Heterogeneous Preference for Forest Ecosystem Management Focused on Species Richness in Japan

We conduct the online survey of choice experiments (CE) to investigate whether people are truly able to evaluate the number of rare/familiar species around Japanese forest ecosystem, whether and how preference heterogeneity is caused. Profile attributes of CE are consisted with rare/familiar species of animal/plant and the forest charge as price attribute. Then we don ft present species f name to focus on pure responses to numerical changes. When analyzing stated preference data, attributes positional order effects are tested to verify reliabilities of results of attributes except the price attribute. Aiming the occasion of synthesizing our results with a geographical information system (GIS), we present the number of species in the range of 5km mesh forest area. A random parameter logit Model (RPL) with repeated data is adopted in order to grasp preference for species richness as a whole. As a result, every attribute is significantly evaluated without any attribute positional order effects excluding the price attribute. Then, it is concluded that 1) the rare animal is positively considered more valuable than the rare plant, 2) the familiar plant is positively evaluated while the familiar animal is not significantly evaluated at the mean parameter estimate, and 3) preference heterogeneities are confirmed on every kind of species. It is analyzed what sources influence those heterogeneities with latent class model (LCM). A membership function is estimated by employing principal components of residential attitudes for environment. It is demonstrated that there is certain influence with residential attitudes, and that it should be paid more attention to residential emotions than demographics when conducting the management on, at least, species richness.