A New Twist on Sequence-Stratigraphic Correlations in Fluvial Strata: Applying the Buffer Concept to Identification and Correlation of "Stable" Low-Accommodation Intervals

Alexandrowicz, Neal; Holbrook, John

With their high degree of dimensional and temporal variability, and general lack of consistent markers, fluvial deposits are notoriously difficult to correlate. Where persistent marker beds are rare, and chronostratigraphic/biostratigraphic control is coarse, sequence stratigraphy is commonly applied. Building on the assumptive principle that the degree of channel-belt amalgamation is determined by rate of accommodation change versus sediment supply, sequence stratigraphy can be used to identify and correlate high-accommodation (mixed lithology, dispersed channel belt) and low-accommodation (sand dominated, amalgamated channel belt) systems tracts. Rather than directly correlating sandstone units within a low accommodation interval, the buffer approach identifies and correlates an interval of space (a "buffer zone") which is suggestive of protracted base-level stability. Buffer zones are defined by a set of upper and lower longitudinal profiles, influenced by upstream controls, that represent the highest surface of aggradation and the lowest depth of incision, respectively, that a river could achieve during the period of base-level stability, and can constitute intervals of one to multiple channel-belt thicknesses. The long term stability of a buffer zone will promote lateral reworking of rivers to form channel belt and/or multivalley sheets as they fill the buffer space. Stratigraphic intervals recording stable buffers are here identified by anomalously high lateral occurrence of channel sand bodies, defined by an "exceedance factor" which is the ratio by which sand occurrence in wells or sections exceeds that which would occur randomly on an aggrading floodplain. A "confidence factor" is then determined to assess the degree to which the correlated fluvial facies are indeed within this same stratigraphic interval, and is the ratio between the thickness of the buffer zone and the thickness between overlying and underlying independently determined correlation lines. The buffer approach has the advantage of correlating a zone of channel body probability instead of directly correlating lithology. Thus, it does not require similarity of sand bodies laterally, nor does it require that sand fill the entire interval or even be present to correlate the interval through a section. Though it does not identify all potential sequences and cannot provide correlations for non-stable intervals, it does provide a conceptual basis for additional correlations.

AAPG Search and Discovery Article #90163©2013AAPG 2013 Annual Convention and Exhibition, Pittsburgh, Pennsylvania, May 19-22, 2013