ML20151H017
| ML20151H017 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 04/12/1988 |
| From: | Algermissen S, Leyendecker E INTERIOR, DEPT. OF, GEOLOGICAL SURVEY |
| To: | |
| Shared Package | |
| ML20151H012 | List:
|
| References | |
| NUDOCS 8808010108 | |
| Download: ML20151H017 (11) | |
Text
.
ESTIMATION OF POTENTIAL EARTHQUAKE LOSSES IN THE PUGET SOUND, WASHINGTON AND PORTLAND, OREGON AREAS S. T. Algermissen and E. V. Leyendecker U.S. Geological Survey Denver, Colorado 14SS STUDIES The U.3 Geological Survey (including studies by personnel from other agencies transferred to the USGS) has also, for a number of years, carried out research on the estimation of monetary losses and casualties asnociated with earthquakes in the United States.
These personnel are currently in the Branch of Geologic Elsk Assessment.
Traditionally, leading experts in earthquake engineering and damate analysis from outside the Survey have participated in these studies and contributed greatly to them.
Investigations have ranged from monetary loss estimates for various regions and urban areas of the United States to disaster preparedness and mitigation studies of four large metropolitan areas San Francisco (1972), Los Angeles (1973), the Puget Sound Area (1975), and Central Utah (1976).
The data base for these and subsequent earthquake loss investigations,was developed from field investigations and scientific and engineering reports of losses associated with a number of significant foreign as well as United States earthquakes.
Most of the resulting published loss estimates by the.
USGS for postulated events have had limited specific ob,lectives such as federal insurance related studies and regional disaster response planning. ~
f list of significant USGS risk publications is included in the "References."
A The conclusion that emerges from this brief summary of the very broad and comprehensive program of earthquake hazards investigations over a period of years is that the results of a tremendous amount of very valuable research is available for application to the present urban hazard study.
The major problem is to design a program that builds on these research results and applies them in a meaningful way to the cencerns and needs of the area.
As previously mentioned work on earthquake losses tu the Puget Sound area is not news however, much has been learned since the 1975 study (Hopper and others, 1975) was published.
These new loss studies are being done both to develop improved estimates of the consequences of earthquakes and to j
develop improved techniques for conducting such studies.
l Earthquake loss studies may be of the "scenario" or "probabilistic" type.
A scenario study might consider the consequences of one or more earthquakes:
frequently this might be the largest likely earthquake. Such a study is very useful,for purposes such as etargency planning.
The probabilistic study considers both magnitude and frequency of occurrence and may be 'more realistic" than the above "worst case" scenario.
Both the scenario and 1
probabilistic types of study are being used in current efforts.
Emphasis is being placed on developing the tools for use in loss studie3 rather than a complete assessment for the region.
Two important types of results will be emphasized in the loss studies:
Estimation of catastrophe potential and of
(
average annual loss.
[CK05o99344 Wd6 f5 P
PDC o
1 r
Catastrophe Potential
(
The estimation;of catastrophe potential requires the estimation of the probability of occurrence and the losses resulting from large earthquakes in the region.
Estimation of catastrophic potential is a three stage process involving:
1 estimation of the probab11P y magnitude distribution of large shocks for each region 2.
estimation of the probability that these larga shocks might occur in a given time period 3.
estimation of the losses (economic and casualty) associated with these earthquakes.
It it well understood that estimates of catastrophe potential based on the historic record of earthquakes in an area may be poor since large earthquakes are rare events.
Because of the emergence of palcoseismic studies (identification and dating of evidence of major earthquake occurrences in the geologic record) it is now possible to extend the record of earthquakes in many arres and thus impenve the estimation of I
both the eagnitude of the largest shock likely to occur and the probability that it will occur.
Thus, the efforts in the Puget Sound area to leprove our understand 1tig of large subduction :ene earthquakes beco,mes important for the estimation of catastrophe potential.
Figure 1 is a schematic diagram showing the elements required for the estimation of.
economic 1csses and casualties (risk analysis) associated with any large regional earthquakes postulated for the estleation of catastrophe potential.
Once the earthquake magnitude associated with a level of probability of occurrence considered critical has been determined, a i
deterministic or "scenario" type loss study can be done for the eaathquake of interest.
Average Annual Loss l
The esti:ation of average annual loss is an important type of loss estimation since it estaulishes the long-term risk in the area.
Essentially, the est!=ation of catastrcphe potential is a subset or p4rt
{
i i
of the determination of average annual loss and the result of the research on catastrophe potential already discussed can be incorporated into
(
i average annual loss.
A number of techniques have been used by USGS (and i
others) to esticate average annual losses.
For exa:ple average annual l
losses can be dpproximated by:
j
)
1.
Steulation of the ground motion associated with the historical record i
of earthquakes in an area. An example is the computation of average 3
j annual loss to dwellings in the San Franetsco Bay area using three j
different time windows of historical earthquakes (Algermissen and i
Steinbrugge. 1976).
i 2.
Simulation of ground motion using the expected ground motion obtained j
from probabilistic seismic hazard analyses (PSHA).
In PSHA studies the shaking history together with the cumulative distribution function 1
(CDF) of ground motion is calculated for each site.
Thus the
{ g distribution of ground shaking at every location of interest can 1
J i
2 1
i
easily be calculated as a byproduat of PSHA.
For example, if the attenuation of percent los (replacement cost) away from an earthquake I
source is known or can be developed. the distribution of percent losses or e.oromte loss (if property values are known) can be determined directly as a byproduct of PSHA.
The use of the attenuation of percent loss away from an earthquake source has been suggested by Steinbrugge, Algermissen and Lagorio (1965).
Probabilistic ground motion maps of the United States have been prepared by Algermissen and Perkins (1976) and updated and revised by Algermissen and others (1982).
Thus the calculation of average annual loss, probability of a certain level of loss or the level of losses for vcrious levels of probatility and exposure tine can be obtained as an additional product of any probabilistic seismic hazard mapping p.*ogran undertaken.
4 A limited area of Seattle, shown in figure 2, has been selected for conducting a detailed study of losses.
It tricludes Vest Seattle, an area with l
considerable damage in 1965, and downtewr. Seattle.
It extends north to about Salmon Bay and Lake Union and south to abcut the Boeing Plant.
This area will be used as a demonstration of some techniques which can be used to conduct a loss stud /.
Most of the work is being done by USGS personnel.
Some work in losses that is being done under the USGS grants program is close11 related and will be included in the study.
within the scope of the grant propvsals.These efforts will be coercinated when it is DATA COLLECTION i
- The inventory of structures, except for housing, has been the weakest link'in developing realistic estimates of total dollar less by building type.
The data collection procedures are centered around (1) a steple system for classifying buildings (2) th? census tract as the basic area for data collection, and (3) machine read "mark-sense" sheets for compiling a comput(r data base.
Each of thase factors is described below.
It should be noted that j
tne procedure can be used for either inventory or damage surveys.
The 1983 Insurance Services Office (ISO) class 1T1 cation system described by Steinbrugge (1982) will be used in the inventory survey of buildings in the census tracts.
This classification uses a systea of 5 eajor classes as shown belcwn Class 1: Wood frame structures Class 2: All-eetal buildings class 3: Steel fraee buildings Class 4: Reinforced concrete buildings, combined reinforced concrete and structural steel buildings Class 5: concrete, brick, or block buildir,s This system was selected for its relative simplicity and ease of use by a lay person with a limited amount of training.
I The census tract will be used as a data collection area in order to simplify the inventory of buildings.
Since census data provide a relatively accurate count of residential construction, this was one component of an inventory that does not have to be complied in detail.
Other types of j
(
3 l
structures require additional inventory work.
These data will be compiled by sampling census tracts.
Data will be collected on "mark-sense" sheets which describe building class and various types of damage if it is a damage survey.
The Elrk-sense sheets are preprinted forms with n.1tiple choice responses that are fillec out with a soft lead pencil.
The "marks" by the soft lead pencil can be "sensed" by an optical scanning device.
Space is also provided for general cor.ments and/or observations.
The presence of this latter type of data can be "sensed" a'.though it must te entered by hand.
These forms are then read into a rc puter data base using an optical scanner.
This process is illustrated in Figure 3 STRONG CROUND MOTION Most loss studies h?ve used MMI as the measure of ground moti.n. Although the limitations of intensity for this use are well recognized, it is also well recognised that there is little information on building damage as a function of some other measure of ground motion.
Thus the use of MMI is expected to continue for some time.
The current werk on ground motion, intensity, and local ofreets by King (King and others, 1958) will t.e included.
DAMAGE DATA As would be expected, the largest data base for earthquake damage exists for California.
Tnese loss data have been categorized by building class and streng ground motion amplitude.
The classification of building stock described earlier has been used to compile existing damage statistics (Steintrugge, 1952).
Of necessity, the MMI scale has been used as a esasure of ground motion.
Due in part to the lack of widespread instrumentation it is anticipated that the use of MMI will continue for some tite.
These data have been cautiously extrapolated for use in other areas of the United States.
Efforts have been made to expand these data for California using an expanded classification systen using expert opinion to create damage statistics (Applied Technology Council, 1985).
FIELD USE The procedu*e fo* data collectlen already discussed was used follcWing the October 1,19BS Whittier Narrows Earthquake.
Census tracts surveyed in Whittica are shown in Figure k.
The open circles represent the coordinates of the center of hou:ang in each tract.
Boundaries of the tracts are omitted for I
simplicity.
The survey procedures worked relatively well.
However, it was concluded that the specific data recorded on the mark-sense sheets, while avequate, should te sleplified as much as passible for use in future i
studies.
i There were also indications that minor modifications were desirable in the building classification system but major changes are not anticipated in the near term.
l 4
SUMMARY
f Procedures for loss studies in the Puget Sound area have been briefly described.
The building classification and survey strategies described have been tested and found usable &nd adequate for describing damage.
Further refinement in the direction of sleplification was found desirable for future use by personnel with training in its use but with limited background in structures.
REFERENCES CITED AAD 0 ELECTED BIBLIOGRAPHY Algernissen, S.T.,1972, Seismic hazards reduction in Alaska:
The great Alaska earthquake of 1964, seismology and geodesy National Academy of Sciences, Washir.gton, D.C.,
- p. 553-556.
Alger issen, S.T., and others: Steinbrugge, K.V., (principal consultant) and others,1973, A study of earthquake losses in the Los Angeles, California area:
U.S. Department of Conceree, NCAA, 331 p.
Algereissen, S.T., McGratn.
M.B.,
and Hanson, S.L., 1978, A technique for the development of rapid estication of carthquake losses:
U.S. Os. logical Survey Open-File Report 78-440, 31 p. (plus appendix).
Alger:issen, S.1., and Perkins, D.M., 1976, A probabilistic estimate of enximum 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 Bendar, b.L.,1982, Probabilistic esticates of maximum acceleration and velocity in rock in the contiguous United States:
U.S. Geological Survey Open-File Report $2-1033, 99 p.,
6 pl.
Algereissen, S.T.,
Rinehart, W.A.,
- Dewey, J., Steinbrugge, K.V.,
and other consultants,1972, A study of earthquake losses in the San Francisco Bay U.S. Department of Conne ce, NCAA, 220 p.
a*ea Algermissen, S.T, Rinehart, W.A., and Stepp, J.C., 1972, A technique for seistic zoning:
Economic censiderations:
International Conference on Microtonation Proceedings. Seattle, Washington, p. 943-956.
Algereissen, S.T., and Steinbrugge, K.V., 1978. Earthquake losses to buildings in the San Francisco Bay area Internat1eral Conference on Micro:onation for Safer Construction Research and Application, 2nd, 1978, Procee' ings, 13 e.
1951 Estimation of earthquake losses, in Proceedings of Conferer.ee on the Evaluation of Regional Selstic hazards and Risk:
U.S. Geological Su? vey Open-Flae Report 81-437 1964, Seisefe hazard and risk assessment:
Some ease histories:
The Geneva Papers on Nisk and Insurance, vol. 9, no. 30, Association Internationale pour l' Etude de l' Econo:le de L' Asu-ance, Geneve, p. 8-26.
I 5
l r
1 Algermissen S.T., Steinbrugge, K.V.,
and Lagorio H.J., 1976. Estimetion of
(
earthquake losses to buildingr (except singit-family dwellings):
U.S.
Geological Survey Open-File Report 78-441, 146 p. (plus appendices).
Algermissen, S.T., Steinbrugge, K.V., Rinehart W A., and Schnabel, R.E.,
1974 Estimation of earthquake losses in Californiat American Association for the Advancement of Science,1koth Annual Meeting Program, San i
i Francisco, ';alifornia Feb. 24-Mar. 1, 1974, p. 63 Algermissen, S.T., Stepp, J.C., Rinehart, W. A., and Arnold, E.P.,
1969 Studies in seismicity and earthquake damage statistics Appendix B:
U.S.
Department of Ccumerce, ESSA, Coast and Geodetic Survey, 68 p.
Applied Technology Counell, Earthquake Damage Evaluation Data for California, l
ATC-13, 1965, 492 p.
Applied Technolog) Counc10, 1978, Tentative provisions ter the development of seismic reg 41ations for buildings:
National Science Foundation Publication 78-8. 505 p.
Hopper M.G., and Algereissen, S.T.,1964 Tvoes of damage that could result from a great earthquake in New Madrid, M1.'sourt, seismic zone:
U.S.
Geological Survey, MF-1713 Hopper, M.G., Algermissen, S.T., and Dobrovolny E.E.,
1983. Estimation of.
?arthquake effects associated with a great earthquake in the New Madrid seismic :ene U.S. Geological Survey Open-File Report 83-179.
Hopp'er, M.G.,
- Langer, C.J.,
- Spence, V.J., Rogers, A.M., Algerm!ssen, S.T.,
Olsen, B.C.,
Lagorio H.J., and Steinbrugge, K.V.,1975 Study of eatthquake losses in the Puget Sound, Washington area:
U.S. Geological l
Survey Open-File Repert 75-375, 29B p.
Hopper, M.G.,
Thenhaus, P.C.,
Barnhard, L.M., and Algermissen, S.T.,
1983.
Damage Survey in Coalinga, California, for the earthquake of May 2,1963, in Bennett, J.H., and Sherburne, R.W., eds., The 1983 Coalinga, California, earthquakes Publication 66, p. 5-8.
California Division of Mines and Geology Special
- King, K.,
- Tarr, A., Carver, D., Williams, R., a*.c Worley, D.,
Geophysical Studies in Support of Seisnic Hazards Assessment of Seattle and Olyepta, Washington, this volume, 1986.
Leyendecker, E.V., Hignland, L.M.,
Hopper, M.G., Arnold, E.P.,
Thenhaus P.C.,
and Powers, P.S., Early Results of isoseismal Studies and Damage Surveys for the Whittier Narrows Eaathquake, Spectra, Earthquake Engineering Research Institute, 1958.
- Rinehart, W.,
Algereissen S.T., ard Gibtons, Mary,1976, Estimation of earthquake Icsses to single family dwellings:
Available from the National Technical Informa'. ion Service, Springfield, Y!rginia 22161, 195 p.
as Report PB-251 677/AS, t
6 i
RoEers. A.M.,
Algermisse.',,
S.T.,
- Hays, W.W.,
and Perkins, D.M., (geological and seismological pertion): Van Strein. D.O., HuEhes. H.C.
HUEhes, R.C.,
(
Lagorio.
H.J., and steinbrugge, K.V. (engineering analysis portion), 1976 A study of earthquake losses in the Salt Lake City, Utah, areat U.S.
Geological Survey Open-File Report 76-89, 357 p.
SteinbruSEe, K.V.,1978, Earthquake insurance and microzonation:
Proceedings Second International Conference Microzonation, I:203/13 1982. Earthquakes, volcanoes, and tsunamis--An anatomy cf hazards:
Skandia Insurance Group, New York City.
SteinbruSEe, K.V., and Altermissen S.T., 1978. Relationship of Modified Mercalli intensity values to wood frame dwelling earthquake insurance:
Wind and seismic effects:
Proceedings of the Ninth Joint Panel Conference of the US-Japan Cooperative ProEram in Natural Resources NBS special Publication $23, VII-16. VII-28.
i e
t
\\
l 7
I
i t
INVENTORY (Structures, population)
YULNERABILITY 1
HA2ARD Esttrnation of Det ermini s t ic Ground Motion and Probabilistic Geologic llazards l
I 1
i s
LOSS TO ECON 0 Hic CRlilCAL CA$UALTIES l
LOSS FACILITIES
_x-Figure 1.
Ele::ents in seiscie risk analysis.
l 1
f
{
l 1
8
[
(
(
(X w o
L SEATT_I
(
Lake Washington Elliot Bay i
/
\\.5 k\\a I
2 3Mi
, (4 km 0
1 2
l s
8#' 2.
Planned earthquake loss study demonstrat!On area.
i i
i
1
(
i HARDWARE:
/
j DATA PRINTER
' SHEET
~
1 D
L TO VAX l
MODEM l
PC 7
l OPTICAL f*******'***eog l
. SCANNER HP PLOTTER O
~
DIGITIZING TABLET l
' LOCAL STREET MAP l
I l
SOFTWARE:
I 4
j PROGRAMS TO-1
- 1. DIGITIZE LOCAL MAPS f
- 2. GRID, SURFACE, CONTOUR
- 3. DRAW MAP
- 4. COMMUNICATE WITH VAX 4
l l
- 5. DRAW GRAPHS l
- 6. SCAN OPTICAL FORMS i
j Figure 3 Acquisition syste: for inventory develop-ent and earthquake less data.
T 10 l
i
i r
(%
o
- 3 4.t...
~
l 0]
qk _ }'x erzeaal urbank i
'y
,y
-r L -
, Mf /
MDp 7,,, 9 a.x a. 39 s.re m
/
st o g
s UN
]@
))
, %e.,,
Sp
\\ $ v
.o
\\ 0 l
-Q ! hV-jby a
-Q NW]\\
%,,,> :s; y
-33e p
7
./
)
7 c-ton
'esch Q
~
)
l MI 5 0
5 to 15 s
KM 5
o 3
to 15 i
",s e use its2 i t s, o t i r.a i
1 Figure li.
Fre11:tinary reElonal Modified Mercalli intensity isosels-J l
mais in the Los Angeles area for the earthquake of October 1, 1987.
)
Open circles represent the centers of census tracts surveyed.
The circled star is the eain shock epicenter.
The star to th. north-west of it is the epicenter of the largest aftershock.
I i
i i
11 i
-