HR: 1330h
AN: S42C-0178    [PDF]
TI: Source Parameters of Reservoir-Induced Small Earthquakes: Can We Resolve Rupture Velocity?
AU: * Tomic, J
EM: yelenat@bu.edu
AF: Boston University Dept. of Earth Sciences, 685 Comm. Ave, Boston, MA 02215 United States
AU: Abercrombie, R E
EM: rea@bu.edu
AF: Boston University Dept. of Earth Sciences, 685 Comm. Ave, Boston, MA 02215 United States
AU: do Nascimento, A F
EM: aderson@dfte.ufrn.br
AF: Universidade Federal do Rio Grande do Norte, Departmento de Fisica Teorica e Experimental, Natal, RN 1596 Brazil
AB: Abercrombie and Leary (1993) noted that the average stress drops of induced earthquakes appear to be lower than those of tectonic earthquakes. This difference may result from different tectonic setting or the shallow depth of induced earthquakes. Alternatively, it may reflect uncertainties resulting from single station recording, and assumption of rupture velocity and source model. We analyze earthquakes induced by the A\c{c}u reservoir, NE Brazil for comparison with tectonic and mining earthquakes. Do Nascimento (2002) located 286 earthquakes (M$\leq$2.8) recorded between 1994 and 1997 on a closely spaced network of 8 3-component seismometers. The locations and focal mechanisms are consistent with activity on NE trending right-lateral faults. The short source-to-receiver distances ($<$5km) and shallow hypocentral depths ($<$3km) of the earthquakes, and the low attenuation of the Precambrian basement rock allowed energy at frequencies up to the Nyquist (100 and 250 Hz) to be well recorded at the surface. Starting with the largest earthquakes we search for groups of collocated events. We use 3 different approaches to calculate the source parameters of the largest 2 earthquakes in the first cluster ($M$2.2 and $M$1.8), at 4 stations with good azimuthal distribution. We fit the individual spectra using an $\omega^{-2}$ source model to find corner frequency ($f_c$), frequency-independent $Q$, and long period amplitude. We use the small earthquakes as empirical Green's functions (EGF) and calculate the spectral ratios of the largest earthquake to the smaller events in the cluster. Using the $\omega^{-2}$ model we fit the ratios solving for the $f_c$ of the largest earthquake. We then determine the relative source time functions using the EGF method. We will also measure radiated energy. The various estimates of source duration and $f_c$ imply stress drops in the range of 1 to 10 MPa. These are similar to tectonic earthquakes suggesting hypocentral depth and strain rate do not affect stress drop. The source-time functions vary systematically with azimuth and imply rupture toward the NE, with rupture velocity consistent with that of large tectonic earthquakes.
DE: 7209 Earthquake dynamics and mechanics
DE: 7215 Earthquake parameters
SC: Seismology [S]
MN: 2003 Fall Meeting