A1 Vertaisarvioitu alkuperäisartikkeli tieteellisessä lehdessä
Landscape effects of disturbance, habitat heterogeneity and spatial autocorrelation for a ground beetle (Carabidae) assemblage in mature boreal forest
Tekijät: F Guillaume Blanchet, J A Colin Bergeron, John R Spence, Fangliang He
Kustantaja: WILEY-BLACKWELL
Kustannuspaikka: HOBOKEN; 111 RIVER ST, HOBOKEN 07030-5774, NJ USA
Julkaisuvuosi: 2013
Lehti:: Ecography
Tietokannassa oleva lehden nimi: Ecography
Lehden akronyymi: Ecography
Numero sarjassa: 5
Vuosikerta: 36
Numero: 5
Aloitussivu: 636
Lopetussivu: 647
Sivujen määrä: 12
ISSN: 0906-7590
DOI: https://doi.org/10.1111/j.1600-0587.2012.07762.x
Tiivistelmä
Ground beetle (Carabidae) assemblages are speciose and frequently employed as indicators of ecosystem function. It is thus important to understand the factors that affect their spatial distributions so as to better interpret how ecosystem variation influences the structure of their assemblages. We evaluated how anthropogenic disturbances, habitat heterogeneity, and spatial autocorrelation (unmeasured spatially structured habitat variables and/or dispersal limitation) influence the distribution of carabids on a mature boreal forest landscape. We worked with a large sample of carabids from 194 sites from a near-regular grid covering 70 km2. Data about forest floor cover, vegetation structure, soil drainage, and topography were associated with each site. We modelled the structure of carabid assemblages using these variables together with Moran's eigenvector maps (MEM) as spatial constraints. Overall, our model explained about half of the variation in ground beetle assemblages. Forest floor cover, soil drainage, and vegetation structure were the principal factors useful for explaining carabid assemblages. The spatial patterns described by the MEMs for the ground beetle assemblages seem to result from spatial autocorrelation in soil drainage, floor cover, topography, and vegetation structure; however, this spatial autocorrelation independently and uniquely explained only 1.4% of the variation in assemblages. This was much less important than the explanatory power of the environmental variables considered; e.g. ground cover descriptors and vegetation structure by themselves accounted, respectively, for 12.1 and 4.5% of the variation in the carabid data. The spatial distribution of ground beetles in undisturbed forest was only mildly affected by surrounding anthropogenic disturbances, but clearly small patches were invaded by species characteristic of open habitat. We conclude that to help conserve ground beetle diversity at the landscape scale in boreal forests, mature forest patches should be large and connected to other patches and the surrounding forest.
Ground beetle (Carabidae) assemblages are speciose and frequently employed as indicators of ecosystem function. It is thus important to understand the factors that affect their spatial distributions so as to better interpret how ecosystem variation influences the structure of their assemblages. We evaluated how anthropogenic disturbances, habitat heterogeneity, and spatial autocorrelation (unmeasured spatially structured habitat variables and/or dispersal limitation) influence the distribution of carabids on a mature boreal forest landscape. We worked with a large sample of carabids from 194 sites from a near-regular grid covering 70 km2. Data about forest floor cover, vegetation structure, soil drainage, and topography were associated with each site. We modelled the structure of carabid assemblages using these variables together with Moran's eigenvector maps (MEM) as spatial constraints. Overall, our model explained about half of the variation in ground beetle assemblages. Forest floor cover, soil drainage, and vegetation structure were the principal factors useful for explaining carabid assemblages. The spatial patterns described by the MEMs for the ground beetle assemblages seem to result from spatial autocorrelation in soil drainage, floor cover, topography, and vegetation structure; however, this spatial autocorrelation independently and uniquely explained only 1.4% of the variation in assemblages. This was much less important than the explanatory power of the environmental variables considered; e.g. ground cover descriptors and vegetation structure by themselves accounted, respectively, for 12.1 and 4.5% of the variation in the carabid data. The spatial distribution of ground beetles in undisturbed forest was only mildly affected by surrounding anthropogenic disturbances, but clearly small patches were invaded by species characteristic of open habitat. We conclude that to help conserve ground beetle diversity at the landscape scale in boreal forests, mature forest patches should be large and connected to other patches and the surrounding forest.
Turun yliopiston Tutkimustietojärjestelmän portaali