Volumetric Uncertainty
Learning objectives
- Treat each volumetric input as a distribution, not a number
- Run a Monte Carlo to get the STOIIP distribution
- Read P90, P50, and P10 and the P10/P90 ratio
- Explain why the volume distribution skews to the high side
A Number Becomes a Distribution
Every term in the volumetric equation is an estimate with a range. The area comes from a contour map, the net thickness from a cutoff, the porosity and saturation from logs with their own error bars. Feed each as a distribution and the STOIIP is no longer a number but a distribution of its own, found by Monte Carlo: draw one value from each input, compute the oil in place, and repeat thousands of times.
P90, P50, P10
The result is reported as three numbers, the percentiles of one distribution rather than three hand-picked scenarios. The P90 is the low case, the value the true volume exceeds nine times in ten; the P50 is the middle; the P10 is the high case. An asset is booked against these, with the P90 the conservative number a lender will trust and the P10 the upside the explorationist dreams of.
Why It Skews
Because the equation multiplies its inputs, their uncertainties compound and the distribution skews, with a long tail to the high side. The single honest measure of how well the prospect is known is the ratio of the P10 to the P90: near one, the volume is pinned down and the appraisal is mature; three or more, it is a wildcat whose value will not be known until it is drilled. Widening the input uncertainty fans the whole spread out.
References
- Capen, E. C. (1976). The difficulty of assessing uncertainty. Journal of Petroleum Technology, 28(8).
- Society of Petroleum Engineers (2018). Petroleum Resources Management System (PRMS).