reports/inputParSummary.md

In Pacific-salmon-assess/samSim: Closed-loop salmon dynamics simulator

Summary of Input Parameters

Describes current primary input parameters for recoverySim.R and, where applicable, how they were derived

1. simPar - defines simulation processes

• NOTE: justification for specific sim values (largely sigma values) are summarized in reports/refParameterValues.xlsx and parameterJustification.docx
• nameOM - operating model name
• nameMP - management procedure name
• species - currently tested with sockeye and chum; pink and coho will be supported eventually
• timeVaryA - set to FALSE (i.e. stable), decrease, increase or cycle
  • Currently specified at simulation run level, but could be swapped for CU- specific values
• simYears
• harvContRule - fixedER or TAM
• canER - Canadian mixed CU fishery (i.e. non-terminal marine)
• corrrelCU - correlation among CUs in recruitment deviations
• tauCatch - interannual variation in age structure; estimated multiple ways using proxy data (details in fraserRiverDoc.md and exploreTau.R)
  • Not possible to parameterize with Nass data
• obsSig - estimated spawner abundance error
• mixOUSig - mixed (i.e. marine) fisheries outcome uncertainty (assumed to be the same for both US and Canadian mixed)
• singOUSig - single fisheries outcome uncertainty
  • Currently specified at simulation run level, but could be swapped for CU- specific values
• obsMixCatchSig - estimated catch error in mixed-CU fisheries
• obsSingCatchSig - estimated catch error in single-CU fisheries
• ageErr - error assigning recruits to a brood year
• extinctThresh - pseudo-extinction threshold
• minA - minimum productivity level at which declines stabilize

2. cuPar - defines CU-specific parameters

• manUnit
• stkName
• stk
• cuNameOM/cuNameMP - vector that is only used when multiple simulations are run with varying CU-specific traits (e.g. differences in single CU exploitation rate)
• model - ricker or larkin (latter Fraser only)
• usER - US exploitation rate; occurs before Canadian or single-CU exploitation
• canER - Canadian mixed-stock exploitation rate
  • Note currently overridden by simPar input, but could be adjusted
• tauCycAge - error term used to generate interannual deviations from mean age
  • For the Fraser these values are the mean of cycle line specific values estimated in exploreTau.R
  • For the Nass these values could not be directly estimated (age structure is not regularly estimated), so arbitrary moderate value used
• alpha - median Ricker's alpha
  • For Fraser provided by AMH/FRSSI group (mean values from rickerLL - summary stats file)
  • For Nass estimated externally using ncStockRecModels.R and then median calculated from 1000 samples
• beta0 - median Ricker's beta; values as above for alpha with exception of Lower Nass CU (both median and sampled values are not estimated in SR model, but based on median spawner abundance (see script for details))
• sigma - median Ricker's sigma; values estimated as above
• meanRecX - mean number of recruits in each age class (estimated from time series for Fraser, static values used for Nass)
• medianRec/QRec - median, lower, and upper quartiles for recruit abundance; added to some output plots
• meanDBE/sdDBE - mean and interannual variation in "difference between estimates", a proxy for en route mortality
  • Fraser values estimated since 2000 from postSeasonMortality_2017.csv
  • Nass fixed at NA
• meanForecast - mean forecast error
  • Fraser estimated from data provided by PSC ManagementTables2005-2017.xls
    • Given relatively small sample sizes, pooled most data, but kept summer estimate separate because persistently conservate, while other MUs' forecasts are overestimated
  • Nass fixed at NA
• sdForecast - variation in in-season forecast error; standard deviation among years

Pacific-salmon-assess/samSim documentation built on Jan. 20, 2025, 11:17 a.m.