 6 lies directly above scarps, commonly referred to as sag ponds, the glacial outwash and Holocene soils. the near-vertical contact between units 3 and provide an excellent environment to record past 1—the main differences between units 5 and 6 earthquakes (Fumal et al., 1993; Meghraoui and Canyon Creek Scarp Excavation are an increase in the amount of gravel in unit Doumaz, 1996; Sieh, 1978). Flooded areas and 6 and a moderately developed sub-blocky soil sag ponds act as catchments for relatively con- The stratigraphically lowest and oldest units structure in unit 5. A set of dark-colored, peb- tinuous sedimentation and also tend to preserve in the excavation (unit 1) are light-colored (light bly, silty sands, with sandy loam textures (units organic material useful for radiocarbon dat- browns and tans) well-bedded sands with thin 7 and 8), blankets the surface of the stratigraphic ing. Field reconnaissance of the Kendall scarp gravelly sand layers (Table 3). We correlate sequence exposed by the trench. Charcoal from showed that the scarp blocked the fl ow of a unit 1 with undivided Pleistocene to Holocene near the base of unit 7 has a calibrated radio- small tributary stream fl owing southward into glacial deposits, including glacial outwash, on carbon age of 4240–3990 cal. yr B.P. (Beta- the North Fork Nooksack River (Figs. 5 and 10). the Mount Baker 1:100,000-scale quadrangle 248647, 3760 ± 40 14C yr B.P.; Table 1). We surmise that wetland stratigraphy records (Tabor et al., 2003). A layer of dark-colored silty Folding of strata in the trench is best shown scarp growth in the Holocene. Thus we inves- sand with gravel (unit 2) overlies unit 1 in the by anticlinal warping of distinct gravelly sands tigated the stratigraphy of the wetland formed northernmost part of the trench, and is absent and thinly laminated silty sands in unit 1 (Fig. adjacent to the scarp with hopes of refi ning the throughout the rest of the trench. Unit 2 exhibits 9B). The apparent wavelength of the anticline is history of scarp growth. a loamy-sand texture and a weakly platy struc- ~12–13 m (almost the length of the excavation) We collected a series of gouge cores along a ture with thin clay fi lms. A southwest-dipping because beds at both ends of the excavation are transect across the wetland to document the stra- contact juxtaposes a small part of unit 1 over nearly horizontal. Vertical relief on the anticline tigraphy and geometry of the wetland deposits the southern end of unit 2. Subsequent erosion is probably less than a meter in the southern end (Fig. 10). A leveling survey related the eleva- removed this contact and the upper surfaces of of the trench and at least 1.5 m in the northern tion of each core location to a common arbitrary unit 1 and 2 in the northern part of the trench, end of the excavation near the steepest part of benchmark. We described each core using stan- and boulder-rich pebbly sand (unit 3) overlies the scarp. A small normal fault (F3) and fi s- dard notation for soils and organic sediments both units. sure offsets the crest of the anticline, and sev- (Schoeneberger et al., 2002; Troels-Smith, Unit 3 is a wedge-shaped set of deposits at eral small warps deform the limbs of the larger 1955), and analyzed core samples for microfos- the toe of the scarp. The wedge is thickest where anticline. sils, plant macrofossils, and ashes. Stratigraphy unit 3 abuts unit 1 along a near-vertical contact Two reverse faults in the excavation offset the of the wetland cores showed cyclical deposition (shown as a scarp free face on Fig. 9B). Unit 3 gravels, soils, and the oldest colluvium. Fault F1 of gyttja and peat layers overlying a basal min- is thickest at the near-vertical contact between strikes 334°, dips 39° SW, and offsets distinct eral deposit (each set of gyttja and peat layers the southernmost part of unit 3 and unit 1, and sand and gravel beds in unit 1 and the buried is herein termed a “couplet”; Figs. 11 and 12). thins toward the northern end of the trench. This soil (unit 2) formed on unit 1. Unit 3 overlies the thinning creates a wedge-shaped set of depos- top of fault F1. We are unsure how much, if any, Basal Deposit its at the toe of the scarp with primary dips of the anticlinal folding accompanied move- Intercalated basal mineral deposits include closely mimicking the slope of the scarp above. ment of fault F1. Fault F2 is near F1 but strikes dark greenish gray to very dark gray clay loam A boulder-rich layer within unit 3 (unit 3A) sits 85° and is nearly vertical. Offset of individual to sandy loam, more than a meter thick in directly on unit 1 and 2 at the toe of the scarp, beds of gravelly sand across F2 shows tens of some locations, and thickest in the middle of and laterally changes abruptly into gravelly sand centimeters of vertical separation but the beds the wetland. Layers of clay loam are typically layers with steep primary dips that are somewhat change thicknesses abruptly, suggesting a lateral 2–7 cm thick and alternate with sandy loam lay- steeper than the scarp slope above. This abrupt component of slip. ers averaging 4–15 cm thick. The clay loams lithologic change coincides with a near-vertical In summary, we correlate stratifi ed sandy typically have yellowish brown mottles and the contact that separates the boulder-rich unit 3A gravels in the Smuggler trench with similar uppermost sandy layers are very fi ne grained from the remainder of unit 3 (units 3B–3D). This deposits mapped as late Pleistocene glacial out- while deeper layers coarsen with depth. Over- same vertical contact also separates units 3B–3D wash deposits by Tabor et al. (2003). Overly- lying these basal inorganic deposits is a dark from unit 4, a dark-colored, pebbly sandy silt ing these outwash deposits is a sequence of two gray to black silt loam. This silt loam is more that sits on the upper surface of unit 3A and unit dark-colored, loamy sands containing charcoal than 30 cm thick in places and organic content 2 in the northern part of the trench (Fig. 9B). with radiocarbon ages between 13,760 and 3990 increases upwards to form a detrital peat. At Charcoal from near the base of unit 4 has highly cal. yr B.P.; wedge-shaped deposits of stratifi ed the southern edge of the wetland the silt loam variable radiocarbon ages likely due to root dis- and boulder-rich sandy gravel intercalate with exhibits forest soil horizonation.

Geosphere, August 2013 839

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al. 1 t i rst earthquake and n u n i d e p o Genesis and l e unit interpretation v e d n o z i r o h degradation of scarp after folding and faulting of unit 1B on F1 and F3 (oldest earthquake) of scarp following oldest earthquake (folding and movement on F1 F3) and then is slope colluvium from 3 m on; deposited after youngest earthquake likely developed in sandy colluvium (unit 3C) the second earthquake gravelly colluvium (unit 5); soil horizon was in the time between fi B Glacial outwash sand Glacial outwash A horizon of modern soil A Scarp colluvium between 1 and 2.75 m (See unit 3B) Glacial outwash C y b t u c d n a d Tectonic Tectonic e d deformation l F1, cut by F2 F1, F2, and F3 o Cut by F2 Scarp colluvium deposited during degradation Cut by F2 Scarp colluvium deposited during collapse and Cut by F2 Buried BC soil horizon developed in upper part F Cut by F2 Scarp colluvium (see unit 3C for more) Cut by F2 horizon buried after youngest event by a thin BA r a †† l e e l l u g g n n n a i i r s s g , , lms , e e e v v i i n s s s s ne, granular ne, granular; fi fi grain grain clay fi to sbk to sbk grain grain massive a a oldest to youngest). Labels for subunits that correspond with A, B, or E horizons include oldest to youngest). Labels for subunits that correspond with , , soil structures M Massive; single 1, f, pl, very thin 1 <1, fi B 1, f, granular 2 Massive, single M 1, f, granular (See unit 6) Soil horizon formed in glacial outwash (unit 1B) Depositional and/or v s , , , f f f , , , , , 1 2 2 2 2 ; v ; ; ; v ; s v v v s v , s s s f , , - , , , Organic f f f o o f material** c v v v c v m, sv m-co, sv sv; 1, co, sv co, sv sv; 1, m, sv sv; 1, co, v vf-f, sv , , , , , , 1 1 3 1 1 1 4 3 4 3 1 4 2 / / / 3 3 / / / / # / / 3 4 4 4 3 2 4 5 4 R R R R R R R Y Y Y Y Y Y Y Y Y 5 5 . . color 5 5 5 Matrix 0 0 0 0 . . . 2 2 1 1 1 1 7 7 7 § % % % % 0 0 0 0 2 5 3 % % % % 1 – – – 1 5 0 0 – 30% 7.5YR4/4 1, vf-f, sv; 4 1 < < % % % % 0 0 0 Pebbles 5 1 3 2 20%–30% 10YR4/3 1, vf-f, sv Planar bedding faulted by Folded; 30%–40% 10YR3/3 1, vf, sv; 2, f, § % 2 % % – 2 2 < < % 1 Cobbles § 5% 1%–2% 10% 10YR4/4 vf-f, sv; 1, 1, TABLE 3. DESCRIPTIONS OF STRATIGRAPHIC UNITS OBSERVED IN THE SMUGGLER’S TRENCH THE SMUGGLER’S IN UNITS OBSERVED 3. DESCRIPTIONS OF STRATIGRAPHIC TABLE Boulders m † a o l y d d d d d d d d Matrix texture n n n n n n n n (70%– to sand (med-crs) 80%) loamy sand A / a a a a a a a a S Sand S Sand Loamy sand S S S N S Sand to S S d ne n d d d a n n n s a a a s s s e n y y y l l l b b b b b b medium–fi medium– coarse sand gravelly silty sand well-sorted sand with boulders medium– coarse sand sand sand sand gravel e e e P P P b b B B B C C C Terms for soil horizon properties follow Natural Resources Conservation Service notation and description. Terms Texture terms follow Natural Resources Conservation Service notation and description. Texture Estimate of area covered by clasts using size charts. Clasts were mostly subrounded to rounded, and occasionally faceted. Primary color is dominant Munsell of matrix, taken dry (moist in parentheses if taken). *Units shown on the trench log are designated by a unit code based on lithology, stratigraphic position, and inferred age (from *Units shown on the trench log are designated by a unit code based lithology, † § # of organic material present. Root terms follow Natural Resources Conservation Service notation and description. **Type †† A C D 1Ac Brownish gray 1Ab Brown pebbly 2 BCb Reddish brown 6 3A CBb Pebbly sand 3 3B CBb Cobbly pebbly the appropriate soil horizon designation. 4 BAb Silty fi 7 Bw Pebbly silty 3 8 A Pebbly silty Unit* Horizon Lithology 1Aa Brown sandy 5 Pebbly cobbly

840 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

122.098°W 122.096°W122.094°W 122.092°W 122.090°W concentrations of organic matter and silt loam, A BlueBlue aarearea sshowshows pprobablerobable 220 m and is very dark gray to very dark brown and maximummaximum extentextent ofof pondedponded areaarea ranges from 4 to 20 cm thick. Common com- ponents include wood, charcoal, and pebbles. 1185 The gyttja component decreases upward where 8 PPhotographhotograph llocationocation (B,(B, below)below) 5 it grades into a very dark brown to black detrital (cone(cone sshowshows pperspective)erspective) 180 m 0 m peat (2 to 30 cm thick) that includes Tsuga 48.914°N heterophylla (western hemlock) needles, Thuja SSedge/Spireaedge/Spirea FForestedorested

Lake plicata (western red cedar) scales, Spiraea wetlandwetland 11848 depth~4 Elevation m wetlandwetland 4 douglasii (hardhack) fruits, Rubus sp. (black- TT33 berry) seeds, herbaceous roots, wood, and char- 1183 83 coal. This sequence of gyttja with diatomite overlying detrital peat is the stratigraphically HHornetornet SScarpcarp ttrenchrench lowest and oldest of three gyttja-peat couplets de in the wetland. ra g OOutletutlet throughthrough ad lro ai Couplet 2 scarpscarp 85 RRailroad grade 48.912°N 1185 An abrupt contact (≤1 mm) separates gyttja-

m a peat couplet 2 from the ﬁ rst couplet. The lower e r t CContourontour intervalinterval = 5050 cmcm 0 100 m unit in the second couplet is a dark brown to SStream dark grayish brown gyttja, 1 to 14 cm thick, that B contains diatomite in several cores. The gyttja grades into a 2- to 23-cm-thick, very dark gray to black detrital peat with silt loam. Charcoal layers, 1 to 4 mm thick, make this unit conspic- uous compared to bounding units. Wood and roots of herbaceous plants are common. A sample of Thuja plicata leaves and conifer needle fragments from the top of the peat in couplet 2 yielded an age of 3320–2990 cal. yr B.P. A sample of Thuja plicata leaves and conifer needle fragments from the gyttja immediately overlying the abrupt contact yielded an age of 3340– 3080 cal. yr B.P. (Table 1).

Couplet 3 Above the contact between the second and third couplets is a pronounced dark brown to very dark brown gyttja, 2 to 11 cm thick, sometimes with diatomite at its base (Figs. 11 and 12). Plant macrofossils are common above and below the contact, with Tsuga heterophylla Figure 10. (A) Map of the wetland adjacent to the Kendall scarp. Blue polygon shows the needles , Thuja plicata scales, and bark marking inferred area of the lake when the stream is fully blocked by the scarp. (B) Photograph of the boundary between the underlying detrital the wetland looking west from near the north end of the Hornet trench. peat of the second couplet from overlying gyttja that contains Polygonum sp. (knotweed) and Scirpus microcarpus (sedge) seeds. Above this Couplet 1 the previously described silt loam, but it is usu- uppermost gyttja is the youngest detrital peat, A light-brown diatomaceous gyttja, or sapro- ally separated from the top of the silt loam by 10 to 40 cm thick. This very dark brown wetland pelic mud, overlies the basal black silt loam a thin layer of gyttja or diatomite. Immediately peat includes herbaceous and shrubby roots, a and detrital peat (Figs. 11 and 12). The contact above the ash is a light-brownish-gray diato- wetland seed mixture, wood, charcoal, and silt between the gyttja and the underlying deposits mite, and immediately above the diatomite is a loam. Seeds and plant material from the base of is abrupt (~1 mm in width). A yellowish brown gyttja. However, in one core, laminated diato- gyttja in couplet 3 yielded a calibrated radiocar- volcanic ash (~2 cm thick in most places) mite lies below the ash (Fig. 11). The distribu- bon age of 920–720 cal. yr B.P. (Table 1). Seeds intercalates between layers of gyttja. Electron tion of diatomite varies from a concentrated from the uppermost peat/soil layers immedi- microprobe analyses indicate that this ash is layer a few centimeters thick to diffuse diatom ately below the gyttja yielded an age of 930–740 from the climactic eruption of Mount Mazama deposition throughout the base of the gyttja. cal. yr B.P. in southern Oregon (Mazama ash, similarity In some cores, a 1-cm-thick charcoal layer In summary, cores from the wetland con- coefﬁ cient = 0.98–0.99; 7780–7480 cal. yr separates the ash and overlying diatomite. The tain a sequence of mineral and organic depos- B.P.). At some locations the ash is found within gyttja that overlies the diatomite contains equal its. The basal sandy loam ﬁ nes upward and is

Geosphere, August 2013 841

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

A 0 South North Troels-Smith sediment type

Th – Rooted herbaceous peat Ag/As – clay and silt 20 Sh – Homogenous humified organic material Ga – fine to medium sand Dg – fragments of wood and herbaceous plants, 2.0 – 0.1 mm Tephra

c c 40 c Lh/Ld – sapropelic ooze, diatomite c Charcoal c c

60 8 (pattern indicates dominate sediment type)

Wetland deposits pinch out on scarp Wetland ?

100 Missing ? Wetland deposits pinch out on slope Wetland Depth (m) 120 12 Core number 15 140 50

20 160 18

22 180 48 24

200 See Figure 12 for details 44 and photographs of 32 cores 18 and 30 30 36 220 40 10 20 30 40 50 Distance (m) Scarp B Downstream South Explanation

Surface soil 920–720 cal yrs BP North Gyttja 3316–2990 cal yrs BP 930–740 cal yrs BP Buried soil 8 Elevation point Mazama tephra CCoupletouplet 3 Sand CoupletCouplet 2 12 Basal couplet contact 15 3344–30793344–3079 calcal yrsyrs BPBP CoupletCouplet 1 50 (dashed where inferred) 20 18 BBurieduried fores 22 forest soilsoil 24 48 1m 7777–7477 cal yrs BP 30 32 44 36 40 Vertical exaggeration = 8x Core location 4m Figure 11. (A) Simplifi ed sediment classifi cation and correlation of contacts observed in cores from a transect across the wetland adjacent to the Kendall scarp. Elevations of marsh surface and core locations are related to an arbitrary benchmark. (B) Simpli fi ed cross section of stratigraphy of the Kendall scarp wetland showing grouping of strata into couplets interpreted as indicators of hydrologic changes in the wetland following past earthquakes.

842 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington *

T3, 30m T3, 18m Troels-Smith sediment type Paleo- Environment Munsell color Troels-Smith sediment type Paleo- Environment 0 Munsell color 3 1 Dg Th Wetland 2 1 Wetland 10YR 2/2 + 70 10YR 5/1 Dg Ag Dg soil soil 10 10YR 2/2 Ld4 Th+ Lake Couplet 3

20 10YR 6/1 Ld4 Th– Lake

Wetland 10YR 3/1 Sh3 Ag1 soil 30 Couplet 1 Depth (cm)

Couplet 2 2.5Y 6/1 Ld2 Ag2 Figure 12. Photomosaics of 40 5Y 7/1 two cores from the wetland 80 core transect, accompanied 10YR 4/2 Ld4 Th+ Lake 7.5YR 6/4 Tephra Mazama by observed sediment colors, 50 lithology and organic sediment 10YR 8/1 Ld2 Ag2 descriptions (Troels-Smith, Wetland 1955), and inferred paleoenvi- Sh3 Ag1 Wetland 5YR 6/1 Dg2 Ag2 60 10YR 2/2 Soil ronment. Core T3-30m shows Th+ Ga– soil the entire wetland section and

core T3-18m shows a 20-cm- Couplet 1 Depth (cm) 10YR 3/1 70 Ld4 Th+ Lake 2 2 long section spanning the depo- 10YR 2/1 5Y 9/1 Ga Ag Floodplain sitional interval for Mazama

10YR 4/4 Tephra Mazama Pre-earthquake strata tephra, showing a thin layer 80 10YR 2/1 of lake sediments below the Ag/As2 Wetland Thin layer of lacustrine sediment underlying Mazama 2 Soil tephra. 10YR 2/1 Sh tephra suggests that the first earthquake predates deposition of Mazama by an unknown but probably 90 short period of time.

Troels-Smith sediment type Ag3 Ga1 100 Gap Sh – homogenous humified organic material Dg – fragments of wood and herbaceous plants, 2.0 – 0.1 mm 2.5Y 3/1 110 Floodplain Lh/Ld – sapropelic ooze, diatomite

Pre-earthquake strata Ag/As – clay and silt G4/10Y Ga2 Ag2 120 Ga – fine to medium sand

10YR 5/6 Ag3 Ga1 Tephra

130 2 2 * – Munsel colors for T3, 18m estimated from digital photographs using G4/10Y Ga Ag mColorMeter for iPad, calibrated with Munsell color chips

capped with a black silt loam; we interpret DISCUSSION local base level in scarp-adjacent wetlands as these as stream deposits capped by a thin soil. manifestation of past earthquakes (Fig. 13). Overlying this thin soil is a series of alternating We interpret past earthquakes and localized We found evidence for three earthquakes— diato maceous sapropelic muds (gyttja) abruptly environmental changes along the Boulder Creek named A, B, and C (from oldest to youngest) overlying dark-colored detrital peats, grouped and Canyon Creek faults from the scarp exca- in subsequent discussions—in the past 7700 into three couplets of mud overlying peat. The vations and wetland stratigraphy. Evidence of years on the Boulder Creek and Canyon Creek lowest sapropelic mud contains a thin layer of past earthquakes in trenches consists primarily faults. Our OxCal modeling provides age esti- Mazama ash. We interpret these muds as lake of folded and faulted late Quaternary sediments mates for these past earthquakes, with the best deposits and the peats as wetland soils, indi- and soils, accompanied by deposition of fault limiting ages for each earthquake coming from cating that hydrology of the wetland changed scarp colluvium. Similarly, we interpret local the marsh sequence (Fig. 14). The ﬁ rst earth- abruptly at three times in the Holocene. environmental changes from ﬂ uctuations in quake (A) resulted in folding of glacial outwash

Geosphere, August 2013 843

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

B Hydric Soils Tectonic Hydrosere Wetland deposit (peat) Interseismic Hydric soils accumulate in areas where earthquake scarps block Lacustrine deposit (gyttja) Postseismic EQ2 EQ stream drainage during a seismic Wetland deposit (peat) Interseismic cycle. Stratigraphy and paleontol- ogy of the soils demonstrates Lacustrine deposit (gyttja) Postseismic repeated hydrologic changes in EQ1 Wetland/pond inception EQ the wetland along the scarp. Fluvial sand/forest soil Preseismic

A Tectonic Hydrosere Development

Interseismic Phase • Vegetation dominated by western redcedar, Spirea sp., Flow Rubus sp., and various wetland herbs • scarp breached by stream and pond drained • wetland vegetation established • wetland persists only while water table is high • forest vegetation takes over and forest soil develops • final phase of tectonic hydrosere succession

Wetland plant seeds and detritus

LacustrineLacustrine organismsorganisms

Postseismic Phase • Algae (e.g., diatoms) and aquatic vegetation dominate • scarp blocks stream flow • pond forms over forest soil • pond deposits contain aquatic fossils (gyttja) • beginning of tectonic hydrosere succession • oldest seismic event marks inception of wetland

Flow Time

Preseismic Phase • Douglas fir, western redcedar, western hemlock, alder, and forest shrubs and herbs dominate • no scarp or highly eroded scarp present • stream flow follows topography • little to no wetland

Flow

Figure 13. (A) Block model illustrating the formation and evolution of tectonic hydroseres in drainages blocked by fault scarps formed by vertical motion on faults during earthquakes. (B) Hypothetical stratigraphic proﬁ le of wetland soils formed by tectonic hydroseres as a result of periodic hydrologic changes in a wetland created by movement on a fault. EQ—earthquake.

844 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

Figure 14. OxCal model of age constraints Historic constraint AD1850; 110–90 cal yrs BP on prehistoric surface-rupturing earthquakes on the Boulder Creek fault, using OxCal version 4.1.7 (Bronk Ramsey, 2009) Wetland: KST3-22-22 920–730 cal yrs BP and IntCal09 atmospheric data (Reimer et al., 2009). All ages are from the Hornet EQ C Modeled age: 910–740 cal yrs BP trench and adjacent wetland. Ages are listed in a sequence (in stratigraphic order but Phase (Pre-EQ C soil) with no depth constraints) and two phases (stratigraphic order unknown). Historic Wetland: KST3-22-29 920–740 cal yrs BP constraint is lack of surface-rupturing earthquakes since European settlement of Whatcom County around A.D. 1850 (http:// Hornet: HT8 1120–930 cal yrs BP www.co.whatcom.wa.us/history.jsp). White Wetland: KM-FG-55-57A 3320–2990 cal yrs BP areas are prior probability distribution functions (PDFs) of calibrated radiocarbon EQ B Modeled age: 3190–2980 cal yrs BP samples. The dark gray areas are posterior PDFs resulting from Bayesian analysis of Phase (Pre-EQ B soil) age constraints. Black areas (highlighted by horizontal gray bands) are modeled PDFs of Hornet: HT2 3200–2890 cal yrs BP prehistoric earthquakes. The modeled age for earthquake (EQ) A is best approximated by the age of Mazama ash; we rounded the Wetland: KM-FG-53-55A 3340–3080 cal yrs BP age to the nearest century and expanded the error by 50 years to account for addi- EQ A Modeled age: 8050–7250 cal yrs BP tional uncertainty. Reported radiocarbon Smuggler: ages listed in 14C yr B.P.; OxCal-calibrated BS6 13760–13400 cal yrs BP ages listed in cal. yr B.P. (Table 1); all ages are rounded to nearest decade. Brackets 14000 12000 10000 8000 6000 4000 2000 0 below each PDF are 2σ uncertainties. Modeled date (BP)

deposits and warping of the ground surface in reverse faulting. A ~2–10-m-thick veneer of gla- eral lakes sometimes form after earthquakes the early Holocene at about the same time as cial deposits covers the bedrock in the vicinity where surface deformation impedes fl ow in deposition of Mazama ash in the area. Two of the scarps. rivers and streams. An excellent example of this younger earthquakes (B and C) each ruptured Outwash terraces in the Columbia Valley phenomenon is fl ooding along the Oued Fodda the ground surface, displacing the surface soils (Fig. 4) grade up-valley and merge with ice- River following the 1980 El Asnam earthquake at the time of the earthquake. The kinematics of contact deposits near Cultus Lake (Kovanen and at Chlef, Algeria (Meghraoui and Doumaz, the Kendall and Canyon Creek scarps are com- Easterbrook, 2001). The Columbia Valley ter- 1996). During the earthquake, a pressure ridge patible; reverse faulting on the east-west–trend- races are inset into the terrace in the Nooksack formed across the river and temporarily blocked ing Kendall scarp would result in reverse and/or River valley at Kendall, suggesting that the ter- water fl ow, causing extensive fl ooding. To better dextral slip on the SE-trending Canyon Creek races in the Columbia Valley are younger than explain the Kendall wetland sequence, we devel- scarp. Stratigraphy observed beneath each scarp the terrace system in the Nooksack River valley. oped a simple conceptual model of aquatic and is also broadly similar: a series of Holocene soils Elevation profi les across the scarps using LiDAR wetland environmental succession as a response and colluvial deposits overlying late Pleisto cene data show that the Kendall scarp deformed both to changing hydraulic conditions before, during, glacial outwash. terraces an equal amount (a minimum of ~2.5 m and after blockage of a stream by a fault scarp Trenches did not expose Tertiary bedrock, but and a maximum of ~4 m; Fig. 8), suggesting that (Fig. 13). We term these wetland environments we infer sandstone of the Chuckanut Formation the earthquakes that formed the scarp occurred that respond to earthquake-induced hydrologic is in the shallow subsurface because outcrops after deposition of the sandy gravels beneath the changes “tectonic hydroseres” to differentiate near each trench are composed of Chuckanut Columbia Valley and Nooksack River valley ter- them from climate-driven changes or changes Formation rocks (Lapen, 2000; Tabor et al., races (~12,000 cal. years old). driven by eutrophication. Each of these tectonic 2003). Barnett (2007) inferred strongly mag- hydroseres in turn lead to distinctive wetland netic rocks near the surface from anomalies Interpretation of Wetland Stratigraphy and lacustrine deposits, such as peat and gyttja in magnetic profi les across the Boulder Creek and Alternatives couplets. Abrupt contacts between couplets fault in the vicinity of the Kendall scarp. These marked the time of rapid ponding of the water magnetic anomalies likely represent faulted We interpret the wetland stratigraphy as behind a fault-scarp dam. serpentinite conglomerate within the Chucka- recording a response to repeated blockage of In the pre-seismic phase, relatively dry forest nut Formation or a sliver of serpentinite caught a small stream by growth of the Kendall scarp conditions and upland vegetation and soils occu- in the fault zone and brought to the surface by during surface-rupturing earthquakes. Ephem- pied sites along the fault (Fig. 13). In the short

Geosphere, August 2013 845

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

post-seismic phase, the scarp dammed streams axis. Berm height ranged from ~0.3 to 1.5 m on ing but not faulting the glacial outwash (except at the site, and a pond or small lake formed. the downstream side. Occasionally, fi ne-grained perhaps bending moment faulting on F3); (3) a Aquatic vegetation and algae deposited at the sediment fi lled the pond so that only the down- forest soil (unit 2) forming on the surface of the site resulted in deposition of organic deposits stream face of the dam was exposed, creating glacial outwash; (4) reverse or oblique reverse (peat, gyttja, and diatomite) containing fresh- a ramp-like feature up to 2 m in height. Most faulting on fault F1 during earthquake B, cutting water fossils that confi rm aquatic conditions often, deposits behind beaver dams were seen units 1 and 2; (5) deposition of the oldest scarp after the earthquake. In the interseismic phase, in outcrops; for example, Wohl (2006) indi- colluvium (unit 3)—the scarp cut across a steep the stream breached the scarp dam and reestab- cated that beaver dam abandonment resulted slope so unit 3 likely contained a component of lished drainage, the pond drained, and a wetland in channel incision through accumulated pond slope colluvium as well (unit 3B), causing an formed. Wetland vegetation (including sedges, sediments and upstream alluvium, and in turn, increase in the thickness of unit 3; (6) formation rushes, grasses, and obligate wetland shrubs and resulted in a ravine cut through the beaver dam of a soil (unit 4) on the surface of unit 3 collu- herbs) occupied the site and peat was the main that exposed older beaver pond deposits. John- vium and draped across the scarp; (7) strike-slip deposit preserved in the wetland. Eventually, the son et al. (1987) described Holocene fl uvial and or oblique reverse movement on fault F2 during stream further breached the scarp, the wetland wetland deposits created by beaver dams, an earthquake C, offsetting units 3 and 4; (8) depo- drained, and forest conditions invaded the for- interpretation driven in part by observations of sition of the youngest scarp colluvium (units 5 mer wetland to complete the cycle. beaver-chewed sticks in exposures of wetland and 6); and fi nally, (9) continued development The cycle between pond and forest is par- deposits in a creek incised through the former of forest soil (units 7 and 8). tially complete today; ephemeral standing water beaver dams. We observed no evidence of past occupies the western end of the impoundment beaver activity at the Kendall scarp. No aban- Earthquake A—8070–7240 cal. yr B.P. during wet times of the year—erosion of the doned prehistoric beaver dams were seen at the stream outlet is not complete (Fig. 10B). Stand- wetland nor does the stream incise the wetland, We infer that an earthquake along the Ken- ing water ~1 m deep likely overtops the scarp as expected in an abandoned beaver pond. In dall scarp folded the glacial outwash and over- during wetter periods and drains down the fact, the Kendall wetland deposits are all below lying forest soil (Fig. 15; earthquake A). After stream. The only outlet for the wetland is the the modern wetland surface and stacked on top glacial ice left the area in the early Holocene, stream across the scarp, and when the stream is of each other, which shows that the wetland a thin forest soil developed at the site (unit 2). not fl owing, the wetland is internally drained. aggraded to the modern surface—an unlikely This soil was only preserved in the footwall of Local landslides, regional climate changes, or result of beaver damming, where each breach- the present-day fault zone and was tilted almost beavers could alternatively drive changes in wet- ing would have led to incision of the deposits the same amount as the underlying glacial out- land plant succession and stratigraphy. LiDAR impounded by the earlier dam rather than wash strata (Fig. 9A). Reverse faults separated images of the wetland area show no evidence aggradation. distinct dip panels of glacial outwash exposed for local landslides that could have blocked the in the excavation. The dip panels showed that stream in the recent past. Climate changes are Sequence of Events at Hornet Trench movement on the faults folded glacial deposits not a likely source for the stratigraphic changes and created fold relief of 0.8–1.4 m (Table 4) we observed in the wetland based on regional We interpret the stratigraphy and structures during the earliest earthquake along the Ken- climate studies. Pellatt (2001) inferred four from the Hornet trench in the following series dall scarp (Barnett, 2007). Evidence for an early regional climatic periods from analysis of pol- of events (Fig. 15): (1) glacial outwash (unit 1) Holocene earthquake on the Canyon Creek len assemblages in varved cores from Saanich being deposited after ca. 12 ka; (2) a forest soil scarp also consists of folded glacial deposits. We Inlet on Vancouver Island, British Columbia. (unit 2) forming on the outwash surface; (3) fold- infer that some of the folding predated faulting Pellatt’s (2001) climate intervals are: (1) an ing of the glacial outwash during earthquake A; on F1 and F2 in the Smuggler’s trench because early Holocene warm and/or dry period from (4) reverse or oblique reverse faulting on faults both faults cut a buried soil (unit 2) that formed 11,450 to 8300 yr B.P., (2) a warm interval with F3, F4, and F5 during earthquake B, displacing a slight angular discordance with the underly- mild winters from 8300 to 7040 yr B.P., (3) a the outwash and forest soil (unit 2) formed on ing sandy gravel. We note that the scarp just east period of transitional middle Holocene climate the outwash surface; (5) deposition of unit 3, the of the Hornet trench was only ~2 m high (see from 7040 to 5750 yr B.P., and (4) establish- fi rst scarp-derived colluvium and formation of a Fig. 5 and Stellar profi le in Fig. 8E). However, ment of a relatively cool and/or wet late Holo- forest soil (unit 4) on unit 3, draping the scarp; if we include the possibility for broad folding cene climate after 5750 yr B.P. Modern conifer (6) reverse or oblique reverse faulting during extending as much as 200 m south of the scarp, forests and oak savannas occupied the region earthquake C on fault F1, offsetting units 3 and then linear fi ts to the terrace surface allow the by ca. 3800 yr B.P. None of the stratigraphic 4, and deposition of unit 5 in a small fi ssure total displacement to increase to ~3.2 m, agree- changes in the Kendall wetland temporally cor- created during earthquake C; (7) deposition of ing with scarp heights we see elsewhere along relate with inferred regional climate shifts, nor a second scarp colluvium (unit 6); (8) contin- the Kendall scarp. are the cycles of wetting and drying observed in ued scarp erosion and soil development across We interpret earthquake A in the wetland the climate record similar to those inferred from scarp; and fi nally, (9) continued development of adjacent to the Kendall scarp based on the sharp the wetland. modern forest soil (unit 7). stratigraphic contact between a black silt loam Beaver activity is another a possible agent for at the top of the basal mineral deposits and a driving the wetting and drying cycles inferred Sequence of Events at Smuggler Trench gyttja at the base of couplet 1. This sharp con- from the Kendall wetland stratigraphy. Persico tact recorded an abrupt environmental change at and Meyer (2009) described the predominant We interpret the stratigraphy and structures about the same time as deposition of Mazama morphologic expression of beaver dams as a from the Smuggler’s trench in the following ash (Figs. 11 and 12). The abrupt stratigraphic berm typically 5–50 m long built across a fl ood- series of events (Fig. 16): (1) deposition of change suggests that near-surface folding and plain, approximately perpendicular to the valley glacial outwash (unit 1); (2) earthquake A fold- warping during earthquake A grew the scarp

846 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

Soil (unit 2) Deposition of glacial outwash (~13–12 ka). Glacial outwash 1) and 2) Soil develops on surface of glacial outwash.

Earthquake A – blind reverse fault (with possible oblique motion) folds outwash 3) and soil.

Disrupted strata Earthquake B – surface rupture on reverse faults (F3, F4, F5) thrusts glacial outwash FF44 over unit 2 (with possible oblique motion). 4) FF33 Shearing and ground shaking disrupt FF55 stratification in outwash.

Soil on scarp (unit 4) Colluvium (unit 3) Scarp partially collapses and erodes, leaving scarp colluvium (unit 3) with a newly 5) formed soil on scarp (unit 4).

Earthquake C – surface rupture on a reverse fault (F1) cuts across colluvium and soil developed following Earthquake B. Small fissures in 6) hanging wall disrupt units 2, 3, and 4.

Colluvium (unit 6) Unit 5 Deposition of unit 5 as fissures collapse and voids fill with mixture of material from units 2, 7) 3, and 4. Colluvium (unit 6) deposited as scarp collapses and erodes.

Scarp continues to erode as modern soil develops across scarp. 8)

Modern soil (unit 7)

Scarp completely draped with modern soil 9) (unit 7).

Figure 15. Diagram showing retrodeformation sequence for the Hornet trench.

high enough to block the stream, changing the 4.5 ha (~45,600 m2) and a maximum depth of The age of earthquake A is best estimated local hydrology of the wetland from a riparian 3–4 m based on adjacent scarp height (Fig. 5). from the position of Mazama ash relative to the zone and/or forest to a small lake or pond just We interpreted the stratigraphic change from basal contact of couplet 1. The ash is intercalated shortly before deposition of Mazama ash ca. basal sands and forest soil to gyttja as evidence in gyttja just above the basal contact, suggest- 7780–7480 cal. yr B.P. (Zdanowicz et al., 1999). for wetland and/or lake inception and evidence ing that the earthquake created the scarp and We used a GIS lake-ﬁ lling algorithm to estimate of the ﬁ rst Holocene earthquake along the Boul- blocked the stream just prior to eruption and the surface area of the lake at a maximum of der Creek fault. deposition of Mazama ash (Figs. 11 and 12). We

Geosphere, August 2013 847

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

Bending moment normal fault Postglacial forest soil

1) Deposition of glacial outwash (~13–12 ka). 2) Earthquake A – blind reverse fault (with Soil develops on surface of glacial outwash. possible oblique motion) folds outwash and soil, causes bending moment faulting in hanging wall.

Accumulation of clasts at base of scarp

3) Formation of colluvium and forest soil after 4) Earthquake B – reverse or oblique faulting folding. on F1, possibly accompanied by additional folding.

5) and 6) Steep original slope causes rapid 7) Earthquake C – lateral or reverse oblique (?) scarp degradation and deposition of scarp and movement on F2. Only indications of faulting are slope colluvium. Formation of soil across scarp. abrupt changes in unit thickness across fault.

8 and 9) Deposition of youngest scarp colluvium and formation of forest soil across scarp and on slope.

Figure 16. Diagram showing retrodeformation sequence for the Smuggler trench.

848 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

TABLE 4. VERTICAL SEPARATION ESTIMATES FOR PALEOEARTHQUAKES FOR THE KENDALL SCARP from reverse movement on F1 in earthquake B to Maximum Minimum oblique or lateral motion in earthquake C hints vertical vertical Paleo- separation separation at possible partitioning of slip from the Boulder earthquake (cm) (cm) Evidence for vertical separation estimation Creek fault onto the Canyon Creek fault. C 70 40 Vertical separation on the youngest buried soil (unit 4) and oldest In our OxCal model, we base pre-earthquake scarp colluvium (unit 3) along F1 in Hornet trench radiocarbon ages on a sample of charcoal from B7025 Vertical separation on the oldest buried soil (Hornet trench unit 2) unit 4 in the Hornet trench that yielded an age 170* 70* of 1120–930 cal. yr B.P., and plant macrofossils A 140 80 Estimated from folded strata in Hornet trench (Barnett, 2007) from below the basal contact of couplet 3 in the Note: Estimates of vertical separation of unit 2 from Stellar’s Jay trench (Barnett, 2007). wetland that yielded an age of 920–740 cal. yr B.P. For our post–earthquake C sample, we used plant macrofossils from the base of couplet 3 infer that the earthquake and ash deposition were attributed the wide range of ages to mixing of in the wetland that yielded an age of 920–720 closely spaced in time—the earthquake possibly soil by roots and recycling of charcoal from cal. yr B.P. Together, these samples produced an occurred within a few decades or centuries prior older units (e.g., Nelson et al., 2003). OxCal-modeled age for earthquake C of 910– to ash deposition. Because of this close temporal An abrupt change in the wetland stratigra- 740 cal. yr B.P. relationship between earthquake A and Mazama phy from detrital peat at the top of couplet 1 to ash deposition, we used an approximate age gyttja at the base of couplet 2 also signal a sec- Reactivation of Tertiary Faults of Mazama ash for the age of earthquake A in ond Holocene earthquake (Figs. 11 and 12). We our OxCal model (age of eruption rounded to infer that uplift of the hanging wall along the The Boulder Creek fault likely formed in the the nearest century plus 50 additional years of scarp during earthquake B blocked drainage of mid-Tertiary as a normal fault (Miller and Misch, error), yielding a modeled age for earthquake A the forested wetland and created a lake suitable 1963; Tabor et al., 2003), but we infer that the of 8070–7240 cal. yr B.P. (Fig. 14). for depositing diatomaceous gyttja. fault is now aligned in the modern stress fi eld to We used our OxCal model to calculate the accommodate north-south shortening as a reverse Earthquake B—3190–2980 cal. yr B.P. age of earthquake B from charcoal in the Hornet or oblique reverse fault. On the basis of the prox- excavation and plant macrofossils from the wet- imity of the Kendall scarp to the mapped trace Earthquake B resulted in reverse or oblique land cores (Fig. 14). Pre–earthquake B samples of the Boulder Creek fault (Figs. 3–5), we infer reverse displacement along the Kendall scarp of charcoal from unit 2 in the Hornet trench that the fault associated with the Kendall scarp that thrust glacial outwash over Holocene soils yielded an age of 3200–2890 cal. yr B.P. and is the Holocene surface expression of the Boul- and colluvium. In the Hornet trench, we esti- plant macrofossils from the wetland yielded an der Creek fault. As shown in Figures 2B and 3, mated ~25–70 cm of vertical separation on the age of 3340–3080 cal. yr B.P. Post–earthquake B map relationships suggest that the Boulder Creek oldest buried soil (unit 2). We do not infer much macrofossil samples from the wetland yielded an fault is a normal fault with Eocene Chucka nut vertical separation in the second earthquake in age of 3320–2990 cal. yr B.P. Combining these Formation deposits in the hanging wall and pre- the Hornet trench (minimum of 25 cm), so it is ages in our OxCal model yielded a modeled age Tertiary ultramafi c and metamorphic rocks in possible that units 3 and 5 are the same stratum for earthquake B of 3190–2980 cal. yr B.P. the footwall, and probably formed by extension but now a separated by a fault with only a small in a dextral shear zone in the early or mid-Ter- amount of offset. Barnett (2007) estimated ~70– Earthquake C—910–740 cal. yr B.P. tiary (Johnson, 1985). Reverse-fault reactivation 170 cm of reverse vertical separation on unit 2 of dormant normal faults is now recognized during earthquake B in a nearby trench (Table 4). Movement on fault F1 in the Hornet trench throughout many active fold-and-thrust belts The second earthquake at Canyon Creek cut records the youngest earthquake along the Ken- around the world (Jackson, 1980; Sibson, 1985; the folded outwash deposits along F1 and off- dall scarp (Fig. 15). Oblique reverse or reverse Sykes, 1978; Wiprut and Zoback, 2000). Thus, set the ground surface (unit 2). We estimated motion on F1 during earthquake C cut both the the Boulder Creek fault and portions of the Bell- that vertical separation of unit 2 was at least 1 scarp-derived colluvium created by earthquake ingham Basin may be vestigial, normal-faulted m because erosion likely removed all of unit 2 B (unit 3) and the overlying soil (unit 4). Vertical extensional features now actively deforming in from the hanging wall of F1. The unusual thick- separation on the youngest buried soil (unit 4) a north-south shortening strain system. Further- ness of unit 3 suggests either a single large rup- and oldest scarp colluvium (unit 3) along F1 is more, other structural basins and thrust faults ture or a series of smaller ruptures, neither of ~40–70 cm (Table 4). We estimated a minimum bordering migrating Cascadia forearc blocks may which we had evidence for. Alternatively, we vertical separation of ~45 cm because remnants share similar complex histories. suggest that contributions of both scarp and of the cut-off soil were not evident in the hang- The Canyon Creek fault is more enigmatic. slope colluvium at the toe of the scarp following wall of F3. Tabor et al. (2003) inferred a fault beneath the ing earthquake B caused the unique thickness The last earthquake in the Smuggler trench Nooksack River valley, immediately downslope of unit 3. In both trenches, the glacial outwash involved fault F2, which offset the youngest of, and having the same trend as, the Canyon deposits were very loose and subject to collapse buried soil (unit 4) in the trench (Fig. 9B). Beds Creek scarp. Their map shows a fault with right- on exposed surfaces with steep faces, suggest- of gravelly sand in unit 1 showed tens of centi- lateral and down-to-the-east separation south of ing that the outwash deposits could form thick meters of vertical separation, but the beds also Canyon Creek that places Chuckanut Formation colluvial deposits over a short period of time. changed thickness abruptly across the fault, sug- rocks against pre-Tertiary rocks (see southwest Radiocarbon ages on charcoal from three sam- gesting a component of lateral slip juxtaposed corner of Fig. 4 and Grouse Butte on Tabor et al., ples from unit 4 in the Smuggler trench poorly parts of the same strata with varying thickness. 2003). It is possible that oblique motion on this constrain the minimum age of earthquake B to Faulting of unit 4 against unit 3B required at least inferred fault created the Canyon Creek scarp, between ca. 13,700 and ca. 260 cal. yr B.P. We 30 cm of vertical separation. An apparent change but the scarp could also refl ect a reactivated ,

Geosphere, August 2013 849

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

previously unmapped, short splay of the Boul- together, then the maximum length of the sur- However, GPS-derived shortening rates in excess der Creek fault or Glacier Extensional fault. face rupture visible on the LiDAR is ~12 km. of zero extend across the United States–Canada Additional LiDAR surveys in eastern Whatcom Maximum displacements per earthquake are border to latitudes just north of Vancouver, Brit- County may shed light on the style of Holocene difﬁ cult to estimate in the Smugglers Trench ish Columbia, suggesting that shallow faults in the faulting, if any, along the remainder of the Can- on the Canyon Creek fault, but Barnett (2007) North American plate in southern British Colum- yon Creek fault. estimated vertical separations of 80–140 cm bia are likely active as well. The 1946 Vancouver We infer that the three earthquakes on the and 70–170 cm, respectively, for earthquakes Island earthquake, as well as smaller more recent Boulder Creek fault are the same as the three A and B on the Kendall scarp (Table 4). We earthquakes in the region, suggest that shallow earthquakes on the Canyon Creek scarp because estimated a vertical separation of 40–70 cm for upper-plate faults in southern British Columbia of the proximity and similar trends of the fault earthquake C from the Hornet trench (Table 4). are indeed active and pose a hazard to the northern scarps. Lack of intervening scarps between the We used these vertical separations and ages for Puget Lowland (Cassidy et al., 2000; Rogers and two mapped scarps could simply be the result each paleoearthquake from our OxCal modeling Hasegawa, 1978). of erosion by the North Fork Nooksack River. to derive vertical slip rates (Table 5). We also The Boulder Creek fault is capable of produc-

We do not see scarps along the Boulder Creek used the vertical separations, ages, and three ing Mw 6.3 earthquakes, based on empirical rela- fault where the fault veers to the north and fol- estimates for fault dip to calculate horizontal tionships between rupture length and magnitude lows Boulder Creek toward the United States– shortening rates according to methods summa- (Wells and Coppersmith, 1994). Empirical rela- Canada border. Instead, it appears that surface rized by Meghraoui et al. (1988). tionships between maximum displacement and ruptures of the Boulder Creek fault continue Long-term slip rates calculated using the total magnitude indicate the fault is capable of larger

eastward and link with the Canyon Creek fault. scarp height on late Pleistocene glacial outwash earthquakes (Mw 6.6–6.8) (Wells and Copper- We envision surface rupture initiating on one terraces (Columbia Valley and Nooksack River smith, 1994). Earthquakes of this size would fault and transferring to the other fault in the valley) and ages of the terrace surfaces are 0.3 ± produce ground motions sufﬁ cient to cause vicinity of the junction of Boulder Creek and 0.1 mm/yr for the Kendall scarp and 0.65 ± 0.1 damage to urban areas in northern Washington the Nooksack River across an area no wider than mm/yr for the Canyon Creek scarp. Rates calcu- and southern British Columbia, including Bell- several hundred meters. Such fault interactions lated for intervals between earthquakes A–C, A–B, ingham, Vancouver, Victoria, and Abbottsford. between adjacent faults are now widely known and B–C are 0.3 ± 0.1 mm/yr, 0.4 ± 0.1 mm/yr, from modern earthquakes. For example, sur- and 0.5 ± 0.2 mm/yr, respectively (Table 5). Con- CONCLUSIONS face rupture during the M7.9 Denali earthquake verting these rates based on vertical separation (Alaska) in 2002 jumped from the Denali fault to horizontal shortening yields shortening rates We interpret observations from scarp exca- to the Totshunda fault in the Little Tok River Val- between 0.5 ± 0.1 mm/yr and 1.0 ± 0.01 mm/yr vations and coring of a wetland adjacent to the ley, and paleoseismological evidence suggests for a fault dipping 30° (north-south–directed hori- Kendall scarp as evidence for three Holocene this type of interaction happened repeatedly in zontal shortening rates are assumed and increase earthquakes in the northern Puget Lowland. the recent past (Eberhart-Phillips et al., 2003; with lower fault dips, thus our calculations using One of these scarps lies along the Boulder Haeussler et al., 2004; Schwartz et al., 2012). a 30° dip likely approximate the highest rates Creek fault, originally mapped as a Tertiary nor- expected for the Boulder Creek and Canyon mal fault but now reactivated in the Holocene Slip Rates and Comparison with Creek faults). Horizontal shortening rates calcu- as a reverse or oblique reverse fault attributed Geodetic Strain Rates lated from GPS observations average ~1 mm/yr to north-south forearc contraction. These ﬁ nd- for sites at the same latitude (49°N) as the Boul- ings place active faulting much further north The Boulder Creek fault presents a seismic der Creek fault (Mazzotti et al., 2002). Thus, our than previously considered and raises questions hazard to population centers in the Fraser-Puget calculated horizontal shortening rates and GPS- about the possible presence of additional active Lowland. If we assume the two scarps rupture derived horizontal shorting rates agree quite well. crustal faults in southwestern British Columbia.

TABLE 5. SLIP RATES FOR KENDALL AND CANYON CREEK SCARPS

Vertical separation Time span Vo Sh Sh Sh Slip rate description* (m)† (cal. yr)§ (mm/yr)# (30° dip, mm/yr)** (45° dip, mm/yr)** (60° dip, mm/yr)** Vertical separation on Columbia Valley terrace (CVT), 4.0* 12,540 ± 142† 0.3 ± 0.1 0.6 ± 0.1 0.3 ± 0.1 0.2 ± 0.1 Kendall scarp Vertical separation on Nooksack Valley terrace (NVT), 3.4* 12,540 ± 142† 0.3 ± 0.1 0.5 ± 0.1 0.3 ± 0.1 0.2 ± 0.1 Kendall scarp Vertical separation on Canyon Creek scarp 6.9* 12,540 ± 142§ 0.65 ± 0.1 1.0 ± 0.01 0. 6 ± 0.1 0.3 ± 0.1 Three-interval rate using earthquake A to earthquake C 2.1 ± 0.7 6840 ± 313 0.3 ± 0.1 0.5 ± 0.2 0.3 ± 0.1 0.2 ± 0.1 Two-interval rate using earthquake A to earthquake B 1.6 ± 0.5 4570 ± 420 0.4 ± 0.1 0.6 ± 0.3 0.4 ± 0.1 0.2 ± 0.1 Two-interval rate using earthquake B to earthquake C 1 ± 0.4 2260 ± 147 0.5 ± 0.2 0.8 ± 0.3 0.5 ± 0.2 0.3 ± 0.1 *Both the Columbia Valley terrace (CVT) and Nooksack Valley terrace (NVT) are surfaces developed on Sumas advance outwash (Kovanen and Easterbrook, 2001). CVT is not well dated but the surface likely post-dates NVT because CVT cuts out NVT in the lower Columbia Valley; thus, we use the age of NVT as a maximum age for CVT. Only a maximum rate is calculated here. †Vertical separation for scarp estimated from LiDAR proﬁle shown in Figure 9. §Maximum ages of scarps are estimated from glacial outwash terraces. NVT terraces grade in an up-valley direction; approximate age of terrace is from charcoal in sands near the terrace surface at Maple Falls. Faulting and folding in the Canyon Creek trench deforms Nooksack gravel (Smuggler unit 1). The age of NVT probably best approximates the age of outwash gravel in the lower North Fork of the Nooksack River (Kovanen and Easterbrook, 2001). # Vo is the vertical slip rate calculated by dividing vertical separation by time span. Average rates are shown. We account for error by taking the difference between the maximum and minimum rates and dividing by 2 (for a ± error).

**We calculate horizontal shortening rates (Sh) according to the following relationship from Meghraoui et al. (1988): To = Vm + Sh, where To is the calculated dip slip and Vm θ θ is the total vertical slip. Parameters To and Vm were calculated using the following equations: | To | cos( ) = | Sh | and | To | sin( ) = | Vm |. We make three different assumptions about fault dip (θ)—30°, 45°, and 60°—to account for variation in dip of the master fault at depth. Actual dip of master fault is unknown. Average rates are shown—we account for error by taking the difference between the maximum and minimum rates and dividing by 2 (for an average ± error).

850 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Holocene earthquakes in western Washington

ACKNOWLEDGMENTS Mountains of Washington and the Cascade Mountains Kovanen, D., and Easterbrook, D., 2001, Late Pleistocene, of Washington and Oregon: Quaternary Glaciations in post-Vashon, alpine glaciation of the Nooksack drain- We thank landowners from The Glen at Maple the Northern Hemisphere, v. 5, p. 145–159. age, North Cascades, Washington: Geological Society Falls, Cowden Gravel, and the Washington State Eberhart-Phillips, D., and 28 others, 2003, The 2002 of America Bulletin, v. 113, p. 274–288, doi:10.1130 Department of Natural Resources for granting us Denali fault earthquake, Alaska: A large magnitude, /0016-7606(2001)113<0274:LPPVAG>2.0.CO;2. slip-partitioned event: Science, v. 300, p. 1113–1118, Lapen, T.J., 2000, Geologic map of the Bellingham access to their property to excavate trenches and doi:10.1126/science.1082703. 1:100,000 quadrangle, Washington: Washington Divi- collect wetland cores. Stephen Personius, Keith England, T., and Calon, T., 1991, The Cowichan fold and sion of Geology and Earth Resources Open File Report Knudsen, and Thomas Pratt provided helpful reviews thrust system, Vancouver Island, southwestern British 2000-5, scale 1:100,000. of an earlier draft. Richard Blakely provided informa- Columbia: Geological Society of America Bulletin, Lienkaemper, J.J., and Bronk Ramsey, C., 2009, OxCal: tion on the size and shape of the Bellingham Basin. v. 103, no. 3, p. 336–362. Versatile tool for developing paleoearthquake chronol- Constructive comments from two anonymous reviews Fumal, T.E., Pezzopane, S.K., Weldon, R.J., II, and ogies–A primer: Seismological Research Letters, v. 80, greatly improved the manuscript. Schwartz, D.P., 1993, A 100-year average recurrence p. 431–434, doi:10.1785/gssrl.80.3.431. interval for the San Andreas fault at Wrightwood, Mazzotti, S., Dragert, H., Hyndman, R.D., Miller, M.M., California: Science, v. 259, p. 199–203, doi:10.1126 and Henton, J.A., 2002, GPS deformation in a region REFERENCES CITED /science.259.5092.199. of high crustal seismicity: N. Cascadia forearc: Earth Gavin, D., 2001, Estimation of inbuilt age in radiocarbon and Planetary Science Letters, v. 198, p. 41–48, doi: Armstrong, J.E., and Brown, W.L., 1954, Late Wisconsin ages of soil charcoal for fi re history studies: Radiocar- 10.1016/S0012-821X(02)00520-4. marine drift and associated sediments of the lower Fraser bon, v. 43, p. 27–44. McCaffrey, R., Qamar, A.I., King, R.W., Wells, R., Khaz- Valley, British Columbia, Canada: Geological Society Haeussler, P., Schwartz, D., Dawson, T., Stenner, H., Lien- aradze, G., Williams, C.A., Stevens, C.W., Vollick, J.J., of America Bulletin, v. 65, p. 349–364, doi:10.1130 kaemper, J., Sherrod, B., Cinti, F., Montone, P., Craw, and Zwick, P.C., 2007, Fault locking, block rotation /0016-7606(1954)65[349:LWMDAA]2.0.CO;2. P., and Crone, A., 2004, Surface rupture and slip dis- and crustal deformation in the Pacifi c Northwest: Geo- Armstrong, J.E., Crandell, D.R., Easterbrook, D.J., and tribution of the Denali and Totschunda faults in the physical Journal International, v. 169, no. 3, p. 1315– Noble, J.B., 1965, Late Pleistocene stratigraphy 3 November 2002 M 7.9 earthquake, Alaska: Bulletin 1340, doi:10.1111/j.1365-246X.2007.03371.x. and chronology in southwestern British Columbia and of the Seismological Society of America, v. 94, no. 6B, Meghraoui, M., and Doumaz, F., 1996, Earthquake-induced northwestern Washington: Geological Society of Amer- p. S23–S52, doi:10.1785/0120040626. fl ooding and paleoseismicity of the El Asnam, Algeria, ica Bulletin, v. 76, p. 321–330, doi:10.1130/0016-7606 Harrington, J.A., and Clark, D.H., 2011, Reassessing a pro- fault-related fold: Journal of Geophysical Research, (1965)76[321:LPSACI]2.0.CO;2. posed large late-glacial alpine advance near Mt. Baker, v. 101, p. 17,617–17,644, doi:10.1029/96JB00650. Barnett, E.A., 2007, Active faulting at the northeast mar- Washington: Geological Society of America Abstracts Meghraoui, M., Philip, H., Albarede, F., and Cisternas, A., gin of the greater Puget Lowland: A paleoseismic and with Programs, v. 23, no. 4, p. 88. 1988, Trench investigations through the trace of the magnetic-anomaly study of the Kendall fault scarp, Hopkins, W.S., 1968, Subsurface Miocene rocks, British 1980 El Asnam thrust fault: Evidence for paleoseis- Whatcom County, northwest Washington [M.S. thesis]: Columbia–Washington, a palynological investigation: micity: Bulletin of the Seismological Society of Amer- Arcata, California, Humboldt State University, 65 p. Geological Society of America Bulletin, v. 79, p. 763–768, ica, v. 78, p. 979. Blakely, R.J., Wells, R.E., Weaver, C.S., and Johnson, S.Y., doi:10.1130/0016-7606(1968)79[763:SMRBCA]2.0 Miller, G.M., and Misch, P., 1963, Early Eocene angular 2002, Location, structure, and seismicity of the Seattle .CO;2. unconformity at western front of the Northern Cascades, fault zone, Washington: Evidence from aeromagnetic Hyndman, R.D., Mazzotti, S., Weichert, D., and Rogers, Whatcom County, Washington: AAPG Bulletin, v. 47, anomalies, geologic mapping, and seismic-refl ection G.C., 2003, Frequency of large crustal earthquakes p. 163–174. data: Geological Society of America Bulletin, v. 114, in Puget Sound–southern Georgia Strait predicted Mustard, P.S., and Rouse, G.E., 1994, Stratigraphy and evo- p. 169–177, doi:10.1130/0016-7606(2002)114<0169: from geodetic and geological deformation rates: lution of the Tertiary Georgia Basin and subjacent Late LSASOT>2.0.CO;2. Journal of Geophysical Research, v. 108, 2033, doi: Cretaceous strata of the Greater Vancouver area, British Booth, D.B., 1994, Glaciofl uvial infi lling and scour of the 10.1029/2001JB001710. Columbia, in Monger, J.W.H., ed., Geology and Geo- Puget Lowland, Washington, during ice-sheet glacia- Jackson, J.A., 1980, Reactivation of basement faults and logical Hazards of the Vancouver Region, Southwestern tion: Geology, v. 22, p. 695–698, doi:10.1130/0091-7613 crustal shortening in orogenic belts: Nature, v. 283, British Columbia: Geological Survey of Canada Bul- (1994)022<0695:GIASOT>2.3.CO;2. p. 343–346, doi:10.1038/283343a0. letin 481, p. 97–161. Brocher, T.M., and 22 others, 2001, Upper crustal structure Johnson, S.Y., 1985, Eocene strike-slip faulting and non- Nelson, A.R., Johnson, S.Y., Kelsey, H.M., Wells, R.E., in Puget Lowland, Washington: Results from the 1998 marine basin formation in Washington, in Biddle, K.T., Sherrod, B.L., Pezzopane, S.K., Bradley, L.-A., seismic hazards investigation in Puget Sound: Journal and Christie-Blick, N., eds., Strike-Slip Deformation, Koehler, R.D., III, and Bucknam, R.C., 2003, Late of Geophysical Research, v. 106, p. 13,541–13,564, Basin Formation, and Sedimentation: Tulsa, Okla- Holocene earthquakes on the Toe Jam Hill fault, Seattle doi:10.1029/2001JB000154. homa, Society of Economic Paleontologists and Min- fault zone, Bainbridge Island, Washington: Geological Bronk Ramsey, C., 1995, Radiocarbon calibration and eralogists, p. 283–302. Society of America Bulletin, v. 115, p. 1388–1403, doi: analysis of stratigraphy: The OxCal program: Radio- Johnson, S.Y., Otton, J.K., and Macke, D.L., 1987, Geology 10.1130/B25262.1. carbon, v. 37, p. 425–430. of the Holocene surfi cial uranium deposit of the north Osborn, G., Menounos, B., Ryane, C., Riedel, J., Clague, Bronk Ramsey, C., 2009, Bayesian analysis of radiocarbon fork of Flodelle Creek, northeastern Washington: Geo- J.J., Koch, J., Clark, D., Scott, K., and Thompson dates: Radiocarbon, v. 51, p. 337–360. logical Society of America Bulletin, v. 98, p. 77–85, Davis, P., 2012, Latest Pleistocene and Holocene gla- Bucknam, R.C., Hemphill-Haley, E., and Leopold, E.B., doi:10.1130/0016-7606(1987)98<77:GOTHSU>2.0 cier fl uctuations on Mount Baker, Washington: Qua- 1992, Abrupt uplift within the past 1700 years at .CO;2. ternary Science Reviews, v. 49, p. 33–51, doi:10.1016 southern Puget Sound, Washington: Science, v. 258, Johnson, S.Y., Potter, C.J., and Armentrout, J.M., 1994, /j.quascirev.2012.06.004. p. 1611–1614, doi:10.1126/science.258.5088.1611. Origin and evolution of the Seattle fault and Seattle Pellatt, M., Hebda, R., and Mathewes, R., 2001, High-reso- Cassidy, J.F., Rogers, G.C., and Waldhauser, F., 2000, basin, Washington: Geology, v. 22, p. 71–74, doi: lution Holocene vegetation history and climate from Characterization of active faulting beneath the Strait 10.1130/0091-7613(1994)022<0071:OAEOTS>2.3 Hole 1034B, ODP leg 169S, Saanich Inlet, Canada: of Georgia, British Columbia: Bulletin of the Seismo- .CO;2. Marine Geology, v. 174, p. 211–222, doi:10.1016 logical Society of America, v. 90, p. 1188–1199, doi: Johnson, S.Y., Blakely, R.J., Stephenson, W.J., Dadisman, /S0025-3227(00)00151-1. 10.1785/0120000044. S.V., and Fisher, M.A., 2004a, Active shortening of Persico, L., and Meyer, G., 2009, Holocene beaver dam- Daneš, Z.F., Bonno, M.M., Brau, E., Gilham, W.D., Hoff- the Cascadia forearc and implications for seismic haz- ming, fl uvial geomorphology, and climate in Yellow- man, T.F., Johansen, D., Jones, M.H., Malfait, B., Mas- ards of the Puget Lowland: Tectonics, v. 23, TC1011, stone National Park, Wyoming: Quaternary Research, ten, J., and Teague, G.O., 1965, Geophysical investiga- doi:10.1029/2003TC001507. v. 71, p. 340–353, doi:10.1016/j.yqres.2008.09.007. tions of the southern Puget Sound area, Washington: Johnson, S.Y., and 10 others, 2004b, Evidence for late Holo- Porter, S., and Swanson, T., 1998, Radiocarbon age con- Journal of Geophysical Research, v. 70, p. 5573–5580, cene earthquakes on the Utsalady Point fault, northern straints on rates of advance and retreat of the Puget lobe doi:10.1029/JZ070i022p05573. Puget Lowland, Washington: Bulletin of the Seis- of the Cordilleran ice sheet during the last glaciation: Dragovich, J.D., Norman, D.K., Haugerud, R.A., and Prin- mological Society of America, v. 94, p. 2299–2316, Quaternary Research, v. 50, p. 205–213, doi:10.1006 gle, P.T., 1997a, Geologic map and interpreted geo- doi:10.1785/0120040050. /qres.1998.2004. logic history of the Kendall and Deming 7.5-minute Kelsey, H.M., Sherrod, B.L., Blakely, R.J., and Haugerud, Reimer, P.J., and 27 others, 2009, IntCal09 and Marine09 quadrangles, western Whatcom County, Washington: R.A., 2012, Holocene faulting in the Bellingham radiocarbon age calibration curves, 0–50,000 years cal Washington State Division of Geology and Earth forearc basin: Upper-plate deformation at the north- BP: Radiocarbon, v. 51, p. 1111–1150. Resources Open-File Report 97-2, 39 p, 3 plates. ern end of the Cascadia subduction zone: Journal of Rogers, G.C., and Hasegawa, H.S., 1978, A second look at the Dragovich, J.D., Zollweg, J.E., Qamar, A.I., and Norman, Geophysical Research, v. 117, B03409, doi:10.1029 British Columbia earthquake of June 23, 1946: Bulletin of D.K., 1997b, The Macaulay Creek thrust, the 1990 /2011JB008816. the Seismological Society of America, v. 68, p. 653–675. 5.2-magnitude Deming earthquake, and Quaternary Kovanen, D.J., and Easterbrook, D.J., 1998, 10,600-yr-old Schoeneberger, P.J., Wysocki, D.A., Benham, E.C., and geologic anomalies in the Deming area, western What- age of post–Cordilleran-Ice-Sheet alpine glaciation in Broderson, W.D., 2002, Field book for describing and com County, Washington: Cause and effects?: Wash- the Nooksack Valley, North Cascades, WA: Geologi- sampling soils, version 2.0: Lincoln, Nebraska, Natural ington Geology, v. 25, no. 2, p. 15–27. cal Society of America Abstracts with Programs, v. 30, Resources Conservation Service, National Soil Survey Easterbrook, D., 1985, Stratigraphy and chronology of qua- no. 7, p. A-165. Center, 298 p. ternary deposits of the Puget Lowland and Olympic

Geosphere, August 2013 851

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021 Sherrod et al.

Schwartz, D.P., Haeussler, P.J., Seitz, G.G., and Dawson, Sibson, R.H., 1985, A note on fault reactivation: Journal of Wells, D.L., and Coppersmith, K.J., 1994, New empirical T.E., 2012, Why the 2002 Denali fault rupture took Structural Geology, v. 7, p. 751–754, doi:10.1016/0191 relationships among magnitude, rupture length, rupture the Totschunda fault: Implications for fault branching -8141(85)90150-6. width, rupture area, and surface displacement: Bul- and seismic hazards: Journal of Geophysical Research, Sieh, K.E., 1978, Prehistoric large earthquakes produced by letin of the Seismological Society of America, v. 84, v. 117, B11304, doi:10.1029/2011JB008918. slip on the San Andreas Fault at Pallett Creek, Califor- p. 974–1002. Seidlecki, E.M., 2008, The geometry and earthquake history nia: Journal of Geophysical Research, v. 83, p. 3907– Wells, R.E., and Simpson, R.W., 2001, Northward migration of the Boulder Creek fault zone, Whatcom County, 3939, doi:10.1029/JB083iB08p03907. of the Cascadia forearc in the northwestern U.S. and Washington [M.S. thesis]: Bellingham, Western Wash- Stuiver, M., and Polach, H., 1977, Discussion: Reporting of implications for subduction deformation: Earth, Planets, ington University, 81 p. 14C data: Radiocarbon, v. 19, p. 355–363. and Space, v. 53, p. 275–283. Sherrod, B.L., 2001, Evidence for earthquake-induced Sykes, L.R., 1978, Intraplate seismicity, reactivation of Wells, R.E., Weaver, C.S., and Blakely, R.J., 1998, Fore- subsidence about 1100 yr ago in coastal marshes of preexisting zones of weakness, alkaline magmatism, arc migration in Cascadia and its neotectonic signifi - southern Puget Sound, Washington: Geological Soci- and other tectonism postdating continental fragmenta- cance: Geology, v. 26, p. 759–762, doi:10.1130/0091 ety of America Bulletin, v. 113, p. 1299–1311, tion: Reviews of Geophysics and Space Physics, v. 16, -7613(1998)026<0759:FAMICA>2.3.CO;2. doi:10.1130/0016-7606(2001)113<1299:EFEISA>2.0 p. 621–688, doi:10.1029/RG016i004p00621. Wiprut, D., and Zoback, M., 2000, Fault reactivation and .CO;2. Tabor, R.W., Haugerud, R.A., Hildreth, W., and Brown, fl uid fl ow along a previously dormant normal fault in Sherrod, B.L., Brocher, T.M., Weaver, C.S., Bucknam, E.H., 2003, Geologic map of the Mount Baker 30- by the northern North Sea: Geology, v. 28, p. 595–598, R.C., Blakely, R.J., Kelsey, H.M., Nelson, A.R., and 60-minute quadrangle, Washington: U.S. Geological doi:10.1130/0091-7613(2000)28<595:FRAFFA>2.0 Haugerud, R.A., 2004, Holocene fault scarps near Survey Geologic Investigations Series I-2660, scale .CO;2. Tacoma, Washington, USA: Geology, v. 32, p. 9–12, 1:100,000. Wohl, E., 2006, Human impacts to mountain streams: Geo- doi:10.1130/G19914.1. Troels-Smith, J., 1955, Characterization of unconsolidated morphology, v. 79, p. 217–248, doi:10.1016/j.geomorph Sherrod, B.L., Blakely, R.J., Weaver, C.S., Kelsey, H.M., sediments: Copenhagen, Geological Survey of Den- .2006.06.020. Barnett, E., Liberty, L.M., Meagher, K.L., and Pape, K., mark, Series IV, 73 p. Zdanowicz, C.M., Zielinski, G.A., and Germani, M.S., 2008, Finding concealed active faults: Extending the Washington Division of Geology and Earth Resources staff, 1999, Mount Mazama eruption: Calendrical age veri- southern Whidbey Island fault across the Puget Low- 2005, Digital 1:100,000-scale geology of Washington fi ed and atmospheric impact assessed: Geology, v. 27, land, Washington: Journal of Geophysical Research, State, version 1.0: Washington Division of Geology p. 621–624, doi:10.1130/0091-7613(1999)027<0621: v. 113, B05313, doi:10.1029/2007JB005060. and Earth Resources Open File Report 2005-3. MMECAV>2.3.CO;2.

852 Geosphere, August 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/4/827/3345917/827.pdf by guest on 27 September 2021