--> Introduction of a Rock Typing Methodology in Crystalline Basement Reservoirs (Yemen), Steckhan, Jan; Sauer, Roman, #90100 (2009)

Datapages, Inc.Print this page

Introduction of a Rock Typing Methodology in Crystalline Basement Reservoirs (Yemen)

Steckhan, Jan1
 Sauer, Roman1

1OMV E&P, Vienna, Austria.

It is essential to investigate the petrographical composition of a reservoir particularly with regard to the response of the relevant rock types to the borehole logs before petrophysical properties can be evaluated accurately. This applies in particular to basement reservoirs since their lithological content is very much different from typical sedimentary reservoirs. The mineralogical composition of crystalline rocks in the Yemen Habban basement containing metamorphic, volcanic as well as igneous rocks is very variable and can hardly be correlated in between wells. Consequently the reservoir delineation based on standard techniques was not successful and a specific rock typing approach had to be invented.

Since the variety of different rock-building minerals was to big to be handled by a standard multi-mineral model and responses of conventional logging measurements have not been characterized to most of the minerals present in the basement it was decided to rather run a rock type model than a mineral model. This rock type model has been based on a supervised neural network which was trained on the petrographical analysis of cores, sidewall-cores and cuttings.

The lithological inventory has been grouped into five dominant rock types which proved to be most representative for the reservoir. Founded on the neural network model these rock types could be identified successfully from the logging data in subsequent wells. This way the key petrophysical parameter could be evaluated and linked to the basement lithology in a very cost efficient way since only conventional logging data has been used for this rock typing approach.

AAPG Search and Discover Article #90100©2009 AAPG International Conference and Exhibition 15-18 November 2009, Rio de Janeiro, Brazil