The upcoming of internet of things draws interest of many companies and leads to the creation of smart environments. The foundation necessary for this purpose lies in the integration of sensors, which continuously provide context data of their environment. Based on this context, changes of state in the environment, i.e., situations, can be detected. However, with the huge amount of heterogeneous context and its processing, new challenges arise. Simultaneously, the dynamic behavior of the environment demands automated mechanisms for applications to adapt to the situations automatically and in a timely manner. To meet this challenge, we present (1) the situation model as a data model for integrating all data related to situation recognition, and (2) the management and provisioning of situations based on this situation model to further decouple situation recognition and applications adapting to recognized situations. Furthermore, we present a prototypical implementation of the situation model and its management.