Abstract
Ecosystem-based Fisheries Management is a holistic, place-based approach to managing living marine resources. In order for it to be successful we need to improve our understanding of species interactions and other ecosystem processes. One way that can be achieved is through the use of ecosystem models. Ecosystem models are a mathematical representation of the complex and sometimes chaotic living world. Ecosystem models encompasses a wide range of approaches to understanding that world. One useful suite of models are known as mass balance models. Mass balance models describe the ow of energy through the system and are adapt at identifying direct and indirect impacts of management decision on the system. This approach has been popularized by the software package Ecopath with Ecosim. In the Northeast United States there have been numerous studies using the Ecopath with Ecosim software, the most notable being the Energy Analysis and Modeling eXercise (EMAX) conducted by the Northeast Fisheries Science Center. While this package has many strengths, it is also very dicult for practitioners to customize without help from the developers and can suer from a lack of reproducibility. Here I describe an open source alternative version of the mass balance algorithms called Rpath. Rpath is written using a combination of R and C++ with the majority of the code readily accessible to be modied by practitioners. The new package is further expanded to serve as an operating model in a closed-loop management strategy evaluation process. As part of this functionality, the model can be paused after each time step to evaluate an external model which then modies the Rpath model parameters before the original model resumes. The closed-loop simulation is demonstrated on the EMAX version of Georges Bank, a region of the Northeast US Continental Shelf Large Marine Ecosystem that has distinct physical and oceanographic properties that make it ideally suited for place-based management. The EMAX version of Georges Bank is too aggregated to be useful for management. Therefore a more disaggregated version was developed that also includes better resolution of shing eets, making the model much more useful in a management context. This new version will be available for future management strategy evaluation processes utilizing the capabilities of Rpath, thus improving the ecosystem modeling toolbox.