Delft University of Technology
Coupling Stable Isotope Analysis with Gas Push-Pull Tests to Derive In Situ Values for the Fractionation Factor αox Associated with the Microbial Oxidation of Methane in Soils
Gebert, Julia; Streese-Kleeberg, J. DOI
10.2136/sssaj2016.11.0387 Publication date
2017
Document Version
Accepted author manuscript Published in
Soil Science Society of America. Journal
Citation (APA)
Gebert, J., & Streese-Kleeberg, J. (2017). Coupling Stable Isotope Analysis with Gas Push-Pull Tests to Derive In Situ Values for the Fractionation Factor αox Associated with the Microbial Oxidation of Methane in Soils. Soil Science Society of America. Journal, 81(5), 1107-1114.
https://doi.org/10.2136/sssaj2016.11.0387 Important note
To cite this publication, please use the final published version (if applicable). Please check the document version above.
Copyright
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons. Takedown policy
Please contact us and provide details if you believe this document breaches copyrights. We will remove access to the work immediately and investigate your claim.
Coupling stable isotope analysis with gas push-pull tests
1to derive in-situ values for the fractionation factor
α
ox2
associated with the microbial oxidation of methane in
3soils
4KEYWORDS 5
Landfill cover soil, greenhouse gas emission, methane oxidation, isotopic fractionation 6
ABSTRACT 7
Prerequisite for the application of stable isotope fractionation for the quantification of the 8
methane oxidation efficiency of landfill covers is that the fractionation factor αox is known or can
9
be estimated with adequate accuracy. So far, αox has only been determined in laboratory
10
experiments. For the first time, αox was determined under in situ conditions in the field by
11
coupling two independent methods, gas push-pull tests and stable isotope analysis, to assess 12
biological fractionation of methane isotopologues in landfill cover soils. On six landfills with 13
nine points of investigation, 22 measurements were carried out, covering a wide range of 14
environmental conditions such as soil temperature and moisture and observed oxidation rates. 15
Values for αox varied between near 1, indicating only little fractionation, and 1.151. Correlation
16
of αox with the methane oxidation rate found by gas push-pull tests revealed a clear asymptotic
relationship with low rates being associated with high values for αox and high rates resulting in
18
only little fractionation. Values for αox varied between the different landfills, but also between
19
the individual points of investigation on the same landfill. The latter is assumed to reflect the 20
spatial variability of methanotrophic activity due to spatial variability in soil moisture and hence 21
air-filled porosity, as well as the spatial variability of gas fluxes. Further, significant variation of 22
αox was observed also for the same sampling point, presumably reflecting the temporal
23
variability of factors influencing methanotrophic activity. These effects could include seasonally 24
changing environmental conditions such a soil temperature and moisture, but also the temporal 25
variability of gas fluxes through the landfill soil cover, changing exposure of methanotrophs to 26
methane and oxygen and hence their activity. The quantification of the methane oxidation 27
efficiency using fractionation of stable isotopes is very sensitive to αox. Assuming a value
28
constant in time and space and transferring this value from laboratory experiments to field 29
settings entails significant uncertainty regarding the quantification of methane oxidation. 30
1 INTRODUCTION
31
Landfills are estimated to be the second to third largest source of anthropogenic methane 32
emissions (EEA, 2009; USEPA, 2016). Sanitary landfilling including active extraction and 33
subsequent treatment or energetic utilization of landfill gas is mandatory in many countries. The 34
microbial oxidation of methane in optimized landfill covers, windows and filters is an option to 35
complement technical measures and mitigate residual methane fluxes for which technical 36
treatment is no longer feasible (Bogner et al., 2007; Huber-Humer et al., 2008; Scheutz et al., 37
2009). It is also an option for the many old landfills that are not equipped with a gas extraction 38
system or for landfills where gas production is low from the start. The potential of biological 39
treatment is widely recognized. Both operators and regulators, however, require accepted 40
methods to prove and quantify the mitigation effect. 41
The fractionation of stable isotopes (SI), naturally occurring as a result of the oxidation process, 42
offers one of the few approaches to assess the efficiency of microbial methane oxidation (fox) in
43
situ (Chanton et al., 1999, 2011; Börjesson et al., 2001, 2007; Cabral et al., 2010; Abichou et al., 44
2011; Widory et al., 2012). The method is based on the fact that carbon naturally exists in the 45
form of two stable isotopes (12C and 13C) whose size, steric properties and diffusion coefficient 46
differ, leading to unequal behavior at enzyme binding sites (Jahnke et al., 1999) and to different 47
transport rates. Due to the preferential oxidation (Barker and Fritz, 1981) and faster diffusive 48
transport (Tyler et al., 1994; Reeburgh et al., 1997; De Visscher et al., 2004) of the lighter 49
isotope, the heavier isotope is enriched in the remaining gas phase. The factors by which 50
oxidation and transport fractionate, αox and αtrans, respectively, are used to calculate methane
51
oxidation efficiency from the difference in isotopic signature between the landfill gas and the gas 52
emitted from the soil surface or any other point of interest. In the past, values of αox have been
determined in laboratory under standardized conditions and have then been used to calculate 54
methane oxidation efficiencies from field isotopic signature data. However, as αox is known to
55
vary with oxidation rate (Templeton et al., 2006; Chanton et al., 2008), temperature (Chanton et 56
al., 2008), and possibly the composition of the methanotrophic community (Jahnke et al., 1999; 57
Templeton et al., 2006), the value was hypothesized to be subject to site-specific and seasonal 58
variation. 59
In order to determine αox in situ, the oxidation efficiency must also be determined in situ. This is
60
possible using the gas push-pull test (GPPT) methodology (Gómez et al., 2009; Urmann et al., 61
2009; Streese-Kleeberg et al., 2011). During a GPPT, a defined volume of the reactive gas of 62
interest (in this case methane and oxygen) and a conservative tracer with similar transport 63
properties (e.g. argon) are injected into the soil’s vadose zone. Following injection, the mixture 64
of reactant, tracer and soil gas is extracted via the same tube. During extraction, the gas mixture 65
is sampled periodically in order to obtain breakthrough curves of reactant and tracer. The 66
reaction rate is calculated from the difference in the breakthrough curve of the reactive gas 67
compared to the one of the tracer. The combination of GPPTs with SI analyses allows the in situ 68
determination of αox by simultaneously collecting data on the change of the isotopic composition
69
and on the oxidation efficiency. The aim of the study was to verify the range of fractionation 70
factors previously published from laboratory experiments and to investigate the hypothesized in-71
situ variation of the fractionation factor resulting from to the variation in field oxidation rates. 72
2 MATHEMATICAL BACKGROUND OF THE STABLE ISOTOPE APPROACH
73
The shift in the isotopic ratio of 13C to 12C through microbial methane oxidation or through 74
diffusive gas transport, signified in the delta notation as δ13C, is described by the following 75
equation: 76
δ13
C [‰] = ((Rsample/Rstandard) – 1)*1000 Eq. 1
77
Where Rsample= 13C/12C ratio in the sample
78
Rstandard= 13C/12C ratio of the carbon reference standard VPDB (Vienna Peedee
79
Belemnite, 13C/12C = 0.0112372). 80
The fractionation factors effective during the oxidation process and during gas transport are 81
termed αoxand αtrans, respectively, and are defined as the oxidation or transport rate constant for
82
12
CH4 divided by the oxidation or transport rate of 13CH4:
83
αox, trans = k12/k13 Eq. 2
84
No fractionation is assumed when gas transport is advective, then αtrans is equal to 1.
85
The exact factors for oxidative and transport fractionation need to be known to quantify methane 86
oxidation efficiency from the measured shift in stable isotope ratios as follows (Blair et al., 87
1985): 88
𝑓𝑓𝑜𝑜𝑜𝑜 =1000 × (𝛼𝛼(𝛿𝛿𝑠𝑠−𝛿𝛿𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎−𝛼𝛼𝑡𝑡𝑡𝑡𝑎𝑎𝑎𝑎𝑠𝑠) ) Eq. 3
Where fox = fraction of methane load oxidized at the particular sampling location, also
90
defined as methane oxidation efficiency 91
δs = δ13C value in sample
92
δanox = δ13C value at the chosen point of reference, often the anaerobic zone of
93
methane production 94
αox, αtrans = factors for oxidative and diffusive fractionation.
95
As a fraction of 1 or a percentage (if multiplied by 100), fox represents a relative value, not an
96
absolute oxidation rate. 97
3 MATERIALS AND METHODS
98
2.1 Site description
99
Investigations were carried out on five old non-sanitary municipal solid waste landfills (landfills 100
A, D, L, H and K) in north western Germany, all filled in the 1960s to the 1980s. The landfills 101
were part of a baseline study on methane oxidation in non-optimized cover soils, conducted 102
within the framework of the MiMethox project (Rachor et al., 2009). The landfills were covered 103
with whatever soil was available to the operator at landfill closure, mostly of sandy to loamy 104
texture, cover thickness varied between 0.1 m and > 2 m. The vegetation that established since 105
the placement of the soil cover consisted of grass, shrubs and small trees. None of the sites had a 106
surface sealing so that landfill gas migrated freely from the waste body through the soil cover. 107
Potential methane oxidation rates as determined in laboratory ranged between 25 and > 2000 g 108
CH4 m-2 d-1 (Gebert et al., 2016). For the purpose of the MiMethox project, all landfills were
equipped with gas probes, temperature and moisture sensors in 5, 15, 40, and 80 cm depth of the 110
soil cover on three locations per landfill (details in Gebert et al., 2011a). These locations had 111
been chosen based on a preliminary soil and soil gas survey. This study used the same locations. 112
The composition of the methanotrophic community was dominated by Methylocystis, 113
Methylobacter and Methylococcus species among type II, Ia and Ib methanotrophs, respectively,
114
on all of the five investigated landfills (Gebert et al., 2009). Detailed information on the landfills, 115
on the properties of the cover soils, and on the gas fluxes through and from the soil cover is 116
given in Rachor (2012). 117
Further measurements were performed on two methane oxidation cover test cells (cells C and G) 118
on a Dutch landfill (landfill W), constructed with a loamy sand (topsoil) above a loam (subsoil; 119
classification according to FAO, 2006) and a forced gas load to the base of the methane 120
oxidation layer of up to 41 g CH4 m-2 d-1. The test cells had a size of 1.060 m2 each. They served
121
to investigate the combination methane oxidation and measures to limit drainage water 122
infiltration. The latter was realized by a capillary barrier (cell C) and a drainage mat (cell G). 123
Short grass vegetation was maintained on these test cells by regular mowing. More information 124
on the test cell setup is given in (Röwer, 2016a; Geck et al., 2016). 125
2.2 Gas push-pull tests
126
The application of the gas push-pull test to quantify methane oxidation rates in landfill cover 127
soils including a detailed description of the methodology and the nature and interpretation of 128
results on 50 GPPTs is described in depth in Streese-Kleeberg et al. (2011). This section 129
summarizes the principal approach of the GPPT. A gas mixture of 30 l, containing methane, 130
oxygen and argon (10% CH4, 20% Ar in air), were injected at the designated soil depth (usually
between 40 and 60 cm below surface) at a flow rate of 2.5 l/min using a pump connected to a 132
mass flow meter and a tube with a perforated tip. The depth was chosen so that the methane 133
background concentration would not exceed 5% and the oxygen background concentration 134
indicated oxidative conditions in the soil. Hereby, it was prevented that a high methane 135
background confounded the GPPT results. Soil gas composition was verified beforehand by on-136
site analysis of the soil gas phase using a biogas analyser (BM2K2-E000, Geotechnical 137
Instruments (UK) Ltd.) for measurement of CH4 (infrared detector), CO2 (infrared detector) and
138
O2 (electrochemical detector). The detection limit for all three gases was 0.1 vol.%, Instrument
139
calibration was performed with certified calibration gases in the laboratory and validated with 140
parallel gas-chromatographic analyses. 141
Following injection, the mixture of reactant, tracer and soil gas were extracted from the same 142
tube. The extraction flow rate was 0.3 to 0.5 l/min. During the extraction phase the gas was 143
streamed through 40 ml glass vials with septum caps which were exchanged every 1 to 4 144
minutes. The time interval depended on the rate of decline of the methane concentration. This 145
was continuously measured in the extracted gas stream with an infrared cell. These samples were 146
analysed for methane and argon concentrations. The vials contained a barrier solution (200 g 147
NaCl l-1 and 5 g citric acid l-1 in water) to prevent gaseous diffusion through the vial septa and 148
were stored in an inverted position until analysis. In addition, gas samples intended for carbon 149
stable isotope analyses were collected from the gas stream using a syringe and injected into vials 150
containing a supersaturated salt solution. Supersaturation served to prevent dissolution of gas 151
into the liquid phase. Vials were also stored in an inverted position until stable isotope analyses 152
took place. Three to eleven stable isotope samples were retrieved during each GPPT. The 153
average δ13
C signature of the injected gas was -41.56 ± 0.89‰ (standard deviation). 154
Data evaluation of the GPPT was performed according to Yang et al. (2007). First, reactant and 155
tracer were corrected for background concentrations for each sampling point. The corrected 156
concentrations were then normalized to the concentration of the injected reactant and tracer. 157
Thereafter, the normalized concentrations were plotted over time elapsed since the 158
commencement of the extraction phase. From the normalized tracer concentrations, the mixing 159
factor was obtained for each sampling point as follows: 160
𝑓𝑓 = (Ctr− Ctrb)/(Ctr0 − Ctrb) Eq. 4
161
Where 𝑓𝑓 = mixing factor 162
Ctr = current tracer concentration
163
Ctrb = background tracer concentration
164
Ctr0 = injected tracer concentration.
165
All concentrations are to be expressed in the same unit. 166
The normalized concentrations decrease at a higher rate than those of the tracer and so-called net 167
mass transfer takes place if the reactant is consumed, i.e. in this case is microbially degraded. Net 168
mass transfer thus represents the change in reactant (methane) concentration that cannot be 169
attributed to dilution and therefore represents methane consumption. The decrease in 170
concentration is due to dispersion in the soil only if the slopes of reactant and tracer are similar. 171
The net mass transfer is calculated as follows: 172
𝑁𝑁𝑁𝑁𝑁𝑁 = 𝐶𝐶𝑟𝑟 – [𝑓𝑓 × 𝐶𝐶𝑟𝑟0+ (1 − 𝑓𝑓) × 𝐶𝐶𝑟𝑟𝑏𝑏] Eq. 5
Where 𝑁𝑁𝑁𝑁𝑁𝑁 = net mass transfer 174
𝐶𝐶𝑟𝑟 = current reactive gas concentration
175
𝑓𝑓 = mixing factor 176
𝐶𝐶𝑟𝑟0= injected reactive gas concentration
177
𝐶𝐶𝑟𝑟𝑏𝑏 = background reactive gas concentration.
178
All concentrations are to be expressed in the same unit, the net mass transfer then assumes this 179
unit. 180
Prerequisite for the validity of the approach is that reactant and tracer gases exhibit similar 181
transport characteristics in the soil gas phase. In the case of methane, argon meets this condition 182
(Gómez et al., 2009). The methane oxidation rate is derived as a zero order kinetic constant from 183
the slope of the net mass transfer over time: 184
𝑘𝑘0 = 𝑁𝑁𝑁𝑁𝑁𝑁/𝑡𝑡 Eq. 6
185
Where 𝑘𝑘0 = zero order rate constant and 𝑡𝑡 = time. 186
The zero order rate constant yields the methane oxidation rate in mass unit CH4 per volume unit
187
soil air per unit time. It has to be considered that this rate is to be interpreted as a potential rate, 188
as both oxygen and methane are supplied at optimum concentration levels during the test. 189
2.3 Analysis of stable isotopes
190
The δ13C ratio was analysed in triplicate using a GC-IRMS (gas chromatography isotope ratio 191
mass spectrometry, Delta Plus, ThermoScientific, Dreieich, Germany) equipped with a 25 m 192
capillary column (Poraplot, 0.32 mm). The reference standard NGS3 8561 (δ13CVPDB-NGS3 193
8561 = -73.27‰; NIST, Gaithersburg, MD, USA) was used to express the methane-derived δ13C 194
ratio relative to the VPDB (Vienna Pee Dee Belemnite) standard. Analytical replicate precision 195
was < 0.2‰. 196
2.4 Calculation of fractionation factor αox 197
Values for αox were obtained using GPPT and methane-derived δ13C-data based on the approach
198
derived by Coleman et al. (1981) from Rayleigh (1896): 199 𝛿𝛿13𝐶𝐶 𝑡𝑡≅ 1000 ∗ �∝1𝑎𝑎𝑎𝑎− 1� × ln �𝑀𝑀𝑀𝑀0� + 𝛿𝛿13𝐶𝐶𝑡𝑡=0 Eq. 7 200 Where 201 δ13
Ct = δ13C value of CH4 remaining at time t at the particular soil depth during the GPPT
202
M/M0 = fraction of CH4 (corrected for dilution) remaining at time t at the particular soil
203
depth related to the initial methane concentration M0 (also corrected for dilution)
204
δ13
Ct=0 = δ13C value of CH4 at the initial time, i.e. of the mixture injected in to the soil.
205
The slope of the line δ13
Ct - δ13Ct=0 versus ln(M/M0) (see also Figure 2) is equal to 1000*(1/αox
-206
αtrans), so that αox was calculated as follows:
207 1 𝛼𝛼𝑎𝑎𝑎𝑎 = 1 + �𝛿𝛿13𝐶𝐶𝑡𝑡−𝛿𝛿13𝐶𝐶𝑡𝑡=0� ln�𝑀𝑀0𝑀𝑀�×1000 Eq. 8 208
Pushing and pulling the gas into and out of the soil results in advective gas transport. Methane 209
concentrations can be assumed to be spatially uniform within the injected and extracted volume a 210
given point in time. Hence, gas transport driven by a concentration gradient was assumed to be 211
insignificant within the time frame of the test. Therefore, fractionation due to diffusive gas 212
transport (Barker and Fritz, 1981) was neglected and αtrans was assumed to be equal to 1. Values
213
for αox were only accepted when the correlation coefficient of the slope was significant on a 99%
214
confidence level. 215
4 RESULTS AND DISCUSSION
216
Table 1 summarizes the methane oxidation rates derived from the GPPTs, the details of the 217
respective environmental conditions, the soil properties as well as the calculated values αox.
218
Some of the information can also be found in Streese-Kleeberg et al. (2011) who introduce the 219
GPPT method adapted for landfill cover soils and present data on more than 50 GPPTs. The 220
following sections discuss the data on environmental conditions, soil properties, and the 221
variability of αox in relation to the oxidation rate.
222
Table 1: Soil properties, environmental conditions, fractionation factors and oxidation rates for each site. n.d. 223
= not determined. Temp. = temperature, moist. = moisture. Site ID notation: H, K, D, L, A = different MSW 224
landfills in north western Germany; WC, WG = test cell C and test cell G on landfill W in The Netherlands. 225
P1-5 = individual sampling points. 226
4.1 Environmental conditions, soil properties and methane oxidation potential
227
Both the environmental conditions and soil properties such as porosity varied between the 228
individual points of investigation but also for the same point over time. Over all landfill sites and 229
points of investigation, soil temperature spanned between 1.0 and 20 °C, and soil moisture 230
ranged between 3.5 and 26.1 vol%. Air-filled porosity, predominantly determining the soil’s gas 231
transport properties, varied between 10.2 and 32.9 vol%. 232
The seasonal variation of soil temperature and moisture influences the microbial process of 233
methane oxidation (e.g. Gebert et al., 2003; Spokas and Bogner, 2011) and hence also emission 234
(e.g. Tecle et al., 2009; Rachor et al., 2013; Geck et al., 2016). The variability of texture, 235
compaction and moisture strongly impacts soil gas diffusivity and gas conductivity (Møldrup et 236
al., 2000; Gebhardt et al., 2009; Gebert et al., 2011b; Röwer et al., 2016b) and thereby modulates 237
the methane oxidation rate in landfill soil covers by influencing the ingress of atmospheric air. 238
Further variability of methane oxidation rates is introduced by the varying level of exposure of 239
the methanotrophic community to methane (Röwer et al., 2011; Schroth et al., 2012), leading to 240
methane oxidation rates spanning several orders of magnitude on one or between different 241
landfill sites (Spokas and Bogner, 2011; Gebert et al., 2016). 242
Given the above, significant spatial and temporal variability of the methane oxidation rate in 243
landfill cover soils can be expected a priori and was also observed during the gas push-pull tests 244
conducted in this study (Table 1). For the six investigated landfill cover soils the range of 245
oxidation rates found with the GPPTs ranged between 2.3 and 144 g m3soil air-1 h-1, corresponding
246
to 0.32 and 18 g m3soil-1 h-1 when values for total porosity and soil moisture at the time of
247
measurement were considered. These oxidation rates relate to efficiencies of 3% and 100% at the 248
end of the GPPT. An extraordinarily high rate of 1110 g m3soil air-1 h-1 or 354 g m3soil-1 h-1 was
249
found for location WC1. Sites WC1 and WC4 are located within 10 m of each other on a test cell 250
investigating the suitability of biocovers for the mitigation of landfill methane. As indicated by 251
gas profile measurements at both sites (data not shown), WC1 receives high concentrations of 252
methane whereas due to spatial heterogeneity of gas fluxes the exposure to landfill gas at WC4 is 253
low. As a result, WC1 developed a very high potential activity of 1110 g CH4 m-3soil air h-1,
254
compared to only 50 g CH4 m-3soil air h-1 at WC4. Evolution of methanotrophic potential strongly
depends on the supply of the substrate methane (Röwer et al., 2011; Spokas and Bogner, 2011; 256
Schroth et al., 2012) and can hence span orders of magnitude in relation to the spatial 257
heterogeneity of fluxes. In line with this, GPPT-based oxidation potentials for the cover soil on a 258
Swiss municipal solid waste landfill were found to be as high as 210 g m3soil-1 h-1 for sites where
259
high methane concentrations were measured in the soil gas phase. The oxidation potential was 260
below the GPPT’s detection limit where soil methane concentrations were very low (Schroth et 261
al., 2012). 262
4.2 Shift of the methane δ13C signature with oxidation efficiency
263
During the time course of each GPPT, the δ13C signature in the sampled soil gas phase shifted 264
towards less negative values, i.e. the heavier isotopologue 13CH4 became enriched as the lighter
265
isotopologue (12CH4) was preferentially consumed by the methanotrophic bacteria. In Figure 1,
266
the change in isotopic signature is plotted versus the oxidation efficiency (fox) determined for
267
each individual sampling point during each of the 22 GPPTs. For the entire data set, the 268
efficiency ranged between near 0% and near 100%. The maximum observed enrichment in δ13C 269
was 62‰ (from -41 to +21‰ for GPPT no. 26 at site HP5, compare also Table 1), with large 270
shifts suggesting vigorous methane oxidation activity. As expected, the degree of enrichment 271
generally increased with increasing oxidation efficiency. However, some data points showed 272
little enrichment in relation to the oxidation efficiency while for others, the extent of enrichment 273
was substantially higher than would be expected from the relationship between fox and δ13C for
274
the bulk data, indicating different extents of isotopic fractionation during the oxidation process. 275
Figure 1: Relationship between methane oxidation efficiency fox and respective shift in carbon isotopic 276
signature of methane relative to the injected gas. Data from 22 individual GPPTs with three to six individual 277
points per GPPT (compare also Table 1). 278
Figure 2 shows how the change in isotopic signature over time (δ13Ct - δ13Ct=0) relates to the
279
natural logarithm of the remaining methane (ln(M/M0)), using an example of a GPPT yielding a
280
low oxidation rate (#51, site DP1) and one yielding a high oxidation rate (#59, also site DP1). 281
According to Eq. 8, values for αox were calculated from the slope of this line, if significant on a
282
99% level. 283
Figure 2: Relationship between the fraction of methane remaining during the course of a GPPT, expressed as 284
ln(M/M0), and the change in isotopic signature, expressed as δ13Ct - δ13Ct=0, for GPPTs 51 (left panel, point 285
DP1) and 59 (right panel, also point DP1). Line = linear fit. 286
4.3 Calculated values for αox and their relationship to the methane oxidation rate 287
The literature reports values for αox between 1.018 and 1.049 (Chanton et al., 1999; Chanton and
288
Liptay, 2000; Börjesson et al., 2001, 2007; De Visscher et al., 2004; Capanema and Cabral, 289
2012), a range also represented by 64% of the measurements in this study (Table 1, Figure 3). 290
Two GPPTs showed lower and six of the 22 GPPTs higher isotopic fractionation. The 291
relationship between the in-situ methane oxidation rate derived from the GPPTs and the values 292
for αox is shown in Figure 3. Values for αox were high at low oxidation rates and decreased
293
nonlinearly with increasing methane oxidation rate, approximating the value of αox = 1 (i.e. no
294
fractionation) at very high rates (sampling point WC1). Hardly any discrimination of the heavier 295
isotopologue was observed under these conditions of high methanotrophic activity (GPPTs #53 296
at site KP2, #103 at site WC3). Vice versa, a low activity allows for a high extent of 297
discrimination (e.g. GPPTs #41 at site DP1, #44 at site LP1). The observed fractionation factor 298
not only varied between the different sites but also for different GPPTs conducted at the same 299
landfill and the same sampling point, presumably reflecting the temporal variability of process 300
controls (e.g. temperature, moisture and hence oxygen supply) over the seasons. The maximum 301
span of values for αox observed for one individual point of investigation was 1.035 to 1.100
302
(point DP1, Table 1, Figure 3). 303
Figure 3: Relationship between methane oxidation rate and fractionation factor αox. n = 22. For site IDs see
304
heading of Table 1. 305
The general relationship between environmental conditions, the methane oxidation rate and the 306
fractionation factor αox can be deduced from the largest single data set available from site DP1 (n
307
= 7), shown in Figure 4. It is seen that cooler conditions in the winter coincide with increased 308
soil moisture, which is typical for the temperate European climate with the lower methane 309
oxidation potentials and wit higher values for αox. Vice versa, the highest oxidation potentials
310
and the lowest extent of fractionation are found when the soil is warmer and dryer. The same 311
seasonal pattern is apparent for sited HP5 and KP2. All in all, the individual data sets are too 312
small for a multivariate statistical analysis of the factors of influence, which, however, have been 313
previously described (e.g. Scheutz et al., 2009). 314
Figure 4: Relationship between environmental parameters, methane oxidation rate and fractionation factor 315
αox as found at site DP1 (data from Table 1).
316
The data from sampling point DP1 were also used to assess the effect of the variability of αox on
317
calculated values for the oxidation efficiency fox.by example of GPPT #28. To this end, fox was
318
calculated for the change in isotopic signature observed for this particular test (40.36‰ to -319
16.54‰) employing the seven individual values for αox determined during the seven GPPTs
320
conducted at site DP1 (1.036 to 1.100, see Table 1). The resulting oxidation efficiencies varied 321
between -2.5% and 66%, indicating that it is not possible to assume αox to be constant even in the
322
very same sampling location (in this case, DP1). 323
In another sensitivity analysis, Capanema and Cabral (2012) varied the value of αox by ±0.5% of
324
the value determined in the laboratory on their own samples. The resulting methane oxidation 325
efficiencies (fox) varied between 53 and 88% for the same data set. Increasing αox by 0.5% led to
326
implausible oxidation efficiencies greater than 100%. All examples show that the sensitivity of 327
the method to the correct value of αox is hence very high and that hence the exact value of αox
328
needs to be determined for each individual combination of conditions under which the methane 329
oxidation efficiency of a landfill cover is assessed. 330
The observed variation of isotopic fractionation with oxidation rate corroborates findings from 331
earlier studies. Templeton et al. (2006) showed that most important factor determining 332
fractionation of methane isotopologues was the fraction of total methane oxidized per time unit, 333
with values for αox decreasing with increasing turnover rates and increasing cell densities.
334
Chanton et al. (2008) found αox to decline nonlinearly with increasing maximum methane
335
oxidation rates (vmax), indicating less discrimination at higher turnover rates when the
336
temperature is constant. 337
In summary, it was shown that values for αox varied at the same sampling point, between
338
different sampling points on the same landfill, and between different landfills. Further, the 339
oxidation rate strongly affected the extent of discrimination of the heavy isotopologue and hence 340
the magnitude of αox. Low rates caused a high extent of fractionation whereas at very high rates
341
αox approximated the value of one, i.e. no preferential oxidation of the lighter isotopologue was
342
observed. In conclusion, the exact value for αox and its variability in time and space should be
343
known to reduce uncertainty in the determination of the methane oxidation efficiency. Since soil 344
temperature and moisture as well as the spatial and temporal heterogeneity of gas fluxes through 345
landfill covers result in continuously changing environmental conditions, the extent of both 346
transport and biological fractionation will vary correspondingly. Adopting αox values from the
347
literature or assigning the same α value to the same measurement point over time or assigning 348
the same value to different points on one or several landfills can lead to misjudgment of methane 349
oxidation efficiencies. In order to avoid this, αox would have to be determined for each individual
350
campaign using undisturbed soils samples at the prevailing moisture content. Depending on the 351
desired level of accuracy, these limitations affect the applicability of the stable isotope approach 352
as a robust, practical and economic method for the exact quantification of methane oxidation 353
efficiencies. 354
FUNDING SOURCES 355
The research was funded by the German Federal Ministry for Education and Research (BMBF) 356
within the framework of the project MiMethox (Microbial Methane Oxidation in Landfill 357
Covers). 358
ACKNOWLEDGMENT 359
The authors would like to thank Dr. Ingke Rachor, Dr. Kim-Karen Kleeberg, Volker 360
Kleinschmidt, Cindy Streblow, Simon Klauke and Dr. Christian Knoblauch for their assistance 361
in field and laboratory measurements. 362
REFERENCES 363
Abichou, T., Mahieu, K., Chanton, J., Romdhane, M., Mansouri, I. (2011): Scaling methane 364
oxidation: From laboratory incubation experiments to landfill cover field conditions. Waste 365
Management 31, 978-986. 366
Barker, J.F., Fritz, P. (1981): Carbon isotope fractionation during microbial methane oxidation. 367
Nature 293, 289-291. 368
Blair, N., Leu, A., Munoz, E., Olson, J., Kwong, E., Des Marais, D. (1985): Carbon isotopic 369
fractionation in heterotrophic microbial metabolism. Applied and Environmental Microbiology 370
50, 996-1001. 371
Bogner J., Abdelrafie Ahmed, M., Diaz, C., Faaij, A., Gao, Q., Hashimoto, S., Mareckova, K., 372
Pipatti, R., Zhang, T. (2007): Waste Management. In: Climate Change 2007: Mitigation. 373
Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental 374
Panel on Climate Change. Metz, B., Davidson, O.R., Bosch, P.R., Dave, R., Meyer, L.A. (eds). 375
Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. 376
Börjesson, G., Chanton, J., Svensson, B. (2001): Methane oxidation in two Swedish landfill 377
covers measured with carbon-13 to carbon-12 isotope ratios. Journal of Environmental Quality 378
30, 369-376. 379
Börjesson, G., Samuelsson, J., Chanton, J. (2007): Methane oxidation in Swedish landfills 380
quantified with the stable carbon isotope technique in combination with an optical method for 381
emitted methane. Environmental Science & Technology 41, 6684-6690. 382
Cabral, A.R., Capanema, M.A., Gebert, J., Moreira, J.F., Jugnia, L.B. (2009): Quantifying 383
microbial methane oxidation efficiencies in two experimental landfill biocovers using stable 384
isotopes. Water, Air, and Soil Pollution 209, 157-172. 385
Capanema, M.A., Cabral, A.R. (2012): Evaluating methane oxidation efficiencies in 386
experimental landfill biocovers by mass balance and carbon stable isotopes. Water, Air, and Soil 387
Pollution 223, 5623-5635. 388
Chanton, J.P., Liptay, K. (2000): Seasonal variation in methane oxidation in a landfill cover soil 389
as determined by an in situ stable isotope technique. Global Biogeochemical Cycles 14, 51-60. 390
Chanton, J.P., Powelson, D.K., Abichou, T., Fields, D., Green, R. (2008): Effect of temperature 391
and oxidation rate on carbon-isotope fractionation during methane oxidation by landfill cover 392
materials. Environmental Science & Technology 42, 7818-7823. 393
Chanton, J.P., Rutkowski, C.M., Mosher, B. (1999): Quantifying methane oxidation from 394
landfills using stable isotope analysis of downwind plumes. Environmental Science & 395
Technology 33, 3755-3760. 396
Chanton, J., Abichou, T., Langford, K., Spokas, K., Hater, G., Green, R., Goldsmith, D., Barlaz, 397
M.A. (2011): Observations on the methane oxidation capacity of landfill soils. Waste 398
Management 31, 914-925. 399
Coleman, D.D., Risatti, J.B. (1981): Fractionation of carbon and hydrogen isotopes by methane-400
oxidizing bacteria. Geochimica et Cosmochimica Acta 45, 1033-1037. 401
De Visscher, A., De Pourq, I., Chanton, J. (2004): Isotope fractionation effects by diffusion and 402
methane oxidation in landfill cover soils. Journal of Geophysical Research 109, (D18111). 403
European Environment Agency (EEA). Annual European Community Greenhouse Gas 404
Inventory 1990-2007 and Inventory Report 2007. Submission to the UNFCCC Secretariat. EEA 405
Technical report, No. 4. 2009; Annex Key Category Analysis. 406
http://reports.eea.europa.eu/technical_report_2006_6/en. 407
FAO - ISRIC & ISSS (eds.) (2006): World reference base for soil resources 2006: a framework 408
for international classification, correlation and communication. World soil resources reports, 409
103. Rome. 410
Gebert, J., Rachor, I., Gröngröft, A., Pfeiffer, E.-M. (2011a): Temporal variability of soil gas 411
composition in landfill covers. Waste Management 31, 935-945. 412
Gebert, J., Gröngröft, A., Pfeiffer, E.-M. (2011b): Relevance of soil physical properties for the 413
microbial oxidation of methane in landfill covers. Soil Biology and Biochemistry 43, 1759-1767. 414
Gebert, J., Singh, B.K., Pan, Y., Bodrossy, L. (2009): Activity and structure of methanotrophic 415
communities in landfill cover soils. Environmental Microbiology Reports 1, 414-423. 416
Gebert, J, Rachor, I.M., Streese-Kleeberg, J., Pfeiffer, E.-M. (2016): Methane oxidation in a 417
landfill cover soil under conditions of diffusive and advective flux, assessed by in-situ and ex-418
situ methods. Current Environmental Engineering, 3, 144-160. 419
Gebhardt, S., Fleige, H., Horn, R. (2009): Effect of compaction on pore functions of soils in a 420
Saalean moraine landscape in North Germany. Journal of Plant Nutrition and Soil Science 172, 421
688-695. 422
Geck, C., Scharff, H., Pfeiffer, E.-M., Gebert, J. (2016): Validation of a simple model to predict 423
the performance of methane oxidation systems, using field data from a large scale biocover test 424
field. Waste Management 56, 280-289. 425
Gómez, K.E., Gonzalez-Gil, G., Lazzaro, A., Schroth, M.H. (2009): Quantifying methane 426
oxidation in a landfill-cover soil by gas push-pull tests. Waste Management 29, 2518-2526. 427
Huber-Humer, M., Gebert, J., Hilger, H. (2008): Biotic systems to mitigate landfill methane 428
emissions. Waste Management & Research 26, 33-46. 429
Jahnke, L.L., Summons, R.E., Hope, J.M., Des Marais, D.J. (1999): Carbon isotopic 430
fractionation in lipids from methanotrophic bacteria II: The effects of physiology and 431
environmental parameters on the biosynthesis and isotopic signatures of biomarkers. Geochimica 432
et Cosmochimica Acta 63, 79-93. 433
Møldrup, P., Olesen, T., Schjønning, P., Yamaguchi, T., Rolston, D.E. (2000): Predicting the gas 434
diffusion coefficient in undisturbed soil from soil water characteristics. Soil Science Society of 435
America Journal 64, 1588-1594. 436
Rachor, I.M. (2012): Spatial and temporal patterns of methane fluxes on old landfills: processes 437
and emission reduction potential. Dissertation. Hamburger Bodenkundliche Arbeiten 67. ISSN 438
0724-6382. 439
Rachor, I.M., Streese-Kleeberg, J., Gebert, J. (2009): Spatial and temporal variability of gas 440
emissions from old landfills. Proceedings Sardinia, Twelfth International Waste Management 441
and Landfill Symposium S. Margherita di Pula, Cagliari, Italy; 5 - 9 October 2009. 2009 by 442
CISA, Environmental Sanitary Engineering Centre, Italy. 443
Rachor, I.M., Gebert, J., Gröngröft, A., Pfeiffer, E.-M. (2013): Variability of methane emissions 444
from an old landfill over different time-scales. European Journal of Soil Science 64, 16-26. 445
Rayleigh, J.W.S. (1896): Theoretical considerations respecting the separation of gases by 446
diffusion and similar processes. Philosophical Magazine 42, 493-498. 447
Reeburgh, W.S., Hirsch, A.I., Sansone, F.J., Popp, B.N., Rust, T.M. (1997): Carbon kinetic 448
isotope effect accompanying microbial oxidation of methane in boreal forest soils. Geochimica 449
et Cosmochimica Acta 61, 4761-4767. 450
Röwer, I.U., Geck, C., Gebert, J., Pfeiffer, E.-M. (2011): Spatial variability of soil gas 451
concentrations and methane oxidation in landfill cover soils. Waste Management 31, 926-934. 452
Röwer, I.U., Scharff, H., Pfeiffer, E.-M., Gebert, J. (2016a): Optimized landfill biocover for CH4
453
oxidation I: Experimental design and oxidation performance. Current Environmental 454
Engineering 3, 80-93. 455
Röwer, I.U., Streese-Kleeberg, J., Scharff, H., Pfeiffer, E.-M., Gebert, J. (2016b): Optimized 456
landfill biocover for CH4 oxidation II: Implications of spatially heterogeneous fluxes for
457
monitoring and emission prediction. Current Environmental Engineering 3, 94-106. 458
Scheutz, C., Kjeldsen, P., Bogner, J.E., De Visscher, A., Gebert, J., Hilger, H. A., Huber-Humer, 459
M., Spokas, K. (2009): Microbial methane oxidation processes and technologies for mitigation of 460
landfill gas emissions. Waste Management & Research 27, 409-455. 461
Schroth, M.H., Eugster, W., Gómez, K.E., Gonzalez-Gil, G., Niklaus, O.A., Oester, P. (2012): 462
Above- and below-ground methane fluxes and methanotrophic activity in a landfill-cover soil. 463
Waste Management 32, 879-889. 464
Spokas, K.A., Bogner, J.E. (2011): Limits and dynamics of methane oxidation in landfill cover 465
soils. Waste Management, 31, 823-832. 466
Streese-Kleeberg, J., Rachor, I., Gebert, J., Stegmann, R. (2011): Use of gas push-pull tests for 467
the measurement of methane oxidation in different landfill cover soils. Waste Management 31, 468
995-1001. 469
Tecle, D., Lee, J., Hasan, S. (2009): Quantitative analysis of physical and geotechnical factors 470
affecting methane emission in municipal solid waste landfill. Environmental Geology 56, 1135-471
1143. 472
Templeton, A.S., Chu, K.-H., Alvarez-Cohen, L., Conrad, M.E. (2006): Variable carbon isotope 473
fractionation expressed by aerobic CH4-oxidizing bacteria. Geochimica et Cosmochimica Acta
474
70, 1739-1752. 475
Tyler, S.C., Crill P.M., Brailsford, G.W. (1994): 13C/12C Fractionation of methane during 476
oxidation in a temperate forested soil. Geochimica et Cosmochimica Acta 58, 1625-1633. 477
Urmann, K., Gonzalez-Gil, G., Schroth, M.H., Hofer, M., Zeyer, J. (2005): New field method: 478
gas push-pull test for in situ quantification of microbial activities in the vadose zone. 479
Environmental Science & Technology 39, 304-310. 480
United States Environmental Protection Agency (USEPA), 2016: Inventory of U.S. Greenhouse 481
Gas Emissions and Sinks: 1990-2014. EPA 430-R-16-002. 482
Widory, D., Proust, E., Bellenfant, G., Bour, O. (2012): Assessing methane oxidation under 483
landfill covers and its contribution to the above atmospheric CO2 levels: the added value of the
484
isotope (δ13C and δ18
O CO2; δ13C and δD CH4) approach. Waste Management 32, 1685-1692.
485
Yang, C.B., Park, M., Zhu, C. (2007): A method for estimating in situ reaction rates from push-486
pull experiments for arbitrary solute background concentrations. Environmental & Engineering 487
Geoscience 13, 345-354. 488
List of figure captions
489
Figure 1: Relationship between methane oxidation efficiency fox and respective shift in carbon
490
isotopic signature of methane relative to the injected gas. Data from 22 individual GPPTs with 491
three to six individual points per GPPT (compare also Table 1). 492
Figure 2: Relationship between the fraction of methane remaining during the course of a GPPT, 493
expressed as ln(M/M0), and the change in isotopic signature, expressed as δ13
Ct - δ13Ct=0, for 494
GPPTs 51 (left panel, point DP1) and 59 (right panel, also point DP1). Line = linear fit. 495
Figure 3: Relationship between methane oxidation rate and fractionation factor αox. n = 22. For
496
site IDs see heading of Table 1. 497
Figure 4: Relationship between environmental parameters, methane oxidation rate and 498
fractionation factor αox as found at site DP1 (data from Table 1).
499
List of table captions
500
Table 1: Soil properties, environmental conditions, fractionation factors and oxidation rates for 501
each site. n.d. = not determined. Temp. = temperature, moist. = moisture. Site ID notation: H, K, 502
D, L, A = different MSW landfills in north western Germany; WC, WG = test cell C and test cell 503
G on landfill W in The Netherlands. P1-5 = individual sampling points. 504
Figures
505
506
Figure 1: Relationship between methane oxidation efficiency fox and respective shift in carbon isotopic
507
signature of methane relative to the injected gas. Data from 22 individual GPPTs with three to six individual 508
points per GPPT (compare also Table 1). 509 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 25 30 35 40 45 50 55 60 65
S
h
ift i
n
δ
13C
s
ignat
ur
e [
‰
]
Oxidation efficiency f
ox[%]
510
Figure 2: Relationship between the fraction of methane remaining during the course of a GPPT, expressed as 511
ln(M/M0), and the change in isotopic signature, expressed as δ13Ct - δ13Ct=0, for GPPTs 51 (left panel, point 512
DP1) and 59 (right panel, also point DP1). Line = linear fit. 513 -0.14 -0.12 -0.10 -0.08 -0.06 -0.04 -0.02 0.00 0 1 2 3 4 5 6 7 8 9 δt -δt= 0 [‰ ] ln(M/M0) [v/v] GPPT 51 slope = -46.70 r = -0.995 Oxidation rate 3.6 g m-3 soil air h -1 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0 5 10 15 20 25 30 Oxidation rate 46.2 g m-3soil air h-1 δt -δt= 0 [‰ ] ln(M/M0) [v/v] GPPT 59 slope = -56.87 r = -0.997
0 50 100 150 200 250 300 350 1000 1200 1,00 1,02 1,04 1,06 1,08 1,10 1,12 1,14 1,16 Site ID DP1 AP1 HP5 KP2 LP1 WC1 WG2 WC4 KP3
Frac
ti
onat
ion f
ac
tor
a
ox[-]
CH
4oxidation rate [g CH
4m
-3soil airh
-1]
514
Figure 3: Relationship between methane oxidation rate and fractionation factor αox. n = 22. For site IDs see
515
heading of Table 1. 516
517
Figure 4: Relationship between environmental parameters, methane oxidation rate and fractionation factor 518
αox as found at site DP1 (data from Table 1).
519 0 5 10 15 20 So il t em per atu re (° C) 0 5 10 15 20 Soi l m oi st ur e (v o. %) 0 10 20 30 40 50 CH4 ox id at io n p ot en tia l (g m -3 so il a ir h -1) 1.00 1.02 1.04 1.06 1.08 1.10 1.12 01. 10. 2008 31. 10. 2008 01. 12. 2008 31. 12. 2008 31. 01. 2009 02. 03. 2009 02. 04. 2009 02. 05. 2009 02. 06. 2009 Fr act io na tio n f act or a lpha _o x
Tables
520
Table 1: Soil properties, environmental conditions, fractionation factors and oxidation rates 521
(potentials) for each site. n.d. = not determined. Temp. = temperature, moist. = moisture. Site ID 522
notation: H, K, D, L, A = different MSW landfills in north western Germany; WC, WG = test 523
cell C and test cell G on landfill W in The Netherlands. P1-5 = individual sampling points. 524 GPPT No. Site ID Date Soil temp. Total porosity Soil moist. Air-filled porosity Total shift in δ13 C Background CH4 Background O2 αox CH4 oxidation rate (°C) (vol.%) (vol.%) (vol.%) (‰) (vol.%) (vol.%) (-) (g m-3soil air h-1)
26 HP5 13.10.08 14.9 51.0 24.4 26.6 62.04 0.0 15.0 1.020 91.3 38 HP5 02.02.09 1.0 51.0 23.4 27.6 34.00 0.22 3.1 1.019 50.6 46 HP5 17.03.09 8.1 42.8 24.4 18.4 15.62 0.0 6.6 1.096 6.50 62 HP5 30.06.09 18.0 51.0 24.5 26.5 43.18 0.0 19.1 1.026 104 65 HP5 03.08.09 20.0 51.0 24.9 26.1 23.78 0.0 19.3 1.033 90.9 27 KP2 20.10.08 12.6 34.5 24.3 10.2 52.68 0.0 5.4 1.018 66.5 29 KP2 03.11.08 10.2 34.5 23.3 11.2 36.63 0.0 8.9 1.023 41.9 53 KP2 27.04.09 12.8 34.5 22.0 12.5 17.44 0.24 16.1 1.005 143.5 28 DP1 29.10.08 15.4 42.8 10.3 32.5 23.81 0.008 11.0 1.036 34.2 31 DP1 24.11.08 10.2 42.8 10.0 32.8 24.51 0.0 7.5 1.035 5.50 34 DP1 16.12.08 8.7 42.8 9.9 32.9 20.46 0.0 16.0 1.064 10.0 41 DP1 17.02.09 4.9 42.8 18.2 24.6 11.86 0.0 4.8 1.100 4.30 51 DP1 20.04.09 12.6 42.8 14.9 27.9 8.13 0.0 10.5 1.049 3.56 55 DP1 26.05.09 16.5 42.8 12.4 30.4 25.95 0.0 12.2 1.044 16.4 59 DP1 23.06.09 17.3 42.8 11.5 31.3 26.25 0.0 9.8 1.060 46.2 32 AP1 25.11.08 4.5 n.d. 21.1 n.d. 13.21 0.47 0.6 1.042 3.20 60 AP1 24.06.09 16.4 n.d. 13.9 n.d. 23.57 0.0 13.1 1.042 34.7 44 LP1 09.03.09 4.3 39.8 26.1 13.7 4.16 0.0 17.2 1.151 2.30 103 WC1 13.07.11 17.3 49.2 17.3 31.9 34.19 0.74 4.4 1.006 1110 104 WC4 13.07.11 17.0 49.2 17.1 32.1 32.99 0.0 17.0 1.033 49.5 105 WG2 13.07.11 20.0 43.2 17.1 26.1 8.47 0.0 17.7 1.030 17.3 119 KP3 17.09.12 16.2 32.6 3.5 29.1 24.56 0.0 15.0 1.034 69.1 525
