Return to Homepage |
Research
| Research Group | Teaching | Demian's Departmental Page | G3 Center
| Geosciences
Department
SAN ANDREAS FAULT
STUDIES: FLUIDS, STRENGTH, AND HEAT TRANSPORT
Heat Flow Studies | Pore Pressure and Fault Weakness | Lab Measurements | Papers
ADVECTIVE
HEAT TRANSPORT & THE STRENGTH OF THE SAN ANDREAS FAULT
|
Along the San Andreas Fault in
California, heat flow observations and inferred stress orientations have been
interpreted to indicate that the fault is weak. Specifically, heat flow data
along much of the San Andreas Fault's length show no indication of a
fault-centered heat-flow high, as would be expected from frictional heating
as the fault moves (red curves in the two plots shown below). Stress
orientations, inferred from geological structures, earthquake focal
mechanisms, and borehole breakouts, have also been interpreted to indicate
that the San Andreas is weak in comparison to the surrounding crust. Taken together, the heat flow and stress
observations have led to the conclusion that the San Andreas is a "weak
fault", in that it slips under considerably smaller shear stresses than
predicted by laboratory experiments on a wide range of rocks. This
discrepancy has become known as the San Andreas “stress - heat flow
paradox”. Critics of the weak fault hypothesis
have suggested that the fault actually does generate frictional heat,
but that groundwater flow, driven by topographic relief, redistributes and
effectively “hides” the heat. The idea of heat
redistribution by topographically-driven groundwater flow is attractive
because there is considerable topography associated with the fault itself.
The two scenarios depicted below constitute end-member hypotheses that could
explain observed heat flow (the pink curves). On the left-hand-side, a strong
fault could generate frictional heat, which is then "smeared out"
by groundwater flow. On the right-hand-side, a weak fault would generate only
a small amount of heat. On the left, a strong fault generates heat, which is
carried away by groundwater flowing from high elevation recharge areas to low
elevation discharge areas. On the right, low strength (possibly caused by a
combination of high pore pressures and weak fault gouge) results in a small
amount of heating. |
|
|
|
Our work has shown that topographically driven groundwater flow is
an unlikely explanation for the lack of an observed heat flow anomaly.
Numerical models of coupled fluid flow and heat transport predict decreased
heat flow in areas of groundwater recharge (topographic highs), and elevated
heat flow in areas of groundwater discharge (topographic lows); yet this is
not observed (below, left). Our
results show that the data are most consistent with a “weak
fault”, even in the presence of substantial advective
heat redistribution. Furthermore, the degree of scatter in observed heat flow
is inconsistent with substantial advective heat
transport (below, right). Research on this topic is ongoing,
and we have been funded by NSF to study the problem in more detail.
Currently, graduate students Patrick
Fulton and Maggie
Popek are working with me to understand sources of
scatter in the heat flow data near Parkfield, in order
better constrain the signals from hydrologic processes and frictional
heating. |
|
|
|
PORE PRESSURE AND THE STRENGTH OF THE
SAN ANDREAS FAULT
|
If the San Andreas Fault is
indeed weak, elevated fluid pressures are one likely explanation.
Hypothesized mechanisms for generating elevated fluid pressures along the San
Andreas include dehydration of the Franciscan assemblage (right) and
upward flux of mantle fluids. We are rigorously testing these hypotheses,
using numerical models of fluid flow that incorporate well-constrained and
realistic ranges for the timing of fluid release, and consider a wide range
of crustal permeability architectures. |
after Irwin & Barnes, 1975 |
|
after Fulton
et al., 2008
|
Dehydration
of the Franciscan assemblage generates fluids for only a limited time,
between 0-5 Myr after formation of the San Andreas
fault system (left). The figure below shows a transect parallel to the
fault based on the “crustal conveyor”
geodynamic model of Furlong & Guzofski (2002). The fluid sources are too small and
too short-lived to generate or sustain fluid pressures high enough needed to
explain the apparent weakness of the San Andreas (below).
after Fulton
et al., 2008 Recent
data from the SAFOD borehole suggest that the fault acts as a hydrologic
barrier to depths of ~3 km (left).
These observations include moderate fluid overpressures documented by mud
weights, and a mantle helium signature in mud gases, both seen only to the NE
of the fault (schematic showing these observations at top, geophysical logs at bottom).
Some of our ongoing work is focused on evaluating the potential role of
mantle fluids in generating excess pore pressures either throughout the crust
or localized along the fault in a general sense. We are also conducting
numerical modeling studies to assess the role of mantle fluids in producing
the detailed observations at SAFOD. |
LABORATORY STUDIES OF SAN ANDREAS MATERIALS: PERMEABILITY & STRENGTH
|
|
|
|
Above right: Map showing field sample
locations for outcrop samples, taken to represent the main lithologies in the vicinity of the San Andreas Fault.
Many of these lithologies were not penetrated by
the SAFOD borehole. Above Left:
Detailed cross section near the SAFOD borehole, showing interpreted rock
units in the subsurface (courtesy of R. Arrowsmith
and M. Thayer, Ariz. State Univ.). Below:
Field sampling of lower Great Valley sedimentary rocks on the NE side of the
fault.
|
We are
currently working on a collaborative project to measure the frictional,
permeability, and elastic properties of both the active San Andreas Fault
(SAF) zone sampled during SAFOD Phase 3 drilling, and outcrop samples of lithologies present within the 3-D crustal
volume containing the fault. Our laboratory measurements will help to
constrain: (1) the strength, sliding stability, and healing of major faults,
and (2) the hydraulic behavior of faults - both locally as related to
long-term and dynamic weakening mechanisms and regionally as elements within crustal scale fluid flow systems. The friction and
permeability measurements are being carried out in the Penn State rock and
sediment mechanics laboratory; elastic property measurements are being
conducted by colleague Harold Tobin at the University of
Wisconsin–Madison. Ultimately, this ongoing research will provide
experimental constraints on rock properties for the San Andreas Fault system
that relate to its mechanical strength, sliding stability, potential role as
a barrier to regional fluid flow, and the interpretation of in situ
conditions from geophysical data. |
Return to Homepage | Research
| Research Group | Teaching | Demian's Departmental Page | G3 Center
| Geosciences
Department