Figure 1. NCSN earthquake locations for ~8000 earthquakes on the Calaveras fault near Morgan Hill, California. Three different perspectives of the seismicity are shown. Top panel shows epicenters in map view. The top panel is oriented such that the NW-SE trending surface trace of the Calaveras fault is horizontal. Middle panel shows hypocenters in the plane of the fault (a side view from the SW). Circles are scaled to show the rupture area assuming a circular, constant stress drop source of 3 MPa. Lower panels show a series of cross sections perpendicular to the fault. The numbers denote the distance along the strike of the fault.

Figure 2. Relocations of the same earthquakes shown in Figure 1. The same three perspectives are shown as in the previous figure. These relocations are from [Schaff, D.P, G.H.R. Bokelmann, G.C. Beroza, F. Waldhauser, and W.L. Ellsworth, High resolution image of Calaveras fault seismicity, J. Geophys. Res. , 107, 633, 2002]. They were obtained using a combination of cross-correlation derived arrival time measuremetns [Schaff, D.P., G.H.R. Bokelmann, W. L. Ellsworth, E. Zanzerkia, F. Waldhauser, and G. C. Beroza, Optimizing correlation techniques for improved earthquake location, Bull. Seismol. Soc. Am. , (submitted), 2002] and the double-difference earthquake relocation method [Waldhauser, F. and W.L. Ellsworth, A double-difference earthquake location algorithm: Method and application to the northern Hayward fault, Bull. Seismol. Soc. Am. , 90, 1353-1368, 2000]. Structure that was previously blurred is now brought into focus. The sets of northerly trending faults, which show up clearly in map view, appear to be conjugate to the Calaveras fault, but they are not since they exhibit right-lateral strike-slip mechanisms (as does the Calaveras). The seismicity to the SW of the fault shows up as an NE-dipping plane that verges into the vertical Calaveras fault near the base of the seismogenic zone. Earthquakes in this trend have reverse faulting mechanisms that agree with the orientation of the seismicity. The three-dimensional geometry of the compressional fault offset at 15-20 km along strike is also much clearer. The difference between the two sets of locations is most obvious in side view, where the widely distributed seismicity in the network locations is now highly organized with streaks, sets of repeating earthquakes (multiplets), and holes, where little seismicity occurs. The larger earthquakes have the largest uncertainties in our relocation procedure because their waveforms are strongly clipped and hence more difficult to cross-correlate precisely with other events. The seismicity distribution that we find on the Calaveras fault appears to be complementary to the slip distribution inferred for the 1984 Morgan Hill, California earthquake [Beroza, G. C., and P. Spudich, Linearized inversion for fault rupture behavior: application to the 1984 Morgan Hill, California, earthquakeJ. Geophys. Res., 93, 6275-6296, 1988]. In particular, the slip near the hypocenter at ~9 km depth falls below the micro-seismicity at about -5 km along strike. The strong second source found by imaging mainshock slip using strong motion data falls within a "hole" in microearthquake activity located at about +10 km along strike. This complementarity suggests that the seismicity we locate may delineate parts of the fault that are locked inter-seismically.

Figure 3. Compare the difference between the catalog locations and our relocations in figures 1 and 2 using this "blink movie."