Geodetic, Geological, and Geophysical Observations in Extending Terrains
Presiding: P La Femina, Rosenstiel School of Marine and Atmospheric Sciences, University of Miami; R Bennett, University of Arizona
G23C-01 13:30h
How extensional structures at divergent plate boundaries become oblique to the geodetic spreading vector
Since the establishment of the regional plate-tectonic spreading vectors using geodetic measurements, it has become apparent that many extensional structures at divergent plate boundaries are oblique to the associated spreading vectors. In fact, entire ridges, such as the Mohns Ridge and the Reykjanes Ridge, are oblique to their spreading vectors. The spreading vector normally coincides with the regional direction of sigma 3, the minimum principal compressive (maximum tensile) stress. Since extensional structures should, as a rule, strike perpendicular to sigma 3, the fact that most are oblique to a particular spreading vector implies that the local direction of sigma 3, which largely controls the strike of extensional structures, cannot be parallel to the spreading vector. None of the volcanic zones of Iceland are perpendicular to the regional geodetic spreading vector. Most of the extensional structures within these zones are also oblique to the vector. For example, the general geodetic spreading vector in the Iceland area is about N105° E. By contrast, the general trend of the North Volcanic Zone is about N02° E, that of the West Volcanic Zone N37° E, and that of the East Volcanic Zone N45° E. Thus, the minimum angle between the spreading vector and the volcanic zones is about 77° for the North Volcanic Zone, about 68° for the West Volcanic Zone, and as little as 60° for the East Volcanic Zone. Here I do not consider the Snaefellsnes Volcanic Zone, trending roughly E-W and thus making an angle of about 15° to the spreading vector. The main extensional structures associated with the volcanic zones, particularly the rift zone, are tension fractures, normal faults, dikes, and feeder dikes (volcanic fissures at the surface). Many of these extensional structures occur in swarms, referred to as volcanic systems (fissure swarms being parts thereof), many of which are non-parallel to axes of the plate-boundary segments within in which they occur. To explain the oblique extensional structures and volcanic systems, many boundary-element and finite-element models have been made. The numerical models indicate, first, that the principal stresses, hence the direction of sigma 3, commonly rotate at contacts between the volcanic zones and the surrounding Pleistocence and Tertiary rocks. This rotation is primarily due to two factors. One factor is the difference in mechanical properties (particularly Young's moduli) between the rocks that constitute the volcanic zones and those that form the Pleistocene and Tertiary lava pile. The other factor is the type of loading: here, the spreading vector which is oblique with respect to the trends of the volcanic zones. Second, the numerical models also show that the local stresses within volcanic systems and associated composite volcanoes may be very different from those of the volcanic zones themselves. These local stresses are partly attributable to shallow chambers and deeper reservoirs that are associated with the volcanic systems. Partly, however, the local stresses are due to differences in the mechanical properties of the rocks that constitute the active volcanic systems and those that constitute the other parts of the volcanic zones. The magma chambers as well as the young lava flows and pyroclastic rocks, particularly in the composite volcanoes, act as soft inclusions that generate local stress fields which may be very different from the associated regional stress fields.
G23C-02 INVITED 13:45h
Dripping of a Mantle Lithosphere Root and Tectonics in the Sierra Nevada Region: Reconciling Geodynamic Modeling with Geodetic/Geological Constraints
Recent geophysical and geochemical studies in the Sierra Nevada region indicate that mantle lithosphere may be actively descending as a Rayleigh-Taylor-type instability. It has been postulated that this "drip" instability influenced the structural evolution of the crust and may be responsible for anomalous present-day topography in the area. Here, we consider in more detail this type of geodynamic mechanism in controlling the active and recent tectonics in the southern Sierra Nevada and surrounding region. Specifically, we conduct a series of scaled 2D numerical and 3D physical analogue experiments to examine quantitatively the evolution of the crust in response to underlying mantle instability. In the experiments, an idealized crust-mantle lithosphere-mantle system is configured in an upper mantle-scale solution space and a density perturbation is introduced into the model mantle lithosphere to initiate the gravitational instability. Observables from the experiments include surface topography, accumulated crustal deformation, and instantaneous strain rate. We compare the latter to crustal strain accumulation inferred from GPS measurements. In particular, an attempt is made to isolate the regional velocity field from the larger-scale North America-Pacific relative plate boundary motion, and to determine the contribution of steady (vs. episodic) strain accumulation in the measurements. In conjunction with the geodetic constraints, we reconcile the predicted deformation in the experiments with geological/geophysical observations of crust and mantle lithosphere structure, and compare the pattern of surface topography. The results provide integrated insights into the role of a local tectonic forcing in driving active and recent tectonic activity in the Sierra Nevada region.
G23C-03 14:00h
Modeling the Secular Velocity Field Across a Divergent Plate Boundary: Geodetic GPS and Geology, South Iceland
Detailed GPS observations in south Iceland combined with two-dimensional elastic half-space models and numerical viscoelastic coupling models indicate that spreading rates in the Western Volcanic Zone (WVZ) increase from 2.8±0.9 mm/yr in the northeast to 7.0±0.4 mm/yr in the southwest. Conversely, spreading rates in the Eastern Volcanic Zone (EVZ) decrease from 19.0±2.0 mm/yr in the northeast to 11.0±0.8 mm/yr in the southwest, the direction of ridge propagation. Summed extension rates across the two rift zones are approximately constant and equivalent to the total plate rate (~18-20 mm/yr), consistent with a simple propagating ridge model whereby the WVZ is deactivating in the direction of EVZ propagation. We interpret the locations of maximum surface velocity gradient to indicate the sub-surface locus of magma accumulation and presumably the locations of future major rifting events; these are generally consistent with Holocene patterns. The neovolcanic zone of the slower WVZ, with fewer Holocene eruptions, consists of a narrow (10-20 km wide) axial graben. The faster EVZ has little normal faulting, higher topography and five historical fissure eruptions, reflecting its proximity to the hotspot and its nature as an active propagating ridge. The last major rifting event in the EVZ was the 1783-84 Lakagigar fissure eruption. However, the maximum surface velocity gradient is located 20 km to the east, on the Brdabunga-Veidivtn fissure swarm, which had a small volume eruption in 1862-64. The pattern of current strain accumulation suggests that the historical pattern of intra-ridge jumping of magmatic activity in the EVZ and crustal accretion across a 60 km wide zone is continuing.
G23C-04 14:15h
Anomalous geodetic dilatation at the Central Nevada Seismic Belt: Inferring secular and transient contributions
The Basin and Range province of the western United States exhibits distributed contemporary extensional and transtensional faulting. Near the western edge of the province a sequence of large historic (M6.8-M7.5) earthquakes occurred in a belt that roughly follows the regional trend in faulting and topography. This alignment of ruptures, known as the Central Nevada Seismic Belt (CNSB), exhibits significantly elevated geodetic extension compared to its surroundings. If interpreted solely as secular strain, the geodetic extension implies fault slip rates that are over a factor of two greater than rates inferred from paleoseismology. If interpreted solely as post-seismic viscoelastic relaxation of the lower crust and upper mantle, then this implies that secular strain rates are near zero, which would underestimate geologic rates. Additionally, the CNSB resides in an area that makes this distinction difficult. East of the CNSB geodetic strain rates are almost indistinguishable from zero, but in the northern Walker Lane to the west, the strain rates are among the highest in the province. Because it lies near this boundary, it is not clear whether the CNSB should behave more like the Great Basin or the Walker Lane. Distinguishing how much of this extension is locally due to faster slip rates on faults versus post-seismic processes is central to understanding regional crustal kinematics, rheological stratification of the Basin and Range lithosphere, and to quantifying seismic hazard in the region. We consider the two end-member cases discussed above in order to develop models that are consistent with geodetic, seismological and geological observations. The first explains tectonic horizontal geodetic extension with interseismic loading of locked normal faults whose pattern is a measure of the long-term slip rate on the fault. The second explains all deformation with post-seismic transients following a recent normal faulting event. We iterate over a large number of possible rheological structures, varying the lower crustal and upper mantle viscosity, generating predicted geodetic velocities and strain rates. We then evaluate the misfit of the data to the predicted velocities and strains rates to find the most likely rheological structures. To estimate an intermediate model, we subtract a model of secular strain that is consistent with the paleoseismologically inferred slip rates from the geodetic data. These models are constrained by a regional compilation of horizontal Global Positioning System velocities, a strain rate model inferred from these velocities, and a compilation of paleoseismic and seismological data for the historic and pre-historic earthquakes at the CNSB.
G23C-05 14:30h
Using a geodetic strain rate model for western North America to improve our understanding of the driving forces behind Basin and Range extension
The past and present-day extensional deformation fields in the northern Basin and Range (BR) Province have variably been explained to be the result of a variety of geodynamic forces. Particularly, stresses resulting from gravitational potential energy (GPE) differences (due to lithospheric and/or asthenospheric density variations) have been invoked in many studies to explain all or most of the extensional deformation. An alternative hypothesis is that BR extension could be explained on purely kinematic grounds; i.e., as a result of the Pacific plate pulling away from the Colorado Plateau. To separate the two mechanisms, areas of extension and contraction (ie, kinematic `sources' and `sinks') need to be recognized and compared to GPE gradients. The relative Pacific-North America (PA-NA) motion implies a transform plate boundary zone between the two plates, as is evident along the Queen Charlotte and southern Imperial Valley Faults in the north and south, respectively. Between these faults lies a part of the plate boundary zone, including the BR, that is much wider and more complex. Yet, when Juan de Fuca (JdF) motion is accounted for as a rotating block within the PA-NA boundary, the crustal flow field within the diffuse plate boundary zone must satisfy the condition of being part of a transform plate boundary zone. That is, the net-flux of material in or out the plate boundary zone is expected to be zero. When we apply Gauss' Theorem to a strain rate field derived from geodetic velocities and imposed JdF motion, we show that there is indeed no net-flux; i.e., the net-dilatation within the boundary zero is zero. The constraint of no net-dilatation allows us to compare BR extension with zones of contraction within the plate boundary zone, as indicated by the strain rate model. If the relative position of the identified source and sink are not consistent with GPE gradients, the present-day BR extension is probably of kinematic origin. A correlation, on the other hand, would be consistent with (but not necessarily proof of) a causal relationship between GPE gradients and extension.
http://geodesy.unr.edu
G23C-06 14:45h
Constraints on Lithospheric Rheology From Fault Displacement Rate Histories and Numerical Experiments
Recent displacement rate and geodetic data on the San Andreas, San Jacinto and eastern California shear zone suggest that changes in the geometry and/or the magnitude of the applied forces on the crust (e.g., a general or local change in fault strike relative to plate motion) can generate strain repartitioning within the crust on time scales of millions to thousands of years. The rates over which this repartitioning takes place in response to changing forces are controlled by the rheological evolution of the lithosphere. We investigate the implications of observed fault displacement histories for the rheology of the lithosphere using 2.5 D numerical experiments of deformation in an analogue system. The numerical technique used allows for the spontaneous formation of elastoplastic shear zones and flow in a Maxwell viscoelastic lower crust. The results show that when a strike slip fault is rotated to strike obliquely to the direction of relative plate motion it causes changes in bending and frictional stresses due to the formation of topography. To accommodate these changes, a conjugate system of oblique-striking strike slip faults develops. The total displacement is then slowly distributed over the new fault system on the time scale of mountain building (i.e. million of years). The rate of change is dependent on the strength of the lithosphere as well as the amount of obliquity applied on the initial strike-slip fault. In other numerical experiments we show that in a system of multiple strike-slip fault zones, displacement rate changes can occur over a time scale of about 100 kyr. This time scale corresponds to the Maxwell time at the brittle ductile transition (BDT). In such a system the lithospheric displacement is alternatively distributed (over 100 kyr) in clusters localized in lower crustal channels and over strike-slip fault zones. We show that the clustering time scale is controlled by the ratio of upper to lower crustal strength. This incomplete exercise shows how displacement rates data sets spanning thousands to millions of years can be used to constrain numerical experiments of lithospheric deformation and, in doing so, place new constraints on the rheology of the lithosphere.