Variations in strike of the western end
When constructing the San Gorgonio block, the question of how to “close” up the block in a geologically sensible way had become a reoccurring issue. With help from the SCEC Community Fault Model, the San Gorgonio block boundary began by tracing the Garnet Hill fault and continuing westwards to trace the irregular “sawtooth shaped” San Gorgonio thrust fault. However, the San Gorgonio thrust fault ends just short of the San Jacinto block boundary and the issue arose as to which direction the new fault segments connecting the San Gorgonio thrust fault to the San Jacinto block would go.
Unlike the Banning fault, the Garnet Hill fault does not continue onwards into the San Gorgonio Pass and a NW orientation for the westernmost segment in the boundary does not have as much relevance in this as does the Banning models. However, it would be interesting to see what effect a NW orientation would have on the area. Both NW and SW models were tested and the results are shown below:
When comparing the different models, I looked at the residual magnitudes, the value we get when we take the difference of the modeled velocities from the observed velocities within that model, as well as the residual improvement between two different models. In the figures showing the residual magnitudes, the red dots represent bigger residual velocity vectors, hinting that the region in particular needs improvement, and the blue dots represent smaller residual vectors, meaning that the GPS velocities were well reproduced in that area.
The residual magnitudes of the NW vs SW adjustments for the Garnet Hill trace suggest that a SW orientation generates a better approximation of GPS velocities along the San Gorgonio thrust fault, especially concerning the GPS stations located in that particular area as well as the central part of the Pass, where the San Gorgonio thrust fault merges with the Garnet Hill strand. However, we can also see that while the residual magnitudes along the San Gorgonio thrust seem to improved with a SW orientation on the western end, the residual magnitudes in the surrounding area (in the San Bernardino Mountains and in the San Jacinto Mountains) seemed to have worsened. This can be shown in the figures below, depicting the residual improvement between the two models:
From these figures, we can deduce that a SW orientation on the westernmost fault segment does in fact reproduce GPS velocities along the study area better than the NW orientation, but additionally worsen the surrounding area. These results are very similar to that of the Banning Fault probably because the Banning fault and the Garnet Hill fault follow along a similar fault trace within a close proximity of each other. Because this study is focused on the San Gorgonio Pass, the SW model was further used for tests on dip and locking depth.