Print this pageStatistical 
Analysis
of Natural Fracture Networks:
Implications to Reservoir Permeability
MALMANGER, EVA M., and LAWRENCE W. TEUFEL, New Mexico Institute of Mining and Technology, Socorro, NM
Fluid flow through a naturally fractured reservoir depends on the fracture
spacing, length, aperture and morphology. Optimum characterization of these
fracture parameters requires an integrated approach using core, log, and outcrop
studies. Outcrop studies provide essential information on the spatial
distribution of fractures and fracture length. A geostatistical analysis of
fracture network maps and scanlines of sandstone outcrops in Wyoming, Colorado,
and Texas shows that fracture spacing and length can be approximated by a
log-normal distribution and that the fracture population is dominated by short
unconnected fractures. In general fracture spacing increases with increasing
fracture length. Average fracture spacing ranges from about 1 m for fractures
less than 10 m in length to 4 m for fractures of lengths of 30-40 m. The
relationship between fracture aperture and fracture length has also been
examined using three statistical methods: 1) constant aperture, 2) linear
relationship between fracture length and aperture, and 3) a
natural-log-relationship. Fracture orientation and length have been combined
with these fracture aperture/length relationships to calculate magnitude and
orientation of a permeability tensor for the same total porosity. These
simulations show that the natural log relationship gives the lowest permeability
and the linear relationship gives the highest permeability by an order of
magnitude. Simulations using only fractures with lengths greater than 20 m (less
than 10% of the fracture
population) show little reduction in bulk permeability
and suggests that the fluid flow through a fractured, tight reservoir rock is
dominated by the longest fractures, which have the widest spacing and fewest
number.