Estimating Mature Plays
39
distribution. The Poisson distribution was used as the prior distributions for these examples. BDSCV estimates a posterior distribution
based on the input parameters and the discovery sequence for each
case. The statistical treatment of the BDSCV model is explained in
Appendix A.
The Keg River Shelf
The Rainbow reef play includes all oil trapped in the Keg River pinnacle reefs and the bank-margin reef buildups that accumulated in small
deep basins of the Western Canada Sedimentary Basin (Reinson et al.,
1993). One hundred sixty-one oil pools have been discovered (Fig. 3.7).
The total number of oil pools in this play estimated by NDSCV is 320.
After 83 iterations, Figure 3.8 shows the posterior number-of-pools distribution. The expected value of the distribution is 330, and the following probability statement can be made:
P (304 ≤ N ≤ 354) = 0.9
The range derived by BDSCV includes the point estimate obtained
by NDSCV, but BDSCV presents a range of N values, which provides
more information for petroleum resource assessments.
Remarks
BDSCV has been applied to more than 150 oil and gas plays (Table 3.2).
In all cases, the posterior distributions cover the point estimates derived
by LDSCV or NDSCV, except for a few cases in which LDSCV and
NDSCV do not yield a defi nite answer.
The statistical method of BDSCV is an extension of PETRIMES
methodology within the superpopulation framework. Use of BDSCV
is made by entering the output from LDSCV or NDSCV to estimate the
posterior distribution. On the other hand, geologists can construct a
number-of-pools distribution based on the information about the number of prospects and the exploration risk, as shown by Lee and Wang
(1983a). The prior number-of-pools distribution (either the Poisson distribution or one constructed by geologists) can be evaluated based on
the discovery sequence, and then the posterior number-of-pools distribution can be determined. The assumption of the Poisson distribution
has not yet been verifi ed, but BDSCV can still assist petroleum resource
assessors by using a distribution for the N value instead of a single
39
distribution. The Poisson distribution was used as the prior distributions for these examples. BDSCV estimates a posterior distribution
based on the input parameters and the discovery sequence for each
case. The statistical treatment of the BDSCV model is explained in
Appendix A.
The Keg River Shelf
The Rainbow reef play includes all oil trapped in the Keg River pinnacle reefs and the bank-margin reef buildups that accumulated in small
deep basins of the Western Canada Sedimentary Basin (Reinson et al.,
1993). One hundred sixty-one oil pools have been discovered (Fig. 3.7).
The total number of oil pools in this play estimated by NDSCV is 320.
After 83 iterations, Figure 3.8 shows the posterior number-of-pools distribution. The expected value of the distribution is 330, and the following probability statement can be made:
P (304 ≤ N ≤ 354) = 0.9
The range derived by BDSCV includes the point estimate obtained
by NDSCV, but BDSCV presents a range of N values, which provides
more information for petroleum resource assessments.
Remarks
BDSCV has been applied to more than 150 oil and gas plays (Table 3.2).
In all cases, the posterior distributions cover the point estimates derived
by LDSCV or NDSCV, except for a few cases in which LDSCV and
NDSCV do not yield a defi nite answer.
The statistical method of BDSCV is an extension of PETRIMES
methodology within the superpopulation framework. Use of BDSCV
is made by entering the output from LDSCV or NDSCV to estimate the
posterior distribution. On the other hand, geologists can construct a
number-of-pools distribution based on the information about the number of prospects and the exploration risk, as shown by Lee and Wang
(1983a). The prior number-of-pools distribution (either the Poisson distribution or one constructed by geologists) can be evaluated based on
the discovery sequence, and then the posterior number-of-pools distribution can be determined. The assumption of the Poisson distribution
has not yet been verifi ed, but BDSCV can still assist petroleum resource
assessors by using a distribution for the N value instead of a single