Paleogeography of the Ryulryu Islands

TROPICS Vol. 10 (1): 5-24 Issued May 30, 2000

Paleogeography of the Ryulryu Islands

Masaaki KIIURA Department of physics and Earth sciences, university of the Ryukyus,Senbaru 1, Nishihara, Okinawa 903-O2t3,Iapan

ABSTRACT Recent submarine, geological, and geophysical investigations including diving surveysreveal the geo-history of the Ryukyu Islands and the East China Sea. Tivo stages are fundamental forformation processes of the Ryukyu Arc. The crustal thinning in the westem part of the East China Seaand thus eastward drifting of the Arc may have occuned in the late Miocene to middle pliocene at thefirst stage. Between 1.6 - 1.3 Ma, the East China Sea area, including most of the Okinawa Trough, mayhave been subaerial. At that time, the Ryukyu Arc region may have been a part of the Eurasian continent.Extensive subsidence may have occurred at the second stage, at about 1.3 Ma, in the early pleistocene.The present Ryukyu Arc (Ryukyu Ridge) has been formed since then. The Ryukyu Arc may have beennearly connected to the Chinese continent, through Taiwan as a land bridge, sometime duiing the twomajor development periods (such as sometime during 1.6 - 1.0 Ma, and 0.2 - 0.025 Ma). The paleo-landmay have been submerged step by step since 0.03 Ma by both crustal movement and sea-level rising afterthe last Ice Age. Submarine stalactite caves at 10 - 35 m deep off the Ryukyu Islands were discovered.The caves have subsided since the Wiirm lce Age. Stone tools were also recovered inside one of them.Additionally, archeological evidence in the form of a stepped pyramid, estimated at greater than 6,0(X)years old, has been discovered beneath the sea off Yonaguni Island. Existence of such submarine ruinsprovide indicators of subsidence pro@sses of the Ryukyu Arc,

Key words: Ryukyu paleoland / Okinawa Trough / paleogeographic map / land bridge / submarinestalactite cave / submarine ruins.

In recent years submarine, geological, and geophysical investigations have been canied out in theRyukyu Arc region by multiple international organizations (e.g., Kimura, 19854 b, 1996a, b; JapaneseDELP Research Group on Back-Arc Basins, 1991; Sibuet et al.,1987,1998). The Ryukyu Arc (Fig.1) lies along the eastern margin of East China Sea. This paper discusses geologic environmentalchange of the Ryukyu Arc and the East China Sea using all available data. Seismic reflection profilesof single and multi-channel systems, obtained from most organizations in the world, were compiledfor the present study to clarify geological and stratigraphic sequenoe in the Ryukyu Arc. Dredging anddrilling data were also incorporated to determine the stratigraphy in the studied area. Detailedtopographic data collected with multi-nanow beam, manned anal unmenned submersible, and seismicrefraction experiments were also available to study the geologic and geophysical features in theRyukyu Islands (Nansei-Shoto Islands) areas. Adding to this, diving surveys around the RyukyuIslands, using SCUBA, show detailed submarine topography and geolog5r.

M. KItr,tuRA

1 25'E

Okinawa Trough

Tokara lslands

Tunghai ShelfAmami-Oshima

Ryukyu Ridge

100kmYaeyama tinigaxi-jima

lriomote-jima

Okinawa-jima

Kerama Gap

Miyako-jima

126"

* Submarine stalactite cave

A Submarine ruins

Fig. 1. Submarine topography around the Ryukyu Arc'

MORPHOIOGY AT{D TECTOMC TRAMEWORK OF TTIE RYTJKTU ARC

The morpho+ectonic framework in the study region can be defined from west to east in Fig 2, namely

L) Tirnghai Shelf, II) Tunghai Slope (western rifted margin of the Okinawa Trough), III) Okinawa

Trough, IV) Tokara Belt (eastern rifted margin of the Okinawa Trough), V) Ryukyu Ridge, VI) Arc-

trench gap and, VII) Nansei-shoto Trench (Ryukyu Trench). The Ryukyu Arc includes the Tokara Belt

(IV) and the Ryukyu Ridge (v) (Kimura, 1985b). The width of the present okinawa Trough is about

100 - 150 km. The topography of the Tunghai Slope and Tokara Belt is rough, and both features are

regarded as rifted margins of the Okinawa Trough. Active volcanism is recognized in the volcanic

Paleogeography of the Ryukyu Islands

Legend

1)

-2)

3)^ll4)As)A30" N

24'

"124" E

East China Sea

Taiwan

{r

F

Shimabara

RF

Amami-Oshimaokunoshima

il,H*)sVII

Philippine Sea

200 km

Fig. 2. Map showing morpho-tectonic framework of the study area modified from Kimura etaf. (D88). I: Tunghai Shelf. II: Tunghai Slope. III: Okinawa Trough. IV: Tokara Belt. V:Ryukyu Ridge. VI: Arc-trench gap. VII: Nansei-shoto Trench. TF: Tunghai Shelf Fault. RF:Ryukyu Ridge Fault. BTL: Butsuzo Tectonic Line. Active hydrothermal mounds were foundin central rifts labeled (t)-(a). (t): Torishima Rift (including Minami Ensei Knoll). (2): IheyaRift. (3): Aguni Rift (including Izena Hole). (4): yaeyama Rift. Irgend: 1)= central rift.2)=fault scarp. Tooth down thrown side. Broken line shows a buried fault. 3)=trench. Tooth upthrust side. 4)= historically active volcanoes. 5)= probably active volcanoes (mostlysubmarine).

front of the Tokara Belt. Active rift movement, associated with classic grabens, was identified in themiddle and southern part of the Tokara Belt (Kimura et at., 1975; Sibuet er at., 1998). Deep faults,such as the Ryukyu Ridge Fault and the Tunghai Shelf Fault, were generated in the early pleistocene

along boundaries of morpho-tectonic division.

The Okinawa Trough is a back-arc basin that has developed generally parallel to the Ryukyu Arc.The Okinawa Trough is estimated to have been filled with deposits since late Miocene time and thecentral grabens or rifts develop along the axial part of the trough arranged in an echelon formation

tl

BTL

@)>6 ..,s

\ LtE W ......y'ttko-jima\tt'r\r.,.r...rfhigaki-jima

)

M. KtnauRA

125" E

30" N

Fig.3. Geological map of the Ryukyu Arc region- Legend: 1) Ryukyu Group (1.3 - 0.01 Ma)

(unconsolidated sediments), 2) Ryukyu Group (limestone), 3) Non-marine sediments (B1,

1.5 - 1.3 Ma), 4) Pre-Bl strata'

(Fig. 2). Existence of the central graben was first described in the middle Okinawa Trough area by

Kimura et al. (1975).Intrusive bodies exist beneath the central grabens to form basal highs in many

cases which defines the central graben as the most recent fissure originating in reSponse to magma

uprising. Many active hydrothermal spots including black smokers, have been found since 1986 in the

Torishima Rift (Ensei knoll), Iheya Rift, and Aguni Rift (Izena Hole) in the middle Okinawa Trough

(Fig. 2) (Kimura et al.,1988,1989, 1991; Nakamura et al, L99O; Halback et a1.,1989).

CRUSTAL STRUCTURE AI\D STRATIGRAPHIC SEQUENCE

Seismic refraction experiments using TNT dynamite has been carried out more than four times since

1984. Before that, there were few lines crossing the Ryukyu Arc and the Okinawa Trough. There are

many sites of sono-radio buoy experiments. All of the seismic refraction data show a similar picture


I

| + GreaterOkinawaTrough

-

.l-l-lTunghai Shelf lTunshai

slope I

o*inawa Trough I tot<ara BetrI I hB'\ N-t

Ryukyu Rids.l

sea level M Bp 4

Br -Bz cenrral rifr B, -Bz

gz B, o.S ,B'

,I;J;W;;J!T:\\ /'lv intruiion]ttunghai Shelf Older Okinawar Faulr Trou gh

Ryukyu Ridge Faulr

EI,,T r,ffi

Fig. 4. Cross section of the Ryukyu Arc and its vicinity, representing stratigraphicsequence. Location is shown in Fig. 2. A-c: See table 1 D: Miocene strata, E:Pre-Miocene strata, K: Kunigami Gravel, N: Nakoshi Formation (=ChinenFormation, CH), M: Minatogawa Limestone.

of the crustal structure of the Okinawa Trough from north to south. The depth of the MohoDiscontinuity varies 25 - 16 km from north to south (Iwasaki et al., 1987;Hirata et at., l99l) andthere is no variation from east to west. Chemical analysis of major-and rare-earth elements collectedfrom the Central Rifts of the Okinawa Trough, and those in other portion of the Trough, show island-arc affinity. There are no magnetic anomalies in the Okinawa Trough, indicating that there is nooceanic crust underlying the Trough (Sibuet et al., 1987). Gravity anomalies also indicate that theOkinawa Trough is comparable to a continental crust. These data indicate that there is no oceaniccrust beneath the entire Okinawa Trough.

Based on detailed velocity structure in the southern Okinawa Trough: the layer of 7.5 km/s(kilometer/second) is correlated with that of the mantle and the layer of 6.2- 6.4 km/s with the layer 2 andthe 4.5 - 6.0 km/s layer is conelated with granitic layer (Ihrata et a1.,1991). The thickness of the 4.5 - 6.0km/s and 6.2 - 6.4 km/s layers decreases remarkably beneath the trough. This suggests that crustalstretching occured after the formation of the 4.5 - 6.0 krn/s layer. The 3.6 - 3.9 km/s layer appearsconelated with the Yaeyama Group of the early Miocene, based upon the stratigraphic sequonce and sonicvelocity. The 3.0 - 3.5 km/s layer is conelated with late Miocene strata on the basis of seismic reflectionrecords and drilling data (Marutani & Sato, 1985). The 1.8 - 1.9 km/s layer is correlated with earlyPliocene to early Pleistocene sediment, such as the upper Shimajiri Group and euaternary sediments.

Seismic reflection profiles reveal a deep sedimentary basin resembling a big trough beneath theTbnghai Slope, and the sedimentary layer in the basin shows it is about 6 km thick (Marutani & Sato,1985; Kimura, 1985a, b; htouzey & Kimura, 1986). This trough is tentatively called older Okinawa

10 M. KlrrauRA

ld Yt Epoc Okinawa Trough Kerarna Saddlc I Okinawa Island

I2

4

6

E

l0

20

40

60

80

t00

150

200

-

-

-

-

Holoccnc A Holocene HoIocere

H

gt).-()E

gdJ

r=eo

=lht#thW

B

manne

(Bz)

E litlEfrnrnillItrEOrincn /Formation

shinzato

[uililffiqililililililn|

nestone

Rddish LimcstoncffiFormation

#E

-+Htrl

(Br)non-marine

+urarinc ( ? )

C)trt)TJord

tFlil!l-ql

Fl

go

(f:E'RlE

a

C2 non-Illannpf"l\r,

--? -.marineC1 Yonabanr F,ormation

Table 1. Stratigraphic correlation.

Glacial/Intcrglacial

Itrtlltrl Unconformity

Trough (Figs. 3 and 4). Thus, there are two sedimentary basins, an older, and a younger Okinawa

Trough, in the northern half of the backarc region. In the southern part, a younger one seems to be

coincident with the older one.

tWurrn

I

$ *t*t

F"

tlc,'*

,1,

0Donau


TECTONIC HISTORY

Paleomagnetic evidence shows that from L0 - 4 Ma, the southern part of the Ryukyu Arc rotatedclockwise L9 degrees with respect to Eurasia (Miki er al., l99O; Miki, 1991). This rotation is attributedto the back-arc opening of the southern part of the Okinawa Trough, which occuned by means of the

"wedge" mode after 10 Ma (Miki et al., 1993). Thus, the major cause of lateral movement of the outerterrane in the Ryukyu Arc is thought to be primarily because of both the opening of the OkinawaTrough after the middle Miocene, and the bending of the outer terane in Kyushu and Taiwan in the lateMiocene. Thereafter, the subducting Philippine Sea plate may have been coupled with the continentalplate, at the Ryukyu Trench, while the Eurasian Plate extended and rifted to make a back-arc basin.Uplifting may have occurred in the Okinawa Trough region before its subsidence when the regionalangular unconformity was formed sometime during Middle to Late Miocene.

As a result, crustal stretching may have occurred after formation of the 4.5 - 6.0 and 6.2 - 6.4 km/slayers. Whereas, the 3.0 - 3.5 km/s layer (? late Miocene strata) seems to be a syntectonic sedimentarylayer formed during major extensional movoment of the Okinawa Trough. As a result, the thinnercrust in the back-arc basin was rifted, but not completely lost. The direction ofgeneral bedding trendsand the geologic terrane in the South Ryukyu changed sometime during the middle Miocene /Pliocene. Consequently, pre-middle Miocene terranes were shifted trench ward. The timing ofbending of the Shimanto Terane is thought to be later than the timing of the extension of the JapanSea. Thus the spreading of the Japan Sea should be compatible with the spreading of the OkinawaTrough; both of which were characterized by volcanism called Green-Tuff activity. Following this, theokinawa Trough was filled up with sediment from the upper Shimajiri Group.

Subsequently, Tanegashima Island, located in the northern Ryukyu Arc, has been rotated counter-clockwise by about 30 degrees, with respect to Eurasia and Southwest Japan since about 2 Ma,(Kodama et al., L99l; Kamata & Kodama, 1994). The depocenter of the trough jumped to the east at 2Ma, and is not coincident with the northern-half of the present Okinawa Trough, but is coincident withthe southern Okinawa Trough.

Drilling data from a Japanese oil company show that there are Pliocene to Pleistocene sedimentsof 2,490 m thick in the northern-most part of the Okinawa Trough, of which the lower part isconelated with the Shimajiri Group, and the upper part is correlated with Layers 81 and 82 (RyukyuGroup), as shown in table 1. A thick B1 layer is expected in the central part of the Okinawa Trough aspounding sediment estimated as a non-marine sediment. The sedimentary basin may have anexpanded origin controlled by the clockwise rotation of the Ryukyu Arc since 1.6 Ma. The Beppu-Shimabara Graben in central Kyushu has been active since 1.6 Ma, accompanied by extensivevolcanism, and faulting. The Kuchinotsu Group was deposited in the graben, formed at this time.Contemporaneously, the margin of the central basin of the Okinawa Trough was eroded subaeriallyfrom 1.6 - 1.3 Ma, as shown by the unconformity under l-ayer B2in the Tokara Belt.

Subsiding movement since 1.3 Ma has formed the Ryukyu Ridge. Sea level became relativelyhigh, and the Ryukyu Limestone deposited at the Philippine Sea side from 1.3 - 1.0 Ma. The GreaterOkinawa Trough (fig. a) has been formed since 1 Ma. Since then, rifting in the central graben of theGreat Okinawa Trough has developed.

Available data confirm evidence of volcanic activities in the central rift vallev in the middle

tl

t2 M. KTNAURA

c

30' N o ooo

L1e €r

(t

3

s3 LT

Tokara Ga

;,\

'Miyako

Ryukyu RidgeOkinawa

OKerama Gap

LK

100 km

O Limestone, * Fossil mammal,tO Drill site,-Sparker line

Ftg.5. Ryulryu Ridge and sonic survey lines by means of sparlrer (30,01nioule).

Okinawa Trough (Kimura et al.,1986 Kimura, 1991). Potassium-Argon (IVAr) dates of volcanic

rocks composed mostly of high alumina basalt and decide represent activities younger than 0.5 Ma,

(Kimura et at., 1986\. Some of the central knolls in the rift valley revealed active hydrothermal

mounds ("SHINKAI 2000" Research Group on the Okinawa Trough, 1986). Furthermore, black

smokers were found at Izena Hole in the middle Okinawa Trough (e.g., Nakamuta A al', l9X)).

I"AI\D BRIDGES DURING THE PAST 0.2 MA

There exists a very long, flat terrace on the summit through the entire Ryukyu Ridge: cross sec{ions of

this terrace (Fig. 5) show essentially the same features, such as uncovered soft sediments on the top

and thick sedimentary coverage of side slopes of the terrace, despite surface depths varying from 100

Paleogeography of the Ryukyu Islands L3

WEST

EA

EAST

water depth0km

I

2

,lOkm ,

OkmI

2r lOkm

I

0km

I . ro km,

.r | - - ---'- |

TA

l0kmt. I

, lOkm,

0kmI

2

3

V.E.: X 3 -4 ,l0km ,

Fig. 6. Interpreted_cross section using sparlrer profites (30,0fi)joules) right angte tonrykyu Ridge showrng TA. I-ocations of.E4-Er2are shown in Fig.5. e-e ur"explained in Fig. 4. OK_lx: drilling site.

- 500 m as shown in Figs 6 and7. This terace is named Terrace A (TA). Based on seismic reflectionsuryeys, 6sth single and multi-channel, the TA is offset by normal faults, thus its depth varies fromlfi) m to more than 1,0(x) m in the Tokara and Kerama Gaps as shown in profiles.

L4 M. KIUURA

1 ,300-cPo,q)EL-(t)t-ts- 1,400

1,200m

1,500

sw

RN87D4'

OK-lx

RN87D-5-l

RN87D-5-2

/ D*o-rbDT30-la

2K362

RNE?OK.7

TAwater

depth

om

DT33-l two way

travel

trrue10 sec

E,Ryutyu Limesrone E ,chin"n Formaaion C'shimajiri GrouP D, Prc-Teitiary formation

{< : fossil mammal

Fig. 7. Cross section of the Tokara (LI) and Kerama Gaps (LK). Locations of LT and

LK are shown in Fig. 5, OK-lx: Drilling site. Numbers represent sample numbers

(after Kimura, 1996).

Exposures of bedrocks are recognized broadly on most of TA in the Kerama and Tokara Gaps' as

observed by "SHINKAI 2(X)0", "Nagasaki Maru" and others. outcrops of bedrock continue from

shallows to 100 m depths, suggesting that less depositional time for newer sediments after the erosion

of the TA surface. Thus, the flat plane of Tenace A (TA) is identified as the erosional surface of post-

formation Ryukyu Limestone in the late Pleistocene from 0.2 - 0.02 Ma (Kimura et al', 1992;1994;

1996). As a result, TA has been offset by normal faults from 100 m to over 1,000 m depths, based on the

sonic profiles (e.g., Fig. 7). Both TA shallows and TA depths are identified as the same erosional surface'

Ryukyu Limestone composing the bedrock that was recovered from a flat terrace at 935m in the

Kerama Gap was dated using nanno fossils. The age of the limestone was estimated as younger than

O.27 Ma. The age of the fossil bivalve Amusium japonicum (Tsukihi-gai in Japanese) in the same

Tokara Saddle

2K609-5

? 2k6oe-1 \ 3K142

V.E.:x3o /l^, ll 82? 'rJ

2K54ll

3K149

v3Kilr-3 / Kl3Kllol.3,5-7 SKlll-l-2

Paleogeography of the Ryukyu Islands 15

Table 2. Radiocarbon ages of samples recovered from Kerama and Tokara Gaps (or Saddles).

Kerama Saddle Tokara Saddle

770i60 yr BP (foraminifera2K5403 1,6L0LI70 BP (foram.)

3,530 +70 (foram.)

3K109-5 4,700+200 (foram.)5,320t L00 (foram.)7,750 t 110 (foram.)

7,880t 160 (foram.)2K540-4-4 sand 19,190t 390 (foram.)

620 m deep.

2K540-4-I sand'j,9,7 80+ 330 (foram.)

LAND 27,240+450 (foram.)2K540-r-3 30,960!620 (shell)

600 m deep

2K540-L-5 32,610+700 (shell)

LAND ?

DT30-1a 240,000 (ESR), (shell)1.000 m deep

450,000 - 400,000 (ESR) (shell)'1,,200 m deep

limestone sample was dated with Election Spin Resonance (ESR) methods, providing an age of 0.24Ma or younger. Based on these dates, it was infered that this area was subaerially eroded after 0.24Ma at least since 0.2 Ma. From the summit of the Ryukyu Ridge in the Kerama Gap (Kerama Saddle),calcareous sediment (Kerama Limestone) of about 40,000-30,000 years old was collected. TheKerama Limestone covers the Naha Limestone unconformably, suggesting that this level was dry landfrom 0'2 - 0.04 Ma. Normal sandy sediments covered the Kerama Limestone unconformably. Samplesof this sandstone were collected abundantly from a depth between 500 - 600 m in the Kerama Gapusing "SHINKAI 2fi)0". The sandstone coats unconformably the erosional bedrock broadly. The 14C

age yielded an accelerator mass spectrometry (AMS) date of about 25,000 years of normal sedimentscoating Ryukyu Limestone cropped out on the flat plane TA (Fig. 7 andtable2,s.

on this account, the level TA is estimated to have been fundamentally dry land sometime between0.2 - o.o3 Ma. Additionally, a mammalian fossil bone tooth was recovered by ',sHINKAI 2(x)0,, onthe flat surface of TA in the Kerama Gap at 620 m deep; the bone / tooth has been identified as belongto Proboscider (or fossil elephants). Its age could not be dated because of lack of collagen, but isbelieved to date to younger than 0.2 Ma, probably 40,000 - 30,000 years ago. This suggests that theland bridge continuing to the continent existed sometime then as well, because a large-bodiedmammal could have arrived only by crossing a land bridge (Kimura, 1996 b, c).

From Peifu Channel of the Taiwan Straits, abundant fossil land mammals, namely elephants,deers, and cows were recovered from the bottom of the sea at depths of 120 - 140 m deep at thecentral part of the strait. rac measurement of samples yielded ages of 19,9g0 t 200 - 15,940 + 200Yrbp. Such values are close to the age of the Wiirm glacial maximum, and there are fossils that may

L6 M. KINAURA

l)1.6.1.3Ma 2) 1.3-l.O Ma

3)1.G0.2 llr

@l hnd E wrter

Fig.S.PaleogeographicmapoftheEastChinaseaandtheRyukyuArcrcgion.

be older (Nakamura et al., 1996). On the others, a lot of different mammalian fossils considered to

have come from the continent to Nansei-Shoto Islands in the late Pleistocene have been excavated on

the land (Oshiro & Nohara, 1990; Hasegawa' 1980)'

Exposure of the basement rock Shimanto Tenane dating to 100 Ma, was also identified by diving

investigations of ,'Dolphin 3K" in the Tokara Gap to the north. Subsequently, an extensive outcrop of

Ryukyu Limestone was discovered by the diving investigation of "SHINKAI 20(X)u in the Tokara

Gap. Ryukyu Limestone was recovered from practically the deepest part of the Gap near 1,200 m'

4) 0.2-0.0'lMa


0.04 - 0.02 Ma125'E

Fig. 9. Paleogeography of the

1 30'

30' N

ffi tand [-]Ryukyu Arc region during 0.04 - 0.02

Age of this limestone is dated to 0.45 - 0.4 Ma, by ESR (Kimura, L997).Thesurface of rhis limestonebed is eroded and the surface is correlatable with TA on the Ryukyu Ridge. It is unconformablycovered with normal, sandy sediment that is younger than 0.03 Ma (table 2). It is possible that therewas subaerial erosion sometime between 0.4 - 0.03 Ma.

Tbnghai Continental Shelf, deeper than 100 m, was under erosional conditions during the Wiirmglacial age, based on analysis of sonic survey records. The present continental shelf, shallower than130 m suffered from erosion during the Wiirm (Nasu, 1990), although it is not known whetherseawater chenges contributed to the formation of the cunont continental shelf then or before theWiirm. Fossil deer were recovered from Kuchimino - Se Bank at 122 m deep in the TunghaiContinental Shelf, providing a 14C date (from gelatin collagen of bone) of 26,fiD years old. The

water

Ma.

18 M. KIVTURA

I'i-

-50

Fig. 10. Cross section of Ginama Submarine Stalactite Cave at the northern'mostpart of okinawa jima (after Kimura , 1997). Location marked by star in Fig. 1.

reliability of this value is high at present (Nakamura et al., 1996). Here, Pleistocene strata outcrop at

the surface of the sea floor (judging from the sparker profiling records.

Subsidence of the last Ryukyu land bridge began when seawater invaded it about 25,000 years

ago, judging by the age of normal sediments covering the basement rocks of the Kerama Gap, dated

lac to 19,780 + 330 Yrbp.

Much of the above-mentioned evidence shows that islands were nearly connected from the

Chinese continent through Taiwan, Okinawa, and Amami Islands by the land early stage since 0.2 Ma

(Riss glacial age). And then, the bridge was interrupted by water at least 0.04 Ma, in Tokara and

Kerama Gaps (Figs. 8 and 9).

Recently, submarine stalactite caves were discovered at about 16 m below surface (Fig. 10)

(Kimura & Nitta, 1996). Two pieces of stone tools were recognized in a submarine cave and two more

stalactite caves were found below sea level in the Ryukyu Arc (Kimura et al., 1996\. Artificial ruins

(Fig. 11) (Kimura, 1997, Kimura, 2000) were later found about 20 - 30 m below sea level, which are

estimated to have been constructed much older than 6,0fi) years ago. If these have been submerged

since 9,000 years ago, then average speed of subsidence is almost coincident with the rate of eustatic

movement since L0,000 years ago near the islands'

The major tectonic force for uplifting and subsiding movements since 0.2 Ma may have been

provided by the subducting plate motion in the Nansei-Shoto (Ryukyu) Trench, rifting movement of

the Okinawa Trough back-arc basin and eustatic movement.

Statactite Pillar


monument ,/

-

970327-1-3 (YNG-1) 970

970321 -1 -1 -1

970321 -1-1-2970322.1.4-470322.1-7.s

a?

970327-t-6-2 (YNG-1)

Fig. 11. Westernhalf part of the Iseki Point, probable, ruins in the southern part offYonaguni Island, Okinawa. Dated samples recovered from the constructions arenumbered (Kimura et aL.,L999).

RECONSTRUCTION OF PALEOGEOGRAPHY

Reconstruction of paleogeography since the early Pleistocene was performed and reconstruction ofrivers and water channels were attempted (Figs. 8 and 9).

7.0 'l-6 Ma (Shimqiiri Sea stage) The relative sea level was increased and the Shimajiri Groupdeposited in the sea. This sea is called Shimajiri Sea (Kizaki & Oshiro, 1977;Kizaki, 197g). r:ndarea of the Ryukyu Arc was probably very limited from Yaku-Shima to Amami region in the northand Ishigaki to lriomote-jima to the south.

1.6 ' 13 Ma (Ryukyu paleoland stage) In early Pleistocene times, most of East China Sea driedup yielding vast dry land (Fig. 8). A series of lakes appeared in the future-Okinawa Trough region.One big lake existed at Kuchinotsu in Kyushu. Layer 81 may have deposited in those lakes. Iatersubsidence of the newer okinawa Trough started along the series of lakes.

Rivers originating from the Asian continent (many from the Himalaya Mountains) would havetrended from northwest to southeast, opening to the southeastern coast. Goto Submarine Canyon mayhave been a part of Paleo-Yellow River (Kimura et al., !975), and may have continued to the pacificcoast through the Tokara Gap. The Yangtze River may have opened at the Kerama Gap to the south ofOkinawajima Island.

The unconformity at the base of the Chinen Sandstone Formation (Fm) in the south of Okinawa-jima shows a major tectonic event of this time.

No. 1

970327-3-2

I970327.3.1 |

s21-1-7 ,j, J970720-l970720-1(YNG.3)\

\

Ls-1 ^)

-2-1-2-2

\oo

970329-1-t (YNG-2)

o-O \H, road?

M. KTVTURA

1.3 - 1.0 Ma (Okinawa Tlough stage) At this time water level rose relatively and the sea spread

out. In Kerama Gap, the Chinen Fm of the Shimajiri Group was eroded, which may have been

emerged and under erosional conditions between 1.3 - 1.0 Ma (post-Chinen Fm). The Ryukyu Arc

seems to have been a land bridge connecting China to Amami-Oshima.

It is clear that there is definite lack of strata during and after formation of the Shimanto Terrane

and before that of the Ryukyu Limestone in the Tokara Gap. However, there is outcrops of younger

sediment probably correlated with Kuchinotsu Group in Shimabara, Kyushu below Holocene sandy

sediment. It is assumed that the Tokara Gap was inundated from 1.3 - 1.0 Ma, because marine

deposits distributes in Shimabara which is western inner area to the gap. Seawater invaded Okinawa

Trough at this time. Invasion of seawater toward the inside the Trough is recognized as marine

sediment in the Kuchinotsu Group in central Kyushu. The Kita-Arima Fm of the Kuchinotsu Group

seems to coincide with that of actual transgression time. Seawater invaded the Inland Sea (Setonaikai

Sea) followed by deposition of the Kita-Arima Fm (Okaguchi & Otsuka, 1980). Thus, the Tokara Gap

would have been opened as represented in Fig. 8 - 2)) and it is estimated that it reached the Pacific

through Tokara Strait (Otsuka, 1988). Figure 8 shows distribution of land and water areas.

Subsequently, mud and sand from the continent have been trapped in the trough, followed by

growth of corals in the clear sea along the Nansei-Shoto Islands. These corals formed the lowest part

of Ryukyu Limestone (Reddish limestone) in this stage.

1.0 - 0.2 Ma (Ryulryu Coral Sea Stage) Between 1.0 - 0.2 Ma sea-level rose tremendously, during

which sea area increased. Most land area had been submerged by water (Fig. 8 - 3))' Because the sea

nearly covered the present Ryukyu Islands, land mammals could not reach Ryukyu and Honshu from

the continent at this time. High points were still land areas and corals grew along the shallow water.

Coral reefs were formed broadly along the summit of the Ryukyu Ridge as a result of sea-level

rising; hence the Ryukyu Limestone was formed. However, Ryukyu Limestone is never recognized in

the depths of Kerama and Tokara Gaps, where older strata such as the Shimajiri Group (Kerama Gap)

and the Shimanto Super Group (Tokara Gap) are exposed. Perhaps there were water channels and

under water erosional environments, in which some coarse sediment deposited, preventing formation

of Ryukyu Limestone parts of Kerama and Tokara Gaps. Giintz-Mindel Interglacial period is

represented in this stage. Nonetheless, formation of limestone was completed in the margins of the

Okinawa Trough and peripheral seas.

At a Nansei-Shoto archipelago, an unconformity exists between Pleistocene Naha Limestone and

yomitan Limestone, which possibly correlates with Mindel glacial epoch' Fossil crabs and oysters are

found in a part of the Sonan Fm on a plateau of approximately 40 m above sea level in Sonan and

Kisembaru in central Okinawa. Ages of 520,fi)0 - 300,000 Yrbp were measured using shell fragments

by ESR teclinique (Kimura, 1997). This geologic sequence is identified as transgression sediment

overlying above mentioned unconformity. It might have been a land bridge from the continent to

Okinawa Islands. However, evidence, is a few. Then, the coral grew up in the sea of those days and

formed the Limestone in Mindel glacial epoch.

Sediments were deposited in the sea after the Mindel ice age when it is thought that the land

bridge was cut off by the rising the sea-levels.

Paleogeography of the Ryukyu Islands 2L

0.2 - 0.04 Ma (Final land bridge stage) In the Ryukyu island chain, a large upheaval occurredbetween 0.2 - 0.O2 Ma (Fig. 8). A land bridge appeared after approximately O.2Ma. Strata older than

the Ryukyu Group were eroded and exposed, forming a land bridge. This land bridge was formedbetween the Riss glacial epoch and Wiirm glacial epoch.

Nearly the entire Ryukyu Ridge emerged, increasing land areas. Basement layers of the Ryukyulimestone crop out directly in the central part of the Kerama and Tokara Gaps (or Saddles) (Fig. 7).This suggests that the center of the Gaps may have uplifted and had eroded more than neighboringareas in the Ryukyu Ridge after deposition of the Ryukyu Limestone formations. As a result, a landbridge from Chinese continent to Ryukyu Arc had been formed just since 0.2 Ma. After then, theTokara Gap has been subsided and then the Kerama Gap has been subsided (Fig. S - a)) by normalfaulting movement.

0.04 Ma ' present The Kerama calcareous layer (Kerama Limestone) was deposited 40,000-30'fin years ago. This suggests that at least a part of the land bridge may have been disconnectedfrom the continent. The Kerama Limestone was eroded unconformably because of sea level loweringby the eustatic movement after then. The land bridge to the Ryukyu Arc may have been appear,allowed Minatogawa Man access to the Ryukyus. The flat plane of TA has been offset by normalfaulting since 25,fi)0 Yrbp, during which time the last land bridge began to be inundated, owing tocrustal movement. Fig. 9 shows a paleogeographic map between 25,000 - 20,000 Yrbp. At this time,shallower part than about present 500 m deep may have land areas, because sediments of this agewere recovered from the place deeper from the place than 500 m. Then, relative subsidence hasprogressed by sea level rising since ultimate the ice age. The Kuroshio Cunent began to flow throughwest of the Nansei-Shoto archipelago, finally to result in the state.

Recently, architectural ruins were identified under water of.20 - 30 m deep in Nansei-shoto Islands(Kimura, 2000) (Fig' 11). This may have submerged since 9,fi)0years ago, showing coincident witnpost-glacial eustatic movement.

SUMMARY AIYD CONCLUSION

1) Paleogeographic maps were made to recreate paleoenvironments of the Ryukyu Arc based onreceut submarine, geological, and geophysical investigations.2) Major land bridges from the Asian continent (China) to the Ryukyu Islands have appeared at leastonce each during 1.6 - 1.0 Ma and just after 0.2 Ma.3) Many 14C ages achieved from fossil mammal bones suggest strongly the existence of the ultimateland bridge from the continotrt to Amami Islands may have existed until the Wtirm ice age.4) The last land bridge has been submerged since 0.025 Ma, according to sea level rising after the iceage and crustal movements, averaging L - 2 cm/yr.

ACKNOWLEDGMENTS I thank the operative scientists and crew of "SHINKAI 2000" and

"Dolphin 3Ku of JAMSTEC, as well as the crew of Nagasaki Maru. Comments by Noriyuki Nasuwere very useful for determining, paleoland. Hiroyuki Otsuka and Itsuro Oshiro provided usefulinstruction for fossil mammals to me. Corals submitted to 1aC dating were identified by Masashi

22 M. KTUURA

Yamaguchi and Robert Van Woesik. Additionally, I thank Hiroshi Ujii6, Toshio Nakamura, Koshiro

Kizaki, and Motdi Ikeya for age determination and paleoenvironmental discussions.

REFERENCES

Halback, P., Nakamura, K., Wahsner, M., Iange, J., Sakai, H., Kiielitz, L', Hansen, R., Yamano, M.,

Post, J., Prause, B., Seifert, R., Michaelis, W., Teichmanm, M., Kinoshita, M., Miiten, A',

Ishibashi, J., Crerwinski, S. & Blum, M., 1989. Probable modern analog of Kuroko{ype massive

sulfide deposits in the Okinawa Trough back-arc basin' Nature 338: 496-499.

Hasegawa, Y., 1980. Notes on vertebrate fossils from late Pleistocene to Holocene of Ryukyu Islands,

Japan. Quat. Res. lE (4): 263-267 .

Hirata, N., Kinoshita, H,, Katao, H., Baba, H., Kaiho, Y., Koresawa, S', Ono, Y. & Hayashi, K', 1'991'

Report on DELP 1988 cruises in the Okinawa Trough, Part 3. Crustal structure of the southern

Okinawa Trou gh. Bull. Earthquake Res. Inst' 66:. 37 -7O'

IwaSaki, T., Shimamura, H., Melles, J., Makris, J., Hirata, N., Kanazawa, T., Urabe, T, & Suyehiro,

K., 1987. Crust and upper mantle structure in the Ryukyu Island Arc deduced from DSS-

Abstracts of the l%h General Assembly of IUGG no. t: 326'

Japanese DELP Research Group on Back-Arc Basins (JDRGBB), 1991. Report on DELP L988

cruises in the Okinawa Trough Part 1.. General outline. BuIl. Earthquake Res. Inst,66: l-15.

Kamata, H. & Kodama ,K., !994. Tectonics of an arc-arc junction: an example from Kyushu Island at

the junction of the Southwest Japan Arc and Ryukyu Arc. Tectonoplrysics 233: 69-81.

Kimura, M., 1985a. Seismicity and Crustal Movement: NulEu Arc and Japanese Island, 195 pp.

Kyushu Univ. Press, Fukuoka. (in Japanese)

-- ,1985b. Back-arc rifting in the Okinawa Trough' Mar' Pet' Geol' 2:222'240.

,1991. An origin of backarc basin'. Ertra Chilcyu 3: 7O-74. (in Japanese)

--,1996a. Active rift system in the Okinawa Trough and its northeastern continuation. Ball.

Disaster Preu Res Inst. 45:27-38.

--,1996b. Quaternary paleogeography of the Ryukyu Arc. J. Geogr. LOS (3):259'285. (in

Japanese with English abstract)

--- ,1996c. Paleogeography of the Ryukyu Arc. ChilEu Monthly 18 (8): 488-494' (in Japanese)

lgg7. A Continent Lost in the Pacific Ocean-Riddle of the Submarine ktins in the Ryulryu

Island,28l pp. Daisan-Bunmei Co, Tokyo. (in Japanese)

-2000.

Formation process of the Iseki Point (Submarine ruins ?) in Yonaguni Islands, Okinawa.

- Is it natural or artificial ? -' Chilqu Monthly 22 (2): L01.'1L1. (in Japanese)

---, Hiroshima, T. & Inoue, 8., L975. Geological structure of East China Sea. Ifuiyokagaku 7 (1):

45-52. (in Japanese with English abstract)

--, Kaneoka, I;, Kato, Y., Yamamoto, S., Kushiro, I., Tokuyama, H., Kinoshita, H., Isezaki, N.,

Masaki, H., oshida, A., uyeda, S. & Hilde, T. w. c., 1986. Report on DELF 1984 cruises in the

middle Okinawa Trough, Part 5. Topography and geology of the central graben and their vicinity.

Bull. Earthquake Res. Inst. 6l:. 269'3LO'

---, Uyeda, S,, Kato, Y., Tanaka, T', Yamano, M., Gamo, T', Sakai, H., Kato, S', Izawa, E' &

Oomori, T., 1988. Active hydrothermal mounds in the Okinawa Trough back-arc basin Japan.


Tbctonophy sics 45: 319 -324.

--, Tanaka, T., Kyo, M., Ando, M., Oomori, T., lzawa, E, & Yoshikawa, I., 1989. Study oftopography, hydrothermal deposits and animal colonies in the middle Okinawa Trough

hydrothermal areas using the submersible "SHINKAI 2(X)0" system. JAMSTECTR Deep-Sea Res.

ll4:225-233. (in Japanese with English abstract)

------ , Furukawa, M., Izawa, E., Ishikawa, M., Kuramoto, S., Sakai, H., Uyeda S. & Makris' J., 199L.

Report on DELP 1988 cruises in the Okinawa Trough, Part 7. Geologic investigation of the central

rift in the middle to southern Okinawa Trough. Bull. Earthquake. Res. Inst.66: L79-209.

-------- Matsumoto, T., Nakamura T., Otsuka, H., Nishida, S., Aoki, M., Ono, T., & Danno, K,. 1992'

Diving survey of the Kerama Saddle in the eastern margin of the Okinawa Trough: Probable

sunken land-bridge during the last glacial age. Proc. JAMSTEC Symp. Deep Sea Res. 8: 107-133.

(in Japanese with English abstract)

, Ono, T., Matsuoka, H., Otsuka, H., Nakamura, & T., llliyazaki, Y., 1994. Tectonics in the

eastern margin of the OkiuawaTrotgh.JAMSTEC J. Deep Sea Res.L0:299-32L.

-

& Nitta, K., 1996. Submarine stalactite discovered in Okinawa Island. Chilcyu Monthly 18:.

544-554. (in Japanese)

- , Kato, S., Nitta, K., Bartoll, J., Ikey4 M., Doi, N. & Nakamura, T., L996. Submarine stalactite cave

discovered in Okinawa and its sipificance. Lecture in Soc. Southern Island Hist. (oral) (in Japanese)

---, Nakamura, T., Motoyama, L, Nakamura, M., Ono, T., Oomori, T. & Tanahara, A., 1999.

Research and study of the age of submarine ruins at Yonaguni Island in Okinawa, Japan;

Summaries of Research AMS at Nagoya University (10) Dating and Materials Research Center

Nagoya Univ. pp. 49-166.

Kizaki, K., 1978. Tectonics of the Ryukyu Island Arc. J. Phys. Earth 26: Suppl., s 301-5307.

Oshiro, 1., 1977 . Paleogeography of the Ryukyu Islands. Kniyokagahr 9: 38-45.

Kodama, Ozawa, T., Inoue, K., Maeda, Y. & Takeuchi, T., 1991. Paleo magnetism and post-middle

Miocene counter-clockwise motion of Tanegashima Island off Kyushu, Japan. J. Geomag.

G eoe lectr. 43: 7 2l -7 40.

I*tonzey, J. & Kimura, M., 1986. The Okinawa Trough: Genesis of a back-arc basin developing along

a continental margin. Tbctonophysics 125: 209-230.

Marutani, H. & Sato, T., 1985. Basic test drilling Miyako jima. OkL Bull. Petroleum Technol. Assoc.

50 (1): 25-33.

Miki, M., 1991.. Formation model of the Okinawa Trough on the basis of paleomagnetic evidence.

Chtlryu Monthly no.3:2L2-216. (in Japanese)

--, Matsuda, T. & Otofuji, Y., 1990. Opening mode of the Okinawa Trough: Paleomagnetic

evidence from the South Ryukyu Arc. Tbctonophysics 175: 335-347.-------- Furukawa, M., Otofuji, Y., Tsao, S. & Huang, T., 1993. Palaeomagnetism and K-Ar ages of

Neogene rocks of northern Taiwan: tectonics of the arc junction of Ryukyu and Luzon arcs.

Geophys. J. Int. Ll4:.225-233.

Nakamura, K., Marumo, K. & Aoki, M., 1990. Discovery of a black smoker vent and a pockmark

emittinB CO2-rich fluid on the sea-floor hydrothermal mineralization field at the Izena Cauldron

in the OkinawaTrougfi.JAMSTECTR Deep-Sea Res 6t33-50.---, Otsuka, H., Okuno, M. & Ohta, T., 1996. Accelerator mass spectrometric 14C dating of

M. KttvtuRA

mammalian fossils collected from the sea bottom of continental shelves around the East China Sea

and along the Ryukyu Island Arc. J. Geogr.105 (3): 3O6-3L6.

Nasu, N., 1990., Continental shelves on the basins of geologSr. Ohbayashi 32: 18-27 . (in Japanese)

Okaguchi, M. & Otsuka, H., 1980. Fission track ages of zircon crystals from tuff layers in the

Kuchinotsu Group and from andesite of the Tatsuishi Formation. The Quaternary Res.19 (2):75-85. (in Japanese with English abstract)

Oshiro, I. & Nohara, T., 1990. Fossil mammals. In: Ujiie, H., ed., Nature of Okinawa: Topograplry

and Geology,pp.213-229. Hirugi-Sha, Naha. (in Japanese)

Otsuka, H., 1988. Pleistocene vertebrate fauna in the Ryukyu Islands: Its dispenal and extinction. 1z:

Otsuka, H., ed., Programme and Abstract of an International Symposium: the Ryulqu Islands- The

Arena of Ad.aptive Radiation of Island Fauno,pp.1l-14. Kagoshima University, Kagoshima.

Sibuet, J-C., Irtouzey, J., Barbier, F., Charvet, J., Foucher, J-P., Hilde, T. W. C., Kimura, M., Ling-Yun, C., Marsset, B., Muller, C. & Stephan, J.-F., 1.987. Back-arc extension in the OkinawaTrough.J. Geophys. Res.92 (14): a1-063.

Sibuet, J-C., Deffontaines, B., Hsu, S.-K., Thareau, N., Ir Formal, J.-P., Liu, C.-S. & ACT party,.

1998. Okinawa Trough Backarc basin: tectonic and magmatic evolution. J. Geophys. Res. 103

(30):245-267.

"SHINKAI 2(X)0" Research Group on the Okinawa Trough, 1986. Submersible "SHINKAI 2000"

study on hydrothermal activity in the middle Okinawa Trough: A preliminary rcport. Abstracts ofInternational K4IKO Conference on Subduaion Zones, pp. 144-L45.

Af.Jffiffi Dftf;f;IfiOE&4[

iE+ 0)H^#NA aA Ui6E o ru H. . frr*Mrw+A+Jffinbr#t X L D 5 : e 13 & D, th$fl [*i ] r]X v tffiawRfrH. awe A Brl 6 r\ t3 L tc " ffi*ffi ,a MEET&ET llz: a 7 7 - t rtr#:fiWre b 6 . ffiL 77 -' l"lt, H v J-ffi H ffi&affiffia#lt t Z nrc{+ i ffi t*d[ @ iFtr,: a#gtt]r, F ffi tE&ffiil 6f Sffi tr + tr 0) ffrt34Drc. za'tk, rcDn - 130 7?.+ifirc, Fvtffi&nr,rffit=wl?t /c" .nlc.Ij.&Eai+*19F.770)r'tt D ttfiBil Efrtl. .? o)ffiffi stfi IJ.r- ) > 7

'(w.a -ffiT t> c tc" ? a'r*., Fffi E'ffiffi a+r\ B J5 & t

130 i?'4Btf3, H277-iff#,4.1,tc. zl,Tfrlraf,ffi1fr, (frB:ffirvtfiIcltffixtFfl rttvfr,tt1tc."ffi*lfiIizc a*/effi H rc, 6i4 & iE : r F trtE e Effi & L ( ril*Igffi L T tt tp 2 4 U n 6 (te t Ldr60n -100 7:'+ffie ?nn -2.sfr+ffi)" coth*sEltis&t2.5 E+friuw, tufuagtLlLxtrofiEAfu affiA rc t z 7 t&WffiI=ftzk L ( L t : lc. trr#ila *Y*. 10 - 35 m OEE l3 ltiF&offi fl "ffi nrff F. E n/c" zft b affi^IJ, )v }.xffil\wl=ttKl,tc6 o)Tb 6" z a i b a 1 ?i) 6 ltE*siw 5 i'Llc" E 6lc 6,000 4& D fi u rFff l=Mfr,8 tttc L#E d i't6 H& U 7 i v F a t 5 ttYt& L lc Arfi leiEffi itSn[El E# o FE r\ B #-H, E *tlc " ? 0l I i teiEEF o)Fft lrtft Sd[ aWrFMEAffifr f 5 *itit D A + L 6 +,

otts.6a"

Paleogeography of the Ryulryu Islands

Documents