Skip to contents

DLMtool

The Data-Limited Methods toolkit (DLMtool) is a library of over 100 data-limited management procedures (MPs) for use in fisheries management strategy evaluation (MSE). It is part of the openMSE collection of R packages.

openMSE 2.0

DLMtool 7.0 is part of the openMSE 2.0 release, alongside MSEtool 4.0. MSEtool 4.0 introduces a modular operating model structure, with stock, fleet, observation, and implementation components defined as independent objects, and a closed-loop MSE workflow based on om, hist, and mse objects, replacing the earlier OM, Hist, and MSE classes. The technical documentation covers this workflow with worked examples.

DLMtool’s management procedures (MPs) predate this new structure and are written against the legacy Data and Rec classes. They remain fully supported: MSEtool::Convert() automatically converts between the new data class and the legacy Data class, so existing MPs run unchanged in either engine. See NEWS for details of what changed in this release.

Key Features

  • Large MP library: output control (TAC), input control (effort, size limits, spatial closures), and mixed-control MPs, ranging from simple catch-based rules to model-based estimators
  • Works with MSEtool v4.0+: legacy MPs run directly against data-class objects produced by MSEtool’s current om/data simulation engine, via automatic object conversion (MSEtool::Convert())
  • Works with the legacy engine: MPs remain fully compatible with the original OM/Data classes for existing workflows
  • Plain R functions: every MP is a function that accepts a Data object and returns a Rec (recommendation) object, so new MPs are easy to write and test
  • Diagnostic tools: functions for checking MP behaviour and data requirements before running a full MSE

Not every legacy MP maps cleanly onto the new engine: a small number rely on legacy-only Data slots (e.g., Data@Abun, Data@FMSY_M, Data@BMSY_B0) that MSEtool’s new observation model does not yet populate when converting data to Data. These MPs fail with an informative error rather than silently returning an invalid recommendation; see ?Required for each MP’s data requirements.

Installation

You can install the released version of DLMtool from CRAN with:

install.packages("DLMtool")

And the development version from GitHub with:

# install.packages("pak")
pak::pak("Blue-Matter/DLMtool")

Documentation

  • openMSE.com: general information about the openMSE framework
  • docs.openMSE.com: technical manual with worked examples covering operating model construction, simulation, and MSE analysis
  • dlmtool.openmse.com: full function reference for DLMtool
  • msetool.openmse.com: function reference for MSEtool, the core package providing the new om/data simulation engine