Methods

The methods used in this study are listed in Table 1. These were designed to study all aspects of the processes involved in N²O emissions. A short overview of techniques applied and measuasurements conducted are given below, the details are shown in the articles included in the apendix.

2.2.1 Emission rates of nitrous oxide (N2O)

N²O emission rates were determined at defined time points using the static chamber technique (Heikkinen et al, 2002). N²O fluxes were calculated based on the linear change in N²O concentration during the 40 min of chamber closure and reported as area-based flux rate. Several soil parameters were measured simultaneously, including soil temperature, soil moisture, depths of water table and active layer. An ambient gas sample was also taken associated to the chamber measurements, at about 2 m above the soil surface to determine concetration and isotopic composition of atmospheric N2O. The concentration of N2O was measured by gas chromatography (GC ) System, equipped with a (⁶³Ni)-electron capture detector.

2.2.2 Sampling and isotopic composition of N2O emitted

The sampling for the isotope analysis was performed at the same time as the sampling for the flux measurements. Gas samples for the isotope analysis were collected from the chamber 60 minutes after enclosure (surface) and at defined depth (soil profile) using evacuated 500 mL stainless steel cylinders and a drierite/ascarite trap system, following the protocol in article I. All measurements of δ¹⁵N^bulk, ¹⁵N site preference (SP), and δ¹⁸O of N²O were performed at the University of California, Berkeley, using a Finnigan MAT 252 isotope ratio mass spectrometer (IRMS) operated in continuous flow mode, coupled with an online Finnigan preconcentrator and gas chromatograph (Thermo Finnigan, Bremen, Germany).

Isotope compositions in article I and II are reported using δ notation:

δX = (R^sample / R^standard – 1) x 1000 ‰ (1)

for which R is the ratio of the heavy to the light isotope and X = ¹⁵N, ¹⁸O, ¹⁵N^bulk, ¹⁵N^α or ¹⁵N^β and the 'standard' is atmospheric N² for δ¹⁵N, δ¹5N^α and δ¹⁵N^β, with the latter on the Toyoda and Yoshida (1999) calibration, (Croteau et al., 2010) and Vienna Standard Mean Ocean Water (VSMOW) for δ¹⁸O. The single measurement precision (1σ) on whole air samples containing ∼1.35 nmol of N2O was ± 0.2‰ for δ¹⁵N^bulk and δ¹⁸O and ± 0.8‰ for δ¹⁵N^α (Park et al., 2012).

2.2.3 Gas sampling from the peat profile

For soil profile sampling pits were dug in the study sites. The size of the pits and the sampling depth varied between the sites. In theVenezuela corn field site (VSC), we dug 1 pit of 1.9 m depth in which we inserted 2 m of 1/8 in. O.D. stainless steel tube probes at 10, 25, 50, 75 and 100 cm on one wall of the pit (see full description in article I). In the Russian tundra site, we dug 3 pits in BP and 3 pits in the vegetated peat (VP) soils and gas collectors (perforated PVC tubes, diameter 5 cm, and volume 250 cm³) were inserted horizontally into the pits walls at depths of 5, 10, 25, 35 and 45 cm and the pits were then closed. In addtion sensors were installed in one soil profile from BP and VP to measure soil temperature, soil moisture and oxygen (see full description in article II).

Gas and soil samples were collected from the soil profiles to determine the concentration and stable isotopic compositions of N²O, mineral N (NH⁴⁺ and NO^3-), total carbon (TC) and total nitrogen (TN). Samples from the soil profile were collected on the same day as the gas emissions.

2.2.4 Soil nitrogen analyses

Soil samples were taken from near the flux chambers for mineral nitrogen analysis (NH4+ and NO3−; n=3) and analysis of δ¹⁵N in NH4+ and NO3−. This sampling was done simultaneously with the gas sampling in order to enable calculation of the instantaneous isotope enrichment factors for ¹⁵N in N2O (Mariotti et al., 1981).Values of δ¹⁵N for NH⁴⁺ and NO^3- from the natural abundance approach and the ¹⁵N tracer experiement were determined using the microdiffusion method (Herman et al., 1995).

Briefly, for analysis of atom% ¹⁵N (at %) of NH⁴⁺, the soil extract was placed in a closed container and MgO was added to buffer the solution to yield a pH of ~10.5.

Volatilized ammonia (NH³) was trapped on acidified filter paper (Whatman 589/3, ashless) during an incubation period of eight days at 30 °C with shaking at 200 rpm.

For the measurements of at % ¹⁵N of NO^3-, after the removal of NH⁴⁺ following the procedure above, Devarda's alloy was added to convert NO^3- to NH⁴⁺ which was then trapped similarly to ammonia. After incubation, the filter paper was removed from the solution, dried for 24 hours over an atmosphere of concentrated H²SO⁴ in a des-iccator, and wrapped in a tin capsule. The filters were then analyzed at the University of Eastern Finland by an elemental analyzer coupled to an isotope ratio mass spec-trometer (EA-IRMS), which included a ThermoFinnigan DELTA XP Plus IRMS, Flash EA 1112 Series Elemental Analyzer, and a Conflow III open split interface (Ther-moFinnigan, Bremen, Germany). The ¹⁵N data were expressed as δ¹⁵N (‰) for natural abundance soil samples and at% ¹⁵N excess relative to the natural abundance ¹⁵N content of NO^3- and NH⁴⁺ in the soils for the ¹⁵N labeling experiment. Dried bulk soil samples were also analyzed for total N and ¹⁵N content using the same EA-IRMS, and at% ¹⁵N excess values were calculated as described above (and see equation 2

below). The reproducibility of 10 standard runs was typically better than 0.5‰ (1σ, n=10).

2.2.5 Reduction of N2O to N2 and quantifying N2O consumption in the soil profile

It has been proposed that significant correlations between pairs of isotope composi-tions for N2O (i.e., for δ¹⁸O, δ¹⁵N^Bulk, δ¹⁵N^α) can be indicative of N2O reduction to N2

(Ostrom et al., 2007; Jinuntuya-Nortman et al., 2008; Opdyke et al., 2009; Well & Flessa, 2009). In this work, we proposed that N2O reduction to N2 should primarily affect

15N at the central (α) position and ¹⁸O, due to primary kinetic isotope effects during the breaking of the N-O bond, and the remaining N2O will become isotopically en-riched in oxygen and in nitrogen at the central (α) position. Whereas a potential sec-ondary ¹⁵N isotope effect at the terminal (β) position should be much smaller. Thus, simultaneous enrichments in ¹⁸O and in ¹⁵N at the central nitrogen of N²O should occur if N²O consumption in soils is occurring to a significant degree. Consequently, a positive correlation between δ¹⁵N^α and δ¹⁸O for samples taken throughout the soil profile is expected, while a negligible correlation between δ¹⁵N^β and δ¹⁸O is expected, if N²O is being reduced to N². These correlations together were shown for BP soil profiles (see resuls & discussion section), but not for the Venezuelan cornfield (VSC), suggesting that N²O may indeed be affected by reduction in the natural soils. Hence, the fraction of N²O originally produced and then consumed by reduction can be es-timated using these isotopic compositions. The approach used is a box model for the soil profiles data set (0 – 45 cm; BP) and is described in detail in article I. The meth-odology assumes that there is no downstream flow (i.e., towards greater depths), diffusion is the only mechanism for gas transport, the N²O flux to the surface is not affected by the concentrations of other gases at different depths, and that the soil atmosphere is in isobaric equilibrium with the atmosphere at the surface. In addition, a Rayleigh distillation equation (2) was used to describe the ¹⁵N^α and ¹⁸O isotopic compositions of the remaining N2O (Mariotti et al., 1981):

− = × ln (2)

for which X is either ¹⁵N or ¹⁸O and the subscript 0 indicates initial values for both concentration and isotopic composition. ε is the isotopic enrichment factor (‰) and describe isotopic discrimination during a single unidirectional reaction or the rate limiting step of a multi-step reaction (Jinuntuya-Nortman et al., 2008). In nature, most biological reactions are multi-step reactions, so the isotopic discrimination during a particular process is the net result of the fractionation along a reaction series and is often described by a net isotope effect (NIE) (η) (Jinuntuya-Nortman et al., 2008).

Therefore, in this work, net isotope effect (NIE) (η) is used in equation (2) instead of the isotopic enrichment factor (ε):

− = × ln (3)

The N²O concentration in the numerator in (2-3) corresponds to the remaining N²O, therefore the fraction of N2O consumed (f) can be estimated by (4):

f = 1 − (4)

The net isotope effect (η^x) can then be estimated in the following manner. First, the slope of δ¹⁵N^α versus δ¹⁸O using data from the soil profiles (Figure 4 in results &

discussion section) yields the ratio of η¹⁵N^α/η¹⁸O given in (5):

= 1.3 ± 0.2 (5)

We then assume an average value for η ¹⁸O for reduction of N²O of ~ −15‰, based on values reported in the literature (Menyailo and Hungate, 2006; Ostrom et al., 2007; Vieten et al., 2007; Lewicka-Szczebak et al., 2014; Lewicka-Szczebak et al., 2015), and calculated the net isotope effect of ¹⁵N^α (η ¹⁵N^α) for reduction of N²O spe-cific for BP soils, using equation 5. The η ¹⁵N^α for reduction of N²O for the BP is ~ −19

‰. In addition, the initial values for the isotopic composition can be inferred from the y-intercepts of Keeling plots – i.e., from plots of the isotopic compositions against the inverse of the concentrations for the species of interest. Thus, = −6.7 ± 1.2 ‰ and = 23.6 ± 0.5 ‰.

2.2.6 ¹⁵N tracer experiment

The study took place at the Arctic Russian tundra site during the growing season 2010, starting on July 21 and ending on August 18 (24 days elapsed time). The ¹⁵N tracer experiment was conducted in situ on BP and adjacent VP in replicates of three (n=3). The tracing experiment comprised three different ¹⁵N labelling treatments:

NH415NO3 (Treatment 1; T1), ¹⁵NH4NO3 (Treatment 2; T2) and ¹⁵NH415NO3 (Treatment 3; T3) with each of the labels applied at an enrichment level of 98 at% ¹⁵N. The single and double ¹⁵N- labelling approach was used to quantify N2O emissions produced by denitrification from treatment with ¹⁴NH4¹⁵NO³ (T1) and total nitrification from the difference in ¹⁵N2O between the ¹⁵NH4¹⁵NO3 (T3) and ¹⁴NH415NO3 (T1) treatments.

To account for nitrification-coupled denitrification, ¹⁵NH⁴¹⁴NO³ treatment (T2) was also added (Baggs et.al, 2003; Bagss & Blum, 2004; Wrage et al., 2005).

The ¹⁵N solutions were added to the soil in situ using the virtual core injection technique proposed by Rütting et al., ( 2011b). Emission rates of ¹⁵N²O were deter-mined at the defined time points with the static chamber technique and their isotope content was determined at the Stable Isotope Facility at the University of California, Davis, using a Delta V Plus isotope ratio mass spectrometer (IRMS) operated in con-tinuous flow mode coupled with an online pre-concentrator and a GasBench. ¹⁵ N-N²O flux rates were calculated from the at% excess of the samples, after subtracting the natural abundance at% ¹⁵N values of N²O (i.e., 0.3663 at% ¹⁵N). The ¹⁵N content of mineral N forms were determined using the microdiffusion method (Herman et al., 1995) described above.

In addition to gas (N²O) and soil samples, plants and roots were quantitatively sampled from the labelled VP plots by manually picking them before soil extractions.

Plants were classified into higher and lower plants. The plants and roots were oven dried and weighted at the field and the aboveground biomass and roots were milled to fine powder (RetschMM301, Haan, Germany) in the laboratory. The samples were processed at the University of Eastern Finland (UEF).for total N and ¹⁵N content de-termined by the same EA-IRMS system describe above for soil N analysis. The ¹⁵N data from plants and roots together with ¹⁵N in mineral soil N, ¹⁵N²O and total ¹⁵N in soil was used for the mass balance.

Table 1. Methods used in this thesis. Detailed descrption of the methodos are provided in the article and manuscript.

Method Article

Static chamber measurments I, II, manuscript

Soil gas characterization I, II

Bulk 15N and ¹⁸O I, II

Site preference analysis (SP) I, II Double ¹⁵N- labeled ammonium nitrate method Manuscript Microdiffusion (¹⁵N mineral N ) I, II, manuscript