Climatic changes will affect the occurrence probability of extreme windstorms. Consequently, management of uneven-aged forests can only be optimized correctly if changes in climatic conditions are considered. This article determines the optimal management regime of uneven-aged forests in the presence of climatic changes affecting windstorm occurrence probabilities, and takes into account the risk preferences of the forest owner. This study analyzes optimal harvesting of uneven-aged stands by applying a Markov Decision Processes (MDP) framework and an economic description of uneven-aged forestry. Two management types are considered: the exact uneven-aged forest management model in which the forest owner jointly manages all the different stand plots, and the independent uneven-aged forest management model that assumes that the forest owner separately and independently manages each plot of the forest. The MDP framework is applied to a non-industrial private forest owner located in North-East of France. We show that the forest owner tends to converge toward a forest structure which is very close to a normal forest. We also find that the independent model can be seen as a poor approximation of the exact model because of strong incidence in terms of optimal harvesting policy. We also show that the optimal decisions depend on risk preferences but not on the considered levels of windstorm probability possibly resulting from climate change.