Various forms of software are being used by groups developing Management Strategy Evaluation (MSE) applications. These code, packages in R and libraries in languages such as C++ for the conditioning of Operating Models (OMs), running simulations, building Management Procedures (MPs) and presenting results and summary statistics.
In some cases existing stock assessment methods, such as Multifan-CL and Stock Synthesis, are being used to construct OMs conditioned on stock assessment data sets and to develop Management Procedures (MPs), e.g. ASPIC.
Much of the software is available from github repositories or in the case of R packages from CRAN.
Contributions and updates are welcome and encouraged. The source file for this page is in a GitHub repository so that pull requests are also possible.
RFMOs
CCSBT
The MSE process was fully completed in costume-made software programmed in ADMB which include the operating model (for assessments), projections, and management procedures. The source code and R-package for evaluating outputs is maintained through a private (available to members) git repository. The main link for the commission for reports etc can be found at http://www.ccsbt.org.
IATTC
SS3 is being used as a platform to generated operating models conditioned on stock assessment data sets. The procedure is described in details in Maunder (2014) in an application for Pacific bluefin tuna. http://www.iattc.org/PDFFiles2/StockAssessmentReports/SAR15/7-Management-Strategy-Evaluation.pdf
ICCAT
A variety of R packages are being used to develop MSEs for a number of stocks Albacore MSE http://rscloud.iccat.int/mse/papers/SCRS2013-34.pdf
IOTC
Two MSEs are being conducted for albacore and skipjack, see the IOTC Working Party on Methods Working Group Report http://iotc.org/science/wp/working-party-methods-wpm. For albacore SS is being used for the conditioning of the OM and FLR for running the MSE https://github.com/iagomosqueira/SS3forOM, see https://github.com/iotcwpm/SKJ for the skipjack MSE.
WCPFC
Multifan-CL http://www.multifan-cl.org is being used to run simulations of HCRs and R4MFCL, a package in R is being used to summarise results https://github.com/hoyles/r4mfcl_git
IPHC
A variety of tools are being used and git is being used for version control and code repositories, see
git http://git-scm.com these include
Interactive tools for use by stakeholders
- IPHC MSE Tool https://iphc.shinyapps.io/MSAB/
Links to Software for conducting MSE
- iSCAM (software in development for the Halibut MSE) https://github.com/smartell/iSCAM/tree/IPHC-developer
- Delay Difference Model from ADMB Course in Hawaii https://github.com/smartell/PIFSC/tree/master/opaka/DelayDiff
- Shiny Applications for R http://shiny.rstudio.com
- Using C++ in R http://www.rcpp.org
High power computing CPU & GPU and using makefiles for spawning parallel jobs
- using OpenCL in R http://cran.r-project.org/web/packages/OpenCL/OpenCL.pdf
- Example of using OpenCL in R https://github.com/smartell/DPSAM/blob/master/R/openCL/openCLeg.R
- Makefiles "http://www.gnu.org/software/make/manual/make.html
- pthreads
Tools
There are numerous R packages being used for developing MSE. These are available from the comprehensive R Archive Network (CRAN) or other repositories.
CRAN
- kobe a package for for summarising results from stock assessments and MSE in the Kobe II format.
- r4ss
FLR
Generic MSE software http://flr-project.org/doku.php
- FLCore
- FLBRP
- FLash
- FLXSA
- biodyn
FLa4a
MSE Framework for data poor stocks: http://cran.r-project.org/web/packages/DLMtool/DLMtool.pdf
Shiny Apps
*Some examples of apps summarising MSE results
Tool to explore HCRs for a tuna like species Atlantic bluefin tuna prototype
http://rscloud.iccat.int:3838/gbyp-mse/
Git
Other
- SS3 http://nft.nefsc.noaa.gov/SS3.html
- R4SS http://cran.r-project.org/web/packages/r4ss
- NOAA MSE http://nft.nefsc.noaa.gov/MSE.html
- SS3SIM An R package for fisheries stock assessment simulation with Stock Synthesis http://arxiv-web2.library.cornell.edu/pdf/1312.6450.pdf