Non-microbial methane formation in oxic soils

Abstract. Methane plays an important role as a radiatively and chemically active gas in our atmosphere. Until recently, sources of atmospheric methane in the biosphere have been attributed to strictly anaerobic microbial processes during degradation of organic matter. However, a large fraction of methane produced in the anoxic soil layers does not reach the atmosphere due to methanotrophic consumption in the overlaying oxic soil. Although methane fluxes from aerobic soils have been observed, an alternative source other than methanogenesis has not been identified thus far. Here we provide evidence for non-microbial methane formation in soils under oxic conditions. We found that soils release methane upon heating and other environmental factors like ultraviolet irradiation, and drying-rewetting cycles. We suggest that chemical formation of methane during degradation of soil organic matter may represent the missing soil source that is needed to fully understand the methane cycle in aerobic soils. Although the emission fluxes are relatively low when compared to those from wetlands, they may be important in warm and wet regions subjected to ultraviolet radiation. We suggest that this methane source is highly sensitive to global change.


Introduction
Traditionally, biogenic methane (CH 4 ) was thought to be formed only by methanogens under strictly anaerobic conditions in wetland soils and rice paddies, intestinal tracts of termites and ruminants, human and agricultural waste.However, Keppler et al. (2006) demonstrated that plants produce CH 4 under aerobic conditions.Subsequently, this possibility has been critically debated (Dueck et al., 2007;Ferretti et al., 2007;Kirschbaum et al., 2007;Keppler and R öckmann, 2007;Vigano et al., 2008;Wang et al., 2008;Nisbet et al., 2009;Keppler et al., 2009;Beerling et al., 2008) and some researchers have suggested alternative explanations for the observed release of CH 4 Introduction

Conclusions References
Tables Figures

Previous observations of methane formation in aerobic soils
Whilst aerobic soils are considered to be net CH 4 sinks due to methanotrophic oxidation of CH 4 it has been shown that oxic upland forest soils produce CH 4 .Although observations of CH 4 production in oxic soil are numerous (Megonigal and Guenther, 2008;Hao et al., 1988;Andersen et al., 1998;von Fischer and Hedin, 2007), all have been attributed to methanogenesis.Methane production by oxic eubacteria (Rimbault et al., 1988) and anaerobic microsites, a refuge for methanogens (Peters and Conrad, 1995), were offered as possible explanations even though CH 4 production from eubacteria could only be detected in trace quantities.In experiments by Kammann et al. (2009) soil cores emitted up to 6 µg CH 4 per core even after homogenization, which may be expected to lead to the destruction of anoxic microsites.Von Fisher and Hedin (2007) using stable carbon isotope studies showed that our understanding of CH 4 formation in oxic soils is incomplete and discussed that methanogens as the sole source for CH 4 in oxic soils should be critically reviewed.Introduction

Conclusions References
Tables Figures

Back Close
Full

Possibility of non-microbial methane formation in soil
In this study we tested the previously postulated hypothesis that non-microbial CH 4 formation occurs in soils (Jugold and Keppler, 2009;Hurkuck et al., 2012).Following preliminary observations, we undertook a series of experiments measuring CH 4 formation from soils (see Table 1 and methods section) as a function of temperature, water content and UV-B irradiation.We used five different soils, including one highly organic soil (referred to as peat, Table 1), which had been lyophilised and homogenized prior to the experiments.Humic acid and lignin were used as alternatives for soil organic matter.Additionally, sub-samples of peat and lignin, sterilised using gamma radiation or autoclaving, were also used in our investigations.

Origin of samples and preparation
Four soils and one peat type were used.If present, stones and larger wood particles were removed from the samples before they were lyophilised and then milled using an electronic coffee grinder (Elta UM105).
Soil SL was sampled at the Lerchenberg forest south of Mainz, Germany (49 • 57 47 N, 8 • 11 01 E).The sampling site is a deciduous forest dominated by beech trees (Fagus sylvatica), featuring few oaks and nearly no undergrowth.The sample was collected from the surface after brushing away the layer of leaf litter.
For soil SG the upper 10 cm of a pine forest soil was sampled at Mainz-Gonsenheim, Germany (50 • 0 24.4 N, 8 • 11 50.3E).The soil in this area is rich in medium to coarse sand and powdery clay particles.It also contains rotting wood debris, pine twigs and is densely rooted.Introduction

Conclusions References
Tables Figures

Back Close
Full Soil SHA was topsoil of a terra fusca sampled at the "Nationalpark Hainich", Germany ( 51• 04 46 N, 10 • 27 08 E).The sampling site is a deciduous forest dominated by beech trees.Soil SW was collected from the organic rich O-horizon of a deciduous forest soil.The vegetation is dominated by beech trees.The sampling site is situated south of Minden, Germany (52 • 15 17.4 N, 8 • 52 29.5 E).
Peat PH was sampled at the peat bog "Großes Torfmoor" near Hille, Germany (52 • 19 23.7 N, 8 • 42 34.7 E).The top 10 cm of sphagnum peat was collected as a bulk sample.A subsample was sterilised using gamma irradiation.

Reaction vials
Samples were incubated in glass vials (360 ml); made in-house by modification of a 300 ml Erlenmeyer-flask (Duran group) fitted with the neck of a 40 ml screw top vial (Supelco) sealed with a hole type screw cap (Supelco) containing a PTFE/silicone septum (Supelco).The UV reaction chambers were also custom built; 200 ml glass chambers with a quartz glass lid and a septa sealed side port for headspace sampling.
The irradiated surface was 19.63 cm 2 .

Determination of organic carbon
Organic carbon content of the samples was determined with a SC Analyser (SC-144 DR, LECO) by combustion of 0.1-0.5 g of sample material at 1300 • C. The carbon content was calculated by comparison to a calcium carbonate standard.For soil SW the organic carbon content was determined by loss on ignition.Therefore the weight loss after two hours at 600 Full

Methane measurements
Headspace above samples in the sealed vials were sampled (5 ml) with a Hamilton gas syringe and analysed using a gas chromatograph (Shimadzu GC-14B) with flame ionization detector (GC-FID).Two reference CH 4 standards (containing 8.905 ppm and 1.736 ppm) were used.

Statistical methods
The statistical comparison of different samples was examined with the open-source software "The R Project for Statistical Computing", version 2.11.1 (The R Foundation for Statistical Computing).

Temperature dependence
Sets of non-sterile and sterile peat samples (PH, 5 g per 360 ml screw cap vial, n = 5) as well as non-sterile sets of each soil sample were incubated for 24 h at temperatures ranging from 30 to 90 • C at 10 • C intervals.At the end of the incubation period a sample of the vial headspace was analysed for CH 4 content.

Drying-rewetting cycles
Peat PH (5 g in 360 ml screw cap vials, n = 5) was incubated for 24 h at either 30, 40 or 50 • C. Another set of samples was incubated under the same conditions but supplemented with 5 ml doubly distilled water.After incubation a sample of the headspace was analysed for CH 4 content.The samples were frozen and lyophilised again directly after measurements.After being rewetted and incubated again, headspace samples were analysed again for CH 4 .This cycle was repeated five times.Introduction

Conclusions References
Tables Figures

Back Close
Full In a further experiment dependence of CH 4 release on the water-sample-ratio was investigated.For this samples of peat PH (5 g in 360 ml screw cap vials, n=5) were supplemented with 1, 5 and 10 ml doubly distilled water.

Experiments with H 2 O 2
Samples PH or SHA (5 g in 360 ml vials, n = 3) and 10 ml aqueous solution with varying concentrations of H 2 O 2 (0-25 mM) were added and vials immediately sealed.The samples were incubated for 24 h at 30 • C, after which a sample of the vial headspace was analysed for CH 4 content.The experiment was also repeated for lignin and humic acid with 25 mM H 2 O 2 .

UV irradiation experiments
An Osram Ultra-Vitalux lamp (300 W) served as UV source.The radiation of this lamp shows an UV-A/UV-B content comparable to solar radiation when the source is located at the appropriate distance.The total unweighted UV-B radiation was determined with a UV radiometer (UVlog, sglux, Berlin, Germany) precalibrated for the used lamp type.
For more details of the lamp characteristics we refer to Vigano et al. (2008).The UV lamp was placed above the leak tight UV reaction chambers.The height was adjusted so as to set the UV-B intensities to the desired value between 1 and 4 W m −2 .To exclude undesired UV-C radiation the quartz glass lids were covered with a 95 nm film of cellulose diacetate.Two fans were employed in order to keep the temperatures in the chambers at 30 • C (±2 • C).Temperature was monitored with a thermocouple.All experiments were conducted with 2-5 g of sample material but the data is presented based on irradiated area rather than sample weight.Methane concentrations in the headspace were measured after 0, 24 and 48 h.The difference between 0 and 24 h was used to calculate emission rates.The emissions induced solely by UV-B were calculated by subtracting the CH 4 concentration measured for the control samples from that measured for the UV irradiated Introduction

Conclusions References
Tables Figures

Back Close
Full samples so as to eliminate the temperature effect.The temperature monitored in the vials during UV experiments ranged from 28 to 32 • C. The control samples, which were also placed under the UV lamp, but covered with UV-opaque glass, showed emissions (transferred to ng g −1 (dw) h −1 ) comparable to those observed for the temperature experiments which were incubated in the dark at similar temperatures.

Isotopic data
δ 13 C sample analysis was carried out using gas chromatography combustion isotope ratio mass spectrometry (GC-C-IRMS) which consisted of a cryogenic preconcentration unit directly coupled to an HP 6890N gas chromatograph (Agilent, Santa Clara, USA), which was connected to a Delta PLUS XL isotope ratio mass spectrome- All 13 C/ 12 C-isotope ratios are expressed in the conventional δ notation in per mil versus V-PDB, defined as (Eq.1): 3 Results

Temperature dependence
The first experiment was designed to determine the temperature dependence and the required activation energy of CH 4 formation in a deciduous forest soil (SL), a coniferous forest soil (SG) and a sphagnum peat sample (PH).Samples were incubated at temperatures ranging from 30 to 90 • C. Methane emissions reached 7.11±0.59ng g −1 (dw) h −1 , 1.19±0.15ng g −1 (dw) h −1 and 1.12±0.16ng g −1 (dw) h −1 at 90 • C for PH, SG and SL, respectively (Fig. 1).Whereas CH 4 release could be observed for PH and SL at 30 • C and 40 • C respectively (Table 1), CH 4 release from SG was only measureable above 50 • C. Soil SHA which had a similar organic carbon content to soils SL and SG (Table 1) was also investigated and CH 4 emissions of 0.45 ± 0.02 ng g −1 (dw) h −1 at 70 • C were observed.For all samples the temperature curves showed an exponential increase of CH 4 emissions with temperature.Interestingly, the results found for the soil and peat samples (Fig. 1) showed a similar pattern to those reported by Keppler et al. (2006) and Vigano et al. (2008) for heated plant matter.Whereas biotically mediated reactions usually have their optimum temperatures between 25 and 40 • C (Dunfield et al., 1993) the observed strong increase in CH 4 emissions over the whole temperature range from 30 to 90 • C supports a chemically driven process.Furthermore, sterile peat samples (exposed to γ-radiation) showed similar or slightly higher emissions of CH 4 when compared to untreated peat samples.The fact that the emissions were not reduced in the sterile sample is further evidence for a non-microbial pathway.The slightly higher emissions observed for some of the

Conclusions References
Tables Figures

Back Close
Full sterile samples may possibly be ascribed to CH 4 production during the sterilisation process.
Since humic substances are usually the main constituents of organic-rich soils commercially available lignin and humic acid were investigated for CH 4 release.These substances, with an organic carbon content of 49.5 % and 43.5 % respectively, when similarly heated up to 90 • C showed even higher CH 4 emissions (at 30 • C 0.1 ± 0.01 ng g −1 (dw) h −1 for lignin and at 90 • C 18.3 ± 0.4 and 6.6 ± 0.9 ng g −1 (dw) h −1 for lignin and humic acid, respectively) than the organic rich soil PH.The similar dependence of CH 4 formation in soils and organic soil components on temperature strongly suggests that the organic soil fraction is the source of CH 4 thermally produced in soils.
The experimental data obtained from samples SL, SG and PH were used to draw Arrhenius plots for CH 4 formation (Supplement Fig. S1).The activation energies (E a ) for CH 4 formation, calculated from these plots, yielded values of 50.1 kJ mol −1 for SL, 77.5 kJ mol −1 for SG and 79.2 kJ mol −1 for PH.These activation energies, being higher than 50 kJ mol −1 , provide supportive evidence of an abiotic process (Sch önknecht et al., 2008).Since adsorption/desorption processes of CH 4 can occur with organic materials, it was considered that in this instance desorption might explain the observed emissions upon heating of the soil samples.Therefore, a series of experiments were performed to test such a possibility.From these it was found that a desorption process did not give rise to significant CH 4 fluxes from any of the soil samples employed in this study except when exceptionally high levels of CH 4 were added (12 500 ppm, see Supplement).These results are in accordance with the findings of Kirschbaum and Walcroft (2008) who reported no significant desorption of CH 4 from plant matter and concluded that desorption is not a quantitatively important artefact contributing to observed aerobic CH 4 fluxes in dry plant leaves.Introduction

Conclusions References
Tables Figures

Back Close
Full

Effect of wetting and drying
Many surface soils and sediments are frequently subjected to changing precipitation and evaporation conditions and as a consequence undergo changes in water content.
In extreme cases these conditions range from droughts to flooding events, including anthropogenic influences on the water budget like damming rivers or drainages for land reclamation.It is therefore important to study the effect of sample water content on the release of CH 4 .This was investigated in an experiment where soil samples were exposed to repeated cycles of wetting and drying.The sample PH emitted up to five times more CH 4 after addition of water, compared to the dried sample when incubated at the same temperature (Fig. 2).Interestingly, this increase appeared to be independent of the amount of water added, when the water content of the sample was in the range of 17 to 67 %.In a succession of five wetting-drying cycles no decline in CH 4 release rate was observed.A highly significant rise in emissions was noted with increasing temperature (p < 0.001).Emissions from dry samples doubled when the temperature was increased from 30 to 50 • C and a similarly strong effect was also observed for the wetted samples at these temperatures.
To rule out the influence of CH 4 consuming bacteria on our findings, a selection of measurements was repeated after addition of difluoromethane (DFM) (Miller et al., 1998) as described in the supplementary section.No differences were observed between samples with and without added DFM.Considerable CH 4 emissions could also be detected after wetting samples of lignin and humic acid, where, respectively, 1.9±0.2 and 3.1 ± 0.3 ng g −1 (dw) h −1 were released (Table 1).

Effect of hydrogen peroxide
Reactive oxygen species (ROS) such as hydroxyl radicals (HO in many degradation processes in soils, being abundant due to its release by roots, soil bacteria and white rot fungi (Frahry and Schopfer, 1998;Kersten and Kirk, 1987).We therefore investigated the influence of H 2 O 2 on CH 4 emissions from peat PH and soil SHA.
Interestingly, it was found that peat and soil responded rather differently following addition of H 2 O 2 .A strong increase in CH 4 emissions and a linear relationship (R 2 = 0.99) with increasing amounts of added H 2 O 2 to sample PH (Fig. 3) was observed whereas for soil sample SHA no additional emissions were observed.It is not clear why the soil and peat samples behaved so differently to the addition of H 2 O 2 .One possible explanation might be related to the differences in the composition of soil SHA and peat PH.Peat consists mostly of organic matter and low mineral content which might make it more prone to be attacked by ROS.Soil, on the other hand, contains other major components such as clay minerals and metal oxides that might more efficiently interact with H 2 O 2 .Samples of lignin and humic acid were also treated with H 2 O 2 .Whereas increased CH 4 emissions were observed for humic acid, no elevated emissions were found for lignin.Thus it is evident that the structural composition of the organic matter in soil has a major impact on the CH 4 emissions.

Effect of ultraviolet radiation
Ultraviolet (UV) radiation has been shown to be an important factor for aerobic production of CH 4 from plant tissues and pectin.It was demonstrated that both UV-A (320-400 nm) and UV-B (280-320 nm) induce CH 4 emissions from plant tissue (Vigano et al., 2008;McLeod et al., 2008), with UV-B radiation showing a much stronger effect.Nevertheless, because average UV-A intensities are around 30-fold higher than UV-B values, UV-A is also an important component on a global level for UV induced CH 4 emissions (Bruhn et al., 2009).Thus, the effect of UV radiation on the formation of CH 4 from soil was evaluated.For most experiments we used a total UV-B irradiance of 2 W m −2 , Introduction

Conclusions References
Tables Figures

Back Close
Full typical for mid-latitudes at the surface.In the tropics, where the UV-filtering ozone layer is thinner, ambient UV-B irradiances are about 3.7 W m −2 (Bernhard et al., 1997).
Measurements at 2 W m −2 UV-B and temperatures of 28 to 32 • C showed emissions of 0.25 to 4.92 µg m −2 h −1 (Table 1), which were linear over two a day period.Methane emission rates were also found to be a function of UV-B intensity.With increasing intensities from 1 to 4 W m −2 CH 4 emissions from soil SL increased linearly from 1.33± 0.22 to 7.28 ± 2.75 µg m −2 h −1 .Emissions from soil SG increased from 0.56 ± 0.12 to 2.75 ± 0.69 µg m −2 h −1 over the same intensity range (Fig. 4).
The combined emission rates under the influence of UV and temperature are similar to those reported for plant foliage (Vigano et al., 2008;Keppler et al., 2008).Interestingly, variations in CH 4 emissions under UV are not correlated to soil organic content (Table 1).However, the emission rates might be influenced by organic photo sensitizers, which have been shown to have a positive effect on CH 4 emissions from pectin (Messenger et al., 2009), or by clay minerals, often described as photo-catalysts (Katagi, 1990;Wu et al., 2008;Kibanova et al., 2011)

Stable carbon isotope composition of methane emitted from soil
In addition to CH 4 emission rates, the stable isotope composition (δ 13 C values) of the released CH 4 from soil SHA, peat PH, humic acid and lignin were also measured.for humic acid.In summary, the δ 13 C values of CH 4 emitted from soil differed between substrates and experimental conditions and ranged from −35.5 to 69 ‰ whereas the δ 13 C values for the organic matter of the bulk soil samples were in the range of −22 to −29 ‰ .Thus, it appears that all treatments caused substantial fractionation between the precursor carbon and emitted CH 4 .Similar δ 13 C values and isotope fractionations have been reported for CH 4 emitted from plant foliage due to UV radiation or upon heating (Vigano et al., 2009).Both the isotopic values reported for the chemical formation of CH 4 from soil and vegetation are commonly also found for terrestrial biogenic sources (Vigano et al., 2009).

Conclusions and outlook
Our study shows that there exist several hitherto unknown processes that produce CH 4 in soil and peat which is clearly not related to methanogenic activity.Figure 5 summarizes our results regarding non-microbial CH 4 formation in the aerobic layers of soils and the environmental factors that might control emissions.From our findings we suggest that the abiotic formation of CH 4 through degradation of organic soil matter represents a thus far undiscovered pathway for CH 4 formation in oxic soils.Our results imply that there are at least two different mechanisms for non-microbial CH 4 formation in soils.This can be best distinguished by comparing thermal and UV-B induced CH 4 release.Samples that released only minor amounts of CH 4 when heated or wetted emitted significant amounts when irradiated with UV-B, and vice versa.
The amounts of CH 4 produced at ambient temperatures of 30 • C are small but increase considerably with increasing temperature.Wetted samples during the drying and rewetting cycle experiments showed much higher emissions than the dry sample itself at low temperatures.Assuming that the first five centimetres of the soil horizon account for most of the CH 4 production, the emission rates from dry and wet soil at 30 to 40 • C (Table 1) would correspond to emission rates of 0 to Introduction

Conclusions References
Tables Figures

Back Close
Full peat (Minkinnen and Laine, 1998).These emissions increase up to an order of magnitude when the soil surface temperature reaches 50 to 70 • C.Although these temperatures are often only observed at soil surfaces in tropical and savannah regions, when compared to field measurements from wetlands with observed CH 4 emissions up to 11.9 mg m 2 h −1 (286.5 mg m −2 d −1 ) and calculated average emission rates of 2.1 mg m −2 h −1 (51 mg m −2 d −1 ) (Morrissey and Livingston, 1992;Roulet et al., 1992;Cao et al., 1998), these are relative minor emissions.However, given the large global soil areas and the frequency at which dried and rewetted soils release CH 4 , this source can nevertheless be an important factor in aerobic soil organic matter degradation.
The CH 4 emissions under UV light are consistent with findings by Vigano et al. (2008) and McLeod et al. (2008), who showed that UV irradiation drives CH 4 production from dried plant matter.Thus soil organic matter is most likely the precursor of CH 4 emissions observed in our studies.This is supported by CH 4 emissions that were observed when lignin and humic acid were exposed to UV irradiation under the same conditions as that for the soil samples.However, it is interesting that under UV irradiation there was no apparent correlation between CH 4 production and the soil organic matter content.This indicates that other soil components also play a role in CH 4 formation.Organic photosensitisers such as tryptophan (Messenger et al., 2009) or the mineral soil fraction, e.g., clay minerals and metal oxides (Katagi, 1990;Wu et al., 2008;Kibanova et al., 2011) may catalyze surface reactions of organic matter leading to CH 4 formation.This would also be in agreement with the recent observation that meteoritic matter such as carbonaceous chondrites containing only a few per cent organic matter releases large amounts of CH 4 when exposed to UV irradiation (Keppler et al., 2012).Methane emissions under UV radiation were found to be in the range of 0.25 to 7.28 µg m −2 h −1 for various soils in the UV-B intensity range of 1 to 4 W m −2 .Again, these emission rates are considerably lower than emissions observed from natural wetlands (Morrissey and Livingston, 1992;Roulet et al., 1992;Cao et al., 1998).However, a large fraction of the terrestrial surface is directly exposed to UV radiation, and this might even increase due to anthropogenic activities leading to deforestation and Introduction

Conclusions References
Tables Figures

Back Close
Full desertification.Interesting regions for on-site studies of UV induced CH 4 release could be steppes regions, newly deforested land, and freshly ploughed fields, whereas for water mediated CH 4 release flooding plains and irrigation areas in dry climates would be relevant.However, it has to be considered that more than 90 % of CH 4 formed within soils is oxidised by methanotrophic bacteria before it reaches the atmosphere (King, 1990).Methane uptake into aerated temperate forest soils ranges from 10 to 204 µg m −2 h −1 , depending on soil type, temperature and water saturation (Born et al., 1990;Castro et al., 1995;King, 1997).Field measurements regarding the temperature and water mediated CH 4 emissions may thus be impaired by methanotrophic consumption.In contrast direct photolysis of soil organic matter will occur at the upper soil surface at maximum depths of 0.2 to 0.4 mm and indirect photolysis processes might affect the soil down to 2 mm depth (Hebert and Miller, 1990).Thus CH 4 formation induced by UV irradiation at the soil surface might lead to direct CH 4 emissions to the atmosphere.It will be a challenge to differentiate between microbial and non-microbial sources in the field.
Hydrogen peroxide was found to have a positive effect on CH 4 production from peat.Levels of H 2 O 2 in soils are influenced by the activity of plant roots, fungi and bacteria (Sch önknecht et al., 2008;Miller et al., 1998).As the release of H 2 O 2 from living organisms is often a defence mechanism, the amount released might be affected by organism density in the soil and the level of stress applied by (changing) environmental factors.All effects shown to increase CH 4 production might gain importance in the course of climate change considering predicted changes in temperatures, precipitation levels and evaporation rates.Flood plains and other environments with strong fluctuations in the water budget might be of particular interest.Further investigations will be essential to fully understand the biogeochemical cycle of CH 4 in soils and its relevance for the atmosphere Introduction

Conclusions References
Tables Figures

Back Close
Full Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | ter (ThermoQuest Finnigan, Bremen, Germany) via an oxidation reactor [ceramic tube (Al 2 O 3 ), length 320 mm, 0.5 mm i.d., with oxygen activated Cu/Ni/Pt wires inside, reactor temperature 960 • C] and a GC Combustion III Interface (ThermoQuest Finnigan, Bremen, Germany).The gas chromatograph (GC) was fitted with a GS-Carbonplot capillary column (30 m × 0.32 mm i.d., d f 1.5 µm; Agilent, Santa Clara, USA) and a Po-raPlot capillary column (25 m × 0.25 mm i.d., d f 8 µm; Varian, Lake Forest, USA).Both columns were coupled using a press fit connector.A tank of high purity carbon dioxide (carbon dioxide 4.5, Messer Griesheim, Frankfurt, Germany) with a known δ 13 C value of −23.6 ‰ (VPDB) was used as the working reference gas.All δ 13 C values obtained from analysis of methane were corrected using three CH 4 working standards (isometric instruments, Victoria, Canada) calibrated against IAEA and NIST reference substances.The calibrated δ 13 C values of the three working standards in ‰ vs. V-PDB were −23.9 ± 0.2 ‰, −38.3 ± 0.2 ‰ and −54.5 ± 0.2 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | • ) have been suggested to play an important role in the release of CH 4 from pectin and might be the driving force in the CH 4 release during UV radiation of plant foliage (McLeod et al., 2008; Messenger et al., 2009).Hydrogen peroxide (H 2 O 2 ) as a precursor of HO • is an important reactant Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Heating experiments showed δ 13 C values of −56 to −65 ‰ for lignin, −51 to −56 ‰ for PH and −42 to −52 ‰ for humic acid.Methane emitted from wet samples of lignin, humic acid and peat PH showed δ 13 C values ranging from −53 to −69 ‰ with humic acid again being the substrate with the highest (less negative) CH 4 values (−53.2 ‰ ± 0.3 ‰).The δ 13 C values measured for CH 4 emitted from humic acid and peat PH over a 24 h period following the addition of H 2 O 2 were −54.9 ± 1.2 ‰ and −60.2 ± 4.5 ‰, respectively.The δ 13 C values measured for CH 4 emitted during 48 h under UV irradiation were −56.0 ± 6.0 ‰ for lignin, −63 ± 3.3 ‰ for SHA, −44.2 ± 1.4 ‰ for PH and −35.3 ± 9.4 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper |

Table 1 .
Organic carbon content, pH value and CH 4 emissions from dry and wetted samples heated at 30 and 40 • C and under UV irradiation of different soils and soil components.
Subscript h indicates soil horizon, C org is organic carbon content, PH is peat Hille, Germany; SW is soil H äverst ädt, Wiehen Mountains, Germany; SG is soil Gonsenheim, Germany; SL is soil Lerchenberg, Germany; SHA is soil Hainich, Germany; n.d. is not detectable (rate cannot be provided as increase in headspace CH 4 was less than 0.02 ppm); n.m. is not measured;