(R) MFDP: Multi Fleet Deterministic Projection in R

The goal of R MFDP is to create an R package that can emulate the Multi Fleet Deterministic Projection (MFDP) program made by Lowestoft in 2000.

In the end the R MFDP program is so much more than the original MFDP:

  1. In addition of the “original” MFDP input, R MFDP also supports a single text file input data (cf. inst/sample directory)

  2. R MFDP supports not only Ftarget for running the forecast, but also Fpa and Flim. The detailed results for these different F values will be written into separate output files and their summary will be collated in a single summary table.

  3. R MFDP also can enforce some rules in doing the forecasts, such as the 3 year rule, maximum/minimum yearly TAC variability, and Bpa.

  4. Input and output data in R MFDP are utilizing the FLStock object from the excellent FLR suite https://flr-project.org. This flexible format enable further experiment possibilities.

For more detailed information on the above features, please have a look at the package vignettes.

Installation

You can install the package directly from Github with:

remotes::install_github("REDUS-IMR/mfdp")

Example

There are two ways of running R MFDP:

1. Using the original MFDP input data

The original MFDP input data usually consist of several files that are grouped by a single index file. Some example of these files are available in the inst/sample/nea-had-2020 directory.

library(mfdp)
input <- system.file("sample/nea-had-2020", "Fpa2020ind.txt", package = "mfdp")
output <- mfdp(input, run_name = "test", out_dir = "out")

2. Using the single MFDP input file

In addition to the original MFDP input file, R MFDP also support a simplified single-file input (cf. inst/sample/nea-had-2020.input.csv file). This is file is actually identical to the .prd file from the original MFDP run outputs.

However, to use the above input type, user must also provide a configuration file (cf. inst/sample/nea-had-2020.params.txt file) that contains the F and TAC values, and rules, among others, that will control the forecast run.

library(mfdp)

input_file <- system.file("sample", "nea-had-2020.input.csv", package = "mfdp")
config_file <- system.file("sample", "nea-had-2020.params.txt", package = "mfdp")

# Reading a single file input must be done in two steps
input <- read_mfdp_input_table(input_file)
output <- mfdp(input, configs = config_file, run_name = "test", out_dir = "out")