Historical Variability of Deciduous Trees and Deciduous Forests in Northern Sweden Effects of forest fires, land-use and climate

Hellberg, E. 2004. Historical variability of deciduous trees and deciduous forests in northern Sweden. Effects of forest fires, land-use and climate. Doctor’s dissertation. ISSN 1401-6230, ISBN 91-576-6542-7. The aim of the work underlying this thesis was to study the historical variability of deciduous trees and deciduous forests in northern Sweden. To ascertain the effects of forest fires and human activities on the historical variability of deciduous trees, retrospective studies were carried out at different. These studies were designed to: improve our understanding of the genesis of deciduous forests in the present boreal landscape; the role of forest fires and effects of forest management on the structure and composition of sprucedominated ecosystems; develop methods to separate charcoal fragments of birch based on wood anatomy, to be able to; study the effects of past human activities on forest-limit ecosystems in northern Sweden. Different sets of retrospective methods (i.e. dendroecology, pollen analysis, anthracology and analyses of written historical sources) were used to achieve these aims. The first study revealed that the effects of aggregated disturbances (forest grazing, human interference with fire regimes and logging) were responsible for the formation of deciduous forests in the present landscape. In the study of spruce-dominated ecosystems, forest fire cycles were found to be long, but had profound effects on the structure and composition of the landscapes until the beginning of the 20th century; i.e. deciduous forests were dominating in areas previously affected by fires. Furthermore, forestry had strong negative effects on the coverage of old coniferous forests and deciduous forests. In the third study, the methodology developed showed that dwarf birch and mountain birch can be distinguished from one another and from silver birch and hairy birch based on wood anatomy, while silver birch and hairy birch could not be distinguished from one another. In the forest limit-ecosystem studied, human impact was found to be strong. Due to the exploitation of trees for fuel wood the area was cleared during the Viking-ages, and in combination, with climate the forest has not recovered since then. A general conclusion was that the scale and effects of human impact were ecosystemand regionspecific depending on the prior role of natural disturbances and the nature of the anthropogenic modification.