estgrowth

Valero et al. 2014. Evaluating the impacts of fixing or estimating

growth parameters, across life histories and data availability.

Summary

Simulations with ss3sim to guide users when it is best to estimate growth within Stock Synthesis compared to using external estimates.

Abstract

Monte Carlo simulations were used to determine when it is better to estimate growth internally in a stock assessment model compared to using external estimates. Simulations were ran for a variety of data combinations as well as multiple life-history types.

Cases

• Growth:
  • Fixed at true values
  • Internal estimation (all 5 parameters)
  • External estimation (all 5 parameters)
• Data types:
  • Length
  • Age
  • Conditional length-at-age
  • Mean length-at-age
• Natural mortality (M):
  • Fixed at true
  • Fixed above and below the truth for a few scenarios
• Fishing pattern:
  • Two-way trip
  • Constant @ F_MSY
    • True F_MSY
    • 1.10 * F_MSY to create a truncated age-structure
    • 0.80 * F_MSY for life histories with higher M b/c true F_MSY leads to convergence issues
• Selectivity (OM) parameterized as a double normal:
  • asymptotic

Metrics

• K
• L_min
• L_max
• CV_young
• CV_old
• M
• SSB_terminal
• F_terminal

Todo:

• Standardize model
• Run scenarios
• Choose which scenarios to introduce miss-specification in M
• Create pltos
• Finish manuscript
• Submit to Fisheries