Rare-earth element geochemistry of volcanic rocks from the ...
and Thiel Mountains. Antarctic Map Folio Series, 12: sheet 5. New York, American Geographic Society. Winkler, H. G. F. 1974. Petrogenests of Metamorphic Rocks, 3rd edition. New York: Springer-Verlag. 320p. Zeck, H. P. 1970. An erupted migmatite from Cerro del Hoyazo, SE Spain. Contributions to Mineralogy and Petrology, 26: 225-246.
Rare-earth element geochemistry of volcanic rocks from the Executive Committee Range, Marie Byrd Land
WESLEY E. LEMASURIER Department of Geology University of Colorado Denver, Colorado 80202 PHILIP R. KYLE Institute of Polar Studies The Ohio State University Columbus, Ohio 43210 PETER C. RANKIN Soil Bureau Department of Scientific and Industrial Research Lower Hutt, New Zealand Volcanoes of the Executive Committee Range (1260 W., 760 to 77°S.) were first described by Doumani (1964), and further work has been reported by LeMasurier and Wade (in press). The range is the largest and most petrologically varied volcanic range in Marie Byrd Land; it includes representatives of essentially all the rock types present in the entire petrographic province. Lavas range in composition from alkali basalt and basanite to intermediate compositions such as hawaiite, mugearite, and benmoreite. An extremely wide range of salic rock types, paralleling those in the African rift valleys, also occur and include phonolite, quartz trachyte, alkali rhyolite, comenditic trachyte, comendite, and pantellerite (figure 1). Strontium isotope measurements (Halpern, 1970; Jones and December 1976
Walker, 1972; LeMasurier and Wade, in press) on basaltic, intermediate, and some salic lavas are similar, suggesting they are from a mantle-derived source and may be comagmatic. However, it is difficult to derive this great variety of oversaturated and undersaturated salic rocks from the nepheline normative alkali basalt magma that seems to be the only primary magma type available in the province. Several combinations of high- and low-pressure petrologic processes seem to be called for. Rareearth element (REE) studies of these rocks have focused on the Executive Committee Range because of the variety of rocks represented and the fact that field and chronologic relations are better displayed here than anywhere else in Marie Byrd Land. Thirteen lava samples from the Executive Committee Range together with an alkali basalt, basanite, phonolite, and pantellerite from other parts of Marie Byrd Land were analyzed by Dr. Rankin for REE, cesium, barium, hafnium, lead, thorium, and uranium using spark source mass spectrometry. The analytical technique and precision of the method are given by Howorth and Rankin (1975). REE analyses are plotted normalized to the chondrite abundances given by Price and Taylor (1973). Interpretation of the results is in progress. Results and some preliminary conclusions are discussed below. Basanite (figure 2). The two samples have similar chondrite normalized patterns that show a small positive europium anomaly. Compared to the alkali basalt (figure 3) the basanites are enriched in REE. The REE distribution of the basanites is typical of undersaturated basaltic magmas, which are considered to form by partial melting of a garnet peridotite mantle (Kay and Gast, 1973). Alkali basalt-hawaiite-mugearite-benmoreite (figure 3). There is an increase in REE uranium, thorium, and barium through this series of rocks; an exception is benmoreite 23A, discussed separately below. All samples have a positive europium anomaly. Alkali basalt and basanite (see above) generally have small positive europium anomalies inherited from partial melting and fractionation processes operating in the mantle (Kay and Gast, 1973; Sun and Hanson, 1975). The europium anomalies in the hawaiite, mugeraite, and benmoreite (50A) probably constitute a feature that in turn was inherited from the alkali basalt parent. The progressive increase in REE through the series is consistent with an origin by fractional crystallization from an alkali basalt parent. Benmoreite sample 23A has an exceptionally 263
16
I I I I I I • 54a 71A 2Qd
C
23Ca• 24B4%29 42a
U
a)
0.
23A •70 • 22d 41A
2A • 4.5c •. 32
•48
62 0'44
48 52 56 60 64 68 72 S10 2
(wt. percent)
Figure 1. Alkalis versus Si0 2 diagram of analyzed lavas.
,1.If
1OC
3)
;5c
10 264
La Ce Pr Nd
Sm E Gd T Dy Ho Er
Figure 2. Chondrite norYb malized REE plot of basanite lavas.
ANTARCTIC JOURNAL
I, a, -o
C
0
U 50 a, CL
E
0 'I,
III
Figure 3. Chondrite normalized REE plot of alkali basalt-hawaiite-mugearitebenmoreite lavas.
5
I I I I
La Ce Pr Nd
large positive europium anomaly (EuIEu* = 1.791) and a high barium content. An anomaly of the magnitude observed in 23A is unusual and difficult to explain. The data suggest addition or accumulation of large amounts of feldspar, but the sample is nearly aphyric and shows no evidence of feldspar phenocrysts that could be cumulus in origin. Lowor high-pressure fractionation processes are unlikely to account for the observed europium anomaly. Quartz trachyte (figure 4). Two samples show extremely different REE patterns. Sample 24B has low barium (less than 50 parts per million) and a large negative europium anomaly (EuIEu* =
l EuIEu* = the measured concentration divided by the concentration estimated by interpolating the chondrite normalized samarium and gadolinium values.
December1976
Sm E Gd T D Ho Er
Y
0.37) indicative of feldspar fractionation. The other sample (42A) has an extremely high barium (1,600 parts per million) and a small positive europium anomaly (Eu/Eu* = 1.08); petrogenesis of this sample did not involve feldspar fractionation. Phonolite (figure 5). Three phonolite samples have similar REE patterns except for varying europium contents, and they must have evolved by similar processes. Feldspar fractionation in slightly different abundances would account for the varying negative europium anomalies. Sample 54A is the most fractionated with the lowest barium, highest uranium, thorium, lead, and largest europium anomaly, yet it has a total REE content intermediate between the other two samples. It may have formed from a parent with a lower REE content than samples 20D and 71A. Relative to the analyzed alkali basalt and basanite, some phonolites show a greater enrichment of the heavy REE compared to the light REE. As frac-
265
1000
500
- 100 L)
E In
5C
1
10 I I I
1 111111
I
L Ce P r Nd So E0 Gd T D Ho E, Y
- La
Co Pr Nd
Sm E Gd Tb Dy Ho E r Yb
Figure 4. Chondrite normalized REE plot of quartz trachyte lavas.
Figure 5. Chondrite normalized REE plot of phonolite lavas.
tionation of most mineral phases present in these lavas will not result in a greater enrichment of the heavy REE compared to the light REE, it appears that the phonolites were not derived from a parent like the analyzed alkali basalt or basanite. A basanite with a higher abundance of light REE, or a lower abundance of heavy REE, would be a suitable parent (compare basanite lavas from the McMurdo Sound area) (Sun and Hanson, 1975, 1976; Kyle and Rankin, 1976). Salic lavas (figure 6). Characteristic of all these lavas is their variable chondrite normalized patterns and their large negative europium anomalies (EuIEu* ranges from 0.20 to 0.65). Substantial quantities of feldspar fractionation must have occurred during the formation of these lavas. The parents and processes of fractionation are under investigation.
References
W. E. LeMasurier collected the samples during fieldwork supported by National Science Foundation grant DPP 70-02980. 266
Doumani, G. A. 1964. Volcanoes of the Executive Committee Range, Byrd Land. In: Antarctic Geology and Geophysics (Adie, R. J . , editor). Amsterdam, North Holland Publishing Company. 666-675. Halpern, M. 1970. Rubidium-strontium dates and 87Sr/86Sr initial ratios of rocks from Antarctica and South America a progress report. Antarctic Journal of the U.S., V(5): 159 161. Howorth, R., and P. C. Rankin. 1975. Multi-element characterization of glass shards from strati graphically correlated rhyolitic tephra units. Chemical Geology, 15: 239-250. Jones, L. M., and R. L. Walker. 1972. Geochemistry of the McMurdo volcanics, Victoria Land, part 1. Strontium isotope composition. Antarctic Journal of the U.S., VI1(5): 142144. Kay, R. W.. and R. W. Gast. 1973. The rare earth content and origin of alkali-rich basalts. Journal of Geology, 81: 653682. Kyle, P. R., and P. C. Rankin. In press. Rare-earth element geochemistry of Late Cenozoic alkaline lavas of the McMurdo Volcanic Group, Antarctica. Geochiinica et Cosmochi,nica Acta.
ANTARCTIC JOURNAL
LeMasurier, W. E., and F. A. Wade. In press. Volcanic history in Marie Byrd Land: implications with regard to southern hemisphere tectonic reconstructions. In: Proceedings of the International Symposium on Andean and Antarctic Vol canology Problems, Santiago, Chile (0. Gonzalez-Ferran, editor). Rome, International Association of Volcanology and Chemistry of Earth's Interior. Price, R. C., and S. R. Taylor. 1973. The geochemistry of Dunedin Volcano, East Otago, New Zealand: rare earth elements. Contributions to Mineralogy and Petrology, 40: 195-205. Sun, S. S., and G. N. Hanson. 1975. Origin of Ross Island basanitoids and limitations upon the heterogeneity of mantle sources of alkali basalts and nephelinites. Contributions to Mineralogy and Petrology, 52: 77-106. Sun, S. S., and G. N. Hanson. 1976. Rare earth element evidence for differentiation of McMurdo volcanics, Ross Island, Antarctica. Contributions to Mineralogy and Petrology, 54: 139-155.
H
La Ce Pr Nd
Sm E Gd Tb Dy Ho Er
Figure 6. Chondrite normalized REE plot of salic lavas.
December 1976
Yb
Igneous rocks of Peter I Island THOMAS W. BASTIEN
Ernest E. Lehmann Associates Minneapolis, Minnesota 55403 CAMPBELL CRADDOCK
Department of Geology and Geophysics The University of Wisconsin, Madison Madison, Wisconsin 53706 Peter I Island lies in the southeastern Pacific Ocean at 68°50'S. 90°40'W. about 240 nautical miles off the Eights Coast of West Antarctica. Rising from the continental rise, it is one of the few truly oceanic islands in the region. Few people have been on the island, and little is known of its geology. Thaddeus von Bellingshausen discovered and named the island in 1821, and it was not seen again until sighted by Pierre Charcot in 1910. A Norwegian ship dredged some rocks off the west coast in 1927, and persons from the Norvegia achieved the first landing in 1929. USNS Burton Island put a party ashore in Norvegia Bay in 1960; Craddock collected 29 rock specimens at that time. Peter I Island is a glaciated volcanic island about 20 kilometers long and 1,750 meters above sea level. The few rock exposures are mainly in shore cliffs; the narrow beaches contain large clasts of rock and glacial ice (figure 1). A well-developed marine platform surrounds the island and interrupts the otherwise symmetric profile of a large volcanic seamount. The observed bedrock consists of interbedded flows of basalt and more siliceous lavas; flow thicknesses are mainly less than 10 meters, and apparent dips are 5 degrees or less. Lavas vary from dense to highly vesicular, and a few surfaces show pahoehoe or ropy structure (figure 2). Some trachyandesite flows contain numerous gabbroid inclusions. Dikes and small stocks cut the stratified rocks. A single potassium-argon whole-rock age of 12.5 ± 1.5 million years was obtained on a basalt flow from Norvegia Bay. Broch (1927) described the 175 dredged rocks and identified three rock types: basalt, andesite, and trachyandesite. Andesite is lacking in the 1960 collection, but the trachyandesite from the shore contains inclusions of gabbroid rocks. Most rocks from the island are basalt; other varieties comprise a minor fraction. The basalts are hypo- to holocrystalline, commonly porphyritic, and consist of clinopyroxene, intermediate plagioclase, metallic oxides, and oh267
Rare-earth element geochemistry of volcanic rocks from the Executive Committee Range, Marie Byrd Land
WESLEY E. LEMASURIER Department of Geology University of Colorado Denver, Colorado 80202 PHILIP R. KYLE Institute of Polar Studies The Ohio State University Columbus, Ohio 43210 PETER C. RANKIN Soil Bureau Department of Scientific and Industrial Research Lower Hutt, New Zealand Volcanoes of the Executive Committee Range (1260 W., 760 to 77°S.) were first described by Doumani (1964), and further work has been reported by LeMasurier and Wade (in press). The range is the largest and most petrologically varied volcanic range in Marie Byrd Land; it includes representatives of essentially all the rock types present in the entire petrographic province. Lavas range in composition from alkali basalt and basanite to intermediate compositions such as hawaiite, mugearite, and benmoreite. An extremely wide range of salic rock types, paralleling those in the African rift valleys, also occur and include phonolite, quartz trachyte, alkali rhyolite, comenditic trachyte, comendite, and pantellerite (figure 1). Strontium isotope measurements (Halpern, 1970; Jones and December 1976
Walker, 1972; LeMasurier and Wade, in press) on basaltic, intermediate, and some salic lavas are similar, suggesting they are from a mantle-derived source and may be comagmatic. However, it is difficult to derive this great variety of oversaturated and undersaturated salic rocks from the nepheline normative alkali basalt magma that seems to be the only primary magma type available in the province. Several combinations of high- and low-pressure petrologic processes seem to be called for. Rareearth element (REE) studies of these rocks have focused on the Executive Committee Range because of the variety of rocks represented and the fact that field and chronologic relations are better displayed here than anywhere else in Marie Byrd Land. Thirteen lava samples from the Executive Committee Range together with an alkali basalt, basanite, phonolite, and pantellerite from other parts of Marie Byrd Land were analyzed by Dr. Rankin for REE, cesium, barium, hafnium, lead, thorium, and uranium using spark source mass spectrometry. The analytical technique and precision of the method are given by Howorth and Rankin (1975). REE analyses are plotted normalized to the chondrite abundances given by Price and Taylor (1973). Interpretation of the results is in progress. Results and some preliminary conclusions are discussed below. Basanite (figure 2). The two samples have similar chondrite normalized patterns that show a small positive europium anomaly. Compared to the alkali basalt (figure 3) the basanites are enriched in REE. The REE distribution of the basanites is typical of undersaturated basaltic magmas, which are considered to form by partial melting of a garnet peridotite mantle (Kay and Gast, 1973). Alkali basalt-hawaiite-mugearite-benmoreite (figure 3). There is an increase in REE uranium, thorium, and barium through this series of rocks; an exception is benmoreite 23A, discussed separately below. All samples have a positive europium anomaly. Alkali basalt and basanite (see above) generally have small positive europium anomalies inherited from partial melting and fractionation processes operating in the mantle (Kay and Gast, 1973; Sun and Hanson, 1975). The europium anomalies in the hawaiite, mugeraite, and benmoreite (50A) probably constitute a feature that in turn was inherited from the alkali basalt parent. The progressive increase in REE through the series is consistent with an origin by fractional crystallization from an alkali basalt parent. Benmoreite sample 23A has an exceptionally 263
16
I I I I I I • 54a 71A 2Qd
C
23Ca• 24B4%29 42a
U
a)
0.
23A •70 • 22d 41A
2A • 4.5c •. 32
•48
62 0'44
48 52 56 60 64 68 72 S10 2
(wt. percent)
Figure 1. Alkalis versus Si0 2 diagram of analyzed lavas.
,1.If
1OC
3)
;5c
10 264
La Ce Pr Nd
Sm E Gd T Dy Ho Er
Figure 2. Chondrite norYb malized REE plot of basanite lavas.
ANTARCTIC JOURNAL
I, a, -o
C
0
U 50 a, CL
E
0 'I,
III
Figure 3. Chondrite normalized REE plot of alkali basalt-hawaiite-mugearitebenmoreite lavas.
5
I I I I
La Ce Pr Nd
large positive europium anomaly (EuIEu* = 1.791) and a high barium content. An anomaly of the magnitude observed in 23A is unusual and difficult to explain. The data suggest addition or accumulation of large amounts of feldspar, but the sample is nearly aphyric and shows no evidence of feldspar phenocrysts that could be cumulus in origin. Lowor high-pressure fractionation processes are unlikely to account for the observed europium anomaly. Quartz trachyte (figure 4). Two samples show extremely different REE patterns. Sample 24B has low barium (less than 50 parts per million) and a large negative europium anomaly (EuIEu* =
l EuIEu* = the measured concentration divided by the concentration estimated by interpolating the chondrite normalized samarium and gadolinium values.
December1976
Sm E Gd T D Ho Er
Y
0.37) indicative of feldspar fractionation. The other sample (42A) has an extremely high barium (1,600 parts per million) and a small positive europium anomaly (Eu/Eu* = 1.08); petrogenesis of this sample did not involve feldspar fractionation. Phonolite (figure 5). Three phonolite samples have similar REE patterns except for varying europium contents, and they must have evolved by similar processes. Feldspar fractionation in slightly different abundances would account for the varying negative europium anomalies. Sample 54A is the most fractionated with the lowest barium, highest uranium, thorium, lead, and largest europium anomaly, yet it has a total REE content intermediate between the other two samples. It may have formed from a parent with a lower REE content than samples 20D and 71A. Relative to the analyzed alkali basalt and basanite, some phonolites show a greater enrichment of the heavy REE compared to the light REE. As frac-
265
1000
500
- 100 L)
E In
5C
1
10 I I I
1 111111
I
L Ce P r Nd So E0 Gd T D Ho E, Y
- La
Co Pr Nd
Sm E Gd Tb Dy Ho E r Yb
Figure 4. Chondrite normalized REE plot of quartz trachyte lavas.
Figure 5. Chondrite normalized REE plot of phonolite lavas.
tionation of most mineral phases present in these lavas will not result in a greater enrichment of the heavy REE compared to the light REE, it appears that the phonolites were not derived from a parent like the analyzed alkali basalt or basanite. A basanite with a higher abundance of light REE, or a lower abundance of heavy REE, would be a suitable parent (compare basanite lavas from the McMurdo Sound area) (Sun and Hanson, 1975, 1976; Kyle and Rankin, 1976). Salic lavas (figure 6). Characteristic of all these lavas is their variable chondrite normalized patterns and their large negative europium anomalies (EuIEu* ranges from 0.20 to 0.65). Substantial quantities of feldspar fractionation must have occurred during the formation of these lavas. The parents and processes of fractionation are under investigation.
References
W. E. LeMasurier collected the samples during fieldwork supported by National Science Foundation grant DPP 70-02980. 266
Doumani, G. A. 1964. Volcanoes of the Executive Committee Range, Byrd Land. In: Antarctic Geology and Geophysics (Adie, R. J . , editor). Amsterdam, North Holland Publishing Company. 666-675. Halpern, M. 1970. Rubidium-strontium dates and 87Sr/86Sr initial ratios of rocks from Antarctica and South America a progress report. Antarctic Journal of the U.S., V(5): 159 161. Howorth, R., and P. C. Rankin. 1975. Multi-element characterization of glass shards from strati graphically correlated rhyolitic tephra units. Chemical Geology, 15: 239-250. Jones, L. M., and R. L. Walker. 1972. Geochemistry of the McMurdo volcanics, Victoria Land, part 1. Strontium isotope composition. Antarctic Journal of the U.S., VI1(5): 142144. Kay, R. W.. and R. W. Gast. 1973. The rare earth content and origin of alkali-rich basalts. Journal of Geology, 81: 653682. Kyle, P. R., and P. C. Rankin. In press. Rare-earth element geochemistry of Late Cenozoic alkaline lavas of the McMurdo Volcanic Group, Antarctica. Geochiinica et Cosmochi,nica Acta.
ANTARCTIC JOURNAL
LeMasurier, W. E., and F. A. Wade. In press. Volcanic history in Marie Byrd Land: implications with regard to southern hemisphere tectonic reconstructions. In: Proceedings of the International Symposium on Andean and Antarctic Vol canology Problems, Santiago, Chile (0. Gonzalez-Ferran, editor). Rome, International Association of Volcanology and Chemistry of Earth's Interior. Price, R. C., and S. R. Taylor. 1973. The geochemistry of Dunedin Volcano, East Otago, New Zealand: rare earth elements. Contributions to Mineralogy and Petrology, 40: 195-205. Sun, S. S., and G. N. Hanson. 1975. Origin of Ross Island basanitoids and limitations upon the heterogeneity of mantle sources of alkali basalts and nephelinites. Contributions to Mineralogy and Petrology, 52: 77-106. Sun, S. S., and G. N. Hanson. 1976. Rare earth element evidence for differentiation of McMurdo volcanics, Ross Island, Antarctica. Contributions to Mineralogy and Petrology, 54: 139-155.
H
La Ce Pr Nd
Sm E Gd Tb Dy Ho Er
Figure 6. Chondrite normalized REE plot of salic lavas.
December 1976
Yb
Igneous rocks of Peter I Island THOMAS W. BASTIEN
Ernest E. Lehmann Associates Minneapolis, Minnesota 55403 CAMPBELL CRADDOCK
Department of Geology and Geophysics The University of Wisconsin, Madison Madison, Wisconsin 53706 Peter I Island lies in the southeastern Pacific Ocean at 68°50'S. 90°40'W. about 240 nautical miles off the Eights Coast of West Antarctica. Rising from the continental rise, it is one of the few truly oceanic islands in the region. Few people have been on the island, and little is known of its geology. Thaddeus von Bellingshausen discovered and named the island in 1821, and it was not seen again until sighted by Pierre Charcot in 1910. A Norwegian ship dredged some rocks off the west coast in 1927, and persons from the Norvegia achieved the first landing in 1929. USNS Burton Island put a party ashore in Norvegia Bay in 1960; Craddock collected 29 rock specimens at that time. Peter I Island is a glaciated volcanic island about 20 kilometers long and 1,750 meters above sea level. The few rock exposures are mainly in shore cliffs; the narrow beaches contain large clasts of rock and glacial ice (figure 1). A well-developed marine platform surrounds the island and interrupts the otherwise symmetric profile of a large volcanic seamount. The observed bedrock consists of interbedded flows of basalt and more siliceous lavas; flow thicknesses are mainly less than 10 meters, and apparent dips are 5 degrees or less. Lavas vary from dense to highly vesicular, and a few surfaces show pahoehoe or ropy structure (figure 2). Some trachyandesite flows contain numerous gabbroid inclusions. Dikes and small stocks cut the stratified rocks. A single potassium-argon whole-rock age of 12.5 ± 1.5 million years was obtained on a basalt flow from Norvegia Bay. Broch (1927) described the 175 dredged rocks and identified three rock types: basalt, andesite, and trachyandesite. Andesite is lacking in the 1960 collection, but the trachyandesite from the shore contains inclusions of gabbroid rocks. Most rocks from the island are basalt; other varieties comprise a minor fraction. The basalts are hypo- to holocrystalline, commonly porphyritic, and consist of clinopyroxene, intermediate plagioclase, metallic oxides, and oh267
