Nitrogen transformations and losses following pig slurry applications to a natural soil filter system (Solepur process) in Brittany, France

The soil filter system, Solepur, has been shown to be highly successful at removing organic matter and nitrogen (N) from pig during its first five years of operation. The system involves three operationsapplication of large volumes of pig slurry to a managed fieldcollection and treatment of the nitrate-rich leacheateand irrigation of the treated water over other fields. The purpose of this study was to determine the environmental implications of applying excessive volumes of slurry to the managed field and to improve the understanding of the N cycle in the soil filter system. Nitrous oxide (N20) and ammonia (NH3) emissions were measured from the managed field following two slurry applications under different soil conditions. Denitrification losses, nitrate leaching, N uptake by plants and soil immobilization of N were determined and methane (CH4) emissions were measured. Emmissions were monitored following slurry applications in June and in the following October. Losses of nitrogen (expressed as a percentage of that applied), following the October application, were measured as 6% (ammonia), 23% (as nitrous oxide) and 12% (as dinitrogen gas). Losses as ammonia following the previous June application were greater (31%) but losses as nitrous oxide were lesse (1%) coinciding with lower emissions of methane. Measured losses and transformation of N were in agreement with the estimated N balance made in earlier work. Methods for reducing NH3 and denitrification losses from the system are discussed, e.g. shallow injection of slurry to reduce ammonia losses and the addition of nitrification inhibitors to slurry to reduce the production of a suitable nitrogen source at times when conditions are favourable for denitrification. Le système Solépur a été développé pendant 5 ans et s'est avéré efficace pour la biodégradation par le sol et l'élimination d'azote. Cette étude a été mise en place pour suivre les différentes formes de l'azote (NH3, N20, N2) émises à  la suite de deux campagnes d'épandages de lisier (juin et octobre). Les pertes en azote (exprimées en % de l'N apporté) s'établissent pour l'épandage effectué en octobre à  6% sous la forme ammoniacale, 23% sous forme de protoxyde d'azote et 12% sous forme d'azote moléculaire. Lors de l'apport de lisier en juin, les pertes sous forme d'ammoniac étaient beaucoup plus importantes (31%) mais les émissions de protoxyde d'azote inférieures à  1% ainsi que les émissions de méthane qui étaient très faibles.