16. MAGNETIC PROPERTIES AND ALTERATION IN BASALT, HOLE 504B, DEEP SEADRILLING PROJECT LEG 831

H. Kinoshita, Department of Earth Science, Chiba Universityand

Toshio Furuta and Hodaka Kawahata, Ocean Research Institute, University of Tokyo2

ABSTRACT

We report measurements of magnetic intensity, inclination, initial susceptibility, Koenigsberger's ratio, saturationmagnetization, and Curie temperatures of 68 basalt samples from the Leg 83 section of Hole 504B. As in the upper partof the hole, reversely magnetized units predominate. Intensities of natural remanent magnetization vary widely, but therange of variation is an order of magnitude less than in the upper part of the hole. This and the other properties mea-sured indicate that the magnetic characteristics of basalts from Hole 504B have been strongly affected by hydrothermalalteration, particularly in the deeper, Leg 83 section. The alteration states of the magnetic samples were studies using X-ray diffraction, electron microprobe, X-ray fluorescence, and ion coupled plasma. Our results suggest three alterationzones in Hole 504B: a low-temperature zone (274.5-890 m) and two high-temperature zones (890-1050 m and 1050-1350 m), differing in the number of veins observed in the samples and presumably differing in the volumes of hydro-thermal fluids which reacted with the basalts.

INTRODUCTION

Hole 504B is located on the southern flank of theCosta Rica Rift. The age of the basement at Site 504 isestimated to be 5.9 m.y., based on the magnetic anom-aly pattern. (This area displays reversed remanent mag-netization.) Hole 504B was started by Leg 69 shipboardparty and reached 1350 m beneath the seafloor (BSF)during drilling on Legs 69, 70, and 83. The basementsection consists of alternating sequences of pillow andmassive basalts; only massive formations and sheeteddikes were cored below 1055 m BSF. This report presentsa series of data on rock magnetic parameters and relatedmineralogical alteration of basalt samples from the Leg83 section of Hole 504B (836-1350 m BSF). Furuta(1983) and Furuta and Levi (1983) presented similar da-ta on samples from Legs 69 and 70 basement sections ofHole 504B. In the first section, paleomagnetic, rock mag-netic, and magneto-mineralogical analyses are described.In the second section, data concerning the hydrothermalalteration found in the paleomagnetic rock samples arebriefly presented in light of the paleomagnetic data.

MAGNETIC PROPERTIES OF BASALT SAMPLESMeasurements of magnetic properties were carried out

on 68 homogeneous basalt samples recovered duringLeg 83. Magnetic susceptibility, intensity, and inclina-tion had been measured on board Glomar Challenger onthe same samples (Hole 504B summary chapter, thisvolume); we repeated the measurement of these parame-ters. Methods used for our shore-based studies werethose described by Furuta (1983) and Furuta and Levi

Anderson, R. N., Honnorez, J., Becker, K., et al., Init. Repts. DSDP, 83: Washington(U.S. Govt. Printing Office).

2 Addresses: (Kinoshita) Department of Earth Science, Chiba University, 1-33 Yayoi-cho, Chiba 260, Japan; (Furuta, Kawahata) Ocean Research Institute, University of Tokyo,Nakano, Toyko 165, Japan.

(1983). The results of our measurements of natural re-manent magnetization (NRM), initial susceptibility, andthermomagnetic properties are listed in Table 1.

Intensity of Natural Remanent Magnetization (/„)The intensity of NRM (Jn) is widely scattered, rang-

ing from 0.09 to 36 x 10~4 gauss. A wide variation ofNRM was also observed in the upper part of Hole 504B(Furuta and Levi, 1983). However, the range of Jn in thelower part of the hole is an order of magnitude smallerthan that in the upper part. Jn versus depth is shown inFigure 1. Jn depends little on the depth but rather seemsto depend on grain-size distribution of ferromagneticminerals as well as on the degree of hydrothermal altera-tion of individual rock samples. Several features are con-spicuous: very low Jn in pillow units, and relatively highJn in massive flows near the bottom of the hole.

Pillow lavas usually show high Jn because fine mag-netic minerals are present. However, the lowest Jn of thishole (less than 10~4 gauss) is found in a pillow lava se-quence. Relatively high Jn is found in a massive flownear the bottom of the hole that is composed of coarse-grained doleritic or micro-gabbroic rocks. These excep-tions to the general trend seem to depend upon hydro-thermal alteration of individual rock samples.

Inclination (/s)Magnetic remanence vectors of all the available mini-

cores were remeasured by a spinner magnetometer in theshore laboratory. The stable inclinations plotted in Fig-ure 2 are used to define 16 magnetic units, which in-clude various lithologic units. Table 2 shows the averagemagnetic properties of each unit. Reversely magnetizedunits predominate through the hole, which is consistentwith the polarity of the upper part of the hole (Furutaand Levi, 1983). There are a few polarity changes, ofwhich one steep inclination zone was observed. A simi-

331

H. KINOSHITA, T. FURUTA, H. KAWAHATA

Table 1. Paleomagnetic results, Hole 504B, Leg 83.

Core-Section(interval in cm)

71-1, 127-129

72-2, 105-10773-1, 87-8973-2, 93-9574-1, 94-9675-1,9-1176-1, 24-2677-1, 23-2577-3, 14-1678-1,42-4479-1, 96-9879-3, 110-113

80-1, 122-12481-1, 114-11682-1, 21-2382-2, 111-11383-1, 33-3584-1, 134-136

85-2, 75-77

87-1, 85-87

87-2, 8-10

88-1, 15-1789-1, 100-103

90-2, 101-10490-4, 143-146

92-3, 67-6993-2, 81-8393-3, 66-68

94-1, 107-10994-3, 82-84

95-1, 100-102

98-1, 78-8099-1, 69-71100-1, 21-23

100-3, 20-22

101-1, 120-122101-2, 90-92103-1, 51-53

104-2, 124-126104-3, 27-29105-1, 116-118

107-2, 11-13108-1, 62-64

109-1, 21-23111-1, 73-75112-1, 63-65

113-1, 5-7113-1, 109-111

117-1, 121-123118-1, 25-27121-1, 33-35

123-1, 52-54126-1, 151-153127-1, 83-85128-1, 27-29129-1, 93-95

129-2, 18-20130-1, 77-79

131-1, 129-131132-1, 17-19132-2, 22-24133-1, 88-90133-2, 17-19134-1, 47-49136-1, 82-84137-1, 54-56

138-1, 28-30

Sub-bottomdepth (m)

837.28

846.06853.06854.94862.45870.60879.75888.74891.65897.93905.47908.62

911.23920.65928.72931.12937.84947.80

957.76

967.86

968.69

976.66986.52

997.031000.45

1016.181023.821025.17

1031.581034.33

1040.51

1062.791072.201080.72

1083.72

1090.711091.911108.02

1119.251119.781126.67

1145.121153.13

1153.721162.241166.64

1171.061172.10

1190.721194.261207.84

1223.031251.021254.341261.281270.94

1271.691279.78

1288.801295.181296.731304.891305.681313.481322.831327.55

1332.29

Jn(× 1 0 - 4 gauss)

33.6

14.319.12.6

13.512.76.66.21.87.58.74.5

0.140.330.170.750.170.29

9.3

2.5

2.3

0.090.74

0.090.38

2.50.60.31

7.27.7

10.3

1.5434.56.7

4.6

3.68.50.52

17.89.2

13.2

13.513.5

17.720.524.5

23.013.5

10.911.26.9

11.013.79.3

19.27.8

9.07.0

6.85.03.4

31.220.336.318.522.2

18.1

's(°)80.6

-14 .0-7 .4- 8 . 0

-10.7-32.6-18.1-11.4

(3.5)-16 .0-10 .0-21.8

- 6 . 9-17 .0-12.0-20.0- 6 . 3

-35.0

0

-52.3

16.9

-21 .0-42.8

-75 .0-69.0

- 6 . 8-11 .0- 6 . 2

7.212.7

- 6 . 4

-40 .0-24.1-23 .6

7.2

-12.5-11.5-27.9

1.14.7

12.1

-8 .5-12 .2

8.04.4

23.0

-6 .5- 3 . 4

-22.3-29.0-32.9

14.611.58.89.05.0

- 3 . 0- 7 . 6

2.618.523.414.79.4

19.537.336.3

-45.0

k(× 1 0 - 4 gauss/Oe)

34.4

20.818.626.328.824.420.024.927.523.011.06.4

1.21.90.95.30.91.5

9.4

1.5

1.3

0.40.9

11.01.1

2.22.41.4

10.913.7

6.4

1.49.71.9

2.4

3.713.11.1

15.312.116.9

16.017.8

21.513.812.6

18.812.5

11.312.414.5

12.48.48.8

10.610.7

12.29.8

10.97.6

12.614.910.610.510.213.6

13.0

Qn

2.7

1.92.90.31.31.40.90.70.20.92.21.9

0.30.50.50.40.50.5

2.8

4.6

4.9

0.62.3

2.30.96

3.20.70.6

1.81.6

4.5

3.19.99,7

6.7

2.71.81.3

3.22.12.2

2.32.1

2.34.15.4

3.43.0

2.72.51.3

2.54.52.95.02.0

2.02.0

1.71.80.75.85.39.65.04.5

3.9

MDF(Oe)

120

145240110155145210130130150260280

400260330170300310

250

380

460

350280

230410

270380340

250250

330

250230420

240

195260260

230220220

260340

140170230

180230

190250240

230170180200250

270250

270270260240220210210250

170

(emu/g)

1.14

1.411.501.641.781.640.711.891.562.580.680.40

0.060.120.120.390.080.11

0.78

0.12

0.10

0.090.11

0.190.11

0.110.190.09

1.141.17

0.55

0.091.000.15

0.12

0.121.980.09

1.240.750.70

1.561.18

1.610.701.41

1.471.23

(0.13)1.172.03

1.470.840.971.151.43

0.950.89

0.920.711.261.74

(0.22)1.510.771.32

1.45

Magneticunit

0

la

lb

2a

2b

2c

3a

3b

4

5

6a

6b

7

8

9

10

11

12a

12b

13

14

15

16

Note: Is = stable inclination after available AF demagnetization; magnetic unit based on stable inclination; data in parenthe-ses are doubtful.

332

MAGNETIC PROPERTIES AND ALTERATION IN BASALT, HOLE 504B

Jn (x10"

4 gauss)

10 20 30

.21 Inclination (°c? -2 § -60 -40 -20 0

20 40 60

900-

1000-

1100-

1200-

1300-

1

2

3

4

~

6

-1-

8

9

10

11

12

13

15

16

•

•I #

#

••

*

••

a

•

•

•

••• •

••* *

E

belo

w s

eaflo

or (

epth

α

900-

1000-

1100-

1200-

1300-

1

2

3

5

6

=7 =

8

9

10

11

12

13

1141

15

16

i

•

~l 1•#

;

a

Φ *

1

•

• m

m#

a•

1

Φ

Figure 1. Intensity of NRM plotted against depth in Leg 83, Hole504B samples. From 910 m to 1025 m the intensity of NRM isquite weak, less than 10~4 gauss, and this part is correlated to thehigh hydrothermal alteration zone.

Figure 2. Stable inclinations plotted against depth in Leg 83, Hole504B samples. The present geomagnetic inclination at Site 504C is18.1° (downward). Magnetic units are based on the stable inclina-tion.

Table 2. Average magnetic properties of magnetic units, Hole 504B, Leg 83.

Magneticunit

lalb2a2b2c3a3b456a6b789

101112a12b13141516

TV

11611122321313323235281

•/n(x 10-4 gauss)

8.90.39.32.52.30.44.71.17.4

10.314.24.64.2

13.413.521.018.0

9.712.28.0

17.918.0

h

(°)- 1 4- 1 6

0- 5 2

17- 3 2- 7 2

- 810

- 6- 2 9

7- 1 7

6- 1 0

12- 6

- 2 810

- 520

-45

k(×10-4gauss/Oe)

21.11.99.41.51.30.76.12.0

12.36.44.32.46.0

14.816.916.015.612.710.211.011.413.0

On

1.30.42.84.64.91.51.61.51.74.57.66.71.92.52.23.93.22.23.42.04.33.9

MDF(Oe)

180295250380460315320330250330300240238236300180205227206260241170

h(emu/g)

1.440.150.80.10.10.10.150.131.460.550.40.120.730.91.41.241.351.111.20.921.11.45

Note: Magnetic units based on stable inclination.

333

H. KINOSHITA, T. FURUTA, H. KAWAHATA

lar steep inclination was also recognized in the upperpart of this hole (Cores 61 through 65; Furuta and Levi,1983). Normal magnetization zones are predominantnear the bottom of the hole. These zones consist of coarse-grained doleritic rocks, characterized by high JQ.

Initial Susceptibility (k) and Koenigsberger Ratio (Qa)

The initial susceptibility (k) is primarily controlled bythe volume concentration of ferromagnetic minerals aswell as by the grain size. Values of k vary from 0.9 to34.4 × I0" 4 gauss/Oe (Table 1) through the hole. Themean k of the hole is around 1 x 10 ~3 gauss/Oe. Thereis a significant decrease of k between 910 and 960 mBSF, which seems again to be related to the hydrother-mal alteration.

Values of Qn (the ratio of remanence to induced mag-netization) are listed in Table 1, and generally correlatewell with / n . However, the Qn values of all the samplesare rather low, and the Qn values of 20% of the Leg 83samples are less than 1. The mean Qn value is around 3.The low Qn and relatively high k values suggest that min-erals carrying initial natural remanent magnetizationmight be influenced by some processes.

Saturation Magnetization (/s)

Saturation magnetization depends basically on theamounts of ferromagnetic components. Values of Js forthe minicore samples range from 0.06 to 2.6 emu/g. Js

of massive lava flows is on average larger than that ofpillow lava in the hole. Microscopic observation indi-cates that hydrothermal alteration probably reduced the/ s , as ferromagnetic minerals (titanomagnetite) were ox-idized toward titanomaghemite and/or replaced clay min-erals. Maghemitization and replacement process of tita-nomagnetite in ordinary natural oceanic basalts are be-

ing studied in more detail by the present authors andwill be presented elsewhere.

Thermal Analyses and Curie Temperatures (rc)

Temperature dependence of the saturation magneti-zation was measured on all samples. A small tip of rock(100 mg) was heated at a rate of 6°C/min., kept at650°C for 10 min., and then cooled down to room tem-perature at the same rate. The cycle of heating and cool-ing was accomplished in a vacuum of 10~2 Pa or less.Two types of thermomagnetic curves were recognized:reversible (Fig. 3A) and irreversible (Fig. 3B, C) ones.The latter is commonly observed in submarine pillowbasalts. The Curie temperatures of these samples rangefrom 440° to 585°C, but most of the Curie temperaturesare distributed between 560° and 580°C, showing a pre-dominance of ferromagnetic minerals of nearly puremagnetite composition. The thermomagnetic behaviorof the samples shows the most ferromagnetic mineralsare more or less oxidized through unmixing process ofTi-rich titanomagnetite to reproduce titanium-poor tita-nomagnetites (Buddington and Lindsley, 1964; 0'Reillyand Banerjee, 1967; Ozima and Larson, 1970; Kinoshi-ta and Aoki, 1972; Nishitani and Kono, 1983).

Comparison of the Leg 83 Section and the Upper Partof Hole 504B

The stable magnetic polarization of rocks from theLeg 83 section of Hole 504B carry predominantly re-verse magnetization, with inclination values similar tothe upper part of this hole (Furuta and Levi, 1983). How-ever, several magnetic properties—intensity of NRM,initial susceptibility, and saturation magnetization—aredifferent between the upper and the lower parts. In par-ticular, Jn of the lower part is one order of magnitude

1.0

200 400

Temperature (°C)

600

0.15

0.10

0.05

200 400

Temperature (°C)

600 200 400

Temperature (°C)

600

Figure 3. Typical thermomagnetic curves. A. Thermally reversible with high Js Sample 504B-133-1, 88-90 cm. B. Thermally irreversible with high Js

Sample 504B-72-2, 105-107 cm. Thermally irreversible with low / s , Sample 5O4B-1O3-1, 51-53 cm.

334

MAGNETIC PROPERTIES AND ALTERATION IN BASALT, HOLE 504B

smaller than that of the upper part. This suggests thatthe alteration of magnetic minerals has been more activein the lower part. The boundary between apparently lowand high alteration layers is considered from mineralog-ical studies to be at around 1000 m BSF (Alt et al., thisvolume). In particular, ferromagnetic minerals in pillowbasalts are considerably influenced by hydrothermal al-teration, because the grain-size of ferromagnetic miner-als in pillow basalts is finer than grain size in those frommassive flows. The Jn of pillow basalts from the lowerpart of the hole is two orders of magnitude smaller thanthat of the upper part. Ferromagnetic minerals in pillowsamples could hardly be analyzed under the reflectionmicroscope (most may have been replaced and changedto clay minerals). Saturation magnetization and initialsusceptibility of pillow basalts are also small.

ALTERATION RELATED TO ROCKMAGNETIZATION

It is well accepted that titanomagnetite in submarinebasalts undergoes more or less a maghemitization fairlyshortly after the intrusion (or eruption); in some casesthe time involved is much less than a few Ma. Factors orcatalysts of this process have long been investigated butare not completely understood yet. Some researcherssuggest that the maghemitization is a product of thelow-temperature oxidation reaction of stoichiometric ti-tanomagnetite. The process seems to be activated by ap-plying mechanical or thermal external energies in somecases (Ozima and Sakamota, 1971; Ozima et al., 1974).However, recent experimental studies of maghemitizationby means of direct, quantitative, oxygen analysis using amicrofocused electron probe microanalyzer revealed thatthe maghemitization process is much more complicatedphenomenon than it has been thought to be (Furuta,Otsuki, and Akimoto, in prep.). It is suggested that themaghemitization process starts either from stoichiome-tric titanomagnetite (i.e., high-temperature stable phaseas noted by Buddington and Lindsley, 1964) or in a re-verse manner, starting from a state somewhere betweenpseudobrookite and hematite (Oshima, pers, comm.;Akimoto, Kinoshita, and Furuta, in prep). In the caseof naturally occurring maghemite these processes mightoccur simultaneously because of the possibility that sta-ble, solid phases of titanomagnetite and hemoilmenitecoexist when the basaltic magma is quenched after erup-tion.

The oxidation and/or reduction process seems to behighly influenced by environmental conditions controlledby external factors, such as hydrothermal alterations ofmineral assemblages. Magnetization of ferromagneticminerals is a kind of secondary physical parameter whichis highly influenced by a change in physical and chemi-cal conditions of the magnetized body. As a result of thefeeble pinning mechanisms of ferromagnetic domainstructures within the multidomain size of ferromagneticcrystals, the natural remanent magnetization may varyeven with a scarcely identifiable change in the internalconditions of a magnetized body. The magnetization ofa minicore-sized, seemingly homogeneous rock sample

may change its character from spot to spot, sometimeseven on a microscopic scale. Therefore, the present au-thors have investigated the alteration stages of the indi-vidual samples used for rock magnetic study, in order tounderstand the variations of rock magnetic propertiesof the formation with respect to depth. Our presentmineralogical analyses should supplement the detailedwork of Alt et al. (this volume). Here we will only brief-ly describe the results of analyses of mineral alterationfound in the individual rock magnetic minicores.

Our techniques of analysis were as follows:Basalt and dolerite were used for microscope obser-

vation, mineral chemistry, and bulk analyses. Constitu-ent minerals of altered rocks and veinlets were identifiedunder the microscope, and by means of X-ray diffrac-tion analysis (XRD) and electron microprobe analysis(EPMA).

The basalt samples were washed in doubly distilledwater for analyses of trace elements (heavy metals) byX-ray fluorescence (XRF) and Ion Coupled Plasma(ICP). Powder samples were dried at 1000°C for morethan two hours before the XRF analysis. One rock pow-der was mixed with 5g di-lithium tetraborate anhydrous(Li2B407); 200 mg of lithium bromide (LiBr) were addedand the mixture was heated at about 1100°C in a plati-num crucible. When the mixture had melted homogene-ously, it was quenched. For ICP, 100 mg of powder sam-ple was decomposed by 10 ml of hydrofluoric acid (HF)and 5 ml of sulfuric acid (H2SO4). The solution was di-luted with 5 ml of condensed nitric acid (HNO3) and bi-distilled water to adjust the total volume to exactly100 ml.

OlivineOlivine phenocrysts remain partly fresh down to 850

m BSF but most of them are altered to secondary miner-als, probably saponite. Olivine phenocrysts are alteredto chlorite from 890 to 1350 m BSF, and to chlorite, ser-pentine, or talc from 1200 to 1350 BSF (Fig. 4).

GlassMost of the glasses are altered to green-brown-col-

ored minerals (probably montmorillonite) above 890 mBSF. Fresh glass, part of which is altered to montmoril-lonite, remains fresh down to about 880 m BSF. Glass isaltered to chlorite, which is one of the typical mineralsof greenschist facies, from 890 to 1350 m BSF and to ac-tinolite from 970 to 1350 m BSF. Talc and serpentine ap-pear from about 1150 to 1350 m BSF (Fig. 4).

FeldsparsPlagioclase is dominant in both phenocrysts and in

groundmass. Plagioclase remains fresh from 274 to 890m BSF. There are many microcracks filled with clay min-erals in plagioclase phenocrysts. Hydrothermal altera-tion is more intensive from 890 to 1100 m BSF. Twinningis discernible in unaltered phenocrysts but hydrothermalalteration destroys it. Some relict microlites and pheno-crysts remain fresh. Plagioclase remains rather fresh from1100 to 1350 m BSF (Fig. 4).

335

H. KINOSHITA, T. FURUTA, H. KAWAHATA

1 I I ε •e

O o o CΛ

800 r

900

1000

g 1100

a. 1200oQ

1300

1400 I-

I I ! 1

j

Figure 4. A schematic representation of the basaltic layer and the distribution of original minerals and sec-

ondary minerals in Leg 83 samples. Zeolite is stilbite or laumonite.

Clinopyroxene

Clinopyroxene phenocrysts are less altered than glassand other minerals. Clinopyroxene phenocrysts remainfresh from 274 to 970 m BSF but their rims are partly re-placed by actinolite from 970 to 1350 m BSF (Fig. 4).

Veins

Dominant vein minerals are chlorite, laumonite, quartz,epidote, and calcite. Subordinate minerals are actino-lite, pyrite, chalcopyrite, and sphalerite (Fig. 4).

Content of Heavy Metals

Trace element (heavy metal) analyses are presented inTable 3. The data are also shown graphically in Figure 5.The concentration of zinc is fairly constant, ranging from57 to 95 ppm, both from 800 to 890 m and from 1050 to1350 m BSF, but it is variable from 890 to 1050 m BSF,ranging from 48 to 2970 ppm.

Manganese shows the same distribution pattern withdepth as zinc. Manganese content varies between 0.1 and0.15 wt.% from 800 to 890 m and from 1050 to 1350 mBSF, but it varies between 890 and 1050 m BSF, rangingfrom 0.1 to 0.36 wt.%.

Cobalt content shows small variation (18-48 ppm,with an average of 39 ppm). This suggests that the upperpart of the Leg 83 section of Hole 504B contains a rela-tively constant concentration of cobalt, and that cobaltis immobile in the reactions between basalt and hot sea-water. Although nickel content varies from 8 to 141 ppmand averages 77 ppm, no systematic variation of nicklecontent can be observed.

Alteration Temperature Zones

The alteration temperature distribution is deducedfrom the sequential appearance of actinolite, chlorite,and epidote. Rocks from 274 to 890 m BSF remain freshor are altered by low-temperature water/rock interaction,as shown by occurrence of low-temperature alteration

Table 3. Heavy metal composition, Hole 504B, Leg 83.

Core-Section(interval in cm)

Sub-bottomdepth

(m)Mn

(wt.%)Zn Co Ni

(ppm)

71-1, 127-12972-4, 38-4077-2, 76-7879-3, 145-14780-1, 19-2180-2, 27-2982-1, 21-2383-1, 33-3583-2, 4-684-2, 86-8885-1, 57-5986-1, 13-1589-1, 40-4289-2, 4-790-1, 109-11190-2, 101-10490-3, 65-6791-2, 68-7092-2, 32-3493-3, 32-3494-2, 109-11194-3, 82-8496-1, 15-1797-3, 9-1199-1, 69-71101-2, 90-92104-3, 27-29107-1, 99-101

, 62-6418-24121-123

122-1, 132-134122-2, 17-19123-1, 52-54132-1, 17-19133-1, 80-82133-2, 17-19

108-1,111-1,117-1,

837.28848.39890.77908.96910.20911.78928.72937.84939.05948.87956.08964.64985.91987.06995.60997.03998.26

1004.191014.331024.831033.101034.331048.661057.601072.201091.911119.781144.501153.131161.711190.721214.831215.181223.031295.181304.811305.68

0.1290.1210.1520.1670.1950.2430.2520.2490.1950.1760.3590.1690.1480.1860.1210.1750.1870.2450.1680.1560.1310.1390.1310.1240.1490.1410.1100.1320.1350.1340.1410.1150.1130.1220.1330.1470.145

76

71

102

98

104

1290

115

129

103

89

2970

102

201

89

82

87

146

101

95

81

77

79

48

57

87

74

60

77

77

71

95

72

71

78

76

84

79

41

45

47

39

32

32

35

35

39

42

36

37

46

43

18

42

38

31

39

43

43

40

31

37

42

37

36

40

40

41

41

42

40

44

48

44

41

71

48

55

40

68

96

131

73

94

116

73

45

36

89

25

141

121

97

80

86

120

131

72

77

152

100

27

116

64

101

52

50

8

49

59

46

36

minerals. Preliminary oxygen isotopic work suggests thatthese minerals formed at temperatures between 2 and110°C (Honnorez et al., 1983), which may be the maxi-mum temperature for basalt alteration and is probably

336

MAGNETIC PROPERTIES AND ALTERATION IN BASALT, HOLE 504B

100

Zn(ppm)

200 300 400 0 0.1

Mn(wt.%)

0.2 0.3 0.4 0 100

Cu(ppm)

200 300 0

Co(ppm)

100 0

Ni(ppm)

100 200

•• v •w•

•

*I I I // I

I 1 1 1

•

\

•

%*

• X

1 1 ,/ 1

•

< *

t

•M

1

••

•

1 1

Alterationzone

Sediments:

800

900 -

-^ 1000 -

| 1100 -

_o

s.

Q 1200

1300 h

1400

Figure 5. Downhole variation in concentrations of heavy metals in Leg 83 samples.

the maximum temperature reached above 890 m BSF(see Alt et al., this volume).

Chlorite and laumontite appear in veins below 890 mBSF. Many epidote crystals are found in the veins below910 m BSF. Actinolite is observed below 970 m BSF.Chlorite first appears between 230° and 280°C, replac-ing smectite at higher temperatures, and epidote formsat temperatures as low as 200°C and becomes a majorphase above 260° to 270° C in the Reykjanes geothermalsystem (Tomasson and Kristmannsdottir, 1972, Kist-mannsdottir, 1976). The alteration temperature of actin-olite-bearing assemblages is higher than 320°C in landgeothermal systems (see discussion in Alt et al., this vol-ume).

The alteration temperature deduced from these factsis shown in Figure 6. It seems that the alteration temper-ature increase occurred sharply between 890 and 950 mBSF.

Our data suggest that hydrothermal alteration inHole 504B can be divided into three zones, based on thedistribution of veins, appearance of secondary minerals,and content of trace elements (heavy metals).

Zone 1 (From 274 to 890 m BSF)In Zone 1 rocks are either unaltered or altered by low-

temperature (0-110°C) water-rock interaction. Metal con-tents show no depth dependency in this zone.

Zone 2 (From 890 to 1050 m BSF)Alteration temperature is high. Hydrothermal solu-

tions passed through this zone and reacted with rocks.Vein materials precipitated and some samples show highzinc and manganese contents.

Zone 3 (From 1050 to 1350 m BSF)Alteration temperature in this zone was high. A large

volume of hydrothermal solution probably did not pass

100

Temperature (°C)

200 300 400

200-

400

600-

800-

1000-

1200-

1400-

.-Present loggingtemperature

\\\\\

\

\\\ ::::S:::̂ :

Epidote

. Actinolite

Figure 6. The alteration temperature and present logging temperatureof Hole 504B as a function of depth. Alteration temperature is asfollows. 274-540 m BSF: 2-30°C; 540-890 m BSF: 60-110°C; 890-1350 m BSF: 200-350°C. L = laumontite mineral assemblage(laumonite-chlorite-epidote-albite-quartz). A = actinolite mineralassemblage (actinolite-chlorite-epidote-albite-quartz).

337

H. KINOSHITA, T. FURUTA, H. KAWAHATA

through this zone, however, as few veins are observed inour samples. If a vast volume of hydrothermal solutionhad passed through the cracks, hydrothermal alterationwould have been extensive along the cracks and somedeposits would be observed. Metal contents show no depthdependency in this zone.

DISCUSSION AND CONCLUSIONSOur findings about the appearance of secondary

minerals are somewhat different from the results ob-tained by Alt et al. (this volume). Chlorite appears be-low 890 m BSF in our analysis and below 846 m BSF inAlt et al. (this volume). This discrepancy may be due ei-ther to different analytical methods or to the sparsity ofour rock magnetic samples. Some of the montmorilloni-tes which appear above 890 m BSF were apparently iden-tified as chlorite if observed only under the microscope,but the chemical composition of montmorillonites andchlorite can easily be distinguished by means of EPMA.

It seems clear that the hydrothermal circulation be-neath the seafloor of the Site 504 has caused a sequenceof basalt alteration with greenschist facies in the deeperpart of the hole. The temperature estimated from sec-ondary minerals increases sharply between 890 and 950m BSF. The alteration that occurred is probably due tostrong hydrothermal fluid circulations.

As a result of hydrothermal circulation, magnetic min-erals (titanomagnetite; (0.5-0.7)Fe2TiO4 - (0.5-0.3)Fe304;data from Furuta, 1983) first changed to titanomagne-tite and then formed secondary Ti-poor titanomagnetiteand pyrrhotite or pyrite. The natural remanent magneti-zation below 900 m BSF seems to have been remagne-tized and does not reveal any dependable primary value.As a result of the hydrothermal reaction of ferromagneticminerals, the natural remanent magnetization shows littleindication of the original magnetization. More detailedanalysis of the hydrothermal field and its effect on geo-

logic formations of this area will be needed for furtherdetailed rock magnetic discussion.

REFERENCES

Buddington, A. E , and Lindsley, D. H., 1964. Iron-titanium oxideminerals and synthetic equivalents. /. Petrol., 5:310-357.

Furuta, T., 1983. Magnetic properties of basalt samples from Holes504B and 5O5B on the Costa Rica Rift, Deep Sea Drilling ProjectLegs 69 and 70. In Cann, J. R., Langseth, M. G., Honnorez, J.,Von Herzen, R. P., White, S. M., et al. Init. Repts. DSDP, 69:Washington (U.S. Govt. Printing Office), 711-720.

Furuta, T., and Levi, S., 1983. Basement paleomagnetism of Hole504B. In Cann, J. R., Langseth, M. G., Honnorez, J., Von Herzen,R. P., White, S. M., et al., Init. Repts. DSDP, 69: Washington(U.S. Govt. Printing Office), 697-703.

Honnorez, J., Laverne, C , Hubberten, H. W., Emmermann, R., andMuehlenbachs, K., 1983. Alteration processes in Layer 2 basaltsfrom Deep Sea Drilling Project Hole 504B, Costa Rica Rift. InCann, J. R., Langseth, M. G., Honnorez, J., Von Herzen, R. P.,White, S. M., et al., Init. Repts. DSDP, 69: Washington (U.S.Govt. Printing Office), 509-546.

Kinoshita, H. and Aoki, Y., 1972. The stability of NRM of DeccanTraps of basalts in India. J. Geomag. Geoelectr., 22:459-475.

Kristmannsdottir, H., 1976. Types of clay minerals in hydrothermallyaltered basaltic rocks, Reykjanes, Iceland. Jokull, 26:30-39.

Nishitani, T., and Kono, M., 1983. Curie temperature and lattice con-stant of oxidized titanomagnetite. Geophys. J. R. Astron. Soc,74:585-600.

O'Reilly, W., and Banerjee, S. K., 1976. The mechanism of oxidationin titanomagnetites; a magnetic study. Min. Mag., 36:29-37.

Ozima, M., Joshima, M., and Kinoshita, H., 1974. Magnetic propertiesof submarine basalts and the implications on the structure of theoceanic crust. /. Geomagn. Geoelectr., 26:335-354.

Ozima, M., and Larson, E. E., 1970. Low- and high-temperature oxi-dation of titanomagnetite in relation to irreversible changes in themagnetic properties of submarine basalts. J. Geophys. Res., 75:1003-1017.

Ozima, M., and Sakamoto, N., 1971. Magnetic properties of syn-thesized titanomaghemite. J. Geophys. Res., 67:7035-7046.

Tomasson, J., and Kristmannsdottir, H., 1972. High temperature al-teration minerals and thermal brines, Reykjanes, Iceland. Contrib.Mineral. Petrol., 36:123-134.

Date of Receipt: 28 September 1983Date of Final Acceptance: 29 February 1984

338