Yesterday scaling laws. An important one for seismic hazard analysis

Transcription

1 Yesterday scaling laws An important one for seismic hazard analysis Estimating the expected size of an earthquake expected on a fault for which the mapped length is known. Wells & Coppersmith, 1994 And there are others

2 Complication our survey of earthquakes showed that many if not most earthquakes don t break the entirety of the length of fault on which they occur. Commonly earthquake ruptures are observed to have ends at complexities along the fault zone, which may be characterized as steps in fault trace or bends in fault trace or bifurcations in fault trace, etc

3 There are reasons to consider why this might be so primarily the result of volumetric considerations (which lead to stress concentrations) at discontinuities.

4 Geometrical controls on earthquake rupture propagation. And historical earthquake ruptures mapped by geologists are our tool to investigate

5 Dec 16, 1954 Dixie Valley Earthquake

6 Dec 16, 1954 Dixie Valley Earthquake

7 Oct 16, 1999 Hector Mine

8 Oct 16, 1999 Hector Mine

9 The displacements in historical earthquakes can be measured - in the field as well as with modern imaging methods Oct 16, 1999 Hector Mine

10 Jun 22, 1992 Landers

11 June 28, 1992 Landers

12 Nov 23, 1987 Superstition Hills

13 Nov 23, 1987 Superstition Hills

14 Apr 08, 1968 Borrego Mtn

15 Traces of Active Earthquake Faults Mapped by Geologists Larger Historical Earthquakes of California 15

16 Cumulative Geologic Offset versus Fault Geometry - Fault Complexity can be measured by number of steps per unit fault length

17 Cumulative Geologic Offset versus Fault Geometry - Fault Complexity can be measured by number of steps per unit fault length

18 Anatolian Fault Turkey 18

19 Structural Evolution of Faults - become smoothe 19

20 Can their be a Seismological Evolution as well? same analysis for global selection of faults.

21 Focus on steps in fault trace

22 NORMAL (7) REVERSE (8) 1887 Sonora, MX 1915 Pleasant Valley, NV 1954 Fairview Peak, NV 1954 Dixie Valley, NV 1959 Hebgen Lake, MT 1983 Borah Peak, ID 1987 Edgecumbe, NZ 1896 Rikuu Japan 1945 Mikawa, Japan 1971 San Fernando 1979 Cadoux, Australia 1980 El Asnam 1986 Maryat, Australia 1998 Tenant Creek, Australia 1999 Chi-Chi, Taiwan About 3 dozen for which maps and surface slip both documented STRIKE-SLIP (23) 1857 San Andreas, CA 1906 San Andreas, CA 1891 Neo-Dani, Japan 1930 Kita-Izu, Japan 1939 Ercincan, Turkey 1940 Imperial Valley, CA 1942 Erbaa-Niksar, Turkey 1943 Tosya, Turkey 1943 Tottori, Japan 1944 Gerede-Bolu, Turkey 1967 Mudurnu, Turkey 1968 Borrego Mtn, CA 1979 Imperial Valley, CA 1981 Sirch, Iran 1987 Superstition Hills, CA 1990 Luzon, Philippines 1992 Landers, CA 1999 Fandoqa, Iran 1999 Hector Mine, CA 1999 Izmit, Turkey 1999 Duzce, Turkey 2001 Kunlun, China 2002 Denali, AK

23 With this data set we can begin to examine the slip characteristics

24 Measured and Published Digitized and Interpolated

25 Seismic Moment = Mo = µ x W x L x D

26 Geologic Mo Vs Instrumental Mo Vertical error bar is factor of 3 of value Horizontal error bar is range of estimates of moment determined instrumentally Slope of 1 would be perfect agreement Good news: Geologic and Instrumental Measures of Mo generally agree for the la

27 Shape of Earthquake Slip Distributions 5 curve fits Flat line - Sine function Ellipse function (average) (amplitude) (amplitude) Asymetric Sine (amplitude and slope m) sine function * (1-m*x/L) Asymetric Ellipse (amplitude and slope m) ellipse function * (1-m*x/L) Triangle (slopes of two lines)

28 Black is surface slip - colored are curve-fits to observed distributions

29 Coefficient of Variation Standard deviation of particular curve-fit to observed slip and then divided by average value of slip Triangle and Asymmetric Curve forms consistently yield better fits to observation - no surprise there Lowest values always solid - > asymmetric fits

30 Just how asymmetric is asymmetric But asymmetry depends of how curve fit

31 No systematc relationship between epicenter/rupture initiation to peak of surface slip. Open symbols are epicenter Closed symbols are point of max slip

32 Triangle Asym-Ellipse Asym-Sine Ellipse Sine Curve-fits of maximum amplitude to rupture length For all, best fit for strike-slip is not linear ---> thus slip forms not self-similar A

33 Turn your attention to fault geometry

34 STRIKE-SLIP About 2/3-3/4 endpoints are associated with steps in fault trace or termination of active fault on which they occurred.

35 Is termination of STRIKE-SLIP rupture associated with step in fault trace of dimension >=1km or end of active fault trace?

36 Of those endpoints associated with discontinuities - a transition exists between step dimensions of 4 to 5 km above which rupture fronts have not been observed to propagate through, and that ruptures appear to cease propagating at steps of lesser dimension only about 40% of the time.

37 AND THAT THE ULTIMATE LENGTH OF AN EARTHQUAKE RUPTURE IS CONTROLLED BY THE GEOMETRICAL COMPLEXITY OF FAULT TRACES AND VARIATIONS IN ACCUMULATED STRESS LEVELS ALONG FAULTS THAT ARISE DUE TO THE LOCATION OF PAST EARTHQUAKES.

38 Summary points: Geologic Moment generally agrees with Seismic Moment for largest earthquakes Unilateral and Bilateral rupture both common No correlation of epicenter to location of maximum slip Shape of slip distribution both symmetric and asymmetric Fault Geometry evolves with Displacement Steps in Fault Trace do Correlate to Ends of Earthquake Ruptures - but not always. Earthquakes generally not observed to rupture through steps >4-5 km in dimension Earthquake rupture limited by fault geometry and past history of earthquake ruptures (stress release)

39 The same approach has been used for bends in fault zones -

40

41 Wesnousky, S.G. (1988), Seismological and Structural Evolution of Strike-Slip Faults, NATURE, 335, Wesnousky, S. G. (2006), Predicting the endpoints of earthquake ruptures, Nature, 444, , doi: /nature0525) Wesnousky, S. G., (2008), Displacement and geometrical characteristics of earthquake surface ruptures: Issues and implications for seismic hazard analysis and the earthquake rupture process, Bulletin of the Seismological Society of America, 98, 4, Wesnousky, S. G. and G. P. Biasi (2011), The length to which an earthquake will go to rupture, Bulletin of the Seismological Society of America, 101, , doi: /

42