1

Testing the shorter and variable recurrence interval hypothesis along the

Cholame segment of the San Andreas Fault

2018 Annual Report

Researchers

J Ramón Arrowsmith (Principal Investigator)

&

Alana Williams, Ph.D. pre-Candidate

School of Earth and Space Exploration

Arizona State University

Submitted Collaboratively with

Sinan Akciz, UCLA

Lisa Grant Ludwig, UCI

Tom Rockwell, SDSU

Disciplinary Committee

Earthquake Geology Interdisciplinary

Science Objectives

2a (Improvement of earthquake catalogs and slip-per-event investigation)

1a (Determine slip rates at multiple time scales)

1d (Fault slip rates and deformation patterns to constrain fault slip models)

2

Introduction

The Cholame segment of the southern San Andreas fault (SAF) accommodates the interaction

between the creeping Parkfield segment to the northwest and the locked Carrizo segment to the southeast

(Figure 1). This ~75 km long reach is known to have ruptured during the M7.8 1857 Fort Tejon

earthquake (e.g., Sieh, 1978b), however a complete record of rupture behavior has not been well-

characterized. Classical models of recurrence for this section of the SAF expected repeats of the 1857

event with smaller superposed M7 Cholame-only events (Sieh and Jahns, 1984). The Uniform California

Earthquake Rupture Forecast 3 estimates among the highest system-wide rates for rupture probability for

the Cholame segment (Field, et al., 2015), despite limited paleoseismic data (e.g., Field, et al., 2014). In

addition, the Cholame segment plays a critical role in potential earthquake nucleation after Parkfield

earthquakes (e.g., Sieh, 1978; Toké and Arrowsmith, 2006; Field, et al., 2017).

Results from our team and others over the past decade suggest that event magnitude and

distribution is variable from event to event along the south-central SAF with both smaller events

producing small slip as well as larger full ruptures along the Cholame and adjacent fault segments (e.g.,

Akciz, et al., 2010; Grant, et al., 2010; Zielke, et al., 2010, 2012, 2015; Salisbury, et al. 2018; Scharer, et

al., 2014; Figure 2). However, slip rate appears relatively steady at about 30-35 mm/yr along strike and

over decadal to millennial timescales (Sieh and Jahns, 1984; Toké, et al. 2011; Schmalze, et al., 2006;

Salisbury, et al., 2018 and Salisbury, 2017).

The large gaps between paleoseismic sites along this portion of the SAF limit our ability to

reconstruct the past events. Given its central position, lack of recurrence information, and uncertain

paleoslip reconstruction, seismic hazard evaluations are limited. The length of the segment and large slip

deficit indicate a potential M7 rupture along the full length of the Cholame section. Cholame paleoseismic

site success has been limited in past studies due to agricultural land use and anthropogenic disturbance, as

well as the lack of chronologic control and dateable material.

Results

Cholame has been the focus of extensive offset research (Sieh, 1978a; Lienkaemper, 2001;

Zielke, et al., 2010, 2012) and a few paleoseismic studies (Stone, 2001; Young et al., 2002), showing an

offset of 3 m in Cholame during the 1857 Fort Tejon earthquake. Assuming that the Cholame segment

slips at the average SAF long term slip rate (~33mm/yr), and the uniform slip of Cholame in 1857, there

is a large slip deficit of ~5m that could accommodate a M7 full length rupture without extending into

Carrizo. Young’s measurement of ~3 m is smaller than the offsets recorded in Carrizo for 1857 (4-6 m;

Zielke, et al., 2012; Salisbury, et al., 2018), and considerably larger than Parkfield (<1 m), which suggests

a change in fault behavior and properties (Sieh, 1978b; Hilley et al, 2001). Young and Stone interpreted

2-4 ground rupturing events and 1 ground shaking event post-1857 in Cholame at the LY4 site, which

suggested a much longer recurrence interval than previously suggested (Sieh and Jahns, 1984). However,

both LY4 sites have poorly constrained geochronology.

Despite the challenges we had encountered at LY4 in the early 2000s, we continue to appreciate

the need for a good paleoseismic record for the Cholame segment. Therefore, we developed a new site

starting in 2016 and continuing in 2018 with support from the Southern California Earthquake Center. The

Annette excavation site is ~36 km north of LY4 (Stone, et al., 2002 and Young et al., 2002; Figure 3; Table

1). Fault zone stratigraphy consists of alternating finely bedded sand, silt, and gravel strata, and bioturbated

soil horizons. In 2016, we excavated two fault-perpendicular trenches and one fault-parallel trench. The

strata record evidence for up to 6 earthquakes, of which 4 are of high quality. There is also possible post-

1857 deformation, similar to that recognized at the LY4 and Bidart sites (Young, et al., 2002; Akciz, et al.,

2010), which may represent a strong ground-shaking event. In 2018, we excavated two fault-perpendicular

trenches several meters southwest of the 2016 trenches, in an attempt to bolster our earlier evidence. The

strata recorded evidence for five ground rupturing events (four of good quality) that could be correlated to

the six events seen previously. We infer that E1 in all four fault-perpendicular trenches was the 1857

earthquake. It is represented by several fault splays capped by silty clay sag deposits that fill a depression

in the fault zone that would have been produced by the 1857 surface rupture.

3

We have evaluated event confidence by compiling event indicators for the seven possible events

and assigning quality rankings for specific tectonic and geomorphic evidence of ground deformation in the

style of Scharer et al. 2007 (Table 1). The quality of an event horizon is interpreted by the quality and

frequency of the event indicators, which allows for consistent qualitative interpretation of event horizon

across paleoseismic sites.

Figure 1. A) The Cholame segment is part of the oldest and most highly localized part of the San Andreas

fault system. The UCERF3 forecasts among the highest system-wide rates for rupture probability for the

Cholame segment (Field, et al., 2015), despite limited paleoseismic data (e.g., Field, et al., 2014). B) The

Cholame segment’s adjacency to Parkfield reminds us of both the 1857 foreshock precedent (Sieh, 1978a)

and the likelihood of failure following a future Parkfield earthquake (modified from Field, et al., 2017).

4

Figure 2. Earthquake chronology for the south-central San Andreas Fault (modified from Scharer, et al.,

2014). Colored curves at each site show the probability distributions for earthquake ages with the colored

bars showing the 2σ range. The “0” PDF at Annette shows the ages range for a possible post 1857 ground

deformation event (also noted at LY4 by Young, et al., 2002). The composite age model pushed the event

“1” PDF amplitude low at the obvious most recent earthquake level which we infer to be 1857 (“1 likely”).

Red lines show possible correlated events and their possible extent. Building on the Scharer, et al., 2014

correlations, these show that both large and long 1857-type events as well as shorter lower magnitude events

are needed to explain the event timing.

The Annette site contains abundant detrital charcoal in many of the units and burn horizons at or

near event horizons that provide great potential for bracketing the ages of these paleoearthquakes. We have

sampled many of these layers and fragments. We have processed an initial 25 14C and will process another

35 at the UC Irvine Keck Radiocarbon Facility in September 2018 (with SCEC support).

Discussion

Our preliminary interpretations from the two field seasons at Annette indicate that the timing of the

six ground rupturing events correlate in general with timing of events at the Bidart Fan and Frasier Mountain

sites but extend slightly older than those sites (Figure 2). We assume that event 1 at Annette is 1857 (Figure

2; Table 1) given the widespread evidence of rupture at that level. The Annette age model (from OxCal)

sequence compared with the Bidart data (Akciz, et al., 2010) and the Scharer et al. 2014 reconstructions

suggests that potentially 4 of the 6 events at Annette ruptured with the Carrizo segment, E1, E2, E3, and

E5. Of the 6 ruptures at Annette, the four with the highest quality were E1, E4, E5, E6. This is consistent

in general with the Scharer, et al. reconstructions (Figure 2) for longer events, except for event 4. We are

confident that event 4 was a large rupture based on the tectonic and geomorphic evidence (Table 1), despite

5

the lack of correlation in Figure 2, which could be explained by a rupture that extended north into Parkfield

without reaching the Carrizo or alternatively may not have been preserved at Bidart. If all 6 events are

correct, then the longer recurrence interval suggested by the LY4 site would be incorrect, however if we

only take into account the 4 high quality events the interpretation remains similar (Stone, 2001; Young et

al., 2002). These initial interpretations are from the 2016 excavations only. We will integrate the 2018

excavation dates and event evidence with our preliminary results to strengthen geochronology of the site.

The timing of ruptures at the Annette site suggests that both long, 1857-type and shorter events are required

to explain the observations. The ranking and summation of event evidence ensures a quantitative approach

to evaluating and comparing ground ruptures between paleoseismic sites.

Figure 3. Previous work location in the Cholame segment from 2016 and 2018 excavation is shown with

logging examples. A) B4 lidar of Annette Site with the fault trace mapped in red, box inset of B. B)

Structure from Motion model of the Annette site with the fault zone shaded. Trenches T2A and T2C are

visible from 2016 excavation, T2D and T2E outlines from 2018 are shown in red. C) Partial log of T2C

NW wall displaying the key units and the event horizons (red stars) E0 through E6. D) Inset box showing

the detailed hand logging with color for reference. See Table 1 for list of event dates, event horizon

evidence, and rankings.

6

EQ ID Event

Rating

Location Trench Evidence

E0

1740-1968

AD

16

16

T2C-NW-P9

T2C-NW-

P6

Liquefaction features in Unit 110 (4) across four panels

E1

1857 AD

45

6

8

13

2

3

10

3

T2A-NW-P4

T2A-NW-P5

T2A-NW-P6

T2C-NW-P4

T2C-NW-P5

T2C-NW-P7

T2C-NW-P9

Truncated units (2), fracturing (1), U110 and U100 sag across fault (3)

Vertical separation of units (4), fissure beneath U110 (4)

Decreasing vertical separation upsection (2), fissure below U110 (4), disrupted

U115 (3), shear fabric (4)

Truncated silt units (2)

Shear fabric along truncated layers (3)

Aligned clasts (2), disrupted grass burn layer beneath U110 (3), truncated silt

units (2), vertical separation decreasing upsection (3)

Silt units display decreasing vertical throw upsection (3)

E2

1436-1596

AD

12

2

3

7

T2A-NW-P4

T2C-NW-P5

T2C-NW-P5

Decreased vertical separation between U150 and U120 (2)

Decreasing vertical separation of units capped by conformable grass burn layer

beneath U115 (3)

Block from U130 draped across fault (4), fracture capped by grassy burn layer

and continuous gravels (3)

E3

1270-1383

AD

6

3

3

T2C-NW-P5

T2C-NW-P5

Fault strand capped by U150 (3)

Silty fissure fill beneath U150 truncated silts (3)

E4

1119-1280

AD

34

12

8

6

8

T2A-NW-P4

T2A-NW-P6

T2C-NW-P5

T2C-NW-P9

U190 sag deposit thickens across fault (4), U270 offset 47 cm (2), color change

in soil U200 (4)

U190 sag deposit draped across fault (4), U200 soil thickens across fault (4)

Disrupted and tilted blocks draped across fault stratigraphically equivalent with

U190 (4), apparent diagonal offset decreasing upsection (2)

Draped gravels in U250 vertically separated 44 cm (4), truncated silt units (2),

onlapping sediments offset ~35 cm (4)

E5

1056-1261

AD

15

5

10

T2A-NW-P4

T2C-NW-P9

Shear fabric (1), U270 vertical separation of ~50 cm is 2x greater than U190

separation (4)

Draped gravels in U250 thickens across fault (4), truncated silt units (2),

onlapping sediments offset ~35 cm (4)

E6

738-1159 AD

19

11

8

T2A-NW-P4

T2C-NW-P9

Truncated and rotated gravels beneath draped sag deposit U270 (4), Unit 270

thickens across fault (4), soil color change across fault (3)

Rotated U290 block in fault zone capped by horizontal silt and sand layers (4),

onlapping sands and silts deposited against paleoscarp (4)

Table 1. Event Evidence for the northwest walls of T2A and T2C excavated in 2016, with the Oxcal dates

for each event presented in the first column, and the summed rankings for each panel in the second

column. The bold ranking is the total overall for each event on two of four trench walls. The third column

shows the panel where the evidence is located, panels are roughly 2 meters wide. Fourth column lists all

evidence that indicates a single event with a specific ground surface, like the locations of the red stars in

Figure 2C, and the respective numerical rankings of the quality of that evidence, 1 being the least

convincing and 4 the most convincing.

7

References

Akciz, S. O., Grant Ludwig, L., Arrowsmith, J R., Zielke, O., (2010). Century-long average time intervals

between earthquake ruptures of the San Andreas fault in the Carrizo Plain, California, Geology, 38, 9,

p. 787-790.

Field, E. H., Jordan, T. J., Page, M. T., Milner, K. R., Shaw, B. E., Dawson, T. E., Biasi, G. P., Parsons,

T., Hardebeck, J. L., Michael, A. J., WeldonII, R. J., Powers, P. M., Johnson, K. M., Zeng, Y., Bird,

P., Felzer, K. R., van der Elst, N., Madden, C., Arrowsmith, J R., Werner, M. J., Thatcher, W.R.,

(2017). A New Earthquake Forecasting Model for California, Seismological Research Letters,

Volume 88, Number 5, doi: 10.1785/0220170045, 2017.

Field, E. H., Arrowsmith, J R., Biasi, G. P., Bird, P., Dawson, T. E., Felzer, K. R., Jackson, D. D.,

Johnson, K. M., Jordan, T. H., Madden, C., Michael, A. J., Milner, K. R., Page, M. T., (2015).

Uniform California Earthquake Rupture Forecast, Version 3 (UCERF3)--The Time-Independent

Model, Bulletin of the Seismological Society of America, Vol. 104, No. 3, pp. 1122–1180, doi:

10.1785/0120130164, 2014.

Hilley, G. E., Arrowsmith, J R., and Stone, E. M., Using microseismicity and surface offset data to define

fault segment boundaries along low friction faults, with an example from the Cholame-Carrizo

segment boundary along the San Andreas Fault, Southern California, Bulletin of the Seismological

Society of America, 91, 427-440, 2001.

Lienkaemper, J. (2001). 1857 Slip on the San Andreas Fault Southeast of Cholame, California. Bulletin of

the Seismological Society of America, 1659-1672.

Scharer, K., Weldon, R., Biasi, G., Streig, A., & Fumal, T. (2017). Ground‐rupturing earthquakes on the

northern Big Bend of the San Andreas Fault, California, 800 AD to Present. Journal of Geophysical

Research: Solid Earth, 122(3), 2193-2218.

Scharer, K., Weldon, R., Streig, A., & Fumal, T. (2014). Paleoearthquakes at Frazier Mountain,

California delimit extent and frequency of past San Andreas Fault ruptures along 1857 trace.

Geophys. Res. Lett. Geophysical Research Letters.

Scharer, K. M., Weldon, R. J., Fumal, T. E., & Biasi, G. P. (2007). Paleoearthquakes on the southern San

Andreas fault, Wrightwood, California, 3000 to 1500 BC: A new method for evaluating paleoseismic

evidence and earthquake horizons. Bulletin of the Seismological Society of America, 97(4), 1054-

1093.

Schmalzle, G., T. Dixon, R. Malservisi, and R. Govers (2006). Strain accumulation across the Carrizo

segment of the San Andreas fault, California: Impact of laterally varying crustal properties, J.

Geophys. Res. 111, no. B5, doi: 10.1029/2005JB003843.

Sieh, K. E. (1978a). Slip along the San Andreas fault associated with the great 1857 earthquake, Bulletin

of the Seismological Society of America, 68, 1421–1448.

Sieh, K. E. (1978b). Central California foreshocks of the great 1857 earthquake. Bulletin of the

Seismological Society of America, 68(6), 1731-1749.

Sieh, K., & Jahns, R. (1984). Holocene activity of the San Andreas fault at Wallace Creek, California.

Geol Soc America Bull Geological Society of America Bulletin, 883-883.

Stone, E.M., Grant, L., and Arrowsmith, J R., Recent rupture history of the San Andreas Fault southeast

of Cholame in the northern Carrizo Plain, California, Bulletin of the Seismological Society of

America, 92, 983-997, 2002.

Toké, N., & Arrowsmith, J. (2006). Reassessment of a Slip Budget along the Parkfield Segment of the

San Andreas Fault. Bulletin of the Seismological Society of America, S339-S348.

Williams, A., (2015). Exploration of a new paleoseismic site in the Cholame section of the San Andreas

fault, CA. Poster presented at the Annual Southern California Earthquake Center Conference, Palm

Springs, CA.

8

Young, J., Arrowsmith, J., Collini, L., Grant, L., & Gootee, B. (2002). Three-Dimensional Excavation

and Recent Rupture History along the Cholame Segment of the San Andreas Fault. Bulletin of the

Seismological Society of America, 2670-2688.

Zielke, O., Arrowsmith, J., Ludwig, L., & Akciz, S. (2010). Slip in the 1857 and Earlier Large

Earthquakes Along the Carrizo Plain, San Andreas Fault. Science, 1119-1122.

Zielke, O., Arrowsmith, J., Ludwig, L., & Akciz, S. (2012). High-Resolution Topography-Derived

Offsets along the 1857 Fort Tejon Earthquake Rupture Trace, San Andreas Fault. Bulletin of the

Seismological Society of America, 1135-1154.