HR: 0830h
AN: B31D-0333 [PDF]
TI: A Soil Carbon Cycle Without Life?: The Content and Residence Times of Organic and Inorganic Carbon in
the Atacama Desert of Chile
AU: * Amundson, R G
EM: earthy@nature.berkeley.edu
AF: University of California at Berkeley, Ecosystem Science Division, ESPM
151 Hilgard Hall #3110, Berkeley, CA 94611 United States
AU: Navarro-Gonzalez, R
EM: navarro@nuclecu.unam.mx
AF: Universidad Nacional Autonoma de Mexico, Laboratorio de Quimica de Plasmas y Estudios Planetarios
Instituto de Ciencias Nucleares
Circuito Exterior, Ciudad Universitaria
Apartado Postal 70-543
Mexico D.F. 04510, MEXICO, Mexico D. F., 04510
Mexico
AU: Ewing, S A
EM: saewing@nature.berkeley.edu
AF: University of California at Berkeley, Ecosystem Science Division, ESPM
151 Hilgard Hall #3110, Berkeley, CA 94611 United States
AB:
The central Atacama Desert of Chile is nearly rainless and virtually devoid of biota. Precipitation increases steadily as one
moves to more southern latitudes, providing a natural experiment to assess the role of water in the soil C cycle. We have
established three research sites along this gradient, where the mean annual precipitation varies from nearly 0 to about 15
mm y$^{-1}$. At the driest site, where plants are completely absent and soil microorganisms quite rare, trace quantities of
organic C (OC) are present ($\sim$0.009$+$$/$$-$0.0038$%$), and OC increases slightly with precipitation (and the increasing
presence of vegetation) to 0.053$%$. The apparent radiocarbon age of the organic matter at the driest site is exceedingly
old ($>$ 7,000 y), suggesting C cycling rates on the order of 10$^{4}$ y. The source of the incoming C is being investigated,
and may include a combination of marine aerosols and exceedingly rare cyanobacteria on the undersides of quartz clasts
("hypoliths"). Radiocarbon-based turnover times appear to increase to decadal scales with increasing rainfall, with annually
cycling OC concentrated in coppice dunes (0.32$%$ OC) and hypolith-associated soils (0.39$%$ OC). The radiocarbon age of
co-existing soil carbonate was $\sim$12,000 years at the driest site and thus older than that of the OC, suggesting limited
weathering and incorporation of modern atmospheric CO$_{2}$ with increasing precipitation.
The character of the organic matter present in the soil was analyzed by pyrolysis-GC-MS. The main organic molecules released
at $750\deg$C in an inert atmosphere are benzene and formic acid. Their concentrations in the driest soil are in the ppb
range, and decrease by about an order of magnitude with depth. This suggests that either the environmental conditions in the
past were much more severe or else that there are slow downward fluxes of organic materials accompanied by decomposition
(either biological or abiotic). In contrast, soil organic matter from the other two southern sites released an order of
magnitude more pyrolysis product at depth with respect to the surface, suggesting higher C inputs and differing decomposition
processes. The ratio of formic acid to benzene was highest at the driest site, indicating more oxidized organic matter.
Moving south, this ratio drops rapidly. At the driest site, the high formic acid$/$benzene ratio appears to remain constant
with depth. At the two more southern sites, this ratio is quite low in the surface and then goes up with depth, suggesting
that the organic matter at greater depths is more oxidized than at the surface.
DE: 0315 Biosphere/atmosphere interactions
DE: 1615 Biogeochemical processes (4805)
SC: Biogeosciences [B]
MN: 2003 Fall Meeting