FLR toolset for evaluating multi-annual management plans for ICES WKMFOA
- Iago MOSQUEIRA (WMR) [email protected]
- Jasper BLEIJENBERG (WMR) [email protected]
This TAF repository contains a template for conducting an evaluation of multi-annual management plans. A single management plan, assuming a shortcut of the stock assessment, the standard ICES advice rule and short-term forecast, can be run with different frequencies, that is, how often advice is provide.
The latest version of the required FLR packages, and all their dependencies, can be installed from the FLR R-universe page by calling:
install.packages("icesTAF")
install.packages(icesTAF::deps(), repos=c(
FLR="https://flr.r-universe.dev",
CRAN="https://cloud.r-project.org/"))
There are three version of the data.R script, setup different according to the stock assessment model used to condition the OM.
-
data_ss3.R, uses an Stock Synthesis stock assessment (sol.27.4 as example). The OM contains the stock-recruitment relationship estimated by SS3.
-
data_sam.R, takes the result of a SAM run (). It builds the SRR uncertainty by bootstrap, and add the computed process error in the hindcast projection.
-
data_a4a.R bases the OM on the result of an FLa4a run (). SRR uncertainty is also computed using bootstrap.