Auxiliary Material Submission for Paper: "Imaging the source region of the 2003 San 
Simeon earthquake within the weak Franciscan subduction complex, central California", 
[Paper # 2004GL021049]
Egill Hauksson1, David Oppenheimer2, and Thomas M. Brocher2
1 Division of Geological and Planetary Sciences, California Institute of Technology, 
Pasadena, California, 91125, USA. 
2 USGS, MS 977, Menlo Park, CA 94025, USA

2004GL021049-FigureS1. Seismicity and faults in the vicinity of the 2003 San Simeon earthquake for the period 
1990 to 21 December 2003.  The study region is defined by the black box in the insert (lower left 
corner) and the 4 by 4 km velocity inversion grid is shown as red filled circles.  The epicenter of 
the 2003 San Simeon mainshock is indicated by a red star.  Major faults from Jennings [1994] are 
shown in red (Holocene and Historic) green (Late Quaternary), and black (pre-Quaternary).  The 
receiver sites for the 1986 refraction experiment are shown as blue inverted triangles, and the 
corresponding sources are indicated by blue stars. The approximate location of the PG&E-1 line 
from Miller et al. [1992] is also shown.  SF Ð San Francisco, LA Ð Los Angeles.  

2004GL021049-FigureS2.  Maps of aftershocks comparing the results of the three relocation methods.  (a) The 
NCSN catalog hypocenters; (b) the relocated hypocenters from 3D velocity models; and (c) the 
relocated hypocenters determined with the double difference method. Major faults from Jennings 
[1994] are shown in red (Holocene and Historic) green (Late Quaternary), and black (pre-
Quaternary).

2004GL021049-FigureS3.  Cross sections of the derivative weighted sum (DWS) or ray density through the Vp 
model as shown in Figure 3.  The model is well resolved where the DWS values are high 
[Hauksson, 2000].  The model is not plotted in Figure 3 where it is poorly resolved along the 
edges, at shallow depth and offshore to the southwest, where DWS is less than 100. 

2004GL021049-FigureS4. Three depth sections through the Vp/Vs model with the relocated aftershocks.  The 
locations of the profiles are shown in Figure 1.  The model is not shown where it is poorly 
resolved with (DWS²5).  The red-white dashed contour represents the 0.01 values of the diagonal 
element of the resolution matrix, where 1.0 is prefect resolution and 0.0 is no resolution.  The 
plotted resolution value of 0.01 is apparently small because the Vp model has more data and 
biases the resolution to higher values. 


2004GL021049-FigureS5.  Cross sections of the derivative weighted sum (DWS) or ray density through the 
Vp/Vs model as shown in Figure S4.  The model is well resolved where the DWR values are high 
[Hauksson, 2000]. The Vp/Vs model is not as well resolved as the Vp model because the S picks 
are only 4.5 % of the total dataset.  Thus we use lower DWS values than for the Vp model to 
identify the resolved parts of the model. The model is poorly resolved along the edges, at shallow 
depth, offshore to the southwest, and at depths below the hypocenters.  We do not plot model 
regions with DWS value of 5 or less in Figure S4.  

		Hauksson, et al., p. 2