Deep-marine tidal bottom currents and their reworked sands in modern and ancient submarine canyons

Submarine canyons provide a unique setting for tidal processes to operate from shallow-marine to deep-marine environments. In modern canyons, current-meter measurements at varying water depths (46-4200 m) show a close correlation between the timing of up- and down-canyon currents and the timing of semi-diurnal tides. These tidal bottom currents in submarine canyons commonly attain maximum velocities of 25-50 cm/s. Based on core and outcrop studies of modem and ancient deep-marine deposits, it is proposed here that sand-mud rhythmites, double mud layers, climbing ripples, mud-draped ripples, alternation of parallel and cross-laminae, sigmoidal cross-bedding with mud drapes, internal erosional surfaces, lenticular bedding, and flaser bedding can be used to interpret deposits of deep-marine tidal cur-rents. This approach is an alternative to the conventional approach in which most deep-water traction structures (e.g. climbing ripples and cross-bedding) would be attributed to deposition from turbidity currents. Underwater photographs show active mass flows (i.e. slides, slumps, grain flows, and debris flows) in modern canyons. Box cores taken from modem submarine canyons (e.g. La Jolla, California) and conventional cores and outcrops of ancient canyon-fill facies (Oua lboe, Pliocene, Nigeria and the Annot Sandstone, Eocene-Oligocene, SE France) contain deposits of both tidal processes and mass flows. This facies association in the rock record can be used as a criterion for recognizing submarine canyon settings. In a channel-mouth environment, deep-marine tidal deposits are likely to develop elongate bars that are aligned parallel to the channel axis within the channel, whereas turbidites are more likely to develop depositional lobes that are aligned perpendicular to channel axis. Turbidite depositional lobes are much larger than the channel width, whereas tidal sand bars are much smaller than the channel width. Therefore, the wrong use of a turbidite lobe model with sheet geometry in lieu of a tidal bar model with bar geometry will result in an unrealistic overestimation of sandstone reservoirs in deep-water exploration. (C) 2003 Elsevier Ltd. All rights reserved.