The 15N-Gas flux method for quantifying denitrification in soil: Current progress and future directions

Denitrification in soil is a challenging process to quantify under in situ conditions, which seriously hampers the ability to accurately close or balance the nitrogen budget of terrestrial ecosystems. The 15N Gas Flux method is one of the best-suited techniques for in situ measurement of denitrification. Using a stable 15N-NO3- tracer injected or applied on the surface of soil under a closed static chamber, this method enables the measurement of both N2O and N2 denitrification fluxes. Its main limitation is the poor sensitivity towards N2 emissions, which is a common weakness of all denitrification measurement methods. We also have identified four assumptions upon which this technique relies to be accurate: 1) homogenous distribution of the tracer inside the confined soil volume, 2) absence of hybrid molecule forming processes, 3) quantitative recovery of produced denitrification products inside the flux chamber headspace (no diffusive losses) and 4) no stimulatory impact of nitrate tracer and water additions on the dynamics of the denitrification process. In this review, we revisit the principles of the 15N Gas Flux method, explore its evolution through time and assess the impact of the four assumptions through literature compilation and simulation. Finally, we elaborate and discuss key technical aspects of this method to help the reader in understanding and optimally applying the 15N Gas Flux method for the measurement of denitrification. To this end, a decision tree has been implemented at the end of this study. The outcome of our review shows that in order to address the main limitation of the 15N Gas Flux method (poor N2 sensitivity), a hybrid approach using an artificial N2-depleted atmosphere in addition to 15N isotopic tracer is a promising lead, although only a few studies have used it so far (even less so in the field). In particular, we demonstrate here the existence of a threshold of 10% atmospheric N2 concentration background below which the sensitivity increases significantly. We also show that the four assumptions mentioned above are unlikely to be fully met under field conditions. The non-homogenous distribution of the 15N tracer in soil has been shown by various authors to cause a 25% underestimation of the rate of denitrification at maximum. Through simulation, we show here that the presence of hybrid molecules should have a moderate impact on total fluxes (N2 or N2O similarly) as long as they contribute for no more than 50% of the total