Installation of __datalowSA is most easily done within RStudio.

To install datalowSA on a Windows computer will need to install the latest Rtools.exe files, which contain the required C++ compiler and other tools needed to develop packages. Rtools can be obtained from https://CRAN.R-project.org under the Download R for Windows installation link and subsequent window. The only tricky aspect is that you must add the Rtools binary directory to your Windows path, but the CRAN page provides clear instructions for doing that. On a Macintosh computer you should not need Rtools.exe as the required software is already installed. However, the devtools and Rcpp package will be required.

Then the development version of datalowSA can be installed from GitHub with:

# install.packages("devtools")  #  unhash these if you do not have devtools or
# install.packages("Rcpp")      #  Rcpp installed
devtools::install_github("haddonm/datalowSA",build_vignettes=TRUE)

You will need devtools to install from GitHub and Rcpp to produce the code needed for the catch-MSY functions.

The steps above should install the vignettes as well (use browseVignettes(“datalowSA”).

If you just want the vignettes and FRDC Report then they are to be found in the ‘https://www.github.com/haddonm/datalowSADocs’ repository. Once there in your browser, perhaps the best way forward is to press the green “code” button and ask to ‘download ZIP’, which will download a 4.9Mb file containing all the files in the repository, which you can open in the usual way.

datalowSA is a branch from simpleSA (the original branch called humbleSA has been discontinued).

simpleSA was written originally, and rapidly, during a FRDC project aimed at providing training courses in data poor stock assessment methods (Reducing the Number of Undefined Species in Future Status of Australian Fish Stocks Reports: Phase Two - training in the assessment of data-poor stocks. FRDC Project 2017/102. See Haddon et al, 2019, which is included in the ‘https://www.github.com/haddonm/datalowSADocs’ repository).

After I left CSIRO in August 2018, it would not have been correct to continue the development of simpleSA, as that might have interfered with how CSIRO wanted to progress any development. So I took a complete branch and called it ‘humbleSA’. While this name is trying to remind potential users that such assessment methods need to be treated with both caution and compassion, given their very limited nature, it could be considered a confusing name. Hence, here I will continue the development of the methods included using the more descriptive name of datalowSA. datalowSA instead of ‘datapoorSA’ because some of the included methods (surplus production models, both simple and age-structured) are more like data-moderate methods in their requirement for an index of relative abundance or a biomass estimate or two.

Development of datalowSA has already moved beyond simpleSA, and it now contains code for conducting simple age-structured stock reduction analyses on truly data-poor fisheries such as are found in exploraory fisheries. However, those functions have yet to be completely documented inside a suitable vignette so caution is especiallyurged should they be used (though each function contains worked examples).

• 2020-09-03 datalowSA 0.1.2 Modified the draft and incomplete catch-curve vignette to correct an equation and the plotting code for selectCC. Thanks to Andre for pointing out that the multinomial -ve log-likelihood equation had errors. The R-code functions, selectCC, multLL, and multinomLL are all present and correct.

Haddon, M. Burch, P., Dowling, N., and R. Little (2019) Reducing the Number of Un-defined Species in Future Status of Australian Fish Stocks Reports: Phase Two - training in the assessment of data-poor stocks. FRDC Final Report 2017/102. CSIRO Oceans and Atmosphere and Fisheries Research Development Corpora-tion. Hobart 125 p.