Relationships Between Seismic Data and Facies Proportions in Meandering Fluvial Systems

Emily Albouy, Geology-Geochemistry Department, Institut Français du Pétrole, 1 & 4, Avenue de Bois-Préau, 92852 Rueil Malmaison Cedex, France, phone: + 33 1 47 52 69 08, fax: + 33 1 47 52 70 67, emily.albouy@ifp.fr, Laure Philippe, ENSG Nancy, Nancy, France, Marie-Christine Cacas, Division Geologie-Geochimie, Institut Francais du Petrole, 1-4 avenue de Bois-Preau, Rueil-Malmaison Cedex, 92852, France, Simon Lopez, Centre de Geostatistique, Ecole des Mines de Paris, 35 rue St Honore, 77305 Fontainebleau, France, and Caroline Joseph, Geophysics Department, Institut Français du Pétrole, 1 & 4, Avenue de Bois-Préau, 92852 Rueil Malmaison Cedex, France.

Basin modelling using algorithms based on the physical laws governing sediment transport can provide the general vertical evolution of lithological proportions and depositional slope on grid sizes which are generally kilometric in scale. Deterministic or geostatistical modelling can provide a more detailed, downscaled picture of specific depositional environments respecting the vertical and horizontal proportions provided by more regional modelling and by well or outcrop control.

Stratigraphic simulations of meandering fluvial environments have been carried out in order to reproduce different depositional architectures: using a mixed deterministic-stratigraphic approach, geological processes described by the equations of fluvial hydrodynamics (meander migration, erosional downcutting) are simulated with a random component governing catastrophic events (avulsion, levee breaching…)

The lithological 3D grids which have been simulated, representing constrasting fluvial architectures and lithological compositions, have been used to create synthetic seismic data by 1D convolution. The quality of the correlation between classical seismic attributes such as amplitude and energy and sand body proportion is shown to depend both on the thickness of the observed interval and on the presence of certain secondary facies. Some textural attributes based on statistical crossplots describing the spatial correlation of seismic amplitude over a sliding window can greatly improve the success of prediction of sandbody location.

The comparison of the seismic attributes of the blocks generated by stratigraphic simulations with attributes calculated from real seismic data can allow a quantitative evaluation of the relative success of stratigraphic models in recreating sedimentary architecture in an given depositional environment.