tV W V U A TY O f HAW AII LIBRAKÏ

SE D IM E N TA R Y P A L E O M A G N E T IC EVIDENCE

FOR CO U NTER-CLO CKW ISE R O TA T IO N

OF THE F IJ I P L A T E A U

A THESIS SU BM ITTED TO THE GRADUATE DIVISION OF THE U N IV E R S ITY OF HAW AH IN P A R T IA L F U L F IL L M E N T

OF THE REQ UIREM ENTS FOR THE DEGREE OF

M ASTE R OF SCIENCE

IN O C EAN O G RAPH Y

M A Y 1971

By

Thomas Chapman G ill ia rd

Thesis Committee

A. Malahoff, Chairman J. E. Andrews

A. S. Furumoto

i i

T O f H AW AII

We c e r t i fy that we have read this thesis and that in our

opinion it is sa tis factory in scope and quality as a thesis fo r the

degree o f M aster o f Science in Oceanography.

t'8RAiir

i i i

T A B L E OF CO NTENTS

L IST OF IL L U S T R A T IO N S ............................................................... iv

I N T R O D U C T I O N .............................................................................. 1

TECHNIQUES AND P R O C E D U R E S ............................................... 6

RESULTS, DISCUSSION AND CONCLUSIONS . . . . 12

A P P E N D I X ...................................................................................... 26

L IT E R A T U R E C I T E D ............................................................... 27

LIST OF ILLU S TR A T IO N S

F igu re Page

1. Summary of the Batym etric and Tectonic Features of the F i j iP lateau ...................................................................... 2

2. S tereo Net of Declination and Inclination Values for A. C.Demagnetization to 150 O e ..................................9

3. Changes in Intensity Values onA . C. Demagnetization to 150 Oe . . . . 10

4. P lo ts o f A l l Data for CoresPC 20 and PC 2 1 ..........................................................13

5. P lo ts o f A l l Data for CoresPC 22 and PC 2 3 .......................................................... 14

6. Sedimentation Rate P lo tsfor A l l Four C o r e s .................................................. 15

7. R e la t ive Rotation Versus AgeCores: PC 20 and PC 2 1 ..........................................17

8. R e la t ive Rotation Versus AgeCores: PC 22 and PC 2 3 ......................................... 18

9. Changes in Declination in a Core on a Crustal B lock RotatingC o u n t e r - c l o c k w i s e ........................................... . 20

10. Summary of Re la tive Rotation VersusA ge for A l l Four C o r e s ..................................... 21

11. R e la t ive Rotation o f Cores P lotted on M alahoff ' s Summaries o f the Bathymetry and Magnetic Featureso f the A r e a . . . . . . . . . 23

iv

I. INTRO D U CTIO N

This thesis reports the results o f a detailed sedimentary paleo-

magnetic investigation o f four c lo se ly spaced 7 to 11 m eter piston

co res co llec ted on the F i j i P lateau in the W estern P ac if ic in June,

1970. This plateau, which surrounds the F i j i Islands, has recen tly

been proposed as an example o f a crustal block which has undergone

westward bending and counter-c lockw ise rotation o f up to 180 degrees

since the ea r ly T e r t ia ry (M a lah o ff 1971). This interpretation is based

on severa l lines o f geophysical evidence that are summarized in

f igu re 1. This f igu re is adapted from M alahoff(1 971).

In this figure dashed lines represent bathymetric lows; solid lines bathymetric highs; solid a rrows indicate d irection of crustal movement on the F i j i Plateau. Hollow arrows along the Tonga Trench suggest d irections o f oceanic crustal plate m ovement into the trench. Chequred arrows north and south o f the F i j i P lateau indicate directions o f oceanic crust movements deduced by . . . Malahoff. . . f rom sea floor spread­ing evidence. Dots indicate location of earthquake fo c i i for shallow earthquakes(0-70 Km), from 1961 to 1969. . .

See M alahoff(1 971) fo r a complete discussion o f this figure and the

geophysical evidence fo r rotation o f the F i j i Plateau.

The four cores studied in this thesis w ere co llec ted from the

portion of the F i j i P lateau indicated by the la rge c ir c le in figure 1.

This area o f the plateau has undergone a la rge amount o f counter-

1

PC*

IT’S*

i8T

2

FIGURE 1

c lockw ise motion i f M alahoff ' s interpretation is correc t .

The sed im entary paleomagnetic measurements reported here are

important for they provide an independent means with which to test

this tectonic interpretation. It was expected that the proposed

counter-c lockw ise rotation o f the F i j i Plateau would be observable

as a change in declination of the paleomagnetic vector with depth in

the cores. Rates of observed rotation can be calculated from the

depths at which re v e rsa ls o f inclination(due to re ve rsa ls of the

po la r ity of the earth 's geomagnetic poles of known age) are encountered

in the sediments.

This thesis is also important for it is one of the f irs t times that

variations in the declination vec tor o f m arine cores have been used to

deliniate the tectonic h is tory of an oceanic crustal plate. Recent

studies of sedim entary paleom agnetism have concentrated mainly on

the determination of depths of po lar ity re v e rsa ls as an independent

t im e stratigraphic tool for dating cores. Recent work in low latitudes

has used changes in declination in in ternally oriented cores as a means

o f distinguishing these po lar ity changes(Hays and others 1969, Foster

and Opdyke 1970).

The use o f paleomagnetic methods in a study of this kind is only

possib le because o f rapid developments in this f ie ld in the last decade.

The deliniation o f an accurate and genera lly accepted chronology for

the re v e rsa ls in po la r ity of the earth 's magnetic pole from potassium

3

argon dating o f volcanic rocks (Cox 1969); the extension of this

chronology by studies o f the linear magnetic anomalies on the ocean

f loo r (P itm an and H e ir t z le r 1966, H e ir tz le r and others 1968, Talwani

and others 1971); and the applications of this chronology to sediments

on the ocean floor(Opdyke and others 1966, Hays and others 1969,

F os te r and Opdyke 1970) have been some of the m ore important of

these advances.

Development o f a low speed spinner magnetometer using flux-

gate probes (F os te r 1966) made possible the routine measurement of

the weak paleom agnetism of m arine sediments. P r io r to the in tro ­

duction of these instruments, such an application of paleomagnetism

was exceedingly d ifficu lt because o f the cumbersome and delicate

astatic m agnetom eters then in use (Blackett 1952).

The com para tive ly recent increase in the number and prec is ion

o f paleomagnetic measurements fr o m all areas of the w or ld combined

with the acceptance o f the concepts of continental drift and plate

tectonics have increased the confidence with which paleomagnetic

results have been accepted.

The rem ainder of this thesis is divided into two main sections.

The f i r s t section describes in detail the techniques and procedures

fo llowed in the analysis of the cores . The final section presents the

experim ental results and then discusses the conclusions and im p li­

cations of these results. An appendix is included at the end which

4

5

descr ibes each o f the four cores in greater detail.

II. TECHNIQUES AND PROCEDURES

C O LLE C T IO N OF CORES

Four piston cores w ere co llected on the F i j i Plateau north-

north west of the island o f V it i Levu in June 1970. Pertinent data

fo r the cores is l is ted in the Appendix. A l l of the cores w ere

co llec ted with a piston c o re r using six m eter plastic core l iners.

These six m eter p lastic lin ers w ere p re -m arked with two vert ica l

lines 66° apart which ran down the entire length of the liner. Junctions

between liners w ere alligned by these marks p r io r to coring. These

co res w ere not alligned with respect to the present magnetic f ie ld but

internal allignment within a single six m eter l iner should be accurate

to severa l degrees. Upon re tr ie va l the liners w ere cut into 1. 5 m eter

lengths, capped, and stored under re fr igera tion .

S A M P L IN G AND LOGGING

At the labora tory the liners w ere trans fe rred from a portable

re fr ig e ra ted core locker and split into two equal halves along one of

the v e r t ic a l marks on the liner. The liner half without the remaining

allignment m ark was then sampled for paleomagnetics with 2 cm.

cubic p lastic boxes at 10 cm. in terva ls down the center of the split

core . The other l iner half was logged fo llow ing standard procedures

o f the institute (Andrews 1970) and sealed in a plastic "D " tube. This

sample is stored under re fr ig e ra t ion as an arch ive sample.

6

When analyzed, a ll four cores show an abrupt 66° change in

declination 4, 1. 5 m eter sections up from the bottom. This distance

corresponds to the junction between plastic core liners in the piston

coring equipment and is attributed to m isallignment of core liners

during shipboard operations. In the fo llow ing discussions the top

sections o f core are recalcu lated to rem ove this offset.

Hays and others (1969) and Fos te r and Opdyke (1970) attribute

changes they observed in the declinations o f three Equatorial

P a c i f ic piston cores to twisting o f the coring apparatus. They also

encountered unexplained shifts in declination in the cores which they

suggest are due to coring or handling operations. The cores they

w e re examining w e re co llected with unlined piston co re rs by Lamont-

Doherty Geophysical Labora tor ies . The procedures used to allign

the various sections o f their cores are to cut a groove down the

length o f the unlined co re as it is extruded on board ship and then

to lay a string into this groove. The unlined core is la ter split

along this string. In contrast, the procedures fo llowed in this

labora tory are based on prem arked plastic liners 6 m eters long.

This system entails much less opportunity for gradual rotation to

be a r t i f ic ia l ly engendered by n ecessa r i ly hurried operations aboard

ship with soft m arine sediments. The only poss ib il ity o f an e r ro r

in handling these sections occurs in the allignment between liners

at 6 m eter in tervals .

7

A N A L Y S IS --5 H Z SPINNER M A G N E TO M E TE R

A l l of the paleomagnetic measurements w ere made within three

days of sampling on P e rm a l i 5HZ Spinner Magnetometers using a

pair of an ti-para lle l Schonstet flux-gate probes as sensors. This

equipment has been described in detail by Hammond (1970).

Tests for Magnetic Stability

P r io r to analysis, severa l random samples from each core

w e re tested for magnetic stability by stepwise A. C. Demagnetization

to 150 Oersteds on a P e rm a l i Ax is A. C. Washer. P lo ts of the changes

in declination and inclination of these samples are shown as stero-

plots in figure 2. The intensity of magnetization for these samples

are shown in figure 3. These figures revea l that the magnetic

m ater ia ls in these cores fa ll into two c lasses; in cores 20, 21, and

22, samples with an intensity of magnetization less than about

51 x 1 0 at 50 Oe A. C. demagnetization are much m ore unstable than

samples with g rea ter intensities. This instability is apparent in the

much la rg e r changes in declination and inclination which the lower

intensity samples showed. The stronger c lass of samples show

rem arkab ly constant magnetic vec to rs , and are not destroyed at

150 Oe. On a ll four cores , 50 Oe was chosen as the optimum le ve l

at which to clean the samples of unstable magnetic components.

Measurements that y ie lded intensities o f magnetization less than

8

9

DEC LIN ATIO N <S INCLINATIONA.C. D E M A G N E T IZ A T IO N TO 150 O e .

FIGURE 2

10

AC DEMAGNETIZATION INTENSITY

; PC 2 2150

0.8

—i—0

1 PC 2 3J 50

FIGURE 3

51 x 1 0 at 50 Oe demagnetization w ere discarded and not used in the

analyses.

Analysis

A fte r A. C. Demagnetization at 50 Oe, each sample was spun

on three perpendicular axes and values w ere obtained for the X, Y,

and Z axes. This procedure is described in detail by Hammond

(1970) and is s im ila r to the procedures used by other paleomagnetic

labora tor ies . A standard sample o f known declination, inclination,

and intensity that had been checked with other laboratories was r un

da ily on each instrument to check its perform ance. The values

obtained w ere converted to declination, inclination, and intensity

on a computer p rogram developed by Hammond (1 970).

11

in. RESULTS, DISCUSSION AND CONCLUSIONS

The values o f Declination, Inclination, and Intensity fo r the

four cores as a function o f depth are shown in figures 4 and 5

below. Two of the cores , PC 21 and PC 22, penetrate through the

re v e rsa l at . 95 m il l ion years at the end of the Jaram illo Event.

Cores PC 20 and PC 23 penetrate the re v e rsa l above this at . 89

m ill ion years which marks the beginning of the Jaram illo Event.

SE D IM E N TA T IO N R A TE S FRO M DEPTHS O F M AG NETIC

R E V E R SA LS

The depths at which reve rsa ls occurred are plotted as a

function of age from the Geomagnetic T im e Scale of Cox (1969) in

f igu re 6.

In three of these cores (20, 21, 23) a straight line can be

drawn through these points and the origin. A line drawn this way

fo r core 22 however, in tersects the depth axis in the v ic in ity of 130

to 160 cm. (dashed line in figure 5). It should be noted however,

that core 22 was the only core which was disturbed badly. The

l in er was collapsed by the suction of the piston from 180 to 340 cm.

and contained lit t le sediment. Based on this information, it has

been assumed that there is a 140 cm. break in this core starting at

180 cm. depth. This assumption allows the sedimentation curve in

f igu re 6 to be shifted 140 cm. to the le ft as shown by the arrows.

In the fo llow ing discussion a ll samples fr o m this core below 180 cm.

12

FIGU

RE

4

DEPT

H Cm

.)

LO

FIGU

RE

5

DEPT

H (m

)

FIGU

RE

6

have had 140 cm. subtracted from them to rem ove this break. Thus

this core is considered to be 974 cm. long instead o f 1, 114 cm.

The calculated rates of sedimentation based on the paleomagnetic

data o f these four co res range fro m . 72 to . 89 cm. per 1, 000 years.

Rates fo r the individual cores are listed in the appendix.

C A L C U L A T E D R E L A T IV E R O TA T IO N OF THE PA LE O M A G N E T IC

VECTORS W ITH IN THE CORES

F igures 7 and 8 a re plots o f changes o f declination re la t ive to

the 100 cm. declination within each core versus age. Counter­

c lockw ise rotation is plotted as positive. The sedimentation curves

w e re used to interpolate ages for each sample in these figures.

The 100 cm. depth was a rb it ra r i ly chosen as a ze ro re fe rence

because the top sections of these piston cores w ere disturbed. In

seve ra l of the cores this disturbance extended 70 to 80 cm. down

from the top o f the core . Because the greatest rate of change of

the declination vec to r in the cores occurs after severa l m eters of

co re with r e la t iv e ly constant declination, the selection of the depth

is not v e r y crucial.

The inclination of the present day magnetic dipole in the F i j i

P lateau is 40°S and inclination measurements are sensitive to flow

o f sediments into the co re barre l. Thus figures 7 and 8 do not

include any measurements which contained an inclination m easu re­

ment greater than 45°. Measurements which y ie lded intensities

16

FIGU

RE

7

AGE

(myJ

30°ROTATION

« » . 1 -

60 ° 90 ° ISO° 1 1150 ° I8CP

PC 22

!• V" Ii ROTATION

I ,------ . 1----- 1-----1----- •-----• I0° 30° 60 ° 90°

ROTATION

o f magnetization less than 1 x 1 0 at 50 Oe demagnetization w ere

also excluded fro m these figures for the reasons discussed

ea r l ie r . Using these c r ite r ia , a lmost a m eter and a half of

sediments in co re 22 are disturbed below the break at 180 cm.

None of the other cores a re a ffected by these c r i te r ia except in the

top few centim eters.

DISCUSSION

F igure 9 shows changes in declination that would occur in a

core on a crustal block rotating counter-clockwise. The arrows

extending from the core indicate the magnetic declination vectors

fo r that point on the core i f the top of the core is oriented para lle l

to the earth 's magnetic fie ld . The vectors describe a spiral.

A l l four cores examined in this thesis show counter-clockwise

rotation of the crustal block upon which they w ere collected. F igure

lOsummarizes the re la t iv e rotation rates of the individual cores

with respect to time.

The shapes o f the curves suggest that the rate of rotation of

this portion o f the F i j i P lateau has been irregu la r and that the

rotation was m ore rapid from about 0. 2 to 0. 5 m ill ion years ago.

An a lternative interpretation of these curves calls for a decrease in

the sedimentation rate over this period o f time. The fact that a line

can be drawn through the depths of polarity reve rsa ls in F igure 6

with a reasonably c lose fit, coupled with the la rge d ifferences in

19

5

FIGU

RE

9

rn

c\>o

FIGU

RE

10

amounts o f rotation found in core 21 over this t im e period suggest

an interpretation o f ir regu la r rotation.

The reasons for the much la rg e r rotation recorded in core 21

a re not known. Malahoff (1971) includes two figures which summarize

surveys by Halunen and others of the trends of magnetic anomalies

and bathymetry o f the area where these cores w e re collected. The

results of this thesis are plotted on these figures in figure 11. It is

in teresting that co re 21 was co llected c lose to what appears to be an

in tersection o f two magnetic trends. The la rge rotation in this core

m ay be indicative of motions of a small plate within the plateau while

the 20-50 degree rotations suggested by the other cores probably

r e f le c t a m ore general motion of the plateau.

It does not seem l ik e ly that the apparent rotation of the d ec l i ­

nation vec tor m easured here is due to a spinning piston coring

apparatus. Hays and others (1969) and Foster and Opdyke (1970) as

mentioned above attribute changes which they observed in the

declinations o f three Equatoria l P ac if ic piston cores to twisting of

the coring apparatus although as discussed prev iously they w ere

using unlined, extruded cores.

The poss ib il ity that the coring apparatus was rotating cannot

be to ta lly discounted but severa l reasons can be presented which

indicate that this is not the case. Most o f the measured rotation

occurs over a small in terva l in the m iddle o f the co re after from

22

23

FIGURE 11

one to two m eters in which the declination remains re la t iv e ly constant.

A l l four cores show a counter-c lockw ise rotation while other cores

co llec ted in other areas o f the W estern P ac if ic with the same equip­

ment and procedures have not shown consistent changes of declination

with depth. The sense and amounts of rotation are compatible with

geophysical evidence.

The results f r o m only four cores cannot be considered conclusive

p roo f of the rotation of the entire F i j i Plateau but do o ffe r strong

evidence to support the concept of at least this portion of the plateau

having undergone a discontinuous counter-c lockw ise motion in the

last m illion years. The data suggests that this rotation has been in

the v ic in ity of 20 to 50 degrees with possib ly small crustal blocks

undergoing much la rg e r motions. It would be in teresting to t r y to

co rre la te the increase in rotation rate from . 2 to . 5 m ill ion years

with other tectonic events in this highly se ism ic area. Anomalously

high rates of heat flow have been observed over this plateau (Sclater

and Menard 1967) and the structure of the sub-bottom seen with

sub-bottom p ro f i le rs shows a chaotic com plex of fault blocks and

faulted sediments.

A m ore detailed geophysical survey of the F i j i Plateau is

planned for the fa ll o f 1971. A number of additional piston cores

w i l l be co llected as a part of this program . The sedimentary

paleom agnetism o f these additional cores should provide a strong

24

tool with which to unravel the tectonic h istory of this com plex region

o f the earth 's crust.

25

A P P E N D IX

CORE NO. L A T IT U D E LONGITUDE D E PTH(m)

PC 20 14 39. 8S 177 11. 6W 2840

PC 21 14 4 1 .5S 176 58. 5W 2815

PC 22 14 37 .OS 176 34. 0W 2890

PC 23 14 4 7 .4S 177 02. 9W 2812

LE N G TH(cm )

RE VE R SALS (cm) depth

SED. R A T E cm /1, 000 yrs.

MAX. ROT. M il l ion yrs.

T O T A L ROT. D egrees

822 630 & 780 89 .4 2 - ,60 20-30

755 485, 625, 685 72 .29 - .35 140-160

974 590, 750, 830 85 .1 5 - ,4 0 50*

763 560, 710 80 . 35-, 45 20-25

*Depth values in this co re a re co rrec ted for a 140 cm. gap below 180 cm. as discussed in the text.

27

L IT E R A T U R E CITED

Andrews, J. E. "Sediment C ore Descriptions: Solomon Islands 1968- 1969, and M urray Fracture Zone, 1967." Hawaii Institute of Geophysics-70-25, Data Report No. 16, October, 1970.

Blackett, P . M. S. " A Negative Experim ent Relating to Magnetism and the Earth 's Rotation. " Philosophical Transactions of the Royal Society o f London, Series A, Vol. 245, Decem ber 16, 1952, pp. 309-370.

Cox, A. "Geom agnetic R e v e r s a ls . " Science , Vol. 163, January 17, 1969, pp. 237-245.

Foster , J. H. and N. D. Opdyke. "Upper M iocene to RecentMagnetic Stratigraphy in Deep-Sea Sediments. " Journal of Geophysical R esea rch , Vol. 75, No. 23, September, 1970, pp. 4465-4473.

Foster , J. H. " A Paleom agnetic Spinner Magnetometer Using a Flux- Gate Gradiom eter. " Earth and P lanetary Science L e t te rs , Vol. 1, No. 6, Novem ber, 1966, pp. 463-466.

Hammond, S. R. "Pa leom agnetic Investigations of Deep Borings on the Ewa Plain, Oahu, H aw a ii . " Hawaii Institute o f Geophysics- 70-10, May, 1970.

Hays, J. D. , T. Saito, N. D. Opdyke, and L. H. Burckle. " P l io c en e - P le is tocene Sediments of the Equatorial P a c i f i c - - Their Paleom agnetic , B iostratigraphic, and C lim atic Record. " Geologica l Society of A m er ica B u lle t in , Vol. 80, August, 1969, pp. 1481-1514.

H e ir tz le r , J. R. , G. O. Dickson, E. M. Herron, W. C. Pitman, and X. LePichon. "M ar in e Magnetic Anomalies, Geomagnetic F ie ld R eversa ls , and Motions o f the Ocean F loo r and Continents. " Journal of Geophysical R esea rch , Vol. 73, No. 6, March 15, 1968, pp. 2119-2136.

Malahoff, A. " F i j i P lateau T ec ton ics - -P oss ib le Rotation and Rifting o f the Lithosphere. " (In p ress )

28

Opdyke, N. D. , B. Glass, J. D. Hays, and J. Foster. "Pa leom agnetic Study o f Antarctic Deep-Sea Cores. " Science , Vol. 154,October 21, 1966, pp. 349-357.

Pitman, W. C. , and J. R. H e itz le r . "M agnetic Anom alies Over the P a c i f ic Antarctic Ridge. " Science , Vol. 154, Decem ber 2, 1966, p. 1164.

Sclater, J. G. , and H. W. Menard. "Topography and Heat F low ofthe F i j i Plateau. 11 Nature, Vol. 216, No. 5119, Decem ber 9,1967, pp. 991-993.

Talwani, M. , C. C. Windisch, and M. G. Langseth, Jr. "ReykjanesRidge Crest: A Detailed Geophysical Study. " Journal of Geophysical R esearch , Vol. 76, No. 2, January, 1971, pp. 473-517.