Estimating Mature Plays
37
when the discovery sequence is available. BDSCV provides a probability statement about the N value and a probability measure for each
undiscovered individual pool size.
There are four methods for estimating the number of pools or the
number-of-pools distribution for a play. The fi rst method is as follows:
Geophysicists and geologists obtain the number of prospects from
structural contour maps based on seismic and geological information. The information about the number of prospects is used to construct a number-of-prospects distribution (Lee and Wang, 1990). The
number-of-prospects distribution and the exploration risk (Lee et al.,
1989) are used to derive the number-of-pools distribution (Lee, 1993d;
Lee and Wang, 1983b).
The second method includes the maximum-likelihood methods of
LDSCV and NDSCV, as discussed in the previous sections. However,
both LDSCV and NDSCV can only provide point estimates about
the value of N. Furthermore, the log-likelihood functions of these two
methods occasionally show a fl at profi le about the value of N and yield
a range of N without probability measures.
The third method (Gordon, 1983) is as follows: A successive sample from a fi nite population is divided into two parts to approximate
10
Ϫ8 10
Ϫ7 10
Ϫ6 .00001 .0001 .001
Pool size, 10 6 m 3
Cumulative frequency greater than, %
.01 .1
1
10 100 1000
10
Ϫ9
0
10
20
30
40
50
60
70
80
90
100
A
B
C
Figure 3.6. Pool-size distribution for the Beaverhill Lake play. Line A is derived
by LDSCV, line B is derived by NDSCV, and line C is derived by random
sampling.