Clastic reservoir analogs based on 2-D outcrop studies provide valuable definitions of geometric and petrophysical heterogeneities at interwell scales. Integration of 3-D ground-penetrating radar (GPR) surveys with sedimentological and stratigraphic data provides information on the internal heterogeneities of sedimentary sequences at scales that allow dissection of the 3-D anatomy of clastic depositional systems. Two 3-D GPR data volumes were acquired in the Ferron sandstone of east-central Utah. The data show prominent lenticular features, a variety of lithologies, and structural elements such as channels and shale drapes that match well with those seen at the same stratigraphic levels in adjacent cliff faces.

No abstract

The most extensive Jurassic marine transgression in North America reached its maximum limits during the Oxfordian Age. At this time, siliciclastic sediments were being brought into the North American seaway from an uplifted zone to the west. Within this setting, complexes of sand ridges and coquinoid sands layers were deposited. Coquinoid sandstones appear to fill erosional scours and were interpreted as channel fills. Re‐evaluation of these features in the light of recently discovered attributes of modern shelf sediments and processes has produced a revised model of coquinoid sand deposition in this setting. Coquinoid sandstones which fill ‘channel‐like’ scours in the Oxfordian (Upper Jurassic) rocks of central Wyoming and south‐central Montana, appear to have formed through the migration of sand waves across the crests of inner shelf sand ridges during periods of storm and tidal flow. Erosion in the zone of flow reattachment in the troughs between sand waves resulted in the development of shell lags. Migration of these scour zones as the sand waves advanced resulted in the deposition of sheet‐like coquinoid sandstone bodies. Sand waves crossing the ridge crest tended to migrate more slowly and to be overstepped by later sand waves. Sand wave troughs thus buried have channel‐like geometries with apparent epsilon bedding.

In recent years, a new model for deposition of sand bodies in a shelf environment has appeared. This model, known as the shelf sand‐plume model, is hypothesized to result from storm‐driven currents that are deflected around a deltaic headland, stripping sand from the headland and redepositing it in a downcurrent ‘plume’ on the inner shelf. The modern analogue for this model is considered to be an arcuate shelf sand body located off the Damietta branch of the Nile Delta. However, the distribution of older deltaic and shoreline sands probably controls the arcuate outline of the sand body. The present current system has certainly reworked these sands into ridges and large‐scale bedforms but is not responsible for the overall outline of the sand body. Grain‐size range and distribution of sand on the shelf demonstrate that the source of sand in the Nile shelf sand body is not the modern Damietta headland as postulated by the shelf sandplume model. In our view, the shelf sand‐plume model is presently unsubstantiated and has orginated as a misapplication of the original Nile example. As a geological model, the shelf sand‐plume model lacks a set of observable, consistently applicable criteria. The only common denominator to the model is the ‘plume’ geometry of a sand body located off a deltaic promontory. However, workers postulating the existence of shelf sand‐plumes have neither clearly established a ‘plume’ geometry nor shown the juxtaposition of these bodies with respect to coeval deltaic headlands in their outcrop or subsurface examples. The model does not provide criteria to distinguish a ‘shelf sand‐plume’ from other classes of shelf sand bodies, notably sand ridges and storm‐generated sheet‐like sands. Its application to the rock record should be re‐evaluated.

This paper was prepared for presentation at the 1998 SPE Asia Pacific Conference on Integrated Modelling for Asset Management held in Kuala Lumpur, Malaysia, 23-24 March 1998.