Waste Management 31 (2011) 914–925

Contents lists available at ScienceDirect

Waste Management
journal homepage: www.elsevier.com/locate/wasman

Observations on the methane oxidation capacity of landfill soils

Jeffrey Chanton a,⇑, Tarek Abichou b, Claire Langford a, Kurt Spokas c, Gary Hater d, Roger Green d,
Doug Goldsmith e, Morton A. Barlaz f
a
Department of Earth, Ocean and Atmospheric Science, Florida State University, Tallahassee, FL 32306-4320, United States
b
Department of Civil and Environmental Engineering, FAMU-FSU College of Engineering, 2525 Pottsdamer Street, Tallahassee, FL 32310, United States
c
US Dept. of Agriculture (USDA) – Agricultural Research Service (ARS), St. Paul, MN, United States
d
Waste Management Incorporated, 2956 Montana Ave., Cincinnati, OH 45211, United States
e
Alternative Natural Technologies, Inc., 1847 Whittaker Hollow Road, Blacksburg, VA 24060-1076, United States
f
Dept. of Civil, Construction, and Environmental Eng., North Carolina State University, Box 7908, Raleigh, NC 27695-7908, United States

a r t i c l e i n f o a b s t r a c t

Article history: The objective of this study was to determine the role of CH4 loading to a landfill cover in the control of
Received 28 May 2010 CH4 oxidation rate (g CH4 m2 d1) and CH4 oxidation efficiency (% CH4 oxidation) in a field setting. Spe-
Accepted 18 August 2010 cifically, we wanted to assess how much CH4 a cover soil could handle. To achieve this objective we con-
Available online 2 October 2010
ducted synoptic measurements of landfill CH4 emission and CH4 oxidation in a single season at two
Southeastern USA landfills. We hypothesized that percent oxidation would be greatest at sites of low
CH4 emission and would decrease as CH4 emission rates increased. The trends in the experimental results
were then compared to the predictions of two differing numerical models designed to simulate gas trans-
port in landfill covers, one by modeling transport by diffusion only and the second allowing both advec-
tion and diffusion. In both field measurements and in modeling, we found that percent oxidation is a
decreasing exponential function of the total CH4 flux rate (CH4 loading) into the cover. When CH4 is sup-
plied, a cover’s rate of CH4 uptake (g CH4 m2 d2) is linear to a point, after which the system becomes
saturated. Both field data and modeling results indicate that percent oxidation should not be considered
as a constant value. Percent oxidation is a changing quantity and is a function of cover type, climatic con-
ditions and CH4 loading to the bottom of the cover. The data indicate that an effective way to increase the
% oxidation of a landfill cover is to limit the amount of CH4 delivered to it.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction States (USEPA, 2007). Landfills may be thought of as point

sources of CH4 to the atmosphere and therefore they make good
The ubiquitous nature of methanotrophic activity in natural targets for mitigation. Landfills are a greenhouse gas source that
bodies of water, wetlands, and human-created rice fields began can be reduced without causing undue societal pressure. At older
to be fully appreciated in the 1970’s (Rudd and Hamilton, 1975; and smaller landfills without gas collection systems some frac-
Reeburgh, 1976; Rudd et al., 1976; Reeburgh and Heggie, 1977; tion of CH4 generated in the landfill passes through the soil
Rudd and Taylor, 1980; Hanson, 1980; Kiene, 1991; King, 1992; where it is oxidized by soil methanotrophic bacteria (Chanton
Hanson and Hanson, 1996). The importance of methanotrophs in and Liptay, 2000; Stern et al., 2007; Abichou et al., 2006a,b).
landfill soils was first recognized by Whalen et al. (1990) who Passive vents at these sites can be treated with biofilters (Powelson
observed the highest rates of CH4 oxidation in any soils examined et al., 2006, 2007; Gebert and Groengroeft, 2006). At modern
to date. Whalen et al. (1990) estimated that methanotrophs were landfills, gas capture for power generation or flaring reduces
able to consume roughly 50% of the CH4 escaping the landfill CH4 emissions considerably. Nonetheless, some fraction of the
surface. Methanotrophs are also able to consume non-methane produced CH4 also escapes these landfills through the soil and
organic compounds (Kjeldsen et al., 1997; Scheutz and Kjeldsen, through leaks in the gas collection system (Spokas et al., 2006;
2005, 2004; Scheutz et al., 2008). Börjesson et al., 2007). Recently, the technique of enhancing
Landfills are responsible for 3–7% of global total CH4 emis- the activity of soil methanotrophic bacteria to further reduce
sions (Lelieveld et al., 1998; Bogner and Matthews, 2003) and fugitive emissions has received considerable attention including
are among the largest anthropogenic CH4 sources in the United recognition from environmental agencies in Finland, Germany
and Canada (Humer and Lechner, 1999, 2001; Huber-Humer
⇑ Corresponding author. Tel.: +1 850 644 7493. and Lechner, 2002; Huber-Humer, 2004; Scheutz et al., 2009).
E-mail address: [email protected] (J. Chanton). Currently, the default value for the IPCC and the USEPA for

0956-053X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.wasman.2010.08.028
 J. Chanton et al. / Waste Management 31 (2011) 914–925 915

landfill cover CH4 oxidation is set at a relatively low value, 2. Methods

between 0% and 10% of emitted CH4 (IPCC, 2006; USEPA, 2004).
This value was based on seasonal results for a New Hampshire 2.1. Methane oxidation rate and percent oxidation
landfill as determined by the studies of Czepiel et al.
(1996a,b). Recently Chanton et al. (2009) reviewed the literature Methane emission rate and stable isotope measurements to
and compiled CH4 oxidation results for 42 determinations of the determine methane oxidation were conducted at two landfills in
fraction of CH4 oxidized following and including Czepiel’s land- the Southeastern USA. Because we wanted to contrast variations in
mark study and reported a mean value of 36 ± 6% for this param- oxidation with variations in emission rate across the surface of a par-
eter. Fifteen seasonal studies ranging from latitude 30° to 55°N ticular cover in several synoptic observations, we used the chamber
yielded a similar value of 35 ± 6%. technique. Landfill X is located in Northwest Florida at 31 degrees
Many factors control rates of CH4 oxidation, among them, pH, North latitude. The climate is humid subtropical. Average annual
moisture, temperature, and nutrient levels. Several investigators rainfall is on the order of 150 cm yr1 and the annual mean temper-
have documented the effect of nitrogen levels on CH4 oxidation ature is 20 °C. Landfill Y is located at 38 degrees North latitude, in a
rates (Boeckx and Van Cleemput, 1996; Boeckx et al., 1996; Bender more temperate but still humid climate. Average rainfall is
and Conrad, 1995; Hilger et al., 2000b). It has recently been 110 cm yr1 and the annual mean temperature is 14 °C. The areas
observed that diurnal temperature cycling has a dynamic impact tested at landfill Y were operated as bioreactors. Soil temperatures
on the rate of oxidation (Spokas and Bogner, 2011). These results at Landfill X varied from 24 to 32 °C during May measurements
suggest that rates may not be adequately predicted from equiva- and 29–36 °C in June measurements. Percent water varied from 8%
lent isothermal incubations (e.g. average daily temperature) and to 21% by weight over both periods. At landfill Y, soils temperatures
may need to be accounted for by calculating the rate through varied from 28 to 37 °C and moisture varied from 5% to 21%.
the actual temperature ramps. Several studies have examined At landfill X, CH4 emissions from the soil were measured for
the effect of different levels of moisture on soil oxidation (Bender three different cells. The North area was a closed cell with a final
and Conrad, 1995; Boeckx and Van Cleemput, 1996; Boeckx et al., composite cover consisting of 45 cm of compacted clay (low per-
1996; Park et al., 2002; Scheutz and Kjeldsen, 2004). Basically, if meability layer) overlain with 1.52 mm (60 mil) LLDPE geomem-
the soil is too dry, the activity of methanotrophic bacteria is brane. A 1 m thick sandy loam soil and a dense vegetative layer
inhibited, and if the soil is too wet, the diffusion of oxygen can were over the geomembrane. Both the flat top and the sloped side
limit their activity. were examined. A second closed cell was a Beanie cap, consisting
The first objective of this study was to examine the extent to of the same composite cover on the top but no geomembrane on
which the rate of CH4 delivery to the soil methanotrophic zones the slopes. A Beanie cap has a geomembrane on the flat surface
controls CH4 oxidation rates (in g CH4 m2 d1) and CH4 oxidation but no geomembrane on the side slopes. The cover was densely
efficiency (% CH4 oxidation) in a field setting. Specifically, we vegetated and a 1 m thick sandy loam soil and a vegetated layer
wanted to assess how much CH4 a cover soil could handle. To covered the flat top and the slopes. The third (South) cell was an
achieve this objective we conducted landfill CH4 emission and intermediate covered area with 30–45 cm of sandy loam soil cover
CH4 oxidation measurements in several synoptic sampling events and sparse to no vegetation. The north area and the beanie cap area
in a single season. We hypothesized that percent oxidation would had completed active gas collection systems. However, the gas col-
be greatest at sites of low CH4 emission and would decrease as CH4 lection system in the south area of landfill X was under develop-
emission rates increased. The trends in the experimental results ment at the time of testing, and additional gas wells are planned.
were then compared to predictions by two differing numerical The existing wells at the south area were screened only in lower
models designed to simulate gas transport in landfill covers, one portions of the waste mass. Emission measurements were con-
by modeling transport by diffusion only and the second allowing ducted at all three areas, and CH4 oxidation measurements were
both advection and diffusion. conducted using the stable isotope approach (Chanton and Liptay,
Ultimately we hope that our work can result in the design of 2000) at the south area.
better landfill covers. For this reason, we evaluated the potential At landfill Y, chamber flux measurements were conducted on
for soils to mitigate CH4 emissions following landfill closure in the flat tops and side slopes of two bioreactor landfill cells (sandy
two cases, with and without gas collection systems. clay) and in one daily covered area. The bioreactors had 2 m of soil
Our second objective was to assess the reproducibility of CH4 overlying the waste, while the daily cover had only 15 cm of sandy
emissions as measured with static chambers and to compare rates loam overlying the waste. These were intermediate covered areas
of emission from sloped areas of the landfill with those from flat and no geomembranes had been installed. Isotopic measurements
top areas. We hypothesized that sloped areas would have higher were conducted at two areas, bioreactor 2 and on the daily cover.
CH4 emissions relative to the flat top areas for two reasons. The first The chambers used in this study were constructed of polished
reason is that rainfall erodes the slopes to a greater degree than the aluminum with a size of 0.63  0.63  0.2 m (covering an area of
flat top areas. A second reason that sloped areas might have greater 0.4 m2). Within them was a small fan to circulate air. Chambers
CH4 emissions has to do with the manner in which waste is placed in were sealed to the ground by clamping them to pre-installed col-
landfills. Waste is placed in horizontal layers, with a daily soil cover lars. Total volume was 80–100 l. Five to six methane samples were
overlying each layer. This type of placement allows for greater lateral collected from each chamber sequentially over a 25 min period
movement of gas, relative to vertical movement. using 60 mL disposable syringes (Becton, Dickinson, and Co.) fitted
These measurements were made in summer when CH4 oxida- with plastic stopcocks (Cole Parmer Instrument Co.). Samples were
tion in the Southeastern USA has been shown to be at a maximum analyzed on a gas chromatograph equipped with a flame ionization
(Chanton and Liptay, 2000). As our objective was to determine the detector within 20 h of collection. Methane flux was determined
maximum amount of methane that a landfill could oxidize, a sum- from concentration data (C in ppmv) plotted versus elapsed time
mer measurement program was appropriate to achieve our objec- (t in minutes). The CH4 concentration within the chambers gener-
tives. We assumed that % moisture, temperature, nutrient levels, ally increased linearly, in which case dC/dt is the slope of the
and other factors that control CH4 oxidation would be relatively fit to the data. This change in volumetric concentration was con-
constant across the synoptic measurements that were made at a verted to a mass flux by using the ideal gas law. The CH4 flux, F
single site. (g CH4 m2 d1), was calculated as follows:
916 J. Chanton et al. / Waste Management 31 (2011) 914–925

F ¼ PVMUðdC=dtÞ=ðATRÞ ð1Þ exceptions were made if there were few measurements of similar
emission value. In the May data from Landfill X, the point at 1330
where P is pressure (1 atm), V is chamber volume (80 L), M is the
on the x-axis only contains two measurements and the point at
molar mass of CH4 (16 g CH4/mol), U is the units conversion factor
2125 g m2 d1 only contains one measurement. For the June land-
(0.00144 L min/(lL d)), A is the surface area covered by the chamber
fill X data the two highest points in terms of emission only have
(0.4 m2), T is chamber temperature (Kelvin), and R is the gas con-
two values in each bin. For landfill Y data the two highest points
stant (0.08205 L atm/(K mol)). The slope of the line, dC/dt, was
in terms of emission only have two values in each bin.
determined by linear regression between CH4 concentration and
The fractionation factor (aox) was determined from the mea-
elapsed time. Following the general approach of Barlaz et al.
sured soil temperature (T, °C) using the regression equation for
(2004), a non-zero flux was reported only if the there was 90% con-
aox with temperature (Chanton et al., 2008a). The parameter atrans
fidence (p < 0.1) in the correlation between CH4 concentration and
was assumed to be 1. This assumption requires that CH4 transport
time, otherwise a zero-flux is reported. The zero value flux rates
is dominated by advection, a process that does not cause isotopic
resulted from measurements which showed little increase or
fractionation (Bergamaschi et al., 1998; Liptay et al., 1998). Recent
decrease in CH4 concentration over time and thus no correlation
laboratory experiments and field studies have shown that this
of CH4 with time.
approach can underestimate CH4 oxidation by as much as a factor
Chambers were placed on a grid pattern with chambers 10–
of 2 by not taking into account diffusive flux (De Visscher et al.,
15 m apart. The geospatial mean CH4 emission rate for the area
2004; Chanton et al., 2008b). Thus the oxidation values reported
was determined by use of inverse distance weighing (IDW) using
here represent lower limits of CH4 oxidation.
‘‘Surfer,” developed by Golden Software, Inc., Golden, CO (Surfer,
The rate of CH4 oxidation, Rox (g CH4 m2 d1) was calculated
2002; Spokas et al., 2003; Abichou et al., 2006a,b). Abichou et al.
from flux and percent oxidation using the following equation:
(2006b) compared the IDW approach with point kriging and found
that the geospatial means in this area obtained by both methods
F
 
were almost identical. Both results were similar to the arithmetic Rox ¼ fox ð5Þ
1  fox
mean. Spokas et al. (2003) reported similar findings.
Methane oxidation was determined from the stable isotope
approach. Stable isotopes for initial and final samples from each where fox is the fraction oxidized (% oxidized/100), calculated from
chamber were collected using 60 mL disposable syringes fitted Eq. (4), and F is flux (g CH4 m2 d1), calculated from Eq. (1). Oxi-
with plastic stopcocks and immediately transferred to evacuated dation rate can only be calculated when a positive flux is mea-
glass vials. Samples were only analyzed when the flux was positive sured, as the d13C value of the residual positive flux is required
to determine the carbon isotopic composition of residual CH4 fol- (Eq. (4)) to obtain fox. Because areas where zero or negative fluxes
lowing oxidation as it passed through the soil beneath the cham- were observed may be indicative of high rates of CH4 oxidation
ber. The d13C of residual CH4 was determined from the equation: (100% oxidation of CH4 from below), the rates calculated with
Eq. (5) are lower limits.
ðdF  C F Þ  ðdI  C I Þ Stable carbon isotopes values were measured by direct injection
dR ¼ ð2Þ
CF  CI into a Hewlett–Packard Gas Chromatograph coupled via a combus-
tion interface to a Finnigan Mat Delta S Isotope Ratio Mass Spec-
where dR is the d13C value of the residual CH4 emitted from the
trometer (GCC-IRMS) following Merrit et al. (1995). Samples with
landfill, dI and dF are the initial and final d13C values of CH4 mea-
small concentrations (<4000 ppm) were cryogenically focused
sured at the initiation and completion of the flux measurement,
using a device coupled to the front end of the GC. Replicates were
and CI and CF are the initial and final CH4 concentrations.
analyzed for most samples, yielding a standard deviation of
The d13C values for dR and anoxic zone CH4 (dA), that is unex-
approximately 0.15‰. Values are reported in the ‘‘d” scale in ‰ rel-
posed to methanotrophic bacteria can be used to calculate the per-
ative to the standard, VPDB (Vienna Pee Dee Belemnite).
centage of CH4 oxidized, provided we know the carbon isotopic
fractionation factor for bacterial oxidation. This parameter, a, is a
measure of the bacteria’s preference for the light isotope over the 2.2. Diffusion-based model
heavy isotope, given by:

aox ¼ kL =kH ð3Þ A comparison to a one dimensional mathematical diffusion/oxi-

dation model was conducted at Landfill X south intermediate cover
12
where kL and kH refer to the rate constants of the light ( CH4) and area to compare the field and model response of CH4 oxidation rate
heavy (13CH4) isotopes. to variation in rates of CH4 emission. The model accounts for cli-
The fraction of CH4 (fox) oxidized (CH4 oxidation efficiency) in matic influence on soil CH4 oxidation activity (Bogner et al.,
upward transit through the landfill cover soil is then given by 2009). As mentioned above, soil moisture and temperature play
(Chanton and Liptay, 2000; De Visscher et al., 2004): critical roles in determining the resulting rate of oxidation. This
ðdR  dA Þ model utilizes the average annual cycle of air temperature, solar
fox ¼ ð4Þ radiation and precipitation patterns coupled to a soil heat and
1000  ðaox  atrans Þ
water transport model (STM2) (Spokas and Forcella, 2009) to arrive
where dR is calculated using Eq. (2) and dA is the carbon isotopic at the soil microclimate conditions (soil temperature and mois-
content of anoxic CH4 sampled from gas wells, and aox and atrans ture) through the landfill cover soil for each minute of the year.
are the isotope fractionation factors for methane oxidation and By coupling these results with a gas diffusion–oxidation model,
associated with transport of CH4, respectively. the potential climatic influence for any particular landfill can be
To achieve greater clarity of presentation, examine the mean estimated from these predictions of the soil conditions. In this
trends in the data, and estimate variability, the % oxidation data fashion, the climatic influence of oxidation can be accounted for
were binned according to emission rate and then averaged and at the site. Various flux rates of CH4 through this intermediate cov-
standard error calculated as (standard deviation/ðsqrtðnÞÞ. At low er were simulated by altering the CH4 concentration at the base of
fluxes we binned all data within the range of 0–1, 1–2, 2–3, 3–4, cover from 100 ppm to 50% CH4. Since this is a diffusion-based
4–5 g CH4 m2 d1 and beyond that, with increasing emission model, changing the concentration gradient leads to differing flux
rates, we grouped the data to generally include 3 values. Some rates of CH4 into the cover soil.
 J. Chanton et al. / Waste Management 31 (2011) 914–925 917

2.3. Advection–diffusion-based model introduced for the biological oxidation parameters similar to
De Visscher and Van Cleemput (2003). Biological oxidation was also
In a landfill setting, water content, temperature, and barometric corrected for moisture content based on Boeckx et al. (1996).
pressure are constantly changing depending on climate conditions, A constant mass flux escaping the gas collection system was set
soil type, cover thickness, and vegetation. We also applied a numer- as the lower boundary of the gas transport model. An equal flux of
ical model to attempt to re-produce the data from Southeastern USA CO2 (volume based) was also assumed to be transported with CH4.
landfills. This model combines water and heat flow with a gas trans- The gas transport model then used the daily soil water and temper-
port and oxidation model (Abichou et al., 2008, 2009; Yuan et al., ature profiles along with the daily flux into the bottom of the land-
2009). One of the key input terms in this model is the landfill gas fill cover to estimate CH4 oxidation in the landfill cover and
pressure and CH4 concentration in the waste at the base of the cover. therefore estimate surface emissions. Simulations were performed
Based upon the gas conductivity of the soil, which is determined as a for different values of flux into the bottom of the soil cover (bottom
function of volumetric water content, CH4 advection towards the flux).
surface is modeled and attenuated by CH4 oxidation to predict sur-
face emissions. This model does not (yet) include the effect of 3. Results
macro-pores or duel phase gas permeability.
The volumetric water content and the temperature profile of 3.1. Results of emission measurements
the landfill cover were resolved with simulations of water and heat
flow in variably saturated soils using HYDRUS1D v3.0 (Simunek Representative replicate determinations of CH4 emissions are
et al., 2005). Volumetric water content and temperature were gen- shown for landfill X in Figs. 1 and 2. For the South flat top area mea-
erated at each node each day by HYDRUS1D. The gas transport surements were conducted on May 23, May 24, and June 14, 2006.
model used these dynamic results to simulate CH4 emission and The geospatial means were 22.5, 40.4, and 22.7 g CH4 m2 d1,
oxidation. Continuity and mass balance equations were then used respectively (Table 1). For the South side slope area measurements
to describe the gas flow and reaction within the porous media. The were conducted on May 30 and May 31, 2006. The geospatial means
reaction component of the gas transport equation was assumed to of these measurements were 199 and 236 g CH4 m2 d1, respec-
be in accordance with the reaction rate of methanotrophic bacteria tively. The arithmetic means were similar. As described previously
calculated by Michaelis–Menton kinetics (De Visscher and Van gas collection wells in this portion of the landfill were screened only
Cleemput, 2003). at depth. Gas migrated laterally which explains the high fluxes mea-
The CH4 oxidation rate is calculated by dual-substrate Michae- sured on the slopes of this area. Repeated flux measurements for the
lis–Menten kinetics from incubation experiments using the exact same areas were also conducted at the flat tops and slopes of
equation, two final covered North and Beanie areas as described above, and
CCH4 CO 2 at a second site along the slope of the intermediate covered South
r CH4 ¼ V max ð6Þ area.
K m½CH4  þ CCH4 K m½O2  þ CO2
At landfill Y, repeated measurements were conducted at one
where Vmax is the maximum methane consumption rate site. The mean of the replicate measurements for all 8 sites where
(nmol s1 g1dry soil) and Km is the half saturation constants for oxy- measurements were repeated at the two landfills is shown (Ta-
gen and methane (mol m3). Vmax and Km are estimated from incu- ble 1) and the coefficient of variation (CV). The average CV for
bation experiments in the laboratory. The Michaelis–Menten the 8 cover types was 32%.
parameters, Vmax and Km, are corrected for variation in soil temper- At landfill Y, the geospatial mean values were 12.8 g CH4 m2 d1
ature and water content (Abichou et al., 2009). In the model, the air for the daily cover and 2.45 and 82.6 g m2 d1 for the slopes on bio-
filled porosity e, diffusion coefficient D, and gas permeability k are a reactors 1 and 2. On the bioreactor top flat area, emissions were 9.7
function of volumetric water content. A temperature correction was and 2.8 g CH4 m2 d1, respectively.

2.88 13.94 10.61

580 0 0.534 7.29 580 0 1.13 4.114 580 72.76 74.77 1.599

0.159 -0.0807 1.34 14.3 1.507 -0.07713 3.171 142.7 4.429 0.219 49.95 66.73
1500 480
560 W 126 460 560 W 126 560 W 126 460
440 1400 440
420 420
0.42 -0.179 0 2.64 0.8965 0 1.438 1.608
1300 2.797 0 0.1256 0.4863
400 400
380 1200 380
360
North (m from origin)

North (m from origin)

540 540 540 360

North (m from origin)

340 1100
340
9.67 0.145 -0.0266 0 320 14 -0.3134 0 -0.0534 8.11 6.525 0 0.2408
1000 320
300 300
280 900 280
260 800 260
520 0.206 31.6 0.722 0.243 240 520 0 11.35 1.022 -0.1249 520 -0.2316 16.09 2.312 0.05478 240
220 700 220
200 200
600
180 180
4.37 1.09 5.07 -0.2 9.062 0.07588 2.582 -0.05428 1.239 0.1225 5.453 -0.1393
160 500 160
500 140 500 500 140
120 400
120
W 125 W 125 W 125
100 300 100
-0.369 0 0.186 2.6 0.1796 0.1013 0.2731 4.736 11.5 27.35 0 11.48
80 80
60 200 60
480 40 480 100 480 40
20 20
0 0 -0.125 3.89 -0.1503 0 0.2111 3.013 0.09726 0 0 0.5963
10 25 10
5 0
0 0
460 0.563 0 854 0 460 0.5042 0 1548 -0.04075 460 0 0.1674 471.2 -1.541

440 440 440

50 60 70 80 90 100 50 60 70 80 90 100 50 60 70 80 90 100
East (m from origin) East (m from origin) East (m from origin)

Fig. 1. Replicated fields of chamber measurements made at landfill X, South flat top area May 23, May 24, and June 14, 2006. The geospatial means of these measurements
were 22.5, 40.4, and 22.7 g CH4 m2 d1, respectively.
918 J. Chanton et al. / Waste Management 31 (2011) 914–925

0 254.3
620 620
0.2875
5.751

-0.4922 -1.08
49.56 43.55
600 0.8724 600 1.62
20.88 -0.08192 -0.08377 56.23
2400 2600
29.28 2300 29.45
64.17
2200 55.57
2400
5.869 0.246 2100 5.888 0.2126
580 2000 580 2200
1900
1800 2000
2251 152.4 2313
8.03 1700 152.2
2.815
0 1800
North (m from origin)

0
1600

North (m from origin)

1500 1600
560 560
207.9
1400
20.82
1300 1400
288.5 426
0 27.78 1200 0.2772 27.86
1100 1200
16.69 1000 14.85

540
20.82
900 17.2 1000
-1.417 540 0.3472
800
286 271 800
700
600 600
0 500 307
187
30.79 0 400 195
28.5 -0.4454 400
520 300 520
200 200
100
53.1
3.685
50 19.22
3.628 50
0 2337 2.362 2557
0
0
500 500

334.3 1192
0.2477 3.232 1320
0.6072

480 480
0 10 20 30 40 50 0 10 20 30 40 50
East (m from origin) East (m from origin)

Fig. 2. Replicated fields of chamber measurements made at landfill X, south side slope area May 30 and May 31, 2006. The geospatial means of these measurements were 199
and 236 g CH4 m2 d1, respectively. Elevated emissions were due to an incomplete gas collection system.

3.2. Results of oxidation measurements to increasing substrate concentrations. The rate increases linearly
initially, and the levels off to a relatively ‘‘flat” value. In Fig. 5C,
At the sites where CH4 oxidation was determined with the iso- the emission rate does not reach sufficient magnitude to allow
tope technique, the % oxidation was observed to vary from near observation of this plateau in the oxidation rate. The results in
100% to near 0% (Figs. 3A, 4A and 5A). Higher % oxidation was gen- Fig. 5D, lower panel, appeared to fall into two populations, emis-
erally associated with lower emission rates, while at higher emis- sion that passed through the soil and was exposed to oxidation
sion rates % oxidation values were lower. At low emission rates, % (closed symbols), and that which apparently without exposure to
oxidation values ranged from 0% to 100%, but as CH4 emissions oxidation (opens symbols). The values (open symbols) that fell
increased, the range in % oxidation decreased and the higher % along the x-axis were excluded from the regressions. The justifica-
oxidation values were not observed. Binned data (see methods) tion for this exclusion was that these data did not appear to exhibit
are plotted as the mean of the % oxidation values within each bin CH4 oxidation and we wanted to determine the relationship for
(y-axis) and their standard error (Figs. 3B, 4B and 5B). Midpoint those samples that did. The cover here was 1.5–2 m thick, so it
and range of each bin are shown on the x-axis. The data were then may be that cracks could allow for bypassing of the main zone of
fit to exponential curves. The exponential relationships between oxidation, as well as allowing rapid transit of CH4 across the cover.
decreasing % oxidation and increasing CH4 emissions were The isotope technique, applied with chambers, is apparently able
significant, p values varied from <0.01 to <0.001, confirming our to capture this phenomenon.
hypothesis that at lower rates of emission, % oxidation would be
greater. Overall, the exponential fits were constrained by low % 3.3. Model results
oxidation results at higher fluxes to yield y intercepts ranging
from 0.21 to 0.43. However, the average % oxidation values for Fig. 6 illustrates the results from the diffusion-based mathemat-
emission rates below 10 g CH4 m2 d1 for Figs. 3–5 were 39.4 ± 6.2%, ical modeling conducted for the intermediate area at landfill X. In
41.9 ± 6.0% and 54.5 ± 8.3%, consistent with our hypothesis that % Fig. 6A, the exponential trend that was observed in the field mea-
oxidation varied inversely with increasing emission rate. It should surements (Figs. 3B, 4B, and 5B) is also seen in the model results for
be stressed that these are absolutely minimum values for % oxida- varying flux rates of CH4 (set by varying concentrations of CH4 at
tion as no correction was applied to the isotope data for the effects the base of the cover soil). The reason for this behavior in the mod-
of diffusion. el is the finite capacity of the soil to oxidize CH4. For the current
We also plotted the CH4 oxidation rate, calculated with Eq. (5) model, a maximum CH4 oxidation rate of 400 ug CH4 gsoil1 d1
above, and found significant positive correlations between the was used, based on laboratory incubations (Spokas and Bogner,
oxidation rate and the emission rate at lower emission rates, e.g., 2011). Using a bulk density of 1.6 g cm3 for the cover soil and
below 300 g CH4 m2 d1 (Fig. 3C), 500 g CH4 m2 d1 (Fig. 4C), optimum conditions present throughout the cover, this would
and 305 g CH4 m2 d1 (Fig. 5D). The criteria we used to select result in 195 g CH4 m2 d1 ultimate oxidation capacity for this
these ‘‘cut-off” points for the linear regressions was that beyond intermediate landfill cover. This value is close to the field data,
the ‘‘cut-off” x-axis values the methane oxidation rates (y-axis) which illustrate a plateau value of about 200 g CH4 m2 d1
reached a plateau or decreased. Figs. 3C and 4C resemble a Michae- (Figs. 3C and 4C), particularly when the field emissions saturated
lis–Menten enzyme kinetic plot of a biological reaction’s response the oxidation capacity. The diffusion model accounting for soil
 J. Chanton et al. / Waste Management 31 (2011) 914–925 919

Table 1 100%
Geospatial mean of chamber-determined CH4 emission rates for different cover types
at landfills X and Y. The mean of the replicate measurements is shown and the 80% A

% oxidation
coefficient of variation (CV) is calculated as 1/2 the range of the measurements
divided by the mean times 100. The average of the CV for the 8 cover types was 32%. 60%

Landfill area CH4 emission CH4 emission CV 40%

rate rate mean
(g m2 d1) (g m2 d1) 20%

Landfill X 0%
North top 0 500 1000 1500 2000 2500
15-May 26.1 20.0 30.8 -2 -1
CH 4 emission g m d
5-Jun 13.8
North slope
22-May 111.3 62.7 77.5 60%
14.1
Beanie top
50% B

% oxidation
17-May 1.3 1.2 5.8 40% y = 0.31e -0.0009x
8-Jun 1.1 r = 0.80, n = 18, p < 0.0001
Beanie slope 30%
18-May 0.6 0.4 37.1
8-Jun 0.3 20%
South flat
10%
23-May 22.5 28.6 21.1
24-May 40.5 0%
14-Jun 22.7 0 500 1000 1500 2000 2500
South slope 1
30-May 199.0 217.6 8.5
CH 4 emission g m-2 d -1
31-May 236.1

-2 -1
South slope 3 300
1-Jun 123.1 83.4 47.7 oxidation rate g m d 250
y = 0.39x + 4.98 r = 0.67, n = 56, p<0.0001
12-Jun 43.6 C
200
Landfill Y
Bioreactor 1, flat top
150
29-Jun 6.95 9.7 28.2 100
1-July 12.4 50
Bioreactor 1, side slope 0
13 July 2.45 0 500 1000 1500 2000 2500
Daily covered area
2-July 12.8 CH 4 emission g m-2 d -1
Bioreactor 2, flat top
3 and 5 July 2.8 Fig. 3. Measurements in May 2006, in first round of sampling, landfill X, South area.
Bioreactor 2, side slope Panel A, % CH4 oxidation versus CH4 emission rate, individual chamber fluxes. Panel
10 and 11 July 82.6 B, % oxidation versus CH4 emission rate, binned data, see text. Bins indicated by x-
Average CV % 32.1 axis error bar. Panel C, CH4 oxidation rate versus CH4 emission rate. Solid diamonds,
n=8 fit to linear regression represent data with emission rates less than
300 g CH4 m2 d1. Open symbols were not included in the regression. The criteria
we used to select these ‘‘cut-off” points for the linear regressions was that beyond
the ‘‘cut-off” x-axis values the methane oxidation rates (y-axis) reached a plateau or
decreased.
temperature and moisture conditions predicted a seasonal vari-
ability in net surface emission of 150–325 g CH4 m2 d1.
As shown in Fig. 6B, the overall annual average for the oxidation one dimension diffusion transport assumptions and be solely a
rate predicted by the model is around 100 g CH4 m2 d1. How- limitation of the current model not accounting for advective gas
ever, there is a discrepancy in the predicted percent oxidation transport and preferential flow, which could also impact soil mois-
between the model [illustrating 100% oxidation at low CH4 flux ture availability.
rates (<100 g CH4 m2 d1)] and field observations not always Fig. 7 shows the results obtained from the simulations per-
illustrating complete oxidation at these low flux rates, with the linear formed using the advection-based model. Fig. 7a shows the same
increase in CH4 oxidation rate as a function of net flux rate, as trend observed in field measurements and in the diffusion-based
discussed above. The field data (Fig. 3C) has a slow gradual increase model. When a curve fit is added to the simulation results
to the plateau, and the model results (Fig. 6B) possess a virtual step (Fig. 7a), the following equation is obtained (r2 = 0.97):
function behavior (no gradual increase to the plateau). This differ-
ence could indicate that despite the fact that conditions are favor- Fraction oxidized ¼ 0:9823  expð0:0044  surface emissionsÞ
able for oxidation at a particular depth; the CH4 in the field setting
ð7Þ
is not being oxidized to the extent predicted by the model. As pos-
tulated above, these reduced oxidation rates could indicate the One can think of this equation as a unique relationship between
presence of macro-pore flow (e.g. cracks, etc.), less than optimal surface emissions and fraction oxidized for an interim landfill cov-
rates of oxidation due to past environmental conditions, the pres- er located in the same climatic conditions. Abichou et al. (2010)
ence of exopolymeric substances (EPS) (Hilger et al., 2000b) that has developed such relationships for several cover types for the dif-
reduces oxygen availability, or even complications due to diffusion ferent climates of the state of California. The results from the field
differences between the isotopes (De Visscher et al., 2004). These data and the modeling results indicate that percent oxidation
diffusion differences would be further complicated by temporal should not be considered as a constant (10% or any other single
differences in the fractionation factor as a function of temperature value). Percent oxidation is a changing quantity and is a function
and soil moisture at different depths within the CH4 oxidation of cover type, climatic conditions and CH4 loading to the bottom
zones. Additionally, this discrepancy could also result from the of the cover.
920 J. Chanton et al. / Waste Management 31 (2011) 914–925

100% 100%

80% 80% A
A

% oxidation
% oxidation

60% 60%

40% 40%

20% 20%
0%
0%
0 100 200 300 400 500 600
0 500 1000 1500 2000
CH 4 emission g m-2 d -1
CH4 emission g m-2 d -1

60% 60%

50% B 50% B
40%
% oxidation

% oxidation
40% y = 0.21e-0.0009x
y = 0.43e -0.0074x

r = 0.52, n = 23, p < 0.01 r = 0.74, n = 17, p = 0.0007

30%
30%
20%
20%
10%
10%
0%
0% 0 100 200 300 400 500 600
0.00 500.00 1000.00 1500.00 2000.00
-2 -1 CH 4 emission g m-2 d -1
CH4 emission g m d

180
-2 -1

250 C
oxidation rate g m d

y = 0.24x + 3.26 r = 0.62, n = 82, p < 0.0001 160

y = 1.57x - 3.44, r = 0.75, n = 26, p < 0.0001
-2 -1

200 140
oxidation rate g m d

C 120
150
100
100
80
50 60
0 40
0 500 1000 1500 2000 20
CH4 emission g m-2 d -1 0
0 10 20 30 40 50 60 70

Fig. 4. Measurements in June 2006, in second round of sampling, landfill X, South

CH 4 emission g m-2 d -1
area. Panel A, % CH4 oxidation versus CH4 emission rate, individual chamber fluxes.
Panel B, % oxidation versus CH4 emission rate, binned data. Bin range indicated by x- 70
axis error bar. Panel C, CH4 oxidation rate versus CH4 emission rate. Solid diamonds,
60 y = 0.17x + 3.25
fit to linear regression represent data with emission rates less than r = 0.91
-2 -1
oxidation rate g m d

500 g CH4 m2 d1. Open symbols were not fit to the regression. The criteria we 50 n = 23
used to select these ‘‘cut-off” points for the linear regressions was that beyond the p < 0.0001
40
‘‘cut-off” x-axis values the methane oxidation rates (y-axis) reached a plateau or
decreased. 30
20 D
10
Fig. 7b shows a similar trend as was observed in the field data. The
maximum oxidation rate was approximately 120 g CH4 m2 d1. 0
0 100 200 300 400 500 600
The modeled maximum oxidation rate is lower than that observed
CH 4 emission g m-2 d -1
in the field because the model results are an average daily rate for
an average year of climatic conditions. Measured higher oxidation Fig. 5. Measurements in July 2006 at landfill Y. Panel A, % CH4 oxidation versus CH4
rates might be due to higher than average temperatures during the emission rate, individual chamber fluxes. Open symbols were from the daily
field sampling campaign. The oxidation rate declines in Fig. 7b due covered area, closed symbols were from bioreactor 2. Panel B, binned data, %
oxidation versus CH4 emission rate from both areas. Bins indicated by x-axis error
to the upwards flow of CH4 from the landfill limiting the diffusion
bar. Panel C, CH4 oxidation rate versus CH4 emission rate for soil daily cover. Panel
of oxygen into the soil. Such effects were not observed in the diffu- D, CH4 oxidation rate versus CH4 emission rate for the bioreactor clay cover, solid
sive model (Fig. 6). diamonds, fit to linear regression represent data with emission rates less than
305 g CH4 m2 d1. Open symbols were not fit to the linear regression (see text).

4. Discussion
repaired and emissions reduced by approximately 60% (Nathan
4.1. Methane oxidation rate and percent oxidation Swan, Cygnus Environmental, personal communication, 2010.)
At landfill ‘‘Y” emissions ranged from 2 to 83 g m2 d1. Four of
Geospatial mean landfill CH4 emissions ranged from 0.3 to the areas examined exhibited emissions below 15 g m2 d1, ele-
236 g m2 d1 as a function of cover type (Table 1). At landfill X, vated emissions were observed only at one side slope 83 g m2 d1.
CH4 emissions varied from 0.4 to 63 g CH4 m2 d1 on the final The bioreactor cells at landfill ‘‘Y” were not elevated relative to the
covered (North area and Beanie) areas, and from 29 to cells at landfill ‘‘X”. The best maintained final covers only allowed
218 g CH4 m2 d1 at the intermediate covered area. The North cell the emission of small quantities of CH4 (1.3 g m2 d1) while
of the landfill had a number of cover penetrations for gas wells and covers in different stages of development or disrepair obviously
leachate recirculation wells and CH4 emission occurred primarily released more CH4. Five landfill flat tops were compared with
around these penetrations. These penetrations were subsequently slopes in this study, and in 3 of 5 cases, the slopes had greater
 J. Chanton et al. / Waste Management 31 (2011) 914–925 921

Fig. 6. Diffusion based model output reproduction of field data. Panel A, % oxidation
versus surface emitted CH4, Panel B, oxidation rate versus surface emitted CH4.

emissions by a factor of 3, factor of 7, and a factor of 30. In two

cases the slopes emitted less CH4 than did the tops. Thus while
slopes have the potential to contribute higher emissions relative
to flat tops, this is not always the case. Well maintained side slopes
where the effects of erosion have been either prevented or repaired
are not necessarily greater sources of CH4 than top areas.
As previously reported, (Bogner et al., 1997; Czepiel et al.,
1996b) CH4 emissions from the landfill surface were dominated Fig. 7. Results of simulations performed using pressure based model. Panel A, %
oxidation versus surface emitted CH4. Panel B, oxidation rate versus surface emitted
by ‘‘hotspots”. But this study offers clear evidence that these fea- CH4.
tures persist at the same location, at least on timescales of weeks
(Figs. 1 and 2). The persistence of these features indicates that with
increased monitoring some hot spots could be identified, repaired
and the cover emissions attenuated by coverage with bio-cells (Figs. 2 and 6). These model results further support the hypothesis
(Abichou et al., 2006a, b). Furthermore, the reproducibility of the that the higher observed surface emissions (>350 g CH4 m2 d1)
chamber technique for measuring methane emission patterns result from non-diffusive transport through the entire cover or
averaged 32%, n = 8. This reproducibility includes measurement non-oxidized CH4 (e.g. macro-pore flow).
error and short term temporal variability. These observations pertain The advection-based model shows similar behavior for % oxida-
to this specific period of measurement under these climatic condi- tion versus emission (Fig. 7a). However in this model, oxidation
tions and this cover soil. Diurnal variations in flux measurements rate decreases as emissions continue to increase (Fig. 7b). In this
from landfills have not been found to be discernable (G. Hater, model, the gas pressure below the landfill cover controls the flux
2010, unpublished data). of CH4 into the cover soil. As pressures increase the upward flow
Based on the field measurements of landfill cover soils and of CH4 reduces the diffusion of oxygen from the upper boundary
confirmed by both models (Figs. 3–7), there is a finite limit to and thus reduces the amount of CH4 that can be oxidized. The dif-
the absolute soil CH4 oxidation capacity. If this is expressed in a fusive model described above does not account for these advective
percentage, the percent oxidation is a decreasing exponential transport mechanisms. However, both models have similar predic-
function of the total CH4 flux rate into the bottom of the cover. tions for the percent CH4 oxidation occurring in the cover soil as a
Once this oxidation limit is reached, increasing the delivery of function of CH4 emission (Figs. 6A and 7A).
CH4 to the soil does not continue to increase the rate of oxidation, Landfill cover soil oxidation is conducted by a biological system
which stays constant. and it behaves as such. When substrate (CH4) is supplied, the cov-
The diffusion model predicted a seasonal variability in net sur- er’s rate of CH4 uptake is linear to a point, and then the system
face emission of 150–325 g CH4 m2 d1, as a function of the vari- apparently becomes saturated. This is a classic enzyme response.
able soil temperature and moisture. These predicted values are The reaction rate increases linearly with substrate concentration.
very close to the geospatial means of two landfill X field measure- Methanotrophs in the oxic zone of the soil are limited by CH4.
ments (199 and 236 g CH4 m2 d1 – Fig. 2). Overall, the diffusion- When the oxidation rate levels off, at higher emission rates, the
based mathematical model duplicates the observations in the methanotrophic community is apparently limited by some other
behavior between % oxidation and observed net emission rate factor, presumably either oxygen or microbial population.
922 J. Chanton et al. / Waste Management 31 (2011) 914–925

Additionally, the response could be due to hysteresis in micro- et al., 2007; Huber-Humer, 2004; Huber-Humer et al., 2008; Huber
bial response to the presence of CH4 (whether concentrations are et al., 1999). There is also evidence that a healthy vegetated cover
increasing or decreasing; Spokas and Bogner, 2011). These driving can enhance biotic CH4 uptake by providing an optimum environ-
factors vary from one landfill climate type to another. However, as ment for bacteria and by facilitating oxygen transport into the soil
seen in the modeling results, the value of 400 ug CH4 gsoil1 d1 via an increased surface/atmosphere contact (Hilger et al., 2000a;
appears to be near a value for ultimate capacity of landfill soils Maurice, 2001; Wang et al., 2008). Organic matter amended and
at least from the Southeast US sites used in this study. Incidentally, vegetated soils have the added benefit of reducing cracks and fis-
this oxidation rate also is the ultimate rate observed in soils from sures in clay soils, thus reducing the input of gas to the bottom
Western US (California) sites (Spokas and Bogner, 2011). of covers and allowing for more efficient gas collection by extrac-
The two trends in the data may seem contradictory. If CH4 oxi- tion systems (Stern et al., 2007).
dation rate is a constant function of emission rate, how can the % For example, Scheutz et al. (2009) compiled and tabulated results
oxidation decrease with emission rate? The behavior is due to a from laboratory column studies that determined CH4 oxidation %
non-zero intercept in the oxidation rate versus emission rate data, and rate from the difference between the CH4 input and output at
and also possibly due to the considerable scatter in the data. the column top. Mineral soils and organic rich soils (compost) which
An additional complicating factor is the degree to which the CH4 are often used for bio-covers were compared. These results are pre-
is able to short circuit exposure to the soil microbial community by sented graphically in this paper in Fig. 8. For both soil types there was
bypass flow through cracks, and other conduits. As observed in no relationship between the column inflow rate and % oxidation. The
Fig. 5D, at this particular soil cover, at emission rates above lack of a relationship for mineral soils is possibly due to a dearth of
100 g CH4 m2 d1 a subset of the emissions experience no CH4 studies conducted at low flow rates, below 150 g CH4 m2 d1. For
oxidation. We hypothesize that this CH4 is bypassing the methan- organic materials a range of values was obtained but under the best
otrophic bacteria. Since these emissions were captured in cham- of conditions it is apparently possible to attain quantitative oxida-
bers, these chambers must have covered cracks or fissures of tion at flow rates as high as 600 g CH4 m2 d1. Similar results have
sufficient size to allow the CH4 to pass through the cover with min- been obtained in biocover simulations at a larger scale (Cabral et al.,
imal interaction with methanotrophic bacteria. In addition, it could 2010). A plot of oxidation rate as a function of column input rate
be hypothesized that just as CH4 is able to migrate through cracks, (Fig. 8B) shows the typical falling off of oxidation rate at higher
there could be occurrences of CH4 being trapped within the soil CH4 flux values in some cases, however. Nonetheless, it appears that
and then being oxidized to a higher extent than the CH4 that is free landfill cover soils can be improved to increase CH4 oxidation
to move through the soil. Gas can be trapped or its rate of transport percentage by amending them with organic matter and vegetation.
slowed due to variation in soil conductivity and tortuosity. This As noted above, this will also serve to reduce desiccation cracks
leads to a scatter on both sides of the average flux value. However,
overall the geospatial means of the flux rates were within the same
order of magnitude as the flux predicted from the diffusion-based 100%
model. This leads us to suggest that these phenomenons (crack
transport and trapped CH4) are not the dominant processes in 80%
landfill CH4 transport, at least at these sites. The exact contribu-
tions of advection and crack transport to overall site emissions
% oxidation

60%
are unknown and requires further research.
Our results indicate that one effective way to increase the % oxi-
40%
dation of a landfill cover is to limit the amount of CH4 delivered to
it. Obviously when CH4 production rates are high, in the early life
20%
of a landfill, limiting the amount of CH4 delivered to the cover is
accomplished by an efficient gas collection system. Early installa-
tion of these systems is warranted (Barlaz et al., 2010). The pres- 0%
0 200 400 600 800
ence of a gas collection system reduces the concentration and -2 -1
CH4 column input g m d
pressure of CH4 at the base of the cover and thereby reduces the
source strength of CH4 entering the cover system. However, the
y = 0.45x + 30.1, r = 0.61, n = 31, p = 0.0003
cover may be relied upon to consume a portion of the CH4 so that 700
the extraction strength of gas collections systems can be adjusted
downward to obtain landfill gas with an elevated CH4 composition 600
-2 -1
oxidation rate g m d

to better supply energy generating systems. Obviously % oxidation 500 y = 0.36x + 28.7
will be greater in landfills with gas collection than in those without r = 0.46
gas collection if other factors such as the age and thickness of the 400 n = 20
waste are equal. A cover with an efficient well-functioning gas col- p = 0.03
300
lection system will achieve a higher % oxidation than a cover with a
gas collection which is inefficient and poorly maintained. 200
Over time, as the waste within a landfill matures and the CH4
production rate decreases, it will at some point become possible 100
to turn off gas collection systems and rely upon the landfill cover 0
for the consumption of the remaining gas produced. The soils that 0 200 400 600 800
we examined consumed minimum estimates of 39–54% of the CH4 -2 -1
CH4 column input g m d
delivered to them at emission rates below 10 g CH4 m2 d1. At
50% oxidation, this means that the delivery rate of CH4 to the base Fig. 8. Laboratory column studies of CH4 oxidation percent (upper panel) and CH4
of the oxidation zone (e.g. the bottom flux) was 20 g CH4 m2 d1. oxidation rate (lower panel) at differing rates of CH4 loading to the bottom of the
column. These literature results were compiled by Scheutz et al. (2009) and are
These soils were low organic and non-vegetated. Amendment of presented graphically here. Open symbols represent organic rich materials while
landfill cover soils with organic materials can enhance the soil’s filled diamonds are column studies that utilized landfill mineral soil. The regression
capacity to hold water which may enhance oxidation (Stern of the organic soils had an r of 0.61, while the mineral soil regression r was 0.46.
 J. Chanton et al. / Waste Management 31 (2011) 914–925 923

reducing bypassing of the soil and thus further increasing % oxida- 180
tion. But organic matter amendments are not without problems
themselves, including degradation of the organic matter, settlement, 160 A
loss of permeability, CH4 formation and so on. Healthy vegetation

Methane Produciton (gm/m 2-day)

140
can maintain an organic treatment, however.
120
4.2. Hypothetical case study
100
An illustrative case study was developed to evaluate when fugi-
tive inputs to the cover are within the soil’s capacity to attenuate 80
k=0.04
CH4 emissions by biological oxidation. The CH4 production rate
60
was calculated for a landfill that received 106 Mg of waste per year
for 10 years by using EPA’s LandGem model (USEPA, 2005). Meth- 40 k=0.1
ane production and emissions are presented at the AP-42 default
decay rate of 0.04 yr1as well as at 0.1 yr1 to simulate accelerated 20
decomposition (USEPA, 1998a,b). The ultimate CH4 yield (L0),
63.9 m3 Mg1, was calculated from waste component specific CH4 0
yield data and statewide waste composition data as presented in 10 20 30 40 50 60 70 80 90 100
Staley and Barlaz (2009). To convert CH4 production and collection Year
rates (m3 yr1) to a flux (g2 yr1), the waste was assumed to have
10
an in place density of 833 kg m3 in a landfill that included 5% cov-
er soil and airspace utilization of 248,656 m3 ha1. These values B
were adopted from a survey of landfill practice (Camobreco et al.,

Uncollected Methane(gm/m 2-day)

1999) and represent a landfill with an average height of 25.8 m 8
including the top area and the slide slopes. The resulting CH4 pro-
duction and uncollected CH4 are presented in Fig. 9a and b, respec- k=0.04
tively. The uncollected CH4 (Fig. 9b) represents CH4 input to the 6
bottom of the soil cover (bottom flux) in response to an assumed
landfill gas collection efficiency. This efficiency was adopted from
Barlaz et al. (2010) and is linked to the age of the waste cell. The 4 k=0.1
collection efficiency regime developed by Barlaz et al. (2010) is
as follows: 0% in years 1 and 2, 50% in year 3, 70% in year 4, 75%
in years 5–10 and 95% thereafter. For this illustration, it was as- 2
sumed that a final cover was placed after 10 years of waste dis-
posal, at which time the gas collection efficiency was 95%. Data
in Fig. 9 are only presented after year 10 (the year of closure) at 0
which time the landfill footprint was constant. The CH4 flux is
10 20 30 40 50 60 70 80 90 100
independent of the mass of waste buried since the flux is calcu- Year
lated from a mass of waste buried in a given area.
The scenarios presented in Fig. 9 are illustrative and there are Fig. 9. Panel A, Methane Production in a hypothetical landfill that received 106 Mg
many factors that would shift the production curves up or down. of waste annually for ten years. Factors required to convert the production volume
to a mass per area, and factors to convert production to emissions after collection
A lower L0 would shift these curves down. The effect of decay rate are given in the text. Panel B assumes 95% gas collection efficiency to calculate the
is illustrated by the shift between 0.04 and 0.1 yr1 and higher CH4 flux to the bottom of the soil cover. Note that the x-axis starts at year 10 which
decay rates are possible at bioreactor landfills. is the year of closure.
As illustrated in Fig. 9b, with a gas collection efficiency of 95%
after landfill closure, the input of uncollected CH4 to the soil cover
is well within the range where the soil cover CH4 oxidizing bacteria varied from 21% to 43% effective, which would result in emissions
are under-saturated with respect to CH4. Even if gas collection was ranging from 23 to 31 g CH4 m2 d1. Thus, emissions of 8 to
only 85% and the input to the cover was three times the values 31 g CH4 m2 d1 represent upper and lower limits to CH4 emis-
shown in Fig. 9B, the CH4 flux would be well within the capacity sion rates once the gas collection system was turned off in this
of the soil for oxidation. With proper management and soil organic scenario.
enhancement including vegetation, oxidation should be nearly
quantitative and emissions should approach zero as was found 5. Conclusions
on the beanie landfill ‘‘X” (Table 1). This analysis assumes even dis-
tribution of CH4 over the landfill cover which is not completely The results from the field data and the modeling results indicate
accurate. Nonetheless, a more even distribution would be expected that percent oxidation should not be considered as a constant 10%
over time as the CH4 production rate decreases and a final cover is or any other single value. Percent oxidation is a changing quantity
in place. Examination of Fig. 9A shows that at the higher decay and is a function of cover type, climatic conditions and CH4 loading
rate, the landfill CH4 production will be at or below to the bottom of the cover.
40 g CH4 m2 d1 after 25–30 year which represents 15–20 year Increasing methane loading to a landfill soil cover reduces the
post closure, since closure occurred in year 10. If the gas collection efficiency of methane oxidation. Methane oxidation rate behaves
system was turned off at this time, CH4 would migrate at this rate in a fashion similar to enzyme kinetics, increasing in a linear man-
towards the bottom of the landfill cover. Based on column studies, ner with increasing substrate at relatively low substrate levels and
oxidation could be above 80% effective in this range, resulting in then flattening out with continued substrate addition. The methan-
emissions on the order of 8 g CH4 m2 d1. Based on the field stud- otrophic community is CH4 limited at low CH4 loading rates, and
ies, which were organic poor mineral soils, % oxidation at this rate then as loading increases, some other factor, either oxygen or
924 J. Chanton et al. / Waste Management 31 (2011) 914–925

microbial populating becomes limiting. The data indicate that the Czepiel, P.M., Mosher, B., Crill, P.M., Harriss, R.C., 1996a. Quantifying the effect of
oxidation on landfill methane emissions. J. Geophys. Res. 101 (D11), 16721–
best way to increase the % oxidation of a landfill cover is to limit
16729.
the amount of CH4 delivered to it. Czepiel, P.M., Mosher, B., Harriss, R.C., Shorter, J.H., McManus, J.B., Kolb, C.E., Alwine,
As previously reported, CH4 emissions from the landfill surface E., Lamb, B.K., 1996b. Landfill methane emissions measured by enclosure and
were dominated by ‘‘hotspots.” Persistence of these features was atmospheric tracer methods. J. Geophys. Res. 101, 16711–16719.
De Visscher, A., De Pourcq, I., Chanton, J., 2004. Isotope fractionation effects by
observed indicating that with increased monitoring such hot spots diffusion and methane oxidation in landfill cover soils. J. Geophys. Res. 109,
could be identified, repaired and the cover emissions attenuated by D18111. doi:10.129/2004JD004857.
with bio-covers and bio-cells (Abichou et al., 2006a,b). De Visscher, A., Van Cleemput, O., 2003. Simulation model for gas diffusion and
methane oxidation in landfill cover soils. Waste Manage. 23, 581–591.
Gebert, J., Groengroeft, A., 2006. Performance of a passively vented field-scale
biofilter for the microbial oxidation of landfill methane. Waste Manage. 26,
Acknowledgements
399–407.
Hanson, R.S., 1980. Ecology and diversity of methanotrophic organisms. Adv. Appl.
This work was supported by Waste Management, Inc. Microbiol. 26, 3–39.
Hanson, R.S., Hanson, T.E., 1996. Methanotrophic bacteria. Microbiol. Rev. 60, 439–
471.
References Hilger, H.A., Wollum, A.G., Barlaz, M.A., 2000a. Landfill methane oxidation response
to vegetation, fertilization, and liming. J. Environ. Qual. 29, 324–334.
Abichou, T., Johnson, T., Mahieu, K., Chanton, J., Romdhane, M., Mansouri, I., 2010. Hilger, H.A., Liehr, S.K., Barlaz, M.A., 2000b. Exopolysaccharide control of methane
In: Fratta, D., Muhunthan, B., Puppala, A. (Eds.), Developing a Design Approach oxidation in landfill cover soil. J. Environ. Eng., (ASCE) 125 (12), 1113–1123.
to Reduce Methane Emissions from California Landfills, Geotechnical Special Huber-Humer, M., Gebert, J., Hilger, H., 2008. Biotic systems to mitigate landfill
Publication No. 199. ASCE, Baltimore. methane emissions. Water Manage. Res. 26, 33–46. doi:10.1177/0734242X07
Abichou, T., Mahieu, K., Yuan, L., Chanton, J., Hater, G., 2009. Effects of compost 087977.
biocovers on gas flow and methane oxidation in a landfill cover. Waste Manage. Huber-Humer, M., 2004. International research into landfill gas emissions and
29, 1595–1601. mitigation strategies – IWWG working group ‘‘CLEAR”. Waste Manage. 24, 425–
Abichou, T., Yuan, L., Chanton, J., 2008. In: Khire, M., Alshawakbeh, A., Reddy, K. 427.
(Eds.), Estimating Methane Emission and Oxidation from Earthen Landfill Huber-Humer, M., Lechner, P., 2002. Proper bio-covers to enhance methane
Covers”, Geotechnics of Waste Management and Remediation, Geotechnical oxidation – findings from a two year field trial. In: Proc. of the Solid Waste
Special Publication (GSP) No. 177. ASCE Baltimore, pp. 80–87. Association of North America, 25th Annual Landfill Gas Symposium, Monterey
Abichou, T., Chanton, J., Powelson, D., Fleiger, J., Escoriaza, S., Lei, Y., Stern, J., 2006a. CA, March 25–28, 2002.
Methane flux and oxidation at two types of intermediate landfill covers. Waste Humer, M., Lechner, P., 2001. Microbial methane oxidation for the reduction of
Manage. 26, 1305–1312. landfill gas emissions. J. Solid Waste Technol. Manage. 27, 146–215.
Abichou, T., Powelson, D., Chanton, J., Escoriaza, S., Stern, J., 2006b. Characterization Humer, M., Lechner, P., 1999. Alternative approach to the elimination of greenhouse
of methane flux and oxidation at a solid waste landfill. J. Environ. Eng. 132 (2), gases from old landfills. Waste Manage. Res. 17, 443–452.
220–228. doi:10.1061/_ASCE_0733-9372_2006_132:2_220. IPCC, 2006. Guidelines for national greenhouse gas inventories. <http://www.ipcc-
Barlaz, M.A., Chanton, J.P., Green, R.B., 2010. Controls on landfill gas collection nggip.iges.or.jp/public/2006gl/index.html>.
efficiency: instantaneous and lifetime performance. J. Air Waste Manage. Assoc. Kiene, R.P., 1991. Production and consumption of methane in aquatic systems.
59, 1399–1404. ISSN: 1047-3289. doi:10.3155/1047-3289.59.12.139. Methane fluxes from terrestrial wetland environments. In: Rogers, J.E.,
Barlaz, M.A., Green, R., Chanton, J.P., Goldsmith, C.D., Hater, G.R., 2004. Evaluation of Whitman, W.B. (Eds.), Microbial production and consumption of greenhouse
a biologically active cover for mitigation of landfill gas emissions. Environ. Sci. gases. American Society for Microbiology, Washington, DC, pp. 91–109.
Technol. 38, 4891–4899. King, G., 1992. Ecological aspects of methane oxidation, a key determinant of global
Bender, M., Conrad, R., 1995. Effect of CH4 concentrations and soil conditions on the methane dynamics. Adv. Microbial Ecol. 12, 431–468.
induction of CH4 oxidation activity. Soil Biol. Biochem. 27, 1517–1527. Kjeldsen, P., Dalager, A., Broholm, K., 1997. Attenuation of methane and non-
Bergamaschi, P., Lubina, C., Konigstedt, R., Fischer, H., Veltkamp, A.C., Zwaagstra, O., methane organic compounds in landfill gas affected soils. J. Air Waste Manage.
1998. Stable isotopic signatures (d13C, dD) of methane from European landfill Assoc. 47, 1268–1275.
sites. J. Geophys. Res. 103 (D7), 8251–8266. Lelieveld, J., Crutzen, P.J., Dentener, F.J., 1998. Changing concentration, lifetime and
Boeckx, P., Van Cleemput, O., 1996. Methane oxidation in a neutral landfill cover climate forcing of atmospheric methane. Tellus 50B, 128–150.
soil: influence of moisture content, temperature, and nitrogen-turnover. J. Liptay, K., Chanton, J., Czepiel, P., Mosher, B., 1998. Use of stable isotopes to
Environ. Qual. 25, 178–183. determine methane oxidation in landfill cover soils. J. Geophys. Res. 103, 8243–
Boeckx, P., Van Cleemput, O., Villaralvo, I., 1996. Methane emission from a landfill 8250.
and the methane oxidizing capacity of its covering soil. Soil Biol. Biochem. 28, Maurice, C., 2001. Bioindication and Bioremediation of Landfill Emissions. Doctoral
1397–1405. Thesis, Lulea University of Technology 2001:29, Department of Environmental
Bogner, J., Matthews, E., 2003. Global methane emissions from landfills: new Engineering, Division of Waste Science and Technology.
methodology and annual estimates 1980–1996. Global Biogeochem. Cycles 17, Merritt, D.A., Hayes, J.M., Des Marais, D., 1995. Carbon isotopic analysis of
1065. doi:10.1029/2002GB001913. atmospheric methane by isotope ratio monitoring gas chromatography mass
Bogner, J., Meadows, M., Czepiel, P., 1997. Fluxes between landfills and the spectrometry. J. Geophys. Res. 100 (D1), 1317–1325.
atmosphere, natural and engineered controls. Soil Use Manage. 13, 268–277. Park, S., Brown, K., Thomas, J., 2002. The effect of various environmental and design
Bogner, J., Spokas, K., Chanton, J., Franco, G., Young, S., 2009. A new field-validated parameters on methane oxidation in a model biofilter. Waste Manage. Res. 20,
inventory methodology for landfill CH4 emissions. In: Proceeding SWANA 434–444.
Landfill Gas Symposium, Atlanta, GA, March 2009, published by SWANA, Silver Powelson, D., Chanton, J., Abichou, T., Morales, J., 2006. Methane oxidation in
Spring, MD. compost and water spreading biofilters. Waste Manage. Res. 29, 1–9.
Börjesson, G., Samuelsson, J., Chanton, J., 2007. Methane oxidation in Swedish Powelson, D.K., Chanton, J.P., Abichou, T., 2007. Methane oxidation in biofilters
landfills quantified with the carbon isotope technique in combination with an measured by mass-balance and stable isotope methods. Environ. Sci. Technol.
optical method for emitted methane. Environ. Sci. Technol. 41, 6684–6690. 41, 620–625.
doi:10.1021/es062735. Reeburgh, W.S., 1976. Methane consumption in Cariaco Trench waters and
Cabral, A.R., Capanema, M.A., Gebert, J., Moreira, J.F., Jugnia, L.B., 2010. Quantifying sediments. Earth Planet Sci. Lett. 28, 337–344.
microbial methane oxidation efficiencies in two experimental landfill biocovers Reeburgh, W.S., Heggie, D.T., 1977. Microbial methane consumption reactions and
using stable isotopes. Water, Air, Soil Pollut. 209, 157–172. their effect on methane distributions in fresh water and marine environments.
Camobreco, V., Repa, E., Ham, R.K., Barlaz, M.A., Felker, M., Rousseau, C., Clark-Balbo, Limnol. Oceanogr. 22, 1–9.
M., Rathle, J., Thorneloe, S., 1999. Life Cycle Inventory of a Modern Municipal Rudd, J.W.M., Hamilton, R.D., 1975. Factors controlling rates of methane oxidation
Solid Waste Landfill, Environmental Research and Education Foundation, and the distribution of methane oxidizers in a small stratified lake. Arch.
Washington, DC, June, 1999. Hydrobiol. 4, 522–538.
Chanton, J.P., Powelson, D.K., Green, R.B., 2009. Methane oxidation in landfill cover Rudd, J.W.M., Furutani, A., Glett, R.J., Hamilton, R.D., 1976. Factors controlling
soils, is a 10% default value reasonable? J. Environ. Qual. 38, 654–663. methane oxidation in shield lakes: the roles of nitrogen fixation and oxygen
Chanton, J.P., Powelson, D.K., Abichou, T., Hater, G., 2008a. Improved field methods concentration. Limnol. Oceanogr. 21, 337–348.
to quantify methane oxidation in landfill cover materials using stable carbon Rudd, J.W.M., Taylor, C.D., 1980. Methane cycling in aquatic environments. Adv.
isotopes. Environ. Sci. Technol. 42, 665–670. Aquat. Microbiol. 2, 77–150.
Chanton, J.P., Powelson, D.K., Abichou, T., Fields, D., Green, R.B., 2008b. Effect of Scheutz, C., Kjeldsen, P., Bogner, J.E., De Visscher, A., Gebert, J., Hilger, H.A., Huber-
Temperature And Oxidation Rate on Carbon-Isotope Fractionation During Humer, M., Spokas, K., 2009. Microbial methane oxidation processes and
Methane Oxidation by Landfill Cover Materials Environmental Science and technologies for mitigation of landfill gas emissions. Waste Manage. Res. 27,
Technology No. 42, pp. 7818–7823. doi:10.1021/es80122y. 409–455. doi:10.1177/0734242X09339325.
Chanton, J., Liptay, K., 2000. Seasonal variation in methane oxidation in landfill Scheutz, C., Bogner, J., Chanton, J.P., Blake, D., Morcet, M., Aran, C., Kjeldsen, P., 2008.
cover soils as determined by an in situ stable isotope technique. Global Atmospheric emissions and attenuations of non-methane organic compounds
Biogeochem. Cycles 14, 51–60. in cover soils at a French landfill. Waste Manage. 28, 1892–1908.
 J. Chanton et al. / Waste Management 31 (2011) 914–925 925

Scheutz, C., Kjeldsen, P., 2005. Biodegradation of trace gasses in simulated landfill Stern, J.C., Chanton, J., Abichou, T., Powelson, D., Yuan, L., Escoriza, S., Bogner, J.,
soil cover. J. Air Waste Manage. Assoc. 55, 878–885. 2007. Use of a biologically active cover to reduce landfill methane emissions
Scheutz, C., Kjeldsen, P., 2004. Environmental factors influencing attenuation of and enhance methane oxidation. Waste Manage. 27, 1248–1258.
methane and hydrochlorofluorocarbons in landfill cover soils. J. Environ. Qual. USEPA, 2007. Inventory Of U.S. Greenhouse Gas Emissions And Sinks: 1990-2005 USEPA
33, 72–79. #430-R-07-002. http://epa.gov/climatechange/emissions/usinventoryreport.html.
Simunek, J., Genuchten, M.T.V., Sejna, M., 2005. The HYDRUS-1D Software Package USEPA, 2005. Landfill Gas Emissions Model (LandGEM) Version 3.02 User’s guide;
for Simulating the Movement of Water, Heat, and Multiple Solutes in Variably USEPA: Washington, DC.
Saturated Media, Version 3.0, HYDRUS Software Series 1. Department of USEPA, 2004. Direct emissions from landfilling municipal solid waste, US
Environmental Sciences, University of California Riverside. Riverside, California, Environmental Protection Agency, Washington, DC. http://www.epa.gov/
USA. stateply/documents/resources/protocol-solid_waste_landfill.pdf.
Spokas, K., Bogner, J., 2011. Limits and dynamics of methane oxidation in landfill USEPA, 1998. Greenhouse gas emissions from management of selected materials in
cover soils. Waste Manage. 31 (5), 823–832. municipal solid waste, EPA 530-R-98-01, 1998.
Spokas, K.A., Forcella, F., 2009. Software tools for weed seed germination modeling. USEPA, 1998. AP-42 Emission Factors for Municipal Solid Waste Landfills –
Weed Sci. 57, 216–227. Supplement E; November 1998; US EPA: Washington, DC http://www.epa.
Spokas, K., Graff, C., Morcet, M., Aran, C., 2003. Implications of the spatial variability gov/ttn/chief/ap42/ch02/final/c02s04.pdf.
on landfill emission rates on geospatial analyses. Waste Manage. 23, 599– Wang, Y., Wu, W., Ding, Y., Liu, W., Perera, A., Chen, Y., Devare, M., 2008. Methane
607. oxidation activity and bacterial community composition in a simulated landfill
Spokas, K., Bogner, J., Chanton, J.P., Morcet, M., Aran, C., Graff, C., Moreau-Le Golvan, Y., cover soil is influenced by the growth of Chenopodium album L. Soil Biol.
Hebe, I., 2006. Methane mass balance at three landfill sites: what is the efficiency Biochem. 40, 2452–2459.
of capture by gas collection systems? Waste Manage. 26 (5), 516–525. Whalen, S.C., Reeburgh, W.S., Sandbeck, K.A., 1990. Rapid methane oxidation in a
Surfer, 2002. Surfer 8 User’s Guide. Golden Software, Inc., Golden, CO. landfill cover soil. Appl. Environ. Microbiol. 56, 3405–3411.
Staley, B.F., Barlaz, M.A., 2009. Composition of municipal solid waste in the U.S. and Yuan, L., Abichou, T., Chanton, J., DeVisscher, A., 2009. Long term numerical
implications for carbon sequestration and methane yield. Journal of simulation of methane transport and oxidation in a compost biofilter. J. Hazard.
Environmental Engineering, accepted for publication. Waste Manage. 13 (3), 1–7.