Modelling total soil respiration in agricultural soils.

Soil respiration is a process which results in CO2 release from the soil to the atmosphere. It comprises two main components. The first one is heterotrophic respiration: CO2 is produced by soil microorganisms while decomposing the substrate. The second one is autotrophic respiration in which CO2 originates from roots and rhizospheric organisms. All the CO2 is then transported to the surface by diffusion (see Goffin et al., this session). Many biotic and abiotic factors play a role in soil respiration, making this process complex to analyze and understand. Temperature often appears as the most important driving variable.
Besides that, interest in the future CO2 emissions from agricultural soils has been growing. Indeed, these ecosystems are a major concern from environmental, economic and social points of view. In particular, the choice of cultural practices and residue management techniques has a strong influence on CO2 emissions from agricultural systems.
This work aims at getting to a better understanding of soil respiration in agricultural soils. To reach this goal, many semi-mechanistic models have been previously developed at very different spatio-temporal scales. We intend to adapt such an existing model to crop soils, within a spatial scale of a cultivated field and an annual temporal scale. The model will be validated by using flux measurements carried out at three different crop sites situated in the Hesbaye region in Belgium (Lonzée) and in the South West of France (Lamasquère, Auradé).
The study was focused first on soil heterotrophic respiration. Within this part, short term sensitivity of this component to temperature was studied by means of a laboratory incubation experiment. This one was performed with soil samples taken at the Lonzée site. Among the many interesting results we got, it showed a clear sensitivity of soil heterotrophic respiration to short term temperature changes. In parallel, the soil heterotrophic model was calibrated on soil chamber measurements taken at the Lonzée site (Belgium). Next steps in this part of the work will be to calibrate the model using the data from the French sites, and finally to validate the model on the three sites.
Afterwards, an autotrophic respiration submodel will be implemented and the results compared to field measurements carried out at the three sites. A further development could consist in simulating agricultural practices to take their impacts on CO2 emissions from crops into account.