spatialSim: multi-species spatiotemporal size-structured operating model for management strategy evaluation

TL;DR

spatialSim introduces a computationally efficient multi-species spatiotemporal model for sedentary fisheries, enabling realistic management strategy evaluation at true spatial scales with localized depletion insights.

Contribution

the paper presents a novel, scalable operating model combining Gaussian Markov Random Fields with areal harvesting to improve spatial management simulations.

Findings

realistic catch-per-unit-effort data generated

localized depletion effects captured

model applied successfully to New Zealand surfclam fishery

Abstract

Spatiotemporal processes have the potential to be one of the most influential factors governing how fisheries targeting sedentary species respond to harvesting. Despite this, management strategy evaluation often fails to account for space or does so at low resolutions due to compute constraints. In this paper, a multi-species spatiotemporal size-structured operating model for sedentary species is presented. The model combines a spatially continuous Gaussian Markov Random Field model of the population dynamics with an areal harvesting model that supports preferential targeting and site selection constraints (e.g., economic constraints). This approach is very compute efficient, which makes it feasible to simulate realistic fisher dynamics and catch data at true spatial scale (e.g., the swept area of a dredge). The New Zealand surfclam fishery was used as a case study to demonstrate the versatility of the operating model and to showcase the simulation of localized depletion, which was manifest in the generation of realistic catch-per-unit-effort data that were uncorrelated with the trends in population abundance. The model is available as part of the open-source R package spatialSim.