Syn-rift clastic sedimentary systems preserve a complicated stratigraphic architecture that records the interplay of tectonics, eustatic sea level and storage and routing of sediments. Previous conceptual models describe and explain changes in depositional stacking patterns along a fault segment. However, stacking patterns, and the nature of key stratigraphic surfaces, is challenging to predict accurately with conventional sequence stratigraphic models that do not consider the three-dimensional interplay of subsidence, sedimentation, and eustasy. We present a novel, geometric, 3D sequence stratigraphic model (‘Syn-Strat’), which applies temporally- and spatially-variable, fault-scale tectonic constraints to stratigraphic forward modelling, as well as allowing flexibility in the other controls in time and space. Syn-Strat generates a 3D graphical surface that represents accommodation. Although the model has the capacity to model footwall variation, here we present model results from the hangingwall of a normal fault, with temporal and spatial (dip and strike) predictions made of stacking patterns and systems tracts for a given set of controls. Sensitivity tests are tied to the depositional architecture of field-based examples from the Loreto Basin, Gulf of California and Alkyonides Basin, Gulf of Corinth. Here, the relative influence of major sedimentary controls, different subsidence histories, varying sedimentation distribution, including along-strike variation in stacking patterns, are assessed and demonstrate the potential of Syn-Strat for reducing subsurface uncertainties by resolving multiple scenarios. In addition, the model demonstrates the nature of diachroneity of key stratigraphic surfaces that can arise in syn-rift settings, which could be represented by a bypass surface (sequence boundary) or reservoir seal (maximum flooding surface) in the rock record. Enabling a quantitative assessment of these surfaces is critical for prospect analysis in hangingwall half-graben-fills, where these surfaces are heavily relied upon for well correlations that are used for hydrocarbon volume and production rate predictions.
Read full abstract