52. PETROCHEMISTRY OF BASALTS AND PLUTONIC ROCKS, LEG 37, DEEP SEA DRILLING PROJECT Leonid Dmitriev, Institute of Geochemistry, Academy of Sciences of the USSR, Moscow, USSR INTRODUCTION During Leg 37 oceanic basement was drilled at four sites (332, 333, 334, and 335) on the west flank of the Mid Atlantic Ridge at approximately 37°N latitude. This report deals with the basaltic rocks recovered from Hole 332B and with a plutonic sequence encountered at Site 334. Hole 332B was drilled about 30 km from the ridge crest in basement approximately 3.5 m.y. old. This hole penetrated 583 meters of acoustic basement with ap proximately 20% recovery. The basaltic sequence was divided by the shipboard party into 11 lithologic units largely on the basis of phenocryst content. At Site 334 a plutonic complex was encountered beneath approximately 50 meters of basalt. The com plex is a layered sequence of gabbro, troctolite, and lherzolite about 67 meters thick. Numerous breccia zones consist of plutonic clasts set in a matrix of nan nofossil chalk. BASALTIC ROCKS OF HOLE 332B Two hundred seventy six analyses of fresh basalt with CO2 less than 1 wt % were selected from Table 10B, Chapter 2 (this volume?. These were plotted against depth, and paired correlations were calculated for the major oxides. A special computer program was used for the calculations which divided all analyses into two principal groups distinguished by different correla tion trends among the major oxides. The differences between the two groups have been demonstrated to be statistically significant by A.V. Garanin (Institute of Geochemistry, USSR Academy of Sciences). Analytical differences between different laboratories (see Wright, this volume) are distinctly less than the differences between the two groups. The paired correlations are shown in Figure 1 (a g) where the different analyses are identified by subbottom depth. These diagrams clearly show the existence of the two main magma types and provide evidence that the observed chemical variations are due chiefly to differentiation of these two types. Magma group 1 is characterized by high MgO, relative ly high total iron (FeO), and low SiO 2 and CaO. TiO 2 and AI2O3 exhibit a positive correlation, but MgO has a negative correlation with all other oxides except (FeO). The concentration of (FeO) probably does not depend on MgO content. Chemical variations in magma group 1 reflect frac tionation and accumulation of olivine or olivine + spinel. Petrographically, these basalts range from coarsely phyric picrites to aphyric types. Magma group 2 has relatively high concentrations of Siθ2 and CaO and low MgO and (FeO). Wide variations in AI2O3, CaO, and especially (FeO) and Tiθ2 characterize this group. In contrast to group 1 magmas, Tiθ2 and AI2O3 show a negative correlation and (FeO) and Tiθ2 show a positive correlation (Figure If, lg). iThe chemical variations within magma group 2 are probably related to plagioclase fractionation and accumulation. Two subgroups (2a and 2b) are recognized on the basis of chemistry and mineralogy. Group 2a magmas have higher concentrations of Siθ2, (FeO), and TiO 2 and lower AI2O3 and CaO than those of group 2b. Rocks of group 2a are aphyric or sparsely phyric with phenocryst assemblages plagioclase + olivine + clinopyroxene. Subgroup 2b consists chiefly of coarsely plagioclase phyric basalt. The distribution of group 1 and group 2 magmas with depth is shown schematically in Figure 2, along with shipboard lithologic units and paleomagnetic units. It can be seen that there is generally a close cor respondence between chemical type and lithologic and paleomagnetic units. There is also evidence for cyclic variations in magma chemistry. All of the rocks are belived to be related to two main eruptive cycles, I and II. Both cycles contain all three recognized magma groups, and consist of several subcycles generally beginning with magmas of group 1 and ending with magmas of group 2b. Only the last subcycle is represented completely; the others normally terminate with magmas of group 2a (Figure 2). The boundary between the two main cycles is marked by alteration of the basalts (high CO2 contents) and by interlayering of basalts and sediments at the end of the first cycle. Generally the thickness and abundance of sedimentary interbeds increase upward not only in the section as a whole, but also within individual cycles and subcycles. In most of the cycles and subcycles the different magma types occur in discrete units. However, in the lower part of subcycle I2, group 1 and group 2a magmas are intimately interlayered. This "interlayered com plex" is succeeded upward by magmas of group 2b and appears to be part of a normal cycle. In general the thickness of individual flows and the volume of lava in individual cycles appears to increase upward in the section. Experimental petrology suggests that magmas of group 1 should form by partial melting of mantle material at depths of 40 to 60 km and magmas of group 2 should form at depths of about 20 km. The frequent eruption of these two distinct magma types suggests a 681
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
52. PETROCHEMISTRY OF BASALTS AND PLUTONIC ROCKS,LEG 37, DEEP SEA DRILLING PROJECT
Leonid Dmitriev, Institute of Geochemistry,Academy of Sciences of the USSR, Moscow, USSR
INTRODUCTION
During Leg 37 oceanic basement was drilled at foursites (332, 333, 334, and 335) on the west flank of theMid-Atlantic Ridge at approximately 37°N latitude.This report deals with the basaltic rocks recovered fromHole 332B and with a plutonic sequence encountered atSite 334.
Hole 332B was drilled about 30 km from the ridgecrest in basement approximately 3.5 m.y. old. This holepenetrated 583 meters of acoustic basement with ap-proximately 20% recovery. The basaltic sequence wasdivided by the shipboard party into 11 lithologic unitslargely on the basis of phenocryst content.
At Site 334 a plutonic complex was encounteredbeneath approximately 50 meters of basalt. The com-plex is a layered sequence of gabbro, troctolite, andlherzolite about 67 meters thick. Numerous brecciazones consist of plutonic clasts set in a matrix of nan-nofossil chalk.
BASALTIC ROCKS OF HOLE 332B
Two hundred seventy-six analyses of fresh basaltwith CO2 less than 1 wt % were selected from Table10B, Chapter 2 (this volume?. These were plottedagainst depth, and paired correlations were calculatedfor the major oxides. A special computer program wasused for the calculations which divided all analyses intotwo principal groups distinguished by different correla-tion trends among the major oxides. The differencesbetween the two groups have been demonstrated to bestatistically significant by A.V. Garanin (Institute ofGeochemistry, USSR Academy of Sciences). Analyticaldifferences between different laboratories (see Wright,this volume) are distinctly less than the differencesbetween the two groups. The paired correlations areshown in Figure 1 (a-g) where the different analyses areidentified by subbottom depth. These diagrams clearlyshow the existence of the two main magma types andprovide evidence that the observed chemical variationsare due chiefly to differentiation of these two types.Magma group 1 is characterized by high MgO, relative-ly high total iron (FeO), and low SiO2 and CaO. TiO2
and AI2O3 exhibit a positive correlation, but MgO has anegative correlation with all other oxides except (FeO).The concentration of (FeO) probably does not dependon MgO content.
Chemical variations in magma group 1 reflect frac-tionation and accumulation of olivine or olivine +spinel. Petrographically, these basalts range fromcoarsely phyric picrites to aphyric types.
Magma group 2 has relatively high concentrations ofSiθ2 and CaO and low MgO and (FeO). Widevariations in AI2O3, CaO, and especially (FeO) andTiθ2 characterize this group. In contrast to group 1magmas, Tiθ2 and AI2O3 show a negative correlationand (FeO) and Tiθ2 show a positive correlation (FigureIf, lg). iThe chemical variations within magma group 2are probably related to plagioclase fractionation andaccumulation.
Two subgroups (2a and 2b) are recognized on thebasis of chemistry and mineralogy. Group 2a magmashave higher concentrations of Siθ2, (FeO), and TiO2
and lower AI2O3 and CaO than those of group 2b.Rocks of group 2a are aphyric or sparsely phyric withphenocryst assemblages plagioclase + olivine +clinopyroxene. Subgroup 2b consists chiefly of coarselyplagioclase-phyric basalt.
The distribution of group 1 and group 2 magmaswith depth is shown schematically in Figure 2, alongwith shipboard lithologic units and paleomagneticunits. It can be seen that there is generally a close cor-respondence between chemical type and lithologic andpaleomagnetic units. There is also evidence for cyclicvariations in magma chemistry. All of the rocks arebelived to be related to two main eruptive cycles, I andII. Both cycles contain all three recognized magmagroups, and consist of several subcycles generallybeginning with magmas of group 1 and ending withmagmas of group 2b. Only the last subcycle isrepresented completely; the others normally terminatewith magmas of group 2a (Figure 2). The boundarybetween the two main cycles is marked by alteration ofthe basalts (high CO2 contents) and by interlayering ofbasalts and sediments at the end of the first cycle.Generally the thickness and abundance of sedimentaryinterbeds increase upward not only in the section as awhole, but also within individual cycles and subcycles.
In most of the cycles and subcycles the differentmagma types occur in discrete units. However, in thelower part of subcycle I2, group 1 and group 2a magmasare intimately interlayered. This "interlayered com-plex" is succeeded upward by magmas of group 2b andappears to be part of a normal cycle.
In general the thickness of individual flows and thevolume of lava in individual cycles appears to increaseupward in the section.
Experimental petrology suggests that magmas ofgroup 1 should form by partial melting of mantlematerial at depths of 40 to 60 km and magmas of group2 should form at depths of about 20 km. The frequenteruption of these two distinct magma types suggests a
681
L. DMITRIEV
52α
51
50
49
uπuo 7: + •
OOnJ O]°lnθ ' •* . .
Ml o OoS λtm'm ; 40 0 0) 0
A . >v..
Symbol
•
•
•
0
D
V
+
0
o
Subbottomdepth(m)
615-665
585-595
530-540
405-495
665-720
615-665
540-585
495-530
255-405
595-615
140-255
Numberof
analyses
12
11
4
74
24
10
7
11
72
16
35
Magma
type
1
2a
2b
Oceanic tholeiite (200 an.)
48 - O
47 -
4610 14 16 18 20 22 24 26
MgO
Figure la. Paired correlations between some major elements in basalts from Hole 332B.
high degree of tectonic activity in the area at the time offormation. Because of the tectonic instability, magmasmoved rapidly to the surface rather than accumulatingin large chambers where they would undergo extensivecrystal fractionation. Most probably the observeddifferentiation in these magmas is due to crystallizationof superheated magma during its rise to the surface andseparation of crystals by flow processes.
In general, the thickness of lithologic units, the abun-dance of sedimentary interlayers, and the volume ofmagma erupted in a given cycle jappears to increase up-ward in the section, suggesting that magmatic and tec-tonic activity in the area tapered off gradually ratherthan ending abruptly.
In Figure 3 basalts from Hole 332B are comparedwith basalts dredged from 67 stations along ridge crestsin the Atlantic, Pacific, and Indian oceans (Sharaskin etal., in press) and with continental tholeiites (Manson,1967). Compared to other oceanic basalts, those from
Hole 332B are distinctly lower in TiC>2, (FeO), andNa2θ and higher in CaO and AI2O3. Compared to con-tinental olivine tholeiites, Hole 332B basalts are moreuniform in composition and are lower in (FeO) andK2O and higher in CaO and AI2O3. It is also apparentthat some 332B basalts are significantly enriched inMgO, CaO, and AI2O3, reflecting accumulation ofolivine or plagioclase. It is interesting that the unusualcharacter of these rocks is not reflected in the other ox-ides.
PLUTONIC COMPLEX AT SITE 334
The plutonic complex at Site 334 has been studied indetail because it may shed light on the composition andorigin of oceanic Layer 3. Approximately 67 meters ofinterlayered gabbro, troctolite, and lherzolite werepenetrated before the hole was terminated. The averagechemical composition of each lithologic type is given inTable 1 (columns 2, 3, and 4). Columns 1 and 5 give
682
PETROCHEMISTRY OF BASALTS AND PLUTONIC ROCKS
0.2
0.010 12 14 20 22 2416 18
MgO
Figure lb. Paired correlations between some major elements in basalts from Hole 332B. See Figure la for legend.
TABLE 1Average Composition of Lherzolite, Troctolite, and Gabbro From Site 334 (wt %)
SiO 2
TiO 2
A1 2O 3
F e2°3FeOMnOMgOCaONa2OK 2 O
1
X
49.691.48
15.672.488.050.187.97
11.362.630.23
S
0.990.471.551.870.410.041.090.810.370.14
2
X
50.690.42
16.22
7.280.149.81
13.431.390.11
S
0.480.431.34
2.740.041.771.600.620.10
3
X
48.090.07
10.14
6.760.13
22.8011.100.320.04
S
1.280.012.06
0.740.013.451.270.090.02
4
J
44.470.064.90
9.100.14
37.193.040.120.03
S
0.960.010.64
0.860.012.291.700.040.02
5
X
45.990.163.475.123.810.15
38.632.260.340.07
S
1.680.101.581.261.500.093.121.260.370.06
Note: 1 = Mid-oceanic ridge tholeiite (average of 200 analyses, Sharaskin et al., 2 = Gabbrofrom Site 334 (8 analyses); 3 = Troctolite from Site 334 (8 analyses); 4 = PlagioclaseLherzolite from Site 334 (12 analyses); 5 = Lherzolite from mid-oceanic ridges (69analyses, Dmitriev, et al., 1972); The sum in columns 2, 3, and 4 is less than 100% be-cause Fe 2 θ3 n a s been recalculated to FeO.
683
L. DMITRIEV
24
22
20
18
16
14
12
10
oo
o o
oo°o
•;+ V Λ . A A.A Ay ^ ^ \ A AA A
>W
4 6 8 10 12 14 16 18 20 22 24 26
MgO
Figure lc. Paired correlations between some major elements in basalts from Hole 332B. See Figure la for legend.
28
average analyses for oceanic tholeiite and oceanic lher-zolite, respectively. Figure 4 (a,b) is a graphical com-parison of the plutonic rocks to average oceanictholeiites and lherzolites. It can be seen that Site 334lherzolites are quite similar to average oceanic lher-zolite but have somewhat lower concentrations of SiCh,TiO2, Na2O, and K2O and higher AI2O3 and CaO. Gab-bro from Site 334 is compositionally similar to basaltfrom Hole 332B and has somewhat lower (FeO), TiCh,Na2θ, and K2O and higher AI2O3 and CaO than theaverage oceanic tholeiite. These differences probablyreflect regional differences in the ocean basins.
Troctolites are compositionally intermediate betweengabbro and lherzolite, and one might conclude that allthree rock types in this complex had a common origin.
However, the chemical relationship shown in Figure 4and the range of composition of each type suggest thatthe troctolites are not related to the other rocks by asimple fractionation scheme. Formation of the troc-tolites probably required metasomatic transfer ofNa2O, K2O, TiO2, AI2O3, and CaO.
In Table 2 the average concentration of selected traceelements in gabbro, troctolite, and lherzolite from Site334 are compared with concentrations in averageoceanic tholeiite and lherzolite. Figure 4 shows that Site334 troctolites are significantly enriched in Au, Cu, Ba,and Li compared to the associated gabbros and lher-zolites. These data confirm the conclusion that the troc-tolites were formed independently of gabbros in anopen system.
684
PETROCHEMISTRY OF BASALTS AND PLUTONIC ROCKS
Figure Id. Paired correlations between some major elements in basalts from Hole 332B. See Figure la for legend.
TABLE 2Distribution of Some Trace Elements in Lherzolites, Troctolites,
and Gabbro From Site 334 (all elements in ppm, Au-ppb)
CrVNiCoSrßaLiRbCuAu*
30331411342130235.61.697
0.77
1
(116)(91)(116)(87)(100)(88)(32)(43)(94)(32)
2
1031 (8)131(5)379 (7)66(7)14(7)22(7)5.0 (5)2.0 (5)86(8)4.7 (8)
3
1700 (6)123 (6)820 (8)62(6)10(7)38(6)8.0 (4)2.3 (8)107 (7)30.2 (6)
4
4800 (5)110(8)
1844 (12)96(9)5(9)29(9)6.1 (4)2.1 (9)45 (12)6.3 (8)
5
4400 (37)47 (37)
2500 (37)117(37)
8(16)4.1 (37)0.5 (37)40 (37)1.4(35)
Note: 1 = Mid-oceanic ridge tholeiite (Dmitriev et al., in press); 2 =Gabbro from Site 334; 3 = Troctolite from Site 334; 4 = Plagio-clase Lherzolites from Site 334; 5 = Lherzolite from mid-oceanicridges (Dmitriev et al., in press). Number of analyses is in paren-theses.
685
L. DMITRIEV
1 ö
17
16
15
14
13
12
11
10
9
8
7
O
O o °°onOo o
O^oO Λ nθ8°O
π 0°+°°+ o
+Λ + +
π+ πftA+dW"
v # ; o
0
. i i i ,
o °00
ooπ
/o 0
0
•D°.
•
D
i
o
V A> . •A *i? AA:
' " . A ,
i i
•
> A »
•
I i i i i i i
4
i i i i i i i4 6 8 10 12 14 16 18 20 22 24 26
MgO
Figure le. Paired correlations between some major elements in basalts from Hole 332B. See Figure la for legend.
28
REFERENCES
Dmitriev, L., Ukhanov, A., and Sharaskin, A., 1972. Com-position of the upper mantle: Geochimia, no. 10. (Englishtranslation in Geochem. International, v. 9, p. 781-792).
Manson, V., 1967. Geochemistry of basalts. In Hess, H. , andPoldervaart, A. Ed., Basalts v. 1, New York (Inter-science), in press.
Sharaskin, A., Charin, G., and Dmitriev, L., in press.Geochemistry of mid-ocean ridge basalts.
686
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
PETROCHEMISTRY OF BASALTS AND PLUTONIC ROCKS
0.0
+v
•. . Δ•• •••••π• U,A Φ
0 0
o o o o o 0
oo
, 1
ocf
. 1 1 1
10 12 14 16 18 20 22 24
A1 2 O 3
Figure If. Paired correlations between some major elements in basalts from Hole 332B. See Figure la for legend.
687
L. DMITRIEV
1.6
1.4-
1.2-
1.0-
0.8
0.6
0.4
0.2
0.0
• o
θ o o ° o ^ θ °o °
00
0
o
i
0
A
Oo
)
A
0
o
0
0
1
iΛ
n
•
*
V+
+ v + w^+
+l D D ^ D
• u ••••••
• i* *
• A .• *
, 1
ö++
+D
D
0 000 v
]0+
+ +
•
•
0
on0
v 0
1 1
10 12 14FeO
Figure lg. Paired correlations between some major elements in basalts from Hole 332B. See Figure la for legend.
688
PETROCHEMISTRY OF BASALTS AND PLUTONIC ROCKS
co2 Core Brief petrographic description
500-
550-
600-
650-
700-
1-4 I-II 2BM
1
Coarsely phyric to aphyricplagioclase basalts inter-layered with soft sediments(empty space in section).
6-15 fπA/I I l B
2BM
2
Sparcely phyric to aphyricbasalts (pl>ol>py).Sedimentary layers.
16-25 IV/V 2BM3,4
Coarsely phyric toaphyric olivine-richbasalts.Sediments.
26-28 VI 2BM
4
Sparcely phyric to aphyricbasalts (pl>ol>py).
29 VI 2BM-5 Olivine-rich basalts.
30-34 VI 2BM Sparcely phyric to aphricbasalts (pl>ol>py).
35 V1T Olivine-rich basalts.
36-37 VIII 2BM7
Coarsely phyric to aphyric plagioclasebasalts. Numerous sedimentary layers.
38-42 IX/XInterlayered ,olivine-richbasalts and basalts with(pl>ol>py).
43-48 X/XISparsely phyric to aphyricbasalts. Microdoleritic basalts(pl>ol>py).
Figure 2. Schematic section of Hole 332B showing the distribution of lithologic and chemical units.
689
L. DMITRIEV
40-
30-
20-
10-
0-
/ \ A
/ V i/ I *
1 \ \1 '1 \1 ' 1 ^/ /I \
/ * i V' ^r •' / \ • λ " .
• / \ ^ \ •"
TiO2
i i i i' —
-
/
/
//
A• j//
\
4\
CaO
10 12 14 16 18 20
0.4 1.2 2.0 2.8 3.6 4 . 0
12 14 16 18 20 22
1.4 2.2 3 . 0 3 . 8 4 . 6 5 . 4
10 12 14 16 18 20
1.
I \
3.
Oxides (wt. %)
Figure 3. Histograms of the major elements in oceanic and continental basalts. 1 = Basalts from Hole 332B;2 = Mid-ocean ridge basalts (200 analyses) (Sharaskin et al, in press); 3 = Continental olivine tholeiites(182 analyses) (Manson, 1967).
690
PETROCHEMISTRY OF BASALTS AND PLUTONIC ROCKS
A1 2 O 3 ; FeO; CaO
15 H
10-
SiO,
50-
45-
5-
O O SiO,
A12°3
II
× FeO
A CaO
I I I OL
q
10MgO wt%
T40
Figure 4a. Correlation between MgO and other major elements in basalts, gabbros, troctolites, and Iherzolites. Data fromTable 1. I = Gabbro from Site 334; II = Trocolite from Site 334; HI = Lherzolite from Site 334; OT = Average oceanictholeite; OL = Average oceanic Iherzolite. Ovals show the standard deviation.
691
L. DMITRIEV
'O
J OT
2.5-
2.0-
1.5-
1.0-
0.5-
0.0
+ + TTO,
\
- i s , .
~r
Na20
D
\
\
\II
\
\
\
A
- f e -10
120
1 r30
0L
III
-i——r40
MgO vit%
Figure 4b. Correlation between MgO and other major elements in basalts, gabbros, troctolites, and Iherzo-lites. Data from Table 1. I = Gabbro from Site 334; II = Troctolite from Site 334; III = Lherzolite fromSite 334; OT = Average oceanic tholeiite; OL = Average oceanic Iherzolite. Ovals show the standarddeviation.
692
PETROCHEMISTRY OF BASALTS AND PLUTONIC ROCKS
MgO vt%
Figure 5. Correlation between MgO and some trace elements in basalts, gabbros, troctolites, and Iherzolites. Data from Table2. I = Gabbro from Site 334; II = Troctolite from Site 334; III = Lherzolite from Site 334; OT = Average oceanic tholeiite;OL = Average oceanic lherzolite.
693
Related Documents
