T41D-1592
Mathematical Modeling Deformation Processes in Continental Margins
Mathematical modeling of marginal structures is a part of the NSF science plan MARGINS (MARGINS Science plan, 2004). It is outlined in this plan that models will provide the framework within which diverse datasets may be synthesized to illuminate intricacies of geodynamic processes such as strain localization and evolution of fault systems. Mathematical modeling is crucial for assessment of the relative importance of plate-tectonic forces versus regionally-derived forces in driving deformation processes in the continental margins. Four models of a continental margin were formulated and studied: (1) a heavy layer with variable thickness, rheology, and density; (2) an anomalous density body within a margin; (3) a superposition solution of (1) and (2) models, and (4) the geophysical deep section in the Asia-Pacific continental margin. Mathematical methods and computer algorithms were developed for evaluating stress and strain tensor components (strain rate tensor components in the case of the viscous rheology) in the formulated models. These models show that the continental margin is the area of high concentration of stress and strain and is the area of very active deformational processes. Using the Coulomb's law (as well as the energy of deformation criterion), areas of ultimate state of the material in continental margin were discovered. Calculations showed that the additional horizontal compression or extension applied to the layer changes pattern of the areas of ultimate state of the material. Compression makes the ultimate state zones be directed toward the 'continental' region, while additional extension makes these zones be directed toward the 'oceanic' part of the model. This result illustrates how interaction of plate-tectonic forces and locally-derived forces can influence deformation processes occurring in the continental margins. Superposition solution of the first and second problems showed how stresses induced by the anomalous density body within a margin can disturb the stress field produced by topography and change the material's ultimate state zones pattern. The comprehensive analysis of the geophysical deep section in the Asia-Pacific continental margin revealed a remarkable correspondence between calculated stresses, seismological and electrical models, gravitational and magnetic fields, and heat flow data of the section.
T41D-1593 INVITED
Similar and Contrasting Response of Rifting and Transtension in the Gulf of California and Walker Lane to Preceding Arc Magmatism
The Gulf of California (GC) and Walker Lane (WL) have undergone strikingly similar development with strike-slip faulting following initial extension. They differ significantly in the amount of Pacific-North American plate motion taken up by each: essentially all relative motion in the GC and ~25% in the WL. In both areas, ancestral arc magmatism preceded and probably focused deformation, perhaps because heating and/or hydration weakened the lithosphere. However, differences in migration of the Rivera (RTJ) and Mendocino triple junctions (MTJ) related to differences in the orientation of plate boundaries determined how strike-slip faulting developed. Abrupt southward jumps in the RTJ led to abrupt cessation of magmatism over arc lengths of as much as 1000 km and initiation of east-northeast extension within the future GC. The best known jump was at ~13 Ma, but an earlier jump occurred at ~18 Ma. Arc magmatism has been best documented in Baja California, Sonora, and Nayarit, although Baja constituted the most-trenchward fringe of the ancestral arc. New and published data indicate that Sinaloa underwent a similar history of arc magmatism. The greatest volume of the arc immediately preceding RTJ jumps was probably in mainland Mexico. Arc magmatism shut off following these jumps, extension began in the future GC, and strike-slip faulting either followed or accompanied extension in the GC. In contrast, the MTJ migrated progressively northward. New and published data indicate magmatism generally shut off coincident with this retreat, but distinct nodes or zones of magmatism, presumably unrelated to subduction, persisted or initiated after arc activity ceased. We have suggested that the WL has grown progressively northward, following the retreating arc, and that the northern WL is its youngest part. However, the timing of initiation of strike-slip faulting in most of the WL is poorly known and controversial. Testing our hypothesis requires determining initiation and magnitudes of total slip across different parts. Despite the progressive migration of the MTJ, arc magmatism ceased abruptly at the latitude of Lake Tahoe ($39.2\deg$) at about 3 Ma, and the southern end of the active Cascade arc jumped ~160 km northward to Lassen Peak ($40.5\deg$), where it remains. Geologic data indicate strike-slip faulting began between these two areas immediately following the end of arc magmatism. The southern Cascade arc is undergoing ~east-west extension, which was the case for the northern Walker Lane immediately before strike-slip faulting began. Further progression or steps in magmatism and strike-slip faulting will likely follow further northward migration of the MTJ.
T41D-1594
Surface Wave Tomography Beneath the Gulf of California Rift Zone
The Gulf of California (GofC) links the San Andreas strike-slip fault system in California to the oceanic spreading system of the East Pacific Rise as the plate boundary of two important tectonic elements, the North American and Pacific plates. Although a lot of studies have been done, due to lack of data the state of the underlying mantle is still largely unknown. We applied the two-station-method using earthquake data recorded by the NARS-Baja array in the GofC. The study is conducted for fundamental mode Rayleigh wave phase velocities along station pairs in the period range from 10 to 100 seconds. One-dimensional shear velocity inversions have been performed to give an idea of the shear velocity structure along several paths. Furthermore, a set of phase velocity maps (including anisotropy, grid spacing 100 km) have been obtained, and resolution analyses were carried out. The results indicate strong structural variations. Although the whole region shares the character of a very thin lithosphere on top of a hot asthenosphere, we still see the lithosphere thinning towards the mouth of the gulf. Also, the lithospere beneath the peninsula is slightly thicker than beneath the Mexican mainland. A prominent low phase velocity anomaly is observed directly beneath the gulf at periods of 25-30 seconds. This anomaly is indicative of melting under the lithosphere along the rift in the gulf. The low velocity anomaly migrates towards the north-east at longer periods (i.e. larger depths). The two-theta terms of anisotropy show some coherent patterns. In northern Baja California the fast axes of the phase velocity maps are parallel to the strike of the plate boundary (or the direction of Pacific Plate motion) at periods of 14 to 20 seconds. At longer periods (30-40 s) this direction changes to ENE to become predominantly E-W at periods larger than 50 s. The E-W direction corresponds to the fast direction of SKS-splitting measurements for southern California and northern Baja California and seems related to asthenospheric flow. South of this region (at 28N-30N) and beneath the GofC our fast phase velocity directions are mostly perpendicular to the strike of the plate boundary over the entire period range of 14 to 100 s. This direction approximately corresponds to Miocene ENE extension or the current North American absolute plate motion direction.
T41D-1595
3-D tomography of the Gulf of California extensional province by Rayleigh wave inversion
The Gulf of California, separating the Pacific and North American plates, is an excellent area to study the continental rifting process, including magmatism, lithosphere deformation and crustal thickness distribution. To create a 3-D image of this rift zone, we carry out a Rayleigh wave dispersion study, which can sense velocity structure at different depths using different period surface waves. We use fundamental modes waves ranging in period from 20 to 140 s generated by teleseismic sources and recorded by the NARS-Baja seismic array and stations in southern California. Surface waves traveling along the Gulf or Baja California have typically undergone waveform distortion before they reach the study area by propagating along continental margins and subduction zones that induce scattering and multipathing. Further waveform distortion is caused by heterogeneities in the extensional province. In other words, the incoming waves are no longer well represented by a simple plane wave. Instead, we employ an array-processing method (Forsyth and Li, 2005), which describes the incoming wave field as the interference of two plane waves which travel along slightly different paths from the great circle paths and employs finite frequency response kernels (Zhou et al. 2004) to represent scattering within the study area. In southernmost California, there is a low-velocity anomaly in the shallow upper mantle dipping eastward from beneath the Peninsular Range toward the Salton Sea. At depths greater than 150 km, there is a high velocity anomaly beneath the Peninsular Range. Together, this pattern resembles the anomaly pattern beneath the southern Sierra Nevada, where the lower lithosphere is thought to have delaminated and been replaced by upwelling asthenosphere. One of the primary goals of this study is to determine how far south this pattern continues. Another goal is to detect remnants of the Farallon plate that may still be attached to the lithosphere beneath Baja California, left over from the stalling of spreading before the ridge subducted. Preliminary results indicate that the slowest anomaly in the asthenosphere is centered at about 28.5 N beneath the Gulf of California.
T41D-1596
Upper-Mantle Shear-Velocity Structure Beneath the Gulf of California
The MARGINS initiative on Rupturing Continental Lithosphere is focused on understanding the process of continental rifting and its evolution into sea-floor spreading. The upper mantle likely plays an important role in this process; stresses associated with mantle flow likely drive the rifting process, and the thermal and compositional state of the mantle establish the environment within which the crustal deformation takes place. The Gulf of California (GoC) focus area provides an excellent opportunity to evaluate mantle control on the rifting process. The GoC sits on top of a broad region of low mantle velocities that suggest pervasive high-temperature conditions in the upper mantle. However, gulf extension over the last 10 Myr has been highly localized, with a spatial transition in extensional style from seafloor spreading in the south, to continental extension in the north. Using data from MARGINS-funded temporary broadband arrays spanning the Gulf, we evaluate the degree to which upper-mantle seismic velocities and thus temperature correlate with degree and style of extension. We utilize over 30 regional earthquakes recorded by the NARS-Baja stations around the Gulf of California in 2002- 2005. The frequency-dependent travel times surface waves (Rayleigh and Love) from these earthquakes were measured relative to a standard tectonic model TNA using a cross-correlation analysis. A preliminary analysis of these travel times suggests substantial variation in upper-mantle velocities across the region. The Gulf appears to be significantly slower than the Mexican mainland to the east. Within the GoC, the northern gulf appears significantly slower (and thus hotter) than the southern Gulf, perhaps due to more efficient cooling of the upper mantle due to the seafloor spreading. Travel-times for paths traveling along Baja show a strong frequency dependence, possibly due to the combination of a thickened crust underlain by a high-velocity lithosphere. The difference between the average slowness of the Rayleigh and Love waves indicate the presence of seismic anisotropy in the upper-mantle beneath the entire region. We will quantify and expand upon these initial results by inverting for 3D models upper-mantle shear-velocity across the region. If the data are available, we will also incorporate new observations from the SCOOBA OBS array, to be collected in October 2006.
T41D-1597
The Heterogeneity of Southern California Mantle: Insights From Basalts and Mantle Xenoliths Along the Eastern Margin of the Salton Trough
Southern California has been in an active plate margin setting since the Permian, and has possibly had a very dynamic mantle history. Episodes of normal subduction, shallow subduction, slab breakoff, and the transition to transtension could have imparted different geochemistry and scales of heterogeneity to the lithospheric mantle and asthenosphere. A series of xenolith-bearing basalt dikes and flows exposed along the San Andreas fault in Joshua Tree National Park (JTNP) provide a northwest to southeast transect of southern California mantle in Pliocene time. Non-xenolith bearing alkali basalts exposed on the eastern side of the park provide an east to west transect of the mantle. The xenolith and non-xenolith bearing basalts differ based on their Fe, Mg, and Sr content suggesting that the mantle source of these basalts becomes more enriched along the transect from east to west. The differences in chemistry and the presence and absence of xenoliths suggest that there may have been a boundary in the mantle between the basalt reservoirs. This boundary may have been emplaced in the Mesozoic to Cenozoic due to normal and flat-slab subduction. The question becomes, does this trend and subducted slab boundary continue along the northwest to southeast transect through the xenolith-bearing basalt dikes and flows. The five xenolith-bearing basalt sites contain Group I xenoliths, dominated by spinel lherzolite (52%), but are multi-modal including harzburgite (29%), dunite (5%), and wehrlite (14%). Mantle-derived olivine ranges Fo$_{87-91}$, and clinopyroxene (cpx) ranges Wo$_{42.51-50.91}$ En$_{47.39-53.30}$ Fs$_{1.72-4.83}$. Due to the Fo content and modal mineralogy, these samples more closely resemble previously studied Proterozoic mantle than Archean mantle (cf. Lee, 2001). The cpx compositions yield a quenching temperature range of 900 � 1100 $\deg$C at 10 kb using the geothermometer projection scheme from Lindsley (1983). The quenching temperatures increase to the southeast approaching the central Salton trough. This implies that the southeastern xenolith-bearing basalts may have originated at a greater depth consistent with a deeper depleted mantle source, whereas the northwestern xenolith-bearing basalts may have originated at a shallower depth consistent with a higher Fe content in the basalt. A second model implies that the increasing temperatures to the southeast represent a lateral thermal gradient reflecting a heat source associated with early rifting in the Salton trough consistent with a higher Mg content in the southeastern basalts. Preliminary results of Principle Components Analysis detect few differences in cpx geochemistry. This implies that the subducted slab boundary evidenced by the basalt geochemistry must be east of the xenolith-bearing basalt sites. This boundary must have a depth component which suggests that the flat slab subduction did not completely destroy the cratonic mantle.
T41D-1598
Transpressional deformation style and AMS fabrics adjacent to the southernmost segment of the San Andreas fault, Durmid Hill, CA
In the Salton Trough, the trace of the San Andreas Fault (SAF) ends where it intersects the NNW-trending Brawley seismic zone at Durmid Hill (DH). The topographic relief of DH is a product of faulting and folding of Pleistocene Borrego Formation strata (Babcock, 1974). Burgmann's (1991) detailed mapping and analysis of the western part of DH showed that the folds and faults accommodate transpression. Key to Burgmann's work was the recognition that the ~2m thick Bishop Ash, a prominent marker horizon, has been elongated parallel to the hinges of folds and boudinaged. We are mapping in detail the eastern portion of DH, nearer to the trace of the SAF. Folds in the eastern part of DH are tighter and thrust faulting is more prominent, consistent with greater shortening magnitude oblique to the SAF. Boudinage of the ash layer again indicates elongation parallel to fold hinges and subparallel to the SAF. The Bishop Ash locally is < 1m thick along fold limbs in eastern DH, suggesting that significant continuous deformation accompanied the development of map-scale features. We measured anisotropy of magnetic susceptibility (AMS) fabrics in the Bishop Ash in order to assess continuous deformation in the Ash at DH. Because the Bishop Ash at DH is altered, consisting mainly of silica glass and clay minerals, samples from DH have significantly lower magnetic susceptibilities than Bishop Ash samples from elsewhere in the Salton Trough. With such low susceptibilities, there is significant scatter in the orientation of magnetic foliation and lineation in our samples. Still, in some Bishop samples within 1 km of the SAF, magnetic foliation is consistent with fold-related flattening. Magnetic lineation in these samples is consistently sub-parallel to fold hinges, parallel to the elongation direction inferred from boudinage. Even close to the trace of the SAF, this correlation breaks down in map-scale zones where fold hinge lines change attitude, fold shapes change, and the distribution and orientations of fractures and veins changes. These zones of structural complication separate broader regions of more uniform deformation patterns. Together, the geometry of structures and AMS fabrics suggest that deformation in eastern DH occurs by the distortion and reorientation of more or less coherent blocks separated by narrow zones where structural elements change orientation.
T41D-1599
Fault History and Architecture of the Southernmost San Andreas Fault and Brawley Seismic Zone: New Constraints from CHIRP Data Acquired in the Salton Sea
The Salton Trough is the boundary between spreading-center dominated extension in the Gulf of California and dextral strike-slip deformation along the San Andreas Fault (SAF) system. The Salton Trough provides an ideal opportunity to image this transition in modes of deformation. The critical portion of this system, namely the intersection of the SAF and the Brawley Seismic Zone (BSZ) in the southern Salton Sea has not been imaged by geophysical methods. To address this problem, we conducted a pilot, high-resolution seismic CHIRP survey in the Salton Sea offshore Bombay Beach. CHIRP imaging, together with onshore field mapping and paleoseismic investigations, has the potential to define the interaction between the SAF and the BSZ, as well as delineate fault architecture and strain partitioning in the central Salton Trough. Preliminary onshore examination of Lake Cahuilla sediments reveal lake-level changes and earthquake event chronology for the last ~1,000 years, and suggest a relatively long period of seismic quiescence for the southern SAF preceded by several events with shorter recurrence intervals. Fault excavations have revealed several lake episodes separated by terrestrial horizons that include distinct features such as mud-cracks. New CHIRP data show potential correlation of faulted offshore stratigraphy with paleoseismic deformation documented at an excavation site 15 km to the north adjacent to Salt Creek. Profiles image a well-defined fault trending obliquely to the strike of the onshore SAF, and the observed trend is sub-parallel to the BSZ. Offset stratigraphy across the fault imaged in CHIRP profiles increases with depth, with a maximum vertical offset of ~6-8 m. Relief of ~.5 m exists across the post-1905 surface and most likely represents deposition mantling an older scarp. Assuming high amplitude reflectors observed in CHIRP data correlate with lowered lake levels associated with weathering and/or desiccation horizons, then we can correlate the offshore CHIRP with the five most recent lake episodes observed at the excavation site. Predicted sedimentation rates from the CHIRP data, based on this assumption, are consistent with rates determined from nearby URS borings in the Salton Sea. The incremental, down-section increase in offset shown in CHIRP profiles is consistent with three to four events over the last four lake cycles, which is nearly equivalent to the timing and number of events observed at the Salt Creek paleoseismic site. If our interpretations are correct, the onshore excavation and offshore CHIRP data suggest that faults in the BSZ rupture in concert with the southern SAF. Furthermore, based on the limited imaging offshore, we suggest that the BSZ consists of a series of relatively short en-echelon normal-dominated faults. Such onshore-offshore studies can potentially answer important questions regarding linkage and slip partitioning between the southernmost San Andreas Fault, Brawley Seismic Zone, Superstition Hills Fault and cross faults such as the Elmore Ranch Fault, and will strengthen interpretations of the individual data sets.
T41D-1600
Chronostratigraphy of the Fish Creek-Vallecito Basin, SW Salton Trough: A High-Fidelity Record of Slip on the West Salton Detachment Fault and Subsidence in its Upper Plate
The Fish Creek-Vallecito basin contains a 5.1-km thick section of sedimentary rocks in the SW Salton Trough that range in age from 8.1 to 0.9 Ma. The section preserves a record of basin subsidence related to slip on the West Salton detachment fault (WSDF), which formed the main western rift-flank structure of the Salton Trough. We obtained a well-constrained chronology from compilation of existing (Johnson et al., 1983) and new paleomagnetic data, ages of two tuffs high in the section, and thicknesses calculated from the geologic map of Winker (1987) and our work in the lower 1.3 km. The tuffs yielded SHRIMP U-Pb ages of 2.56 $\pm$ 0.09 and 2.54 $\pm$ 0.09 Ma from single zircons. Geohistory analysis, corrected for paleobathymetry and global sea-level change, yields a decompacted subsidence curve with 5 segments bounded by abrupt changes in subsidence rate: (1) 0.46 mm/yr from 8.1 to 5.5 Ma; (2) 1.8 mm/yr from 5.5 to 5.2 Ma; (3) zero subsidence or slight uplift from 5.2 to 4.6 Ma; (4) 1.9 mm/yr from 4.6 to 3.2 Ma; and (5) 0.4 mm/yr from 3.2 to 0.9 Ma. The base of the Elephant Trees Fm, dated here at 8.1 Ma, provides the earliest well dated record of extension in the SW Salton Trough. Earliest marine incursion is dated at 6.3 Ma, and the first appearance of Colorado River sand coincides closely with the Miocene-Pliocene boundary (5.33 Ma). Because the base of the marine Imperial Group does not coincide with a change in subsidence rate, we suggest that initial marine incursion resulted from a latest Miocene global sea-level highstand superposed on steady subsidence. Thus, the inflections at 8.1 and 5.5 Ma are the two most likely ages for onset of slip on the WSDF, but 4.6 Ma is also possible. Variations in subsidence rate are not predicted by models for extensional detachment faults, and may reflect episodic pulsed fault slip and/or long- wavelength folding related to dextral-wrench tectonics. Rapid subsidence in segment 4 began during progradation of the Colorado River delta into the Imperial seaway, and continued during deposition of the fluvial Diablo and Olla formations. It ended at 3.2 Ma, 0.4 m.y. prior to the end of Colorado River sand input and transition to locally-derived Hueso Fm. This abrupt change in sediment composition occurred during slip on the WSDF, and may have been driven by an increase in sediment supply related to climate change. The entire basin has been inverted and completely exhumed, suggesting a rock uplift rate of ca. 6 mm/yr during the past 0.9 m.y. Rapid uplift coincides with modern transpressional deformation and strike-slip faults that cross-cut and terminated slip on the WSDF starting at 1.1-1.3 Ma (Steely, 2006; Lutz et al., in press; Kirby et al., in press).
T41D-1601
Evolution of Rotations in the Fish Creek Vallecito Basin, Western Salton Trough, CA
Rocks in the Western Salton Trough region record the history of slip on the transtensional West Salton detachment fault and initiation of younger strike-slip faults in this plate boundary zone. Spatial and temporal patterns of vertical axis rotations as determined by paleomagnetism can be used to provide valuable constraints on the structural-tectonic evolution of this area. Prior work includes the magnetostratigraphy of Plio-Pleistocene sedimentary rocks in the Fish Creek-Vallecito Basin (FCVB) (Opdyke et al., 1977; Johnson et al., 1983), who found that these rocks contain a complete record of geomagnetic field reversals spanning Pliocene-Pleistocene time. Johnson et al. (1983) also concluded that the FCVB had undergone $35\deg$ of CW rotation during the past 0.9 Ma. We resampled and reanalyzed their section, and sampled additional sedimentary and plutonic rocks in the Western Salton Trough in order to better document the history of vertical axis rotation recorded by these rocks. Results from 29 sites in the FCVB have well-defined magnetizations with two components. The first removed component in all samples is unblocked between 90 and 220 $\deg$C, and the second-removed components are unblocked between 300 and 590 $\deg$C. The second-removed components have either normal or reversed polarity. Sites from the Diablo Fm are predominantly reversed and have a mean of D = 204, I = -48.3, k = 37, $\alpha$$_{95}$ = $12.7\deg$, N = 5. Sites from the middle of the section (Olla and Tapiado Fms) are predominantly normal and have a mean of D = 8.1, I = 48, k = 32, $\alpha$$_{95}$= $8.7\deg$, N = 10. Sites from the upper portion of the section (Hueso Fm) have predominantly reversed polarity with means of D = 179.6, I = - 43.4, k = 82, $\alpha$$_{95}$ = $10.2\deg$, N = 4. Results from weakly-magnetized and deformed rocks of the La Posta pluton, on the south side of Whale Peak, have well-defined magnetizations with a group mean direction of D = 16.3, I = 37.3, k = 44, $\alpha$$_{95}$ = $7.4\deg$, N = 10. The stratigraphic distribution of declination anomalies combined with analysis of secondary magnetization appears to record ca. $24\deg$ CW rotation between ~4.0 and 2.5 Ma with little or no rotation since about 2.0-2.5 Ma, and no rotation or tilting since ~0.5 Ma. Comparing the directions from the footwall rocks at Whale Peak with published compilations from undeformed portions of the La Posta pluton (Symons et al, 2003) suggests that CW rotation of $36\deg$ $\pm$9 CW may be present in older rocks. Together, these results suggest that the FCV basin may have experienced moderate CW rotation during slip on the detachment and deposition of the fluvial Diablo and Olla formations. Crustal rotation is not predicted for a purely extensional detachment fault, and may reflect a component of dextral-wrench tectonics in this oblique transtensional system. Our paleomagnetic and bedding-dip data further suggest that southwestward tilting began around 1.0-1.2 Ma, during deposition of the upper Hueso Fm, and ended by ~0.6 Ma, a period for which Johnson et al. (1983) reported ~$35\deg$ of vertical-axis CW rotation. SW-ward tilting in the FCV basin is related to slip on the southern Elsinore fault, and thus its initiation appears to coincide closely with that of the San Jacinto and San Felipe faults.
T41D-1602
The West Salton Detachment Fault, Salton Trough, California: a Primary Low-Angle Normal Fault in an Evolving Dextral Wrench Zone
The west Salton detachment fault (WSDF), bounded the W rift flank, and was largely coeval with the southern San Andreas fault (SSAF). The WSDF is exposed in ~E-trending folds: broad, apparently primary corrugations S. Santa Rosa Mts., Borrego Valley-Pinyon Mts., Whale Peak, Vallecito Valley, and Tiera Blanca Mts) and narrow, post-WSDF folds (e.g., adjacent to San Felipe and Earthquake Valley faults). WSDF slip may have begun at ~12+, ~8.1, 5.5 or 4.6 Ma and was probably rapid from ~5 to 2 Ma. Two (U-Th)/He vertical transects from the WSDF footwall show rapid cooling since 12 Ma, and very rapid cooling between ~5.5-4.5 and ~2 Ma. Subsidence curves from the Fish Creek � Vallecito basin (FCVB; Dorsey et al., this session) show increased rates at ~8.1 Ma, 5.5, and 4.6 Ma. Syntectonic conglomerate (base ~8.1 Ma) there records earliest extension, but may have been only local. Widespread marine deposits (~6.3 to 4.25 Ma) locally contain syndetachment fault-scarp facies; eustatic sea level rise may have controlled initial marine flooding. Subsidence was most rapid from ~4.6 to 3 Ma. Upper-plate normal faults are rare but folds formed locally. At Borrego Mtn. a WNW-trending anticline formed by ~6 Ma and persisted until after 4 Ma, coeval with WSDF slip. Folding at Split Mtn may have begun earlier. The WSDF has at least 5 km of E or NE slip, from offset basement but higher WSDF strands carry syntectonic conglomerates some additional distance. (U-Th)/He apatite ages from the upper and lower plates suggest ~2.4 km of footwall exhumation, yielding 5-15 km of slip, depending upon dip assumed. WSDF striae scatter widely, but concentrate at 090-110, probably the main or most recent slip direction. CW vertical-axis rotations have occurred (Housen et al., this session): ~3-4 m.y. old FCVB strata are rotated 19$\deg$ $\pm$ 12$\deg$, and footwall La Posta pluton at Whale Peak rotated perhaps 36$\deg$ (relative to the Peninsular Range La Posta). Similar rotations were common in N Baja CA in latest Miocene-Pliocene time and that belt of rotation may continue into the western Salton Trough, but the FCVB rotations may reflect local Quarternary dextral shear and the Whale Peak data can be explained by SW tilting. In contrast, 2 to 0.6 Ma strata in the FCVB record no vertical-axis rotation but show syndepositional SW tilting toward the Elsinore fault. Syn-WSDF deposition (Hueso Fm.) in Vallecito Valley continued until ~0.9 Ma. At ~1.1 Ma, the supradetachment Borrego sub-basin was reorganized and uplifted, recording a switch to dextral faulting. Thus, the southern Elsinore, Earthquake Valley, San Felipe, and San Jacinto fault zones were dominant since ~1 Ma.
T41D-1603
Late Quaternary Range-Front Fault Scarps in the Western Sierra El Mayor, Baja California, Mexico: A Geomorphologic Expression of Slip Across an Active Low-Angle Normal Fault
The western margin of the Sierra El Mayor (SEM), in northeastern Baja California, is controlled by an active, top-to- the-west, low-angle normal fault named the Canada David detachment (CDD) that accommodates part of the extensional component of shearing between the Pacific and North American plates. The CDD has a length of 60 km and shows a curvilinear trace with two major antiformal and synformal megamullion pairs. Late Quaternary slip has produced a broad array of Quaternary scarps cutting alluvial fans along nearly the entire length of the CDD. Detailed mapping reveals eight regional strath terraces distinguished by surface weathering characteristics, soil profile development and relative elevation. Relative height between terraces increases in domains where the CDD and basin deposits are being uplifted due to either the basinward migration of faulting (e.g., rolling hinge) or flexural uplift in antiformal megamullion domains. Linear diffusion analysis of 46 synthetic fault scarps, with a calculated angle of repose $\Theta$$_{o}$ = $28.75\deg$, reveal fault scarp domains exhibiting both multi-modal and unimodal distribution of diffusion ages (kt). Uni-modal domains are typically younger, but there is no systematic variation in scarp age with distance along the CDD. Scarps yielding negative kt ages (i.e. scarps steeper than $\Theta$$_{o}$) are common in the north, near inferred locations of important historic seismic events. Microseismicity drops off significantly adjacent to these very young scarp arrays, which likely reflects a recent post-seismic stress drop. Domains of high seismic risk are identified by high microseismicity and lack of young scarps. Minimum estimates of the diffusivity constant (k) are calculated by coupling scarp diffusion ages and $^{10}$Be surface exposure ages of the faulted deposits. In the southernmost SEM a Q6 terrace with a minimum surface exposure age t = 233$\pm$6.6 ky (weighted mean of six rock samples) is cut by scarps with an average kt = 11.25$\pm$9.31 m$^{2}$, which yields a minimum k = 0.05$\pm$0.04 m$^{2}$/ky. Scarps cutting Q4 terraces yield diffusion ages (kt = 0.16$\pm$0.04 m$^{2}$) consistent with the inferred post-glacial age of the four youngest terraces.
T41D-1604
Imaging the Seismic Structure Beneath Gulf of California Determined From MCS Data (CORTES-P96)
The Gulf of California (GC) is a right lateral transform fault zone developed as a series of short spreading segments that delineates the boundary between the Pacific and North American plates. From April to May 1996, over 1500 miles of seismic reflection data (3000 c.i. and shot rate of 30s), potential field, seabeam and backscattering have been acquired along and across the length of the GC within the multidisciplinary experiment CORTES-P96 (R/V Hesperides). The data acquisition was aimed to increase our current understanding of the process of continental rifting and break up, mainly in actively extending and thinning areas enough to generate new crust. The main objective consist to resolve the complex geodynamics of the GC by imaging the most significant structures: the fracture zones (transform faults, escarpments), oceanic ridges, trapped continental blocks related to the San Andrea-GC transform system, ancient volcanic arcs. We present 5 transects crossing the GC from the east coast of the peninsula till the mainland and 4 transects extending along the coast showing the main morphological features. The multichannel seismic data show detailed images of the structure of the GC, which seems to be dominated by faulting. These profiles provide a good database to study the crustal structure and recent history/activity and to interpret the tectonics in the area surveyed.
T41D-1605
A Paleomagnetic Investigation of Large-Scale Vertical Axis Rotations in Coastal Sonora: Evidence for Transtensional Proto-Gulf Deformation
A paleomagnetic investigation into possible vertical axis rotations has been conducted in the Sierra el Aguaje and Sierra Tinajas del Carmen, Sonora, Mexico, in order assess proposed styles for oblique continental rifting in the Gulf of California. Two styles of rifting have been proposed; (1) strain partitioning (Stock and Hodges, 89), and (2) transtension (Gans, 97), for the Proto-Gulf period of the Gulf of California. The presence of large-scale vertical axis rotations would lend weight to the argument for transtension. The Sierra el Aguaje and Sierra Tinajas del Carmen are located in southwestern coastal Sonora, Mexico. The ranges represent the eastern-rifted margin of the central Gulf of California. This is one of the few areas of that margin which is entirely above water, with new ocean crust of the Guaymas basin lying immediately offshore of the western edge of the ranges. The ranges are composed of volcanic units and their corresponding volcaniclastic units that are the result of persistent magmatic activity between 20 and 8.8 Ma, including three packages of basalt and andesite that make excellent paleomagnetic recorders. Based on cross cutting relations and geochronologic data for pre-, syn-, and post-tectonic volcanic units, most of the faulting and tilting in the Sierra El Aguaje and Sierra Tinajas del Carmen is bracketed between 11.9 and 9.0 Ma, thus falling entirely within Proto-Gulf time. Existing field relations suggest the presence of large (>45°) vertical axis rotations in this region. This evidence includes: a) abrupt changes in the strike of tilted strata in different parts of the range b) ubiquitous NE- SW striking faults with left lateral-normal oblique slip, that terminate against major NW-trending right lateral faults, and c) obliquity between the general strike of tilted strata and the strike of faults. The results of the paleomagnetic investigation are consistent with the field evidence and show large clockwise rotations between ~30º and ~100º with no discernable translation. Such large-scale rotations lend credence to the theory that the area inboard of Baja California was experiencing transtension during the Proto-Gulf period, rather than the pure extension that would have been the result of strain partitioning.
T41D-1606 INVITED
Symmetric Rifting of Cold Lithosphere in the Southern Gulf of California
Modeling of an 881 km seismic refraction profile across conjugate rifted margins in the southern Gulf of California indicates an overall pure shear mode of extension. Forward modeling of crustal structure shows significant Moho topography, which has been verified by a tomographic inversion. This Moho structure is interpreted to be the result of extension of cold lithosphere: extension was accommodated by brittle deformation through the whole crust, and the preservation of Moho structure indicates no significant lower crustal flow in this region. Average lower crustal velocities of 6.3 km/s indicate no significant magmatism during rifting. However, there do appear to be magmatic intrusions in the areas of highest stretching (beta>4). Total NW-SE opening across the southern Gulf of California is estimated to be 495 km, which at the Pacific-North America spreading rate of 50 mm/yr indicates that extension must have begun at least 10 Ma. This is contrast to previous models of tectonic evolution in the Gulf of California, which suggested significant extension did not start until 6 Ma.
T41D-1607
Crustal Structure of the Southern Gulf of California and subducting Rivera plate
We present two new seismic velocity models that compose the most detailed picture to date of the crustal structure of the southeastern Gulf of California. Wide-angle and multi-channel seismic (MCS) data were collected in 2002 using the R/V Maurice Ewing and a 480-channel, 6 km-long streamer, ocean bottom seismometers and land seismometers each spaced 10-15 km apart along transects. Both transects discussed here cross the Jalisco continental rifted margin. Compared with its conjugate rifted margin, the Cabo section of Baja California, the Jalisco margin possesses a wider zone of transitional/extended crust. The oceanic crust across the mouth of the Gulf has a thickness of ~6 km, but the thickness of the Jalisco continental crust has yet to be determined. Wide-angle modeling of the Jalisco margin indicates the presence of the Rivera plate subducting under mainland Mexico. In addition, the trench between the down going slab and rifted margin shows up clearly on prestack depth migrated images from multi-channel seismic data. Previous workers have shown from hypocenters of microearthquakes that the southern Rivera plate dips ~10 degrees down to 20 km depth and then increases to ~50 degrees by 40 km depth; however, little is known about the depth of the plate under the Jalisco rifted margin. Understanding the limits of the lateral and vertical extent of the subducting plate will constrain plate tectonic configurations and timing of events at the mouth of the Gulf.
T41D-1608
Mid-Miocene to Early Pliocene Upper Crustal Extension in the Southern Gulf of California
600 km of multi-channel seismic data across conjugate rifted margins in the southern Gulf of California show many rifted basins. Analysis of these basins shows two distinct styles: the first consists of large basins with no obvious basin-bounding fault or large syn-rift sedimentary sequences and only small-scale basement faulting. The second style consists of smaller, simple half-grabens, mostly with lower syn-rift sequences and upper post- rift sequences. All recent sediments imaged appear to be post-rift, indicating no significant active faulting in this area at present. Rough estimates of basin age based on sedimentation rates indicate that these two styles correspond to two phases: the larger basins may have been formed beginning 14 Ma, whereas the simpler half- grabens are more likely formed beginning around 8 Ma. A highly reflective ropey layer is imaged at the top of basement along much of the transect; modeling of mommon-midpoint (CMP) supergathers indicates that this layer has a velocity of about 2.5 km/s and a thickness of 250-500 m. We interpret this layer as being the 20-11 Ma Comondu formation. In the Tamayo Bank and Trough region, adjacent to the southeast continent-ocean transition, this Comondu layer appears to continuously mantle the bank and trough, masking the large-scale faulting that created the trough, because despite the large size of this basin (1500 m sedimentary thickness) there are no syn-rift sedimentary sequences recording its creation. We interpret this as evidence for overlap between the end of the Comondu Arc and onset of NW-SE extension in the Gulf of California. Along with the estimated ages of the basins, all evidence points toward an alternative tectonic history for the Gulf of California, where significant NW-SE oriented extensions began possibly as early as 14 Ma.
T41D-1609 INVITED
Evolution of the mouth of the Gulf of California based on rift-related seismic sequences and onshore basins
Two related types of studies have produced major advances in understanding the evolution of the mouth of the Gulf of California: detailed studies of onshore basins with micropaleontologic age control is beginning in the San Jose del Cabo and Tres Marias basins and seismic reflection lines across at least 12 offshore basins with no direct age control, but crude ages from estimated sedimentation rates. We present a new scheme for development of the mouth of the Gulf based on these data. Additional seismic data and data from drilling are necessary to make further advances. The western Gulf margin along Baja California is narrow and basins are mainly partially starved. The onshore Cabo basin has marine strata from 8 Ma (possibly 10 Ma) and younger; the basin initiated at10-12 Ma. Older strata are tilted along a secondary fault in the east-central basin while younger strata are gently tilted. The eastern Gulf margin has a 40 � 100-km wide marine shelf. Maria Madre Island at the outer edge of the shelf has 1100 � 1600 m of Late Miocene (8 Ma) to Pleistocene strata. The section changes abruptly from basal marginal marine sandstone into thick mudstone and diatomite formed on a slope to deep marine, then shallows up to upper slope, and is capped by nonmarine Pleistocene. Two angular unconformities are in the Pliocene section. Offshore basins imaged in seismic are of two kinds. Older, more complex basins are 1-2 km thick and began at 10-12 Ma. These are either sag basins with many small normal faults or syn-rift basins with typically 1-5 km spacing on faults; there is one strike-slip fault. The upper parts of these basins are post rift. The other basins are simple half graben typically 400-600 m thick with lower syn-rift and upper post-rift strata; these formed ca. 4 - 6 Ma. The only active faults are on the western margin. Our synthesis suggests this sequence of events: formation of rift (and transtensional?) and sag-rift basins across the entire mouth of the Gulf at 12 � 10 Ma; syn-rift basin development with closely spaced normal faults continues to ca. 6-8 Ma; from ca. 6 to 4 Ma numerous smaller half graben form; at 3-4? Ma faulting on the eastern margin dies and basins become post-rift as the eastern margin enters drift stage; at 3-4 Ma the western basins near the center of the Gulf also become post-rift and Alarcon spreading ridge commences, but faulting continues at low rates along the western margin of the southern Gulf. The pattern of offshore faults, 3 onshore basins, and uncommon folding suggest that transtensional basins were common.
T41D-1610
New high-res multichannel profiles from the southern Gulf of California: Preliminary results
48-channel seismmic reflection profiles recently collected across continental margins and young oceanic basins clarify: (1) The pattern and timing of deformation in the obliquely ruptured and now submerged continental crust; (2) The nature and locations of the transitions from rifted and sheared continental crust to "immature" (sediment/sill complex) oceanic crust; (3) The extent of syn-rift eruptive volcanism and post-rifting intrusive volcanism; and (4) The structural effects of the highly asymmetric sediment supply to the southern Gulf of California. In addition to enriching geologic interpretation of recent multibeam surveys and rock-sampling campaigns, correlating the new data set with the several thousands km of single-channel seismic profiles collected from this region in past decades increases the usefulness of the extensive archival data.
T41D-1611
Evolution of the east-central San Jose del Cabo basin, Baja California Sur, Mexico
The San Jose del Cabo basin at the southern tip of the Baja California peninsula records the early tectonic evolution of the west side of the Gulf of California. This study focused on the east central margin of the basin. The basal La Calera Formation unconformably overlies Cretaceous granite and consists of conglomerate, pebbly sandstone and conglomerate, and sandstone deposited in alluvial fans and fan-deltas. Deposition of the La Calera Formation was from ca. 9-14 Ma. The lower member of the Trinidad Formation was deposited beginning ca. 9-13 Ma and consists of sandstone, mudstone, and shelly mudstone deposited in nearshore and estuarine environments. These age estimates are based on sedimentation rates and foraminifera and coccoliths from the NN 11A nannozone (7.4 � 8.6 Ma, GTS 2004). The middle member of the Trinidad Formation consists of deeper water mudstones deposited by turbidity currents and suspension settling in a shelf to slope and conglomerates deposited by submarine debris flows on the shelf. The basin began earlier than previously thought. The oldest marine rocks are ca.9-13 Ma, while sedimentation on the east side began at ca. 9-14 Ma, synchronous with estimates of initiation of offset on the San Jose del Cabo fault. The Zapote fault is a down-to-the-east normal and sinistral-oblique fault that exposes a wedge of granite and older strata in the footwall to the west. The fault was active during sedimentation in the late Miocene and possibly later. The fault divides the study area into an eastern hanging wall subbasin and western footwall subbasin. The eastern subbasin formed an embayment in the eastern margin of the Cabo basin. A regional flooding surface (ca. 8 Ma) can be correlated across the fault that marks a major marine incursion. Depositional systems evolved rapidly from coarse-grained terrestrial systems to fine-grained marine and estuarine systems. The Cabo basin provides an excellent analogue for comparison with offshore basins, which are broadly similar with more faulting in lower strata and fewer or no faulting in upper strata. Offshore seismic data show older, deformed syn-rift strata in half graben overlain by younger, undeformed post-rift strata. The normal faults with 1-5 km spacing cut the basement rock and oldest sedimentary units. The eastern margin of the Cabo basin has older, growth strata cut by the Zapote fault that are overlain by simpler strata. Smaller scale normal faults in the Cabo basin are no longer active while the Cabo fault remains active.
T41D-1612
Late Quaternary faulting in the Cabo San Lucas-La Paz Region, Baja California
While Baja California drifts, active deformation on and just offshore indicates that spreading is not completely localized to the rift axis in the Gulf of California. Using on and offshore data, we characterize normal faulting- related deformation in the Cabo San Lucas-La Paz area. We mapped sections of the north trending faults in a 150 km long left-stepping fault array. Starting in the south, the San Jose del Cabo fault (east dipping) bounds the ~2 km high Sierra La Laguna. It is >70 km long with well defined 1-10 meter fault scarps cutting the youngest late Quaternary geomorphic surfaces. Our preliminary mapping along the north central section exhibits extensive late Quaternary terraces with riser heights of tens of meters above Holocene terraces. The San Jose del Cabo fault trace becomes diffuse and terminates in the area of Los Barriles. Moving northward, the fault system steps to the west, apparently transferring slip to the faults of San Juan de Los Planes and Saltito, which then step left again across the La Paz basin to the NNW trending Carrizal Fault. It has an on shore length of > 60 km. We produced a 25 km detailed strip map along the northern segment. It is embayed by convex east arcs several km long and 100 m deep. In the south, few-m-high scarps cut a pediment of thin Quaternary cover over tertiary volcanic rocks. The escarpment along the fault is hundreds of meters high and scarps 1-10 m high where it goes offshore in the north. Near Bonfil, a quarry cut exposes the fault zone. It comprises a 5-10 m wide bedrock shear zone with sheared tertiary volcanic units. On the footwall, the lower silty and sandy units have moderately well developed pedogenic carbonate, whereas the upper coarse gravel does not. These late Quaternary units appear to be faulted by one to three earthquakes. Finally, we mapped the Saltito fault zone NNE of La Paz. It is a NW trending structure with well developed 5-10 meter high bedrock scarps defining its NW 5 km and slightly concave east with a 500 m left. Along all the fault zones studied, offset geomorphic surfaces indicate late Pleistocene to Holocene offset. These surfaces can be exploited to determine slip rates and produce a regional chronosequence to test for synchroneity of climatically modulated variations in sediment supply and transport capacity. In addition, a shallow marine geophysics and coring extends our mapping and provides important age control and improved stratigraphic assessment of fault activity.
T41D-1613
Numerical Simulation of Fault Interaction in a Trans-Tensional Setting, the La Paz �Los Cabos Region, Baja California, Mexico.
A number of medium to large normal faulting earthquakes occurred in the La Paz-Los Cabos region, at the tip of Baja California, within the last four decades. They (along with the tectonic geomorphology of the fault zones on the Peninsula) demonstrate that the existing structures in the area are active and capable of hazardous earthquakes. The goal of this study is to understand how the individual active faults in this region affect the behavior of the fault system as a whole. What role does fault interaction (i.e., stress transfer) and earthquake triggering play in the La Paz-Los Cabos region? Do we know all the significant, active and therefore hazardous structures that are part of the fault system? Are these structures capable of releasing the tectonically accumulated strain? What role does that fault system play in the regional, trans-tensional setting? To approach these questions we utilize a numerical model, based on derivations by Okada (1992), with which we compute the strain distribution and Coulomb failure stress for a given (frictionless) displacement along a rectangular fault patch and its interactions with other faults of the fault array. Beginning with simple geometric models of the fault system in the La Paz-Los Cabos region, we investigate under what conditions individual earthquakes may have triggered subsequent events. We focused on the M =5.6 event on April 4th 1969 that may have had an effect on the timing of the M=6.2 event on June 30th 1995. The proximity of these two earthquakes (epicenters only 60km apart) supports the idea that stress transfer caused by the 1969 event may have altered the seismic cycle of the fault activated in 1995. Because fault geometry and slip distribution during these two events are not well known, we explore the parameter space to learn under what conditions the 1969 event may have triggered the 1995 event. We apply the empirical relations among magnitude, fault geometry, and displacement, derived by Wells&Coppersmith (1994), to limit the parameter space to realistic scenarios. Preliminary results suggest that the 1995 earthquake has not experienced a clock-advance due to the 1969 event �mainly because of the relatively small average displacements, associated with a M =5.6 earthquake. However, larger earthquakes may have been caused by the same structures, depending on the actual fault size. Therefore, fault interaction as a significant process in this fault system cannot be ruled out unless the fault geometries (i.e., maximum earthquake size) of the structure involved are better known.
T41D-1614
Framework stratigraphy and chronology of the Comondu Group from Loreto to La Paz, Baja California Sur: regional correlation for use in across-Gulf of California correlations
The eastern coast of Baja California Sur along the Gulf of California has excellent exposure of the Comondu Group for ~250 km from near Loreto to La Paz (~26.5� to ~24�N). The Comondu Group is a 600 � 1500 m thick succession of volcanic, volcaniclastic, and sedimentary rocks that formed in a terrestrial volcanic arc and forearc basin from ~25 to 12 Ma. The Comondu Group is the precursor to formation of the Gulf of California and therefore is the best tool for regional correlation across the Gulf. Our 1:50,000 (and locally more detailed) mapping and stratigraphic analysis has produced a regional correlation supported by 20 ages, 17 recent 40Ar/39Ar ages. Three structural rift segments have pre-Comondu rocks at the base. In the two accommodation zones between the rift segments, the lower Comondu is at sea level. The Comondu Group gradationally overlies tuffaceous shallow marine strata in the south that changes to terrestial strata in the northern segment. The initiation of terrestrial deposition of the Comondu Group is at ~24 Ma in the south, at 25 Ma in the central segment, and at 22- 24 Ma in the north. We divide the Comondu Group into 3 main units. The lower unit is dominated by fluvial sandstone and conglomerate with minor eolian sandstone in the north and local volcanic breccias in the central and southern areas. Tuffs are common in the south (up to 6-7), while there are 0 to 1 in the north. The tuffs in the lower unit range from 25 to 19 Ma, with the upper San Juan tuff of Hausback now dated precisely at 19.37+/-0.06 Ma. Sedimentation rates in the lower unit are 100 m/m.y. The ages of the lower tuffs matches the range of caldera ages in the southern Sierra Madre Occidental reported by others and suggests potential across-Gulf correlations. The lower unit is overlain everywhere by 100 � 500 m of volcanic breccia and local lava flows of the middle unit. Ages from near the base of the middle unit suggest it is older (18.64) in the south and younger (16.6 Ma) in the north. There is a widespread erosional unconformity between the middle and lower units and the contact is an abrupt upward change to massive, coarse breccia. This may be a major westward jump of the volcanic arc at ~18.5 to 17 Ma. The upper Comondu Group is missing from the top of the rift escarpment in the south and north either due to major erosion or lack of deposition. Where the upper unit is preserved, it is up to 200 m of conglomerate in the central areas, and thick lava flows in the north; the lavas have 4 ages from 14.67 to 12.36 Ma and may represent a final westward jump of the arc.
T41D-1615
Truncated lithotectonic belts and initial rupture within the southern Gulf of California
While a multitude of factors were likely influential in controlling the site of intial lithospheric rupture to form the Gulf of California and spawn the Baja microplate, specific zones of weakness may have been locally important. Oft- cited factors include the position of the Miocene Comundu arc and the geometry of the medial Cretaceous batholith. Previous areomagnetic imaging (e.g., Langenheim and Jachens, 2003) has shown that the well- defined expression of the highly magnetic zone of the western Peninsular Ranges batholith (PRB) extends continuously southward from the Santa Ana Mountains (34 N) to the Todos Santos region just north of the southernmost tip of the peninsula (23 N). Our own field, petrologic, and geochemical studies have documented continuity of the eastern zone of the PRB including representatives of the ca. 95 La Posta tonalite- trondjhemite suite south to the Las Cruces region of the NE Los Cabos block. The boundary between these 100-90 Ma plutons (in the NW) and younger (65-90 Ma) Late Cretaceous intrusives (in the SE) extends SW across the Los Cabos block to the Todos Santos area where it appears to be truncated along the coast. Similarly, a NE-SW buttress unconformity expressed by Eocene and younger strata of the Magdalena platform also strikes offshore in this region. The apparent truncation of lithologic trends in the Todos Santos area corresponds with a marked change in the bathymetry of the offshore region where the Magdalena platform in the north gives way to the Todos Santos basin further south. Finally, both our and previous work with the formerly adjacent batholithic rocks of the mainland indicate that volumetrically significant elements of the 120-90 Ma batholith appear absent south of Puerto Vallarta area. We postulate that Eocene (or later) tectonic removal of the forearc region and 120-90 Ma portion of the of the Cretaceous batholith that affected the SW Los Cabos block gave rise to a significant mechical weakness that was highly influential in governing the location along the continental margin where initial rupture across the Cretaceous batholith occured.
T41D-1616
Results From NICLAKES Survey of Active Faulting Beneath Lake Nicaragua, Central American Volcanic Arc
In May of 2006 we used a chartered ferry boat to collect 520 km of seismic data, 886 km of 3.5 kHz subbottom profiler data, and 35 cores from Lake Nicaragua. The lake covers an area of 7700 km2 within the active Central American volcanic arc, forms the largest lake in Central America, ranks as the twentieth largest freshwater lake in the world, and has never been previously surveyed or cored in a systematic manner. Two large stratovolcanoes occupy the central part of the lake: Concepcion is presently active, Maderas was last active less than 2000 years ago. Four zones of active faulting and doming of the lake floor were mapped with seismic and 3.5 kHz subbottom profiling. Two of the zones consist of 3-5-km-wide, 20-30-km-long asymmetric rift structures that trend towards the inactive cone of Maderas Volcano in a radial manner. The northeastern rift forms a 20-27-m deep depression on the lake bottom that is controlled by a north-dipping normal fault. The southwestern rift forms a 25-35-m deep depression controlled by a northeast-dipping normal fault. Both depressions contain mound-like features inferred to be hydrothermal deposits. Two zones of active faulting are associated with the active Concepcion stratovolcano. A 600-m-wide and 6-km-long fault bounded horst block extends westward beneath the lake from a promontory on the west side of the volcano. Like the two radial rift features of Maderas, the horst points roughly towards the active caldera of Concepcion. A second north-south zone of active faulting, which also forms a high, extends off the north coast of Concepcion and corresponds to a localized zone of folding and faulting mapped by previous workers and inferred by them to have formed by gravitational spreading of the flank of the volcano. The close spatial relation of these faults to the two volcanic cones in the lake suggests that the mechanism for faulting is a result of either crustal movements related to magma intrusion or gravitational sliding and is not related to more regional tectonic forces. We did not find evidence for a regional fault parallel to the active volcanic axis or faults in the 10-30 km long offsets between the major stratovolcanoes. Core data is currently being analyzed and will improve constraints on the age of faulting.
T41D-1617
Results from NICLAKES Survey of Active Faulting Beneath Lake Managua,Central American Volcanic arc
Lake Managua covers an area of 1,035 km2 of the Central American volcanic arc and is enclosed by three major stratovolcanoes: Momotombo to the northwest was last active in AD 1905, Apoyeque in the center on the Chiltepe Peninsula was last active ca. 4600 years BP, and Masaya to the southeast was last active in AD 2003. A much smaller volcano in the lake (Momotombito) is thought to have been active <4500 yrs B.P. In May of 2006, we used a chartered barge to collect 330 km of 3.5 kHz profiler data along with coincident 274 km of sidescan sonar and 27 km of seismic reflection data. These data identify three zones of faulting on the lake floor: 1) A zone of north- northeast-striking faults in the shallow (2.5-7.5 m deep) eastern part of the lake that extends from the capital city of Managua, which was severely damaged by shallow, left-lateral strike-slip displacements on two of these faults in 1931 (M 5.6) and 1972 (M 6.2): these faults exhibit a horst and graben character and include possible offsets on drowned river valleys 2) a semicircular rift zone that is 1 km wide and can be traced over a distance of 30 km in the central part of the lake; the rift structure defines the deepest parts of the lake ranging from 12 to 18 m deep and is concentric about the Apoyeque stratocone/Chiltepe Peninsula; and 3) a zone of fault scarps defining the northwestern lake shore that may correlate to the northwestern extension of the Mateare fault zone, a major scarp- forming fault that separates the Managua lowlands from the highlands south and west of the city. Following previous workers, we interpret the northeast-trending group of faults in the eastern part of the lake as part of a 15- km-long discontinuity where the trend of the volcanic arc is offset in a right-lateral sense. The semi-circular pattern of the rift zone that is centered on Chiltepe Peninsula appears to have formed as a distal effect of either magma intrusion or withdrawal from beneath this volcanic complex. The Mateare fault forms the southern edge of the large depression in which the volcanoes and lake have formed. Its onshore segment west of Managua strikes to the WNW while the submerged part of the fault beneath the lake strikes to the WNW and mimics the change in the orientation of the valley at this point. We see no obvious linkages between the faults in the three provinces and suggest that at least the central province may be controlled by a distinct volcano rather than reflecting regional tectonic motions.
T41D-1618
Active rifting in the continental backarc of North Island, New Zealand
Active rifting in the Central Volcanic Region (CVR) of North Island, New Zealand, has thinned the crust by at least 50%. This volcanic area includes the Taupo Volcanic Zone and is a direct extension of Tonga-Kermadec oceanic backarc rifting into continental lithosphere. Interpretations from refraction and wide-angle reflection seismic data provide evidence for a magmatic underplate, with 6.8 km/s velocities, starting at ~15 km depth. Velocities then increase with depth to mantle Pn velocities of 7.4 km/s at 20 km depth. These 7.4 km/s velocities are interpreted as anomalous upper mantle as beneath this level passive seismic studies show similar Pn velocities that increase slowly to ~7.8 km/s at ~80 km depth. Despite the thin crust of the CVR, the region is elevated above sea level. Buoyant support for the thin crust from a mantle density anomaly of -66 kg/m$^3$ is inferred from rock uplift studies. This density anomaly, combined with Pn velocities that are 8% less than regular mantle, implies the mantle lid here may have been removed. Strong reflections from ~34 km depth in the upper mantle are recorded from an interface directly beneath the volcanically, and geothermally, active eastern side of the CVR. On the basis of reflection amplitudes, the mantle reflections are most readily explained by an interface with a 30-70% drop in S-wave speed compared to the surrounding mantle. A ~16 km wide layer of 4-9% partial melt is inferred to be the source of the reflections. These interpretations highlight the degree the CVR's crust and upper mantle have been altered by extension and volcanism. High heat output, ~13 times the continental average, indicates a significant portion of the crust in the CVR consists of intrusions in various states of cooling. Yet to the north, backarc rifting in the ocean Tonga-Kermadec arc has produced normal backarc heat flow. A key to understanding the high heat output and prolific rhyolitic volcanism of the CVR come from knowledge of the history of the large scale lithospheric evolution of the continental convergent margin, from subduction initiation, convergence and crustal thickening, to eventual backarc rifting.