Moose, trees, and ground-living invertebrates: indirect interactions in Swedish pine forests

Citation
O. Suominen et al., Moose, trees, and ground-living invertebrates: indirect interactions in Swedish pine forests, OIKOS, 84(2), 1999, pp. 215-226
Citations number
29
Categorie Soggetti
Environment/Ecology
Journal title
OIKOS
ISSN journal
00301299 → ACNP
Volume
84
Issue
2
Year of publication
1999
Pages
215 - 226
Database
ISI
SICI code
0030-1299(199902)84:2<215:MTAGII>2.0.ZU;2-5
Abstract
The role of moose in structuring the boreal forest ecosystem has been studi ed extensively in recent pears. This research has focused mainly on the eff ects of moose on vegetation and soil dynamics. However, the extent to which these effects influence animal communities has received little attention. We studied how invertebrate assemblages on the forest floor of two Swedish pine forests were affected by the presence of moose. Our approach was to tr ap invertebrates using pitfall traps in unbrowsed exclosures and browsed re ference plots, estimate the percentage cover of herb-layer plants, measure the height and density of tree species, and measure the accumulation of lea f litter. The effect of moose on tree strata differed between the two areas . In Sunnas, where the stand was a mixture of deciduous trees and pines, th e main effect of moose was to eliminate most of the deciduous trees from th e canopy. Here the amount of deciduous leaf litter was ten times higher in unbrowsed exclosures compared with browsed plots. The stands on the dry sit es in the other area, Furudal, were almost pure pine. Here, moose browsing reduced the density of taller pines, but increased the number of pine sapli ngs. At Sunnas, the herb-layer plant cover did not differ between browsed a nd unbrowsed plots. However, there was a difference between exclosures and browsed plots in terms of the ground-dwelling invertebrate assemblage. Numb er of individuals of most invertebrate groups that differed between exclosu res and browsed plots were higher in unbrowsed plots and the total number o f invertebrates was also higher in unbrowsed plots, but diversity of invert ebrate fauna was higher in browsed plots. At Furudal both the vegetation an d invertebrate assemblages differed between browsed and unbrowsed plots. Br owsed plots had a higher cover of ground lichens; while unbrowsed plots had more mosses. Among the invertebrate taxa that differed between treatments, the number favored by browsing and the number disfavored by browsing were roughly equal, and diversity was not affected by moose. We suggest that by affecting the density and composition of tree species, moose indirectly inf luenced the structure of invertebrate communities. At Sunnas, the reduction in the amount of deciduous leaf litter was probably the key effect. Vegeta tion was not affected much, but the differences in litter layer composition had affected the microclimate on the forest floor while also providing an important additional trophic resource and habitat for ground-dwelling inver tebrates. Al Furudal the main effect of moose was to reduce the density of the pine canopy: thus increasing the amount of light penetrating to the for est floor. As a consequence, the floral composition and microclimate was al tered which, in turn, affected the invertebrate fauna.