ML20151H116
| ML20151H116 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 04/12/1988 |
| From: | Algermissen S INTERIOR, DEPT. OF, GEOLOGICAL SURVEY |
| To: | |
| Shared Package | |
| ML20151H012 | List:
|
| References | |
| NUDOCS 8808010148 | |
| Download: ML20151H116 (9) | |
Text
.
F G
ESTIMATION OF GROUND SHAXING IN THE PACIFIC NORTHWEST S. T. Algermissen U.S. Geological Survey
[
Denver, Colorado INTRODUCTION The expected ground idotion in the Pacific Northwest has been estimated in U.S. Geological Survey studies as part of national probabilistic ground motion maps produced in 1976 (Algermissen and Perkins, 1976), 1982 (Algermissen and i
1 otherr,1982) and in a revision of the 1983 national probabilistic maps to be I
published this year (Algermissen and others,1988).
Only a 50 year, 10 l
percent probability of exceedance acceleration map was produced in 1976.
In 1982, acceleration and velocity maps (in rock) were developed for periods of time of interest (exposure times) of 10, 50 and 250 years, with a 10 percent 3
chance of exceedance.
The 1982 maps were produced using mean values of ground motion attenuation and fault rupture length.
The national probabilistic hazard maps to be published this year are based on the probabilistic model, seismic source zones and attenuation used in the 1982 maps but they include parameter variability in attenuation and in fault rupture length.
4 The seismic source tones used in the 1982 and 1988 maps are identical and 1
are shown in Figure 1.
/1gure 2 shows the 50 year, 10 percent chance of exceedance acceleration map produced in 1982 for the Pacific Northwest.
Figu e 3 shows the 50 year acceleration map recently prepared (1988) that includes parameter variability.
MATURE OF THE SEISMIC HAZARD The earthquakes important in seismic analysis for the Puget Sound area (1) The possibility of a large subduction zone shocks (2) the recurrence are:
of historical earthquakes up to magnitude 7.1 that have caused damage in the i
Puget Sound area and occur at depths of about 40-70 km: and (3) the possible J
occurrence of damaging shallow shocks.
1 Recently, Heaton and Kanamori (1984) have suggested the possibility of very large, shallow subduction zone earthquakes at the Juan de Fuca-America plate boundary.
No historical large plate boundary earthquakes are known in j
this region but paleeseismic data are emerging to support this view (Atwater, 1987).
Historically, all of the recent damaging earthquakes (1939, 1946, 1949, '965) are believed to have occuared at depths of 40-70 km either within a region of bending of the subducted Juan de Fuca plate or near the plate l
interface.
l Very little attention has been given to the possibility of a large M =7.0, shallow earthquake, even though one is believed to have occurred.
3 1
There is other evidence of recent significant shallow activity.
Evidence of the occurrence of an earthquake in 1872 east of the Cascades with a magnitude
- (
has been extensively reviewed by a number of l
of appaoximately 7.0 Ms 1
investigators, most recently by Hopper and others (1982) who believe that the i
i 8808010148 880615 I
PDR ADOCK C5000344 s
P PDC i
7 j
'(
earthquake was located near Lake Chelan, Washington and had a shallow focus.
I Other recent significant shallow activity has occurred in the Elk Lake (Grant and others,1984) and Goat Rocks (Zollweg and Crosson,1981) areas of Washington, and there is evidence of Holocene faulting west of the Hood Canal (Gower, 1978). The sources of uncertainty in earthquake origins is summarized in Table 1.
Table 1.--Uncertainties in ground motion hazard assessment in the Puget Sound area J
Hypothesis Evidence 1
Very large plate boundary No known historical evidence but possible j
earthquakes M -8.5 - 9.0 paleoseismic evidence from recent studies w
3 might occur.
(Atwater, 1987). Conflicting views regarding the rate of subduction of the Juan de Fuca plate and the accumulation of strain. Ground motion attenuation relationships for such an earthquake not well known.
Large, shallow (H =7.0)
Evidence of an H -7.0 shock near Lake Chelan s
3 1
earthquakes might occur in 1872 but location and magnitude very 4
)
onshore.
uncertain (Hopper and others, 1982).
Evidence of Holocene faulting west of the Hood Canal (Gower, 1978). Very limited available seismotectonic or seismological
(
data to identify possible source areas of j
large shallow shocks.
l Large (M =7.0) earthquakes Well-documented historical shocks.but s
occur at depths of 40-70 km.
the possible spatial distribution is uncertain.
J 4
The portion of the 1982 acceleration map shown in Figure 2 and the recently prepared (1988) acceleration maps shown in Figure 3 represent a more conservative modeling of shallow earthquakes in the Puget Sound area than was
{
taken by Algermissen and Perkins (1976).
For the national ground motion maps developed in 1982, 25 percent of the earthquakes with H magnitudes greater than 6.5 were assumed to occur at shallow depth. The choice of 25 percent was s
very arbitrary. All large shocks were assumed to occur at depths of 50 km in I
the development of the 1976 national map.
None of the maps (1976, 1982, 1988) consider the possibility of a subduction zone earthquake. Thus, there is considerable uncertainty in probabilistic ground motion assessment in the Puget Sound area becaum of the difficulty in quantifying the occurrence of large subduction zone earthquakes and the occurrence of damaging shallow g-shocks.
1 F
Approximate calculations of the effect of the occurrence of a large (H =8.5) earthquake at two possible locations in the subduction zone were w
undertaken in an attempt to roughly estimate the effects of such an earthquake in the Seattle urban area. The position of these postulated earthquakes are shown in Figure 1.
The fault rupture length assumed in both cases is 225 km, although the rupture length is not critical for the present discussion as long as it is at least 100 km.
The offshore earthquake is assumed to occur at a l
depth of about 20 km.
The earthquake beneath Puget Sound is assumed to occur at a depth of about 50 km.
In both cases, calculations of probable ground motion were made assuming average recurrence times for the earthquakes of 300, 600 and 900 years. Each earthquake, in turn, was included in the probabilistic model used to compute the 1982 national maps and in the 1988 model which included parameter vzriability in attenuation and fault rupture length. The results give only a general idea of the influence on expected ground motion since the subduction zone earthquake was only approximately i
modeled and since our future work in modeling ground motion in the Pacific Northwest will include a careful modeling of all source zones, new attenuation relations, etc. Nevertheless, the results are interesting. A large subduction zone earthquake offshore (Figure 1) would not appreciably affect the 50 year acceleration or velocity at Seattle for average recurrence times of 300 through 900 years. The accelerations and velocities (in rock) computed for a 50 year exposure time and 10 percent chance of exceedance at Seattle did not appreciably differ from the values shown in Figures 2 and 3 (for the 1982 l
and 1988 models).
4 It should be understood that the occurrence of a large subduction zone earthquake would in all probability causo damage in Olympia, Tacoma and 4
i Seattle.
Damage would most likely occur to unreinforced masonry of any height and selectively to other buildings principally in the range of 5-25 stories in I
height.
The important point is that the ground motion caused by such an earthquake with an average recurrence time of greater than 300 years doesn't l
l contribute significantly to the expected peak ground motion in a 50 year period.
The principal damaging ground motions contributing to the 50 year peak motion appears to be associated with the rather frequent occurrence of l
shocks in the range 6.0<M <7.5 that occur beneath the three urban areas w
considered at a depth of about 50 km (for example in 1939, 1946, 1949 and 1965).
i A large subduction zone earthquake at a depth of 50 km beneath Seattle would not appreciably change the 50 year, peak acceleration or velocity at r
Seattle for an average recurrence time of 900 years but it would appr oximately
)
double the peak velocity at Seattle for an average recurrence time of 300 i
years for a subduction zone earthquake.
It should be understood that the suggested doubling of the 50 year velocity at Seattle is only a rough approximation since no entirely suitable attenuation relations are available for such an earthquake.
DISCUSSION i
The probabilistic model for the estimation of ground motion in the Pacific Northwest must be considered very tentative at present since two of i
]
the three major poasible sources or rates of damaging earthquakes (large
(
subduction zone shocks and shallow damaging shocks) are not very well l
unde rs tood.
Much new paleoseismic research data are becoming available i
1 3
i i
{[
concerning recurrence rates for subduction zone shocks but very little has N
been done to resolve the problem of possible damaging shallow shocks. These uncertainties in defining the sources of earthquakes and the emergence of new data on attenuation of subduction zone earthquakes (for example from studies of the 1985 earthquakes in Chile and Mexico) guarantees that probabilistic ground motion models for the Pacific Northwest will undergo extensive revision ov,er the next few years.
PROPOSED RESEARCH PROGRAM FOR THE PACLFIC NORTHWEST The USGS urban hazards program in the Pacific Northwest currently underway will address a number of important problems critical to earthquake hazard analysis and to the estimation of expected gaound motion. The strategy is outlined briefly here:
1.
Incorporation of paleoseismic evidence for large subduction zone earthquakes into probabilistic ground motion models Major paleeseismic studies are underway to investigate the possibility of subduction zone earthquakes. These studies are crucial to disaster preparedness, but the contribution of these earthquakes to the expected maximum ground motion appears to depend heavily on the location of the shock (at least for exposure times <50 years ).
2.
Improvement in attenuation relations Significant new data on the attenuation of seismic waves froc large, subduction zone earthquakes has become available recently (for example, as a result of the 1985 earthquake affecting Central Chile and Mexico City). These data need to be considered in developing a new model for ground motion hazard assessment.
3 Site response Recent large earthquakes have clearly shown the importance of site response in ground motion assessment. New models for ground motion assessment developed in the current USGS urban hazard program will include site response in the urban areas of the Puget Sound region i
and will be based on the site response data currently being recorded and analyzed (Xing and others, presentation at this meeting).
4 Shallow earthquakes Attention needs to be given to the problem of the possibility of damaging shallow earthquakes in the area. Shallow earthquakes of quite modest magnitude (for example, the 1986. M 5.5 earthquake that 3
occurred at a depth of about 5 km beneath the city of San Salvador, i
cousing major damage to the city) could cause great damage.
j 4
t I
(
n 4
I
P
SUMMARY
The development of ideas concerning the assessment of earthquake ground motion in the Puget Sound area has been outlined and areas where significant improvements in these assessments are believed possible have been suggested.
These areas of uncertainty and opportunities for improvement in hazard as,sessment are currently major research efforts in the USGS urban hazards program underway in the Puget Sound area.
Success in this research will lead to much improved probabilistic models for the assessment of ground shaking and in estimates of ground shaking.
l REFERENCES _ CITED Algermissen, S. T., Perkins. D.H., Thenhaus, P. C., Hanson, S. L., ard Bonder,
B. L.,1982, Probabilistic estimates of maximum acceleration and velocity in rock in the contiguous United States:
U.S. Geological Survey Open-File Report 82-1033, 99 p.,
6 pl.
Algermissen, S. T., and Perkins, D. M., 1976, A probabilistic estimate of i
maximum acceleration in rock in the contiguous United States U.S.
Geological Survey Open-File Report 76-416, 45 p.
Algermissen, S.
T.,
Perkins, D.
M., Thenhaus P. C.
Hanson, S. L., and i
Bender, B.
L., 1988. Probabilistic estimates of maximum acceleration and velocity in rock in the contiguous United States (including parameter availability), U.S. Geological Survey Miscellaneous Field Inv. Map (in preparation) 1:5,000,000, 4 sheets.
Atwater, B.
F., 1987, Evidence for great Holocene earthquakes along the outer coast of Washington state: Science, v. 236, p. 942-944.
Gower H.
D., 1978. Tectonic map of the Puget Sound region, Washington:
U.S.
Geological Survey Open-File Report 78-426, 22 p.
t Grant, W. C., Weaver, C. S., and Zollweg, J. E., 1984, the 14 February 1981 4
l Elk Lake, Washington, earthquake sequence: Seismological Society of
~
America Bulletin, v. 74, p. 1289-1309.
Heaton, T.
H., and Kanamori, H.,1984, Seismic potential associated with subduction in the northwestern United States: Seismological Society of j
America Bulletin, v. 74, p. 933-941.
Heaton, T.
H., and Hartzell, S. H. 1986, Source characteristics of hypothetical subduction earthquakes in the Northwestern United States:
Seismological Society of America Balletin, v. 76, p. 675-708 Hopper, M. G., Algermissen, S.
T., Perkins, D.
M., Brockman, S. R., Arnold, E.
P.,1982. The earthquake of December 14, 1872, in the Pacific Northwest:
Program of the Annual Seismological Society of America Meeting.
1 1
!I a
Weaver, C.
S.,
and Smith, S. W., 1983, Regional tectonic and earthquake hazard
(
1mplications of a crustal fault zone in southwestern Washington Journal of Geophysics Research, v. 88, p. 10371-10383 Zollweg, J. E., and Crosson, R. S.,1981, The Goat Rocks wilderness, Washington, earthouake vf 28 May 1981: EOS, v. 62, no. 45, p. 966.
6
125 120
/
I 4" '
/
(
G,)
~~~
.a t.
lympi i
I l
I I
o
)
45 s
(
8 kh
~
~ ~ ~ ~ ~ ~..
T 4
/
4 i
ry l
'N I
I I
40 '
/
Seismic Source Zones in the Pacific Northwest
?
290.__
490 100 su Kil' meters o
Figure 1.
Selsmic source zones used in the developnent of national probabilistic ground motion maps in 1902 (Algermissen and others,1982)
I and in 1988 (Algermissen and others, 1988).
The two heavy black lines superimposed on the source zones represont f aults used to model the locat f on of subduction zone earthquakes discussed in this paper.
k
~
125 120 l'
(
Sg 2
l 9
9 b Se attle 6t ol la
)
gl l
/
45 %
Y I
- O_s, 6
7 i
8
<4 (10
~
14 4
l 4
N 5
62' A7]b in C2nb 1982 50-Year Acceleration Map t._. i 890 _ _. 90 d90 3
Kilomelet s
~
~
Figure 2.
Pacific Northwest portion of the 50 year, exposure time 10 percent chance of exceedance, map of maximum acceleration In rock (Algermissen and others, 1982).
{
f 8
I
('
dw
~
s, 6
l o '-
3 :'.
'y k
5 9 t Ll" -
\\
~'
k,..
\\
w ymp f'p '. e
~~
l n
4_
in to
- ~
10
,,,, ' 'm
^-
16 g6
--- - # ~
I
3
..,s...
'N, p,
r D
s.. '
/,
7
\\
E 4
.4 s
'E
]
/
u a
(js2 j f F ~.
7.s A..&
B O
'O s
1982 50-Year Acceleration Map with Parameter Variability (1988)
L_.J?
890._ ??o d?o Kilom e le t e Figure 3 Pac!rto Northwest portion or a 50 year, to percent chance or exceedance, map or max! mum acceleration in rock that is under develo;qent at the present time ( Algermissen and others, 1988).
The model is exactly the same as used to produce the map in Figure 2, except that variability is included in attenuation and fault rupture length.
The variation in attenuation of log, or acceleration (7 ) 1s 0.62 and in log o of fault 4
g rupture length (Tg) is 0.52.
(
9
__ __ ______ - _ _