This item is only available as the following downloads:
1 A GEOCHEMICAL AND VOLCANOLOGICAL INVESTIGATION OF PLUME RIDGE INTERACTION AT THE DISTAL ENDS OF THE GALPAGOS SPREADING CENTER (86 89.5W AND 97.5W) By KATRINA A LYCE GARMAN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012
2 2012 Katrina A lyce Garman
3 To my family
4 ACKNOWLEDGMENTS I am eternally grateful for my friends, family, and colleagues who have supported me whether financially, academically, or emotionally through the course of my masters research. I thank Dr. Mike Perfit wholeheartedly for his encouragement and mentorship, as well as financial support for the writing of this thesis. I also tha nk Dr Dave F oster and Dr. Ray Russo for valuable input and guidance in interpreting data and shaping my research appr oach as members of my committee, and Dr. Liz Screaton for stepping in at the last minute as a substitute for my defense. I would also like to acknowled ge Dr. George Kamenov for his assistance and unfailing patience in instrument oper ation, Dr. Matt Smith for his insight into geochemical data analysis and Dr. Rachel Walters for countless hours of assistance with geochemical modeling and thesis writing (a nd afternoon tea ) Finally, my degree would not have been completed without the constant support of all of my family and friends.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 2 BACKGROUND ................................ ................................ ................................ ...... 16 3 GEOCHEMICAL INVESTIGATION OF PLUME RIDGE INTERACTION ON THE DISTAL ENDS OF THE GALPAGOS SPREADING CENTER ............................. 22 Research Goals ................................ ................................ ................................ ...... 22 Methods ................................ ................................ ................................ .................. 22 Results ................................ ................................ ................................ .................... 25 Major Elements ................................ ................................ ................................ 25 Trace Element Variations ................................ ................................ ................. 26 Discussion ................................ ................................ ................................ .............. 28 G eochemical Modeling ................................ ................................ ..................... 28 Petrogenesis on the GSC ................................ ................................ ................. 31 4 LAVA MORPHOLOGY AT 97.5W ON THE GALPAGOS SPREADING CENTER ................................ ................................ ................................ ................. 50 Mid ocean Ridge Lava Morphology ................................ ................................ ........ 50 Methods ................................ ................................ ................................ .................. 51 Argo II Camera and Bathymetric Surveys ................................ ........................ 51 Image Analysis ................................ ................................ ................................ 51 Slope Analysis ................................ ................................ ................................ .. 52 Results ................................ ................................ ................................ .................... 53 Lava Morpholo gies and Relative Ages ................................ ............................. 53 Morphology and Slope ................................ ................................ ..................... 54 Discussion ................................ ................................ ................................ .............. 55 5 CONCLUSIONS ................................ ................................ ................................ ..... 70
6 LIST OF REFERENCES ................................ ................................ ............................... 72 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 79
7 LIST OF TABLES Table page 3 1 Major element concentrations of GSC lavas. ................................ ..................... 33 3 2 Trace element concentrations of GSC lavas. ................................ ..................... 34 3 3 Partition coefficients for fractional crystallization calculations. ............................ 38 4 1 Distribution of lava flow morphologies in regions of increasing slope. ................ 60 4 2 Relative sediment cover of western GSC lavas. ................................ ................. 61
8 LIST OF FIGURES Figure page 1 1 Map of Galpagos Region ................................ ................................ ................. 15 3 2 K/Ti ratio of Galpagos Spreading Center (GSC) lavas ................................ ..... 41 3 3 Spider diagram of GSC lava samples ................................ ................................ 42 3 4 Rare earth element diagram of GSC lava samples ................................ ............ 43 3 5 Nb/Zr vs. Ce/Yb of distal GSC lavas. ................................ ................................ .. 44 3 6 Mantle normalized Nb/La vs Longitude of distal GSC lavas ............................. 45 3 7 Major element fractional crystalliza tion model of eastern GSC lavas ................. 46 3 8 Major element fr actional crystalliza tion model of western GSC lavas ................ 47 3 9 Trace element fractional crystallization models of eastern GSC lavas. .............. 48 3 10 Trace element fracti onal crystallization models of western GSC lavas .............. 49 4 1 Location and bathymetry of the Galpagos Spreading Center ...... 62 4 2 Lava morphologies of the western GSC. ................................ ............................ 63 4 3 Lava morphology distribution at 97.5 W on the GSC. ................................ ........ 64 4 4 Slope of terrain at 97.5 W on the GSC. ................................ .............................. 65 4 5 Maps of lava morphology distribution at 97.5W ................................ ............... 66 4 7 Sediment cover at 97.5 W on the GSC. ................................ ............................. 68 4 8 Lava morphology distribution by slope ................................ ............................... 69
9 LIST OF ABBREVIATION S AHA Autonomous Hydrophone Array AMC axial magma chamber Cpx clinopyroxene EGSC Eastern Galpagos Spreading Center E MORB enriched mid ocean ridge basalt ESC electronic still camera GSC Galpagos Spreading Center LLD liquid line of descent MOR mid ocean ridge MORB mid ocean ridge basalt N MORB normal mid ocean ridge basalt Ol olivine Plag plagioclase REE rare earth element T MORB transitional mid ocean ridge basalt WGSC Western Galpagos Spreading Center
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science A GEOCHEMICAL AND VOLCANOLOGICAL INVESTIGATION OF PLUME RIDGE INTERACTION AT THE DISTAL ENDS OF THE GALPAGOS SPREADING CENTER (86 89.5W AND 97.5W) By Katrina Alyce Garman August 2012 Chair: Michael Perfit Major: Geology The Galpagos region of the equatorial eastern Pacific is a site of historically documented interaction between a hotspot and a mid ocean ridge. This study serves to fill in the gaps in the existing literature, which has focused primarily on the central portions of the Galpagos Spreading Center (GSC) those closest to the locus of the Galpagos H otspot near 91W longitude to examine whether the hotspot may be contributing significant volumes of melt to the lavas that are erupted at the distal ends of the ridge. Major and trace element concentrations of lavas from 97.5W on the western GSC and 86 89.5W on the eastern GSC, in conjunction with the morphology of erupted lavas at 97.5W, are used as indicators of the petrogenesis and magmatic supply in these portions of the ridge. The central portions of the GSC have been found to exhibit evidence of an enriched mantle source and elevated magmatic supply relative to other ridges of similar spreading rate, with a waning influence in both directions away from 91W. Our results are consistent with these findings, as the study region nearest the plume 89.5 W is characterized by relatively high incompatible element ratios (K/Ti, Nb/Zr, Nb PM /La PM )
11 and greater degrees of fractionation than the distal regions as evidenced by major element variations, low MgO content, and f ractional crystallization model calcul ations In contrast, the study areas located farthest from the hotspot 97.5W and 86W are found to produce normal mid ocean ridge basalt (N MORB), with mafic compositions and depleted incompatible element signatures. The westernmost study region is also c haracterized by lava morphologies pillows and talus that are typically found on spreading ridges with relatively low magma supply. However, we do find evidence of occasional lobate flows and rare sheet flows, even at this distal portion of the ridge, which suggests that local variations in effusion rate and possibly underlying slope may be playing a role in determining lava morphology on the GSC.
12 CHAPTER 1 INTRODUCTION ocean ridge (MOR) system represents the most magmatically active setting on the planet, yet the dynamics and complexities of MORs are still poorly understood. The Galpagos region is a particularly interesting geolog ic setting, as it is a location where the interaction between a mid ocean ridge and a mantle plume occurs. The Galpagos Spreading Center (GSC) extends westward from approximately 85W to beyond 100W in the equatorial Pacific, and the Galpagos Hotspot is currently centered at 91W (Detrick et al., 2002) ( Figure 1 1 ). The central segments of the GSC, located closest to the plume locus, have been the focus of research for several decades ( e.g., Schilling et al., 1976 ; Laverne and Vivier, 1983 ; Christie and Sinton, 1986 ; Canales et al., 2002; Detrick et al., 2002 ; Sinton et al., 2003 ; Cushman et al., 2004 ; Rotella et al., 2009 ), whereas the distal ends of the ridge have not been studied nearly as extensively. Schilling et al. (1976) suggested that east of 87 W and west of 95W, the geochemical and petrologic signature of the hotspot wanes; however, comprehensive data that extend to these regions are scarce. This paper fills the gap in the existing literature and construct s a more complete record of the geochemical and volcanic signature over the spatial extent of the GSC. The structure of a mid ocean ridge has been linked to spreading rate ( Macdonald 1982 ; Perfit and Chadwick, 1998 ; Sinton et al. 2002 ), but many features of magmatism, both on axis and off, remain poorly understood. The morphology of GSC lavas reflects the physical conditions that govern magmatism at the spreading center. Lava flow morphology has been attributed to both spreading rate (Perfit and Chadwick, 1998) and magma supply (White et al., 2008 ; Colman et al., in press ); processes that may in fact
13 be directly related. The presence of the Galpagos Hotspot provides a potential source for increased magmatic supply at the GSC, which may shape the spatial variation in morphology of the r idge and lavas erupted there (Behn et al., 2004; Christie et al., 2005; White et al., 2008). Here, we use major and trace element compositions, bathymetric data, and seafloor photographs collected by deep towed surveys to examine the sources, distribution and morphology of lava flows from the eastern most (86W 89.5W) and westernmost (97.5W) portions of the GSC ( Figure 1 1 ). The eastern GSC was investigated by the Galpagos Rift Exploration cruise from May 24 June 4, 2002 to revisit the site of the first discovered hydrothermal vent communities. Rock samples, digital images, bathymetry, magnetics, and bottom water properties were collected between 86 89.5 W. The western portion of the Galpagos Spreading Center investigated here was chosen for study as par t of the AHA Nemo Leg 2 cruise (March 24 May 10, 2000), following detection of possible seafloor eruptions by the Autonomous Hydrophone Array (AHA) in the eastern Pacific. Both dredge samples and rock cores were collected between in addition t o bathymetric data and photographs of the seafloor. According to expectations raised by previous studies ( Sinton et al., 2003; White et al., 2008 ) the Galpagos H otspot should not be geochemically detectable within these regions of the GSC, and lava morph ologies here should be dominated by pillow basalt flows and volcanic cones. Here we show that on the eastern GSC, in the portion of the study area nearest the plume (89.5W), MOR lavas show evidence of an incompatible element enriched source component. Far ther from the hotspot, at both 86W and
14 97.5W, the geochemical signature of GSC lavas reflects normal mid ocean ridge basalt (N MORB) composition. Lava morphologies found at 97.5W are predominantly pillow lavas and talus, consistent with the models and h ypotheses proposed by previous studies. However, lobate flows are found in moderate abundance, and the presence of sheet flows has been confirmed which suggests that factors influencing lava morphology vary on a local scale Evidence for the range of hots pot influence on the GSC will help us to discern petrogenesis and the dynamics of mid ocean ridge plumbing systems and also the ways in which hotspot material is distributed spatially, which will lead us to a better understand ing of mantle dynamics on a global scale.
15 Figure 1 1 Region map of the central eastern Pacific showing the Galpagos Spreading Center (GSC) modified from Rotella, M.D., J.M. Sinton, J.J. Mahoney and W. Chazey, III (2009). Geochemical evidence for low magma supply and inactive propa gation at the Galapagos 93. 25 W overlappi ng spreading center. Geochemi stry Geophysics Geosystems. 10. Q 09005 The GSC is traced in blue, and the Galpagos Islands lie directly to the south Orange circle denotes current location of the hotspot at 91 92W, as illustrated by Detrick et al. (2002). Red brackets indicate the western study region (97.5W), explored by the AHA Nemo 2 cruise in 2000; blue brackets denote the eastern study region (86 89W) investigated by the Galpagos Ri ft Exploration cruise in 2002. Large a rrows indicate direction of relative plate motion.
16 CHAPTER 2 BACKGROUND The center of plume ridge interaction along the GSC has been the focus of geochemical and geophysical research since the mid 1970s (e.g., Schilling et al., 1976; Canales et al., 2002; Detrick et al., 2002; Schilling et al., 2003; Sinton et al., 2003; Behn et al, 2004). The opposing ends of the spreading center have not received nearly as much attention. Sinton et al. (2003) Cushman et al. (2004) and Colman et al. (in press) discussed plume ridge interaction along the western GSC that extended to nearly 98 W, and the region from 85.5W 91W has been the focus of a number of investigations (Allmendinger and Riis, 1979; Ballard et al., 1979; van Andel and Ballard, 1979; Ballard et al., 1982; Emmerman and Hubberten, 1983; Fornari et al., 1983; Perfit and Fornar i, 1983; Perfit et al., 1983; Christie et al., 2005), however few studies have attempted to synthesize both the chemical and physical characteristics of the distal reaches of the GSC. The region of maximum chemical and physical influence of the Galpagos H otspot on the ridge lies between 91W 92W (Schilling et al., 1976; Detrick et al., 2002; White et al., 2008; Ingle et al., 2010), just north of the presumed current plume locus beneath Fernandina Island (Figure 1 1) The GSC spreading rate ranges from 4 8 mm/yr at 98W to a maximum of 6 2 mm/yr at 86W (DeMets et al., 2010 ). Based on these rates, the GSC was classified as an intermediate spreading ridge by Wilson and Hey (1995). The segments farthest from the Galpagos H otspot (west of 95W) resemble a slow spreading ridge with a characteristic rift valley. Nearer to the hotspot (east of ~92.7W), a transition to an axial high is apparent in the bathymetry, which resembles a fast spreading ridge (Detrick et al., 2002). Sinton et al. (2003) and White et al. ( 2008) divide d the region into three
17 morphological provinces, with a transitional region between ~95W and 92.7W. Seismic investigations indicate that the oceanic crust thickens approaching 91W from the west, and gravity anomalies suggest reduced mantle d ensity in the same locations (Canales et al., 2002). Colman et al (in press) argue that the average crustal magma supply increases between 95W and 92W. Geochemistry and volcanology. Schilling et al. (1976) first systematically sampled and described the g eochemistry of lavas erupted along the GSC, and found ocean ridge basalts (N MORB), characterized by depleted, chondrite normalized, light rare earth element (REE) patterns, are only found east of 87W and west of 95W along the ridge. Fu enriched ocean ridge basalts (E MORB) dominate the region adjacent to the Galpagos Archipelago, which they attributed to a mantle plume rising beneath the Galpagos platform. Nearly the entire range (>87%) in isotopic variabil ity along the western GSC from 91W 98W can be attained by invoking two mantle sources: an incompatible element enriched component, which dominates the central portion of the ridge near 91 92W and wanes with increasing westward distance; and a depleted c omponent, which dominates the western end of the ridge beyond 95.5W (Ingle et al., 2010). The incompatible element ratio K/Ti (K 2 O/TiO 2 ) is used to track the extent of enrichment of MORB along the axis (Detrick et al., 2002; Cushman et al., 2004; Ingle et al., 2010). Lavas west of 95.5W had K/Ti ratios < 0.09, and thus are classified as N MORB. Between 95.5W and 92.7W, the lavas are transitional (T MORB), with K/Ti of 0.09 0.15. East of 92.7W to 91W, E MORB prevails, with K/Ti > 0.15 (Detrick et al., 2002). These data, in addition to increasing H 2 O contents (Cushman et al., 2004) and incompatible element ratios
18 (Ingle et al., 2010) from west to east along the GSC approaching the hotspot ~91W, support the early conclusions of Schilling et al. (1976), a nd provide geochemical evidence of the influence of the Galpagos Hotspot on the mid ocean ridge. Detrick et al. (2002) also suggested that decreasing average lava Mg# (Mg 2+ /(Mg 2+ +Fe 2+ )) from west to east across the western GSC was a consequence of increas ed crystal fractionation in magmas with the greatest plume influence. Recent studies and compilations have shown that the supply of melt to a given portion of the ridge is correlated with the geochemical signatures of lavas erupted (Rubin et al., 1998; Rub in et al., 2001; Rubin and Sinton, 2007), and often melt supply is fundamentally linked to spreading rate. In general, regions with high melt supply (usually fast spreading ridges) produce relatively highly differentiated (e.g., low MgO) lavas, but with a narrow range in mantle source characteristics (e.g., incompatible trace element and isotopic ratios) (Sinton and Detrick, 1992; Rubin and Sinton, 2007). In contrast, low melt supply regions often consist of more mafic lavas (higher MgO) derived from mantle with a wider range of compositions. This inverse relationship between differentiation and source composition is attributed to the thermal conditions of the upper crust: in fast spreading regions with high local magma supply, the crust is young, relatively hot, and weak. Here, low volume eruptions occur frequently after accumulation of relatively small batches of magma (Rubin et al., 2001). Slow spreading regions, however, have comparatively cooler and stronger crust, where eruptions are less frequent and p roduce greater volumes of lava. Additionally, small, steady state melt lenses have been seismically imaged at shallow depths beneath ridges with high magma supply, whereas low supply areas have melt lenses that are either deeper or
19 not detectable, and that very rarely reach steady state (Perfit and Chadwick, 1998). The presence of a magma lens allows for greater likelihood of attaining compositional homogeneity, while allowing for differentiation to occur over time of residence in the crust (Rubin et al., 2 001). those ridges that are not influenced by magmatic input from a nearby hotspot. However, on the Galpagos Spreading Center, magmatic input from the Galpagos Hotspot generates variations in magma supply along the ridge, even where spreading rate is nearly constant (DeMets et al., 2010). Recent research on the GSC has found that areas that have high magma supply (nearest the hotspot) are characterized by relatively differentiated lavas, as expected for high supply regions, however the compositional range is higher than in areas of low magmatic supply (Colman et al., in press). Additionally, only the central portions of the GSC, nearest the hotspot where melt supply is greater, hav e seismically detected melt lenses (Blacic et al., 2004). With increasing distance to the east and west of 91W, imaged melt lenses progressively deepen, and eventually become undetectable. A survey of the western GSC, with particular emphasis on the overl apping spreading center (OSC) at 93.25W, found that parental geochemical compositions are not identical along axis. Neighboring dredge samples along the western GSC were found to vary in parental melt composition and extent of fractionation over distances of 11 km or less; therefore, eruptions at this location tap magma chambers that are either small or poorly mixed and chemically heterogeneous (Rotella et al., 2009).
20 The bathymetry of the GSC appears to be symmetrical about the ridge hotspot intersection and the eastern ridge axis transitions from axial high to rift valley from west to east away from 91W similar to bathymetry changes observed away from the hotspot to the west (Sinton et al., 2003; White et al., 2008; Shorttle et al, 2010). However, in c ontrast to the physical symmetry, the compositional signature of the ridge is not symmetrical about the Galpagos H otspot with more incompatible element depleted signatures along the eastern limb than in the west To explain this difference, Shorttle et a l. (2010) invoke d plume flow paths of different lengths on either side of the plume center, with longer transport distances leading to the relative depletion to the east Laboratory experiments using polyethylene glycol (PEG) have been conducted to a nalyze the effect of slope on submarine lava morphology (Gregg and Fink, 2000). flow s, somewhat akin to submarine sheet flows that commonly form lava channels, were generated, as the wax could flow farther before solidifying. Based on these results, it was assumed that slope can play a pivotal role in the emplacement of submarine lava flo ws. We would expect to find lobate and sheet flows on steep slopes, and pillow lavas on shallowly dipping terrain (Gregg and Fink, 2000). However, later studies of opposit of the solid ifying basaltic crust must play a greater role in determining lava morphology
21 than the laboratory experiments demonstrate, as the chilled surface of the lava produces a solid margin that maintains the integrity of the lava structure and inhibits flow over great distances (Gregg and Smith, 2003).
22 CHAPTER 3 GEOCHEMICAL INVESTIGATION OF PLUME RIDGE INTERACTION ON THE DISTAL ENDS OF THE GALPAGOS SPREADING CENTER Research Goals The geochemical signatures of lavas erupted on the Galpagos Spreading Center are invo ked to try to assess the impact the Galpagos H otspot asserts on the magmatic system of the ridge. In recent years, the relationship between magma supply, lava morphology, and the composition of lavas erupted at MOR have been determined (e.g., Rubin et al. 2001; Rubin and Sinton, 2007), and scientists are now beginning to investigate the implications of complicating MOR magmatic systems by the additional melt influx from a nearby hotspot (Colman et al., in press). Here, we use major and trace element conce ntrations of lavas from the distal ends of the GSC (86 89.5W and 97.5W) to show that at the farthest reaches of the Galpagos Spreading Center, geochemical variability largely reflects episodic, low magma supply eruptions, whereas locations more proximal to the Galpagos H otspot (here, 89.5W) exhibit geochemical fingerprints that are much more likely to be the result of magmatic influx from the plume. Methods The eastern study area (86W 89.5W) was investigated by the Galpagos Rift 2002 cruise, part of the NOAA Ocean Exploration Program, in May June 2002 on board the R/V Atlantis (http://oceanexplorer.noaa.gov/explorations/02Galpagos/galapagos.html). This c ruise included a revisit to a previously sampled location near 86W where some of the et al., 2002; Shank et al., 2003), in addition to expanding westward along the GS C to 89.5W. Seventeen individual rock samples were collected for geochemical analysis by nine dives with the deep submergence vehicle Alvin The autonomous underwater vehicle
23 ABE (Autonomous Benthic Explorer) was used to conduct high resolution nighttime mapping of the water column and seafloor to direct Alvin surveys to locations of thermal anomalies and structures of interest. Alvin traverses totaled 12.5 km of seafloor, and ABE documented ~3 km 2 area (Shank et al., 2003). The western study area, from 97 Nemo 2 cruise in April May 2000 on board the R/V Melville Thirteen rock dredges and eighteen rock cores were c ollec ted in the western GSC during the cruise. Dredge locations were determined either by assuming tha t directly determining the location using transponder navigation. Rock wax core locations were selected near 97.5W for sampling of individual morphological features, such as volcanic cones. The posi code GPS position at the time of impact of the rock corer. Major and trace element analysis. Glass chips from quenched lava rinds from both 86 89W and 97.5W were selected for major and trace element analys is. Glass fragments were cleaned for 5 7 minutes in an H 2 O HCl H 2 O 2 acid mixture in an esonifier, and hand picked to avoid phenocrysts, alteration, or Mn oxide crust using a stereoscope. Major element concentrations of lavas were determined by analyzing g lass chips on a JEOL 8900 Electron Microprobe at the USGS Microbeam Laboratory in Denver, Colorado using USGS mineral standards to calibrate the analyses, and full ZAF corrections were applied. Seven to ten individual points were analyzed per sample. The JdF house standard ALV 2392 9 from the East Pacific Rise were used for secondary normalizations and to monitor instrument drift (Smith et al., 2001). Variations in analyses of ALV 2392 9 and repeat analy sis of
24 3% for most elements except MnO, K 2 O, and P 2 O 5 which can have higher error where concentrations are low (<0.2 wt%). To prepare glass chips for trace element analysis, ~50 mg of selected glass chips (2 6 mm in diameter) were added to 7 ml Teflon Savillex hexavials that had been pre cleaned in 14 N HNO 3 for 24 hours and then in 14 N HCl for 24 hours, followed by 24 hour hotplate reflux in 6 N HCl. Vials were washed between cl eaning steps with 4X quartz distilled deionized H 2 O. The glass chips were leached in 1 mL 2N optima grade HCl + 2 mL 4X H 2 O for ten minutes on a hot plate at 120C. The liquid was then removed, 2mL of 4X H 2 O were added and the vials were returned to the ho t plate for another ten minutes. The liquid was removed and the leached glass chips were then dried down. The glass chips then were dissolved in 1 mL of optima grade HF + 2 mL of optima grade HNO 3 for 24 hours at 120C. After the glass was completely disso lved the liquid w as evaporated until only the dry samples remained For trace element analysis, ~4.5 mL of 5% HNO3+100 ppm HF spiked with 8 ppb of Re and Rh were added to the residue of each sample which was capped and dissolved on a hotplate overnight. A procedural blank was prepared along with the samples to monitor contamination effects and purity of acids used. Unknowns were diluted 2000x and analyzed using a high resolution magnetic sector Element2 Inductively Coupled Plasma Mass Spectrometer (ICP MS ) at the University of Florida, Department of Geological Sciences following the methods of Goss et al. (2010). Calibration was performed using a combination of in house basalt standards (Endeavor Ridge MORB, ENDV, and ALV 2392 9, East Pacific Rise) and USG S (AGV 1, BIR 1 and BCR 2) standards (Kamenov et al.,
25 2007; Goss et al., 2010). To assess data accuracy and reproducibility as well as instrumental drift, standard 2392 9 was analyzed multiple times, with long term reproducibility error < 5 relative percent (1 sigma) for most elements. Trace elements with extremely low concentrations in MORB lavas have long term reproducibility error that is higher: Rb (11%), Zr (9%), Ba (10%), Pb (21%), Th (10%), U (10%). Each sample was analyzed twice wit h separate dilutions. The average concentration of the two analyses is reported here. Results Major Elements Major element concentrations of 86 89.5 W and 97.5 W lavas are presented in Table 3 1. Western GSC samples display limited ranges of total major element variation compared to eastern samples, yet the lava compositions of both regions fall well within the ranges of existing data for basalts from the GSC (PetDB database, Lehnert et al., 2000 ; Ingle et al., 2010 ) (Figure 3 1). Eastern GSC lava samples are more evolved and exhibit a wider range in MgO content, ranging from 5.59 8.15 weight percent MgO (average 7.01%, median 7.23%), whereas western GSC lavas are significantly more MgO rich, ranging from 7.89 9.22 weight percent (average 8.67%, median 8.7 2%) (Table 3 1). Eastern GSC lava compositions largely produce linear to curvilinear trends on MgO variation diagrams (increasing TiO 2 Na 2 O, K 2 O and FeO (total) ; decreasing Al 2 O 3 CaO as MgO decreases). In contrast, western GSC lavas are more variable at a given value of MgO. Lavas from 89.5 W have the highest concentrations of FeO and TiO2 (up to 13.37 wt.% and 2.16 wt.%, respectively), although these samples are not nearly as evolved as those described by Fornari et al. (1983) and Perfit et al. (1983) alo ng the Eastern Galpagos Rift/Inca
26 Transform/Ecuador Rift, which include MORB, FeTi basalts, basaltic andesites, and andesites. Major element concentrations of s amples in this study from 97.5W are consistent with the findings of Cushman et al. (2004) : her e, Na 2 O = 2.07 2.46% and TiO 2 = 1.02 1.57%, although P 2 O 5 is slightly higher, at 0.08 0.15%. Eastern samples have lower concentrations of the incompatible minor element K 2 O (~0.05 wt.%) than western samples (0.07 0.12 wt.%) in the least evolved lavas (MgO >8%). K/Ti [(K 2 O/TiO 2 )*100] values in the western GSC range from about 5.4 7.8, whereas eastern samples range from 3.9 11.7; the most depleted K/Ti values of any field area in this study were found at 86 W (Figure 3 2). Cushman et al. (2004) described the suite of N MORB lavas from the WGSC as being characterized by low concentrations of elements that are incompatible in mantle melting, such as Na 2 O (1.67 2.56 wt.%), TiO 2 (0.77 1.68%), and P 2 O 5 (0.05 0.12%). Trace Element Variations Trace element concentrat ions of 86 89.5 W and 97.5 W lavas are presented in Table 3 2. Western GSC lavas display compositional ranges falling within normal mid ocean ridge (N MORB) values, whereas EGSC lavas range from N MORB to slightly E MORB (Figures 3 3 and 3 4). On primitive mantle (Sun and McDonough, 1989) and chondrite (McDonough and Sun, 1995) normalized diagrams, incompatible trace element and rare earth element (REE) patterns display concave downward curves that flatten out in the least incompatible elements (Fi gures 3 3 and 3 4). A marked shift in the slope of EGSC samples on the normalized diagrams indicates the presence of two suites of lavas: one, more depleted, mostly consisting of lavas from 86 W; the other, more enriched in the highly incompatible elements including only lavas from 89.5W. One sample from 89.5W, sample 93 1WR, lies outside the range of relatively enriched samples from this sample
27 site, and falls within ranges similar to the more depleted lavas from 86 and 97.5W (Figures 3 3 and 3 4). We stern GSC (97.5 W) samples are more depleted than eastern GSC samples in the most incompatible elements ( e.g., Rb, Ba, Th, U) and heavy rare earth elements. S ome eastern lavas have concentrations of these highly incompatible elements that are 10 times grea ter than primitive mantle, whereas western lavas have near primitive mantle values (Figure 3 3). Following the pattern seen in the major elements, the eastern samples represent a wider range of trace element concentrations than the western samples. Negativ e Eu and Sr anomalies exist in lavas from both regions. Samples from 97.5 W are depleted in light rare earth element (LREE) concentrations relative to chondrite s as these lavas are characterized by (La/Sm) N ranging from 0.48 0.49. Lavas from 86W are simi larly depleted, with (La/Sm) N ranging from 0.49 0.58, whereas most 89.5W samples have LREE ratios that reach values near chondritic ( (La/Sm) N = 0.64 1.1 ) The only sample at 89.5W with significantly depleted LREE signatures is 93 1WR, as described above (all other samples: (La/Sm) N = 0.86 1.09)). Of the three suites of lavas, none are remarkably depleted in the most incompatible elements with respect to chondrite, but 89.5W lavas are slightly enriched. (Sm/Yb) N ranges from: 0.94 1.10 (97.5W); 0.90 0.93 (86W); 0.96 1.29 (89.5W). Incompatible trace element ratios in basalts from the distal ends of the GSC (86 W and 97.5W) exhibit some variation in comparison to 89.5 W basalts. For example, Nb/Zr values range from 0.019 0.029 in the west (97.5 W) and 0. 027 0.030 in the east (86 W) in comparison to 0.040 0.078 in eastern samples (89.5 W) (Figure 3 5). These results are consistent with previously published data for the western GSC, where samples west of 95 W exhibited Nb/Zr ratios <0.04, whereas lavas fro m 92.7 95.5W were characterized by
28 Nb/Zr values as high as 0.8 (Detrick et al., 2002). Normalized Nb/La ratios, which have been found to correlate strongly with Nd ( e.g., Kurz and Geist, 1999), are significantly higher in 89.5 W lavas (Nb PM /La PM = 0.96 1.41) than western lavas (Nb PM /La PM = 0.543 0.742) or eastern lavas (Nb PM /La PM = 0.762 0.838) (Figure 3 6). Discussion Geochemical Modeling All three suites of lavas examined in this study exhibited some degree of geochemical variab ility in both major and trace element compositions. The program PETROLOG (Danyushevsky and Plechov, 2011) was used to model the variation in major element variability of GSC lavas and to evaluate whether the observed trends can be produced by fractional cr ystallization, or whether some other igneous process(es) must be invoked. The most incompatible element depleted rock sample from the 97.5W study area (RC 29) which is also one of the most MgO rich (9.06 wt %) was chosen as the starting parental melt comp osition to model WGSC crystallization. Due to the apparent presence of two suites of lavas in the EGSC study area, as evidenced by incompatible element variations discussed above, two separate crystallization calculations were required here with parental m elt compositions for the depleted and enriched suites (3791 2 and 3792 4WR, respectively). All petrogenetic models were attempted at reasonable pressures for crustal differentiation processes: 2 kbar, 1 kbar, and 0.5 kbar; the pressure that generated the c losest fit to the observed trends is reported below. Both fractional crystallization calculations for the eastern GSC were executed at a constant pressure of 0.5 kbar under anhydrous conditions at the quartz fayalite magnetite oxygen buffer. The model requ ired 35% fractionation to replicate the liquid line of descent (LLD) in the more depleted suite, and a separate 45% fractionation to replicate the
29 observed LLD in the enriched suite (Figure 3 7) Results predict a crystallization sequence of Ol at 1195 fo llowed immediately by Ol + Cpx and Ol + Cpx + Plag at 1183C in the depleted series, whereas the enriched series followed a crystallization sequence of Plag starting at 1172C followed by Plag + Ol and Plag + Ol + Cpx at 1162C. Fractional crystallization calculations to model the WGSC were executed at a constant pressure of 2 kbar, at the quartz fayalite magnetite oxygen buffer under anhydrous conditions. Here, the best results predict a crystallization sequence of Ol as the liquidus phase at 1226C follow ed by Ol + Plag, and Ol + Plag + Cpx at 1205C. A total of 40 wt. % fractionation is required to model the closest approximation to the major element variations of the suite. However, WGSC lavas exhibit relatively scattered compositions and do not appear t o follow a single LLD, thus no single major element model was able to reproduce the results closely (Figure 3 8). This suggests there were multiple parental liquids a conclusion that is supported by the variable incompatible trace element ratios in the WG SC lavas. To test the fractional crystallization models further trace element concentrations in these suites of lavas can be predicted based on the Rayleigh Fractionation equation: C i L /C i o = F (D 1) (3 1) where C i L represents the concentration of element i (in ppm) in the liquid, C i o represents the original concentration of trace element i in the magma, F represents the fraction of melt remaining in the system, and D is the bulk distribution coefficient for the crystallizing assemblage (Rollinson, 1993). Trace element concentrations were calculated at steps of 5% fractionation using the cry stallization order and phase proportion constraints provided by major element model calculations and published distrib ution coefficients (Table 3 3).
30 Incompatible element abundances in the 86 and 89.5W (EGSC) suites of samples can be closely reproduced b y fractional crystallization of the selected parental source s in each region (Figure 3 9) providing additional support for the range of lavas having formed from crystal differentiation at shallow levels in the crust. The most incompatible elements (Rb, Ba Th, U, Nb) in both suites are most difficult to reproduce, which may be in part a consequence of the difficulty in constraining the partition coefficients of these elements. This reproducibility by fractional crystallization calculation is in agreement w ith the close fit of the major element models around the two inferred LLDs in EGSC lavas, suggesting that fractional crystallization plays a role in generating the geochemical variation in lavas t here. The signature of crystallization is superimposed upon variation s that must be attributed to melt input from the Galpagos H otspot, as evidenced by the need to invoke separate parental melt compositions o ne, incompatible element enriched near the hotspot and the other, incompatible element depleted at a greate r distance from the plume to replicate the total variation in the region. The overall trace element compositions of WGSC (97.5W) lavas cannot entirely be reproduced by fractional crystallization modeling, as the most incompatible elements are more enriche d in the samples compared to the calculations by ~50% (e.g., Nb PM = 3.23 in sample D 34A vs. Nb PM = 2.16 at 40% crystallization), yet the least incompatible elements are more depleted in the lavas compared to the calculated abundances (e.g., Yb PM = 6.75 in sample D 34A, vs. 7.64 at 40% crystallization). This is consistent with major element results, as no single LLD is apparent for the suite of 97.5W lavas. Therefore, the western end of the GSC appears to be supplied by a mantle source that is variable, at least over the distance and/or time for which our survey is representative.
31 H igher pressure c onditions must be invoked in order to generate the closest fit to the western GSC lava suite (2 kbar vs. 0.5 kbar in the east). This is significant, in that it su ggests that fractionation processes occurred at greater depths in the WGSC than in the EGSC. These inferences are consistent with the results of seismic surveys conducted on the Galpagos Spreading Center, which have successfully imaged shallow (1.0 2.5 km deep) melt lenses beneath the ridge in the regions of 91.3 92.5W and slightly deeper (2.5 4.5 km) lenses between 92.7 94.7W but not west of 94.7W (where our study area is located) (Blacic et al., 2004). These authors attribute the deepening and eventu al disappearance of the melt lens to the decreasing melt supply and thermal contribution with increasing distance from the Galpagos plume. Therefore, when fractional crystallization is taking place in the crust as far west at 97.5W, it is likely occurrin g deeper or over a greater range of depths than in locations closer to the hotspot. Petrogenesis on the GSC The K/Ti ratios of GSC lavas have been used to infer how much influence the Galpagos H otspot exerted upon a given segment of the ridge (Detrick et al., 2002; Cushman et al., 2004; Christie et al., 2005; Ingle et al., 2010). These authors have MORB compositions with little to no plume component, whereas (K/Ti)*100 > 15 indic ates MORB with significant plume source driving the incompatible K 2 O content MORB. Our results show that all sampled lavas in the regions farthest from the plume (97.5W and 86W) are well within t he N and the easternmost lavas are characterized by lower K/Ti values than the westernmost, consistent with the findings of Christie et al. (2005). No samples in our study reached E MORB K/Ti values; however,
32 nearly half of the lavas from 89.5W are within the T MORB range. This spatial distribution supports the hypothesis that the influence of the Galpagos H otspot decreases outward from the plume center, with the distal regions investigated here rece iving little to no enriched mantle contribution (Figure 3 2). Our findings are consistent with previous results for the Galpagos Spreading Center, as western (97.5W) GSC samples are more mafic than eastern (86 89.5W) GSC lavas, which are located closer to the hotspot source, and eastern samples span a much wider range of MgO content. Some geochemical variability, beyond the capabilities of fractional crystallization, is evident in both major element and trace element concentrations at the western end of the GSC (97.5W) consistent with low magmatic supply to this region (Rubin et al., 2001; Rubin and Sinton, 2007). However, eastern GSC (86 89.5W) lavas exhibit the greatest variation in source characteristics (e.g., CaO, Na2O, K/Ti, Nb/La), with the great est distribution occurring at 89.5W. This finding is inconsistent with the effects of increased melt supply alone, and must be attributed to the introduction of enriched source material or melts to the ridge from the influence of the Galpagos H otspot.
33 Table 3 1. Major element concentrations of GSC lavas (wt. %) Sample # SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O K2O P2O5 Total Galapagos Rift Exploration: 3789 2 51.1 1.25 14.1 10.8 0.19 8.03 12.0 2.17 0.05 0.10 99.79 3789 3 51.1 1.21 14.2 10.5 0.19 8.04 12.1 2.18 0.05 0.12 99.77 3790 1 51.4 1.43 13.9 11.3 0.21 7.46 11.4 2.34 0.07 0.15 99.71 3790 2 51.5 1.40 13.9 11.2 0.20 7.52 11.5 2.33 0.07 0.14 99.66 3791 1 51.1 1.17 14.2 10.5 0.20 8.11 12.2 2.13 0.05 0.09 99.75 3791 2 51.2 1.19 14.1 10.6 0.20 8.16 12.2 2.12 0.05 0.09 99.85 3792 1 51.4 1.41 13.7 11.5 0.21 7.43 11.5 2.33 0.07 0.15 99.72 3792 2 50.6 2.01 13.4 13.4 0.24 6.31 10.6 2.83 0.16 0.18 99.64 3792 4 50.5 1.92 13.7 12.5 0.21 6.63 11.1 2.77 0.15 0.19 99.75 3792 5 50.5 1.90 13.7 12.5 0.21 6.78 11.1 2.77 0.16 0.18 99.84 3793 1 50.8 1.49 13.9 12.1 0.22 7.23 11.5 2.57 0.09 0.11 100.05 3793 2 50.3 2.00 14.1 12.0 0.21 6.61 11.4 2.83 0.20 0.19 99.73 3794 1 51.6 2.16 13.4 12.5 0.22 5.63 9.93 3.56 0.24 0.32 99.51 3794 2 51.6 2.11 13.7 12.2 0.21 5.59 9.96 3.56 0.25 0.34 99.45 3795 1 51.5 2.10 13.6 12.3 0.20 5.59 9.96 3.52 0.24 0.31 99.37 AHA Nemo 2: D33A 50.4 1.06 15.4 9.15 0.17 8.87 12.6 2.07 0.07 0.10 99.87 D33B 50.8 1.06 15.2 9.17 0.17 8.89 12.5 2.09 0.06 0.09 100.01 D34A 49.6 1.40 15.8 9.80 0.19 8.26 11.9 2.46 0.10 0.12 99.65 D34B 50.3 1.45 15.5 9.85 0.19 8.30 11.5 2.44 0.10 0.14 99.86 D35A 50.5 1.12 15.4 9.32 0.18 9.22 11.8 2.15 0.06 0.10 99.89 D35B 50.4 1.12 15.5 9.39 0.17 8.97 12.2 2.17 0.07 0.10 100.06 D36A 50.9 1.20 15.2 9.40 0.19 8.64 12.0 2.23 0.08 0.11 99.97 D36B 50.4 1.22 15.5 9.37 0.18 8.82 12.0 2.19 0.08 0.12 99.83 D37A 51.0 1.05 15.1 9.20 0.17 8.94 12.4 2.07 0.08 0.11 100.12 D37B 50.8 1.36 14.9 10.0 0.20 8.42 11.7 2.37 0.10 0.15 100.00 D38A 50.6 1.35 15.1 10.1 0.18 8.40 11.8 2.35 0.08 0.12 100.02 D38B 50.9 1.24 14.9 9.60 0.19 8.93 11.6 2.24 0.09 0.13 99.81 D39A 50.8 1.23 14.9 9.64 0.16 9.06 11.7 2.28 0.09 0.12 100.00 D39B 50.7 1.04 15.1 9.12 0.19 8.95 12.6 2.10 0.07 0.10 100.06 D40A 51.6 1.17 14.4 9.70 0.19 8.04 12.6 2.22 0.08 0.11 100.07 D40B 51.5 1.16 14.5 9.34 0.18 8.21 12.6 2.25 0.09 0.13 99.96 D45C 50.6 1.23 15.5 9.45 0.17 8.86 12.2 2.32 0.08 0.10 100.40 RC28 50.5 1.02 15.5 9.13 0.16 9.07 12.3 2.11 0.07 0.08 99.97 RC29 50.6 1.02 15.6 9.10 0.17 9.06 12.0 2.08 0.08 0.08 99.78 RC30 50.1 1.38 15.3 10.1 0.20 8.68 11.0 2.41 0.11 0.12 99.40 RC31 50.9 1.42 14.8 10.2 0.18 7.89 11.8 2.36 0.08 0.12 99.80 RC32 50.3 1.48 15.1 10.1 0.18 8.47 11.4 2.42 0.10 0.15 99.76 RC33 50.5 1.08 15.6 9.29 0.18 8.96 12.0 2.14 0.08 0.09 99.88
34 Table 3 1. Continued Sample # SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O K2O P2O5 Total RC34 50.9 1.30 15.2 9.65 0.18 8.48 11.5 2.31 0.09 0.11 99.67 RC35 51.2 1.29 14.6 9.78 0.19 8.22 11.9 2.27 0.07 0.11 99.64 RC36 50.5 1.20 15.7 9.21 0.17 8.61 11.9 2.41 0.08 0.10 99.88 RC37 50.2 1.33 15.3 9.73 0.18 8.96 11.6 2.36 0.09 0.10 99.90 RC38 50.6 1.57 14.9 10.3 0.20 8.19 11.2 2.42 0.12 0.14 99.70 RC39 50.8 1.33 14.8 9.90 0.18 8.85 11.2 2.30 0.08 0.10 99.63 RC40 50.7 1.43 14.9 10.3 0.19 8.43 11.4 2.39 0.08 0.13 99.93 RC41 50.4 1.16 15.4 9.32 0.18 8.89 12.2 2.20 0.08 0.11 99.98 RC42 50.6 1.17 15.6 9.08 0.18 8.72 12.0 2.27 0.07 0.10 99.87 RC43 50.4 1.16 15.7 9.33 0.17 8.82 12.1 2.20 0.07 0.10 99.97 RC44 50.9 1.14 15.2 9.24 0.17 8.58 12.1 2.15 0.07 0.11 99.66 RC45 51.0 1.14 15.2 9.18 0.19 8.68 12.0 2.15 0.07 0.09 99.74
35 Table 3 2. Trace element concentrations of GSC lavas. Sample # 3789 3 3790 1 3791 2 3792 2 WR 3792 4 WR 3792 5 3793 1 WR 3793 2 Galapagos Rift Exploration: Li 5.3 6.3 5.4 7.8 7.2 7.5 6.1 7.1 Sc 41 41 41 44 42 43 44 40 Ti 6639 7559 6431 11461 10342 10403 8036 10553 V 298 306 298 382 348 363 330 334 Cr 243 192 239 55 73 77 72 121 Co 41 40 40 43 41 42 42 39 Ni 75 66 74 45 51 55 57 57 Cu 82 75 81 83 82 93 74 98 Zn 79 85 78 103 94 95 88 90 Ga 14 15 15 18 17 17 16 17 Rb 0.67 1.1 0.69 3.2 3.3 3.3 1.5 4.3 Sr 45 48 45 86 87 88 72 112 Y 30 39 29 44 38 39 33 37 Zr 56 88 56 114 105 106 74 114 Nb 1.5 2.6 1.6 7.1 6.7 6.9 3.0 8.9 Cs 0.009 0.013 0.009 0.033 0.031 0.032 0.018 0.040 Ba 5.6 8.3 5.6 27 27 27 13 37 La 1.9 3.0 1.9 5.7 5.7 5.7 3.0 7.1 Ce 6.2 9.6 6.3 16 15 15 8.9 18 Pr 1.1 1.7 1.1 2.4 2.3 2.3 1.5 2.6 Nd 6.4 9.2 6.4 13 12 12 8 13 Sm 2.4 3.3 2.4 4.1 3.9 3.9 2.9 4.1 Eu 0.87 1.1 0.87 1.4 1.3 1.4 1.1 1.4 Gd 3.5 4.6 3.4 5.3 5.1 5.1 4.0 5.0 Tb 0.66 0.87 0.65 0.98 0.92 0.93 0.75 0.90 Dy 4.5 5.9 4.5 6.6 6.2 6.2 5.1 5.9 Ho 0.98 1.3 1.0 1.4 1.3 1.3 1.1 1.2 Er 2.8 3.8 2.8 4.1 3.8 3.8 3.2 3.5 Tm 0.45 0.59 0.43 0.63 0.58 0.58 0.49 0.53 Yb 2.9 3.9 2.8 4.1 3.8 3.7 3.3 3.4 Lu 0.44 0.59 0.43 0.63 0.57 0.57 0.50 0.51 Hf 1.7 2.4 1.6 3.0 2.8 2.8 2.1 2.9 Ta 0.09 0.16 0.09 0.38 0.39 0.39 0.18 0.50 Pb 0.07 0.13 0.07 0.32 0.32 0.32 0.19 0.39 Th 0.08 0.15 0.08 0.34 0.36 0.35 0.15 0.47 U 0.02 0.05 0.02 0.15 0.13 0.13 0.09 0.18
36 Table 3 2. Continued. Sample # 3794 2 WR 3795 1 D 34A D 35A D 36A D 38B D 39B1 D 40A AHA Nemo2: Li 6.6 6.9 5.2 4.6 5.0 5.5 4.7 4.7 Sc 37 36 38 36 37 38 37 38 Ti 11396 11120 8186 6230 6762 7468 5804 5964 V 299 291 287 261 269 286 254 260 Cr 150 145 323 357 364 296 379 313 Co 34 35 40 41 41 39 38 38 Ni 49 50 121 156 147 120 107 85 Cu 47 70 70 72 71 67 74 81 Zn 80 84 79 74 75 79 70 69 Ga 19 18 15 14 14 15 13 14 Rb 4.8 5.2 1.1 0.66 0.88 0.96 0.72 0.78 Sr 93 94 89 66 72 70 66 80 Y 58 62 33 26 28 31 25 25 Zr 208 204 84 57 64 72 54 58 Nb 14 14 2.3 1.3 1.7 1.7 1.3 1.6 Cs 0.04 0.05 0.01 0.01 0.01 0.01 0.01 0.01 Ba 40 39 9.6 6.2 8.5 8.4 6.8 7.6 La 9.3 11 3.2 2.0 2.3 2.4 1.9 2.1 Ce 25 29 9.9 6.5 7.6 7.7 6.2 6.8 Pr 3.7 4.2 1.7 1.2 1.3 1.4 1.1 1.2 Nd 19 21 9.2 6.8 7.4 7.9 6.3 6.6 Sm 5.9 6.5 3.2 2.4 2.6 2.9 2.2 2.3 Eu 1.8 1.9 1.1 0.89 1.0 1.0 0.83 0.86 Gd 7.4 8.0 4.4 3.5 3.7 3.9 3.1 3.2 Tb 1.4 1.4 0.81 0.64 0.69 0.73 0.59 0.60 Dy 9.4 10 5.4 4.3 4.5 4.9 3.9 3.9 Ho 1.9 2.0 1.2 0.94 0.97 1.0 0.84 0.84 Er 5.7 5.9 3.3 2.7 2.8 3.0 2.4 2.4 Tm 0.88 0.91 0.51 0.42 0.44 0.46 0.37 0.37 Yb 5.9 6.0 3.3 2.7 2.8 3.0 2.5 2.4 Lu 0.90 0.91 0.51 0.42 0.43 0.47 0.37 0.36 Hf 5.3 5.1 2.4 1.7 1.9 2.0 1.6 1.6 Ta 0.76 0.74 0.15 0.09 0.11 0.11 0.09 0.10 Pb 0.33 0.35 0.25 0.13 0.16 0.18 0.11 0.16 Th 0.75 0.76 0.13 0.07 0.09 0.09 0.07 0.08 U 0.29 0.29 0.04 0.02 0.03 0.02 0.02 0.02
37 Table 3 2. Continued. Sample # RC 29 RC 30 RC 31 RC 32 RC 35 RC 36 RC 43 Li 4.6 5.6 5.8 5.9 5.4 5.1 4.7 Sc 36 37 41 37 39 37 36 Ti 5585 7698 7552 8243 6988 6706 6442 V 250 280 294 296 288 260 263 Cr 360 308 313 277 334 336 375 Co 39 40 38 40 39 39 39 Ni 142 138 98 127 116 125 140 Cu 71 74 69 66 72 72 72 Zn 69 79 78 80 75 72 73 Ga 14 15 15 15 15 14 14 Rb 0.49 1.1 0.96 0.93 0.78 0.77 0.59 Sr 62 86 74 83 72 85 68 Y 24 32 32 33 29 27 27 Zr 49 80 72 84 66 64 58 Nb 0.9 2.3 1.7 2.3 1.5 1.5 1.2 Cs 0.01 0.01 0.01 0.01 0.01 0.01 0.01 Ba 4.6 9.3 8.0 9.0 7.2 7.1 5.3 La 1.7 3.0 2.5 3.1 2.4 2.4 2.0 Ce 5.5 9.2 8.2 9.8 7.5 7.4 6.7 Pr 1.0 1.6 1.5 1.7 1.4 1.3 1.2 Nd 5.8 8.5 8.2 9.1 7.5 7.2 6.6 Sm 2.1 2.9 2.8 3.1 2.7 2.6 2.4 Eu 0.81 1.0 1.0 1.1 1.0 0.93 0.87 Gd 3.0 4.0 3.9 4.2 3.7 3.4 3.4 Tb 0.57 0.75 0.74 0.79 0.69 0.65 0.64 Dy 3.8 4.9 4.9 5.2 4.7 4.3 4.2 Ho 0.83 1.0 1.1 1.1 0.99 0.89 0.92 Er 2.4 3.0 3.1 3.2 2.9 2.6 2.7 Tm 0.36 0.46 0.47 0.50 0.44 0.40 0.41 Yb 2.5 3.0 3.2 3.2 2.9 2.6 2.7 Lu 0.37 0.46 0.47 0.49 0.44 0.38 0.41 Hf 1.5 2.1 2.0 2.2 1.8 1.7 1.6 Ta 0.07 0.14 0.11 0.14 0.10 0.10 0.08 Pb 0.10 0.23 0.18 0.24 0.16 0.17 0.12 Th 0.05 0.11 0.09 0.11 0.08 0.08 0.06 U 0.01 0.04 0.03 0.04 0.02 0.02 0.02
38 Table 3 3. Partition coefficients for fractional crystallization calculations. Element Olivine CPX Plag Rb 0.0003 0.0004 0.016 Sr 4.00E 05 0.091 1.204 Y 0.0082 0.47 0.019 Zr 0.001 0.26 0.001 Nb 5.00E 05 0.0089 0.023 Ba 1.00E 05 0.0003 0.206 La 0.0002 0.054 0.069 Ce 7.00E 05 0.086 0.065 Pr 0.0003 0.15 0.059 Nd 0.0003 0.19 0.052 Sm 0.0009 0.27 0.040 Eu 0.0005 0.43 0.332 Gd 0.0011 0.44 0.03 Tb 0 0 0.025 Dy 0.0027 0.44 0.021 Ho 0.01 0.4 0.018 Er 0.0109 0.39 0.015 Tm 0 0 0.013 Yb 0.024 0.43 0.011 Lu 0.016 0.56 0.01 Hf 0.0029 0.33 0.011 Ta 0.00012 0.013 0.033 Th 7.00E 06 0.0021 0.04 U 9.00E 06 0.001 0.019 Primary Reference Wanless et al., 2010 Wanless et al., 2010 Bedard, 2006 Secondary Reference Halliday et al., 1995 Halliday et al., 1995
39 Figure 3 1. Major element variation diagrams for GSC lavas. Existing data were acquired from the PetDB database of Lehnert, K., Y Su, C.H. Langmuir, B. Sarbas, and U. Nohl (2000). A global geochemical database structure for rocks. Geochemistry Geophysics Geosystems 1(5), 1012 1025 a nd Ingle, S., G. T.Ito, J. J. Mahoney, W. Chazey III, J. Sinton, M. Rotella, et al. (2010). Mechanism s of geochemical and geophysical variations along the western Galpagos Spreading Center. Geochemistry Geophysics Geosystems 1 1,
40 Q04003. and include all glass analyses of basalt samples from the spreading center All values in weight percent.
41 Figure 3 2. K/Ti (K 2 O/TiO 2 )*100 ratio of GSC lavas. Existing data were acquired from the PetDB database of Lehnert, K., Y Su, C.H. Langmuir, B. Sarbas, and U. Nohl (2000). A global geochemical database structure for rocks. Geochemistry Geophysics Geosystems 1(5), 1012 1025 and inclu de all basalt ic glass s amples from the spreading center.
42 Figure 3 3. Spider diagram of 23 GSC lava samples. Shaded fields represent suites of GSC lavas identified by Ingle et al. (2010): light gray (top), E MORB; medium gray (middle), T MORB; dark gray (bottom), N MORB. Sample concentrations normalized to primitive mantle accordi ng to Sun, S. .S, and W.F. McDonough (1989). Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes. Geological Society, London, Special Publications, 42, 313 345
43 Figure 3 4. Rare earth diagram of 23 GSC lava samples. Shaded fields represent suites of GSC lavas identified by Ingle et al. (2010): light gray (top), E MORB; medium gray (middle), T MORB; dark gray (bottom), N MORB. Sample concentrations normalized to chondrite values from McDonough, W. F., and S. S. Sun (1995). The composition of the earth. Chemical Geology 120(3 4), 223 253
44 Figure 3 5. Nb/Zr vs. Ce/Yb of distal GSC lavas. Elevated ratios of Nb/Zr suggest a more incompatible element enriched source for EGSC lavas relative to the WGSC.
45 Figure 3 6. Mantle normalized Nb/La vs. Longitude of distal GSC lavas. In Kurz, M. D., and D. Geist (1999). Dynamics of the Galpagos Hotspot from helium isotope geochemistry. Geochimica Et Cosmochimica Acta, 63(23/24), 4139 4156 a strong correlation was found between Nb/La and Nd, suggesting this ratio is a strong proxy for source composition.
46 Figure 3 7. Major element fractional crystallization mo del of eastern GSC lavas. Blue triangles represent observed compositions. B lack lines depict best fit PETROLOG fractional crystallization calculations at 0.5 kbar; dotted line represents calculated fractionation at 1 kbar; dashed line at 2 kbar Due to the apparent presence of two suites of lavas, a mafic end member was chosen as the parental melt composition for each suite, and two separate models were executed.
47 Figure 3 8. Major element fractional crystallization model of western GSC lavas. Red tria ngles represent observed compositions. B lack lines depict best fit PETROLOG fractional crystallization calculations at 2 kbar; dotted line represents calculated fractionation at 0.5 kbar; dashed line at 1 kbar The range of major element concentrations at a given MgO suggests lavas at 97.5 W did not follow a single LLD.
48 Figure 3 9. Trace element fractional crystallization models of EGSC lavas. Blue lines represent observed compositions; black lines represent Rayleigh fractionation calculations using a ma fic end member parental melt from each of the two suites of lavas (shown here by the change in slope), mineral proportions as given by major element model output, and a model step of 5% crystallization.
49 Figure 3 10. Trace element fractional crystallizat ion models of WGSC lavas. Red lines represent observed compositions; black lines represent Rayleigh fractionation calculations using mafic end member parental melt, mineral proportions as given by major element model output, and a model step of 5% crystall ization.
50 CHAPTER 4 LAVA MORPHOLOGY AT 97.5W ON THE GALPAGOS SPREADING CENTER M id ocean Ridge Lava Morphology Intermediate spreading rate ridges may display axial morphology and lava morphologies similar to either fast or slow spreading ridges (Macdonald 1982; Perfit and Chad wick, 1998; Sinton et al. 2002). T he western GSC, which is the slowest spreading portion of the ridge, displays an axial valley characteristic of slow spreading (Detrick et al., 2002; Sinton et al., 2003; White et al., 2008). Submarine lava flow m orphology has been shown to vary with both spreading rate (Perfit and Chadwick, 1998) and magma supply (White et al., 2008); however, on the scale of this project (approximately 40 linear kilometers along the western GSC) these parameters are expected to b e relatively constant As described above, the geochemical compositions, crystallinities, and the inferred temperatures of lavas erupted at 97.5W are also relatively uniform, and thus the viscosities of these lavas should not vary substantially ( e.g., Bon atti and Harrison, 1988; Griffiths and Fink, 1992a; Gregg and Fink, 1995). Consequently one would predict that the seafloor in the study area would be covered with pillow basalts and related talus both common at slow spreading centers ( Perfit and Chadwick 1998) Despite these assumptions, classification of the morphology of lavas based on our analysis of Argo II towed camera photographs reveals a wide range of morphologies including sheet flows and lobate flows as well as the expected talus and pillows. Therefore, it appears that some other factor(s) must play a role in determining the morphology a lava flow assumes upon eruption at the GSC. P roximity to the Galpagos H otspot may result in increased magmatic supply at the GSC, which could generate a broad er range of morphologies (Behn et al., 2004; Christie et al., 2005; White et al.,
51 2008; Colman et al., in press). One other factor that may contribute, based upon terrestrial observations and laboratory experiments ( e.g., Gregg and Fink, 2000; Gregg and Sm ith, 2003) is the slope of the terrain. Here we use bathymetric data (Figure 4 1), Argo II images, and ESRI ArcGIS software to show that the western GSC is predominantly comprised of pillow lavas an d talus, as expected for a slow to intermediate spreading center without substantial magmatic influx from the hotspot. However, because lava flows found within the ridge axis are not restricted to these two types, other short term spatial or temporal factors may have exerted more control on eruptive processes and magma supply. Methods Argo II Camera and Bathymetric Surveys Two lowerings of the Argo II towed imaging and mapping vehicle were performed during the AHA Nemo 2 cruise in the 97 W region of the GSC, covering approximately 61 nautical miles of sea f loor. A total of more than 22,000 digital images were captured by an electronic still camera (ESC) and transmitted back to the ship via fiber optic cable. Photographs cover an area of ~4 m x 5 m of seafloor and were captured at 10 20 second intervals with rare overlap between images. Layback navigation accuracy for both Argo II lowerings was ~50 100 m. Bathymetric data were collected using DSL 120 side looking sonar and Argo II optical and acoustic mapping systems. Image Analysis Approximately one out of every eight Argo II photographs (a total of 2,779) was classified for lava morphology and relative age, as determined by sediment cover. Lava morphology was determined based upon the dominant morphology within the field of
52 view (ap proximately 15 20 m 2 ). Morphological categories include: lineated sheets, ropy sheets, hackly flows, lobate flows, pillows, mixed pillows and lobes, talus, and total sediment cover/morphology not visible (Ballard and van Andel, 1977; Ballard et al. 1982; K ennish and Lutz, 1998; Kurras et al., 2000) (Figure s 4 2 and 4 3 ). Sediment cover was evaluated on a relative scale from 1 (no sediment cover; completely glassy) to 4 (total sediment cover; pockets completely filled with sediment). Slope Analysis ESRI ArcG IS 9.3 software was utilized to determine whether the slope of the terrain is correlated with the morphology of lava emplaced and preserved on the GSC. First, the database containing lava morphology photograph locations and classifications was imported int o ArcGIS and displayed as point data with colored symbols to denote morphology. Five multibeam bathymetric data files, available at the AHA Nemo 2 data archive website ( http://science.whoi.edu/ahanemo2/ ), were imported into ArcGIS as point data and converted to raster files. These five rasters were merged into a single bathymetric raster file, on which a slope analysis was performed. The output raster (Figure 4 4 ), which displays the slope of a given reg ion calculated from the difference between adjacent cell depths, was filtered to highlight regions of specific slope: greater than 15, greater than 30, and greater than 45. These selected regions were converted to polygon shapefiles and used to clip the layer containing lava morphology distribution, so that only the sample locations within the regions of desired slope were displayed (Figure 4 5 ).
53 Results Lava Morphologies and Relative Ages Pillow lavas and talus (angular, blocky pieces likely derived fro m local pillow flows) dominate the western GSC, comprising 39% and 32% of the total classified area (1080 and 890 images), respectively. An additional 15% (405) of the photographs are fields of mixed pillows and lobate flows, likely reflecting transitions from one morphology to another. The remaining areas are composed of 11% lobate flows (307 photos) and 0.3% (8 photos) ropy sheet flows (Table 4 1 Figure 4 3 ). No lineated sheets or hackly flows were confirmed anywhere in the field area. In some cases, sca rps could be seen in photographs, as evidenced by steep walls with underlying talus (Figure 4 2). At least seven major conical or rounded plateau shaped structures are apparent within the axis in this study area (Figure 4 1), as indicated by bathymetric da ta. All Argo II images were characterized by at least a moderate sediment cover with pockets filled with sediment, and many exhibited heavy cover. Of the 2,779 classified photographs, 116 (4%) were covered in a light moderate layer of sediment, with small pockets between forms beginning to fill. Twenty one percent (596) exhibited total sediment cover, with filled pockets. The remainder (2,067 photographs, or 74%) were heavily sedimented (see Table 4 2). No fresh, glassy surfaces were revealed. Due to the a bundance of sediment, lava flow contacts were often difficult to distinguish and in some cases, lava morphology was entirely concealed. Three percent (89) of the total images analyzed contain sediment cover that is sufficiently heavy to obscure the lava mo Table 4 1, Figure 4 5 ).
54 The distribution of sediment within the ridge axis does not appear to be systematic (Figure s 4 6 and 4 -7 ). One plateau just west of the center of the fi eld area is completely covered by sediment, and the last series of photographs that lead away from the ridge axis at the northwest edge of the study area is dominated by the heaviest layer of sediment. The rest of the study area reveals areas of light sedi ment cover scattered throughout regions of both moderate and heavy sediment cover, without any obvious spatial pattern. Morphology and Slope Sixty five percent (1,804) of the photographs classified were taken in locations with a slope greater than 15. Pi llows and talus dominate these regions, comprising 40% and 33%, respectively. An additional 11% were lobate, and 16% were fields of mixed lobate and pillows. Rare (0.3%) sheet flows were documented here, as were 1). Twenty four percent (668) of the total photographs are from areas with >30 slope. The major constituents of these areas are pillows (42%) and talus (34%). Fourteen percent of the photographs in regions sloping >30 were mixed lobate flows and pillows, and 10% were lobate flows. Only one sheet flow was documented in this slope range (Table 4 1). Only 7% of the 2,779 photographs were captured in locations with a slope of 45 or higher, with 50% of that portion occurring as pillow lavas, 22% as talus, and another 17% as pillows and lobate flows within one frame. No sheet flows were found in regions of steepest slope, nor were any unidentifiable fields due to heavy sediment cover (Table 4 1). Figure 4 8 shows the distribution of lava morphologies within eac h slope range.
55 Discussion In this study, we found that a variety of lava flow morphologies are generated on the Galpagos Spreading Center at 97.5W, but this region is dominated by pillow lavas and talus as predicted by tectonic and bathymetric analyses o f other slow spreading MORs (e.g., Kennish and Lutz, 1998; Perfit and Chadwick, 1998). The prevalence of pillows and talus is in contrast with the dominant morphologies found closer to the Galpagos H otspot on the western GSC, where sheet flows and lobate flows are more common (White et al., 2008; Colman et al., in press). The same pattern was found by a similar study conducted on the to the east of the Galpagos H e predominantly pillows whereas farther west, sheet flows dominate (Ballard et al., 1982). The spreading rate of the GSC varies by nearly 1.5 cm/year along its axis (DeMets et al., 2010 ); however, in our narrow study region the variation in spreading rate should be undetectable. Spreading rate is therefore not a likely source for the observed variations in lava morphology. Mid ocean ridge lavas have been found to vary little in viscosity (Perfit and Chadwick, 1998; Gregg and Fornari, 1998; Gregg and Smith, 2003), particularly when they are as chemically homogeneous as those of the western GSC, so viscosity is also an unlikely cause of the variety of morphologies here. Our results are in agreement with the field observations of Gregg and Smith (2003), as no sheet flows were found in the steepest terrain, and of the true flow morphologies, pillow and lobate flows are the most common. However, in our field area, talus contributes a significant proportion of the terrain at all slopes, not only the steepest. Thi s is likely a result of the frequent tectonic faulting (as evidenced by the numerous seismic events recorded by the Autonomous Hydrophone Array data) due to
56 spreading of the ridge, which differs significantly from an intraplate hotspot setting such as Hawa ii that is largely constructional. The ridge morphology of slow to intermediate spreading centers is controlled primarily by tectonism, in contrast to fast spreading centers, where magmatism is the primary control in shaping the ridge axis (Macdonald, 1982 ; Perfit and Chadwick, 1998; Sinton et al., 2002). The bathymetry of the western GSC clearly shows an axial valley bounded by steep walls produced by normal faults ( e.g., Sinton et al., 2003; White et al., 2008). Therefore, we presume fault generated talus is abundant within the ridge axis as spreading creates new fault scarps and fissures. Much of the observed talus is found near structures that we interpret to be fault scarps, due to the obvious change in depth visible in the darkness of the ESC images, t he nearly vertical nature of the walls, and the abundance of talus at the base of the drop offs (Figure 4 2D). The widespread talus, abundance of fault scarps, and apparent lack of fresh lava flows as inferred from the sediment cover throughout the region suggest that the seismic events detected in the years leading up to the AHA Nemo 2 cruise were related to tectonism and faulting, rather than eruptions on the seafloor as predicted. One important caveat to our evaluation of slope as a factor influencing la va morphology is that the slope determined here is by necessity calculated after lava emplacement. Therefore, the results of our survey may not speak directly to the underlying slope over which the lavas flowed, or to post emplacement tectonism, but rather to the slope of the terrain produced by those lavas. This effect may be reflected in the fact that no sheet flows were found in regions with >45 slope, suggesting that steep terrain is built by accumulation of lava flows and therefore sheet flows, by nat ure much
57 lower relief features that appear to be more fluid, are incapable of producing such steep ( e.g., Kennish and Lutz, 1998, Stakes et al., 2006; Colman et al., in press) that can be reflected in the bathymetry of a ridge. shallow slopes: more than 90% of the photos with undetermined morphology (those with too much sediment overlying the b asalt to determine morphology conclusively) were found in areas sloping less than 15. This finding has three possible implications. First, the unidentifiable morphology may speak to the accumulation of sediment upon the surface of the seafloor, and these are simply flows that are much older than their surroundings, in order to allow sufficient sediment to be deposited to completely obscure the underlying basalt. Alternatively, these unidentifiable morphologies may result where the local relief of the lava flow itself is low, so that even a fine layer of sediment will produce a smooth surface. If the latter is true, then perhaps sheet flows are more abundant in this location than initially determined, but are simply more easily masked by modest sediment cove r. If all of the photographs with unidentified morphologies classified by this study actually represent sheet flows, that brings the sheet flow population of the 97.5W GSC to 3.5%, which is still a very small proportion, and does not change the conclusion s of this study. A final consideration may be the possibility that steeply dipping slopes and high relief may inhibit the accumulation of sediment and thus leave the crust better exposed, explaining the absence of unidentifiable morphologies in these regio ns.
58 Observations based on laboratory simulations (Griffiths and Fink, 1992b) and field surveys (Gregg et al., 1996; Kennish and Lutz, 1998; Perfit and Chadwick, 1998; Fundis et al., 2010; Colman et al., in press) indicate that lava morphology varies with e ffusion rate and magma supply. Pillow lavas have been found to be representative of low effusion rates and point source volcanism, in contrast to high effusion, fissure fed eruptions that generate sheet flows In this region of the GSC, the abundance of pi llow lavas (and associated pillow talus) dwarfs the occurrence of sheet flows. The relationship between lava morphology, eruptive vent morphology and effusion rate has also been linked to the abundance of seamounts within the axis of the GSC (Ballard et al ., 1982; Detrick et al., 2002; Sinton et al., 2003; White et al., 2008; Colman et al., in press): seamounts composed primarily of pillow lavas, manifestations of point source volcanism, become more abundant with increasing distance from the Galpagos H otsp ot in either direction along the GSC, as axial morphology shifts to a broad valley. This symmetrical pattern seems to indicate that the magmatic influx from the hotspot influences the eruptive style in the central segments of the GSC by increasing the effu sion rate by virtue of increased magma supply. Additionally, shallow melt lenses have been seismically detected along the western GSC, but only as far west as 95.5W (Blacic et al., 2004); no melt lens has been detected within our field area. This is consi stent with both the axis and eruptive vent morphology, as the presence of a steady state melt lens has been found to correlate with magmatic supply The presence of conical or rounded plateau shaped structures visible in the bathymetry of the axis in this study area (Figure 4 1) indicates a propensity for point source eruptive activity at 97.5W as predicted by earlier studies (Sinton et al., 2003; White et al., 2 008; Colman et
59 al., in press). The lack of systematic distribution of sediment cover, which is assumed to represent relative age of the lava flows, also speaks to the likelihood that point source eruptions are the norm in this region of the WGSC, and occur throughout the ridge axis. A final piece of evidence for the low magmatic budget of the far western GSC includes the mafic nature of lavas from this area (Figure 3 1 ) (Rubin et al., 2001; Rubin and Sinton, 2007), as described above. Samples from 97.5W sh ow little sign of extensive differentiation, which would be expected in regions of high magma supply and long lived melt lenses. Therefore, the eruptions that produce these lavas are likely episodic and characterized by low effusion rates and low magma sup ply. Quantification of effusion rate on a local scale is beyond the scope of this project and difficult to estimate with any accuracy; however, recent evidence has shown that rates of extrusion and local magma supply can vary both spatially and temporally, even within individual eruptions (Colman et al., in press). Therefore, future work to determine effusion rates of individual lava flows in the region of 97.5W on the GSC will be vital to understanding whether this factor accounts for the variability in l ava morphologies found here beyond the regional scale.
60 Table 4 1. Distribution of lava flow morphologies in regions of increasing slope. All photos: > 15: > 30: > 45: Morphology # % Morphology # % Morphology # % Morphology # % Ropy Sheet 8 0.3 Ropy Sheet 5 0.3 Ropy Sheet 1 0.1 Ropy Sheet 0 0.0 Lobate 307 11.0 Lobate 194 10.8 Lobate 64 9.6 Lobate 19 10.3 Pillow 1080 38.9 Pillow 718 39.8 Pillow 277 41.5 Pillow 93 50.5 Talus 890 32.0 Talus 597 33.1 Talus 229 34.3 Talus 40 21.7 Mixed 405 14.6 Mixed 281 15.6 Mixed 95 14.2 Mixed 32 17.4 Unknown 89 3.2 Unknown 9 0.5 Unknown 2 0.3 Unknown 0 0.0 Total: 2779 100.0 Total: 1804 100.0 Total: 668 100.0 Total: 184 100.0
61 Table 4 2. Relative sediment cover of western GSC lavas. Relative sediment index # photographs Percent of total No sediment; vitreous 0 0.0 Light sediment; small pockets 116 4.2 Heavy sediment 2067 74.4 Total cover 596 21.4 Total: 2779 100.0
62 Figure 4 1. Location and bathymetry of the Galpagos Spreading Center 97 W Inset map produced by GeoMapApp; white box indicates location of study area. Possible point source construct sites (plateaus, small seamounts) are designated by blac k arrows.
63 Figure 4 2. Lava morphologies of the western GSC. (A) Pillow lavas; (B) Lobate flows; (C) Talus, which is often found near a scene like (D), interpreted to be a fault scarp due to the steep wall and talus at the bottom; (E) Ropy sheet flow; (F) Unknown morphology due to heavy sediment cover. Horizontal s cale of each photograph is ~3 5 m. A B C D E F
64 Figure 4 3. Lava morphology distribution at 97.5 W on the GSC.
65 Figure 4 4 Slope of terrain at 97.5 W on the GSC, as calculated from the multibeam bathymetry raster.
66 Figure 4 5 Maps showing the distribution of lava morphologies within entire field area. Each point represents one analyzed photograph. B athymetric base map from multibeam bathymetry collected by AHA Nemo 2 cruise.
67 Figure 4 6 Sediment distribution in the region of 97.5 W.
68 Figure 4 7. Sediment cover of photographed locations at 97.5 W on the GSC.
69 Figure 4 8 Maps showing the distribution of lava morphologies in the western GSC according to slope of terrain. Slope > 4 5 Slope > 30
70 CHAPTER 5 CONCLUSIONS Through careful examination of lava geochemical signatures, bathymetric data, and lava flow morphology, this study provides a detailed analysis of the petrogenesis and volcanism of the Galpagos Spreading Center in three regions: 97.5 W on the western GSC and 89.5 and 86 W on the eastern GSC. From these analyses, we find that the study location nearest the current locus of the Galpagos H otspot 89.5 W exhibits geochemical evidence of influx of plume derived mantle source material, which is recognizable in the elevated incompatible element ratios, such as K/Ti, Nb/ Zr, and Nb PM / La PM These lavas are also characterized by the greatest extents of fractionation as inferred from major element concentrations (most notably MgO) and fractional crystallization models. In contrast, lavas from the distal ends of the GSC (both 97.5 and 86 W) s how little geochemical evidence of influence from the Galpagos H otspot, as here incompatible element ratios fall in the N MORB range lavas are, on average, more mafic, and show little evidence for elevated melt volume for an intermediate spreading ridge. Further evidence to support the waning influence of the Galpagos H otspot at the distal reaches of the GSC, at least in the westward direction, was found in the morphology of the lavas in the 97.5 W study area. From our analyses of multibeam bathymetric data and towed camera photographs, we find that this portion of the western GSC is dominated by pillow lavas ( 39%) and talus ( 32%) on par with many slow to intermediate spreading centers without elevated magmatic supply due to a nearby hotspot. This pattern holds true regardless of local slope. In addition to the abundant pillows and talus, however, are many transitional regions (15%), lobate flows
71 (11%) and even sheet flows (0.3%). Local variations in effusion rate and u nderlying slope are the most likely candidates for influencing these small scale variations in lava morphology, however further investigation is required to determine exactly what role these factors play in individual eruptive sequences.
72 LIST OF REFERENCES Allmendinger, R. W., and F. Riis, (1979). The Galpagos rift at 86W 1. R egional morphological and structural analysis. Journal of Geophysical Research, 84 (B10), 5379 5389. Ballard, R. D., R. T Holcomb, and T. H. van Andel, (1979). The Galpagos R ift at 86W: 3. S heetflows, collapse pits and lava lake of the rift valley. Journal of Geophysical Research, 84 5407 5422. Ballard, R. D., T. H., van Andel, and R. T. Holcomb, (1982). The Galpagos R ift at 86W 5. V ariations in volcanism, structure, and hydrothermal activity along a 30 kilometer segment of the rift valley. Journal of Geophysical Research, 87 (B2), 1149 1161. Bdard, J. H. (2006). Trace element partitioning in plagioclase feldspar. Geochimica Et Cosmochimica Acta, 70 (14), 3717 3742. doi:10 .1016/j.gca.2006.05.003 Behn, M. D., J. M. Sinton, and R. S. Detrick, (2004). Effect of the Galpagos H otspot on seafloor volcanism along the Galpagos S preading C enter (90.9 97.6 W). Earth and Planetary Science Letters, 217 (3 4), 331 347. doi:10.1016/S00 12 821X(03)00611 3 Blacic, T. M., G. Ito J.P Canales R.S. Detrick and J. Sinton, (2004). Constructing the crust along the Galpagos S preading C enter 91.3 95.5W: Correlation of seismic layer 2A with axial magma lens and topographic characteristics. Journal of Geophysical Research, 109 B10310. Blacic, T. M., G. Ito, A.K. Shah J.P. Canales, and J. Lin (2008). Axial high topography and partial melt in the crust and mantle beneath the western Galpagos S preading C enter. Geochemistry Geophysics Geosyst ems, 9 Q12005. doi:10.1029/2008GC002100 Bonatti, E., and C.G.A. Harrison (1988). Eruption styles of basalt in oceanic spreading ridges and seamounts effect of magma temperature and viscosity. Journal of Geophysical Research Solid Earth and Planets, 93 ( B4), 2967 2980. doi:10.1029/JB093iB04p02967 Canales, J. P., G. Ito R.S. Detrick, and J. Sinton (2002). Crustal thickness along the western Galpagos S preading C enter and the compensation of the Galpagos H otspot swell. Earth and Planetary Science Letters 203 (1), 311 327. doi:10.1016/S0012 821X(02)00843 9
73 Christie, D. M., and J.M. Sinton (1986). Major element constraints on melting, differentiation and mixing of magmas from the Galpagos 95.5 W propagating rift system. Contributions to Mineralogy and Pe trology, 94 (3), 274 288. doi:10.1007/BF00371437 Christie, D. M., R. Werner, F. Hauff, K. Hoernle, and B.B. Hanan (2005). Morphological and geochemical variations along the eastern Galpagos S preading C enter. Geochemistry Geophysics Geosystems, 6 Q01006. doi:10.1029/2004GC000714 Colman, A., J.M. Sinton, S.M. White, J.T. McClinton, J. Bowles, K. Rubin, et al. (in press). Effects of variable magma supply on mid ocean ridge eruptions: Constraints from mapped lava flow fields along the Galpagos spreading cen ter. Cushman, B., J. Sinton, G. Ito, and J.E. Dixon (2004). Glass compositions, plume ridge interaction, and hydrous melting along the Galpagos S preading C enter, 90.5 degrees W to 98 degrees W. Geochemistry Geophysics Geosystems, 5 Q08E17. doi:10.1029/2 004GC000709 Danyushevsky, L. V. (2001). The effect of small amounts of H2O on crystallisation of mid ocean ridge and backarc basin magmas. Journal of Volcanology and Geothermal Research, 110 (3 4), 265 280. doi:10.1016/S0377 0273(01)00213 X Danyushevsky, L. V., and P. Plechov (2011). Petrolog3: Integrated software for modeling crystallization processes. Geochemistry Geophysics Geosystems, 12 (7), Q07021. Demets, C., R.G. Gordon, D.F. Argus, and S. Stein (1994). Effect of recent revisions to the geomagnetic reversal time scale on estimates of current plate motions. Geophysical Research Letters, 21 (20), 2191 2194. doi:10.1029/94GL02118 Demets,C., R.G. Gordon, and D.F. Argus (2010). Geologically current plate motions. Geophysical Journal International .181(1), 1 80. d oi: 10.1111/j.1365 246X.2009.04491.x Detrick, R. S., J.M. Sinton, G. Ito, J.P. Canales, M. Behn, T. Blacic, et al. (2002). Correlated geophysical, geochemical, and volcanological manifestations of plume ridge interaction along the Galpagos S preading C enter. Geochemistry Geophysics Geosystems, 3 8501. doi:10.1029/2002GC000350 Eason, D., and J. Sinton (2006). Origin of high al N MORB by fractional crystallization in the upper mantle beneath the Galpagos S preading C enter. Earth and Planetary Science Letters, 252 (3 4), 423 436. doi:10.1016/j.epsl.2006.09.048 Emmermann, R., and H.W. Hubberten (1983). Geochemistry of basalts erupted at the Galpagos S preading C enter between 85 degrees W and 87 degrees W. Initial Reports of the Deep Sea Drilling Project, 70 (APR), 409 418.
74 Fink, J. H., and R.W. Griffiths (1992). A laboratory analog study of the surface morphology of lava flows extruded from point and line sources. Journal of Volcanology and Geothermal Research, 54 (1 2), 19 32. doi:10.1016/0377 027 3(92)90112 Q Fornari, D., S. Humphris, D. Scheirer, D. Yoerger, A. Bradley, T. Shank et al. (2002). High resolution mapping and near bottom investigations of the Galpagos rift between 86W and 89.5W using ABE, A lvin and a new towed digital camera [Abstrac t]. American Geophysical Union Fall Meeting, T11C 1256. Fornari, D., et al. (2000). AHA Nemo2 shipboard data web site compiled during R/V melville NEMO expedition, leg 2. 2010, from science.whoi.edu/ahanemo2 Fornari, D. J., M.R. Perfit, A. Malahoff, and R. Embley (1983). Geochemical studies of abyssal lavas recovered by dsrv A lvin from eastern Galpagos R ift, I nca T ransform, and E cuador R ift .1. M ajor element variations in natural glasses and spa t ial distribution of lavas. Journal of Geophysical Research 88 (NB12), 519 529. doi:10.1029/JB088iB12p10519 Fundis, A. T., S.A. Soule, D.J. Fornari, and M.R. Perfit (2010). Paving the seafloor: Volcanic emplacement processes during the 2005 2006 eruptions at the fast spreading E ast P acific R ise, 9 50 N. Geoch emistry Geophysics Geosystems, 11 Q08024. doi:10.1029/2010GC003058 Goss, A. R., M.R. Perfit W.I. Ridley, K.H. Rubin, G.D. Kamenov, S.A. Soule, et al. (2010). Geochemistry of lavas from the 2005 2006 eruption at the E ast P acific R ise, 9 46 N 9 56 N: Implications for ridge crest plumbing and decadal changes in magma chamber compositions. Geochemistry Geophysics Geosystems, 11 Q05T09. doi:10.1029/2009GC002977 Gregg, T. K. P., and J.H. Fink (1995). Quantification of submarine lava flow morphology t hrough analog experiments. Geology, 23 (1), 73 76. doi:10.1130/0091 7613(1995)023<0073:QOSLFM>2.3.CO;2 Gregg, T. K. P., and J.H. Fink (2000). A laboratory investigation into the effects of slope on lava flow morphology. Journal of Volcanology and Geothermal Research, 96 145 159. Gregg, T. K. P., and D.J. Fornari (1998). Long submarine lava flows: Observations and results from numerical modeling. Journal of Geophysical Research, 103 27517 27531. Gregg, T. K. P., D.J. Fornari, M.R. Perfit, R.M. Haymon, and J.H. Fink (1996). Rapid emplacement of a mid ocean ridge lava flow on the E ast P acific R ise at 9 46' 51'N. Earth and Planetary Science Letters, 144 (3 4), E1 E7. doi:10.1016/S0012 821X(96)00179 3
75 Gregg, T. K. P., an d D. K. Smith (2003). Volcanic investigations of the P una R idge, H awai'i: Relations of lava flow morphologies and underlying slopes. Journal of Volcanology and Geothermal Research, 126 (1 2), 63 77. Griffiths, R. W., and J. H. Fink (1992a). Solidification and morphology of submarine lavas a dependence on extrusion rate. Journal of Geophysical Research Solid Earth, 97 (B13), 19729 19737. doi:10.1029/92JB01594 Griffiths, R. W., and J. H. Fink (1992b). The morphology of lava flows in planetary environments predictions from analog experiments. Journal of Geophysical Research Solid Earth, 97 (B13), 19739 19748. doi:10.1029/92JB01953 Halliday, A. N., D. Lee, S. Tommasini, G. R. Davies C. R. Paslick, J. G Fitton et al. (1995). Incompatible trace elements in OIB and MORB and source enrichment in the sub oceanic mantle. Earth and Planetary Science Letters, 133 379 395. Haymon, R. M., E. T. Baker, J. A. Resing S. M. White K. C. MacDonald and GaIAPAGoS TEAM. (2007). Hunting for hydrothermal vents along the Galpagos S preading C enter. Oceanography, 20 (4), 100 107. Ingle, S., G. T Ito J. J. Mahoney W. Chazey III, J Sinton M. Rotella, et al. (2010). Mechanisms of geochemical and geophysical variations along the western Galpagos S preading C enter. Geochemistry Geophysics Geosystems, 11 Q04003. doi:10.1029/2009GC002694 Kamenov, G. D., J.A. Saunders W.E. Hames, and D.L. Unger (2007). Mafic magmas as sources for gold in middle miocene epithermal deposits of the northern G reat B asin, U nited S tates: Evidence from a isotope compositions of native gold. Economic Geology, 102 (7), 1191 1195. doi:10.2113/gsecongeo.102.7.1191 Kennish, M. J., and R.A. Lutz (1998). Morphology and distribution o f lava flows on mid ocean ridges: A review. Earth Science Reviews, 43 (3 4), 63 90. doi:10.1016/S0012 8252(98)00006 3 Kurras, G. J., D.J. Fornari, M.H. Edwards M.R. Perfit, and M.C. Smith (2000). Volcanic morphology of the E ast P acific R ise crest 9 49 52 ': Implications for volcanic emplacement processes at fast spreading mid ocean ridges. Marine Geophysical Researches, 21 (1 2), 23 41. doi:10.1023/A:1004792202764 Kurz, M. D., and D. Geist (1999). Dynamics of the Galpagos H otspot from helium isotope ge ochemistry. Geochimica Et Cosmochimica Acta, 63 (23/24), 4139 4156. Laverne, C., and G. Vivier (1983). Petrographical and chemical study of basement basalts from the Galpagos S preading C enter, leg 70. Initial Reports of the Deep Sea Drilling Project, 70 (A PR), 375 389.
76 Lehnert, K., Y. Su, C.H. Langmuir, B. Sarbas, and U. Nohl (2000). A global geochemical database structure for rocks. Geochemistry Geophysics Geosystems, 1 (5), 1012 1025. Macdonald, K. C. (1982). Mid ocean ridges fine scale tectonic, volcanic and hydrothermal processes within the plate boundary zone. Annual Review of Earth and Planetary Sciences, 10 155 190. doi:10.1146/annurev.ea.10.050182.001103 McDonough, W. F., and S. S. Sun (1995). The composition of the earth. Chemical Geology, 120 (3 4), 223 253. Ocean Explorer Webmaster. (2010). Ocean explorer: Galpagos rift. Retrieved 06/25, 2012, from http://oceanexplorer.noaa.gov/explorations/02Galpagos/Galpagos.html Perfit, M. R., and W.W. Chadwick, Jr. (1998). Magmatism at mid ocean r idges: Constraints from volcanological and geochemical investigation. In W. R. Buck, P. T. Delaney, J. A. Karson and Y. Lagabrielle (Eds.), Faulting and magmatism at mid ocean ridges (pp. 59 116). Washington, D.C.: American Geophysical Union. Perfit, M. R ., and D.J. Fornari (1983). Geochemical studies of abyssal lavas recovered by dsrv A lvin from eastern Galpagos R ift, I nca T ransform, and E cuador R ift .2. P hase chemistry and crystallization history. Journal of Geophysical Research, 88 (NB12), 530 550. doi: 10.1029/JB088iB12p10530 Perfit, M. R., D.J. Fornari A. Malahoff, and R.W. Embley (1983). Geochemical studies of abyssal lavas recovered by dsrv A lvin from eastern Galpagos R ift, I nca T ransform, and E cuador R ift .3. T race element abundances and petrogenesis. Journal of Geophysical Research, 88 (NB12), 551 572. doi:10.1029/JB088iB12p10551 Rollinson, H. (1993). Using geochemical data: Evaluation, presentation, interpretation London: Longman Group. Rotella, M. D., J.M. Sinton, J.J. Mahoney, and W. Chazey, III (2009). Geochemical evidence for low magma supply and inactive propagation at the Galpagos 93.25 W overlapping spreading center. Geochemistry Geophysics Geosystems, 10 Q09005. doi:10.1029/2009GC002445 Rubin, K. H., and J.M. Sinton (2007). I nferences on mid ocean ridge thermal and magmatic structure from MORB compositions. Earth and Planetary Science Letters, 260 257 276.
77 Rubin, K. H., M.C. Smith, E.C. Bergmanis, M.R. Perfit J.M. Sinton, and R. Batiza, (2001). Geochemical heterogeneity wi thin mid ocean ridge lava flows: Insights into eruption, emplacement and global variations in magma generation. Earth and Planetary Science Letters, 188 349 367. Rubin, K. H., M.C. Smith, M.R. Perfit, D.R. Christie, and L.F. Sacks (1998). Geochronology a nd geochemistry of lavas from the 1996 N orth G orda R idge eruption. Deep Sea Research II, 45 2571 2597. Schilling, J. G., R.N. Anderson, and P. Vogt (1976). Rare earth, fe and ti variations along the Galpagos S preading C entre, and their relationship to the Galpagos mantle plume. Nature, 261 108 113. Schilling, J. G., D. Fontignie, J. Blichert Toft, R. Kingsley, and U. Tomza (2003). Pb hf nd sr isotope variations along the Galpagos S preading C enter (101 83 W): Constra ints on the dispersal of the Galpagos mantle plume. Geochemistry Geophysics Geosystems, 4 8512. doi:10.1029/2002GC000495 S chilling J. G., R.H. Kingsley and J.D. Devine (1982). Galpagos hot spot spreading center system .1. S patial petrological and geo chemical variations (83 W 101 W). Journal of Geophysical Research, 87 (NB7), 5593 5610. doi:10.1029/JB087iB07p05593 Schrader, E. L., and S.H. Stow (1983). Geochemistry and mineralogy of fresh and altered basalts from the Galpagos R ift. Initial Reports of the Deep Sea Drilling Project, 70 (APR), 391 408. Shank, T., D. Fornari, D. Yoerger S. Humphris, A. Bradley, S. Hammond, et al. (2003). Deep submergence synergy: Alvin and ABE explore the Galpagos R ift at 86 W. EOS, 84 (41), 425 433. Shorttle, O., J. Maclennan, and S.M. Jones (2010). Control of the symmetry of plume ridge interaction by spreading ridge geometry. Geochemistry Geophysics Geosystems, 11 Q0AC05. doi:10.1029/2009GC002986 Sinton, J., E. Bergmanis, K. Rubin, R. Batiza, R.K.P. Gregg, K. Gronvold, et al. (2002). Volcanic eruptions on mid ocean ridges: New evidence from the superfast spreading E ast P acific R ise, 17 19 S. Journal of Geophysical Research Solid Earth, 107 (B6), 2115. doi:10.1029/2000JB000090 Sinton, J., R. Detrick, J.P. Can ales, G. Ito and M. Behn (2003). Morphology and segmentation of the western Galpagos S preading C enter, 90.5 98 W: Plume ridge interaction at an intermediate spreading ridge. Geochemistry Geophysics Geosystems, 4 8515. doi:10.1029/2003GC000609
78 Sinton, J. M., and R.S. Detrick (1992). Mid ocean ridge magma chambers. Journal of Geophysical Research, 97 197 216. Smith, M. C., M.R. Perfit, D.J. Fornari, W.I. Ridley M.E. Edwards, G. Kurras et al. (2001). Magmatic processes and segmentation at a fast spreading mid ocean ridge: Detailed investigation of an axial discontinuity on the E ast P acific R ise at 937'N. Geochemistry Geophysics Geosystems, 2 (10), 1040. Stakes, D. S., M.R. Perfit, M.A. Tivey D.W. Caress, T.M. Ramirez, and N. Maher (2006). The cleft revealed: Geologic, magnetic, and morphologic evidence for construction of upper oceanic crust along the southern J uan de F uca R idge. Geochemistry Geophysics Geosystems, 7 Q04003. doi:10.10 29/2005GC001038 Sun, S. S and W.F. McDonough (1989). Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes. Geological Society, London, Special Publications, 42 313 345. van Andel, T. H., and R.D. Ball ard (1979). The Galpagos R ift at 86W: 2. V olcanism, structure, and evolution of the rift valley. Journal of Geophysical Research, 84 (B10), 5390 5406. Wanless, V. D., M.R. Perfit, W.I. Ridley, and E. Klein (2010). Dacite petrogenesis on mid ocean ridges: Evidence for oceanic crustal melting and assimilation. Journal of Petrology, 51 (12), 2377 2410. doi:10.1093/petrology/egq056 White, S. M., J.D. Meyer, R.M. Haymon, K.C. Macdonald, E.T. Baker, and J.A. Resing, (2008). High resolution surveys along the hot spot affected Galpagos S preading C enter: 2. I nfluence of magma supply on volcanic morphology. Geochemistry Geophysics Geosystems, 9 Q09004. doi:10.1029/2008GC002036 Wilson, D. S., and R.N. Hey (1995). History of rift propagation an d magnetization intensity for the C ocos N azca S preading C enter. Journal of Geophysical Research Solid Earth, 100 (B6), 10041 10056. doi:10.1029/95JB00762
79 BIOGRAPHICAL SKETCH Katrina A lyce Garman was born in 1987 in H arvard, Illinois She grew up i n Poplar Grove, IL and graduated from North Boone High School in 2006. She earn ed her B achelor of A rts in geology with a research focus in volcanology and igneous geochemistry from Colgate University (Hamilton, N ew Y ork ) in 2010. She then spent three months as an intern with the U nited S tates Geological Survey at the Cascades Volcano Observatory in Vancouver, W ashington before moving to Gainesville, F lorida to b egin her graduate work in the University of Florida Department of Geo logical Sciences. Kat ie received her Master of Science degree from UF in the summer of 2012.