ModelRisk example · Oil and gas
Volumetric Resource Estimate
How much gas and condensate could the Falcon discovery hold? P90, P50 and P10 resources from a volumetric calculation.
From the model’s saved run
Resource range
P90 = low estimate (90% chance of being exceeded), P10 = high estimate.
| Measure | Unit | P90 (low) | P50 | P10 (high) | Mean |
|---|---|---|---|---|---|
| Gas initially in place | Bcf | 165.5 | 280.2 | 478.3 | 306.1 |
| Recoverable gas | Bcf | 117.9 | 201.0 | 344.8 | 219.5 |
| Recoverable condensate | MMbbl | 2.84 | 5.10 | 9.13 | 5.67 |
| Total recoverable resources | MMboe | 22.7 | 38.6 | 66.5 | 42.2 |
Probability of at least each value of Chance of at least this volume, P90 / P50 / P10 from the saved simulation.
The business question
Two appraisal wells have confirmed a gas-condensate accumulation in the Falcon structure. Before a development concept can be chosen, the team needs a range for the recoverable resources rather than a single number. Each input to the volumetric equation — area, thickness, net-to-gross, porosity, water saturation, gas expansion factor, recovery factor and condensate yield — is uncertain.
A common shortcut is to build a 'low case' by multiplying every input's low value together. This model shows why that shortcut is misleading: the chance of every input being low at the same time is tiny, so the 'all-low' case sits far below the true P90.
- Who it is for
- Geoscientists, reservoir engineers and anyone preparing resource estimates for reporting or for a development decision.
- Main outputs
- Gas initially in place · recoverable gas · condensate · total resources in barrels of oil equivalent, as P90 / P50 / P10 · comparison with deterministic low, mid and high cases · inputs ranked by influence.
How the model works
Three sheets after this one, read from left to right. Each one feeds the next.
Inputs sheet
Area is entered as a Lognormal from P90 and P10 estimates; the other inputs are PERT ranges (minimum, most likely, maximum).
Volumes sheet
The volumetric equation in steps: gross rock volume, net rock volume, pore volume, hydrocarbon pore volume, gas initially in place (GIIP), then recoverable gas, condensate and total barrels of oil equivalent. A second table builds deterministic low, mid and high cases by setting every input to the same percentile.
Results sheet
Simulation gives the true P90 / P50 / P10 of the resources. They are compared with the deterministic cases, and the inputs are ranked by how strongly they drive the total.
ModelRisk techniques
Find similar models
- Product
- Type
- Application
- Uses
The results in detail
Deterministic cases against the simulation
The 'all-low' and 'all-high' cases are far more extreme than the P90 and P10: all inputs are rarely low (or high) together.
| Case | Unit | Deterministic | Simulated | Chance below |
|---|---|---|---|---|
| Low case vs P90 | MMboe | 11.8 | 22.7 | 0.2% |
| Mid case vs P50 | MMboe | 39.4 | 38.6 | 52.0% |
| High case vs P10 | MMboe | 120.9 | 66.5 | 99.7% |
Read 'Chance below' as: how often the real resources would fall below that deterministic number.
What drives the resource range
Lowest and highest mean total resources over ten equal groups of each input (VoseSimSensitivityMean), drawn from the overall mean.
Tornado chart: the inputs ranked by how far the mean output moves across each input’s range.
The ranking as a table 8 inputs
| Input (ranked by influence) | Lowest mean | Highest mean | Rank |
|---|---|---|---|
| Productive area | 21.3 | 75.3 | 1 |
| Gross thickness | 32.2 | 53.2 | 2 |
| Net-to-gross | 34.6 | 48.9 | 3 |
| Porosity | 35.1 | 49.0 | 4 |
| Gas recovery factor | 38.2 | 47.0 | 5 |
| Gas expansion factor | 38.0 | 45.5 | 6 |
| Water saturation | 38.9 | 45.6 | 7 |
| Condensate-gas ratio | 40.9 | 44.4 | 8 |
How to read these results
- The P90, P50 and P10 are the volumes with a 90%, 50% and 10% chance of being exceeded. They come from the simulation, so they account for how all the inputs vary together.
- The deterministic low case multiplies every input's 10th percentile (or 90th for water saturation). The chance of a result that low is usually well under 1%, so it is not a P90. The same is true of the high case at the other end.
- The mid case is not the mean either: multiplying medians gives roughly the median of a product, while the mean is higher because the distribution is skewed to the right.
- The inputs at the top of the ranking are where better data — another appraisal well, a better seismic depth conversion — would narrow the range most.
See it move. Open the workbook, change an input on the Inputs sheet and run the simulation: every table and chart on this page updates in Excel.
Inputs and assumptions
Every assumption in the model, with its uncertainty and where it comes from. The figures are illustrative.
Reservoir geometry 3 inputs
Size of the accumulation: the productive area and the thickness of reservoir rock within it.
| Assumption | Unit | Uncertainty | Low | Most likely | High |
|---|---|---|---|---|---|
| Productive areaMapped closure within the gas-water contact; P90 and P10 from the low and high depth-conversion cases. | acres | Lognormal · P90, P10 | 2,000 | – | 5,000 |
| Gross reservoir thicknessSeen in the two appraisal wells; thins towards the flanks. | ft | PERT | 80 | 120 | 170 |
| Net-to-gross ratioShare of the gross interval that is reservoir-quality sand. | fraction | PERT | 45% | 60% | 75% |
Rock and fluid properties 3 inputs
How much of the net rock is pore space filled with gas, and how much the gas expands at the surface.
| Assumption | Unit | Uncertainty | Low | Most likely | High |
|---|---|---|---|---|---|
| PorosityCore and log porosity from the appraisal wells. | fraction | PERT | 14.0% | 19.0% | 24.0% |
| Water saturationLog-derived saturation; uncertain away from the wells. | fraction | PERT | 20.0% | 28.0% | 40.0% |
| Gas expansion factorStandard cubic feet per reservoir cubic foot, from the fluid sample (pressure and temperature). | scf / rcf | PERT | 180 | 210 | 240 |
Recovery 2 inputs
Share of the gas that can be produced, and the condensate that comes with it.
| Assumption | Unit | Uncertainty | Low | Most likely | High |
|---|---|---|---|---|---|
| Gas recovery factorDepletion drive in analogue gas fields. | fraction | PERT | 60% | 72% | 82% |
| Condensate-gas ratioBarrels of condensate per million scf of produced gas (well test). | bbl / MMscf | PERT | 15 | 25 | 40 |
Constants 2 inputs
Unit conversions. Do not change.
| Assumption | Unit | Value |
|---|---|---|
| Square feet per acreUnit conversion. | ft² / acre | 43,560 |
| Gas per barrel of oil equivalentEnergy-equivalence convention used for reporting. | scf / boe | 6,000 |
Units and conventions
- Volumes
- Bcf = billion standard cubic feet; MMbbl = million barrels; MMboe = million barrels of oil equivalent (6,000 scf of gas = 1 boe).
- Percentiles
- Petroleum convention: P90 is the low estimate (90% chance of being exceeded), P10 the high estimate.
- Gas expansion
- Standard cubic feet of gas per cubic foot of reservoir pore space (the inverse of the gas formation volume factor).
Key assumptions and limitations
- Independence
- The inputs are treated as independent. In practice porosity and water saturation are often correlated; OG-05 shows how to add that with a copula.
- Single zone
- One reservoir zone with average properties; no compartments or gas-water contact uncertainty.
- Recovery
- The recovery factor and condensate yield do not depend on the size of the accumulation.
- Data
- All figures are illustrative. Falcon is not a real discovery.
How to use it
- Change the yellow cells on the Inputs sheet to describe your own accumulation.
- Press F9 a few times and watch the Volumes sheet: each recalculation is one possible accumulation.
- Run a simulation from the ModelRisk ribbon (10,000 samples with a fixed seed are set).
- Read the Results sheet, which fills in automatically, and open the Results Viewer for more charts.
Simulate this model with ModelRisk’s classic engine: it does not run in Turbo mode yet.
ModelRisk functions used
- VoseLognormalAlt
- Productive area, entered as P90 and P10 estimates.
- VosePERT
- Expert ranges (minimum, most likely, maximum) for the other inputs.
- VosePERT(…, U), VoseLognormalAlt(…, U)
- With U fixed at 0.1, 0.5 or 0.9 the same distributions return a chosen percentile — used to build the deterministic cases.
- VoseInput, VoseOutput
- Name the inputs and outputs for sensitivity analysis and the Results Viewer.
- VoseSimPercentile, VoseSimProbability
- Simulation statistics on the Results sheet.
- VoseSimSensitivityMean
- The tornado chart: every VoseInput ranked by how far the mean resources move across its range.
Further reading
- Sensitivity analysis in Excel: a tornado chart step by step (Vose guide)
- SPE / WPC / AAPG / SPEE / SEG / SPWLA / EAGE (2018). Petroleum Resources Management System (PRMS), probabilistic methods.
- Vose, D. (2008). Risk Analysis: A Quantitative Guide, 3rd edition. Wiley — on why percentiles do not multiply.
- OG-02 adds the chance of success; OG-05 builds a full exploration economics model on the same volumetric logic.
Download the workbook and run it yourself. It opens in Excel with ModelRisk. The free trial runs every example model.
Related example models
- ModelRisk★★★★★Production Forecast with Decline UncertaintyHow much more will well K-12 produce?Fits a decline to production history, carries the uncertainty of the fit into the forecast and adds a risk event.
- ModelRisk★★★★★Oil Price Forecast from HistoryWhich price deck should the five-year plan use?Fits three time-series models to price history, weighs them by the evidence and averages their forecasts.
- ModelRisk★★★★★Chance of Success with Shared RiskOne well, two stacked targets that share the same source rock. What are the real chances of success, and what is the well worth?Dependent chance factors: one shared risk and two independent ones, with risked volumes and value.
An illustrative example: the figures are invented for teaching and do not describe a real company, project or asset. Last updated 9 October 2026.