Geomorphologic Delineation of Deepwater Hydrocarbon Reservoirs with Pre-Stack Seismic Inversion

Contreras, Arturo J.¹, Torres-Verdin Carlos², William L. Fisher², William E. Galloway² (1) Chevron ETC, Houston, TX (2) The University of Texas at Austin, Austin, TX

This paper describes the successful application of a new inversion methodology to quantitatively integrate pre-stack seismic data, well logs, geologic data, and geostatistical information for the spatial delineation of deepwater Miocene reservoirs in the Marco Polo Field, located in the central Gulf of Mexico.

Inversion results clearly evidence the existence of lobate geo-bodies associated with prospective reservoir areas (M-series). Deterministic inversion results show high-porosity geo-bodies detected from the P-impedance model, and low “LambdaRho” anomalies characteristic of hydrocarbon-bearing sands. Similarly, lithotype distributions obtained from stochastic inversion confirm the presence of lobate-shaped sand bodies.

Furthermore, co-simulated 3D models of petrophysical properties consistently reveal the existence of internal heterogeneities (minor internal lobes) within the main lobes.

We use the channel/lobe depositional model introduced by Galloway (1998), which describes similar lobate-shaped deposits as the characteristic geomorphic feature of these base-of-slope sediments. Furthermore, the model predicts presence of internal heterogeneities (small lobes) within the main lobe, which is clearly confirmed by the petrophysical models. In addition, the gamma-ray log-motif associated with this model clearly matches the logs recorded in the study area.

The geometry and relatively small size of the geo-bodies obtained from inversion results, suggest that the M-series reservoir sands can be described as “mounded” lobes: sandy turbidite deposits, with low-moderate channelization, and small areal extent (~ 1 km²). These reservoirs have been finally interpreted as stacked terminal turbidite lobes within an overall fan complex; the interpretation is consistent with previous core-data interpretations and regional stratigraphic/depositional studies.