Multispecies, trait and community size spectrum ecological modelling in R (mizer)

Abstract

Size spectrum ecological models have emerged as a conceptually simple way to model a large community of individuals which grow and change trophic level during their lives. They are a subset of physiologically structured models where growth (and thus maturation) is food dependent, and processes are formulated in terms of individual level processes. A key feature is that of a size spectrum, where the total abundance of individuals at size scales negatively with size: there are more small things than big things. Mizer is a software package for implementing size spectrum ecological models using the R statistical programming environment. The package has been developed to model marine ecosystems that are subject to fishing. Roughly speaking there are three versions of the size spectrum modelling framework of increasing complexity: The community model, in which only one \species is resolved; the trait-based model, in which all the species-species parameters are the same, except for the asymptotic size which determines other life-history parameters such as the size at maturity; and the multispecies model in which multiple \real species are resolved, each of which can have differing species-specific traits. The community and trait-based models can be considered as simplifications of the multispecies model. mizer is able implement all three model versions using the same set of tools. Mizer contains routines and methods to allow users to set up the model community, and then project it through time under different fishing strategies. The results of the simulations can then be explored using a range of plots and summary methods, including plots of size spectra and the calculation of community indicators such mean weight in the community and the slope of the size spectrum. The models created can be quite flexible and there are many options for setting up and running simulations. For example, different stock-recruitment relationships can be implemented and fishing gears with different selectivity patterns can be set up so that di_erent species are caught by different gears. However, mizer aims to make setting up the models relatively simple. For example, easy to use \wrapper functions are provided so users can set up community and trait-based models with the minimum of R gymnastics. Additionally, most of the methods and functions in the package have default options and assumptions. These can be changed by the user once they are familiar with the models. This vignette starts by summarising the principles and assumptions of size spectrum models. The mizer package is then introduced by showing how to set up the simplest types of size spectrum models: the community and trait-based model. These sections give a basic overview of the classes and methods used by mizer. Simple examples are given that demonstrate how trophic-cascades can be simulated. There then follows a more detailed description of the mizer classes and methods using a multispecies model of the North Sea as an example. Finally, there is a detailed example of a multispecies size spectrum model of the North Sea, including running projections using historical fishing patterns. z LB -N A -289-EN -N As the Commission’s in-house science service, the Joint Research Centre’s mission is to provide EU policies with independent, evidence-based scientific and technical support throughout the whole policy cycle. Working in close cooperation with policy Directorates-General, the JRC addresses key societal challenges while stimulating innovation through developing new standards, methods and tools, and sharing and transferring its know-how to the Member States and international community.