Jordan Bureau soutiendra sa thèse le vendredi 7 avril 2017 à 13h au Centre INRA, 2163 Avenue de la pomme de pin, 45075 Orléans, bâtiment CEES dans la salle Dominique King.

Titre de la thèse : Nitrous oxide emissions by agricultural soils: effect of temperature dynamics; up-scaling measurements from the plot to the landscape 

Membres du jury :

Butterbach-Bahl, Klaus - Professor, IMK-IFU, Garmisch-Partenkirchen, Germany 

Catoire, Valéry - Professor, University of Orléans 

Ceschia, Eric - Senior lecturer, CESBIO, Toulouse 

Fléchard, Christophe - Researcher, INRA Agrohydrologie, Rennes 

Hénault, Catherine - Research director, INRA Science du sol, Orleans 

Nicolardot, Bernard - Professor, AgroSup, Dijon 

Résumé : The greenhouse gas N2O is mainly emitted by soils. Soil emissions are characterized by considerable spatial and temporal variabilities that make their quantification very difficult. While soil N2O emissions are studied on an agricultural area in the Central France by the UR SOLS since 2008, we specifically studied in the laboratory the effect of temperature on these emissions and also developed a method for upscaling N2O emissions from the plot to the landscape scales. Surprisingly, N2O emissions were observed not to increase with temperature. Q10 values, describing N2O emission sensitivity to temperature, were observed to change over time. The use of acetylene for inhibiting N2O reduction has revealed that the biological processes involved in the N2O production and its consumption respond differently to temperature variations. N2O fluxes measured in the field using several methods covering different scales of the landscape gave consistent results. The mean measured N2O fluxes were 43 μg N-N2O m-2 h-1 for the eddy covariance mast, 37 μg N-N2O m-2 h-1 for the fast-box over a similar area, while it was 71 μg N-N2O m-2 h-1 by the automatic chambers over a fertilized wheat field. Flux attribution methods were developed to determine both the spatial and temporal variability of the N2O flux over a 1-km landscape, resulting in original maps of N2O emissions at the landscape scale. All these results could be further used for developing ecosystem models. Both these ecosystems models and the methodologies hereby proposed for upscaling N2O emissions will help in soil N2O emission quantification at large scales, relevant to the inventories and mitigation strategies.