A Practical Approach to Hierarchical Marginal Marine Classification Illustrated With Modern and Ancient Examples

Abstract

Marginal marine systems show a significant degree of variability that is not described in most widely used depositional models. A careful examination of a large number of modern and ancient systems demonstrates that: (1) mixed-influence systems (Fw, Ft, Fwt, Ftw, Wf, Wt, Wft, Wtf, Tw, Tf, Twf, Tfw) are much more common in the geological record than fluvial-dominated (F), wave-dominated (W), and tide-dominated (T) end members; (2) there is often a significant process and architectural variability within the same regressive or transgressive sediment package (intra-parasequence scale); (3) there can be a continuum between different depositional environment types in the same system; (4) shoreline deposition can be affected by fluvial distributive networks formed by avulsions on the delta plain or avulsions further upstream as part of larger megafan systems, which can result in different stratigraphic architecture; (5) a system can contain different scales of discontinuities, which can be either regional or of limited lateral extent; (6) the same system can contain lobate shoreline components, linear shoreline components and transitional linear-to-lobate components. A practical way of dealing with such variability is adopting a hierarchical classification framework, where issues occurring at specific architectural scales can be addressed. Such an approach allows describing architecture deposited over hundreds of thousands to millions of years (e.g., the nature of stacking of regressive and transgressive packages), thousands to tens of thousand of years (e.g., extent of individual transgressive or regressive packages; major changes in process or sediment depocenters within such packages), tens to hundreds of years (e.g., progradation pulses and internal facies association variability within individual delta lobes), months to years (formation of individual mouth bars). The variability within each hierarchy level is then described in terms of a limited set of architectural categories that are related to process. Investigating the possible links between categories belonging to different hierarchy levels allows a component of prediction, where parent, child and sibling units can be identified. We show a number of worked modern and ancient examples that describe how the classification system can be used in practice, and how this approach can be used for prediction, uncertainty management, and compiling geospatial data and databases.

AAPG Datapages/Search and Discovery Article #90216 ©2015 AAPG Annual Convention and Exhibition, Denver, CO., May 31 - June 3, 2015