The Chatham Rise is a highly productive deep-sea ecosystem that supports numerous substantial commercial fisheries, and is a likely candidate for an ecosystem based approach to fisheries management in New Zealand. We present the first end-to-end ecosystem model of the Chatham Rise, which is also to the best of our knowledge, the first end-to-end ecosystem model of any deep-sea ecosystem. We describe the process of data compilation through to model validation and analyse the importance of knowledge gaps with respect to model dynamics and results. The model produces very similar results to fisheries stock assessment models for key fisheries species, and the population dynamics and system interactions are realistic. Confidence intervals based on bootstrapping oceanographic variables are produced. The model components that have knowledge gaps and are most likely to influence model results were oceanographic variables, and the aggregate species groups ‘seabird’ and ‘cetacean other’. We recommend applications of the model, such as forecasting biomasses under various fishing regimes, include alternatives that vary these components.

Calibration of complex, process-based ecosystem models is a timely task with modellers challenged by many parameters, multiple outputs of interest and often a scarcity of empirical data. Incorrect calibration can lead to unrealistic ecological and socio-economic predictions with the modeller's experience and available knowledge of the modelled system largely determining the success of model calibration. Here we provide an overview of best practices when calibrating an Atlantis marine ecosystem model, a widely adopted framework that includes the parameters and processes comprised in many different ecosystem models. We highlight the importance of understanding the model structure and data sources of the modelled system. We then focus on several model outputs (biomass trajectories, age distributions, condition at age, realised diet proportions, and spatial maps) and describe diagnostic routines that can assist modellers to identify likely erroneous parameter values. We detail strategies to fine tune values of four groups of core parameters: growth, predator-prey interactions, recruitment and mortality. Additionally, we provide a pedigree routine to evaluate the uncertainty of an Atlantis ecosystem model based on data sources used. Describing best and current practices will better equip future modellers of complex, processed-based ecosystem models to provide a more reliable means of explaining and predicting the dynamics of marine ecosystems. Moreover, it promotes greater transparency between modellers and end-users, including resource managers.Please note that this is an author-produced PDF of an article accepted for publication following peer review. The definitive publisher-authenticated version is available on the publisher Web site. Highlights► Best practices for calibrating complex process-based ecosystem models are described. ► We emphasize the importance of understanding the model structure and data sources. ► Strategies to estimate core parameters for an Atlantis ecosystem model are defined. ► Key diagnostic routines to help modelers identify erroneous parameters are outlined. ► A pedigree to evaluate ecosystem model confidence and data quality is proposed.

Ecosystem models require the specification of initial conditions, and these initial conditions have some level of uncertainty. It is important to allow for uncertainty when presenting model results, because it reduces the risk of errant or non-representative results. It is crucial that model results are presented as an envelope of what is likely, rather than presenting only one instance. We perturbed the initial conditions of the Chatham Rise Atlantis model and analysed the effect of this uncertainty on the model’s dynamics by comparing the model outputs resulting from many initial condition perturbations. At the species group level, we found some species groups were more sensitive than others, with lower trophic level species groups generally more sensitive to perturbations of the initial conditions. We recommend testing for robust system dynamics by assessing the consistency of ecosystem indicators in response to fishing pressure under perturbed initial conditions. In any set of scenarios explored using complex end-to-end ecosystem models, we recommend that associated uncertainty analysis be included with perturbations of the initial conditions.

Ecosystem and multi-species models are used to understand ecosystem-wide effects of fishing, such as population expansion due to predation release, and further cascading effects. Many are based on fisheries models that focus on a single, depleted population, and may not always behave as expected in a multi-species context. The spawning stock recruitment (SSR) relationship, a curve linking the number of juvenile fish to the existing adult biomass, can produce dynamics that are counter-intuitive and change scenario outcomes. We analysed the Beverton–Holt SSR curve and found a population with low resilience when depleted becomes very productive under persistent predation release. To avoid implausible increases in biomass, we propose limiting recruitment to its unfished level. This allows for specification of resilience when a population is depleted, without sudden and excessive increase when the population expands. We demonstrate this dynamic and solution within an end-to-end ecosystem model, focusing on myctophids under fishing-induced predation release. We present one possible solution, but the specification of stock-recruitment models should continue to be a topic of discussion amongst multi-species and ecosystem modellers and empiricists going forward.