Bulletin of the Seismological Society of America, Vol. 84, No. 4, pp. 974-1002, August 1994 New Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement by Donald L. Wells and Kevin J. Coppersmith Abstract Source parameters for historical earthquakes worldwide are com- piled to develop a series of empirical relationships among moment magnitude (M), surface rupture length, subsurface rupture length, downdip rupture width, rupture area, and maximum and average displacement per event. The resulting data base is a significant update of previous compilations and includes the ad- ditional source parameters of seismic moment, moment magnitude, subsurface rupture length, downdip rupture width, and average surface displacement. Each source parameter is classified as reliable or unreliable, based on our evaluation of the accuracy of individual values. Only the reliable source parameters are used in the final analyses. In comparing source parameters, we note the fol- lowing trends: (1) Generally, the length of rupture at the surface is equal to 75% of the subsurface rupture length; however, the ratio of surface rupture length to subsurface rupture length increases with magnitude; (2) the average surface dis- placement per event is about one-half the maximum surface displacement per event; and (3) the average subsurface displacement on the fault plane is less than the maximum surface displacement but more than the average surface dis- placement. Thus, for most earthquakes in this data base, slip on the fault plane at seismogenic depths is manifested by similar displacements at the surface. Log-linear regressions between earthquake magnitude and surface rupture length, subsurface rupture length, and rupture area are especially well correlated, show- ing standard deviations of 0.25 to 0.35 magnitude units. Most relationships are not statistically different (at a 95% significance level) as a function of the style of faulting: thus, we consider the regressions for all slip types to be appropriate for most applications. Regressions between magnitude and displacement, mag- nitude and rupture width, and between displacement and rupture length are less well correlated and have larger standard deviation than regressions between magnitude and length or area. The large number of data points in most of these regressions and their statistical stability suggest that they are unlikely to change significantly in response to additional data. Separating the data according to extensional and compressional tectonic environments neither provides statisti- cally different results nor improves the statistical significance of the regressions. Regressions for cases in which earthquake magnitude is either the independent or the dependent parameter can be used to estimate maximum earthquake mag- nitudes both for surface faults and for subsurface seismic sources such as blind faults, and to estimate the expected surface displacement along a fault for a given size earthquake. Introduction Seismic hazard analyses, both probabilistic and de- terministic, require an assessment of the future earth- quake potential in a region. Specifically, it is often nec- essary to estimate the size of the largest earthquakes that might be generated by a particular fault or earthquake source. It is rare, however, that the largest possible earthquakes along individual faults have occurred during the historical period. Thus, the future earthquake poten- 974
93
Embed
New Empirical Relationships among Magnitude, Rupture ...€¦ · Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 975
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Bulletin of the Seismological Society of America, Vol. 84, No. 4, pp. 974-1002, August 1994
New Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement
by Donald L. Wells and Kevin J. Coppersmith
Abstract Source parameters for historical earthquakes worldwide are com- piled to develop a series of empirical relationships among moment magnitude (M), surface rupture length, subsurface rupture length, downdip rupture width, rupture area, and maximum and average displacement per event. The resulting data base is a significant update of previous compilations and includes the ad- ditional source parameters of seismic moment, moment magnitude, subsurface rupture length, downdip rupture width, and average surface displacement. Each source parameter is classified as reliable or unreliable, based on our evaluation of the accuracy of individual values. Only the reliable source parameters are used in the final analyses. In comparing source parameters, we note the fol- lowing trends: (1) Generally, the length of rupture at the surface is equal to 75% of the subsurface rupture length; however, the ratio of surface rupture length to subsurface rupture length increases with magnitude; (2) the average surface dis- placement per event is about one-half the maximum surface displacement per event; and (3) the average subsurface displacement on the fault plane is less than the maximum surface displacement but more than the average surface dis- placement. Thus, for most earthquakes in this data base, slip on the fault plane at seismogenic depths is manifested by similar displacements at the surface. Log-linear regressions between earthquake magnitude and surface rupture length, subsurface rupture length, and rupture area are especially well correlated, show- ing standard deviations of 0.25 to 0.35 magnitude units. Most relationships are not statistically different (at a 95% significance level) as a function of the style of faulting: thus, we consider the regressions for all slip types to be appropriate for most applications. Regressions between magnitude and displacement, mag- nitude and rupture width, and between displacement and rupture length are less well correlated and have larger standard deviation than regressions between magnitude and length or area. The large number of data points in most of these regressions and their statistical stability suggest that they are unlikely to change significantly in response to additional data. Separating the data according to extensional and compressional tectonic environments neither provides statisti- cally different results nor improves the statistical significance of the regressions. Regressions for cases in which earthquake magnitude is either the independent or the dependent parameter can be used to estimate maximum earthquake mag- nitudes both for surface faults and for subsurface seismic sources such as blind faults, and to estimate the expected surface displacement along a fault for a given size earthquake.
Introduction
Seismic hazard analyses, both probabilistic and de- terministic, require an assessment of the future earth- quake potential in a region. Specifically, it is often nec- essary to estimate the size of the largest earthquakes that
might be generated by a particular fault or earthquake source. It is rare, however, that the largest possible earthquakes along individual faults have occurred during the historical period. Thus, the future earthquake poten-
974
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 975
tial of a fault commonly is evaluated from estimates of fault rupture parameters that are, in turn, related to earthquake magnitude.
It has been known for some time that earthquake magnitude may be correlated with rupture parameters such as length and displacement (e.g., Tocher, 1958: Iida, 1959; Chinnery, 1969). Accordingly, paleoseismic and geologic studies of active faults focus on estimating these source characteristics. For example, data from geo- morphic and geologic investigations of faults may be used to assess the timing of past earthquakes, the amount of displacement per event, and the segmentation of the fault zone (e.g., Schwartz and Coppersmith, 1986; Schwartz, 1988; Coppersmith, 1991). To translate these source characteristics into estimates of earthquake size, rela- tionships between rupture parameters and the measure of earthquake size, typically magnitude, are required.
Numerous published empirical relationships relate magnitude to various fault rupture parameters. Typi- cally, magnitude is related to surface rupture length as a function of slip type. Additional relationships that have been investigated include displacement versus rupture length, magnitude versus maximum surface displace- ment, magnitude versus total fault length, and magni- tude versus surface displacement times surface rupture length (Tocher, 1958; Iida, 1959; Albee and Smith, 1966; Chinnery, 1969; Ohnaka, 1978; Slemmons, 1977, 1982; Acharya, 1979; Bonilla and Buchanon, 1970; Bonilla et al., 1984; Slemmons et al., 1989). Other studies relate magnitude and seismic moment to rupture length, rup- ture width, and rupture area as estimated from the extent of surface deformation, dimensions of the aftershock zone, or earthquake source time functions (Utsu and Seki, 1954; Utsu, 1969; Kanamori and Anderson, 1975; Wyss, 1979; Singh et al., 1980; Purcaru and Berckhemer, 1982; Scholz, 1982; Wesnousky, 1986; and Darragh and Bolt, 1987).
The purpose of this article is to present new and re- vised empirical relationships between various rupture pa- rameters, to describe the empirical data base used to de- velop these relationships, and to draw first-order conclusions regarding the trends in the relationships. Specifically, this article refines the data sets and extends previous studies by including data from recent earth- quakes and from new investigations of older earth- quakes. The new data provide a much larger and more comprehensive data base than was available for previous studies. Additional fault characteristics, such as subsur- face rupture length, downdip rupture width, and average fault displacement, also are included. Because the new data set is more comprehensive than those used for pre- vious studies, it is possible to examine relationships among various rupture parameters, as well as the relationships between rupture parameters and magnitude. An impor- tant goal of this article is to present the observational data base in a form that is sufficiently complete to enable
the reader to reproduce our results, as well as to carry out subsequent analyses.
The following sections describe the observational data base, present the statistical relationships developed be- tween magnitude and fault rupture parameters, and then evaluate the relationships in terms of their statistical sig- nificance, relative stability, and overall usefulness.
Data Base
A worldwide data base of source parameters for 421 historical earthquakes is compiled for this study. The data include shallow-focus (hypocentral depth less than 40 km), continental interplate or intraplate earthquakes of mag- nitudes greater than approximately 4.5. Earthquakes as- sociated with subduction zones, both plate interface earthquakes and those occurring within oceanic slabs, are excluded. For each earthquake in the data base, we compiled seismologic source parameters and fault char- acteristics, including seismic moment, magnitude, focal mechanism, focal depth, slip type, surface and subsur- face rupture length, maximum and average surface dis- placement, downdip rupture width, and rupture area.
In general, the data presented in this article are ob- tained from published results of field investigations of surface faulting and seismologic investigations. For many earthquakes, there are several published measurements of various parameters. One objective of this study is to identify the most accurate value for each parameter, or the average value where the accuracy of individual val- ues could not be determined. Special emphasis is placed on identifying the sources and types of measurements reported in the literature (e.g., rupture area based on af- tershock distribution, geodetic modeling, or teleseismic inversion). All data are then categorized by type of mea- surement, and the most accurate value is selected for fur- ther analysis. The data selection process for each rupture parameter is described in detail in the following sections.
From the larger data base, 244 earthquakes are se- lected to develop empirical relationships among various source parameters. For these earthquakes, which are listed in Table 1, the source parameters are considered much more reliable than the source parameters for the other earthquakes. Earthquakes that are evaluated but ex- cluded from further study because of insufficient infor- mation or poor-quality data are provided on microfiche (Appendix A). Each earthquake listed in Table 1 is iden- tified by location, name (geographic descriptor or as- sociated fault), and date of origin in Coordinated Uni- versal Time (UTC). Each source parameter given in Table 1 is discussed below.
Slip Type
Past studies have demonstrated that the slip type or style of faulting is potentially significant for correlating earthquake magnitude and rupture parameters (e.g.,
Slemmons, 1977; Bonilla et al., 1984). To categorize the dominant slip type for each earthquake in our data base, we use a simple classification scheme based on the ratio of the horizontal component of slip to the vertical component of slip. The horizontal-to-vertical slip ratio is calculated from all estimates of the components of slip, including, in order of priority, surface displacement, geodetic modeling of surface deformation, and the rake from earthquake focal mechanisms.
Published earthquake focal mechanisms were re- viewed to compare the nature of surface deformation, such as surface fault displacements and regional subsi- dence, uplift, or lateral deformation, with the seismo- logic data for each earthquake. For some earthquakes, there are several published focal mechanisms, including those derived from waveform inversions, P-wave first motions, and moment tensor inversions. Because focal mechanisms derived from waveform inversion of long- period P and SH waves usually are considered more rep- resentative of the primary style of co-seismic slip than are short-period P-wave first-motion solutions, the for- mer generally are preferred (Aki and Richards, 1980). Theoretically, because the nature and amount of slip at the surface is at least partly controlled by the depth of the focus and the nature of surface geologic conditions, categorizing slip based solely on the slip components measured at the surface may not correspond to the slip type indicated by seismologic data. In practice, how- ever, we find that the dominant sense of slip at the sur- face is representative of the overall sense of slip mea- sured from the rake of earthquake focal mechanisms.
Slip types for the earthquakes in Table 1 reflect the following scheme, which is based on the ratio of hori- zontal (HZ; strike slip, S) to vertical (VT; reverse, R, or normal, N) slip:
HZ:VTSIip >2:1 2 : l t o l : l l : l t o l : 2 < 1 : 2
Slip Type S S-R, S-N R-S, N-S R, N
In Table 1, the strike-slip component is characterized as right lateral (RL) or left lateral (LL), depending on the sense of horizontal displacement. For 60 oblique-slip earthquakes, the subordinate sense of slip is listed after the primary slip type. For the regressions, each earth- quake is assigned to one of three slip types: strike slip, normal, or reverse. Earthquakes having a horizontal-to- vertical slip ratio greater than 1 to 1 are considered strike slip; those having a horizontal-to-vertical slip ratio of 1 to 1 or less are considered normal or reverse, depending on the sense of vertical displacement.
The earthquakes in Table 1 also are categorized by other characteristics to evaluate potential differences in rupture parameter correlations. Earthquakes are charac- terized with respect to whether they occurred within a compressional environment (one that is characterized by
compressional or transpressional tectonics), or within an extensional environment (one that is characterized by ex- tensional or transtensional tectonics). Slemmons et al. (1989) proposed a similar classification for their data base and found no significant differences between regressions developed for the two environments. The earthquakes also are separated according to whether they occurred within an active plate margin or within a stable conti- nental region. Stable continental regions are regions of continental crust that have no significant Cenozoic tec- tonism or volcanism (Electric Power Research Institute, 1987; Johnston and Kanter, 1990); active plate margins include all other regions in our data base.
Magnitude and Seismic Moment
Estimates of moment magnitude (M) and surface- wave magnitude (Ms) are listed in Table 1. Most pre- vious studies of earthquake source parameters compiled M s estimates, because these are the most commonly cited magnitudes for older instrumental earthquakes. There are, however, several problems associated with using Ms to analyze source parameter relationships. Because Ms is a measure of seismic-wave amplitude at a specific period (approximately 18 to 22 sec), it measures only the en- ergy released at this period. Although Ms values gen- erally are very stable between nearby stations, signifi- cant variations in Ms may occur between distant stations. These variations are related to azimuth, station distance, instrument sensitivity, and crustal structure (Panza et al., 1989). Furthermore, for very large earthquakes (Ms > 8.0), the periods at which Ms is measured become sat- urated and no longer record large-scale faulting char- acteristics (Hanks and Kanamori, 1979). A similar prob- lem with saturation of measured seismic waves also occurs for scales such as local or Richter magnitude (ML) and body-wave magnitude (mb). For small earthquakes (Ms < 5.5), 20-sec surface-wave amplitudes are too small to be recorded by many seismographs (Kanamori, 1983). Thus, traditional magnitude scales are limited by both the frequency response of the Earth and the response of the recording seismograph.
A physically meaningful link between earthquake size and fault rupture parameters is seismic moment, M0 = /~/9 A, where ~ is the shear modulus [usually taken as 3 × 1011 dyne/cm 2 for crustal faults (Hanks and Kan- amori, 1979)];/9 is the average displacement across the fault surface; and A is the area of the fault surface that ruptured. In turn, M0 is directly related to magnitude [e.g., M = 2/3 * log M0 - 10.7 (Hanks and Kanamori, 1979)].
Seismic moment (M0) also is considered a more ac- curate measure of the size of an earthquake than are tra- ditional magnitude scales such as Ms and mb because it is a direct measure of the amount of radiated energy, rather than a measure of the response of a seismograph to an earthquake (Hanks and Wyss, 1972). It is com- puted from the source spectra of body and surface waves
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 983
(Hanks et al., 1975; Kanamori and Anderson, 1975) or is derived from a moment tensor solution (Dziewonski et al., 1981). Furthermore, there is a larger variability in the value of Ms than of Mo measured at different sta- tions. For any earthquake, Ms values from stations at different azimuths may differ by as much as 1.5 mag- nitude units, whereas Mo values rarely differ by more than a factor of 10, which is equivalent to a variability of 0.7 in M values. Thus, M is considered a more re- liable measure of the energy released during an earth- quake (Hanks and Kanamori, 1979).
For earthquakes that lack published M s estimates, other measures such as Richter magnitude (NIL) or body- wave magnitude (mb) are listed in Table 1. Because there are several methods for calculating Ms, values calculated by comparable methods are listed where possible. Ac- cording to Lienkaemper (1984), Ms calculated by the Prague formula, which is used for Preliminary Deter- mination of Epicenters (PDE--U.S. Geological Survey monthly bulletin), is directly comparable to MaR calcu- lated by Gutenberg and Richter (1954). On the average, Ms computed by Abe (1981), Gutenberg (1945), and Richter (1958) differ systematically from Ms (PDE) and MaR (Lienkaemper, 1984). Comparable Ms values listed in this report are taken from the following sources, listed in order of preference: Ms (PDE), M s (Lienkaemper, 1984), and MaR (Gutenberg and Richter, 1954). Addi- tional sources for magnitudes are listed in the footnotes to Table 1.
To arrive at a single estimate of seismic moment for each earthquake in the data base, we calculate an av- erage seismic moment from all published instrumental seismic moments, including those measured from body waves, surface waves, and centroid moment tensor so- lutions. Noninstrumental estimates of seismic moment, such as those based on estimates of rupture dimensions or those estimated from magnitude-moment relation- ships, are not used to calculate average seismic moment. Moment magnitudes are calculated from the averaged seismic moment by the formula of Hanks and Kanamori (1979): M = 2/3 * log M0 - 10.7. The values of M calculated from/140 are shown to two decimal places in Table 1 to signify that they are calculated values; these values are used for the regression analyses. When con- sidering individual estimates of moment magnitude, however, these values are considered significant only to one decimal place, and should be rounded to the nearest tenth of a magnitude unit.
Previous studies of the relationship between Ms and M indicate that these magnitudes are approximately equal within the range of Ms 5.0 to 7.5 (Kanamori, 1983). Our data set shows no systematic difference between Ms and M in the range of magnitude 5.7 to 8.0 (Fig. 1). In the range of magnitude 4.7 to 5.7, Ms is systematically smaller than M, in agreement with the results of Boore and Joy- net (1982). The standard deviation of the difference be-
tween each pair of Ms and M values in Figure 1 is ap- proximately 0.19. This standard deviation is less than the standard deviation of 0.28 calculated by Lienkaem- per (1984) for residuals of all single-station Ms estimates for individual earthquakes. Based on these standard de- viations, the difference between the magnitude scales (Ms and M) is insignificant for the earthquakes of magnitude greater than 5.7 listed in Table 1.
For regressions of magnitude versus surface rupture length and magnitude versus maximum displacement, previous studies excluded earthquakes with magnitudes less than approximately Ms 6.0 (Slemmons, 1982; Bon- ilia et al., 1984; Slemmons et al., 1989). These authors noted that earthquakes of Ms less than 6.0 often have surface ruptures that are much shorter than the source length defined by aftershocks, and that possible surface ruptures for these earthquakes may be less well studied than those for earthquakes of larger magnitude. Fur- thermore, surface faulting associated with earthquakes of magnitude less than 6.0 may be poorly expressed as discontinuous traces or fractures, showing inconsistent or no net displacement (Darragh and Bolt, 1987; Bon- illa, 1988). We evaluate regression statistics for mag- nitude versus surface rupture length and magnitude ver- sus surface displacement for earthquakes of magnitude less than 6.0 (Ms or M), and conclude that elimination of the magnitude cutoff expands the data sets without significantly compromising the regression statistics. Thus, several well-studied surface-rupturing earthquakes of
v
7 "13
° m ¢..
a
J I ' I ' J ' I . / ,
176 EQs o. / $ I S /
X o / ~
, .
. , , L f J I I r i I I
5 6 7 8 9
Moment Magnitude (M)
Figure 1. Surface-wave magnitude (Ms) ver- sus moment magnitude (M) for historical conti- nental earthquakes. Segmented linear regression shown as solid line, with segment boundaries at M 4.7, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, and 8.2. Short dashed lines indicate 95% confidence interval of regression line. Long dashed line indicates equal magnitudes (1 to 1 slope).
984 D.L. Wells and K. J. Coppersmith
magnitude less than 6.0 (e.g., 1979 Homestead Valley and 1983 Nunez-Coalinga, California) are included in the data base.
For the regressions on subsurface rupture length and on rupture area, the lower bound of magnitude is set at M 4.7 because aftershock sequences for earthquakes of lower magnitude rarely are the subject of detailed in- vestigations. Aftershocks and source parameters of nu- merous recent earthquakes of moderate magnitude (M 4.7 to 6.0) have been studied in detail (e.g., 1984 North Wales, England; 1986 Kalamata, Greece; and 1988 Pas- adena and 1990 Upland, California). It is appropriate to use these moderate-magnitude earthquakes to evaluate subsurface rupture length, rupture width, and rupture area relationships, because the use of subsurface character- istics eliminates the problems associated with the incom- plete expression of rupture at the surface usually asso- ciated with moderate-magnitude earthquakes (Darragh and Bolt, 1987).
Instrumentally measured magnitudes (Ms or M) do not exist for all the earthquakes listed in Table 1. For these earthquakes, magnitudes are estimated from re- ports of felt intensity (MI), or are estimated from the rup- ture area and displacement using the definition of seis- mic moment [M0 = /x/5 A (Hanks and Kanamori, 1979)]. The earthquakes that lack instrumental magnitudes are included for use in displacement-to-length relationships, which do not require magnitude.
Surface Rupture Length
The length of rupture at the surface is known to be correlatable with earthquake magnitude. This study re- views and reevaluates previously published surface rup- ture lengths for historical earthquakes and expands the data set to include recent earthquakes and new studies of older events. Published and unpublished descriptions of surface rupture are reviewed to evaluate the nature and extent of surface faulting for 207 earthquakes. Rather than relying on values reported in secondary data com- pilations, we reviewed original field reports, maps, and articles for each earthquake.
Rupture lengths measured from maps and figures are compared to the lengths reported in descriptions of sur- face faulting. Descriptions of surface faulting also are reviewed to evaluate whether the ruptures are primary or secondary. Primary surface rupture is defined as being related to tectonic rupture, during which the fault rupture plane intersects the ground surface. Secondary faulting includes fractures formed by ground shaking, fractures and faults related to landslides, and triggered slip on sur- face faults not related to a primary fault plane (e.g., slip on bedding plane faults or near-surface slip on adjacent or distantly located faults). Because identifying primary tectonic rupture is particularly difficult for smaller-mag- nitude earthquakes (less than approximately Ms or M 6.0), these events are included in regression analyses only when
the tectonic nature of the surface rupture is clearly es- tablished (e.g., the 1966 Parkfield, California, earth- quake, but not the 1986 Chalfant Valley, California, earthquake). Discontinuous surface fractures mapped be- yond the ends of the continuous surface trace are con- sidered part of the tectonic surface rupture and are in- cluded in the calculation of surface rupture length.
Major sources of uncertainty in reported measure- ments of surface rupture length are as follows. (1) In- complete studies of the rupture zone. Less than the entire surface rupture was investigated and mapped for any of various reasons, such as inaccessibility, discontinuity of the surface trace along strike so the entire rupture was not identified, or the fault trace was obscured before postearthquake investigations were undertaken. Consid- erable uncertainty in the extent of rupture is assessed for investigations completed years to decades after an earth- quake. (2) Different interpretations of the nature and ex- tent of surface deformation. Interpretations may differ on the extent of primary surface rupture, the differentia- tion of primary and secondary surface rupture, and the correlation of surface rupture on different faults to in- dividual earthquakes for multiple event sequences. (3) Unresolvable discrepancies between lengths reported by different workers. These discrepancies are related to level of effort in field investigations, method of measuring fault traces, or lengths reported in text versus the lengths drawn on maps.
Earthquakes are selected for regression analyses in- volving surface rupture length if the data met all of the following criteria: (1) uncertainty in the rupture length does not exceed approximately 20% of the total length of the rupture; (2) at least one estimate of the amount of surface displacement is reported; and (3) the lengths of ruptures resulting from individual events in multiple earthquake sequences are known.
Subsurface Rupture Length, Downdip Width, and Rupture Area
Subsurface source dimensions, both rupture length and rupture area (length times downdip width), are eval- uated for more than 250 earthquakes. Wyss (1979) com- piled a smaller data base of rupture areas for continental and subduction zone earthquakes, and Darragh and Bolt (1987) compiled subsurface rupture lengths for moder- ate-magnitude strike-slip earthquakes. We expand the data base and relate these rupture parameters to moment mag- nitude.
The primary method used to estimate subsurface rupture length and rupture area is the spatial pattern of early aftershocks. Aftershocks that occur within a few hours to a few days of the mainshock generally define the maximum extent of co-seismic rupture (Kanamori and Anderson, 1975; Dietz and Ellsworth, 1990). Because the distribution of aftershocks may expand laterally and vertically following the mainshock, the initial size of the
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 985
aftershock zone is considered more representative of the extent of co-seismic rupture than is the distribution of aftershocks occurring within days to months of the mainshock. Furthermore, detailed studies of aftershocks of several recent earthquakes (such as the 1989 Loma Prieta, California) suggest that early aftershocks occur at the perimeter of the co-seismic rupture zone, and that the central part of this zone is characterized by a lack of seismicity for the first few hours to days after the
Figure 2. Surface rupture length versus sub- surface rupture length estimated from the distri- bution of early aftershocks of historical continen- tal earthquakes.
1.4
t -
1.2
d 1
o
~"- .8 k_
r~ :s .6
(/3
OJ .4 U 0 ~c ~ .2
I I r I
53 Earthquakes
• • •
l e o %, • • t R i O
~o o • •
" . . ~ ° " o~O
0 a I i I i I i I 4 5 6 7 8
Magni tude (M)
Figure 3. Ratio of surface to subsurface rup- ture length versus magnitude.
mainshock (Mendoza and Hartzell, 1988; Dietz and Ells- worth, 1990). This observation suggests that even the rupture area defined by early aftershocks may be slightly larger than the actual co-seismic rupture zone (Mendoza and Hartzell, 1988).
We estimate subsurface rupture length using the length of the best-defined aftershock zone. The accuracy of the size of the aftershock zone depends on the accuracy of the locations of individual aftershocks, which depends, in turn, on the azimuths and proximity of the recording stations and the accuracy of the subsurface structure ve- locity model. The largest uncertainty typically is in- curred in calculating the depths of the hypocenters rather than the areal distribution of epicenters (Gubbins, 1990). Earthquakes are excluded from regression analysis if only a few aftershocks were recorded, or if the aftershock lo- cations were very uncertain.
Alternative but less satisfactory methods to assess the extent of subsurface co-seismic rupture include con- sidering the surface rupture length, geodetic modeling of surface displacement, and modeling of the earthquake source time function. Comparisons for this study suggest that the surface rupture length provides a minimum es- timate of the subsurface rupture length. For example, for 53 earthquakes for which data on both surface and sub- surface rupture length are available, surface rupture length averaged about 75% of subsurface rupture length (Fig. 2). However, the ratio of surface rupture length to sub- surface rupture length appears to increase with magni- tude (Fig. 3). Thus, we conclude that surface rupture length is a more reliable estimator of subsurface rupture length as magnitude increases.
Estimates of rupture length calculated from geodetic modeling of vertical and horizontal changes at the ground surface, or from corner frequencies of seismograms (source time functions for circular, unilateral, or bilat- eral ruptures) also are compiled from the literature. For some earthquakes, rupture lengths estimated from these methods are much shorter than rupture lengths measured from the distribution of aftershocks (Mendoza and Hart- zell, 1988). Thus, these measures of rupture length may not represent the extent of co-seismic rupture in the same way that aftershocks do. In this study, estimates of sub- surface rupture length based on geodetic modeling or source time functions are accepted for regression anal- ysis only when independent estimates of rupture length are available for corroboration.
Downdip rupture widths are estimated from the depth distribution of the best-defined zone of aftershocks. Where the downdip width of rupture is unknown from the dis- tribution of aftershocks, it is estimated from the depth (thickness) of the seismogenic zone or the depth of the hypocenter and the assumed dip of the fault plane. For most earthquakes of magnitude 5 1/2 or larger, the mainshock typically occurs at or near the base of the seismogenic zone (Sibson, 1987). Estimates of rupture
986 D.L. Wells and K. J. Coppersmith
width based on hypocentral depth of the mainshock or width of the seismogenic zone are used to calculate rup- ture area only for earthquakes for which detailed infor- mation on regional seismicity is available, or for which detailed studies of the hypocentral depth and focal mech- anism have been performed.
Major sources of uncertainty for measuring subsur- face rupture parameters are as follows: (1) accuracy of aftershock locations in three dimensions; (2) interpreta- tion of the initial extent (length and downdip width) of the aftershock sequence; (3) temporal expansion of the aftershock zone; (4) interpretation of the length of mul- tiple earthquake rupture sequences; (5) identification of the strike and dip of the rupture plane from aftershocks; and (6) reliability of geodetic and seismologic modeling.
Earthquakes are selected for regression analyses in- volving subsurface rupture length, rupture width, and rupture area if the data met the following criteria: (1) subsurface rupture length and width are measured from an aftershock sequence of known duration; and (2) af- tershocks were recorded by a local seismograph net- work, or many aftershocks were recorded at teleseismic stations. In cases where information on aflershock dis- tribution is lacking, the earthquake is included in the analysis if (1) consistent subsurface rupture lengths are calculated from at least two sources such as geodetic modeling, source time functions, or surface rupture length, and (2) rupture width can be estimated confidently from the thickness of the seismogenic zone or the depth of the mainshock hypocenter.
Maximum and Average Surface Displacement
Observational data from field studies of faults as well as theoretical studies of seismic moment suggest that earthquake magnitude should correlate with the amount of displacement along the causative fault. In contrast to the published information on surface rupture length, dis- placement measurements for many earthquakes often are poorly documented. In this study, we attempted system- atically to compile information on the amount of co-seis- mic surface displacement and to identify the maximum and the average displacement along the rupture.
The most commonly reported displacement mea- surement is the max imum observed horizontal and/or vertical surface displacement. We reviewed published measurements of displacement, including components of horizontal and vertical slip to calculate a net maximum displacement for each earthquake. Because the majority of displacement measurements reported in the literature were measured weeks to years after the earthquake, these displacement estimates may include post-co-seismic slip or fault creep. For events where displacements were measured at several time periods, we generally select the first measurements recorded after the earthquake to min- imize possible effects of fault creep. For several recent events in our data base (such as 1992 Landers, Califor-
nia), we note that little or no postearthquake creep was observed. Thus, displacement measurements recorded several weeks or longer after the earthquake may rep- resent the actual co-seismic slip, except for a few regions where post-co-seismic slip has been documented (e.g., Parkfield and Imperial Valley regions of California).
The net displacement is calculated from the vector sum of the slip components (horizontal and vertical) measured at a single location. Commonly, the maxi- mum horizontal displacement and the maximum vertical displacement occur at different locations along a rupture. In those cases, unless the subordinate component is re- corded at the sites of the maxima, a net slip vector can- not be calculated. Furthermore, it is difficult to recog- nize and measure compression and extension across a fault, even for the more recent, well-studied earth- quakes.
Average displacement per event is calculated from multiple measurements of displacement along the rup- ture zone. For most earthquakes, the largest displace- ments typically occur along a limited reach of the rupture zone. Thus, simple averaging of a limited number of dis- placement measurements is unlikely to provide an ac- curate estimate of the true average surface displacement. The most reliable average displacement values are cal- culated from net displacement measurements recorded along the entire surface rupture. Figure 4 shows a sur- face displacement distribution for the 1968 Borrego Mountain, California, earthquake, a relatively well-stud- ied event. The average displacement may be calculated by several graphical methods, including a linear point- to-point function, a running three-point average, or an enveloping function that minimizes the effects of anom- alously low or high displacement measurements (D. B. Slemmons, 1989, personal comm.). The average-dis- placement data base reported in this study includes events examined by Slemmons using graphical techniques, and
I-- 400 '~
350 ~
300 -
250 -
~£ 200- LU 150 -
<1~ 100 -
50 -
o 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
H O R I Z O N T A L D I S T A N C E (km)
=~ i - - - -Cent ra l Break ---}=~outh Break-oq North Break
Figure 4. Distribution of right slip measured in April 1968 for the 9 April 1968 Borrego Moun- tain, California, earthquake. Dashed line indicates estimated displacement for April 1968 (modified from Clark, 1972).
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 987
events for which data were obtained from the published literature or calculated from individual measurements of displacement for these earthquakes. Specifically, we in- clude estimates of average displacement that we calcu- late from a minimum of 10 displacement measurements distributed along the surface rupture, or were reported from extensive studies of the entire surface rupture.
For the average-displacement data set, the maximum surface displacement is about twice the average surface displacement, although the ratio of average to maximum surface displacement ranges from about 0.2 to 0.8 (Fig. 5). In addition, for a subset of earthquakes with pub- lished instrumental estimates of seismic moment, the ra- tio of average to maximum displacement does not vary systematically as a function of magnitude (Fig. 5).
A matter of interest is the relationship of co-seismic surface displacement to "subsurface" displacement that occurs on the fault plane within the seismogenic crust (as given in the definition of seismic moment). To eval- uate the relationship of surface displacement to average subsurface displacement, we calculate an average dis- placement from the seismic moment and the rupture area for all earthquakes having acceptable estimates of max- imum and average surface displacement, seismic mo- ment, and rupture area. The calculated values of sub- surface displacement are compared with the observed maximum and average surface displacements in Figures 6 and 7. The ratio of average subsurface displacement to maximum surface displacement ranges from 0.14 to 7.5; the ratio of average subsurface displacement to av- erage surface displacement ranges from 0.25 to 6.0. These ratios do not appear to vary as a function of magnitude (Figs. 6a and 6b).
To evaluate the distribution of data, we calculate re-
E o
c/)
E
E x
.8
.6
I ,4 -
.2
0 4
I I I I
• • • • •
• ° O • go
oetdb • • • o05, ,oB °
eeO% • • O 0
• oqllJ
• QO g o •
57 Earthquakes
, J , I , I , I 5 6 7 8
Magnifude (M)
Figure 5. Ratio of average surface to maxi- mum surface displacement versus magnitude.
siduals for the ratios and find that the distribution is con- sistent with a normal distribution of data. Because of this and because of the large range of data, we believe that the mode provides an appropriate measure of the distri- bution of ratios. For 44 earthquakes for which we have estimates of both maximum displacement and subsurface displacement, the mode of the distribution of the ratios of average subsurface displacement to maximum surface displacement is 0.76 (Fig. 7a). This indicates that for most earthquakes, the average subsurface displacement is less than the maximum surface displacement. For 32 earthquakes for which we have estimates of both average displacement and subsurface displacement, the mode of the distribution of the ratios of average subsurface dis- placement to average surface displacement is 1.32 (Fig. 7b). Thus, for the earthquakes in our data set, average subsurface displacement is more than average surface displacement and less than maximum surface displace- ment. Furthermore, for these earthquakes, most slip on the fault plane at seismogenic depths is manifested at the surface.
The major sources of uncertainty in the displace- ment data set reflect the following: (1) documentation of less than the entire fault rupture trace; (2) lack of suitable features (e.g., stratigraphy, streams, or cultural features) for measuring displacement; (3) distribution of displace- ment along multiple fault strands, or distributed shearing over a broad fault zone; (4) modification of the fault scarp by landsliding or erosion; (5) increase in displacement due to afterslip; (6) inadequately documented locations of slip measurements; and (8) measurements of slip on geomorphic features displaced by repeated earthquakes or postearthquake creep.
Earthquakes are selected for regression analyses in- volving displacement if the data met all of the following criteria: (1) type of displacement (strike slip, reverse, normal) and nature of measurement (maximum or av- erage surface slip) are known; (2) slip occurred primarily on a single fault, or the total slip across a zone of faults is known; (3) net maximum displacement is calculated from horizontal and vertical components of slip mea- sured at a single locality; and (4) the measured displace- ment can be attributed uniquely to the most recent earth- quake. In addition, for average displacement, the estimate is calculated from the sum of numerous contempora- neous displacement measurements, or was reported in literature by researchers who investigated the entire length of the surface rupture.
Regression Models
Numerous regression models exist for evaluating the relationship between any pair of variables, including models for linear or nonlinear relationships and normal (Gaussian) or nonparametric distributions of data. Most previous studies of fault rupture parameters used a sire-
988 D.L. Wells and K. J. Coppersmith
pie linear regression model such as ordinary least squares. Other models considered for this study included least- normal squares and reduced major axis (Troutman and Williams, 1987). These models have the advantage of providing a unique solution regardless of which variable is chosen to be the dependent variable. Although this unique solution provides the best fit to all the data, and thus the most accurate interpretation of the relationship between variables, it does not minimize the error in pre- dicting any individual variable. An ordinary least-squares model, however, calculates a nonunique solution that minimizes the error in predicting the dependent variable from the independent variable (Troutman and Williams,
1987). Thus, because we are interested in predicting pa- rameters to evaluate seismic hazard, and to make our new empirical relationships comparable to previously determined relationships, we use an ordinary least-squares regression model for all analyses.
A further consideration in selecting a regression model is how it treats uncertainties in the data. Based on their detailed analysis of the "measurement" uncertainties as- sociated with magnitudes (Ms), surface rupture lengths, and maximum displacements, Bonilla et al. (1984) noted that for any given earthquake, the stochastic variance (earthquake-to-earthquake differences) in these rupture parameters dominates errors in measurement. Specifi-
7 o
o 0
, , i I , i , , j , , , , , ,
(a)
o o o
o O°oe o o
o o
~ go o
0 O 0 0 0 0
0
0
0 0
0
0
' ' ' ' ' ' I , , , i , ,
(b)
o o
o o o o o
oo o
o O o ~ D o o o o o o o
0008 o
o
44 Earthquakes 32 Earthquakes
i i i i i i i i J i i i i i i i i I I i i i i i J i i i i i i i
0 - I 1 1010 -1 1 10
Ave S u b s u r f a c e / M a x Su r face Disp Ave S u b s u r f a c e / A v e Su r face Disp
Figure 6. (a) Ratio of average subsurface to maximum surface displacement versus magnitude. (b) Ratio of average subsurface to average surface displace- ment versus magnitude. Average subsurface displacement is calculated from the seismic moment and the rupture area.
.Q E z
20
18
16
14
12
10
8
6
4
2
0 10 -1 1 1010 -1 1
Ave S u b s u r f a c e / M a x Su r face Disp Ave S u b s u r f a c e / A v e S u r f a c e Disp
F i g u r e 7 . (a) Histogram of the logarithm of the ratio of average subsurface to maximum surface displacement. (b) Histogram of the logarithm of the ratio of average subsurface to average surface displacement. Average subsurface dis- placement is calculated from the seismic moment and the rupture area.
10
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 989
cally, they observed that a weighted least-squares model, which incorporates estimated measurement errors as a weighing factor, provides no better correlations than does an ordinary least-squares regression model. Similarly, Singh et al. (1980) analyzed the effects of data errors on solutions from linear and quadratic regressions. They concluded that there are significant difficulties in esti- mating the errors in source parameters, and that includ- ing estimated errors did not significantly improve the statistical correlations.
Although earthquake-specific uncertainties in the measured data are not listed in Table 1, the uncertainty in each listed parameter falls within the limits of ac- ceptability defined by the selection criteria, except for those parameters shown in parentheses. The parameters shown in parenthesis are excluded from the regression analyses because the uncertainties in the values are too large; however, these values are included in the data set for the sake of completeness. Thus, we consider the measurement uncertainties during the data selection pro- cess, but not for the regression analyses. For the 244 earthquakes included in the analyses, the uncertainties in measurements for any given earthquake are consid- ered much smaller than the stochastic variation in the data set as a whole.
One assumption of ordinary least-squares models is that the residuals have a normal distribution. Because many geologic and seismologic variables do not have a normal distribution, it is necessary to transform the data to a logarithmic form; this transformed data typically has a normal distribution (Davis, 1986). To test the as- sumption that the data sets have a (log) normal distri- bution, we calculate residuals between the empirical data and the predicted independent variable from each regres- sion equation. We complete X 2 tests for binned and un-
binned data sets for each set of residuals. We compute the optimum number of bins for each data set using the method of Benjamin and Cornell (1970). The X 2 tests indicate that the distribution of residuals for all data sets is consistent with a normal distribution of data at a 95% significance level. We also examine the distribution of residuals for each data set to evaluate the fit of the data to the regression model. Because the distribution of re- siduals shows no obvious trends, a linear regression model provides a satisfactory fit to the data (Fig. 8).
One significant change from the methods and results of most previous studies is that our analyses present regressions based on moment magnitude (M) rather than surface-wave magnitude (Ms). During preliminary anal- ysis of the regression relationships, we observed that the standard deviation of magnitude is consistently smaller for relationships based on M than for relationships based on Ms. In addition, the correlation coefficient generally is slightly higher for M relationships than for Ms rela- tionships. One advantage, however, to using Ms-based relationships is that the number of events in each rela- tionship is increased. We consider the smaller standard deviations and generally improved correlations for M- based relationships more important than increasing the size of the data set. We present only regressions based on M; for different applications, however, Ms-based re- lationships may be calculated from the data set.
Regression Results and Statistical Significance
Ordinary least-squares regression analyses (Tables 2A and 2B) include regression of M and lOgl0 of surface rupture length, subsurface rupture length, downdip rup- ture width, rupture area, maximum surface displace- ment, and average surface displacement as a function of
i i I l l l , , I , I , ~ l H I I I l l I H d I I , , , , , , i
1 0 1 0 0 1 0 ~ 1 0 4
Rupture Area (kin 2)
Figure 8. (a) Residuals for surface rupture length regression versus observed surface rupture length. (b) Residuals for rupture area regression versus observed rupture area.
990 D.L. Wells and K. J. Coppersmith
slip type. Regressions of surface rupture length and max- imum and average displacement also are presented (Ta- ble 2C). Regression descriptors include number of events, regression coefficients (a and b), standard error of the coefficients, standard deviation of the dependent vari- able (s), correlation coefficient (r), and data range. The empirical relationships have the form y = a + b * log (x) or log (y) = a + b * log (x), where y is the dependent variable and x is the independent variable. Two plots are presented for each pair of parameters. The first shows the data, the "all-slip-type" regression line (i.e., the regression fit to all of the data), and the 95% confidence interval (Figs. 9a through 16a). The second shows the regression lines for individual slip types (Figures 9b through 16b). The length of the regression line shows the range of data for each empirical relationship.
We calculate t statistics for the Correlation coeffi- cient to evaluate the significance of each relationship. A t distribution estimates a probability distribution based
on the size of the data set. We use a t test to calculate critical values of t, then compare these values to critical values of t for a selected significance level. We evaluate significance levels for a two-tailed distribution, because the correlation may be positive or negative. All rela- tionships are significant at a 95% probability level, ex- cept for the reverse-slip relationships for maximum and average displacement. These relationships are not sig- nificant because the position of the regression line is poorly constrained by the data; they are shown in brackets in Table 2 because they are not considered useful for pre- dicting dependent variables. Furthermore, we exclude them from comparisons to regression lines for other re- lationships. The results of our analyses indicate a poor correlation between surface displacement and other rup- ture parameters for reverse-slip earthquakes. The re- verse-slip relationships excluded from further analysis include maximum displacement versus magnitude, av- erage displacement versus magnitude, surface rupture
Table 2A Regressions of Rupture Length, Rupture Width, Rupture Area, and Moment Magnitude (M)
Coefficients and Standard Correlation Standard Errors
Slip Number of Deviation Coefficient Magnitude Length/Width Equation* Typet Events a(sa) b(sb) s r Range Range (kin)
M = a + b * l o g ( S R L )
l og (SRL) = a + b * M
M = a + b * l o g ( R L D )
l o g ( R L D ) = a + b * M
M = a + b * l o g ( R W )
l o g ( R W ) = a + b * M
M = a + b * l o g ( R A )
l o g ( R A ) = a + b * M
SS 43 5.16(0.13) 1.12(0.08) 0.28 0.91 5.6 to 8.1 1.3 to 432 R 19 5.00(0.22) 1.22(0.16) 0.28 0.88 5.4 to 7.4 3.3 to 85 N 15 4.86(0.34) 1.32(0.26) 0.34 0.81 5.2 to 7.3 2.5 to 41 All 77 5.08(0.10) 1.16(0.07) 0.28 0.89 5.2 to 8.1 1.3 to 432 SS 43 -3 .55(0 .37) 0.74(0.05) 0.23 0.91 5.6 to 8.1 1.3 to 432 R 19 -2 .86 (0 .55 ) 0.63(0.08) 0.20 0.88 5.4 to 7.4 3.3 to 85 N 15 -2 .01 (0 .65 ) 0.50(0.10) 0.21 0.81 5.2 to 7.3 2.5 to 41 All 77 -3 .22(0 .27) 0.69(0.04) 0.22 0.89 5.2 to 8.1 1,3 to 432 SS 93 4.33(0.06) 1.49(0.05) 0.24 0.96 4.8 to 8.1 1.5 to 350 R 50 4.49(0.11) 1.49(0.09) 0.26 0.93 4.8 to 7.6 1.1 to 80 N 24 4.34(0.23) 1.54(0.18) 0.31 0.88 5.2 to 7.3 3.8 to 63 All 167 4.38(0.06) 1.49(0.04) 0.26 0.94 4.8 to 8.1 1.1 to 350 SS 93 -2 .57 (0 .12 ) 0.62(0.02) 0.15 0,96 4.8 to 8.1 1.5 to 350 R 50 -2 .42 (0 ,21 ) 0.58(0.03) 0.16 0.93 4.8 to 7.6 1.1 to 80 N 24 -1 .88 (0 ,37 ) 0.50(0.06) 0.17 0.88 5.2 to 7.3 3.8 to 63 All 167 -2 .44(0 .11) 0.59(0.02) 0.16 0.94 4.8 to 8,1 1.1 to 350 SS 87 3.80(0.17) 2.59(0.18) 0.45 0.84 4.8 to 8.1 1.5 to 350 R 43 4.37(0.16) 1.95(0.15) 0.32 0.90 4.8 to 7.6 1.1 to 80 N 23 4.04(0.29) 2.11(0.28) 0.31 0.86 5.2 to 7.3 3.8 to 63 All 153 4.06(0.11) 2.25(0.12) 0.41 0.84 4.8 to 8.1 1.1 to 350 SS 87 -0 .76(0 .12) 0.27(0.02) 0.14 0.84 4.8 to 8.1 1,5 to 350 R 43 -1 .61(0 .20) 0.41(0.03) 0.15 0.90 4.8 to 7.6 1.1 to 80 N 23 -1 .14(0 .28) 0.35(0.05) 0.12 0.86 5.2 to 7.3 3.8 to 63 All 153 -1 .01(0 .10) 0.32(0.02) 0.15 0.84 4.8 to 8.1 1.1 to 350 SS 83 3.98(0.07) 1.02(0.03) 0.23 0.96 4.8 to 7.9 3 to 5,184 R 43 4.33(0.12) 0.90(0.05) 0.25 0.94 4.8 to 7.6 2.2 to 2,400 N 22 3.93(0.23) 1.02(0.10) 0.25 0.92 5.2 to 7.3 19 to 900 All 148 4.07(0.06) 0.98(0.03) 0.24 0.95 4.8 to 7.9 2.2 to 5,184 SS 83 -3 .42(0 .18) 0.90(0.03) 0.22 0.96 4.8 to 7.9 3 to 5,184 R 43 -3 .99 (0 .36 ) 0.98(0.06) 0.26 0.94 4.8 to 7.6 2.2 to 2,400 N 22 -2 .87 (0 .50 ) 0.82(0.08) 0.22 0.92 5.2 to 7.3 19 to 900 All 148 -3 .49(0 .16) 0.91(0.03) 0.24 0.95 4.8 to 7.9 2.2 to 5,184
*SRL- - su r f ace rupture length (km); R L D - - s u b s u r f a c e rupture length (kin); R W - - d o w n d i p rupture width (km), R A - - r u p t u r e area (kmZ).
t S S - - s t r i k e slip; R - - r eve r se ; N - - n o r m a l .
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 991
length versus maximum displacement, and surface rup- ture length versus average displacement. We also eval- uate regressions between Ms and displacement; we ob- serve similar trends in correlation coefficients and standard deviations for each slip type.
Analysis of Parameter Correlations
The empirical regressions for all-slip-type relation- ships (Table 2) as well as the data plots (Figs. 9a through 16a) enable us to evaluate the correlations among var- ious rupture parameters. The strongest correlations (r = 0.89 to 0.95) exist between magnitude (M) and surface rupture length, subsurface rupture length, and rupture area. These regressions also have the lowest standard devia- tions (s = 0.24 to 0.28 magnitude units). Magnitude ver- sus displacement relationships have lower correlations (r = 0.75 to 0.78) and higher standard deviations (s = 0.39 to 0.40 magnitude units). Displacement versus length relationships have the weakest correlation (r = 0.71 to 0.75), with standard deviations of 0.36 to 0.41 magni- tude units. These results indicate that displacement and rupture length generally correlate better with magnitude than with each other. The weaker correlations may re- flect the wide range of displacement values (variations as great as 1 1/4 orders of magnitude) observed for rup- tures of the same length (Figs. 12a and 13a).
In general, the relatively high correlations (r > 0.7) and low standard deviations for all the regressions in- dicate there is a strong correlation among the various rupture parameters, and that these regressions may be used confidently to estimate dependent variables.
Because our relationships are based on M rather than
Ms, a quantitative comparison with most regressions cal- culated for previous studies cannot be made. For the sur- face rupture length and maximum displacement regres- sions based on Ms that we calculated during our preliminary analyses, we observed that the correlation coefficients generally were slightly higher, and the stan- dard deviations were lower, than for the regressions cal- culated by Bonilla et al. (1984), Slemmons (1982), Slemmons et al. (1989), and Wesnousky (1986). We also observed that our regressions typically provided similar magnitude estimates to the relationships of Slemmons, and slightly lower magnitude estimates than the rela- tionships of Bonilla et al. (1984). The coefficients for our all-slip-type rupture area regression are similar to the coefficients estimated by Wyss (1979) for an M versus rupture area relationship. Further, because the data sets we use to calculate regressions typically are much larger than the data sets used for previous studies, even qual- itative comparisons among results of different studies are difficult to evaluate.
Effects of Slip Type on Regressions
By comparing the regressions for various slip types (Figs. 9b through 16b), we may evaluate the differences in magnitude or displacement that will result from a given fault parameter as a function of the sense of slip. The sensitivity of the regressions to the sense of slip greatly affects their application, because estimating the sense of slip of a fault may be difficult. If the regressions are insensitive to slip type, such a determination would be unnecessary, and using the all-slip-type regression would be appropriate. A further advantage to using all-slip-type
Table 2B Regressions of Displacement and Moment Magnitude (M)
Coefficients and Standard Correlation Standard Errors
Slip Number of Deviation Coefficient Magnitude Equation* Typet Events a(sa) b(sb) s r Range
Displacement Range (km)
M = a + b * log (MD) SS 43 6.81(0.05) 0.78(0.06) 0.29 0.90 5.6 to 8.1 {R~ 21 6.52(0,11) 0.44(0.26) 0.52 0.36 5.4 to 7.4 N 16 6.61(0.09) 0.71(0.15) 0.34 0.80 5.2 to 7.3 All 80 6.69(0.04) 0.74(0.07) 0.40 0.78 5.2 to 8.1
log (MD) = a + b * M SS 43 -7 .03(0 .55) 1.03(0.08) 0.34 0.90 5.6 to 8.1 {R 21 -1.84(1.14) 0.29(0.17) 0.42 0.36 5.4 to 7.4 N 16 -5 .90(1 .18) 0.89(0.18) 0.38 0.80 5.2 to 7.3 All 80 -5 .46(0 .51) 0.82(0.08) 0.42 0.78 5.2 to 8.1
M = a + b * log (AD) SS 29 7.04(0.05) 0.89(0.09) 0.28 0.89 5.6 to 8.1 {17 15 6.64(0.16) 0.13(0.36) 0.50 0.10 5.8 to 7.4 N 12 6.78(0.12) 0.65(0.25) 0.33 0.64 6.0 to 7.3 All 56 6.93(0.05) 0.82(0.10) 0.39 0.75 5.6 to 8.1
log (AD) = a + b * M SS 29 -6 .32(0 .61) 0.90(0.09) 0.28 0.89 5.6 to 8.1 {R 15 -0.74(1.40) 0.08(0.21) 0.38 0.10 5.8 to 7.4 N 12 -4 .45(1 .59) 0.63(0.24) 0.33 0.64 6.0 to 7.3 All 56 -4 .80(0 .57) 0.69(0.08) 0.36 0.75 5.6 to 8.1
0.01 to 14.6 0.11 to 6.5} 0.06 to 6.1 0.01 to 14.6 0.01 to 14.6 0.11 to 6.5} 0.06 to 6.1 0.01 to 14.6 0.05 to 8.0 0.06 to 1.5} 0.08 to 2 . l 0.05 to 8.0 0.05 to 8.0 0.06 to 1.5} 0.08 to 2.1 0.05 to 8.0
* M D - - m a x i m u m displacement (m); A D - - a v e r a g e displacement (M). t S S - - s t r i k e slip; R - - r eve r se ; N - - n o r m a l . $Regressions for reverse-slip relationships shown in italics and brackets are not significant at a 95% probability level.
992 D . L . Wells and K. J. Coppersmi th
Table 2C Regressions of Surface Rupture Length and Displacement
Equation*
Coefficients and Standard Correlation
Standard Errors Slip Number of Deviation Coefficient Displacement
Type# Events a(sa) b(sb) s r Range (m)
Rupture Length Range (kin)
log (MD) = a + b * log (SRL)
log (SRL) = a + b * log (MD)
log (AD) = a + b * log (SRL)
log (SRL) = a + b * log (AD)
SS 55 -1.69(0.16) 1,16(0.09) 0.36 0.86 0.01 to 14.6 {R$ 21 -0.44(0.34) 0.42(0.23) 0.43 0.38 0.11 to 6.5 N 19 -1.98(0.50) 1.51(0.35) 0.41 0.73 0.06 to 6.4 All 95 -1.38(0.15) 1.02(0.09) 0.41 0.75 0.01 to 14.6 SS 55 1.49(0.04) 0.64(0.05) 0.27 0.86 0.01 to 14.6 {R 21 1.36(0.09) 0.35(0.19) 0.39 0.38 0.11 to 6.5 N 19 1.36(0.05) 0.35(0.08) 0.20 0.73 0.06 to 6.4 All 95 1.43(0.03) 0.56(0.05) 0.31 0,75 0,01 to 14,6 SS 35 -1.70(0.23) 1.04(0.13) 0.32 0.82 0.10 to 8.0 {R 17 -0.60(0.39) 0.31(0.27) 0.40 0.28 0.06 to 2.6 N 14 -1.99(0.72) 1.24(0.49) 0.37 0.59 0.08 to 2.1 All 66 -1.43(0.18) 0.88(0.11) 0.36 0.71 0.06 to 8.0 SS 35 1.68(0.04) 0.65(0.08) 0.26 0.82 0.10 to 8.0 {R 17 1.45(0.10) 0.26(0.23) 0.36 0.28 0.06 to 2.6 N 14 1.52(0.05) 0.28(0.11) 0.17 0.59 0.08 to 2.1 All 66 1.61(0.04) 0.57(0.07) 0.29 0.71 0.06 to 8.0
1.3 to 432 4 to 148} 3.8 to 75 1.3 to 432 1.3 to 432 4 to 148} 3.8 to 75 1.3 to 432 3.8 to 432 6.7 to 148} 15 to 75 3.8 to 432 3.8 to 432 6.7 to 148} 15 to 75 3.8 to 432
*SRL--surface rupture length (km); MD--maximum displacement (m); AD--average displacement (m). §SS--strike slip; R--reverse; N--normal. SRegressions for reverse-slip relationships shown in italics and brackets are not significant at a 95% probability level.
12
7 r-
6 r- ID
E O
5
I 1 I i It I i i i i i I i i I I i i i i i I i i i i i i i
o Strike Slip (a) [] Reverse . . o - ° z~ Normal ~ n ~ . , o
77 EOs %~'.0 ZX Z ~ C ~ ~ s "
M = 5.08 + 1.16*Iog(SRL)
. . . . I , , , 1 , ,
- - + Sfrike Slip - - * Reverse - - ,, Normal
' ' ' ' " 1
J
j " /
i J i i t |
(b)
J
I I I I t ' ' ' 1 , i t t l l t l l I , , ~ , , , R , , , , t , i I t I a I = l t l l , , , , , L
1 0 1 0 0 1 0 3 1 1 0 1 0 0 1 0 3
S u r f a c e Rupfure Lengfh ( k i n ) S u r f a c e R u p f u r e Lengfh ( k m )
Figure 9. (a) Regress ion of surface rupture length on magni tude (M). Regres- sion line shown for all-slip-type relat ionship. Short dashed line indicates 95% confidence interval. (b) Regress ion lines for strike-slip, reverse, and normal-sl ip relationships. See Table 2 for regression coefficients. Length of regression lines shows the range of data for each relat ionship.
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 993
s
"12
7 c u~
c
E o
~ 5
4 10 -2
t t I I I I I I1 I I I I [ I r I I I I
o S t r i k e S l ip
[] Reverse
,~ N o r m a l
8 0 E s o °.xr Q o [ ]
O n ~ , ~ "- 0
~" ° - / . °" o _ ><.-'- 2, []
o S .
I M = 6.69 + 0.74*Iog(MD) I I l l l l l J I J i = = = = = J i J = = = l u l l
10 -1 1 10
. . . . . . . I /
(a) |
o o-I
i Maximum Displacement (m)
, , i r~f,, r i = E Er,=,j
- - + Str ike Slip
- - * Reverse
- - ', N o r m a l
/
i i I i i i , , j
(b)
/
Jr
/
10 -2
I I Z I Z I I z i I I ~ ~ l z z z J I , , , , , , , I
10 -1 1 10
Maximum Displacement (m)
Figure 10. (a) Regression of maximum surface displacement on magnitude (M). Regression line shown for all-slip-type relationship. Short dashed line in- dicates 95% confidence interval. (b) Regression lines for strike-slip, reverse, and normal-slip relationships. See Table 2 for regression coefficients. Length of regression lines shows the range of data for each relationship.
¢1 "0
o m t'- 12)
c- G) E o
9
10 - z
i i
o
[]
A
56
Strike Slip (a)
Reverse Normal o o~
0 . ' ~ EQs o- . [ ] A @," ,, " " " [] 0 ,, ~ " "
°
[] o
M = 6.9,3 + 0 . 8 2 * l o c j ( A D )
10 -1 1 10
Average Displacement (m)
. . . . . . . I . . . . . . . . I
- - + S t r i k e S l i p
- - * R e v e r s e
~ N o r m a l
/ ~ b / I
/
I I [ I I I I I I
(b)
/
0 - 2 l ~ f l i l J J i = = l l l l l J I I I l l l l l ]
10 -1 1 10
Average Displacement (m)
Figure 11. (a) Regression of average surface displacement on magnitude (M). Regression line shown for all-slip-type relationship. Short dashed line indicates 95% confidence interval. (b) Regression lines for strike-slip, reverse, and normal- slip relationships. See Table 2 for regression coefficients. Length of regression lines shows the range of data for each relationship.
994 D.L. Wells and K. J. Coppersmith
regressions is that the range of application for the regres- sions is larger than for single-slip type regressions.
Visually, there is little difference in the position of the regression lines as a function of the sense of slip for surface rupture length, subsurface rupture length, or rup- ture area (Figs. 9b, 15b, and 16b). Other relationships show larger differences between the position of the regression lines (Figs. 10b through 14b). To evaluate the statistical significance of the differences in the results, we use t statistics to compare the regression coefficients for individual slip-type data sets to the coefficients for the rest of the data (i.e., SS to N + R, N to R + SS, and R to SS + N). We also evaluate individual slip re- lationships to each other (SS to R, SS to N, R to N). We use the statistical analysis to evaluate whether regression coefficients differ at high levels of signifi- cance (generally 95%). In some cases, as discussed be- low, we examine the coefficients at higher levels of sig- nificance (e.g., 99%). In the following discussion, the difference between regression coefficients is considered negligible if they are not different at a 95% significance level. The difference between regression coefficients be- comes appreciable if they are different at higher levels of significance.
We observe no difference as a function of slip type at a 95% significance level (i.e., the regression coeffi- cients do not differ at a 95% significance level) for re-
lationships between surface rupture length and magni- tude and subsurface rupture length and magnitude. For these relationships, using the all-slip-type relationship is appropriate because it eliminates the need to assess the type of fault slip. Furthermore, the uncertainty in the mean is smaller for the all-slip-type relationship than for any individual slip-type regression, because the data set is much larger.
For rupture area versus magnitude, we observe no difference in the coefficients of strike slip and normal regressions at a 95% significance level. The reverse regression coefficients differ from normal and strike-slip coefficients at all levels of significance. For downdip rupture width versus magnitude, the coefficients of re- verse and strike-slip regressions differ at all levels of sig- nificance. Normal and strike-slip coefficients, and re- verse and normal coefficients do not differ at 95 to 98% significance. These results indicate that the reverse-slip regression may be most appropriate for estimating mag- nitude, rupture width, or rupture area for reverse-slip faults, whereas the all-slip-type regression may be ap- propriate for other fault types.
We note, however, that even though the regression coefficients may differ at various levels of significance, the actual difference between the expected magnitudes that the regressions provide typically is very small. For example, for an expected rupture area of 100 km 2, strike-
o Strike Slip ~ 9 / / o Reverse oUO , ~ 9 ~ " (a) z~ Normal z ~ e(~,~9/.'" ~o
9s
o :,% OOoOO ,,~ StO A A 0
o []
/ ,'& J
• 0 s
Iog(~D) = -1 .58 + 1.02*IoQ(SRL) I I I , , , L , I , l i L l I K L l I I ' ' ' '
10 100 103
Surface Rupture Length (km)
i i i i i i i i i I I i i i I I I Z I 7 T D I I I I
m
- - + Strike Slip / (b) - - * Reverse *
v ° / - - ~ Normal x / /
I X
/
/ /
I I I I I I I I I I I , . I I I I I I I I = I l a l
10 100 103
Surface Rupture Length ( k m )
Figure 12. (a) Regression of surface rupture length on maximum displace- ment. Regression line shown for all-slip-type relationship. Short dashed line in- dicates 95% confidence interval. (b) Regression lines for strike-slip, reverse, and normal-slip relationships. See Table 2 for regression coefficients. Length of regression lines shows the range of data for each relationship.
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 995
10 E
V
, , , p -
C -
(9 E (9 0 1 12 Q. l/)
a
® 17) 0_ I 0 1
> <
10 - 2
0 []
z~
66
I i i J l l j I I J i i l l ~ l I t 1 I I I I I
. (a) Strike Slip R e v e r s e 0 0~"::~'~" N o r m a l o o . .~ , ,~ - . . .
% oO ,~'4 "'5 EOs
• A • [ ]
log(AD) = - 1 . 4 3 + 0.88*Iog(SRL) I I I I I I I I ] I i i i I I l l I I , I I I I I
10 100 103
Sur face Rupture Length ( k m )
i i i , | i 1 , , . . . . . . . . i . . . . . . . . I k (b) ll - - + Strike Slip
X - - * Reverse / zx N o r m a l /
/ /
x
x
, I t
,It
f
I I I I I I I I J t I I I I l l l J I I I I I il
10 100 103
Sur face Rupture Length ( k m )
Figure 13. (a) Regression of surface rupture length on average displacement. Regression line shown for all-slip-type relationship. Short dashed line indicates 95% confidence interval. (b) Regression lines for strike-slip, reverse, and normal- slip relationships. See Table 2 for regression coefficients. Length of regression lines shows the range of data for each relationship.
9
8
m 7 -0
17) 6 E]
5
4
o S t r i k e Sl ip
o R e v e r s e
zx N o r m a l
167 EQs Z~
' o
J .z o
(a)
M = 4.38 + 1.49*Iog(RLD)
10 100 10 ~
S u b s u r f a c e Rupture Length ( k m )
. . . . . . . . I . . . . . . . . I
- - * Strike Slip - - * R e v e r s e
- - " Normal / s /2"
I T ! I r l r l
(b)
/ x
/
, , , , , , , , I I I I l l l l l J
10 100
Subsu r face Rupture Length
Figure 14. (a) Regression of subsurface rupture length on magnitude (M). Regression line shown for all-slip-type relationship. Short dashed line indicates 95% confidence interval. (b) Regression lines for strike-slip relationships. See Table 2 for regression coefficients. Length of regression lines shows the range of data for each relationship.
i i i i j
10 3
996 D.L . Wells and K. J. Coppersmith
V
® 7
"13 3
" l " O m
E El) 6 13
5
4
i l i i ~ i i i [ , i i
o Str ike Slip (a) [] R e v e r s e
z~ Norma l o
153 EQs ° u 0 0 ~ o . ~
o []
., ;1~2° oOO-
M = 4 . 0 6 + 2 . 2 5 * l o 9 ( W I D )
I I , I I I I I ,I I , I I I
10
Subsurface Rupture Width (kin)
' ' ' ' ' ' ' ' I
- - * Str ike Slip
• R e v e r s e
A Normal
/ ' / / / " / /
i i
(b)
I [ I I I I I I [ i I i
10
Subsurface Rupture Width (kin)
Figure 15. (a) Regression of downdip rupture width on magnitude (M). Regression line shown for all-slip-type relationship. Short dashed line indicates 95% confidence interval. (b) Regression lines for strike-slip, reverse, and normal- slip relationships. See Table 2 for regression coefficients. Length of regression lines shows the range of data for each relationship.
=i V
7
"(3
C O~ 6 o
o S~r{ke Slip (a) [] R e v e r s e
z~ Norma l j~o~, °
148
[] o~c~.~ ~o
M = 4 .07 + 0.98*Iog(RA) I I I I I l l l ~ t I I I I l l l ~ I ~ I ~ 1 1 1 1 t t I I t t
10 100 103 10 4
R u p t u r e A r e a ( k m 2 )
~ i i , i , , I , ~ , , J , , L I
- - + Strike Slip * R e v e r s e
A Normal
, , , , , , q
J f
. J = / / "
/ / "
I t I | l t l l ] ] I I I I 1 | 1 ] I I I I I I l l [
10 100 10 ~
Rupture Area (km 2) Figure 16. (a) Regression of rupture area on magnitude (M). Regression line shown for all-slip-type relationship. Short dashed line indicates 95% confidence interval. (b) Regression lines for strike-slip, reverse, and normal-slip relation- ships. See Table 2 for regression coefficients. Length of regression lines shows the range of data for each relationship.
(b)
/
1 0 4
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 997
slip regressions indicate an expected magnitude of M 6.0, whereas reverse and normal regressions indicate M 6.1 and M 6.0, respectively. For an expected rupture area of 5000 km 2, all regressions indicate an expected mag- nitude of M 7.7 to 7.8. Differences of more than 0.2 magnitude units occur only at magnitudes less than M 5.0. Because the difference in these magnitude estimates is small, the all-slip-type relationship for rupture area versus magnitude is appropriate for most applications. The difference between magnitude estimates for rupture width versus magnitude relationships also is small, thus, the all-slip-type relationship again is preferred for most applications.
In contrast, regressions for displacement relation- ships show larger differences as a function of slip type. Visually, the positions of regression lines for normal and strike-slip data sets vary somewhat for magnitude versus maximum displacement and magnitude versus average displacement relationships (Figs. 10b and 1 lb). Apply- ing t statistics to these relationships shows that strike- slip and dip-slip (normal plus reverse) coefficients differ at all significance levels. Normal-slip coefficients do not differ from strike-slip plus reverse coefficients at a 95% significance level. Because strike-slip relationships are well correlated and have low standard deviations (r => 0.89 and s =< 0.29), using these regressions (magnitude versus maximum or average displacement) may be ap- propriate when the expected slip type is assessed with a high degree of confidence. For situations in which the slip type is uncertain, or for normal and reverse-slip faults, the all-slip-type regression may provide the most reliable results.
Small differences occur in the position of normal and strike-slip regression lines for relationships between dis- placement and surface rupture length (Figs. 12b and 13b). Evaluation of t statistics for displacement versus surface rupture length relationships shows that normal and strike- slip coefficients do not differ at a 95% significance level. Because the strike-slip regression has the highest cor- relation (0.86 and 0.82) and the lowest standard devia- tion (0.36 and 0.32) of the three slip types, for maxi- mum and average displacement regressions, respectively, it may provide the most reliable results when the ex- pected slip type is assessed with a high degree of con- fidence. The all-slip-type relationship may be appropri- ate for other situations.
Effects of Data Selection
We evaluated the relative stability of individual re- lationships with respect to changes in the data set (i.e., addition or deletion of events or changes in the source parameters). We tested the sensitivity of the correlations by removing two data points at random from each data set and recalculating the regression coefficients. Rela- tionships that include more than approximately 14 data points are considered stable because there is no differ-
ence at a 95% significance level between the regression coefficients for both data sets. We consider relationships that are based on fewer than 10 data points to be unsta- ble, because changes in these smaller data sets may pro- duce significant changes in the regression coefficients. We also observe that larger data sets typically have higher correlations and lower standard deviations.
It is interesting to note that although there are far more data points for subsurface rupture length and rup- ture area relationships (for all-slip-type regressions) than for surface rupture relationships, they have only. slightly higher correlation coefficients and slightly lower stan- dard deviations (Table 2). This suggests that these three regressions are very stable and are unlikely to change significantly with additional data. Because the surface and subsurface rupture parameters are measured by dif- ferent techniques, the similar statistical correlation also implies that the variability in the data sets is stochastic in nature, and does not result from errors in measure- ment techniques. It is expected that variable expression of subsurface ruptures at the surface might result in a weaker correlation between surface rupture length and magnitude than between subsurface rupture length and magnitude. However, both relationships are well cor- related and have similar statistical variability.
Effects of Tectonic Setting
Recent studies relate magnitude to rupture length and to displacement and relate seismic moment to rupture length for regions of different geographic setting, tec- tonic setting, or regional crustal attenuation character- istics (e.g., Acharya, 1979; Wesnousky et al., 1983; Bonilla et al., 1984; Nowroozi, 1985; Khromovskikh, 1989; Slemmons et al., 1989; dePolo et al., 1991; John- ston, 1991). One goal of this study is to evaluate whether the tectonic setting of a region might have a greater ef- fect on regressions than does the type of fault slip. The results of Slemmons et al. (1989) suggest that separating data by compressional and extensional settings is insig- nificant for rupture length relationships, but may be sig- nificant for displacement relationships. The data in Ta- ble 1 are separated into compressional and extensional settings, and regression coefficients are calculated for each all-slip-type relationship (excluding average displace- ment). We use t statistics to compare the coefficients (a and b) of extensional and compressional regressions, and we observe no difference between the coefficients at a 95% significance level for any of the relationships. Thus, the difference between the extensional and compres- sional coefficients is insignificant.
Johnston ( 1991) calculated regressions of magnitude versus surface rupture length and magnitude versus max- imum displacement for data from stable continental re- gions (SCR's). His results were not significantly differ- ent from regressions for non-SCR data sets. We also calculate all-slip-type regressions for the SCR earth-
998 D.L. Wells and K. J. Coppersmith
quakes in our data base and compare these results to data from the rest of the world. Because the SCR data sets for surface rupture length and displacement relationships contain only six to seven earthquakes and the correla- tions are low (r < 0.75), these relationships are not sig- nificant at a 95% probability level and are not considered further. Relationships for magnitude versus subsurface rupture length, magnitude versus rupture width, and magnitude versus rupture area comprise 18, 17, and 17 earthquakes, respectively, are well correlated (r > 0.9), and are significant at a 95% probability level. Compar- ing SCR regression coefficients to non-SCR coefficients shows that the rupture area regressions differ at a 95% significance level, whereas the subsurface rupture length and rupture width regression coefficients do not differ at a 95% significance level. We note, however, that the difference in expected magnitudes generally is small (less than 0.2 M) for these regressions (Fig. 17). These results indicate that subdividing our data set according to var- ious tectonic settings or geographic regions does not greatly improve the statistical significance of the regres- sions.
Discussion
The primary purpose of developing regression re- lationships among various earthquake source parameters is to predict an expected value for a dependent parameter from an observed independent parameter. Because we
calculate the regressions by the method of ordinary least squares, the coefficients presented in Table 2 are for es- timating the dependent variable. The independent and dependent variables will depend on the application--either the expected magnitude for a given fault parameter, or the expected fault parameter for a given magnitude. Ta- ble 2 gives the normal and inverted regression coeffi- cients as a function of the sense of slip.
Note that the values of dependent variables derived from these regression formulas are expected values. Thus, the calculated values are expected to be exceeded in 50% of the earthquakes associated with the given value of the independent variable. Bonilla et aI. (1984) discuss tech- niques for evaluating dependent variables at lower ex- ceedance probabilities. In addition, the formulas in Ta- ble 2 are not applicable to values of the independent variable that lie outside the data range listed for each regression.
The empirical relationships presented here can be used to assess maximum earthquake magnitudes for a partic- ular fault zone or an earthquake source. The assumption that a given magnitude is a maximum value is valid only if the input parameter, for instance the rupture length, also is considered a maximum value. For example, sup- pose we are interested in assessing the maximum mag- nitude that a fault is capable of generating, and that we have sufficient data to estimate the possible length and downdip width of future ruptures. Evaluating the seg- mentation of a fault zone (e.g., Schwartz and Copper-
=E 8 V
"13
- . - 7 1- 17)
6 ,dl.- r- O3
E o 5 ~E
4
i I l l l l l [ I I l l l ; l l l I l I ~ I I I I ~
/ I l I I I ' l l l t I I I I I I I I I I I I l l l l l I I I f l u _
10 100 10 + 10 +
Rupture Area (km 2)
Figure 17. Regression lines for stable continental region (SCR) earthquakes and non-SCR continental earthquakes. (a) Regression of surface rupture length on magnitude (M). (b) Regression of rupture area on magnitude (M).
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 999
smith, 1986) provides a basis for assessing the maximum length of future ruptures. The depths of earthquake hy- pocenters, together with the dip of the fault, limit the maximum downdip width of future ruptures. Given that the length and width are assessed to be maximum values, empirical relations between magnitude and rupture length and rupture area will provide the expected maximum magnitudes. These are expected maximum magnitudes for the given maximum fault parameters. However, be- cause there is dispersion associated with the statistical relations, both higher and lower magnitudes are possible for any single event having the given rupture parameters. The standard deviation for each regression provides a measure of that dispersion.
Regarding regressions between magnitude and sub- surface rupture length and rupture area, previous studies indicate that the size and depth of the earthquake, as well as the nature of near-surface materials, have a significant effect on whether the subsurface rupture is partly or fully expressed by faulting at the surface (e.g., Amaike, 1987; Berberian and Papastamatiou, 1978; Bernard and Zollo, 1989; Bonilla, 1988). In addition, the absence of surface rupture during some large-magnitude earthquakes (greater than M 7), and the occun~nce of surface rupture for some smaller-magnitude earthquakes (less than M 5.5), show that there are large variations in rupture at the surface. Thus, variation in the geologic conditions and the hy- pocentral depths of future earthquakes will have uncer- tain effects on the extent of future surface ruptures. In contrast, subsurface rupture length and rupture area, which are estimated from the spacial distribution of after- shocks, are not subject to these uncertainties. For ex- ample, in the subsurface, earthquakes typically appear to rupture individual fault segments, and the segment boundaries are defined at the surface by various geo- metric, structural, or geologic features (Knuepfer, 1989). During some earthquakes, however, even though an en- tire segment ruptures in the subsurface, the rupture may not propagate over the full length of the segment at the ground surface. Thus, we believe that subsurface rupture length regressions are appropriate for estimating mag- nitudes for expected ruptures along single or multiple fault segments. Where the extent of previous ruptures at the surface can be evaluated, however, surface rupture length regressions are appropriate for estimating ex- pected magnitudes. Applying subsurface rupture length and rupture area relations to estimating magnitudes may help to overcome uncertainties associated with estimat- ing the surface rupture length for some seismic sources.
The regressions for subsurface rupture length and rupture area also provide a basis for estimating the mag- nitudes of earthquakes that may occur on subsurface seismic sources such as blind thrust faults, which cannot be evaluated from surface observations. Furthermore, regressions on subsurface parameters include data for moderate-magnitude earthquakes (in the range of mag-
nitude 5 to 6), allowing the characterization of relatively small seismic sources that may not rupture the surface.
The use of empirical regressions to assess maximum magnitudes typically involves developing several mag- nitude estimates from which a maximum magnitude value is selected or an uncertainty distribution is constructed. Various segmentation models have been proposed to de- fine the reaches of a fault zone that are relatively con- tinuous and behave similarly (Schwartz and Copper- smith, 1986; Schwartz, 1988). Estimates of the possible lengths of future ruptures involve considering the pos- sibilities that one or more of these segments might rup- ture. Alternative rupture scenarios and associated rupture lengths result in multiple estimates of earthquake mag- nitude using a single regression relationship, such as sur- face rupture length versus magnitude or subsurface rup- ture length versus magnitude. Further, if the downdip geometry of a fault zone is known, the rupture width and rupture area relationships provide additional magnitude estimates. Detailed geologic studies along a fault zone can result in estimates of the maximum and average dis- placement associated with individual paleoseismic events along the fault zone. These displacement estimates also may be used with the appropriate regressions to assess expected magnitudes. Ultimately, developing a maxi- mum magnitude estimate involves judging which rupture scenarios are most credible, which rupture parameters (e.g., rupture length, area, and displacement) represent maximum parameters, and the relative preference for the various regressions (perhaps based on the dispersion as- sociated with each regression). For probabilistic seismic hazard analyses, these considerations and estimates may be combined into a probabilistic distribution of the max- imum magnitude (Coppersmith, 1991).
In addition to assessing maximum magnitudes, the regressions presented in this study have other potential engineering applications. For example, seismic design criteria for facilities such as pipelines and tunnels require estimates of the amount of displacement that might occur where the facility crosses a fault. The regressions of dis- placement on magnitude provide the expected values for a given earthquake magnitude. In particular, the average displacement regression provides the mean displacement along the length of a rupture, and the maximum displace- ment regression provides the expected largest slip at a point along a rupture. In most applications, the average displacement is desired because it is unknown, prior to a rupture event, whether the facility lies at the point where the maximum displacement will occur. The maximum displacement regression might be used to provide a con- servative upper bound for engineering design.
Conclusions
The data base reveals that surface rupture length typ- ically is equal to 75% of the subsurface rupture length,
1000 D.L. Wells and K. J. Coppersmith
and the average surface displacement typically is equal to one-half of the maximum surface displacement. The ratio of surface rupture length to subsurface rupture length increases slightly as magnitude (M) increases. There is no apparent relationship between the ratio of average displacement to maximum displacement and magnitude (M). We calculate the average subsurface displacement on the fault plane from the rupture area and the seismic moment; this is more than the average displacement and less than the maximum displacement measured at the surface. Thus, for many earthquakes in our data base, most slip on the fault plane at seismogenic depths prop- agates to the surface. We also note that there is no sys- tematic difference between Ms and M for the events in the data base over the range of magnitude 5.7 to 8.0. However, Ms is systematically smaller than M for mag- nitudes less than 5.7.
The empirical regressions show a strong correlation between magnitude and various rupture parameters, which enables us confidently to use these relationships to es- timate magnitudes or rupture parameters. The regres- sions between magnitude and surface rupture length, subsurface rupture length, downdip rupture width, and rupture area are well determined in most cases, having correlation coefficients of about 0.84 to 0.95 and stan- dard deviations of about 0.24 to 0.41 magnitude units. Relationships between displacement and rupture length or magnitude are less well correlated (correlation coef- ficient about 0.71 to 0.78).
In most cases, the empirical regressions do not vary significantly as a function of the sense of slip. The t statistics show that the regression coefficients are not different at high significance levels for regressions be- tween magnitude and surface rupture length, and mag- nitude and subsurface rupture length. Relationships be- tween magnitude and rupture area, and magnitude and rupture width, are different at a 95% significance level. The regression coefficients are similar, however, and differences in parameters estimated from these regres- sions typically are small. This conclusion suggests that the all-slip-type regression may be used for most situa- tions, and is especially significant for evaluating ex- pected magnitudes for poorly known faults or blind faults that lack clear surface expression. The regressions of displacement versus magnitude show a mild dependency on the sense of slip in some cases; however, these re- lationships have the weakest statistical correlations.
Analysis of data sets of various sizes shows that regressions containing approximately 14 or more data points are insensitive to changes in the data. Smaller data sets (less than 10 to 14 data points) generally are sen- sitive to changes in the data, and correlations may not be significant. The regressions for subsurface rupture length and rupture area are based on the largest data sets, yet show statistical correlations similar to those of the smaller data set for surface rupture length regressions.
This suggests that the relationships based on large data sets (more than 50 earthquakes) are unlikely to change significantly with the addition of new data.
In evaluating dependency of the relationships on tec- tonic setting we compare the coefficients (a and b) of extensional and compressional regressions for each re- lationship using t statistics. We observed no difference between the coefficients at a 95% significance level for any of the relationships; thus, the difference between the extensional and compressional coefficients is small. We calculate all-slip-type regressions for the SCR earth- quakes in our data base and compare these results to data from the rest of the world. Comparing SCR regression coefficients to non-SCR coefficients shows that the rup- ture area regressions differ at a 95% significance level, whereas the subsurface rupture length regressions do not differ at this significance level. These results indicate that subdividing the data set according to various tec- tonic settings or geographic regions occasionally may provide slightlY different results, but typically does not improve the statistical significance of the regressions.
Because of the larger number of data and good sta- tistical correlations, we believe that the all-slip-type regressions are appropriate for most applications of these regressions. The use of the regressions for subsurface rupture length and rupture area may be appropriate where it is difficult to estimate the near-surface behavior of faults, such as for buried or blind faults. Reliable estimates of the maximum expected magnitude for faults should in- clude consideration of multiple estimates of the expected magnitude derived from various rupture parameters.
Acknowledgments
We thank Pacific Gas & Electric Company (San Francisco) for fi- nancial support of this study. Dr. Robert Youngs (Geomatrix Con- sultants) provided extensive assistance in the statistical analysis of the data sets. Dr. David Burton Slemmons and Mr. Zhang Xiaoyi (Uni- versity of Nevada, Rent) contributed the preliminary results of their studies of average surface displacements for historical earthquakes. We are grateful for the expertise Dr. Slemmons provided in evaluating the surface rupture lengths and displacements for many of the earth- quakes in the data base. We also wish to thank Dr. William Savage and Dr. Janet Cluff (Pacific Gas & Electric Company), Dr. Slem- mons, and an anonymous reviewer for careful evaluations of drafts of this article.
References
Abe, K. (1981). Magnitudes of large shallow earthquakes from 1904- 1980, Phys. Earth Planet. Interiors 27, 72-92.
Abe, K. and S. Noguchi (1983a). Determination of magnitude for large shallow earthquakes 1898-1917, Phys. Earth Planet. In- teriors 32, 45-59.
Abe, K. and S. Noguchi (1983b). Revision of magnitudes of large shallow earthquakes, 1897-1912, Phys. Earth Planet. Interiors 33, 1-11.
Acharya, H. K. (1979). Regional variations in the rupture-length
Empirical Relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displacement 1001
magnitude relationships and their dynamical significance, Bull. Seism. Soc. Am. 69, 2063-2084.
Aki, K. and P. G. Richards (1980). Quantitative Seismology, Volume 1, W. H. Freeman, San Francisco, 512 pp.
Albee, A. L. and J. L. Smith (1966). Earthquake characteristics and fault activity in southern California, in Engineering Geology in Southern California, R. Lung and D. W. Proctor (Editors), As- sociation of Engineering Geologists, Los Angeles Section, 9 - 34.
Amaike, F. (1987). Seismic explorations of the buried fault associated with the 1948 Fukui earthquake, J. Phys. Earth 35, 285-308.
Ambraseys, N. N. (1975). Studies in historical seismicity and tec- tonics, in Geodynamics Today, The Royal Society, London, 7 - 16.
Ambraseys, N. N. (1988). Engineering seismology, Earthquake Eng. Struct. Dyn. 17, 1-105.
Benjamin, J. R. and C. A. Coruell (1970). Probability, Statistics, and Decision for Civil Engineers, McGraw-Hill, New York, 684 pp.
Berberian, M. and D. Papastamatiou (1978). Khurgu (north Bandar Abbas, Iran) earthquake of 21 March 1977--a preliminary field report and a seismotectonic discussion, Bull. Seism. Soc. Am. 68, 411-428.
Bernard, P. and A. Zollo (1989). The Irpinia (Italy) 1980 earth- quake--detailed analysis of a complex normal faulting, J. Geo- phys. Res. 94, 1631-1647.
Bonilla, M. G. (1988). Minimum earthquake magnitude associated with coseismic surface faulting, Bull. Assoc. Eng. Geologists 25, 17-29.
Bonilla, M. G. and J. M. Buchanon (1970). Interim report on world- wide historic surface faulting, U.S. Geol. Surv. Open-File Rept. 70-34, 32 pp.
Bonilla, M. G., R. K. Mark, and J. J. Lienkaemper (1984). Statistical relations among earthquake magnitude, surface rupture length, and surface fault displacement, Bull. Seism. Soc. Am. 74, 2379- 2411.
Boore, D. M. and W. B. Joyner (1982). The empirical prediction of ground motion, Bull. Seism. Soc. Am. 72, $43-$60.
Chinnery, M. A. (1969). Earthquake magnitude and source parame- ters, Bull. Seism. Soc. Am. 59, 1969--1982.
Clark, M. M. (1972). Surface rupture along the Coyote Creek fault, in The Borrego Mountain Earthquake of April 9, 1968, U.S. Geol. Surv. Profess. Pap. 787, 55-86.
Coppersmith, K. J. (1991). Seismic source characterization for en- gineering seismic hazard analysis, in Proc. 4th International Conference on Seismic Zonation, Vol. I, Earthquake Engineer- ing Research Institute, Oakland, California, 3-60.
Darragh, R. B. and B. A. Bolt (1987). A comment on the statistical regression relation between earthquake magnitude and fault rup- ture length, Bull. Seism. Soc. Am. 77, 1479-1484.
Davis, J. C. (1986). Statistics and Data Analysis in Geology, Second Ed., Wiley, New York, 646 pp.
dePolo, C. M., D. G. Clark, D. B. Slemmons, and A. R. Ramelli (1991). Historical surface faulting in the Basin and Range prov- ince, western North America: implications for fault segmenta- tion, J. Struct. Geol. 13, 123-136.
Dietz, L. D. and W. L. Ellsworth (1990). The October 17, 1989, Loma Prieta, California, earthquake and its aftershocks: geom- etry of the sequence from high-resolution locations, Geophys. Res. Lett. 17, 1417-1420.
Duda, S. J. (1965). Secular seismic energy release in the circum- Pacific belt, Tectonophysics 2, 409-452.
Dziewonski, A. M., T.-A. Chow, and J. H. Woodhouse (1981). De- termination of earthquake source parameters from waveform data for studies of global and regional seismicity, J. Geophys. Res. 86, 2825-2852.
Electric Power Research Institute (1987). Seismic hazard methodol- ogy for the central and eastern United States--Volume 1: Meth-
odology, Report NP-4726, prepared for Seismicity Owners Group and Electric Power Research Institute under research projects P101- 38, -45, -46, 2256-14.
Gubbins, D. (1990). Seismology and Plate Tectonics, Cambridge University Press, Cambridge, England, 339 pp.
Gutenberg, B. (1945). Amplitudes of surface waves and magnitudes of shallow earthquakes, Bull. Seism. Soc. Am. 34, 2-12.
Gutenberg, B. and C. F. Richter (1954). Seismicity of the Earth and Associated Phenomena, Second Ed., Princeton University Press, Princeton, New Jersey, 310 pp.
Hanks, T. C. and H. Kanamori (1979). A moment-magnitude scale, J. Geophys. Res. 84, 2348-2350.
Hanks, T. C. and M. Wyss (1972). The use of body-wave spectra in the determination of seismic-source parameters, Bull. Seism. Soc. Am. 62, 561-589.
Hanks, T. C., J. A. Hileman, and W. Thatcher (1975). Seismic mo- ments of the larger earthquakes of the southern California re- gion, Geol. Soc. Am. Bull. 86, 1131-1139.
Iida, K. (1959). Earthquake energy and earthquake fault, Nagoya University, J. Earth Sci. 7, 98-107.
Johnston, A. C. (1991). Surface rupture in stable continental regions, EOS 72, 489.
Johnston, A. C. and L. R. Kanter (1990). Earthquakes in stable con- tinental crust, Scientific American 262, 68-75.
Kanamori, H. (1983). Magnitude scale and quantification of earth- quakes, Tectonophysics 93, 185-199.
Kanamori, H. and D. L. Anderson (1975). Theoretical basis of some empirical relations in seismology, Bull. Seism. Soc. Am. 65, 1073- 1096.
Khromovskikh, V. S. (1989). Determination of magnitudes of ancient earthquakes from dimensions of observed seismodislocations. Tectonophysics 166, 269-280.
Knuepfer, P. L. K. (1989). Implications of the characteristics of end- points of historical surface fault ruptures for the nature of fault segmentation, in Proc. of Conf. XLV, Fault Segmentation and Controls of Rupture Initiation and Termination, D. P. Schwartz and R. H. Sibson (Editors), U.S. Geol. Surv. Open-File Rept. 89-315, 193-228.
Lee, W. H. K., F. T. Wu, and S. C. Wang (1978). A catalog of instrumentally determined earthquakes in China (magnitude > 6) compiled from various sources, Bull. Seism. Soc. Am. 68, 383-398.
Lienkaemper, J. J. (1984). Comparison of two surface-wave magni- tude scales--M of Gutenberg and Richter (1954) and Ms of "pre- liminary determination of epicenters," Bull. Seism. Soc. Am. 74, 2357-2378.
Mendoza, C. and S. H. Hartzell (1988). Aftershock patterns and main shock faulting, Bull. Seism. Soc. Am. 78, 1438-1449.
Nowroozi, A. A. (1985). Empirical relations between magnitudes and fault parameters for earthquakes in Iran, Bull. Seism. Soc. Am. 75, 1327-1338.
Ohnaka, M. (1978). Earthquake-source parameters related to mag- nitude, Geophys. J. R. Astr. Soc. 55, 45-66.
Panza, G. F., S. J. Duda, L. Cernobori, and M. Herak (1989). Gu- tenberg's surface-wave magnitude calibrating function: theoret- ical basis from synthetic seismograms, Tectonophysics 166, 35- 43.
Purcaru, G. and H. Berckhemer (1982). Quantitative relations of seis- mic source parameters and a classification of earthquakes, in Quantification of Earthquakes, S. J. Duda and K. Aki (Editors), Tectonophysics 84, 57-128.
Richter, C. F. (1958). Elementary Seismology, W. II. Freeman, San Francisco, 768 pp.
Rothe, J. P. (1969). The Seismicity of the Earth, 1953-1965. UNESCO, Paris.
Scholz, C. H. (1982). Scaling laws for large earthquakes: conse- quences for physical models, Bull. Seism. Soc. Am. 72, 1-14.
1002 D . L . Wells and K. J. Coppersmith
Schwartz, D. P. (1988). Geology and seismic hazards: moving into the 1990's, in Earthquake Engineering Soil Dynamics I f - -Re- cent Advances in Ground Motion Evaluation. Vol. 20, J. L. Van Thun (Editor), American Society of Civil Engineers Geotech- nical Special Publication, New York, 1-42.
Schwartz, D. P. and K. J. Coppersmith (1986). Seismic hazards-- new trends in analysis using geologic data, in Active Tectonics, National Academy Press, Washington, D.C., 215-230.
Sibson, R. H. (1987). Effects of thult heterogeneity on rupture prop- agation, in Proc. of Conf. XXX1X, Directions in Paleoseismol- ogy, A. J. Crone and E. M. Omdahl (Editors), U.S. Geol. Surv. Open-File Rept. 87-673, 362-373.
Singh, S. K., E. Bazan, and L. Esteva (1980). Expected earthquake magnitude from a fault, Bull. Seism. Soc. Am. 70, 903-914.
Slermaaons, D. B. (1977). Faults and earthquake magnitude, U.S. Army Corps of Engineers, Waterways Experimental Station, Miscel- laneous Papers S-73-1, Report 6, 1-129.
Slemmons, D. B. (1982). Determination of design earthquake mag- nitudes for microzonation, Proc. of the Third International Earthquake Microzonation Conf. Vol. 1, U.S. National Science Foundation, Washington, D.C., 119-130.
Slemmons, D. B., P. Bodin, and X. Zang (1989). Determination of earthquake size from surface faulting events, Proc. of the Inter- national Seminar on Seismic Zonation, Guangzhou, China, State Seismological Bureau, Beijing, 13.
Tocher, D. (1958). Earthquake energy and ground breakage, Bull. Seism. Soc. Am. 48, 147-153.
Troutman, B. M. and G. P. Williams (1987). Fitting straight lines in the earth sciences, in Use and Abuse of Statistical Methods in the Earth Sciences, W. B. Size (Editor), Oxford University Press, New York, 107-128.
Utsu, T. (1969). Aftershocks and earthquake statistics (I), some pa- rameters which characterize an aftershock sequence and their interrelations, J. Faculty Sci., Series VII, Vol. III, Hokkaido University, Japan, 129-195.
Utsu, T. and A. Seki (1954). A relation between the area of after- shock region and the energy of main-shock, J. Seism. Soc. Ja- pan 7, 233-240.
Wesnousky, S. G. (1986). Earthquakes, Quaternary faults, and seis- mic hazards in California, J. Geophys. Res. 91, 12587-12631.
Wesnousky, S. G., C. H. Scholz, K. Shimazaki, and T. Matsuda (1983). Earthquake frequency distribution and mechanics of faulting, J. Geophys. Res. 88, 9331-9340.
Wu, F. T. (1968). Parkfield earthquake of 28 June 1966--magnitude and source mechanism, Bull. Seism. Soc. Am. 58, 689-709.
Wyss, M. (1979). Estimating maximum expectable magnitude of earthquakes from fault dimensions, Geology 7, 336-340.
Geomatrix Consultants, Inc. San Francisco, California 94111
51. Am6ra5ey5, 1970, 1975, 1988; Am6ra5ey5 and 2at0pek, 1969; 8arka and Kad1n5ky-Cade, 1988; Ey1d09an, 1988
52. Chen and M01nar, 1977; F10ren50v and 5010nenk0, 1965; Knuepfer, 1989; M01nar and Den9, 1984; 0ka1, 1992; 0ka1, 1976; 7app0n1er and M01nar, 1979
53. And0, 1977; 8en-Menahem, 1977, 1978; 8en-Menachem and 70k502, 1963; Kanam0r1, 1977; Ke11eher and 5av1n0, 1975; N15henk0 and Jac06, 1990; 0ka1, 1992; P1afker and 0ther5, 1978; 5tauder, 1960; 70cher, 1960; Ut5u, 1962
54. 8arr1ent05 and 0ther5, 1987; D05er, 1985; D05er and 5m1th, 1989; Ha11 and 5a610ck, 1985; Knuepfer, 1989; Meyer5 and Ham11t0n, 1964; 5ava9e and Ha5t1e, 1966; 5tewart and 0ther5, 1964
55. D05er and 5m1th, 1989; Wa11ace and 0ther5, 1981; We5taway and 5m1th, 1989; We5taway and 0ther5, 1989
56. Am6ra5ey5, 1963, 1975; Am6ra5ey5 and Me1v111e, 1982; M0hajer and P1erce, 1963; N0wr0021, 1985; Petre5cu and Purcaru, 1964
57. A6e, 19746; Ut5u, 1969 58. Arn6ra5ey5, 1975; 8a1ak1na and 0ther5, 1968; N0rth, 1977; 5h1r0k0va, 1968 59. 801t and Herra12, 1983; Evan5 and McEv111y, 1982; Ud1a5, 1965; Ut5u, 1969 60. A6e, 1975; Ak1, 1966; 80yd and 0ther5, 19847M091 and 0ther5, 1964; M0r1 and 80yd,
1985; Nakamura and 0ther5, 1964; 5atake and A6e, 1983; 75u60kawa and 0ther5, 1964 61. McEv111y, 1966; Ut5u, 1969 62. McEv111y and Ca5aday, 1967; Ut5u, 1969 63. Archu1eta and Day, 1980; 8r0wn and 0ther5, 1967; 8r0wn and Vedder, 1967; Eat0n and
0ther5, 1970; L1ndh and 800re, 1981; 7r1funac and Udwad1a, 1974; 75a1 and Ak1, 1969; Wa11ace and R0th, 1967; Wu, 1968
64. Ara6a52 1991; 80ucher and 0ther5, 1967; L1e6ermann and P0mer0y, 1970; Pa9e, 1968 65. Am6ra5ey5, 1975, 1988; Am6ra5ey5 and 2at0pek, 1968; 8arka and Kad1n5ky-Cade, 1988;
Kud0, 1983; N0rth, 1977; Wa11ace, 1968 66. D05er and 5m1th, 1989; 6reen5fe1der, 1968; Hehn6er9er and En9en, 1980; Kachad00r1an
and 0ther5, 1967; Rya11 and 0ther5, 1968; 75a1 and Ak1, 1970; Wa11ace and 0ther5, 1981 67. Chen and M01nar, 1977; Huan9 and Chen, 1986; M01nar and Den9, 1984; M05kv1na,
1978; 0ka1, 1976 68. Am6ra5ey5, 1970, 1975, 1988; Am6ra5ey5 and 2at0pek, 1969; 8arka and Kad1n5ky-Cade,
1988; Ey1d09an, 1988; Hank5 and Wy55, 1972; Kad1n5ky-Cade and 8arka, 1989; Kud0, 1983; N0rth, 1977; Ut5u, 1969
69. Am6ra5ey5, 1975; N0rth, 1977; 5u15tar0va and K0c1aj, 1980 70. N0rth, 1977; Pav11de5 and 7ran05, 1991; 7ayma2 and 0ther5, 1991
8-3
71. A11en and N0rd4u15t, 1972; 8urd1ck and Me11man, 1976; 8urf0rd, 1972; 8ut1er, 1983; C1ark, 1972; E6e1 and He1m6er9er, 1982; Ham11t0n, 1972; Hank5 and Wy55, 1972; Heat0n and He1m6er9er, 1977; K1kuch1 and Kanam0r1, 1986; Peter50n and 0ther5, 1991; Wy55 and Hank5, 1972a
72. Adam5 and 0ther5, 1971; Adam5 and L0wry, 1971; 8erryman, 1984; 8ev1n and 0ther5, 1984; D0wr1ck, 1991; Len5en and 0tway, 1971; R061n50n and 0ther5, 1975; 5hepherd and 0ther5, 1970
73. Am6ra5ey5 and Me1v111e, 1982; Am6ra5ey5 and 7cha1enk0, 1969; 8ayer and 0ther5, 1969; Cramp1n, 1969; Han1~5 and Wy55, 1972; Jack50n and F1tch, 1979; McEv111y and N1a21, 1975; N1a21, 1968; N0rth, 1977; N0wr0021, 1985; 7cha1enk0 and 8er6er1an, 1975; 7cha1enk0 and Am6ra5ey5, 1970
74. Denham and 0ther5, 1980; Fredr1ch and 0ther5, 1988; 60rd0n, 1971; 60rd0n and Lew15, 1980; Lan95t0n, 1987; V09fj0rd and Lan95t0n, 1987
75. 6edney and 0ther5, 1969; Huan9 and 815wa5, 1983 76. Am6ra5ey5, 1975, 1988; Am6ra5ey5 and 7cha1enk0, 1972; Arpat and 81n901, 1969;
Ey1d09an and Jack50n, 1985; Jack50n and F1tch, 1979; Kud0, 1983; N0rth, 1977; We5taway, 1990
77. Peter50n and 0ther5, 1991; 5ander5 and Kanam0r1, 1984; 7hatcher and Ham11t0n, 1973 78. De2a, 1971; Lander, 1969; Ph111p and Me9ard, 1977; 5e6r1er and 0ther5, 1988; 5uare2
and 0ther5, 1983 79. 8rant1ey and Chun9, 1991 80. 1ma9awa and 0ther5, 1984; M1kum0, 1973a 81. 6reen and 810ch, 1971; 6reen and Mc6arr, 1972; Maa5ha and M01nar, 1972; 5hud0f5ky,
1985; 50merv111e, 1986; Wa9ner and Lan95t0n, 1988, 1989 82. De2a, 1971; Lander, 1969a, 19696; Ph111p and Me9ard, 1977; 5e6r1er and 0ther5, 1988;
5uare2 and 0ther5, 1983 83. 6an and 0ther5, 1978; 6e0det1c 5urvey 8r19ade, 1975; M01nar and Den9, 1984; Wan9
and 0ther5, 1978; 2han9 and Lu1, 1978; 2h0u and 0ther5, 1983a 84. Am6ra5ey5, 1975, 1988; Am6ra5ey5 and 7cha1enk0, 1972; Ey1d09an and Jack50n, 1985;
Jack50n and F1tch, 1979; Kud0, 1983; N0rth, 1977; 7a5dem1r091u, 1971; We5taway, 1990 85. Ha5e9awa and 0ther5, 1975; M1kum0, 1974 86. A11en and 0ther5, 1973, 1975; Can1te2 and 70k502, 1972; Hank5, 1974; Heat0n and
He1m6er9er, 1979; Heat0n, 1982; Kam6 and 0ther5, 1971; Lan95t0n, 1978; M1kum0, 79736; 5ava9e and 0ther5, 1975; 5harp, 1975, 1981; 7r1funac, 1974; U.5. 6e01091ca1 5urvey 5taff, 1971; Wy55 and Hank5, 19726
87. Am6ra5ey5, 1975, 1988; Ke19ht1ey, 1975; Kud0, 1983; 5eymen and Ayd1n, 1972 88. E115w0rth, 1975; J0hn50n and McEv111y, 1974; Kur1ta, 1976 89. E115w0rth, 1975; J0hn50n and McEv111y, 1974 90. Am6ra5ey5, 1975; Am6ra5ey5 and 0ther5, 1972; Am6ra5ey5 and Me1v111e, 1982; Dewey
and 6rant2, 1973; Jack50n and F1tch, 1979, 1981; N0rth, 1977; 5ava9e and 0ther5, 1977; 5060ut1 and 0ther5, 1972
8-4
91.
92. 93. 94. 95.
96.
97.
98.
99. 100.
101. 102.
103. 104.
105.
106. 107. 108. 109.
110. 111.
112.
113.
114.
Ke11eher and 5av1n0, 1975; Lander, 1973; N15henk0 and Jac06, 1990; Pa9e, 1973; Pere2 and Jac06, 1980; 5che11 and Ruff, 1986, 1989 Jack50n and Y1e1d1n9, 1983 J0hn50n and McEv111y, 1974; Kur1ta, 1976; We550n and E115w0rth, 1972 8akun, 1984; J0hn50n and McEv111y, 1974; Kur1ta, 1976; We550n, 1987 8r0wn and 0ther5, 1973; Dewey and 0ther5, 1973; Lan9er and 0ther5, 1974; Matum0t0 and Latham, 1973; P1afker and 8r0wn, 1973; Ward and 0ther5, 1974 A11en and 0ther5, 1991; 8eck, 1989; M01nar and Den9, 1984; Q1an, 1986; 7an9 and 0ther5, 1976; 7an9 and 0ther5, 1984; 2h0u and 0ther5, 1983a, 19836 8ent and He1m6er9er, 19916; 800re and 5t1erman, 1975, 1976; Ca5t1e and 0ther5, 1977; E115w0rth and 0ther5, 1973; 5t1erman and E115w0rth, 1976 M01nar and Den9, 1984; M01nar and Chen, 1983; 51n9h and 6upta, 1979; 51n9h and 0ther5, 1978 A11150n and 0ther5, 1978 A6e, 1978; Mat5uda and Yama5h1na, 1974; 0hnaka, 1978; 7ake0, 1989; 2akhar0va and 0ther5, 1978 A6e, 1978 Jack50n and 0ther5, 1979; Lan95t0n and Dermen91an, 1981; Ne150n and 0ther5, 1986; N1 and 6uan9we1, 1989; 2akhar0va and 0ther5, 1978 J0hn50n and Had1ey, 1976; 5harp, 1976 Chun9 and 8rant1ey, 1989; C1par, 1979; 6e0det1c 5urvey 8r19ade, 1978; 6u and 0ther5, 1976; J0ne5 and 0ther5, 1982; L1n and 0ther5, 1979; M01nar and Den9, 1984; Q1an9 and 2han9, 1984; Ra1e19h, 1977; 5tewart and 0ther5, 1976; Wu and 0ther5, 1976; 2akhar0va and 0ther5, 1978 Ara6a52 and 0ther5, 1981; 8ache and 0ther5, 1980; D05er and 5m1th, 1989; Wa11ace and 0ther5, 1981; W1111am5, 1979 Hatanaka and 7ake0, 1989; Hatanaka and 5h1ma2ak1, 1988; Mura1 and Mat5uda, 1975 Fu15, 1976; H111 and 8ee6y, 1977; Knuepfer, 1989 8ache and 0ther5, 1980; D05er and 5m1th, 1989; P1tt~and 0ther5, 1979 8ufe and 0ther5, 1976; C1ark and 0ther5, 19•16; Hart and Harp5ter, 1978; Hart and Rapp, 1975; Hart and 0ther5, 1977; Lahr and 0ther5, 1976; Lan95t0n and 8ut1er, 1976; Le5ter and 0ther5, 1975; Rya11 and Van W0rmer, 1975; 5ava9e and 0ther5, 1977 Franke1, 1984; Hart2e11 and 8rune, 1979 Am6ra5ey5, 1988; Arpat, 1977; Ey1d09an, 1980; Jack50n andMcKen21e, 1984; Kud0, 1983; Na6e1ek and 70k502, 1978a; 70k502~and Arpat, 1977 8ucknam and 0ther5, 1978; Dewey and Ju11an, 1976; Kanam0r1 and 5tewart, 1978; K1kuch1 and Kanam0r1, 1982; Lan9er and 80111n9er, 1979; L150w5k1 and 7hatcher, 1981; P1afker, 1976; P1afker and 0ther5, 1976; Y0un9 and 0ther5, 1989 Ey1d09an and 0ther5, 1985; Hart2e11, 1980; Kre5tn1k0v and 0ther5, 1980; Kr15ty and 0ther5, 1980; 5hteyn6er9 and 0ther5, 1980 Amat0 and 0ther5, 1976; 8r101e and 0ther5, 1986; Ca9nett1 and Pa54ua1e, 1•979; C1par, 1980, 1981; F1nett1 and 0ther5, 1979; Mart1n15, 1976; 70kuyama, 1976
8-5
115.
116.
117. 118. 119. 120. 121.
122. 123.
124. 125. 126. 127.
128. ~
129. 130. 131.
132.
133.
134. 135.
136. 137. 138.
139.
Ey1d09an and 0ther5, 1985; Hart2e11, 1980; Kre5tn1k0v and 0ther5, 1980; Kr15ty and 0ther5, 1980; 5hteyn6er9 and 0ther5, 1980 8ut1er and 0ther5, 1979; Chan9, 1979; Chen and 0ther5, 1979; Chen and 0ther5, 1988; Jenn1n95, 1980; K1kuch1 and Kanam0r1, 1986; M01nar and Den9, 1984; Na6e1ek and 0ther5, 1987; Q1an9 and 2han9, 1984; 5hed10ck and 0ther5, 1987; Wu and 0ther5, 1981; X1e and Ya0, 1991; Y0n9 and 0ther5, 1988; 2han9 and 0ther5, 1980; 2h0u, 1987 J0ne5 and 0ther5, 1984; M01nar and Den9, 1984 A6e, 1978 J0ne5 and 0ther5, 1984; M01nar and Den9, 1984 J0ne5 and 0ther5, 1984; Ke19ht1ey, 1975; M01nar and Den9, 1984 Am6ra5ey5, 1988; 8arka and Kad1n5ky-Cade, 1988; 6u1kan and 0ther5, 1978; K1kuch1 and Kanam0r1, 1986; Kud0, 1983; Na6e1ek and 70k502, 19786; 70k502 and 0ther5, 1977, 1978 60n2a1e2 and 0ther5, 1984; Nava and 8rune, 1983 8er6er1an and Papa5tamat10u, 1978; 8er6er1an and 0ther5, 1977; Jack50n and F1tch, 1981; N0wr0021 and M0hajer A5hja1, 1985 R1chard50n, 1989 Carver and 0ther5, 1978; Carver and 0ther5, 1981; Carver and 0ther5, 1983 Warren and 0ther5, 1978, 1985 8arker, 1993; Ca5tan0, 1982; Kad1n5ky-Cade, 1985; Kad1n5ky-Cade and 0ther5, 1985; Lan9er and 80111n9er, 1988 Am6ra5ey5 and 0ther5, 1979; Am6ra5ey5 and Me1v111e, 1982; 8er6er1an and 0ther5, 1979; N0wr0021 and M0hajer-A5hja1, 1985; 20h00r1an 12adpanah and 0ther5, 1981 K1kuch1 and 5ud0, 1984; 5ack5 and 0ther5, 1981; 5h1ma2ak1 and 50merv111e, 1979 Ye11n and Cr0550n, 1982 8arker and Lan95t0n, 1981; 8ru5t1e and Mu11er, 1983; Karaka1515 and M1kuma, 1993; Ku1hanek and Meyer, 1979; Merc1er and 0ther5, 1979; Merc1er and 0ther5, 1983; Papa2ach05 and 0ther5, 1979; 50uf1er15 and 5tewart, 1981; 50uf1er15 and 0ther5, 1982 8ent and He1m6er9er, 19916; C0r6ett and J0hn50n, 1982; Lee and 0ther5, 1978; Wa11ace and 0ther5, 1981; Wh1tc0m6 and Hutt0n, 1978 8ru5t1e and Mu11er, 1983; Hae551er and 0ther5, 1980; 5cher6aum and 5t011, 1983; 7urn0v5ky and 5chne1der, 1982 50merv111e and 0ther5, 1980 Am6ra5ey5 and Me1v111e, 1982; 8er6er1an, 1979, 1982; 8er6er1an and 0ther5, 1979; Hart2e11 and Mend02a, 1991; N1a21 and 5h0ja-7aher1, 1985; N1a21 and Kanam0r1, 1981; N0wr0021 and M0hajer-A5hja1, 1985; 5harp and 0ther5, 1978 Pepp1n and 0ther5, 1989 Hauk550n and 5a1d1var, 1986 Ek5tr0m and D21ew0n5k1, 1985; H111 and 0ther5, 1980; Hutt0n and 0ther5, 1980; 5te1n and L150w5k1, 1983 800re and 0ther5, 1981; 8ru5t1e and Mu11er, 1983; C0n501e and Fava11, 1981; Kanam0r1 and 61ven, 1981; 75e1ent15 and 0ther5, 1988
8-6
140. 141.
142. 143. 144.
145.
146.
147. 148.
149. 150. 151.
152.
153. 154. 155. 156.
157.
158.
159. 160.
Denham and 0ther5, 1987; Fredr1ch and 0ther5, 1988; Lew15 and 0ther5, 1981 Arm5tr0n9, 1979; 80uch0n, 1982; Ek5tr0m and D21ew0n5k1, 1985; Herd and 0ther5, 1979; K1n9 and 0ther5, 1981; Lee and 0ther5, 1979; Lu1 and He1m6er9er, 1983; Rea5en6er9 and E115w0rth, 1982; Uhrhammer, 1980 Ha5e9awa and Wetm111er, 1980 De5champ5 and 0ther5, 1984 Archu1eta, 1982; Archu1eta, 1984; D05er and Kanam0r1, 1986; E5p1n05a, 1982; Hart2e11 and Heat0n, 1983; Hart2e11 and Hehn6er9er, 1982; J0hn50n and Hutt0n, 1982; Kanam0r1 and Re9an, 1982; 0150n and Ap5e1, 1982; Re111n9er and Lar50n, 1986; 5harp, 1982; 5harp and 0ther5, 1982; 511ver t~nd Ma5uda, 1985 Ha9h1p0ur and Am1d1, 1980; N1a21 and Kanam0r1, 1981; N0wr0021 and M0hajer-A5hja1, 1980, 1985 Ha9h1p0ur and Am1d1, 1980; N1a21 and Kanam0r1, 1981; N0wr0021 and M0hajer-A5hja1, 1980, 1985 Marr0w and R06ert5, 1985 80atwr19ht and 800re, 1982; 801t and 0ther5, 1981; 80n111a and 0ther5, 1980; Ek5tr0m and D21ew0n5k1, 1985; 5che1mer and 0ther5, 1982 Franke1, 1984; 5ander5 and Kanam0r1, 1984 6a9nepa1n-8eyne1x and 0ther5, 1982 8arker and Lan95t0n, 1983; Cramer and 70pp02ada, 1980; Ek5tr0m and D21ew0n5k1, 1985; 61ven and 0ther5, 1982; Ju11an and 51pk1n, 1985; L1de and Rya11, 1985; Uhrha1nmer and Fer9u50n, 1980 Ander50n and 8rune, 1991; Ander50n and 51m0n5, 1982; Ek5tr0m and D21ew0n5k1, 1985; Mun9u1a and 8rune, 1984; Nakan15h1 and Kanam0r1, 1984; 5harp, 1981; 511ver and Ma5uda, 1985; W0n9 and Fre2, 1982 15h1da, 1984; L1nde and 0ther5, 1982; Mat5uura, 1983; 7ake0, 1988 Am6ra5ey5 and Jack50n, 1990; Papa2ach05 and 0ther5, 1983 Hermann and 0ther5, 1982; Mauk and 0ther5, 1982; 50merv111e, 1986 Am6ra5ey5, 1981; C15terna5 and 0ther5, 1982; De5champ5 and 0ther5, 1982; K1n9 and Y1e1d1n9, 1984; K1n9 and V1ta-F1n21, 1981; Na6e1ek, 1985; 0uyed and 0ther5, 1981; 0uyed and 0ther5, 1983; Ph1111p and Me9hra0u1, 1983; Rue99 and 0ther5, 1982; F. 5wan, per5. c0mm. 1992; Y1e1d1n9, 1985; Y1e1d1n9 and 0ther5, 1981 Amat0 and 0ther5, 1989; 8ernard and 20110, 1989; 8ru5t1e and Mu11er, 1983; Cr0550n and 0ther5, 1986; De1 Pe220 and 0ther5, 1983; De5champ5 and K1n9, 1983; De5champ5 and K1n9, 1984; Pant05t1 and Va1en515e, 1990; Vaccar1 and 0ther5, 1990; We5taway, 1987; We5taway and Jack50n, 1984, 1987 Den9 and 2han9, 1984; M01nar and Ly0n-Caen, 1989; M01nar and Den9, 1984; Q1an, 1986; 7an9 and 0ther5, 1984a; 7an9 and 0ther5, 19846; 2h0u and 0ther5, 19836 6rant and 0ther5, 1984 8e22e9h0ud and 0ther5, 1986; Jack50n and 0ther5, 1982; K1m and 0ther5, 1984; K1n9 and 0ther5, 1985; 5tavrakak15 and 0ther5, 1991
8-7
161.
162.
163.
164.
165.
166. 167. 168. 169. 170.
171. 172.
173. 174.
175. 176. 177.
178.
179. 180. 181.
182. 183.
8e22e9h0ud and 0ther5, 1986; Jack50n and 0ther5, 1982; K1m and 0ther5, 1984; K1n9 and 0ther5, 1985; 5tavrakak15 and 0ther5, 1991 8e22e9h0ud and 0ther5, 1986; Jack50n and 0ther5, 1982; K1m and 0ther5, 1984; K1n9 and 0ther5, 1985; 5tavrakak15 and 0ther5, 1991 8er6er1an and 0ther5, 1984; 6he1tanch1 and 0ther5, 1990; N0wr0021 and M0hajer-A5hja1, 1985; 51ev1n and Wa11ace, 1986 8er6er1an and 0ther5, 1984; 6he1tanch1 and 0ther5, 1990; N0wr0021 and M0hajer-A5hja1, 1985; 51ev1n and Wa11ace, 1986 Ch0y and 0ther5, 1983; N9uyen and Herrman, 1992; 50merv111e, 1986; 5uare2 and Na6e1ek, 1983; Wetm111er and 0ther5, 1984 Franke1, 1984; 5ander5 and Kanam0r1, 1984 Ek5tr0m and D21ew0n5k1, 1985; 5te1n and Ek5tr0m, 1992 Ch0y and K1nd, 1987; 1.an9er and 0ther5, 1987; P1afker and 0ther5, 1987; 51pk1n, 1986 L0mn1t2 and Ha5h12ume, 1985 Ch0y, 1990, 51pk1n and Needham, 1990; -Eat0n, 1984; Eat0n, 1990; E6erhart-Ph1111p5 and Rea50n6er9, 1990; Ek5tr0m and D21ew0n5k1, 1985; Feh1er and J0hn50n, 1989; Hank5 and 800re, 1984; Hart2e11 and Heat0n, 1983; Kanam0r1, 1983; Mc6arr and 0ther5, 1990; R1a1 and 8r0wn, 1983; 5her6urne and 0ther5, 1983; 5te1n, 1985; 7an1m0t0 and Kanam0r1, 1986; Uhrhammer and 0ther5, 1984 Chen and Wan9, 1984 801t and Herra12, 1983; Eat0n, 1990; Eat0n and 0ther5, 1985; E6erhart-Ph1111p5 and Rea50n6er9, 1990; Hart and McJunk1n, 1983; Rymer and 0ther5, 1985; Uhrhammer and 0ther5, 1984; Na6e1ek and 5uare2, 1989; N9uyen and Herrman, 1992 8arr1ent05 and 0ther5, 1985; 8arr1ent05 and 0ther5, 1987; 80atwr19ht, 1985; Cr0ne and 0ther5, 1987; D05er and 5m1th, 1985; Ek5tr0m and D21ew0n5k1, 1985; R1ch1n5, 1985; 5te1n and 8arr1ent05, 1985a, 19856; 7an1m0t0 and Kanam0r1, 1986 Am6ra5ey5, 1988; 8arka and Kad1n5ky-Cade, 1988; 151am1, 1986; L1 and 0ther5, 1987 Ah0rner and Pe121n9, 1985; A5p1na11 and K1n9, 1985; Came16eec~k and De8ecker, 1985 Am6ra5ey5 and Adam5, 1986; D0r6ath and 0ther5, 1984; Jen5en and 0ther5, 1989; Lan9er and 0ther5, 1987; 5u1e1man and 0ther5, 1993 8akun and 0ther5, 1984a; 8akun and 0ther5, 19846; 8er02a and 5pud1ch, 1988; C0ckerham and Eat0n, 1985, 1987; Ek5tr0m, 1984; 61adw1n and J0hn5t0n, 1986; Hart2e11 and Heat0n, 1986; H005e, 1987; Pre5c0tt and 0ther5, 1984a, 19846; Uhrhammer an(1 Darra9h, 1984 Hae551er and 0ther5, 1988 K0nd0r5kaya and 0ther5, 1989; We5taway and 0ther5, 1989 An5e11 and 0ther5, 1986; Marr0w and Wa1ker, 1988; 7r0dd and 0ther5, 1985; 7ur61tt and 0ther5, 1985 Lahr and 0ther5, 1986 M120ue and 0ther5, 1985; 7ake0 and M1kam1, 1987; 7ake0, 1987; Yama5h1na and 7ada, 1985
8-8
184. 185.
186. 187. 188. 189. 190. 191.
192. 193.
194. 195. 196. 197. 198.
199. 200.
201.
202. 203.
204. 205. 206.
207. 208.
Lan9er and 0ther5, 1991; N9uyen and Herrman, 1992 8arker and Wa11ace, 1986; D05er and 5m1th, 1989; 6r055 and 5ava9e, 1985; J0hn5t0n and 0ther5, 1987; Pr1e5t1ey and 0ther5, 1988 Ca5tan0, 1985; 1NPRE5, 1985 M011 and 0ther5, 1987 M0r1, 1989 Eat0n, 1985; Ek5tr0m, 1986; Ek5tr0m and 0ther5, 1992; Ek5tr0m and 5te1n, 1989 Ka15er and Duda, 1988; K0nd0r5kaya and 0ther5, 1989; N1 and 6uan9we1, 1989 8arker, 1989; Ch0y and 80atwr19ht, 1988; Ha5e9awa and 0ther5, 1989; H0rner and 0ther5, 1989; H0rner and 0ther5, 1990; K0nd0r5kaya and 0ther5, •989; Wetm111er and 0ther5, 1988 80un1f and 0ther5, 1987; De5champ5 and 0ther5, 1991 8arker, 1989; Ch0y and 80atwr19ht, 1988; Ha5e9awa and 0ther5, 1989; H0rner and 0ther5, 1989; H0mer and 0ther5, 1990; K0nd0r5kaya and 0ther5, 1989; Wetm111er and 0ther5, 1988 51mp50n and 0ther5, 1988; Wy55 and Ha6ermann, 1988 61a55m0yer and 80rcherdt, 1990; N1ch0150n and 0ther5, 1988; N9uyen and Herrman, 1992 R09er5 and 0ther5, 1990 Fredr1ch and 0ther5, 1988; Machette and 0ther5, 1993; McCue and 0ther5, 1987 801t and Uhrhammer, 1986; 0ppenhe1mer and Mac6re90r-5c0tt, 1991; 2h0u and 0ther5, 1989; 2h0u and McNa11y, 1990; 2h0u and 0ther5, 1993 Ca6rera and 0ther5, 1991; Merc1er and 0ther5, 1992; Yeat5 and 0ther5, 1994 Chen and Wan9, 1986, 1988; Chen and 0ther5, 1988; Hwan9 and Kanam0r1, 1989; L1aw and 0ther5, 1986; Pe220pane and We5n0u5ky, 1989; 5a126er9 and 0ther5, 1988; 5h1n and 0ther5, 1989; Wu and 0ther5, 1989; Yeh and 0ther5, 1990; Yu and Lu1, 1986 Hart2e11, 1989; J0ne5 and 0ther5, 1986; L150w5k1 and 6r055, 1987; Mend02a and Hart2e11, 1988, N1ch0150n and 0ther5, 1987; Pachec0 and Na6e1ek, 1988; 5ee6er and 0ther5, 1987 Hauk550n and J0ne5, 1988; Pachec0 and Na6e1ek, 1988 C0ckerham and C0r6ett, 1987; deP010 and 0ther5, 1991; deP010 and Rame111, 1987; D05er and 5m1th, 1989; 6r055 and 5ava9e, 1987; J0hn5t0n and 0ther5, 1987; Kah1e and 0ther5, 1986; Knuepfer, 1989; L1enkaemper and 0ther5, 1987; Pachec0 and Na6e1ek, 1988; Pre5c0tt and 0ther5, 1988; 5m1th and Pr1e5t1ey, 1987 Ly0n-Caen and 0ther5, 1988; Papa2ach05 and 0ther5, 1988 Har10w and 0ther5, 1993; Rymer, 1987; Wh1te and 0ther5, 1987 Chen and Wan9, 1988; Chen and 0ther5, 1988; 601d5te1n and Archu1eta, 1991; Hwan9 and Kanam0r1, 1989; Kanam0r1, 1988; Pe220pane and We5n0u5ky, 1989; 5a126er9 and 0ther5, 1988; Wu and 0ther5, 1989 75ukuda and 0ther5, 1989 60n2a1e2-6arc1a, 1991
8-9
209.
210. 211. 212. 213. 214.
215.
216.
217.
218.
219.
220. 221.
222.
223.
224. 225.
226.
Ander50n and 0ther5, 1990; Ander50n and We66, 1989; 8ean1and and 0ther5, 1989; 8ean1and and 0ther5, 1990; Dar6y, 1989; 6rape5, 1987; New 2ea1and Department 0f 5c1ent1f1c and 1ndu5tr1a1 Re5earch, 1987; Pender and R06ert50n, 1988; 2han9 and 0ther5, 1989 Maeda, 1991 Lan9er and 80111n9er, 1991; 7ay10r and 0ther5, 1989 Le1 and 0ther5, 1991; We1 and Chun9, 1993 Pechman and 0ther5, 1992 8arker, 1988; 8ent and Hehn6er9er, 1989; 801t and 0ther5, 1989; Hart/e11 and 11da, 1990; Hauk550n and J0ne5, 1989; Hauk550n an(1 0ther5, 1988; L1n and 5te1n, 1989; L1nde and J0hn5t0n, 1989 A9new and Wyatt, 1989; 8ent and 0ther5, 1989; Hudnut and 0ther5, 1989; L150w5k1 and 5ava9e, 1988; Ma915tra1e and 0ther5, 1989; 5harp and 0ther5, 1989; 51pk1n, 1989 A9new and Wyatt, 1989; 8ent and 0ther5, 1988; 8udd1n9 and 5harp, 1988; Hudnut and 0ther5, 1989; Kah1e and 0ther5, 1988; L150w5k1 and 5ava9e, 1988; Ma915tra1e and 0ther5, 1989; Mc6111 and 0ther5, 1989; 5harp and 0ther5, 1989; W1111am5 and Ma915tra1e, 1989 80wman, 1991; 80wman and 0ther5, 1990; Ch0y and 80wman, 1990; Chun9 and.0ther5, 1988; Cr0ne and 0ther5, 1992; J0hn5t0n, 1988; McCaffrey, 1989 80wman, 1991; 80wman and 0ther5, 1990; Ch0y and 80wman, 1990; Chun9 and 0ther5, 1988; Cr0ne and 0ther5, 1992; J0hn5t0n, 1988; McCaffrey, 1989 80wman, 1991; 80wman and 0ther5, 1990; Ch0y and 80wman, 1990; Chun9 and 0ther5, 1988; Cr0ne and 0ther5, 1992; J0hn5t0n, 1988; McCaffrey, 1989 Nava and 0ther5, 1989; Pechman and 0ther5, 1990, 1992 Chen and Q1n, 1991; Chen and Wu, 1989; H01t and Wa11ace, 1989; 1n5t1tute 0f Earth4uake En91neer1n9, 1989; L1 and Na6e1ek, 1989; Ma0 and 2han9, 1991; Wan9 and 0ther5, 1989; Wu, 1989; Yu and 0ther5, 1991 Chen and Q1n, 1991; 1n5t1tute 0f ~trth4uake En91neer1n9, 1989; L1 and Na6c1ek, 1989; Ma0 and 2han9, 1991 ; 2h0u and 0ther5, 1990 Cara6aja1 and 8arker, 1991; Du 8er9er and 0ther5, 1991; N0rth and 0ther5, 1989; 50merv111e and 0ther5, 1990; Wetm111er and 0ther5, 1989 J0ne5 and 0ther5, 1990; Kanam0r1, 1989; Kanam0r1 and 0ther5, 1990 Aref1ev and 0ther5, 1989; 80mmer and Am6ra5ey5, 1989; 80rcher,.1t and 0ther5, 1990; C15terna5 and 0ther5, 1989a, 19896; D0r6ath and 0ther5, 1992; Hae551er and 0ther5, 1989; J1mene2 and 0ther5, 1989; Kanam0r1, 1993; Lan9er and 0the~r5, 1989; Needham and 51pk1n 1989; Pachec0 and 0ther5, 1989; Ph111p and 0ther5, 1989; 5harp, 1989 Pechnaan and 0ther5, 1990, 1992
8-10
227.
228. 229.
230. 231. 232.
233. 234.
235. 236. 237. 238.
239.
240.
241. 242.
243. 244.
8arker and 5a126er9, 1990; Ch0y and 80atwr19ht, 1990; D1et2 and E115W0rth, 1990; D21ew0n5k1 and 2wart, 1990; Kanam0r1 and He1m6er9er, 1990; Kanam0r1 and 5atake, 1990; Lan95t0n and 0ther5, 1990; L150w5k1 and 0ther5, 1990; McNa11y and 0ther5, 1989; M1chae1 and 0ther5, 1990; Na6e1ek, 1990; P1aNer and 6a110way, 1989; Pre5c0tt and 0ther5, 1990; R0man0w1C2 and Ly0n-Caen, 1990; Ruff and 71Che1aar, 1990; 5a126er9 and 0ther5, 1990; 50merv111e and Y05h1mura, 1990; Uhrhammer and 0ther5, 1990; 2han9 and Lay, 1990 Am6ra5ey5 and 0ther5, 1990; Me9hra0u1, 1991 Adam5 and 0ther5, 1991; Adam5 and 0ther5, 1990; 8ent, 1993; Wetm111er and 0ther5, 1991 Fukuyama and M1kuma, 1993 Dre9er and Hehn6er9er, 1991a; Hauk550n and J0ne5, 1991a; Hutt0n, 19906 8er6er1an and 0ther5, 1992; N1a21 and 8020r9n1a, 1992; 7h10 and 0ther5, 1990; 75ukuda and 0ther5, 1991 A6e, 1990; 5harp and Um6a1, 1990; 7h10 and 0ther5, 1990; Y05h1da and A6e, 1990, 1992 deP010 and H0rt0n, 1991; Dre9er and 0ther5, 1991; H0rt0n and deP010, 1992; McNutt and 0ther5, 1991 75ukuda and 0ther5, 1992 Dre9er and He1m6er9er, 19916; Hauk550n and J0ne5, 19916; Wa1d and 0ther5, 1991 McLaren and 5ava9e, 1992; M. McLaren, per5. c0mm. 1993 8arka and Ey1d09an, 1993; 8ennett and 0ther5, 1992; EER1 1993; 7r1f0n0v and 0ther5, 1993 Hauk550n and 0ther5, 1992; Hauk550n and 0ther5, 1993; H0u9h and 0ther5 (1993, 1n rev1ew); N1ch0150n and 0ther5, 1993; Rymer, 1992 8erryman, 1992; Camp1110 and Archu1eta, 1992; Dre9er and He1m6er9er, 1992; Hart and 0ther5, 1993; Hauk550n and 0ther5, 1992; Hauk550n and 0ther5, 1993; H0u9h and 0ther5 1992; Kanam0r1 and 0ther5, 1992; 51eh and 0ther5, 1993 Hauk550n and 0ther5, 1992; Hauk550n and 0ther5 1993; J0ne5 and He1m6er9er, 1993 Ander50n and 0ther5, 1992; Harm0n, per5. c0mm. 1993; 5m1th and 0ther5, 1993; 5heehan and 0ther5, 1993; 2ha0 and Hehn6er9er 1993 Mad1n and 0ther5, 1993; J. Na6e1ek, per5. c0mm. 1993 Hauk550n and 0ther5, 1993; 5. Hecker, per5. c0mm. 1993; J. 5c0tt, per5. c0mm. 1993
Am6ra5ey5, N.N., E1na5ha1, A.5., 80mmer, J.J., Haddar, F., Mada5, P., E19ha20u11, A., and J. V09t (1990). 7he Chen0ua (A19er1a) earth4uake 0f 29 0ct06er 1989, En91neer1n9 5e15m0109Y and Earth4uake En91neer1n9 Re5earch Rep0rt N0. 90-4, 1mper1a1 C011e9e 0f 5c1ence and 7echn0109y, L0nd0n.
Ander50n, H., 5m1th, E., and R. R061n50n (1990). N0rma1 fau1t1n9 1n a 6ack-arc 6a51n-- 5e15m01091ca1 character15t1c5 0f the 1987 March 2 Ed9ecum6e, New 2ea1and, earth4uake (a65.), E05, 71, n0.2, 51-52.
Ander50n, H., and 7. We66 (1989). 7he rupture pr0ce55 0f the 1987 Ed9ecum6e earth4uake, New 2ea1and, New 2ea1and J. 6e01. 6e0phy5., 32, 43-52.
Ander50n, J.6., and J.N. 8rune (1991). 7he V1ct0r1a acce1er09ram f0r the 1980 Mex1ca11 Va11ey earth4uake, Earth4. 5pectra, 7, 29-43.
Ander50n, J.6., and R.5.51m0n5, ed5. (1982). 7he Mex1ca11 Va11ey earth4uake0f 9 June 1980: Earth4uake En91neer1n9 Re5earch 1n5t1tute New51etter 16, 24 p.
An5e11, J., A5p1na11, W., K1n9, 6 . , and R. We5taway (1986). 7he 1984 Ju1y 19 N0rth Wa1e5 earth4uake - a 10wer cru5ta1 c0nt1nent event 1nd1ca1h19 6r1tt1e 6ehav10r at an unu5ua1 depth, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 84, 201-206.
Ara6a52, W.J. (1991). A 5yn0p515 0f the 1966 Ca11ente/C10ver M0unta1n5, Nevada, earth4uake: Unpu6115hed paper, 5upp1ementary data f0r E1ectr1c P0wer Re5earch 1n5t1tute-H19h Leve1 Wa5te Perf0rmance A55e55ment Pr0ject, 26 p.
Archu1eta, R.J. (1984). A fau1t1n9 m0de1 f0r the 1979 1mper1a1 Va11ey earth4uake, J. 6e0phy5. Re5., 89, 4,559-4,585.
Archu1eta, R.J., and 5.M. Day (1980). Dynam1c rupture 1n a 1ayered med1um--the 1966 Parkf1e1d earth4uake, 8u11. 5e15m. 50c. Ant., 70, 671-689.
C-3
Aref1ev, 5.5., 80r1550ff, 8.A., and R.E. 7atev05yan (1989). 50me feature5 0f the ep1centra1 area 0f 5p1tak, Decem6er 7, 1988 earth4uake, 1n 5chenk, v., and 5chenk0va, 2., ed5., Pr0ceed1n95 0f the 4th h~ternat10na1 5ymp051um 0n the Ana1y515 0f 5e15m1c1ty and 5e15m1c R15k, 8echnye Ca5t1e, C2ech0510vak1a, 6e0phy51ca1 1n5t1tute, C2ech0510vak Academy 0f 5c1ence5, Pra9ue, 49-56.
Arm1j0, R., 7app0nn1er, P., and H. 70n911n (1989). Late Cen0201c r19ht-1atera1 5tr1ke-511p fau1t1n9 1n 50uthern 716et, J. 6e0phy5. Re5., 94, 2787-2838.
8a1ak1na, L.M., Ru5tan0v1ch, D.N., and D. Kh0d2h1yev5k1y (1968). 7he f0ca1 mechan15m 0f the after5h0ck5 0f the earth4uake 0f Ju1y 26, 1963, at 5k0pje, h,e5t1a, Phy51c5 0fthe 5011d Earth, 1, 110-114.
8arka, A., and H. Ey1d09an (1993). 7he Er21ncan earth4uake 0f 13 March 1992 1n ea5tern 7urkey, 7erra N0va, 5, 190-194.
8arka, A., 70k502, M.N., Kad1n5ky-Cade, K., and L. 6u1en (1987). 7he 5e9mentat10n, 5e15m1c1ty and earth4uake p0tent1a1 0f the ea5tern part 0f the n0rth Anat011an fau1t 20ne: 5u6m1tted t0 J. 6e0phy5. Re5., 34 p.
8arka, A., and K. Kad1n5ky-Cade (1988). 5tr1ke-511p fau1t 9e0metry 1n 7urkey and 1t5 1nf1uence 0n earth4uake act1v1ty, 7ect0n1c5, 7, 663-684.
8arker, J.5., and C.A. Lan95t0n (1981). 1nver510n 0f te1e5e15m1c 60dy wave5 f0r the m0ment ten50r 0f the 1978 7he55a10n1k1, 6reece, earth4uake, 8u11. 5e15m. 50c. Am., 71, 1423-1444.
8arker, J.5., and C.A. Lan95t0n (1983). A te1e5e15m1c 60dy-wave ana1y515 0f the May 1980 Mamm0th Lake5, Ca11f0rn1a, earth4uake5, 8u11. 5e15m. 50c. Am., 73, 419-434.
8arker, J.5., and D.H. 5a126er9 (1990). L0n9-per10d and 6r0ad-6and te1e5e15m1c 60dy-wave m0de11n9 0f the 0ct06er 18, 1989 L0ma Pr1eta earth4uake, 6e0phy5. Re5. Letter5, 17, 1409- 1412.
8arker, J.5., and 7.C. Wa11ace (1986). A n0te 0n the te1e5e15m1c 60dy wave5 fr0m the 23 N0vem6er 1984 R0und Va11ey, Ca11f0rn1a, earth4uake, 8u11. 5e15m. 50c. Am., 76, 883-888.
8arr1ent05, 5.E., 5te1n, R.5., and 5.N. Ward (1987). C0mpar150n 0f the 1959 He69en Lake, M0ntana, and the 1983 80rah Peak, 1dah0, earth4uake5 fr0m 9e0det1c 065ervat10n5, 8u11. 5e15m. 50c. Am., 77, 784-808.
8arr1ent05, 5.E., Ward, 5.N., 60n2a1e2-Ru12, J.R., and R.5.5te1n (1985). 1nver510n f0r m0ment a5 a funct10n 0f depth fr0m 9e0det1c 065ervat10n5 and 10n9 per10d 60dy wave5 0f the 1983 80rah Peak, 1dah0 earth4uake, 1n 5te1n, R.5., and 8ucknam, R.C., ed5., Pr0ceed1n95 0f W0rk5h0p XXV111, 0n the 80rah Peak, 1dah0, Earth4uake, U.5. 6e01. 5ur. 0pen-F11e Rep0rt 85-290, 485-518.
8ayer, K.C., Keuckr0th, L.E., and R.A. Kar1m (1969). An 1nve5t19at10n 0f the Da5ht-e 8aya2, 1ran, earth4uake 0f Au9u5t 31, 1968, 8u11. 5e15m. 50c. Am., 59, 1793-1822.
8ean1and, 5., 8erryman, K.R., and 6.H. 811ck (1989). 6e01091ca1 1nve5t19at10n5 0f the 1987 Ed9ecum6e earth4uake, New 2ea1and, New 2ea1and J. 6e01. 6e0phy5., 32, 73-91.
8ean1and, 5., 811ck, 6.H., and D.J. Dar6y (1990). N0rma1 fau1t1n9 1n a 6ack arc 6a51n: 9e01091ca1 and 9e0det1c character15t1c5 0f the 1987 Ed9ecum6e earth4uake, New 2ea1and, J. 6e0phy5. Re5., 95, 4693-4707.
8ean1and, 5., and M.M. C1ark (1987). 7he 0wen5 Va11ey fau1t 20ne, ea5tern Ca11f0rn1a, and 5urface rupture a550c1ated w1th the 1872 earth4uake (a65.), 5e15m. Re5. Letter5, 58, 32.
8er6er1an, M. (1979). Earth4uake fau1t1n9 and 0edd1n9 thru5t a550c1ated w1th the 7a6a5-E- 6015han (1ran) earth4uake 0f 5eptem6er 16, 1978, 8u11. 5e15m. 50c. Am., 69, 1861-1887.
8er6er1an, M. (1982). After5h0ck tect0n1c5 0f the 1978 7a6a5-e-6015han (1ran) earth4uake 5e4uence--a d0cumented act1ve ~th1n- and th1~k-5k1nned tect0n1c• ca5e, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 68, 499-530.
8er6er1an, M., and D. Papa5tamat10u (1978). Khur9u (n0rth 8andar A66a5, 1ran) earth4uake 0f March 21, 1977--a pre11m1nary f1e1d rep0rt and a 5e15m0tect0n1c d15cu5510n, 8u11. 5e15m. 50c. Am., 68, 411-428.
8er6er1an, M., A5udeh, 1., 811ham, R.6., 5ch012, C.H., and C. 50uf1er15 (1979). Mechan15m 0f the ma1n 5h0ck and the after5h0ck 5tudy 0f the 7a6a5-E-6015han (1ran) earth4uake 0f 5eptem6er 16, 1978--a pre11m1nary rep0rt, 8td1. 5e15m. 50c. Ant., 69, 1851-1859.
8er6er1an, M., Jack50n, J.A., 6h0ra5h1, M., and M.H. Kadjar (1984). F1e1d and te1e5e15m1c 065ervat10n5 0f the 1981 6016af-51rch earth4uake5 1n 5E 1ran, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 77, 809-838.
8er6er1an, M., Papa5tamat10u, D., and M. Q0ra15h1 (1977). Khur9u (n0rth 8andar A66a5, 1ran) earth4uake 0f March 21, 1977, 1n 8er6er1an, M., ed., C0ntr16ut10n5 t0 the 5e15m0tect0n1c5 0f1ran (Part 111), 6e01. M1n1n9 5ur. h~an, Rep0rt N0. 40, 7-49.
8ernard, P., and A. 20110 (1989). 7he 1rp1n1a (1ta1y) 1980 earth4uake--deta11ed ana1y515 0f a c0mp1ex n0rma1 fau1t1n9, J. 6e0phy5. Re5., 94, 1631-1647.
8er02a, 6.C., and P. 5pud1ch (1988), L1near12ed 1nver510n f0r fau1t rupture 6ehav10r: app11cat10n t0 the 1984 M0r9an H111, Ca11f0rn1a, earth4uake, J. 6e0phy5. Re5., 93, 6275-6296,
C-6
8erryman, K.R. (1984). Late Quaternary tect0n1c5 1n New 2ea1and, 1n Wa1c0tt, R.1., c0mp11er, An 1ntr0duct10n t0 the Recent Cru5ta1 M0vement5 0f New 2ea1and, R0ya1 50c. New 2ea1and M15c. 5er1e5 7, 91-107.
8ev1n, A.J., 0tway, P.M., and P.R. W00d (1984). 6e0det1c m0n1t0r1n9 0f cru5ta1 def0rmat10n 1n New 2ea1and, 1n Wa1c0tt, R.1., c0mp11er, An 1ntr0duct10n t0 the Recent Cru5ta1 M0vement5 0f New 2ea1and, R0ya1 50c. New 2ea1and M15c. 5er1e5 7, 13-60.
8e22e9h0ud, M., De5champ5, A., and R. Madar1a9a (1986). 8r0ad-6and m0de111n9 0f the C0r1nth, 6reece earth4uake5 0f Fe6ruary and March 1981, Anna1e5 6e0phy51cae, 4, n0.83, 295-304.
80atwr19ht, J. (1985). Character15t1c5 0f the after5h0ck 5e4uence 0f the 80rah Peak, 1dah0, earth4uake determ1ned fr0m d191ta1 rec0rd1n95 0f the event5, 8u11. 5e15m. 50c. Am., 75, 1265- 1284.
80atwr19ht, J., and D.M. 800re (1982). Ana1y515 0f the 9r0und acce1erat10n5 rad1ated 6y the 1980 L1verm0re Va11ey earth4uake5 f0r d1rect1v1ty and dynam1c 50urce character15t1c5, 8u11. 5e15m. 50c. Am. , 72, 1843-1865.
801t, 8.A. (1967). 5e15m01091ca1 n0te5--j0tt1n95 fr0m Japan, the 7an90, N061, N119ata and Mat5u5h1r0 earth4uake5 and t11e N1kar1 tra1n, 8u11. 5e15n1. 50c. Am., 57, 133-138.
80rcherdt, R.D., Lan9er, C., F1150n, J.R., 51mp50n, D.W., 61a55m0yer, 6., Andrew5, M., and E. Cran5w1ck (1990). 0n the rupture 20ne and 10ca1 9e01091c effect5 0f the Armen1an earth4uake 0f Decem6er 7, 1988, 1n Pr0ceed1n95 0f the F0urth U.5. Nat10na1 C0nference 0n Earth4uake En91neer1n9, Pahn 5pr1n95, Ca11f0rn1a, V01ume 1, 131-140.
80uch0n, M. (1982). 7he rupture mechan15m 0f the C0y0te Lake earth4uake 0f 6 Au9u5t 1979 1nferred fr0m near-f1e1d data, 8u11. 5e15m. 50c. Am., 72, 745-757.
80un1f, A., Hae551er, H., and M. Me9hra0u1 (1987). 7he C0n5tant1ne (n0rthea5t A19er1a) earth4uake 0f 0ct06er 27, 1985--5urface rupture5 and after5h0ck 5tudy, Earth P1anet. 5c1. Letter5, 85, 451-460.
80wman, J.R. (1991). 6e0det1c ev1dence f0r c0nju9ate fau1t1n9 dur1n9 the 1988 7ennant Creek, Au5tra11a earth4uake 5e4uence, 6e0phy5. J. 1nt., 107, 46-56.
80wman, J.R., 61650n, 6., and 7. J0ne5 (1990). After5h0ck5 0f the 1988 January 22 7ennant Creek, Au5tra11a 1ntrap1ate earth4uake5: ev1dence f0r a c0mp1ex thru5t-fau1t 9e0metry, 6e0phy5. J. 1nt., 100, 87-97.
80yd, 7.M., M0r1, J., and 6. 5uare2 (1984). Fau1t p1ane determ1nat10n 0f the 1964 N119ata, Japan earth4uake (a65.), E05, 65, n0. 45, 1016.
8rant1ey, 8.J., and W.Y. Chun9 (1991). 80dy-wave wavef0rm c0n5tra1nt5 0n the 50urce parameter5 0f the Yan911an9, Ch1na, earth4uake 0f Ju1y 25, 1969: a deva5tat1n9 earth4uake 1n a 5ta61e c0nt1nenta1 re910n, Pure App11ed 6e01)hY5., 135, 529-543.
8r101e, P., de Nata1e, 6., 6au10n, R., P1n9ue, F., and R. 5carpa (1986). 1nver510n 0f 9e0det1c data and 5e15m1c1ty a550c1ated w1th the Fr1u11 earth4uake 5e4uence (1976-1977), Anna1e5 6e0phy51cae, 4, n0. 84, 481-492.
8u11, W.8., and P.A. Pearthree (1988). Fre4uency and 512e 0f 1ate Quaternary 5urface rupture5 0f the P1taycach1 fau1t, r10rthea5t 50n0ra, Mex1c0, 8u11. 5e15m. 50c. Am., 78, 956-978.
8urd1ck, L.J., and 6.R. Me11man (1976). 1nver510n 0f the 60dy wave5 fr0m the 80rre90 M0unta1n earth4uake t0 the 50urce mechan15m, 8u11. 5e15m. 50c. Am., 66, 1485-1499.
8urf0rd, R.0. (1972). C0nt1nued 511p 0n the C0y0te Creek fau1t after the 80rre90 M0unta1n earth4uake, 1n 7he 80rre90 M0unta1n Earth4uake 0f Apr11 9, 1968, U.5. 6e01. 5ur. Pr(~ Paper 787, 105-111.
8ut1er, R. (1983). 5urface wave ana1y515 0f the 9 Apr11 1968 80rre90 M0unta1n earth4uake, 8u11. 5e15m. 50c. Am., 73, 879-883.
8ut1er, R., 5tewart, 6.5., and H. Kanam0r1 (1979). 7he Ju1y 27, 1976 7an95han, Ch1na earth4uake--a c0mp1ex 5e4uence 0f 1ntrap1ate event5, 8u11. 5e15m. 50c. Am., 69, 207-220.
8uwa1da, J.P., and P. 5t. Amand (1955). 6e01091ca1 effect5 0f the Arv1n-7ehachap1 earth4uake, 1n Earth4uake5 1n Kern C0unty, Ca11f0rn1a, Dur1n9 1952, Ca1114. D1v. M1ne5 6e01. 8u11. 171, 41-56.
Ca6rera, J., 566r1er, M., and J.L. Merc1er (1991). P110-Quaternary 9e0dynam1c ev01ut10n 0f a 5e9ment 0f the Peruv1an Andean C0rd111era 10cated a60ve the chan9e 1n the 5u6duct10n 9e0metry: the Cu2c0 re910n, 7ect0n0phy51c5, 190, 331-362.
Ca9nett1, V., and V. Pa54ua1e (1979). 7he earth4uake 5e4uence 0f Fr1u11, 1ta1y, 1976, 8u11. 5e15m. 50c. Ant., 69, 1797-1818.
Came16eeck, 7., and M. De 8ecker (1985). 7he earth4uake5 0f L1e9e 0f N0vem6er 8, 1983 and Decem6er 21, 1965, 5e15m1c Act1v1ty 1n We5tern Eur0pe, 233-248.
Camp1110, M., and R.J. Archu1eta (1992). A rupture m0de1 1"0r the 28 June 1992 Lander5, Ca11f0rn1a, earth4uake, E05, 73, n0. 43, 374.
Can1te2, N., and M.N. 70k502 (1972). 5tat1c and dynam1c 5tudy 0f earth4uake 50urce mechan15m- -5an Fernand0 earth4uake, J. 6e0phy5. Re5., 77, n0. 14, 2583-2594.
Cara6aja1, C.C., and J.5.8arker (1991). 50urce pr0ce55e5 and wave pr0Pa9at10n effect5 0n the N0vem6er.25, 1988 5a9uenay, Que6ec earth4uake (a65.), E05, 72, n0. 17, 202.
Chen, P-5., and J-2. Q1n (1991). 7he rupture pr0ce55 0f Lancan9-6en9ma earth4uake, J. 5e15m. Re5., 14, 95-103.
Chen, W., and P. M01nar (1977). 5e15m1c m0ment5 0f maj0r earth4uake5 and the avera9e rate 0f 511p 1n Centra1 A51a, J. 6e0phy5. Re5., 82, 2945-2969.
Chen, Y., and F.7. Wu (1989). Lancan9-6en9ma earth4uake, a pre11m1nary rep0rt 0n the N0vem6er 6, 1988, event and 1t5 after5h0ck5 (a65.), E05, 70, n0.49, 1527, 1540.
Chen, Y-7., L1n, 8-H., Wan9, X-H., Huan9, L-R., and M-L L1u (1979). A d1510cat10n m0de1 0f the 7an95han earth4uake 0f 1976 fr0m the 1nver510n 0f 9e0det1c data, Acta Academ1a 51n1ca, 22, 201-217.
Ch0y, 6.L. (1990).~.50urce parameter5 0f the earth4uake, a5 1nferred fr0m 6r0ad6and 60dy wave5, M Rymer, M.J, and E115w0rth, W.L. ed5., 7he C0a11n9a, Ca11f0rn1a, Earth4uake 0f May 2, 1983, U.5. 6e01. 5ur. Pr0f. Paper 1487, 193-206.
Ch0y, 6.L., and J. 80atwr19ht (1988). 7e1e5e15m1c and near-f1e1d ana1y515 0f the Nahann1 earth4uake5 1n the N0rthwe5t 7err1t0r1e5, Canada, 8u11. 5e15m. 50c. Am., 78, 1627-1652.
Ch0y, 6.L., and J. 80atwr19ht (1990). 50urce character15t1c5 0f the L0ma Pr1eta, Ca11f0rn1a, earth4uake 0f 0ct06er 18, 1989 fr0m 9106a1 d191ta1 5e15m1c data, 6e0phy5. Re5. Letter5, 17, 1183-1186.
C-10
Ch0y, 6.L., and J.R. 80wman (1990). Rupture pr0ce55 0f a mu1t1p1e ma1n 5h0ck 5e4uence: ana1y515 0f te1e5e15m1c, 10ca1, and f1e1d 065ervat10n5 0f the 7ennant Creek, Au5tra11a, earth4uake5 0f January 22, 1988, J. 6e0phy5. Re5., 95, 6867-6882.
Ch0y, 6.L., and R. K1nd (1987). Rupture c0mp1ex1ty 0f a m0derate-512ed (m 6 6.0) earth4uake-- 6r0ad6and 60dy-wave ana1y515 0f the N0rth Yemen earth4uake 0f 13 Decem6er 1982, 8u11. 5e15m. 50c. Am., 77, 28-46.
Ch0y, 6.L., 80atwr19ht, J., Dewey, J.W., and 5.A. 51pk1n (1983). A te1e5e15m1c ana1y515 0f the New 8run5w1ck earth4uake 0f January 9, 1982, J. 6e0phy5. Re5., 88, 2199-2212.
Chun9, W-Y., and 8.J. 8rant1ey (1989). 7he 1984 50uthern Ye110w 5ea earth4uake 0f ea5tern Ch1na--50urce pr0pert1e5 and 5e15m0tect0n1c 1mp11cat10n5 f0r a 5ta61e c0nt1nenta1 area, 8u11. 5e15m. 50c. Am., 79, 1863-1882.
Chun9, W-Y., 8rant1ey, 8.J., and A.C. J0hn5t0n (1988). 50urce mechan15m5, 5urface rutpure, and re1at1ve 10cat1n5 0f the 22 January 1988 7ennant Creek earth4uake5, centra1 Au5tra11a (a65.), E05, 69, n0. 44, 1301.
C1par, J. (1979). 50urce pr0ce55e5 0f the Ha1chen9, Ch1na earth4uake fr0m 065ervat10n5 0f P and 5 wave5, 8u11. 5e15m. 50c. Am., 69, 1903-1916.
C1par, J. (1980). 7e1e5e15m1c 065ervat10n5 0f the 1976 Fr1u11, 1ta1y, earth4uake 5e4uence, 8u11. 5e15m. 50c. Am., 70, 963-983.
Cr0ne, A.J., Machette, M.N., 80n111a, M.6., L1enkaemper, J.J., P1erce, K.L., 5c0tt, W.E., and R.C. 8ucknam (1987). 5urface fau1t1n9 acc0mpany1n9 the 80rah Peak earth4uake and 5e9mentat10n 0f the L05t R1ver fau1t, centra1 1dah0, 8u11. 5e15m. 50c. Am., 77, 739-770.
Cr0550n, R.5., Mart1n1, M., 5carpa, R., and 5.C. Key (1986). 7he 50uthern 1ta1y earth4uake 0f 23 N0vem6er 1980--an unu5ua1 pattern 0f fau1t1n9, 8u11. 5e15m. 50c. Am., 76, 381-394.
Dar6y, D.J. (1989). D1510cat10n rn0de111n9 0f the 1987 Ed9ecum6e earth4uake, New 2ea1and, New 2ea1and J. 6e01. 6e0phy5., 32, 115-122.
De1 Pe220, E., 1annacc0ne, 6., Mart1n1, M., and R. 5carpa (1983). 7he 23 N0vem6er 1980 50uthern 1ta1y earth4uake, 8u11. 5e15m. 50c. Am., 73, 187--200.
Den9, Q., and P. 2han9 (1984). Re5earch 0n the 9e0metry 0f 5hear fracture 20ne5, J. 6e0phy5. Re5., 89, 5699-5710.
Den9, Q., Wu, D., 2han9, P., and 5. Chen (1986). 5tructure and def0rmat10na1 character 0f 5tr1ke-511p fau1t 20ne5, Pure App11ed 6e0phy5., 124, n0. 1/2, 203-223.
Denham, D., A1exander, L.6., and 6. W0r0tn1ck1 (1980). 7he 5tre55 f1e1d near the 51te5 0f the Mecker1n9 (1968) and Ca11n91r1 (1970) earth4uake5, we5tern Au5tra11a, 7ect0m)phy51c5, 67, 283-317.
Denham, D., A1exander, L.6., Ever1n9ham, 1.8., 6re950n, P.J., McCaffrey, J., and J.R. Enever (1987). 7he 1979 Cad0ux earth4uake and 1ntrap1ate 5tre55 1n we5tern Au5tra11a, Au5tra11an J. Earth 5c1ence5, 34, 507-521.
Dewey, J.W.,~and A. 6rant2 (1973). 7he 6h1r earth4uake 0f Apr11 10, 1972 1n the 2a9r05 M0unta1n510f 50uthern 1ran--5e15m0tect0n1c a5pect5 and 50me re5u1t5 0f a f1e1d rec0nna155ance, 8u11. 5e15m. 50c. Am., 63, 2071-2090.
De2a, E. (1971). 7he Par1ahuanca earth4uake5, Huancay0, Peru: Ju1y-0ct06er 1969, pre11m1nary rep0rt, 1n C0111n5, 8.W., and Fra5er, R., ed5., Recent Cru5ta1 M0vement5, R0ya1 50c. New 2ea1and 8u11. 9, 77-83.
D1et2, L.D., and W.L. E115w0rth (1990). 7he 0ct06er 17, 1989, L0ma Pr1eta, Ca11f0rn1a, earth4uake and 1t5 after5h0ck5: 9e0metry 0f the 5e4uence fr0m h19h-re501ut10n 10cat10n5, 6e0phy5. Re5. Letter5, 17, 1417-1420.
D0r6ath, C., D0r6ath, L., 6au10n, R., 6e0r9e, 7., M0ur9ue, P., Ramdan1, M., R061neau, 8., and 8. 7ad111 (1984). 5e15m0tect0n1c5 0f the 6u1nean earth4uake 0f Decem6er 22, 1983, 6e0phy5. Re5. Letter5, 11,971-974.
D0r6ath, L., D0r6ath, C., R1vera, L., Fuen2a11da, A., C15terna5, A., 7atev0551an, R., Aptekman, J., and 5. Aref1ev (1992). 6e0metry, 5e9mentat10n and 5tre55 re91me 0f the 5p1tak, ~Armen1a)
• Q 3 ••
earth4uake fr0m the ana1y5~5 0f the after5h0ck 5e4uence, 6e0phy5. J. 1nt., 108, 309-328.
C-13
D05er, D.1. (1985). 50urce parameter5 and fau1t1n9 pr0ce55e5 0f the 1959 He69en Lake, M0ntana, earth4uake 5e4uence, J. 6e0phy5. Re5., 90, 4537-4555.
Eat0n, J.P. (1984). 5e15m1C 5ett1n9, 10Cat10n, and f0Ca1 mechan15m 0f the May 2, 1983, C0a11n9a earth4Uake: 1n 5Ch011, R.E., and 5tratta, J.L., ed5., C0a11n9a, Ca11f0rn1a, Earth4Uake 0fMay 2, 1983: Earth4Uake En91neer1n9 Re5earCh 1n5t1tUte Rep0rt 84-03, 18-21.
Eat0n, J.P. (1985). 7he May 2, 1983 C0a11n9a earth4uake and 1t5 after5h0Ck5: a deta11ed 5tUdy 0f the hyp0Center d15tr16Ut10n and 0f the f0Ca1 mechan15m5 0f the 1ar9er after5h0ck5, 1n Rymer, M.J., and E115w0rth, W.L., ed5., MeChan1C5 0f the May 2, 1983 C0a11n9a Earth4Uake; U.5. 6e01. 5Ur. 0pen-F1fe Rep0rt 85-44, 132-201.
Eat0n, J.P. (1990). 7he earth4uake and 1t5 after5h0Ck5 fr0m May 2 thr0U9h 5eptem6er 30, 1983, 1~ Ryrner, M.J, and E115w0rth, W.L. ed5., 7he C0a11n9a, Ca11f0rn1a, Earth4uake 0f May 2, 1983, U.5. 6e01. 5Ur. Pr0f. Paper 1487, 113-170.
Eat0n, J~P., 0•Ne111, M., and J.N. MUrd0Ck (1970). After5h0Ck5 0f the 1966 Parkf1e1d-Ch01ame, Ca11f0rn1a, earth4Uake, 8U11. 5e15m. 50C. Am., 60, 1151-1197.
Ey1d09an, H., and J. Jack50n (1985). A 5e15m01091ca1 5tudy 0f n0rma1 fau1t1n9 1n the Dem1rc1, A1a5eh1r and 6ed12 earth4uake5 0f 1969-70 1n we5tern 7urkey--1mp11cat10n5 f0r the nature and 9e0metry 0fdef0rmat10n 1n the c0nt1nenta1 cru5t, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 81,569- 607.
Ey1d09an, H., Na6e1ek, J., and M.N. 70k502 (1985). 7he 6a211, U55R, 19 March 1984 earth4uake--the mechan15m and tect0n1c 1mp11cat10n5, 8u11. 5e15m. 50c. Am., 75, 661-675.
Feh1er, M.C., and P.A. J0hn50n (1989). Determ1nat10n 0f fau1t p1ane5 at C0a11n9a, Ca11f0rn1a, 6y ana1y515 0f pattern5 1n after5h0ck 10cat10n5, J. 6e0phy5. Re5., 94, 7496-7506.
F1nett1, 1., Ru551, M., and D. 51ejk0 (1979). 7he Fr1u11 earth4uake (1976-1977), 7ect0n0phy51c5, 53, 261-272.
F10ren50v, N.A., and V.P. 5010nenk0, ed5. (1965). 7he 6061-A1ta1 earth4uake, Academy 0f 5c1ence5:0f the U55R: tran51ated.f1~0m Ru551an 6y 15rae1 Pr09ram.f1~r 5c1ent~17c 7ran51at10n5, Jeru5a1ern, 424.
Franke1, A. (1984). 50urce parameter5 0f tw0 M k - 5 earth4uake5 near An2a, Ca11f0rn1a, and a c0mpar150n w1th an h~nper1a1 Va11ey after5h0ck, 8u11. 5e15m. 50c. Am., 74, 1509-1527.
Fredr1ch J., McCaffrey, R. Denham D. (1988). 50urce parameter5 0f 5even 1ar9e Au5tra11an earth4uake5 determ1ned 6y 60dy wavef0rm 1nver510n, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 95, 1-13.
Fu15, 6. (1976). 6r0und 6reaka9e and after5h0ck5 0f the M E = 5.2 6a1way Lake earth4uake, June 1975, M0jave De5ert, Ca11f0rn1a (a65.), E05, 57, n0. 11, 954.
6he1tanch1, M.R., K11kuch1, M., and M. M150ne (1990). Far f1e1d 50urce ana1y515 0f the 1981 6016af-51rch, 50uth-ea5t 1ran, earth4uake, E05, 71, n0. 43, 1480.
60rd0n, F.R. (1971). Fau1t1n9 dur1n9 the earth4uake at Mecker1n9, we5tern Au5tra11a: 14 0ct06er 1968, ~ C0111n5, 8.W., and Fra5er, R., ed5., Recent Cru5ta1 M0vement5, R0ya1 50c. New 2ea1and 8u11. 9, 85-93.
60rd0n, F.R., and J.D. Lew15 (1980). 7he Mecker1n9 and Ca11n91r1 earth4uake5 0ct06er 1968 and March 1970, 8u11. 6e01. 5ur. We5t. Au5tra11a, 126, 229 p.
6reen, R.W.E., and A. Mc6arr (1972). A c0mpar150n 0f the f0ca1 mechan15m and after5h0ck d15tr16ut10n 0f the Cere5, 50uth Afr1ca, earth4uake0f 5eptem6er 29, 1969, 8u11. 5e15m. 50c. Am., 62, 869-871.
C-17
6reen5fe1der, R. (1968). After5h0ck5 0f the 7ruckee, Ca11f0rn1a, earth4uake 0f 5eptem6er 12, 1966, 8u11. 5e15m. 50c. Am., 58, 1607-1620.
6r055, W.K., and J.C. 5ava9e (1985). Def0rmat10n near the ep1center 0f the 1984 R0und Va11ey, Ca11f0rn1a, earth4uake, 8u11. 5e15m. 50c. Am., 75, 1339-1347.
6u1kan, P., 6urp1nar, A., Ce1e61, M., Arpat, E., and 5. 6enc091u (1978). En91neer1n9 rep0rt 0n the Murad1ye-Ca1d1ran, 7urkey, earth4uake 0f 24 N0vem6er 1976: prepared f0r C0mm1ttee 0n Natura1 D15a5ter5, C0mm15510n 0n 50c10techn1ca1 5y5tem5, Nat10na1 Re5earch C0unc11, 32 p.
6uten6er9, 8., and C.F. R1chter (1954). 5e15m1c1ty 0f the Earth and A550c1ated Phen0mena, 2nd ed.: Pr1ncet0n Un1ver51ty Pre55, Pr1ncet0n, New Jer5ey, 310 p.
Hae551er, H., 6au10n, R., R1vera, L., C0n501e, R., Fr09neux, 6a5par1n1, 6., Marte1, L., Patau, 6., 51c111an0, M., and A. C15terna5 (1988). 7he Peru91a (1ta1y)earth4uake 0f29, Apr11 1984: a m1cr0earth4uake 5urvey, 8u11. 5e15n1. 50c. Am., 78, 1948-1964.
Hae551er, H., Cara, M., J1mene2, E., De5champ5, A., and 8. R0man0w1c2 (1989). Rupture pr0ce55 0f the Armen1an earth4uake fr0m 6r0ad-6and and very 10n9 per10d te1e5e15m1c rec0rd5, E05, 40, n0.43, 1199.
Hae551er, H., H0an9-7r0n9, P., 5ch1ck, R., 5chne1der, 6., and K. 5tr06ack (1980). 7he 5eptem6er 3, 1978 5wa61an Jura earth4uake, 7ect0n0phy51c~, 68, 1-14.
Ha9h1p0ur, A., and M. Am1d1 (1980). 7he N0vem6er 14 t0 Decem6er 25, 1979 6haenat earth4uake5 0f n0rthea5t 1ran and the1r tect0n1c 1mp11cat10n5, 8u11. 5e15m. 50c. Am., 70, 1751- 1757.
Ha11, W.8., and P.E. 5a610ck (1985). C0mpar150n 0f the 9e0m0rph1c and 5urf1c1a1 fractur1n9 effect5 0f the 1983 80rah Peak, 1dah0 earth4uake w1th th05e 0f the 1959 He69en Lake, M0ntana, earth4uake, 1n 5te1n, R.5., and 8ucknam, R.C., ed5., Pr0ceed1n95 0f W0rk5h0p XXV111 0n the 80rah Peak, 1dah0, Earth4uake, U.5. 6e01. 5ur. 0pen-F11e Rep0rt 85-290, 141-152.
Ha5e9awa, H.5., Wetm111er, R.J., and M. Lam0nta9ne (1989). A c0mpar1510n 0f the three 1ar9e5t Nahann1 earth4uake5 (1985-1988) and the 5e15m0tect0n1c env1r0nment, 5e15m. Re5. Letter5, 60, 29.
Hatanaka, Y., and K. 5h1ma2ak1 (1988). Rupture pr0ce55 0f the 1975 centra1 01ta, Japan, earth4uake, J. Phy5. Earth, 36, 1-15.
Hatanaka, Y., and M. 7ake0 (1989). Deta11ed rupture pr0ce55 0f the 1975 centra1 01ta, Japan, earth4uake 1nferred fr0m near-f1e1d data, J. Phy5. Earth, 37, 251-264.
C-19
Hauk550n, E. (1990). 7he 1933 L0n9 8each earth4uake and 1t5 after5h0ck5, 5e15m. Re5. Letter5, 61, 42.
Hauk550n, E., and 5.6r055 (1991). 50urce parameter5 0f the 1933 L0n9 8each earth4uake, 8u11. 5e15m. 50c. Am., 81, 81-99.
Hauk550n, E., and L.M. J0ne5 (1988). 7he Ju1y 1986 0cean51de (M E = 5.3) earth4uake 5e4uence 1n the c0nt1nenta1 60rder1and, 50uthern Ca11f0rn1a, 8u11. 5e15m. 50c. Am., 78, 1885-1906.
Hauk550n, E., and L.M. J0ne5 (1989). 7he 1987 Wh1tt1er Narr0w5 earth4uake 5e4uence 1n L05 An9e1e5, 50uthern Ca11f0rn1a--5e15m01091ca1 and tect0n1c ana1y515, J. 6e0phy5. Re5., 94, 9569-9589.
Hauk550n, E., and L.M. J0ne5 (1991a). 7he 1988 and 1990 Up1and earth4uake5:1eft-1atera1 fau1t1n9 adjacent t0 the centra1 7ran5ver5e Ran9e5, J. 6e0phy5. Re5., 96, 8143-8165.
Hauk550n, E., and L.M. J0ne5 (19916). 7he 1991 (M L "-- 5.8) 51erra Madre earth4uake 1n 50uthern Ca11f0rn1a: 5e15m01091ca• and tect0n1c ana1y515, E05, 72, n0.44, 319.
Hauk550n, E., J0ne5, L.M., Hutt0n, K., and D. E6erhart-Ph1111p5 (1993). 7he 1992 Lander5 earth4uake 5e4uence: 5e15m01091ca1 065ervat10n5: J. 6e0phy5. Re5., 99, n0. 811, 19,835- 19,858.
Hauk550n, E., and 0ther5 (1988). 7he 1987 Wh1tt1er Narr0w5 earth4uake 1n the L05 An9e1e5 metr0p011tan area, Ca11f0rn1a, 5c1ence, 239, 1409-1412.
Hauk550n, E., Hutt0n, K., Kanam0r1, H., 8ryant, 5., Q1an, H., D0u91a55, K., J0ne5, L.M., E6erhart-Ph1111p5, D., M0r1, J., and 7.H. Heat0n (1992). 0verv1ew 0f the 1992 (M6.1,7.5,6.6) Lander5 earth4uake 5e4uence 1n 5an 8ernard1n0 C0unty, Ca11f0rn1a, E05, 73, n0. 43,357.
H0mer, R.8., Wetm111er, R.J., Lam0nta9ne, M., and M. P10uffe (1989). 7he Nahann1, NW7, earth4uake 5e4uence, 1985-1988, 5e15m. Re5. Letter5, 60, 28.
H0rner, R.8., Wetm111er, R.J., Lam0nta9ne, M., and M. P10uffe (1990). A fau1t m0de1 f0r the Nahann1 earth4uake5 fr0m after5h0ck 5tud1e5, 8u11. 5e15m. 50c. Am., 80, 1553-1570.
H0rt0n, 5., and D. Dep010 (1992). 7he 0ct06er 24, 1990 Lee V1n1n9, Ca11f0rn1a earth4uake and 0ther recent m0derate earth4uake5 1n the we5tern 6a51n and ran9e, 5e15m. Re5. Letter5, 63, 39.
H0u9h, 5.E., M0r1, J., 5em6era, E., 61a55m0yer, 6., Mue11er, C., and 5. Lydeen (1993, 1n rev1ew). 5urface rupture a550c1ated w1th the 6/28/92 M7.4 Lander5 earth4uake: D1d 1t a11 happen dur1n9 the ma1n5h0ck•: Unpu6115hed Paper.
15h1da, M. (1984). 5pat1a1-temp0ra1 var1at10n 0f 5e15m1c1ty and 5pectrum 0f the 1980 earth4uake 5warm near the 12u Pen1n5u1a, Japan, 8u11. 5e15m. 50c. Am., 74, 199-221.
Jack50n, J.A., and 7.J. F1tch (1979). 5e15m0tect0n1c 1mp11cat10n5 0f re10cated after5h0ck 5e4uence5 1n 1ran and 7urkey, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 57, 209-229.
Jack50n, J.A., and 7.J. F1tch (1981). 8a5ement fau1t1n9 and the f0ca1 depth5 0f the 1ar9er earth4uake5 1n the 2a9r05 m0unta1n5 (1ran), 6e0phy5. J. R. A5tr. 50c. L0nd0n, 64, 561-586.
Jack50n, J.A., 6a9nepa1n, J., H0u5eman, 6., K1n9, 6.C.P., Papad1m1tr10u, P., 50uf1er15, C., and J. V1r1eux (1982). 5e15m1c1ty, n0rma1 fau1t1n9, and the 9e0m0rph01091ca1 deve10pment 0f the 6u1f 0f C0r1nth (6reece): the C0r1nth earth4uake5 0f Fe6ruary and March 1981, Earth P1anet. 5c1. Letter5, 57, 377-397.
Jack50n, J.A., and D. McKen21e (1984). Act1ve tect0n1c5 0f the A1p1ne-H1ma1ayan 6e1t 6etween we5tern 7urkey and Pak15tan, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 77, 185-264.
Jack50n, J.A., and 6. Y1e1d1n9 (1983). 711e 5e15m1c1ty 0f K0h15tan, Pak15tan: 50urce 5tud1e5 0f the Hamran (1972.9.3), Dare1 (1981.9.12) and Patan (1974.12128) earth4uake5, 7ect0n0phy51c5, 91, 15-28.
Jack50n, J.A., M01nar, P., Patt0n, H., and 7. F1tch (1979). 5e15m0tect0n1c a5pect5 0f the Markan5u Va11ey, 7adj1k5tan, earth4uake 0f Au9u5t 11, 1974, J. 6e0phy5. Re5., 84, 6157-6167.
J0hn5t0n, A.C. (1988). 065ervat10n5 0f the 5urface rupture 0f the 22 January 1988 7ennant Creek earth4uake 5e4uence, centra1 Au5tra11a: Center f0r Earth4uake Re5earch and 1nf0rmat10n 5pec1a1 Rep0rt 88-1.
C-22
J0hn5t0n, M.J.5., 80rcherdt, R.D., 61a55m0yer, 6., and A.7. L1nde (1987). 5tat1c and dynam1c 5tra1n dur1n9 the Ju1y 21, 1986, Cha1fant earth4uake near the L0n9 Va11ey Ca1dera, Ca11f0rn1a, 5e15m. Re5. Letter5, 58, 20.
J0ne5, L.E., and D.V. He1m6er9er (1993). 50urce parameter5 0f the 1992 819 8ear earth4uake 5e4uence, E05, 74, n0. 16.
J0ne5, L.M., Han, W., Hauk550n, E., J1n, A., 2han9, Y., and 2. Lu0 (1984). F0ca1 mechan15m5 and after5h0ck 10cat10n5 0f the 50n9pan earth4uake5 0f Au9u5t 1976 1n 51chuan, Ch1na, J. 6e0phy5. Re5., 89, 7697-7707.
Len5en, 6.J., and P.M. 0tway (1971). Earth5h1ft and p05t-earth5h1ft def0rmat10n a550c1ated w1th the May 1968 1nan9ahua earth4uake, New 2ea1and, ~ C0111n5, 8.W., and Fra5er, R., ed5., Recent Cru5ta1 M0vement5, R0ya1 50c. New 2ea1and 8u11. 9, 107-116.
Mend02a, C., and 5.H. Hart2e11 (1988). 1nver510n f0r 511p d15tr16ut10n u51n9 te1e5e15m1c P wavef0rm5--N0rth Pa1m 5pr1n95, 80rah Peak, and M1ch0acan earth4uake5, 8u11. 5e15m. 50c. Am., 78, 1092-1111.
Merc1er, J.L., M0uyar15, N., 51meak15, C.¢ R0und0yann15, 7., and C. An9e11dh15 (1979). 1ntra- p1ate def0rmat10n: a 4uant1tat1ve 5tudy 0f the fau1t5 act1vated 6y the 1978 7he55a10n1k1 earth4uake5, Nature, 278, 45-48.
Merc1er, J-L., Carey-6a11hard15, E., M0uyar15, N., 51meak15, K., R0und0yann15, 7., and C. An9he11dh15 (1983). 5tructura1 ana1y515 0f recent and act1ve fau1t5 and re910na1 5tate 0f 5tre55 1n the ep1centra1 area 0f the 1978 7he55a10n1k1 earth4uake5 (n0rthern 6reece), 7ect0n1c5, 2, 577-600.
Merc1er, J.L., 5e6r1er, M., Lavenu, A., Ca6rea, J., 8e111er, 0., Dum0nt, J.F., and J. Machare (1992). Chan9e5 1n the tect0n1c re91me a60ve a 5u6duct10n 20ne 0f Andean type: 7he Ande5 0f Peru and 8011v1a dur1n9 t11e P110cene-P1e15t0cene, J. 6e0phy5. Re5., 97, 11,945-11,982.
Meyer, 8., 7app0nn1er, P., 6audemer, Y., Pe1t2er, 6., and A. 81u550n (1989). 1932 Chan9 Ma ( M - 7.6) earth4uake 5urface 6reak5 and ne0tect0n1c5 0f n0rthern 716et-Qu1n9ha1 H19h1and5 (a65.), E05, 70, n0.43, 1350.
M1chae1, A.J., and U.5.6.5.8ranch 0f 5e15rn0109y, 1990, 5e15m09en1c 5tructure and 5e15m1c1ty 0f the 1989 L0ma Pr1eta, Ca11f0rn1a 5e4uence, E05, 71, n0. 8, 291.
M1kum0, 7. (1973a). Fau1t1n9 mechan15m 0f the 61fu earth4uake 0f 5eptem6er 9, 1969, and 50me re1ated pr061em5, J. Phy5. Earth, 21, 191-212.
M1kum0, 7., and M. And0 (1976). A 5earch 1nt0 the fau1t1n9 mechan15m 0f the 1891 9reat N061 earth4uake, J. Phy5. Earth, 24, 63-87.
M120ue, M., Nakamura, M., 5et0, N., 5aka1, K., K06aya5h1, M., Haneda, 7., and 5. Ha5h1m0t0 (1985). A c0ncea1ed fau1t 5y5tem a5 1nferred fr0m the after5h0ck act1v1ty acc0mpany1n9 the 1984 We5tern Na9an0 prefecture earth4uake 0f M6.8, 8u11. Earth4. 915t1t. 70ky0, 60, 199-220.
M091, A., Kawamura, 8., and Y. 1wa6uch1 (1964). 5u6mar1ne cru5ta1 m0vement due t0 the N119ata earth4uake 1n 1964, 1n the env1r0n5 0f the Awa 51ma 151and, Japan 5ea, J. 6e0det1c 5ur. Japan, 10, n0. 3-4, 180-186.
M011nar1, M. (1984). Late Cen0201c 5tructura1 9e0109y 0f 5tewart and M0nte Cr15t0 va11ey5, Wa1ker Lane 0f we5t centra1 Nevada, 1n L1nt2, J., Jr., ed., We5tern 6e01091ca1 Excur510n5, 6e01. 50c. Am. F1e1d 7r1p 6u1de600k, 4, 219-231.
M01nar, P., and W.-P. Chen (1983). F0ca1 depth5 and fau1t p1ane 501ut10n5 0f earth4uake5 under the 716etan p1ateau, J. 6e0phy5. Re5., 88, 1180-1196.
M01nar, P., and Q. Den9 (1984). Fau1t1n9 a550c1ated w1th 1ar9e earth4uake5 and the avera9e rate 0f def0rmat10n 1n centra1 and ea5tern A51a, J. 6e0phy5. Re5., 89, 6203-6227.
M01nar, P., and H. Ly0n-Caen (1989). Fau1t p1ane 501ut10n5 0f earth4uake5 and act1ve tect0n1c5 0f the 716etan P1ateau and 1t5 mar91n5, 6e0phy5. J. 1nt., 99, 123-153.
M0r1, J. (1989). 7he New 1re1and earth4uake 0f Ju1y 3, 1985 and a550c1ated 5e15m1c1ty near the Pac1f1c-5010m0n 5ea-815marck 5ea tr1p1ejunct10n, Phy5. Earth P1anet. 1nter10r5, 55, 144-153.
M0r1, J., and 7. 80yd (1985). 5e15m01091ca1 ev1dence 1nd1cat1n9 rupture a10n9 an ea5tward d1pp1n9 fatu1t~ p1ane f0r the 1964 N119ata, Japan earth4uake, J. Phy5. Earth, 33, 227-240.
M0r1, J., McKee, C., and H. Let2 (1987). 7he centra1 New 8r1ta1n earth4uake 0f May 10, 1985: ten510na1 5tre55e5 1n the fr0nta1 arc, Phy5. Earth P1anet. 1nter10r5, 48, 73-78.
M05kv1na, A.6. (1978). F0ca1 mechan15m5 and parameter5 0f the M090d earth4uake 0f January 5, 1967, and 1t5 after5h0ck5: Earth Phy51c5, 14, 1-10.
Mun9ufa, L., and J.N. 8rune (1984). L0ca1 ma9n1tude and 5ed1ment amp11f1cat10n 065ervat10n5 fr0m earth4uake5 1n the n0rthern 8aja Ca11f0m1a-50uthern Ca11f0rn1a Re910n, 8u11. 5e15m. ~50c. Am., 74, 107-119.
Mura1, 1., and 7. Mat5uda (1975). 7he earth4uake 0f 1975 1n the centra1 part 0f 01ta Prefecture, Kyu5hu, 8u11. Earth4. Re5. 1n5t. 70ky0, 50, 303-327.
Na6e1ek, J. (1985). 6e0metry and mechan15m 0f fau1t1n9 0f the 1980 E1 A5nam, A19er1a, earth4uake fr0m 1nver510n 0f te1e5e15m1c 60dy wave5 and c0mpar150n w1th f1e1d 065ervat10n5, J. 6e0phy5. Re5., 90, 12,713-12,728.
Na6e1ek, J., and 6. 5uare2 (1989). 7he 1983 600dn0w earth4uake 1n the centra1 Ad1r0ndack5, New Y0rk--rupture 0f a 51mp1e, c1rcu1ar crack, 8u11. 5e15m. 50c. Am., 79, 1762-1777.
Na6e1ek, J., and M. 70k502 (1978a). 7he 50urce mechan15m 0f the 5ept. 6, 1975 7urk15h earth4uake, Earth4. N0te5, 49, n0.4, 82.
Na6e1ek, J., and M. 70k502 (19786). 50urce5 pr0pert1e5 0f the 1976 earth4uake 1n E. 7urkey, Earth4. N0te5, 49, n0. 1, 82.
Na6e1ek, J., Chen, W.P., and H. Ye (1987). 7he 7an95han earth4uake 5e4uence--1t5 1mp11cat10n5 f0r the ev01ut10n 0f the n0rth Ch1na 8a51n, J. 6e0phy5. Re5., 92, 12,615-12,628.
Nakamura, K., Ka5a11ara, K., and 7. Mat5uda (1964). 711t1n9 and up11ft 0f an 151and, Awa5h1ma, near the ep1centre 0f the N119ata earth4uake 1n 1964, J. 6e0det1c 5ur. Japan, 10, n0. 3-4, 172-179.
Nata11, 5.6., and M.L. 56ar (1982). 5e15m1c1ty 1n the ep1centra1 re910n 0f the 1887 n0rthea5t 50n0ra earth4uake, Mex1c0, 8u11. 5e15m. 50c. Am., 72, 181-196.
Nava, F.A., and J.N. 8rune (1983). 50urce mechan15m and 5urface wave exc1tat10n f0r tw0 earth4uake5 1n n0rthern 8aja Ca11f0rn1a, Mex1c0, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 73, 738- 763.
Needham, R.E., and 5.A. 51pk1n (1989). 7e1e5e15m1c 50urce parameter5 0f the 7 Decem6er 1988 Armen1an earth4uake, E05, 70, n0.43, 1200.
Ne150n, M.R., McCaffrey, R., and P. M01nar (1986). 50urce parameter5 f0r 17 earth4uake5 1n the 71en 5han, centra1 A51a, determ1ned 6y P and 5H wavef0rm 1nver510n (a65.), E05, 67, n0. 16, 305.
New 2ea1and Department 0f 5c1ent1f1c and 1ndu5tr1a1 Re5earch (1987). 7he March 2, 1987, earth4uake near Ed9ecum6e, N0rth 151and, New 2ea1and, E05, 68, n0. 44, 1162-1171.
N1, J.F., and F. 6uan9we1 (1989). Fau1t p1ane 501ut10n5 0f earth4uake5 and act1ve tect0n1c5 0f the Pam1r-K0rak0rum re910n (a65.), E05, 70, n0.43, 1226.
N1a21, M. (1968). Fau1t rupture 1n the 1ran1an (Da5ht-e-8aya2) earth4uake 0f Au9u5t 1968, Nature, 220, 569-570.
N1a21, M., and H. Kanam0r1 (1981). 50urce parameter5 0f 1978 7a6a5 and 1979 Qua1nt, 1ran, earth4uake5 fr0m 10n9-per10d 5urface wave5, 8u11. 5e15m. 50c. Am., 71, 1201-1213.
N1a21, M., and J. 5h0ja-7aher1 (1985). 50urce 9e0metry and mechan15na 0f 1978 7a6a5, 1ran, earth4uake fr0rn we11 10cated after5h0ck5, 7ect0n0phy51c5, 115, 61-68.
N1ch0150n, C., Kanam0r1, H., and C.R. A11en (1987). C0mpar150n 0f the 1948 and 1986 earth4uake5 a10n9 the 50uthern 5an Andrea5 fau1t, C0ache11a Va11ey, Ca11f0rn1a (a65.), E05, 68, n0. 44, 1362.
N1ch0150n, C., R0e10ff5, E., and R.L. We550n (1988). 7he n0rthea5tern 0h10 earth4uake 0f 31 January 1986:wa5 1t 1nduced~, 8u11. 5e15m. 50c. Am., 78, 188-217.
N1ch0150n, C., Harr15, R.A.2~ and R.W. 51mp50n (1993). Chan9e5 1n att1tude-chan9e5 1n 1at1tude: what happened t0 the fau1t5 1n the J05hua 7ree area 6ef0re and after the M7.4 Lander5 ma1n5h0ck, 5e15m. Re5. Letter5, 64, 34.
N15henk0, 5.P., and K.H. Jac06 (1990). 5e15m1c p0tent1a1 0f the Queen Char10tte-A1a5ka-A1eut1an 5e15m1c 20ne, J. 6e0phy5. Re5., 95, 2511-2532.
N0rth, R. 6. (1977). 5e15m1c m0ment, 50urce d1 men510n5, and 5tre55e5 a550c1ated w1 th earth4uake5 1n the Med1terranean and M1dd1e Ea5t, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 48, 137-161.
N0rth, R.6., Wetm111er, R.J., Adam5, J., An911n, F.M., Ha5e9awa, H.5., Lam0nta9ne, M., Du 8er9er, R., 5ee6er, L., and J. Arm6ru5ter (1989). Pre11m1nary re5u1t5 fr0m the N0vem6er 25, 1988 5a9uenay (Que6ec) earth4uake, 5e15m. Re5. Letter5, 60, 89-93.
N0wr0021, A.A., and A.M. M0hajer-A5hja1 (1980). Fau1t1n9 0f Kur12an and K011 (1ran) earth4uake5 0f N0vem6er 1979, a f1e1d rep0rt, 8u11. du 8ureau de Rechereche5 6e010914ue5 et M1n1ere5 (Deux1eme 5er1e), 5ect10n 1V, 6e01091c 6enera1, n0. 2, 91-99.
N0wr0021, A.A., and A.M. M0hajer-A5hja1 (1985). Fau1t m0vement5 and tect0n1c5 0f ea5tern 1ran--60undar1e5 0f the Lut p1ate, 6e0phy5. J. R. A5tr.:50c. L0nd0n, 83, 215-237.
0hnaka, M. (1978). Earth4uake-50urce parameter5 re1ated t0 ma9n1tude, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 55, 45-66.
0ka1, E.A. (1976). A 5urface-wave 1nve5t19at10n 0f the rupture mechan15m 0f the 6061-A1ta1 (Decem6er 4, 1957)earth4uake, Phy5. Earth P1anet. 1nter10r5, 12, 319-328.
0ka1, E.A. (1992). U5e 0f the mant1e ma9n1tude M M f0r the re55e55ment 0f the m0ment 0f h15t0r1ca1 earth4uake5, Pure App11ed 6e0phy5., 139, 17-57.
0150n, A.H., and R.J. Ap5e1 (1982). F1n1te fau1t5 and 1nver5e the0ry w1th app11cat10n5 t0 the 1979 1mper1a1 Va11ey earth4uake, 8u11. 5e15m. 50c. Am., 72, 1969-2001.
0tuka, Y. (1933). 7he 9e0m0rph0109y and 9e0109y 0f n0rthern 1du Pen1n5u1a, the earth4uake f155ure5 0f N026, 1930, and the pre- and p05t-5e15m1c cru5t def0rmat10n5, 8u11. Earth4. Re5. 1n5t. 70ky0, 11,530-574.
0uyed, M., Me9hra0u1, M., C15terna5, A., De5champ5, A., D0re1, J., Frechet, J., 6au10n, R., Hat5fe1d, D., and H. Ph111p (1981). 5e15m0tect0n1c5 0f the E1 A5nam earth4uake, Nature, 292, 26-31.
0uyed, M., Y1e1d1n9, 6., Hat2f1e1d, D., and 6.C.P1 K1n9 (1983). An after5h0ck 5tudy 0fthe E1 A5nam (A19er1a) earth4uake 0f 1980 0ct06er 10, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 73, 605- 639.
P1afker, 6., Hud50n, 7., 8run5, 7., and M. Ru61n (1978). Late Quaternary 0ff5et5 a10n9 the Fa1rweather fau1t and cru5ta1 p1ate 1nteract10n5 1n 50uthern A1a5ka, Canad1an J. Earth 5c1., 15, 805-816.
P1afker, 6., A9ar, R., A5ker, A.H., and M. Han1f (1987). 5urface effect5 and tect0n1c 5ett1n9 0f the 13 Decem6er 1982 N0rth Yemen earth4uake1 8u11. 5e15m. 50c. Am., 77, 2018-2037.
Pre5c0tt, W.H., L150w5k1, M., J0hn5t0n, M.J.5., 5chu12, 5.5., and J.C. 5ava9e (1990). Def0rmat10n 6ef01"e, dur1n9 and after the L0ma Pr1eta earth4uake 0f 0ct06er 1989, E05, 71, n0. 8, 290.
Pre5c0tt, W.H:, 5ava9e, J.C., and M. L150w5k1 (1988). Cru5ta1 5tra1n, 1n Nat10na1 Earth4uake Ha2ard5 Reduct10n Pr09ram, 5ummar1e5 0f 7echn1ca1 Rep0rt5, v. XXV, U.5. 6e01. 5ur. 0pen-F11e Rep0rt 88-16, 274-281.
Pr1e5t1ey, K.F., 5m1th, K.D., and R.5. C0ckerham (1988). 7he 1984 R0und Va11ey, Ca11f0rn1a, earth4uake 5e4uence, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 95, 215-235.
Purcaru, 6., and H. 8er~khemer (1982). Quant1tat1ve re1at10n5 0f 5e15m1c 50urce parameter5 and a c1a551f1cat10n 0f earth4uake5, 1n Duda, 5.J., and Ak1, K., ed5., Quant1f1cat10n 0f Earth4uake5, 7ect0n0~phy51c5, 84, 57-128.
Q1an, H. (1986). Recent d15p1acement5 a10n9 X1an5hu1he fau1t 6e1t and 1t5 re1at10n w1th 5e15m1c act1v1t1e5, J. 5e15m. Re5., 9, 601,613.
C-35
Q1an9, 2., and L. 2han9 (1984). 7he c1a551f1cat10n 0f Quaternary act1ve fau1t5 1n n0rth Ch1na: Earth4uake Pred1ct10n Re5earch, 2, 267-276.
Ra1e19h, C.8. (1977). Pred1ct10n 0f the Ha1chen9 earth4uake, E05, 58, n0. 5, 236-272. Rea5en6er9, P., and W.L. E115w0rth (1982). After5h0ck5 0f the C0y0te Lake, Ca11f0rn1a,
earth4uake 0f Au9u5t 6, 1979, J. 6e0phy5. Re5., 87, 10637-10655. Re111n9er, R. (1984). C05e15m1c and p05t5e15m1c vert1ca1 m0vement a550c1ated w1th the 1940 M
7.1 1mper1a1 Va11ey, Ca11f0rn1a, earth4uake, J. 6e0phy5. Re5., 89, 4531-4537. Re111n9er, R., and 5. Lar5en (1986). Vert1ca1 cru5ta1 def0rmat10n a550c1ated w1th the 1979 M = 6.6
R1a1, J.A., and E. 8r0wn (1983). Wavef0rm m0de11n9 0f 10n9 per10d p-wave5 fr0m the C0a11n9a earth4uake 0fMay 2, 1983, 1n 8ennett, J.H., and R.W. 5her6urne, ed5., 7he 1983 C0a11n9a, Ca11f0rn1a Earth4uake5, 1983, Ca11f. D1v. M1ne5 6e01. 5pec1a1 Pu611cat10n 66, 247-259.
R1chard50n, W.P. (1989). 7he Matata earth4uake 0f 1977 May 31: a recent event near Ed9ecum6e, 8ay 0f P1enty, New 2ea1and, New 2ea1and J. 6e01. 6e0phy5., 32, 17-30.
5ack5, 1.5., L1nde, A.7., 5n0ke, J.A., and 5. 5uyeh1r0 (1981). A 510w earth4uake 5e4uence f0110w1n9 the 12u-05h1ma earth4uake 0f 1978, 1n 51mp50n, D., and R1chard5, P.6., ed5., Earth4uake Pred1ct10n, An 1nternat10na1 Rev1ew, Amer1can 6e0phy51ca1 Un10n, Maur1ce Ew1n9 5er1e5 4, 617-628.
5a126er9, D.H., Cara6aja1, C.C., 8arker, J.5., and F.7. Wu (1990). Pre11m1nary 50urce character15t1c5 0f the 0ct06er 18, 1989 L0ma Pr1eta ma1n5h0ck 6a5ed 0n te1e5e15m1c P and 5 wavef0rm5, E05, 71, n0. 8, 290.
5a126er9, D.H., Wu, F., 8arker, J., McCaffrey, R., Wan9, J., and K.C. Chen (1988). 5e15m1c1ty, f0ca1 mechan15m5 and tect0n1c5 re1ated t0 three 1986 earth4uake5 1n the v1c1n1ty 0f 7a1wan, E05, 69, n0. 16, 400.
5ander5, C.0., and H. Kanam0r1 (1984). A 5e15m0tect0n1c ana1y515 0f the An2a 5e15m1c 9ap, 5an Jac1nt0 fau1t 20ne, 50uthern Ca11f0rn1a, J. 6e0phy5. Re5., 89, 5873-5890.
5ander5, C., Ma915tra1e, H., and H. Kanam0r1 (1986). Rupture pattern5 and pre5h0ck5 0f 1ar9e earth4uake5 1n the 50uthern 5an Jac1nt0 fau1t 20ne, 8u11. 5e15m. 50c. Am., 76, 1187-1206.
5atake, K., and K. A6e (1983). A fau1t m0de1 f0r the N119ata, Japan, earth4uake 0f June 16, 1964, J. Phy5. Earth, 31,217-223.
5ava9e, J.C., and L.M. Ha5t1e (1966). 5urface def0rmat10n a550c1ated w1th d1p-511p fau1t1n9, J. 6e0phy5. Re5., 71, n0.20, 4897-4904.
5ava9e, J.C., and L.M. Ha5t1e (1969). A d1510cat10n m0de1 f0r the Fa1rv1ew Peak, Nevada, earth4uake, 8u11. 5e15m. 50c. Am., 59, 1937-1948.
5her6urne, R., McNa11y, K., 8r0wn, E., and A: A6urt0 (1983). 7he ma1n5h0ck-after5h0ck 5e4uence 0f2 May 1983: C0a11n9a, Ca11f0rn1a, 1n 8ennett, J.H., and 5her6urne, R.W., ed5., 7he 1983 C0a11n9a, Ca11f0rn1a Earth4uake5, 1983, Ca11f. D1v. M1ne5 6e01. 5pec1a1 Pu611cat10n 66, 275-292.
5h1, J., Fen9, X., 6e, 5., Yan9, 2., 80, M., and J. Hu (1984). 7he Fuyun earth4uake fau1t 20ne 1n X1nj1an9, Ch1na, 1nn A C011ect10n 0f Paper5 0f the 1nternat10na1 5ymp051um 0n C0nt1nenta1 5e15m1c1ty and Earth4uake Pred1ct10n, 5e15m0109Y Pre55, 8e1j1n9, Ch1na, 325-346.
5h1h, C.L., Huan, W.L., Ya0, K.K., and Y.7. H51e (1978). 0n the fracture 20ne5 0f the Chan9ma earth4uake 0f 1932 and the1r 9ene515, Ch1ne5e 6e0phy51c5, 1, 17-45.
5h1ma2ak1, K., and P. 50merv111e (1979). 5tat1c and dynam1c parameter5 0f the 12u-05h1ma, Japan, earth4uake 0f January 14, 1978, 8u11. 5e15m. 50c. Am., 69, 1343-1378.
5h1n, 7.-C., Chan9, 2.-5., and 6.-K. Yu (1989). 7he c0mp1ex rupture 0f the 20th May, 1986, 7a1wan earth4uake, Pr0c. 6e01. 50c. Ch1na, 32, 233-253.
5h1r0k0va, Y.1. (1968). F0ca1 mechan15m 0f the earth4uake 0f Ju1y 26, 1963, at 5k0pje: Phy51c5 0f the 5011d Earth (12ve5t1a, Earth Phy51c5), 104-109.
50uf1er15, C., and 6.5. 5tewart (1981). A 50urce 5tudy 0f the 7he55a10n1k1 (n0rthern 6reece) 1978 earth4uake 5e4uence, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 67, 343-358.
50uf1er15, C., Jack50n, J.A., K1n9 6.C.P., 5pencer, C.H., and C.H. 5ch012 (1982). 7he 1978 earth4uake 5e4uence near 7he55a10n1k1 (n0rthern 6reece), 6e0phy5. J. R. A5tr. 50c. L0nd0n, 68, 429-458.
5te1n, R.5. (1985). Ev1dence f0r 5urface f01d1n9 and 5u65urface fau1t 511p fmm 9e0det1c e1evat10n chan9e5 a550c1ated w1th the 1983 C0a11n9a, Ca11f0rn1a, earth4uake, 1n Rymer, M.J., and E115w0rth, W.L., ed5., Mechan1c5 0f the May 2, 1983, C0a11n9a Earth4uake, U.5. 6e01. 5ur. 0pen-F11e Rep0rt 85-44, 225-253.
5te1n, R.5., and 5.E. 8arr1ent05 (1985a). P1anar h19h-an91e fau1t1n9 1n the 8a51n and Ran9e-- 9e0det1c ana1y515 0f the 1983 80rah Peak, 1dah0, earth4uake, J, 6e0phy5. Re5., 90, 11,355- 11,366.
5te1n, R.5., and 5.E. 8arr1ent05 (19856). 7he 1983 80rah Peak, 1dah0, earth4uake-- 9e0det1c ev1dence f0r deep rupture 0n a p1anar fau1t, 1n 5te1n, R.5., and 8ucknam, R.C., ed5., Pr0- ceed1n95 0f W0rk5h0p XXV111 0n the 80rah Peak, 1dah0, Earth4uake, U. 5. 6e01. 5ur. 0pen- F11e Rep0rt 85-290, 459-484.
5te1n, R.5., and M. L150w5k1 (1983). 7he 1979 H0me5tead Va11ey earth4uake 5e4uence, Ca11f0rn1a--c0ntr01 0f after5h0ck5 and p05t5e15m1c def0rmat10n, J. 6e0phy5. Re5., 88, 6477-6490.
5te1n, R.5., and W. 7hatcher (1981). 5e15m1c and a5e15m1c def0rmat10n a550c1ated w1th the 1952 Kern C0unty, Ca11f0rn1a, earth4uake and re1at10n5h1p t0 the Quaternary h15t0ry 0f the Wh1te W01f fau1t, J. 6e0phy5. Re5., 86, 4913-4928.
5tewart, 6.5., 8ut1er, R., and H. Kanam0r1 (1976). 5urface and 60dy wave ana1y5e5 f0r the Fe6.4, 1975, Ha1chen9 and Ju1y 27, 1976, 7an95han ch1ne5e earth4uake5 (a65.), E05, 57, n0. 11,953-954.
5tewart, 5.W.,H0f1nann, R.8., and W,H. D1ment (1964). 50me after5h0ck5 0f the He69en Lake earth4uake, U.5. 6e01. 5ur. Pr0./•. Paper 435-D, 19-24.
Japan, 42, 59-66. 7ake0, M., and N. N1kam1 (1987). 1nver510n 0f 5tr0n9 m0t10n 5e15m09ram5 f0r the 50urce
pr0ce55 0f the Na9an0ken-5e16u earth4uake 0f 1984, 7ect0n0phy51c5, 144, 271-285. 7an9, R.-C., Huan9, 2., Q1an, H., Den9, 7., J1an9, L., 6e, P., L1u, 5., Ca0, Y., and C. 2han9
(1984). 0n the recent tect0n1c act1v1ty and earth4uake 0f the X1an5hu1he fau1t 20ne, 1n A C011ect10n 0f Paper5 0f t11e 1nternat10na15ymp051um 0n C0nt1nenta15e15m1c1ty and Earth4uake Pred1ct10n, 5e15m01091ca1 Pre55, 8e1j1n9, Ch1na, 347-369.
7an9, R-C., Q1an, H., Chan9, W., Chan9, C., Ca0, Y., and 5. L1u (1984). 0n the 5e15m09e01091c 5ett1n9 and c0nd1t10n5 0f 5e15m09en1c 5tructure5 0f 1981 Da0fu earth4uake, 5e15m0109Y 6e0109Y, 6, 33-40.
7an9, R-C., Wen, D-H., Den9, 7-6., and 5-M. Huan9 (1976). A pre11m1nary 5tudy 0n the character15t1c5 0f the 9r0und fracture5 dur1n9 the Lu-Hu0 M = 7.9 earth4uake, 1973, and the 0r191n 0f the earth4uake, Acta 6e0p10,51ca 51n1ca, 19, 17-27.
7an1m0t0, 7., and H. Kanam0r1 (1986). L1near Pr09ramm1n9 appr0ach t0 na0ment ten50r 1nver510n 0f earth4uake 50urce5 and 150me te5t5 0n the three-d1men510na1 5tructure 0f the upper mant1e, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 84,,,413-430.
7app0nn1er, P., and P. M01nar (1979). Act1ve fau1t1n9 and Cen0201c tect0n1c5 0f the 71en 5han, M0n9011a, and 8ayka1 re910n5, J. 6e0phy5. Re5., 84, 3425-3459.
7r1funac, M.D., and F.E. Udwad1a (1974). Parkf1e1d, Ca11f0rn1a, earth4uake 0f June 27, 1966: a three d1men510na1 m0v1n9 d1510cat10n, 8u11. 5e15m. 50c. Am., 64, 511-533.
7r0dd, H., War6ut0n, P., and C.1. P001ey (1985). 7he 9reat 8r1t15h earth4uake 0f 1984 a5 5een fr0m afar, 6e0phy5. J. R. A5tr. 50c. L0nd0n, 83, 809-912.
75a1, Y.-8., and K. Ak1 (1969). 51mu1tane0u5 determ1nat10n 0f the 5e15m1c m0ment and attentuat10n 0f 5e15m1c 5urface wave5, 8u11. 5e15m. 50c. Am., 59, 275-287.
75a1, Y.-8., and K. Ak1 (1970). 50urce mechan15m 0f the 7ruckee, Ca11f0rn1a, earth4uake 0f 5eptem6er 12, 1966, 8u11. 5e15m. 50c. Am., 60, 1199-1208.
75u60kawa, 1., 09awa, Y., and 7. Haya5h1 (1964). Cru5ta1 m0vement5 6ef0re and after the N119ata earth4uake, J. 6e0det1c 5ur. Japan, 10, n0. 3-4, 165-171.
75ukuda, 7., 5aka1, K., K06aya5h1, M., Ha5h1m0t0, 5., Haneda, 7. (1989). 50urce pr0ce55, character15t1c5 0f a550c1ated 5e15m1c1ty and 5e15m0tect0n1c 1mp11cat10n5 0f the 1986 0mach1 earth4uake 0f M 5.9 1n the n0rthwe5tern part 0f Na9an0 Prefecture, centra1 Japan, 8u11. Earth4. Re5. h~5t. 70ky0, 64, 433-456.
75ukuda, 7., 5aka1, K., Ha5h1m0t0, 5., 6he1tanch1, M.R., 501tan1an, 50., M02affar1, P., M02affar1, N., Aka5heh, 8., and A. Javaher1an (1991). After5h0ck d15tr16ut10n 0f the 1990 Rud6ar, n0rthwe5t 1ran, earth4uake 0f M7.3 and 1t5 tect0n1c 1mp11cat10n5, 8u11. Earth4. Re5. 1n5t. 70ky0, 66, 351-381.
75ukuda, 7., 5aka1, K., Ha5h1m0t0, 5., Haneda, 7., and M. K06aya5h1 (1992). 5tructura1 feature5 0f the precur50ry 5e15m1c 9ap and after5h0ck re910n 0f the 1990 50uthern N119ata earth4uake 0f M 5.4, 8u11. Earth4. Re5. h15t. 70ky0, 67, 361-388.
75uya, H. (1950). 7he Fuku1 earth4uake 0f June 28, 1948--rep0rt 0f the 5pec1a1 c0mm1ttee f0r the 5tudy 0f the Fuku1 earth4uake: Japan 5c1ence C0unc11, 5pec1a1 C0mm1ttee, 70ky0, 197 p.
7ur61tt, 7., 8arker, E.J., 8r0w1tt, C.W.A., H0we115, M., Marr0w, P.C., Mu550n, R.M.W., Newmark, R.H., Redmayne, D.W., Wa1ker, A.8., Jac06, A.W.8., Ryan, E., and V. Ward (1985). 7he N0rth Wa1e5 earth4uake 0f 19 Ju1y 1984, J. 6e01. 50c. L0nd0n, 142,567-571.
7urn0v5ky, J., and 6.5chne1der (1982). 7he 5e15m0tect0n1c character 0f the 5eptem6er 3, 1978, 5wa61an Jura earth4uake 5er1e5, 7ect0n0phy51c5, 83, 151-162.
Ud1a5, A.5.J. (1965). A 5tudy 0f the after5h0ck5 and f0ca1 mechan15m 0f the 5a11na5-Wat50nv111e earth4uake5 0f Au9u5t 31 and 5eptem6er 14, 1963, 8u11. 5e15m. 50c. Am., 55, 85-106.
Uhrhammer, R.A., L0max, A., and E.R. C0111n5 (1990). 8D5N rec0rd1n9 0f 5anta Cru2 M0unta1n5 (L0ma Pr1eta) earth4uake5, June 1988 t0 N0vem6er 1989,,E05, 71, n0. 8, 290.
Wan9, K., Ya0, 2., 6a0, L., and 7.C, Wa11ace (1989). 50urce mechan15m 0f the 1988 Lancan9- 6en9ma, Ch1na, earth4uake, E05, 70, n0. 43, 1218.
C -45
Ward, P.L., 61665, J., Har10w, D., and A6urt0, Q.A. (1974). After5h0ck5 0f the Mana9ua, N1cara9ua, earth4uake and the tect0n1c 519n1f1cance 0f the 715capa fau1t, 8u11. 5e15m. 50c. Am., 64, 1017-1029.
Ward, 5.N., and 6.R. Va1en515e (1989). Fau1t parameter5 and 511p d15tr16ut10n 0f the 1915 Ave22an0, 1ta1y, earth4uake der1ved fr0m 9e0det1c 065ervat10n5, 8u11. 5e15m. 50c. Am., 79, 690-710.
Warren, D.H., 8ufe, C., C0ak1ey, J., and 5. Mark5 (1978). After5h0ck5 0f the N0vem6er 22, 1977, earth4uake near W1111t5, Ca11f0rn1a, Earth4. N0te5, 49, n0.4, 95.
Warren, D.H., 5c0f1e1d, C., and C.6. 8ufe (1985). After5h0ck5 0f the 22 N0vem6er 1977 earth4uake at W1111t5, Ca11f0rn1a, act1v1ty 1n the Maacama fau1t 20ne, 8u11. 5e15m. 50c. Am., 75, 507-517.
We5n0u5ky, 5.6., 5ch012, C.H., and K. 5h1ma2ak1 (1982). Def0rmat10n 0f an 151and arc--rate5 0f m0ment re1ea5e and cru5ta1 5h0rten1n9 1n 1ntrap1ate Japan determ1ned fr0m 5e15m1c1ty and Quaternary fau1t data, J. 6e0phy5. Re5., 87, 6829-6852.
We550n, R.L. (1987). M0de111n9 after5h0ck m19rat10n and after511p 0f the 5an Juan 8aut15ta, Ca11f0rn1a, earth4uake 0f 0ct06er 3, 1972, 7ect0n0phy51c5, 144, 215-229.
We550n, R.L., and W.L. E115w0rth (1972). Pre11m1nary hyp0centra1 data f0r the 5t0ne Cany0n earth4uake 0f 5eptem6er 4, 1972, Earth4. N0te5, 153, n0. 3, 13-15.
We5taway, R. (1987). C0mment 0n ••7he 50uthern 1ta1y earth4u~ake 0f 23 N0vem6er 1980--an unu5ua1 pattern 0f fau1t1n9" 6y Cr0550n, R.5., Mart1n1, M., 5carpa, R., and R. Key, 5.C., 8u11. 5e15m. 50c. Am., 77, 1071-1074.
We5taway, R. (1990). 810ck r0tat10n 1n we5tern 7urkey, J. 6e0phy5. Re5., 95, 19,857-19,884. We5taway, R., and J. Jack50n (1984). 5urface fau1t1n9 1n the 50uthern 1ta11an Campan1a-8a5111cata
earth4uake 0f 23 N0vem6er 1980, Nature, 312, 436-438. We5taway, R., and J. Jack50n (1987). 7he earth4uake 0f 1980 N0vem6er 23 1n Campan1a-
8a5111cata (50uthern 1ta1y), 6e0phy5. J. R. A5tr. 50c. L0nd0n, 90, 375-443. We5taway, R., and R.8. 5m1th (1989). 50urce parameter5 0f the Cache Va11ey (L09an), Utah,
earth4uake 0f 30 Au9u5t 1962, 8u11. 5e15m. 50c. Am., 79, 1410-1425. We5taway, R., 6awth0rpe, R., and M. 70221 (1989). 5¢15m01091ca1 and f1e1d 065ervat10n5 0f the
198J~,La2~0~.A~u220 ear~h4uak~e5~1mp11cat10n5 f0r the act1ve tect0n1c5 0f 1ta1y, 6e0phy5. J. I [ . I I ~ J I . I I I J I I b I I . , I I . I I I I k 4 I l I | . , ] U , - - I~U~ ' - -~ I I ' T .
Y0n9, C., 7501, K.-L,, Fe161, C., 2henhuan, 6., Q1j1a, 2., and C. 2han911 (1988). 7he 7an95han earth4uake--5e15m01091ca1 feature5, Chapter 3 1n Y0n9, C., 7501, K.-L., Fe161, C., 2henhuan, 6., Q1j1a, 2., and 2han911, C., ed5., 7he 6reat 7an95hat• Earth4uake 0f1976, An Anat0my 0fD15a5ter, Per9am0n Pre55, E1m5f0rd, New Y0rk, 96-127.
Y05h1da, A., and N. Hamada (1991). Redeterm1nat10n 0f hyp0center5 0f f0re5h0ck5, ma1n 5h0ck, and after5h0ck5 0f the K1ta-12u earth4uake and the 1t0 earth4uake 5warm 0f 1930, J. Phy5. Earth, 39, 329-344.
Y05h1da, Y., and K. A6e (1990). Mechan15n1 0f the Lu20n, Ph111pp1ne earth4uake 0f Ju1y 16, 1990, E05, 71, n0. 43, 1441.
Y05h1da, Y., and K. A6e (1992). 50urce mechan15m 0f the Lu20n, Ph111pp1ne5 earth4uake 0f Ju1y 1990, 6e0phy5. Re5. Letter5, 19, 545-548.
Yu, 5-8., and C-C. Lu1 (1986). C05e15m1c def0rmat10n a550c1ated w1th the May 1986 Hua11en earth4uake, 8u11. 1n5t1t. Earth 5c1ence5, Academ1a 51n1ca, 6, 73-84.
Yu, W.X., Ca1, 7.J., and X.Y. H0u (1991). Def0rmat10n 20ne 0f M = 7.6 Lanchan9 earth4uake, 5e15m0109Y 6e0109Y, 13, 343-352.
2akhar0va, A.1., 5tar0v01t, 0.E., and L.5. Chepkuna5 (1978). 5e15m1c m0ment and 1t5 determ1nat10n 1n pract1ce 0f data 9enera112at10n 0f un1f1ed 5y5tem 0f 5e15m1c 065ervat10n5 (U550) 0f the U.5.5.R., 7ect0n0phy51c5, 49, 247-253.
2han9, J., and 7. Lay (1990). 50urce parameter5 0f t he 1989 L0ma Pr1eta earth4uake determ1ned fr0m 10n9-per10d Ray1e19h wave5, 6e0phy5. Re5. Letter5, 17, 1195-1198.
2han9, J., Ander50n, J.6., K1n9, 6., Pr1e5t1ey, K., and R. R061n50n (1989). Later after5h0ck5 0f the March 2, 1987 Ed9ecum6e, New 2ea1and, earth4uake, E05, 70, n0. 43, 1210.
C-48
2han9, P., Ma0, F., and D.8. 51emm0n5 (1989). 6e0metry and d15p1acement 0f the 5urface rupture 20ne a550c1ated w1th the 1954 D1x1e Va11ey, Nevada, earth4uake, 5e15m. Re5. Letter5, 60, 30.
2han9, P., M01nar, P., 8urchf1e1, 8.C., R0yden, L., Wan9, Y., Den9, Q., and F. 50n9 (1988). 80und5 0n the H010cene 511p rate 0f the Ha1yuan fau1t, n0rth-centra1 Ch1na: Quaternary Re5earch, 30, 151-164.