THREE DIMENSIONALITY OF LOWER MIOCENE SEQUENCES, US MID-ATLANTIC MARGIN

The US Mid-Atlantic region continues to be intensely studied as a natural laboratory of passive margin development. Several generations of seismic data and boreholes have supplied data with which to evaluate the margin's growth. Previous studies suggested uniform progradation of Oligocene sequences but marked along-strike variability of middle-upper Miocene sequences. We show that along strike variability is evident in the early Miocene as well, and that it is due to increased siliciclastic sediment supply distributed to the shelf by isolated deltaic systems. Recently collected high-resolution multichannel seismic data (short-offset, shallowly towed GI airgun yielding ~5 m vertical resolution) clearly image 11 lower Miocene sequences defined by erosional truncation, marine onlap, and downlap. However, no single dip line displays a clinoform inflection point for each of these 11 units, and geometric features across each surface vary throughout the seismic grid (166 km along strike, 5-45 km offshore, 8 to 40 m water depth). For example, sequence boundaries in various dip-line profiles lose their characteristic sigmoidal shape, form more gently seaward-dipping surfaces, and are truncated by and/or concatenated with over/underlying boundaries in both updip and downdip directions; the corresponding sequences show dramatic changes in thickness along strike. Isopach maps indicate that at least two different point sources created regionally separated wave-dominated deltas. Sequences in the northern half of the grid are generally thicker, have better developed lowstand deposits, prograde shorter distances across the shelf, and contain more extensive erosional features than they do in the southern half. We conclude that these contrasts in sequence stratigraphic architecture are due to north-south variations in sediment supply and subsidence. Thus, the number and significance of sequences detected within a single sedimentary basin may depend on which profile is analyzed, and to minimize errors in extracting a record of eustatic change, an extensive along-strike seismic grid is clearly advisable.