Photo: Bert Knottenbeldt
A new modeling approach shows that management strategies that implement learning-by-doing can deal with high uncertainty at a very low cost when compared to a pure optimization approach.
Accounting for unexpected is a huge challenge in the management of renewable resources (such as fish stocks). The many effects of climate change exacerbate this challenge even more. For instance, slow change affecting fish growth rates, or more rapid impacts on stock size, can disrupt management plans and timely responses may be difficult. So how do you obtain long-term sustainable management under these uncertain conditions? A small group of researchers from the Stockholm Resilience Centre and the KTH Royal Institute of Technology might be getting closer to an answer. In a recent study published in The Royal Society they present a novel computational approach to understand how to simultaneously manage and learn about a system when its dynamics are unknown and constantly changing. They find that a learning-by-doing strategy in the management of renewable resources, can obtain high resilience to shocks and disturbances at a very low cost compared to optimizing for a specific type of change.
Cross-fertilizing AI methods with sustainable resource management
The researchers used methods from reinforcement learning to create an artificially intelligent (AI) model. The model allows the researchers to study how to iteratively learn and manage a renewable resource to obtain sustainable outcomes and resilient management strategies during turbulent times. The model is made up of three learning (or management) components: a learning model, a mental model and a decision-making model. The influence of each component, for the learning-by-doing process, is tested by varying a parameter linked to each component in different scenarios of environmental change. The model builds on classic growth models for renewable resources, and the scenarios are informed by real world cases.
Key aspects to consider amid increasing uncertainty
Uncertainty is inherent to the dynamics of renewable resources, thus we can never accurately predict how the resource dynamics will change. To accommodate this, the researchers propose three key aspects renewable resource managers should consider amid increasing uncertainty:
The researchers conclude that this study makes a strong first step to operationalize selected core principles in resilience and sustainability science, such as unexpected events, surprise and radical uncertainty.
Based on the following article: Learning by doing
Read the scientific publication at the Royal Society: Strategies for sustainable management of renewable resources during environmental change; Emilie Lindkvist, Örjan Ekeberg and Jon Norberg. March (2017).