A Bayesian approach to ecosystem service trade-off analysis : the case of the Curonian Lagoon in Lithuania

Abstract

Ecosystem services are the basis of all life on the planet and through their interaction with natural systems and with human civilization, are the main contributor to human well-being. As such, ecosystem services are subject to anthropogenic interferences aiming to manage, utilize and ideally optimize their benefits for humans. However, anthropogenic maintenance of an ecosystem service is likely to create trade-offs with the other consumers of those services. Thus, the study of ecosystem services is attracting considerable attention as it is recognized to be difficult to account for all the tangible and intangible values of such services. The aim of this study is to extend the understanding of ecosystem service trade-offs by analyzing the applicability of Bayesian belief networks in ecosystem service valuation to the Curonian Lagoon in Lithuania. Bayesian belief networks function on the basis of causal dependencies to perform probabilistic inferences between factors of interest. Initially, a network was constructed on knowledge derived from the literature. This was followed by a survey of experts which challenged the literature based network and resulted in an updated expert informed network. Experts were further surveyed for prior probabilities to parameterize the final model. Scenario analyses, conducted to assess the models’ ability to produce the intended outputs, revealed divergent views among the experts. Results further suggest high uncertainties allocated to the generated posterior probabilities of the ecosystem services under investigation. Thus, this study demonstrates the potential of using Bayesian belief networks for ecosystem service valuation in information restricted environments while shedding new light on the ecosystem services provided by the Curonian Lagoon.