We fit a linear trend to the data, and sites closer to the rupture plane show a clear nonlinear postseismic trend. We fit a constant velocity to the data because they were sampled nonuniformly and infrequently. We fit the velocities to the 1.5 years following the earthquake because all of the data were collected within a month or two after the earthquake and in the summer of 1995, and the timeframe appears representative of the immediate postseismic process. For stations with a larger postseismic signal (e.g., LNCH) this assumption breaks down, however, at most of the stations the postseismic motion is small enough that it is difficult to distinguish between a linear and nonlinear function to within the errors. We estimated the postseismic velocity field by subtracting the contribution of shear from the San Andreas and other strike slip faults in southern California from the observed horizontal postseismic velocities. We assumed that the vertical velocities were near zero prior to the earthquake and used the vertical results directly from our solution.
Back || GPS Observations || Data || Behavior || Aftershocks || Inversions || SAR Observations || SAR Images of Northridge || Comparisons || Next