Effect of biocover equipped with a novel passive air diffusion system on microbial methane oxidation and community of methanotrophs

Abstract

A novel biocover with passive air diffusion system (PADS) was designed in this study. Its effect on landfill gas components in the macrocosms of simulated biocover systems was also investigated. The results show that O₂ concentration increased in the whole profile of the macrocosms equipped with PADS. When simulated landfill gas (SLFG) flow rate was no more than 40 mL min⁻¹, the methane oxidation rate was 100%. The highest CH₄ oxidation capacity reached to 31.34 mol m⁻³ day⁻¹. Molecular microbiology analysis of the soil samples taken from the above macrocosm showed that the growth of type I methanotrophs was enhanced, attributable to enhanced air diffusion and distribution, whereas the microbial diversity and population density of type II methanotrophs were not so affected, as evidenced by the absence of any difference between the biocover equipped with PADS and that of the control. According to a phylogenic analysis, Methylobacter, Methylosarcina for type I, and Methylocystis, Methylosinus for type II, were the most prevalent species in the macrocosm with PADS.

Implications

Semiaerobic landfill technology was reported to be an effective in situ CH₄ emission reduction method. A semiaerobic status is achieved through a continuous supply of fresh air sucked passively through the open end of a leachate collection pipe, with a temperature gradient created by organic matter degradation. In this research, a novel passive air diffusion system (PADS) was designed, and O₂ concentration increased in the whole profile of the microcosms equipped with PADS. The constant supply of fresh air reduced CH₄ production and enhanced biodegradation. Through molecular microbiology analysis, it was found that growth of methanotrophs was enhanced, and species of prevalent methanotrophs were found.

Supplemental Materials

Supplemental materials are available for this article. Go to the publisher's online edition of the Journal of the Air & Waste Management Association for a list of abbreviations and PCR protocol.

Introduction

Methane (CH₄) is an important greenhouse gas with a global warming potential that is 25 times greater than that of carbon dioxide (CO₂), according to Intergovernmental Panel on Climate Change 2007 (IPCC). Its contribution to escalating global warming is estimated at 15% (Themelis and Ulloa, 2007). Landfill sites are thought to contribute 20 to 70 Tg of CH₄ annually to global CH₄ emissions and are the fourth largest anthropogenic source of CH₄ worldwide (Boeckx et al., 1996). According to the Annual Report on the State of the Environment in China (S.E.P. Administration, 2009), CH₄ production from landfill account for 11% of the total amount in anthropogenic CH₄ production in 2005, and it is predicted that the ratio will exceed 14% in 2015. CH₄ production and emission from landfill in Beijing accounts for 39.48% (in 2005) of the total CH₄ emission in the city, which is significantly higher than other sources. That is because landfill is the primary way (>88%) for municipal solid waste (MSW) treatment in Beijing.

Together with CH₄ emissions from the fossil fuel sector, landfill CH₄ emission requires the most urgent mitigation to stabilize CH₄ concentration in the atmosphere (Stern and Kaufman, 1996). The oxidation of CH₄ by bacteria residing in soil constitutes an important terrestrial CH₄ sink. These bacteria are particularly abundant in the topsoil of landfill caps due to the high CH₄ concentration supply from the landfill below. Whalen et al. (1990) first reported the high CH₄-oxidizing capabilities of landfill cover soil, which is capable of oxidizing CH₄ at a rate of 2.8 mol m⁻² day⁻¹. Since then, various studies have focused on the biotransformation of CH₄ in the landfill cover soils. Documented laboratory and field studies showed that landfill cover soil exposed to CH₄ can develop high capacities for CH₄ oxidation by indigenous methanotrophic microorganisms (Bogner et al., 1997; Chanton et al., 1999; Christophersen et al., 2000; Czepiel et al., 1996; Kightley et al., 1995; Liptay et al., 1998). In recent years, landfill biofilter designed to optimize CH₄ oxidation have demonstrated the potential to mitigate landfill CH₄ emissions as high as 1900 g m⁻³ day⁻¹ (Scheutz et al., 2009).

Traditionally, methanotrophs are classified into two general groups (type I and II) based on several characteristics, such as cell morphology, membrane arrangement, carbon assimilation pathway, and predominant phospholipid fatty acids (Hanson and Hanson, 1996). Type I methanotrophs, which belong to γ-proteobacteria, are composed of Methlyomonas, Methylococcus, Methylomicrobium, Methylosarcina, Methylosphaera, Methylothermus, Methylosoma, Methylohalobius, Methylocaldum, and Methylobacter (Hanson and Hanson, 1996; Rahalkar et al., 2007). A recent study (Lee, 2008) divided type I methanotrophs into two different groups, i.e., type Ia and Ib; the latter is composed of Methylococcus, Methylocaldum, and Methylothermus, whereas the rest of previous type I belong to type Ia methanotrophs. Genera that are members of type II methanotrophs belong to α-proteobacteria, including Methylosinus, Methylocella, Methylocapsa, and Methylocystis.

Type I methanotrophs whose thriving is much influenced by O₂ grow faster than type II and thus type I methanotrophs are more sensitive to the environment, such as O₂ concentration and nitrogen; by contrast, higher CH₄ concentrations favor the growth of type II methanotrophs. Crossman et al. (2004) confirmed that type I methanotroph is responsible for CH₄ oxidation for the top section of the landfill cap, whereas type II methanotroph population is responsible for CH₄ oxidation in the deepest section of the landfill.

The application of landfill cover layers for the CH₄ oxidation is mostly influenced by the available O₂ concentration (Nikiema et al., 2007). In order to reduce problems related to O₂ diffusion in landfill covers and open biofilters, many authors prefer the use of multilayer beds (Berger et al., 2005; Scheutz et al., 2009), compost (Streese and Stegmann, 2003), soil (Parker et al., 2002), a mixture of compost and soil (Haubrichs and Widmann, 2006), etc. Indeed, these porous materials can improve O₂ transfer and distribution. However, they cannot overcome the problem of insufficient O₂ supply because diffusion of atmospheric O₂ is limited and, generally, an oxygenated zone of only 0.60–0.80 m is observed (Crossman et al., 2004; Kallistova et al., 2005; Perera et al., 2002; Tagaris et al., 2003). In other words, improvement of CH₄ oxidation in a landfill cover soil depends on continuous O₂ supply rather than limited atmospheric diffusion.

Semiaerobic landfill technology was first developed more than 20 years ago in Japan (Matsufuji and Sinha, 1990) and has since been reported to be an effective in situ CH₄ emission reduction method. A semiaerobic status is achieved through a continuous supply of fresh air sucked passively through the open end of a leachate collecting pipe (diameter ≥0.6 m), with a temperature gradient created by biodegradation of organic matter. The constant supply of fresh air not only enhances biodegradation and reduces CH₄ production but also simultaneously improves leachate quality. It is a proven technology that has been tested in many places in Japan and in a few developing countries, such as Malaysia, Iran, and China (Chong et al., 2005). Obviously, a semiaerobic landfill provides more favorable conditions for the placement of an in situ O₂ supply into the cover layer than a conventional landfill, and the CH₄ oxidation can be expected. Unfortunately, no research has been carried out on the effect of O₂ supply to the landfill top cover, so the feasibility of CH₄ oxidation improvement in the landfill cover layer by this method remains unknown.

In this paper, a novel passive air diffusion system (PADS) for biocover layer was developed. The CH₄ oxidation efficiencies under different simulated landfill gas flow rate were reported. Methanotrophic population structure and activity were also investigated to reveal the effect of O₂ on the microbial population. It aims to provide a feasible air supply system without energy consumption in the landfill biocover layer to promote the growth of methanotrophic bacteria, and finally to reduce the CH₄ emission from landfill site.

Materials and Methods

Microcosms of simulated biocover

The experimental setups of the simulated biocover microcosms were each made of polyvinyl chloride (PVC) pipes with a size of 0.45 m × 0.38 m (height and inner diameter). The microcosm that was equipped with PADS to enhance the air supply in the column was labeled “macrocosm of simulated novel biocover system” (MSNBS), whereas the macrocosm that simulates conventional biocover system (MSCBS) without PADS was set as the control. A diagram of the macrocosms is shown in Figure 1. As can be seen, the PADS made from Unplasticized-PVC consists of three parts: the bottom drainage pipe (inner diameter = 0.057 m), the vertical venting pipe (inner diameter = 0.046 m, length = 1.000 m), and the air-distributing head wheel (inner diameter = 0.30 m). Whereas the air-distributing head wheel and part of the vertical venting pipe (0.15 m) was placed inside the macrocosm, the rest of the vertical venting pipe and the bottom drainage pipe were kept outside of the macrocosm. The vertical venting pipe outside the macrocosm was heated, and the air temperature in the pipe was kept at 50 ± 1 °C by a heating controller. This was carried out to simulate the microbial self-heating inside the landfill site during organic waste biodegradation. In this setup, the air was sucked in passively through the open ends of the bottom drainage pipe because of the temperature difference between the air inside the vertical pipe and the fresh air outside of the pipe. Subsequently, the drawn-in air was transported upwards through the vertical venting pipe and filled to the macrocosm through the head wheel.

Figure 1. Simulated landfill biocover microcosms (left, MSCBS; right, MSNBS).

Yard waste compost and topsoil from Beishenshu municipal solid waste (MSW) Landfill (Beiijing, China) were mixed homogeneously at a ratio of 7:3 (w/w). This mixture was used as the material for the simulated biocover macrocosms. The height of the filling material was 0.35 m. Both macrocosms have gas inlet ports at the bottom. Simulated landfill gas (SLFG; CH₄ and CO₂, 1:1 v/v) was pumped in through the gas inlets at flow rates of 0 (no gas pumped), 10, 20, 40, 60, and 80 mL min⁻¹, respectively, to simulate landfill gas interference in the landfill body right below the biocover layer. At the bottom of the column, a 0.05-m layer of gravel was placed to ensure a homogenous distribution of the gas in the macrocosms. Each macrocosm had seven gas sampling ports with a 0.05-m interval down through the column and an outlet port on the tops.

The landfill gas in the macrocosms was sampled once a day. Each microcosm had eight gas sampling ports with a 0.05-m interval down through the column. The gas samples were collected from these sampling ports with syringes; CH₄, O₂, and CO₂ were analyzed using gas chromatograph (Agilent 6890N; Agilent Technologies, Palo Alto, CA, USA) with a thermal conductivity detector. The detection program consisted of a carrier gas (H₂) flow quantity of 30 mL min⁻¹, oven temperature of 120 °C, injector temperature of 160 °C, and detector temperature of 160 °C.

Sampling

Two types of samples were collected from each macrocosm at the end of the experiment for molecular biology analysis of methanotrophs. Samples A1 and B1 represent the bottom layers (10 cm up the gravel of the SLFG-distributing layer) of MSCBS and MSBNS, respectively, and samples A2 and B2 represent the top layers of the MSCBS and MSBNS (10 cm down the top of the biocover material), respectively. Each sample had three replicates.

DNA extraction and PCR

First, DNA was extracted and purified directly from the fresh samples using the FastDNA SPIN Kit for Soil (MP Biotechnology, Solon, OH, USA) according to the manufacturer's instructions. To amplify type I methanotrophs present in the environmental DNA, the primers MethT1bR and MethT1dF (Table 1) were used, and MethT2R and the bacterium-specific primer 27F were used to amplify type II methanotrophs. Polymerase chain reaction (PCR) was conducted in a thermocycler (TC-3000; High System, Barloworld Scientific, Staffordshire, United Kingdom). The presence and size of amplification products were determined by agarose gel electrophoresis and ethidium bromide staining. For PCR protocols see Supplementary Material.

Table 1. Methanotroph-specific primers and probes

Primer/Probe	Sequence	Target Genera
MethT1bR	5'-GATTCYMTGSATGTCAAGG-3'	Type I methanotrophs
MethT1dF	5'-CCTTCGGGMGCYGACGAGT-3'
MethT2R	5'-CATCTCTGRCSAYCATACCGG-3'	Type II methanotrophs
27F	5'-AGAGTTTGATCMTGGCTCAG-3'
358F	5'-CCTACGGGAGGCAGCAG-3'	All Bacteria
517R	5'-ATTACCGCGGCTGCTGG-3'
Notes: Y represents C or T; R represents A or G; M represents A or C; S represents C or G.

DGGE

A nested-PCR approach was used to profile the type I and type II methanotrophic communities by denaturing gradient gel electrophoresis (DGGE). The PCR-amplification in the above-mentioned steps were used as templates for PCR with the primers GC358F (with a 40-bp GC clamp at the 5' end of 358F) (Yan et al., 2011) and 517R (Table 1), which span variable region 3 (V3 region) of the 16S rRNA gene. Hot-start and touchdown PCRs were performed to reduce primer-dimer complexes.

The PCR products were analyzed on agarose gels to confirm the presence of a single amplification of expected size. Denaturing gradient electrophoresis gels were 6.5% polyacrylamide with a denaturant gradient from 20% to 70% and 30% to 60% for type I sequences and type II sequences, respectively. Gels were run in 1× Tris‐acetate‐EDTA (TAE) buffer for 12 hr at 100 V at 60 °C and visualized by SYBR-Gold staining.

Specific PCR-DGGE bands were manually excised from the gel, suspended in 50 μL of sterile water, and incubated overnight at 4 °C. The PCR-DGGE protocol was repeated using the above solution as template, until only a single band was detected. A final PCR step was performed without the GC-clamp attached to the forward primer (358F). The resulting PCR products were sent to SinoGenoMax Co. Ltd. (Beijing, China) for purification and nucleotide sequencing.

Sequencing and phylogenetic analysis

Basic Local Alignment Search Tool (BLAST) program was used to search the National Center for Biotechnology Information (NCBI) sequence database (http://www.ncbi.nlm.nih.gov/BLAST/) for sequence similarity. All nucleotide sequences were aligned using the CLUSTAL X program (Aymerich et al., 2003). Phylogenetic trees were constructed with the neighbor-joining method using MEGA 5.0 software (http://www.megasoftware.net). Reference sequences were obtained from the GenBank database and were included in the phylogenetic trees for comparison. All nucleotide sequences were optimally aligned prior to tree construction.

Band patterns from PCR-DGGE fingerprints were analyzed using the unweighted pair group method (UPGMA), in which the arithmetic mean used DNAMAN v. 4.1 (Lynnon Biosoft, Vaudreuil, Quebec, Canada). Binary sequences were generated for individual fingerprints by determining band number (R) and position compared with the total number of band positions using Labworks 4.6 software (UVP Products, Upland, CA, USA). The scanned gel images were analyzed using Labworks 4.6.

To evaluate the microbial community in the different height of the macrocosms, richness (R), diversity (H), and dissimilarity (D) indices were calculated from the DGGE band profiles. The Shannon–Weaver index of general diversity (H) was calculated according to the following equation (Shannon and Weaver, 1963):

(1)

where P_i represents the relative probability of the bands in a gel lane. H is calculated on the basis of the bands on the gel lane that are applied for the generation of the dendrograms by using the intensities of the bands as judged by peak heights in the densitometric curves (Ampe and Miambi, 2000; Nübel et al., 1999). P_i is calculated as

(2)

where n_i refers to the height of a peak and N is the sum of all peak heights in the densitometric curve.

The dissimilarity index (D) is adopted to show the difference among the communities (Hiraishi et al., 1991; Luxmy et al., 2000). This is given by

(3)

where == 1.00, and x_ik and x_jk represent the ratio of relative band intensity for samples i and j in the location of band number k. Dissimilarity indices were expressed numerically within a range of 0 (perfectly similar) to 1.00 (completely dissimilar).

Accession numbers for nucleotide sequences

The partial sequences of the 16S rRNA gene of methanotrophs obtained in this study are available from the NCBI database under accession numbers JN166453–JN166465 and JN166487–JN1666496.

Results

CH₄ oxidation capacity of biocover equipped with PADS

CH₄ oxidation capacity and O₂ concentration in MSNBS and MSCBS were compared. Results show that the introduction of PADS in MSNBS significantly increased O₂ concentration and improved its distribution in the biocover material either with or without SLFG interference (Figure 2). In the first stage without SLFG interference, O₂ concentration was higher than 18% in different MSNBS layers, whereas those in MSCBS varied within a range of 10–20%. When SLFG interfered in the macrocosms and the flow rate increased from 10 to 80 mL min⁻¹, the O₂ content maintained a steady level in MSNBS, where the average O₂ concentration detected from each layer was higher than 15%. Those in MSCBS decreased significantly as SLFG flow rate increased, and the average O₂ concentration was lower than 10%. More importantly, compared with MSCBS, the O₂ concentrations in MSNBS showed less fluctuation during the experimental duration. In other words, the introduction of PADS can significantly change the O₂ distribution profile in the biocover matrix.

Figure 2. O₂ content in different microcosm layers at different SLFG flow rates. (a) MSCBS; (b) MSNBS.

Correspondingly, the CH₄ oxidation capacity was significantly higher in MSNBS than that in MSCBS (Figure 3). In MSNBS, when SLFG flow rate was no more than 40 mL min⁻¹, CH₄ could not be detected at the outlet port, indicating a CH₄ oxidation rate of 100%. The maximum CH₄ oxidation capacity obtained at a flow rate of 40 mL min⁻¹ was 31.34 mol m⁻³ day⁻¹, which is the highest value generated. The CH₄ concentration at the outlet increased slightly (<6%) as SLFG flow rate rose. Those at other layers were no more than 10% even at high SLFG flow rates. The MSCBS CH₄ concentration changed drastically with increasing SLFG flow rate, with the highest concentration of 55% observed at the outlet when interfered by SLFG at a flow rate of 80 mL min⁻¹. Those at the other layers were significantly higher than that of MSNBS as well. Similar trends were observed in the CO₂ concentration distribution of the two macrocosms. Compared with MSCBS, the CO₂ concentration in biocover layers of MSNBS was kept at a lower level with PADS performance. For instance, the highest CO₂ concentration was less than 15% when SLFG flow rate reached its highest, whereas that in MSCBS was around 45% (Figure 4).

Figure 3. CH₄ content in different microcosm layers at different SLFG flow rates. (a) MSCBS; (b) MSNBS.

Figure 4. Fingerprints of type I (a) and type II (b) methanotrophic bacterial communities in the microcosms as generated by nested PCR-DGGE (denaturing gradient: 30–60%). Bands designated with an asterisk were excised and subjected to sequencing and identification.

The above results prove that PADS promoted CH₄ oxidation by providing sufficient O₂ to the biocover system. This system can overcome the problem of limited O₂ transportation in biocover layers, thus allowing methanotrophic populations to thrive in the macrocosm.

Numerical analysis of microbial community diversity of methanotrophs

The changes in the methanotrophs community structure in the two macrocosms are investigated and the result is shown in Figure 5. Since each band in the DGGE profiles is likely to be derived from one phylogenetically distinct population, R is equal to the total number of the species in the profile. Thirteen (i.e., R = 13) and 12 (i.e., R = 12) species were obtained for type I and type II methanotrophs, respectively, indicating that there were 13 dominate type I methanotrophs and 12 dominate type II methanotrophs in the samples. For type I methanotrophs, around 10 dominate species (R = 10) were generated with samples from MSNBS, whereas those generated from MSCBS were around 9 (R = 9). For type II methanotrophs, around 11 dominate species (R = 11) were generated with samples from MSNBS, whereas those generated from MSCBS were around 10 (R = 10).

Figure 5. Neighbor-joining tree depicting phylogenetic relationships of type I methanotrophic populations detected in microcosms by nested PCR-DGGE.

Moreover, the H and D indices, reflecting the structural diversity of the methanotrophic community and the banding-pattern dissimilarity, respectively, were calculated on the basis of the number and relative intensities of bands on the gel track (Table 2). Results show that there was a significant difference in microbial diversity between MSNBS and MSCBS for type I methanotrophic bacterial communities. In the MSCBS samples, a significant difference in H value between two layers of the sample (2.04 ≤ H ≤ 2.73) was found, reflecting O₂ concentration effect on type I methanotroph diversity. Although there was no such phenomena in MSNBS, widespread occurrence of high populations of type I methanotrophs, as evidenced by a similar H value (2.78 ≤ H ≤ 2.80), indicated that there was sufficient O₂ in the entire profile of the system. However, no significant difference of microbial diversity of type II methanotrophic bacterial communities were found among them, as the H value changed in a narrow range (2.84 ≤ H ≤ 2.89). This reflects that O₂ concentration in the experiment conditions has little effect on type II methanotroph structure and population density.

Table 2. Comparison of the microbial diversity and dissimilarity of type I and type II methanotrophics between reactors A and B

Type	Lane No.	Richness (R)	Microbial Diversity(Shannon-Weaver Index, H)	Dissimilarity (D)
Type I	B2	10	2.80	D(B1, B2) = 0.26
	B1	7	2.78	D(A1, B1) = 0.34
	A2	9	2.73	D(A2, B2) = 0.32
	A1	3	2.04	D(A1, A2) = 0.38
Type II	B2	11	2.89	D(B1, B2) = 0.31
	B1	11	2.89	D(A1, B1) = 0.33
	A2	10	2.86	D(A2, B2) = 0.33
	A1	10	2.84	D(A1, A2) = 0.35
Notes: The value R indicates the number of bands in the DGGE data analyzed with Labworks 4.6. software.

The band patterns of type I methanotrophs show that there was a lower dissimilarity (D(B1, B2) = 0.26) between the upper and lower MSNBS layers, indicating even O₂ distribution in the system as shown in the text above. The dissimilarity was higher (D(A1, A2) = 0.38) in MSCBS, however, reflecting the difference in O₂ concentration between the two sampling layers. Low O₂ concentration was clearly the limiting factor for type I methanotrophic growth. The band patterns of type II methanotrophs showed lower dissimilarity (0.31 ≤ D ≤ 0.35) in the two macrocosms, confirming that O₂ concentration in the study has little influence on type II methanotrophs.

Phylogenetic relationships analysis

Thirteen bands were excised from denaturing gels containing 16S rRNA gene fragments of type I methanotrophic bacteria; these were then sequenced. The detected populations were related to several methanotrophic genera, including Methylobacter, Methylococcus, and Methylosarcina. Among them, Methylobacter were the most prevalent species. Species that were not closely related to any known methanotrophic bacteria were also found in our study, such as band I-5 (JN166457), band I-12 (JN166464), and band I-13 (JN166465) (Figure 6). The community structure of type I methanotrophs in MSNBS was also found to be different from that in MSCBS. For example, Methylosarcina sp. (band 4, JN166456) and Methylobacter sp. (band 9, JN166461) substituted uncultured γ-proteobacterium (band 5, JN166457) and Methylococcus sp. (band 8, JN166460). Additionally, uncultured Methylococcaceae bacterium (band 7, JN166459) and β-proteobacterium bacterium (band 13, JN166465) were found almost exclusively in MSNBS.

Figure 6. Neighbor-joining tree depicting phylogenetic relationships of type II methanotrophic populations detected in microcosms by nested PCR-DGGE.

Ten different bands were excised from the denaturing gels containing 16S rRNA gene fragments of type II methanotrophic bacteria; these were then sequenced. The detected populations were related to Methylosinus sp., Methylocystis sp., and Methylocella sp. Among them, Methylocystis sp. was the most predominant species. Populations that were not closely related to any known methanotrophic population were also detected such as band 1(JN166487) and band 10 (JN166494).

Discussion

Since CH₄ will continuously produce and release in landfills even at stabilized stage, development of in situ CH₄ reduction technology is urgent to alleviate greenhouse effect origin from landfill site. Moreover, in situ oxidation is an optimum choice regardless of whether the landfill site is equipped with gas collection system or not. The above results show that passive air diffusion system has significantly effect on the greenhouse gas emission reduction. And it is technically feasible especially for semiaerobic landfill: the bottom drainage pipes that simultaneously used for air inlet and leachate drainage can be the air supply source, with the function of the air-distributing wheel that connected via the vertical venting pipe, fresh air can be transported and distributed up to the biocover layer thus to increase the O₂ concentration nearby. The air-distributing wheel can be substituted by the gravel path during engineering application. An extra advantage of this novel biocover system is to reduce other nocuous gases in the LFG such as gaseous ammonia and sulfur alcohol. Thus alleviate the odors in the landfill site. Demonstration engineering with a size of 1 × 10⁴ m³ has now been carried out by the authors to test the efficiency of PADS at an up-scaled level.

The use of PADS support higher O₂ diffusion than the level of atmospheric diffusion reported in other studies (Crossman et al., 2004; Kightley et al., 1995). Undoubtedly, PADS became a second O₂ source for the biocover in addition to the natural atmospheric diffusion. In this study, methanotrophic populations thrived in MBNBS, since PADS provided sufficient O₂ to the whole profiles of it. It is also well documented by another study that improved O₂ condition stimulates the growth of aerobic microbes including methanotrophs (Liptay et al., 1998).

The analysis of methanotroph-biased nested PCR-DGGE indicated that sufficient O₂ promoted the growth and influenced structure of type I methanotrophs but had little effect on population density as well as community structure of type II methanotrophs, reflecting that O₂ is a limiting factor for type I methanotrophs, which grow faster than type II. It also means that type I methanotrophs are more sensitive to the environment. Both type I and type II methanotrophs produce a membrane-bound (particulate) CH₄ monooxygenase enzyme (pMMO), which is thought to proliferate under low CH₄-to-O₂ ratios (Auman and Lidstrom, 2002; Costello et al., 2002; Hanson and Hanson, 1996). On the other hand, most type II methanotrophs produce not only the pMMO enzyme but also a more reactive cytoplasmic (soluble) CH₄ monooxygenase enzyme (sMMO), which explains the fast acclimation of type II methanotrophs to the environment, and their existence in a wide range of CH₄-to-O₂ ratios (Hanson and Hanson, 1996). Zhang et al. (2008) also found that there is a higher abundance of type II than that of type I methanotrophs in fertilization paddy soils.

In this study, the most abundant population among the isolates are Methylobacter (type I) and Methylocystis (type II), which confirmed the results of other studies that demonstrated these two species were extremely abundant in the top layer of landfill covers, as well as the topsoil where O₂ was sufficient (Bogner et al., 1997; Mcdonald and Murrell, 1997; Stralis-Pavese et al., 2006; Vallaeys et al., 1997; Wise et al., 1999).

Studies showed that Methylomonas was an active species in the paddy soils (Zheng et al., 2008) besides Methylobacter, that's because this genus is adapted to the condition of high CH₄ concentration and low O₂, whereas Methylocystis is more frequently detected in a condition with high O₂ and low CH₄, such as in upland agricultural soil and forestry soil (Mei et al., 2011; Mohanty et al., 2006). In our system, however, the structure of the dominate species of methanotrophs were different; for instance, species such as Methylomonas and Methylocysitis were not detected, whereas some unknown methanotrophic population were present both for type I and II methanotrophs, indicating the unique condition in MSNBS that provides high concentration of both CH₄ and O₂.

Landfill is the primary way (>80%) for MSW treatment in China. There are totally more than 400 landfills in China, but only 13 are equipped with LFG extraction and utilization facilities (mostly gas to electricity generation) (Y. F. Nie, personal observation). Landfill gas pollution is thus the key issue of environmental impacts due to direct or fugitive emission. The passive in situ O₂ supply that has been proved effective both by CH₄ degradation kinetics and microbiological study provide an all-purpose CH₄ emission reduction method for the landfills regardless of whether it is equipped with a gas collection system or not. We believe that could be a technically as well as economically feasible approach for the remediation of LFG pollution during MSW landfilling.

Conclusions

1.	The O2 concentration increased in the whole profile of the macrocosms equipped with PADS. The highest CH4 oxidation capacity at 31.34 mol m−3 day−1 was obtain by this system.
2.	Molecular microbiology analysis showed that the growth of type I methanotrophs was enhanced in the macrocosm because of enhanced air diffusion and distribution; Methylobacter and Methylosarcina were the most prevalent species of type I methanotrophs.
3.	The microbial diversity and population density of type II methanotrophs were not affected by O2, however, as evidenced by the absence of any difference between the biocover equipped with PADS and that of the control; Methylocystis and Methylosinus were the most prevalent species of type II methanotrophs.

Acknowledgments

This research project was funded by the National High Technology Research and Development Program of China (grant no. 2007AA06Z350) and the National Natural Science Foundation of China (grant no. 41101225).