This paper focuses on the characterization of stationary states for a multiple-stand forest that is subject to forest fires and managed by a producer who has expected utility preferences. An analytical and tractable characterization of the stationary rotation age is established on the basis of Karush–Kuhn–Tucker conditions. The rotation age is shown to be unique and to depend on the risk probability but not on producer's risk preferences. A numerical application, using these analytical findings and based on the forestry of maritime pine in southwestern France is conducted. This numerical application consists in designing an optimal carbon sequestration policy based on financial incentives aiming at extending forest rotation ages. Results show that forest fire probability has a significant decreasing impact on sequestration costs.