Observed travel times for Multiply-reflected ScS waves from a deep-focus earthquake A . P . ESPESOSA ( * ) Ricevuto il 26 Agosto 1966 SUMMARY. — Tlie deep-focus Argentinean earthquake of December 8 1962, generated multiply reflected ScS phases which were recorded very clearly at stations of the IGY and the TJSC&GS standardized worldwide networks and at Canadian stations. The data gathered from this earth- quake for the multiply-reflected ScS and sScS were used to construct the travel times and to extend them to shorter epicentral distances. These new data brought to light an error in published travel times for the 2(ScS) phase. RIASSUNTO. — 11 terremoto profondo avvenuto in Argentina l'8 Di- cembre 1962, ha originato onde ScS riflesse più volte, che sono state registrate molto chiaramente dalle stazioni sismiche della rete mondiale standard del- l ' I G Y e d e h ' U S C & G S e dalle stazioni Canadesi. 1 dati raccolti mediante lo studio di questo terremoto per le onde ScS e sScS, sono stati usati per costruire la curva dei tempi di tragitto ed estesi a distanze epicentrali più brevi. Questi nuovi dati, inoltre, hanno messo in luce un errore relativo ai tempi di tragitto già pubblicati per le onde 2(ScS). INTRODUCTION Multiply-reflected ScS and their surface images f r o m the December 8, 1962, deep-focus earthquake in Argentina were unusually well record- e d b y long-period seismographs at stations of the IGY and the TJSG& GS standardized network, and at Canadian stations. This investigation describes a study made on the identification of the multiply-reflected ( * ) Environmental Science Services Administration, U.S. Coast and Geodetic Survey. Rockville, Maryland. Fig. 1. - Map showing location of epicenter, stations, and a great circle path through Palisades. OBS KR V KD T l i A V K L T I M E S j,\>K M U L T I P I . Y - R F.FLF.CTED. ETC. 4 1 7 phases, and the construction of their travel-times as recorded on long- period instruments. T h e observations of the travel-times of the 2(ScS) at shorter cpicentral distances have permitted the extension of the travel- t i m e curve f o r this phase down to an cpicentral distance of 7°. Pre- viously in the literature, the travel-time curve f o r 2(Sc8) has been given f o r an cpicentral distance greater than 90°. T h e n(8cS) f a m i l y of phases has been identified without any doubt b y the presence of the n(s8cS) phases on the recordings. Identification of phases traveling in the m a j o r circle path is shown. I n the process of plotting the ob- served t r a v e l times of the 2(ScS) phase, an error in the published travel- t i m e curve f o r this phase has been found, f o r a depth of focus of 600 k m . INTERPRETATION OF THE DATA T h e epicenter, origin time, and depth of focus f o r the December 8, 1962, earthquake were located b y using Bolt's (1960) program f o r the 7090 IBM computer. One hundred fifty readings of impulsive P phases, .14 pP and 10 PKP readings, were used in the relocation. T h e pertinent data are listed in T a b l e I . Table I Epicenter Origin Time (GMT) Depth Magnitude h m s km 25.8" S 63.4° W 21 27 22.2 620 7 PAS USC&GS 25.78° S 62.15° W 21 27 18.0 580 6 3/4 PAL T h e data used in this study are f r o m the stations shown in Figure 1. T h e solid line in this figure is a great circle path through Palisades, N . Y . , and the station code i d e n t i f y i n g each of the stations corresponds to the three letter code assigned b y the U8G&G8. Readings of the phases which constitute the data were done f r o m long-period recordings having a natural period of 30 sec and 100 sec f o r the seismometer and galvanometer, respectively. F i g u r e 2 shows a graphical description of the phases which have undergone four reflections at the core-mantle boundary, denoted as 418 A . F . E S P I N O S A 4(ScS), through the minor circle path and recorded at SGH. A l s o in the same figure is shown a phase which travels through the m a j o r circle path and has been reflected three times at the core-mantle boundary, denoted as 3(ScS)~, and also recorded at the same station. Pig. 2 - Graphical example of ray paths and notations used for the multiply- reflected phases which have traveled along the minor and major circle paths. F i g u r e 3 shows a tracing of the E-W component seismogram at GAB, w i t h an epicentral distance of 36.24° and an azimuth f r o m the station to the epicenter of 174.61°. T h e original recording was done on a recorder w i t h a 30 mm/min. revolution rate. I n this figure, the tracing has been reduced b y a factor of four. T h e portion of this recording is between the arrival t i m e of the S phase and approximately a minute after the arrival t i m e of the 4(8c8). On the original seismo- gram, the peak-to-peak amplitude of the ScS phase is 357 m m . T h e n(ScS) and n(sScS) are indicated on the seismogram. T h e identifi- cation of the n(ScS) series has been made certain b y the arrival of its surface images on the recordings. O 6 3 7 0 k ( S H ) S S c S 2(ScS) 3(ScS) 4 ( S c S ) 2(sScS) 3(sScS) 4(sScS) MIN C A R E - W ( 3 0 - I 0 0 ) F i g . 3 - Seismogram f r o m G A R , December 8, 1962, E - W long-period component. (SH) s ScS s S c S 2(ScS) 2 ( s S c S ) 3 ( S c S ) 3 ( s S c S ) 4 ( S c S ) 4 ( s S c S ) 1 • - i MIN B E C E - W ( 3 0 - I 0 0 ) T i g . 4 - Seismogram from BEG, December 8, 1962, E-W long-period component. 4 2 0 A . F . ESIM N o f j A A n o t h e r typical sei.sinogr.un tracing is shown in Figure 4. This figure is the E-W component, at, BEG with an epioentral distance of 57.86° and an azimuth of 17*.6" and shows an especially clear recording of the n(ScS) and the nlsScN) series. On this record, the two series are indicated from n - 1 to an order I; however, at, the end portion of this seismograni, the 5(KcS) can be .s<-<-n very clearly. This phase was also well recorded at st.it.ions close to the epicenter, but there was difficulty in picking its arrival time since the background noise is high in that portion of the seismograni. On inspection of the N-iS' component seis- mograni, the, horizontally polarized shear phases are v e r y small in amplitude, and in the vertical component these phases were not discernible. F i g u r e 5 shows a composite of seismograms for nine stations. E i g h t are due, N o r t h of the epicenter, and the ninth is due East. These record- ings were made; on long-period instruments; the E - W component is the one shown in this figure, liiiforfiinatcly, t he variation of speed in the revolution of the drum is different I'or sonic of these stations; some have a 15 irim/min. and others 30 inm/niin. revolution rate. W h e n reduction was done, on each of the seismograms in this composite figure, a one-to-one correspondence was lost. T h e picture length is made up of four parts put together, which can be seen in the breaking of the time axis. This f a c t also has introduced distortion, and because of these distortions, no a t t e m p t is made to draw a travel-time curve which connect one phase with another for a given series. T h e ANT, BAA, and ABE seismograms were omitted in this figure since their locations are too close, to the time, axis. T h e RDJ, GAB, BEG are tracings of the ori- ginals, and the PAL, HAL, SGtt, VIC, BES, and ALE are reduced pictures of the original seismograms at these stations. T h e t i m e axis is in minutes, with an interval of 2 minutes, and the, distance axis is in degrees. TRAVEL-TIME CURVES. T h e seismograms obtained f r o m 38 of the IGY, USC&GS, and the Canadian stations w e r e read (sec F i g u r e 1) and a total of 23 station re- cords were used in the filial gathering of the data which is used to con- struct the travel-time curves f o r the multiply-reflected phases. T o inter- pret the phases on the recordings, the travel times of impulsive phases were picked, tabulated, and plotted. These readings were done on all three 130 3 5 0 3 4 0 3 3 0 3 2 0 310 3 0 0 2 9 0 2 8 0 2 7 0 2 6 0 2 5 0 2 4 0 o K 120 110 100 90 80 70 c 6 0 E - c o L L ) 50 > < < r 40 30 20 10 0 0 2 0 3 0 j 4 0 5 0 . > s X } v u I - < 2 < < CD Q OC cr < o u -J ^ LLI < < m Q- i 1 6 0 7 0 8 0 9 0 100 110 120 x o CO DISTANCE ( d e g r e e s ) ai to o ^ cj c!