Abstract Maintaining biodiversity in boreal forest landscapes in conjunction with forestry is a challenging task. This requires ecological understanding that is based on empirical research. In this thesis, I examined spatial and temporal occupancy patterns as well as predictability of the occurrence of the Siberian flying squirrel (Pteromys volans L.) in Finland. I used thematic maps which matched habitat requirements of the flying squirrel in forested landscapes and data on species presence and absence, which were gathered in suitable forest habitats. The results of this thesis provide applications for landscape management. First, the preferred habitat characteristics of the flying squirrel were linked to available forest data. In addition, some predictive habitat models could be used to estimate the distribution of the flying squirrel within a region. Second, based on a five year study the forests were classified as continuously occupied, continuously unoccupied and variable-occupancy patches. The dynamic occupancy pattern emphasizes the need for repeated surveys to also locate the seldom-used suitable habitats in a landscape. Third, a comparison of simulated future scenarios in long-term forest planning suggested that flying squirrel habitat might be maintained without considerable loss of timber in a landscape. Thus, a combination of ecological and economic goals in forestry planning is an encouraging alternative. Fourth, there were more polypore species in forests occupied by the flying squirrel. This suggests that conservation of the flying squirrel habitats would protect other naturally co-occurring species, and thus the flying squirrel could be assigned as an umbrella species in mature spruce-dominated forests. Based on these findings, I suggest that the flying squirrel could be used as one of the target species for forest management in boreal forest landscapes. Further research challenges are related to the examination of habitat thresholds and to the projection of future scenarios where ecological, economic and social aspects are combined to assist in complex decision making processes.