Use of Experimental Design for Uncertainty Assessment in Deep Offshore Gulf of Mexico

Abstract

The lower tertiary Wilcox trend, which extends from Alaminos Canyon(AC) to Keathley Canyon (KC), is attracting more attention from the oil&gas industry (Vassilellis, et al., 2013) and more production and final investment decision (FID) are expected in the near future according to WoodMac latest reports. The range of reserves of the lower tertiary Wilcox trend is between 3 and 15 billion barrels (Dusterhoft, Strobel, & Szatny, 2012). Buckskin field is among the fields located in this trend at approximately 290 miles south-east of Houston (Ait-Ettajer, 2017). The Buckskin discovery well (yacimiento_golfo_mexico.aspx, 2009) and the appraisal wells drilled afterwards, encountered more than 1,000 foot hydrocarbon column between 25,000 and 27,000 feet depth and 6000 feet of water, in sand bearing turbidite, located in different compartments bounded by a complex fault network. Despite, the subsurface studies performed following the drilling campaign, significant uncertainties remained. Several approaches were assessed in order to study the impact of those uncertainties on the hydrocarbon storage and mobility of this field. Probabilistic methods based on experimental design and Monte-Carlo simulations, were chosen in order to optimize the development plan and the reserves assessment, due to their robustness and their flexibility. Those probabilistic methods, combined with experimental design technique, allowed the identification of the key uncertainties and the definition of response model that predicts the range of oil in place and the optimization of the development plans, without the need to run hundreds of thousands realizations.