NearSource Observations of Earthquakes Implications for Earthquake Rupture
Near-Source Observations of Earthquakes: Implications for Earthquake Rupture and Fault Mechanics Bill Ellsworth U. S. Geological Survey and Kaz Imanishi Geological Survey of Japan A. I. S. T.
A central goal of seismology is to understand the physical conditions under which earthquakes occur Key question seismology can address in the near-source region include: • Minimum size of earthquakes • Magnitude dependence of source parameters • Rupture nucleation D r Illustrations by Peter Shearer
A central goal of seismology is to understand the physical conditions under which earthquakes occur Key question seismology can address in the near-source region include: • Minimum size of earthquakes • Magnitude dependence of source parameters • Rupture nucleation . u(t) σd ES Dc
Deep Geophysical Observatories in California San Andreas Fault Observatory at Depth (SAFOD) 2. 7 km deep Long Valley Exploratory Well (LVEW) 2. 7 km deep
Earthquakes at 300 m distance to seismometer in LVEW with magnitudes as small as Mw -2. 5 Seconds
May 2006 Multiplet at SAFOD This multiplet occurred at distance of about 600 m (S-P time is 0. 1 s). M 1. 12 M 1. 25 0. 1 s • High signal-to-noise ratio • High frequency energy is observed. Spectrogram of M 1. 12 event
Static Stress Drop Mw: 0. 38 and 0. 11 Sato & Hirasawa (1973) Eshelby (1957) Multi-Window Spectral Ratio Method (Imanishi & Ellsworth, 2006) • Stress drops range from approximately 1 to 100 MPa. • For any given cluster, the stress drops are nearly constant within a factor of 2 -3.
Aftershocks of M 1. 8 “Hawaii” Target (August 11, 2006) Spectral ratios relative to EV 1 Mw – 2. 7 Mw – 2. 5 Mw – 2. 6 Spectral ratios are almost constant. Mw – 2. 6 • Corner frequencies of these events are beyond the frequency band • Or all the events have the same corner frequency
Static Stress Drop Measurements at SAFOD Hickman & Zoback (2004) Average value is near the strength of the rock.
Stress Drop Scaling
Stress Drop Scaling
Stress Drop Scaling The existence of picoearthquake implies lab-like values for Dc
Apparent Stress Scaling (Radiated Energy / Seismic Moment) Ide and Beroza (2001) SAFOD Pilot Hole Apparent Stress Measurements (Imanishi and Ellsworth, 2006)
Dynamic Stress Drop Kostrov (1964) solution for a growing circular crack: s. D = r vp 3 r ü /(k vr 2 Vs)
Dynamic Stress Drop Kostrov (1964) solution for a growing circular crack: s. D = r vp 3 r ü /(k vr 2 Vs) 1. 1 MPa 3. 2 MPa 5. 0 MPa 2. 2 MPa Some events begin simply, while others have a nucleation phase.
August 11, 2006 M 1. 8 Hawaii Repeat Seismometer at 2. 65 km depth at a distance of 120 m
Dynamic Stress Drop in Initial Millisecond s. D = r vp 3 r ü /(k vr 2 Vs) Dynamic Stress Drop = 4 – 7 MPa
Critical Dimension for Instability in Rate & State Theory Laboratory h* = G Dc /(s-P 0)(b-a) August 11, 2006 “Hawaii” M 1. 8 at 0. 5 ms.
Conclusions • Stress Drop and Apparent Stress are scale-invariant for Mw>0. We do not as yet have measurements for smaller earthquakes. • The dynamic stress drops in the first 1 -2 ms (rupture dimension ~2 -4 m) are typically in the range from 1 – 10 MPa and are comparable to the event stress drops. • Earthquakes as small as Mw -3. 5 (picoearthquakes) occur along the San Andreas Fault at SAFOD and in Long Valley Caldera. If there is a minimum earthquake magnitude, we have not yet seen it. • Lab values of Dc are consistent with the occurrence of picoearthquakes at SAFOD. • The August 11, 2006 Mw 1. 8 “Hawaii” earthquake began without a Slow Initial Phase and has (b-a) ~ 0. 01.
- Slides: 19