1.The　Generation　of　Tsunamis　and　the　Focal　Mechanism　of　Earthquakes

ABSTRACT

　The　generation　of　tsunamis　is　discussed　in　connection　with　the　type　of　faulting　associated　with　the　focal　mechanism　of　tsunamigenic　earthquakes.The　focal　mechanism　of　tsunamigenic　earthquakes　and　of　earthquakes　not　accompanied　by　tsunamis,occurring　in　or　near　Japan　during　the　period　1926　to　1968,were　investigated　on　the　basis　of　radiation　pattern　of　P−waves　and　for　some,of　S−waves.Only　earthquakes　having　a　magnitude　greater　than　6.3　were　taken　into　consideration.
　The　faulting　inferred　from　the　focal　mechanism　of　tsunamigenic　earthquakes　is　mostly　the　dip−slip　type;the　generation　of　large−magnitude　tsunamis　is　especially　associated　with　dipslip　type.The　earthquakes　associated　with　strike−slip　faulting　are　almost　never　accompanied　by　tsunamis,even　though　the　magnitude　of　the　earthquake　is　nearly　seven.Also,most　after−shocks　are　not　associated　with　the　generation　of　tsunamis.
　The　generation　of　tsunamis　is　closely　related　to　the　type　and　dimension　of　dislocation　in　the　source,being　associated　with　characteristics　of　the　earthquake,such　as　magnitude,focal　depth,and　water　depth　at　the　epicenter.

INTRODUCTION

　The　generation　of　tsunamis　associated　with　the　focal　mechanism　of　tsunamigenic　earthquakes　occurring　in　or　near　Japan　was　discussed　by　the　present　author（Lida,1967）at　the　eleventh　Pacific　Science　Congress　at　Tokyo　in　1966.The　generation　of　tsunamis　has　been　found　to　be　greatly　dependent　on　characteristics　of　an　earthquake,such　as　magnitude,focal　depth,focal　mechanism,aftershock　activity,and　water　depth　at　the　epicenter.This　paper　is　a　progress　report.The　present　investigation　is　concerned　with　tsunamigenic　earthquakes　during　the　period　1900　to　1968.However,since　the　more　accurate　data　are　those　from　1926　to　1968,during　which　the　focal　mechanism　of　all　earthquakes　having　a　magnitude　greater　than　6.3,even　those　not　accompanied　by　tsunamis,was　investigated,this　time　interval　was　specially　studied.

CATALOGUE　OF　TSUNAMIGENIC　EARTHQUAKES　AND　TSUNAMI　MAGNITUDE

　A　catalogue　of　tsunamigenic　earthquakes,compiled　from　various　reports　and　papers　and　from　mareographic　records　of　tidal　stations　distributed　along　the　coasts　of　Japan,is　presented　in　Table1　along　with　tsunami　magnitudes.Tsunami　magnitudes　m　were　determined　by　the　following　formula
　　　m＝　log_2　h　or　h　＝　2.0^m,（1）
in　which　h　is　the　maximum　height,in　meters,measured　at　a　coast　10　to　300km.from　the　tsunami　origin.
　The　geographic　distribution　of　epicenters　of　tsunamigenic　earthquakes　from　1900　to　1968　is　shown　in　Fig.1,classified　according　to　tsunami　magnitudes.From　Fig.1,it　is　apparent　that　most　of　these　tsunami　sources　are　located　on　the　Pacific　side　in　northeastern　Japan.Figure2　shows　the　relationship　between　the　earthquake　magnitude　and　focal　depth　of　an　earthquake　accompanied　by　a　tsunami.The　magnitude　of　a　tsunami　is　shown　in　round　numbers　in　Figure2.Earthquakes　occurring　off　the　coast　accompanied　by　tsunamis　and　not　accompanied　by　tsunamis　are　depicted　by　filled−in　circles　and　by　plain　circles,respectively.Data　for　the　magnitudes　and　focal　depths　of　the　earthquakes　were　taken　mainly　from　the　「Catalogue　of　Major　Earthquakes　Which　Occurred　In　and　Near　Japan」published　by　the　Japan　Meteorological　Agency（1958,1966）.
　Figure2　shows　that　there　is　an　approximate　linear　boundary　between　earthquakes　accompanied　by　tsunamis　and　those　that　are　not　accompanied　by　tsunamis.This　boundary　is　considered　to　be　a　limiting　magnitude　for　earthquakes;below　this　magnitude　tsunamis　do　not　occur,as　reported　in　previous　papers（Iida,1958,1963,1965）.This　limiting　magnitude　for　tsunamigenic　earthquakes　may　be　generally　expressed　by　the　straight　line　as
　　　M＝6.3　＋　0.005　H,（2）
where　the　focal　depth　H　is　in　kilometers.This　straight　line　is　shown　as　the　full　line　in　Figure2.The　coefficient　of　H　in　equation　2　is　somewhat　different　from　that　in　previous　papers（Iida,1958,1963）.From　Fig.2,it　can　be　seen　that　there　are　three　tsunamigenic　earthquakes　located　on　the　left　side　of　the　straight　line　expressing　the　limit　of　equation　2.Taking　those　three　into　consideration,the　limit　may　be　expressed　by
　　　M＝5.6＋0.01　H.（3）
This　relation,expressing　the　smallest　magnitude　of　a　tsunamigenic　earthquake,is　the　same　as　before.
　The　limit　for　disastrous　tsunamis,having　a　magnitude　of　more　than　2,may　also　be　determined　by
　　　M＝7.7＋0.005　H,（4）
as　shown　by　the　broken　line　in　Figure2.Thus　the　earthquake　accompanied　by　tsunamis　having　magnitude　of　less　than　2　may　generally　be　located　between　the　full　and　broken　lines　in　Figure2.It　is,however,noticed　that　four　tsunamis（1940　Hokkaido,1953　Boso,1963　Iturup,and　1964　Niigata）having　a　magnitude　of　2　or　3　are　located　on　the　left　side　of　the　broken　line　of　equation　4.It　is　noted　that　the　magnitude　of　the　1963,Iturup　tsunami　was　especially　great　compared　with　its　earthquake　magnitude.
　As　seen　in　Fig.2,even　though　the　magnitude　is　greater　than　that　located　on　the　right　side　of　the　line　representing　equation　2,there　are　a　number　of　earthquakes　not　accompanied　by　tsunamis.The　number　of　these　earthquakes　for　the　period　from　1926　to　1968　totaled　79;the　epicenters　are　shown　in　Figure3.These　earthquakes　are　classified　into　three　groups:（1）the　aftershocks,32　in　number,（2）the　relatively　deepfocus　shocks,15　in　number,and（3）others,32　in　number.Groups（1）and（2）of　these　earthquakes　are　generally　considered　not　to　generate　tsunamis.Therefore,the　reason　group（3）is　not　accompanied　by　tsunamis　must　be　explained.For　this　investigation,the　mechanism　of　earthquakes　is　considered.

MECHANISM　OF　EARTHQUAKES　AND　TYPES　OF　FAULTING

　To　see　the　difference　between　tsunamigenic　earthquakes　and　non−tsunamigenic　earthquakes　with　magnitude　greater　than　6.3,focal　mechanism　is　investigated.There　are　a　number　of　papers　concerning　the　focal　mechanism　of　earthquakes.Most　of　the　data　concerned　were　taken　from　the　publications　of　the　Dominion　Observatory,Ottawa　by　Wickens　and　Hodgson（1965,1967）and　publications　by　Stauder　and　Bollinger（1964,1965）.Further,some　of　the　data　were　taken　from　the　technical　reports　of　the　Japan　Meteorological　Agency,especially　from　the　report　on　individual　earthquakes（1969）and　from　the　papers　by　Ichikawa（1966）and　others（Hirasawa,1965）.
　Some　examples　of　focal−plane　solutions　of　tsunamigenic　and　non−tsunamigenic　earthquakes　are　shown　in　Fig.4,which　shows　the　distribution　patterns　of　compression　and　dilatation　of　P　initial　motions.Some　radiation　patterns　of　S−waves　were　also　considered,together　with　those　of　P−waves,especially　in　the　results　by　Stauder　and　Bollinger（1964,1965）.From　these　radiation　patterns,not　only　the　nodal　planes　and　their　dips　can　be　determined,but　also　the　strike　and　inclination　of　the　nodal　planes.Assuming　that　the　faulting　occurs　along　the　nodal　plane,the　dip　and　strike　components　of　the　unit　vector　of　the　movement　due　to　faulting　can　be　calculated.When　the　strike　component　thus　obtained　is　greater　than　the　dip　component,the　faulting　is　called　strike−slip　type;when　the　dip　component　is　greater　than　the　strike　component,the　faulting　is　called　dip−slip　type.In　this　way　the　type　of　faulting　was　investigated　for　each　of　the　earthquakes　mentioned　above.The　results　are　listed　in　Tables　1　and　2.It　can　be　seen　that　more　than　60　percent　of　the　tsunamigenic　earthquakes　are　from　faulting　of　dip−slip　type.
　The　geographic　distribution　of　the　epicenters　of　tsunamigenic　earthquakes　classified　according　to　the　type　of　faulting　is　shown　in　Figure5.From　this　figure,it　can　be　determined　that　there　are　few　tsunamigenic　earthquakes（only　8　percent）produced　by　faults　of　strike−slip　type.Further,there　are　a　number　of　earthquakes　on　the　northern　Pacific　Ocean　side　for　which　the　types　of　faulting　are　not　determined.Figure6,which　corresponds　to　Fig.3,shows　the　geographic　distribution　of　the　epicenters　of　submarine　earthquakes　not　accompanied　by　tsunamis,classified　according　to　the　type　of　faulting.From　Fig.6　it　is　seen　that　most　earthquakes　show　the　strike−slip　type　fault,whereas　few　earthquakes　are　from　dip−slip　type　faulting.

TSUNAMI　MAGNITUDE　AND　TYPES　OF　FAULTING

　The　relationship　between　tsunami　magnitude　and　the　magnitude　of　a　tsunamigenic　earthquake　was　investigated　in　connection　with　the　types　of　faulting　as　shown　in　Figure7.From　Fig.7　it　can　be　generally　concluded　that　the　greater　the　magnitude　of　an　earthquake,the　larger　is　the　tsunami　magnitude.Also,most　large　tsunamis　occur　in　association　with　dip−slip　type　faulting.The　tsunamigenic　earthquakes　marked　by　N　in　Fig.7　are　from　the　normal　fault　type,whereas　those　without　N　are　from　reverse　fault　type.Therefore,it　may　be　concluded　that　most　earthquakes　accompanied　by　tsunamis　in　or　near　Japan　are　of　the　reverse　fault　type　and　that　the　relation　of　the　tsunami　magnitude　to　the　reverse　and　the　normal　fault　type　is　not　clearly　distinguishable.

DISCUSSION

　As　is　generally　known,tsunamis　accompany　only　those　earthquakes　whose　epicenters　are　at　shallow　depths,less　than　100　km.,beneath　the　ocean　or　close　to　the　shore;most　tsunamis　are　generated　by　abrupt　deformation　of　the　sea　bottom,either　fault　displacements　or　more　general　deformations　caused　by　earthquakes,including　deformations　due　to　submarine　volcanic　explosions　or　landslides.Tsunami　magnitude　can　be　estimated　from　the　generation　mechanism　of　tsunamis,the　dimensions　of　the　area　of　tsunami　origin,the　speed　and　amount　of　displacement,and　the　water　depth　at　the　tsunami　source.The　dimensions　and　degree　of　crustal　deformation　of　an　ocean　bottom　is　closely　related　to　the　earthquake　magnitude,its　focal　depth,and　the　focal　mechanism　of　the　earthquake.Further,the　area　of　tsunami　origin　is　closely　related　to　the　area　deformed　by　an　earthquake　and　is　approximately　equal　to　the　area　of　the　aftershock　activity.
　Tsunamigenic　earthquakes　are　always　followed　by　a　number　of　aftershocks.The　generation　of　tsunamis　accompanying　earthquakes　is　closely　correlated　with　dip−slip　type　faulting,i.e.,the　faulting　of　most　tsunamigenic　earthquakes　is　of　the　dip−slip　type.The　faulting　of　most　non−tsunamigenic　earthquakes　having　magnitudes　greater　than　6.3　is　of　strike−slip　type.The　process　of　tsunami　generation　is　schematically　illustrated　in　Figure8.
　Most　aftershocks　are　generally　not,accompanied　by　tsunamis,but　there　are　some　aftershocks　which　are　accompanied　by　tsunamis（such　as　1938　Fukushima,1952　Hokkaido,1960　Sanriku,1961　Ibaraki　and　Kushiro,1963　Iturup,and　the　1968　Aomori）,although　the　tsunami　magnitudes　are　usually　comparatively　small.
　Abrupt　deformation　of　a　sea　bottom　is　really　caused　by　a　main　shock.By　the　change　in　stress　or　strain　due　to　this　deformation,aftershocks　are　caused.Consequently,we　consider　an　aftershock　not　to　partake　in　the　production　of　deformation　of　the　sea　bottom.Since　a　tsunami　results　from　sea−bottom　deformation,which　an　aftershock　does　not　produce,aftershocks　are　not　accompanied　by　tsunamis.Therefore,if　an　earthquake　which　is　considered　to　be　an　aftershock　is　accompanied　by　a　tsunami,that　earthquake　is　regarded　as　an　independent　shock　and　not　as　an　aftershock　of　the　main　shock.
　The　crustal　deformation　due　to　a　large　earthquake　may　be　classified　into　two　types:the　deformation　may　completely　finish　during　the　vibration　of　the　main　shock,or　the　deformation　may　continue　for　several　hours　or　several　months　after　the　end　of　the　vibration　of　the　main　shock.The　former　type　of　deformation　is　called　the　first　kind　of　deformation　and　the　latter　type,the　second　kind　of　deformation.Any　tsunami　is　attributed　to　the　first　kind　of　deformation.True　after−shocks　are　generated　by　the　release　of　the　stress　secondarily　produced　in　the　area　of　the　crustal　deformation　caused　by　the　main　shock.Therefore,it　is　concluded　that　an　aftershock　does　not　produce　new　deformation　of　a　sea　bottom　in　the　area　deformed　by　the　main　shock　and,consequently,an　aftershock　does　not　generally　produce　a　tsunami.The　conditions　of　tsunami　generation　are　greatly　dependent　on　the　focal　mechanism　of　earthquakes.

BIBLIOGRAPHY

Hirasawa,T.1965.Source　mechanism　of　the　Niigata　earthquake　of　June　16,1964,as　derived　from　body　waves.J.Phys.Earth.Vol.13,pp.35−66.
Ichikawa,M.1966.Mechanism　of　earthquakes　in　and　near　Japan,1950−1962.Papers　in　Meteorology　and　Geophysics.16,pp.201−229.
Ichikawa,M.1966.Statistical　investigation　of　mechanism　of　earthquakes　occurring　in　and　near　Japan　and　some　related　problems.Kenkyu−Jiho.18,pp.1−72.
Iida,K.1967.Earthquakes　and　tsunamis（abstract）,Symposium　on　Tsunami　and　Storm　Surges,the　Eleventh　Pacific　Science　Congress,Tokyo　1966.Proc.PSC　Tsunami　and　Storm　Surges　Symposium.
Iida,K.1958.Magnitude　and　energy　of　earthquakes　accompanied　by　tsunami,and　tsunami　energy.J.Earth　Sciences.Nagoya　Univ.,6,pp.101−112.
Iida,K.1963.Magnitude,energy,and　generation　mechanisms　of　tsunamis　and　a　catalogue　of　earthquakes　associated　with　tsunamis.Proc.Tenth　Pacific　Science　Congress　Symposium.IUGG　Monograph　24,pp.7−18.
Iida,K.1965.Earthquake　magnitude,earthquake　fault,and　source　dimensions.J.Earth　Sciences.Nagoya　Univ.,13,pp.115−132.
Japan　Meteorological　Agency　1958.Catalogue　of　major　earthquakes　in　and　near　Japan（1926−1956）.Seismological　BU■■.Supplementary　volume,No.1,pp.1−91.
Japan　Meteorological　Agency　1966.Catalogue　of　major　earthquakes　in　and　near　Japan（1957−1962）.Seismological　Bu■■.Supplementary　volume,No.2,pp.1−49.
Japan　Meteorological　Agency　Seismological　Bull.1963−1968.
Japan　Meteorological　Agency　1969.Report　on　the　Tokachi−oki　earthquake,1968.Technical　Report.No.68.
Stauder,W.and　Bollinger,D.A.1964.The　S−wave　project　for　focal　mechanism　studies,earthquake　of　1962,AFAFOSR−62−458,Dept.Geophysics　and　Geophysical　Eng.,Inst.Technology,Saint　Louis　Univ.
Stauder,W.and　Bollinger,G.A.1965.The　S−wave　project　for　focal　mechanism　studies,earthquake　of　1963,AFAFOSR−62−458,Dept.Geophysics　and　Geophysical　Eng,,Inst.Technology,Saint　Louis　Univ.
Wickens,A.J.and　Hodgson,J.H.1967.Computer　re−evaluation　of　earthquake　mechanism　solutions　1922−1962.Pub■.Dominion　Obs.Ottawa.33.