- Email: [email protected]
Impact of preferential methane flow through soil on microbial community composition
European Journal of Soil Biology 69 (2015) 8e16
Contents lists available at ScienceDirect
European Journal of Soil Biology journal homepage: http://www.elsevier.com/locate/ejsobi
Original article
Impact of preferential methane flow through soil on microbial community composition Julia Gebert a, *, Mirjam Perner b a b
University of Hamburg, Institute of Soil Science, Allende-Platz 2, 20146 Hamburg, Germany University of Hamburg, Biocenter Klein Flottbek, Molecular Biology of Microbial Consortia, Ohnhorststraße 18, 22609 Hamburg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 September 2014 Received in revised form 21 March 2015 Accepted 23 March 2015 Available online 28 March 2015 Handling editor: C.C. Tebbe
The anaerobic microbial degradation of waste organic fractions in landfills constitutes one of the principal anthropogenic methane sources. Microbial oxidation of methane in optimized landfill covers or biofilters has been listed as key mitigation technology for the reduction of methane fluxes from landfills that are no longer suitable for energy recovery or flaring. Therefore, it is vital to understand what influences distribution of methane oxidizers and their activity in landfill soils. Here we describe the impact of gas fluxes through preferential pathways (hotspots) in the cover soil of a municipal solid waste landfill in north-western Germany on the soil properties and the microbial communities that colonize the upper soil crust in these environments. Two sites with high surface methane concentrations (>14,000 ppm), two sites with moderate surface methane concentrations (~400 ppm) and two sites without measurable methane emissions at the surface were investigated. It was found that elevated average soil methane concentrations coincided with increased levels of TOC and TN and the TOC/TN ratio in the topsoil. The increase of the latter posits a change in the composition of the organic matter towards increasing levels of nitrogen-poor components as for example EPS, which were observed in the samples with higher TOC/ TN ratios. Elevated average soil methane concentrations were also accompanied by a decrease in the overall bacterial diversity. The community at these sites were dominated by a few lineages such as methanotrophs, particularly of type II, Burkholderiales, Rhodospirillales and Bradyrhizobiaceae. This dominance may have contributed to the purple discoloration at the soil surface at the sites with the highest surface methane concentrations. © 2015 Elsevier Masson SAS. All rights reserved.
Keywords: Landfill cover soil Microbial community Diversity Methane oxidation Gas emission
1. Introduction Under the anaerobic conditions prevailing in landfills, the microbial degradation of the waste organic fraction yields a gas mixture typically composed of 55e60% v/v CH4 and 40e45% v/v CO2, accompanied by various trace gases. Along with agriculture, fossil fuel extraction and biomass burning, waste management and disposal, i.e. landfills, constitute one of the principal anthropogenic methane sources. They are estimated to be responsible for emissions as high as 35e69 Tg CH4/a and are ranked as second largest anthropogenic methane emission source in Europe, making up 22% of the total anthropogenic methane emissions at approximately 3.6 Tg/a (~90 Tg CO2 equivalents/a; [1]). The microbial oxidation of
* Corresponding author. E-mail addresses: [email protected] (J. Gebert), [email protected] (M. Perner). http://dx.doi.org/10.1016/j.ejsobi.2015.03.006 1164-5563/© 2015 Elsevier Masson SAS. All rights reserved.
methane in optimized landfill covers or biofilters has been listed as key mitigation technology for the reduction of methane fluxes from landfills that are no longer suitable for energy recovery or flaring [2]. Methane oxidation thus presents an “end-of-pipe” option both for older, non-sanitary sites but also for sanitary landfills after active gas extraction has ended but gas production continues for decades or even centuries on a lower level. As a result of the high source strength of methane, landfill soil covers harbor an abundant methanotrophic community, usually dominated by Methylosinus and Methylocystis species [3e6], classified as type II methanotrophs. Landfill cover soils show potential methane oxidation rates in the order of several hundred g CH4 m2 d1 [7,8] which, however, are subject to large spatial variability [9], resulting from the spatial variability of gas fluxes through the cover soil, caused by preferential flow paths in the soil. These are formed by differences in soil texture and compaction, through soil aggregation, rootage [10] or animal burrows, providing pathways of enhanced gas permeability and hence for preferential gas [11,12]
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
and water [13] flow. Thus, methane emissions are not spatially homogeneous but occur through small-scale areas, often referred to as emission hotspots [14], creating small-scale patterns of habitats with varying environmental conditions. Hotspots of gas emission can often be identified visually, for example by a change, damage or lack of vegetation, likely due to the shortage of oxygen in the root zone as a result of strong advective flow of gas from the underlying waste body [15]. Moreover, a distinct discoloration of the topsoil towards purple colors is frequently encountered at locations of increased gas emission (author's personal observation). In contrast, such discoloration was never observed at methanotrophically very active sites, harboring an abundant methanotrophic population but not showing high emissions [16]. It is hypothesized that the discoloration stems from the particular microbial community established as a result of the specific habitat posed by the enhanced gas fluxes. Characteristics of these habitats comprise, for example, excess availability of methane-derived carbon and of hydrogen sulfide (both components of landfill gas) and possible changes between anaerobic and aerobic conditions over very short distances on the micrometer to centimeter scale (soil pore system to secondary macropores such as cracks). However, to date, the microbial community of such discolored patches has never been investigated. The aim of this study was to elucidate possible differences between microbial community compositions from methane emitting hotspot and neighboring non-methane-emissive locations, consisting of the same original type of soil. The focus lay particularly on those parts of the community that were affected directly or indirectly by methane concentrations. 2. Materials and methods 2.1. Study site The investigated site is a 1.5 ha old municipal solid waste landfill located approximately 70 km southwest of the city of Hamburg in northwestern Germany that was operated between 1970 and 1982. The waste, a mixture of household (37 mass-%), construction and demolition (27 mass-%), industrial (10 mass-%), other inert waste (11 mass-%) and sludges (15 mass-%) [17], was disposed of in a former sand pit on top of which waste was further piled up to form a mound. Base or surface liners are missing so that landfill gas can migrate freely from the landfill body through the cover. The gas extraction system was operated only in the first few years after closure. Cover soil properties and cover thickness vary strongly and basically reflect whatever material was available to the landfill operator at the time of cover construction. Soil texture is mostly sandy, however, bulk density, pores size distribution, as well as soil chemical properties such as carbon and nitrogen content vary strongly (data given in Ref. [9]). The site is vegetated, mostly with grassland interspersed with shrubs, bushes and small trees. Previous analyses of the methanotrophic community using diagnostic microarray revealed a preponderance of type II methanotrophs [6] and a 1 methane oxidation rate of 0.3e19.5 mg CH4 g1 [9]. Assuming a dw h thickness of the methane oxidation horizon of 40 cm and a dry bulk density of the soil of * g cm3, these oxidation rates translate to 0.17 and 9.80 g CH4 m2 d1. Actual gas production was estimated at 5700 m3 CH4 per year, using the IPCC (Intergovernmental Panel on Climate Change) gas production model [17,18], yielding a hypothetical spatial load to the cover soil of 0.72 g CH4 m2 d1, assuming an even spatial distribution of this load to the landfill cover. However, due to the above mentioned preferential nature of gas flow, emissions escaped through hotspots with fluxes of up to 155 g CH4 d1, emanating from areas of a few square centimeters in
9
size [14]. 2.2. Sampling locations and sampling strategy Gas and soil samples were collected on the 26th of April, 2012 from two areas (1 and 2) at the south eastern slope of the landfill. On this slope, several hotspots of gas emission had been detected during a previous monitoring campaign [14]. The two areas were 3 m apart on the same height of the slope. Emissions in this area, monitored by means of static chambers, ranged between 2.7 and 62.5 g CH4 d1 with an average of 32.1 g CH4 d1 (n ¼ 31, measured biweekly to monthly from 2008 to 2010). The methodology of the chamber measurement is described in Ref. [14]. Within these two areas six sampling locations were chosen: two were hallmarked by distinct purple discoloration of the soil (1A and 2A), a lack of vegetation and very high surface methane concentrations. Within a maximum distance of 80 cm from 1A to 2A, two further sampling sites, namely 1B and 1C (near 1A) and 2B and 2C (near 2A), were selected, which did not exhibit this typical purple discoloration, bore fresh grass vegetation and showed no detectable surface methane concentration. Sites 1C and 2B exhibited intermediary features, showing lower levels of surface methane concentrations. For details of sampled soils and depths see Table 1. 2.3. Soil properties and gas composition Soils were analyzed for texture according to DIN ISO 11277 [19], organic carbon and total nitrogen according to DIN ISO 10694 [20] and DIN ISO 13878 [21] using a CHN analyser vario Max (Elementar Analysensysteme GmbH). Measurement error was 0.08% TOC and 0.01% TN. Moisture was determined by gravimetric analysis following oven-drying at 105 C. Composition of the soil gas phase was sampled using a gas probe that was inserted into 4, 10, 20, 30 and 40 cm below soil surface. Gas was collected by probing through the probe septum with a needle connected to a 60 ml syringe and analyzed on-site using a biogas analyzer (BM 2000, Geotechnical Instruments (UK) Ltd.; detection limit ¼ 0.1%). The concentration of N2 was calculated by subtracting the sum of the measured concentrations of CH4, CO2 and O2 from 100. Methane concentrations at the soil surface were analyzed using a portable flame ionization detector (FID; Sewerin GmbH, Gütersloh). The reading was recorded when the FID gave a stable value. The lower detection limit was 1 ppm, the upper detection limit 14,000 ppm of volatile organic compounds. 2.4. Microbial community analyses Following collection, the soil samples were frozen at 20 C immediately until microbial community analyses were performed. 2.4.1. DNA extraction, PCR, cloning, and sequencing Around 0.5 g soil from eight samples, i.e. 1A, 1B (consisting of parallels 1B1 and 1B2), 1C1, 1C2, 2A, 2B and 2C (for soil depths see Table 1) was used to extract DNA with the UltraClean Soil DNA Isolation Kit (MoBio, Solana, CA, USA) according to the manufacturer's instructions. Bacterial 16S rRNA genes were PCR-amplified in three parallels using the oligonucleotide primer sets consisting of 27F and 1492R [22]. All amplifications were performed in triplicate, products pooled and separated by gel electrophoresis. Purified PCR products were ligated into the pGEM®-T vector system (Promega, Madison, WI, USA) and transformed into competent Escherichia coli DH5a. PCR products of the correct size were partially sequenced using the primer 27F [ [22], sequence length ¼ 331 to 856 base pairs]. Sequencing was performed with
10
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
Table 1 Selected soil properties, surface and soil methane concentrations, total organic carbon (TOC), total nitrogen (TN) and moisture content of the collected soil samples. Sample ID
1A-1 1A-2 1B 1C-1 1C-2 2A 2B 2C
Depth (cm) for
Soil properties
Methane concentration
Microbiology analyses
Abiotic properties
Purple discoloration
EPS
Vegetation
Moisture (% dw)
TOC (%)
TN (%)
TOC/TN ()
At surface (ppm)
Avg. in soil (vol. %)
In 5 cm depth (vol. %)
0e3 e 0e7 0e3 4e7 0e2 0e2 0e2
0e3 8e10 0e7 0e3 4e7 0e2 0e2 0e2
Yes
Yes
No
0 400
0 12.1
0 0
Yes No No
Yes Yes No
No No Yes
15.7 13.0 11.5 13.8 16.6 16.8 14.5 13.7
22.3
Yes No
1.035 0.634 0.307 0.445 1.124 0.694 0.4 0.244
23.9
No Yes
16.3 8.2 3.5 6.1 18.6 11.7 5.8 3.3
>14,000
No No
53.6 32.8 30.4 26.8 61.7 18.1 35.8 27.9
>14,000 380 0
17.7 20.1 5.0
7.1 19.0 0
EPS ¼ extrapolymeric substance; TOC ¼ total organic carbon; TN ¼ total nitrogen; dw ¼ dry weight.
the 3730xl DNA Analyzer (Applied Biosystems) and sequences yielded between 401 and 870 nucleotides. Detailed procedures for amplification, amplicon purification, cloning and sequencing have been described [23]. 2.4.2. Phylogenetic and statistical analyses All sequences were edited and clustered to operational taxonomic units (OTUs, similarity 93%) with the Lasergene Software SeqMan (DNAStar Inc., Madison, WI, USA). For soil sample 1A 34 consensus sequences (93 single sequences), for soil sample 1B1 94 consensus sequences (144 single sequences), for soil sample 1B2 71 consensus sequences (128 single sequences), for soil sample 1C1 69 consensus sequences (130 single sequences), for soil sample 1C2 56 consensus sequences (127 single sequences), for soil sample 2A 42 consensus sequences (106 single sequences), for soil sample 2B 66 consensus sequences (132 single sequences) and for soil sample 2C 75 consensus sequences (129 single sequences) were generated. The 16S rRNA gene sequences were compiled using the ARB software (www.arb-home.de) [24]. Maximum-likelihood based trees and 100 bootstrap replicates were constructed using PhyML [25]. Trees were imported into ARB and shorter sequences were added subsequently to the trees without changing its topology. Procedures for editing, comparing, compiling and aligning sequences and constructing maximum-likelihood trees have been reported [23]. Distance matrices of the aligned single sequences from the eight soil samples were created by using DNADIST from the PHYLIP version 3.65 package (J. Felsenstein, University of Washington, Seattle). These distance matrices served as input to DOTUR [26] for clustering DNA sequences into OTUs (97%), calculating collector's curves and diversity indices (Simpson and Shannon diversity index). The evenness index, which describes the extent of the dominance of individual OTUs, was estimated according to Pielou (1966) [27].
exopolymeric substances (EPS). Contrastingly, the soil at sites 1B and 2C had a dark brown, humic appearance (Fig. 1, right), bore fresh grass vegetation, were moist at the soil surface and had no EPS. Here, no methane could be detected at the surface (Table 1). Features of intermediary nature were observed at sites 1C and 2B, lacking discoloration but also lacking vegetation, showing a dry soil surface and lower surface methane concentrations. However, here also EPS were observed below the soil surface. Sites 1A and 2A (high surface methane concentrations) showed some of the highest TOC and TN contents, elevated up to four-fold relative to the low-emissive, non-vegetated sites 1C and 2B and up to almost five-fold compared to the non-emissive, vegetated reference sites 1B and 2C (Table 1). At these reference sites, the organic carbon and nitrogen levels fell within the range of typical values for natural terrestrial humic topsoils in Northwestern Europe [30]. TOC and TN were highly correlated (R2 ¼ 0.988, P < 0.01). However, the ratio of TOC to TN also increased with elevated organic carbon content (R2 of linear regression ¼ 0.64, P < 0.05). 3.2. Soil gas composition At sites 1A and 2A (purple discoloration, high surface methane concentrations, no vegetation, presence of EPS; see Table 1) the soil gas phase was characterized by high concentrations of methane and carbon dioxide and low concentrations of oxygen up to the shallowest depth (Fig. 2, upper panel). The concentration of the landfill gas constituents remained almost unchanged from bottom to top. Also, at both sites, no or hardly any increase in the ratio of
2.4.3. Nucleotide sequence accession numbers The sequence data have been submitted to DDBJ/EMBL/GenBank databases under accession numbers KJ633283 e KJ633789. 3. Results 3.1. Soil properties and surface methane concentrations The top 10 cm of the soil at all sampled sites were of identical texture, classified as sandy loam [28]. At locations 1A and 2A a purple discoloration of the soil surface was observed (Table 1 and Fig. 1, left; Munsell value 5R 2.5/3) [29]. The soil also appeared to be dry, coinciding with high surface methane concentrations and a lack of vegetation (Table 1). Below the surface the soil was aggregated to form lumps of slimy consistency, likely to be caused by
Fig. 1. Discoloration of sample 1A (left) in comparison to sample 1C (right).
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
11
Fig. 2. Composition of the soil gas phase (soil gas profiles) from sampling sites 1A and 2A with high surface methane concentrations (upper panel), sites 1C and 2B with medium surface methane concentrations (middle panel) and sites 1B and 2C where no methane could be detected at the surface (lower panel).
carbon dioxide to methane was observed, indicating hardly any methanotrophic activity which is otherwise characterized by a strong increase in the ratio due to consumption of methane and production of carbon dioxide [31]. The top 5 cm at site 2C were better aerated, accompanied by a slight increase in the ratio of carbon dioxide to methane, indicating a low level of methanotrophic and/or respiratory activity. Site 2B was initially classified as a site with features of intermediary nature, lacking purple discoloration at the surface and lacking vegetation, showing a lower level of surface methane concentrations around 400 ppm but exhibiting presence of EPS. However, the soil gas profile was very similar to that at site 1A and 2A with high concentrations of methane and carbon dioxide and low concentrations of oxygen, which remained almost unchanged from bottom to top of the soil cover. Also at site 2B, the ratio of carbon dioxide to methane remained almost constant, indicating hardly any methanotrophic activity. Sites 1A, 2A and 2B also exhibited the highest average methane concentration in the soil gas phase and the highest methane concentration in 5 cm depth (Table 1). In contrast, at site 1B (no discoloration, no methane measured at
surface, fresh grass vegetation) no methane was found in the soil gas phase (Fig. 2, lower left panel). The soil gas profile showed a high extent of aeration of the soil, even in greater depths, indicated by both high oxygen and near-atmospheric nitrogen concentrations. Similar to site 1B, site 2C (also lacking purple discoloration, surface methane and EPS, and also bearing fresh grass vegetation) had a well-aerated topsoil with methane concentrations below the detection limit. However, different from site 1B, methane was found in the subsoil at concentrations up to 8 vol. % (Fig. 2, lower right panel). Although with respect to the presence of EPS, vegetation and surface methane concentrations some differences between 1C and 2C in soil properties were apparent, their soil gas profiles had some evident similarities (Fig. 2, middle left and lower right panel). Also at site 1C, a well-aerated topsoil and increased methane concentrations in the subsoil were found, with up to 24 vol. % at a higher level, though. 3.3. Community composition All soil samples had a high bacterial diversity (Table S1) as is commonly found for soil environments [32]. However, the
12
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
communities were not sampled for complete coverage of the bacterial community (Fig. S1). The communities of the soil samples consisted of bacteria across much of the bacterial domain (Fig. 3) including members of the Alphaproteobacteria including methanotrophs of type II (Fig. S2), Betaproteobacteria, Gammaproteobacteria including methanotrophs of type I, Planctomycetes, Nitrospira (Fig. S3), Bacteroidetes, Deltaproteobacteria, Verrucomicrobia, Acidobacteria, Firmicutes, Actinobacteria, Chloroflexi, and candidate divisions, e.g. TM6 or BRC1. All samples had relatively large but varying proportions of methanotrophic and non-methane utilizing methylotrophic bacteria (Fig. 3B) [33]. In samples 1A, 2A and 1C2, sequences clustering with Methylocella/Methylocapsa of the alphaproteobacterial type II methanotrophs predominated (Figs. 3B, S2). In these samples no or hardly any gammaproteobacterial type I methanotrophs were identified. Relatively even proportions of type I and type II methanotrophs were found in samples 1C1 and 2B. In samples 1B1, 1B2 and 2C type I methanotrophs prevailed over type II methanotrophs. Methylotrophs not capable of using methane were found in all samples readily (Fig. 3B). Among the more abundant lineages this included particularly Burkholderia-related organisms in sample 2A, 1C1 and 2B, Bradyrhizobium in sample 1A, 1C2 and 1B1 and Planctomycetes that
are suggested to be capable of methylotrophy [33,34]. 3.3.1. High surface methane concentration (sites 1A and 2A) The lowest bacterial diversity was in the soil samples for which the highest surface methane concentrations and the highest TOC and TN content were measured, namely 1A and 2A (Tables 1, S1). However, the two methane-rich soil samples appeared to host different proportions of methanotrophic bacteria (Figs. 3, S2 and S3). At site 1A, no type I methanotrophs were identified, but 9% of the 16S rRNA gene sequences were affiliated with type II methanotrophs. By contrast, at site 2A, 3% could be grouped with type I methanotrophs and 18% of the sequences were attributed to the type II methanotrophic bacteria. Members of purple non-sulfur bacteria were recognized in soil sites 1A and 2A and included Rhodosprillum and Roseomonas of the Rhodospirillales (Figs. 3, S2). Also, at site 2A a large proportion of Janthinobacterium of the Burkholderiales was found (Figs. 3, S3). 3.3.2. Intermediate surface methane concentration (sites 1C and 2B) Relative to site 1A and 2A, at site 1C and 2B the bacterial composition appeared to shift from a few dominant microbial groups to a broader diversity of different lineages, which did not dominate the overall community composition (Fig. 3). Methylococcales of the type I methanotrophs and Beijerinckiaceae and Methylocystaceae of the type II methanotrophs appeared most abundant in the soil sample 1C (11% and 21%, respectively, Figs. S3 and S4). In the soil sample 2B 4% type I and 8% type II methanotrophs were identified. Planctomycetes made up a relatively large proportion in the 1C and 2B soil bodies (8% of sequences). 3.3.3. No surface methane concentration (sites 1B and 2C) The highest bacterial diversities were noted for 1B1 and 2C where no methane was measured at the surface and in 5 cm depth and where TOC contents were the lowest (Tables 1 and S1). In the soil sample 1B, 12% and 3% of the sequences were affiliated to 16S rRNA genes of type I and type II methanotrophs, respectively (Figs. S2 and S3). The sample from site 2C comprised 7% and 1% type I and type II. In these soils Planctomycetes accounted for 12% and 8% of the 16S rRNA gene sequences at site 1B and 2C, respectively. 3.4. Linkages between abiotic factors and the microbial community
Fig. 3. Proportions of 16S rRNA genes for each site for selected microbial lineages. Details of relationships can be found in the phylogenetic trees in the Supplementary materials in Figs. S2 and S3.
Regression analyses between abiotic and biotic properties showed that in most cases the data points cluster quite strongly, for example, around high or low evenness or a high or low Simpson index (Fig. 4; for the entire results set of regression analyses see Supplementary material). Therefore, in spite of the fact that values for the coefficient of determination (R2) and likelihood of error (p) are given, the depicted relationships should be carefully interpreted as a qualitative indication of the relationship between the respective parameters. The average soil methane concentrations are negatively related with the Shannon diversity index and species richness (Table S2). Hence, the higher the methane concentrations in the soil were, the lower the diversity of the community seemed to be (Fig. 4). Besides correlating with the average methane concentrations, richness also correlated negatively with levels of TOC and TN, and with the TOC/TN ratio (Table S2). Thus, the higher the average methane soil concentrations were, the higher TOC and TN was and the lower species richness was at a given site (Fig. 4). The Simpson diversity index in turn appeared to be inversely related to the presence of type I methanotrophs and Planctomycetes and positively with the relative abundance of clone sequences affiliated with type II methanotrophs (Fig. 4, Table S2).
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
13
Fig. 4. Relationships between different abiotic and microbiological properties of the samples with distinct methane surface concentrations. The linkages of Simpson diversity index, the evenness index, OTUs, TOC, TN and average methane soil concentrations, type I and II methanotrophs and Planctomycetes are displayed. The lines mark the linear regression.
4. Discussion 4.1. Soil gas composition The observed gas profiles at sites 1A, 1B and 2B are typical for
hotspots of gas emission, showing hardly any changes in the gas composition from bottom to top and only little or no aeration [34]. These sites are usually characterized by advective (i.e. pressuredriven) gas transport, impeding the diffusive ingress of atmospheric oxygen into the soil. The vertical stratification of the soil gas
14
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
composition is usually stable over time, i.e. there are no seasonal differences as typical for sites with diffusion-driven gas transport [34]. Unfavorable conditions for methanotrophic activity are caused by restricted aeration in addition to an excessive methane load, usually too high for complete oxidation, even if the level of aeration would suffice. As a result, considerable methane fluxes emit from these sites [14]. In contrast, the vertical stratification of soil gas composition at sites 1C and 2C indicates mainly diffusive upward transport of landfill gas, allowing for ingress of atmospheric oxygen from the soil surface and thus for favorable conditions for methanotrophic activity in the upper part of the soil cover. Typical patterns, as also seen at sites 1C and 2C, comprise elevated concentrations of landfill gas components in the subsoil and their strong decrease towards the surface, while the concentration of atmospheric components (oxygen, nitrogen) increases. Methane oxidation is indicated by strongly increasing ratios of carbon dioxide to methane towards the soil surface [31]. Characteristically, these gas composition patterns are subject to pronounced seasonal variation, reflecting the change of parameters impacting methanotrophic activity, such as soil temperature and moisture [14,34]. As a result, emissions are usually observed under cooler and moister but not under warmer and dryer conditions. At the time of sampling, the top 40 cm at site 1B were fully aerated and not influenced by landfill gas at all. Elevated carbon dioxide concentrations were either due to methane oxidation at depths below 40 cm or resulted from soil respiratory activity or both.
the result of methanotrophic activity under conditions of excess carbon and oxygen deficiency [35,36], it is posited that the EPS at site 1A and 2A arises mainly from type II methanotrophic activity. The higher proportion of methanotrophs in general as well as the presence of type I methanotrophs in upper soil crust at site 2A may be related to the better aeration of the soil in the upper 5 cm at this site (Fig. 2, upper panel), due to the different oxygen demands discussed above [41e43]. Conditions towards considerably lower oxygen concentrations in the upper soils of site 1A are also supported by elevated amounts of 16S rRNA genes of the Rhodospirillales (Figs. 3, S2), which typically thrive under microaerophilic or anaerobic conditions [45]. Members of purple non-sulfur bacteria were recognized in soil sites 1A and 2A and included Rhodosprillum and Roseomonas of the Rhodospirillales (Figs. 3, S2). These organisms may be responsible for the purple discoloration observed at sites 1A and 2A (compare Fig. 1) [46,47]. Alternatively, at site 2A a large proportion of Janthinobacterium of the Burkholderiales was found (Figs. 3, S3), which may also contribute to the purple discoloration [48]. In methanotrophic enrichment cultures, redcolored heterotrophic, mostly nocardioformic microflora accompanying methanotropic enrichment cultures are frequently found. However, in this study no sequence associated with this group of bacteria was detected. These lineages are thus either absent in the sampled environments or, if present, in extremely low abundances. Hence, nocardioforme bacteria are unlikely to be responsible for the observed coloring in the samples. 4.4. Community composition at intermediate surface methane sites (1C, 2B)
4.2. Soil C and N Sites 1A, 1C, 2A and, to a lesser extent, 2B, were characterized by strongly elevated levels in organic carbon and nitrogen (Table 1). Since vegetation was missing at these locations, plant biomass cannot have contributed to the organic fraction of the soil. Rather, elevated nitrogen levels likely reflect a more prolific microbial biomass and elevated organic carbon contents may be attributed to a higher microbial biomass and to EPS. EPS consist mainly of polysaccharides; their occurrence has been interpreted as the result of methanotrophic metabolism under conditions of excess carbon, oxygen deficiency and nitrogen limitation [35,36]. Formation of EPS in relation to high methane fluxes have been observed in laboratory studies by a number of authors [37e40], however, so far it has not been described for field settings. The fact that the ratio of TOC to TN increased with increasing TOC contents (Table 1) indicates a change in the nature of the organic matter towards components containing more carbon and less nitrogen, as for example nitrogen-free polysaccharide compounds referred to as EPS. Indeed, EPS were observed in the soils collected from sites 1A and 2A as well as sites 1C and 2B (Table 1). 4.3. Community composition at high surface methane sites (1A, 2A) At sites 1 A and 2A, type II methanotrophs appear to have outcompeted type I methanotrophs, which is in line with the fact that type II methanotrophs have a competitive advantage under methane-rich and oxygen limited conditions hallmarking these sites [[41e43]; Fig. 1, upper panel]. Previous investigations in landfill cover soils and landfill biofilters have also revealed a dominance of type II methanotrophs [6,44]. The exceptionally high TOC/TN-ratios indicate a trend towards nitrogen-free organic matter like EPS, which was also detected in these soil samples (Table 1). Given that EPS was observed in these soils, that type II methanotrophs constitute a large proportion of the retrieved 16S rRNA genes and that the formation of EPS has been interpreted as
The higher numbers of clone sequences affiliated with methanotrophs at site 1C relative to site 2B are likely a reflection of better aerated soils above 10 cm at site 1C (Fig. 2, middle panel). High methane concentrations were found in the soil, in case of site 1C up to 40 cm depth and in case of size 2B up to at least 5 cm depth (similar to sites 1A and 2A discussed above). High methanotrophic activity at site 1C was indicated by a rapid increase in the ratio of carbon dioxide to methane above 20 cm depth. At site 2B, high methanotrophic activity can also be expected above 5 cm depth. Given the high soil methane concentrations and the methanotrophic activity evidenced by the change in the ratio of carbon dioxide to methane, it is assumed that similar to the sites with high surface methane concentrations discussed above EPS at sites 1C and 2B is also produced by methanotrophs in reaction to excess carbon supply [35,36]. Most of the characterized Planctomycetes isolates are slow growing, aerobic or facultative chemoheterotrophs specialized on carbohydrate metabolisms. Since some Planctomycetes have been shown to contain genes in the tetrahydromethanopterin-linked C1 transfer pathway, the potential for methylotrophy has been suggested [cf. [49]]. Some members of this deeply diverging lineage also appear to be capable of chemolithoautotrophic growth through anaerobic oxidation of ammonium [see Ref. [49]]. Conditions in these soils seem to be favorable for their successful colonization. 4.5. Community composition at no surface methane sites (1B, 2C) The highest bacterial diversities were noted for 1B1 and 2C where no methane was measured at the surface and in 5 cm depth and where TOC contents were the lowest (Table 1 and S1). These findings supported the trend that with decreasing methane concentrations the community changes from a few dominant groups towards a broader phylogenetic microbial community. In both samples, the methanotrophic community was dominated by type I
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
15
organisms (Figs. 3, S2 and S3). Since both soils are rich in oxygen and poor in methane (Fig. 2, lower panel), these findings coincide with the known type I methanotrophs' ability to outcompete type II strains under oxygen-rich and methane-deficient conditions [41e43]. The lack of EPS in these soils in conjunction with very low type II methanotrophic sequences is supportive of our assumption of EPS formation at the other sites (discussed above) being related to type II methanotrophic activity. In the ‘low methane’ soils at sites 1B and 2C, Planctomycetes accounted for 12% and 8% of the 16S rRNA gene sequences, respectively, matching the prevailing aerobic conditions.
ecological driver posed by enhanced landfill gas fluxes at sites of preferential pathways, caused by the small-scale patterned gas permeability of the soil. The presence and activity of the sitespecific microbial community in turn significantly influences the soil properties, in the case of this study e.g. by increasing and changing the nature of the soil organic matter through formation of extrapolymeric substances. The changed microbial community at high methane flux sites is also likely to be responsible for the specific discoloration of the soil.
4.6. Linkages between soil properties and community composition
We would like to thank Annica Beuerbach for excellent technical assistance. The presented work was funded by the German Federal Ministry for Education and Research (BMBF), grant number 01 LS 05095.
The correlation analyses suggest that high average soil methane concentrations and corresponding low levels of soil aeration, caused by advective gas flux from the waste body, are accompanied by high TOC and TN levels in the topsoil. The higher TOC and TN levels are, the more type II methanotrophs and the less type I methanotrophs and Planctomycetes can be detected in the soils. Additionally, the bacterial diversity and evenness of species distribution appear reduced. This is suggestive of the development of a more specialized microbial community that appear to become dominated by a few lineages such as methanotrophs, particularly of type II methanotrophs, Burkholderiales, Rhodospirillales and Bradyrhizobiaceae with enhanced gas fluxes. Under specific conditions methanotrophs may accumulate methanol outside of the cell [50] which in turn may be utilized by methylotrophs such a members of the Bradyrhizobiaceae. Members of the latter three may be responsible for the distinct purple discoloration of the soil observed at the emissive sites. This in turn suggests that preferential pathways of gas act as a strong driver for the development of more specialized, less diverse communities. In a study examining soil properties and microbial community composition along preferential pathways of water flow in forest soil, an increase in organic carbon and nitrogen contents in the soil along the preferential flow paths was also found, alongside with elevated microbial biomass compared to the bulk soil [51]. Also, a different community structure was detected for some bacterial groups, however, for others not. The authors concluded that only few communities with the ability to thrive under aerobic and anaerobic conditions can compete at sites of preferential flow paths. Within the methanotrophs, it is the type II organisms that are better suited to survive adverse (anaerobic) conditions resulting in restricted growth rates. These have thus been reported to form communities that are more stable over time, compared to the faster growing type I methanotrophs that are more adapted to react to changing environmental conditions and that show more variability in community composition [43,52,53]. Examples for such adverse conditions are low or zero oxygen levels even near the soil surface under conditions of strong advective landfill gas flux, as is typical for preferential pathways of gas flux (see Fig. 2 and [34]). It thus appears plausible that at the sites with elevated gas fluxes (1A, 2A) the community is dominated by type II methanotrophs. 5. Conclusions The scope of this study was to identify the impact that landfill gas fluxes along preferential pathways (hotspots) in landfill soil covers have on soil properties and the respective microbial community. Soil properties and the microbial community composition are posited to be a result from the specific habitat created by the differing magnitude of landfill gas fluxes. The study demonstrates how the composition of the microbial community shifts towards a less diverse, more specialized population as a result from the strong
Acknowledgments
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ejsobi.2015.03.006. References [1] European Environment Agency (EEA), Technical Report No.2/2011. Annual European Union Greenhouse Gas Inventory 1990e2009 and Inventory Report, Submission to the UNFCCC Secretariat, 27 May 2011. [2] J. Bogner, M. Abdelrafie Ahmed, C. Diaz, A. Faaij, Q. Gao, S. Hashimoto, K. Mareckova, R. Pipatti, T. Zhang, Waste management, in: B. Metz, O.R. Davidson, P.R. Bosch, R. Dave, L.A. Meyer (Eds.), Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2007. [3] N. Stralis-Pavese, A. Sessitsch, A. Weilharter, T. Reichenauer, J. Riesing, J. Csontos, J.C. Murrell, L. Bodrossy, Optimization of diagnostic microarray for application in analysing landfill methanotroph communities under different plant covers, Environ. Microbiol. 6 (2004) 347e363. [4] N. Stralis-Pavese, L. Bodrossy, T.G. Reichenauer, A. Weilharter, A. Sessitsch, 16 S rRNA based T-RFLP analysis of methane oxidising bacteria e assessment, critical evaluation of methodology performance and application for landfill site cover soils, Appl. Soil Ecol. 31 (2006) 251e266. ry, A.C. Singer, D. Kumaresan, L. Bodrossy, N. Stralis-Pavese, J.I. Prosser, [5] M. He I.P. Thompson, J.C. Murrell, Effect of earthworms on the community structure of active methanotrophic bacteria in a landfill cover soil, ISME J. 2 (2008) 92e104. [6] J. Gebert, B.K. Singh, Y. Pan, L. Bodrossy, Activity and structure of methanotrophic communities in landfill cover soils, Environ. Microbiol. Rep. 1 (2009) 414e423. [7] C. Scheutz, P. Kjeldsen, J. Bogner, A. De Visscher, J. Gebert, H. Hilger, M. HuberHumer, K. Spokas, Processes and technologies for mitigation of landfill gas emissions by microbial methane oxidation, Waste Manag. Res. 27 (2009) 409e455. [8] J. Streese-Kleeberg, I. Rachor, J. Gebert, R. Stegmann, Field quantification of methane oxidation in landfill cover soils by means of gas push-pull tests, Waste Manag. 31 (2011) 995e1001. € wer, C. Geck, J. Gebert, E.-M. Pfeiffer, Spatial variability of soil gas [9] I.U. Ro concentration and methane oxidation capacity in landfill covers, Waste Manag. 31 (2011) 926e934. [10] S. Bohn, P. Brunke, J. Gebert, J. Jager, Influence of vegetation on microbial methane oxidation e a column study, Waste Manag. 31 (2011) 854e863. [11] A. Tuli, J.W. Hopmans, D.E. Rolston, P. Moldrup, Comparison of air water permeability between disturbed and undisturbed soils, Soil Sci. Soc. Am. J. 5 (2005) 1361e1371. [12] S.E. Allaire, J.A. Lafond, A.R. Cabral, S.F. Lange, Measurement of gas diffusion through soils: comparison of laboratory methods, J. Environ. Monit. 10 (2008) 1326e1336. [13] S. Gebhardt, H. Fleige, R. Horn, Effect of compaction on pore functions of soils in a Saalean moraine landscape in North Germany, J. Plant Nutr. Soil Sci. 172 (2009) 688e695. € ngro €ft, E.-M. Pfeiffer, Variability of methane emis[14] I. Rachor, J. Gebert, A. Gro sions from an old landfill on different time scales, Eur. J. Soil Sci. 64 (2013) 16e26. €ngro €ft, E.-M. Pfeiffer, Temporal variability of soil gas [15] J. Gebert, I. Rachor, A. Gro composition in landfill covers, Waste Manag. 31 (2011) 935e945. €ngro € ft, G. Miehlich, Kinetics of microbial landfill methane [16] J. Gebert, A. Gro oxidation in biofilters, Waste Manag. 23 (2003) 609e619. [17] A.M. Yemaneh, Assessment of Biological Waste Stability and Estimation of
16
[18]
[19]
[20]
[21]
[22]
[23]
[24] [25] [26]
[27] [28]
[29] [30]
[31]
[32]
[33] [34] [35]
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16 Methane Formation Potential in Old Landfills e Case of Three Landfills Near the City of Hamburg (Master thesis), Hamburg University of Technology, Institute for Environmental Technologies and Energy Industry, 2010. IPCC (Intergovernmental Panel on Climate change), in: S. Eggleston, K. Buendia, T. Ngara, K. Tanabe (Eds.), 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Waste, vol. 5, 2006. http://www.ipcc-nggip.iges. or.jp/public/2006gl/vol5.html (accessed 01.02.14.). DIN ISO 11277, Bodenbeschaffenheit. Bestimmung der Parti€ßenverteilung in Mineralbo €den. Verfahren mittels Siebung und Sedikelgro mentation, Deutsches Institut für Normung e.V. Beuth Verlag, Berlin, 2002. DIN ISO 10694, Bodenbeschaffenheit: Bestimmung von organischem Kohlenstoff und Gesamtkohlenstoff nach trockener Verbrennung (Elementaranalyse), Deutsches Institut für Normung e.V. Beuth Verlag, Berlin, 1996. DIN ISO 13878, Bodenbeschaffenheit: Bestimmung des Gesamt-Stickstoffs durch trockene Verbrennung (Elementaranalyse), Deutsches Institut für Normung e.V. Beuth Verlag, Berlin, 1998. D.J. Lane, 16S/23S rRNA sequencing, in: E. Stackebrandt, M. Goodfellow (Eds.), Nucleic Acid Techniques in Bacterial Systematics, John Wiley & Sons, Chichester, England, 1991, pp. 115e175. €nberg, M. Perner, W. Bach, M. Hentscher, A. Koschinsky, D. Garbe-Scho W.R. Streit, H. Strauss, Short-term microbial and physico-chemical variability in low-temperature hydrothermal fluids near 5 S on the Mid-Atlantic Ridge, Environ. Microbiol. 11 (2009) 2526e2541. W. Ludwig, O. Strunk, R. Westram, ARB: a software environment for sequence data, Nucleic Acids Res. 32 (2004) 1363e1371. S. Guindon, O. Gascuel, A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood, Syst. Biol. 52 (2003) 696e704. P.D. Schloss, J. Handelsman, Introducing DOTUR, a computer program for defining operational taxonomic units and estimating species richness, Appl. Environ. Microbiol. 71 (2005) 1501e1506. E.C. Pielou, The measurment of diversity in different types of biological collections, J. Theor. Biol. 13 (1966) 131e144. FAO e ISRIC & ISSS (Ed.), World Reference Base for Soil Resources 2006: a Framework for International Classification, Correlation and Communication, 2006. World soil resources reports 103. Rome. Munsell Color, Munsell Soil-color Charts, Revised and produced in 2009, Michigan, USA, 2009. R.J.A. Jones, R. Hiederer, E. Rusco, P.J. Loveland, L. Montanarella, The Map of Organic Carbon in Topsoils in Europe, Version 1.2, September 2003: Explanation of Special Publication Ispra 2004 No.72 (S.P.I.04.72), European Soil Bureau Research Report No.17, Office for Official Publications of the European Communities, Luxembourg, 2004. EUR 21209 EN, 26 pp. and 1 map in ISO B1 format. €wer, H. Scharff, C.D.L. Roncato, A.R. Cabral, Can soil gas profiles J. Gebert, I.U. Ro be used to assess microbial CH4 oxidation in landfill covers? Waste Manag. 31 (2011) 987e994. J. Gans, M. Wolinsky, J. Dunbar, Computational improvements reveal great bacterial diversity and high metal toxicity in soil, Science 309 (2005) 1387e1390. S. Kolb, Aerobic methanol-oxidizing bacteria in soil, FEMS Microbiol. Lett. 300 (2009) 1e10. €ngro €ft, E.-M. Pfeiffer, Temporal variability of soil gas J. Gebert, I. Rachor, A. Gro composition in landfill covers, Waste Manag. 31 (2011) 935e945. J.D. Linton, J.P. Watts, R.M. Austin, D.E. Haugh, H.G.D. Neikus, The energetics and kinetics of extracellular polysaccharide production from methanol by
[36]
[37]
[38] [39]
[40]
[41]
[42] [43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51] [52] [53]
microorganisms possessing different pathways of C1 assimilation, J. Gen. Microbiol. 132 (1986) 779e788. W. Babel, Peculiarities of methylotrophs concerning overflow metabolism, especially the synthesis of polyhydroxyalkanoates, FEMS Microbiol. Rev. 103 (1992) 141e148. H.A. Hilger, D.F. Cranford, M.A. Barlaz, Methane oxidation and microbial exopolymer production in landfill cover soil, Soil Biol. Biochem. 32 (2000) 457e467. J. Streese, R. Stegmann, Microbial oxidation of CH4 from old landfills in biofilters, Waste Manag. 23 (2003) 573e580. J.H. Wilshusen, J.P.A. Hettiaratchi, V.B. Stein, Long-term behaviour of passively aerated compost methanotrophic biofilter columns, Waste Manag. 24 (2004) 643e653. R. Haubrichs, R. Widmann, Evaluation of aerated biofilter systems for microbial methane oxidation of poor landfill gas, Waste Manag. 26 (2006) 408e416. D.W. Graham, J.A. Chaudhary, R.S. Hanson, R.G. Arnold, Factors affecting competition between type I and type II methanotrophs in two-organism, continuous-flow reactors, Microb. Ecol. 25 (1993) 1e17. J.A. Amaral, R. Knowles, Growth of methanotrophs in methane and oxygen counter gradients, FEMS Microbiol. Lett. 126 (1995) 215e220. T. Henckel, P. Roslev, R. Conrad, Effects of O2 and CH4 on presence and activity of the indigenous methanotrophic community in rice field soil, Environ. Microbiol. 2 (2000) 666e679. J. Gebert, N. Stralis-Pavese, M. Alawi, L. Bodrossy, Analysis of methanotrophic communities in landfill biofilters by means of diagnostic microarray, Environ. Microbiol. 10 (2008) 1175e1188. G.M. Garrity, D.J. Brenner, N.R. Krieg, J.T. Staley (Eds.), Bergey's Manual of Systematic Bacteriology, The Proteobacteria, Part C: the Alpha-, Beta-, Delta-, and Epsilonproteobacteria, vol. Two, Springer, New York, New York (, 2005, ISBN 978-0-387-24145-6, pp. 354e361. S. Hou, K.S. Makarova, J.H. Saw, Complete genome sequence of the extremely acidophilic methanotroph isolate V4, Methylacidiphilum infernorum, a representative of the bacterial phylum Verrucomicrobia, Biol. Direct 3 (2008) 26. J.H. Saw, B.W. Mountain, L. Feng, Encapsulated in silica: genome, proteome and physiology of the thermophilic bacterium Anoxybacillus flavithermus WK1, Genome Biol. 9 (2008) R161. H.J. Tripp, J.B. Kitner, M.S. Schwalbach, J.W. Dacey, L.J. Wilhelm, S.J. Giovannoni, SAR11 marine bacteria require exogenous reduced sulphur for growth, Nature 452 (2008) 741e744. D.H. Buckley, V. Huangyutitham, T.A. Nelson, A. Rumberger, J.E. Thies, Diversity of Planctomycetes in soil in relation to soil history and environmental heterogeneity, Appl. Environ. Microbiol. 72 (2006) 4522e4531. J. Xin, J. Cui, J. Niu, S. Hua, C. Xia, S. Li, L. Zhu, Production of methanol from methane by methanotrophic bacteria, Biocatal. Biotransformation 22 (2004) 225e229. M. Bundt, F. Widmer, M. Pesarob, J. Zeyer, P. Blaser, Preferential flow paths: biological ‘hot spots’ in soils, Soil Biol. Biochem. 33 (2001) 729e738. R.S. Hanson, T.E. Hanson, Methanotrophic bacteria, FEMS Microbiol. Rev. 60 (1996) 439e471. M.S. Vecherskaya, V.F. Galchenko, E.N. Sokolova, V.A. Samarkin, Activity and species composition of aerobic methanotrophic communities in tundra soils, Curr. Microbiol. 27 (1993) 181e184.
Contents lists available at ScienceDirect
European Journal of Soil Biology journal homepage: http://www.elsevier.com/locate/ejsobi
Original article
Impact of preferential methane flow through soil on microbial community composition Julia Gebert a, *, Mirjam Perner b a b
University of Hamburg, Institute of Soil Science, Allende-Platz 2, 20146 Hamburg, Germany University of Hamburg, Biocenter Klein Flottbek, Molecular Biology of Microbial Consortia, Ohnhorststraße 18, 22609 Hamburg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 September 2014 Received in revised form 21 March 2015 Accepted 23 March 2015 Available online 28 March 2015 Handling editor: C.C. Tebbe
The anaerobic microbial degradation of waste organic fractions in landfills constitutes one of the principal anthropogenic methane sources. Microbial oxidation of methane in optimized landfill covers or biofilters has been listed as key mitigation technology for the reduction of methane fluxes from landfills that are no longer suitable for energy recovery or flaring. Therefore, it is vital to understand what influences distribution of methane oxidizers and their activity in landfill soils. Here we describe the impact of gas fluxes through preferential pathways (hotspots) in the cover soil of a municipal solid waste landfill in north-western Germany on the soil properties and the microbial communities that colonize the upper soil crust in these environments. Two sites with high surface methane concentrations (>14,000 ppm), two sites with moderate surface methane concentrations (~400 ppm) and two sites without measurable methane emissions at the surface were investigated. It was found that elevated average soil methane concentrations coincided with increased levels of TOC and TN and the TOC/TN ratio in the topsoil. The increase of the latter posits a change in the composition of the organic matter towards increasing levels of nitrogen-poor components as for example EPS, which were observed in the samples with higher TOC/ TN ratios. Elevated average soil methane concentrations were also accompanied by a decrease in the overall bacterial diversity. The community at these sites were dominated by a few lineages such as methanotrophs, particularly of type II, Burkholderiales, Rhodospirillales and Bradyrhizobiaceae. This dominance may have contributed to the purple discoloration at the soil surface at the sites with the highest surface methane concentrations. © 2015 Elsevier Masson SAS. All rights reserved.
Keywords: Landfill cover soil Microbial community Diversity Methane oxidation Gas emission
1. Introduction Under the anaerobic conditions prevailing in landfills, the microbial degradation of the waste organic fraction yields a gas mixture typically composed of 55e60% v/v CH4 and 40e45% v/v CO2, accompanied by various trace gases. Along with agriculture, fossil fuel extraction and biomass burning, waste management and disposal, i.e. landfills, constitute one of the principal anthropogenic methane sources. They are estimated to be responsible for emissions as high as 35e69 Tg CH4/a and are ranked as second largest anthropogenic methane emission source in Europe, making up 22% of the total anthropogenic methane emissions at approximately 3.6 Tg/a (~90 Tg CO2 equivalents/a; [1]). The microbial oxidation of
* Corresponding author. E-mail addresses: [email protected] (J. Gebert), [email protected] (M. Perner). http://dx.doi.org/10.1016/j.ejsobi.2015.03.006 1164-5563/© 2015 Elsevier Masson SAS. All rights reserved.
methane in optimized landfill covers or biofilters has been listed as key mitigation technology for the reduction of methane fluxes from landfills that are no longer suitable for energy recovery or flaring [2]. Methane oxidation thus presents an “end-of-pipe” option both for older, non-sanitary sites but also for sanitary landfills after active gas extraction has ended but gas production continues for decades or even centuries on a lower level. As a result of the high source strength of methane, landfill soil covers harbor an abundant methanotrophic community, usually dominated by Methylosinus and Methylocystis species [3e6], classified as type II methanotrophs. Landfill cover soils show potential methane oxidation rates in the order of several hundred g CH4 m2 d1 [7,8] which, however, are subject to large spatial variability [9], resulting from the spatial variability of gas fluxes through the cover soil, caused by preferential flow paths in the soil. These are formed by differences in soil texture and compaction, through soil aggregation, rootage [10] or animal burrows, providing pathways of enhanced gas permeability and hence for preferential gas [11,12]
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
and water [13] flow. Thus, methane emissions are not spatially homogeneous but occur through small-scale areas, often referred to as emission hotspots [14], creating small-scale patterns of habitats with varying environmental conditions. Hotspots of gas emission can often be identified visually, for example by a change, damage or lack of vegetation, likely due to the shortage of oxygen in the root zone as a result of strong advective flow of gas from the underlying waste body [15]. Moreover, a distinct discoloration of the topsoil towards purple colors is frequently encountered at locations of increased gas emission (author's personal observation). In contrast, such discoloration was never observed at methanotrophically very active sites, harboring an abundant methanotrophic population but not showing high emissions [16]. It is hypothesized that the discoloration stems from the particular microbial community established as a result of the specific habitat posed by the enhanced gas fluxes. Characteristics of these habitats comprise, for example, excess availability of methane-derived carbon and of hydrogen sulfide (both components of landfill gas) and possible changes between anaerobic and aerobic conditions over very short distances on the micrometer to centimeter scale (soil pore system to secondary macropores such as cracks). However, to date, the microbial community of such discolored patches has never been investigated. The aim of this study was to elucidate possible differences between microbial community compositions from methane emitting hotspot and neighboring non-methane-emissive locations, consisting of the same original type of soil. The focus lay particularly on those parts of the community that were affected directly or indirectly by methane concentrations. 2. Materials and methods 2.1. Study site The investigated site is a 1.5 ha old municipal solid waste landfill located approximately 70 km southwest of the city of Hamburg in northwestern Germany that was operated between 1970 and 1982. The waste, a mixture of household (37 mass-%), construction and demolition (27 mass-%), industrial (10 mass-%), other inert waste (11 mass-%) and sludges (15 mass-%) [17], was disposed of in a former sand pit on top of which waste was further piled up to form a mound. Base or surface liners are missing so that landfill gas can migrate freely from the landfill body through the cover. The gas extraction system was operated only in the first few years after closure. Cover soil properties and cover thickness vary strongly and basically reflect whatever material was available to the landfill operator at the time of cover construction. Soil texture is mostly sandy, however, bulk density, pores size distribution, as well as soil chemical properties such as carbon and nitrogen content vary strongly (data given in Ref. [9]). The site is vegetated, mostly with grassland interspersed with shrubs, bushes and small trees. Previous analyses of the methanotrophic community using diagnostic microarray revealed a preponderance of type II methanotrophs [6] and a 1 methane oxidation rate of 0.3e19.5 mg CH4 g1 [9]. Assuming a dw h thickness of the methane oxidation horizon of 40 cm and a dry bulk density of the soil of * g cm3, these oxidation rates translate to 0.17 and 9.80 g CH4 m2 d1. Actual gas production was estimated at 5700 m3 CH4 per year, using the IPCC (Intergovernmental Panel on Climate Change) gas production model [17,18], yielding a hypothetical spatial load to the cover soil of 0.72 g CH4 m2 d1, assuming an even spatial distribution of this load to the landfill cover. However, due to the above mentioned preferential nature of gas flow, emissions escaped through hotspots with fluxes of up to 155 g CH4 d1, emanating from areas of a few square centimeters in
9
size [14]. 2.2. Sampling locations and sampling strategy Gas and soil samples were collected on the 26th of April, 2012 from two areas (1 and 2) at the south eastern slope of the landfill. On this slope, several hotspots of gas emission had been detected during a previous monitoring campaign [14]. The two areas were 3 m apart on the same height of the slope. Emissions in this area, monitored by means of static chambers, ranged between 2.7 and 62.5 g CH4 d1 with an average of 32.1 g CH4 d1 (n ¼ 31, measured biweekly to monthly from 2008 to 2010). The methodology of the chamber measurement is described in Ref. [14]. Within these two areas six sampling locations were chosen: two were hallmarked by distinct purple discoloration of the soil (1A and 2A), a lack of vegetation and very high surface methane concentrations. Within a maximum distance of 80 cm from 1A to 2A, two further sampling sites, namely 1B and 1C (near 1A) and 2B and 2C (near 2A), were selected, which did not exhibit this typical purple discoloration, bore fresh grass vegetation and showed no detectable surface methane concentration. Sites 1C and 2B exhibited intermediary features, showing lower levels of surface methane concentrations. For details of sampled soils and depths see Table 1. 2.3. Soil properties and gas composition Soils were analyzed for texture according to DIN ISO 11277 [19], organic carbon and total nitrogen according to DIN ISO 10694 [20] and DIN ISO 13878 [21] using a CHN analyser vario Max (Elementar Analysensysteme GmbH). Measurement error was 0.08% TOC and 0.01% TN. Moisture was determined by gravimetric analysis following oven-drying at 105 C. Composition of the soil gas phase was sampled using a gas probe that was inserted into 4, 10, 20, 30 and 40 cm below soil surface. Gas was collected by probing through the probe septum with a needle connected to a 60 ml syringe and analyzed on-site using a biogas analyzer (BM 2000, Geotechnical Instruments (UK) Ltd.; detection limit ¼ 0.1%). The concentration of N2 was calculated by subtracting the sum of the measured concentrations of CH4, CO2 and O2 from 100. Methane concentrations at the soil surface were analyzed using a portable flame ionization detector (FID; Sewerin GmbH, Gütersloh). The reading was recorded when the FID gave a stable value. The lower detection limit was 1 ppm, the upper detection limit 14,000 ppm of volatile organic compounds. 2.4. Microbial community analyses Following collection, the soil samples were frozen at 20 C immediately until microbial community analyses were performed. 2.4.1. DNA extraction, PCR, cloning, and sequencing Around 0.5 g soil from eight samples, i.e. 1A, 1B (consisting of parallels 1B1 and 1B2), 1C1, 1C2, 2A, 2B and 2C (for soil depths see Table 1) was used to extract DNA with the UltraClean Soil DNA Isolation Kit (MoBio, Solana, CA, USA) according to the manufacturer's instructions. Bacterial 16S rRNA genes were PCR-amplified in three parallels using the oligonucleotide primer sets consisting of 27F and 1492R [22]. All amplifications were performed in triplicate, products pooled and separated by gel electrophoresis. Purified PCR products were ligated into the pGEM®-T vector system (Promega, Madison, WI, USA) and transformed into competent Escherichia coli DH5a. PCR products of the correct size were partially sequenced using the primer 27F [ [22], sequence length ¼ 331 to 856 base pairs]. Sequencing was performed with
10
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
Table 1 Selected soil properties, surface and soil methane concentrations, total organic carbon (TOC), total nitrogen (TN) and moisture content of the collected soil samples. Sample ID
1A-1 1A-2 1B 1C-1 1C-2 2A 2B 2C
Depth (cm) for
Soil properties
Methane concentration
Microbiology analyses
Abiotic properties
Purple discoloration
EPS
Vegetation
Moisture (% dw)
TOC (%)
TN (%)
TOC/TN ()
At surface (ppm)
Avg. in soil (vol. %)
In 5 cm depth (vol. %)
0e3 e 0e7 0e3 4e7 0e2 0e2 0e2
0e3 8e10 0e7 0e3 4e7 0e2 0e2 0e2
Yes
Yes
No
0 400
0 12.1
0 0
Yes No No
Yes Yes No
No No Yes
15.7 13.0 11.5 13.8 16.6 16.8 14.5 13.7
22.3
Yes No
1.035 0.634 0.307 0.445 1.124 0.694 0.4 0.244
23.9
No Yes
16.3 8.2 3.5 6.1 18.6 11.7 5.8 3.3
>14,000
No No
53.6 32.8 30.4 26.8 61.7 18.1 35.8 27.9
>14,000 380 0
17.7 20.1 5.0
7.1 19.0 0
EPS ¼ extrapolymeric substance; TOC ¼ total organic carbon; TN ¼ total nitrogen; dw ¼ dry weight.
the 3730xl DNA Analyzer (Applied Biosystems) and sequences yielded between 401 and 870 nucleotides. Detailed procedures for amplification, amplicon purification, cloning and sequencing have been described [23]. 2.4.2. Phylogenetic and statistical analyses All sequences were edited and clustered to operational taxonomic units (OTUs, similarity 93%) with the Lasergene Software SeqMan (DNAStar Inc., Madison, WI, USA). For soil sample 1A 34 consensus sequences (93 single sequences), for soil sample 1B1 94 consensus sequences (144 single sequences), for soil sample 1B2 71 consensus sequences (128 single sequences), for soil sample 1C1 69 consensus sequences (130 single sequences), for soil sample 1C2 56 consensus sequences (127 single sequences), for soil sample 2A 42 consensus sequences (106 single sequences), for soil sample 2B 66 consensus sequences (132 single sequences) and for soil sample 2C 75 consensus sequences (129 single sequences) were generated. The 16S rRNA gene sequences were compiled using the ARB software (www.arb-home.de) [24]. Maximum-likelihood based trees and 100 bootstrap replicates were constructed using PhyML [25]. Trees were imported into ARB and shorter sequences were added subsequently to the trees without changing its topology. Procedures for editing, comparing, compiling and aligning sequences and constructing maximum-likelihood trees have been reported [23]. Distance matrices of the aligned single sequences from the eight soil samples were created by using DNADIST from the PHYLIP version 3.65 package (J. Felsenstein, University of Washington, Seattle). These distance matrices served as input to DOTUR [26] for clustering DNA sequences into OTUs (97%), calculating collector's curves and diversity indices (Simpson and Shannon diversity index). The evenness index, which describes the extent of the dominance of individual OTUs, was estimated according to Pielou (1966) [27].
exopolymeric substances (EPS). Contrastingly, the soil at sites 1B and 2C had a dark brown, humic appearance (Fig. 1, right), bore fresh grass vegetation, were moist at the soil surface and had no EPS. Here, no methane could be detected at the surface (Table 1). Features of intermediary nature were observed at sites 1C and 2B, lacking discoloration but also lacking vegetation, showing a dry soil surface and lower surface methane concentrations. However, here also EPS were observed below the soil surface. Sites 1A and 2A (high surface methane concentrations) showed some of the highest TOC and TN contents, elevated up to four-fold relative to the low-emissive, non-vegetated sites 1C and 2B and up to almost five-fold compared to the non-emissive, vegetated reference sites 1B and 2C (Table 1). At these reference sites, the organic carbon and nitrogen levels fell within the range of typical values for natural terrestrial humic topsoils in Northwestern Europe [30]. TOC and TN were highly correlated (R2 ¼ 0.988, P < 0.01). However, the ratio of TOC to TN also increased with elevated organic carbon content (R2 of linear regression ¼ 0.64, P < 0.05). 3.2. Soil gas composition At sites 1A and 2A (purple discoloration, high surface methane concentrations, no vegetation, presence of EPS; see Table 1) the soil gas phase was characterized by high concentrations of methane and carbon dioxide and low concentrations of oxygen up to the shallowest depth (Fig. 2, upper panel). The concentration of the landfill gas constituents remained almost unchanged from bottom to top. Also, at both sites, no or hardly any increase in the ratio of
2.4.3. Nucleotide sequence accession numbers The sequence data have been submitted to DDBJ/EMBL/GenBank databases under accession numbers KJ633283 e KJ633789. 3. Results 3.1. Soil properties and surface methane concentrations The top 10 cm of the soil at all sampled sites were of identical texture, classified as sandy loam [28]. At locations 1A and 2A a purple discoloration of the soil surface was observed (Table 1 and Fig. 1, left; Munsell value 5R 2.5/3) [29]. The soil also appeared to be dry, coinciding with high surface methane concentrations and a lack of vegetation (Table 1). Below the surface the soil was aggregated to form lumps of slimy consistency, likely to be caused by
Fig. 1. Discoloration of sample 1A (left) in comparison to sample 1C (right).
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
11
Fig. 2. Composition of the soil gas phase (soil gas profiles) from sampling sites 1A and 2A with high surface methane concentrations (upper panel), sites 1C and 2B with medium surface methane concentrations (middle panel) and sites 1B and 2C where no methane could be detected at the surface (lower panel).
carbon dioxide to methane was observed, indicating hardly any methanotrophic activity which is otherwise characterized by a strong increase in the ratio due to consumption of methane and production of carbon dioxide [31]. The top 5 cm at site 2C were better aerated, accompanied by a slight increase in the ratio of carbon dioxide to methane, indicating a low level of methanotrophic and/or respiratory activity. Site 2B was initially classified as a site with features of intermediary nature, lacking purple discoloration at the surface and lacking vegetation, showing a lower level of surface methane concentrations around 400 ppm but exhibiting presence of EPS. However, the soil gas profile was very similar to that at site 1A and 2A with high concentrations of methane and carbon dioxide and low concentrations of oxygen, which remained almost unchanged from bottom to top of the soil cover. Also at site 2B, the ratio of carbon dioxide to methane remained almost constant, indicating hardly any methanotrophic activity. Sites 1A, 2A and 2B also exhibited the highest average methane concentration in the soil gas phase and the highest methane concentration in 5 cm depth (Table 1). In contrast, at site 1B (no discoloration, no methane measured at
surface, fresh grass vegetation) no methane was found in the soil gas phase (Fig. 2, lower left panel). The soil gas profile showed a high extent of aeration of the soil, even in greater depths, indicated by both high oxygen and near-atmospheric nitrogen concentrations. Similar to site 1B, site 2C (also lacking purple discoloration, surface methane and EPS, and also bearing fresh grass vegetation) had a well-aerated topsoil with methane concentrations below the detection limit. However, different from site 1B, methane was found in the subsoil at concentrations up to 8 vol. % (Fig. 2, lower right panel). Although with respect to the presence of EPS, vegetation and surface methane concentrations some differences between 1C and 2C in soil properties were apparent, their soil gas profiles had some evident similarities (Fig. 2, middle left and lower right panel). Also at site 1C, a well-aerated topsoil and increased methane concentrations in the subsoil were found, with up to 24 vol. % at a higher level, though. 3.3. Community composition All soil samples had a high bacterial diversity (Table S1) as is commonly found for soil environments [32]. However, the
12
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
communities were not sampled for complete coverage of the bacterial community (Fig. S1). The communities of the soil samples consisted of bacteria across much of the bacterial domain (Fig. 3) including members of the Alphaproteobacteria including methanotrophs of type II (Fig. S2), Betaproteobacteria, Gammaproteobacteria including methanotrophs of type I, Planctomycetes, Nitrospira (Fig. S3), Bacteroidetes, Deltaproteobacteria, Verrucomicrobia, Acidobacteria, Firmicutes, Actinobacteria, Chloroflexi, and candidate divisions, e.g. TM6 or BRC1. All samples had relatively large but varying proportions of methanotrophic and non-methane utilizing methylotrophic bacteria (Fig. 3B) [33]. In samples 1A, 2A and 1C2, sequences clustering with Methylocella/Methylocapsa of the alphaproteobacterial type II methanotrophs predominated (Figs. 3B, S2). In these samples no or hardly any gammaproteobacterial type I methanotrophs were identified. Relatively even proportions of type I and type II methanotrophs were found in samples 1C1 and 2B. In samples 1B1, 1B2 and 2C type I methanotrophs prevailed over type II methanotrophs. Methylotrophs not capable of using methane were found in all samples readily (Fig. 3B). Among the more abundant lineages this included particularly Burkholderia-related organisms in sample 2A, 1C1 and 2B, Bradyrhizobium in sample 1A, 1C2 and 1B1 and Planctomycetes that
are suggested to be capable of methylotrophy [33,34]. 3.3.1. High surface methane concentration (sites 1A and 2A) The lowest bacterial diversity was in the soil samples for which the highest surface methane concentrations and the highest TOC and TN content were measured, namely 1A and 2A (Tables 1, S1). However, the two methane-rich soil samples appeared to host different proportions of methanotrophic bacteria (Figs. 3, S2 and S3). At site 1A, no type I methanotrophs were identified, but 9% of the 16S rRNA gene sequences were affiliated with type II methanotrophs. By contrast, at site 2A, 3% could be grouped with type I methanotrophs and 18% of the sequences were attributed to the type II methanotrophic bacteria. Members of purple non-sulfur bacteria were recognized in soil sites 1A and 2A and included Rhodosprillum and Roseomonas of the Rhodospirillales (Figs. 3, S2). Also, at site 2A a large proportion of Janthinobacterium of the Burkholderiales was found (Figs. 3, S3). 3.3.2. Intermediate surface methane concentration (sites 1C and 2B) Relative to site 1A and 2A, at site 1C and 2B the bacterial composition appeared to shift from a few dominant microbial groups to a broader diversity of different lineages, which did not dominate the overall community composition (Fig. 3). Methylococcales of the type I methanotrophs and Beijerinckiaceae and Methylocystaceae of the type II methanotrophs appeared most abundant in the soil sample 1C (11% and 21%, respectively, Figs. S3 and S4). In the soil sample 2B 4% type I and 8% type II methanotrophs were identified. Planctomycetes made up a relatively large proportion in the 1C and 2B soil bodies (8% of sequences). 3.3.3. No surface methane concentration (sites 1B and 2C) The highest bacterial diversities were noted for 1B1 and 2C where no methane was measured at the surface and in 5 cm depth and where TOC contents were the lowest (Tables 1 and S1). In the soil sample 1B, 12% and 3% of the sequences were affiliated to 16S rRNA genes of type I and type II methanotrophs, respectively (Figs. S2 and S3). The sample from site 2C comprised 7% and 1% type I and type II. In these soils Planctomycetes accounted for 12% and 8% of the 16S rRNA gene sequences at site 1B and 2C, respectively. 3.4. Linkages between abiotic factors and the microbial community
Fig. 3. Proportions of 16S rRNA genes for each site for selected microbial lineages. Details of relationships can be found in the phylogenetic trees in the Supplementary materials in Figs. S2 and S3.
Regression analyses between abiotic and biotic properties showed that in most cases the data points cluster quite strongly, for example, around high or low evenness or a high or low Simpson index (Fig. 4; for the entire results set of regression analyses see Supplementary material). Therefore, in spite of the fact that values for the coefficient of determination (R2) and likelihood of error (p) are given, the depicted relationships should be carefully interpreted as a qualitative indication of the relationship between the respective parameters. The average soil methane concentrations are negatively related with the Shannon diversity index and species richness (Table S2). Hence, the higher the methane concentrations in the soil were, the lower the diversity of the community seemed to be (Fig. 4). Besides correlating with the average methane concentrations, richness also correlated negatively with levels of TOC and TN, and with the TOC/TN ratio (Table S2). Thus, the higher the average methane soil concentrations were, the higher TOC and TN was and the lower species richness was at a given site (Fig. 4). The Simpson diversity index in turn appeared to be inversely related to the presence of type I methanotrophs and Planctomycetes and positively with the relative abundance of clone sequences affiliated with type II methanotrophs (Fig. 4, Table S2).
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
13
Fig. 4. Relationships between different abiotic and microbiological properties of the samples with distinct methane surface concentrations. The linkages of Simpson diversity index, the evenness index, OTUs, TOC, TN and average methane soil concentrations, type I and II methanotrophs and Planctomycetes are displayed. The lines mark the linear regression.
4. Discussion 4.1. Soil gas composition The observed gas profiles at sites 1A, 1B and 2B are typical for
hotspots of gas emission, showing hardly any changes in the gas composition from bottom to top and only little or no aeration [34]. These sites are usually characterized by advective (i.e. pressuredriven) gas transport, impeding the diffusive ingress of atmospheric oxygen into the soil. The vertical stratification of the soil gas
14
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
composition is usually stable over time, i.e. there are no seasonal differences as typical for sites with diffusion-driven gas transport [34]. Unfavorable conditions for methanotrophic activity are caused by restricted aeration in addition to an excessive methane load, usually too high for complete oxidation, even if the level of aeration would suffice. As a result, considerable methane fluxes emit from these sites [14]. In contrast, the vertical stratification of soil gas composition at sites 1C and 2C indicates mainly diffusive upward transport of landfill gas, allowing for ingress of atmospheric oxygen from the soil surface and thus for favorable conditions for methanotrophic activity in the upper part of the soil cover. Typical patterns, as also seen at sites 1C and 2C, comprise elevated concentrations of landfill gas components in the subsoil and their strong decrease towards the surface, while the concentration of atmospheric components (oxygen, nitrogen) increases. Methane oxidation is indicated by strongly increasing ratios of carbon dioxide to methane towards the soil surface [31]. Characteristically, these gas composition patterns are subject to pronounced seasonal variation, reflecting the change of parameters impacting methanotrophic activity, such as soil temperature and moisture [14,34]. As a result, emissions are usually observed under cooler and moister but not under warmer and dryer conditions. At the time of sampling, the top 40 cm at site 1B were fully aerated and not influenced by landfill gas at all. Elevated carbon dioxide concentrations were either due to methane oxidation at depths below 40 cm or resulted from soil respiratory activity or both.
the result of methanotrophic activity under conditions of excess carbon and oxygen deficiency [35,36], it is posited that the EPS at site 1A and 2A arises mainly from type II methanotrophic activity. The higher proportion of methanotrophs in general as well as the presence of type I methanotrophs in upper soil crust at site 2A may be related to the better aeration of the soil in the upper 5 cm at this site (Fig. 2, upper panel), due to the different oxygen demands discussed above [41e43]. Conditions towards considerably lower oxygen concentrations in the upper soils of site 1A are also supported by elevated amounts of 16S rRNA genes of the Rhodospirillales (Figs. 3, S2), which typically thrive under microaerophilic or anaerobic conditions [45]. Members of purple non-sulfur bacteria were recognized in soil sites 1A and 2A and included Rhodosprillum and Roseomonas of the Rhodospirillales (Figs. 3, S2). These organisms may be responsible for the purple discoloration observed at sites 1A and 2A (compare Fig. 1) [46,47]. Alternatively, at site 2A a large proportion of Janthinobacterium of the Burkholderiales was found (Figs. 3, S3), which may also contribute to the purple discoloration [48]. In methanotrophic enrichment cultures, redcolored heterotrophic, mostly nocardioformic microflora accompanying methanotropic enrichment cultures are frequently found. However, in this study no sequence associated with this group of bacteria was detected. These lineages are thus either absent in the sampled environments or, if present, in extremely low abundances. Hence, nocardioforme bacteria are unlikely to be responsible for the observed coloring in the samples. 4.4. Community composition at intermediate surface methane sites (1C, 2B)
4.2. Soil C and N Sites 1A, 1C, 2A and, to a lesser extent, 2B, were characterized by strongly elevated levels in organic carbon and nitrogen (Table 1). Since vegetation was missing at these locations, plant biomass cannot have contributed to the organic fraction of the soil. Rather, elevated nitrogen levels likely reflect a more prolific microbial biomass and elevated organic carbon contents may be attributed to a higher microbial biomass and to EPS. EPS consist mainly of polysaccharides; their occurrence has been interpreted as the result of methanotrophic metabolism under conditions of excess carbon, oxygen deficiency and nitrogen limitation [35,36]. Formation of EPS in relation to high methane fluxes have been observed in laboratory studies by a number of authors [37e40], however, so far it has not been described for field settings. The fact that the ratio of TOC to TN increased with increasing TOC contents (Table 1) indicates a change in the nature of the organic matter towards components containing more carbon and less nitrogen, as for example nitrogen-free polysaccharide compounds referred to as EPS. Indeed, EPS were observed in the soils collected from sites 1A and 2A as well as sites 1C and 2B (Table 1). 4.3. Community composition at high surface methane sites (1A, 2A) At sites 1 A and 2A, type II methanotrophs appear to have outcompeted type I methanotrophs, which is in line with the fact that type II methanotrophs have a competitive advantage under methane-rich and oxygen limited conditions hallmarking these sites [[41e43]; Fig. 1, upper panel]. Previous investigations in landfill cover soils and landfill biofilters have also revealed a dominance of type II methanotrophs [6,44]. The exceptionally high TOC/TN-ratios indicate a trend towards nitrogen-free organic matter like EPS, which was also detected in these soil samples (Table 1). Given that EPS was observed in these soils, that type II methanotrophs constitute a large proportion of the retrieved 16S rRNA genes and that the formation of EPS has been interpreted as
The higher numbers of clone sequences affiliated with methanotrophs at site 1C relative to site 2B are likely a reflection of better aerated soils above 10 cm at site 1C (Fig. 2, middle panel). High methane concentrations were found in the soil, in case of site 1C up to 40 cm depth and in case of size 2B up to at least 5 cm depth (similar to sites 1A and 2A discussed above). High methanotrophic activity at site 1C was indicated by a rapid increase in the ratio of carbon dioxide to methane above 20 cm depth. At site 2B, high methanotrophic activity can also be expected above 5 cm depth. Given the high soil methane concentrations and the methanotrophic activity evidenced by the change in the ratio of carbon dioxide to methane, it is assumed that similar to the sites with high surface methane concentrations discussed above EPS at sites 1C and 2B is also produced by methanotrophs in reaction to excess carbon supply [35,36]. Most of the characterized Planctomycetes isolates are slow growing, aerobic or facultative chemoheterotrophs specialized on carbohydrate metabolisms. Since some Planctomycetes have been shown to contain genes in the tetrahydromethanopterin-linked C1 transfer pathway, the potential for methylotrophy has been suggested [cf. [49]]. Some members of this deeply diverging lineage also appear to be capable of chemolithoautotrophic growth through anaerobic oxidation of ammonium [see Ref. [49]]. Conditions in these soils seem to be favorable for their successful colonization. 4.5. Community composition at no surface methane sites (1B, 2C) The highest bacterial diversities were noted for 1B1 and 2C where no methane was measured at the surface and in 5 cm depth and where TOC contents were the lowest (Table 1 and S1). These findings supported the trend that with decreasing methane concentrations the community changes from a few dominant groups towards a broader phylogenetic microbial community. In both samples, the methanotrophic community was dominated by type I
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16
15
organisms (Figs. 3, S2 and S3). Since both soils are rich in oxygen and poor in methane (Fig. 2, lower panel), these findings coincide with the known type I methanotrophs' ability to outcompete type II strains under oxygen-rich and methane-deficient conditions [41e43]. The lack of EPS in these soils in conjunction with very low type II methanotrophic sequences is supportive of our assumption of EPS formation at the other sites (discussed above) being related to type II methanotrophic activity. In the ‘low methane’ soils at sites 1B and 2C, Planctomycetes accounted for 12% and 8% of the 16S rRNA gene sequences, respectively, matching the prevailing aerobic conditions.
ecological driver posed by enhanced landfill gas fluxes at sites of preferential pathways, caused by the small-scale patterned gas permeability of the soil. The presence and activity of the sitespecific microbial community in turn significantly influences the soil properties, in the case of this study e.g. by increasing and changing the nature of the soil organic matter through formation of extrapolymeric substances. The changed microbial community at high methane flux sites is also likely to be responsible for the specific discoloration of the soil.
4.6. Linkages between soil properties and community composition
We would like to thank Annica Beuerbach for excellent technical assistance. The presented work was funded by the German Federal Ministry for Education and Research (BMBF), grant number 01 LS 05095.
The correlation analyses suggest that high average soil methane concentrations and corresponding low levels of soil aeration, caused by advective gas flux from the waste body, are accompanied by high TOC and TN levels in the topsoil. The higher TOC and TN levels are, the more type II methanotrophs and the less type I methanotrophs and Planctomycetes can be detected in the soils. Additionally, the bacterial diversity and evenness of species distribution appear reduced. This is suggestive of the development of a more specialized microbial community that appear to become dominated by a few lineages such as methanotrophs, particularly of type II methanotrophs, Burkholderiales, Rhodospirillales and Bradyrhizobiaceae with enhanced gas fluxes. Under specific conditions methanotrophs may accumulate methanol outside of the cell [50] which in turn may be utilized by methylotrophs such a members of the Bradyrhizobiaceae. Members of the latter three may be responsible for the distinct purple discoloration of the soil observed at the emissive sites. This in turn suggests that preferential pathways of gas act as a strong driver for the development of more specialized, less diverse communities. In a study examining soil properties and microbial community composition along preferential pathways of water flow in forest soil, an increase in organic carbon and nitrogen contents in the soil along the preferential flow paths was also found, alongside with elevated microbial biomass compared to the bulk soil [51]. Also, a different community structure was detected for some bacterial groups, however, for others not. The authors concluded that only few communities with the ability to thrive under aerobic and anaerobic conditions can compete at sites of preferential flow paths. Within the methanotrophs, it is the type II organisms that are better suited to survive adverse (anaerobic) conditions resulting in restricted growth rates. These have thus been reported to form communities that are more stable over time, compared to the faster growing type I methanotrophs that are more adapted to react to changing environmental conditions and that show more variability in community composition [43,52,53]. Examples for such adverse conditions are low or zero oxygen levels even near the soil surface under conditions of strong advective landfill gas flux, as is typical for preferential pathways of gas flux (see Fig. 2 and [34]). It thus appears plausible that at the sites with elevated gas fluxes (1A, 2A) the community is dominated by type II methanotrophs. 5. Conclusions The scope of this study was to identify the impact that landfill gas fluxes along preferential pathways (hotspots) in landfill soil covers have on soil properties and the respective microbial community. Soil properties and the microbial community composition are posited to be a result from the specific habitat created by the differing magnitude of landfill gas fluxes. The study demonstrates how the composition of the microbial community shifts towards a less diverse, more specialized population as a result from the strong
Acknowledgments
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ejsobi.2015.03.006. References [1] European Environment Agency (EEA), Technical Report No.2/2011. Annual European Union Greenhouse Gas Inventory 1990e2009 and Inventory Report, Submission to the UNFCCC Secretariat, 27 May 2011. [2] J. Bogner, M. Abdelrafie Ahmed, C. Diaz, A. Faaij, Q. Gao, S. Hashimoto, K. Mareckova, R. Pipatti, T. Zhang, Waste management, in: B. Metz, O.R. Davidson, P.R. Bosch, R. Dave, L.A. Meyer (Eds.), Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2007. [3] N. Stralis-Pavese, A. Sessitsch, A. Weilharter, T. Reichenauer, J. Riesing, J. Csontos, J.C. Murrell, L. Bodrossy, Optimization of diagnostic microarray for application in analysing landfill methanotroph communities under different plant covers, Environ. Microbiol. 6 (2004) 347e363. [4] N. Stralis-Pavese, L. Bodrossy, T.G. Reichenauer, A. Weilharter, A. Sessitsch, 16 S rRNA based T-RFLP analysis of methane oxidising bacteria e assessment, critical evaluation of methodology performance and application for landfill site cover soils, Appl. Soil Ecol. 31 (2006) 251e266. ry, A.C. Singer, D. Kumaresan, L. Bodrossy, N. Stralis-Pavese, J.I. Prosser, [5] M. He I.P. Thompson, J.C. Murrell, Effect of earthworms on the community structure of active methanotrophic bacteria in a landfill cover soil, ISME J. 2 (2008) 92e104. [6] J. Gebert, B.K. Singh, Y. Pan, L. Bodrossy, Activity and structure of methanotrophic communities in landfill cover soils, Environ. Microbiol. Rep. 1 (2009) 414e423. [7] C. Scheutz, P. Kjeldsen, J. Bogner, A. De Visscher, J. Gebert, H. Hilger, M. HuberHumer, K. Spokas, Processes and technologies for mitigation of landfill gas emissions by microbial methane oxidation, Waste Manag. Res. 27 (2009) 409e455. [8] J. Streese-Kleeberg, I. Rachor, J. Gebert, R. Stegmann, Field quantification of methane oxidation in landfill cover soils by means of gas push-pull tests, Waste Manag. 31 (2011) 995e1001. € wer, C. Geck, J. Gebert, E.-M. Pfeiffer, Spatial variability of soil gas [9] I.U. Ro concentration and methane oxidation capacity in landfill covers, Waste Manag. 31 (2011) 926e934. [10] S. Bohn, P. Brunke, J. Gebert, J. Jager, Influence of vegetation on microbial methane oxidation e a column study, Waste Manag. 31 (2011) 854e863. [11] A. Tuli, J.W. Hopmans, D.E. Rolston, P. Moldrup, Comparison of air water permeability between disturbed and undisturbed soils, Soil Sci. Soc. Am. J. 5 (2005) 1361e1371. [12] S.E. Allaire, J.A. Lafond, A.R. Cabral, S.F. Lange, Measurement of gas diffusion through soils: comparison of laboratory methods, J. Environ. Monit. 10 (2008) 1326e1336. [13] S. Gebhardt, H. Fleige, R. Horn, Effect of compaction on pore functions of soils in a Saalean moraine landscape in North Germany, J. Plant Nutr. Soil Sci. 172 (2009) 688e695. € ngro €ft, E.-M. Pfeiffer, Variability of methane emis[14] I. Rachor, J. Gebert, A. Gro sions from an old landfill on different time scales, Eur. J. Soil Sci. 64 (2013) 16e26. €ngro €ft, E.-M. Pfeiffer, Temporal variability of soil gas [15] J. Gebert, I. Rachor, A. Gro composition in landfill covers, Waste Manag. 31 (2011) 935e945. €ngro € ft, G. Miehlich, Kinetics of microbial landfill methane [16] J. Gebert, A. Gro oxidation in biofilters, Waste Manag. 23 (2003) 609e619. [17] A.M. Yemaneh, Assessment of Biological Waste Stability and Estimation of
16
[18]
[19]
[20]
[21]
[22]
[23]
[24] [25] [26]
[27] [28]
[29] [30]
[31]
[32]
[33] [34] [35]
J. Gebert, M. Perner / European Journal of Soil Biology 69 (2015) 8e16 Methane Formation Potential in Old Landfills e Case of Three Landfills Near the City of Hamburg (Master thesis), Hamburg University of Technology, Institute for Environmental Technologies and Energy Industry, 2010. IPCC (Intergovernmental Panel on Climate change), in: S. Eggleston, K. Buendia, T. Ngara, K. Tanabe (Eds.), 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Waste, vol. 5, 2006. http://www.ipcc-nggip.iges. or.jp/public/2006gl/vol5.html (accessed 01.02.14.). DIN ISO 11277, Bodenbeschaffenheit. Bestimmung der Parti€ßenverteilung in Mineralbo €den. Verfahren mittels Siebung und Sedikelgro mentation, Deutsches Institut für Normung e.V. Beuth Verlag, Berlin, 2002. DIN ISO 10694, Bodenbeschaffenheit: Bestimmung von organischem Kohlenstoff und Gesamtkohlenstoff nach trockener Verbrennung (Elementaranalyse), Deutsches Institut für Normung e.V. Beuth Verlag, Berlin, 1996. DIN ISO 13878, Bodenbeschaffenheit: Bestimmung des Gesamt-Stickstoffs durch trockene Verbrennung (Elementaranalyse), Deutsches Institut für Normung e.V. Beuth Verlag, Berlin, 1998. D.J. Lane, 16S/23S rRNA sequencing, in: E. Stackebrandt, M. Goodfellow (Eds.), Nucleic Acid Techniques in Bacterial Systematics, John Wiley & Sons, Chichester, England, 1991, pp. 115e175. €nberg, M. Perner, W. Bach, M. Hentscher, A. Koschinsky, D. Garbe-Scho W.R. Streit, H. Strauss, Short-term microbial and physico-chemical variability in low-temperature hydrothermal fluids near 5 S on the Mid-Atlantic Ridge, Environ. Microbiol. 11 (2009) 2526e2541. W. Ludwig, O. Strunk, R. Westram, ARB: a software environment for sequence data, Nucleic Acids Res. 32 (2004) 1363e1371. S. Guindon, O. Gascuel, A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood, Syst. Biol. 52 (2003) 696e704. P.D. Schloss, J. Handelsman, Introducing DOTUR, a computer program for defining operational taxonomic units and estimating species richness, Appl. Environ. Microbiol. 71 (2005) 1501e1506. E.C. Pielou, The measurment of diversity in different types of biological collections, J. Theor. Biol. 13 (1966) 131e144. FAO e ISRIC & ISSS (Ed.), World Reference Base for Soil Resources 2006: a Framework for International Classification, Correlation and Communication, 2006. World soil resources reports 103. Rome. Munsell Color, Munsell Soil-color Charts, Revised and produced in 2009, Michigan, USA, 2009. R.J.A. Jones, R. Hiederer, E. Rusco, P.J. Loveland, L. Montanarella, The Map of Organic Carbon in Topsoils in Europe, Version 1.2, September 2003: Explanation of Special Publication Ispra 2004 No.72 (S.P.I.04.72), European Soil Bureau Research Report No.17, Office for Official Publications of the European Communities, Luxembourg, 2004. EUR 21209 EN, 26 pp. and 1 map in ISO B1 format. €wer, H. Scharff, C.D.L. Roncato, A.R. Cabral, Can soil gas profiles J. Gebert, I.U. Ro be used to assess microbial CH4 oxidation in landfill covers? Waste Manag. 31 (2011) 987e994. J. Gans, M. Wolinsky, J. Dunbar, Computational improvements reveal great bacterial diversity and high metal toxicity in soil, Science 309 (2005) 1387e1390. S. Kolb, Aerobic methanol-oxidizing bacteria in soil, FEMS Microbiol. Lett. 300 (2009) 1e10. €ngro €ft, E.-M. Pfeiffer, Temporal variability of soil gas J. Gebert, I. Rachor, A. Gro composition in landfill covers, Waste Manag. 31 (2011) 935e945. J.D. Linton, J.P. Watts, R.M. Austin, D.E. Haugh, H.G.D. Neikus, The energetics and kinetics of extracellular polysaccharide production from methanol by
[36]
[37]
[38] [39]
[40]
[41]
[42] [43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51] [52] [53]
microorganisms possessing different pathways of C1 assimilation, J. Gen. Microbiol. 132 (1986) 779e788. W. Babel, Peculiarities of methylotrophs concerning overflow metabolism, especially the synthesis of polyhydroxyalkanoates, FEMS Microbiol. Rev. 103 (1992) 141e148. H.A. Hilger, D.F. Cranford, M.A. Barlaz, Methane oxidation and microbial exopolymer production in landfill cover soil, Soil Biol. Biochem. 32 (2000) 457e467. J. Streese, R. Stegmann, Microbial oxidation of CH4 from old landfills in biofilters, Waste Manag. 23 (2003) 573e580. J.H. Wilshusen, J.P.A. Hettiaratchi, V.B. Stein, Long-term behaviour of passively aerated compost methanotrophic biofilter columns, Waste Manag. 24 (2004) 643e653. R. Haubrichs, R. Widmann, Evaluation of aerated biofilter systems for microbial methane oxidation of poor landfill gas, Waste Manag. 26 (2006) 408e416. D.W. Graham, J.A. Chaudhary, R.S. Hanson, R.G. Arnold, Factors affecting competition between type I and type II methanotrophs in two-organism, continuous-flow reactors, Microb. Ecol. 25 (1993) 1e17. J.A. Amaral, R. Knowles, Growth of methanotrophs in methane and oxygen counter gradients, FEMS Microbiol. Lett. 126 (1995) 215e220. T. Henckel, P. Roslev, R. Conrad, Effects of O2 and CH4 on presence and activity of the indigenous methanotrophic community in rice field soil, Environ. Microbiol. 2 (2000) 666e679. J. Gebert, N. Stralis-Pavese, M. Alawi, L. Bodrossy, Analysis of methanotrophic communities in landfill biofilters by means of diagnostic microarray, Environ. Microbiol. 10 (2008) 1175e1188. G.M. Garrity, D.J. Brenner, N.R. Krieg, J.T. Staley (Eds.), Bergey's Manual of Systematic Bacteriology, The Proteobacteria, Part C: the Alpha-, Beta-, Delta-, and Epsilonproteobacteria, vol. Two, Springer, New York, New York (, 2005, ISBN 978-0-387-24145-6, pp. 354e361. S. Hou, K.S. Makarova, J.H. Saw, Complete genome sequence of the extremely acidophilic methanotroph isolate V4, Methylacidiphilum infernorum, a representative of the bacterial phylum Verrucomicrobia, Biol. Direct 3 (2008) 26. J.H. Saw, B.W. Mountain, L. Feng, Encapsulated in silica: genome, proteome and physiology of the thermophilic bacterium Anoxybacillus flavithermus WK1, Genome Biol. 9 (2008) R161. H.J. Tripp, J.B. Kitner, M.S. Schwalbach, J.W. Dacey, L.J. Wilhelm, S.J. Giovannoni, SAR11 marine bacteria require exogenous reduced sulphur for growth, Nature 452 (2008) 741e744. D.H. Buckley, V. Huangyutitham, T.A. Nelson, A. Rumberger, J.E. Thies, Diversity of Planctomycetes in soil in relation to soil history and environmental heterogeneity, Appl. Environ. Microbiol. 72 (2006) 4522e4531. J. Xin, J. Cui, J. Niu, S. Hua, C. Xia, S. Li, L. Zhu, Production of methanol from methane by methanotrophic bacteria, Biocatal. Biotransformation 22 (2004) 225e229. M. Bundt, F. Widmer, M. Pesarob, J. Zeyer, P. Blaser, Preferential flow paths: biological ‘hot spots’ in soils, Soil Biol. Biochem. 33 (2001) 729e738. R.S. Hanson, T.E. Hanson, Methanotrophic bacteria, FEMS Microbiol. Rev. 60 (1996) 439e471. M.S. Vecherskaya, V.F. Galchenko, E.N. Sokolova, V.A. Samarkin, Activity and species composition of aerobic methanotrophic communities in tundra soils, Curr. Microbiol. 27 (1993) 181e184.