Simple vs. complex stock assessment models: a comparison of their utility for managing fisheries

学者和从业人员都使用渔业资源评估来估计鱼类丰度和计算渔业捕捞配额。这些评估已演变为涉及数百个参数的复杂模型。这些措施的实施很费时，而且仅限于训练有素的技术人员。此外，这种复杂性使得这些模型对决策者和利益相关者完全不透明。最近一项关于过分简单化的鲑鱼模拟渔业的研究表明，基于较简单模型的评估可能优于使用较复杂模型的评估，即使较复杂模型是正确的，即使使用这些模型可以纳入更多的相关信息。本研究通过模拟东白令海狭鳕种群及其渔业扩展了这一分析。这一分析的结果将证明更简单的模型是否能够提供可靠的估计。狭鳕渔业是美国最大的渔业，平均每年的捕捞量超过100万吨，并且该种群也是白令海生态系统的主要组成部分。本研究的结果将提高该渔业管理的经济和养护绩效。这些结果可能会为美国和其他地方的各种渔业的种群评估程序提供信息。

Both scholars and practitioners use fisheries stock assessments to estimate fish abundance and to calculate fisheries harvest quotas. These assessments have evolved towards complex models involving hundreds of parameters. Their implementation is time-consuming and limited to highly trained technical staff. Additionally, this complexity makes these models completely opaque to decision makers and stakeholders. A recent study of a simplistic simulated salmon fishery showed that assessments based on simpler models could outperform those using more complex models, even if the more complex models were correct and even if their use allowed the incorporation of additional relevant information. This research extends this analysis by simulating the Eastern Bering Sea pollock stock and its fishery. The results of this analysis will demonstrate whether simpler models can provide reliable estimates. The pollock fishery is the largest U.S. fishery, with harvests averaging over a million tons per year, and the stock is also a major component of the Bering Sea ecosystem. The findings of this research will improve the economic and conservation performance of management of this fishery. The results are likely to inform stock assessment procedures in a wide range of fisheries in the U.S. and elsewhere.