 |
| Titel |
Inverse modeling of gas diffusion coefficients and CO2 production rates from steady state gas profiles in a tropical lowland forest soil |
| VerfasserIn |
B. Koehler, E. Zehe, M. D. Corre, E. Veldkamp |
| Konferenz |
EGU General Assembly 2009
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 11 (2009) |
| Datensatznummer |
250025280
|
|
|
|
|
|
| Zusammenfassung |
| Soil respiration is the second largest flux in the global carbon (C) cycle but the underlying
below-ground process, soil carbon dioxide (CO2) production, can not be measured in the
field. Mathematical models may be a useful tool to calculate soil CO2 production rates.
Under the assumptions that gaseous diffusion is the only relevant transport pathway for
CO2 in soils and that measured soil CO2 profiles are at steady state, soil CO2
production has frequently been calculated from the vertical change in CO2 flux. The flux
was determined using gas diffusion coefficients calculated based on established
relationships with soil properties like porosity. Large uncertainties remain regarding the
calculated soil CO2 production rates. Encountered problems were that the calculated
production strongly depended on the chosen function to calculate the diffusion
coefficients; production rates were sometimes negative and a depth-resolution of soil CO2
production was not achieved but rather production was reported as sum per soil
horizon and top soil production was often calculated from mass balance with the
measured soil CO2 efflux. Instead of using a relationship with soil properties to
calculate diffusion coefficients we calculated diffusion coefficients based on fitting
parameters of a steady state function describing the measured CO2 profiles. We
assessed the validity of these diffusion coefficients using radon as a tracer, tested the
model assumptions and demonstrate the model approach using CO2 data from a
seasonal tropical lowland forest. Our calculations produced valid results in that
calculated production profiles decrease monotonically from the top to the deep soil and
production is largest at the end of wet season and smallest at the end of dry season.
Calculated mineral soil production rates agree with measured soil CO2 effluxes during
dry season and the annual difference between wet season measured CO2 efflux
and modeled mineral soil CO2 production is in the same order of magnitude as
the litterfall-C input, and thus attributable to a missing description of litter layer
decomposition/CO2 production. We show that the assumption of steady state should not a
priori be considered valid for all data sets but be tested, that diffusion coefficients
calculated based on soil properties may not be of sufficient accuracy to reliably
calculate soil CO2 production, and that the calculation results depend critically
on the method chosen to interpolate between the measured CO2 concentrations. |
|
|
|
|
|
|