ML19275A617
| ML19275A617 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 08/15/1979 |
| From: | Chinnery M MASSACHUSETTS INSTITUTE OF TECHNOLOGY, CAMBRIDGE |
| To: | |
| Shared Package | |
| ML19275A615 | List: |
| References | |
| NUDOCS 7910110274 | |
| Download: ML19275A617 (76) | |
Text
'
4,",*,Mf::. 4 ' '- hy y,y ,[, 4 , ~ ~. n;; - ;., ,_ , ,. ,, ,,
~
e%. 'hyw$
- .r'.wA+.'?. ~4,- ,1 ' ' *J~ M;3.~:.,
%w.K..,w@. z .m M.e.,;nM;v.y:2Vr. k.
.,W.. > e,, .v:,r .Xw'
~
- s. -s1.s ? ?.s,fs~h ,y:v; , < ..m s
- -fl?v$'Y
.~ 8 g..
~ .cw ~ .. ty . . . A .- .
c~. 'n n a< --
c, . .
j..'D l"l; f.y -9^ r a/. % l'e w .*-.::% .e-=*'.~-
.,2- .. 1.
+.g.
Q.1,i.,:t vc W;.*J; .tC.t :. .
f., vy,cq1,
< <Il.s r ..,, :<;c.$.,v;".
v,. ,
. . *s.2a: "lQ's y- *- -
e4 .-
- i~
.s. *. .r. - %s.st -; .- - -- ~-
- w. . ,a.
,,,'.'*.s~
- 5., 7. f. * . ** Q ; :! n .
t "I . 'c.w., < ~ . * ; < ,
$'*'p !*
, - ~-j ' ' "
s .
'. **r;; , ;i * '* .]* > y .. . ' r- ,, ,,;
{'g ,' , ' ,,,, y j ' '
Q[ ; * }y _ ) r,h' :.-&:~5;.~
- -e
'1 .. g .
- '.....*]
.d,., .- ,, g
' i. i *=.7...y f_,,,...
h; .. . . , . ' _ -
3 .,....-
.g
.. .c- .
.. 'M . *c' ,75 y--e,-
- i.. (,; y &; ,1. * -
- . ,v '
- ~* 4 .?
- f.*n~ ;;j;y; y91 -
'4.'..
.,.-, ((E ,m* .i.v.%.p.q; .v K g.x W.,n..
g,..J .3:'.s. m.a o#. .u v.*Jmfs..m v y ., .; M .?,: ' .u;a;.c.,. <p:.n ~ .. ..: ..
? w$h.M.. py;.~WWEM.;&w.. 4 .. 4., ,,,-
.gm
'a M20W,.V. q> .rM.s,, w~ *. ;tx
- y. ...
Y 44 kW'%;;:7RN0iv:'n : . ~.?;n .. ~- - l4. .
W?%,c-@M::i,% .:@- Q%g::$dl'4 M@M._.gu_Q,%Q'W3h Mu.c M,%.n _ M 4 :-n -
. _D: W-n ,.
u.:~:. ', * ;.s.
', ,m % x -_
.sa:n -
- "n .
N N k;,fg)kbgh..a#.pqm.r.v.M:,,&.+dk k h
.wou qqr w 9.+p k h.;... b.x.b~*..dk.~
n m.. y.y %.
dWy %Wwwmn.1.n. m W.Q,. .. , .c u w. . .a.. wn D QNMM W.w, ..W- W%:. h
~4 .~:;;V Q3M&IN# M D' D Q
M ., $j M m. p M 4 h.n? M @%m m
- p. %
A m ' s qpm . e,. w%.?M
- 9M.,
s W
, w;w..;o M m p M:.
L.5 n
- ,,. .e: w $u w ME. 3d .
p+M,-c.m .ms W %n..eu e p, % @.%.sw%.,,4 Sy.:. y;Nu. 7w: mw.&up%w+c.w.
.m=
w pm . a
.,r: w
- v w y s;.. ;i Jp r. haw *w mm .
, m,, ,,,.,.:,. , 2,.,,,v.n,,,,,.,,mu_imus.,,ns,..- .,
9 Md ,
g W ..,,
dpc .49.0.
,t h e.. .
v . m
" p .< m-
,, -.i .Q.' n w ,6@etN@OJ@@MWMWQF @s;g%+m'MM@
w9 pp?..y %p .u .. -
-w. - -
sw/N6h MWaprL.
km! M.6,...x 9 w w p y,g g, .~$. . DR tc<N:.
e.~ M .w ..n- W.1Chinaer<v1. . :x
.sg. b .~ . . .-
W isn g fk,jtM.#. . m v.
- .i M W k('6n.Qu,pyq%w%sn.ry2 A c:.m'dM.
v r , r / c 4.' 5 Q- .& .N .F -.
-z . :.
~.
?ple Wq"%' 4WWy e -o n -
h $ h ik k49.w .qvQ1&#+1>.sn.s e/.@.g? hp',,m'M k.. sfim y$ NNh.kh-hh hhhf.4.:w'M'0}w':'
e Mn$.M d xqqs jimp, afi J%wApr yelsmolo IcalInput.
W.1 p y w;,w n . w . a ,.n A M. u h ...rW WNy hff.b h4,Jc w h: ...~. ne
- W' ! dy.vN.
u.-
,f W - .w W#:- W o
ke. Qle.e.,g, $w, crM D e -q r . . f-qo F s ,. d dm u .r.! f .r.,.
N ,vw & M 'a Q W J;8 C;1:9c A
Q MGiW.n(d$...},p-+di'p4 g[5%Wm#W.wf - AmQ M - m sMr)v EQ $.. ;mgM'Qs25M.#w 9 %.w W s 1 w ~; %:w w. m ,#fa s p
- k. . .
& 51 Reactor .c up@%g[p;kdsp,X R$Q gW qnpq%ap.En.glaWs. pwM Mc YQhy ahqMwwg sws% m n& &gw wMWn U%&m ~
@kkkb M k MNhbM m h M @h@p#g.V $i'5N
[am%+ gist 195 s mm.t . .;
"~9 w_.m__
4&s. 3 r i.w~$.m..M_wam&_&Www%._neamm.&
d ,..m o We_ _..mmw.+ c+ w d.
&,,Y*. W %: ,a,h,nm kk d G_.,b ,
Y f ' hth {ch*?'k Y
wWm *h'l.,1e ?. Y*Ii.h:- * ~,,?L ;'.
aw
-> : w ' a = m$ '
(A.h.S
< ,y Y- A @Nk $;.:. W y%& 0..,W.,a2.%(&w}hy Rk ;u 1;h'hlE .bprcm$~h h p w!$w[.$+aIh we. ' t
.9 e
W hxN4 Q. '
. .p &k j G:' '. -
PlQ$ M C MM%}Wy%,jpg@M p p V-M@$@kMW@%_y%%%@h_Wk_h5W 5
Md N W M hs WWW@hd1 "
$a&M W U_ _N _
(
n.w:n
- w. .
j,. u .. . a. ' f M e-
.c.
E..e[W o..n.
.,.n~ u *'J' b' .~ w,. . ;' .. u Y.
w .
um, n sia
. . . . w.
,O .*d'* c*,* [. m,h'7'p. .. 4
. n.:.~.e
- * ' ~ -
-n --
E, -
$,#yd.y: .
.lL 4 ,
m.M, 4 g,. .?.f z:: ? .%~ . ,t"*N ~M
> ~ ; --*:.*~.
- a%.
=
.:- . .." 1 Y.*f.
. &,. _~f0. 4'.u&,,.o* . ... S,w ,,h h. -
. " ~!'h$~ - e*.'E*. m
-- E* .' 4 '
' f&. ."** w
.. v .
.,'.*?
- s. 4 j\;
3,.s.N.g*(..*
.,.3,t.-W.
A
e .
C.O 2. f.:".'$.
s *: ' f. ; s.
- '~'. l }h
, - * ~a,.' ,
- Q -
i~ f.;
~
S T- -
\lassachusetts In st;tute of T 'c'.,olco Lincoln Labornery AN INVESTIGATION OF MAXIMUM POSSIBLE EARTHQUAKES Annual Report iI b
( )i Project T:tle. Inves:: gat:cas of :he Se: sac.epea' :nput ta the Safety Des:gn of N;c. ear Power Reactors in New Englanj.
NRC Contrac:. NRC 04 77-019 Pnne:pa' Investigator: \lichael A. Ch:nnerf. 3:aup Leader Applied 3etsreicc, Groep Lt..coln Lateratorj, '.!!T 42 Carleten .treet Cambndge, '.t A 02142 Pened af Centract: 1 Janaa.y 197~ ~ al Decem:er 1977 15 August ' r3
s t P Ab S .. . .o.. .,
'.'k. . d.S '***C"*.
..; . .# a S C ." .# .
. *. S **S*."#..". *a**#^*
. . . . . e' " . ~ ~.. ^. " . ~ . * * . ' ~.'"U.C <.. -
,a
.e l; .,. 7 . _ ' .* o, 4., , .,. e. g ..%..,.
.. a....,. . , A - . va,,,,,.../
-a,~~
. a. ., a,, wm e _;~,~.
, 4
=
J . *. g .4 ' =. A 3 *,3d f ~s.#
.a ,
- o. 7 3 .4 _' g'g .' a.
e.c .' a. +4.'# ...
- a. _ .! .. a. . .-w o . - a. .- .. .w ma
. .r .. 4 . . e., ..%.. a. a. .r.,. ._J.. -.o.. l .
o +' . af.#.3tCL y"C S S .%. ' *.
- a."'. '.".q" a . *. S
. o' u'C'4 S '.'".a*.a*..'
. a.y g .3*. .m b '. a.
. a. .. '.. . 4,3 2 .- a.
ay .4. 4,gi a q A, i
-w' 21*- a' CC,# gb..'*.#**.'
fa . .
- S.a~.# # a. .". ^. ^.
D/ . ,d. ~.'.~.a*.*.V*.". ".'.".a. .
a =.r, i.w .4w- o..' . a. .. b..e. d s 2 - a. y n _' 4 4 43 y a. .w,y ya. .o i ,
. - , ,-w4 g d'-f '.aapoca. oa ~..# . %.
> ,. p .e h. .r.. ,
' a. . .g, ..u.
- s. .,, ,,. e a a .. ,' . . ,13. '.s. a.
- .- a. n. n
, a
. 4.. , -, .
3%a. ,. -a a S . A.. a. ... d. ,y,. .... sp
. g~,_~ ., % a.<
a n., kuo.o.gk a. .o..o..JS , * *o *O
- Xa2..# ". *#
- . ," C S S .~~ #
- . '. '.". e ". a. 3 .# ^
- o '. .. - 7 a " #. 2.. '.. # .* *. .. 3aX.'.UC . .' *.
4,
- a. a . %.A ,., o 4' a. a- .a o.S uC C a. .. c e .r.,.,
2.eCC r.S... a.,e A. ..%..ay. .'
. . . .. 4. a- . . . ,. a..e .. .
21.. . . ... ., e zgn_ : . . A.a. a. c g .' s. a - ..d % a .4 g a e 3 2. , . .- n. de n a. A. ~,,f ,,
4
. .3. n . _.~a....g' s .' t. .y
. w :. . .g~ ..~
, . %2 . .a.
- q mg'r.g
, - .m .y .a..' , q *. ,
.* g ,. ,3a. a. .n % .~ , . g.'r. s. g . ., a. .,.y
..-g .. a.
2 3 %*_a_, ,
- n. . A. 3 .%gv-,
w a.
3 . .
...k.e .w a. g e. .,., o. . -.- - sk
, S e.*.C a. .'.# 2aX # '".~2 -
y C a~ .e '. h. ' *. *a.**.'.'.C,'$.*..*.
. A' f" ." ' g' ".*. m a'
- "C..
-f
. a .2,d. /' C.# '.f e'4 ~'u' e', ' ' .". A.
.'.".'S'. '..".A.
. .",T'y a. ." 1.'*. .
a ". ".C*'*.
.a 4.nc.9 . , A.a A .
.n
I P Table of Contents Abstract iii Introdue ica 1
- 1. .Vaxim m ?cssible Earthquakes: Curren: Status 2 1.1 Introduction 2 1.2 Definitions 3 1.3 Apprcaches to the Proble:a 5 1.h Physical A s'.:=ents 15 Arguments Using F.arthquake statistics 10 1.6 Use of the !.4 vel of Seismic Activig 12 1.7 Pattern Rece6nition Approaches 16 1.8 Other Studies 18 1.9 Discussien and Conclusions 19
- 2. Analysis of Global Catalogs 23 2.1 Characteristics of Global Catalc.3s 23 2.2 hrthqua'ce Statistics 2h 2.3 Saturation of the Magnitude Scale 28 2.h The ISC Catalog 32 2.5 Events in the Aleutians-Kuriles Regica LO 2.6 Inter;retatica A9 2.7 Discussi:n 57 2.3 Ocnclusi:ns 50 References 51 Appendix: hegress Repert: 'Tev Irgland :rus and 'Jpper
. Van-le 3 ructure 5c iv
4 7
e A
- 9
..g.g.A.A..,,.4..
.. w..
.
- v. . 4 m, g w 4 %. p @- *g**** . . , *.
"l.,.Y'_
'."w.4_
. 4 o. yw. . as . . .== a.*, . .a.a .w a. b. ._...'A.A,. +e. A A. v M. . .v#' ..=.=.=.=.=,.e..
v% ,,~ ~ _ ._, ,8 * ; .w, -
- ~~ .O v. *. p.. o. ;-~. w,
.. . .. _. .w , ,. ,
,4 . . 4. .g 4. . A. s...f, a'..u a.*.'.#^-'. "ye '.* '. ., iu .S. .' '.. e A_ **.,
. 4 ' ' .'*
. .d . .e' *. b.._# s- . ^. *. s .s.
,. #-*ed . ***O
.. s ,,*s'. A_ # a. s
- a. v a .' .' o.* *. .e'
- . k..e , o s s -' b .' ' _' *.v ,
c ' e. s '~' a ,...' .g . '..e nax' ~ um , c s s ' *
. _ . a. a__ a . '..s- " .a..k a. . "..a . n ' e .".*.
Se ax,e a._4 v4. * %..4 .. a 3.' ~
' 'e *. .- a. 3' _' ". .
"he first study 20nsisted Of a review and assessment :f available s e .' e. m.* .' .' .'. -_ '._'.
- e -a *."m-a. . n '.'..' .' a- .s y .' . . . :'. ..c a. - um." ^ * ' . ."..a. -. a. s e _a . - . ". .. ' . . . " . ..s area has been car-ied out in the Scviet Union, -his review pr vid.es =.
k'
~. . a s e. r.a , _ ~s . .ma e r a"..e.a. s .i va. s e *... - ' ~. . .' a.. .,~. ~.~. s , a . A. a '..d . c " s s .' . r. . . .. .'. . ..e' various appr: aches which have been tried.
~~he
. 2 a n.o..d -2* ~d*/ "
vas a. a_',,.a._.. . , . , '.c k '. .*
_ a. v .' .'. =. ..'. a. .' ' c a. .
kcund.s .s', =__ a .-*.% suak. a. .e.' .. a. v'_ . "_ .' .
3_' . ' a '. '.c d_/- vava. ca3.._' ~ _a_
- "' ..=.-_',.3a, a --
a.A..u 4. . . , a 4. . . . , ,.
. . . .we _c:a -
m..3
.4,
. . . . s. . ..- . cf, c n . .. . . e A,. 4. .o d
attempt 0 n.erstand *,he scur:es :f bias in the regnitudes listed in
..w 4. 3 . a *. al .+ g , s'... ~ "a.*._' .' '. '.. e .e e = ~. _' s'. e .- s . .. c d .' . _' e_
- , e s _= _' k., _' =. ' . , . ,
a e ar . '.. ' . . ..ar.' ..t.n
, e s a-.' .k.,' a. =.~.a....a. ~ _ ~ .. . ar. sy.' ' .~. s '.'..a. . . ".. a. a e '. ' aa a. e. , .
.v ..w..
. . . . .g.4_ .. ..3. . . 3 . 4. 0 3 v, ., . . , ,
.,, .. , j. ,
.....,3._,j.
- 3. a ,. ~
3 ..5s
. . g e _3. , s a. ... , , .wl
. .-. ..fp a . e O - a . . . A,.,7, 9
.4
.w4
..
- A .o a.g .'*
- e. .C g a.**. *. h. , . ' * * . . , . ."g * * *
..=*3*. g.m. A. .'.
- y bp ..*= .=%..'*_a . -
D..-*_S*****p
..=a .
. a. .g **.*
.s.-. _- e _ .2. s4 , 3..r.
f,.3., jo .7 _4._.
4 . .
. ..e
.w
...a. .g . 45,_4 . ,.j e. , ,
- a. A. ...=
%. 7
.w4
. . ..g., . .
3 **. ,4 2 -* *
- 7 .*= . . .-* a. 3 3 .== ;n. ,e C ** *. .'-D ..
, ' 1 C _ _+ 5_ a. 4 . . . . . ' G. . ..g...= = p ., 4 3. .*t .
P00R OFSR
2 -
paam rge ne .it O .
.tj .rs a*
...m.. O.www_.
ccc o., wep r6%u..g.r < v e n.Wm -tms.m. c i,g ,[ .r
.. .vGs at. v 1.1 'ntr:ductica We vould like to kncv vhether er no: there is a limi Or " upper bound" to the size of earthquakes for a variety of reascas. First, earthquake si:e is usually intended to be a measure :f energy release.
However, energy usually varies strcagly vith size. For sxample, the standard relation between sagnitude M and e:.ergy 3 (in ergs) is log Z = a o
+bMo (1.1)
Bath (1966) reviews several estimates for the constants ac and ba, and shows that b, appears to lie in the range 1.L to 2.0. Since the nanber 3 of earthquakes is usually described by the relation log N = a - tM (1.2) where b is about 1 (see, for example, Ri hter 19c8), the total seismic energy release is dcminated by the largest events. We shall have reason to question both equations 1.1 and 1.2 later in this report, but the conclusion appears to remain valid. Analysis of the energy budge: Of the earth requires kncviedge ^f the rate of occurrence and energy release in the largest events that :ccur.
Second, 3 rune (1968) has shevn hev the relative slip :f teeteni:
pistes can be estL=ated frca earthquake size , and shewed tha- -he c:a1 The ^' ia-a- ='
slip is desinated by the largest evants that Occur.
questi:n of hev ruch tecteni: sction is re' eased in seistic slip levies _
and 3 rune,1971) can :nly te answered lently :nce ve 2nderstani .T.ese large events.
. zag, ...m.. .2_y.,.
.u,.
. ,g.z__g.a
. . .. g ... gx i_ _;n ,. ,. . . . . e_ ,. ,.
...s. ..
,c._
,D..:-,.s._me...wM , ,a._a.
3M., ..3_,..- n., -, 3....
G . . ._. .
- s. . . ., . .w... ,.,,
...g.
Gue .* .w. . e . . .w. 7w
- 42.
~
- a. y a..* *. J .M S.,'T
. . . - . , 2. 7 A. a. .I...#*&.M'.',.a**._,f,
. g *a.
.. l.a 3 .* a. o.
.Sa _m .. 2. .? s. *
"t s
wa *.a.-d *. b a . # s " .ar.' ^. a - y. ab. _' *, .##." w
- a ." '. a .' a- '. d
. . .. .# .' o .'. .# .$..* ...4 .'. # * ' a s a' ' .* b. . S a-
.".U c .'. a. a." C V e." g#..* .".
- a . %. *. .'e . nP 7."~' a s .5...
. 7. . a e* " *. .* . .#."' .". a , .U S . . 90, .l.'r' r~ a. . . A #-X
.x, .
.. ,, . ,g. ...
.me ga.. .:, - ;..- a .r ....,, 2.g -
- 5a >, a .e
..u..e g a. .=...n..-a.,,
... $ .aa. .> .z . a.
.he c.a."a. '. h.u*~d c . ..
. W- - . *, - tah. .o '. e ' . a s a. d .-a a. a. v a.' ..a * '. .m.. . . " ." e " ax' ." u:1 earthquaxe pctentia.1" of an area Cdo:5 ann, '9%). . ~'he purpcse :f the present study is to assess our ability to estimate this quantity.
'4e . an ".s e." ' _' /- .' _' v d.d - e ., k a e o ve.- =__' '_ f rok. .' a._. .".. . o . . o , a.- . a . 7'.s-,
vb- a*. .# s '.he a.vd ' e..c a. . .".a.
. a a_.- *.*..q' uak e s .- .".s '. .' a.. -. . " . = . . -a.=. '.
.z_-ba_' y.".e c - e.e . . .
have a maximum .:ossi'cle si e? And second, how ices this maxist:2 pcssible s .' ~.e va*/' .#~ m - . g
- c n '. n ==.3.^.n.'.
. . . . ~* . ~'k..
. a. .' .' s . - "sa s . .d . ..
. e m..e~ k.. . ~
' '., e n".s.b..
siso.ler to answer than the seccnd, and it is 1:gica2. to exasine it
.-.,g ... u.cyey,.. , gg
. ..a.
. a.u g , .. -aea, ... .t g a..s. s.o.:,2 .
. ~ 3.a y,. . ny . ..a n. .a 3 ,
ansvers to either of these questions.
~.2
. efinitions
". h e .* a. ar a. .vc .' ..y c r .an . .d.e ." # ...i..* ' + n s .'a*. va. --
2",.i a .
a .rf '.- v .- * ".a..'+.-a. "a. . .
.cntt.u ue. %. e .e.t .,.s t 4.a ~.. %.e aae ~.t- mz. . 4 e., n ,, e " max.: .. ,, n
..a . u..a. a.a.,a ,. n. . d .: a.
n.
.u.me a e.e.t n. 4t..on sn.o n s_a,e .
"he term "OaXi212:' is Oct, "n f O r t". .at ely , alva' fs us ed vith -~ he s a.*.',e
..e = 4.' . .e~ . -v""e.Aa..'.'..*..'.~.
- . . " a-
. *.
- e .~ k, v .'. - o. s ..*. , V "._' .' ". ~=..*=..~s..-~.**.'..'..g-=..=..
yO s s .# %. .' e a.va... . ". '.a . ^. a . .' c a.'- "' 5'.# v a. .". ,, k..e y a'f a '.'.$' .'*.'#..#^".s'*.'.".*.
a n.w. n. a. a a. .a. . .x a.cenA
.a .-w 2 a. .r t . 4... 4 g , a ..~
. ws g . t . a. g ..-
.. a g ;. , 2
. .wm_' .a,s., . .u. . a,. ,.. .. . .. a...
.>.w : . w gw4e ..a.--.f. * -* '
.. ., .* a. ". .a-- '- * * * ^
- . .k.ak # ' #- - -- . ./ _ ^ v a. ' .
- a'..' '. a. -s . . ,"' '. eA.
2**** 2 a' ". a. .* _* e-a.g.,; 5,c., an aC.,,. . . . u. .g
- a. . : . ., a. .,. ...b
- e. 2a.: ..
- - G - S--
=., ;, a. .. ... . a ,. . . _ .
.w
-- S - - a .--3s
. . , ~~------3 -* - * = * -
9 s . A. ..% a. 4 a E.: u. a. .a ..% ,g, .6
. . , y a. .t .a .e.t a. A a a.
. .e. a, y %. ..t ., %.. 4. . ..wm e. *. ye . %..
.. w 2 w ...4.,
. . . .f
. a. 7 a.-
.. .saa' m.
.4,.
_..g ..%.. a.
.4
- . . . , a. .... a. 4
.....-.a. a. - ... . , ,. ..,...,...a..
~*%
. g a. .y . . . ,
... a.a*: ;.:. . . .. e. a . w a. .,. a. .,. y 4 2 .-.. ? a. , g..e.., . .g 4
.i ..
2.
.a a..a..
n a g ,. . :.. . *
. . ..a..y %. .p . .', a. .s ' .y ....' a. 4 . . -. . g . _ _ %..a. ;. .. .
ja. *
. a. ,. e.....'..s
,, . .s .. a.
.e.... . % a. .,. . .g %. . g *.. g a.
. ...%. a. . ..
.y
i 4
. r M, for the "true" maxi =t:2 possible magnitude (? for the maximum o
rossible energy, etc), and M~nax
- for the =agni:ude tha: Occurs with 2 n. y a u .t _..<
.. y o. , y.s..< ., u.. a.n. o s. ..e s . .u..n. 2 - - ,, ,-
. . e . a. .- .. 'c a % -- ' _' ' ~. v. o " ." . . e ~ e_' .' 3' '. ". .' .' .' . / " .
As ve shall see in the next section, very different metheds =ust be used o
in the estimation of M and M~ .
2&x 3ax The definition of earthquake " size" is even scre iifficult. There are a large number of quantities which attempt to seasure this size. A partial list includes:
a) Body vave magnitude (23) b) Surface wave magnitude (M )
s c) 100 second period ma6nitude d) seismic nement (M,) ,
e) radiated seismic energy f) elastic potential energy release g) maximum epicentral intensity (!)
h) maximum epicentral acceleration
- 1) local magnitude (. }
The basic problems here are not only to decide which of these reasures of size are the nost appropriate for a given s'tuati:n, but to reccgni:e that the relationships between these seasures are in general pecrly unders ccd and in scme cases demenstrably very ncn-linear. In part -
cular, seme of these quan-ities have built-in ;pper 5:2nds which :an obscure the search for a fundanental -:pper lini- o +arthquake size. W=
shall examine this proble= in cre detail in secti:n 2.
A:
- w*^'
- . . . .* e. n= .' ~.~.~~y.'a..'-,
m . .. v *. ' ". .~. s. '. a~ =..=-- '- **e'..'.=.-c....-=.
- s. . "s ... -:
- wo., J
- a 44 4..- - . . f. , J a ...u.a.. ..%..e s .w i . . 4 m. .a. ., a-
=. e ra m--s-.... ..*.a..n.
.. . ...b.g....
4.
.i g r. e. . . .. . . . ...
j~a._ ,
.....:,3.-
..--... .3. a. - 2. 3 .* - 3e, 4....4 3.
. a. .r....
.. , . 22.
.. .-4 --
. , 5
.' i m.4 *. e d ;- o r +.4 o . 3 " . "..a. . a e.' am4 . s . e ~. ~.. ~un , =- .d. = - a. . .' o a e./- '. .' a. d- .
o ."..a.
e .* c d o ' m a. a s "- a. a. .. . = m .' ~. .r a. ^. . ~'.. e - a. .' a .c .- ~ . ~ ~ . * . . 7 3.
.. ' =. .' .'. .' . /
' ~
... ' c '-
..esa.
.. "ac
. *. . r s .'.._a . ."..a. .=..
. ..a 3 '
"d a. .
~
=...,..a. '. o- .=..' . e s . 'a.
a .a . .' ..e - . u. ,
a.,. 4 ,,,,..,../
. .w . ., . w .
...e.. ..
,w
. a ac.e . .. ..<s.4.,s ,.#
.3.,...
. 3 .... .. . s. . g- , . .L,. .3 e .,.-... . ..a.
^5 ui +. e o .. a. _ --
, . a. .' a. .- a. n c a..o
. .' e _' . s- a' . a.. ' = m.' ci . '.~,v.^ ' a. .- a. a , . "- a. . e. m sagnitude refers to 10 cal =agnitude M.. Of all =easures of magni:ude u
this is one of the 'ca dest to quanti"y. : vas introduced Originally by Richter, and designed for 'ces.1 shocks 'n Calif:rnia. ::s definiti:n is very arbitrary, and refers to the logarithm of the maximum re00ried trace amplitude of 2 specific instrument Necd. Anderson seismeg-aph' at a specific distance (100 2=d . Because the instrument vill receri a vide
.ar.
. a. o# . *-a. . que _- .' a. s _.
.k..a. s k..o *, a, e ' .d ' a..d , =.. .d. ',ec aus e
. .. . .'..a. .- a. '. o- ^
s e '. e..' .' c
. he.s e '. d.e . . - n . ' .". ^. a .# n , '.k. .e a 3 -....' "....=.- - a. . '. .'..a. max.'. mum .- = ^. a.
amplitude is not lear. ?cr small ear.hquales, the maximum trace a=plitude v' ._' ' o #. + e.. -. a. . 'a.- .o '.cd'/ v a _v e .a - . *. ~< . a' s- .a. . s'..o .- . _4 ' s ,, a .-. a. s .
' . ' .. - ' -.=
'.,a.
ear *,hqua'<es, the maxim'.:n trace amplitude vil'. sue'.17 be assceiated. vi;h
.r.damental
^
acde or higher =cde (L phase' su-face waves.
~'h.
ef.'na'al_ . se'"~..a.ss o'
. .. '4 .#s, o' --" se, . 5. .a *. ' . .' o- =
. ..a. a - "_- a.
.e.r 3. .
g und oc '. .# .', r. .# ~. ~. "..O . e ar .#.# a.l w d a * - .
a"e~a.
.'.# . .- *. *v". e Y. . .' *. =- ' . " . ~ * * . -.*
.. .* .o.y g n.*. .c. . .. 4 ... ..b---- - -- g g 34 - 4ewo.4~g3-- ~ ~ *--.7.-
--* - - - *- a- 4e .%g me
-* < - - - - - - < - - - o-J-.
/s.Na.,.a C. 4.. 3
. d .J
. . . . . g3, .f * *"'.' 3 ) . , y . , o 4 . , ,
3 Ic . .,. . s.s..n...
- m. . .=m.e_., .
' . , .. .! . ., " > . a. .*aa . .i ., ,. ..
- .,. .#.4 s. .' .J. n.3.a
- g. 5.J .. . , 4 4 . m.4
. 2. .
J .%
=.
3.4.= . . . J. @. _..J..*..*.
. 4..."a.2.
.1 .
n4 C ,.e .s o , b..m .ae .g . b.. a
. ..m . h *. a 3
'."'% e . , % e .,. . .* ..f-o y p
. .. a J
. .n. . b.a.
. . .# . e w E. . . . .....s. 3..y,,y.,..
. . ., . .. a .
.w >
a...
..g,,
= .. . b. e f.. . %. _* m. ,. .* .%e 3. ,... . m & ...J... .
% e .o .w. . . . #. =.'.g. . o 4. J . 4 . g- a. s.
7
. I %. .*. 3 3. o. . .. . .g,
... .- ,33 3 .'.a ...J s..., .. . ,-_- . - s.. a. .'a..#~.*.,"
. .' .# ."..a..**. .*. ."..a. ~,~e,'.'.
. . ..'.s...' .' . . *
. . .*.g . .6i..,
- gWg
. . . ... ..gg. 'yg
..3 g g g .
. . .e p
.3 q
g........9g.3.
,* ,g gg
[ g a
'm. .j Jg .,,...... q
o 6
.m e ..e .u vn ..r.. .. . ,a.e-f a..
.mese ape,,a ..,ua4 .,
.. , o a.:. ,.
. . . ., d u.. s. e < - . a ..- .-..v.,
read.
A number f a r.creaches to the .r:blem have been .trar.csei Isee for exanple, Shenkeva and Karnik, 19?k). First, there are a nu=ber ;f bread arguments that attenpt to limit the upper size of earthquakes :n the basis of physical principles, including fault geccetry and slip, and the strength of earth materials. 3enerally speaking, these arguments make a convincing case in favor of a global upper bound, but give little indication where this might be. A second apprcach uses earthquake statistics, either in the forn of frequency-cagnitude data or modelled by the theory of er*remes. ~~hese two analytical techniques generally lead to similar results, but both turn out to be severely limited by the definitions Of ma6nitude used. A third approach, which seems very icgical yet whiu lacks any :cavincing physical basis, attempts to relate the size :f the maximum pcssible earthquake to the level of seismic activity in a regi:n.
It vould be very nice if such a relationship vere to exist, bu: there is no clear evidence that it does. More recent act. reaches have tended to focus on information frca non-seismic scur:es, such as geci:si:a1 and seemorphological data. Scce of these approaches are statisti:al, ; sing pattern reccgnition techniques. Others are nere deterministi:, and attempt to link long term geclegi:a1 fault movement Oc shcr: term earthquake slip.
7.4...
. . . . g - ./ g-' oA#
- b. a. S *. 8'y"
. y."*. a~'h
- a~ .w.
~.
'f ". kw-.' *m 'g ." *. d. '-*-# .~. 2 '.
- S . '". e
.n.e.4
. w. a 4
.' . gy . %.c,.
. .. . . gk . a- .
- 4. 3 -. o. . g *. 4. 7. '.). .e %C ."*. -# .7 CS*- * *-." .a * '
.'#. ~ . . ' * .v* ^. .'
"a.a s.....
. . . r,.. .% v-... .- . ' ,Org 3.-.
3 3 .-
. .. g . . ,7
...o*3.a*.'v*.
s.--.. 2"u*
- *2*".
.4 4 p g . , a". ,a
-g. a*e. .# .**w .# .* .* *..Va
. .*1 a" k g*e* A. 1 *rn #
.**- ,* ,*'. e. a. .m. -s---- 2%*.a. #
.t . *
- s.
. ". b. . a.
3 2 . ,*/
-y u*t.,
a -..
3 4- .23 ye. a*%g *--*
- m. e.
a Im a ap a. *#
-- . ..=.=..
- \
. =. -%. e w a. .s .u a.
..#**. . -- : ..-=. %.
- a m ~q s
..e.
. r 7
' e .* .* n '. .. e~
. . *..e a~ ' .~ e ~ .' ' c 3- ~a a. a.- .*..er" =.'.< e s .
. '.'.~.%.e-a.'.'.=.
. . . *. e .~ ~. e s ~a. . /'
4.ifficult to establish empirical iata for maxinun ;cssible earthquakes
.' n. st a a.' .". ~ . *. e~ .' . r. s , a- i. . a. .* e s e ' .s *
. . e~,a_a '. a. v =. .n. . o~ ...a -f '. . a". = . a. *. " . . 2. a. .. c' s of ICCO years er scre. ~41thcut these espirical estimates, it is virtually i=possible to exasine the validity of many preposed approaches.
l.h Physical A. g ments There seems to be universal a greenent that any =easu e of size of an earthqua'<e must have an upper icund. ~'his arpment is :f ten intuitive ,
but it can be refined to scme extent. Certainly equations 1.1 and 1.2 cannot both be valid for indefinitely it.rge M, since this veuld imply an infinite release of seismic energy per unit time 'Nev= ark and Ecsenblue h, 1971). Ecwever, both cf these equatiens a-e peerly defined 2: large magnitudes, so the arg' ment is not Sco helpful.
~n tuitica is often carried into the discussi:n of re'i:nal e ut.t.er bcunds. Ne'c. ark and Ecsenblueth (19~1) remark that earthquakes vith M > 9 in the centinents and M > ~ under the deep oceans are dikely, thcush they admit there is no real basis for these estimates. :n fact, if M is surface wave sagnitude M , we sha i see that M probably i es ac:
s
,x,,.
. e d a cc,,.,. 2 . o anyv..e.,.. u . , u.t. .. .u:s .2a . a a..z~,w... . .e . . e ., , e. : ....A. 2 ,
, a.. a.
. =..d
.. . .^.. a ..-"e "e e 'cund ',aa.-..'^.,n .?.7'.
. . . . , 7.a ." qu aW. . .= ' V
~
a have baaa. bserved severs' times :n the sil-ecean riiges, where the a s. .. if
. a .I . . . . .a . .d .
.~ cap. .. _a e $ 2. . . . .. t.. 2. s .e. .' a . 1g g.e. ..*. a A. %./, . %..e
. . .s
- a. . c.
. ~J
. , g.y
. a.ae . 2 . 3 ~1 .a....
.. 2..a,a a
. . , . . '
- 2; v. /.3*i,
, . ') .'; y.. ,.c.a.
,.. 3
. . .-.y..
. s.co~. A, ..
.w s. o . ..
. .. o..o.v. .
- a. yn .d a. .... w g *. . . e a. s . '*. . # a. s
. #. .". ~.".a.
~~~
.'c a . w .~. '"as '" '"a = a~ .#. * . ~. ~ ' .'. '.."a.s'w'""y~.'.'.'.
g .
..%. . 3 .,. 2 m'.. 3 , .yye a. .e '. (, , . . A
. _. a. sr a. a. .. s , ss
. . . .m .f a. .w g . . . .. .* 3 y.d . %.. .:s.i:. . ~-. ' ..= ; E. 3 *. .
~'
"."g .. . e.3 * ..s.*
....g
.A. .J .:
,J
.%.4....g 7a. .
- a. A. ...w a.3
..J.
- 3*A. . .: a.# .**. *** .
. . . '*=g I
.9 *. b.. .d.
S
. o area (see also Isteva, 1969; Veneziane, 19~5). :t seems likely that this study refle::s a general belief tha: areas Of 1:v seirsicity should have low upper bcunds to earthquake at:e (see section 1.5) .
It is possible to go semevhat bey 0Li intuition. Tsubci l195c) has proposed an upper bound to earthquake energy. He first relates earthquake energy to the volume 7 the strained region around the scurce, then assumes tha: the strain is uniform throu6h eut this volume, and then ases field evidence f:r the maximus strain which the earth's crust can vithstand i
(about 10-4). Then, if 7 is limited by the thickness of the : rust, an o
upper bound to energy of about 5 x 10'k ergs is obtained. !t is hard to assess the validity of the assumptions used in obtaining this result.
A very sLnilar approach has been given by Shebalin (1970}, though it is less 00nvincing. He quotes linear relations between earthquake magnitude and both mean length of focus and vertical extent Of focus, frca an earlier paper (Shebalin, 1971). He then uses limitati:ns :n both length and depth to set an upper bound to magnitude. The validity of his starting relations is very much Open to question.
Similar procedures have teen Out'.ined by Hofmann (197k } , who describes hev sagnitude fault-length relationships (e.g. 3ccilla and and Zuchanan ,
1970) =ay be : sed to assit'n saximum nagnitudes. Obvicusly this tr:e :f approach pretuppcses that we can clearly define the 10 cati:n and length Of all active faults in an area, that breakage bey nd the presen fault len6th is impossible, and that the magnitude-fa"' 'a g;h relaci:n is single valued :his is equivalent Oc pr:;csing cha; all earthquakes have the sa e stress irep). Each of these assumpti:ns is diffi:si; :: .'us;iff Shenk:vs and T.arnik (19~h) raise the ;cssibility ha; the este :f strain securula i:n say set lisits :n ::e naximum energy releas ei in 1r
9 ea-thquake. "' hey indicate , for exemple , that if .:pper and icver te nds can be placed on a 3enioff strain release e rat.h, the maxim m :cssible .
. . . . w...,, . . q-. a. v , - de eme.4.e.e.;
. . , . ... ,,4
. . . . s a , .
r"-^ as^ h. '. a- e * ' e ' a,.s a- -
.' ess .ka.
record of ea-thquakes already contains at least :ce maximum possible event.
"hese studies are s.aical of these attemu. tin s : use physical argments. "'he strength of rock, under various physical :enditions, is not vell kncvn. However, we kncv even less abcut the limitations en the size of the zone of slip, and it is this variable which probably limits the usefulness of physical arguments. The larges: inown fault area is proba'oly the 1960 Chile ear.hquake, which was abcut 10C0 km icn$ and perhaps 200 '.ca vide On a shallev die pin 6 fault plane 'Kana=cri and Cipar , 197!* ) . "'here do not seem to be any convincing arguments why fault breaks could not te larger than this on occasien. Could the entire Aleutian are systes breal at once, for example?
"he e.#,. .. . . o .# s + .,n .,. we ;.,
.. ces os ,..,a,.a... .- ...~.. 2...so. . .
4
_2 ,..,, p . -w
.me basi: -Oroblem can then be f:rmulated as follows: Seis=ic =cment Mn is w a.ot
. . . . . a. A,. %.,7, f,
M - . ==LWn n. _ . y, ;\
.o y e . . . .t 2,
a .%e
. .<g_4A.1.,7,
. . s .%e.m _,,....u..
.3 s_~,.,-s u c z... ...a_~ _a _z .. ,. .. s .a n. ..., \ ,
g .' s. .w. e .g.:..A. . g. ('s a cr.e,..,..
. . . . 3.- .
. 4C y., -
44 ,
2., 4 4 . . m. .e . . , ,
. .w
.a m.,
4 a .. .w, 2...
...g. .
la, . . . ,. c.ase...
- w
. e a.. .es -J.wD '*
w
.a %.#. .#.*.'.A.".
.n sa
- s
.10 = " ----
t
(. ..
7 ,3,. .
. . . . .t a 2 5.e. r. .m .p .. .,. J. ,. ,. .*
.*3...,.
.. .w. .*.% 4 .* w . ../
-.= y4. .% o..' . *,/
. =. .2 - a. .:
.g .'.==.3
. ..'E,, .' .
, - =
.. e ..
. . . J
. .. *f. a. .=*.J
. y . ** g 9 .a , . . .. .. . "' -7 . #.g . ,e *
.m. .e
. 4I. g* . 9 y
- ..a. ' . , 3c ,
. . ;. , .s
.,. . ~ a-
3n
- v So ve say generally write
.-)
.'. i - c _?
_ ac s.
if stress ireps are roughly the same (atcut 0 bars ) f:r all earthquakes ,
as has been suggested (Kanascri and Andersen, 1975), then limitations to seismic mement M de:end only on limitations to the dimensions Of the o
fault area.
However, questions about the constancy of la remain. Scme studies appear to indicate 1: cal stress irops as high as several kiletars (Archam-beau, 1973). In the eastern US, the occurrence Of =cderate sized earthquakes in the lover crust with no surface e. ression of movement would appear to require rather ,n'i fault dimensions and :orrespondingly large stress drcps. To take an example, if a fault area 20 x 20 kn vere possible in an area of stress concentrati:n in the Eastern US, vi-h a stress drop of one 111ctar, equation 1.5 gives a seismic semenn Of :ver
^
10'8 dyne-em (equivalent to ansM of ever .,, see Figure h). "his is probably larger than any earthquakes so f ar Observed in this e. ea.
We conclude, -hen, that while physical argunents support the idea that there sust be an upper bound to earthquake size, and suggest tha:
.w
...e a. - f w.e
- a. a ..s..
a...w 4.--..L- ,
- a. 3 4 r. na ..L.
. 4 ._3..
2 ,
.w4.a . . - , .
myy .. . c.... .4, ... a.
ca -- ..c +. .v e *. .o n s .. *-.' .7 . " a. aerr o". ' a'. e , a- e--- a. . =..- a a-
--- ^ q'- ". . - a. s . ' a . = . "..e sizes of these upper tcunds.
- .L,. s. , n. s
- . s=.~.4 -.
. . .- - ,_. u.,,,,,af.a. q . a...a....
4..<.,
A vari a. .-/ .
1"
-. "c.s " ave a.*.em-,es' y .- us ~a
. . "~ a. a.=.'-.'.=-
a .- * ~ .",.a.-
3a.g
.. 4a..s..d g a
. .# . e *. -.*"*"^
.b
.-s'.'-
a A. ....4 **^^ . o a. s . J' ia *. *. La.'<- ' * *l3 f .e- d s'.- ' '. ." *. ...*-.".-
.,. ' a.
- g. .J.
... 2.
.J mm.
=w. ..
3* - .
3.ew
- a. . b.d.
- e. -v h.....
- a. .w J
..-g s. -.o . . .-g
% 4 . . g '% a. . e. ....s-.m.
3 ." g3
- .ymp *** *
.g.....
w ' . ..y yX,..- 7.'g
- . , gg,x S
.a
.g f s.e q . . .%me- . . m. ..-% .Z3 . , . .. s.. .. .w
.). . .....-j, .t ... . . . . . . .:. ....
J.-- g.
.J .. ..
J' J. a.
o .. ,.J....
A
. .~.
.%4f ,,,
...C.,,.,
.s ., .., . ,. 3.1
,g,,,,
sa.uration of the =agnitude scale and individua' station dete :icn
- cmpletely obscure the presence or absence of upper bounds.
There are two pcssible apprcaches to the analysis of earthquake catalegs. The first involves the use of the frequency-cagnitude cu-ve, which is discussed extensively in rection 2. "'he other is based on Gumbel's (1958) Theory of Extremes. Gumbel described three asympto:i:
distributicas which may be used to model the distribution of largest events occurring in a sequence of equal time peri:ds through the earthquake record. The Type I asymptotic distribution of largest values :crrespcnds to a linear frequency-magnitude relation, with no upper bcund. Se ? fpe II asymptotic distributica includes the case where large events =~ 'aas frequent the.n vould be expected en the basis of : aller events, i.e. a non-linear frequency-sagnitude curve. 'he ?fpe ::: asy=ptcti: listribution specifically includes an upper bound. Algebraie details can be found, for example, in Yegulalp and Kuo (197k).
Applications of the ? pe f distribution generally acecmplish ne more than the use Of linea- frequency ragni ue statistics , and no upper beund is included. Papers using this distribution include Ipstein and iconit: (196o), Sayskiy and Katek (1965), Milne and Cavenper 1968',
Ocnnel'. (1968), Karnik and Hubnerava (1968,19~0 ' , fegula'; 2.nd he (19"i), Shenkova and Ka.nik (197h) and Shab ' '4 M sc: (197"'. 2cugh sc=e Of these papers nen, ion maximum sagnitud.e ea ~. quakes, it is : lea-D
'. " a*. */" & *. .'~
. a' -4#3c'.'s a* e .A .' a' '. .%.e ' '.'
1 a.'. .d.-* */ .Ty, " . " * . 'a.s".'..'*..'<..* * * '4.b. .#..' .". '.a a~ .$.
.. wa.e w i.7
. . . . , m. .e -C,.,..,..,,,
- . .. ', 4 .-.
- 2. .. b. = ,. e
- J.X a. a. ,. . . e. . d.. '. . a. <.
.. .. . .= ..$a.
.~
- 3. .
...e
( . -4.dd g g
.'.w..
- s. g .
g . . ,g pup g, s.
.. . Ay e e. w. 4.. 3 e . h.e N..,m.,.g...***
.. 3. .g . , e ..
.J q Jd . . , ,
2,.4.*.1..',,
3
. o.
,. . . 3.
.. . J. 2 ' ./* == C- *.* a.
P
- **4.. e. *.# ..e* .
. .. . a. "."k o. g. s.. # a.,.a .*.4.a .O ' a. .# . s- b...o a. . 4 3 .a.u *#=.
12 (1973), and Teguisip and Kuo (197h). The firs: er these studies ices net define the magnitude used, while the second is based :n Gu' enterg and Richter's (195h} data. They can both be shown te te f:rrally equivalent to trying to fit the frequency-sagnitude curve vi h a truncated distribution (Cosentino _et _al., 1976, 1977). We note that Kncroff and Kagan (1977) have argued than frequency-=agnitude statistics are to be preferred over extremal statistics since the first uses all of the available data.
To anticipate section 2, there is no doubt that saturation Of the Ms scale be61ns in the ran6e 7-T.5 It is interesting to note tha: most frem these studies are greater than M = 7.5, of the estimates of M s max and the vast majority are greater than M, = 3.0. As icn6 as saturation of the magnitude scale is not considered, there is no vay that the results can be unambiguously interpreted as indicating the presence of an upper bound with re61 onal variations.
1.6 Use of the 'evel of Seismic Activity Perhaps the scst persistent attempts to study the nature of earth-quake upper bcunds have been made in the USSR by Ri:nichenko and his ec-vorkers, beginning vith Riznichenko (1962, 196ha, 196hb). Many assceiated references are listed by Ri:nichenko and 2a6 dasarova (1975).
R1:nichenko's basic postulate is that there is a clear ut :;per bound to the energy released in an earthquake. Setting the : tal energy release 3 = 10k joules, he discusses the trebles in terms of I m
and K . He uses an implied relatienshio. between ener.:7 and the :tserved m
quantity, sagnitude, of the form leg I = a + bM 1.6' The particular tal;es Of a and b usad 3-a c; que:ed 'and are still : pen
,-..a. --v
_._*Soe....
ese ...,s.,
saw eawa .,,
2 awa 4
em ea_,,,.4..:,<..
g e e w sm -s e.e .ma oo e -
de . -am
, t .e e e as. em ..w ..~.
13 He recognized from the beginning that it was difficu.'.: er impcssible to determine K
=ax directly from the Observed earthquake :atalog of an area. He has therefore f: cussed on the possibility Of establishing a relationshio between K and the level of seismic activity A in the max frequency-energy relation log 1. = A - y (K - K )
L o (1 7)
(A is therefore the activity at the reference energy level Xo ). He has discussed the form of the relationship A(Kmax) in several racers (31:ni- - -
chenko 196ha, Rznichenko and 3agdasarova 1976 and :hers). 3riefly, his argn=ent is to relate the energy K of a2 earthquake to a volume radius 3 (for Central Asia he obta.ined 33 = 0.315 10" 7 ~10), to avers 6e the activity A over a circular region cf radius 3 to cbtain I and then determine an f
emrirical relation between I and K max
- . ?cr Central Asia he determined (Riznichenko and 3agdasarova, 1976) log I = 2.Sh + 0.21 (I (1.3)
=ax -15) vhile 'cr Japan he found a better fit with ics I = 2.Sh + 0.39 (K=ax -15) (1 9)
So %
'"hese equations are intended to be val.d for 15<K<19, or 10~"<I<10~~ ergs.
I'he 'crm of these equations vas ierived very artificially (Ri:ni-chenko, 196ha). K vas simply chosen as the largest event for a given max
. , s. '
. .344m. .w. .. r. 3.4 6. -, 33cv .
+* <*
e 3'82"r.s d
- - - - 1 4***"a.# --
- a d # -.' - "- -~"*
ms. ,*e, m e A E ,ea.e . s. ., 7 w aw a
.b4.C-
~- s ' ** , ...
. s . 4.J . ....a. ., . . a 4 2... , 0 , 3A.u2
. 3ax .
..4 .,.e s . . . a. 4..w-
. . . e n .go.3. .* 4 . . e t.i.
.,.w
.. .we .- 2..
3.... .. OaX 7 ./ s.a... . . a. ' o..e
. -..4 ..
. w... . , -.u . 24
~
- gg w .o. ,'. v a , .~ ,T, _' ,' . . n .'. 0, 6 h " .-" e .' .w" s ". a.~.~ a' ^~~._a*.a.'.'.
.a .'". a.'3.#..'.".
- '. . .E J a. ' *.
.S . 0,*s 34.w 6 s 0,6. , .e sv o.s..e . s..g e.
. ...e.
. . . . *4... ....
. w3 ..
. 3 J.
. . s. .......2...........%...
r a..%.
,w . s ,3.. . . . . . .. * . . e .* . . . . J. . . .*
.g.
.. - " .... 4.:..**
. j. . . . .. .. .
r 3
r 2. . . . , C, a
. . ..a. . ,,4 2 . J' . ...'.'-
t
..a 2 .: :. ..-. . . % :..g. .
i .
l'a .
log 'I log 'I = A 'rK
< v= ... . ,
ao g s.,,
w
.. .,, . _ . 2.. 4- .. -. , e nx o ,
.c s..-. . . a ..s a ,. ....,,.._ ,, ...,,,e....;.,..._-
__ v .. .. ...
.~s -cu g .
.. . ..e g- e_, , and assures- a ' '_ .. a a -
. .. a. ., .n. -a._'.=...
'. ~. a. s- .'. es .. =. .. .
"cr energy values belev I .
=ax
Obviously, the .er:bles in this ac..creach is that I needs to be
=ax de . e . ...*..e .4 '_ a. s c m e .- a.. e'
. . ' - . s ' e .". .- a. '. w.. e a . . e .- =. .'__' ov .-4 . '. a. a s _ _a '._v
. . ' ' s . .a. A .
~4e must allev, hevever, the ;cssibility that successive application of the equation in varicus regicns (e.g. Gorbunova, 1969: Drunya and Stepanenke, 1972) may improve the ocastants by an iterative er "tect-strapping" methed. '"he logical basis for the expression 1.3 is not established.
'4hether or not it works in practice is less clear. he authors :cspare 31 la ge earthquakes in v'apan with the predictions of equation 1 9 "venty-.cne are found to be in apeenent , 10 are f0und to be larger than the predicted Kmax, though the authers note that uncertainties in many of the epicenters make it hard to make a fir:n conclusion from this result.
Se situation is far frcm satisfactory. "he existence Of a relaticn between I
=ax and A is not roven, and atrears to be =cre Of a hope than a scientific fact.
~4e should note, in passing, that i' the me.ximus value is defined
- t -. > .w. en .. w . ..,.
.m. s.4 .e., .
, .,recao 4
_2_./.
. . t T s s ,- .i. .y , _ .. ..- ,... ,. . _ a . 4.
n ., e . . . . ._.
- ax , , .. . ...
the maximum value and the rate Of seismic activi;y. ~'his has been described, in a most obscu e vay, by Housner '1970'. His argument may
" , l.' ^vs -
".e .- a. s ~= . e d
- a- - . '_e .s =. s a- , e .= _' d. .la. a - ...S.c -d=.> .".- a. .. u.a. . e , -
=agnitude lav of the form les 'I = a - bM 'l.12
.p a .. .; _ a . .~.g .ga g' g .z..
. . ., . ...,.. , c
. n. y ,. . ., o. , y:_ .u zg3. _ :
. . , _,, .u . .a..,
.. 4 ,,
- - - , 4... . g g
. -. . 4 e , g , . ,4. . . . .
.. .s.-
..,., ,y. . . . . . ,.a.
. 2 a.
. ._. u. e
.. . . ,',. . a. gy _%_,..
.s y s. ... . .
%. a.
. 2. _: _a a,
. a. s. .,,.__:e
_.e_ _ .s. .' ,. ._a'_r
.~~:C. - 5*
_i6 2
' hen l e g 'I =a-bT (1.11) n sax W-
.. , vo '. '.. .'. ' =. .~.. . .~. e~,'.. . r. .=. , v ' ~,. >___' ' ' .' *. .~. . ., a a..d
- v o_ _' .'.e s , v e.
_ . *.. a~. =.
o o
- c3~ '.{ = c , k , S( l z a ,%. .V.~ Is _O )
_a . =ax . ,
n e zax so,
" b 1 =-s'
'o Y=ax (2) = b Y"
=ax (1) + .-o -
(1.12)
.o.
It is reasonable to set b : b: 1, and then 2
o o rmax(2) = Tmax(1) + (a 2 - a, ) -
er ,
c o o si
-a rsax (2) = r=ax(1) + leg . ::.13 )
,I _
.o where 'I o is the number of events with ,ernitude 0, which may te ca?.en as - -
an indication of the level of activity. in a simple example, if a es 2 o
has a seismicity Of one-hundredth of area 1, then the T value f:r
=ax 3
area 2 vill be tvc units ma er than the Tnax for area 1.
The reasca that Hcusner's (1970) ars.::'ent is :tscure is -hat he .
tries to asscciate the above with a true '4 max value, as shcvn in Figure
- 1. Clearly the ana'ysis really refers o Our .mtcunded frequency-sagnitude lav.
.n -2 , , 7 , , ax_4 . 4.
. g _,_<.e....a
. ... a.
s -
,a.__.e= ,..
an.s.o
.. ,.c...._
.~ ~. _ a .' - .". s. %._'
.
- e ' ' n- * .~ *. c " . '_ . - =. * . *.a~o_'< =.
. . "
- e .'. =. =. . s e ' sm_' . _'. ~. .' v .'. ~.~s s .i e , w,u , s".c ~. . s o- . ' . . =. s '. ab _i _' e " _4e _- ~, . S.. e .= . "_~. "._. 4
_ _ m a. v a. . . ' e v a _' _'_' ._' . - . . '
this relati:nsh.i; has been essentially .:n-existen:.
_.~ v
.a..e . . u. ..
.. .e_ _.a... .
...,,..a
. es
- v. a ,,
... 3 . -.
a _, a . . .u...,
. . ., .a. ,e..a-a.>..,.~>,_....a
.. . . a . .e .
=. % 3 a ,a <>
3
.2
,- sm s . %. _' a. a. s . '.. .. . .s '.r a. a. ..
.%., .a. y a. _ ..'.a....;, .
- a,. . . .. . _.
m.a. .' $ =. a. .
,e
.I e .
activity alone there have been several attempts ;0 include a variety of
,w .,.w e. . 3,a.c eh/' s '. a a.' a ..d 3- a. ^. _' . 3.' m _= '. .'.n.."..a*.'~...
34
.. .,n .4c ,. , nx^. a...d -' '" ." .'. .a c' ..e ( .' 9 ?, h, ' .".av a.- . - . , a - a.
. . d. $ - .4 ^ ^. .- a. .' a'.e A. . ". .e estimati:n of I using the level of seismi: activity, the gradient Of max the 3cuguer gravity ancmCy ( suggested by subci,19h0, and Berg _et _al . ,
196h), and the velocity of vertical sevements dete mined by geodetic and sectorphological methods. '"he three estimates vere cembined tcgether to obtain a single est'-ate using weights of 1. 0 for the seismic data, and 0 5 for each of the ::her methods. '"he resu.' s are no more convincing than those based on seismic activity alone. -'his paper is notable, hcvever, for its extensive collection of references.
Shenkova and Karnik (197h) state frequency-energy data are not reliable encush for the es:N icn of Kmax, and rge the inclusien of data on " environmental properties and the rate Of energy accumu'ation" (i.e. Benioff graphs) . Ecvever they give little indicatica hov these
.,.....4 .,. s o e .4 ., o# rma.,. 4 em n a .,c,, 4 .,e
... a. .,3.w
.w..e .
In view of the interest of several Russian secphysicists in pattern
.r.
. ,. sa.4-, ... * ' ^ n '., ob.' ems ', a e a. , ."^...~ a.xamp.a. '
, a.. .' #. = d. . .. = _ ' .,
. . ' ,a ~. 4. ', , '..s'- ..c .
su.-. arising that attemWta he.ve been made to :lv these methcds ;0 the
- a. .. *
.e 4.... ., e rs.4 u., a e..CG s. i.,9 3ax w .aa. .oD . As .
4, 4-s .a a A a ,. . m. , e . j,. . : ._. ,, , . ., .
2
.fa (_,ga::,,i,,
e .d ar. a-y p.' .' c a',,.# w. - . .o .".a. .a.- ya ,,". 4 a .a - .- a. 3 -.' .-. . # a- .' a. 2 . ". a. d. '. '/ .c . '. a ,. v o . ".
7....7a.,e /
s_ gaf). f m. s ,. .
3.. . . .aeo. 4-
.s . o . ,.- c..
. g .. . . se
.,.,,,D a.4... . . . , , .-
.~% c w . s p .,.y a %. .' . . e a .. . ,,. a. s . .u a. ., A 4 , , . : .. . . . > . u. ,. .uwa. en.
. s. a M 2 <
. .3 . . e ., ,. . .. ., .,. e. .%,... . . .
./ . - .S.
. ..w.' . a. .,.L. ,. . a.
.. .. .e -s.24... . A. 4,r_'.
a.
- w. 2, w n .r....
-a a...,.s.,'.,,...
- ....:.,.3.-
.. . c ..4. ,w ,, b. g
.go w.f. ......J. .q y3e.'**....**.',, .
iia..=
2..'. .'c.'~.~.~'*a.
- .. , 2
. - . s. .. p...p . E . .p.4a. ,
g
. o. l..: .un .t
. 3 .n =. 2 >.y+..,3 m,
s 7. . ., *y o . y .' ,
- a. . s. . %.... . s..s..GA
. . .a ,_ ef 2..
._e.g o. .e. e,. g..a. .
f p . A s ,. g, z .
%,8
. a .,
m .J ..a .* s ...sg . % e . p p.
- a. .. .. a. p .%s . .. .. .. 2 ..
- 3.,. #
a.._*
. ...a#
' .f.' . .%
,1
."1 3ax
,o
. %1 n.,.e k.a . ' . ,- -o
. . b ' a.m .~ ' '.'..#. s =.. .a_' v. s '.
. . . .' , "..cvev a. . , *^*
. ~ . =.'_*-a.2=.' '.~ /
. "..e .a." ". .c r a . .. .. +..'a..- --
~ a'ed. u. a. . k. . a. a- --.,.' e.a.
. .- .- . .' .a. '..^ r. s ". . . , .- a -.' .a. a-
, .c . rn '4 ,,..
. . . ... e ..n ..a.a .t . -. a- u.us . .a- a. . a_- .
e,. ..> . a. . 3>o ,..s . =>.w-,.
. . a.. a.
. =ax are not readily available, the authors -;. sed " estimates made by experts".
This introduces such a strongly subjective element into the a.na'ysis that it must be regarded as meaningless.
.3 0.w...e. s..,d o 4..a s m.yo . ace ... e.
.. . a,a ' a =- s hculd ' e - e..*. '. . .ed , '. '..e .".- s * '. o
. . . --- ,<. a. *. =. . .a. s s-and the second because it has an interesting approach to the problem.
Caputo (1977) has proposed a ecmplex sedel which purports not :nly to account for the linearity of the frequency-nagnitude relati:n, bu* 00 predict the maxi =um seismic magnitude and sc=ent. The assumpti:ns :n which the auther bases his analysis appear Oc te :capletely =reascnatie, and the paper is meanin6 1ess.
Smith (1970', ; On the other hand, has propcsed asing geol:sical data to obtain a mean rate of siiP for a fault :cce over the rast 10's of .
thousands Of years er 'cnger. _ Then, if the arequency-=ccent relati:n-ship for the area is linear, and can be defined (see Chinnery and :icr;h, 9 # c, . 3.4.wi
, . . . 3 sa. .. ..e , 4.. .,.ss 43.. . .us',, .5en
. .w..e a. a, . . .. . . 2.
2pper bcund sement that is consistent vi h tserted slip (3rme, 1969).
Smith uses geci:gical data of Hamil:en 9~5' to obtain these u:per
Cund - Cme . .s- '. .'~" . .~.>. " a..~~. n v =..= '. ". ac h .a ..rg e . '.~c ut.A . e ~ . . . . ~.'.e -s e'
= .' a *. Ve e or .w. e 3c s '. .* a. .a a w- s-a '.a. '..av a. s a a. .. , ' . " . .
%. ,,13 3,g..g.o.w
. , . . a. .,. .
. . .-a.
e e s. t..z.a. .. 2- . a. '~
a eenv_'amg y.
w a ... .t .' ' , . ,3.4 .e . n. eva. . $; a. a .t r o .t e v ~ 2. ~..a.s e o. .~ - *
- 2
.t4 .. a. .
.. . . . e 2..ao.,e g. . cf. . ,C wme .. . a..' 2.
. 2. .
.o.u.a ..' . . , . .a .. a. .s . . :.
.f a. . .e ..4 e
s -- .. , a. a. .. ,g. .a. A , w y. a. . .. ,,..,,, ..
. . . . . a.
.a. . . , s . 4 .. .- e
.4 4 . .. . . . . , , . . .
g.a.g *. g4 g* .* ** * .
.i .n. .~.a. .J.a.. . a. .w.
. . . y.w a. . g . 4.....n.
4
. 3 . .i7. 4 9. .. 3. . .
47..n. .. . . b a. 3. .' J.,.u g g. . . . .
s Ly
.ea,al.. s./s.,.,.s %aa. _/ sep. ,. n _4_ s + _ s 3. .. . .A. . - . . .. . .,- a. 5 e .,. .... 4 a. e ._.- ,. e ... . .
. .-. 4
.,.. c.cg<,a.
, _ , . ,, sm ,,.. .. .. ., >_s _a.,,,/.
, . . . . >..a .. > ><. .a,,a_
. . , ._r,.a_.s .
. ,.,a--._/ ,,,
much c,.h4s . . . s. _,p .,,,, a.f .e.4vd v.
c .4,e,
.- , nd . .. . . , 4.
.s,-.,,,,A w
.,e .
a,s,
..A to other neighboring faults. To put this e.nother way, Smith's (1976) approach requires that the earthquake precess be stationary over the pericd of the geolcgical data en each faul: Ocnsidered. Sis is a questionable assumption for the fault scne as a whcle, and may be invalid for individual faults within the system. And, of ccurse, there appears to be no way to apply Smith's se:hed Oc regions such as the Eastern US, where geological informatica on fault slip is net available.
1.9 Discussion and Ocnclusions The basic prcbles '.a attempting to dete :ine the maximun ;cssible earthquake ih a regica :an be stated quite simply. 'f the earthquake record for the region has a length T years, then evidence is available that bears on the eartna,uakes that have mean return periods Of up to T years, or a probabi'.ity of occurrence down o 1/~' per year. ~%is evidence is not necessarily geed evidence, for the largest earthquakes in -he samt. z, e .
- m. m, e.., c e -n.. . . - ,. e ,a. *.hr'uak a. s 2.~,
3 .
s .o,,
- a. .=.. 2 -. . . , '..e ' a s -. '.'. e d -. -.' .=.
veil by a Poissen listribution (Ipstein and icnnits, 196c; .ccr.1: : ,
- 9 c,c, , . "--e 'e . . '.. a* ' ' ' . / '. "..a '. =.* .'. =. a a- ~. ,.n=.=..=....v'.'.=.=...=.'
- . ,.-=.a'.~.'.'.'.-'
, o
. _.l, v. .y. ,. , ,
.,c ,...,.
-. .p- . _.a , a ,. e.,. 4. , g ., .o
- s. 2
.2 y *l tv. .. .
.=o , 4. .o .. m. , .u..,. .. ., .. .a. ..
- z _ . ./
w s . .. . .u< .. . .ee
. . . , a,
. .. o. . ..c.. _ = ... -.. . .. .
.,.w4 a
4
,e
, . . 1-
.g.. ... 3...A..A
.e
.w.
. . . . . .' 3..'.,.....s.,.
. .eg _3 . 3.y.o y% ...- - . .
a . .* f , J -
=#
o e , ./
- 3=3 a a. 7 a. .. *. .* .
40 "o phrase this another way, a 100 year record :f earthquakes vil'.
only give reliable inf rnatica (at the 9CT level) f:r those earthquakes v4.1
. . . -.- _ e o .,. . a. . 2 , , e. . e.d . . a 'oc'a ,. i C s a...a.- s s .- _' a. s s , ..- a.
2 . . .a. _a ' . . *. -
bability of . 225 er more. In practice, of course, the '.ength of the earthqua%e record is often 00nsiderably less than 100 yea s, and this applies to mest of the regions of the '.'SSR studied in the quoted literature, and to California and other active :enes. Clearly, then, a 100 year record of seismicity is only adequate for the determinati:n of maximum possible earthquakes if the mean return periods of these ee. hqMes are significantly less than 50 years. '"his implies that the maximum possible earthquake =ust have cccurre<1 several times during the pericd of observation.
In all of the literature that has been surveyed, there is no case of a specific region where a maxist:a ;ossible earthquake :an be : lea-ly defined. 2ven when all regions are :casidered together in a global earthquake record, the apparent upper tcund to surface v?.ve magnitude M S
can easily te accounted for on the ': asis of saturatica of the magnitude scale (Chinnery and North, 1975). Perhaps the ses sefu' :ontribution to this area that could be made at the present time vould be the clear and unambiguous demcnstration of the existence of an ;;per tcund ::
earthquake si:e in just One region, anywhere on the g;:be.
It is necessary to add, here, that we have not a :emp;ei o iefine
, , n .
.. % e .. a. rm + .. a.. 3. w.
.. 4 s .4 1 .%c ..s..
. .,.,1.,
. ,. t, s e a. , .
.. . ,x_, ,_a, 2. ,. _4 a.,
. .a
.se7. s.. , . y , 3 , -~
- y. 4...,'n
. '.a s '.e e n =m +. y a a- .. ' a. .d k. .v .'e .e m ".=..-. ...'
.,v.'..-."
yk .a ., %.. 3 y, e p. o . .-J 3
.. ..e pLC . .2...
u
- a. g 2 d
.. .7. . b. -*2 . d-.. m-..
- O. gy .. .. .Y. . 43 s' c-4r . .-- .c A -
- gg .'.a-.'..P.*.
J.
'. .#3 s* '.' s s #. *. .
- '6**...e..'* '.*a,
. . . . 2. .'C ' *. y '. *. o
. .*.*w~. a. .'..$~.
y .# * * . . - '. 2 '.a-'*.*.'
y 4
- a. y . a..c..
. g .Jw. * * . * . %. 4 a. gw..%.g. g.. .g*p a. s.
J
.T . ..
- 3
.:a. .i .3 3.4 .*
g.
.w a. .
-j . y ..c. .g %.
3...,-.... g .' a.
i J.A. . A. A. =. *. .*a.**
$ . . .e. * 'a.g *. .q. .e,." . 5. *. *. g'?
- a. n. .a. *. .; .I .e. aJ
~
.v .i ..r.s 3 w g. a. a. y a. ... ~%..
. a. ^#.*
.=..'. .
3 s. g .' a ,= .. J. >. 73 y. . . . a.
.a g g . J. .. p
. .gh.
. ." a. .
s
, . 91 7 +. a ,, e .=.. s , '. " e.n , '.k..e. . a. x.' s *. '. . . e a e _' . .' ^. 2a '. o- =.~ " ".a x_
_ . .' a. . e- .' ov a / ,_,3... .w . ..A 4
. . .n . . .. , . se s . ... 2 , o, ,, o . . . .,.. . .x.2..e
.a. .. c.
. _ . . s4 e . , ..<<_. 2 ,-,.,a.s..,._.. . .
earthquakes. In spite of the deep seated belief of many seis=clegists and a.a *.'..qua%. a. a. ..3_' - '
" .a e -s '. ..a'.
. ,,e .- '.c'~. ' A s . .u --s . a.x's+., 'e .'.-/ -a. a s c nab'_ t'.
approach, given Our current state of kncviedge, is Oc assume that these upper bounds are at rather high levels in all areas.
'4e ar a. -'~ e.- a..",. a. .'o.~. e d .#. .'.o .k..e ^s _' a s o-. 4. - me .' od o '. =-.' . ,_' a. a. .r. .- a. -
pelation of linear frequency-sagnitude r frequency-intensity relation-shit. s . his raises an additional or:blen which deserves discussion. .
In the context of the evalue.tica of the seismic risk to critical structures such as nuclear power plants, ve would like to establish a vay to determine the size of the earthquake that Occurs with seme fixed risk probability v' thin a given region. Fo11cving McGuire '19"O and others, we nay 2se"u117 set this fixed prchability at 10 " per year. If the earthquake prccess is stationary over 1:ng periods of time, such an earthquake vil' have a mean return peried Of 10,0C0 years. f the process is non-staticnary, this state =ent is meaningless. Ecwever, in ve . nave very _,]. t.,e a:. ternatire but to assume .na; :ne ava:._- . .
pract,ce .
ab'A a. .- a. a. ^* . d .- '. a. .a _"qua'<. a. s _' a- .- a. , .~. s a .. . a ,, _' v a. ~ '
. m" a. .- a . e e.. .
- . " - . a. . c a o.' *c h .==al' ' . ^ '.*..3*. . _
~
=. ar *.* q".ak=. o- _ ' . . ~ . * * . _ _ _ - '-'a^'a.e ,as. 2 '. ~.'.a.
. . ~na^.-
- 4. .a. e... ..... . ..
.w e r.,_en . .. __a., : 2 ...
_.,.e.a..,,. . ,. . n
. 2. .. . ,a _../ . _a. v. . a o _4.e_ . . ..
. s .. ao 4_ ..4.. .. ...
... / _ .. ., 5 . ., , _4 _ e.. ..... < . . . s ^. .' a. a..' . '..'y' '.. = 'e a s . ' . . .'_-a. 'd.e d _' . a. . .- a...a = ..
.=~..a .'."_
., .w. 4 .,. a ' . e _ .3.. . ,... .-.a o. 44.a s. .3S.A
. . w.r
.; ..... 3... .
2 0,4. .e 3. .
, -c ' :- . .g as
- 3 . . w 3 ., .
5 . . - .- - .
~ - -e
,wg 5..a.
1
..~.4.
e
..g. , . p.
. .3._e,2.s--.3 3 ..ya w. . > . A.
..a g .. =.. -.
. . . . . --3
.:. e.
. . .o... == .4 .! . ... '{.y
." ..s;_ .3 .
2 .; 'g ej a. .,; _2 J *..3==
,b
. -. 3 .
..g
>.. . g ..O g^ g*., .a J. ..
_.. .g.
4..
s..a.,
- O,
- 1 ., :% d .f. s- 32 a
' ,.w.~ J ..,., f --
3
.4 7.
. _ , . . ,e..,
/
22 e .
. u.. ., .g,< .3 ._2 / s ,. . '_ a . .= '. . . ' . " . .
.e
- 4. , . _'sa
. ._,4_. y _a _
.as 4 _. 4 . ._./ .2
. . _ 3,.
3 . . . ,
sc e __-~3
_ . ~.. r. *. ~ c e' e '_ *v o_' v .' .c o. ' a' n _ . . . f .*. . .' c *.. _' a- 7 0 .-~, __..=.~".....m
_..u. e .- c =. . . .
. a- *. .e. .". ..~ ~ . ~ ^ . ~ ^ . _ ' a =a ' _' a. -
- v =. ". . a*..
_ . . . . ' a- '.c..~'-..'..=..-.s*.a.-->..
_ __ y . .
into the stati:narity of earthquake processes in varicus teetenic environ-ments Ocntinue.
The cst premising avenues f:r future investigs.ti ns into maximum possib'a a=- ' quakes vould appea 10 lie in three areas. Tirst, 'e need more infer-2.ti:n on the nature :f the strain and stress fie'is in seisnie _
zones. Seccnd, ve need to improve sur understanding of he ultimate strength f crustal materials in a vareity of tectoni: settings. 't seems lile'y that the true upper icund is controlled by -he si:e of the
- . = c ~_ mul a*.*..v.o s'. -a s s , .'....d ."..a. a x'. _' _' *. ,f ^. #. . S. .a. .. _ a-'. _' .- . c '.e. o
. e 3_' e.n _ .
.f ..g.s.*.d .< .$..a. -. ess.. -'h.
. i .wd'", ."e '...'.^ ~._a.'.on '-^.m 3
-a^
..'.e'^.=._' a..d gecmorphci:gi:al data on long te 3 f au'.t slip, where surface f aulting is visible, sust place scme :enstraints On the largest possitie earthquakes (Smith, 1,076). his apprcach needs further development , though -he ques-ica f stationarity may lini its usefulness.
a
- 9. . .p. L.u.. . .,a. . ,_ev. ,7 . -
c.% .2. n, . , , , , ,
.n. ..Go,
.a.a'..s 2..* ,. k..o
.. ,.a . a.
. 4.4
.o.3 s o .e a.i_' c h. e._i a
s %
A 'cgin.a_' aca. .. .o sea.k ."s.- 'n.'.. a..*.n
. .^.r. . k e a. x .* - . e...~. '. .'
, ,- e .-
bounds to earthquake size, and the variatier of these upper tcunds with tectenic regica, is within earthque.ke catalogs. '"here are basically two 11nds c.' aa.a3s, *...ose ~ ' ^ ~y.4_'ad. '^., r ' s '4 - .s '. ..e .s.a '.-.m
- i
.4 . e d - a. 3 .' .-a o
'cca' e vc. .. , .' .d ,h. o s e .-a.m .d _' a d ."o. . "e wk.M.a. .. v^e. _'d. .s ..4 . a 3 oka_' '
network of stations. 'de have chosen to begin this study by analyzing the global earthqua'ce cataleg, since this seems nest likely :: :entain evidence. for regiccal variaticns, if they exist.
In order to be useful for this study, a global :stalog nust have two important charseteristics. First, it must te ::splete, particu'arly for large earthquakes, and preferably for medi.:m-sized eveni;s as well.
Second, it sust use a clearly defined reasure of earthquake sagnitude which is uniformly applied to all events. As ve sh&11 see, this turns
.o '.,e .=. su-h ~c e .es* .'-*..'va.
.ut m. . . .o-di .'.n .kman '. =,-,-ea s ..', . '.a. =. .
.e.s S ., s .4g.u. .
C eva. . al _'3s* *w a.' .a.al "5s ara. . .=.v a. ' '. .=.v. a. . '. ..e s .a .' .. v ' " 4 .' ..5- s. v a.. . . o-sanca. . weu a a.- .' /. ~;, C O ' s .' .. C _' "_^. a. A ~
~
. .=..*. e.~3 au~ u*..* '. . ' . . = . - ' _' y' ,% ,1 , .,~u'.a , _' ;, o' ' ' -
a ,d .>.v..% ,. ' .y _ a s.,,; ,i . %. .s. # .....a.e./.
. , .w...e g_ w. a.- 2 4 . 4 % . . . ., .'
24e,
.a s . a. .... ,, S .go 3
.. a. .*/ ,.C C .,. .. .n+. .: .. _ , 6 0 , =. .. .'
m" .a .<. e . .= . a. . 5 3 -3 - - - = - - * - - - -a- -' --- - - >-
- ~*
.45 .~
.'=S.~=.=. . ' . ' ..c u- k..c cc g~ ~. .~. ~. .' ~.~.r . *4'.*..'.*.*.s.='.'.,_'.'a'.~.=... ..' . " e A .- .' '.
- g 4 4. . .aA
. .s a A. . .,4 .: C y.,.e . %, - .m
."1 e . .gw ',
- ,*g.C O.*f .' .
4 ., .w
. a. s. o '.; . . .*^4* y J. ' , 3 ,,..%
m C .,. a. .,..i,a *a
.J .e.*. a *a*
3 . .=*.e s. 3 D . . b. p a. e *'s.
2.*d .o. .# *..s.'..a.. .'.a . w c .-%. a. .m. . b. . .' 3
.==a..*.'.a'#.,
...a...a..
.3 . . . . .J. =. % 5 7a. / 4....
. 4 .g.~ . ; %. ~. 4. ...%.. 4.,. .. . . .. a s.,a.J .
- a. .,,. a. 3
- g. ;. . a .'
. %. */ , y a,.,
I g ** *
.g.a ...a.J .. . . J.
. "."% a. . . .R . . a3. ..
- 3. A. %...
. 4a..J . .
. .
- w*d*
..a.
- a. J , 2 a.4
..3
.-'.'..a..=...-3
.; ,. ...i a. ' *. .?....a..,. 1. a. * ;
..%..a. . . . . s. a_ .
, . . . . . . %. . a. .... .%. , -
- . . :. 3 . .3.. . . . ~ a. j. . , .. ea . .a. ;.......s.
. ..s
a e
ch .
- o. s... _, e . --. . , ,.a . L. _ _ e .. .. a. ?.u..e
- s. ~ ,-, e-. 2..e_
a.a ., .
.,..e.-
. <.a.
... ,, r . .
<.,.,...e.g'
. s s... A. ...w. . .e 2../
. .. .. ,,. o s. c u,.. ... _,c n....s w a..e
...... . . ..-, , .. .a m. -s..,.
o n...e-......... t n. a~_ s.- c..,a.-.
c e..e ,,
,,, ~.,... .e . t ec\
s._ , u_ a s .,u..c.... .- . . .. - ,n.__,n... a.~
~
ue
.e . e _ e -w .u..a* ,, .*+_.-,- .* e* .s ' va.
a..a._
4
- . c_. a.a.a, . 4. . c3..,.u.
. .w on . .' .. , a..d '. .=s".a. c *. a.
. s*.a_' 3.- *jpic a9 4e1ays
- n
. . ."..a. .. ".b '. .' ^ a . 4. -..a. - #. . "..e C
-v"
- a+.a.' r - a e-a*
a - - - .-
fr:m two to three years. 3cth the PDE and SC :atalog began consisten
.cu.<.e u.L,_,_e.<
.a p,.,duc.ien r n a.. . . u. ye , .s. .s . ,..e . e. _~ c. ch , ,
,-A s < ..c , o.u..e . .
have saintained the production of very uniform catalegs.
3cth catalogs, since 196h, have recorded a bcdy vave magnitude ab for essentially all events. This magnitude is based on the maximus peak
, e gy. ,.- 3,1.' '." da. '.n '.k- e '. ' . s . '. a.v a a. a...- a.d s- . '. ."e T
. ".ava. '-. ' va'_ . - . .
s" c. *. .a.. '. ^.d '.a..s . . ;ma. n *w s fo , e. a . _' a g .' a.
s . *- . a . ".. a. .- ..a.- . .v .'. a. s" a. r.e
- . .v-' a- - d.
centered at about i h ). Surface wave magnitudes M S
.a a pericd Of about 20 secccds) vere recorded very irregularly, and Only in the las; ye. a .- .- - .vo '-ava. .**. '. em y '. s '. a. a.a- - ada. *-
" ea s"-a. . V. . r, a - " '. .'
. . a. '. a s .' a .
.s
- 7. e .,u,_4
.s o. men.. .wa. . . .he .a.a. . . g. ws e .., p _, . e , . . , a _.
..- . . .. s .. ., o... -c . s - n . .we bo Qa, yse. =gs.- .<. . . sae .
v
. c. .~=.asc.s- vk- .' ~. "-
. 1.- a. '--v ~. _' *. .. a. e a' ~ . *..e .. e .c.
Se3.
4
.. w .s, .wis-e.
os ..c t 4.eS.4 .w -... ., wL.
.w,,,
e
. s _ 4 _ .,.... , . > a.. , e
.a- s.g ...,
o w
- C ". . #..
. A*. *. Cn"7*. s- w" ". *. _' ' *. e '.4 .w^ 3. "a'/*.
- "Cvn. -* '_ ' . 3-a ***-
s O eOT a. X-y_
. e,n,a, .l5. 3 , . ' C, ?, ~? )' .
a ."e u aa.c.*.-s . u .".a. "o_'_.v .
va. a. '.a -- .- . u- - . . . . - . . s. .- . . ".. e . .- ". - o . a _' . s-de * *
.# .,. w g +7 e. .e/ ey.- w s , *. .' m. --o' .o-ee
.a- . .l a. .J g 3* f o- ' _* S-b . '. a. -- 4..a4.a.4_ ..e. m g s* .*.,. . J. .* *2.g
/.w..-. ... 2 , . & .4 _ ,. A. e. so m -
.4.4. a . . .s 4.., .C.1
- .t.
wt.
. .. . a
.a a 4. . .e.3.,,.- . 3
.4.....
.. ..)
.s .e .w
~
g,. g_.f.= e 3 . e . .e.., . -y .'.3** -. **C" *. .^ #. '. a . a . n ^. . a. .* - - *. A. .
.? . _0 Ts- s .- . -g b - o W e. ..-.'D.'a.) o.**. -
. J. - ... ..s. r..' a* ..* .& *
- O ***
- a. .c..g ..
- 9h.p 3 33 ..-* ... k. g a. E. ,. . g. .. w. 3.
.%.-* 2.
- E. 2 * #...S.. .k---..A.E..**..J. .T
- # b 3
-- 4. _.
g ....%.
.- r. - 3.Q s. q
..-. 2. _
- s* . .-.
- 4
- e. S.
- 3 V . *. k.
. *. b. . a. - e. . 3
- 5..* .=. .C . . .d m,
. k.e.*.*.*2.c..=.
i, e d.
a.- s. * "..C.' ' '~'. a.
. *. *g'.' e .% C *f ' "..d *- * .'** " *v' .' a. . * " o ~b-*.-" "#'Ae*
~~* ^*"3"
- - - . * < * - - '*#'*---#'eS -
.,~up*1
-- .a
' * ~
s _ ,, ,: w ]' o.u..ec .e,y er a. C . a. , . S ,
. . 1. . A.
4
_S ,m e.n.. a. ..e .4 .e..'/.
.. a J. o. s. . .y. . .e .' a. 5 n a. S ..
.4 -
7.r. .. . .g. . w a ,s... a.
- 3. fen ..4. o. y e .. 4.. w+ ',4 . ~".".C'.'.e#.".
. '. .'*. *. S e .
- o ..y y- .- EC . . *. s- rye . .-w w
.e */... */
44 w..s.e. . . ....,
. . . . w..e.f-3 ..
.4 4 ... ./.. j. . a ."1.4 .* a - .. s- . * . .u.o
~w o. 1 ecd. -ry--.A. s .e
.v .ve game Ag*.g o.o* ,. . (' gee, #^p .w s. x m a,y 1
. , ak . 4 . a.e.w.f ,
- g. d T..Jea*s %e..., y . 1., , .o...d W.q a'g g *.
ap.d .%. '.go. o ,- , .'. y, a*r *t } . .o,e a. g. u . g e w .o r.. '.n..a a . , . .,..d w. e n. .o.. , 3,- .%. .-- .r- a. c.ue..
. ,f ..g S. ..4...- . . . ? s.
gy y ...w . g,,w.. ., .a e 5 ' . ' ' . .' *
.# .*. a *. * *. E' .' $' , #. '. #.*, ~ . . ^ '. a. p"".*..#^""a.#.
w f '. '.". A. .# a . a . . . . . .
3 ..d.
2
- 4. .s g
- Q3Cyv CA .'.* an A,. *Q g, o.e.
f
( 3 9"1 ")
_ % aV e n
s-ye en. 4 .9 4. .e a. .' g. . s %..e,e4n
. %..E.. . .X. w a.r. g*
.. .e.
a.w .. mu w4 'n 4..m..P o . .,. .w.4 m ,. in SCI'1p . .o.SeS. *:%
.w. ,. . .n.4a . a,. a s ewn.. , .4. a.
ok o..a..
- 9 m.e .b.o.
.e,.
. g a. n uan ac./ _,ge.,. ..: + o A.e 3 y.ev .egw,w.* . '. .-. .e.g, w.C . ... L. .
J4* , + e ek a s e . 3 a-o. A, 3. 4s . ., w..,. C.,. o.
fs .ae e . . d o. w..., e .,. , 9 c O)
. , ' 4.. , .C e, ... . . ., . a. ..s. a.
.wCa.
a a.-.o . %.. s Uai. a. S .. .
- ". *al# #. C"n# & ."ef* a.d- a #.* a. w.
a m.' a. .".w y..m ar.'. .' .". .J a. ". a. . o.
. . # . n wa .# ."..a.
'Crs:
,,. <e.,'.)
.wg ,I .
- -O - s.M t
.gw e,.. '{ .. .
4
.3 6.% e 3,M e - .*
- s. ......g.
. w a , a'4 *. *: .
~4. '. ak-N *'. .# '. '.' A *. S .#.". * *
.Ca-'
"a".e*
. . e
..,.e..d
.. . 3 .'4, and a a ,. d *3 1. . . o n ,.,4 S .,. 2 . . ....a. . w. a=. .. ' , . . ,., .e .w.
. . . . m. ,., . . n.
- .. . 4. w .. 4. o-O I
....,. . S . aa.'* #..' *j* .d.a # ' c '* * *.. .a
. ( '. .". a. '
C a S *. E.".*. 1, .# a..". e
. ".d. e... ' . " . . "_ e
- a.* '. a. ' #.
. . " a. Ca5"- .' ~. ".d. a.
.... f- ea . .gw .s ,,. '.q .w
...e o. g. % g.
.af.o. g 2..
-. g .o. ., . r
. e s e .= ,
4.
- 4. . . . 2
. . 3. A. ' .
. . . . a..s ..
4a. "1C . .. .0 - - ,/ . ..3.,. .. . .O .,ee
.a .w.
"L, a 3. ..< 4.. .. .e .."*2 .
.wd
- y ,
- o- . %.)f IS O'
.o g g.. 4 . ~.a.a. -.u14 g .- a...s.
.g.u.e. a. , .. n..., e '.( ,
4
.3 ., s. - .
p ..o we. a. .. s..y a. . :
... .ga.... . - . .
.s . a . .
-wa- **# '
- . w
.'*.-.#."."."".S, : " P-.".2 .' *.
. .. . s 3. ,g, . . . :. . . ". . ".a */ '. e ** *. .' e .'* #. a. .d. -o" 2
.... , ".*.A. .' a'
.,w, . a.- . , 4..J . .4.3
. . . . . . aw
. 3g .. .s..,w...
3 .w=..
. . . . , . .. * .o..,.4,
.g..... .. . 4. , * ..2*. 4 4,
. = . w.
. 2. .
- s. . . g....'n..
... .f .g 3.
- g w m *I a.g 3. ..
J '.g
, . ..bg . 3p
.gi. J . * .
..a.
y
- s. , . *. g *. s.. ]'g *. s.e .
9.
... . ..%.a. .4..3.G.
..s
.e h., .,,,=.4o..*.*j.
ef y *-.3 3.
. ...gg ..
.w ...
,* * * - * *
- 4 *b..c.e..==a.*,**.*3.* J * *
. . # .%. w
. . .. . . . o. . .*
.C..,., ,,
.4..,.a.
.. .i*4 .a.. 3w . . . . b. c O . 3 .'a.**. .=. .' #..=. c. o **
..*****...s;..d**..#.*.*.*.J.
- a., **..4 4* * * * *
- a. 1 . . . . - .. * * ' . * . * . *. A. ** e y* .3.==
. .. 4 a. A. 2 .*
a...*: .#. . 3... . e. .e.
4 5-log N c
linear 'requency-nap.itude lav
\
s V8 9.
-,4- .,A
- w. .
M,m ._ aX 9 *
.re. . ,e '.. C ,
.p4*f. ?.4 . 8 ...... .. L., .p., t,..
. A. v-
?!L"~ hquake Cap.. Tace , J S i n g O u'2'.L a*. _a7e e..
..3 .e"C. ,7- Za '"..
e .' '. "ad e s' *. * *. #. a* *. ' ' S .
. 27 13-2-12586 1000- _
l_
r i
l 10 0 '--
m -
O L o I N L e :
C i_
Q) '
> { /
~
o i 1 0 t--
o .'
~
l Z '
r
- w - .
I
@ l_ LOG jg N = 7.66 -0.9 3 M s N.
W L Cr \. I i
u_ !
W 1.0 - \.\*
l 2 E x*
s :-
< r \.
J -
D 2
_ \.\ !
i ,
D o !
r \ ;
1 h l 0.1 -
\
_ k r
I
\
\
l 1 0.01: -
6.0 7.0 8.0 9.0 MAGNITUDE (M3 )
n s. :: : c a : = :-= e-:: e s e i = i :.= e r ;= .
28
.,g. t
.~ - s e. .-, , . :...n_a- a
.a., .-, ... . ,. . .<.., a. . . a. _~ . .>.. . , , o 3 >. . .. e ., . . . . . < , .
- a. s , . n. ..
.~,a
... . ya..,/
.o ..e
. . . . .,sa..s
. o.
..a.
. s .e ., , _, ._. e .a.s.., . = s ,. s n.a 2. . . e =. . .,.. .. s. . . s
. a. .,o--
.i.
In an ideal world, the presence of an apper tcund to earthquake magnitude vill reveal itself by a departure frce linearity at the upper end. Figr e 2 shows an idealised representation of this ncn-linearity.
Unfortunately, there e.re two Other effects that :an also lead ;c a curve similar to Figure 2. First, any seasre of sagnitude based on a limited s*eectral band has a built-in saturation .erot.erty. "'his is discussed in the nex. section. And second, seismic instr'ments frequently have a limited dynamic ran6e, and the sagnificatien is Often se: to reccrd medium sized es-thquakes. In this :aac, isrge earthquekes vill cause the ins : ment to go Off-scale, and a measure Of =agn' ude is impossible.
As a result, there say te a purely instrumental upper-tcuni to sessreable sagnitude for a given instrument. he effect of this :n ne:verk iete~nina-tiens f event =agnitude 's discussed in later sections.
2.3 Saturation of the Magniturie Scale Several authors (hinnery and North, 1975; Kanescri and Andersen, 96p, . ... . . s, .av e
.c e n... , y ,.o .4 .. ed ,. . .. . .,., a.. e s a.,. , . , .. w. ..e
. . s. e c . . . o .. . . . . .
spectrum Of the radiaticn emit ed by an ee.rthquake seu-:e, any =easrenen e . ~.. ~ s _ . r . ^ '. .' '~. =. -s" a. .. s's-.-"s..~a~.'.~.=..*..
~.*-a3'...'."*.
'. a .= e .* . n * ' L1" .~. e ' 2,
?:r ex:spie, M is usua2y seasr ed at about 20 seccnds peri:d. *'* hen 3
a.,,.
v ...3
,a ,s.. ..C,,e... . a.
, , . , ..u.
., a.o
. . . . . ..,s., .. . . .
o.. . . . , .
....g..
..-% a.e.
. .' S e . .. .n.
. .4. s , . %. e 3 3m, .'.& . . ..
. . A. a. m
. .e' .'
a *)t.% .e .p. ... .e.
, ,.a 944 4-
- 3. .... e. ..I..*'
. e.
..v.
. .a 5-
.J.%
J J , a,q--*--"---*y s
- J ,a
9
.by...b
---*t**
i.* J,.
- * * * - - - a.J.
'4.a
- - - - .g: *-- *
.g a. .p ,.*.a . . .* . w e .
. . ... a a.. e .s . , ......3
. 4. 4 3 . . 3 s. a. A. .; .
. -%4 .a.,.,
. , .,,4
.%t ..w.
. . ,, - :.1 7 4. s _.. .. ;
. a
.w . .g3 .s.
. . . ,.-..3.4...
- -...*. e........
...s..;
, 4 . . 4 ,, . . . . . . .
s c
e *
.g y. 7..... .a . .' s *, e d ' ... . ". .e .* .* .*. s s '. .' '
- s. .
s- . *.* 4 y . .# 'Ju'. e .'.' b. a.. ** 3
- O.#...'..*.
... .. f .' ,0 " b ' .
, s
.. 3 ::.
. , 3,.. .. .... . . . .... . .4 3.,2 . ..1 3 '." .4.s a.-a. ".a.-~.,
. . . .' . s e . o : .- ". . e. .. . .da.y .V..
a values '1'vernden,1970) .
~~his dief: rem has Often teen .: sed as a basis f r discussing the existence Of an upper bcund to earthquake sagnitude 'see, f0r example, Ecusner, 19~C). *t is, hcVever, ;cssible to interpre: ,his c'rve in another way. .?icre h shows a ::cpila 'On of recent data relatins .
surface wave magnitude M to the seissi: =c::'ent M . ~he highes: vc s o points correspcnd to the 1960 hile and 196h Alaska earthque.kes. 3cch have been extensively studied and seem reasonably reliable. Se observa-tienal data alcarly indicate a saturation of the M scale which seems to a
begin at about M3=7 5, and be :cmplete at about M =3.5 2e sclid line S
in Figure a is a rough form Of the M -M relatica.
s o At this point we can legitimately ask if the fall-eff in Firre 3 can te wholly attributed to this sa ' ration. *4e :an say this such: if
.w '.a . a .'.. . - a. 1 a .e .- a.'s
. . . a. ?'. 3." . .
.' a . e >. ' .. .o a '- a mg-
- e". i ..c ~.e.. . 3.".p k. , . b.e
.3.,*.-. 4. a. 7. ,.i. *J.,egr (s.e
.. . . . . . 7.4... 3-.
. -i e
v.'ane
. w.. 3 g 3 A ..a.Cn. .
s f y~ ~ c, '/ % 8.7. .s ..A .s&. *w. *,.......,,..
. e...
3.4 . . .. ..g . .s ww p.-y n aou,.
,,. s, . -4 w..
.4 .a .. ,.I.4.w ... . a ..,.,
.3 .*
.e, . , e.. ms. O' i , 4 .e ,- .- .. ....g..w,,,.a/.
. s w..a./,.
- ..".e s c. e. s--..- a. s s .4.. p. '.', ,'.a..-..'^... s a. em s .- . a s -..a' . .' a. .- ,.es.
.a. . "..a
.w,.
.- ~. w~~. 4% a.
.' s ..'.a. '. .'. .3 e , .* . d. -- -d
. ' . . . = . .. .". e ' Ju . e..*w *. . g-
.3 ww 2.'. .- 2'..."..=..- . . .*'%.'.
w
(/ .; 4 3~. . . ,' ,, .
. .e .Xp.a.4.eA
. 3. a.
. 3 a .. . . 3. . e . .
. . . . .s - .w ...e .. y .
s a.. ...
3
-'w,,..
. . . . .. . ...o
.. .4.y c . .a..,. . . .C .e
..o
.3.. 3 e.s.. e.
. 3 ..4
. .. . a
- a. ..;/. .r e. S .. .
. 3 .
$. .a.
.. .o.... . w. . .
3 - .
.O ... 43 .- . . . . .d
. .. d. .J a. .* ., .. .#.-, a
. . .y y .
. -G ,...4 . w-
- * *^~
3 ...# ..: -=..# . . . . * *.** . . . . *,
- k..w. .*.5s . *b -
4..
'.I.w . '. s = = f .
be2 3.. g- ..a. .J. . - ., . .s .. .*.=. e , * ...
'*2* **^* *
,.....4-3
. w a. ... .. 3. c,
- ; .. %. g . . g . s.
. . . 3. .. .q.. .J .. #. .3 .J .,. .* ). . .. a. ...
%. ,.3 %. ..; g.. %.c .... g ...
. . . . .a , ..
.3 J.
- v . . .. "' J. g . = a. ,E .
30 18 2 2585 31 1 10
- CHINNERY AND NORTH 1975 o CHEN AND MOLNAR 1977 10~O -
29 .
10 --
. o E o u 28 e. o NC 10 - *
- o x
3 8 o H .o z 27
- w 10 - o 2 .
o 2 . l 26 3 10 -
.. I
- 1. .
l l . o
- an.
25l !
10 ;7 l,,a .
l 8
. *.. l j
I 24 * ' '
10 .
i
_J 5.0 6.0 7.0 8.0 9.0 MAGNITUDE (M3 )
Fig. h: Ocmplilat'.:n of 37 published esti =.tes :f seissi: c re..
ss s .^metica .'f surfice va ce e.g.itude 'i s.
n 18-2-12587 1000" - :
r I, r-r 100 _-
n b
, I,
- b. ; i c3 i e L i X
s -
x C
- e . .
e 10 -
>- ,C U ,
z - =
w '
n D I -
o W l
~
g * .
u.
w 1.G ;-
- L
& r-
< r ,
=j [ LOG 10
- 2 N = 17.47 -0.61 LOG 10 'Mo l l 1 a _
U 0.1 -:- ,
t u
i.
i I
i O.01' ' '
24 25 26 27 28 29 30 31 LOG 10 IM **00 n.: s 7:equen:7 - :::en- s:src ::ns: u ei :' : 7-ree s 3 ed -
a2 .
Seccnd, the i=per:ance cf =agnitude saturatica is demenstrated.
fee.2 v.. . .~.me
. .,3 . .c an.s , e,m'
- $ a.' . a . a.'. e .e. .. s. .' .e
. ~ '."e _' '.'. . , - . a __' a. , v a. ~s -
.e ex*, e *. .= a*. ~ a'. .' . r. '.c .a c - -" a *. ' o v e. _.a e.
- . '.*." da. s . ~5'-
. . .". _' _' . .' a. _a . _' .v =a >. a. .
the problem Of trying to estimate regional variations in maximum earth-qua'<es very diffi: ult.
2.h 'fhe ISC "a alog An incremental f equency magnitude pict Of data in -he :SC :a alog for the period 1966-70 is shown in the left"a-d portion of Figure 5.
Al* hough !5C data are available for a longer period, we have chosen to limit ourselves to this 5-year span in order, as ve shall see, to :0 spare the overall catalc6 vith certain special stations that vere :nly Operating during this ,ime.
The resulting plot is typical of all frequency-=b i"t" #=#r*=t17 available (e.g. 3 razee and Stover, 1969, 3ra:ee, 1969). "here is no clear linear portion to the graph, and this has led seme authers to propose a acn-linee. relation (e.g. Shlien and Toksos,197^; Aer: and Cornell, 1973; Stevart, 197h). It is therefore very diffi:u'_t to determine a unique b-value, though typice.1 attempts to dc this lead c high values c.,.ip . .o _.,: . ., c .,. . rssee w-
..eJ.a. 61 , A*, _'. v.ae..' ." A.e . . . a ./- a. v a. .. . a- =_.-.
c*. . a. ,o r .ed_ , =.d ."e f'o*s cu: ta. s .d.e
. ~..v.a.. _d s . .%' - ~* " e k .' 3" a... A. , .'.a....-"_a.-
4.-.e. . .. . . 3 .c 3, .w -
. . a. 3.a,h *fcea
_ , s .o a'.ea. -
, e_ , _=_-d a._-d _ea.- '_.,=#.., ~ .. ~
a 9 e ,
o.3. .to .7.
2,
. . . ...an ..,, a,,ea -., ." e
.-a.a_
s '.."
.'..s ."' a .'~.a.
fe .. 4,d.
.. 3 e.mc>. . . n .e
. _asa. . o
- -_. . .< .4 .e . w ,. s e , . . 3..
s..a.._ .. . e .._a...a 4 .: ,.
_3.._ e.
. .f .ggf. . w. g .e.....
.w. ..g.4.
.a . . . . . a. e.4... e ..a
. . . . .r . .w, .. - . g _. . 4 .
- 4. a.
1, . . . . . .
j 3 s.
. m.g
. o. ,a .= . g ,._8., e a. J.' a. y4 4 ge.g 4.3 .m
.. =.%..,. *2
_.v ,
mp..
j
.. =. %..g3 . .. s. .ss. .
w.a.3
- e a.n x _3,4 ... a. =..-
% _4 a . s, , e a. ...J m , .j a. se*a. a.a ..'...a
= .. b'
. m . a.4 _3 .t .. . =. a=. e.. . . .' .* E. .. . b.
. . - . ..e; ,. . _.4.....-;..,,.,- . ..,....s.,.,..
...a,... . y ,r,
. .' -., ,.4
< ..w. .....C . . .,..._3.,.--;
e, 13
'. *.. e . = C . ***
.. a- . s ,, '. a s ' ~~ ^ ..
.* s e ' a.- *. .' . a . *. a '. ' . ' * * . . ' . = .. '. v..a..,,...'."..*=..=.,'a.-=. .
..CCm"u*.eA f a. -a . b..e .
a.v a. . a e- a w .# '. '. c s e ." ^. ,C . . e A '.*y .k a. .Os U .- . .'*. .' .'o" s , 5 .. C' '.
. ,qu.4 .se..*.
. . '.h. a*. a . .' a. a s *,, .1 o .' *. he s '. a *. .* .. - s ~." - *. "a . a. . =. , c .* . e d
. ."..e =.v=....
vas superimposed. The resulting frequency-sagnitude graph is shevn in
.u
.ue <3. ,q.s..a..g
. ,.c.n. s. n c e. va.n.v
. . . 6 ( . .y. e .,,o., .s .a .. c <.... .,. a. \ . ,
n s e n c n.a. a. . . .a a ...
vas formed by applying the station segnitude tisses determined by :lerth
(.,.97e) to .he 6
o,n. s.a.,. . .n ..e.v,,e .x. .w_e...s. ..a a . ou.wvn
.. . ai .s
. .en
. ,4...
The 28 station ne:vork shows very sisi'ar characteristics to the ca*alog as a vk..o' . e. .n , art .# cular , ". ..e ,a. "..=.- a..' -"- a ,, ur ..,f ..%.. 3. -,.- w and the fall-of" at high magnitudes a e preservei. -'his is convenien.
s.<,c.. ..<+_.. ,,n.es .,a, a .o s...
.u
.me o 3 c.a.<
.. ..w n _....s .. .. ,..< .a.,c.ea.2
_e. n. .o .n e -,..
- stalog.
There are reascns to suspect that biases nay be in*r: duce. d in 0 the e*.vark =aEn.# ."da. s '-*./ .".A -
. e s s ,~., .# -
c."*. e d 3 ,, a '. ^.
a v=.. - ae .' ..3 .%. .a. -. = .
. ,a3m. .a .a. .u . g . ~.vh t s f o.n n.,y . g 2 y a. ., ., ye a <gcn.gse.a .> . c , . a...a _- <.
~
.s... .
s e c *. ..* ^ .. s . .' . 'a.-i s .- =. ,~ c . . . .. s t.' gg a. s *,, a , "...va.v a. . , . w.. a * .' . ,., . .a-f..~.^..'..~.".'_'=.
.' ~.,c k .' .~. g a . . u- %. .'.=.'g.'e.~.~.~-3. / .<. .ar '- w . e . .' a '. .' ,'. s
. . . '..e -
=. v a. ..'. a ~. =. ,~ c . . e ^. ~-. "g o
4nd.4 ./.8 A... . a.M ..a.4..Ons.
a.
74.g.. .
. t gh..eya. .-
- g. 3 ..a . o .c s .wg
... ..f,...,.a
. . , C *'" * #- -/
'#=v** Y'
'"-d
-. ;f* (C',,
i r./ .
a 4,.,.
. .w. me . .. e., 2.,.
. .. , . .t,,u. . s. .e
.o,3 A,/ . -
.,...a. . 'A 4.s .s..e e ,,y....
. * : . . C' . .
,e v.
5 .
. . ., 4
.w - r. ...
.J.n, . G9 4 ,
.a .ew. s .
.w
.. . .a. 4-.,s
+ a . ,,,.i.C,,
v "...' ~.b e. =. .' . . ^. e , e ..w' =. .. . ' -
.. a c ur ~. *. .' . c s . . .- . .
. ..'es~=..~~..~.=.-.=..=.....-
s.- e
..a.4
- a. ..
-f .w. e .3...-:..,.J, .. . . .e 2
a.: . a.a.2.. 3..- ~;....4 J 44
.a...
.,..,..e 4 . -
.w, . ... .- .. ,.y e
%. 4 e .w.. .... 6. a .. ..S. .f....w v . . . . . 44..&
- a. ..
. . . ...g ..
.. 4..,
-4 4 .w ..* '*
-w...
A. . . ..w. ,. ..S .w. ..a..a. .=.4. . .. . T J. 5 ..
. a w ga . . ......a
. . . . . : ... . 2. .. 4a. . 44 .....
.c..... .
- 3. .e
.3 24 J e
.:wy, *;..g ...s,, *
- 3. . 3. 4..a ..2 . . .O .
. . . . . , .. . ~ . , 3.,..
. . ..w
.. 7 2. . . ,. . ..
3h TA3I.E 1: 28 5 32:0N : C 4CEK STATICN CODE LCCATICN ,
3IAS (Nor-h, 1977)
ALQ Albuquerque, N.M. -0.20 3HA Broken Hill, Zambia -0.28 3MO Blue-Mtos., Oresen -0.29 BNS Bensberg, Germany +0.20 BUL Bulavayo, Rhodesia -0.C7 CAN Canberra, Australia -0.02 CLK Chileka, Malawi -0.27 COL College, Alaska +0.01 COP Copenha6en, Denmark +0.36 ZUR Eureka, Nevada -0.2h KEY ~ ~ Kevo, Finland +0.02 KHC Czechoslovakia +0.10 KJN Kajaani, Finland +0.Ih IJU Ljubljana, Yugoslavia +0.29 MBC M- Mould Bay, Canada +0.Ih MOX .Voxa, Germany +0.02 NCR Nord, Greenland -0.11 NP- Northwest Territories, Canada 0.00 NUR Nurmijarvi, Finland +0.19 PMG Port Moresby, New Guinea +0.10 PR3 ~ Pretoria, South Africa -0.07 FRU Czechoslovakia +0.Ch RES Resolute, Canada +0.13 SJG San Juan, Pue.no Rico +0.2h TFO Tonto Forest, Ari:ena -0.32 IUC Tucson, Ari:ona -0.13 UB0 Uinta 3asin, Utah -0.11 VIN Vindhoek, Scuth Africa -0.C9
22- 3 85 1966-70 3000 ALL EVENTS _ 28 STATION NETWORK
- 3 STATION DETECTION 1000 -
- o
..e o .
SLOPE 1.49 : '
- WITHOUT STATION BIAS
, ; .o o o WITH STATION BIAS IOO '
< E N
SLOPE 1.47 a
10 *
. g
_o .
o
) __..! _L l_ l 1 I I J l 1 Ia- t 35 4.0 4.5 5.0 5.5 6.0 G.5 7.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 rn b '" b V i p, . t.: tereiluency inagn i tiale dut.u for t.he IUC catalog, for ull listed events (left.), arid for a y selec t.ed net. work or L'8 ut.utions ( ri ght. ) . The 28 st.ut, ion net. work is list.ed in Table 1.
4 O
36 m C o o 4 p
- 4
\
n & >
- 4 /
e 4 O h J
- e V m s. z o
O a \o.
e o -
p V
- c &.4
'o M .
?
3 O d f
% % ~,
e c >
e y-O e
. 4 A v
t- . g yi ,.. ,.
- o. -
ld @ 3 W0 e p . - -- ... -
q ,
e--
m j JO $
E _ _ _
g ,: j b Z O-e g W o e
> J -t c.
W D e N O:
d Y g J e --
o '4 =
4 N :3
+
C
.c -
O
- r l x
-* 4
\ T,
%e w e 10
~
e c
Y i
O O
O o
O O
~
O
~
Z
4 m
PMG c22-33m.
1966- 70
)ggg _ . . _ _ _ . . . . _ __
u ALL EVENTS -
30 < A < 90*
', ~
/t'%w -
f* .
/y
/
SLOPE O.97
/
. SLOPE 1.03 10 0 -
t -
j i f
. .I N , , , , ,
\
\ .
10 \" . \
7 .\
o A
\ -
\u
.\ 5
\
~
i
\ * \
l \
-~ ~
05 )0 l5 20 25 3d 3$ 5!U 5!5-~~s.U-~6I5~ 7I~0 LOG A/T STATION m b
Fig. 8: Freiluency niugnit.ude dat.a for Port, Morestiy , New Guinea.
38 .
We have : mpiled similar plots for all of the stations in :he 29 station network. A vide variety of behavier is seen. :f attempts are
=ade to fit the frequency-n plots with a straight line, slopes are found to lie anywhere vithin the range 0 9 to 1.5 ?igures 7 and 3 shov elearly the differences that are observed.
There are two possible interpretations of these data. If the differences in b-value are real, this could indicate an important regional variation in seismicity characteristics (clearly FMG and KEY sample different portions of global seisEicity). The second alternative is that station reporting characteristics vary considerably, and the data are not good enough to define a true t-value.
Perhaps the most surprising result is obtained when frequency-station plots are made for the U.S. 7ZIA observatories. These are 3MO (Blue Mountains, Oregon), UB0 (Uinta 3asin, Utah), TFO (Tonto Forest, Arizona) and WMO (Wichita Mountains, Oklahcma). '"he four plots are superimposedinFigbe9. Zach station has been adjusted heri: ental'y according to the station biases of North (1977} , and s=all vertical adjustments have been sade to improve coincidence, recognizing that there are small differences in the seismicity sanpled by each stati:n.
Again, only events in the distance range 30 to 90 are included.
Snarkably, :hese data are all consistent vith a seismicity :urve that is linear, vith a slope of about 0.7, up to s =5.3, and then the b
curve tends devnvards and approaches the vertical in the range n =~.
to 7.3 This relation, indicated as a sclii line :n ?igure ?, s r*~e-kabl/
sisilar to the 3utenberg-Richter M, :urve ' Figure 3' in shape. E:vever, it iiffers iraca:i: ally from those :tserved by n:rnal stati:ns. 3 :i:e, f:r example, tha; these :tservatories ree:rd nany even s in the range n"'
b # ' 2' #U*r**3
- E#'* *#* 1i8'*d i 'h* AC OSI^125-
- 39
- CCCe '
til 2 :15?5 'l l
l l I
} o c op,Wh
? x' g CcC'- s .
C o
- VELA ARRAYS l
- xa ADJUSTED OR STAT'ON SIAS .
, AND SE!SYtC:TY LEVEL o x* 1966 ~C '970 i_
a i o *
.- x 0
a 4 I 06
'CC 7 l
~
y \ SLCPE O93
\
-o a N
_ N
. eMO
.\ N O
\x ae ao\
o UBO i \
x o x TF0 oe \
- 0 T _
a \
A WMO
- _- aa g \
r =x \
o g
'- ta \
f i I ' l I i i
!$ 40 4' 'O 55 60 SS 'O ?'
m b
3'ig . 9 : h eque".0f -307tiOO % .010t3 f 0 r f 0'O "- . 2-
'/ D Ob3er7130 ries.
ho -
There are a .uaber f i=portant differences between the '7ILA arrays snd the average anal:g seismic statien. The Operaters of the 'TILA arrays were highly trained specialists, who made an unusual attempt to measure magnitudes :arefully and consistently. Mere important, each Of the arrays was equipped vith a low gain channel, which Save the arrays a much larger dynamic range than the aver!Ie statien. These points stre:6ly suggest that the 7EIA data may be more reliable than regular station reports. An additional suggestion that this is the case is cbtained frem the Large Aperture Seismic Array (LASA) in 31111ngs , ..icntana.
Figure 10 shows data frem this aEray for a ecmpletely differen time period (1971). The seismicity curve shown in Figure 9 is an excellent fit to this data set (in'?igure 10 this seismicity curve has teen adjusted s
vertically for a best fit). ~
In order to investigate this problem in =cre detail, it vculd clearly be advantagecuo to limit the 6eographical region within which
. i the events are located. In this case ve may expect a well defined seisnicity curve, and we can test the ability of various network to detect this curve. This is done in the next section.
2.5 Events in the Aleutian-Kuriles Regi:n The analysis of the previous sectice vas repeated for events in the Aleutian-Kurile Island area (defined by icngitudes 135*3 to DO N , and latitudes 30*-90 }. The i=portant seismi:ity Of this area lies vithin the 30 to 90' range :f staticas in both Itrepe and the U.S.
Figure IL shcws the Octal :SC data base f:r this area fer 1966 ' .
The frequency-magnitude ista do not disagree 3 r : gly vith the seismi:i y u-ve sheva., which is tha, shevn in Figure 9 adjusted terti:al k f:r a best fit. Open 21:ser examina,ien, it ranspires -ha: the :atal:s f:r
C22-5623 1000_. _
LASA BULLETIN
, 1971 ASSUMED BIAS = - O.25 100 *-
N -
10 -
1 I i i i i i 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7. 0 MAGNITUDE (mb)
?1e. u - ,2=,.=r-=ae.: =1e 11 a := ., n ;, ;;,- = ,
ge . -_ . - :--27 ; ; 3A' :- v . :.3 . . . .
2., 3 0 i 1 ' i".? i3 I.".9 3,i 30i 0 i!;* 2 =.***f , 3 .". 0 *=** in Ii[ , ?.
h2 022-5621 1000_ -
ALEUTIAN-KURIL EVENTS
- 1966-70 O
ALL ISC O
O 9
O 100 -
~
S
~ #
/
./
N _ ;
4
_G .
9 s,*
6 10 --
G e O
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 MAGNITUDE ( m b )
7ts. u: 7 equenci-=asat=ee is a --
re: s := - .e Aleu-itn ^ d
=-as '.is e1 1.- ..e IC :a al:g, 1966 70.
, h3 this aree. is heavi".y biased by the reperts from the 'Tl :bservs. cries,
, a.- . .' ' ' ar " y- . " . - .' v ... . 4 no > d a,. - =.*. e . v a. .. . a .
'.-u.e
. sa ...a.s
. .. .n n .4 ,*ar4.e.t.A e 2 24... . .'.'), y' .4 n 'n sn s%. ys ~.%.e
. . .ta.. ..
.m . .a twenty-fite statien ne work (this is the sane network as that listed in Table i, with the 'GLA sites EMO, "FC and UEO removed) . As before, three station detection is required before e.n event is included. Ncv
- he
- shape ..'. *.be ".e
- v~,.->.. c ~ve i s . ' ea. .' y va. / 4 4 .'...# a. ." a. .". . '.- a .'..e se..'s ..d.4**./
curve of Figure 3 'n fact, it is very difficult to locate the seismicity curve in any "best fit" position by ver-ical movemen't.
On the other hand, data frcm the ~73".A arrays for this area shev excellent agreesent vith the global seis 1:ity :urve, as shewn in Tigre
- 13. Notice again that the 7ELA arrays record =any avents vi h nagnitudes between o.5 and 7.3, while the 25 station .e,vork shows acte (Figure 12). it is not possible to attribute this effect to ,he secgraphical location of the stations used, since there are o North A=eri:an stati:ns included in the 25 station network.
We can accentuate the problem "u-ther by considering Only stations in Europe. Firre ik shcws the same data f:r a 10 stati:n European
..e *,,vo . >.. , vhi " .' a- .' .' s ,. e d. 4..2 *. ab .' a. .o . ~.e a ' a. d . .* .* n . .' . S--a
.. %. .d a a e a . .'
North (197~) do not change the disagree =en; in shape with the "I'.A statices, but they do reduce many of the ne:verk sagnitudes. 2.is
. St.,.,...a. ..C ,,
. , e 5... .
. . . . /. pc s < ...
. . <. ./. 2.as
, m.,,.
. -- . e.n a.a.,. ..
. . e.
a ......a. :
s
.g .w. e , c . ., .. .,.
..7' .e . ... ,
+ '
. . ,.., A. s .?.S4 ,. . . . p a 0.4 .e.3.4..., 4 .f. ,,,,f.
.~ . .. o. .
. r. s ....
- .%.. e. .w a. 3.=
. a. *. a.. .o '.,y 2.. .w a.,g .agg yJ. a. h . .h..g 4 g.* J.=. .4.g o m.,e. g .n =. 4. A %. # *
. .=. 4 4 * . J....o.s.'
f =. '_. a.
a.g.4aw.
.~
... a .J .
..p
. '. . . .. a. . ye. .w 'g. .g .3..n. a..y g 7 a., ."" J. g , . a.
a.%qg.g . b.. a *
., . %. 4. *
- a =. .J
..a
- a,,
.% .# .s . . . J.. .g .., .*.*. .~ t
- 4
. . 2. .t m. ,,. . ,..%,,./ ..4. . . . .. ,. .. *"".'."4' */. a ..
. . . , . 4. .%. _ g A ' , .g J g.
. .. .'.. .a 4 ., .
J C . 3 a a A .- ,. v . ,. .
'. .r.., ..w- ,,.
..g. . s .,
. . . ..y.c . a. A. ,,g..:..
. * ...w..
4 , .
3 w.a....
.g
a-
- 22-5627 1000
_ ALEUTIAN -KURll EVENTS 1966-70 TWENTY-FIVE STATION e s, NETWORK e .
~
o "
~ - e e
~ e e
N - '.
e 10 -
- e e
- e -
e
_ e eee
] 1 I I I l l 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 (MAGNITUDE (mb )
713 12: Prequen:7-:apit 2:le data .*:r a 2 5 s .e.-i:0 netv 9.
'the sta-ices listed in Table 1, e_ 2.N'C , TF0 and *2O :::1i- ed'.
~
h5 022-562k 1000 - .
- ALEUTI AN -KURIL EVENTS
- o o
, ze 1966-70 OA 3 . UBO o o TFO
.g a A BMO 100 -
0 A 4
-eo N -
A b
O o o 10 _-* O 9 O 0 A 0 .
m A Ao i -
\
, i i i i o o . o a. .
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 MAGNITUDE ( mb) ris. a: vreque :r-=sg.i use ic e :v:= : r a tr:1rs fOr Aleu~.iir.-Tu-il 97?0 3. 2.e 301.1 Ir'~e is :he sa:e as -hat in Fi. r e .0, ti'us 9d
.er;i:111f f0r 1 'Oes~. fi'.
~
a6 C22-5625 1000- _
ALEUTI ANS-KURIL EVENTS 1966-70
- TEN STATION EUROPEAN NET O
- e.
- WITHOUT BIAS O
- O. o WITH BIAS 100 - *
. .c. 0*
3 o.
- O N -
- o. .
10 _-
e o O o
- O o .
Oe l I c. . I l I I n .
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 MAGN ITUDE ( mb)
Fig. '.4: hequency-sagni ude ilta :':: 2
'.C s a , i : .
Zur:;esn .e: vers. 2.e 3:1- i :n s 23ed tre listed in Table 2.
'7 a
e TA3LE 2: 10 S*ATION IGOPEA'I 'E7,'CRX S*_ATION CODE 1CCATICN 3*AS (:forth,1977 }
3NS Bensberg, Germany +0.20 CCP Copenhagen, Denmark +0.36
- CI Kevo, Finland +0.C2
- GC C::echoslovakia +0.10 Li Kajaani, Finland + 0. l'4
~ 'U m Ljubljana, Yu6oslavia +0.29 MOX Moxa, East Ge. any +0.02 JUR Nurmijarvi, Finland +0.19 PSU C::echoslovakia, +0.Ch S?J StuttgaM , Germany +0.29
kB C22-5626 1000 -
ALEUTI AN -KURIL EVENTS
_ KEV
_ 1966-70 100 -
o e
- e e
N -
e 10 .
se
- o
- o
- e e
- os ee
] l m l l l l la 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 MAGNITUDE ( m b I Fig. 15: .Weque..cy-sag.itude ia a f:r eve :3 i.. - he
.Ceatian- hri; area, m.s :eserved at F~e re ,
Finland. "he solid : rve is the sar.e 13 these in Tigres 11-1 .
kg
.. o . ~, ases, 3.. . . .. e . . .%. s.e.._',... ., '
.'.ad_
.. g~ .-. .- a. , c. . m a .s" ' a .
e.a. a. v a.. . . s .
For the reascas discussed in -he next see ,i:n, the latter explanation seems Cost likely.
2.6 n terpretati:n At this point ve are " aced with two possibili-ies. Fither the '.I. S .
VELA arrays (and perhaps *ASA, too) have a poorly :alibrated 1:v sain channel, which leads to the systemati: overestinati:n :f the magnitudes of large events, or the magnitudes of these large events is systematically
~*
underestinated by the sicbal network of analog seismic stations. *e have been unable to find any independent evidence f:r the firs: Of these alternatives, and it must te censidered miikely. 3 is pessible, however, to suggest an explanation 'cr the second of -hese alternatives, based on the dynamic range of typical analcg stations, and the process of averaging which is used to obtain a netverk magnitude.
Any seismi: station can be described by a detecti:n probability c u:-re . 'he general form of this curve, and :he pe.re.=e ers necessar/ to de"ine i , are shcvn in Figure 16. For our presen pu ;cses, since ve are exs:sining an earthquake ca. al:g, ve should regard this as the :u.rve d
e s -.
s . ' b '. ..e, .'.a.e g~.~4. .ab .' .' .# .-] . ' .a*. . . ". .e a- .a . .' . v '. ' ' - a.
. - . c .- ., a . .
2 ' r . ' .. ". ". a. ...'
s.... w.. . _. . ..q,;ag o. .o . .
w
...e anzu..s4s/. ...e
. . . . . s'...g. . .w
...e .
- C',. .. ', . ' . . - '
- a. x '_ , _ a. ,
.w. e a.3 .. 4..
ces .o.. wr ,,,e . a... ,.
a Cr_.4 . ., .2. .
... .. ... y .. .. a3. 4 .... .. . . . .
fe ...,.4, . . , .
~'PLCi3um prOhabili*./ ? vill be less *hM 1. 3.
e ;.'* 0 b a'.' .' .' .' *./* .- '.' ."v a 'S3 * # .# , , . a *. '.c *. .-*. ' ^. v M g .. .' *. '.' .4
- S '.'..a..-a. .'..*.
3 .43.. 3 _- .a. C.. . . e a a. . . . . D . ... '
. 3.
4
- 3. . w a .w
...g e...-
.ag. - . . . . ,s .. ... .. .w.
. . . . .r . .. ..,...
.e4. .,....s e . D w. a
?
.%.. . : Vss .9 4,.....
a
..y,. J 3 .....:...-
...d . . . . . . . . .
w ,s .J.4 a ..
4..'.=. a. 3 3 '. 'f e. .. ..3..'.-=.*.
4 4 =' *.
- a * *
. *.b. . a. c.a 's . . . ..=s
. .a.g.
. .#. . 3/*
- ds...a.
. . .a '3 . #. .a. *-1**.#.4
.w
.. .... .g. .. :...., . . .* .w
...3.,
,a. ;. ,g . . ...a 7a. * -. 4..y...
4 4 ..
- a. ,
. 3 ....
4.e.
. w ,a
.._2
50 13-2 '2521 E-d p_ _ -
R DETECTION SATURATION Q
cr LI PING Q.
z o
') X P
o I j -
I w l l-- l W I I
o i !
~
G Gs STATION m b d
. STATION DETECTION PARAMETERS G
d Yd SPREAD OF DETECTION CURVE Gs 50% SATURATION THRESHOLD 7s SPREAD OF SATURATION CURVE B STATION. MAGNITUDE BIAS P PROBABILITY OF REPORTING R
rir. .6: 7:m : ee re e:-::., 7::t en t- .- := , e for a seis:li s a-.::.
p.
,*.u....
-. . o ,. e. .. g , . . ~.u..
. tg a. g g.a g.a a. .a..n
. . a . ..a., .n n,.
, . .a . ...
. a ~. . .+>,u
. . . . . a o. ..g .-a- o
--d e ~ ., .. u..o gg. .a. . agy. s., e ,. --.. *.* .' *. *.~ c s + ' ~ ,~ c -.~ *. a.. *. =. .* .* =. c ~. a~
- ^* ' ' ~
' . . . =. . e .~ . . .~. s the dynamic range f:r amplitude reperting-Amplitudes are generally =easured with a rule on the seis=cgram, which is traced by a beam of light on photcgraphie paper. ~*he s=alles; amplitude seasurable depends on the line thickness, which is typically abcut 1 :n. One vould expect a=plitudes of a fev milli.e ers o be easily measurable. *dith larger events, hevever, probless arise. Mes; operntors record the amplitude, :ero te peak, Of the first sving of the trace. When this intersects the edge of the paper, ses: 0;erators vili cot report aa amplitude. Also, when the trace a.mplitude iecomes nore than a fev om> the ability of an :nerator to 10cate the t'.: Of the .cesk . .
(or Orcugh) vill depend on the quality of the photographic recordins, which is usually quite variable. And very large events, even if they de not go Off-scale, are usually difficul; to measure.
On purely gecmetrical grounds, :ne vculd expect the dynemi: range of a=plitude reper~ ting to be between 2 and 3 Orders of =agni ude (i.e.
between 2 and 3 bm units). As ve shall see, hevever, 1* se ms o be s.e.4 e. ~. A,d .,.srde.-s O. "
afn.<....a.we >n .sr C..>,4.,
o
,.s
-- -,..,...",,.v...
--- .. ... . .e3 o
g#r gg.1.J . .e.sa.e. g ,she #. '. M ," L'**. 3.# *be.a.*.,e'.'*..'.*.*.'"..".ak..'.'..'~.,v.'~'"*/^,'
d d ,
.# a~
-J
.me ** . *J sg 4.. . a.4 A . _6 'pqe *
..% a.we . . e .w a.p .o. . a. .# .SJ. .... . d. a. *
. .e. . ...J g .e , .% G..J a.
ia.
. . . ..e .%.
e.
- a. G @ .
- h "**% #
4 a. =. b e s '.a*. .s. #'7 . s . a'O . .' .' *. ]* .~/*. "4'. .' .'
a' ' .
- n .* *.
. . 2.%. a" ' .' .* . a. -
)
as.g. .g.: ** d en 3 y ..:s -. .-.eg.4..
4.,,. e
. ._ ., . ,.%.ae 3 . 3.e
... . . . . . . 22. .w.
.. : .. .e..... s..e.a.
.. O
'.a. g . a *. .*&. .** 2 .'* e. 3*A. .' 4.w" 3 * -Q .** e. *. h G *4 1 .# a. 'g * *#
s . .
- .e. *.*6 .k2 ..
3 . *
.u.'3 g* * *.. .E a. ...'
. A. 4
- % g 3 . .3..
J
.4 J
.a. a.......J. .
. g" J *. . ..g e ,a. 3 .g .Wg. . . . . .... . 4 . . , . J. 4 2. .,. 3. 4 J... . %.a. .* .J$.
4 D'.3.....J..
..G.=b . . . , a. .a
- ..% a. .g 4 . ..%. a. 3. 3
. w
. .e *"%
...3. a.2 ..
.......2.. ..a.a .
52 in F16ure 17. 3ec&use of scattering, an event of magnitude s vill lead to a distributien of observed magnitudes at a ne:vork of stati:ns. This distributien is often roughly ncr=al vith standard deviation sicut 0.3 a
b units (7 n Seegern, 1973), and its mean (in the absence of station bias) vill be an estimate of s. However, when the magnitude of the event approaches either the detection threshold or the clipping threshold of the stations, the distribution beccmes skeved.
Effects near the detection threshold have been discussed by Ringdahl (1976) and by Christoffersson el s]. (1975).
Those stations where scatter-ing produces s lov amplitude vill not report, whereas these where a large amplitude occurs will report. This leads to a net ;csitive bias when the station reports are averaged to produce a network magnitude.
Methods can te devised for including the fact that scme stations lid not report an event (the maximum likelibccd method) but these methods are cumberscme, sad require a detailed knowledge of the detecti:n probability curves. It does not appear possible to apply them to a data set such as the ISC catalog.
An equivalent bias arises at the : lipping threshold of staticns, sithough this has not been discussed in the literature. It is, Of course, reversed in sign. When a large event occurs, those ati:ns where scattering produces a large ssplitude vill usually not re;cri, while those stations that receive 5 1:v amplitude vill repers. The result is s negative bias to the netverk nagnitudes reported f:r large events. This negative bias vill be quite substantial, up :: :.? Or i sagnitude init, and sn adequately ac::unt for the differen:e :eiveen the W seisni:ity :urve a.nd the :SC :stal:g sei3=i:ity :u-ve.
em da e
Number of Observations I
i l
Approximately ncr:a1 l 3 - 2.3 ab Units l
I E
I t
m Observed a.. . ..
. . . .,a3...z..,2.
24,3 - ,.. ". h e a . .'a. c .- ' .' 2- a+. . +. . .' g n .- .= '.a .- , =. 4.
g p.34..,4,3.
. .... .g .y.......e o ge..4..4,
~
,.ad
.. .o
. a 4..,st.4.,2.a
... ... n ..,
_m 3.. . .... .2. . a. ..
a .- e..sn.
. s.a.<
. . . . ..a.
. . . z . .as...>.,..<..
u...a .a........
is usuCly appr:ximately ec nal.
5h f
Ve :an illustrate our arrrent by using data frem a single stati:n.
Figure 13 shcws the data for IUR (Eureka, :Tevada) . "he left hand porti:n of this figure shows a :cnventional interpretation of the reporting characteristics of this station. An arbitrary straight line is fitted to the data, and detection and clipping thresholds (indicated by arrows) are determined at g=h.5 and 6.3 respectively. In the right hand portion of the figure, the 7ELA seismicity curve is used (IUR is quite close to the cbservatory UBO). In this interpretation the station fails to report many events for g greater than 5 5 The thresholds are nov4'.3 and 6.1, and "ecmplete" reporting is limited to the range 's.7 to 5 5 A similar interpretation for staton .EV using Figure 15 suggest that this station carries out ":cmplete" reporting over an even smaller range, perhaps as little as 0.3 g units (frca 5.2 to 5 5).
A different representation of the same phencmenen for station ICR is shown in Figure 19 Here, for each interval of 0.1 g units of CB0 reported magnitudes, ve have averaged the difference in reported magnitude between IUR and UB0 for events in the ISC catalog during the pericd 1966-70. The theoretical interpretation of such a data set has been discussed in detail by Chinnery and Lacoss (1976). :f the detection probability curve for IUR vere hori: ental (Figure 16} then this ple should be horizontal :co. The presence of a detection thresheid shevs as prcnounced positive biases as lov sagnitudes. "'here is a hin: Of a flat portica of the curve in the vicinity :f 5.0-3.5, and then the data continue teccming nere negative. "his must be interpreted as being due
,o a : lipping threshcil. In general erss, Firre '.9 is entirely :ensis-
- ent vi h the righ- %-d preferred inte pretati:n f Firre 13.
s c22-SS87 EUR 1966-70 1000: [ :
j eON !
[4 100 _- :
- \
F : %
- . _. .g
.\
~
. .- . .y
\
10 : 's : i-
\ :
\
\ n\
\
~
~
\
\
- A. 7 j..
1 _4 i _i_ u___t _.h -
i , , , i _!1 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 3.5 4.0 4.5 5.0 5.5 6.0 ~ 6.5 7.0 STAT!ON m STATION m b b Vig. til: Two int. :rpretut.lons or t.he report,ing frequency or ut,ut.lon EUlt (Eureka, Nevmlu).
The right hund litterpret.ullon is preferred.
56 C22-5h61 e
^
0.5 - EUR - UBO o 1966-70 co .
D e e W
O O e F__-
e z e (9 e 2 e 1 0.0 -
e ct D -. e e W
ee e e W '
O e 3 ee g e e Z
O e 2 - 0.5 -
e s 1 I I i 1 3.5 4.0 4.5 5.0 5.5 6.0 M AGNITUDE (UBO)
Fig. 19: ?.ach pein: .s the rterage difference ~:e een ;'.e sta. i:n zagnizades : f ?J3 s=d l3C '~ ~ 1' =*ren s Mszet in z'.e ~ic estal:g, ;;;; ed as 1 Pr.cti:n :f .de 13C cap.i;ude .
. s L
2.7 Discussien
".-e. - =. s ul ' .~ ' e s c .- * *. e^- .* *.,cv ". ~.. .~ ~.
. . v'. .' =. .' ~.nv ' .~. w' .' .~.3
~
=. v --
' A = ~. . c =. . '_.c. -.....c.-..- .
m e,m +.a-_ . l.< fp g
- ..e,.o ,na.,n,g s.+a i.ns _as a.
. . .. <.,c .an.
. . --nu,_,,,
f.
> .u..a.. ... u.e magnitudes of larger events published in the ISC Cata'eg are biased lov and unreliable. A corollary to this cenclusion is that it is virtually impossible to study the seismicity characteristics of differen: regicns using this (or similar) estalogs, since each region is "nonitored' by a different set of stations, with different operating and reper;ing character-istics.
The 7ELA arrays appear to be unique in *. heir vide dynasi: range, and, until a global network Of digital stations beccmes available and has accumulated a substantial data set, the 7ElA data is the :nly reliable source of infor=ation on upper bcunds. So far, we have not discovered any evidence for regional variations in seismicity using these arrays.
As an example, Figure 20 shews data for shallev seissicity along the South American subduction zone. The 310bal cu: te (Figure 91 is again an
,x,,._,.,....,,.
If we assu=e that the */?.'A seismicity curve is valid and represents saturation of the m. scale , ve can use similar arg'.:=ents t: these f a
.'". . r. a. . / a.d .7c . . .h. l .' 97>~ ) . o ~. . n s .. ~ . . an ' -:c 3 me - . .- = ' a . '. .' .. .= ".. .' .. .
. s s,.:n_< 3, . e... . ... e... ..,a.2. no..a - ,- . e......,
..... ,.d. A, a.. , ....._ .a3. .... ,2....... .2 3 s
'3. + v,.). <3 .1 o =.s 1 _. :
s , -
n' ,f:%,
( =. g ,. , ~ . ,
.. ,..,z .a , . ,. .
- _,p f e... ,*g ., . ~. g ..- .'. .s
.a.. .
.. ....~.a.
..c . .
Ourve is as shCV~ in Figu"e 21. 3C3e i: D abC u t he 2 ". s t a.'. ~. i .~.
,,,..a..s..,..... ... ,3 .s .. 3, .,
- a. .u..,. .n.,.a....,-...-.n.,.,
,, ...g.
..~
. u ,. . ,
. . . . . .~ .
.~
.a ..
. . . .. . _ a. ..
- 2. ,2. .
i 58 C22-57h3 1000 .
SOUTH AMERICAN EVENTS 1966-70 ohgA& . UBO e o^ o TFO
- ^ O A BMO 100 r o
- 2 . A N ~
O
_ O A g 0 O o
10 -
A
~
o O
A d o 1 i i i i i ,,,,s 3.5 4.0 4.5 5.0 5.5 6.0 6.5 6.0 MAGNITUDE (mb)
Frequen:7 agnitude ia 1 :':r scu-h Ameri:12 e r ::s ::se rei
~
rig, 20-1, 3 *sT A 177173. 2.9 3 0 ' i l O'ir'r? 13 *he
. 3'~^ $3 ~ h i'. in e... . 2 4a.. a4 f.-,'g*'r f ; -. g - '93, -i,,
s e
. /9 e
4 C22-5622 9
Ms ~ 8.6 l 8-M3~ 7. 2 W
O 7 -
- b
\
-[
mb~ 7. 3 i
3
_H_
m b~ 5.8 1
z \,
o6 -
4 2 MS 5
4 -
l i i i i ! 1 !
22 23 24 25 26 27 28 29 30 31 LOG (moment)
Fig. 21. An emp'-*
- _3-nczent relaticeship ::: sis en vi-h -he 7ILA seismici;y :urve ,' Figure 3}. -
ne Ms - CC* 0 r*1lti -
ship frem hinne 7 ar.d North (1975' is shevn for : mpariser
60 .
The interpretation that the curve in the ' TELA reisnicity relati:nship is due entirely to saturation of the "b scale seems reascnable. The shape of the a - cment curve in Figure 21 is similar to that Of the M -
b 3 soment relation, and (at least qualitatively) m appears to saturate at about the expected magnitude. It therefore seems unlikely that any information about upper bound magnitudes :an be obtained fr0m the existing global a b "t"1 E8' 2.3 Conclusions The conclusions of this study are very negative. It does not appear that the test earthquake catalog iata can shed any light :n the problem of the existence or the regional variation of maximum earthquake si:e. This leaves only the much less ecmprehensive catalogs of Gutenberg and Richter (195h) and others, collected before 1960. 'Jhile these Older catalogs are useful for event times and locations, there are greving indications that the assigned magnitudes in these catalegs are unrelie.ble (e.g. Chen and Molnar, 1977). At least part of this unreliability probably arises frem the instrumental problems described above.
01 w:. o.a
.=.-.. r. ,C v .a-
' References ra-ked by an asterisk =. e in:1:ded f:r ccmpleteness, but vere not used during this study.
Many have co; teen translated into Inglish.
Aki, K. , Scaling lav of earthquake source time-function, Geo :hys. J. ,
_31 3-26, 1972.
'Anan'in, I. 7., Assessment of the seismic activity and the naximun possible energy of earthquakes in individual seis=cgeni: :enes in the Caucasus regi:n, in Seis=cgeni Structu es and Seismic Dislocations, 7N!! 3eoficika, Moscow, 1973.
Archambeau, C., Estimation of ncn-hydrostatic stress in the earth by seismic metheds: lithospheric stress levels a'Ong the Pacific and Nazca plate subduction :cnes, manuscript, in press, 1973.
Sath, M., Earthquake energy and magnitude, in Physics and Chemistry Of the Earth, Vol'2:le ~, Pergemen Press ,1966.
3 erg, J. W. , Gaskell, R. , and Rineha:-t, 7. -i, Iarthquake energy release and ,sostasy,
. 3m,.,. seism. e,cc. Am., _?4,
. :- ,e v a ,.,64 y i
- 3cnilla, M. G., and 3uchanan, J. M., Interim reper: :n vorldvide his:Ori:
surface faults. Open file report, NCER, U. S. 3eological Survey, 1970.
2crisov, 3. A., and Reysner, G. I., Seiste-tecteni: pregnosis of the maximum magnitude Of earthquakes in the Ca ;athian Regien, I:vestia, Earth Physics, no. , 21-31, 1976.
'3crisev, 3. A., Reysner, 3. I. and Sholpo, 7. N., On the prepe.rr.:icn and use of geological-secphysical data for the identification of :enes with different . % values in the cuter :ene :f the A' pine f: lied region, Symp. on Search for Iarthquake Fredi:: rs (Abstrac s',
MGGGS, MASF"N, Tashkent, 197h.
3 razee, R. J. , Further re,:crting on the distributi:n f earthquakes with respect to magnitude sb, Zarthqua%e N::es, -C. 51, 1969
_ .,,,,- ~ . , .o. . v . , a.nu^ o,~~'~v=..~, '4 . ,
- e '. .' o- ,. ~" ~ '. ..
.'=_'.'..~sak=.-."..'.
~
respect to magnitude 23, 3u'_1. Seism. Ecc. Am., E , ;^15-1C1", 1,269
. _. ., i., . .I . , .= e < s 3.< ,, ..C ne . . , s , .< .w.w . ., .<.j. .. , , , a. ..,..- . .,
a_.<f .e_ . ..g-._e..
fau t :CCes,v. ,eC:h7s.
. J ,es., y)s, i . .24,,,, . . - -
-;;O.
+3=e , 7. :., Kirill:va, :. 7., ?r.an'in, :. 7., "redenskaya, 3. A .,
Ecysner, 's. 2., and Ihci,:c, 7. 3., A :e=p: :: estina e the naxi: =
= p_ 33 e,
.g
= .p . . . ' -a
'a a. .- . .= , .:. _ _'_2
. ~ _ _ . . . . -'T-.
_.# =.__-_.
no. ia, Science Nauf.a . Press, Mcse:v, 19~1.
s .
e2 .
- Sune, 7. I., and ?clyak va, T. ?., Correlation of the saxt=un possible earthquakes in the Caucasus region and Asia Minor vith seismic activity, in Investigation of Seismic Conditions, "Stiintsa",
Kishinev, 1974
- Sune, 7. I., Turbovich, I. T., 3crisov, 3. A., Oitis, 7. G., Reysner,
- 3. I., and Yurkov, 3. ?., Method of develcpment of a relationship between the earthquake magnitude and the tectenic parameters of a region, Proc. Acad. Sci. USSR, 21h, 197h.
Bune, 7. I., Turbovich, !. T., Borisov, 3. A., Gitis, 7. G., Reysner, G. I. , and Yurkov, 3. 7. , Method of prognosticating the maxtnua magnitude of earthquakes, Izvestia, Earth Physics, no. 10, 31 43, 1975 Caputo, M., A mechanical =odel for the statistics of earthquakes, magnitude, sement and fault distribution, Sull. Seism. Sec. Am., 6_7., 8h9-861, 1977 Chen, W.-P. , and Molnar, ?. , Seismic mcments of =ajor earthquakes and the avera6e rate of slip in Central Asia, J. Gecphys. Res., "2, 29h5-2970, 1977 Chinnery, M. A., Theoretical fault models, Pubs. Ocm. Obs. Ottava, 27, 211-223, 1967.
Chinnery, M. A., and Lacoss, R. T., Magnitude differences between station pairs, in Seismic Discrimination, Semi-Annual Technical Su:mnary, Lincoln Laboratory, M.I.T. , 30 June 1976.
Chinnery, M. A. , and North, R. G. , The frequency Of very large earth-quakes, Science, 190, 1197-1198, 1975 Chinnery, M. A. , and Ecdgers, D. A. , Earthquake statistics in Southern NewEhgland,EarthauakeNotes,hh,89-103, 1973 Christoffersson, L. A. , Lacoss, R. T. , and Chinnery, M. A. , Statistical models for magnitude determinatics, in Seismi: Discrimination, Semi-Annual Technical Summary, Lincoln abora: cry, M.:.T. , 31 Dec ster 1975 Octnell, O. A., Engineerin6 seismic risk analysis, Bull. Seism. Sec. Am.,
j8., 1583-:.6c6, 1968.
2:nnell, O. A., and Merz, 3. A., Seismic risk analysis Of Boston, ASC3 N> cnal Strue- -=' ?ngineering Meeting, incinnati, Ohio, April _,.h.
Cennell, C. A., and Merz, 3. A., Seismic risk analysis cf Bes:On, J. Struct. Civ., ASCI,101, no. ST10, 202 -2Ca , '; 5 _
csentino, ?., ?i:arrs, 7., and lucio, O., Truncated expcnential frequency-=agnitude rell .i:nship in ea thq.af.e statis .i:s , Eu_1.
- e t- .n1. _=c. e. . .x2 . , :, _ O_ _ :_ - :_.2,., _,:v~.
k .
63 Co s e.". * .' ..c , 3. . , a ..d ~ "w '.,s '_ a , 3 . , A 3' e. .a. .- s_' ' e, ,a . ' ~s .. . . ' .'..a .'.- a. g" a. . . . /- .. %a ..' .,", d. a.
,. . ' a ., ' _o .n. .i n *. ".e
... . "../;c +.. b.e s .' s .. .' = ax_' _. .::1 . '
- a.-3 .' , ..a _ ..a z. . '" . ,_4, , W. .
Geofis. /3cce), 29, 1-2, 19~6. _
Oavies , G. F. , and 3 rune, J. N. , Regicnal and gicbal f aul; slip rates frcm seismicity, Nature, 229, '01-107, 1971. _
Orumya, A. 7., and Stepanenko, N. Y., Mao. Of the maxinun ecssible earth- .
quakes of the Vrancea seisnic regica, ::vestia, Earth ?hysics , no.10, 77-78, 1972.
Duda, S. J., Secular seismic energy reles,se in the Circus-Facift: telt, Tectenophysics, 2_, h09 h52, 1965
'D hiblad:e, E. A. , Seismic activity and the T.aximum earthquakes in the Territory of Georgia and its vicinity, in Study of Seis=ic Danger, " Fan", Tashkent, 1971.
Epstein, 3., and icmmitz, C., A model f:r the occurrence Of large earth-quakes, Nature, 211, 95h-956, 1966.
- Esteva, 1., Seisnicity prediction: a 3ayesian approach, Free. -th
'J.C.S.3., i, Santiago, Chile, 1969 Esteva, 1., Seismic risk and seismic design decisions, in Seissic resign for Nuclear ?cuer Plants, M.I.T. Press, 1970.
Evernden, J. F. , Study of regional seismicity and related problems ,
Bull. Seism. Soc. Am., _60, 393-ah6, '.970.
"aa/* s kif , 7. .Y. , ' d '.sa'.,ok. , A. 3. , "..*.e.=.,~e_'_'.a.,,d..
. . . _. #. ." e ."..eo --./ . . '
s extreme values to the proble.as of recurrence Of larga =a -' quakes (in Russian), in Dynamics of the Ea :h's " rust, AN SSSR, Nauka, Moskov, _.n6;. y .
Ge.3 .,and , .. M., .au>erman, e.. A., and ...s e :...,s . __
,cro.. < , .I . -., -attern
... a c ..4.,e,.,n
. _ . , a y .,.,.4,,a
. . o . a. + ..
. . w .q..a.L. .. .4. . . ..,,e,.s . .s....__....4.,,
2 . . . -
- a. =uya.
. .r .a .n 2,an. _ _n+,.., _,_,, ac. _s,,, _-~~6.
. yi
.:orbuncva , _. . <., A na c_, mum-earthquake na; .,:r :.ne .J . erthern , e: .. :..:an, I:vestia, Earth Physics, nc. l'., 3-la, '.969 a,.mee , . u., 2. .. .a. a_ -. 2 . <_ , s o .o N. . . . .n. s ,
- a. ,_..- ,,._> a ~..a... s_:... =.. , u , . :. :. .: .
au.,e me w .3, ., 2.d v... >.. ... , . v... : ..
<.s___o , , , _ . . .e
.. _e
. . . _a.,... , a.-
. . . a.
o.. w..
. . .. c o. .. .. .. , O..
4
. . . .n. . _ .,., g *,,*3 4 7... 3 .J .~ / 0. . . 33, - ,s :4 73%4*., ?., 3...d.. .9,,y 4
. / , v. - . 4,. , w. ?. , ,
r _ . .
.0 4 4. ......s. n. .e. , -o. .'
. .s.2 7. 3.
a...
- J . .:. . , *. . ,, . 2.n. e. .' . g,2_ .s .o- :~ ..
. . . ,.y y , '..f._ta.. .. 7 4. a. _ _A
. . A 'f '.r. _ l. _- ' ,
4.r. a. _4 _*. .._ _ ,
T1
- v. , .? : _ a .-2. g '.._7
- 3. .g s. .... .:. .., c ., . ,-4_a.a_-w
. .3..._..
. .,._c_,
._a_ .., 3' _ .g e 4._4 .c
. . .S. . ,:] , g.,4. ..
,4. 9 .._. _, - ;, .
. 3 . ,C a _ .'; . ,,f.:. , _',. :
Oh .
- Hofmann, 3. 3., State of the - #-- assessing earthquake ha:ards in the United States, Sept. Mc. 3, U. S. Army Ing. Ja:ervay Exp. Stati:n, 7icksburg, Miss., Misc. Paper S-73-1, 197h.
Ecusner, 3. W., resign spectrum, _a Earthquake Engineering, ed. P. 1.
Wiegel, Frentice-Hall, Inc., 1970.
'Kallaur, T. I., Seismic activity and the energy of maximum earthquakes in scme regions of Tarkmenia, in Study of Seismic Cancer, " Fan",
Tashkent, 1971.
Kanameri, H., and Anderson, D. L., Theoretical tasis for seme empirical relations in seismology, 3u11. Seism. Sec. Am., 11, 1373-1c95, 1975 Kanamori, H. , and Cipar, J. J. , ?ccal precess of the great Chilean Iarthquake ,
May 22, 1960, Phys. Earth Plan. Int., 2, 128-136, 197h.
Kanamori, H., and Jennings, P. C., Determination of local magnitude ML from strong-motion acceleregrams, Bull. Seism. Ecc. Am., 58,, k"l h85, 1978.
Karnick, 7. , and Hubnerova, 2. , The probability of ec:urrence of largest earthquakes in the European area, Nre and A;pl. 3eophys., 3 ,
61-73, 1968.
Knopcff, L. , and Kagan, 'f. , Analysis of the theory of extrenes as applied to earthquake problems., J. Geophys. Res., 12,, 56h7-5657, 1977 Kcgan, L. A., and Shakirova, T. D., Assessment of Kmax by means f Gumbel's first marimum distribution, in Problems in the Assessmen of the Seismic Danger, ',3auka ,n Moscow, -197k.
- 1ambert, 3. G., Tolstoy, A. I., and Becker, E. S., 7ela Ne verk ?valuati:n and Autcoatic Processing Research, Techni:a1 Report No. ~, Texas Inst.~.:nents Inc. , 9 December 197a.
lee, ~4. N. K. , and 3rillinger, 3. R. , A preliminary analysis of the Thinese carthquake history, paper presented at the U.S. 3eol:gical Survey Conference on Seismi: Oaps, 3cston, May 1973.
Ic= nit:, C., Statistical predi: tion of earthquakes, Rev. hophys., i, 27~-
39h, 1966.
McGarr, A., Upper limit to earthquake size, Nature, 262, 27 -379, 19'd.
McGuire, R. K. , Methcdel:gy f:r incorpcrating para e er incertainties into seismic ha:ard analysis for 1:v risk design intensities, presented a: Int. Symp. :n Iarthq. Struct. Eng., 5 . ;cuis, August
_y,...
Mer:, H. A. , and Connell, . A. , Seis=i: risk anal / sis tased :n a quaira-i:
crgnitude-frequency '_av, 3u11. Seira. Ecc. 22., il, 1999- 2006, 1.:~ 2
~
. 4-O 1
Milne,~4. G., and Cavenport, A. 3., Earthquake probability, Pfos. Icm.
Obs. ttava, Seism. S eri e_ s_, 1963-~, 19pp., 1968.
Neunhofer, H. , Non-linear energy frequency curves in statistics Of earthquakes, Pageoph, 2, 76-33,1969 _
Newmark, 'i. M. , and F.osenblueth, 3. , Fundementals of Ia-thcuake 7r.gineerine, Frentice-Hall Inc., 1971.
Nordquist, J. M. , Theory of largest values applied to earthquake magnitudes, h ans. Am. Geophys. 'Jnion, 26, 29-31, 19h5 North, 3. 3., Station Magnitude 31as, !:s Dete-mination, Causes and Iffe :s, Lincoln laboratory, M.I.T., Technical Note 1977-24, 19~7.
Otsuka, M., Cut-off of seismic ener3,v, J. Phys. Iarth, -
2_1_,
119-123, 1973 Papa:achos, 3. C. , Cependence of the s::is..ic taraneter b :n the magnitude . _
range, Pageoph, 112, 1059-lC65, 197h.
?'ei-shan, ". , and Fang-Lui, L. , An applicatica of statistical theory Of extreme values to mcderate and long-interval earthquake .:redi::icn, Acta Geophys. Sinica, 6,, 6-2h,1973 (Plenum oublishing 20rp.
translation, 1975).
- ?erkins, D. , '"he search for maximum ragnitude, NCAA Earthq. Inf. 3u' ' . ,
18-23, July 1972.
- ?utrushevskiy, 3. A., On the relationship between the earthquake Of maximum a ,s intensity and the geci gical state, Bull. 00uncil Seism.,
no. w, yo0.
Richter, C. ?. , Elementar-f Zeismclegy, ~4. H. ? eeman and Oc=pany,1953, a..vx .4..c.x,.,
. ,.., x.a.2. . s . 4. ., s o , .,., , . 4 _. a_.
. ,_, ...,.a .4 , .-
. .u..e..,,.....
. ... , 2 pro'cability of earthquale Occur ence, -'ectenophysics , 26,1-21,
.n,g7(s.
. ..gda.', ?., .V.axim'.:n ' _' '.'.e.' ' "c c d
. _ a. s . .' ma*. ' *. .a.. ^ #. .-a.'.".4.-
. =s '.."=.,
- _2_" ' _' .
Seism. Sce. Am., _66, 789-SC2, 1976.
2,_a .p. < +,
. . . .4 e .. r.
. , .i. f , , =c s .s " b .4. .'. .' . '. a. s .' .' ~. ~c ' .~ " ' a . ' .. e~. ~ > x ' ~ ~~ =. c..~ ~. ". .v.. - " .* >. =. = ,
'rans. (Sudy) -nst. thys. Trr.h, Acad. Sci.
- S S S. , M,, 192, 1962. .
2.4 .,.n.4 ., u..e - ,r .,.. f., .o.q.* a. . 4. , 3 w 24 . ...
. . . .j a. . . . k. g . . e =.
7 , ..
e
. w. .a =-..-..a..-..
.e u.,s, ._. a r a. 3 ., g s.
- a. g . .. gt. a.3_4. ,
a....a..y,
. 'c .s.1 2 2..
y .Ys
- - a
- A . .qa.. . .r. . ~ >:: . .- :, ,
-:g,,: -::a, -scaa.
2.4 .2
. . u..a. ~ ,. , .r . f.,me 4.,.3.a g.a ... ..). .g
. .. ~. .....y
..3,,
>=
._z.o ,,.....,y, s.,:
..s..... ....
.2.,, ,o a ,. 4 . , :c .., ,,.:. . . .. .... , ~.,.,a..
.u
. . a c. 2. a.
..~.
C ...-:
- ,, a s,..,.. .::::
- -:::, .:C4 3.
s c6 '
s
'Ri:nichenko, T. 7., Seismic activity and the energy of the largest earthquakes, in Problems Of the 3eophysics of Scviet Central Asia Kasakhstan, Science 'Nauka) Press, Mcscow, 1967.
"Ri:nichenko , Y. 7. , The generalized lav of earthquake Occurrence ,
Boll. di Geofis. Theer. ed. Applic., 12,, no. i8, 1970.
'Ri:nichenko, Y. V., The strengest ;cssible earthquakes, Zarth and Universe (2emlya i 7esiennaya), no. 5, 1971.
- Ri:nichenko, Y. V., Determination of seismic danger, in Preblems in the Quantitative Assesr=ent of Seismi: Canger, "Nauka", Mcscow, 1971 Ri:nichenko, Y. 7., and 3agdesarova, A. M., The strangest ;cssible earth-quakes of Japan, I:vestia, Earth Physics, no. 11, ik-32, 1975
'R1:nichenko, Y. Z., Drunya, A. 7., Stepanenko, N. Y., Crustal seismic activity and the maxLnum ;ossible earthquakes in the Carpathian-3alkan Region, in Regional Studies en Seismic Regime, Shtinitsa, Kishinev, 197h.
Ri:nichenko, Y. 7. , and 0:hiblad e, 3. A. , Oeterminatien Of the maxisus possible earthquakes on the basis of the ecmprehansion data of the Caucasus Region, I:vestia, Earth Physics, no. 5, 6h-85, 197h.
Ri:nichenko, Y. V. , and Zakhareva, A. I. , Generalized lav Of earthquake occurrence, Izvestia, Zarth Physics, no. 3, 29-38, 1971.
Rothe, J. P. , The Seismicity of the Earth 1953-196 , 7NESCO ?ublication, 1969 Shakal, A. F., and Tekso:, M. N., Earthquake hazard in New Ingland, Science, 195, 171 '73, 1977 -
Shehalia, N. 7. , The sansum sagnitude and saximu= scale intensity :f an earthquake, I vestia, Earth Physics, no. 6, 12-20, 1970.
Shebalin, N. 7. , Esti=atica of the size and ;csition Of the focus of the Tashkent earthquake from sacr seirn#- ' ' ' strunental data ,
in The Tashkent Earthc_uake Of 1966, Uzbek Srinch, Acad. Sci.
USSR Press ,1971.
Shetalin, N. W., Assessnent of the maximen seismi: danger in the ' '-=a-Tamansk region, in Seismicity, Seismic Tanger in the ' '-as and Seismestability of 5tructures , "Naukovsys :u:ka", Kiev, 1,0 2 .
5henk:va, 2. , and Karnik, 7. , The pr:hability ' -^^"---*c e :f large s; earthquakes in the Iuropean a ea - Fir ::, Fure and A;:1. 3e :hys ,
a
_o_, -
- , 3 -e _-, ,en.
. , -., -- ., , g 7.L- .. a w
-, ,., .smy o a --- - ..> - ,.....s.-
> a,,
a<--
e the larges possible earthquakes, ::vestia, Isrch Physi:3, nc. 11 ,
.: -- e,,
..... :~3
. O ,g s ..
Shiien, S., and Toksos, M. N., Frequency-nagnitude statistics of earth-quake occurrence, Earthquake Nctes, L' , 5-13, 19 0.
Smith, S. '4. , retermination of maximum earthquake magnitude, Oeophys.
Res. letters, 3, 351-35h, '976.
Stewart, I. C. ?. , On the use Of the maximum likelihecd estimator for recurrence curves, Earthcuake Notes, h5, 17-22, 1975 __
- Tsuboi, 0. , ,- Isostasy and maximum earthquake energy, ?r:c. I=o. Acad.
varan, 32,, 19ho.
Tsuboi, C., Earthquake energy, earthquake volume, aftershock area and strength of the earth's crust, J. Phys. Zarth, d, 63-66, 1956.
'Jspenskaya, T. A. , Experience obtained in calculating the map of mny d mum possible earthquakes in the Pribaikal region, in Study Of Seismic Daneer , " Fan", Tashkent , 1971.
Veneziano, D., ?robabilistic and Statistical Medels for Seis=ie Risk Analysis, M.I.T. Dept. Of Civil Eng. , Publicati:n R~5-3h ,1973 Von Seggern, J. , Joint magnitude determination and analysis of variance for explosion magnitude estimates, 3u11. Seism. Soc. Am., $1, S27-ah5, 1973 Yegulalp, 7. M., and Kuo, J. ?., Statistical predia:10c of the Occurrence of maximum magnitude earthquakes, Bull. Seism. Scc. Am.,jk,393-klh, 197h.
- akhareva, A. I., Ccmputer progres for calcu'ating T.,,ax maps, in _
Investigation of Seismi: ~cndi:10ns, "Stiintsa", Kishinev, 197h.
- Zenin, ?. A. , and Novoseleva, M. ?. , ?rediction :f 1:ng cers seismi:
- .4..r .4 . ../ .,n ._e
.as_4, o
. _ e 3 ..mm.
- ,n . . .c _ . g , ,_.o__- 2_.
_m2 3,r ,sc,.g
,o-parsmeters of the ?ribaikal region, in Problems in the quantitative Assessment of Seismic 2 anger, "Nauka",.N secv, 19~2
o8 ,.
APPENDIX tregress Report: New England Crust and *.*;;er :.'.antle S ruc tu e
~'he recent establish =ent of the northeastern seismic array has allowed us to construct a preliminary model of the crust and upper "A .'tle structure beneath Nev England. Because the array has only been in full operation for approximately 2 years, the istase; is limited, and we have analysed the data using a variety cf techniques including:
- 1. observati:ns of relative J3 residuals
- 2. a time ters analysis using ? arrivals 3 three-dimensional modeling using teleseismic ?-vaves 4 analysis of array diagrams 5 refraction studies Preliminary results indicate a crustal thickening under central New Hampshire coupled with a slight crustal thickening vestward Ocvards the North American craton. here is also scme sugges-ion Of a region Of relatively low velocity in the upper mantle beneath central New Ea=pshire and scuthern Maine.
Metheds of Analysis and Results 1".:e relative arrival times of teleseismi: ? vaves were read fr:n enlarged :opies of 13 mm devel corder film. :n general, the firs fev
.f. ..es .x.w, ac,, .ss .,e 4.
.. -sne ..,..
.. ... . , .. f 2.-
.a..,......,.4
. . . .. 7a. _. ,,.,.. ..... . ...
., . a a l e.a '- m a - - m' - a- . ,e a . r ~. .~ ^4- " ". ' .s
. - '. .. . '.. a. 2. '-e- a' .
. ..-.=
precedu e vas required for a nuster of veaEy recorded ;eleseisms in
,,,2 e .
,,.s.
s.. ..,c ...
. ..o
.. .me . ....
- 3. .
.. . .s,..
.. .2
. /. ..,. . .,. . _ . .. ,
.;g.j. , ,,.,.s r- .. $2,. 5 .,...< .,. 2 ego....,a
. m .
., ~.-
. : , y ,..... a .,. 1 . ....,
-. . . . . i. .o.
..... .,.. -,.2 2,.,
. . ./ .=. 2.,
,s ,
3
, . . , , , -..... . . a .
. .v
.....,s.... f.... ., . .
.s....,
. .4
.. 4. 4 ....
.,44 3
.wa. . . . 2..,... . ..e ...,...
~.. .....,
.. ,. ....... s.
. ~ 09 Ab=.o.'".e +. ava' . a . ' e - a. s 4 4." .a ' a- va-a.
. .. . . .a ' .. " ' .= . e A
. v4 . "-
. -a.e.,e..
,o v7
+w ah 7.s. s e .e.A. ,.o 4.a. .a.J. e.e A ., v*
. * - %e 3 J3 _ , obs . ., J3 ij .ij .ij where 3,,*3 is the absolute residual vith respect to '3 tables for
.w s *.a*. *.^ n4. , a.ve n*.
- obs w ,4 , . 4s 'k..e ^.,b s e.-vad '. . a~. a. .'. . ima. . s * .".e' ^. . .' 3' ' n times frem ?':E bulletins ; ~'., a,J3is the theoretica' travel time th cugh a J3 earth.
The residuals were reduced by calculating relative residuals vith respect to a mean residual ecmputed for each event; 3
1 R 44 = R,.,J3 -
- =y .T R<<J3
'd 'O 43 'u Where N is the number of stations reporting ? arrivals f:r a given event. he utilization of relative residua's reiuces scu :e effects a.nd mislocatica errors, removes errors in origin time, and reduces effects of travel path through an inhcmegeneous mant' e. .
= this vay, pcsitive residuals represent late arrivals where the vaves have been sicved in the crust er upper mantle beneath the array.
- m. .,,. ,, . . a.
sese,..,1 .,,,ns.<a-.e.. . .,.a. . ds n . .. ....,,a. m e.< _;,. .> .sa.te residuals which suggest the presence of large sca'. ' regi:nal structu es
.' n . he . .-" s *,, = ..d " y, r e. tar. ..' a. k. a...e .c . h. *."..e = ~. =.f .
. . ~k e d..a. .a .- " c v '.c ,."..
=. .imu+.ha.' v' .' * *,. 4 . n.s .' .. .- a. s .' dual ", a ' ..a. e, , .=-4 v a '. .=. ....' ~. r. - ...
a". =. . = 3
~a statica residua's across the array.
%. , 2., a sa.. a. .' ..v a. . . a. .d , , ' , a .* a. y- .. ". o .' .'., *sc.,
~
b .. c ' ..e-
. . . a. . 5 . a.. .= .' im a. . -.
a..
3 . n e.a.~ . .C awa. .' .2 .6, <-aCv.2
- * - + - *.ue o
.,a. ed.e a. .. ' ,. . ( ' 37, " '.
,. . r.a..e u a y,s o.u ,. . .w.a ., .c . .. s.
- 2. ....g -.w, '#
- 3.4 . . . , . . .J.. .3 .. .y." * .J* e ..C *. .. 2
- * $' .# . n a.'
.. ..""..a. . ' . '*-a...=.#".'.'",f .' . v
.., ,, 4 . .f .r . .w9
.e y . . ... s. .
.a.g.w .....,..- ,
. . . ~
.pa. .e. .e
.e y .
w
.: . . 34-- ....s.
'.,(, 4 * *
- a...*.*'..*,,*
- a. . ".*%..
. ' 2 *...- e. a. a..** .a.*a. . 2 . .* *W , ,f a
..W**- ,
. , **====a..a.a*J
.... .-,.,..d
- .3.*.
e
-1/.2.
..m
. a. ..
s o A.., -4 m.
.y
,'). *...p..i
... .. . e e.....E.o .: a., . ;g .. 2. ~.
.. .. ..%,y
.3 .w g..=. .. % .g .. . * ;g
,F 70 -
scurce of these intrusive ::=plexes is deep-seated ' Chap =an, 1976), and it is possible that this anc aly is related to the f:rnati:n :f these pil*.ons.
S time term analysis sing ? arrivals indicates that the variati:ns in average station residuals may be due to variati:ns in crustal thickness and/or velocity. This is in centrast to the Observed s:i=uthal distribu-tion of residuals for each statica which is probably due to deeper effe s. It was assu=ed that the distribution of average residuals is caused by crustal thickness variations, and the data were inverted to find a crustal thickness zap of New England. The resulting map suggests a crustal thickening beneath central New Hampshire, with scre acr=al thicknesses in Massachusetts and Maine. The ecc: curs of the map pa-o ai the northeasterly trend of the Appalachians.
The variations in crustal thickness observed acrcss the network are also supported by analysis Of array diagress. P ese are sterec6raphic projections of sicvness and a i=uth ancmalies Observed fr:m a plane wave fi: to the vavefront traversing the ne:vork. These studies indicate a Mcho which dips 2 or less to the northwest. This is ne surprisin6 because it is expected tha: the crust vould thicken fres the centinental
=argin cvaris the North American cra cn.
- n additien to the abcve centicned studies , an average crustal velocity scdel has been :cepiled for eastern Massachusetts and southern New Zampshire by ecmbining results from timed qua :/ ilasts with the
.a-- e .,...2 17.az.]g .; 3 . vu,
--. -- aue, - ,a
.sn. . . . . . . . + . _ , .] . e < _ s- ----
. - -- . : - , s . u..n,----.er ,
- 3. s =.
V... . .'.. ' -- %'
. ' = - _= s .' . _' _' .v s -.
,. 7.,
u layer (:ca) ? veiccity (km/sec:
3 - 7.3 5.68
,s.,,ne.1 ne 40 C.40 26.1-38.0 7.33 Moho 3.13 Future Studies Studies for the next year vill be aimed at improving the preliminary crust and upper mantle model for :Tev Ingland. '"his will be achieved by using additional teleseismic ? and pk? data. The database is currently being expanded to include readings frem short period stati:ns i: Ocnnecticut and eastern : lev Terk.
The st'uctural models derived fren the residual studies vill be ecmpared to these frcm 1ca 6 period surface wave dispersica studies.
Phase velocities are presently being :cmputed as a functica :f azimuth frca the ^uebe:-Maine border event of June 15, 1973, and simple crustal
=cdels vill be developed. Phase velocities vill also be ressured sing the two stati:n technique.
More elaterate medels vill be generated by perferning a. sinu':aneous inversion of phase velocity and attenuation f011 ving the techniques of i s
.ee and e.c_,:=cn ,_. i,e;.
.4 a. .,
. . , ; ., .. .. e - '
-s -hasa.,
e a a~ ..c . . , e . ' .-d " .' s.".e.-
. . .c d a. '~ v a. ~.. =.v
. a,."'- '
%e 4 on.4 4..a.24. . . ..v whp co' **,
w ,.Cn
. f & e. . . m.. .'.so. n.. ..
..33 ,
. 3.o. rs _ . g .4 w,. a ..
. .v
. . .: *-g,
' g *.a f..' .' ..4
~
m.=
y e m .a.s~ o. v.
-g . i ~ ~. . ". % a. . %. e. n. .~ .' .' s. .s =. a
. A.
a . .s .%- = a. o. e. r. m c .*. a. .. . ,
.L.' s~ 4. *.. n .'
..w C . ' # .~ g~ ' ' ' * * *
... . ****^
~- .**
..~.w.a . v *. .' ~~*^
w y . . ' ~.'V.. .
T2 .
c <
References Aki, K., A. Christoffersson, and I. S. Eusebye, Determinati:n Of the three-dimensional seisaic structure of the lithosphere , i. Geophys .
Res,j2_,277-296,1977 Chapman, 2. A., Structural evolutien of the '4hite Mountain magma series, aeos. acc. ,m n , u em., ao , c-a1-;u0, ,94o.
.,i Lee,*4. 3. and S. C. Solcmon, Inversion schemes for surfree wave sttenua-tion sad Q in the crust and the mantle, Geophys. [< l. Astron. Sec.,
33,'47-71,1975