Built by a geologist who has run these workflows himself: a sedimentologist turned structural geomodeler.
Will this fault seal, or leak? Slide the throw and watch. The lab draws the Allan diagram for a synthetic well, maps shale gouge ratio (SGR) on every sand-on-sand window and turns it into simulator transmissibility multipliers. Override Vsh or Vcl on any interval; every equation is cited below. It is a teaching and screening tool, not a substitute for a calibrated fault seal study on your data.
Synthetic demonstration data.
The lab
Runs the throw from zero to the maximum while the diagram and the table update.

SGR on the fault: blue below the 20% threshold (potential leak), red above (sealing). Hatched: the calibrated range, 15 to 20%.
- Sand on sand: full color
- Against shale: dimmed
- Footwall boundary
- Hanging-wall boundary
- Leak point at this throw
- Chance of seal, with 10, 50, 90% contours (dotted, solid, dashed)
Blocky sand: five clean, sharp-based tabular sands between thick shales. Each column is one throw; the cursor marks yours.
Synthetic demonstration data: all datasets shown were built solely for demonstration purposes and are not client data.
Loading the lab.
Faults in series
A path crosses every fault where sand meets sand, then the next block's connected sand.
By interval, at the current throw
Type a Vsh or Vcl into a row to override the log. Window: net sand against net sand. Multiplier: the window mean. Seal capacity: at the weakest point, flat above 40% SGR.
Multiplier classes, a lab convention (set by Ben)
- Open T ≥ 0.01
- Moderate baffle 0.001 ≤ T < 0.01
- Heavy baffle 0.0001 < T < 0.001
- Seal T ≤ 0.0001
| Loading. |
Monte Carlo
Each interval's Vsh is a triangular range, by default the log mean ± 0.1. The seed reproduces a run and travels in the share link.
More ranges: throw, fault-rock thickness, threshold
| Interval | Min | Most likely | Max | Reset |
|---|
CSV and PDF are made in your browser; no data leaves it.
How to read it
Depth runs down the footwall fault plane; throw grows to the right. Footwall beds are bands, hanging-wall beds diagonals one throw deeper. Where throw grows steadily from a fault tip, each column is a point on that fault, so this triangle juxtaposition diagram reads as an Allan diagram (Allan 1989).
Color is SGR, the shale (or clay) fraction of the section that slid past each point. Sand-on-sand windows, where fluid can cross, are full strength; juxtaposition seals are dimmed.
The right-hand strip is the fault at your throw; blue ticks mark leak points.
Allan 1989
Allan, U. S., 1989, Model for hydrocarbon migration and entrapment within faulted structures: AAPG Bulletin, v. 73, no. 7, p. 803–811. doi:10.1306/44b4a271-170a-11d7-8645000102c1865d (opens in a new tab)
What it simplifies
- Flat, layer-cake beds; both walls carry one well (in series, one per block).
- SGR gives fault-rock composition, not smear continuity, lenses or fine structure.
- Single-phase multipliers; capillary and two-phase effects can restrict flow more (Sperrevik et al. 2002).
- Threshold and seal envelope are calibrations from other basins; real studies use local pressure data.
- Vsh from the linear gamma-ray index. Not yet: LAS upload (GR, GAMMA, VSH, VSHALE), Larionov transforms, 3D fault-plane view.
Sperrevik et al. 2002
Sperrevik, S., Gillespie, P. A., Fisher, Q. J., Halvorsen, T., and Knipe, R. J., 2002, Empirical estimation of fault rock properties, in Hydrocarbon Seal Quantification: Norwegian Petroleum Society Special Publication 11, p. 109–125. doi:10.1016/s0928-8937(02)80010-8 (opens in a new tab)
Equations and where they come from
Each equation says how it was checked. References open to the full citation and DOI, each checked against its publisher or Crossref.
-
Shale gouge ratio and the seal threshold
SGR = Σ(Vsh × Δz) / throw × 100%
Summed over the beds that slid past the point, a window as tall as the throw, integrating the log exactly. Vcl gives the clay form; overrides give the zoned form.
About 20% SGR separates minimal from significant across-fault pressure difference (Yielding, Freeman & Needham 1997): a depth- and lithology-dependent calibration, not a law.
Threshold calibrations
Freeman et al. (1998) report 18% holding about 8 bar. Above 40 to 50% SGR fault rock is smear-dominated; below 15 to 20%, disaggregation zones or cataclasites (Yielding, Bretan & Freeman 2010, as reported in Miocic et al. 2019).
Yielding, Freeman & Needham (1997); typography checked in Pei et al. (2015) and Miocic et al. (2019, Eq. 4).
How this applies to your field
SGR is the first screen: windows below the threshold are likely flow and pressure paths (expect shared contacts); windows above are candidate seals needing multipliers and a seal-capacity check.
Illustrative schematic, not a simulation result.
Yielding et al. 1997
Yielding, G., Freeman, B., and Needham, D. T., 1997, Quantitative fault seal prediction: AAPG Bulletin, v. 81, no. 6, p. 897–917. doi:10.1306/522b498d-1727-11d7-8645000102c1865d (opens in a new tab)
Freeman et al. 1998
Freeman, B., Yielding, G., Needham, D. T., and Badley, M. E., 1998, Fault seal prediction: the gouge ratio method: Geological Society, London, Special Publications, v. 127, p. 19–25. doi:10.1144/gsl.sp.1998.127.01.03 (opens in a new tab)
Yielding et al. 2010
Yielding, G., Bretan, P., and Freeman, B., 2010, Fault seal calibration: a brief review: Geological Society, London, Special Publications, v. 347, p. 243–255. doi:10.1144/sp347.14 (opens in a new tab)
Miocic et al. 2019
Miocic, J. M., Johnson, G., and Bond, C. E., 2019, Uncertainty in fault seal parameters: implications for CO2 column height retention and storage capacity in geological CO2 storage projects: Solid Earth, v. 10, no. 3, p. 951–967 (open access). doi:10.5194/se-10-951-2019 (opens in a new tab)
Pei et al. 2015
Pei, Y., Paton, D. A., Knipe, R. J., and Wu, K., 2015, A review of fault sealing behaviour and its evaluation in siliciclastic rocks: Earth-Science Reviews, v. 150, p. 121–138. doi:10.1016/j.earscirev.2015.07.011 (opens in a new tab); open accepted manuscript (opens in a new tab)
Further reading (3)
- Lindsay, N. G., Murphy, F. C., Walsh, J. J., and Watterson, J., 1993, Outcrop studies of shale smears on fault surfaces, in The Geological Modelling of Hydrocarbon Reservoirs and Outcrop Analogues: International Association of Sedimentologists Special Publication 15, p. 113–123. doi:10.1002/9781444303957.ch6 (opens in a new tab)
- Yielding, G., 2002, Shale gouge ratio: calibration by geohistory, in Hydrocarbon Seal Quantification: Norwegian Petroleum Society Special Publication 11, p. 1–15. doi:10.1016/s0928-8937(02)80003-0 (opens in a new tab)
- Vrolijk, P. J., Urai, J. L., and Kettermann, M., 2016, Clay smear: review of mechanisms and applications: Journal of Structural Geology, v. 86, p. 95–152 (open access). doi:10.1016/j.jsg.2015.09.006 (opens in a new tab)
-
Fault-rock permeability
log10 kf = −4 SGR − ¼ log10(D) (1 − SGR)5
kf in mD, D in meters (throw / sin dip), SGR a fraction; capped at host permeability.
Manzocchi, Walsh, Nell & Yielding (1999): the widely quoted form, not yet checked against the typeset 1999 equation.
How it was checked
It reproduces the paper's own worked example: D = 20 m and SGR = 0.2 give a median kf of 0.12 mD.
kf = 80 000 exp{−[19.4 SGR + 0.00403 zmax + (0.0055 zf − 12.5)(1 − SGR)7]}
Alternative: zmax deepest burial, zf depth at faulting (m). Switching algorithm tests uncertainty (Grant 2026).
Sperrevik et al. (2002), as reported in Miocic et al. (2019, Eq. 9); conventional units (Miocic et al. print none).
How this applies to your field
The grid inherits kf: once fault rock controls a connection, the multiplier scales with it, order for order. Algorithm and burial history can decide whether a history match needs a sealing or leaky fault.
Illustrative schematic, not a simulation result.
Manzocchi et al. 1999
Manzocchi, T., Walsh, J. J., Nell, P., and Yielding, G., 1999, Fault transmissibility multipliers for flow simulation models: Petroleum Geoscience, v. 5, no. 1, p. 53–63. doi:10.1144/petgeo.5.1.53 (opens in a new tab)
Sperrevik et al. 2002
Sperrevik, S., Gillespie, P. A., Fisher, Q. J., Halvorsen, T., and Knipe, R. J., 2002, Empirical estimation of fault rock properties, in Hydrocarbon Seal Quantification: Norwegian Petroleum Society Special Publication 11, p. 109–125. doi:10.1016/s0928-8937(02)80010-8 (opens in a new tab)
Miocic et al. 2019
Miocic, J. M., Johnson, G., and Bond, C. E., 2019, Uncertainty in fault seal parameters: implications for CO2 column height retention and storage capacity in geological CO2 storage projects: Solid Earth, v. 10, no. 3, p. 951–967 (open access). doi:10.5194/se-10-951-2019 (opens in a new tab)
Grant 2026
Grant, N. T., 2026, Modelling how stochastic parameter ranges impact fault seal uncertainty: a review and worked example: Petroleum Geoscience, v. 32, no. 4, article petgeo2026-018. doi:10.1144/petgeo2026-018 (opens in a new tab)
Further reading (2)
- Fisher, Q. J., and Knipe, R. J., 1998, Fault sealing processes in siliciclastic sediments: Geological Society, London, Special Publications, v. 147, p. 117–134. doi:10.1144/gsl.sp.1998.147.01.08 (opens in a new tab)
- Fisher, Q. J., and Knipe, R. J., 2001, The permeability of faults within siliciclastic petroleum reservoirs of the North Sea and Norwegian Continental Shelf: Marine and Petroleum Geology, v. 18, no. 10, p. 1063–1081. doi:10.1016/s0264-8172(01)00042-3 (opens in a new tab)
-
Fault-rock thickness
tf = D / 170
A harmonic average of outcrop data; real faults scatter widely, so the ratio is a control here and a range in Monte Carlo.
Manzocchi et al. (1999), as reported in Shipton et al. (2002, Eq. 6).
How this applies to your field
Simulators see thickness only as tf / L. Once fault rock dominates, doubling tf about halves the multiplier, and finer cells lower it too. Treat 170 as a range.
Illustrative schematic, not a simulation result.
Manzocchi et al. 1999
Manzocchi, T., Walsh, J. J., Nell, P., and Yielding, G., 1999, Fault transmissibility multipliers for flow simulation models: Petroleum Geoscience, v. 5, no. 1, p. 53–63. doi:10.1144/petgeo.5.1.53 (opens in a new tab)
Shipton et al. 2002
Shipton, Z. K., Evans, J. P., Robeson, K. R., Forster, C. B., and Snelgrove, S., 2002, Structural heterogeneity and permeability in faulted eolian sandstone: implications for subsurface modeling of faults: AAPG Bulletin, v. 86, no. 5, p. 863–883. doi:10.1306/61eedbc0-173e-11d7-8645000102c1865d (opens in a new tab); open copy (opens in a new tab)
-
Transmissibility multiplier
T = [1 + (tf / L) ((km − kf) / kf)]−1
Applied to one cell connection across the fault: L is the cell-center spacing, km the host permeability. As a harmonic average, tight fault rock matters less in large cells.
Where the class cutoffs come from
The four classes in the table and the figure (Open T ≥ 0.01, Moderate baffle 0.001 ≤ T < 0.01, Heavy baffle 0.0001 < T < 0.001, Seal T ≤ 0.0001) are a lab convention (set by Ben). The papers cited here give the multiplier itself; none of them names multiplier classes or sets cutoffs for them. By the equation, T is 1 with no fault rock and falls toward 0 as the fault rock tightens.
Manzocchi et al. (1999, their Eq. 7), as reported in Shipton et al. (2002, Eq. 5); two-cell form in Michie, Yielding & Fisher (2018, Eq. 1).
How this applies to your field
This is the value on each fault face of the simulation grid. Low values delay pressure support and cross-fault flow; zero isolates a compartment. Recompute it when the grid changes.
Illustrative schematic, not a simulation result.
Manzocchi et al. 1999
Manzocchi, T., Walsh, J. J., Nell, P., and Yielding, G., 1999, Fault transmissibility multipliers for flow simulation models: Petroleum Geoscience, v. 5, no. 1, p. 53–63. doi:10.1144/petgeo.5.1.53 (opens in a new tab)
Shipton et al. 2002
Shipton, Z. K., Evans, J. P., Robeson, K. R., Forster, C. B., and Snelgrove, S., 2002, Structural heterogeneity and permeability in faulted eolian sandstone: implications for subsurface modeling of faults: AAPG Bulletin, v. 86, no. 5, p. 863–883. doi:10.1306/61eedbc0-173e-11d7-8645000102c1865d (opens in a new tab); open copy (opens in a new tab)
Michie et al. 2018
Michie, E. A. H., Yielding, G., and Fisher, Q. J., 2018, Predicting transmissibilities of carbonate-hosted fault zones: Geological Society, London, Special Publications, v. 459, p. 121–137. doi:10.1144/SP459.9 (opens in a new tab); open accepted manuscript (opens in a new tab)
Further reading (1)
- Manzocchi, T., Childs, C., and Walsh, J. J., 2010, Faults and fault properties in hydrocarbon flow models: Geofluids, v. 10, no. 1–2, p. 94–113. doi:10.1111/j.1468-8123.2010.00283.x (opens in a new tab)
-
Faults in series
Tpath = N / Σ (1 / Ti)
Resistances add: with N equal unfaulted connections the path multiplier is the harmonic mean, so one tight fault controls it.
Open, partially sealed, sealed
Open: a sand-on-sand route stays under the threshold at every fault. Partially sealed: every route crosses fault rock at or above it. Sealed: no sand-on-sand route.
Series combination of the multiplier above (arithmetic, no further assumption).
How this applies to your field
Links to an aquifer or injector often cross several faults, and the tightest sets the path: one multiplier of 0.001 chokes a route the others leave open.
Illustrative schematic, not a simulation result.
Manzocchi et al. 1999
Manzocchi, T., Walsh, J. J., Nell, P., and Yielding, G., 1999, Fault transmissibility multipliers for flow simulation models: Petroleum Geoscience, v. 5, no. 1, p. 53–63. doi:10.1144/petgeo.5.1.53 (opens in a new tab)
Shipton et al. 2002
Shipton, Z. K., Evans, J. P., Robeson, K. R., Forster, C. B., and Snelgrove, S., 2002, Structural heterogeneity and permeability in faulted eolian sandstone: implications for subsurface modeling of faults: AAPG Bulletin, v. 86, no. 5, p. 863–883. doi:10.1306/61eedbc0-173e-11d7-8645000102c1865d (opens in a new tab); open copy (opens in a new tab)
-
Seal capacity
AFPD = 10(SGR/27 − C) bar, h = AFPD / (Δρ g)
SGR in percent; C = 0.5 shallower than 3 km burial, 0.25 at 3.0 to 3.5 km, 0 deeper. h is the buoyant column the window's weakest point holds, with SGR held at 40%.
Why 40%, and the range
The calibration spans about 20 to 40% SGR, and seal strength stops rising above about 40% (Bretan & Yielding's 2002 conference abstract, the preview of the 2003 paper), so the lab holds SGR at 40% in the envelope. In Monte Carlo mode the column is quoted as a range with its mode (Grant 2026).
Bretan, Yielding & Jones (2003), as reported in Miocic et al. (2019, Eq. 5). The column height is hydrostatics.
How this applies to your field
This caps the column a fault-bounded trap holds. If it is shorter than the column to spill, the fault sets the contact, and volumes should use the shorter column.
Illustrative schematic, not a simulation result.
Bretan et al. 2003
Bretan, P., Yielding, G., and Jones, H., 2003, Using calibrated shale gouge ratio to estimate hydrocarbon column heights: AAPG Bulletin, v. 87, no. 3, p. 397–413. doi:10.1306/08010201128 (opens in a new tab)
Miocic et al. 2019
Miocic, J. M., Johnson, G., and Bond, C. E., 2019, Uncertainty in fault seal parameters: implications for CO2 column height retention and storage capacity in geological CO2 storage projects: Solid Earth, v. 10, no. 3, p. 951–967 (open access). doi:10.5194/se-10-951-2019 (opens in a new tab)
Grant 2026
Grant, N. T., 2026, Modelling how stochastic parameter ranges impact fault seal uncertainty: a review and worked example: Petroleum Geoscience, v. 32, no. 4, article petgeo2026-018. doi:10.1144/petgeo2026-018 (opens in a new tab)
Further reading (2)
- Yielding, G., 2012, Using probabilistic shale smear modelling to relate SGR predictions of column height to fault-zone heterogeneity: Petroleum Geoscience, v. 18, no. 1, p. 33–42. doi:10.1144/1354-079311-013 (opens in a new tab)
- Grant, N. T., 2020, Using Monte Carlo models to predict hydrocarbon column heights and to illustrate how faults influence buoyant fluid entrapment: Petroleum Geoscience, v. 27, no. 2, article petgeo2019-156. doi:10.1144/petgeo2019-156 (opens in a new tab)
-
Monte Carlo
Each realization draws every interval's value from its triangular range; throw, thickness and threshold can be drawn too. P10 is the 10th percentile. Methods follow Grant (2020) and Wilson (2023); Grant (2026) finds seal capacity negatively skewed, so quote a range.
How a realization is drawn
A draw rescales its interval's log by draw / most likely, so a zero-width range reproduces the deterministic answer exactly. Percentiles run low to high. The seed fixes the draw order, so a seed always gives the same run.
How this applies to your field
Carry the spread into decisions: low, mid and high multiplier cases in simulation, a column-height range in volumes. Where P10 and P90 straddle the threshold, new core, log or pressure data pays off.
Illustrative schematic, not a simulation result.
Grant 2020, gouge zones
Grant, N. T., 2020, Stochastic modelling of fault gouge zones: implications for fault seal analysis: Geological Society, London, Special Publications, v. 496, p. 163–197. doi:10.1144/sp496-2018-135 (opens in a new tab)
Grant 2020, Monte Carlo
Grant, N. T., 2020, Using Monte Carlo models to predict hydrocarbon column heights and to illustrate how faults influence buoyant fluid entrapment: Petroleum Geoscience, v. 27, no. 2, article petgeo2019-156. doi:10.1144/petgeo2019-156 (opens in a new tab)
Wilson 2023
Wilson, P., 2023, A data science approach to uncertainty evaluation in fault-seal analysis for exploration applications in siliciclastic reservoirs: SPE Journal, v. 28, no. 4, p. 1912–1924. doi:10.2118/214330-pa (opens in a new tab)
Grant 2026
Grant, N. T., 2026, Modelling how stochastic parameter ranges impact fault seal uncertainty: a review and worked example: Petroleum Geoscience, v. 32, no. 4, article petgeo2026-018. doi:10.1144/petgeo2026-018 (opens in a new tab)
Further reading (2)
- Hesthammer, J., and Fossen, H., 2000, Uncertainties associated with fault sealing analysis: Petroleum Geoscience, v. 6, p. 37–45. doi:10.1144/petgeo.6.1.37 (opens in a new tab)
- Zheng, X., and Espinoza, D. N., 2022, Stochastic quantification of CO2 fault sealing capacity in sand-shale sequences: Marine and Petroleum Geology, v. 146, article 105961. doi:10.1016/j.marpetgeo.2022.105961 (opens in a new tab)
Try this
- Watch a leak close. Blocky sand at 100 ft: blue windows. Sweep the throw: they shrink, shale drags in, they turn red.
- Thin beds still smear. No shale is thick, yet at a few hundred feet of throw SGR nears the package's shale fraction.
- The template is missing a sand. This well lost 48 m, Sand C with it. Note Sand D, then tick Restore the missing section: Sand C returns and Sand D's weakest SGR drops from 60 to 46%. The as-drilled template overstates its seal.
- A range, not a number. Heterolithic well, Monte Carlo, Vsh ± 0.1: Heterolith C's column roughly doubles from P10 to P90, its multiplier with it.
- Three faults, one path. From Sand D: A and C each pass about half the flow, B under a thousandth, so the path is choked.
- Same SGR, different rock. At default burial, Sperrevik et al. gives multipliers one to three orders of magnitude below Manzocchi et al.: burial matters.
