The processes responsible for transporting and depositing thick sections of coarse-grained terrigenous clastics on the abyssal floor and for forming associated sedimentary structures are still conjectural. Many workers attribute coarse deep-sea sediments and their probable lithified equivalent, the graywackes of flysch deposits to some type of density movement. Deductions concerning the processes operating in a density flow generally are made from flume studies--in which an artificial situation may develop, or from lithified units--where the magnitude of post-depositional change is unknown. Both approaches contribute to our knowledge, but the unconsolidated elastics themselves should contain a unique key to understanding the dynamics of abyssal sedimentation. To test this theory, divisions of parallel lamination, found in deep-sea sand and silt, were selected for analysis. Since individual laminae closely approach discrete populations of particles assembled under contrasting conditions, their use carries environmental sampling to its practical limits. Northeast Pacific sediments of late Pleistocene and Holocene age, from deep-sea channel and abyssal plain environments, and representing two or three provenances were studied. A total of 115 light-colored and 84 dark-colored laminae were sampled from eight sequences at five locations. Samples averaged about 0.8 gram and were quantitatively processed using quarter-phi calibrated sieves and decantation techniques. Statistical evaluation of the procedure shows better than 95 percent sample recovery, and indicates that textural variance between laminae is significantly greater than within-sample variance. The classic concept of density transport--that coarsest material is carried by the nose of the current, and that clastic size grades tail-ward and upward in a uniformly decreasing manner--is not substantiated by moment measures, sand-silt-clay percentages or factor analysis of grain-size distributions, at least during deposition of the coarse division of parallel lamination. Coarse abyssal lamination develops within a narrow range of current velocity, the limits of which are defined texturally. Absolute velocity values for these limits can only be related, at the present time, to the few flume or in situ bottom current measurements available. Texture indicates that while the total amount of sand carried in suspension varies, lamination does not begin to form until a current is essentially depleted of all material coarser than fine sand--establishing an upper competency limit. At that time, coarse suspended material is distributed throughout the flow mostly in large eddies or vortices whose velocities are estimated on the order of about one meter/sec. Mean current velocity must be sufficient to maintain a dispersed traction carpet without deformation of bedform into ripples. This is postulated at about 50 cm/sec. A current model, based on textural evidence, is proposed to account for lamination. It is suggested that the critical stage in the formation of coarse abyssal lamination occurs while sediment is being dragged along the bottom as bedload. The flowing clastic traction carpet acquires kinetic energy as the current bypasses material lost from suspension. In turn, this energy results in grain shear. When the concentration of granular material in traction is large, it dissipates the energy of bottom shear mostly in collision contacts between gliding grains. The dispersive stresses developed tend to maintain grain separation and prevent settling. Eventually, turbulence in seawater entrapped between grains is suppressed and the net path of grans impelled by repeated collisions becomes quasi-laminar. Within this quasi-laminar traction system, dispersive pressure causes some migration of finer sizes toward the base of the carpet and a concentration of coarser grains in the upper bedload. As new material is introduced in large quantities from suspension, the zone of internal shear--the base of the moving carpet--is displaced progressively upward. As it passes, sediment compacts to a fraction of its dispersed thickness and a population of grains with a slightly finer size distribution than the carpet load comes to rest. This is buried by new deposition and a densely-packed, dark layer continues to accrete upward as long as a moving traction carpet is sustained and a dense rain of clastics is contributed from suspension. When a sand-laden eddy impinges on the bottom, it releases its coarsest load into traction and the dark layer then accreting increases significantly in grains larger than 44 microns. Any eddy, whether laden or not, on striking bottom adds to, or deducts its velocity from the velocity of the traction carpet and either increases or decreases bottom shear. Additional impulse given to tractive shear by eddies merely results in more effective size sorting. However, an eddy whose velocity of rotation is opposed to current movement may reduce shear below the critical necessary to maintain a thick carpet by dispersive pressure, The dispersed carpet collapses and instantaneously ceases moving. This less-densely packed layer has a slightly higher sand content than the accreted material below. When partially dried or weathered, alternate layers exhibit different moisture retention properties--the less-porous, accreted layers appearing dark and the more loosely packed layers appearing light. / Graduation date: 1970

Moored instrument records, drogue displacements, and hydrographic observations are used in describing the coastal currents and upwelling off Peru. The data were obtained over the continental shelf near 15°S. during a two week study in late March and early April of 1969. First order statistics and graphical representations of current meter time series indicate that the longshore flow was poleward during most of the study period, interrupted by a three day 'event' of equatorward flow. The similarity of flow at all current meters indicates that the field of flow as quasi-barotropic. The depth, extent, and transport of poleward flow indicated by current meter time series and geostrophic sections were similar to those described in the literature for the Peru-Chile Undercurrent. The observations suggest that this flow moved further offshore as equatorward flow appeared over the shelf. Power spectral analyses performed on current meter records indicate the existence of semidiurnal tidal currents in the longshore direction. The magnitude of these currents is estimated at 10% to 15% of period mean speeds. Ten meter drogue displacements are compared with 25 m recorded currents and with winds. The observations indicate that: the drogues were affected by both the 25 m flow and the wind; the depth of the wind drift layer was between 10 m and 25 m; the drogue displacements were in the sense expected from the Ekman model. Vertical sections of sigma-t, oxygen, and nitrate indicate the existence of conditions consistent with upwelling. Surface maps of temperature, nitrate, and chlorophyll 'a' over the shelf are used to define the horizontal field of upwelling and its variations in time. The distributions suggest that upwelling existed throughout the period and underwent temporal and spatial modulations in intensity. The possibility of a causal mechanism between observed current and upwelling variations is examined. Vertical salinity sections indicated the presence of a weak salinity minimum between the surface and 100 m. It is suggested that this minimum manifests the remnants of a tongue of Subantarctic Water embedded in a much larger mass of Equatorial Subsurface Water. The occurrence of the minimum only in conjunction with poleward flow suggests that the water was advected or mixed coastward somewhere north of the area studied, was entrained in the Peru-Chile Undercurrent, and was carried south again. / Graduation date: 1970

An airgun refraction line along the length of the caldera of Axial Seamount as recorded by three Ocean Bottom Seismometers has been analyzed using the tau-zeta inversion technique (Dorman and Jacobson, 1981). Five resulting velocity profiles show that the seismic velocities of the upper 1.4 km of the crust are low and similar to Mid-Atlantic Ridge upper crustal velocities. The low velocities and failure to observe shear waves are thought due to a combination of a thick section of pillow basalts and sheet flows, and a high porosity. The porosity is believed mostly due to a large fracture density. Hammond (pers. comm., 1986) has proposed the caldera is the site of overlapping spreading centers. A high fracture density in the upper crust of the caldera could be due to stresses generated in the area of overlap (Macdonald et al., 1984). No systematic variation in velocity structure along the length of the caldera could be resolved. Shallow water (about 1560 m), smooth bathymetry and the absence of sediments allowed direct measurement of the surface velocities of the caldera floor. An average surface velocity of 3.05 km/s is observed. Assuming a highly simplified crustal model, this velocity translates into a minimum porosity of 30% near the surface of the caldera floor. This minimum porosity is predicted to gradually decrease to about 0% at 1.4 km depth. Upper limits on the porosity can not be found with the available information. No magma chamber has been observed to a depth of 1.4 km. A compressional wave attenuation source in the northwest corner of the caldera below a depth of 1.4 km, however, might be a small magma body or alternatively an anomalously highly fractured area. / Graduation date: 1987

A free-air gravity anomaly map of the continental margin of Peru between 12° and 18° S. Lat. shows a -110 to -220 mgl anomaly associated with the Peru-Chile Trench, a -60 mgl anomaly over the Pisco Basin on the continental shelf, and -120 mgl anomaly over the Mollendo (or Arequipa) Basin on the upper continental slope. Anomalies observed over the continental slope and shelf consist of slope and basin anomalies superposed on a very large, broad regional anomaly. The approximately zero mgl anomaly observed in the region of the Nazca Ridge indicates the ridge is isostatically compensated. A structural model constrained by the observed gravity anomalies and seismic refraction data indicates that compensation is due to a crust approximately 8 km thicker and about 0. 04 g/cm³ less dense than the oceanic crust on either side of the Nazca Ridge. Gravity anomalies are consistent with mass distributions expected at the Peru-Chi1e Trench as a consequence of subduction of the Nazca Ridge and the Nazca Plate. Crustal and subcrustal cross sections constrained by free-air gravity anomalies, seismic refraction data, and geologic information indicate approximately 2 km of crustal thinning seaward of the trench on the southeast side of the Nazca Ridge but no crustal thinning on the northwest side of the ridge. Crustal thickness increases from approximately 10 km near the trench to about 25 to 30 km under the southwestern flank of the Andes and to approximately 70 km under the Andes. The crust is inferred to be 33 km thick under the Amazon Basin. A cross section north of the Nazca Ridge suggests a rupture of the crust at depth under the coast mountains, and earthquake hypo centers projected onto this cross section indicate a relatively shallow, nearly horizontal Benioff zone under the Andes and the Amazon Basin. A cross section south of the Nazca Ridge does not show these features, hence a different subduction process on each side of the Nazca Ridge is indicated. Free-air gravity anomalies indicate a structural high extending northwest from 17° S. Lat, along the coast, the Paracas Peninsula and nearly 100 km offshore along the edge of the continental shelf. Computations based on gravity data suggest the Pisco Basin immediately east of this structural high contains approximately 2. 2 km of sediment. A similar computation for the Mollendo Basin yields a sediment thickness of approximately 1.4 km. Gravity anomaly patterns are consistent with uplift beneath the continental shelf edge and upper slope and suggest a continental margin composed of compacted, dewatered sediments of both continental and oceanic origin. / Graduation date: 1976

In April of 1974, Oregon State University conducted a geophysical survey of the Orozco fracture zone, a Left-lateral transform fault which offsets the East Pacific Rise off the coast of Mexico near 15°N, 105°W. Magnetic, gravity, bathymetric, and seismic reflection data were collected during a four day period. This survey is combined with previous surveys by Oregon State University and other institutions to provide a geophysical interpretation of the Orozco fracture zone and the surrounding area and to develop a tectonic history of the northern Cocos plate. The Orozco fracture zone is characterized by a typical zone of seismicity and an offset in the magnetic anomaly pattern. There is, however, a conspicuous absence of a well defined topographic trough. This appears to be a result of the small age offset of the ridge crest, a reorientation of the fracture zone trend, and a possible southward migration of the fracture zone down the ridge axis. Three crustal and subcrustal cross sections over the Orozco fracture zone are constructed from the gravity data. One, across the active portion between the ridge offset, shows the active troughs to be underlain by a broad, low-density root extending two kilometers into the mantle. Two gravity cross sections across the East Pacific Rise show a thinning of oceanic layer 3 of nearly 2 kilometers at the rise crest and a corresponding 0.5 kilometer thickening of layer 2. A large magnetic anomaly of over 1300 gammas is found at the intersection of the Orozco fracture zone and the East Pacific Rise. A comparison with a very similar observation at the intersection of the Juan de Fuca ridge and the Blanco fracture zone in the northeast Pacific suggests that the East Pacific Rise is "leaking" into the fracture zone in this area. Many features have been observed on the northern Cocos plate which cannot be accounted for by present Pacific-Cocos motion the northeast strike of the eastern extension of the Orozco fracture zone, an apparent fanning of magnetic anomalies, and the northeast strike, as well as the origin, of the Tehuantepec ridge. Several possible schemes are examined to explain these observations and all but one are completely eliminated. The proposed explanation supposes a reorientation of the spreading center after a large change in the Pacific- Cocos pole of rotation resulting in the Zed pattern described by Menard and Atwater (1968). / Graduation date: 1976

University deployed two arrays of three sonobuoys over the Blanco fracture zone and two arrays of four sonobuoys over the Gorda ridge to detect and locate microearthquakes. Microearthquake activity predicted by plate tectonic theory for the Blanco fracture zone and Gorda ridge was observed by these arrays to originate from these features. Microearthquake activity of one event per hour was observed along the southern flank of the ridge associated with the Blanco fracture zone near 128°10'W. Similar seismicity was observed in Cascadia Gap and associated with the fault scarp and basin of the gap. Cascadia Gap, which is located near the center of the Blanco fracture zone, exhibits ridge-like seismic activity based on a b-value of 1.5 determined from microearthquake observations. The seismic activity on the Gorda ridge, including swarm events, averaged 3.5 events per hour. The microearthquakes originated from the median valley floor, valley walls, and on top of the crestal hills. Other events, which could not be located, appeared to originate from the surrounding hills with a predominance of events from west of the Gorda-Blanco intersection. Focal depths at the intersection area are 6.5 to 9 km below a 3.5 km datum, while those farther to the south at 42°41'N range from 2.5 to 3 km below the datum. A composite fault plane solution for the events on the Gorda ridge indicates high angle faulting on the eastern valley wall with the eastern side down. Since microearthquake and earthquake activity are usually closely associated, the data obtained by this study suggests that the 20-30 km eastward offset of large earthquake epicenters from the topographic features is due to an incorrect use of excessively high lithospheric velocities for seismic stations lying predominantly to the east. / Graduation date: 1976

A deep source-receiver seismic refraction experiment was conducted on the upper part of the Monterey Deep Sea Fan. The aim of this thesis is to construct the velocity structure of the upper Monterey Fan and to examine the lateral seismic velocity variations among the upper, middle and lower fan subunits. Using primary waves and whispering gallery phases (the multiply-reflected refraction waves), the sediment velocity structure was modeled by the tau-zeta travel time inversion process. The changes in velocity gradients with depth of the upper Monterey Fan are morphologically similar to that found on both the Central Bengal Fan and the Nicobar Fan, an abandoned lower fan of the Bengal Fan Complex. The velocity gradient of the upper Monterey Fan at depth, 0.59 s⁻¹ is significantly lower than both the middle Bengal Fan (0.68 s⁻¹) and the Nicobar Fan (0.81 s⁻¹). The upper fan subunit, which is closer to its sediment source, is characterized by higher porosities caused primarily by a higher sedimentation rate than the lower fan subunits. Since seismic velocity is inversely related to porosity, the upper fan subunit should have lower velocity gradients and seismic velocities than the other fan subunits. If porosity and velocity variations exist, then these variations can be used to constrain various models of deep sea fan formation. No definite conclusion can be drawn at this time due to a fault within 1 km of the Nicobar Fan site; however, a systematic velocity variation pattern of deep sea fans is revealed. Some portions of the Monterey Fan data contain refracted waves which have bottomed within the underlying acoustic basement structure. The entire velocity structure was solved by both the general and the "stripping" solving schemes. The results of basement structure show a velocity ranging from 3.4 to 5.8 km/s indicating that the uppermost part may be pre-existing continental rise sediments. / Graduation date: 1986

Graduation date: 1962

Graduation date: 1963

Examination of polished sections of manganese micronodules from metalliferous sediments from the Bauer Deep reveals sequences of ferromanganese deposition which are consistent for micronodules from a single sediment sample and imply a common depositional history for members of such assemblages. The relatively simple 'stratigraphy' of manganese micronodules, as compared to macronodules, makes stratigraphic correlation of depositional histories easier and more conclusive for micronodules than macronodules. Quantitative microprobe analysis and X-ray mircodiffraction patterns indicate that the major authigenic silicate component of metalliferous sediment is an iron- and magnesium-rich, low-aluminum nontronite. Microprobe and X-ray microdiffraction analyses of manganese micronodules establish todorokite as the predominant crystalline component. Comparison of the distributions of lanthanum, cerium, and samarium in micronodules and phosphatic fish debris indicates that bulk sediment REE distribution is determined by the relative abundances of cerium-enriched micronodules and cerium-depleted biogenic phosphate. / Graduation date: 1974