In flr/ss3om: Tools for Conditioning Fisheries Operating Models Using Stock Synthesis 3

library(knitr)
opts_chunk$set(echo=FALSE, message=FALSE, warning=FALSE,
  fig.width=5, fig.height=4.5)

The ss3om package has been developed to bring the output of the Stock Synthesis (SS) integrated stock assessment platform [@Methot_2013] into the FLR framework [@Kell_2007]. The package initial objective was to allows the use of SS model runs to construct Operating Models (OM) for Management Strategy Evaluation (MSE) of candidate Management Procedures (MP). The outputs of model runs are used to construct object of the various classes employed when running evaluation of management procedures using the FLR methods, particularly those defined in the mse package. Initial and current work on these code is related to the evaluation of management procedures for some stocks managed by the Indian Ocean Tuna Commission (IOTC), but has been extended for other stocks and model configurations.

The MPs evaluated using these OMs are generally based on simpler estimators of stock status or trends, so no attempt is made here to provide facilities for running SS as part of a management procedure. For simulations involving SS as a status indicator, please take a look at the ss3sim package.

This package has been used on a number of SS models (see below), for both versions 3.24 and 3.30. Stock synthesis is a very flexible framework, able to allow construction of a large number of different models. Not all of them can currently be loaded into FLR using ss3om. If your model fails to load, or the objects do not quite hold the right information, please do not hesitate contacting us to see if the code could be extended to work on it. However, development is now focused on models coming from SS 3.30, including some of the new features it has brought. Nevertheless, interaction with the output and input files is taken care by the r4ss package. So as long as r4ss can load your model, there is a good chance we can load it into FLR.

Installation

The package is available from the FLR repository, and can be installed, including all its dependencies, by calling

install.packages("ss3om", repos=structure(c(CRAN="https://cloud.r-project.org/",
  FLR="https://flr-project.org/R")))

You can also install the development version directly from the github repository. For this an updated version of r4ss is needed:

remotes::install_github("r4ss/r4ss")
remotes::install_github("flr/ss3om")

Using read functions to load SS3 results

library(ss3om)

dir <- system.file("ext-data/herring", package="ss3om")

FLStock

albio <- readFLSss3(dir, name="HER")
range(albio) <- c(minfbar=3, maxfbar=7)

plot(albio)

| FLR class | From folder | From list | | |-----------|-------------|-------------|---------------| | FLStock | readFLSss3 | buildFLSss3 | buildFLSss330 | | FLSR | readFLSRss3 | buildFLSss3 | buildFLSss330 |

        SS_output         buildFL*
folder -----------> list ---------> FL*
       `----------- readFL -------->|

Loading a full OM grid using load

Flow of information SS3 - FLR

Tested SS models

Checking an operating model

• Convergence
• Estimates of B0
• Recruitment residual trends

Auxiliary functions

• runtime
• convergencelevel,
• extractFbar, extractSSB, extractRec, extractZatage
• packss3run

flr/ss3om documentation built on March 20, 2024, 4:18 a.m.