wise, s. w., jr., schlich, r., et al., 1992 proceedings of ...w. jr. schlich r. et al. 1992 proc....

Wise, S. W., Jr., Schlich, R., et al., 1992Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 120

11. GEOCHEMISTRY OF CENOZOIC ASH LAYERS FROM THE KERGUELENPLATEAU (LEG 120): A FIRST STEP TOWARD A TEPHROSTRATIGRAPHY OF THE SOUTHERN

INDIAN OCEAN1

W. Morche,2 H.-W. Hubberten,2 A. Mackensen,2 and J. Keller3

ABSTRACT

Geochemical investigations were conducted on 10 discrete ash layers and 22 samples of dispersed ashaccumulations from Sites 747, 749, and 751 of Ocean Drilling Program (ODP) Leg 120 to the Kerguelen Plateau inthe southern Indian Ocean. The chemical data obtained from some 400 single-grain glass analyses allow thecharacterization of two rock series. The first consists of transitional to alkali basalts; the second, mainly of trachyteswith subordinated rhyolites, all reflecting the characteristic magmatological evolution of the Kerguelen Plateau asa hotspot-related volcanism. Chemical correlation with possible source areas indicates that the ashes were mostprobably erupted from the Kerguelen Islands. The investigated ash layers clearly reflect the Oligocene toQuaternary changes in the composition of the volcanic material recorded from the Kerguelen Islands. In addition tothe Kerguelen Islands, Heard Island, Crozet Island, and other sources may have contributed to deposition of thetephras. Pleistocene tephras of "exotic" calc-alkaline composition are most probably derived from enhancedmagmatic activity during that time span at the South Sandwich island arc.

When using data obtained from tephras of the ODP Leg 119 Kerguelen sites, several eruptive periods can becorrelated through the composition of the deposited ashes. Some of them are widely distributed over the KerguelenPlateau and are seen as a first step toward a southern Indian Ocean tephrostratigraphy.

INTRODUCTION

The Kerguelen Plateau is situated between 46° and 64°S inthe southern Indian Ocean and trends northwest for approxi-mately 2500 km. It is about 500 km wide and rises 2-4 kmabove the adjacent deep-sea basins. The Kerguelen Plateau isthe workTs largest submarine plateau and can be divided intotwo distinct domains (Schlich, 1975; Houtz et al., 1977): theNorthern Kerguelen Plateau or Kerguelen-Heard Plateau,located between 46° and 54°S, and the Southern KerguelenPlateau.

Five sites were drilled on the Southern Kerguelen Plateauduring Leg 120 of the Ocean Drilling Program (ODP): 747, 748,749, 750, and 751 (Schlich, Wise, et al., 1989; Fig. 1).Particularly for our study of volcanic ash geochemistry, weselected Site 747 from the Central Kerguelen Plateau, wherenumerous distinct ash layers were encountered. In addition,the only ash layers drilled at Sites 749 and 751 on the SouthernKerguelen Plateau were studied. At Site 747, volcanic ash wasencountered, either dispersed within the sediments or asdiscrete ash layers; this occurrence reflects the influence ofnumerous volcanic eruptions through time. In contrast, Sites749 and 751, located south of Site 747, yielded only one ashlayer each and only small amounts of dispersed glass grains.At Sites 748 and 750 no volcanic ash was recorded.

The purpose of this study is to determine the chemicalcomposition of the ash layers. Furthermore, using these data,we discuss the provenance of the volcanic material. Becauseof the prevailing westerly winds and eastward-flowing oceansurface currents in this region, source areas of the tephrasdistributed by wind, ocean currents, and ice rafting are most

1 Wise, S. W., Jr., Schlich, R., et al., 1992. Proc. ODP, Sci. Results, 120:College Station, TX (Ocean Drilling Program).

2 Alfred Wegener Institute for Polar and Marine Research, Colum-busstrasse, D-2850 Bremerhaven, Federal Republic of Germany.

Mineralogisch-Petrographisches Institut, Universitàt Freiburg/Breisgau,Albertstrasse 23b, D-7800 Freiburg, Federal Republic of Germany.

likely (but not exclusively) located to the west of the drillsites. During the time interval in which the deposits examinedin this study were formed (Oligocene to Quaternary), volcanicactivity in the vicinity of the drill sites has been recorded fromKerguelen and Heard islands and Crozet, the last of whichwas active only in Quaternary time (Girod and Nougier, 1972).

The magmas erupted on the Kerguelen Islands display awide compositional range characterized by two main rockseries: an older transitional to alkali basalt series that probablyformed since Eocene time, and a younger trachyphonoliticseries (Nougier, 1972a, 1972b; Giret and Lameyre, 1983; Giretet al., 1987; Watkins et al., 1974; Nougier et al., 1983). Thevolcanic products of Heard Island belong to both an alkalibasaltic and a trachytic series, as on the Kerguelen Islands,and are presumed to have formed at least since the lateMiocene (Stephenson, 1972; Clarke et al., 1983). The Quater-nary volcanic rocks of Crozet Island, in contrast, belong toone basaltic rock suite that shows much less compositionalvariability (Chevallier et al., 1983; Gunn et al., 1972).

Additional information is available from volcanic ash lay-ers that occur in Leg 119 Sites 736 and 737, which are locatedcloser to the assumed eruption centers (Morche et al., 1991).With the help of the geochemical data available from thisstudy, we will try to correlate characteristic ashes between theLeg 119 and 120 sites.

MATERIAL AND METHODSTen discrete ash layers were investigated from Holes 747A,

747B, 749B, and 751 A. In addition, 22 accumulations ofdispersed ashes from Holes 747A and 747B were studied toobtain more information about volcanic activity in the Ker-guelen Plateau area from the Oligocene to the present (Tables1 and 2).

Site 747 (54°48.68S, 76°47.64E; 1697-m water depth) pene-trated to a total depth of 350 m below sea floor (mbsf). Theoldest sediments recovered have a Santonian age (Schlich,Wise, et al., 1989). Volcanic ash and pumice accumulationsare restricted to the uppermost 170 m of Holes 747A and 747B

151

W. MORCHE, H.-W. HUBBERTEN, A. MACKENSEN, J. KELLER

50

ANTARCTICA

600 km

Measured at 55°S

• • • • • • • mean summer sea-ice edge

• • mean winter sea-ice edgeFigure 1. Location of Leg 120 drill sites, as well as Sites 736, 737, 745, and 746 drilled during Leg 119in the southern Indian Ocean. The bathymetry (in meters) is from GEBCO (Hayes and Vogel, 1981;Fisher et al., 1982). The position of the Polar Frontal Zone has been placed according to Whitworth(1988); the position of sea-ice coverage has been diagrammed according to Dietrich and Ulrich (1968).

and range from early Oligocene to Pliocene age. We investi-gated all seven discrete layers recovered from Site 747 (Table1 and Fig. 2).

Sites 749 (58°43.03'S, 76°24.45'E; 1069.5-m water depth;252-m penetration) and 751 (57°43.56'S, 79°48.89'E; 1634-mwater depth; 170-m penetration) recovered lower Eocene toQuaternary sediments (Barron, Larsen, et al., 1989). Only onediscrete ash layer was obtained from Hole 749B, and one fromHole 751.

All tephra samples were dried at 50°C. Carbonate wasremoved by adding 10% hydrochloric acid. The carbonate-

free residue was rinsed with distilled water and dried. Thesamples were then sieved to obtain 200-µ,m fractions. Only the 63-200-µm fraction was usedfor further study. The samples were first described by meansof a petrographic microscope. Textural features were thendocumented with the help of a scanning electron microscope(SEM). The glass shards were separated using a heavy liquidand a Frantz Isodynamic magnetic separator. Polished grainmounts were prepared for microprobe analyses and petro-graphic studies. The geochemical analyses were performedwith an energy-dispersive KEVEX equipment (EDX) at-

152

GEOCHEMISTRY OF CENOZOIC ASH LAYERS

Table 1. SamplesLeg 120.

Core, section,interval (cm)

120-747A-3H-1, 28-303H-2, 122-1243H-6, 59-617H-3, 137-1397H-3-CC

16H-5,31-3216H-6, 5-7

120-747B-3H-4, 65-67

120-749B-2H-2, 0-2

120-751A-4H-6, 86-88

from discrete

AWIno.

22

2387-186-2482526

24

87-2

27

tephra layers,

Age

late Pliocene

late Plioceneearly Pliocenemiddle Miocenemiddle Miocenelate Oligocenelate Oligocenelate Pliocene

Quaternary

late Pliocene

Notes: Ocean Drilling Program (ODP) sample num-bers and corresponding Alfred Wegener Institute(AWI) laboratory numbers are given.

tached to an SEM at Freiburg University. To avoid uncon-trolled loss of Na2O, glass particles were analyzed under adefocused electron beam of 10 × 10 m minimum (acceleratingvoltage =15 kV, beam current = ~5 nA, counting time =100s). Matrix corrections and calibration with mineral and glassstandards from the Smithsonian Institution were made(Jarosewich et al., 1980). Selected analyses of used standardsare given in Table 3 for comparison. To obtain additionalinformation about volcanic activity, the sand-size fraction(>63 m) of the carbonate- and opal-free nonbiogenic materialfrom Holes 747A and 747B was studied with respect to thecontent of dispersed ash. Twenty-two samples containing>30% glass shards were selected for geochemical analysis.

RESULTSThe most important results of our study were obtained

from the analyses of the discrete tephra layers from Holes747A and 747B. These ash layers are Pliocene, middle Mi-ocene, and late Oligocene in age, and display either alkalibasaltic or trachytic composition. The basaltic ash layerrecovered at Hole 749B is of Quaternary age, and the trachyticash from Hole 751A occurs in late Pliocene sediments. No ashlayers were recovered from Sites 748 and 750 on the SouthernKerguelen Plateau, nor did the nonbiogenic >63-m fraction ofthe samples from these sites contain significant amounts ofvolcanic glass, thus showing limits of distribution of tephratoward higher southern latitudes.

Most of the glass shards examined are dark to pale brown,although some samples consist of populations characterizedby colorless glass. Glass textures range from dense, poorlyvesicular, blocky shards to highly vesicular particles with anelongated, fibrous habit. Tiny mineral inclusions, such asFe-Ti oxides and pyroxenes, are common. The mineral assem-blages within the discrete ash layers are all thought to bephenocrysts (euhedral crystals, adhering glass). The compo-sitional difference between the glasses is also reflected inspecific mineral assemblages. The trachytic tephras are char-acterized by the occurrence of Fe-Ti oxides, aegirine-augite,Plagioclase, alkali feldspar, and apatite. In addition, heden-bergite, biotite, and augite occur in some of these tephras. Thetephras that display a transitional to alkali basaltic composi-tion lack alkali pyroxenes and alkali feldspars. Plagioclase,augite, and, in some cases, olivine are the only phenocrysts.

The mean values of chemical data of about 135 analyses ofindividual glass shards from the Leg 120 discrete tephra layers

are presented in Table 4. The complete data set, including thecomposition of the dispersed ashes, is listed in the Appendix(on microfiche in back pocket). The geochemistry of thetephras shows a large variation in most of the major elements,not only between, but also within different layers. Twodistinctively different compositional groups can be recognizedin the plot of total alkalis vs. silica (Fig. 3A):

1. The first group comprises ashes of trachytic compositionand is restricted almost exclusively to the Miocene, Pliocene,and Pleistocene ashes of Holes 747A, 747B, and 751 A. Mostof them plot in a narrow field with silica contents from 60% to69% and total alkali contents between 11% and 14%. Ash layer87-1 consists of two glass populations: one typical for thistrachytic group, and another with less silica and alkalis(trachyandesitic).

2. The second group is mainly represented by the Oli-gocene tephras that display transitional to alkali basalticcompositions. Nevertheless, both Oligocene tephra layers (25and 26) consist of two glass populations, one of basaltic andone of trachytic to rhyolitic composition. Thus, they representa nearly complete fractionation series, saturated with respectto silica. Only one younger ash layer, the Quaternary tephrafrom Hole 749B (87-2), falls within the same basaltic group.

The dispersed tephras in general show the same compositionaltrends as the discrete tephra layers (Fig. 3B). Additionalglasses of trachyandesitic composition occur during Miocene(154-2 and 155-1), Pliocene (153 and 154-2), and Pleistocene(149-1 and 149-2) time. Furthermore, tephras of low-K calc-alkaline compositions are found in Quaternary sediments(Samples 149-1, 149-2, 159-1, etc.; see Table 5). Their "exot-ic" composition in the framework of a hotspot-related mag-matism corresponds to island-arc activity. Figure 4 shows thealmost complete overlap in the K2O-SiO2 diagram with com-positions of rocks from the South Sandwich Islands.

DISCUSSIONIn an earlier paper, Morche et al. (1991) concluded that

tephras from Sites 736 and 737 (Leg 119) were dominantlyderived from the Kerguelen Islands. Site 736 is located about70 km east, and Site 737 about 150 km southeast of theKerguelen Islands (Fig. 1). Site 747 is located more or less inthe same direction with respect to the Kerguelen Islands,however, the distance is much greater (about 900 km). Nev-ertheless, the Kerguelen Islands should be considered themost likely source for the tephras.

The geochemical variations of the two ash series encoun-tered within the sediments of Site 747 show that both repre-sent an almost complete compositional suite. The compositionof the analyzed glass shards may be compared with publishedanalyses of rocks from the Kerguelen Islands (Giret andLameyre, 1983; Morche et al., 1991). The basaltic and tra-chytic rock series almost completely overlap and, therefore,provide a strong argument for a Kerguelen Islands prove-nance. The subdivision of the tephras in the Oligocene basaltic(and subordinated trachytic to rhyolitic) group and a youngertrachytic group should also be reflected in the eruptive se-quence of the Kerguelen Islands if this area was the eruptioncenter. On the Kerguelen Islands, volcanic activity in lateOligocene to early Miocene time was dominated by theeruption of products of transitional to alkali basalt composi-tion; the only dated early rhyolitic formation yielded an age of26-30 Ma (Nougier, 1972a; Watkins et al., 1974; Giret et al.,1987). This coincides precisely with the ash compositionfound in the Oligocene section of Holes 747A and 747B (25and 26), as well as with the Oligocene tephras encountered

153


149-1 Pleistocene

149-2 Pleistocene

148-2 Pleistocene

150-1 late Pliocene

Table 2. Samples of dispersed tephra and their relative concentra-tions, Leg 120.

Core, section, AWIinterval (cm) Abundance Color no. Age

120-747A-1H-2, 108-112 + D1H-2, 144-148 + D1H-3, 0-4 + D L1H-3, 36-40 + D L1H-3, 72-76 + D L1H-4, 72-76 + D1H-4, 108-112 + D1H-4, 142-146 + D1H-5, 0-4 + D1H-5, 36-40 ++ DL1H-5, 72-76 ++ D L 148-1 Pleistocene1H-5, 141-145 + D L1H-6, 2-6 + D L1H-6, 36-40 + D L2H-1, 0-4 + L2H-1, 36-40 + L2H-1, 72-76 + L2H-1, 108-112 + L2H-1, 144-148 + L2H-2, 0-4 ++ L2H-2, 36-40 + L2H-5, 36-40 + D L2H-5, 72-76 + D L2H-6, 0-4 + D L2H-6, 36-40 + D L2H-6, 72-76 + D L2H-6, 108-112 + DL2H-6, 144-148 ++ DL2H-7, 0-4 ++ DL2H-7, 36-37 + + + + D L2H-7, 72-76 + + L2H-7-CC + + + L3H-l,0-4 + + + + DL3H-1, 36-40 + + + + DL3H-1, 72-76 + D L3H-1, 108-112 + + + DL3H-1, 144-148 + L3H-2, 0-4 + L3H-2, 72-76 + + + + DL3H-2, 108-112 + + + + DL3H-2, 144-148 + + + D L3H-3, 0-4 + + + D L3H-3, 36-42 + + + + D L3H-3, 72-76 + + + + D L3H-3, 108-112 + + + L3H-3, 144-148 ++ L3H-4, 0-4 ++ L3H-4, 36-40 + + + DL3H-4, 72-76 + + + DL3H-4, 108-112 + + + DL3H-5, 0-4 + + + DL3H-5, 36-40 + + + DL3H-5, 72-76 + + + DL3H-5, 108-112 + + + DL3H-5, 144-148 + + + + DL3H-6, 0-4 + + + + DL3H-6, 36-40 + + + + DL4H-l,0-4 ++ L4H-1, 36-40 + + + L4H-1, 72-76 + + + L4H-1, 108-112 + + + L4H-1, 144-148 + + + L4H-2,0-4 + + + L4H-2, 36-40 + + + L4H-2, 72-76 + + + + D L4H-2, 108-112 + + + + DL4H-2, 144-148 + + + D L4H-3, 0-4 + + + DL4H-3, 36-40 ++ DL4H-3, 72-76 ++ DL4H-3, 108-112 ++ D L4H-4, 0-4 + + + D L4H-4, 36-40 + + + + D L4H-4, 72-76 + + + + D L4H-4, 108-112 + + + + D L

Table 2 (continued).

150-2 late Pliocene

151 early Pliocene

152-1 early Pliocene

153 early Pliocene

Core, section,interval (cm) Abundance Color

AWIno. Age

4H-4, 144-1484H-5, 0-44H-5, 36-404H-5, 72-764H-5, 108-1124H-5, 144-1484H-6, 0-44H-6, 36-404H-6, 72-764H-6, 108-1124H-6, 144-1444H-7, 0-44H-7-CC4H-7, 36-404H-7, 72-765H-4, 72-765H-6, 72-765H-6-CC6H-2, 72-766H-4, 72-76

16H-5,72-7616H-CC17X-2, 72-7617X-5,72-7617X-CC18X-2, 72-7618X-2, 72-7618X-CC

120-747B-1H-2, 0-41H-2, 36-402H-2, 108-1122H-3, 144-148

D LD LD LD LD LD LD LD LD LD LD LD LD LD LD LD LD LD L

LL

D LD LDDDDDD

DDD

L

154-1

154-2

155-1

155-2156-1156-2

157-1157-2

158-1158-1

158-2159-1159-2

early Pliocene

late Miocene

late Miocene

late Miocenelate Miocenelate Oligocene

late Oligocenelate Oligocene

early Oligoceneearly Oligocene

PleistocenePleistocenePleistocene

152-2 early Pliocene

Notes: + = l%-5%, + + = 5%-20%, + + + = 20%-50%, + + + + =>50%). D = brown to dark brown, L = colorless to light yellow.

within Holes 737A and 737B of Leg 119 (Morche et al., 1991).In contrast to the Oligocene ash layers analyzed from Site 737,where all glasses are exclusively of basaltic composition, thetwo Site 747 ashes (25 and 26) show more or less bimodalcompositions ranging from basalts to trachytes and rhyolites.In the case of Sample 26, a continuous tapping of a zonedmagma chamber is suggested.

Most of the Miocene, Pliocene, and Pleistocene tephras ofSites 747 (Fig. 2) and 751 belong to the trachytic to rhyoliticseries, reflecting enhanced explosive volcanic activity. Be-cause volcanic rocks displaying trachytic composition haveerupted at the Kerguelen Islands predominantly during theMiocene to Quaternary time span (Nougier 1972a, 1972b;Giret et al., 1987), we suggest that the Kerguelen Islands haveto be regarded as the source of the younger ashes found atSites 747 and 751.

Other possible sources of the tephras could be HeardIsland, about 450 km southeast of the Kerguelen Islands, and,less likely, Crozet Island in the far northwest. Heard Island,which is located much closer to Site 747 than to Site 737,shows many similarities with Kerguelen in terms of thecomposition of the magmatic rock series and evolution duringtime (Stephenson, 1972; Clarke et al., 1983). Crozet Island, onthe other hand, forms a single Quaternary alkali-basalt petro-graphic province (Gunn et al., 1972; Chevallier et al., 1983)and cannot be the source of the Oligocene basaltic or Quater-nary trachytic tephras found in sediments at Sites 747 and 751.

We conclude, therefore, that almost all of the discretetephra layers studied were erupted from the Kerguelen Is-lands. Heard Island has to be considered as another importantsource of volcanic ash. Transport of additional tephra to thesite of deposition by ice rafting is also possible, because all of

154


Φ

OHole g Hole747A s '"•"» A 8 β

1 6 0 -

747B

Tephra Tephralayers dispersed

158-2* * 148-1/2* * 149-1/2

159-1/2

t * * * 150-1

* * * * 150-2

152-1/2ffft 153Hi 154-1/2**

* * 155-1

155-2

156-1

*** *

*

156-2

157-1/2

158-1

Figure 2. Schematic stratigraphic column, Site 747. The position of the discrete tephras investi-gated (solid lines) as well as their Alfred-Wegener-Institute (AWI) laboratory numbers are shown.Accumulations and relative abundance of dispersed tephras are marked with an asterisk.

155


Table 3. Analytical results obtained by EDX at Freiburg University on Smithsonian Institution reference glass

and mineral standards.

USNM

111240Basaltglass

113498Basaltglass

113716Basaltglass

72854Rhyoliteglass

115900Plagioclase

SmithsonianFreiburg

SmithsonianFreiburg

SmithsonianFreiburg

SmithsonianFreiburg

SmithsonianFreiburg

(%)

(mean)(mean)(SD)

(mean)(mean)(SD)

(mean)(mean)(SD)

(mean)(mean)(SD)

(mean)(mean)(SD)

SiO2

51.1150.860.15

51.6851.600.07

51.8451.970.42

77.1977.510.20

51.2151.510.07

TiO2

1.861.860.04

4.123.910.07

1.311.320.03

0.120.040.08

0.050.000.00

A12O3

14.1414.210.02

12.6712.700.16

15.4915.420.08

12.1412.190.15

30.8930.770.10

FeO

11.9112.110.15

13.4013.700.12

9.129.150.15

1.241.060.09

0.460.360.04

MnO

0.220.220.03

0.000.270.08

0.000.150.11

0.030.000.00

0.000.040.08

MgO

6.756.690.07

5.154.760.15

8.267.820.12

0.090.000.00

0.140.000.00

CaO

11.1811.210.08

9.449.170.14

11.3911.320.10

0.500.510.03

13.6313.360.25

K2O

0.190.180.01

0.830.900.03

0.090.150.03

4.925.070.07

0.180.140.02

Na2O

2.642.750.12

2.702.960.15

2.492.770.11

3.773.620.05

3.453.690.14

Notes: Means of the Freiburg analyses are based on 4 -8 single analyses. SD = standard deviation. The reference analysesgiven (numbers = USNM standards) are taken from Jarosewich et al. (1980).

the drill sites are situated in an area characterized by a highfrequency of drifting icebergs (Barron, Larsen, et al., 1989).Sea-ice rafting may also have been an important process oftransport for the volcanic ashes, because the sites aresituated far south of the mean winter sea-ice edge (Fig. 1).This seems to be the case especially for the low-K calc-alkaline Pleistocene tephras, which can be geochemicallycorrelated with the island-arc activity of the South SandwichIslands.

TEPHROCHRONOLOGY

In this section, we will discuss our geochemical results tocorrelate ash layers between ODP sites from Legs 119 and 120on the Kerguelen Plateau. Data and results from Morche et al.(1991) are used for correlation. A preliminary gëochronologi-cal synopsis, comparing all the tephra layers studied from theKerguelen Plateau (Legs 119 and 120) with the rock series ofthe Kerguelen Islands, is shown in Figure 5. As discussed

Table 4. Mean values of chemical data from the discrete ash layers studied, Leg 120.

AWIno.

22N = 923N = 924N = 1425(A)N = 325(B)N = 1726N = 24(total)26(B)N = 827N = 1048N= 986-2N = 987-1 (A)N = 487-l(B)N = 1187-2(A)N = 687-2(B)N = 887-2(C)N = 2

(%)

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

MeanSD

Mean

P2O5

0.000.000.000.000.000.000.530.280.000.000.170.25

0.000.000.000.000.170.090.000.000.140.160.000.000.040.090.000.000.00

SiO2

66.051.04

62.430.53

65.031.56

53.936.02

69.132.80

59.106.92

67.671.46

67.511.04

63.353.87

64.451.09

55.441.70

60.430.98

50.721.36

60.380.78

68.58

TiO2

0.540.130.480.120.720.312.941.170.490.101.921.13

0.530.080.530.110.600.210.480.122.150.210.640.233.810.200.670.240.26

A12O3

14.620.65

17.850.17

15.290.81

14.760.71

12.681.42

14.530.29

14.530.34

13.240.95

17.162.22

16.850.76

17.840.60

19.281.03

14.650.53

19.060.97

14.86

FeO

5.500.124.060.255.480.10

12.922.786.141.159.853.46

5.390.915.890.334.420.354.600.158.090.434.480.49

13.911.754.770.633.93

MnO

0.220.140.110.130.170.120.110.200.140.140.210.11

0.100.120.220.101.530.470.040.090.080.090.020.070.140.110.000.000.00

MgO

0.000.000.000.000.000.002.871.570.000.001.731.69

0.000.000.000.000.000.000.000.001.320.330.000.003.620.480.000.000.00

CaO

1.160.241.160.141.480.487.532.841.350.675.593.03

2.000.480.810.140.070.101.290.185.790.381.740.299.200.801.710.300.82

K2O

5.580.286.080.255.790.432.041.645.240.313.011.57

5.020.665.230.206.440.636.230.334.500.516.580.491.260.386.550.135.60

Na2O

6.260.287.670.555.970.302.901.174.710.734.030.80

4.680.496.390.246.260.635.840.144.560.326.640.602.640.456.680.785.54

Notes: Ash layers in which different geochemical groups of glass shards can be discriminated are indicatedas (A), (B), and (C), respectively. N = number of individual glass shards analyzed from a single ashsample. SD = standard deviation.

156


above, all of the discrete tephra layers encountered in theOligocene to Quaternary sediments from the Kerguelen Pla-teau may have been derived from the Kerguelen Islandvolcanic rock series as basaltic volcanism occurred over theentire time span and trachytic magmatic activity took placesince the late Oligocene.

An analogue to the numerous Quaternary trachytic tephrasfrom Site 736 is found in the tephras of Site 747A (149-1,149-2, 159-1, and 159-2) and in a more distal position at Site749 (87-2). The latter shows bimodal composition with anadditional basaltic part, which can also be found in Site 747A(158-2). The oldest Quaternary ash of Site 736 (77) shows abimodal glass composition with one trachytic group thatcorresponds to the aforementioned tephras and can thereforebe assumed to have formed through the same eruptive eventfrom either the Kerguelen or Heard islands. The glasses oflow-K calc-alkaline composition from Layer 77 correspondwith Pleistocene glasses from Site 747 (149-1, 149-2, 158-2,and 159-1; Table 5). These tephras most probably werederived from the South Sandwich Islands. The distance ofabout 6500 km indicates sea-ice rafting as an importantmechanism of transport. Furthermore, these data can becompared with analyses of Quaternary glasses from ODPLeg 114 (Table 5). At Site 701 the thick proximal facies oftephras from the South Sandwich Islands have been recog-nized as much as several centimeters thick (Hubberten etal., 1991).

All of the Pliocene tephras found in Sites 737, 747, 751, and745 display the same trachytic composition and may becorrelated. They are probably the result of a period of majorexplosive eruptions at the Kerguelen Islands because theyoccur as far from the assumed eruption center as the deepestKerguelen site, Site 745 in the Australian-Antarctic Basin.This period of numerous volcanic eruptions is also docu-mented in the increased concentrations of glass shards withindispersed ash layers studied from Site 747 (Fig. 2 and Table 1).The Miocene tephras from Sites 737 and 747 may be corre-lated because most of them occur at the same time and theybelong to the same trachytic group. However, Ash Layers 80and 81-1 display higher silica and lower alkali contents thanthose from Site 747 (86-2 and 48). The transitional-basalticupper Oligocene tephra layers from Site 737 (81-2 and 82) andSite 747 (25 and 26) are chemically identical and occur in thesame time period, suggesting the same eruptive events. TheKerguelen Islands are assumed to be the source region forthese ashes, as both sites are within the major wind and oceancurrent directions from those islands. Trachytic to rhyoliticglasses from Ash Layers 25 and 26 were not found at Site 737and are absent either because of a hiatus or the lack ofrecovery during drilling. Alternatively, Heard Island could bethe source because of their closer geographic position. Theearly Oligocene transitional-basaltic ashes from Site 737 werefound in the dispersed glasses at Site 747B (e.g., 158-1).Higher glass concentrations not investigated here occur at thatsite in the lower Oligocene and may correspond to alkali-basaltic Ash Layers 84-1 and 84-2 from Site 737.

CONCLUSIONSThe volcanic ash layers studied from the Leg 120 sites

belong to two different rock series: a basaltic series and atrachytic one. The tephras of basaltic composition occurpredominantly in the Oligocene sediments; only one basalticash layer was found in the Quaternary. The Miocene,Pliocene, and most of the Quaternary tephras belong to atrachytic group. When compared with the volcanic productsof the Kerguelen Islands, it is obvious that most of the tephraswere formed through eruptive events at that locality. A less

important contribution from Heard, Crozet, or other islandshas to be considered for some of the tephras. At least fivedistinct eruptive events can be detected in more than one ofthe Leg 119 and 120 Kerguelen Plateau sites: (1) a Quaternarytrachytic tephra correlated between Site 736 and 747A; (2) aQuaternary basaltic tephra encountered at Sites 736 and 749;(3) low-K calc-alkaline volcanic ash in Quaternary sediments,probably derived from the South Sandwich island arc bysea-ice rafting; (4) several Pliocene and Miocene trachytictephras that are distributed over the entire Kerguelen Plateauand are deposited as far as the Antarctic-Australian Basin(Site 745); and (5) late and early Oligocene basaltic tephralayers occurring at Sites 737 and 747. The correlation ofdistinct tephra layers between different sites on the KerguelenPlateau area may be regarded as a first step toward a tephro-stratigraphy of the polar Indian Ocean. Future investigationson characteristic trace elements and on glass populations ofsmaller grain size within dispersed ash concentrations will beneeded to establish this tephrostratigraphy.

ACKNOWLEDGMENTSWe wish to thank Heike Ostermann (AWI) and Erika Lutz

(Freiburg University) for technical help and assistance duringsample processing and microprobe measurements. Financialsupport by the Deutsche Forschungsgemeinschaft (Grants Fu119/14 and Hu 378/2) is acknowledged. This is AWI Publica-tion No. 346.

REFERENCES

Baker, P. E., 1978. The South Sandwich Islands: III. Petrology of thevolcanic rocks. Sci. Rept. Br. Antarct. Surv., 93.

Barron, J., Larsen, B., et al., 1989. Proc. ODP, Init. Repts., 119:College Station, TX (Ocean Drilling Program).

Chevallier, L., Nougier, J., and Cantagrel, J. M., 1983. Volcanologyof Possession Island, Crozet Archipelago (TAAF). In Oliver, R. L.,James, P. R., and Jago, J. B. (Eds.), Antarctic Earth Science:Cambridge (Cambridge Univ. Press), 652-658.

Clarke, I., McDougall, I., and Whitford, D. J., 1983. Volcanicevolution of Heard and McDonald Islands, southern IndianOcean. In Oliver, R. L., James, P. R., and Jago, J. B. (Eds.),Antarctic Earth Science: Cambridge (Cambridge Univ. Press),631-635.

Dietrich, G., and Ulrich, J., 1968. Atlas zur Ozeanographie: Mann-heim (Bibliographisches Institut).

Fisher, R. L., Jantsch, M. Z., and Comer, R. L., 1982. GeneralBathymetric Chart of the Oceans (GEBCO), Scale 1:10,000.000.5-9: Ottawa (Canadian Hydrographic Service).

Giret, A., and Lameyre, J., 1983. A study of Kerguelen plutonism:petrology, geochronology and geological implications. In Oliver,R. L., James, P. R., and Jago, J. B. (Eds.), Antarctic EarthScience: Cambridge (Cambridge Univ. Press), 646-651.

Giret, A., Lameyre, J., Beaux, J.-F., Gautier, I., Verdier, O., Chotin,P., and Cantagrel, J.-M., 1987. Géologie, deux siècles de re-cherche dans les lies Kerguelen. Actes Colloq. Rech. Fr. TerresAustrales, 345-355.

Girod, M., and Nougier, J., 1972. Volcanism of the Sub-antarcticislands. In Adie, R. J. (Ed.), Antarctic Geology and Geophysics:Oslo (Universitetsforlaget), 777-789.

Gunn, B. M., Abranson, E. C , Watkins, N. D., and Nougier, J.,1972. Petrology and geochemistry of lies Crozet: a summary. InAdie, R. J. (Ed.), Antarctic Geology and Geophysics: Oslo (Uni-versitetsforlaget), 825-829.

Hayes, D. E., and Vogel, M., 1981. General Bathymetric Chart of theOceans (GEBCO), Scale 1:10,000.000. 5-13: Ottawa (CanadianHydrographic Service).

Houtz, R. E., Hayes, D. E., and Markl, R. G., 1977. KerguelenPlateau bathymetry, sediment distribution and crustal structure.Mar. GeoL, 25:95-130.

Hubberten, H.-W., Morche, W., Westall, F., Fütterer, D. K., andKeller, J., 1991. Geochemical investigations of volcanic ash layersfrom southern Atlantic Legs 113 and 114. In Ciesielski, P. F.,

157


Kristoffersen, Y., et al., Proc. ODP, Sci. Results, 114: CollegeStation, TX (Ocean Drilling Program), 733-749.

Jarosewich, E., Nelen, J. A., and Norberg, J. A., 1980. Referencesamples for electron microprobe analysis. Geostandards News-lett., 4:43-41.

MacDonald, G. A., and Katsura, T., 1964. Chemical composition ofHawaiian lavas. J. Petrol., 5:82-133.

Morche, W., Hubberten, H.-W., Ehrmann, W. U., and Keller, J.,1991. Geochemical investigations of volcanic ash layers from Leg119 to Kerguelen Plateau. In Barron, J., Larsen, B., et al., Proc.ODP, Sci. Results, 119: College Station, TX (Ocean DrillingProgram).

, 1972a. Geochronology of the volcanic activity in liesKerguelen. In Adie, R. J. (Ed.), Antarctic Geology and Geophys-ics: Oslo (Universitetsforlaget), 803-808.

., 1972b. Volcanic associations in lies Kerguelen. In Adie,R. J. (Ed.), Antarctic Geology and Geophysics: Oslo (Univer-sitetsforlaget), 809-815.

Nougier, J., Pawlowski, D., and Cantagrel, J. M., 1983. Chrono-spatial evolution of the volcanic activity in southeastern Ker-guelen (T.A.A.F.). In Oliver, R. L., James, P. R., and Jago, J. B.(Eds.), Antarctic Earth Science: Cambridge (Cambridge Univ.Press), 640-645.

Schlich, R., 1975. Structure et age de 1'océan Indien occidental. Mem.Hors-Ser. Soc. Geol. Fr., 6:1-103.

Schlich, R., Wise, S. W., Jr., 1989. Proc. ODP, Init. Repts., 120:College Station, TX (Ocean Drilling Program).

Stephenson, P. J., 1972. Geochemistry of some Heard Island igneousrocks. In Adie, R. J. (Ed.), Antarctic Geology and Geophysics:Oslo (Universitetsforlaget), 793-801.

Taylor, S. R., Arculus, R., Perfit, M. R., and Johnson, R. W., 1981.Island arc basalts. Basaltic Volcanism on the Terrestrial Planets.Basaltic Volcanism Study Project: New York (Pergamon Press),193-213.

Watkins, N. D., Gunn, B. M., Nougier, J., and Baksi, A. K., 1974.Kerguelen: continental fragment or oceanic island? Geol. Soc.Am. Bull., 85:201-212.

Whitworth, T., Ill, 1988. The Antarctic circumpolar current.Oceanus, 31:53-58.

Date of initial receipt: 16 January 1989Date of acceptance: 22 October 1990Ms 120B-163

1 4 -

1 2 -

S- 10-

co"

O 8CM

6 -

4 -

87-2

86-248

Alkali-basalts

44 46 48 74I

76

SiO,

Figure 3. Total alkali vs. silica diagrams for discrete ash layers (A) and accumulations of dispersed ash layers (B) investigated in this study.Numbers given in the diagram correspond to the A WI sample numbers. The discriminating line between the alkali and tholeiitic basalts has beenplotted according to MacDonald and Katsura (1964). Fields of discrete ash layers are shown in black for comparison in Figure 3B.

158


14 -

12-

èi 10

CO

8 -

6 -

4 -

44 46 48 50 52 54 56 58 60 62 64 66 68 70 72 74 76

SiCX

Figure 3 (continued).

Table 5. Average composition of all Leg 119 and 120 calc-alkaline tephras.

(%) SiO2 TiO2 A12O3 FeO MnO MgO CaO K2O Na2O

Leg 120N 15

aLeg 119N = 24

South Sandwich IslandsN = 15

cLeg 114N = 15

Notes: Tephras are from low-K series and are shown in comparison with South Sandwich Islands volcanic rocks andthe Quaternary Leg 114 Tephra 114-701A-1H-4, 58-60 cm.

a Morche et al. (1991).b Baker (1978).c Hubberten et al. (1990).

MeanSD

MeanSD

MeanSD

MeanSD

57.753.43

56.872.90

55.502.95

56.640.79

1.100.32

1.120.21

1.000.28

1.090.10

15.411.52

15.400.98

16.131.37

15.560.41

10.482.57

10.961.40

10.961.48

11.170.62

0.220.07

0.200.11

0.190.03

0.250.09

3.201.47

3.251.46

3.851.37

3.410.43

7.921.26

8.491.28

8.701.53

8.270.25

0.580.14

0.580.18

0.550.26

0.360.06

3.340.91

3.030.54

3.000.44

3.200.32

159


4.0

High-K-series

dispersed tephraoLeg 119 Leg 120 South Sandwich Islands

Figure 4. Comparison of dispersed glasses from Legs 119 and 120 with the probable source area of the SouthSandwich Islands in the K2O-SiO2 diagram. Classifications are after Taylor et al. (1981); data are from Baker (1978)and Morche et al. (1991).

KerguelenIslands

N N W -Hole 736A

(70 km)Holes 737A

and 737B(150 km)

Holes 747Aand 747B(900 km)

Hole749B

(1100 km)

Hole751A

(1200 km)

SSEHoles 745B

and 746(1500 km)

πo2

Quaternary

Pliocene

late

Miocene

early

late

Oligocene

early

Eocene

27

85

86-1

Figure 5. Geochronological synthesis of discrete tephra layers from Legs 119 and 120 compared with dated rock series from the KerguelenIslands. The time scale for the Kerguelen Islands shows dated volcanic/intrusive activity (data from Nougier, 1972a, and Giret et al., 1987).Distances from the holes to the Kerguelen Islands are indicated in parentheses.

160

wise, s. w., jr., schlich, r., et al., 1992 proceedings of ...w. jr. schlich r. et al. 1992 proc....

Documents