View on GitHub


R package for the VRAP program

VRAP (Viability and Risk Assessment Procedure)

To use



The Viability and Risk Assessment Procedure (VRAP) model was adapted from the Risk Assessment Procedure (RAP) original developed by Jim Scott (WDFW) and Bob Hayman (Skagit Coop) in the 1990’s. The original purpose of the model was to assess the impact of various harvest rates on salmon populations based on the variance in returns and in management precision, including a harvest rate scale based on size of the returns. The model was based on the assumption of density dependence in recruits per spawner and introduced stochasticity in the estimates with variance estimates around the fit of the data to the predicted spawner-recruit curve and in management error. The model has since been expanded to assess the level of productivity and capacity (the two parameters of density dependent spawner recruit functions) needed to sustain a viable population given a level of desired harvest and the estimated variance in returns and management precision.

Population viability is a measure of survival based on abundance, productivity, and/or other conditions that give a population a low risk of extinction over some defined period of time. The population viability criteria used by NOAA in evaluating salmon recovery describe the number, productivity, spatial structure and diversity required for a naturally self-sustaining population to have either a 0.95 or 0.99 probability of persistence over a 100 year time period.

The VRAP Model

The Viability and Risk Assessment Procedure (VRAP) model is a stochastic simulation model originally written in Visual Basic and later converted to R. VRAP was developed to address the complex life history of Chinook salmon, returning to spawn at multiple ages and being susceptible to fisheries in multiple years. The model may be used to assess various harvest rates given a spawner-recruit function or to determine the parameters of a spawner-recruit function that would satisfy the viability assumption given a desired harvest rate.

VRAP allows one to choose among the Hockey stick, Ricker, or Beverton-Holt spawner recruit functions to determine the production from spawning escapement. Harvest schemes are then applied to estimate progeny spawners based on cohort analysis. Uncertainty may be introduced at several levels in the analysis. The variance of estimated recruits (based on observed data of escapement and catch) around the spawner recruit curve is used to stochastically determine recruits. In addition, patterns and random variability due to environmental conditions may be introduced by way of two covariates to the analysis. Also, error in achieving a targeted harvest rate may also be introduced.

VRAP uses risk assessment analysis to look at viability status and/or effects of various harvest rates on viability/rebuilding of salmon based on predicted returns from the number of spawners using density dependent spawner-recruit functions and estimated error around the estimates.

Two Types of VRAP Simulations

VRAP’s two modes of operation are:

For more information

See the VRAP User Guide in the doc folder of the package.