Page 207 - Geology of Carbonate Reservoirs

188 FRACTURED RESERVOIRS

Nelson (2001) , fracture widths of “ crystalline carbonates ” and chalks at simulated
− 1
− 6
− 1
depths of 10,000 feet are 10 and 10 to 10 cm, respectively. If k r and k f are known

at a given confining pressure, the earlier equations will provide results that are more
realistic for subsurface reservoir conditions. Fracture spacing, D , also infl uences
fracture permeability; however, determining reliable values for D is not easy, espe-
cially in the subsurface where observations are usually limited to core samples.

Nelson (2001) uses the definition of Parsons (1966) for fracture spacing; that is, “ the

average distance between regularly spaced fractures measured perpendicular to a
parallel set of fractures of a given orientation. ” Fracture spacing can be calculated

with statistical and geometrical methods from measurements made on cores or
outcrops, but those methods are beyond the scope of this book.
7.2.2 Fracture Porosity
Calculating fracture porosity is similar to calculating matrix porosity in that φ f is the
percentage of total rock volume made up of fracture pores. Whereas matrix porosity

is represented by φ r = ( V p / V t ) × 100, fracture porosity is expressed in terms of frac-
ture width and spacing, or φ f = [ e /( D + e )] × 100
where φ r = Matrix porosity in percent
V p = Pore volume in intact (matrix) rock

V t = Total rock volume
φ f = Fracture porosity
e = Average effective fracture width
D = Average spacing between parallel fractures

Note that φ f for a constant e varies as a function of distance between fractures,
indicating that fracture porosity is scale dependent . Matrix porosity is not. If frac-
2
ture spacing is measured over a small area (e.g., cm ) and only one fracture is
counted, the resulting fracture porosity value will be high because the proportion
of porosity contributed by the one fracture in a small area is a large part of the
total porosity. If the measurement area is increased in size, matrix porosity remains
fixed, and more fractures are counted in the larger area, then φ f will be propor-

tionately smaller because the contribution of matrix porosity is greater over the
larger area and the proportion of fracture porosity to matrix porosity becomes
smaller. For this reason, it is important to utilize a large enough measurement area
to encounter several fractures. It also means that obtaining statistically valid sam-

pling is more difficult when it is not possible to measure fracture spacing directly
as on outcrops.
Measurement of fracture porosity also requires larger samples. Conventional
1 - inch diameter “ perm plugs ” that are used for routine core analyses do not have
enough volume to adequately sample both matrix and fracture porosity compo-
nents; therefore full - diameter core segments are necessary. It is also important to
have experienced geologists examine core samples before measuring porosity in
order to distinguish between induced fractures and natural fractures. And it is
helpful to identify which natural fracture sets are likely to be the principal contribu-
tors to fracture porosity, especially if several fracture sets are present but only one