Characterization of heterolithic deposits using electrofacies analysis in the tide-dominated Lower Jurassic Cook Formation (Gullfaks Field, offshore Norway)

Citation
R. Gupta et Hd. Johnson, Characterization of heterolithic deposits using electrofacies analysis in the tide-dominated Lower Jurassic Cook Formation (Gullfaks Field, offshore Norway), PETR GEOSCI, 7(3), 2001, pp. 321-330
Citations number
27
Categorie Soggetti
Earth Sciences","Geological Petroleum & Minig Engineering
Journal title
PETROLEUM GEOSCIENCE
ISSN journal
13540793 → ACNP
Volume
7
Issue
3
Year of publication
2001
Pages
321 - 330
Database
ISI
SICI code
1354-0793(200108)7:3<321:COHDUE>2.0.ZU;2-F
Abstract
A range of heterolithic facies, comprising thinly interbedded (millimetre-d ecimetre scale) sandstones and mudstones, characterizes the heterogeneous L ower Jurassic Cook Formation, including the productive Cook-3 reservoir in the Gullfaks Field. These heterolithic facies were deposited in a tide-domi nated estuarine to deltaic setting and show up as massive 'shaly-sands' on conventional wireline logs. This makes the recognition and discrimination o f different heterolithic facies types virtually impossible, which severely limits detailed reservoir geological and petrophysical predictions. This pr oblem has been addressed by undertaking a high resolution electrofacies ana lysis using core facies as a 'training set' and applying this, through mult i-variate statistical techniques, to the interpretation of the conventional logs. An electrofacies database was created comprising five genetically linked ro ck types (ranging from lenticular-wavy bedding, through flaser bedding and into clean/ massive sandstones). This electrofacies scheme was validated wi th reference to c. 125 m of cored section from five wells using gamma-ray, density and neutron logs. Multivariate statistical techniques have enabled probabilistic discrimination of the different types of heterolithic facies down to intervals of only 0.25 to 0.5 m thick, which is considerably greate r than could be achieved using conventional well-log evaluation techniques alone.