Despite an intense focus on the effects of habitat fragmentation on biodiversity, evidence for the direct effects of habitat fragmentation and patch area on species occurrence remains surprisingly ambiguous. Many studies were conducted in landscapes where the amount of habitat was never reduced below the point where fragmentation effects would be evident. Also, matrix type and composition of surrounding habitat patches are not always quantified, and species differences are often not taken into account. Intensively managed forests cover over 140 million ha globally and management practices often result in small, isolated, forest patches surrounded by a homogeneous matrix of younger forests. As a result, these systems provide outstanding opportunities to study effects of habitat loss and fragmentation.
This research employs a randomized complete block design to manipulate location and aggregation of retention trees within clearcut forests in Western Oregon, USA. The five treatments follow a fragmentation gradient ranging from multiple small patches dispersed throughout the clearcut area to large retention blocks on the clearcut edge. We deployed grids of consistent size within and outside retention patches to quantify the response of small mammals to habitat fragmentation.
Results/Conclusions
We collected preliminary data on 2 of 10 research blocks, and captured three species in sufficient numbers to estimate density by habitat arrangement and type. Peromyscus maniculatus, a habitat generalist, showed no response to treatment (β5=0.02, 95% CI=-0.51-0.54). Tamias townsendii, a forest dependent species, showed higher densities within retained patches compared to clear cuts (β=2.15, 95%CI=1.63-2.68), and higher densities in more fragmented treatments (β5=2.87, 95% CI=2.09-3.65). Neotoma cinerea, a large-bodied species with specific thermal requirements, had much higher densities in retained patches than harvested areas (β=1.91, 95%CI=0.65-3.17), but showed no treatment effect, likely due to small sample size (n=16). Rarefaction curves indicated that treatment had no effect on species richness within patches (Range 7.34-8.60 species/50 individuals). However, species turnover between treatments was high (0.72, Sorenson dissimilarity = 0.84) indicating that each treatment, although maintaining similar levels of richness, retained different components of regional small mammal diversity. These results indicated that species traits are important in shaping response to habitat fragmentation and that a variety of tree retention strategies may be required to conserve forest small mammal communities in the Pacific Northwest.