The effects of habitat changes, conservation measures and interspecific interactions on forest-dwelling hawks

Human-caused habitat changes are a major reason for the loss of biodiversity and population declines of many species. Habitat changes, such as habitat loss, fragmentation and habitat degradation, have direct and indirect impacts on species and their interactions with other species, which all affect their occurrence, survival and reproduction. In Northern Europe, intensive forestry has transformed boreal forests and worsened living conditions of especially those species dependent on mature and old forests. Populations of many forest-dwelling species have declined but the mechanisms by which habitat changes affect these species are often not known.
In this dissertation, I examined habitat changes, habitat-change effects and habitatassociated breeding performances of three declining forest-dwelling hawks, the northern goshawk Accipiter gentilis (goshawk), the common buzzard Buteo buteo and the honey buzzard Pernis apivorus. These species can compete for nest sites, but are also involved in intraguild predation (goshawk can predate the buzzards). The hawks may lack nest sites in managed forests on account of which their nesting possibilities have been supported by constructing artificial nests. I studied the breeding success of the hawks in artificial nests since the benefits of this intervention as a conservation measure had not been previously analysed. Finally, since dominant species in interspecific interactions can exclude subordinate ones from the remaining habitat patches, I studied the conditions when this could occur.
According to my results, the area of old forests has decreased throughout southern Finland while young forests have increased. These habitat changes are particularly adverse for the goshawks because their breeding success improves with an increasing proportion of old spruce forests and a decreasing proportion of young forests around their nests. The breeding performances of the common buzzard or honey buzzard were not significantly associated with habitats, probably because factors other than habitat have a greater effect on their reproduction. However, common buzzards were not eager to re-occupy nests that were surrounded by a vastness of old forests.
The breeding success of all three species was lower in artificial nests than in natural ones, although the difference in breeding success between the nest types was not significant for the honey buzzard. This suggests that conservation measures aimed at enabling the breeding of hawks in managed forests conflict with their conservation objectives. I discuss the possible contributing factors of this unexpected result and directions for further investigations. In the meantime, artificial nests could be used to replace fallen natural nests if the re-construction of natural nests is not possible for the hawks.
Interspecific interactions with other raptors proved to be a decisive factor in whether or not a subordinate raptor species occupied a territory. The occupancy stage is critical for common buzzards, most likely because a breeding failure later in the season (due to a disadvantageous occupancy decision) would involve a considerable waste of breeding investment. Common buzzards prefer to occupy safe territories, where the threat of intraguild predation by the goshawk and interference competition with other raptors are small. My results also suggest that adverse interspecific interactions with other raptors can impede a subordinate raptor species from fully exploiting the periodic food peaks. Although high prey levels would benefit common buzzards, they often preferred to avoid food-rich areas if that abundance had attracted interspecific raptors. These results suggest that other raptors that are predators or interspecific competitors, can exclude subordinate raptors from their territories, and thus contribute to the occurrence and population levels of subordinate raptors. Finally, my dissertation shows that for each species, it is crucial to know the most relevant spatial and temporal scales in order to identify where external disturbances affect them most. Therefore, long-term data is of paramount importance to detect these scales, which can differ even for similar-sized species with rather similar ecological requirements.