Abstract - The power-law, or fractal, distribution of
fault sizes enables densities of sub-seismic faults of specified sizes
to be predicted by extrapolation of population curves of seismically imaged
faults. A variety of normal fault datasets mainly from the North Sea have
been used to establish and test procedures for measuring populations of
seismically imaged faults. The recommended fault size parameter for measurement
is the throw on a fault where it is intersected by a sample line. In most
cases multi-line sampling and plotting is necessary to obtain sufficient
data points for good characterisation of the fault population. Measurements
may be made either directly from workstation interpretations of seismic
sections or from horizon maps, and throw can be expressed in units of either
depth or TWT. A good characterisation requires a range of throw measurements
of at least 1.5 orders of magnitude. A range of clastic and carbonate sequences
has been investigated and the lithology is shown to have little effect
on the general character of population curves for seismically imaged faults,
except in the case of salt. When more than one fault set is present, as
determined by orientation, each set should be independently measured. Only
tectonic fault systems have been fully investigated and gravitational fault
systems are unlikely to have such highly ordered population distributions.
Techniques and limitations of sub-area sampling are examined. Integration
of seismic and well fault data is possible but requires more detailed fracture
logging of cores and well images than is now usual. The principal limitation
of fault population measurements concerns the difference in size between
the area or volume for which a fault population prediction is required
(e.g. for reservoir simulation studies), and that for which it can be measured.
Sub-areas that can be characterised by seismic data are often larger than
required and the converse is true of well data.
In: North Sea Oil and Gas Reservoirs III, (edited by
Aasen, J. O. et al.) Kluwer Academic Publishers, 141-155, 1994.