| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology | Volume 43 | Number 5 | Pages 859-883 | 2002
© Oxford University Press 2002
Petrology and Geochemistry of Volcán Cerro Azul: Petrologic Diversity among the Western Galápagos Volcanoes
1DEPARTMENT OF GEOLOGY, UNIVERSITY OF ALASKA ANCHORAGE, ANCHORAGE, AK 99508, USA
2DEPARTMENT OF GEOLOGY, UNIVERSITY OF IDAHO, MOSCOW, ID 83844, USA
3DEPARTMENT OF MARINE CHEMISTRY AND GEOCHEMISTRY, WOODS HOLE OCEANOGRAPHIC INSTITUTION, WOODS HOLE, MA 02543-1541, USA
Received April 5, 2001; Revised typescript accepted November 22, 2001
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGY OF CERRO AZULRELATIVE AGE OF THE...SAMPLING AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Cerro Azul, one of the large shield volcanoes in the western Galápagos archipelago, has erupted a wide range of tholeiitic to alkalic basalts. These diverse compositions include some of the most primitive yet reported from the western archipelago and are unlike those of the other, well-studied, neighboring volcanoes of Sierra Negra and Alcedo, which have erupted basalt of fairly uniform composition. Major- and trace-element modeling shows that Cerro Azul, Alcedo and Sierra Negra share a similar depth of melting and source composition. Modeling also reveals that there are small, systematic differences in the extent of partial melting between the volcanoes that can be related to their distance from the proposed plume center below the westernmost island of Fernandina. However, even though melts segregating from the plume in the western Galápagos reflect a narrow range of temperatures and source compositions, there are wide variations in the enrichments of major and trace elements between Cerro Azul, Alcedo and Sierra Negra that cannot be attributed to mantle processes. We believe the observed intershield geochemical differences result from magma supply and cooling rates that are unique to each volcano, and reflect the variations in lithospheric transport and storage processes across the western archipelago.
KEY WORDS: basalt; Galápagos; magma supply; mantle plume; ocean island
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF CERRO AZULRELATIVE AGE OF THE...SAMPLING AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The Galápagos archipelago is a hotspot-generated collection of island volcanoes that differs in many respects from more familiar examples, such as Hawaii. The archipelago is located adjacent to the Galápagos Spreading Center and constructed atop a shallow volcanic platform on young, thin ocean lithosphere (Fig. 1). Although it has been long recognized that the Galápagos islands reflect intraplate volcanism resulting from a mantle plume (Morgan, 1971
), a number of tectonic and petrologic features conflict with a simple mantle plume model. These include: (1) a non-linear arrangement of volcanoes; (2) a wide distribution of simultaneously active volcanoes; (3) a distinct geographic pattern to the petrologic and geochemical diversity of the volcanoes; (4) a lack of any consistent petrological and geochemical evolutionary patterns like those documented for Hawaii (Macdonald & Katsura, 1964; Frey & Clague, 1983; Chen & Frey, 1985) and Réunion (Albarède et al., 1997). Much of the geochemical variability between volcanoes in the Galápagos has been ascribed to mantle processes connected with a well-defined, archipelago-wide pattern of depleted and enriched mantle source compositions (Geist et al., 1988; Geist, 1992; Graham et al., 1993; White et al., 1993; Harpp, 1995; Kurz & Geist, 1999; Harpp & White, 2001). These geochemical patterns are proposed to result from both intrinsic differences in plume composition and the effects of entrainment and mixing of enriched plume and depleted asthenosphere (mid-ocean ridge basalt; MORB) during plume ascent. In addition to compositional effects imparted by mantle-source compositions, Geist et al. (1998) presented an archipelago-wide intershield geochemical comparison that focused on the control of lithospheric processes in the production of the morphological and geochemical variations among the Galápagos volcanoes. Geist et al. (1998) proposed that Galápagos magmas evolve by fractional crystallization at systematically different depths in the crust and mantle across the archipelago. Both types of comparisons are important in assessing the dynamics of the Galápagos plume as it impinges on the base of the young ocean lithosphere, in order to determine both the magma source distributions and the relationship of regional stresses and structures to the geochemical variations and the spatial distribution of volcanic vents. However, archipelago-wide comparisons are complicated by a wide range of tectonic and magmatic settings across the region, such as lithospheric age and thickness, source composition, distance from the plume center, depth of melting, and age of volcanism. To limit these variables, this investigation is restricted to the southwestern part of the archipelago, where variations in crustal age and thickness, source composition, and depth of melting are limited (Geist, 1992; White et al., 1993; Feighner & Richards, 1994). Furthermore, the seven large active shield volcanoes forming the western islands of Fernandina and Isabela are nearly equivalent in age (Naumann & Geist, 2000) and aligned at a high angle to the modern Nazca plate motion vector, precluding a simple chain-like temporal progression for their origin (Fig. 1).
|
We report here new and detailed major, trace and isotopic analyses from Cerro Azul, document the petrologic processes responsible for the geochemical variation, and compare Cerro Azul with the well-studied neighboring volcanoes of Alcedo (Geist et al., 1995) and Sierra Negra (Reynolds & Geist, 1995). These three volcanoes share many geomorphologic and petrologic features, yet in detail each is unique. The diverse basalt compositions of Cerro Azul are unlike those of Alcedo and Sierra Negra, although all three volcanoes share a similar age, depth of melting and, to a lesser extent, source composition and degree of melting. These results suggest that the compositional variations among these volcanoes reflect a range in magma supply, storage, and cooling rate, which in turn are controlled by the total supply rate of primitive magma to each volcano.
Tectonic setting
The Galápagos archipelago is located on the Nazca plate 
1000 km west of South America (Fig. 1). The islands rise from a shallow volcanic platform that forms the western end of the aseismic Carnegie Ridge. The east–west-trending Galápagos Spreading Center (GSC) lies directly to the north of the archipelago and forms the boundary between the Nazca and Cocos plates.
The proximity of the Galápagos hotspot and the GSC has strongly influenced the tectonic and magmatic history of both features. The GSC is characterized by numerous jumps, ridge propagations, oblique transforms, and positional instabilities. These features are believed to result from the thermal disturbance of the nearby plume (Wilson & Hey, 1995).
Another plume–ridge interaction is manifest in the thickness of the lithosphere beneath the archipelago. Gravity modeling by Feighner & Richards (1994) has revealed that the center of the archipelago is underlain by thin, weak lithosphere, with an elastic thickness of 6 km or less and being close to Airy compensation, whereas the western and southern parts of the platform are flexurally supported by a lithosphere with an effective elastic thickness of 12 km. Crustal thickness across the archipelago was modeled by Feighner & Richards (1994), and reaches a maximum of 18 km in the southwestern Isabela island (the focus of this study) and thins to 10 km at the edges of the Galápagos platform.
Another unusual feature of the archipelago was described by Darwin (1860), who recognized that the physiography of the Galápagos archipelago is dominated by two prominent structural lineaments (northwest and ENE), which include the alignment and spacing of volcanoes, seamounts, and troughs. Northwestern alignments are dominant in the northern and western parts of the archipelago and include the islands of Wolf, Darwin, Pinta, and Marchena along the Wolf–Darwin lineament (Feighner & Richards, 1994) and the large shield volcanoes of Isabela and Fernandina (Fig. 1). Cerro Azul and Sierra Negra are aligned along a conjugate ENE trend that strikes 65° from the main axis of Isabela.
Although direct evidence for lithospheric faults or fractures is concealed by the voluminous lavas that have constructed the Galápagos platform, the alignment patterns of the Galápagos volcanoes at a high angle to the Nazca plate motion vector suggest that this trend reflects the regional stress field and that lithospheric stresses, faults or fracture zones have controlled the location and spacing of volcanoes, rather than their systematic development as a result of the passage of the Nazca plate over the Galápagos hotspot. Darwin (1860), Banfield et al. (1956), McBirney & Williams (1969), Nordlie (1973), White et al. (1993), Feighner & Richards (1994), and Chadwick & Dieterich (1995) all described the ‘Darwinian’ lineaments and attributed the origin of the volcano alignments to the presence of some ‘megatectonic’ (Nordlie, 1973) stress. Chadwick & Howard (1991) also discussed the lineaments and their possible role in the alignment of the individual volcanoes but discounted their influence over modern volcano eruptive fissures, which they believe are controlled by intravolcano stresses resulting from magma intrusion.
|
GEOLOGY OF CERRO AZUL |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF CERRO AZULRELATIVE AGE OF THE...SAMPLING AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
A detailed volcanic history of Cerro Azul has been presented in Naumann & Geist (2000)
, and is briefly summarized below. Cerro Azul forms the SW end of Isabela island and has a maximum subaerial dimension of 34 km x 22 km and a subaerial volume of 172 km3 (Naumann & Geist, 2000) (Fig. 2). It rises to 1640 m and is second in elevation only to Volcán Wolf (1710 m). Ten documented eruptions have occurred between 1932 and 1998, an average of one eruption approximately every 6·5 years. Although Cerro Azul has been constructed primarily by effusive Hawaiian-style eruptions, explosive hydrovolcanic eruptions have occurred intermittently from vents on the caldera floor and southern summit rim (Naumann & Geist, 2000).
|
The ‘inverted soup plate’ profile unique to the western Galápagos shield volcanoes is well displayed at Cerro Azul. Most sectors of the volcano have shallow-dipping lower flanks (generally <4°), steep upper flanks (ranging from 12° to 30°), and a wide flat summit platform 5 km in diameter (Fig. 2). A multiply nested caldera of 450 m depth contains numerous benches, eruptive fissures, and cones, and is surrounded by a flat, summit rim of 1 km width (Fig. 2). The entire summit rim is surrounded by steep slopes, and in most sectors of the volcano slopes descend to 600 m above mean sea level. Below that height on the northern, eastern, and southern flanks, shallow-dipping lower slopes form a constructional apron up to 15 km wide. On the SW sector, however, the slopes are steeper (up to 30°) and plunge from the summit rim (1600 m in elevation) directly to the sea within a horizontal distance of 4·5 km (Fig. 2). Here, a 20 km2 embayment (Caleta Iguana) interrupts the otherwise convex outline of the volcano and represents a departure from the normal morphologic profile of the western Galápagos volcanoes. This embayment is produced by a series of steep, north- to NW-trending landslide scarps up to 100 m high that expose the steeply dipping (up to 30°) lava flows of the SW flank. The failure of the southwestern flank of Cerro Azul results from its construction directly adjacent to the steep submarine escarpment, of 3 km height, that defines the westernmost edge of the Galápagos platform (Fig. 1). On this sector of the volcano recent lower flank vents are absent owing to the gravitational stresses imposed by the SW-directed flank failure, which has redirected intrusions to the NW or SE (Naumann & Geist, 2000). This has resulted in a concentration of recent eruptions from a NW–SE rift zone on the summit platform and within the caldera that intersects earlier eruptive fissures at a high angle.
Owing to frequent eruptions, a dry climate and the resulting lack of fluvial erosion, there is no deep dissection of the volcano surface, except within the caldera. Two samples, one from the base of the eastern caldera wall at a depth of 450 m and another from the upper part of a 300 m section of lavas that filled the caldera and form the intra-caldera benches, yield 40Ar/39Ar plateau ages of 69 ± 17 ka and 82 ± 17 ka (Naumann & Geist, 2000). Cosmogenic 3He surface exposure ages from samples representing a complete range of surface weathering were collected at the summit rim, north coast, and near Caleta Iguana (Kurz & Geist, 1999). The ages indicate that the surface of the volcano is young and has been resurfaced in 5000 years. We estimate that at least 25% of the volcano is covered by flows <2000 years old and calculate an average eruption rate of 0·30 x 106 m3/yr based on the surface area (600 km2) covered by flows of 2 m thickness. By comparison, Reynolds & Geist (1995) estimated that the surface of Sierra Negra is <7000 years old and computed a long-term growth rate of 1·1 x 106 m3/yr.
|
RELATIVE AGE OF THE WESTERN GALÁPAGOS VOLCANOES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF CERRO AZULRELATIVE AGE OF THE...SAMPLING AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Volcanoes have been active on most of the Galápagos islands during the Quaternary, but the maximum measured ages of the volcanoes generally increase from Fernandina in the west to San Cristobal in the east (Fig. 1) (White et al., 1993
; Sinton et al., 1996), as predicted by the eastward motion of the Nazca plate. Determining the relative ages of the western shield volcanoes of Isabela and Fernandina is problematic for the following reasons: (1) they are aligned nearly perpendicular to the direction of the Nazca plate motion, precluding a simple temporal progression for their origin as the Nazca plate moved above the Galápagos hotspot; (2) the flanks of each volcano are completely surfaced by young lavas; (3) owing to a dry climate and the lack of erosion, there is no incision into the volcano flanks. The deepest exposures come from the caldera walls, but Alcedo and Sierra Negra have only shallow calderas (Fig. 3).
|
The lack of deep erosion or drill holes precludes precise age determinations for the volcanoes, but Naumann & Geist (2000) determined that their emergent (subaerial) ages were similar by comparing their time-averaged eruption rates with computed subaerial volumes. Naumann & Geist (2000) determined that Cerro Azul (0·05 x 106 m3/yr; 175 km3) would have become emergent at 350 ka, Sierra Negra (1·1 x 106 m3/yr; 588 km3) at 535 ka, and Alcedo (1·0 x 106 m3/yr; 234 km3) at 313 ka. Thus, all three volcanoes have been contemporaneously growing throughout their emergent histories.
|
SAMPLING AND ANALYTICAL TECHNIQUES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF CERRO AZULRELATIVE AGE OF THE...SAMPLING AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Rock samples were ground in a Spex tungsten-carbide ring mill and analyzed at the Washington State University geoanalytical facility according to the procedures of Johnson et al. (1999)
. Major elements, Ni, Cr, Sc, V, Ba, Sr, Rb, Zr, Y, Nb, Cu and Zn were determined by XRF techniques. Samples were examined in thin section for weathering and phenocryst accumulation, and a subset of lavas (20) were analyzed for the rare earth elements (REE), Ba, Th, Nb, Y, Hf, Ta, U, Pb, Rb, and Cs by inductively coupled plasma-mass spectrometry (ICP-MS). Estimates of relative precision are based upon analyses of standards and replicate analyses and are <2% for major-element oxides and <5% for all trace elements except Cr and Rb (17% at these levels). Sr and Nd isotopes were analyzed in a subset of 11 samples in the Chemistry Department of Woods Hole Oceanographic Institution according to the method of Kurz et al. (1995). He isotopes were analyzed in a subset of five samples according to the methods reported by Kurz & Geist (1999). Errors for isotopic values are reported as 2
. |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF CERRO AZULRELATIVE AGE OF THE...SAMPLING AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Petrography and mineral compositions
Olivine, clinopyroxene and plagioclase are the dominant phenocryst minerals in Cerro Azul lavas, and petrographic examination and chemical analyses define three lava series, as follows.
- High-MgO basalts (>10% MgO), which are sparsely porphyritic. These lavas contain 2–5% euhedral to subhedral olivine phenocrysts (>0·5 mm) in an intergranular groundmass of olivine, augite, plagioclase, and Fe–Ti oxides. Olivine phenocrysts range in composition from Fo89 to Fo85 (Fig. 4), and single grains are either unzoned or exhibit weak normal zoning, with rims 4% less forsteritic than the cores. The lack of disequilibrium textural features and the sparse amounts and consistent proportions of phenocrysts suggest that most of these lavas have compositions close to those of natural liquids. There are two exceptions: sample CA-31 (13% MgO), which contains >10% olivine phenocrysts (>0·5 mm) that are interpreted to have accumulated, and CA-46 (17·2% MgO), which is a gabbroic xenolith containing 30% olivine phenocrysts. Both are excluded from the chemical plots.
- Intermediate-MgO basalts, defined as having between 7 and 10% MgO. These lavas are by far the most abundant type of lava on Cerro Azul and contain variable amounts of olivine + plagioclase ± clinopyroxene phenocrysts. Olivine phenocrysts range in composition from Fo87 to Fo83 (Fig. 4), and single grains show weak normal zoning, with rims 3% less forsteritic than the cores. Plagioclase phenocrysts are typically subhedral and range in composition from An92 to An80 (Fig. 5). Many plagioclase phenocrysts are normally zoned with rims up to 10% lower in anorthite component than the core, or in rare occurrences, showing weak oscillatory zoning. Clinopyroxene phenocrysts range from subhedral to anhedral and are limited to the samples with <8% MgO. They range in composition from Wo47En45Fs8 to Wo45En46Fs9 (Fig. 4), and many exhibit weak, normal zoning. Clinopyroxene and plagioclase make up glomeroporphyritic clots in many of these intermediate lavas. The groundmass mineral assemblage consists of olivine, plagioclase, clinopyroxene, opaques, and sparse needles of apatite, and textures vary from hyalopilitic to ophitic, although fine- to medium-grained intergranular texture is the most common. Inclusion relationships indicate that olivine, plus Cr-spinel, is the initial liquidus phase, followed by plagioclase and then clinopyroxene.
- Evolved basalts are those with MgO <7%. These contain phenocrysts of olivine + plagioclase with sparse clinopyroxene. Olivine phenocrysts range in composition from Fo71 to Fo69 (Fig. 4). Plagioclase phenocrysts are euhedral to subhedral and range in composition from An71 to An57 (Fig. 5). Many plagioclase phenocrysts in the evolved basalts are reversely zoned with rims up to 12% higher in anorthite component than the core. Several specimens have a few small grains of clinopyroxene (Wo40En43Fs16) (Fig. 4). Two samples from Cerro Azul (CA-20, CA-48) are unusually rich in Al2O3 (>16%) relative to the rest of the samples and contain between 10 and 20% plagioclase phenocrysts. Because the relative proportions of clinopyroxene and olivine in these samples are similar to those in the other lavas, they are interpreted to have accumulated plagioclase and are excluded from the major-element chemical plots.
|
), transitional (+) and alkaline (
) basaltic magma series at Cerro Azul. Symbols for each series are used in all successive geochemical plots. Olivine compositions lie along the bottom of the quadrilateral.
|
Major elements
Cerro Azul has erupted a wide compositional range of basaltic lavas, some of which are the most primitive compositions known from the western Galápagos archipelago (Table 1; Figs 6 and 7). On a plot of total alkalis vs silica, lavas from Cerro Azul plot as a continuous range from tholeiitic to transitional to alkaline (Fig. 6). These fields (tholeiitic, transitional, and alkaline) closely follow the three magma series defined by the petrography and mineral chemistry above. Primitive lavas (>10% MgO) all plot well within the tholeiitic field. Most intermediate lavas (7–10% MgO) straddle the dividing line of Macdonald & Katsura (1964) and plot as transitional between alkalic and tholeiitic compositions, and evolved samples (<7% MgO) all plot within the alkaline field.
|
|
|
Tholeiites represent the majority of samples (48%), whereas alkaline and transitional samples are 25% and 27%, respectively. Curiously, 12 of the 13 alkaline samples were collected from the SW sector of the volcano and include older flows from the upper and lower flanks, as well as recent lavas from a NW–SE-trending summit rift zone that cross-cuts earlier circumferential and radial fracture zones on this sector of the volcano.
With the exception of SiO2, the more evolved transitional and alkaline lavas of Cerro Azul typically plot as well-defined trends below 8% MgO (Fig. 7). TiO2, FeO, Na2O, K2O, and P2O5 are inversely correlated with MgO and steadily increase with decreasing MgO. There is a change in slope in all oxides at 8% MgO. The CaO/Al2O3 ratio of Cerro Azul lavas steadily decreases below 8% MgO and is lowest in the most evolved alkaline samples (Fig. 8).
|
Although Cerro Azul samples have a continuous range of MgO, the concentrations of many of the major-element oxides vary widely at a similar MgO content (e.g. 6–9% MgO) (Fig. 7). For example, the SiO2 content of the less evolved tholeiitic series ranges from 48 to 50% at 8% MgO. FeO, Na2O and K2O also have wide ranges at 8% MgO content. A particularly unusual primitive olivine tholeiite (11·7% MgO) from Cerro Azul (CA-36) lies well off the trend for the other tholeiitic samples and has much higher SiO2 (51·5%) and lower TiO2, CaO, K2O, and P2O5.
Trace elements
The Sc/Y ratio, which acts as a proxy for the CaO/Al2O3 ratio, is constant in lavas with MgO >8% but decreases steadily below 8% MgO (Fig. 8). The abundances of the compatible elements Ni and Cr decrease exponentially as a function of decreasing MgO (Fig. 9). Abundances are highest (Ni >300 ppm and Cr >800 ppm) in the primitive tholeiites and lowest in the evolved alkaline lavas (Ni <20 ppm and Cr <25 ppm).
|
Cerro Azul lavas have a wide range in the enrichments of the most incompatible trace elements (e.g. Ba ranges from 40 to 200 ppm), and the abundances of Ba, U, Th, Rb, Zr, Sr and the REE are positively correlated (Fig. 10). The highest concentrations are in the most evolved alkaline lavas and systematically decrease to the lowest abundances in the most primitive tholeiite (CA-36). All incompatible trace elements define steep linear trends on variation diagrams (plotted against Ba), with the exception of Sr, which remains at 350 ppm at concentrations >120 ppm Ba. (Fig. 10). Incompatible trace-element ratios of Ba/Zr (Fig. 11) increase slightly but systematically from the tholeiitic to the alkaline suites whereas the ratios of Sr/Nd decrease.
|
|
Chondrite-normalized REE patterns of the lavas from Cerro Azul define steep, parallel, light REE (LREE)-enriched patterns that are systematically stacked from the tholeiites (lowest values) to the alkaline samples (Fig. 12). There is a single anomalous lava (CA-36), which has a chondrite-normalized REE pattern different from that of any analyzed lava from the western archipelago. It has La/Yb of 2·6 and defines a flat trend from La to Eu, then steepens and crosses the trends of other Cerro Azul lavas from Gd to Lu (Fig. 12). This pattern is similar to that of several samples reported from Santa Cruz in the central Galápagos archipelago, where a wide range in the percent of partial melting has produced wide chemical variability (Bow, 1979; White et al., 1993).
|
Radiogenic isotopes
Isotopic ratios of Sr and Nd are plotted in Fig. 13. Eight Cerro Azul lavas define a small range in 87Sr/86Sr (0·70330–0·70334 ± 0·00002) and 143Nd/144Nd (0·512913–0·512950 ± 0·000014), representing some of the most enriched values in the Galápagos. CA-36, which is enriched in SiO2 and depleted in the LREE and other incompatible trace elements, also has higher 143Nd/144Nd (0·512976) than the other Cerro Azul lavas.
|
Magmatic 3He/4He isotopic ratios from Cerro Azul define a small range (13·7–14·2 Ra ± 0·2), which are intermediate between the highest (Fernandina, 30 Ra) and lowest (Santa Cruz, 8·6 Ra) values from the Galápagos (Graham et al., 1993; Kurz & Geist, 1999).
Comparison of Cerro Azul with Alcedo and Sierra Negra
Major elements
Comparisons between Cerro Azul, Sierra Negra and Alcedo are complicated by the fact that Alcedo has erupted 1 km3 of rhyolite late in its volcanic history (Geist et al., 1994). These evolved lavas are volumetrically minor when compared with the bulk of Alcedo (233 km3), which has been constructed by basaltic lava of more restricted composition over the past 350 ky. Thus, we compare only basaltic compositions (MgO >5%) from Alcedo with Cerro Azul and Sierra Negra, to identify the predominant, long-term intershield differences. The implications of the highly evolved lavas at Alcedo will be considered later when we compare Alcedo’s crustal magmatic system with those of Cerro Azul and Sierra Negra.
The wide range of major-element compositions from Cerro Azul contrasts strongly with the narrow range of basaltic compositions erupted at Alcedo (Geist et al., 1995) and Sierra Negra (Reynolds & Geist, 1995), where all lavas have MgO <6·5%. On an alkali–silica diagram (Fig. 6), the wide spectrum of Cerro Azul compositions (2·6–4·8% Na2O + K2O) encompasses and exceeds the total variation at both Alcedo (2·8–4·0% Na2O + K2O) and Sierra Negra (3·3–4·5% Na2O + K2O) together.
Tholeiitic lavas from Cerro Azul extend to much more primitive compositions (13% MgO, Ni >320 ppm, mg-number = 70) than lavas from Alcedo (7·2% MgO) or Sierra Negra (7·8% MgO) and represent near-primary compositions on the basis of their high MgO and Ni contents (Hess, 1992). Alcedo and Sierra Negra lavas form trends parallel to the Cerro Azul liquid line of descent below 8% MgO, and with increasing degrees of differentiation (8% > MgO > 4%) there is little intershield variation in the slopes or levels of enrichment or depletion of major-element concentrations and ratios (Figs 7 and 8). Cerro Azul lavas are richer in Na2O and K2O and have consistently lower CaO/Al2O3 and Sc/Y ratios at a given MgO than either Sierra Negra or Alcedo (Fig. 8).
Trace elements
The range of the compatible trace elements Cr and Ni is much more restricted at Sierra Negra (5–60 ppm Ni, 20–220 ppm Cr) and Alcedo (10–60 ppm Ni, 20–180 ppm Cr) compared with Cerro Azul (10–330 ppm Ni, 10–800 ppm Cr) (Fig. 9).
With the exception of U, Sr and Rb, incompatible element concentrations are consistently higher in Alcedo and Sierra Negra lavas at a given Ba content than in Cerro Azul lavas. La, Th, Zr, and Yb are good intershield discriminators and define narrow fields with similar slopes for all three volcanoes. Throughout the range of Cerro Azul compositions, the concentrations of the highly incompatible elements La and Th increase by a factor of >6, whereas throughout the range of compositions from Alcedo and Sierra Negra, the concentrations increase by factors of only 3 and 2·5. The moderately incompatible elements Zr and Yb increase by factors of 3·3 and 2·6 in the Cerro Azul lavas, and by factors of 2 and 1·8 in the Alcedo and Sierra Negra suites, respectively (Fig. 10). This intershield difference in the enrichment of incompatible trace elements is repeated in the incompatible major-element oxides at Cerro Azul, where K2O varies by a factor of 4·9, whereas at Sierra Negra and Alcedo it is enriched by factors of only 2 and 2·8 (Fig. 7).
The concentration of Sr in the Alcedo and Sierra Negra suites remains constant (300 ppm) in lavas with >70 ppm Ba and forms narrow, overlapping trends. In contrast, Cerro Azul tholeiites with low Ba concentrations define a steep positive trend that intersects the Alcedo and Sierra Negra trends near 70 ppm Ba, and the transitional and alkaline lavas of Cerro Azul are richer in Sr (>350 ppm) than the Alcedo and Sierra Negra suites.
Sr/Nd and Ba/Zr ratios are generally higher in Cerro Azul tholeiites than Alcedo and Sierra Negra at similar Ba concentrations, although the most primitive Cerro Azul tholeiites have ratios similar to those of Sierra Negra. Sierra Negra values are higher than Alcedo (Fig. 11).
Even though REE patterns are similar, in detail there are small intershield variations (Fig. 12). Chondrite-normalized La/Yb ratios (La/YbN) range from 4·3 to 6·1 (highest in alkaline samples, lowest in tholeiites) for Cerro Azul with an average of 5·0, Alcedo ranges from 3·5 to 4·1 with an average of 3·6, and Sierra Negra ranges from 4·2 to 5·0 with an average of 4·6. The overall range of enrichment represented by the spread of the stacked chondrite-normalized plots is narrowest at Sierra Negra whereas Cerro Azul shows the widest range.
Isotopic ratios
Sr and Nd isotopic ratios for the Galápagos archipelago define a wide range of compositions between those typical of MORB and values typical of the ocean island basalts (White et al., 1993). Despite this wide range of isotopic compositions throughout the archipelago, Cerro Azul and Sierra Negra lavas are isotopically similar and form a very narrow range barely beyond analytical uncertainty. The lavas from Alcedo have more depleted Sr and Nd isotopic ratios than Cerro Azul and Sierra Negra. Two exceptions are CA-36, which plots close to Alcedo values, and a single Alcedo lava that plots with Cerro Azul and Sierra Negra (Geist et al., 1995) (Fig. 13).
He isotopic ratios from the samples reported here overlap with data reported by Kurz & Geist (1999) for Cerro Azul. The previous samples come from a different part of the volcano, and overall the data are consistent with a homogeneous mantle source. The average He isotopic ratios are lower at Cerro Azul (14 Ra) than Sierra Negra or Alcedo, which both have an average value of 16 Ra (Kurz & Geist, 1999).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF CERRO AZULRELATIVE AGE OF THE...SAMPLING AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Origin of Cerro Azul lavas and comparisons with Sierra Negra and Alcedo
Mantle source composition
The archipelago-wide pattern of depleted-to-enriched source compositions in the Galápagos has been well documented by Geist et al. (1988)
, Geist (1992), Graham et al. (1993), White et al. (1993), Harpp (1995), and Harpp & White (2001). Although the distribution of these different source compositions explains much of the isotopic variability throughout the archipelago, it does not explain the systematic variations in the enrichment of major and trace elements between the neighboring volcanoes of Cerro Azul, Sierra Negra, and Alcedo. Sr, Nd, and He data from Cerro Azul and Sierra Negra define an extremely limited isotopic variability. This suggests that these two volcanoes are derived from very similar source compositions (Fig. 13) and therefore source compositional differences cannot be responsible for the wide variations in major and trace elements between these volcanoes. Alcedo’s magmas have come from a slightly more depleted source than those of Sierra Negra and Cerro Azul, but the difference is small when compared with the archipelago-wide range (Fig. 13).
Degree and depth of melting
To determine the degree of melting required to produce the lavas of each volcano, partial melting of a hypothetical garnet lherzolite source was modeled using the methods of Johnson et al. (1990). Both batch melting and aggregated fractional melting were modeled and the results are presented in Table 2, with the melting curves for aggregated fractional melting plotted in Fig. 12. Both types of melting produce very similar results, with fractional melting requiring several percent lower overall extents of melting to reproduce the chondrite-normalized patterns. Whereas the absolute extents of melting depend on the mostly unconstrained choice of mantle composition and mode, with the exception of one Cerro Azul sample (CA-36), the combined compositional range of all three volcanoes can be reproduced by 2–5% aggregated fractional melting of a garnet lherzolite source (Figs 12 and 14). The low and consistent abundances of the heavy REE (HREE) require the inclusion of garnet in the source mineralogy, but to preserve La/Sm ratios the mode of garnet must be low (2%).
|
|
According to the fractional melting model, small differences in the degree of melting produce strong variations in the LREE/HREE ratio, explaining the systematic intershield variations in La/Yb ratios between Cerro Azul, Alcedo and Sierra Negra (Fig. 14). The average La/Yb ratio is highest at Cerro Azul (7·2), indicating that these lavas were produced by the lowest degrees of melting (2–3%) when compared with Alcedo and Sierra Negra. The suites of lavas from Sierra Negra and Alcedo have average La/Yb ratios of 6·6 and 5·3, respectively, indicating that Alcedo’s lavas have been produced by the highest degrees of partial melting (5%), and that the percent of partial melting required to produce Sierra Negra’s lavas is intermediate (4%). Although these La/Yb variations could be explained by differing depths of melting (and therefore different amounts of residual garnet in the sources), we feel that the similar isotopic compositions between these lavas preclude a large range in source composition or depth of melting. However, because of Alcedo’s more depleted source composition (as indicated by Nd isotopes), the extent of partial melting may be slightly lower than predicted by this model. The important point is that the range of La/Yb compositions from all three volcanoes is produced by only a 2–3% variation in the degree of melting, which according to Baker & Stolper (1994) corresponds to a 10–20°C temperature variation across the zone of melting. This suggests that the plume beneath Cerro Azul, Alcedo and Sierra Negra is both similar in composition and well regulated in temperature. The slight decrease in the extent of partial melting from Alcedo to Sierra Negra to Cerro Azul is consistent with their respective positions relative to the center of the Galápagos mantle plume, which is proposed to lie beneath Fernandina (Graham et al., 1993; White et al., 1993; Kurz & Geist, 1999). The highest temperatures, along with the largest degree partial melts, would be expected closest to the plume center (Alcedo). Lower temperatures and smaller degree melts should produce a wider range of enriched compositions and would be expected for the edge of the plume (Cerro Azul).
Fractionation or differentiation at Cerro Azul
Systematic variation in the major-element oxides and trace elements, parallel REE plots, and constant incompatible trace-element ratios all suggest that the tholeiitic, transitional, and alkaline basalts of Cerro Azul are genetically related by fractional crystallization. Naumann & Geist (1999) presented a model for the production of Cerro Azul’s alkaline basalts from the primitive tholeiites by fractional crystallization of augite + plagioclase + olivine at pressures between 5 and 3 kbar, and it is briefly summarized here. A mass-balance calculation using the least-squares method (Bryan et al., 1969) and microprobe analyses of phenocrysts within the lavas indicates that a hypersthene-normative tholeiitic magma with 9·42% MgO (sample CA-28) could have produced a transitional basalt with 7·01% MgO (sample CA-37) through removal of 17% plagioclase, 14% augite, and 7% olivine, leaving 62% residual magma (Table 3). The sum of the squares of the residuals of the 10 major elements is 0·48. A nepheline-normative alkali–olivine basalt with 4·7% MgO can be modeled by the further removal of 15% plagioclase, 7% augite, and 5% olivine, the sum of the squares of the residuals being 0·25 (Table 3). The close correspondence of the mass-balance calculation to the compositional variations exhibited by these lavas provides further evidence that the alkali–olivine basalts are related to the tholeiites by fractional crystallization; this example suggests a total of 65% crystallization by weight.
|
The compositional variation of the Cerro Azul suite indicates that augite also became saturated with olivine and plagioclase as the parental tholeiitic magmas cooled and evolved to an MgO concentration of 8%. Below 8% MgO, the trends of major-element oxides and the CaO/Al2O3 and Sc/Yb ratios for Cerro Azul sharply change slope reflecting the addition of clinopyroxene as a fractionating phase (Fig. 8).
Comparison with Sierra Negra and Alcedo
The intershield similarities of major- and trace-element concentrations and ratios at MgO <8% suggest that similar magmatic evolution, involving the fractionating assemblage ol + pl + cpx, controlled the compositional evolution of basaltic compositions at all three volcanoes. At Alcedo, Geist et al. (1995) determined that the range of basaltic compositions (6·8–4% MgO) requires 40% fractionation of ol + pl + cpx, from an olivine tholeiite parent. At Sierra Negra, Reynolds & Geist (1995) determined that the range of basaltic compositions required 36% fractionation involving ol + pl + cpx from the most primitive lava to reproduce the observed compositional range. These intershield differences in the predicted degree of fractionation between Cerro Azul, Alcedo and Sierra Negra closely match the relative degree of enrichments represented by the range of chondrite-normalized REE plots in Fig. 12; Cerro Azul (65%) > Alcedo (40%) > Sierra Negra (35%). However, when the total range of compositions from Alcedo is considered, an additional 50% fractionation is required to produce the rhyolitic lavas (72% SiO2, 0·32% MgO). Thus, the range in fractionation is smallest at Sierra Negra. Both Alcedo and Cerro Azul show wide variations although the range of compositions at Cerro Azul trends to high MgO, whereas the compositional range of Alcedo’s lavas trends to low MgO.
Using a series of projection schemes that show changing liquidus assemblages with pressure (Grove et al., 1993; Yang et al., 1996), Geist et al. (1998) showed that magmas from Cerro Azul, Alcedo and Sierra Negra equilibrated at a range of pressures between 1 and 5 kbar (3·5–17·5 km). Both Alcedo and Sierra Negra compositions plot over a narrow range of pressures between 3 and 5 kbar. In contrast, Cerro Azul compositions plot over a wide range from near-surface pressures to >5 kbar, although the bulk of samples cluster between 1 and 3 kbar.
Implications of the intershield geochemical variations
The petrological and geochemical diversity between Cerro Azul, Alcedo and Sierra Negra records systematic variations in the melting and crystallization conditions and reflects the range of magmatic systems active in the western Galápagos today. Clearly, some process prohibits either the formation or eruption of primitive (MgO >6·5%) compositions at Alcedo and Sierra Negra but not at Cerro Azul.
Fractionation of ol + pl + cpx at pressures between 1 and 5 kbar has controlled liquid line of descent at all three volcanoes (Geist et al., 1998), but the highly differentiated compositions (rhyolite) at Alcedo can occur only where magma supply rates are low enough to isolate a large crustal magma chamber from input by mantle-derived basalt (Geist et al., 1995, after Christie & Sinton, 1981). Long-lived magma chambers capable of homogenizing melts have been predicted for the Galápagos volcanoes based on geochemical data (White et al., 1993; Geist et al., 1995, 1998; Reynolds & Geist, 1995; Allan & Simkin, 2000) and structural studies of caldera and fissure formation (Chadwick & Howard, 1991; Chadwick & Dieterich, 1995). White et al. (1993) argued that the uniform major-element and isotopic compositions of many Galápagos volcanoes are the result of magma chambers serving to homogenize melts passing through them. With the accumulation and mixing of multiple melt batches in a long-lived chamber, temperatures remain high and the composition of erupted magmas is evolved but restricted. We believe this is the case at Sierra Negra and the pre-rhyolite stage at Alcedo. If magmas are stored in smaller chambers and not thoroughly mixed, a range of compositions results. This is the case at Cerro Azul.
Petrogenetic model
In developing a petrogenetic model, we begin with the premise that the relatively restricted range of lava compositions at Sierra Negra reflects a well-regulated thermal environment in which a high supply of primitive magma is buffered as it passes through an ol + pl + cpx mush zone (e.g. Sinton & Detrick, 1992; Albarède et al., 1997) or a steady-state magmatic system in which magma mixing, combined with relatively constant resupply and eruption rates, effectively buffers the range of major-element compositions (O’Hara, 1977). The distinct geochemical and petrological trends seen at Cerro Azul and Alcedo are related as transient stages resulting from lower magma supply rates that are either leading towards or decaying from this steady state. Throughout this range, variations in the geometry of magma storage (small, disconnected chambers to large single chambers) and cooling rates (slow to fast) are controlled by the total supply rate of primitive magma to each volcano. Similar models were developed by Christie & Sinton (1981) and Sinton & Detrick (1992) for MORB, Meyer et al. (1985) for Iceland, and Frey et al. (1990) for Hawaii.
Evidence for variable magma supply rates between Cerro Azul, Sierra Negra and Alcedo may also be reflected in the shape and size of their calderas (Fig. 3), which are presumed to represent their magma chamber dimensions. If this is true, then a larger chamber means a higher magma supply. At 4 km x 5 km, Cerro Azul has the smallest caldera in the western Galápagos, and it has been constructed by the collapse of multiple, small (1–2 km width) chambers, resulting in a multi-nested, scalloped outline (Naumann & Geist, 2000). Alcedo’s caldera is intermediate in size (5 km x 7 km) with a smooth oval outline reflecting a single chamber (Geist et al., 1994), and, at 9 km x 11 km, Sierra Negra’s caldera is the widest in the western archipelago and also reflects the existence of a single large chamber (Reynolds & Geist, 1995).
Christie & Sinton (1981) showed how cooling rates of crustal magma bodies and magma supply rates must change as the magmatic system evolves behind the tip of a propagating rift and the effects of these changes on shallow fractionation. Throughout this evolution, magma supply rates gradually increase and cooling rates of crustal bodies decrease, resulting in a wide range of melt compositions directly at the propagating tip (low supply, fast cooling) to buffered Fe–Ti basalts of fairly uniform composition well behind the tip (high supply, slow cooling). A direct correlation can be made with these spreading ridge environments and those at Cerro Azul, Alcedo, and Sierra Negra (Fig. 14). At Cerro Azul, small volumes of high-MgO tholeiitic magmas are stored in multiple, small, ephemeral chambers at a wide range of depths between 3 and 15 km, which differentiate to variable extents, depending on the intrusion and structural histories of the individual chambers. Further, on occasion, odd primary magmas such as CA-36 bypass the currently active chambers and are not homogenized into the main melt batch. Some of the reason for this disrupted magma system may lie in the fact that with its location on the platform margin, Cerro Azul remains unbuttressed on two sides and has experienced numerous flank failures. This may allow for a deep-seated fracturing of the edifice, resulting in a more distributed set of magma conduits. This type of structural control may also explain the high concentration of alkaline lavas on this sector of the volcano. At Sierra Negra (and the pre-rhyolite stage of Alcedo), the volcano has been constructed by constant high resupply rates of magma to a large, steady-state chamber with low cooling rates, buffering the major-element compositions of the lavas and resulting in the eruption of Fe–Ti basalts with a narrow compositional range (4–7% MgO). At Alcedo, lower resupply rates more recently have resulted in an intermediate cooling rate and long residence times of magmas. This thermal history causes increased cooling and the evolution of compositions as differentiated as rhyolite. In the case of Alcedo, high degrees of crystal fractionation are favored only when the cooling rate exceeds the resupply rate of primitive magma (Fig. 14). The origin of odd lavas similar to CA-36 by larger extents of melting may occur at Sierra Negra and Alcedo but goes unnoticed because of their homogenization in a large chamber. It is only at Cerro Azul, with a plexus of smaller chambers, that compositional variability can be preserved.
|
A systematic variation in the rate of magma supply between the volcanoes helps to support the idea that the magma supply rate is the dominant control over the petrologic and geochemical variation between the volcanoes. A high magma supply rate at Sierra Negra is confirmed by the eruption rate calculations of Reynolds & Geist (1995)
(0·15 x 106 m3/yr); the supply rate of primitive magma from the mantle must exceed this value two-fold to account for crystallization of primary magma. In addition, subaerial volume and emergent age calculations for Cerro Azul, Sierra Negra, and Alcedo (Naumann & Geist, 2000) indicate that Sierra Negra is the most voluminous volcano in the Galápagos (588 km3), containing nearly 25% of the subaerial volume of the entire archipelago, and, because it is similar in age to the others, it must therefore have received the greatest supply of magma. We predict that the opposite is true for Cerro Azul (175 km3), which is <30% of the volume of Sierra Negra and has a much lower eruption rate (0·05 x 106 m3/yr; Naumann & Geist, 2000).
Naumann & Geist (2000) also proposed a model to explain the morphologic variation between these volcanoes that is based on differences in the magma supply rate to each volcano. If the western Galápagos volcanoes are similar in age, then the morphologic variations among them cannot represent a uniform and simple path of evolution as suggested by Nordlie (1973). Instead, each volcano has been constructed simultaneously under different magma supply rates, which, in turn, are reflected in their morphologic variability. Thus, a simple pattern of morphologic development related to temporal evolution may be absent in the Galápagos.
The control of magma supply rate in the Western Galápagos
The results presented here suggest that the compositional variations among Cerro Azul, Sierra Negra, and Alcedo reflect variations in magma supply rate of primitive magma to each volcano, but what is controlling the different rates? At Hawaii, the geochemical transitions (alkaline to tholeiitic magma series) are linked to age and position relative to the Hawaiian hotspot. At mid-ocean ridges (including Iceland), mantle temperature (and the resultant magma supply rate) is controlled by the spreading rate, the position relative to the propagating ridge tip, or the proximity to the cold edge of a transform fault (Meyer et al., 1985; Sinton & Detrick, 1992). White et al. (1993) recognized that magma supply rates have influenced the enrichment of Galápagos magmas and related the development of permanent chambers, homogenization of melts, and extreme fractionation to volcano age and crustal thickness. White et al. (1993) predicted that the western shields evolved through a series of stages that eventually resulted in the production of differentiated compositions like those of Alcedo. A problem with this explanation is that the western volcanoes are all young and aligned at a high angle to the Nazca plate motion, precluding a simple temporal progression for the differences in their petrologic variation. In addition, no systematic chemical trends (tholeiitic to alkaline) have yet been defined for any Galápagos volcano. Clearly, some factor other than volcanic age is controlling the magma supply rate.
A decrease in magma supply reflects a decrease in the degree of melting in the postshield volcanism on Haleakala (Chen & Frey, 1985), Mauna Kea (Frey et al., 1990) and Kohala (Lanphere & Frey, 1987). In the Galápagos, the calculated percent of partial melting does not simply relate to the volume of the erupted magmas. The average degree of partial melting systematically increases from Cerro Azul (2–3%) at the plume edge to Sierra Negra (3–4%) and Alcedo (5%), which are closer to the center of the plume. These melt differences represent only a 10–20°C temperature variation within the plume and are slight when compared with differences in eruption rate between the volcanoes, which differ by a factor of three. A 1–2% difference in the degree of partial melting between Cerro Azul and Sierra Negra cannot account for the differences in their eruption rates. This decoupling between the percent of melting and eruption rates in the Galápagos was also recognized by Geist et al. (1995) for the rhyolites at Alcedo, where a 10-fold decrease in the eruption rate was not accompanied by a drop in the overall degree of partial melting. It seems reasonable to assume that if differences in magma supply rate among the western shield volcanoes cannot be linked to the degree of melting then these differences may instead be linked to plume flow and shallow-level magma transport mechanisms.
When melt production from the rising Galápagos contacts the base of the lithosphere, rising melts would be focused to the surface by faults or fractures in the overlying lithosphere (Fig. 15). This ‘lithospheric filter’, superimposed above an eastward expanding plume, would promote a broad area of volcanism and would permit the simultaneous development of widely distributed volcanic centers, aligned at a high angle to the plate motion vector. A similar model is suggested for Iceland, where melts are focused into rifts as a result of lower lithostatic forces caused by fracturing (Meyer et al., 1985). Geophysical modeling by Feighner & Richards (1994) indicates that the volcanoes along the Wolf–Darwin lineament are also coeval and suggests that the Wolf–Darwin lineament is a crustal-scale fault produced by loading of the lithosphere, which has served to focus the passage of melts. In addition, Feighner & Richards (1994) speculated that similar features may provide pathways for magma and control the transport time of magma to the surface, allowing deep-generated magma to rise quickly to the surface. Such spatial patterns can be seen in other mid-plate volcanic chains that do not form clear time-transgressive chains, such as coeval Pacific Ocean seamounts near the East Pacific Rise, whose origin and morphologic evolution result from tectonic processes and thermal regimes at fracture zones (Batiza & Vanko, 1983), the Austral Islands, where the volume and location of midplate volcanism is also controlled by lithospheric stresses (McNutt et al., 1997), or the North Arch volcanic field near Hawaii, whose origin is ascribed to the flexure and cracking of the Pacific plate as a result of the Hawaiian swell (Clague et al., 1990).
|
Tectonic controls are also responsible for compositional variations associated with propagating rifts of the Galápagos Spreading Center (Christie & Sinton, 1981
), MORB (Sinton & Detrick, 1992), and Iceland (Meyer et al., 1985), where magma supply rate is linked to the thermal maturation of conduit systems. Thermal maturation of conduits is observed to take place on active fissure eruptions (Spera, 1980) where increased magma supply promotes a positive feedback as higher velocities create low-pressure regions and further temperature rises draw in more magma, leading to ever-increasing supply. In this way Sierra Negra, which contains >25% of the subaerial volume in the Galápagos (Naumann & Geist, 2000), may have become dominant at the expense of the surrounding magmatic systems (Fig. 15).
Focusing of melts along fissures also promotes the regular spacing of volcanoes. As melts migrate laterally towards areas of greatest flow, they focus into nodes that help draw more melt in, giving rise to the regular spacing. This is analogous to the segmentation of spreading ridges, where the greatest volume of flow is in the central portion of each segment (Whitehead et al., 1984).
The schemes shown in Fig. 16 are conceptual and illustrate a four-stage development of the western Galápagos archipelago:
- Stage 1: structural weaknesses (fractures?) are imposed on the thin lithosphere of Nazca plate as a result of its proximity to the nearby Galápagos Spreading Center (Fig. 16a).
- Stage 2: widespread fissure-type eruptions begin creation of the Galápagos platform. The location of eruptive fissures is controlled by the pre-existing structure of the Nazca plate (Fig. 16b).
- Stage 3: voluminous fissure eruptions continue to construct the Galápagos platform, which only slightly pre-dates the emergence of the volcanoes. The Nazca plate is flexed isostatically to compensate for the additional load. Isotopic variations documented by Graham et al. (1993)
, Harpp (1995), and Harpp & White (2001) show that compositions of lavas dredged from the surrounding platform match those at the nearby volcanoes (Fig. 16c).
- Stage 4: fissure eruption changes into focused, point source eruptions along the eruptive fissures that have controlled the passage of melts to the surface. Continuous eruptions produce the volcanoes of Isabela and Fernandina, whose alignment is due to the initial fracture zones and whose spacing is controlled by Rayleigh–Taylor instabilities above the melt segregation zone in the upwelling mantle plume. Variations in eruptive volumes and thermal maturation are controlled primarily by lithospheric weaknesses. Currently, Sierra Negra is receiving the highest supply. Roca Redonda (Fig. 1) has become emergent without creating a platform morphology similar to the rest of the western archipelago, possibly indicating that it represents a starved center where there has been low magma supply throughout the growth cycle of the other volcanoes (Standish et al., 1998
) (Fig. 16d).
- Stage 2: widespread fissure-type eruptions begin creation of the Galápagos platform. The location of eruptive fissures is controlled by the pre-existing structure of the Nazca plate (Fig. 16b).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF CERRO AZULRELATIVE AGE OF THE...SAMPLING AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Cerro Azul has erupted a wide range of tholeiitic to alkalic basalts. These diverse compositions are some of the most primitive yet reported from the western archipelago and are unlike those of other, well-studied, neighboring volcanoes of Sierra Negra and Alcedo, which have erupted basalt of fairly uniform composition. Major- and trace-element modeling shows that the source for Cerro Azul and Sierra Negra is identical, whereas the source for Alcedo’s basalts is slightly more isotopically depleted. Modeling also reveals that all three volcanoes share a similar depth of melting in the garnet stability field, and there are small (2–3%) systematic differences in the extent of partial melting between the volcanoes that can be related to their distance from the proposed plume center below the westernmost island of Fernandina. However, erupted volumes do not correlate with the variations in melting. Even though melts segregating from the plume in the western Galápagos reflect a narrow range of pressures and limited degree differences in the extent of partial melting, there are wide variations in the enrichments of major and trace elements between Cerro Azul, Alcedo and Sierra Negra, which result from different extents of cooling and crystallization in the lithosphere. The observed intershield geochemical differences result from magma supply and cooling rates that are unique to each volcano, and reflect the variations in lithospheric transport and storage processes across the western archipelago. Cerro Azul basalts have a wide range of major- and trace-element compositions but overall are less evolved than basalts from Alcedo and Sierra Negra. Variable degrees of fractionation at different depths have produced small volumes of tholeiitic through alkaline magmas, which are stored in multiple, ephemeral chambers, preserving the compositional variability of the primary magmas. At Alcedo, low supply rates combined with a long-established crustal magma body have resulted in the generation of compositions as differentiated as rhyolite. At Sierra Negra, a high supply of magma replenishes a large, thermally steady-state chamber that is well mixed and periodically tapped, buffering the composition of the lavas. Overall, the wide compositional range of Cerro Azul lavas indicates the range of magmatic conditions imposed by the position of Cerro Azul at the leading edge of the Galápagos plume, and a relatively low supply of magma when compared with the other volcanoes. We believe the unique alignment, spacing, morphology and chemical variations among the western shield volcanoes are not the result of their systematic development owing to the passage of the Nazca plate over the Galápagos plume, but rather the systematic control of a ‘lithospheric filter’, which either promotes or inhibits the passage of primary melts from the underlying plume.
|
ACKNOWLEDGEMENTS |
|---|
We thank the staff of the Charles Darwin Research Station and the Galápagos National Park for their permission and logistical support during fieldwork on the islands. Special thanks go to our guide and friend Eduardo Villema, without whom our expeditions would have failed. Thanks go also to Pablo Samaniego and Rommel Villagomez for able assistance in the field. Our friends at the Washington State University geoanalytical laboratory provided great service as always. Brian Cousens and Dave Christie provided constructive and thoughtful reviews. Finally, we wish to recognize the editorial prowess of Big George Bergantz. This research was funded by NSF research grants 91-17640 and 94-054462 to D.G. This is WHOI Contribution 10556.
|
FOOTNOTES |
|---|
*Corresponding author. Telephone: 907-786-6846. Fax: 907-786-6850. E-mail: aftrn{at}uaa.alaska.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF CERRO AZULRELATIVE AGE OF THE...SAMPLING AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Albarède, F., Luais, B., Fitton, G., Semet, M., Kaminski, E., Upton, B. G. J., Bachelery, P. & Cheminée, J.-L. (1997). The geochemical regimes of Piton de la Fournaise volcano (Réunion) during the last 530 000 years. Journal of Petrology 38, 171–201.
Allan, J. F. & Simkin, T. (2000). Fernandina Volcano’s evolved, well-mixed basalts; mineralogical and petrological constraints on the nature of the Galapagos plume. Journal of Geophysical Research 105, 6017–6041.
Baker, M. B. & Stolper, E. M. (1994). Determining the composition of high-pressure mantle melts using diamond aggregates. Geochimica et Cosmochimica Acta 58, 2811–2827.[Web of Science]
Banfield, A. F., Behre, C. H., Jr & St Clair, D. (1956). Geology of Isabela (Albemarle) Island, Archipelago de Colon (Galápagos). Geological Society of America Bulletin 67, 215–234.
Batiza, R. & Vanko, D. (1983). Volcanic development of small oceanic central volcanoes on the flanks of the East Pacific Rise inferred from narrow-beam echo-sounder surveys. Marine Geology 54, 53–90.
Bow, C. S. (1979). The geology and petrogenesis of the lavas of Floreana and Santa Cruz islands: Galápagos islands. Ph.D. dissertation, University of Oregon, Eugene.
Bryan, W. B., Finger, L. W. & Chayes, F. (1969). Estimating proportions in petrographic mixing equations by least-squares approximation. Science 163, 926–927.
Chadwick, W. W. & Howard, K. A. (1991). The pattern of circumferential and radial eruptive fissures on the volcanoes of Fernandina and Isabela islands, Galápagos. Bulletin of Volcanology 53, 259–275.
Chadwick, W. W. & Dieterich, J. H. (1995). Mechanical modeling of circumferential and radial dike intrusion on Galápagos volcanoes. Journal of Volcanology and Geothermal Research 66, 37–52.[Web of Science]
Chen, C. Y. & Frey, F. A. (1985). Trace element and isotopic geochemistry of lavas from Haleakala volcano, East Maui, Hawaii: implications for the origin of Hawaiian basalts. Journal of Geophysical Research 90, 8743–8769.
Christie, D. M. & Sinton, J. M. (1981). Evolution of abyssal lavas along propagating segments of the Galápagos spreading center. Earth and Planetary Science Letters 56, 321–335.
Clague, D. A., Holcomb, R. T., Sinton, J. M., Detrick, R. S. & Torresan, M. E. (1990). Pliocene and Pleistocene alkalic flood basalts on the seafloor north of the Hawaiian Islands. Earth and Planetary Science Letters 98, 175–191.
Darwin, C. (1860). The Voyage of the Beagle. Doubleday: New York.
Feighner, M. A. & Richards, M. A. (1994). Lithospheric structure and compensation methods of the Galápagos archipelago. Journal of Geophysical Research 99, 6711–6729.
Frey, F. A. & Clague, D. A. (1983). Geochemistry of diverse basalt types from Loihi seamount, Hawaii: petrogenetic implications. Earth and Planetary Science Letters 66, 337–355.
Frey, F. A., Wise, W. S., Garcia, M. O., West, H., Kwon, S. T. & Kennedy, A. (1990). Evolution of Mauna Kea volcano, Hawaii: the transition from shield building to the alkalic cap stage. Journal of Geophysical Research 95, 1271–1300.
Geist, D. J. (1992). An appraisal of the melting process and the Galápagos hotspot: major and trace element evidence. Journal of Volcanology and Geothermal Research 52, 65–82.
Geist, D. J., White, W. M. & McBirney, A. R. (1988). Plume–asthenosphere mixing beneath the Galápagos Archipelago. Nature 333, 657–660.
Geist, D. J., Howard, K. A., Jellinek, A. M. & Rayder, S. (1994). Volcanic history of Volcán Alcedo, Galápagos Archipelago: a case study of rhyolitic oceanic volcanism. Bulletin of Volcanology 56, 243–260.[Web of Science]
Geist, D., Howard, K. A. & Larson, P. (1995). The generation of oceanic rhyolites by crystal fractionation: the basalt–rhyolite association at Volcan Alcedo, Galápagos Archipelago. Journal of Petrology 36, 965–982.
Geist, D., Naumann, T. R. & Larson, P. L. (1998). Evolution of Galápagos magmas: mantle and crustal fractionation without assimilation. Journal of Petrology 39, 953–971.
Ghiorso, M. S. & Sack, R. O. (1995). Chemical mass transfer in magmatic processes. IV. A revised and internally consistent thermodynamic model for the interpolation and extrapolation of liquid–solid equilibria in magmatic systems at elevated temperatures and pressures. Contributions to Mineralogy and Petrology 119, 197–212.[Web of Science]
Graham, D. W., Christie, D. M., Harpp, K. S. & Lupton, J. E. (1993). Mantle plume helium in submarine basalts from the Galápagos platform. Science 262, 2023–2026.
Grove, T. L., Kinzler, R. J. & Bryan, W. B. (1993). Fractionation of mid-ocean ridge basalt (MORB). Geophysical Monograph, American Geophysical Union 71, 281–310.
Harpp, K. S. (1995). Magmatic evolution at hotspots and mid-ocean ridges: isotopic and trace element studies from the Galápagos Islands, the East Pacific Rise and Mid-Atlantic Ridge. Ph.D. dissertation, Cornell University, Ithaca, NY.
Harpp, K. S. & White, W. M. (2001). Tracing a mantle plume: isotopic and trace element variations of Galápagos seamounts. Geochemistry, Geophysics, Geosystems (electronic article) 2000GC000137.
Hess, P. C. (1992). Phase equilibria constraints on the origin of ocean floor basalts. Geophysical Monograph, American Geophysical Union 71, 67–102.
Johnson, D. M., Hooper, P. R. & Conrey, R. M. (1999). XRF analysis of rocks and minerals or major and trace elements on a single low dilution Li-tetraborate fused bead. Advances in X-ray Analysis 41, 843–867.
Johnson, K. T. M., Dick, H. J. B. & Shimizu, N. (1990). Melting in the oceanic upper mantle: an ion microprobe study of diopsides in abyssal peridotites. Journal of Geophysical Research 95, 2661–2678.
Kurz, M. D. & Geist, D. J. (1999). Dynamics and evolution of the Galápagos hotspot from helium isotope geochemistry. Geochimica et Cosmochimica Acta 63, 4139–4156.
Kurz, M. D., Kenna, T. C., Kammer, D., Rhodes, J. M. & Garcia, M. O. (1995). Isotopic evolution of Mauna Loa volcano: a view from the submarine southwest rift. Geophysical Monograph, American Geophysical Union 92, 289–306.
Lanphere, M. & Frey, F. A. (1987). Geochemical evolution of Kohala Volcano, Hawaii. Contributions to Mineralogy and Petrology 95, 100–112.
Macdonald, G. A. & Katsura, T. (1964). Chemical composition of Hawaiian lavas. Journal of Petrology 5, 82–133.
McBirney, A. R. & Williams, H. (1969). Geology and Petrology of the Galápagos Islands. Geological Society of America, Memoir 118, 197 pp.
McNutt, M. K., Caress, D. W., Reynolds, J., Jordahl, K. A. & Duncan, R. A. (1997). Failure of plume theory to explain midplate volcanism in the southern Austral islands. Nature 389, 479–482.[Web of Science]
Meyer, P. S., Sigurdsson, H. & Schilling, J.-G. (1985). Petrological and geochemical variations along Iceland’s neovolcanic zones. Journal of Geophysical Research 90, 10043–10072.
Morgan, W. J. (1971). Convective plumes in the lower mantle. Nature 230, 42–43.
Naumann, T. R. (1998). Geology and petrology of Cerro Azul volcano, Isabela Island, Galápagos Archipelago. Ph.D. dissertation, University of Idaho, Moscow, 153 pp.
Naumann, T. R. & Geist, D. J. (1999). Generation of alkaline basalt by crystal fractionation of tholeiitic magma. Geology 27, 423–426.
Naumann, T. R. & Geist, D. J. (2000). Physical volcanology and structural development of Cerro Azul volcano, Isabela Island, Galápagos: implications for the development of Galápagos-type shield volcanoes. Bulletin of Volcanology 61, 497–514.[Web of Science]
Nordlie, B. E. (1973). Morphology and structure of the western Galápagos volcanoes and a model for their origin. Geological Society of America Bulletin 84, 2831–2956.
O’Hara, M. J. (1977). Geochemical evolution during fractional crystallization of a periodically refilled magma chamber. Nature 266, 503–507.
Reynolds, R. W. & Geist, D. J. (1995). Petrology of lavas from Sierra Negra volcano, Isabela Island, Galápagos archipelago. Journal of Geophysical Research 100, 24537–24553.[Web of Science]
Sinton, J. M. & Detrick, R. (1992). Mid-ocean ridge magma chambers. Journal of Geophysical Research 97, 197–216.
Sinton, C. S., Christie, D. M. & Duncan, R. A. (1996). Geochronology of Galápagos seamounts. Journal of Geophysical Research 101, 13689–13700.
Spera, F. J. (1980). Aspects of magma transport: In: Hargraves, R. (ed.) Physics of Magmatic Processes. Princeton, NJ: Princeton University Press pp. 265–323.
Standish, J., Geist, D., Harpp, K. & Kurz, M. (1998). The emergence of a Galápagos shield volcano, Roca Redonda. Contributions to Mineralogy and Petrology 133, 136–148.
White, W. M., McBirney, A. R. & Duncan, R. A. (1993). Petrology and geochemistry of the Galápagos Islands: portrait of a pathological mantle plume. Journal of Geophysical Research 98, 19533–19563.
Whitehead, J., Dick, H. J. B. & Schouten, H. (1984). A mechanism for magmatic accretion under spreading centers. Nature 312, 146–148.
Wilson, D. S. & Hey, R. N. (1995). History of rift propagation and magnetization intensity for the Cocos–Nazca spreading center. Journal of Geophysical Research 100, 10041–10056.
Yang, H. J., Frey, F. A., Rhodes, J. M. & Garcia, M. O. (1996). Evolution of Mauna Kea volcano: inferences from lava compositions recovered in the Hawaii Scientific Drilling Project. Journal of Geophysical Research 101, 11747–11767.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. J. GEIST, T. R. NAUMANN, J. J. STANDISH, M. D. KURZ, K. S. HARPP, W. M. WHITE, and D. J. FORNARI Wolf Volcano, Galapagos Archipelago: Melting and Magmatic Evolution at the Margins of a Mantle Plume J. Petrology, November 1, 2005; 46(11): 2197 - 2224. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||