Search

Western Mexico's second largest river apparently built its delta across the narrow continental shelf so that the river pours sediment directly onto the continental slope. As in other cases of shelf encroachment, the slope is characterized by numerous valleys, of which one with canyon-sized dimensions can be traced seaward to the Mid-America Trench. Evidence for turbidity currents was established by current-meter records and observations from scuba and vehicle diving. A dense grid of accurately located sounding and seismic lines shows the nature of deposition seaward of this delta, but in a zone off both sides of the main river mouth, seismic profiles of the slope suggest the presence of basement rock. However, it seems somewhat more likely that, despite some rock outcrops where erosion has exceeded deposition in Necesidad Canyon, the slopes are underlain mostly by an appreciable thickness of sediments that have been homogenized by extensive slumping which has destroyed any internal reflectors. Farther seaward from this zone, where slopes are gentler, there is a broad area in which good reflectors show clear evidence of natural levees, cross-bedding, valley fills, and other signs of slope progradation. Despite all the evidence of progradation of the slope beyond the delta, we have clear indications that much sediment is being bypassed along the two main canyons of the area. Turbidity currents partly initiated by submarine landslides are eroding locally the steep canyon heads and transporting a large amount of sediment toward the Mid-America Trench. These flows are powerful enough to move coarse sediment and coarse plant debris, and to leave large flow ripples along their broad, flat floors.

Side-scan sonar records along the moderate- to high-energy Pacific Coast of Mexico show ripple-marked bands normal to the surf. These bands have been traced as far seaward as 1,500 m from the beach in water as deep as 30 m. There is strong evidence that these ripple bands are produced by rip currents. Earlier studies suggested that rip currents do not extend to the bottom beyond the breaker zone. Our interpretations indicate that rip currents—probably those that develop under severe wave conditions—transport sediment, shape the bottom, and produce internal sedimentary structures much farther seaward than was previously believed.

Certain shelf areas have been resurveyed with side-scanning sonar, fathometer, and accurate navigational control for studies of the rate of ice gouging and sediment reworking by ice in the Beaufort Sea, Alaska. Analysis of one such survey, initially done in 1973 and rerun in 1975 along 14 km of track-line in 6 to 14 m of water depth, suggests complete reworking of the bottom to an average depth of 20 cm in 50 yr. Only 3 cm of sediment accumulated during this time interval, making the rate of reworking much higher than rate of deposition.

Barrier islands in arctic Alaska are composed of sandy gravel, whereas lower-latitude barrier islands are composed of predominantly sandy material. This pronounced grain-size difference needs explanation. Because wind-winnowing of island surfaces is effective during most of the yearly cycle, while sediment supply by waves is effective for about 2 mo at most, eolian sand deflation might be a mechanism to explain the coarse sediments of arctic barrier islands. In late winter, plumes of windblown sand are found on the fast ice downwind of barrier islands. More than 300 t (tonnes) of sand were removed by wind action from one small island near Prudhoe Bay in one winter; this loss suggests a deflation rate of 2 mm for that period.

A previously unreported process was observed at the head of a tributary to the Rio Balsas submarine canyon system in Mexico. During a period of large surf, river discharge deflected a pulsating longshore current [peaking at over 7 km/hr (2 m/sec)] seaward over the tributary heading in the surf zone. This pulsating flow occasionally entered the river mouth, causing rhythmic fluctuations with amplitudes of at least 30 cm and a period of about 3 minutes within the mouth, as recorded by a partially filtered tide gage. In diving to the bottom of the tributary at a depth of 18 m, we encountered current pulses with estimated velocities of 4 km/hr (more than 1 m/sec) transporting large amounts of suspended sand down an axial slope of 26°. This bottom flow was at least 3 m thick, and was characterized by pulses separated by quiet periods in phase with the surface rip current. The upper 20 cm of the canyon fill during this time consisted of sand smoothly laminated parallel to the bottom, indicating net deposition on the steeply sloping floor. Estimated water budget suggested that the entire water column below the rip current at 18 m was not flowing seaward during the pulses. The bottom flow landward of the dive site probably separated from the surface flow and was propagated downslope as a turbidity current. Its magnitude was sufficient t o erode the canyon walls. These observations substantiate that rip currents play a role in the formation of some submarine canyons. Surf-beat induced rhythmic flushing of the river mouth, however, did not cause density currents in the main canyon head. The need for future canyon studies under extreme conditions is pointed out.

A series of seismic reflection profiles was made during the summer of 1964 in the northeastern part of Prince William Sound, Alaska. Three rock units recognized in the profiles are (1) the ice-sculptured metamorphic basement, overlain by (2) probable glacial drift, and (3) Holocene marine sediments. Channels in the basement may have been formed in part by faulting as well as by glacial scour; they have been extensively modified by fill of rock units (2) and (3), so that the present sea floor does not closely represent the effects of glaciation. The Holocene strata have been faulted, but the vertical separation is small, suggesting that sound reacts to regional tectonism without severe Recent faulting.

Reconnaissance observations were made along the Beaufort Sea coast of Alaska in 1970 during a 4-week period after initiation of river flow into the ice-covered ocean. During the first few days following river breakup, the fast ice was inundated by fresh-water overflow for as far as 10 km from shore. The water drained through widely distributed drain holes (strudel) in the ice creating scour depressions in the sea bottom. Later the rivers flowed directly into growing open-water areas surrounding distributary mouths and thence seaward below the fast ice. The presence of the sea ice during river flood greatly restricts the cross-sectional area available for the seaward flow out to at least the 2-m contour, where the bottom slope suddenly steepens. At the Colville Delta, the 2-m bench extends 5 to 8 km from shore and presumably represents an area of sediment bypassing. It appears that only insignificant amounts of sediment are rafted away from arctic Alaska9s deltas during the final breakup of the sea ice.