Subsurface Volumetric Estimation & Probabilistic Risk Quantification
Determining the volume of hydrocarbons trapped within an oil or natural gas reservoir is the foundation of field appraisal, commercial sanction (Final Investment Decision / FID), and reserve auditing. In early exploration and appraisal phases, geological parameters—including structural trap area, net reservoir thickness, effective porosity, fluid saturation, and recovery efficiency—carry wide ranges of uncertainty.
Single-point "deterministic" calculations often produce misleading impressions of precision. Modern petroleum evaluation governed by the Petroleum Resources Management System (SPE-PRMS) mandates probabilistic analysis using Monte Carlo simulation to establish statistical distribution percentiles.
1. Volumetric Formulations: STOIIP and OGIP
The volumetric equation calculates the geometric volume of hydrocarbon-bearing pore space at subsurface reservoir pressure and temperature, subsequently dividing by the formation volume factor to convert to standard surface stock-tank conditions:
7,758 is the conversion constant from acre-feet to barrels (43,560 ft²/acre / 5.6146 ft³/bbl). B_oi is initial oil formation volume factor (rb/stb).
43,560 converts acre-feet to cubic feet. B_gi is initial gas formation volume factor (rcf/scf).
2. The Petroleum Exceedance Convention (P90 < P50 < P10)
In general statistics, percentiles are frequently indexed in ascending order (where the 90th percentile exceeds the 10th percentile). However, international petroleum engineering strictly observes the Exceedance Convention:
- P90 (1P Proved): There must be at least a 90% probability that the actual quantity recovered will equal or exceed the estimate. This represents the high-confidence, conservative baseline for banking finance.
- P50 (2P Proved + Probable): There must be at least a 50% probability that the actual quantity will equal or exceed the estimate. This is the median / "Best Estimate" utilized for project budgeting.
- P10 (3P Proved + Probable + Possible): There is a 10% probability that the actual quantity recovered will equal or exceed the estimate. This reflects optimistic upside potential.
Consequently, under the SPE-PRMS exceedance framework, P90 < P50 < P10 is universally maintained.
Frequently Asked Questions (Answer-First)
Why is P90 smaller than P10 in petroleum reserves?
Petroleum reserves follow an exceedance convention. P90 represents the volume that has a 90% probability of being equaled or exceeded, making it conservative and small. P10 represents an upside case that has only a 10% probability of being equaled or exceeded, making it large.
What distribution should be used when reservoir data is sparse?
The Triangular or Modified PERT distribution is recommended during exploration when only minimum, most likely (mode), and maximum estimates are available from seismic and analog wells. Lognormal distributions are standard for net pay thickness and permeability.
Why does the product of average values differ from the Monte Carlo mean?
Subsurface parameters are multiplied together non-linearly. In probability theory, the expectation of a product of random variables equals the product of their expectations only if perfectly independent. Deterministic multiplication of modal numbers almost always underestimates or distorts the true expected mean.