It’s Not the Destination, It’s the Journey: Multispecies Model Ensembles for Ecosystem Approaches to Fisheries Management

Abstract

As ecosystem-based fisheries management becomes more ingrained into the way fisheries agencies do business, a need for ecosystem and multispecies models arises. Yet ecosystems are complex, and model uncertainty can be large. Model ensembles have historically been used in other disciplines to address model uncertainty. To understand the benefits and limitations of multispecies model ensembles (MMEs), cases where they have been used in the United States to address fisheries management issues are reviewed. The cases include: (1) development of ecological reference points for Atlantic Menhaden, (2) the creation of time series to relate harmful algal blooms to grouper mortality in the Gulf of Mexico, and (3) fostering understanding of the role of forage fish in the California Current. Each case study briefly reviews the management issue, the models used and model synthesis approach taken, and the outcomes and lessons learned from the application of MMEs. Major conclusions drawn from these studies highlight how the act of developing an ensemble model suite can improve the credibility of multispecies models, how qualitative synthesis of projections can advance system understanding and build confidence in the absence of quantitative treatments, and how involving a diverse set of stakeholders early is useful for ensuring the utility of the models and ensemble. Procedures for review and uptake of information from single-species stock assessment models are well established, but the absence of well-defined procedures for MMEs in many fishery management decision-making bodies poses a major obstacle. The benefits and issues identified here should help accelerate the design, implementation, and utility of MMEs in applied fisheries contexts.

Exploring the role of Northeast Atlantic cod in the Barents Sea food web using a multi‐model approach

“…This is particularly appealing because even when an individual model omits a species, the method statistically predicts behaviour of that species based upon interspecies relationships that can be obtained from other models in the ensemble, and ultimately this gap filling allows quantitative comparison across an ensemble of somewhat dissimilar models. Overall, our exploration of structural uncertainty is a step towards ‘mingling models’ (Reum et al, 2021; Townsend et al, 2014), not fully achieving formal ensembles but nonetheless using multiple models to strengthen inference and qualitatively compare predictions from models that span a range of complexity. This study also has the strength of being able to look at model responses mechanistically, which might be hidden by a statistical ensemble.…”

Oceans of plenty? Challenges, advancements, and future directions for the provision of evidence-based fisheries management advice

“…Focus should be placed on the main system drivers and additional complexity should only be added if it is supported by the data. As complexity increases, multiple model structures should be maintained to enable thorough exploration of sensitivity to key assumptions, especially where limited data or information exist to inform plausible trophic or environmental relationships, or to allow development of ensemble approaches (Spence et al 2018 ; Reum et al 2021 ). Systematic review should follow best practices (e.g., analysis of fits to observed data and exploration of model sensitivities), but it should also include in-person review panels (e.g., typical of stock assessments) that are more in-depth (Rose and Cowan 2003 ; Kaplan and Marshall 2016 ).…”

“…With increasing, often georeferenced, data availability, there will be a continued trend away from spatially aggregated single-species modeling efforts towards spatially explicit assessment approaches and data-conditioned MICE, where random effects act as a unifying statistical tool utilized across disciplines. Ensemble and multi-model approaches will become more commonplace, allowing an improved recognition of ecosystem functioning and associated uncertainty in model recommendations (e.g., Drew et al 2021 ; Howell et al 2021 ; Reum et al 2021 ). Bio-socioeconomic modeling initiatives will also continue to advance, helping to elucidate the processes driving harvest patterns and technical interactions in multispecies fisheries (e.g., Russo et al 2019 ).…”

Estimating the Impact of a Seasonal Fishing Moratorium on the East China Sea Ecosystem From 1997 to 2018

“…Since 2004, the minimum mesh size of fishing nets in the China Sea has been 5 cm, but B. pterotum, the dominant planktivore species, has only a 3.5-cm maximum body length (Fishbase); thus, this species can benefit from this policy. These low-TL fishes have a short growth period and high reproductive rate that allows them to adapt to intense fishing pressure (Reum et al, 2021). Jiang et al (2009) also showed that after a fishing moratorium, the community tends to be dominated by fast-growing small groups, which facilitates the expansion of fish in the community.…”

“…Ecopath with Ecosim (EwE), a widely used tool to support ecosystem-based fisheries management, prioritizes the ecosystem rather than a single species population (Pikitch et al, 2004;Halpern et al, 2008;Surma et al, 2019;Reum et al, 2021), and it can explore the long-term performances of multiple FMPs under different scenarios (Li, 2009;Russo et al, 2017;Papapanagiotou et al, 2020;Wang et al, 2020;Paradell et al, 2021). Therefore, to verify the hypothesis on FMPs' effects, this work attempted to estimate the interannual variation of the ECS ecosystem from 1997 to 2018 with EwE and explored the long-term effects of the SFM under the actual fishing pressures present during the two decades.…”