Key words: Oceanic crust, spreading ridge, backarc, segmentation, faulting, volcanism. Eulàlia Gràcia* and Javier Escartín ** Abstract. - PDF

CONTRIBUTIONS to SCIENCE, 1 (2): (1999) Institut d Estudis Catalans, Barcelona Crustal accretion at mid-ocean ridges and backarc spreading centers: Insights from the Mid-Atlantic Ridge, the Bransfield

Please download to get full document.

View again

of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

Study Guides, Notes, & Quizzes

Publish on:

Views: 62 | Pages: 13

Extension: PDF | Download: 0

CONTRIBUTIONS to SCIENCE, 1 (2): (1999) Institut d Estudis Catalans, Barcelona Crustal accretion at mid-ocean ridges and backarc spreading centers: Insights from the Mid-Atlantic Ridge, the Bransfield Basin and the North Fiji Basin Eulàlia Gràcia* and Javier Escartín ** Institut de Ciències de la Terra Jaume Almera. Consell Superior d Investigacions Científiques (CSIC), Barcelona Abstract Mid-Ocean Ridges are a natural laboratory for the study of magmatic and tectonic processes and their interactions. Owing to their relatively simple structure and geodynamic history, they have been studied by highly active projects, such as RIDGE (USA), followed by the international initiative InterRidge, which have yielded enormous advances during the last two decades. By these ridge-oriented initiatives, a large portion of the global ridge system has been explored, mapped and sampled. These data have constrained models of melt production and dynamics of the mantle, formation of the lithosphere, magmatic accretion of the crust, tectonic deformation, and volcanic processes at shallow levels, as well as their interactions. Geophysical observations (bathymetry, acoustic backscatter, and gravity) from several sites along mid-ocean ridges (Mid-Atlantic Ridge) and two back-arc basins (Bransfield and North Fiji) summarized here provide constraints on the surface expression of ridge tectonism, volcanism, and the density structure of the oceanic lithosphere in depth, and their bearing on the accretionary processes at ridges. Resum Les dorsals mèdio-oceàniques són laboratoris naturals per a l estudi dels processos volcànics i tectònics, i les seves interaccions. La estructura i història geodinàmica relativament senzilla de les dorsals oceàniques ha fet que aquéstes siguin el focus d actius projectes pluridisciplinars, tals com RIDGE (EUA), i InterRidge. Fonamentats en aquestes iniciatives en els darrers vint anys s han produit avanços significatius, en l exploració, cartografia i mostreig d una elevada porció del sistema global de dorsals oceàniques. Aquestes dades han servit per a constrényer models de dinàmica del mantell i el seu percentage de fusió, de formació de la litosfera, d acreció magmàtica, de deformació tectònica i dels processos volcànics en nivells superficials. Aquest és un article de síntesi en el qual agrupem observacions i interpretacions basades en la nostra pròpia experiència en diverses localitats al llarg de dorsals medio-oceàniques (Dorsal Mèdio-Atlàntica) i conques de rerearc (Bransfield i Nord- Fiji). Les eines utilitzades estan basades en els mètodes de geofísica marina comunment utilitzats en l estudi de dorsals, com són la batimetría multifeix, la retrodifusió acústica, i la gravimetría. Els resultats d aquests estudis ens proporcionen informació sobre la expressió superficial de la tectònica, volcanisme, així com de l estructura en profunditat de la litosfera oceànica i els processos que hi tenen lloc. Key words: Oceanic crust, spreading ridge, backarc, segmentation, faulting, volcanism * Author for correspondence: Eulàlia Gràcia, Institut de Ciències de la Terra Jaume Almera (CSIC), Departament de Geofísica. Lluís Solé i Sabarís s/n Barcelona, Catalonia (Spain). Tel , Fax: ** Now at: Laboratoire de Pétrologie, CNRS. Paris The oceanic crust, that covers more than 60% of the total Earth s surface, is the result of magmatic accretion which takes place along the global Mid-Ocean Ridge (MOR) system (Fig. 1) [1]. MORs, with a total length km, are the locus of the most active and voluminous magmatic activity on the Earth, with a total production of more than 20 km 3 of oceanic crust every year, and contributes to two thirds of the total internal heat lost by the Earth. This magmatism directly results from the passive upwelling of the mantle and decompression melting as plates separate along the ridge axis. The ridge system is also one of the most tectonically active areas on Earth. Plate separation is taken up primarily by magmatic accretion (formation of oceanic crust), but also by tectonic extension of the lithosphere near the MOR, which produces an ubiquitous shallow seismicity belt along ridges, modifies the structure of the crust and morphology of the 176 Eulàlia Gràcia and Javier Escartín seafloor. The morphology of the oceanic seafloor and structure and composition of the crust are thus a result of the interplay between magmatism, tectonism and hydrothermalism, and their study can provide constraints on the different processes occurring at MORs. Here, we present data from two geodynamic settings with continuous generation of oceanic crust: mid-ocean ridges and backarc basins (Fig. 1). Regarding the first setting, we have selected three portions of the Mid-Atlantic Ridge (MAR), which show strong differences in shallow and deep structure according to the magmatic supply and the distance from hotspots. Regarding the second, two contrasting backarc basins have been selected, the Bransfield Basin (BB), with continental rifting and incipient accretion of oceanic crust, and the mature North Fiji Basin (NFB), where seafloor spreading processes are similar to those observed at MORs (Fig. 1). Overview of Oceanic Crust Accretion The relevance of the oceanic crust to understand plate tectonics and mantle convective processes, which was shown in the 1960 s, raised intense research on mid-ocean ridges. The first surveys showed that MORs are a linear chain elevated 2-3 km above surrounding, older ocean basins as a result of thermal uplift due to mantle convection. New, more sophisticated surveying techniques, have revealed that the ridge is not continuous but dissected by a series of fracture zones [1,2]. With new swath-bathymetric systems, other ridge discontinuities at smaller scales [3] have been identified (e.g., overlapping spreading centers (OSCs) [4], nontransform offsets (NTOs) [5], and propagators [6]). These studies have also shown that fast and slowspreading ridges display notable differences in segmentation, seafloor morphology, crustal structure, lithospheric composition, and nature of magmatic and tectonic processes. Spreading rate seems to be an important parameter controlling mantle upwelling, melt production and lithospheric cooling under ridge axis: Ridge segmentation. One of the first observations revealed the presence of ridge segments, typically 40 km long (between 10 and 100 km) at the slow-spreading MAR, with transform discontinuities and NTOs displacing the axis laterally by 10 km and several hundred kilometers [e.g., 3]. In contrast, fast-spreading ridges show long, straight ridge sections ( 100 km) with small lateral offsets (OSCs and propagators) or large offset fracture zones, and rotating microplates [e.g., 7]. Seafloor morphology. Slow-spreading ridges are characterized by a prominent rift valley (Fig. 2a), 1 km high typically, bounded by inward-facing normal faults [8]. The result of spreading and tectonic extension is the formation of abyssal hills sub-parallel to the ridge axis, with vertical reliefs ranging from several hundreds of meters and 2 km. In contrast, fast-spreading ridges are characterized by an axial high with a central summit caldera 50 m in vertical relief (Fig. 2a). Faulting seems to be less prominent, with abyssal hills of lesser amplitude. Figure 1. Seafloor age map [1] showing the location of the areas described in the text. Stars correspond to the study areas: CSR, Central Spreading Ridge; SPR, South Pandora Ridge; CBB, Central Bransfield Basin. Names of tectonic plates: PAC, Pacific; NAZ, Nazca; C, Cocos; SAM, South America; NAM, North America; EUR, Eurasia; AFR, Africa, AUS, Australia; ANT, Antarctica. Crustal accretion at mid-ocean ridges and backarc spreading centers: Insights from the Mid-Atlantic Ridge, the Bransfield Basin and the North Fiji Basin 177 Crustal structure and composition. Slow-spreading ridges show high lateral variability in crustal composition and thickness. The crust is systematically thicker at the center of segments and thins towards the ends [e.g., 9-11] (Fig. 2b). Peridotites and gabbros are commonly dredged near the end of segments, suggesting that deep lithospheric levels are exposed on the seafloor by tectonic extension [12-14] and/or that the crust is not fully magmatic but highly heterogeneous and composed of gabbros and peridotites [15,16]. In contrast, the crustal thickness at fast-spreading ridges does not seem to vary substantially for hundreds of kilometers [e.g., 17,18] (Fig. 2b), and the scarcity of peridotite outcrops together with gravity and seismic data suggests that the crust is fully igneous and relatively homogeneous, with layers of gabbro, diabase and basalt (from bottom to top) overlying the mantle [19]. Magmatic and tectonic processes. The differences described above are caused by the effect of spreading rate on a) the thermal structure of the lithosphere and therefore on the mechanical properties and extensional tectonics at the ridge, and b) on the style of mantle upwelling and therefore, melt supply to the ridge axis. High melt supply to the ridge axis at fast-spreading ridges is indicated by the presence of a steady-state magma chamber [20,21] and by the evidence of abundant off-axis volcanism [e.g., 22], and volcanic eruptions [e.g., 23]. There is also a marked change in the style of magmatic accretion, from a laminar mode at fastspreading ridges to a segmented, plume-like mode at slowspreading ridges [24] (Fig. 2b). This may arise from buoyant segmentation of mantle upwelling at slow-spreading ridges [e.g., 9,24,25,26]. Alternatively, segmentation may occur at shallow levels in the melt extraction process [25,27]. Differences in the thermal regime also explain the transition observed from an axial high at fast spreading ridges to an axial valley at slow spreading ridges [28,29,30] (Fig. 2a). Data and Methods Geophysical methods and geological observations are essential tools to study mid-ocean ridges and backarc spreading centers. We can consider three main working scales for the most usual methods: Figure 2. The morphology of the ridge crest (a) and the mode of mantle and/or melt flow and crustal accretion below the mid-ocean ridge (b) depend largely on spreading rate. a) Fast-spreading ridges (full-spreading rate 60 mm/yr) typically show an axial high ( 500 m) with a small summit caldera ( 1 km wide). In contrast, slow-spreading ridges (full-spreading rate 30 mm/yr) show a marked rift valley (typically 1 km deep or more), while intermediate ridges have transitional morphology. Modified from [5]. b) Slowspreading ridges are characterized by transform discontinuities and NTOs that are segmented along the ridge axis at scales of km. Segments show systematic variations in crustal thickness along their length as a result of focused magmatic accretion in depth (mantle convection and/or melt focusing). Fast-spreading ridges, in contrast, are characterized by a relatively constant crustal thickness and an inferred two-dimensional passive mantle upwelling and melt supply to the ridge axis. Modified from [24]. 1) Global. Satellite altimetry data provide the gravity field over the oceans, including large areas of uncharted seafloor, such as southern oceans and Antarctic margins [31]. 2) Regional. Shipboard data (swath-bathymetry/acoustic backscatter, magnetics, gravity, seismics, electromagnetic) provide high-resolution ( m) coverage over regions at 100-km scales. These data allow the identification of detailed volcanic and tectonic features of the seafloor, as well as deep structure (i.e., velocity, density) of the oceanic lithosphere. 3) Local. Remotely operated vehicles (ROVs), and autonomous underwater vehicles (AUVs), such as manned submersibles provide in situ data ( 1-10 m resolution) for studying magmatic and tectonic features at local scales (1-10 km). A summary of the main methods and systems used in mid-ocean ridge surveys is presented on Table 1. In this paper, we focus on the seafloor structures and on the shallow density structure of the oceanic lithosphere at ridges, characterized from swath-bathymetry acoustic backscatter data and from gravity data, respectively. Satellite gravity data provide a regional perspective of the tectonic structures and setting for each of the study areas: the slow-spreading MAR (South of the Azores, South of Atlantis FZ, and Fifteen-Twenty, Table 2), and three backarc areas (South Pandora Ridge and Central Spreading Ridge in the North Fiji Basin, and Central Bransfield Basin, Table 2). 178 Eulàlia Gràcia and Javier Escartín Table 1. Methods commonly used in mid-ocean ridge surveys, especially the ones followed here. Method Measurement Data output Main results on M.O.R. Systems commonly used accuracy in M.O.R. surveys satellite altimetry 1 km sea-surface height global scale gravity GEOSAT/GM, ERS-1/GM, and bathymetry SEASAT, TOPEX shipborne swath-bathymetry m water-depth, morphostructure, Simrad EM1000 and EM12 (Norway) acoustic backscatter seafloor facies Seabeam 2000 (USA), Thomsom (Fr) Furuno HS10 (Jap.), Hydrosweep (USA) side-scan sonar m acoustic backscatter seafloor facies, GLORIA (UK) tectonic/magmatic mapping HMR-1 (USA) gravity 1 km gravity field / anomalies (FAA, MBA) relative crustal thickness Bodenseewerk KSS330 magnetics magnetic field / anomalies spreading rate, EGG-Geom. G866, BGM3 crustal magnetization Barringer M244 proton magnetometre near-bottom deep tow 1-10 m bathymetry, backscatter, high-res. morphostructure, TOBI (UK), SAR (Fr), DSL-120 (USA) magnetics, video, photo detailed geophysical surveying ROVs* m bathymetry, backscatter, magnetics, seafloor monitoring, sampling, Jason (USA), Ropos (Can.), Victor (Fr.) video, photo, rock and water sampling in situ geophysics AUVs** / m bathymetry, backscatter, magnetics, direct observation, Alvin, ABE (USA), Cyana, Nautile (Fr), submersibles video, photo, rock and water sampling geological mapping / sampling Shinkai 2000, Shinkai 6500 (Japan), hydrothermal activ. exploration MIR-I, MIR-II (Rus.) *ROV: Remotely Operated Vehicle **AUV: Autonomous Underwater Vehicle, such as manned submersibles, which allow scientists to go up to m deep. The Slow-Spreading Mid-Atlantic Ridge The MAR has been extensively studied during the last 25 years, from the Equatorial MAR to the Iceland hotspot. We have chosen three sites along the MAR: a hotspot-influenced magmatic ridge section (South Azores), a «normal» ridge section away from hotspots (South Atlantis FZ) and a «cold» section (Fifteen-Twenty) were spreading is mostly amagmatic (Figs. 1, 3a). The spreading rate in the three areas is very similar (24-26 mm/yr full rate, Table 2), and variations in morphology, tectonic structure and crustal composition maybe directly linked to magma supply. MAR South of the Azores, 38 N N The MAR south of the Azores is especially interesting for the study of ridge-hotspot interactions (Figs. 1, 3a). First, it is close to the Azores hotspot and so affected by the associated thermal and chemical flux [32]. Second, at a smaller scale, the ridge comprises several varying ridge segments linked by systematic left-lateral NTOs [33,34]. The regional pattern of MAR between 38 N and N has been outlined by [34]. Along this ±500 km long section of the ridge, with a general trend of 050, and average full spreading rate of 22 mm/yr, there is a relatively constant along-axis topographic gradient from the Azores platform towards the south [34,35] (Fig. 3b). In a regional sense, the MAR south of the Azores is segmented by two large discontinuities, the Pico and Oceanographer Fracture Zones, which include seven second-order segments separated by large and wide NTOs. The northernmost segments are strongly influenced by the Azores hotspot, and the rift valley, which typifies most slowspreading ridges, disappears. All the ridge segments are associated with Mantle Bouguer anomaly (MBA) gravity lows, but show less along-axis variation than in other sections of the MAR [34,35]. In order to determine the interaction and balance between extrusive volcanic activity and extensional tectonics, the segments and NTOs of the MAR south of Azores (38 N to N) were imaged using the Towed Ocean Bottom Instrument (TOBI) [36] (Table 1). Volcanic-tectonic terrains present various degrees of tectonic dismemberment and deformation, a range of sedimentary cover and thickness, and relationships between lava flows, providing us with a relative age between them. In many cases, groundtruthing of these terrains was carried out with a combination of submersible observations, deep-tow camera traverses and dredge sampling [37-39]. The three main geological categories (volcanic, tectonic and sedimentary) of the TOBI seafloor textures [40] are shown in Figure 4a. The segment presented in this work, named Lucky Strike, is characterized by rift valley morphology, showing a fairly constant width of km throughout its 60 km of length (Fig. 3b). The dominant bathymetric feature is the shallow platform area, which occupies 50 km 2 of the central axial floor between 2100 and 1800 m water depth. Its gently sloping upper surface is topped with four small volcanoes/volcanic ridges [39]. On the TOBI sidescan sonar image of this Crustal accretion at mid-ocean ridges and backarc spreading centers: Insights from the Mid-Atlantic Ridge, the Bransfield Basin and the North Fiji Basin 179 Figure 3. a) Location map of the three selected areas along the Mid-Atlantic Ridge: b) South of Azores, extending from the Azores to the Oceanographer FZ, c) 29 N, extending from Atlantis FZ to Kane FZ and d) around the N FZ. Data in b-d) are based on satellite gravity map [31]. Location of Figs. 4b, 4c and 5 are also depicted. segment (Fig. 4b), the central volcano cluster, characterized by volcanic breccia and pillow constructs, is cut by several fault/fissure lineaments which continue across the platform [37,38]. The platform surface, characterized by a mottled/patchy backscatter pattern, is cut through almost everywhere by a fabric of closely spaced faults and fissures, with less than 25 m vertical throw [40]. To the north and south of the platform, TOBI data show well-defined linear clusters of constructional volcanic ridges consisting of numerous pillow mounds [38]. This mounded terrain is locally cut by dense arrays of faults, but the style of faulting varies systematically when the distance from the segment center increases (Fig. 4b). Away from the centre, the number of faults decreases, show a larger spacement (more than 250 m), and have greater individual throws (between 50 and 75 m) [40]. Faulting in the northern domain follows a similar pattern, but is not as widespread. Both domains contain small lava flows, either pervasively faulted or completely undeformed. In both areas, overlapping flow margins provide unequivocal evidence for complex, multiple extrusive events. MAR South of Atlantis Fracture Zone, 29 N The Broken Spur segment is ~80 km long, limited by the N and N NTOs (Fig. 3a,c), and shows the characteristics of a «typical» slow-spreading segment. Traces of the NTOs bounding the segment to the north and south [41], demonstrate that the offsets have migrated along the axis. Several geophysical studies have described its morphology [33,42-44], gravity and crustal structure [9,41], and tectonic evolution [45]. The crust at the segment center is thicker than at the end [9]. In addition, there is an across-axis asymmetry at the segment end, with thicker crust at the outside corner (OC) than at the inside corner (IC) of ridge-offset intersections [9,41], reflecting the combined effect of focused magmatic accretion on axis [e.g., 9,46] and tectonic thinning at ICs [47]. The IC terrain is characterized by larger faults ( 1 km in throw) than the OC or segment center (SC) ( 500 m) [43,48], and is associated with diffuse microseismic activity [49].
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks