Using machine learning to inform harvest control rule design in complex fishery settings

TL;DR

This paper applies reinforcement learning and Bayesian optimization to design harvest control rules for complex, partially observed fish populations, specifically addressing the highly variable Walleye fishery in Alberta, Canada.

Contribution

It introduces a novel approach using machine learning tools to optimize harvest policies in complex ecological settings, surpassing traditional reference point methods.

Findings

Optimized policies outperform standard reference point policies.

Including mean fish weight observations improves policy effectiveness.

Reinforcement learning effectively handles ecological complexity and uncertainty.

Abstract

In fishery science, harvest management of size-structured stochastic populations is a long-standing and difficult problem. Rectilinear precautionary policies based on biomass and harvesting reference points have now become a standard approach to this problem. While these standard feedback policies are adapted from analytical or dynamic programming solutions assuming relatively simple ecological dynamics, they are often applied to more complicated ecological settings in the real world. In this paper we explore the problem of designing harvest control rules for partially observed, age-structured, spasmodic fish populations using tools from reinforcement learning (RL) and Bayesian optimization. Our focus is on the case of Walleye fisheries in Alberta, Canada, whose highly variable recruitment dynamics have perplexed managers and ecologists. We optimized and evaluated policies using several complementary performance metrics. The main questions we addressed were: 1. How do standard policies based on reference points perform relative to numerically optimized policies? 2. Can an observation of mean fish weight, in addition to stock biomass, aid policy decisions?