Ocean degradation has become a pressing concern worldwide, requiring strategic interventions to safeguard and sustainably manage marine biodiversity and resources. Marine spatial planning (MSP) is a crucial tool in this endeavor, facilitating systematic temporal and spatial allocation of marine activities, and biodiversity protection. Here I assess the feasibility and potential of employing artificial intelligence in the development of MSP for New Zealand, by using reinforcement learning implemented in the software CAPTAIN (Conservation Area Prioritization Through Artificial Intelligence). My approach seeks to improve and speed up the MSP process, informing the generation of spatial plans that prioritize conservation objectives, by integrating available data on species distributions and life-history traits, as well as data on resource use and opportunity costs. I establish a proof-of-concept that can be used to integrate artificial intelligence-driven insights with human decision-making, accelerating the generation of marine spatial plans tailored to each nation. By identifying conservation priority areas within the New Zealand region and comparing current marine protected areas, decisions on where to set MPAs can change drastically using CAPTAIN. I propose that, while artificial intelligence holds promise in addressing challenges such as species loss minimization, it remains imperative to acknowledge the complementary role of human expertise in MSP.
promotor/supervisor feedback
nothing yet
