Although there is increasing evidence that fluids may play a significant ro
le in the earthquake rupture process, direct observation of fluids in activ
e fault zones remains difficult. Since the presence of an electrically cond
ucting fluid, such as saline pore water, strongly influences the overall co
nductivity of crustal rocks, electrical and electromagnetic methods offer g
reat potential for overcoming this difficulty. Here we present and compare
results from high-resolution magnetotelluric (MT) profiles across two segme
nts of the San Andreas Fault (SAF) which exhibit very different patterns of
seismicity: Parkfield, which has regular small earthquakes and creep event
s, and in the Carrizo Plain, where the fault is seismically quiescent and a
pparently locked. In both surveys, electric fields were sampled continuousl
y, with 100 m long dipoles laid end-to-end across the fault. From 100 to 0.
1 Hz the data from both profiles are consistent with a two-dimensional (2-D
) fault-parallel resistivity model. When both transverse electric and magne
tic (TE and TM) mode data are included in the interpretation, narrow (simil
ar to 300-600 m wide) zones of low resistivity extending to depths of 2-4 k
m in the core of the fault are required at both locations. However, at Park
field the conductance (conductivity thickness product) of the anomalous reg
ion is an order of magnitude larger than at Carrizo Plain, suggesting much
higher concentrations of fluids for the more seismically active Parkfield s
egment. We also image structural differences between the two segments. At C
arrizo Plain, resistive, presumably crystalline, rocks are present on both
sides of the fault at depths below 3-4 km. In particular, we clearly image
resistive basement extending similar to 10 km or more east of the SAF, bene
ath the Elkhorn Hills and Temblor Range. At Parkfield the situation is quit
e different with a resistive block of Salinian granite west of the fault an
d an electrically conductive, presumably fluid rich Franciscan complex to t
he east. It is possible that these structural differences control the diffe
rence in mechanical behavior of the fault, either directly by affecting fau
lt strength or indirectly by controlling fluid supply.