THE PORT HUENEME EARTHQUAKE OF MARCH 18, 1957 By G. W. HOUSNER and D. E. HUDSON

Transcription

1 Bulletin f the Seismlgical Sciety f America. Vl. 48, pp April 1958 THE PORT HUENEME EARTHQUAKE OF MARCH 18, 1957 By G. W. HOUSNER and D. E. HUDSON ABSTRACT The Prt Hueneme earthquake f March 18, 1957, was the first recrded strng-mtin earthquake fr which the grund mtin cnsisted essentially f a single pulse. Since all the energy f the earthquake was cncentrated in ne pulse, the grund acceleratins and the respnse spectrum values were cnsiderably larger than fr mre typical Pacific Cast earthquakes f equivalent magnitude. These abnrmally high values are reflected in damage reprts, which indicated an unusual amunt f damage fr a shck f magnitude 4.7. VENTURA PORT HUENEME EARTHQUAKE N OF 18 MARCH 1957 MLES ' US lol OXNARO \ % U.S.C.G.S. "~,t,,~,,f ST R ON G MOTON p R T'~"'x.A. CGELEROMETER,HUENEME ~ ~_~ ~ 19 15' (~ EPGENTER... -~.~... P 'Nu Fig. 1. Lcatin f epicenter and strng-mtin accelermeter. THE EARTHQUAKE at Prt Hueneme, Califrnia, n March 18, 1957, had certain unusual features that are f interest t engineers cncerned with earthquakeresistant cnstructin. Accrding t Prfessr C. F. Richter f the Seismlgical Labratry f the Califrnia nstitute f Technlgy, in Pasadena, the shck was rated at magnitude 4.7 and the epicenter was lcated at 34 06' N, ' W, with prbable accuracy f five miles. A United States Cast and Gedetic Survey strngmtin accelermeter was apprximately five miles nrthwest f the pint just mentined, as is shwn in figure 1. The maximum recrded grund acceleratin was 18 per cent g, which was exceptinally large fr an earthquake f magnitude 4.7, and this was reflected in the damage reprts, which were relatively numerus fr a shck f this magnitude. The smallest earthquakes fr which strng-mtin recrds Manuscript received fr publicatin January 6, [ 163 ]

2 164 BULLETN OF THE SESMOLOGCAL SOCETY 0~ AMERCA f interest t structural engineers had previusly been btained were f magnitude 5.3. n figures 2 and 3 are shwn the recrded grund acceleratins and the integrated grund velcities and grund displacements. The initial acceleratin pulse in the east-west directin was prly defined n the accelergram and its amplitude was therefre determined by the requirement that the velcity and displacement shuld g t zer at the end f the shck. t is shwn in figures 2 and 3 that the strng ACGELERATON TME z~ ~ ~: ~-.~, io B R AGOELERATON ~ /'~ h TM p VW/ V,5,sE.- - ~ (~ ~ SEG, -'Ld >" d VELOCTY ~ VELOOTY J ~ ,0 > SEO. ~ ~n [ ~J ~ f'~l '~ " ~ los, -,5 20 P - ~J ~ \ SEO..; "~-J.~-~,.'5 a ~ =,,, L ~- L SEO. '.,. L 0'----'--~ <J SEG, ~ k3 (3 F RESPONSE OF ~)~;~ (~l~ J ~ 2.48 SEG. PEND. ~_ RESPONSE 2 OF ~- z SEG 2.0 (3 L XJ Fig. 2 (left). East-west cmpnents f grund mtin. Fig. 3 (right). Nrth-suth cmpnents f grund mtin. grund mtin cnsisted essentially f a single displacement pulse with its crrespnding acceleratins. The traces f the cmbined hrizntal grund mtins are shwn in figure 4, where it is seen that the grund displacement cnsisted f an almst straight thrw in the nrthwest directin fllwed by a slightly curved path back t the riginal psitin. The grund mtin was thus surprisingly simple, fr the usual strng-mtin recrds shw a much mre cmplicated mtin. The strng mtin f the Prt Hueneme shck lasted apprximately ne secnd and cnsisted f three acceleratin peaks. This may be cmpared with a mre typical recrd such as the Glden Gate Park accelergram f the March 22, 1957, San Francisc earthquake. This earthquake had a magnitude f 5.3 and its epicenter was 6 miles frm the accelermeter. The strng mtin lasted fr 3 t 4 secnds and there were twenty acceleratin peaks with a maximum acceleratin f 12 percent g. This San Francisc earthquake was itself f unusually shrt duratin, as cmpared with the

3 PORT ttueneme EARTHQUAKE OF 2~ARCtt 18, t 20 secnds duratin f many Pacific Cast strng-mtin earthquakes. ts recrd has been selected fr cmparisn since f all the available strng-mtin recrds this ne mst nearly matches in magnitude and distance the Prt Hueneme earthquake. Recrded alng with the strng-mtin aceelermeter traces were the respnses f tw pendulums having perids f apprximately 21/~ secnds. The recrds prduced by these instruments are shwn in figures 2 and 3. t shuld be nted that the scale fr these 2~ secnd pendulum respnses differs by a factr f 2 frm that used fr the integrated displacement curves. Cmparing the recrded respnses f the 2~ secnd pendulums with the integrated velcity and displacement curves f 2: fig ACCELERATON VELOCTY DSPLAGEMENT Fig. 4. Cmbined hrizntal grund mtins. figures 2 and 3, it will be nted that the 2~ secnd instrument measures smething in between the velcity and the displacement. t can thus be cncluded that the pendulum perid f 2~ secnds is nt sufficiently large as cmpared with the perids invlved in the present grund mtin fr the instrument t serve as a displacement meter. Judging frm the grund displacement shwn in figure 4, the slip tk place alng a fault riented N W. Assuming that the present earthquake is assciated with typical large-scale grund mtins in the Califrnia regin, which are knwn t invlve a slw drift f the western sectin f the state in a nrthwesterly directin, it wuld seem that the accelermeter must have been suthwest f the fault line. Since the accelermeter was clse t the shre line, the mtin it recrded riginated at the fault and traveled ut under the cean, whereas the mtin that prduced damage inland riginated n the ther side f the fault and was nt recrded. As a further means f cmparing the present earthquake with past Pacific Cast strng-mtin earthquakes, respnse spectrum curves fr the tw cmpnents have been calculated. 1 Figures 5 and 6 shw the relative velcity respnse spectrum curves, and figures 7 and 8 give the acceleratin spectrum curves. 1 D. E. Hudsn, "Respnse Spectrum Techniques in Engineering Seismlgy," Prc. Wrld Cnference n Earthquake Engineering, Earthquake Engineering Research nstitute, San Francisc, 1956; G. W. Husner, R. R. Martel, and J. L. Alfrd, "Spectrum Analysis f Strng-Mtin Earthquakes," Bull. Seism. Sc. Am., Vl. 43~ N. 2 (April, 1953).

4 166 BULLETN OF THE SESMOLOGCAL SOCETY OF AMERCA to hi O~ ' >- - ra to 0 _ l.ij =) VELOCTY RESPONSE SPECTRUM PORT HUENEME- 18 MARCH 1957 WEST DAMPNG = 0,2.5,5,10,15,20% / CRTCAL , UNDAMPED NATURAL PEROD-SECONDS Fig. 5. East-west cmpnent. Maximum relative velcity respnse spectrum. to Ll 2.0 " 1.6 '-- h ).- ~ _J w > 0.8 VELOCT'Y RESPONSE SPECTRUM PORT HUENEME - 18 MARCH 1957 SOUTH DAMPNG = 0,2.5,5,10,15,20% CRTCAL x,,~ 0.4 (.D ,0 1,5 ~_ UNDAMPED NATURAL PEROD-SECONDS Fig. 6. Nrth suth cmpnent. Maximum relative velcity respnse spectrum.

5 PORT /:UENEME EARTHQUAKE OF :MARCH 18, ~ 0.4L - n.- tu w,d e,d '< 0.2 f ACCELERATON RESPONSE SPECTRUM PORT HUENEME-18MARCH 195"7 WEST DAMPNG= 0,2.5,5,10,15,20% CRTCAL UNDAMPED NATURAL PEROD-SECONDS Fig. 7. East-wes~ cmpnent. Maximum abslute acceleratin respnse spectrum. 0.5 z 0.4 l- re"._ "' 0.3 Ll D (D O.2 x UNDAMPED NATURAL PEROD-SECONDS Fig. 8. Nrth-suth cmpnent. Maximum abslute acceleratin respnse spectrum. 5.0

6 168 BULLETN OF THE SES~0LOGCAL SOCETY OF AMERCA The shapes f the spectrum curves d nt differ markedly frm thse f mre typical earthquakes, but the rdinates are suprisingly large fr a shck f 4.7 magnitude. Fr example, the rdinate f the 20 per cent damping curve is apprximately 0.8 fr the nrth-suth cmpnent and 0.4 fr the east-west cmpnent. These are cmpared in the fllwing table with spectra f the San Francisc earthquake f March 22, Epicentral Magultude distance 20 per cent S~ (miles) (ft./sec.) Prt Hueneme ± 0.8; 0.4 Glden Gate Park ; 0.18 State Building ; 0.22 Alexander Building ; 0.12 These figures shw that if the energy released by an earthquake is cncentrated in a single displacement pulse the effect n structures will be much mre severe than if the character f the grund mtin is like that f the San Francisc shck f March 22, Single-pulse earthquakes have been recrded by seismlgists befre this, but these were shcks f small magnitude. The Prt Hueneme accelergram is the first strng-mtin recrding f a single-pulse shck. t is thus nw knwn that such exceptinal strng grund mtins can ccur, at least fr shcks up t 4.7 magnitude. This raises the questin whether it is pssible that shcks f larger magnitude culd be f this type, fr if a shck f magnitude 6 r greater culd be f the singlepulse type, it wuld require a revisin f engineering thinking with respect t pssible intensities f grund mtins. On the basis f past experience it wuld appear that the prbability f a strng single-pulse shck is relatively small, and hence engineers shuld cntinue t be guided by typical earthquake grund mtins. The pssibility f strng single-pulse shcks must, hwever, be kept in mind. EARTHQUAKE ENGNEERNG ]~ESEARC- LABORATORY~ DVSON OF ENGNEERNG~ CALFORNA ]NSTTUTE OF TECHNOLOGY~ PASADENA~ CALFORNA