"'INEEE.'. I BODY NAVE NAGNITUDES AND LOCATIONS OF FRENCH /

Size: px

Start display at page:

Download ""'INEEE.'. I BODY NAVE NAGNITUDES AND LOCATIONS OF FRENCH /"

Diana Georgina Summers
5 years ago
Views:

1 I-8166 I I 248 BODY NAVE NAGNITUDES AND LOCATIONS OF FRENCH / UNDERGROUND EXPLOSIONS AT TN.(U) ATONIC HEAPONS RESEARCH ESTABLISHMENT RLDERNASTON (ENGLRND) UNCLASSIFIED P D MARSHALL ET AL. NOV 85 URE--12/85 F/G /1I NL "'INEEE.'.

2 V ' 7 L;- 1~ L L.2

3 UK UNLIMITED 9793 " 00 AWRE 0 12/85 - '(.( :-: ATOMIC WEAPONS RESEARCH ESTABLISHMENT AWRE REPORT No. 0 12/85 Body Wave Magnitudes and Locations of French Underground Explosions at the Mururoa Test Site P D Marshall R C Lilwall Penelope J Warburton C, ~JinIAPR * Available from H.M. Stationery OfficeE PRICE 4.00 NET 3AWRE. MOEXPE). Aldermaston, Berks. UK UNLIMITED November 1985

4 UK UNLIMITED -.. ATOMIC WEAPONS RESEARCH ESTABLISHMENT AWRE REPORT NO. 012/85 Body Wave Magnitudes and Locations of French Underground Explosions at the Mururoa Test Site P D Marshall R C Lilwall Penelope J Warburton Aceession For TI-S GRA&I DTIC TAB Unannounced -l Justi ication 13y Recommended for issue by Distribution/ _,. Approved by A Douglas, Superintendent Availability Codes Avail and/or Dist Special.I---! B L Elphick, Head of Division /' UK UNLIMITED

5 UN LIM~ITED ISBN CONTENTS SUMMARY 3 I. INTRODUCTION 3 2. JOINT EPICENTRE RELOCATIONS4 3. DETERMINATION OF MAGNITUDE4 4.SEISMOGRAMS ACKNOWLEDGMENTS 5 REFERENCES 6 TABLES I -3 7 FIGURES MILL]

6 . UN LIMITED SUMMARY Since 1975 the French have conducted underground nuclear weapon tests beneath the Mururoa atoll in the Tuamotu Archipelago in the Southern Pacific. A least squares joint estimate of origin time and epicentre together with a least squares joint estimate of magnitude is presented for presumed explosions in this area. These are based on data taken from bulletins of the International Seismological Centre. Some seismograms, recorded in Canada, are included to illustrate the nature of the P waves from these presumed explosions. 1. INTRODUCTION Basic source information (location, origin time, depth, yield, etc) about underground nuclear explosions is important to seismologists interested in fundamental properties of the structure of the earth, as well as those interested in discrimination between earthquake and explosion from their seismic signals. Numerous scientists have appealed for the release of epicentral details of explosions to aid research programmes (Piullen, 1958, Griggs and Press, 1961, Teller, 1963). In response, Springer and Kinnaman (1971, 1975) published the basic epicentre details for all underground nuclear explosions, US and British, detonated in the USA from 1961 to Numerous yield estimates were also included. The origin time and precise epicentres of French underground nuclear explosions in the Sahara between 1961 and 1966 have been published by Duclaux and Michaud (1970). However no comparable data are available for underground explosions in the USSR or the French Pacific Test Site at the Tuamotu Archipelago. Several international data centres collect seismic wave arrival times from all over the world and compute estimates of the origin time, epicentre, depth and size for seismic disturbances. Bulletins containing these data are published by the US National Earthquake Information Centre (NEIC) in Colorado, USA, and the International Seismological Centre (ISC) in Newbury, UK. A similar service is provided by the Institute of Physics of the Earth in Moscow, but the-- Soviet bulletin does not report data on Soviet nuclear explosions. The French produce a bulletin for their metropolitan and Polynesian seismological networks but the bulletin contains no information on French explosions. To provide seismologists with improved estimates of some of the basic source information about French presumed underground nuclear explosions, seismic wave arrival time and amplitude data collected by the ISC have been analysed using a joint epicentre technique (JED) to relocate the epicentres, and a least squares analysis (LSMF) of amplitude data to provide consistent estimates of the seismic magnitude. A similar study has already been made for explosions at the Soviet test site near Semipalatinsk, East Kaa'<h and has '-en published by Marshall et al (1984). W. i. -U _

7 It is planned to carry out a similar study of other Soviet test sites such as Novaya Zemlya, North Caspian Sea region and the Nevada Test Site (NTS) USA. 2. JOINT EPICENTRE RELOCATIONS The Joint Epicentre Determination (JED) method described by Douglas (1967) was employed to relocate the events using P & PKP arrival time data taken from International Seismological Centre (ISC) bulletins. To fix the overall location of the group, one of the epicentres must be restrained to a * predetermined value. The restrained epicentre chosen was that for the event on 25 July 1979, the largest and most widely recorded. No true epicentre has been.- - published and so the location must be fixed using other evidence. The strategy used here was to shift the restrained epicentre until the overall pattern fitted centrally over the island (figure 1). Taking the median latitude and longitude as representative of the group, only when the restrained epicentre is located near the southern coast does this median lie on the island and its central lagoon. The adopted epicentre ( , W) gives the minimum deviation of the median location from the lagoon centre (taken as 71.83S, W). Confidence in the restrained location is gained by the fact that the ISC location (21.86S, 139.0W) is approximately 6 km WNW of the chosen position, a bias similar to that expected when station travel time corrections are not used (Lilwall and Underwood, 1970). Arrival time readings were weighted to remove gross errors and allow for variation in the quality of arrival time measurement between stations. The effects of gross errors were reduced using the method of uniform reduction (Jeffreys, 1961). Where a station reported sufficient events (here set at 10) the standard deviation of the residuals was calculated and the station's arrival time readings weighted to account for variation in this. This technique also permitted the incorporation of a large body of PKP data which would all normally be given zero weight. Table I gives the relocated epicentres, origin times and the 95% confidence limits. All depths were constrained to a surface focus source. Figure I shows the epicentres plotted on a map of the atoll. Although the standard deviation of the residuals (0.28s) is small the poor azimuthal distribution of stations results in relatively large limits, the majority (22 out of 28) however are less than 10 km. All but three (events 11, 22, 24) have 95% limits which intersect the island. 3. DETERMINATION OF MAGNITUDE The size of a seismic source is measured by its magnitude. For short - ~period (SP) seismic P wave data the Gutenberg and Richter definition is used:-. rn.= log 10 A/T + B(A~) + S,... (1) where A is the amplitude of the P wave in nm, T its predominant period in seconds, B(A) a distance normalising term and S a station correction. 4A

8 ',' ' " Consider n explosions recorded at some or all of q stations. Then if m is the magnitude of the ith explosion recorded at station j, we can write M.. = b. + s. + E... (2) where b. depends on the seismic size of the explosion, s. is a station term and E.. is an error term. Least squares can be used to estimate b i and s. if it is assume4 that 1b. ),S. = 0. By making the further assumption that the errors c.. are normally distributed, confidence limits can be determined for b. and s.. T AJs is the procedure used to obtain the least squares estimate of the magnitud4 of explosions (Douglas, 1966). The basic input data are taken from the ISC bulletins, either as mb or in the form of log,, A/T from which m. is calculated. The stations used are located in the distance range 60 to 90.bThe mean magnitude, (a least squares estimate) of the explosions is given in table I. Station corrections are given in table? together with the 95% confidence limits and the number of observations. The statistics associated with the least squares analysis are given in table SEISMOGRAMS The short period seismological array station located at Yellowknife, Canada, appears to be well placed to detect explosions at Mururoa. Several examples of the beamed-array P wave are illustrated in figure 2. It would appear that the seismograms fall into one of two categories. For example, explosion number 19 has no large second trough whereas explosion number 24 has a clear, large second trough; the presence or absence of the second large amplitude trough defines the two categories. The presence of the second trough may be a "depth" effect caused by the interaction of the direct P with the free-surface reflection (pp) since most of the explosions which' exhibit this feature are of.. relatively large magnitude compared to the lower magnitude explosions for which the second trough is absent. 5. ACKNOWLEDGMENTS The authors would like to express their appreciation to Mrs E Bradley 107- for her help in the preparation of the data for subsequent analysis.

9 [s..'.... ~.-...,2 REFERENCES 1. K E Bullen: "Seismology in our Atomic Age". Compt Rend Xlc Assembly Gen UGGI, (1958) 2. D T Griggs, F Press : "Probing the Earth with Nuclear Explosions". 3 Geophys Res, 66, 237 (1961) 3. E Teller: "Plowshare". Nucl News, 6, 1-13 (1963) 4. D L Springer, R L Kinnaman: "Seismic Source Summary for US Underground Nuclear Explosions ". Bull Seism Soc Am, 61, (1971) 5. D L Springer, R L Kinnaman: "Seismic Source Summary for US Underground Nuclear Explosions ". Bull Seism Soc Am, 65, (1975) 6. F Duclaux, L Michaud: "Conditions Experimentales des Tirs Nucleaires Souterrains Francais au Sahara, ". C R Acad Sc Paris, 270, Serie B, (1970) 7. P D Marshall, T C Bache, R C Lilwall: "Body wave magnitudes and locations of Soviet Underground Explosions at the Semipalatinsk Test Site". AWRE Report No. 016/84, HMSO, London (1984) 8. A Douglas: "3oint Epicentre Determination". Nature 215, (1967).. 9. R C Lilwall, R Underwood: "Seismic Network Bias Maps". Geophys 3 R Astr Soc, 20, (1970) 10. H Jeffreys: "The Theory of Probability". 3rd Ed, Oxford University Press (1961) i1. A Douglas: "A Special Purpose Least Squares Programme". AWRE Report No. 054/66, HMSO London (1966) UUUpRT,,oIL alt NO1 Nf.CIE1AI I,.-., / LE t. "0 E~t.l$ 0: T.E PUBLIC ''""- T,.C.ANt 6

10 CUN 0 -rslc' N U'\ tf\ (Z. a' N 00 L. 0 C. W 0-00 C, 00 QU)- -3 N... r: Ifu. U - L LU. ' L _;. IC'.. _ If'ul Lr ID CUN N P\l: N L4 Cr- N 0-4V % f"\ L -t NA? L - SA. r- C 01 - C pla ' 0 L.3 IC'. OC3 0Ai. *~~~~~~' W'. N ALN N? C 0-'~0C c' 1.. \ N N N- W. LIN' -4 c- 0 LAN W -:r L U-.- i W>z 00~ N2 C.. r"c N C%~v~a'e.a L) U:TC - LAO NO C N C. r - -7 N a', 0. K &. 0 N-aw o oc Nc- LA ' c A ON, r- c\nn 0 O 0- CC ' C.?. C a '\L C' 'C N, 0 CC 0) CC a' Ll.. -C '.; r- CC: 0 a- cl r$ r r -...cp L ~ C -. - N' r-a'. C N z P"\ PC\-. a' 4'~~~ -. a a CC CC K.0 +1i I L C Pr, N' r.. 0 N- ac C. N r' r A.N a' C,. - V I a' 7 ON N CN- LA a '% C- L N. -. N a-- ao N it. ON r,' 3 a r. r\ r- r, CN n' r.. 0 ' LAPA " -7 a' 4 0 N L N.. C LA a IAIQ 8 T N \ C o-.4 C- 0' 4 T, q -4 CC C- p- I '..Z 'r. N N r' L("C L.", ' \ U\ N... r0 N C\ C. C -.; N PA. z. -. N, NP"?..N c N C\. r- '. NZ C r IC', NOCI -C a, 0P C, N, PC 0A Na- 30 '- N 4 r.4 -- N r N. F. r4 r.fz o C,0t o L

11 a. % -41 to 0 La C cc N -Z 00 C U ('.2 - c- 1. C. 4 - _ C -ee0 Qi i K- - C.)+1 to o~ o o *U -Ci-.N 0 9D~ m.j Lf, 00

12 ITERATION 0 TABLE '(.3) Station Terrs for Explosions at Mururoa COMPUTED4 95, PZYCENTr STATION V ALU Z CO0NiD!:CE LIMITS NUMBER Al.iz P 16 ARE ?80 2 ASP BCT -C).~ ?? BFD -0.0,27 C BY BA-0.40? +/ BNH -0.09, COL '-."2, 1? C TA C C?~4 DUG -0.2/C FBA 0. C) I7 C.15" 7 FFC -C.-00 /0.9 GIL 0.Ocd OLE-0.0'C 7 C GCL i C.1' 9 NH p7 JC T / LD z5 5 LON0.0& C LPB -0.20c (' LPS 0.0.*' C7.2,07 4 MA ~ C.25 / M50 0).'/ +Q 0.105r N 0.C NNA C7.1,80 2 plm0.2~ i FIT C.1? EL 5 S'DV SES 0. '0.,- ". - c 9 4 SPA ~.-C?( S OO C., ~ 7,?2 TOV -0.Cz 37."~0 ITJL C fi UAV 0.2 :1' 0.2 7,0 UCT o use as st--t-zr. corre:tizn s the si;gn cf th.e station terrn must be re~erse (s"el enuatior.2)~ 9

13 ~ ~ ~ ~ ~ ~. --. W ~ L -. " W.7. * TABLE 2(b) IT~AIC:;Station Corrections STA0N C?.tP'ED 95 PIRcL:: VALUE CONF:DENCE- LIMITS AI; AR ASP 7-012?7 G.~4 3 IL 'A" C -01C 22 B-A-C.29,- 4/.0 57 B F2-. 4 G.~.10L. 4 Bi-k C.1 C.4r BM; -c z ,4k' C ~ 4-0.0C<7 12 c:ac' 9.c8 l 4.U L C - 0.0c 5 TEc. 045., C ,-.14c 2 GOL -C.2>,-0.82 ~ic. 2- :9 C7.121 H~i C 1C JC' -C. C4~ -1 C.1125 L: *-0. C('5 C,. CU- Lp /C..c LPS C.214 MA. C2. / C.Cr. pc.231 C.C 4/C.C 07- c.7-. :.-- 2.C.'& C c7.5.1 C.. 1 C S C2- :.. TC IV,1 ~ - C. 10 C.'- 0..0' :0: -o.1e... *.1 10

14 TABLE 3 Statistical and Other Variables Used - French Test Site Iteration 0 Number of Readings = 296 Number of Unknowns 76 Sum of Squares of Residuals = Mean Square of Residuals = Number of Degrees of Freedom = 220 Students T 1.97 Sum of Residuals 0.36E-12 Sum of Squares Due to Stations = Mean Square Due to Stations Number of Effective Stations 47 Sum of Squares Due to Events = Mean Square Due to Events = Number of Effective Events 28 Mean Station-Event Effect = _ Sum of Dummy Variables = 0.32E-14 Squares Due to Dummy Variables = 0.84E-45 II. I..

15 TABLE 3 (continued) '. Iteration 3 Number of Readings = 296 Number of Unknowns = 76 Sum of Squares of Residuals Mean Square of Residuals Number of Degrees of Freedom = 220 Students T = 1.97 Sum of Residuals E-03 Sum of Squares due to Stations = Mean Square due to Stations = Number of Effective Stations 47 Sum of Squares due to Events = Mean Square due to Events Number of Effective Events - 28 Mean Station-Event Effect ± Sum of Dummy Variables Squares due to Dummy Variables = -0.32E E-15 12

16 ~~2 it * y <x <.L,lk--*: N ~ ~ C < F L 1~~1* I 0XE VI4 13

17 '. t, " No No V,; b 5.01 CLIPPED No , " , ' ' - - No. 4.7 M b 4.9?2 No C ~ - - j-o; -~~~ o ': No. A "-, b. Time markers 5 seconds FIGURE 2. EXAMPLES OF THE P-WAVEFORMS RECORDED AT YELLOW--MINX- KNIFE, CANADA FROM UNDERGROUND EXPLOSIONS AT MURUROA. 14

18 No. 24..El."L No M b. C'9 'A'V"':': , No I- No o t.-, - - n~ 5.. No second No ", ',", /I,.- ' " ",;-''--"-"- "'..-"".-.5.., Tine markers 5 seconds...,:?.:.:

19 -.~ - - o % DOCUMENT CONTROL SHEET %x Overall security classification of sheet... UN.CL. ASS. FIE... (As far as possible this sheet should contain only unclassified information. If it is necessary to enter classified information, the box concerned must be marked to indicate the classification eg (R), (C) or (S)). I. DRIC Reference (if known) 2. Originator's Reference 3. Agency Reference 4. Report Securit.. " ~~~~~classification "..'.". -AWRE REPORT 012/85 - UK UNLIMITED 5. Originator's * Code 6. Originator (Corporate Author) Name and Location VL1, (if known) - Atoinic Weapons Research Establishment, Aldermaston, Berkshire Sa. Sponsoring Agency's 6a. Sponsoring Agency (Contract Authority) Name and Location - Code (if known) l 7. Title Body Wave Magnitudes and Locations of French Underground Explosions at the Mururoa Test Site.?a. Title in Foreign Language (in the case of Translation) 7b. Presented at (for Conference Papers). Title. Place and Date of Conference 8. Author l.surname, Initials 9a. Author 2 9b. Authors Date pp ref Marshall P D Lilwall R C Warburton P J October I1 11. Contract Number 12. Period 13. Project 14. Other References 15. Distribution Statement 16 Descriptors (or Keywords) (TEST) Proving Grounds Underground Nuclear Explosions Magnitude Least Squares Method Seismic Waves Since 1975 the French have conducted underground nuclear weapon tests beneath the Mururoa atoll in the Tuamotu -\rchipelago in the Southern Pacific. A least Squares joint estimate of -" origin time and epicentre together with a least squares joint estimate of magnitude is presented for - - presumed explosions in this area. These are based on data taken from bulletins of the International.V Seismological Centre. Some seismograms, recorded in Canada, are included to illustrate the nature of the P waves from these presumed explosions. Abstract

20 Lj p i n-.' -. 9 * p. ~1

Latitude (North) Longitude (East) Latitude (North) Longitude (East)

Latitude (North) Longitude (East) Latitude (North) Longitude (East) BODY-WAVE MAGNITUDES AND LOCATIONS OF PRESUMED EXPLOSIONS AT THE CHINESE TEST SITE, 1967-1996 Aoife O'Mongain (1), Alan Douglas, John B. Young AWE Blacknest (1) on placement from the British Geological