AimOne of the few general laws in ecology is that species richness is a positive function of area. However, it has been proposed that area would merely be a proxy for energy. Additionally, habitat heterogeneity has been found to be an important factor determining species richness. Yet the relative importance of those relationships is little known, and it is still unclear how they are brought about. We aimed to dissect which factors drive the species richness of boreal forest birds, and to identify the most probable mechanisms. Location Forested protected areas in Finland.Methods Using bird line census data collected in 104 protected areas, we ran simultaneous autoregressive models to explain the species richness of forest birds. We explored the value of forest area, tree volume, tree growth, mean degree days and habitat heterogeneity as explanatory variables and used the species richness within different species groups, based on the predictions of hypothesized mechanisms, as a response variable.Results Energy, rather than area or habitat heterogeneity, seems to be the main driver of species richness in boreal forest birds. More specifically, productive energy was a better predictor of total species richness than solar energy. Among the tested hypothetical mechanisms, the sampling hypothesis received strong support. After accounting for sampling, solar energy had an effect on species richness.Main conclusions As productive energy, such as tree volume, is associated with species richness, high-energy areas should be prioritized in forest conservation planning. Reductions in productive energy may first lead to the disappearance of the rarest species due to the random sampling process. Climate change may result in increased species richness due to increasing amount of productive and solar energy in forests. However, the range shifts of bird species may not be fast enough to keep up with the temperature increases.

Understanding large-scale variation in species richness in relation to area, energy, habitat heterogeneity and anthropogenic disturbance has been a major task in ecology. Ultimately, variation in species richness results from variation in individual species occupancies. We studied whether the individual species occupancy patterns are determined by the same candidate factors as total species richness. We sampled 26 boreal forest ponds for dragonflies (Odonata) and studied the effects of shoreline length, water vascular plant species density (WVPSD), availability of nutrients, intensity of forestry, amount of Sphagnum peat cover and pH on dragonfly species richness and individual dragonfly species. WVPSD and pH had a strong positive effect on species richness. Removal of six dragonfly species experiencing strongest responses to WVPSD cancelled the relationship between species richness and WVPSD. By contrast, removal of nine least observed species did not affect the relationship between WVPSD and species richness. Thus, our results showed that relatively common species responding strongly to WVPSD shaped the observed species richness pattern whereas the effect of least observed, often rare, species was negligible. Also, our results support the view that, despite of the great impact of energy on species richness at large spatial scales, habitat heterogeneity can still have an effect on species richness in smaller scales, even overriding the effects of area.

No abstract