Balancing conservation and economic gain: a dynamic programming approach

Abstract We optimize the trade-off between economic and ecological concerns in conservation biology by using a novel method to link a spatially explicit individual-based model to a dynamic programming model. To date, few optimality models have been presented to optimize this trade-off, especially when the common currency cannot be easily measured in dollars. We use a population simulation model (e.g. spatially explicit individual-based model) to model a hypothetical forest bird population's response to different cutting and planting regimes. We then link these results to a dynamic programming model to determine the optimal choice a manager should make at each time step to minimize revenue foregone by not harvesting timber while maintaining a given population of birds. Our results show that if optimal management choices are made further back in time, future (terminal) reward may be greater. As the end of the management period approaches, past management practices influence the terminal reward more than future practices can. Thus if past revenue lost is high, the future reward will be low as compared to when past revenue lost is low. The general strategy of setting some minimum viable population size and then using a population simulator linked to a dynamic programming model to ask how to maintain such a population size with minimum economic loss should have nearly universal applicability in conservation biology.