Print this page
Click
to view presentation in PDF format.
Detecting
Vertical Hydrocarbon Expulsion from Source 
Rock
in Seismic Data*
David L. Connolly1
Search and Discovery Article #40331 (2008)
Posted October 15, 2008
*Adapted from oral presentation at AAPG Annual Convention, San Antonio, TX, April 20-23, 2008
1dGB USA, Sugar Land, TX. ([email protected])
Modeling petroleum systems accurately in a basin requires a good
understanding of the areal distribution of source rocks, the stratigraphic
intervals in which source rocks occur, and their thermal maturity. These models
have depended on assumptions about source rock presence based on regional
outcrop information or limited well information because wells are rarely
drilled in the hydrocarbon kitchen lows. These models also must be based on
assumptions about regional heat flow which have a high degree of uncertainty.
Direct detection of vertical hydrocarbon expulsion from source rock in seismic
data is a promising approach to constrain geologic models. The methodology
involves detecting vertically aligned low-energy chaotic zones in the seismic
data (gas chimneys) through a supervised, multi-attribute, neural network
approach. By determining the origin of these chimneys, we can infer both the
stratigraphic interval from which the hydrocarbons originate, the distribution
of source prone facies, and the distribution of thermally mature source rock.
The chimney information can also be used to show migration pathways into
potential reservoir intervals and possible leakage via faults or top seal
failure from these reservoirs. This information can then be input into the
basin model to constrain model inputs. Similarly basin models can constrain the
results of chimney processing. Case studies are shown from the North Sea, South
Atlantic, and Gulf of Mexico.
