摘要：This paper studies a land use switching model as a measure of adaptation to climate change in tart cherry industry in Michigan. In order to capture the effects of extreme events, we employ a real options land conversion model where an underlying stochastic process allows a Poisson-type jump component. We compare land use decisions under the increased frequency and severity of extreme weather events as well as well-known gradual climate change. We find that when decision makers are allowed to optimize dynamically and to learn, gradual changes and extreme events can lead to different likelihoods of adaptation occurring as well as different adaptation incentives even when traditional net present value (NPV) calculations are equal. We suggest that although gradual change imposes higher incentive to switch than the extreme events, the realized action may be dominated by the extreme events, especially extreme magnitude change.
