Conservation of Biodiversity

This chapter examines two biodiversity conservation applications of Markov decision problems. First, it examines the problem of managing an endangered species, the Sumatran Tiger (Panthera tigris sumatrae), which is difficult to observe. Second, the chapter investigates the problem of how to recover two endangered species which interact as predator and prey. Northern abalone (Haliotis kamtschatkana) are the preferred prey of sea otters (Enhydra lutris), both co-habiting along the pacific northwestern coast of Canada and United States. It provides for the first time an optimal recovery strategy for these two species which takes into account their functional relationship using two types of reinforcement learning algorithms over a finite-time horizon. Finally, the chapter discusses the need for further research development in the MDP community to solve challenging optimization problems in conservation biology.