N 2 O production in soils – Natural abundance approach

Hot spots have been proved to have a great importance in the overall N²O emissions from different ecosystems (Davidson & Schimel, 1995; Boyer et al., 2006; Savage et al., 2014; Cowan et al., 2015). Here we were mainly interested in identifying the dominant processes responsible for the N²O emission from hot spots from the soils included to this work. In that sense, the emission-weighted average SP value is more representative of the dominant processes responsible for the N²O emission, because it takes into account the contribution of individual SP values (possible different processes) related to the magnitude of the N²O fluxes. All the available data on emission - weighted average SP values of N²O emitted in situ (-6‰ to 38 %), including SP values from this work, are largely within the range of SP values reported from soil incubation studies, supporting the conclusion that it is better to compare the isotope results from field studies with the isotope values obtained in incubation studies instead of pure culture bacteria studies (Figure 4) (Perez, 2001; Well et al., 2005;

Opdyke et al., 2009; Ostrom et al., 2010; Toyoda et al., 2011; Kato et al., 2013; Yano et al., 2014). Moreover, most of these data, including the SP value from the BAF soils (9.7 ‰ ± 7.9 ‰) and VSC (1.5 ‰± 21.0 ‰), lie within the SP range found for soil denitrification in incubations studies (Figure 4) (Perez et al., 2006; corrected values;

Well et al., 2006; Well & Flessa, 2009; Meijide et al., 2010; Lewicka-Szczebak et al., 2014). While this was surprising for the fertilized agricutural soil (VSC), the high water content (WFPS > 60 %), the positive correlation between the N²O emissions and soil NO3- concentrations and N2O/NO ratios greater than one (Marquina et al., 2015), suggest indeed that denitrification is the main process responsible for N2O produc-tion in this cornfield.

The SP emission-weighted average value for the BP (-6.0 ‰ ± 8.1 ‰) is the lowest SP emission-weighted average value (i.e., isotopically depleted) reported from in situ studies, and is overlapping with the SP values for nitrification/nitrifier-denitrification (−16.8 ‰ ± 8.4 ‰) obtained by Perez et al. (2006) in soil incubation experiments.

However, as mentioned above, there is a great variation in the SP values reported from soil incubation studies, particularly for nitrification, and the exceptionally low value of Perez et al. (2006) for nitrification contrasts the results from pure culture and other soil incubations studies. It has been recently argued that the low SP values from Perez et al. (2006) may be due to the predominance of nitrifier-denitrification (Ostrom et al. 2010) a process similar to classical denitrification SP values around zero or less in pure cultures bacteria studies (Sutka et al., 2006, 2003, 2004).

It is possible that the N²O emissions in 2011 from the BP soils, which were consider-ably lower than the ones measured previously (Repo et al., 2009, Marushchack et al., 2011), were derived from nitrification and/or nitrifier-denitrification. Nitrification has shown to be important in the N²O production in non-saturated peat (Liimatainen et al., 2014). Nitrification has been proven to occur in these soils (Pitkämäki, 2010) and is supported by high nitrogen mineralization rates (Marushchack et al., 2011) and high NO^3- content in peat. On the other hand, these bare permafrost peat soils have a high denitrification potential and high abundance of denitrifiers (Palmer et al.,

2011). Anaerobic microsites may well occur in this peat allowing denitrification to take place simultaneously with nitrification. In wetter years with high N²O emissions from BP (Repo et al., 2009, Marushchack et al., 2011), the dominant process governing the N²O release from BP may well be different, e.g. denitrification.

Figure 4. Isotopic composition of N2O from in-situ soil emissions of natural soils (squares open;

closed; semi open) and agricultural soils (triangles and circle closed). Pink areas indicate the range values reported for nitrification/niftrifier denitrification from soil incubations studies, and blue areas indicates the range values reported for denitrification from soil incubations studies:

(a) Well et al., 2008; (b) Well et al., 2006; (c) Well & Flessa 2009; (d) Perez et al., 2006 cor-rected values/article I; (e)Meijide et al., 2010; (f) Lewicka-Szczebak et al., 2014. The closed stars show the δ¹⁵N^bulk, SP and δ¹⁸O values of tropospheric N2O (Park et al., 2012).Data points are average, emission-weighted average or individual measurements. References: Sub-Arctic permafrost peatland (BP, article II); Alpine grassland (Kato et al., 2013); Brazilian Amazon forest (BAF, article I; Park, 2005);Temperate region: (1-2) Cropland/rice paddy (Yano et al.,

2014); (3) Grassland cultivated (Ostrom et al., 2010); (4) Grassland/ urine amendment (Ya-mulki et al., 2001); (5) Grassland/ KNO3 amendment (Bol et al.,2003); (6) Grassland/ aban-doned arable field, (7) Cropland/Soybean-conventional tilled/ unfertilized, (8) Cropland/Corn- conventional tilled/ manure amendment (Opdyke et al., 2009); (9) Cropland/ Paddy rice, soy bean, upland rice/ urea amendment (10) Cropland/Vegetables / poultry manure, pelleted poul-try manure amendment, (11) Cropland/Vegetables / (NH4)2SO4 amendment, (12) Cropland/Vegetables / poultry manure, pelleted poultry manure, (NH4)2SO4 amendment (Toyoda et al., 2011); (13) Sub-tropical region: Cropland/ Wheat, urea amendment (Perez et al., 2001); (14) Tropical savanna: Cropland/ Corn, NPK amendment (VSC; article I). (Modify from article II).

Our results suggest that denitrification is the most important microbial process for N²O production in the three soil profiles The SP values for N²O measured in the soil profiles ranged from 3 ‰ to 30 ‰ for the BP, -3.5 ‰ to 21‰ for VSC and 1 ‰ to 21 ‰ for BAF soils (Figure 3 in article I and II). The SP values of our two soil profiles and the BAF profile (Park 2005) overlap with the SP values obtained in soil incubation studies for denitrification (1‰ to 21‰) (Perez et al., 2006; corrected values; Well et al., 2006; Well & Flessa, 2009; Meijide el al., 2010; Lewicka-Szczebak et al., 2014). They are within the ranges published for soil profiles of natural ecosystems (Koehler et al., 2012; Kato et al., 2013). Denitrification dominance is also supported by the soil properties including high WFPS, high NO^3- concentration and a positive correlation between water content and N²O concentrations.

However, there are few SP values observed that fall outside the range of SP values for denitrification (e.g -3.5 ‰ in the VSC and + 30 ‰ in BP). This trend can be ex-plained by the concentration fluctuations in depth in the soil profile. In the VSC site, the low SP values were measured after fertilization (with NH⁴NO3) when the N²O concentration was the highest (∼15000 ppb), indicating possible changes in the rela-tive contribution of the N²O production processes and enhancement of nitrification (or nitrifier-denitrification) and denitrification after fertilization. The high SP values observed from the BP soil profile corresponded with a decline in N2O concentration in this profile, and might be attributed to N²O reduction to N². In the BP profile, nitrogen and oxygen in N2O were generally more enriched in ¹⁵N and ¹⁸O near the surface than deeper in the soil profile (Figure 3 in article II). The negative correlation of N2O concentration with δ¹⁵N^bulk and δ¹⁸O over the profile, as well as the lowest δ¹⁵N^bulk value measured generally at the depth with the highest N2O concentration, suggest that N²O reduction to N² was the main factor controlling the N²O isotopic shifts with depth in these soils. N²O reduction could also occur in the BAF soils but to a lesser extent (see below).

SP values of N²O emitted from the BAF and VSC soils were generally in the range of SP values measured in the soil profiles, same trend has been observed before in other soils (Koehler et al., 2012; Kato et. al, 2013). Surprisingly, for BP soil there was a large difference between the surface SP values and the SP values deeper in the soil profile (Figure 3 in article II). This difference between the surface and deeper SP values could be attributed mainly to following factors: (1) high production of N²O at

the surface, (2) different sources of N²O at the surface and deeper soil profile, (3) low reduction of N²O to N² at the surface and (4) deeper N²O had minor contribution to the N²O emissions resulting from slow diffusion rate of N²O and reduction to N² in the soil profile (see below).