In an old growth coniferous forest located in the central Cascade Mountains, Oregon, we added or removed aboveground litter and terminated live root activity by trenching to determine sources of soil respiration. Annual soil efflux from control plots ranged from 727 g C m À2 year À1 in 2002 to 841 g C m À2 year À1 in 2003. We used aboveground litter inputs (149.6 g C m À2 year À1 ) and differences in soil CO 2 effluxes among treatment plots to calculate contributions to total soil efflux by roots and associated rhizosphere organisms and by heterotrophic decomposition of organic matter derived from aboveground and belowground litter. On average, root and rhizospheric respiration (R r ) contributed 23%, aboveground litter decomposition contributed 19%, and belowground litter decomposition contributed 58% to total soil CO 2 efflux, respectively. These values fall within the range of values reported elsewhere, although our estimate of belowground litter contribution is higher than many published estimates, which we argue is a reflection of the high degree of mycorrhizal association and low nutrient status of this ecosystem. Additionally, we found that measured fluxes from plots with doubled needle litter led to an additional 186 g C m À2 year À1 beyond that expected based on the amount of additional carbon added; this represents a priming effect of 187%, or a 34% increase in the total carbon flux from the plots. This finding has strong implications for soil C storage, showing that it is inaccurate to assume that increases in net primary productivity will translate simply and directly into additional belowground storage.

No abstract

We assessed microbial community composition as a function of altered above- and belowground inputs to soil in forest ecosystems of Oregon, Pennsylvania, and Hungary as part of a larger Detritus Input and Removal Treatment (DIRT) experiment. DIRT plots, which include root trenching, aboveground litter exclusion, and doubling of litter inputs, have been established in forested ecosystems in the US and Europe that vary with respect to dominant tree species, soil C content, N deposition rate, and soil type. This study used phospholipid fatty-acid (PLFA) analysis to examine changes in the soil microbial community size and composition in the mineral soil (0-10 cm) as a result of the DIRT treatments. At all sites, the PLFA profiles from the plots without roots were significantly different from all other treatments. PLFA analysis showed that the rootless plots generally contained larger quantities of actinomycete biomarkers and lower amounts of fungal biomarkers. At one of the sites in an old-growth coniferous forest, seasonal changes in PLFA profiles were also examined. Seasonal differences in soil microbial community composition were greater than treatment differences. Throughout the year, treatments without roots continued to have a different microbial community composition than the treatments with roots, although the specific PLFA biomarkers responsible for these differences varied by season. These data provide direct evidence that root C inputs exert a large control on microbial community composition in the three forested ecosystems studied.

No abstract