By Mark Zoback, Stephen Hickman, William Ellsworth, and the safod science Team doi: 10. 2204/iodp sd. 11. 02. 2011

Scientific Drilling, No. 11, March 2011  

23

Science Reports

track was abandoned and cemented off after retrieving 

Core 1 due to a drilling mishap. A second sidetrack was 

undertaken that enabled us to obtain Cores 2 and 3 (Table 2) 

across the SDZ and CDZ. After obtaining the cores across 

the active shear zones, the hole was slightly enlarged to 

allow for installation of 18-cm-diameter casing and eventual 

deployment of the SAFOD observatory. The casing was 

installed and cemented to a measured depth of 3214 m (as 

measured in the Phase 3 hole), which is ~17 m beyond the 

center of the SDZ as extrapolated from the Phase 2 to the 

Phase 3 holes. The casing could not be installed to greater 

depth in the Phase 3 hole due to progressive borehole insta-

bility and bridging.

When the cores reached the surface, they were carefully 

cleaned, labeled, and photographed, and they have been 

stored at 4°C to prevent desiccation and microbial activity. 

The core is currently stored at the IODP Gulf Coast 

Repository (GCR) at Texas A&M University. High-resolution 

photographs and descriptions of all Phase 3 cores (as well as 

supplemental information including thin-section analysis, 

results from preliminary XRD analysis and core-log depth 

integration) are presented in a comprehensive Core Atlas 

(Table 1). One page of the core atlas is presented in Fig. 6, 

which shows a section of the core that crosses the SDZ. The 

foliated gouge matrix is highly altered, both chemically (e.g., 

there is much less silica and different clay mineralogy than 

observed in the rocks outside the fault zone) and mechani-

cally (e.g., there is pervasive shearing observed on planes of 

varied orientation within the core). Clasts of various types of 

rock are seen in the gouge matrix, most notably clasts of 

serpentinite including a large piece of sheared serpentinite 

with calcite veins. 

Zone (near or above the 

weight of the over- 

burden) has been one of 

 

the leading hypotheses to 

explain its low frictional 

strength (Rice, 1992). Two 

lines of evidence indicate 

 

an absence of severely 

elevated pore pressure 

(near-lithostatic, or greater) 

within the fault zone 

required to explain the low 

frictional strength of the 

San Andreas. Highly eleva-

ted fluid pressures were not 

observed during drilling in 

the fault zone. Such pres-

sures would have resulted 

in influxes of formation fluid 

into the wellbore if the pore 

pressure was appreciably 

greater than the drilling 

mud pressure. While the 

density of the drilling mud 

was about 40% greater than hydrostatic pore to stabilize the 

borehole, in the strike slip/reverse faulting stress state that 

characterizes the SAFOD area (Hickman and Zoback, 2004), 

pore pressures within the deforming fault zone would have 

to exceed the overburden stress in Rice’s model (1992) for a 

weak fault in an otherwise strong crust. In addition, analysis 

of the rates of formation gas inflow during periods of no dril-

ling (Wiersberg and Erzinger, submitted) shows no evi-

dence of elevated pore pressure within the fault zone relative 

to the country rock, and the Vp/Vs ratio is relatively uniform 

(~1.7) across the ~200-m-wide damage zone and the local-

ized shear zones within it (Fig. 4B). As Vp decreases seve-

rely at very elevated pore pressure (i.e., at very low effective 

stress), Vs would not be affected as much, and the Vp/Vs 

ratio would be expected to decrease (Mavko et al., 1998). 

Altogether, none of these observations indicate the presence 

of anomalously high pore pressure in the fault zone. 

Phase 3 – Coring the San Andreas Fault 

Zone

During Phase 3 the SAFOD engineering and science 

teams successfully exhumed 39.9 meters of 10-cm-diameter 

continuous core, including cores from the two actively de-

forming traces of San Andreas Fault Zone (the SDZ and 

CDZ; Zoback et al., 2010). Figure 5 shows the sidetracks 

drilled laterally off the SAFOD main borehole in map and 

cross-sectional views. Note the position of the cores with 

respect to the various contacts and shear zones described 

above. As shown, Core 1 was obtained close to the contact 

between the arkosic sandstones and conglomerates of the 

Salinian Terrane and the shales, mudstones and siltstones 

associated with the Great Valley Formation. The first side-

Figure 5.

 [A] Map view and [B] cross-section of the trajectory of the rotary-drilled SAFOD main borehole as it 

passed through the San Andreas Fault Zone at a depth of ~2700 m, as well as the trajectories of the sidetrack 

boreholes used to obtain core samples along the actively deforming traces of the fault during Phase 3. Note 

the positions of the SDZ, CDZ, and NBF and the extent of the damage zone as defined in Figs. 2, 3, and 4. 

Also shown are single-station locations of the aftershocks of the 11 August 2006 Hawaii target earthquake 

recurrence; these were made using a seismometer in the main hole at a true vertical depth of 2660 m.  The “C” 

and “D” symbols refer to the polarity of the P-wave from each aftershock. Because the borehole seismometer 

is offset to the northeast from the fault trace, the transition from “C” to “D” occurs where expected for right 

lateral slip on the fault.

650

700

750

800

850

950

1000

1100

Easting (m)

Northing (m)

C

C

C

C

D

D

C

D

C

C

C

CC

C

C

C

C

C

C

C

C

C

1150

1250

1350

−2800

−2700

−2600

Cross-Section 

North 45 East (m)

True Vertical Depth (m)

C

C

C

C

D

D

C

D

C

C

C

CC

C

C

C

C

C

C

C C

C

3192 SDZ

SALINIAN

TERRANE

SALINIAN

TERRANE

GREAT

VALLEY

FORMATION

Damage Zone

1050

GREAT

VALLEY

FORMATION

1200

3192 

 SDZ

3302 

 CDZ

3413 

Core 1

Core 1

Core 2

Core 3

Core 3

SAFOD Main Borhole 

SAFOD Main Borhole 

Hawaii

Aftershocks

Hawaii 

Aftershocks

SAFOD

  Observatory

Core 2

Damage Zone

SAFOD

3302 CDZ

3413 NBF 

NBF

  

Observatory

A

B

---