A growing interest has been shown recently in the dymanics of nitrogen in agricultural and natural ecosystems. This has been caused by increasing demands for food and fibre by a rapidly expanding world population, and by a growing concern that increased land clearing, cultivation and use of both fertilizer and biologically fixed nitrogen can have detrimental effects on the environment. These include effects on water quality, eutrophication of surface waters and changes in atmospheric composition all caused by increased cycling of nitrogenous compounds. The input and availability of nitrogen frequently affects the productivity of farming systems more than any other single management factor, but often the nitrogen is used inefficiently. Much of the fertilizer nitrogen applied to the soil is not utilised by the crop: it is lost either in solution form, by leaching of nitrate, or in gaseous forms as ammonia, nitrous oxide, nitric oxide or dinitrogen. The leached nitrate can contaminate rivers and ground waters, while the emitted ammonia can contaminate surface waters or combine with atmospheric sulfur dioxide to form aerosols which affect visibility, health and climate. There is also concern that increased evolution of nitrous oxide will deplete the protective ozone layer of the stratosphere. The possibility of a link between the intensity of agricultural use of nitrogen, nitrous oxide emissions and amounts of stratospheric ozone has focussed attention on these interactions.

In recent years concern has grown over the contribution of nitrogen (N) fertilizer use to nitrate (NO3−) water pollution and nitrous oxide (N2O), nitric oxide (NO), and ammonia (NH3) atmospheric pollution. Characterizing soil N effluxes is essential in developing a strategy to mitigate N leaching and emissions to the atmosphere. In this paper, a previously described and tested mechanistic N cycle model (TOUGHREACT-N) was successfully tested against additional observations of soil pH and N2O emissions after fertilization and irrigation, and before plant emergence. We used TOUGHREACT-N to explain the significantly different N gas emissions and nitrate leaching rates resulting from the different N fertilizer types, application methods, and soil properties. The N2O emissions from NH4-N fertilizer were higher than from urea and NO3−-N fertilizers in coarse-textured soils. This difference increased with decreases in fertilization application rate and increases in soil buffering capacity. In contrast to methods used to estimate global terrestrial gas emissions, we found strongly non-linear N2O emissions as a function of fertilizer application rate and soil calcite content. Speciation of predicted gas N flux into N2O and N2 depended on pH, fertilizer form, and soil properties. Our results highlighted the need to derive emission and leaching factors that account for fertilizer type, application method, and soil properties.