Revised Earthquake Hazard of the Hat Creek Fault, Northern California: a Case Example of a Normal Fault Dissecting Variable-Age Basaltic Lavas

Home , Hat Creek (California)

Revised earthquake hazard of the Hat Creek fault, northern California: A case example of a normal fault dissecting variable-age basaltic lavas

Matthew W. Blakeslee and Simon A. Kattenhorn* Department of Geological Sciences, University of Idaho, 875 Perimeter Drive, MS 3022, Moscow, Idaho 83844-3022, USA

ABSTRACT INTRODUCTION histories and earthquake potential for this fault style, as well as to provide an improved regional Normal faults in basalt have distinctive Normal faults are prevalent in basalt environ- seismic hazard assessment related to this fault. surface-trace morphologies and earthquake ments in response to the common association These techniques provide a viable alternative to evidence that provide information about between basaltic volcanism and rifting. Such traditional paleoseismologic analyses, such as the slip behavior and earthquake potential. faults have distinctive surface morphologies trenching, which are ill-suited for the analysis The 47-km-long Hat Creek fault in northern where they cut through near-surface lavas (Pea- of faulted lavas. California (USA), a useful case example of cock and Parfi tt, 2002; Grant and Kattenhorn, The Hat Creek fault is located within a vol- this fault style, is a segmented fault system 2004; White and Crider, 2006; Rowland et al., canic corridor between Mount Shasta and Las- located along the western margin of the 2007; Ferrill et al., 2011) and remain active dur- sen Peak, near the southern end of the Cascade Modoc Plateau that is a regional earthquake ing volcanic periods such that variably aged lava Range and its associated underlying subduction hazard. In response to interaction with spo- fl ows cut by the fault can be used as temporal system (Fig. 1) (Wills, 1991; Muffl er et al., radically active volcanic systems, surface markers of slip rates and slip history. We use the 1994; Blakely et al. 1997; Walker, 2008). Nor- ruptures have progressively migrated west- case example of the Hat Creek fault in north- mal faulting and recurring volcanic activity ward since the late Pleistocene, with older eastern California (USA) to illustrate the effi - from more than 500 vents over the past 7 m.y. scarps being successively abandoned. The cacy of using offset lava fl ows to constrain slip created a pervasively faulted volcanic region most recent earthquake activity broke the surface through predominantly ca. 24 ka basaltic lavas, forming a scarp with a maximum throw of 56 m. Past work by others identifi ed 7–8 left-stepping scarp segments with a combined length of 23.5 km, but did not explicitly address the throw characteristics, fault evolution, slip history, or earthquake potential. We address these defi ciencies in our understanding of the fault system with new fi eld observations and mapping that suggest the active scarp contains 2 addi- Figure 1. Terrain map of north- tional segments and is at least 6.5 km longer ern California with the location McCloud than previously mapped, thus increasing the of the Hat Creek fault (box) knowledge of the regional seismic hazard. relative to populated areas and Our work details scarp geomorphic styles Fall River Cascades volcanoes Mount Mills and slip-analysis techniques that can be Shasta and Lassen Peak. CA— applied to any normal-faulted basalt envi- California; NV—Nevada; ID— Burney Modoc ronment. Applied to the Hat Creek fault, Idaho; OR—Oregon. Plateau we estimate that a surface-breaking rup-

ture could produce an earthquake of ~Mw Old Station (moment magnitude) 6.7 and a recurrence interval of 667 ± 167 yr in response to a rapid slip rate in the range 2.2–3.6 mm/yr, creating a moderate risk given a lack of historical earthquake events.

*Corresponding author

Geosphere; October 2013; v. 9; no. 5; p. 1397–1409; doi:10.1130/GES00910.1; 9 ﬁ gures. Received 11 February 2013 ♦ Revision received 10 July 2013 ♦ Accepted 13 August 2013 ♦ Published online 13 September 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Blakeslee and Kattenhorn

in the vicinity of Lassen Peak (Muffl er et al., fault in the region. It is highly segmented and is talus piles that refl ect gradual geomorphic mod- 1994). The fault is located in the extending arc composed of three subparallel systems of scarps ifi cation of the scarps to slope angles of ~30°– and/or backarc transition of the Cascadia sub- of different ages (Fig. 2A) that accrued a cumu- 45°. There is no evidence of disruption of the duction zone, marking the approximate west- lative throw in excess of 600 m (Muffl er et al., talus slopes by recent surface rupture, consistent ern margin of a Miocene and younger volcanic 1994; Walker, 2008). The oldest and largest with these scarps refl ecting an older, abandoned highland called the Modoc Plateau (White and system of scarps, referred to as the Rim, has as portion of the fault system. Lava fl ows at the top Crider, 2006). The Modoc Plateau hosts numer- much as ~350 m of throw and defi nes the eastern- of the footwall east of Murken Bench have been ous Neogene and Quaternary normal faults, most extent of the fault system. The 47-km-long K-Ar dated as 924 ± 24 ka (Clynne and Muf- similar in style to the Basin and Range prov- Pleistocene Rim consists of seven right-step- fl er, 2010), constraining the maximum age of ince to the east (LaForge and Hawkins, 1986). ping, northwest-oriented segments ranging in the fault system as Calabrian (late Pleistocene). The north-northwest-trending, west-dipping length from ~1–16 km (Walker, 2008). These The intermediate-aged fault scarps west of Hat Creek fault is the most prominent normal scarps are heavily vegetated and have prominent the Rim are colloquially referred to here as the

′

′ A ′ B

°25

1°3

°25 °30 40°55′ N

121

40°55′ N Cinder Butte

Murken Bench 40°50′ N

40°50′ N 2

Rim Previously 4 mapped Pali 40°45′ N Active Scarp 40°45′ N Active Scarp Newly 5 mapped Active Scarp

6 40°40′ N Elevation (m) Elevation (m) High : 2104 High : 2104

Low : 870 7 Low : 870 40°40′ N 0 4 8 km 0 2 4 km

Figure 2. Digital elevation model of the Hat Creek fault (elevation range in meters) derived from the U.S. Geological Survey 30 m national elevation data set. (A) Traces of west-dipping fault scarps in the Hat Creek fault system. The three scarp system components are informally named (from oldest to youngest) the Rim, the Pali, and the Active Scarp. (B) Enlargement of the Active Scarp fault trace. Seven identiﬁ ed segments are numbered and shown in black. Newly mapped additions to the Active Scarp in the north (red) follow the base of one of the Rim segment scarps.

1398 Geosphere, October 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Earthquake hazard of the Hat Creek fault

Pali (a Hawaiian term for eroded basaltic cliffs) Although this region of California is seis- slip characteristics of the Hat Creek fault were and have accrued as much as ~175 m of throw mically active (Fig. 3), including small events documented in the USGS Quaternary fault data- (Walker, 2008). The Pali, also Pleistocene in (M<3.5) in the vicinity of the Hat Creek fault, base by Sawyer (1995) (http://geohazards.usgs age, extends for ~24 km and is made up of fi ve the Active Scarp has not experienced a surface- .gov/cfusion/qfault/); however, that analysis also left-stepping segments with generally north ori- breaking earthquake event in recorded history combines multiple faults into one system (59 km entations in the southern part of the fault system (~200 yr for northern California). Nonetheless, cumulative length), and the details of the fault where the Pali intersects the Rim, but chang- in the southern portion of the fault system, the slip history are poorly constrained, with a sug- ing to north-northwest orientations in the north Active Scarp offsets glacial deposits (younger gested recurrence interval in the range 1000– where the Pali segments approach the volcanic than 15 ka) by 20 m (Muffl er et al., 1994; U.S. 3000 yr and a slip rate of 1–5 mm/yr. edifi ce at Cinder Butte (Fig. 2A). Many of the Geological Survey, 1996), indicating that 35% Our investigation of the Hat Creek fault segments are overlapping and exhibit physically of the total maximum throw in the northern part tightly constrains offset and timing history that connected (i.e., breached) relay ramps, creating of the Active Scarp and perhaps as much as can be used to refi ne and advance seismic haz- a mechanically continuous system of interacting 80% of the average throw along the entire fault ard assessment. In conjunction with a revised segments. length has accrued since these deposits formed, cumulative length of the Active Scarp, the his- The youngest system of scarp segments, and suggesting that motion along the fault likely tory provides a more accurate estimate of the referred to here as the Active Scarp, has a continued from the late Pleistocene into the earthquake potential of that portion of the fault maximum displacement of 56 m just north of Holocene. system that is likely to rupture in a single event. Murken Bench (Fig. 2B) and exhibits evidence The U.S. Geological Survey (USGS) Quater- At risk are numerous local towns (Burney, Fall of repeated earthquake activity since the late nary fault database (http://earthquake.usgs.gov River Mills, Susanville, Red Bluff, and Redding Pleisto cene. Surface-breaking ruptures of the /hazards/qfaults) currently gives an incomplete are all within 90 km of the Hat Creek fault), a Active Scarp follow the older Pali scarps within a picture of the fault and its recent activity. The hydroelectric infrastructure within 20 km of few tens of meters of the base of the Pali, except Hat Creek fault is listed as having been active the fault system, the Allen Telescope Array at the southernmost Active Scarp segment, which in the past 15 k.y.; however, the database does the Hat Creek Radio Observatory, only 1.5 km occurs a few tens of meters from the base of a not discriminate between different parts of the west of the Active Scarp, and Lassen Volcanic southern Rim segment (Fig. 2A). The majority fault system that have been active at different National Park, 25 km to the south. of the Active Scarp offsets the Hat Creek Basalt, times (i.e., Rim, Pali, and Active Scarp). The The long-term geometric and kinematic a low-potassium olivine tholeiite that covers USGS also provides various earthquake hazard evolution of the fault system in the context of much of the hanging wall valley fl oor west of the estimates for the Hat Creek fault system in the regional tectonics was described in Walker fault scarp. The basalt originated from a cluster 2008 National Seismic Hazards Maps database (2008) and Walker and Kattenhorn (2008), out- of vents 30 km to the south of the throw maxi- (http://geohazards.usgs.gov/cfusion/hazfaults lining temporally variable tectonic and mag- mum and has been 40Ar/39Ar dated as 24 ± 6 ka _search/disp_hf_info.cfm?cfault_id=8,%209); matic infl uences that affected fault trace geom- (Turrin et al., 2007). These lavas fl owed north however, the earthquake magnitude estimates etries. For example, whereas the Rim is made down the Hat Creek valley, which is bounded (which range from M6.5–7.2) combine three up of seven right-stepping segments, the later on its western side by a ca. 500–800 ka volcanic different faults (Hat Creek, McArthur, and May- scarps of the Pali and the Active Scarp all have escarpment (there is no western bounding fault fi eld faults; 97 km cumulative length) that are a consistent left-stepping en echelon pattern antithetic to the Hat Creek fault). unlikely to rupture in tandem. Details about the (Fig. 2). This change in geometry likely refl ects

Legend Figure 3. Seismic events from 1968 to 2011 in the general M 2.0–2.49 area of the Hat Creek fault. Data obtained from the North- M 2.5–2.99 Hat Creek fault ern California Earthquake M 3.0–3.49 Data Center (http://quake.geo .berkeley.edu/). Despite being a M 3.5–3.99 seismically active area, the Hat Creek fault itself has produced > M 4.0 very few events and no historic large earthquakes. White Fault Trace star shows a moderate (M5.7) earthquake on 23 May 2013. Scale 0 30 60 km M5.7 (5/23/13)

Geosphere, October 2013 1399

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Blakeslee and Kattenhorn

the infl uence of regional tectonic patterns, par- We recorded all observation locations along ticularly the Walker Lane belt to the southeast, the fault using a handheld Trimble GeoExplorer which transferred dextral shear into the Lassen 6000 GPS device, which was also used to D T region (Blakely et al., 1997; Muffl er et al., record throw along the Active Scarp with a 2008), creating the left-stepping segment pat- vertical precision of as little as ~10 cm. Throw θ terns and potentially driving the ongoing activity profi les for the majority of the Active Scarp of the later scarps (Walker, 2008). are based on data derived from stereo imagery (Walker, 2008); however, GPS elevations were ACTIVE SCARP OBSERVATIONS also obtained to derive throw profi les along portions of the Active Scarp by physically walking Field and Analytical Methods out both the hanging wall and footwall cutoffs and collecting GPS data points at 5 s intervals. Figure 4. Relationship between throw (T) Our observations of the morphology of the All data were imported into an ArcGIS environ- and displacement (D) along a normal fault surface trace of the Hat Creek fault are based ment and superposed on a 30 m digital elevation with dip, θ. GPS field measurements of on seven separate fi eld visits to all segments model basemap (Fig. 2). footwall and hanging wall cutoff elevations of the fault (Rim, Pali, and Active Scarp; Fig. GPS fi eld data record locations and elevations indicate T. Slip rates along faults are based 2A) between 2003 and 2012. The scarp can be and thus only account for the vertical compo- on the cumulative amount of displacement, reached via numerous access roads and hik- nent of fault motion, or throw (T). Our analysis determined using D = T/sin θ. ing trails that lead off of State Route 89, which of the fault motion history also considers the comes within 1 km of the fault at segment 6 component of motion in the plane of the fault (Fig. 2B). A gravel road that is along much of itself, referred to as the slip or displacement 2009). Lava fl ows from Cinder Butte covered the top of the Rim provides numerous vantage (D). These parameters are related using the fault the northern end of the Pali scarp; however, the points along the Rim footwall. These visits dip in the subsurface (θ) such that T = D sin θ Active Scarp subsequently dissected these lavas allowed us to collect extensive fi eld descriptions (Fig. 4). and curved toward the center of Cinder Butte of the scarp morphology along the entire length (Fig. 5), indicating more recent motion on the of the fault, with emphasis on the most recent Scarp Geometry and Surface Morphology fault relative to Cinder Butte volcanism. The surface ruptures along the Active Scarp. fi nal stage of Cinder Butte volcanism ended at In all locations visited along the Active Scarp, The Active Scarp was previously mapped 38 ± 7 ka with the eruption of the basaltic ande- we made detailed observations of the manner in as consisting of 7 (Walker, 2008) or 8 (Muffl er site of Cinder Butte (Turrin et al., 2007). which the fault had interacted with the youngest et al., 1994) left-stepping segments spaced from The majority of the Active Scarp ruptures lava fl ows. The two main elements of the fault 0.79 to 1.83 km apart and with a cumulative through the 24 ± 6 ka Hat Creek Basalt, which trace, hanging wall monoclines and vertical length of 23.5 km. The discrepancy in number covers much of the Hat Creek valley and accu- scarp faces (described in the following), were of segments documented in past work simply mulated at the base of preexisting scarps of the examined to unravel the progressive sequence refl ects the interpretative nature of separating Rim (in the south) and the Pali (further north). of disruption of the lavas by repeated fault out connected segments. We adopt the inter- Rupture of the Active Scarp through the Hat motions. We noted the dimensions of columnar pretation of 7 segments along the previously Creek Basalt resulted in surface features that blocks of lava that had been disrupted or moved identifi ed extent of the Active Scarp (Fig. 2B). are characteristic of active normal faults cut- out of place along both the vertical scarp and the The strike of the Active Scarp is subparallel to ting basaltic lava fl ows (Gudmundsson, 1987a, upper surface of the hanging wall monocline. the Rim and Pali segments in the southern third 1987b, 1992; Grant and Kattenhorn, 2004; Relative exposure ages of different portions of of the fault system (segments 5–7 in Fig. 2B) Martel and Langley, 2006), including verti- the vertical scarp were qualitatively linked to the where the Rim and Pali converge; however, in cal scarps where the fault breached the surface amount of lichen coverage of exposed columns the northern part of the fault system, both the along columnar joints and subsequently accrued in the scarp face. Pali and the Active Scarp diverge from the domi- throw, dilational cracks and fi ssures along the We did not date samples of lava fl ows cut by nant trend of the older scarp system, follow- base of the scarp, and as much as ~40 m wide the Active Scarp, relying instead on robust ages ing a northwest trend toward the ~335-m-tall, zones of basalt rubble at the base of the scarp provided by previously published works. Gen- 8.5-km-wide shield volcano Cinder Butte created by repeated episodes of fault rupture eral fl ow directions of lavas were determined (Fig. 2). The westward migration of successive (Fig. 6A). based on fl ow extents relative to known source scarp system segments of the Hat Creek fault Although geomorphic processes have affected vents for fl ows of different ages, also docu- suggests that Cinder Butte and its underlying scarps of all ages along the Hat Creek fault, mented previously (Muffl er et al., 1994; Turrin magmatic system may have focused the devel- evident in the rubble piles that collected along et al., 2007). These fl ow directions were used opment of the active portion of the fault in the Rim and Pali scarps in response to mass wast- to identify where lavas had fl owed toward, and proximity of Cinder Butte. Such a model is con- ing, earthquake-related ground shaking effects draped across, existing portions of fault scarps sistent with documented examples of magmatic are the dominant geomorphic modifi er along the located at lower elevations than the source vents. systems affecting fault growth and orientation in relatively youngest Active Scarp. For example, Where lavas of different ages abut or overlap, response to local stress perturbations related to signifi cant disaggregation of the network of we attempted to discern the locations of the lava magma pressure (Clifton and Schlische, 2003; small columns (generally <20 cm across and contacts in the fi eld; however, the vegetation Clifton and Kattenhorn, 2006; Rowland et al., <30 cm tall) within the rapidly cooled upper density commonly makes such distinctions dif- 2007; Gudmundsson et al., 2009) or the topo- surface of the Hat Creek lavas is common even fi cult, in which case aerial photos and Google graphic and mechanical attributes of volcanic where lava surfaces are subhorizontal, imply- Earth were used to approximate the contacts. constructs (Friese, 2008; Jenness and Clifton, ing strong ground shaking (Fig. 6C). In some

1400 Geosphere, October 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021

Earthquake hazard of the Hat Creek fault

Creek lava Creek limit of Hat Hat of limit ek and rps are rps are the Active the Legend Pali Previously mapped Previously Scarp Active Rim Newly mapped Active Scarp maximum throw on maximum throw scarp (175 m) Pali 24 ± 6 ka Hat Creek Basalt Hat Creek maximum throw on maximum throw Rim scarp (350 m)

maximum throw on maximum throw Scarp (56 m) Active

eek lava eek

Hat C Hat limit of limit maximum throw on northernmaximum throw Scarp (30 m) of Active segments

ava

salt l 53.5 ± 2 ka

Six Mile Hill CA 89 basalt west of basalt west

s mit of Hat Creek Ba li 38 ± 7 ka Figure 5. Perspective view down and to the northeast of the northern portion of the Hat Creek fault system, showing the Hat Cre 5. Perspective view down and to the northeast of northern portion Hat Creek Figure the Cinder Butte lavas in relation to the Active Scarp (red), the Pali (green), and the Rim (blue). The newly mapped portion of and the Rim (blue). the Pali (green), Active Scarp (red), to the Butte lavas in relation the Cinder points on the various sca Butte. Locations and amounts of maximum throw Scarp (yellow) traces the Rim scarp northeast of Cinder as shown. Image courtesy of Google Earth Pro. Cinder Butte

der Butte lava basaltic andesite of basaltic andesite

limit of Cin Cassel/Fall River Mills road Mills River Cassel/Fall

Geosphere, October 2013 1401

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Blakeslee and Kattenhorn

A instances, individual columns were lifted up out vertical scarp of place and onto the surface of the fl ow, leaving cavities within the fl ow top that can be matched to individual column blocks like puzzle pieces, implying ground accelerations suffi cient to

overcome the weight of the blocks (i.e., >1 gn). Large basalt columns (1–2 m wide and several meters tall) within the hanging wall have tilted, fallen over, or broken apart adjacent to the rubble from scarp, suggesting these areas have undergone collapsed signifi cant ground shaking. This shaking also monocline caused similar-sized columns in the footwall to fall from the vertical scarp and infi ll the fi ssure below, exposing less-weathered and relatively lichen-free surfaces along the scarp wall. Older surfaces are coated by lichen of species Rhizo- B carpon with maximum diameters of ~24 mm that imply an exposure time of at least 250 yr (e.g., Bull, 2000), although lichenometric dating techniques are not robust. The Active Scarp is also characterized by a hanging wall fault-trace monocline (Fig. 6B), representing the near-surface fl exing of the Hat Creek Basalt above an upward-propagating fault tip prior to initial surface breaching by the fault. Such monoclines are common along normal faults that rupture to the surface through young basalt fl ows (Grant and Kattenhorn, 2004; Martel and Langley, 2006; White and Crider, 2006; Rowland et al., 2007). Along the Active Scarp, the monocline accounts for as much as 33 m of throw (Walker, 2008), imply- C ing numerous fault slip events prior to breaching of the surface along the upper hinge line of the fold. Once the surface was breached by the fault, monocline growth ceased, and all subsequent surface throw accumulation was directed along the vertical scarp. Progressive disaggregation of the monocline occurred along the Active Scarp in response to the local effects of repeated fault rupture, with the fi nal stage of monocline history being its complete collapse, rendering the monocline to a pile of rubble (Fig. 6A). The rubble pile is commonly composed of intact columns of basalt with individual dimensions of several meters and with large open cavities between the blocks. Monocline disaggregation is most advanced along fault segments with the largest Figure 6. (A) View to the southeast of the northern end of segment 6 (Fig. cumulative throws (segments 2–6 in Fig. 2), and 2B) of the Active Scarp. The 22-m-high vertical scarp represents surface is least advanced where throw decreases toward breaching of a monocline along its upper hinge line. Progressive destruc- the segment tips or at relay zones, implying a tion, interpreted to be the result of earthquake shaking, has reduced the relationship between cumulative fault activity monocline to a pile of rubble in this location. (B) Intact monocline just and monocline breakdown. north of Murken Bench, along segment 2 of the Active Scarp (Fig. 2B). Taken together, these characteristics of Monocline height here is ~21 m (Walker, 2008). (C) Disruption of basalt monoclines in various stages of breakdown sug- blocks on the upper surface of the Hat Creek lavas at the top of the mono- gest repeated earthquake-induced ground shak- cline in B. Inferred earthquake-induced ground accelerations in excess ing events. Intact or only partially disaggregated

of 1 gn have resulted in ejected blocks (arrow) on top of the lava ﬂ ow breached monoclines can be seen along portions surface. Scale at center of image is 15 cm long. of the fault where the scarp height is at least

1402 Geosphere, October 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Earthquake hazard of the Hat Creek fault

6 m, implying that multiple surface-breaching of measurement to ascertain the total tectonic a result of emplacement across the southward earthquake events are needed to induce mono- offset. In the case of the Hat Creek fault, how- sloping surface of the older lavas of the basalt cline disaggregation or destruction after initial ever, the Active Scarp postdates both the Cinder west of Six Mile Hill (Fig. 5). breaching of the monocline upper hinge line. Butte and Hat Creek lava fl ows and formed on Young ruptures through the lavas along the the hanging wall side of the older Pali scarps. base of the northern Rim scarp show striking Maximum Fault Throw Hence, any offset of the Cinder Butte lavas similarities to the previously documented sur- across the Active Scarp must be purely tectonic. face ruptures along the fault segments south of The northern portion of the Active Scarp The southern margin of the south-fl owing Cin- Cinder Butte. This portion of the fault was not (segments 1 and 2; Fig. 2B), north of Murken der Butte lavas on both the footwall and hang- previously identifi ed as part of the Active Scarp, Bench, crosses the northern extent of the Hat ing wall sides of the Active Scarp, where it is with presumed late Pleistocene most recent Creek Basalt along the fault (Fig. 5), beyond onlapped by the north-fl owing Hat Creek lavas, activity in the USGS Quaternary fault database. which the fault ruptures through basaltic ande- exhibits at least 70 m of cumulative fault throw. The young ruptures along the base of the north- site (38 ± 7 ka) originating from several vents A second measurement of 83 m of throw was ern Rim scarp motivate us to test if they are part concentrated at Cinder Butte (Turrin et al., taken ~150 m north of the fi rst measurement. of the Active Scarp system (i.e., rupture in tan- 2007). The maximum observed throw of 56 m In this location, the Active Scarp and Pali scarp dem with previously identifi ed active segments), (determined from GPS data with 10 cm verti- appear to merge (Fig. 7E); therefore, a portion resulting in a longer fault length that should be cal precision) offsets the northern extent of the of this 83 m offset may be geomorphic, indicat- considered in a seismic hazard analysis, or if Hat Creek lavas at the transition to older Cinder ing that 70 m is the more reliable measurement they represent independently rupturing parts of Butte lavas. Larger offsets are apparent for the of minimum total tectonic throw for the Cinder the overall Hat Creek fault system. upper surface of the Cinder Butte lavas (at least Butte lavas. The combination of geomorphic To map the young scarps along the northern 70 m but potentially as much as 83 m), indicat- and tectonic offset of young lavas highlights the Rim northeast of Cinder Butte, we walked the ing 14–27 m of throw accrual along the fault importance of incorporating the relative tim- entire length of the rupture (both footwall and during the ~14 k.y. period between the erup- ing and propagation directions of fault-growth hanging wall), mapping it with differential GPS tion of the Cinder Butte lavas and the Hat Creek events and lava-fl ow episodes and advancement to capture the segment geometries and throw lavas near the end of the Pleistocene. directions into fault-offset analyses in tectoni- distribution. We identifi ed two new rupture seg- Southward-fl owing lavas from Cinder Butte cally active volcanic settings in order to obtain ments (Fig. 2B) extending a total of 4 km, break- were deformed and offset by the northern por- appropriate slip rates. ing the surface ~50 m west of the base of the tions of the Pali (and later the Active Scarp), Rim scarp with a maximum throw of 30 m in which propagated northwest toward Cinder Revised Active Scarp Length the southern segment. The surface rupture loca- Butte as the fault system seemingly responded tion relative to the base of the Rim is identical to the Cinder Butte magmatic system (Figs. The previously identifi ed northern extent to the previously documented Active Scarp seg- 7A, 7B) (Walker, 2008). The uncertainty in the of the Active Scarp terminates within Cinder ments south of Cinder Butte, which break verti- amount of Active Scarp offset of the Cinder Butte (Muffl er et al., 1994). Our investigation cally to the surface through the Hat Creek lavas Butte lavas derives from the fact that later lavas of the fault system and fi eld mapping northeast several tens of meters from the base of the Pali of this eruptive period fl owed south along the of Cinder Butte suggests the continuation of scarps. The southern termination of the newly axis of the Pali scarps, resulting in bifurcation young rupture activity, implying the length identifi ed young rupture is located 4.5 km north- of lava fl ows into two lobes that fl owed onto the of the Active Scarp has been underestimated. In east of the northern termination of the previously fault footwall and down onto the hanging wall this region, the base of the Rim scarp created identifi ed Active Scarp (segment 1) within Cin- (Fig. 7C). The resultant vertical offset of the a buttress for lavas fl owing across the hanging der Butte (Fig. 2B). The northern termination is two lobes of the lava fl ow was thus not tectonic, wall of the Rim (Fig. 5). Some of these lavas ~0.8 km north of where the fault is crossed by but geomorphic. Subsequent activity along the fl owed eastward away from eruptive center at the Cassel–Fall River Mills road. Additional Rim fault after the eruption of the Cinder Butte lavas Cinder Butte, covering an older fl ow surface of segments of the Hat Creek fault continue for at requires some portion of the total offset to be the basalt west of Six Mile Hill, dated as 53.5 ± least 6 km north-northwest beyond this point to fault-related throw. In contrast, Hat Creek lavas 2 ka (Muffl er et al., 2012). The lavas of the just north of the Pit River; however, no evidence fl owed from the south and so accumulated only basalt west of Six Mile Hill fl owed southward of recent activity was confi rmed by this study. in the hanging wall of the Pali scarp (Fig. 7D). along the base of the Rim, following a local Similar to the geometry of cracks along the These lavas were subsequently deformed as the slope toward the throw maximum along the Active Scarp segments south of Cinder Butte, Active Scarp broke to the surface through them, Rim 8.5 km south of the source vents immedi- the newly discovered rupture zone contains such that the entire 56 m of throw within these ately south of the Cassel–Fall River Mills road fracturing that has a left-stepping geometry, lavas is decidedly fault-related (Fig. 7E). (Fig. 5). The surface morphologies of these indicating similar rupture kinematics (normal Ordinarily, it may be impossible to distin- lavas are distinct from older Pleistocene fl ows with a small right-lateral component). Displace- guish between the geomorphic and tectonic exposed in the Rim, defi ning a youthful but ment of the late Pleistocene lavas has created components of offset where lavas have fl owed rugged lava surface with numerous tumuli and features identical to those observed along the along both the footwall and hanging wall sides defl ation pits. The roughness of this lava sur- Active Scarp within Hat Creek Basalt, such as of an active fault because the height of the scarp face complicates the identifi cation of the east- vertical scarps, dilational cracks and fi ssures, would need to be known at the time of lava ern extent of the Cinder Butte lavas. The Cinder the presence of a large fault trace monocline emplacement across the fault (which is unlikely Butte lavas appear to abut the base of the Rim in the hanging wall, and a large rubble zone to be determinable). In such cases, the scarp scarp due east of the high point of Cinder Butte created by ground shaking and the breakdown height during lava emplacement would be sub- but likely did not reach the Rim to the north- of the monocline (Fig. 8). The vertical fault tracted from the cumulative offset at the time east of the high point (Muffl er et al., 2012) as scarp exposes older Pleistocene basalt near its

Geosphere, October 2013 1403

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Blakeslee and Kattenhorn

A Cinder Butte B Cinder Butte

Pali scarp Early Cinder Pali scarp Early Cinder Butte lavas Butte lavas propagation Pleistocene lavas Pleistocene lavas of fault

>40 ka (<53.5 ka) ~40 ka?

C Cinder Butte D Cinder Butte

geomorphic offset of Late Cinder Butte lavas

Late Cinder Pali scarp Late Cinder Pali scarp Butte lavas Butte lavas

Hat Creek lavas lava partially Pleistocene lavas covers scarp Pleistocene lavas

~38 ka ~24 ka

E Cinder Butte

vertical fault scarp (dilational fault) Late Cinder Pali scarp Butte lavas Active Scarp

Hat Creek lavas Pleistocene lav 56 m

as monocline

0 ka

Figure 7. Interpreted sequence of volcanism and fault activity where the Active Scarp cuts into Cinder Butte from the south. (A) Develop- ment of Cinder Butte, the lavas of which overlie the 53.5 ± 2 ka basalt west of Six Mile Hill, resulted in a new portion of the Hat Creek fault (the Pali) branching away from the older Rim scarps and propagating northwest toward Cinder Butte. (B) Fault offset of older Cinder Butte lavas by continued growth of the Pali scarp to the northwest. (C) Eruption of later Cinder Butte lavas at 38 ± 7 ka, emplacing lava ﬂ ows onto both the footwall and hanging wall sides of the fault and creating a geomorphic offset of the lava. (D) Eruption of Hat Creek lavas at 24 ± 6 ka. Northward transport of these lava ﬂ ows along the hanging wall of the Pali scarp resulted in onlap of Cinder Butte lavas. (E) Ongoing fault activity was manifested by a vertical fault (the Active Scarp) cutting up through Hat Creek lavas, forming a fault-trace monocline and ultimately a vertical scarp, producing a total 56 m of vertical offset of Hat Creek lavas.

southern end where late Pleistocene lavas did tion of the basalt west of Six Mile Hill) and pos- than basaltic andesite lavas from Cinder Butte, not reach the base of the Rim; however, where sibly within the past 38.5 ± 7 k.y., if the basaltic with a signifi cantly greater amount of surface the youngest lavas abut the Rim, all of the throw andesite lavas of Cinder Butte reached the Rim lichen. Nonetheless, the fault scarp features in is taken up within those lava fl ows, partially in the vicinity of the young ruptures. the basalt west of Six Mile Hill are similar to by the monocline and partially by throw along The northernmost segment of the newly iden- other portions of the Active Scarp, including a the vertical scarp. Therefore, the accumulation tifi ed scarp defi nitively cuts through the basalt variably disaggregated hanging wall monocline of the total throw must have occurred at least west of Six Mile Hill. Where exposed along the and a vertical fault scarp. The throws within this within the past 53.5 ± 2 k.y. (i.e., since the erup- fault scarp, this lava appears noticeably older unit are as much as 23 m within the northern-

1404 Geosphere, October 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Earthquake hazard of the Hat Creek fault

A features consistent with the Active Scarp, such Rim scarp as vertical scarps fl anked by a fault-trace monocline, the disaggregated appearance due to the collapse of the monocline during earthquakes, and offsets of relatively young lava fl ows, suggest these fault segments have both undergone recent rupture with signifi cant ground shaking surfacesurface tracetrace ofof newlynewly mappedmapped Active Scarp segments and should thus be considered together when evaluating the credible earthquake magnitude and seismic hazard potential of the region. To test this assertion, we explore fault evolution and earthquake scenarios in which the newly mapped segment is fi rst treated as an indepen- B dent fault and then considered to be incorpo- monocline rated into the entire Active Scarp system.

Estimation Method for Earthquake Slip and Recurrence

Empirical relationships between rupture length and maximum surface displacement during discrete earthquake events (e.g., Wells and Coppersmith, 1994; Wesnousky, 2008) per- mit the estimation of slip rates and recurrence intervals between earthquakes if the ages of offset layers are known. For example, Wells and Coppersmith (1994) presented empirical data for normal fault lengths in the range 3.8 km to 75 km to generate a regression line for maxi- Figure 8. Surface morphology of newly mapped segments of the mum displacement (MD, in meters) versus sur- Active Scarp northeast of Cinder Butte. (A) The surface trace of face rupture length (SRL, in kilometers). The the fault within lavas of the basalt west of Six Mile Hill consists relationship is given by log (MD) = –1.98 + of a variably intact, disaggregated, or collapsed monocline fl anking 1.51 × log (SRL) for normal faulting. Maximum a vertical scarp (view to the east). (B) View of an intact portion of displacement per event can be used to deter- the monocline along the newly mapped segments, within the basalt mine the number of events required to accrue west of Six Mile Hill. Cinder Butte is visible on the skyline (view to the cumulative displacement along the fault. If the southwest). the age of the oldest offset unit is known, the number of events in this time interval informs us about the recurrence interval between events. most recently active segment of the fault. The east of Cinder Butte defi ne a 4.5 km right step This method uses the simplifying assumption of combination of evidence based on surface mor- and a 2.5 km along-strike gap within the fault characteristic earthquake events (equal displace- phologies suggests that the Hat Creek fault rup- system, increasing the total length of young ment per event and constant rupture length). tured through at least three late Pleistocene lava ruptures by 6.5 km compared to prior estimates. Nonetheless, it provides a reasonable insight fl ows of different ages, with young surface rup- Although predominantly a left-stepping fault into earthquake activity in the absence of addi- ture both south of Cinder Butte (the previously system in response to dextral-oblique extension, tional information such as paleoseismologic identifi ed Active Scarp) and northeast of Cinder this right step in the fault geometry is interpreted evidence from scarp trenching, which is typi- Butte, along the base of old Rim segments. to result simply from the effect of the Cinder cally limited to only the last several events and Butte magmatic system on the temporal evolu- is not well suited to paleoseismological analysis DISCUSSION tion of the southern portion of the fault. Large in faulted basalt. Moreover, the Active Scarp steps are not uncommon in segmented normal represents the reactivation of an existing fault If the newly mapped rupture trace is the fault systems and are not necessarily hindrances system at depth (the Pali and the Rim), which continuation of the Active Scarp to the north- to earthquake ruptures. For example, steps from controlled the rupture length of the Active Scarp east of Cinder Butte, it indicates that the most 3 to 8 km wide were associated with both the as it developed through young lavas. The surface

recent surface ruptures along the Hat Creek fault 1954 Dixie Valley (moment magnitude, Mw 6.8) trace length thus likely remained approximately

extend further north than previously considered. and the 1915 Fairview Peak (Mw 7.2) earth- constant through time, as did the maximum dis- Therefore, the Hat Creek fault has remained quakes in Nevada (Zhang et al., 1991). placement per event. active along the Rim in the northern part of the There are numerous lines of evidence to Regression relationships are based on maxi- fault system, despite having abandoned the Rim suggest that the newly identiﬁ ed segments are mum surface displacement, not throw (which portion of the fault system further south. The part of the Active Scarp system and rupture in is the vertical component of the fault displace- additional segments of the Active Scarp north- tandem. For example, the various morphologic ment). Faults scarps at Hat Creek are vertical

Geosphere, October 2013 1405

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Blakeslee and Kattenhorn

within ~50 m of the surface, where they break therefore more likely that the newly mapped ever, no clear evidence for young rupture (e.g., through the Hat Creek lavas (Muffl er et al., segments are part of a larger system that rup- through alluvial deposits that cross portions of 1994; Walker, 2008); however, slip along the tures less frequently—specifi cally, the entire the fault) was observed in that area during a fault below this depth occurs along a dipping Active Scarp portion of the Hat Creek fault. reconnaissance survey. fault plane, resulting in a component of dila- Mechanical interaction and resultant slip tion at the surface along the vertical scarp. Slip Rate Analysis partitioning in segmented normal fault systems This phenomenon is typical of dilational faults results in higher throws (both cumulative and in basalt lavas (e.g., Grant and Kattenhorn, As additional evidence in support of this per earthquake event; e.g., Dawers and Anders, 2004; Ferrill et al., 2011). Throw is converted assertion, we consider the throw distribution 1995; Willemse, 1997) and higher slip rates for to displacement assuming a typical subsurface along the entire length of the Active Scarp. It is segments near the centers of fault systems. For normal fault dip of 60° (Anderson, 1951). This well documented that interacting normal fault example, the 387 km Wasatch fault in Utah has assumption is reasonable given that geomor- segments distribute throw throughout the length a slip rate of 1–2 mm/yr in the central segments, phically modifi ed scarps of the Pali and Rim of a fault system (i.e., kinematic coherence; decreasing to 0.5 mm/yr in the distal segments have typical dips of ~45°, indicating originally Figs. 9A, 9B), typically producing the maximum (Machette et al., 1991). Analogously, the active higher dips. throw at the center of the fault trace (e.g., Walsh portion of the Hat Creek fault exhibits variable and Watterson, 1991; Dawers et al., 1993; Cart- slip rates in different segments along its length. Earthquake Potential of Newly wright et al., 1995; Willemse, 1997). The throw To characterize the earthquake potential, includ- Mapped Segments commonly exhibits an approximately elliptical ing slip rates, slip-per-event, recurrence inter- distribution, attenuating from the maximum at vals, and earthquake magnitude along a fault The newly mapped, recent rupture segments the center of the surface trace to zero at the fault system consisting of multiple segments that rup- along the northern portion of the Rim have tips, with local variability at segment bound aries ture in tandem, we use the assumption that the a total rupture length of 4 km (i.e., within the within the fault system related to either fault segment containing the maximum cumulative range used in the Wells and Coppersmith regres- growth history (Childs et al., 1995; Kattenhorn throw should be used. In so doing, we account sion). If these segments rupture independently and Pollard, 2001) or the partial accommodation for the maximum amount of throw that neces- of any other segments of the Hat Creek fault, the of fault throw by relay ramp deformation (Hug- sarily accumulated during a determined time slip rate and recurrence interval must accom- gins et al., 1995; Blakeslee, 2012). interval: in this case, the age of the Hat Creek modate the offset of the 53.5 ± 2 ka basalt west Along the previously identifi ed seven seg- lavas. Segment 2 has the maximum throw (56 m of Six Mile Hill. Given the maximum throw of ments of the 23.5 km Active Scarp, throw is in 24 ± 6 ka lavas) and hence the highest slip 30 m (corresponding to 34.6 m of displacement distributed among segments by mechanical rate averaged since the late Pleistocene: 2.7 along the fault plane; Fig. 4), the newly identi- interaction, whereby the fault segments are mm/yr, or in the range 2.2–3.6 mm/yr given the fi ed young fault scarps would have an associ- affected by the presence of one another despite uncertainty of the Hat Creek lava age. This esti- ated slip rate of ~0.65 mm/yr (or in the range spacings or underlaps between segments of mate assumes a subsurface fault dip of 60° and 0.6–0.7 mm/yr when accounting for lava age hundreds of meters to several kilometers (Fig. so a total displacement of 64.7 m in the plane uncertainty). Using the Wells and Coppersmith 2B). However, the throw versus distance profi le of the fault (Fig. 4). We conservatively assume (1994) regression, a normal fault with a length along these seven segments (Walker, 2008) does pure dip-slip motion along the fault, although of 4 km (which is somewhat low for a surface- not show a maximum throw at the center of the a slight dextral component of motion may be breaking fault rupture) should have an average total length (Fig. 9C). Instead, the throw pro- present based on the presence of left-stepping maximum displacement of only ~8.5 cm per fi le is greatly skewed toward its northern end, fractures along the surface rupture trace. Hence, rupture event (equivalent to ~7.5 cm of throw), where the 56 m maximum throw displaces the actual slip rates may be slightly higher than we implying a low recurrence interval of 134 ± 5 yr Hat Creek Basalt within segment 2, just south calculate. to accrue 30 m of throw since the faulted lava of Cinder Butte (Fig. 5). The throw profi le The maximum throw of the Cinder Butte was erupted. shape suggests that the previously identifi ed lavas (at least 70 m) is less defi nitive; however, Rupture of these fault segments could pro- seven segments of the Active Scarp do not rep- this estimate would require a minimum slip rate

duce a Mw 5.8 earthquake, using the regres- resent the full length of the active fault system, of 2.1 mm/yr (or in the range 1.8–2.6 mm/yr sion relationship between moment magnitude requiring mechanical interaction with more taking into account lava age uncertainty) since and maximum displacement per event (MD, in segments north of segment 1. The addition of these slightly older lavas erupted, approximately meters), given as M = 6.61 + 0.71 × log (MD) the newly mapped, young segments northeast consistent with the slip rate deduced from the (Wells and Coppersmith, 1994). Such an event of Cinder Butte extends the active fault system offset of the Hat Creek lavas. Compared to other would probably be felt regionally in northeast- northward, producing a maximum throw more normal fault slip rates, such as the Wasatch ern California. For example, a M5.7 earthquake centered along the rupture length and an over- fault, the Hat Creek fault slip rate is relatively along a normal fault located ~65 km south- all throw distribution with a more symmetric, high. The 0.65 mm/yr slip rate computed for southeast of the southern end of the Hat Creek somewhat elliptical shape (Fig. 9C). The throw the newly mapped segments northeast of Cinder fault on 23 May 2013 was felt as far south as the pattern implies that the newly mapped segments Butte is consistent with those segments being San Francisco Bay area, as well as north into are kinematically coherent with the previously at the distal end of the coseismic rupture seg- southern Oregon and east into central Nevada identifi ed Active Scarp, rather than being part ments of the Active Scarp. Analogously, at the (USGS earthquake database: http://earthquake of an independent fault. Nonetheless, the throw southern end of the Active Scarp, segment 6 has .usgs.gov). However, no seismic events have profi le is somewhat skewed toward the north an effective slip rate 1.0 mm/yr (or in the range been attributed to the fault in recorded human end, raising the possibility of additional active 0.8–1.4 mm/yr taking into account lava age history in the region (~200 yr), and there is segments farther north (Figs. 2 and 3), where the uncertainty). The observed attenuation of the no fi eld evidence of a very recent rupture. It is Hat Creek fault approaches the Pit River; how- slip rate to the distal segments is thus consistent

1406 Geosphere, October 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Earthquake hazard of the Hat Creek fault

Figure 9. (A) Idealized ellipti- AB cal distribution of throw or slip along the length of a continuous normal fault composed of a single segment. Thick line below

Throw or Slip Throw

graph shows map view of fault or Slip Throw trace. (B) Distribution of throw along a 650-m-long normal fault composed of seven mechanically interacting segments in the Distance Along Fault Trace Distance Along Fault Trace Bishop Tuff in eastern Califor- nia (after Willemse et al., 1997). Dashed curve shows the approx- imation of an elliptical profi le. Distal segments underrepresent this distribution. Actual fault C dip direction was not specifi ed. (C) Cumulative throw profi le 60 2 of the Active Scarp (monocline New segments height plus fault scarp height). Active Scarp Profi les in red are derived 50 from stereo imagery and indicate seven previously identi- fied segments of the Active 40 Scarp (numbered as in Fig. newly mapped 2B). Roughness in the profi les segments 3 refl ects technique uncertainty 30 and surface topography of the Hat Creek lavas. The profi le in 4 6 blue shows the newly mapped (m) Throw 1 segments at the northern end of 20 7 the fault system based on GPS

data gap 5 data collected in the ﬁ eld with 10 cm vertical precision. The 10 new segments cumulatively cre- ate a more elliptical throw pro- ﬁ le with the maximum throw 0 closer to the geometric center 01015202530355 the fault trace. Distance Along Fault Trace (km)

with our assertion that the newly mapped seg- andesite lavas erupted. Although the highest Pali Willemse, 1997), the throw peak along segment ment is part of the Active Scarp system. scarps are ~5.4 km south of the southern extent 2 appears to fall above any choice of ellipse to Slip rate estimates provide some insights of the Cinder Butte lavas (Fig. 5), the geomor- approximate the overall throw profi le. There are into the timing of the long-term evolution of the phic offset of bifurcating lava lobes fl owing many reasons why this throw peak may have fault system. For example, a hypothesized stress onto the footwall and hanging wall blocks of the occurred. One possibility is that throw has been perturbation induced by the Cinder Butte mag- Pali scarp closer to Cinder Butte (Fig. 7C) may underestimated along the Active Scarp seg- matic system caused a new branch of the fault to nonetheless have been quite signifi cant. This ments south of the throw peak as a result of the propagate away from the Rim and redirect fault possibility strengthens the argument that only throw being partitioned onto other fault seg- activity toward Cinder Butte, forming the Pali the throw offset of Cinder Butte lavas across the ments during earthquakes. For example, in the scarp system west of the original Rim scarps. relatively younger Active Scarp can be reliably vicinity of Active Scarp segments 3 and 4, there Given the 175 m maximum throw along the Pali used to estimate fault slip rates. are two north-oriented fault scarps that appear system and the simplifying assumption of a con- One fi nal consideration regarding slip rate to link the Pali with the Rim (Fig. 2). Although stant a slip rate of 2.7 mm/yr, the evolution of analysis relates to the seemingly anomalous we found no clear evidence of recent rupture the Pali may have commenced ca. 65 ka, indi- nature of the throw peak along segment 2 (Fig. along those segments, the Hat Creek lavas did cating ~30 k.y. of activity in the Cinder Butte 9C) where the maximum throw of 56 m was not reach these segments, which may make magmatic system prior to the eruption of the measured by GPS. Although approximately recent rupture evidence diffi cult to identify. youngest lavas at 38 ± 7 ka. In this time inter- elliptical throw profi les have been noted along Another possibility is that the overall pattern val, the Pali scarps could have acquired maxi- other mechanically interacting segmented nor- of throw is highly skewed toward the north- mum throws of ~70 m by the time the basaltic mal fault systems (Dawers and Anders, 1995; ern end of the Active Scarp, with a maximum

Geosphere, October 2013 1407

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Blakeslee and Kattenhorn

at segment 2. Such skewed cumulative throw CONCLUSIONS ACKNOWLEDGMENTS profi les have been noted (e.g., Willemse et al., We thank Erin Walker for developing the fault 1996) where fault segments mechanically Field observations and mapping along the evolution history in our related study, Marie Jackson interact with a nearby perturbing infl uence, Hat Creek fault lead us to propose the existence for contributing the photogrammetry database of the such as another fault (or in this case, perhaps of two previously overlooked northern segments Active Scarp heights of the Hat Creek fault used in Cinder Butte). of the Active Scarp, providing new insights our throw profi les, Nicole Bellino for lichen sam- pling and analysis, Leslie Fernandes and Tom Sawyer Ultimately, throw profi les simply refl ect the about the geometry, evolution, and seismic for fi eld assistance, and Patrick Muffl er and Robert cumulative effects of interactions between dif- potential of the fault. The newly mapped seg- Krantz for helpful discussions. Portions of this work ferent components of a constantly evolving segments are kinematically coherent with the pre- were funded under National Science Foundation grant mented fault system; therefore, no faults have viously identifi ed Active Scarp segments, with EAR-1113677 and a Seed Grant from the University of Idaho. Digital elevation models in Figure 2 were truly elliptical throw profi les, and local peaks a surface morphology in 53.5 ± 2 ka lava fl ows derived from 30 m resolution data obtained from the are not uncommon (Fig. 9B) (cf. Dawers and identical to surface ruptures along the previ- U.S. Geological Survey National Elevation Dataset Anders, 1995; Cartwright and Mansfi eld, 1998). ously mapped Active Scarp within the 24 ± (http://nationalmap.gov/viewer.html). Earthquake data Our careful matching of Hat Creek lavas across 6 ka Hat Creek Basalt to the south. Near the used in the production of Figure 3 were obtained the Active Scarp at segment 2 using fi eld-based center of the reinterpreted Active Scarp sys- from the Northern California Earthquake Data Cen- ter (http://quake.geo.berkeley.edu/). We thank Juliet GPS measurements leave us confi dent in the tem, a maximum of 56 m of offset of Hat Creek Crider and Patrick Muffl er for their thoughtful reviews accuracy of the 56 m throw accumulation (and Basalt implies a late Pleistocene–Holocene slip of the original manuscript. hence our calculated slip rate) since the Hat rate of 2.2–3.6 mm/yr, implying a very active REFERENCES CITED Creek lavas erupted, pooled against the existing extensional fault system with high strain rates, Pali scarp, and were subsequently offset by the possibly refl ecting the contribution of a local Anderson, E.M., 1951, The dynamics of faulting and dyke development of the Active Scarp (Fig. 7). magmatic extension component to the overall formation with applications to Britain: Edinburgh, Oli- ver & Boyd, 206 p. strain budget. Blakely, R.J., Christiansen, R.L., Guffanti, M., Wells, R.E., Revised Credible Earthquake Magnitude Based on earthquake magnitude and slip- Donnelly-Nolan, J.M., Muffl er, L.J.P., Clynne, M.A., per-event scaling relationships for the reinter- and Smith, J.G., 1997, Gravity anomalies, Quaternary vents, and Quaternary faults in the southern Cascade The additional active segments increase the preted 30 km rupture length, we estimate a Range, Oregon and California: Implications arc and originally mapped rupture length by 6.5 km credible earthquake magnitude of M 6.7 backarc evolution: Journal of Geophysical Research, w v. 102, p. 22,513–22,527, doi:10.1029/97JB01516. to a total of 30 km. Applying the surface rup- and a recurrence interval of 667 ± 167 yr to Blakeslee, M.W., 2012, Segmented normal fault evolution: ture length to maximum displacement scaling account for the cumulative throw. As one of Interpretation of relay zone deformation and seismic relationship for normal faults described earlier the most prominent faults in the area, but hazard analysis [M.S. thesis]: Moscow, University of Idaho, 70 p. (Wells and Coppersmith, 1994), a 30 km fault one that lacks historical earthquake events Bull, W.B., 2000, Lichenometry: A new way of dating and should manifest a maximum displacement of (the timing of the last event is unknown but locating prehistorical earthquakes, in Noller, J.S., et al., 1.78 m per rupture event (or 1.54 m of maxi- is likely in excess of 200 yr ago in order to eds., Quaternary Geochronology: Methods and Appli- cations: American Geophysical Union Reference Shelf mum throw assuming a fault dip of 60° in the predate historical records of earthquakes in the Volume 4, p. 521–526, doi:10.1029/RF004p0521. subsurface), implying a recurrence interval of region), the active portion of the fault is thus Cartwright, J.A., and Mansfi eld, C.S., 1998, Lateral displacement variation and lateral tip geometry of normal 667 ± 167 yr for the 56 m throw maximum in a potential signifi cant earthquake hazard for faults in the Canyonlands National Park, Utah: Jour- 24 ± 6 ka lavas. This recurrence interval neces- northeastern California. nal of Structural Geology, v. 20, p. 3–19, doi:10.1016 sitates ~15 Holocene rupture events along the The Hat Creek fault study provides a con- /S0191-8141(97)00079-5. Cartwright, J.A., Trudgill, B.D., and Mansfi eld, C.S., 1995, Active Scarp in addition to the late Pleistocene text for the evaluation of earthquake hazards, Fault growth by segment linkage: An explanation for activity that postdated the eruptions of the basalt slip rate analysis, and recurrence interval scatter in maximum displacement and trace length west of Six Mile Hill, Cinder Butte basaltic determination in tectonically active volcanic data from the Canyonlands grabens of SE Utah: Jour- nal of Structural Geology, v. 17, p. 1319–1326, doi: andesite, and the Hat Creek Basalt. environments. Young basalt lavas can be rea- 10.1016/0191-8141(95)00033-A. The concomitant potential earthquake mag- sonably well dated using Ar-Ar analysis, and Childs, C., Watterson, J., and Walsh, J.J., 1995, Fault overlap zones within developing normal fault systems: nitude is Mw 6.7 based on maximum displace- thus provide useful temporal offset markers. Journal of the Geological Society, v. 152, p. 535–549, ment versus magnitude relationships for normal True tectonic offsets can be distinguished from doi:10.1144/gsjgs.152.3.0535. faults, as previously described (Wells and any geomorphic offset caused by lavas fl owing Clifton, A.E., and Kattenhorn, S.A., 2006, Structural archi- tecture of a highly oblique divergent plate bound- Coppersmith, 1994). Magnitude versus rupture across existing scarps through fi eld analysis of ary segment: Tectonophysics, v. 419, p. 27–40, doi: area relationships (M = 3.93 + 1.02 × log (RA), fl ow directions and characteristics relative to 10.1016/j.tecto.2006.03.016. with RA in square kilometers) will provide the developing fault scarps. Using the simplify- Clifton, A.E., and Schlische, R.W., 2003, Fracture popula- tions on the Reykjanes Peninsula, Iceland: Compari- same result for a seismogenic thickness of 15 km ing assumption of characteristic earthquake son with experimental clay models of oblique rifting: (down-dip rupture width of 17.3 km); however, events during the period of tectonic offset of Journal of Geophysical Research, v. 108, no. B2, 2074, doi:10.1029/2001JB000635, 17 p. greater seismogenic thicknesses will increase young lavas (i.e., constant slip-per-event and Clynne, M.A., and Muffl er, L.J.P., 2010, Geologic map of slightly the maximum credible earthquake. For earthquake recurrence), we provide a reason- Lassen Volcanic National Park and vicinity, California: example, an 18–20 km seismogenic thickness able methodology for evaluating seismic hazard U.S. Geological Survey Scientifi c Investigations Map 2899, 110 p., scale 1:50,000. would only increase the potential earthquake in faulted lavas. This technique is particularly Dawers, N.H., and Anders, M.H., 1995, Displacement-length scaling and fault linkage: Journal of Structural Geology, magnitude to Mw 6.8. Given the relatively low useful given the diffi culty of trenching analy- recurrence interval and the lack of historic earth- sis in lavas and the limitations of other tech- v. 17, p. 607–614, doi:10.1016/0191-8141(94)00091-D. Dawers, N.H., Anders, M.H., and Scholz, C.H., 1993, quake events or knowledge of the timing of the niques such as lichenometry and cosmogenic Growth of normal faults: displacement–length scaling: last event, the Hat Creek fault may thus provide nuclide analysis (which may be incapable of Geology, v. 21, p. 1107–1110, doi:10.1130/0091- 7613 (1993)021<1107:GONFDL>2.3.CO;2. a greater probabilistic seismic hazard than has distinguishing between events that are relatively Ferrill, D.A., Wyrick, D.Y., and Smart, K.J., 2011, Coseismic, been previously realized. closely spaced in time). dilational-fault and extension-fracture related pit chain

1408 Geosphere, October 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021 Earthquake hazard of the Hat Creek fault

formation in Iceland: Analog for pit chains on Mars: Dams, Central Valley Project, California: U.S. Bureau northern California: U.S. Geological Survey Open-File Lithosphere, v. 3, p. 133–142, doi:10.1130/L123.1. of Reclamation Seismotectonic Report, v. 86–1, 132 p. Report 96–705, 27 p. Friese, N., 2008, Brittle tectonics of the Thingvellir and Machette, M.N., Personius, S.F., Nelson, A.R., Schwartz, Walker, E.L., 2008, Evolution of the Hat Creek fault system, Hengill volcanic systems, Southwest Iceland: field D.P., and Lund, W.R., 1991, The Wasatch fault zone, northeastern California [M.S. thesis]: Moscow, Univer- studies and numerical modeling: Geodinamica Acta, Utah—segmentation and history of Holocene earth- sity of Idaho, 112 p. v. 21, p. 169–185, doi:10.3166/ga.21.169-185. quakes: Journal of Structural Geology, v. 13, p. 137– Walker, E.L., and Kattenhorn, S.A., 2008, Slip history and Grant, J.V., and Kattenhorn, S.A., 2004, Evolution of verti- 149, doi:10.1016/0191-8141(91)90062-N. evolution of the Hat Creek fault, northern California cal faults at an extensional plate boundary, southwest Martel, S., and Langley, J., 2006, Propagation of normal [abs.]: Eos (Transactions, American Geophysical Iceland: Journal of Structural Geology, v. 26, p. 537– faults to the surface in basalt, Koae fault system, Union), v. 89, no. 52, Abstract T21B–1976. 557, doi:10.1016/j.jsg.2003.07.003. Hawaii: Journal of Structural Geology, v. 28, p. 2123– Walsh, J.J., and Watterson, J., 1991, Geometric and kine- Gudmundsson, A., 1987a, Geometry, formation and develop- 2143, doi:10.1016/j.jsg.2005.12.004. matic coherence and scale effects in normal fault ment of tectonic fractures on the Reykjanes Penin sula, Muffl er, L.J.P., Clynne, M.A., and Champion, D.E., 1994, systems, in Roberts, A.M., et al., eds., The geometry southwest Iceland: Tectonophysics, v. 139, p. 295–308, Late Quaternary normal faulting of the Hat Creek Basalt, of normal faults: Geological Society of London Spe- doi:10.1016/0040-1951(87)90103-X. Northern California: Geological Society of America cial Publication 56, p. 193–203, doi:10.1144/GSL.SP Gudmundsson, A., 1987b, Tectonics of the Thingvellir fi ssure Bulletin, v. 106, p. 195–200, doi:10.1130/0016-7606 .1991.056.01.13. swarm, SW Iceland: Journal of Structural Geology, v. 9, (1994)106<0195:LQNFOT>2.3.CO;2. Wells, D.L., and Coppersmith, K.J., 1994, New empirical p. 61–69, doi:10.1016/0191-8141(87)90044-7. Muffl er, L.J.P., Blakely, R.J., and Clynne, M.A., 2008, Inter- relationships among magnitude, rupture length, rupture Gudmundsson, A., 1992, Formation and growth of normal action of the Walker Lane and the Cascade Volcanic Arc, width, rupture area, and surface displacement: Seismo- faults at the divergent plate boundary in Iceland: Terra Northern California: Eos (Transactions of the American logical Society of America Bulletin, v. 84, p. 974–1002. Nova, v. 4, p. 464–471, doi:10.1111/j.1365-3121.1992 Geophysical Union), v. 89, no. 53, abs. T23B–2032. Wesnousky, S.G., 2008, Displacement and geometric char- .tb00582.x. Muffl er, L.J.P., Champion, D.E., Calvert, A.T., and Clynne, acteristics of earthquake surface ruptures: Issues and Gudmundsson, A., Friese, N., Andrew, R., Philipp, S.L., Ertl, M.A., 2012, Paleomagnetic, geochronologic, and petro- implications for seismic-hazard analysis and the G., and Letourneur, L., 2009, Effects of dyke emplace- logic data discriminate tholeiitic basalts of the northern process of earthquake rupture: Seismological Soci- ment and plate pull on mechanical interaction between Hat Creek graben, northeastern California: American ety of America Bulletin, v. 98, p. 1609–1632, doi: volcanic systems and central volcanoes in Iceland, in Geophysical Union Fall Meeting, abs. V33B– 2868. 10.1785/0120070111. Thordarson, T., et al., eds., Studies in volcanology: The Peacock, D.C.P., and Parfi tt, E.A., 2002, Active relay ramps White, I.R., and Crider, J.G., 2006, Extensional fault-propa- legacy of George Walker: International Association of and normal fault propagation on Kilauea volcano, gation folds: Mechanical models and observations from Volcanology and Chemistry of the Earth’s Interior Spe- Hawaii: Journal of Structural Geology, v. 24, p. 729– the Modoc Plateau, northeastern California: Journal of cial Publication 2, p. 331–347. 742, doi:10.1016/S0191-8141(01)00109-2. Structural Geology, v. 28, p. 1352–1370, doi:10.1016 Huggins, P., Watterson, J., Walsh, J.J., and Childs, C., Rowland, J.V., Baker, E., Ebinger, C.J., Keir, D., Kidane, T., /j.jsg.2006.03.028. 1995, Relay zone geometry and displacement transfer Biggs, J., Hayward, N., and Wright, T.J., 2007, Fault Willemse, E.J.M., 1997, Segmented normal faults: Corre- between normal faults recorded in coal-mine plans: growth at a nascent slow-spreading ridge: 2005 Dab- spondence between three-dimensional mechanical Journal of Structural Geology, v. 17, p. 1741–1755, doi: bahu rifting episode, Afar: Geophysical Journal Inter- models and fi eld data: Journal of Geophysical Research, 10.1016/0191-8141(95)00071-K. national, v. 171, p. 1226–1246, doi:10.1111/j.1365 v. 102, p. 675–692, doi:10.1029/96JB01651. Jenness, M.H., and Clifton, A.E., 2009, Controls on the -246X.2007.03584.x. Willemse, E.J.M., Pollard, D.D., and Aydin, A., 1996, geometry of a Holocene crater row: A fi eld study from Sawyer, T.L., compiler, 1995, Fault number 9, Hat Creek Three-dimensional analyses of slip distributions on southwest Iceland: Bulletin of Volcanology, v. 71, fault zone, in Quaternary fault and fold database of normal fault arrays with consequences for fault scal- p. 715–728, doi:10.1007/s00445-009-0267-9. the United States: U.S. Geological Survey, http:// ing: Journal of Structural Geology, v. 18, p. 295–309, Kattenhorn, S.A., and Pollard, D.D., 2001, Integrating 3-D geohazards.usgs.gov/cfusion/qfault/. doi:10.1016/S0191-8141(96)80051-4. seismic data, fi eld analogs, and mechanical models in Turrin, B.D., Muffl er, L.J.P., Clynne, M.A., and Champion, Wills, C.J., 1991, Active faults north of Lassen Volcanic the analysis of segmented normal faults in the Wytch D.E., 2007, Robust 24±6 ka 40Ar/39Ar age of a low- National Park, Northern California: California Geol- Farm oil fi eld, southern England, United Kingdom: potassium tholeiitic basalt in the Lassen Region of NE ogy, v. 44, p. 51–58. The American Association of Petroleum Geologists California: Quaternary Research, v. 68, p. 96–110, doi: Zhang, P., Slemmons, D.B., and Mao, F., 1991, Geometric Bulletin, v. 85, p. 1183–1210. 10.1016/j.yqres.2007.02.004. pattern, rupture termination, and fault segmentation LaForge, R.C., and Hawkins, F.F., 1986, Seismotectonic study U.S. Geological Survey Working Group on Northern Cali- of the Dixie Valley-Pleasant Valley active normal fault of northernmost California for Shasta, Keswick, Spring fornia Earthquake Potential, 1996, Database of poten- system, Nevada, USA: Journal of Structural Geology, Creek Debris, Trinity, Lewiston, and Whiskeytown tial sources for earthquakes larger than magnitude 6 in v. 13, p. 165–176, doi:10.1016/0191-8141(91)90064-P.

Geosphere, October 2013 1409

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/5/1397/3345815/1397.pdf by guest on 24 September 2021