Are dual isotope and isotopomer ratios of N _2O useful indicators for N _2O turnover during denitrification in nitrate-contaminated aquifers?

摘要

Denitrifying aquifers are sources of the greenhouse gas N _2O. Isotopic signatures reflect processes of production and reduction of N _2O, but it is not clear to which extent these can be used to quantify those processes. We investigated the spatial distribution of isotopologue values of N _2O (δ ~(18)O, average δ ~(15)N, and ~(15)N site preference, SP) in two denitrifying sandy aquifers to study N _2O production and reduction and associated isotope effects in groundwater. For the first time, we combined this approach with direct estimation of N _2O reduction from excess-N _2 analysis. Groundwater samples were collected from 15 monitoring wells and four multilevel sampling wells and analysed for NO3-, dissolved N _2O, dissolved O _2, excess N _2 from denitrification and isotopic signatures of NO3- and N _2O. Both aquifers exhibited high NO3- concentrations with average concentrations of 22 and 15mgNL ~(-1), respectively. Evidence of intense denitrification with associated N _2O formation was obtained from mean excess-N _2 of 3.5 and 4.3mgNL ~(-1), respectively. Isotopic signatures of N _2O were highly variable with ranges of 17.6-113.2‰ (δ ~(18)O), -55.4 to 89.4‰ (δ ~(15)N ~(bulk)) and 1.8-97.9‰ (SP). δ ~(15)N and δ ~(18)O of NO3- ranged from -2.1‰ to 65.5‰ and from -5‰ to 33.5‰, respectively. The relationships between δ ~(15)N of NO3-, δ ~(15)N ~(bulk) and SP were not in good agreement with the distribution predicted by a Rayleigh-model of isotope fractionation. The large ranges of δ ~(18)O and SP of N _2O as well as the close correlation between these values could be explained by the fact that N _2O reduction to N _2 was strongly progressed but variable. We confirm and explain that a large range in SP and δ ~(18)O is typical for N _2O from denitrifying aquifers, showing that this source signature can be distinguished from the isotopic fingerprint of N _2O emitted from soils without water-logging. We conclude that isotopologue values of N _2O in our sites were not suitable to quantify production or reduction of N _2O or the contribution of different processes to the total N _2O flux, apparently because these values were not only governed by individual pathways but eventually also by the spatial distribution of substrates and activity within the aquifers. These observations could be explained by the dynamics of N _2O production, reduction and transport in water-saturated systems with heterogenic distribution of microbial activity and by a combination of diffusive and enzymatic isotope effects.