Testing the use of a land cover map for habitat ranking in boreal forests.

Habitat loss and modification is one of the major threats to biodiversity and the preservation of conservation values. We use the term ''conservation value'' to mean the benefit of nature or habitats for species. The importance of identifying and preserving conservation values has increased with the decline in biodiversity and the adoption of more stringent environmental legislation. In this study, conservation values were considered in the context of land-use planning and the rapidly increasing demand for more accurate methods of predicting and identifying these values. We used a k-nearest neighbor interpreted satellite (Landsat TM) image classified in 61 classes to assess sites with potential conservation values at the regional and landscape planning scale. Classification was made at the National Land Survey of Finland for main tree species, timber volume, land-use type, and soil on the basis of spectral reflectance in satellite image together with broad numerical reference data. We used the number and rarity of vascular plant species observed in the field as indicators for potential conservation values. We assumed that significant differences in the species richness, rarity, or composition of flora among the classes interpreted in the satellite image would also mean a difference in conservation values among these classes. We found significant differences in species richness among the original satellite image classes. Many of the classes examined could be distinguished by the number of plant species. Species composition also differed correspondingly. Rare species were most abundant in old spruce forests (>200 m3/ha), raising the position of such forests in the ranking of categories according to conservation values. The original satellite image classification was correct for 70% of the sites studied. We concluded that interpreted satellite data can serve as a useful source for evaluating habitat categories on the basis of plant species richness and rarity. Recategorization of original satellite image classification into such new conservation value categories is challenging because of the variation in species composition among the new categories. However, it does not represent a major problem for the purposes of early-stage land-use planning. Benefits of interpreted satellite image recategorization as a rapid conservation value assessment tool for land-use planners would be great.

Impact of busy roads on breeding success in pied flycatchers Ficedula hypoleuca.

The impact of dense traffic on the breeding success in pied flycatchers (Ficedula hypoleuca) was studied using roadside nest-boxes. Nest site selection and breeding success of flycatchers were observed in relation to the distance from the road. The number of occupied territories was no higher closer to the road than it was deeper inside the forest. The distance to the road had no effect on the laying date, clutch size, or brood size. However, nests closer to the road were more likely to fail at the chick stage. The number of broods that were closer to roads and were lost completely was significantly higher than those further away. As a consequence, the number of fledglings per breeding attempt decreased closer to roads. The nestlings typically died as older chicks when the parent birds were providing them with maximum quantities of food. A possible reason for the declining reproductive success was the traffic-related mortality of parent birds, as weather conditions were not especially adverse for raising healthy offspring.