T32B-01
New Insights into the Transition From Magmatic to Tectonic Rifting
Magma plays a major role in the development of many rifts and continental margins. This is particularly clear for some of the more recent continental rifts including the Afro-Arabian Rift System and the breakup of South America from Africa. We are interested in how magma, injected as dikes, may lead to weakening of the lithosphere so that rifting can proceed even if the supply of magma wanes. We use a hybrid numerical model to simulate the effect of dike injection on continental lithopsheric rifting. We have developed a numerical diking simulation where the key diking parameters controlling the input of magma are the magma chamber size, minimum diking interval, and maximum tectonic force. The model includes a 2D finite difference code (FLAC) for tracking long- term stress build-up and strain in a viscoelastic-plactic model lithosphere. A boundary element code is used to simulate the effect of short-duration dike intrusion events that are specified to occur periodically at the center of the model region. The stresses from the finite difference code are applied to the boundary element code to calculate how much a dike opens as a function of depth. If a dike is generated, basaltic-density magma is "injected" into the finite difference model based on the distribution of dike opening obtained from the boundary element code. Diking thermally weakens the lithosphere and changes the lithospheric density structure, both weakening the lithosphere and reduce the force difference needed to continue extension. Varying the diking interval and magma chamber size, changes the rates magma input and lithospheric weakening. The maximum tectonic force effects the rate of magma injection, total magmatic extension, and hence, the timing of the transition from magmatic to tectonic extension. With normal lithospheric thicknesses and thermal structure, this transition may require as little as 3-5 km of magmatic extension before the onset of tectonic rifting.
T32B-02 INVITED
Initiation and Evolution of Magmatic Continental Rifts
Many rifts and rifted margins are associated with large igneous provinces (LIPs), but the igneous province, where there are thick layers of extruded basalt, accounts for a small fraction of the length of the rift. Several factors argue for the importance of magmatic dike propagation from the center of a LIP in the formation of many, if not most, rifted continental margins. Magmatic rifts and margins are remarkably straight. Tectonic rifts may not be straight since the faults that accommodate tectonic stretching can follow regions of pre-existing weakness. Dikes are generally very straight because they, unlike faults, need a source of magma and a connection to that magma source. The open part of the dike is the conduit connecting the source area to the tip of the dike. If the magma source is localized, then dikes can only remain connected to the source if they are straight. Dike opening is a 3D process, since stress, temperature and density variations in all directions can affect dike opening. However, here we consider some simple models of diking, and do not treat lateral propagation of dikes. First, 1D models of the stress needed for fault slip show that the lithospheric force (the vertically integrated stress difference through the lithosphere) needed for tectonic rifting of normal continental lithosphere may be much larger than likely levels of force. The minimum force to drive opening of lithosphere-cutting dikes increases with the square of the mantle lithospheric thickness, and for normal continental lithosphere can be an order of magnitude lower than the force for tectonic extension. To look at the amount of magmatic dike intrusion needed to weaken continental lithosphere and so initiate a successful rift we have developed two-dimensional cross-sectional models of rifts. The new feature of the models is that the dikes open in response to stresses built up in the lithosphere by plate separation. We report the results of numerical experiments in which we vary the applied extensional tectonic force, the initial thermal structure and the rate at which magma is supplied in dikes. The heating caused by as little as a few kilometers of dike opening can lower the tectonic strength of lithosphere by a factor of ten. This may be enough weakening to allow continued rifting even if the magma supply shuts off. We show that magma does not have to reach the surface to weaken lithosphere and that the intrusion of magma into a rift reduces the amount of syn-rift 'tectonic subsidence' and increases the amount of post-rift 'thermal subsidence' consistent with observations.
T32B-03
Crustal Thickness and Continental Lithosphere Thinning Factors for the Woodlark Basin From Gravity Inversions
The Woodlark basin, a young oceanic basin currently propagating westward into the continental Papuan Peninsula, provides an excellent natural laboratory to study continental breakup and the ocean-continent transition. Gravity inversion has been used to determine Moho depth, crustal thickness and continental lithosphere thinning factors. In this study we produce crustal thickness and continental lithosphere thinning factor maps derived from gravity inversions incorporating a lithosphere thermal gravity anomaly correction and different breakup ages to account for the breakup history of different parts of the basin. Moho depths from gravity inversion have been compared with crustal thickness estimates gained from velocity inversions (Zelt et al. 2001). In the Eastern Woodlark the oceanic crustal thickness is predominantly less than 7km thick whereas in the Western Woodlark, to the west of the Moresby Transform, the oceanic crustal thickness is greater than 7km thick. In several locations on the margins of the Eastern Woodlark Basin, the continental crust thins to less than that of the oceanic crust which forms immediately beside it, suggesting that the Eastern Woodlark ocean basin is predominantly non-volcanic; the lack of volcanoes on these margins provides further evidence for this. Crustal thicknesses predicted from gravity inversion immediately to the east of the Moresby Seamount are greater than expected for 'normal' oceanic crust and the bathymetry is shallower than the rift basins north and south of Moresby Seamount. This region may be underlain by thicker oceanic crust due to a more volcanic breakup because of secondary mantle convection (Martinez et al. 1999) or more fertile mantle, or may include remnant thinned continental crust beneath thick volcanics.
T32B-04
The Main Ethiopian Rift: a Narrow Rift in a Hot Craton?
The Main Ethiopian Rift (MER) is a classic example of a narrow rift, but a synthesis of our results from the EAGLE (Ethiopia-Afar Geoscientific Lithospheric Experiment Phase I broadband experiment) and from the EBSE experiment (Ethiopia Broadband Seismic Experiment) suggests the MER formed in thin, hot, weak continental lithosphere, in strong contrast with predictions of the Buck model of modes of continental lithospheric extension. Our joint inversion of receiver functions and Rayleigh-wave group velocities yields shear-wave velocities of the lowermost crust and uppermost mantle across the MER and the Ethiopian Plateau that are significantly lower than the equivalent velocities in the Eastern and Western branches of the East African Rift System. The very low shear-wave velocities, high electrical conductivity in the lower-crust, and high shear-wave splitting delay times beneath a very broad region of the MER and the Ethiopian Plateau indicate that the lower-crust is hot and likely contains partial melt. Our S-receiver function data demonstrate shallowing of the lithosphere-asthenosphere boundary from 90 km beneath the northwestern Ethiopian Plateau to 60 km beneath the MER. Although we lack good spatial resolution on the lithosphere-asthenosphere boundary, the region of thinned lithosphere may be intermediate in width between the narrow surface rift (< 100 km) and the broader zone of strain in the lower crust (~ 300 km). The MER developed as a narrow rift at the surface, localized along the Neoproterozoic suture that joined East and West Gondwana. However, a far broader of lower crust and uppermost mantle remains thermally weakened since the Oligocene formation of the flood basalts by the Afar plume head. If the lithosphere- asthenosphere boundary is indeed a strain marker then lithospheric mantle deformation is localized beneath the surface rift. The development of both the Eastern/Western branches of the East African Rift System to the south and of the MER in the north as narrow rifts, despite vastly different lithospheric strength profiles, indicates that inherited structure, rather than rheological stratification, is the primary control on the mode of extension in these continental rifts.
T32B-05 INVITED
Magmatic Rifting of Pangaea Linked to Onset of South American Plate Motion
Finite element models indicate that the break-up of Pangaea was driven by the motion of the continental cratons and not by the impact of a plume. Several finite element model scenarios related to the break-up were constructed including a large plume impact at the North America-South America-Africa triple junction and various possible continental motions. As static elastic finite element models each result shows the instantaneous stress field resulting from the applied forces (plume, continental motion etc). To determine a plausible sequence of events, each snapshot of the stresses generated by the applied forces was compared with evidence of the break-up preserved along the southeastern margin of North America in the form of dikes and faults. The faults indicate a NW trending extensional stress at 230 Ma followed by a rapidly changing stress field associated with dike intrusion around 198 Ma. Dikes from 198 Ma along the southeastern margin of North America indicate that the stress field in this region rotated from NE extensional to NW extensional within 1-2 million years. The finite model result sequence that best represents this changing stress field begins with North America, South America, and Africa sutured to each other and Africa fixed in space relative to the other continents. A NW trending extensional stress field, which would have resulted in the 230 Ma NE trending normal faults, is best created by the motion of North America to the northwest. If North America continues its motion to the northwest (as would be most tectonically plausible) then the 198 Ma northwest trending dike injection and the onset of magmatic rifting is best explained by the onset of South America's southwestward motion. However, for the stress field to rotate 90 degrees within a short 2 million year time span, South America must quickly become separated from North America by a weak area such that its motion no longer affects the stress field in North America. The onset of magmatic rifting along the eastern margin of North America combined with the probable magmatic activity along the southern gulf coast of North America suggest that the onset of motion by South America triggered magmatic rifting which rapidly weakened the crust between North and South America.
T32B-06
Lithospheric Control on the Initiation of the Columbia River Basalts and Yellowstone Hotspot: Role of the Cretaceous Western Idaho Shear Zone
The plate boundary conditions of the western United States are well known from reconstructions based on oceanic seafloor spreading patterns, making the area an ideal location to address the association of magmatism and continental breakup. A major plate rearrangement occurred at 18 Ma on the west coast of the US, including the cessation of Monterey microplate spreading center, initiation of the San Andreas fault system, and the Pacific plate's capture of western California. Immediately after this time, magmatism associated with the Columbia River flood basalts (17-14 Ma) and the cryptic initiation of the Yellowstone hotspot (~18 Ma) occur in the Idaho-Oregon-Nevada region. In this area, the sharp western margin of North America is defined by the western Idaho shear zone. The crustal portion of this shear zone is offset ~120 km eastward from the mantle portion of the same shear zone due to Late Cretaceous- Early Tertiary Sevier thrust faulting. The feeder dikes of the Columbia River flood basalts follow the mantle portion of this margin. We infer that the Columbia River flood basalts follow the strong fabric of the mantle section of the western Idaho shear zone, which explains the northward extent of the basalt flows. The Yellowstone hotspot correlates with the southernmost known extent of the mantle portion of the western Idaho shear zone. These data suggests chronic reactivation of lithospheric-scale features (lithospheric scars), due to transtensional kinematics, is the major source for both the production of flood basalts and hotspots: The system works in a top-down sense. Further, because the crustal and mantle portions of the western Idaho shear zone are offset, we can clearly determine that it is the mantle portion of shear zone that is responsible for generating and/or guiding the magmatism. The timing clearly indicates that the cessation of magmatism follows the change in kinematics, suggesting that lithospheric tectonics is the cause of the magma intrusion. Further, the reactivation is clearly dextral oblique and the magmatism occurs along potentially the most anisotropic, vertical zone in the western US Cordillera (western Idaho shear zone). Thus, scars in the mantle lithosphere, reactivated in oblique divergence, appear to exert a fundamental control on melt migration and, possibly, melt production.
T32B-07
Origin and Evolution of the Iceland Plateau
Seafloor spreading within the Iceland region has been complex since the opening of the North Atlantic in late Paleocene-early Eocene. Whereas symmetric magnetic anomalies can be traced parallel to the Reykjanes Ridge and Mohns Ridge back to chrons 23-24, anomalies within the Iceland Plateau, west of the Aegir Ridge and along the Greenland-Iceland-Faeroe Ridge are irregular, indicating plate boundary complexities, most likely associated with branched accretion zones. Our 700 km long refraction/reflection and gravity profile, straddling 66.5°N across the Iceland Insular Shelf, Iceland Plateau and western Norway Basin revealed large variations in crustal structure between the presently active Kolbeinsey Ridge and extinct plate boundaries. The westernmost 300 km of the profile lies across the Iceland shelf, considered to have formed by rifting at the Kolbeinsey Ridge whereas the easternmost 400 km lie across the Iceland Plateau and Norway Basin, a region formed by rifting at the Aegir Ridge and possibly containing slivers of older crust rifted off the east Greenland margin along with the Jan Mayen Ridge. Crustal thickness varies from 4-5 km across the Aegir Ridge to 12-15 km beneath the Iceland Plateau and from 24-28 km beneath the outer Iceland shelf, to 12-13 km near the southern tip of the Kolbeinsey Ridge. Pronounced undulations in lower crustal structure and corresponding gravity highs across the Iceland Plateau are most likely associated with extinct spreading centers indicating that branched crustal accretion zones existed west of the Aegir Ridge prior to the westward ridge jump forming the KR at 26 Ma. Slivers of older crust, rifted off the east Greenland margin may exist between individual rift segments. However, more extensive surveying is required in order to reveal details of crustal evolution within the Iceland Plateau.
T32B-08
Middle Cambrian to Ordovician Arc – Back-arc Development on the Leading Edge of Ganderia, Newfoundland Appalachians
Evolution of many modern intra-oceanic and continental arc systems is exemplified by cycles of arc construction, rifting and separation of remnant and active arcs by a back-arc basin cored by oceanic crust. The inherent subductability of rifted arc complexes leaves only a fragmentary record of these processes in the ancient record. Synthesis of recently obtained geochronological, geochemical, isotopic and stratigraphic data is enabling the resolution of the evolution of the Cambro-Ordovician Penobscot - Victoria arc system that developed on the leading edge of Ganderia, a peri-Gondwanan microcontinent. The two stages of arc – back-arc development display distinctly different magmatic and sedimentary histories in a predominantly extensional supra-subduction zone setting. They are separated by an orogenic episode marked by the obduction of back-arc ophiolites onto the Ganderian passive margin. The Cambrian to Lower Ordovician Penobscot arc is characterized by continuous migration of the magmatic front, and the development of multiple volcanically active rift basins. The rift basins display a variety of characteristics ranging from bimodal calc-alkaline magmatism to felsic-dominated incipient rift magmatism to tholeiitic/boninitic supra-subduction zone ophiolites. Comparison to modern analogues suggests that the Penobscot arc developed in a similar setting to the volcanically active Havre Trough and Taupo Volcanic Zone. In contrast, the Victoria arc phase is dominated by development of multiple epiclastic rich volcano-sedimentary basins above tectonically modified Penobscot basement with only sparsely preserved magmatic rocks, characterized by calc-alkaline felsic volcanics and tholeiitic to alkaline back-arc basin basalts. The change in character of the back-arc volcanic rocks over time may reflect a multitude of largely interrelated tectonic factors including the speed of slab retreat, degree of extension in the arc versus the back-arc basin (i.e., the Cambrian Penobscot arc compared to the Ordovician Exploits-Tetagouche back-arc), reactivation of inverted Penobscot extensional faults during Middle Ordovician rifting and/or depletion of fertile components by the Middle Ordovician.