HR: 08:45h
AN: B51E-04 INVITED     [Abstracts]
TI: Are Roots the Source of All Soil Organic Matter? Results From Isotopic Experiments in Temperate Forests
AU: * Torn, M S
EM: mstorn@lbl.gov
AF: Lawrence Berkeley Nat'l Laboratory, One Cyclotron Rd, Berkeley, CA 94720 United States
AB: Plants produce organic detritus through roots and aboveground senescence, mainly litterfall. In soil science and biogeochemistry, the main source of soil carbon (C) inputs has been assumed to be litter. For example, litterfall is used as a measure of primary productivity relevant to belowground processes and decomposition, and properties of litter decay have been used to parameterize soil C models. There is little empirical evidence, however, that aboveground C inputs make a quantitatively important contribution to mineral soil organic matter (SOM). In a series of experiments in Mediterranean conifer and eastern deciduous forests, we used 13-C and 14-C analysis to quantify the contribution of leaf /needle C versus root C into soil organic matter pools (separated by density, physical, and chemical fractionation). Because dissolved organic C (DOC) leaching into soil may be rapidly decomposed by microbes, we also examine incorporation of isotopic tracers into microbial biomass (using chloroform-fumigation extraction, 13-C PLFA, and handpicked ectomycorrhizal fungi). We have found that aboveground inputs make almost no contribution to soil organic matter or microbial biomass in the mineral soil of these forests, at least within five years of substrate deposition. A new model of the litter layer might have the litter layer accumulating and decaying in relative isolation from the mineral soil. In that case, DOC leaching from the litter layer may be providing energy but not biomass to microbes, be mineralized in abiotic reactions with soil minerals, or be moving rapidly in macro pores. We note that these sites have low earth worm populations; sites with more bioturbation might have more surface C input to SOM. We have found that fine root lifetimes are much longer than typical leaf or needle lifespan, such that the two sources must be treated differently in biogeochemical models. It also means that the stock of SOM in these forests is derived from a much smaller flux of C inputs than previously assumed, implying that the efficiency of stabilization (or residence time) of root inputs as SOM is much higher than previously assumed.
DE: 0315 Biosphere/atmosphere interactions (0426, 1610)
DE: 0428 Carbon cycling (4806)
SC: Biogeosciences [B]
MN: Fall Meeting 2005