Probabilistic Inputs
Configure Rock, launch, and contact-property distributions with repeatable sampling and bounds.
JRunout supports variable inputs at three levels: Rock definitions, Rock Source launch conditions, and terrain contact materials.
Supported distributions
| Input family | Normal | Gamma | Uniform |
|---|---|---|---|
| Rock mass and density | Yes | Yes | Yes |
| Rock Source X/Y velocity and angular velocity | Yes | Yes | Yes |
| Material restitution and friction | Yes | Yes | No in the current editor |
- Normal uses the deterministic field as its mean and adds a standard deviation.
- Gamma uses positive shape (
k) and scale (θ) parameters. - Uniform uses an explicit ± range around the deterministic mean.
Each distribution can also have an Absolute Min and Absolute Max. Mass and density retain a positive lower bound; restitution and friction keep their physical bounds.
Configure a distribution
- Open the Material, Rock, or Rock Source editor.
- Select the distribution button beside a supported numeric field.
- Enable probabilistic sampling.
- Choose the distribution and enter its spread or shape/scale.
- Add absolute bounds when the engineering model requires them.
- Review the preview and apply.
The editor rejects non-positive standard deviations, Gamma parameters, and uniform ranges, as well as a minimum greater than its maximum.
Deviation Cutoff
Configure → Scenario Settings → Rockfall → Deviation Cutoff (σ) controls the sample envelope. For Normal inputs, the sampled value is limited to the mean plus or minus the selected number of standard deviations, subject to absolute and physical bounds. Gamma sampling is also clipped to the configured cutoff envelope and bounds. Uniform input retains its explicit ± range within the configured cutoff and bounds.
A cutoff of zero removes probabilistic spread from Normal and Uniform inputs.
Sampling frequency
Rock and Rock Source values are drawn once for each Rock job. Contact-material values use the scenario's Per Simulation, Per Segment, or Per Impact sampling option. See Simulation Models & Settings for the exact redraw boundary.
Repeatability
The scenario Random Seed is combined deterministically with the Rock Source and Rock ordinal. With the same geometry, catalogs, settings, and seed, the job sequence and input samples are repeatable. Changing population order, allocations, or the Random Seed changes those samples.
Probabilistic inputs are model assumptions, not automatic calibration. Choose distribution families, spreads, correlations, and bounds from appropriate site characterization and engineering judgment.