Confirmation of co-denitrification in grazed grassland


Pasture-based livestock systems are often associated with losses of reactive forms of nitrogen (N) to the environment. Research has focused on losses to air and water due to the health, economic and environmental impacts of reactive N. Di-nitrogen (N2) emissions are still poorly characterized, both in terms of the processes involved and their magnitude, due to financial and methodological constraints. Relatively few studies have focused on quantifying N2 losses in vivo and fewer still have examined the relative contribution of the different N2 emission processes, particularly in grazed pastures. We used a combination of a high 15N isotopic enrichment of applied N with a high precision of determination of 15N isotopic enrichment by isotope-ratio mass spectrometry to measure N2 emissions in the field. We report that 55.8 g N m−2 (95%, CI 38 to 77 g m−2) was emitted as N2 by the process of co-denitrification in pastoral soils over 123 days following urine deposition (100 g N m−2), compared to only 1.1 g N m−2 (0.4 to 2.8 g m−2) from denitrification. This study provides strong evidence for co-denitrification as a major N2 production pathway, which has significant implications for understanding the N budgets of pastoral ecosystems.


It has been estimated that at the beginning of the 21st century almost half the global population depended on fertilizer nitrogen (N) for its food supply1. Global population growth is predicted to further increase the demand for food by up to 100% by 20502 and there is a need to meet this in an environmentally and economically sustainable manner3. More specifically, the global demand for meat and dairy products is predicted to increase by over 30%, driven by increased affluence in the developing world4. Pasture-based livestock systems account for 25% of global land area and are inherently ‘leaky’ in terms of N, with less than 30% of the applied N recovered in milk and meat products5. Applying current farming methods to meet increased global food demands is thus likely to result in a further acceleration of the N cycle, due to increased fertilizer use and deposition of animal excreta6. Full recovery of applied N in grassland remains elusive, with 20–40% of applied N often unaccounted for7,8,9 because soil N transformations result in the formation of reactive N (Nr) compounds which include nitrate (NO3), ammonia (NH3) and nitrous oxide (N2O). Globally, livestock are responsible for 65% of N2O emissions, 64% of NH3 emissions and 60% of leached N, with animal excreta being the principal source10,11,12. In grazed pasture systems most Nr losses arise from N deposited in the form of ruminant urine, which results in localized increases in N loadings ranging from 20 to 120 g N m−2 and which generally exceed the pasture plants’ immediate requirements11.

Loss of Nr from grazed pasture systems occurs via inorganic N leaching and overland flow to surface and ground waters, NH3 volatilization and emissions of both N2O and di-nitrogen (N2) via biotic or abiotic mechanisms9. Nitrogen loss as N2, while potentially unacceptable on an economic basis, completes the N cycle and returns N to the atmosphere in an environmentally benign form. The loss of N2 from pasture livestock systems is not nearly as well studied as Nr losses. However, a more complete understanding of factors influencing N2 emissions could further elucidate Nr loss pathways. Isotopic mass balance studies have consistently failed to account for 20% of applied N9,13, with N2 emissions assumed to be the main source of uncertainty.

A commonly used methodology for calculating N2 and N2O fluxes arising from denitrification in 15N balance studies is that of Mulvaney and Boast14, which assumes that 14N and 15N atoms are randomly distributed during generation of the gas of interest (N2 or N2O) and that the NO3 pool, from which N2 is derived, is isotopically uniform. If these assumptions are violated the gas flux may be underestimated15,16.

Di-nitrogen is the end product of conventional or ‘true’ denitrification, also referred to as canonical denitrification, in which NO3 is sequentially reduced, via obligate intermediaries: NO2, NO and N2O. Denitrification is mediated by a range of microorganisms and occurs under anoxic or hypoxic conditions17,18. Nitrifying organisms may also produce N2O and N2 under reduced oxygen conditions in a process known as nitrifier-denitrification while true nitrification only results in N2O emissions18,19.

A process rarely considered in determining gaseous contributions to 15N mass balances is that of co-denitrification. Whilst hybrid N2 production is recognized in microbiology20, it has seldom been quantified in soil N process studies. Co-denitrification produces N2O (N2OCO) and N2 (N2CO) when, during sequential binding, a side reaction occurs between the initial electrophilic enzyme/N species complex and a nucleophile21,22. As a result, co-denitrification results in hybrid N2 and/or N2O molecules that are formed from isotopically non-uniform pools, with one N atom of NO/NO2 derived from an inorganic N source (NO3, NO2 or NO) and another nucleophilic N atom from a co-substrate17,22 (usually N3, NH3 or a monomeric organic N source such as an amine) (Fig. 1). Whereas abiotic N2 production has been shown to occur at low pH (<5.2)23,24, N2 from co-denitrification is recognized as a biotic process occurring under intermediate to high pH conditions (>6)22.

Figure 1

Conceptual model of co-denitrification under urine patches in grassland soils, commencing with urea, the dominant N substrate found in ruminant urine.

In grazed pastures N input from ruminant urine is the main source of N cycling and loss25. Subsequent N transformations within the created urine patch, including elevated pH and inorganic N concentrations and the stimulated microbial activity, drive Nr losses26,27,28. Isotopic mass balance studies have consistently failed to account for 20% of applied N9,13, with N2 emissions assumed to be the main source of uncertainty. Potentially, this is because the contribution of co-denitrification to gaseous N losses from ruminant urine has not yet been investigated under intensively managed grasslands.

In a previous laboratory study, N2 and N2O emissions accounted for 30–65% and <5%, respectively, of the urine N applied to an undisturbed grassland soil over a 30-day period29 but the authors were unable to identify the specific processes contributing to the high N2 emissions. Identification of such processes would enable a better understanding of N use efficiency in agricultural systems. This paper presents a novel approach that combines two 15N flux determination methodologies in order to differentiate the relative contributions of true denitrification and co-denitrification14,23,30.

Therefore, the objective of this paper was to determine the relative contributions of the denitrification and co-denitrification processes to N2 emissions from ruminant urine applied to a pastoral soil.

Results and Discussion

Denitrification of 15N-labelled N pools results in the generation of N2 gas where either one or both N atoms are 15N-labelled, giving mass numbers 29 or 30, respectively. Co-denitrification results in the formation of hybrid N2 where a large proportion of N2 molecules are produced as 29N, relative to 30N and the ratio of Δ29R to Δ30R, derived solely from co-denitrification, will be 27223 (see SI methods for further details). The observed mean ratio of Δ29R / Δ30R in this study was 214, indicating a substantial contribution of co-denitrification to the total N2 efflux. Using the conventional equations of Mulvaney and Boast14, the overall mean 15N enrichment of the pool from which the N2 was derived (15XN) could be quantified and this was calculated to be 0.0214 (2.14 atom% 15N). If true denitrification was occurring, this 15XN value should have been similar to the 15N enrichment of the pool from which N2O was derived (N2OaD). However, this was not the case, as the mean N2OaD value was calculated to be 36 atom% 15N, a far greater enrichment than that of the calculated 15XN. Hence a process other than true denitrification was responsible for the majority of the N2 produced.

It is unlikely that anaerobic ammonium oxidation (ANAMMOX) was responsible for the high N2 emissions, although its contribution cannot be ruled out. While ANAMMOX is recognised as a significant N2 production mechanism in aquatic systems31,32, its contribution to N2 production in terrestrial biomes has to date been recognised only in rice paddy soils and wetland soils32,33, characterised by anoxic conditions. Evidence to date suggests the contribution of ANAMMOX to N2 production from soils might be low, given that the ANAMMOX-specific genetic potential has also been shown to be low in wetland soils33. Trimmer and Purdy34 found an uncharacterised metabolism potentially capable of oxidising organic-N (e.g. NH2 groups) directly to N2, a process that was neither ANAMMOX nor denitrification. We suggest that the process Trimmer and Purdy observed was most likely co-denitrification. Brabandere et al.35 discussed how their experiment may have underestimated ANAMMOX and they suggest that the direct use of amines from dissolved organic N in a process similar to ANAMMOX may be responsible, but that this process has not yet been linked to a specific microbial metabolism. Our paper is the first time an attempt has been made to assess co-denitrification in a pastoral ecosystem where it plays a major role and may be the reason why circa 20% of N is still unaccounted for in N mass balances.

Temporal profiles of gaseous N emissions revealed substantial losses associated with N2CO, with mean daily fluxes of 0.44 g N m−2 d−1 over the four-month experiment (Fig. 2a). In contrast, emissions associated with true denitrification were an order of magnitude lower; with mean daily fluxes of 0.01 g N m−2 d−1 and 0.005 g N m−2 d−1 observed for N2TRUE and N2OTRUE, respectively (Fig. 2b,c). Although N2CO was the predominant loss pathway there was no detectable N2OCO during the course of the experiment. The cumulative gaseous N losses associated with true denitrification were 1.1 and 0.66 g N m−2 for N2TRUE and N2OTRUE, respectively (Fig. 3). Emissions from N2CO were the dominant loss pathway accounting for 55.8 g N m−2 and 97% of total gaseous N loss, equivalent to 56% of the N applied. Comparisons with prior studies are limited due to the fact that only one laboratory study reports co-denitrification, to our knowledge36. Their study36 showed that 92% of the N2 emitted was due to co-denitrification and only 8% due to denitrification after applying 15N enriched NH4NO3. These findings are similar to the findings in the current study.

Figure 2

Mean daily N2 and N2O fluxes (g N m−2 day−1) (n = 4) from (a) co-denitrification (N2CO) and true denitrification (b) N2TRUE and (c) N2OTRUE over a four month period following urine deposition. Error bar is the standard error of the mean (n = 4).

Figure 3

Cumulative N2 and N2O emissions from co-denitrification (N2CO) and true denitrification (N2TRUE and N2OTRUE) over a four month period following urine deposition.

Confidence intervals (95%) (n = 4) from analysis of the treatments, were 38 to 77 g m-2 for N2CO, 0.4 to 2.8 g m-2 for N2TRUE and 0.27 to 0.77 g m-2 for N2OTRUE.

In order to investigate the underlying drivers of the N2 emissions, we assessed the effect of nitrification on both N2TRUE and N2CO, using the nitrification inhibitor dicyandiamide (DCD) which was applied within a subset of 15N-labelled urine treatments. Incorporating DCD with the urine inhibited the first stage of nitrification, the oxidation of NH4+ to NO2 (Fig. 4). Over the first 30 days of the experiment, when the nitrification inhibition was effective, cumulative N2 emissions were significantly (P < 0.05) reduced by 55% for N2CO and no significant changes were found for N2TRUE (Fig. 5). This indicates that reduced NO2 formation, resulting from nitrification inhibition, affected N2CO. This finding, combined with the significantly higher N2CO contribution to the N2 flux than from N2TRUE (Fig. 3), indicates that different N pools supplied each process. The reduced nitrification activity in the presence of DCD and associated reduction in N2CO flux demonstrates that one source pool for co-denitrification (the applied N contribution) was the NO2 produced during nitrification (oxidation of NH4+ to NO2), rather than true denitrification (reduction of NO3 to NO2)37.

Figure 4

The effect of the nitrification inhibitor DCD on soil ammonium and nitrate concentrations in urine-affected soil, measured from aligned small plots receiving the same treatments as lysimeters in the main study.

Error bar is the standard error of the mean (n = 4).

Figure 5

Effect of the nitrification inhibitor DCD on cumulative N2 emissions from true denitrification (N2TRUE) and co-denitrification (N2CO) (g N m-2) over the first month following urine deposition.

Significant differences between “+DCD” and “−DCD” for each process are marked with *(P < 0.05). Error bar is the standard error of the mean (n = 4).

Urine N deposition on pastoral soils may provide the optimal conditions for co-denitrification, due to urine patches containing high inputs of N (mainly urea-N), localised regions of elevated soil pH and because urine increases the supply and turnover of labile soil organic N and C26,30. The formation of both hybrid N2 and N2O has been shown to be promoted by increases in pH38,39. Urine patches represent localized areas of high pH (circa. 8–10) due to hydrolysis reactions40,41. See Supplementary Figure S1 online for pH under urine-affected soil.

During the second stage of nitrification, the oxidation of NO2 to NO3, can be inhibited by elevated pH and high ammonia-N concentrations26,42,43. We applied 1000 μg urea-N g soil−1, with previous studies having shown that 400 μg NH4+-N g soil−1 was sufficient to inhibit NO2 oxidation41. Whereas, in unfertilized soils, NO2 pools are very low (<5 μg NO2-N g soil−1) as the oxidation of NH4+ to NO2 proceeds at a slower rate than the subsequent oxidation of NO2 to NO3. In a controlled incubation study with the same soil type and urine N input (1000 μg urea-N g soil−1), soil pH was elevated from days 0–40, during which NH4+ concentrations peaked soon after urine addition then steadily declined; which confirms expected observations in urine-affected soil27. However, there was a delay in NO2 or NO3 produced from days 10–25 during which time NH4+ concentrations were declining and soil pH remained > 6, suggesting a substantial loss of N.

Elevated soil pH under urine patches results in the hydrolysis of organic matter and increased microbial turnover of organic N and C29. Denitrification rates have been shown to be correlated with the availability of labile organic C for the supply of reductant44 and when combined with large pools of labile, nucleophilic organic N, this may result in the formation of hybrid (N-N linkage) denitrification end-products22. Indeed, this scenario has been hypothesized for both hybrid N2O and N2 formation observed under high C and elevated carbon dioxide (CO2) conditions45,46.

In the current study, N2 rather than N2O was the end product of co-denitrification. The reason for this is not readily apparent, but may be related to the presence of different microorganisms, cofactors such as copper and iron complexes, the form of metabolisable carbon substrates present and the type and oxidative state of the alternative nucleophilic N substrates (e.g., hydrazine, ammonia, ethylene diamine, aniline and/or amino acids)47. Soil pH has also been shown to favour N2O reductase48 activity and therefore promote N2 formation, which suggests the possibility of conversion of hybrid N2O to hybrid N2 under urine-affected soil when the pH is >6, as in this study.

Examples of where soil microorganisms have influenced the ratio of N2O to N2 co-denitrification product ratio include that of Okada et al.49. Where Mesorhizobium spp. under oxic conditions were observed to produce only hybrid N2, with amino acid comprising the unlabelled N source. Grazed pastoral soils have been shown to have high free amino acid levels, accounting for 10–40% of the soluble N in soil solution50 and it is perhaps possible that co-denitrification products favour N2 formation as a consequence of the microbiology that exists. For example, soil fungi have been suggested to be the principal microbial source of N2 emissions from co-denitrification36,51.

The N2CO losses of 55.8 g N m−2, equating to an N loss of 56% of the N applied (in mass balance terms), are greater than all other N loss pathways from the urine patch25. Interestingly, our finding of 56% of applied N lost as N2 fits within the range of loss (as a percentage of N applied) reported by Monaghan and Barraclough29, the only other direct measurement of N2 emissions from urine-affected soil we could find and within which the authors were unable to identify the process(es) responsible. Clough et al.9 estimated N2 loss of a similar magnitude to N2TRUE measured in the current study which suggests that N2 emissions from co-denitrification may have been overlooked. These high rates of N2CO also represent a substantial mobilization of native soil N from a urine patch (27.9 g N m−2), given that half the N2CO is derived from the applied urine N and half from native soil N. Rather than indicating a priming effect or net loss of soil N, the contribution of native soil N to N2 emissions indicates substantial mineralization-immobilization turnover (MIT) beneath a urine patch29,52, over the short term (<12 months) rather than the long term53. Nevertheless, the underlying drivers for the removal of a substantial proportion of applied and native soil N requires further investigation.


Experimental set-up

Intact monolith lysimeters, 0.5 m in diameter and 0.7 m deep, were collected from a permanent grazed grassland soil54 in the south-east of Ireland (8°15′W, 52°9′N, County Cork) and installed in the Johnstown Castle Environmental Research Centre lysimeter facility (52°17′N, 6°30′W) in County Wexford, Ireland. The area sampled had not been grazed or received manure applications for the previous eight weeks. The soil was a free-draining haplic cambisol55 which was typical of those used for dairy production. Total N was 0.42% and total carbon 4.2%. Cattle urine was collected in late autumn 2010 from mixed age Holstein-Friesian dairy cows during milking and analyzed for total N concentration. Urine-N was amended with 98 atom% 15N-labelled urea resulting in a urine N concentration of 10 g N L−1 and a 15N enrichment of 45 atom% excess. The authors confirm that all experiments were performed in accordance with relevant Teagasc guidelines and regulations and all experiments were approved by Teagasc.

Two liters of labelled urine were applied to each lysimeter on 28 December 2010 (late winter), simulating a urine patch in the field with an equivalent N loading rate of 1000 kg N ha−1. To understand the role of nitrification on N fluxes, dicyandiamide (DCD; Sigma Aldrich, Germany) was used to inhibit nitrification56. The DCD was sprayed in solution form (20 mL of 15 g DCD L−1 solution) onto the surface of the lysimeters in two split applications of 15 kg DCD ha−1, one immediately after urine application and again two months later. There were two treatments: (1) 15N-labelled urine (“no DCD”) and (2) 15N-labelled urine with DCD (“DCD”). There were four replicate lysimeters for each treatment. The climatic conditions during the 123 day experimental period 28 December 2010 to 9 May 2011, were typical for the area. The cumulative rainfall input was 241 mm, which was slightly lower than the long-term average for the area (January to April, inclusive) of 288 mm. The average maximum and minimum air temperatures were 10.5 and 4.9 °C, respectively and the average daily air temperature was 7.8 °C, which was slightly higher than the long-term average of 6.8 °C. See Supplementary Figure S2 and S3 show estimated soil water-filled pore space (WFPS) from the lysimeters and daily average air and 100 mm depth soil temperature.

Aligned soil plots

Intact monolith plots, 0.5 m in diameter and 0.2 m deep, were also collected from the same soil type and received the same urine and DCD treatments as the lysimeters. There were four replicates per treatment. These plots were sampled for soil to 75 mm depth every 10 days (on average) during the first 60 days after treatment application. Soil samples were passed through a 4 mm sieve and extracted with 2 M potassium chloride (KCl) solution. The filtrate was analysed for NH4+-N and NO3-N concentrations on an Aquakem 600 A automated analyzer (Thermo Electron, Sweden). Blank KCl samples were used to account for any residual N in the extracting solution and results were expressed as a concentration of N in the soil (μg N g−1 soil).

Gas analysis

A static closed chamber method was used to measure N2O and N2 emissions57. During measurement the outside edge of the chamber was inserted into a water channel built around the top edge of the lysimeter casing in order to ensure a gas-tight seal. At the beginning of each measurement period the chamber was placed slowly on top of each lysimeter and headspace gas samples were taken through a butyl rubber septum fitted into the top of the chamber initially (t0), after 20 minutes (t20), 40 minutes (t40) and 120 (t120) minutes. Samples were transferred from the chamber to a pre-evacuated 7 ml glass vial using a plastic syringe. Headspace samples were taken between 12:00 and 15:00 hours. Samples were taken every 2 to 3 days (n = 39) for 123 days post urine application.

Headspace N2O concentration was quantified using gas chromatography (Varian 3800 GC with ECD detector). Headspace N2 and N2O samples were analyzed for 15N on a Thermo-Finnegan Delta-S isotope ratio mass spectrometry (IRMS) interfaced with a GC capillary column in the UC Davis Stable Isotope Facility ( The limit of quantitation for the isotopic ratio mass spectrometry analysis of N2O and N2 was 150 ppmv (standard deviation of 0.1‰).

True denitrification and co-denitrification to N2 were calculated using the 15N flux method36. In brief, for N2O the ion currents (I) at m/z 44, 45 and 46 enabled concentrations and molecular ratios 45R (45I/44I) and 46R (46I/44I) to be calculated. The sources of N2O were then apportioned into the fraction (d’D) derived from the denitrifying pool of enrichment aD and the fraction d′N = (1 − d′D) derived from the pool or pools at natural abundance58. For N2, the ion currents at m/z 28, 29 and 30 enabled molecular ratios 29R (29I/28I) and 30R (30I/28I) to be determined. Differences between the molecular ratios of enriched and ambient atmospheres were expressed as Δ29R and Δ30R. The flux of N2 was calculated using three different methods:

  1. 1

    Δ29R and Δ30R were used to calculate the enrichment of the denitrifying pool (15XN) and then the N2 flux according to Mulvaney and Boast14;

  2. 2

    using Δ30R data only and the equation of Mulvaney59 assuming that the enrichment of the denitrifying pool was aD30; and

  3. 3

    using Δ29R and Δ30R to calculate a separate contribution due to co-denitrification (N2CO) and true denitrification (N2TRUE) calculated by Method 2.

True denitrification contributes to Δ29R and Δ30R whereas co-denitrification contributes mostly to Δ29R, the Δ29R to Δ30R ratio always being 27223. All of the Δ30R was assumed to be derived from true denitrification, so Δ30R was used to calculate the flux of N2 due to denitrification by Method 2. Using the ‘backsolver’ facility in Microsoft Excel, the value of Δ29R that could be attributed to N2TRUE was then obtained. The difference between the total measured Δ29R and Δ29R due to true denitrification was assigned to co-denitrification. The fraction of the total moles of N2 in the headspace from co-denitrification (dCD) was calculated using Equation 123.

where p1 (0.9963) and q1 (0.0037) are atom fractions of 14N and 15N in the natural abundance pool; p2 and q2 are the atom fractions of 14N and 15N in the enriched NO3 pool from which co-denitrification is assumed to occur. To determine the rates of N2TRUE and N2CO, the total air volume of the chamber, corrected for standard temperature and pressure, was calculated and then the amount of N2-N in the headspace determined. The amount of N2-N derived from true denitrification and co-denitrification was determined by multiplying the total amount of N2-N in the headspace by d and dCD, respectively. Denitrification rates were expressed as g N m−2 day−1.

Statistical analysis

Data analysis was conducted on cumulative fluxes from the full 123 day observation period and from the first month following urine application (to 1-Feb), to test for a DCD treatment effect. Confidence intervals for the final cumulative figures were obtained from the fitted model for the analysis of cumulative emissions. These were conservative estimates as they were based on treatment outcomes (n = 4). Log transformation was used as required to approximate constant variance and normally distributed data. Means from analyses on the log scale, with their confidence intervals, were back-transformed for interpretation on the original data scale and presentation in figures and text.

The analysis of the cumulative data was a t-test of the outcomes from four replicate lysimeters per treatment. The analysis model was fitted with the linear model procedures of the SAS statistical package (SAS Institute Inc., Cary, NC). Residual checks were made to ensure that the assumptions of the analysis were not violated. Influence statistics (e.g. Cook’s D, Restricted Likelihood Distance) were used to check whether one or more points could be creating bias in the analysis outcome.

The cumulative N-gas emissions were calculated by linear interpolation between sampling days and trapezoidal calculation60. Seventeen of the 312 observations were removed where the N2O concentration was below the IRMS limit of detection or where there were excessively high N2 enrichment readings, which caused numerical errors to propagate through the calculations of the response resulting in non-credible outcomes. For the comparisons of a treatment effect in the first month (to 1-Feb) following urine deposition, only one lysimeter had a data point removed. Removal of that data point from the analysis did not affect the finding of significant difference between the treatments.

In order to test the scale of possible errors in total cumulative fluxes due to the removal of the data, the maximum and minimum observed values for the each of the affected lysimeters were substituted for the missing values. The corresponding maximum and minimum cumulative fluxes did not alter the general conclusion based on the magnitude of the cumulative N2CO flux.

Conclusions and Implications

We conclude that co-denitrification was the dominant process producing N2 beneath a urine patch, accounting for 97% of all denitrification-derived gaseous N2 loss. Previously published values of gaseous N emissions from grazed pasture systems which only accounted for emissions of N2O and N2 via conventional denitrification may have grossly underestimated the total N loss. Other workers have shown that either bacterial or fungal co-denitrification can occur in aerobic grassland soils. We contend that co-denitrification is likely to be a ubiquitous loss pathway in grazed grasslands but the magnitude of losses may be dependent on soil factors such as soil pH and organic matter content.

The quantification of significant N loss via co-denitrification has major environmental and economic implications. The large N surpluses identified in grazed grassland systems are often assumed to be lost as environmentally-damaging reactive N emissions. However, this may not always be the case as co-denitrification resulted in large, non-reactive N2 fluxes with dual contributions from the applied urine and native soil N, whereas reactive N2O fluxes were minimal. Nevertheless, the overwhelming size of the flux indicates that this loss of N, whilst environmentally benign, represents a considerable economic loss of soil N required for production of agricultural goods. Further studies on co-denitrification and the factors affecting its magnitude will help us to close the ‘gap’ in N budgets and improve N use efficiency in grazed grassland systems and perhaps in other terrestrial ecosystems on a global scale.

Additional Information

How to cite this article: Selbie, D. R. et al. Confirmation of co-denitrification in grazed grassland. Sci. Rep. 5, 17361; doi: 10.1038/srep17361 (2015).


  1. Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z. & Winiwarter, W. How a century of ammonia synthesis changed the world. Nature Geoscience 1, 636–639, 10.1038/ngeo325 (2008).

    CAS  ADS  Article  Google Scholar 

  2. Godfray, H. C. J. et al. The future of the global food system. Philosophical Transactions of the Royal Society B: Biological Sciences 365, 2769–2777 (2010).

    Article  Google Scholar 

  3. Tilman, D., Balzer, C., Hill, J. & Befort, B. L. Global food demand and the sustainable intensification of agriculture. Proceedings of the National Academy of Sciences of the United States of America 108, 20260–20264 (2011).

    CAS  ADS  Article  Google Scholar 

  4. Alexandratos, N. & Bruinsma, J. World Agriculture Towards 2030/2050: the 2012 Revision. (FAO, Rome, 2012).

  5. Goulding, K., Jarvis, S. & Whitmore, A. Optimizing nutrient management for farm systems. Philosophical Transactions of the Royal Society B-Biological Sciences 363, 667–680, 10.1098/rstb.2007.2177 (2008).

    CAS  Article  Google Scholar 

  6. Fowler, D. et al. The global nitrogen cycle in the Twentyfirst century. Philosophical Transactions of the Royal Society B: Biological Sciences 368 (2013).

  7. Allison, F. E. The enigma of soil nitrogen balance sheets. Advances in Agronomy 7, 213–250, 10.1016/s0065-2113(08)60339-9 (1955).

    Article  Google Scholar 

  8. Ball, R., Keeney, D. R., Theobald, P. W. & Nes, P. Nitrogen balance in urine-affected areas of a New Zealand pasture. Agronomy Journal 71, 309–314 (1979).

    CAS  Article  Google Scholar 

  9. Clough, T. J., Ledgard, S. F., Sprosen, M. S. & Kear, M. J. Fate of N-15 labelled urine on four soil types. Plant and Soil 199, 195–203 (1998).

    CAS  Article  Google Scholar 

  10. Galloway, J. N. et al. The nitrogen cascade. Bioscience 53, 341–356, 10.1641/0006-3568(2003)053[0341:tnc];2 (2003).

    Article  Google Scholar 

  11. Oenema, O., Velthof, G. L., Yamulki, S. & Jarvis, S. C. Nitrous oxide emissions from grazed grassland. Soil Use and Management 13, 288–295 (1997).

    Article  Google Scholar 

  12. Steinfeld, H. Livestock’s long shadow: environmental issues and options. (FAO, Rome, 2006).

  13. Decau, M. L., Simon, J. C. & Jacquet, A. Fate of urine nitrogen in three soils throughout a grazing season. Journal of Environmental Quality 32, 1405–1413 (2003).

    CAS  Article  Google Scholar 

  14. Mulvaney, R. L. & Boast, C. W. Equations for determination of N-15 labelled dinitrogen and nitrous-oxide by mass-spectrometry. Soil Science Society of America Journal 50, 360–363 (1986).

    CAS  ADS  Article  Google Scholar 

  15. Arah, J. R. M. New formulas for mass-spectrometric analysis of nitrous-oxide and dinitrogen emissions. Soil Science Society of America Journal 56, 795–800 (1992).

    CAS  ADS  Article  Google Scholar 

  16. Boast, C. W., Mulvaney, R. L. & Baveye, P. Evaluation of nitrogen-15 tracer techniques for direct measurement of denitrification in soil: I. Theory. Soil Science Society of America Journal 52, 1317–1322 (1988).

    ADS  Article  Google Scholar 

  17. Baggs, E. M. Soil microbial sources of nitrous oxide: Recent advances in knowledge, emerging challenges and future direction. Current Opinion in Environmental Sustainability 3, 321–327 (2011).

    Article  Google Scholar 

  18. Zhu, X., Burger, M., Doane, T. A. & Horwath, W. R. Ammonia oxidation pathways and nitrifier denitrification are significant sources of N2O and NO under low oxygen availability. Proceedings of the National Academy of Sciences of the United States of America 110, 6328–6333, 10.1073/pnas.1219993110 (2013).

    ADS  Article  PubMed  PubMed Central  Google Scholar 

  19. Wrage, N., Velthof, G. L., van Beusichem, M. L. & Oenema, O. Role of nitrifier denitrification in the production of nitrous oxide. Soil Biology & Biochemistry 33, 1723–1732, 10.1016/s0038-0717(01)00096-7 (2001).

    CAS  Article  Google Scholar 

  20. Kumon, Y. et al. Codenitrification and denitrification are dual metabolic pathways through which dinitrogen evolves from nitrate in Streptomyces antibioticus. Journal of Bacteriology 184, 2963–2968, 10.1128/jb.184.11.2963-2968.2002 (2002).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  21. Shoun, H., Kim, D. H., Uchiyama, H. & Sugiyama, J. Denitrification by fungi. Fems Microbiology Letters 94, 277–281, 10.1111/j.1574-6968.1992.tb05331.x (1992).

    CAS  Article  Google Scholar 

  22. Spott, O., Russow, R. & Stange, C. F. Formation of hybrid N2O and hybrid N2 due to codenitrification: First review of a barely considered process of microbially mediated N-nitrosation. Soil Biology & Biochemistry 43, 1995–2011, 10.1016/j.soilbio.2011.06.014 (2011).

    CAS  Article  Google Scholar 

  23. Clough, T. J., Stevens, R. J., Laughlin, R. J., Sherlock, R. R. & Cameron, K. C. Transformations of inorganic-N in soil leachate under differing storage conditions. Soil Biology & Biochemistry 33, 1473–1480 (2001).

    CAS  Article  Google Scholar 

  24. Smith, D. H. & Clark, F. E. Volatile losses of nitrogen from acid or neutral soils or solutions containing nitrite and ammonium ions. Soil Science 90, 86–92 (1960).

    CAS  ADS  Article  Google Scholar 

  25. Selbie, D. R., Buckthought, L. E. & Shepherd, M. A. The challenge of the urine patch for managing nitrogen in grazed pasture systems. Advances in Agronomy 129, 229–292, 10.1016/bs.agron.2014.09.004 (2015).

    Article  Google Scholar 

  26. Clough, T. J. et al. Emission of nitrogen oxides and ammonia from varying rates of applied synthetic urine and correlations with soil chemistry. Australian Journal of Soil Research 41, 421–438, 10.1071/sr02105 (2003).

    CAS  Article  Google Scholar 

  27. Di, H. J. & Cameron, K. C. Sources of nitrous oxide from 15N-labelled animal urine and urea fertiliser with and without a nitrification inhibitor, dicyandiamide (DCD). Australian Journal of Soil Research 46, 76–82 (2008).

    CAS  Article  Google Scholar 

  28. Wachendorf, C., Taube, F. & Wachendorf, M. Nitrogen leaching from N-15 labelled cow urine and dung applied to grassland on a sandy soil. Nutrient Cycling in Agroecosystems 73, 89–100, 10.1007/s10705-005-8313-2 (2005).

    Article  Google Scholar 

  29. Monaghan, R. M. & Barraclough, D. Nitrous-oxide and dinitrogen emissions from urine-affected soil under controlled conditions. Plant and Soil 151, 127–138 (1993).

    CAS  Article  Google Scholar 

  30. Stevens, R. J. & Laughlin, R. J. Lowering the detection limit for dinitrogen using the enrichment of nitrous oxide. Soil Biology & Biochemistry 33, 1287–1289 (2001).

    CAS  Article  Google Scholar 

  31. Dalsgaard, T., Canfield, D. E., Petersen, J., Thamdrup, B. & Acuña-González, J. N2 production by the anammox reaction in the anoxic water column of Golfo Dulce, Costa Rica. Nature 422, 606–608, 10.1038/nature01526 (2003).

    CAS  ADS  Article  PubMed  PubMed Central  Google Scholar 

  32. Shen, J. P., Xu, Z. & He, J. Z. Frontiers in the microbial processes of ammonia oxidation in soils and sediments. Journal of Soils and Sediments 14, 1023–1029, 10.1007/s11368-014-0872-x (2014).

    CAS  Article  Google Scholar 

  33. Ligi, T. et al. The genetic potential of N2 emission via denitrification and ANAMMOX from the soils and sediments of a created riverine treatment wetland complex. Ecological Engineering 80, 181–190, 10.1016/j.ecoleng.2014.09.072 (2015).

    Article  Google Scholar 

  34. Trimmer, M. & Purdy, K. J. Evidence for the direct oxidation of organic nitrogen to N2 gas in the Arabian Sea. ISME Journal 6, 1798–1800 (2012).

    CAS  Article  Google Scholar 

  35. De Brabandere, L. et al. Vertical partitioning of nitrogen-loss processes across the oxic-anoxic interface of an oceanic oxygen minimum zone. Environmental Microbiology 16, 3041–3054, 10.1111/1462-2920.12255 (2014).

    CAS  Article  PubMed  Google Scholar 

  36. Laughlin, R. J. & Stevens, R. J. Evidence for fungal dominance of denitrification and codenitrification in a grassland soil. Soil Science Society of America Journal 66, 1540–1548 (2002).

    CAS  ADS  Article  Google Scholar 

  37. Müller, C., Laughlin, R. J., Spott, O. & Rütting, T. Quantification of N2O emission pathways via a 15N tracing model. Soil Biology and Biochemistry 72, 44–54 (2014).

    Article  Google Scholar 

  38. Leach, S. A., Thompson, M. & Hill, M. Bacterially catalyzed n-nitrozation reactions and their relative importance in the human stomach. Carcinogenesis 8, 1907–1912, 10.1093/carcin/8.12.1907 (1987).

    CAS  Article  PubMed  Google Scholar 

  39. Smith, N. A. & Smith, P. Nitrate reduction and n-nitrosation by obesumbacterium-proteus. Letters in Applied Microbiology 14, 61–64, 10.1111/j.1472-765X.1992.tb00648.x (1992).

    CAS  Article  Google Scholar 

  40. Jarvis, S. C. & Pain, B. F. Ammonia volatilisation from agricultural land. Proceedings of the Fertilizer Society 298, 1–35 (1990).

    Google Scholar 

  41. Watson, C. J., Kilpatrick, D. J. & Cooper, J. E. The effect of increasing application rate of granular calcium ammonium nitrate on net nitrification in a laboratory study of grassland soils. Fertilizer Research 40, 155–161 (1994).

    Article  Google Scholar 

  42. Ali, T. U., Kim, M. & Kim, D. J. Selective inhibition of ammonia oxidation and nitrite oxidation linked to N2O emission with activated sludge and enriched nitrifiers. Journal of Microbiology and Biotechnology 23, 719–723 (2013).

    CAS  Article  Google Scholar 

  43. Monaghan, R. M. & Barraclough, D. Some chemical and physical factors affecting the rate and dynamics of nitrification in urine-affected soil. Plant and Soil 143, 11–18, 10.1007/bf00009124 (1992).

    CAS  Article  Google Scholar 

  44. McCarty, G. W. & Bremner, J. M. Laboratory evaluation of dicyandiamide as a soil nitrification inhibitor. Communications in Soil Science and Plant Analysis 20, 2049–2065 (1989).

    CAS  Article  Google Scholar 

  45. Kammann, C., Mueller, C., Gruenhage, L. & Jaeger, H.-J. Elevated CO2 stimulates N2O emissions in permanent grassland. Soil Biology & Biochemistry 40, 2194–2205, 10.1016/j.soilbio.2008.04.012 (2008).

    CAS  Article  Google Scholar 

  46. Roobroeck, D., Butterbach-Bahl, K., Brueggemann, N. & Boeckx, P. Dinitrogen and nitrous oxide exchanges from an undrained monolith fen: short-term responses following nitrate addition. European Journal of Soil Science 61, 662–670, 10.1111/j.1365-2389.2010.01269.x (2010).

    CAS  Article  Google Scholar 

  47. Garber, E. A. E. & Hollocher, T. C. N-15, o-18 tracer studies on the activation of nitrite by denitrifying bacteria - nitrite water-oxygen exchange and nitrosation reactions as indicators of electrophilic catalysis. Journal of Biological Chemistry 257, 8091–8097 (1982).

    CAS  PubMed  Google Scholar 

  48. Clough, T. J., Kelliher, F. M., Sherlock, R. R. & Ford, C. D. Lime and soil moisture effects on nitrous oxide emissions from a urine patch. Soil Science Society of America Journal 68, 1600–1609 (2004).

    CAS  ADS  Article  Google Scholar 

  49. Okada, N., Nomura, N., Nakajima-Kambe, T. & Uchiyama, H. Characterization of the Aerobic Denitrification in Mesorhizobium sp. Strain NH-14 in Comparison with that in Related Rhizobia. Microbes and Environments 20, 208–215 (2005).

    Article  Google Scholar 

  50. Jones, D. L., Owen, A. G. & Farrar, J. F. Simple method to enable the high resolution determination of total free amino acids in soil solutions and soil extracts. Soil Biology & Biochemistry 34, 1893–1902, 10.1016/s0038-0717(02)00203-1 (2002).

    CAS  Article  Google Scholar 

  51. Long, A., Heitman, J., Tobias, C., Philips, R. & Song, B. Co-Occurring Anammox, Denitrification and Codenitrification in Agricultural Soils. Applied and Environmental Microbiology 79, 168–176, 10.1128/aem.02520-12 (2013).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  52. Jansson, S. L. & Persson, J. In Nitrogen in Agricultural Soils (ed F.J. Stevenson ) 229–252 (Soil Science Society of America, 1982).

  53. Sebilo, M., Mayer, B., Nicolardot, B., Pinay, G. & Mariotti, A. Long-term fate of nitrate fertilizer in agricultural soils. Proceedings of the National Academy of Sciences of the United States of America 110, 18185–18189, 10.1073/pnas.1305372110 (2013).

    ADS  Article  PubMed  PubMed Central  Google Scholar 

  54. Cameron, K. C. et al. Lysimeters without edge flow: an improved design and sampling procedure. Soil Science Society of America Journal 56, 1625–1628 (1992).

    ADS  Article  Google Scholar 

  55. WRB. IUSS Working Group—World reference base for soil resources 2006. (FAO, Rome, 2006).

  56. Di, H. J. & Cameron, K. C. The use of a nitrification inhibitor, dicyandiamide (DCD), to decrease nitrate leaching and nitrous oxide emissions in a simulated grazed and irrigated grassland. Soil Use and Management 18, 395–403, 10.1079/sum2002151 (2002).

    Article  Google Scholar 

  57. Selbie, D. R. et al. The effect of urinary nitrogen loading rate and a nitrification inhibitor on nitrous oxide emissions from a temperate grassland soil. Journal of Agricultural Science 152, S159–S171, 10.1017.S0021859614000136 (2014).

    Article  Google Scholar 

  58. Arah, J. R. M. Apportioning nitrous oxide fluxes between nitrification and denitrification using gas-phase mass spectrometry. Soil Biology & Biochemistry 29, 1295–1299, 10.1016/s0038-0717(97)00027-8 (1997).

    CAS  Article  Google Scholar 

  59. Mulvaney, R. L. Determination of n-15-labeled dinitrogen and nitrous-oxide with triple-collector mass spectrometers. Soil Science Society of America Journal 48, 690–692 (1984).

    CAS  ADS  Article  Google Scholar 

  60. Velthof, G. L. & Oenema, O. Nitrous oxide fluxes from grassland in the Netherlands .1. Statistical analysis of flux-chamber measurements. European Journal of Soil Science 46, 533–540, 10.1111/j.1365-2389.1995.tb01349.x (1995).

    Article  Google Scholar 

Download references


The authors are grateful for the funding that was provided through the Research Stimulus Fund Program administered by the Department of Agriculture & Food under the National Development Plan 2007–2013 RSF 07536. The first author is grateful for the funding provided by Teagasc through the Walsh Fellowship Scheme. The authors thank J. Murphy, D. Hennessey, K. McNamara and A. Cuddihy for their contribution to this research. All 15N analysis was conducted by the University of California Stable Isotope Laboratory, Davis, CA. The authors are grateful to Dr J. Luo, Dr C. de Klein, Dr M. Tourna and Dr P. Christie for their comments on the draft manuscript.

Author information




K.R. and G.L. were the principal investigators for the project funding. D.S., K.R., G.L., H.D. and K.C. designed the experiment. D.S. conducted the flux measurements and R.L. and C.W. carried out the N2 flux calculations. J.G. conducted the statistical analysis. C.S. provided additional supporting data. K.R., G.L., R.L. and D.S. drafted the manuscript and H.D., J.M. and K.C. assisted with the experimental design, supervised D.S. and together with T.C., contributed to the writing of the paper.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Electronic supplementary material

Rights and permissions

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Selbie, D., Lanigan, G., Laughlin, R. et al. Confirmation of co-denitrification in grazed grassland. Sci Rep 5, 17361 (2015).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing