| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Danish Institute of Agricultural Sciences, P.O. Box 50, DK-8830 Tjele, Denmark
b Institute of Agricultural, Environmental and Energy Engineering, University of Agricultural Sciences, A-1180 Vienna, Austria
c GSF-Research Center for Environment and Health GmbH, Institute of Soil Ecology, D-85764 Neuherberg, Germany
* Corresponding author (soren.o.petersen{at}agrsci.dk)
Received for publication September 6, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: CFU, colony-forming units • DCS, digested cattle slurry stored with (DCSC) or without (DCS0) cover • EC, electrical conductivity • OM, organic matter • UCS, untreated cattle slurry stored with (UCSC) or without (UCS0) cover
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Some studies have indicated that a surface crust reduces CH4 emissions from stored slurry (Husted, 1994; Sommer et al., 2000). We hypothesized that the surface crust could act as a sink for CH4 via bacterial oxidation. Methane oxidizing bacteria (methanotrophs) are found in a diverse range of environments including, for example, rice fields (Henckel et al., 2000), sediments (King, 1990), and landfill cover soils (Whalen et al., 1990). They are classified in two phylogenetic groups referred to as Type I and Type II, while a third group (Type X) shares metabolic features with both of the other groups (Hanson and Hanson, 1996). There are significant differences in the growth requirements and cell yields of Type I and Type II methanotrophs, but both types may coexist in heterogeneous or fluctuating environments (Henckel et al., 2000; Macalady et al., 2002). The process of CH4 oxidation is very similar to autotrophic NH3 oxidation, and NH3 serves as a competitive inhibitor that can reduce CH4 oxidation activity, possibly via the NO–2 that is produced (King and Schnell, 1994). Conversely, NH3 oxidizing bacteria can oxidize CH4. However, the specific CH4 oxidation rate by NH3 oxidizers is <5% of the CH4 oxidation rate by methanotrophs, and the affinity for CH4 is much lower (Hanson and Hanson, 1996).
Methane oxidation is not likely to occur in the hyper-osmotic and oxygen-depleted environment of the bulk slurry phase (Schnell and King, 1996; Welsh, 2000), but in a porous surface crust there could well be diffusional constraints on solute mobility combined with access to O2 from the atmosphere, which would allow CH4 oxidation to occur. The objective of this study was to determine whether methanotrophic bacteria inhabit the environment of slurry storage surface crusts, and if so, to consider the importance of CH4 oxidation for greenhouse gas mitigation under practical storage conditions. Surface crust materials were taken from a pilot-scale experiment where untreated and anaerobically digested cattle slurry had been stored with or without a solid cover.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Sampling
Samples of surface crust material (n = 3) were taken from an approximately 1-m2 area in each of the four storage treatments. The UCS surface crust was 6 to 10 cm thick with a loose and ill-defined lower boundary; blocks (approximately 15 x 15 cm) were excavated without disturbing the vertical stratification. The straw layer of DCS was generally loose and unstructured, and only grab samples were obtained. All samples were immediately frozen at –20°C until further processing. Following storage, the natural surface crust samples were sectioned into the following depth intervals without thawing: 0 to 2, 2 to 4, and 4 to 6 cm, while the remaining part was discarded as it was considered to be qualitatively identical to the 4- to 6-cm layer. The frozen straw material from DCS storages was cut to <1-cm pieces. This subsampling strategy is illustrated in Fig. 1
. Samples from each treatment and depth interval were thawed and mixed before subsampling for characterization and incubation experiments.
|
Incubation Experiment with Extracts of Surface Crust Material
To establish if there was a potential for CH4 oxidation in the different storage treatments, extracts of surface crust material were prepared as described above for the determination of CFU numbers, except that 5-g subsamples were extracted. This was done to achieve cell densities in 20 mL of extract corresponding to those obtained with a 100-mL extraction volume in the standard procedure described above. From UCS treatments, only the 2- to 4-cm depth interval was used, since a preliminary incubation experiment had shown similar kinetics with all depth intervals (data not shown). Volumes of extract corresponding to 106 to 107 CFU were transferred to 120-mL serum bottles and then supplemented with salt medium to a total volume of 20 mL. All bottles were closed with bromobutyl rubber stoppers and crimp caps. A control with sterile salt medium was included to check for leakage losses. Methane was injected and mixed with the headspace by repeated pumping with the syringe before the pressure was released via a needle; the final concentration of CH4 was 5.6%. The bottles were incubated on a rotary shaker (80 revolutions min–1) in the dark at 22 ± 1°C. Concentrations of CH4 and O2 were recorded after 0, 0.2, 0.9, 2, 3, 6, 7, 8, 9, 10, 13, and 15 d using a Model 3400 GC (Varian, Palo Alto, CA) equipped with a molecular sieve 5A column (60/80, 35°C) and thermal conductivity detector (150°C). The carrier gas was He at 45 mL min–1.
Incubation Experiment with Surface Crust Material
Having established the presence of CH4 oxidizing bacteria, a second incubation experiment was conducted with surface crust material to evaluate the potential for methanotrophic activity under field-moist conditions, and to examine the effect of fluctuating moisture content. Ten-gram subsamples of the 12 surface crust samples described above were transferred to 300-mL wide-mouthed flasks equipped with septa and screw caps. Then each headspace was amended with 100 µL CH4, corresponding to a final concentration of approximately 350 µL L–1 CH4. The samples were incubated in the dark at 22 ± 1°C. Headspace concentrations of CH4 were determined daily until linear CH4 removal rates were observed; O2 was measured every second day in selected samples to ensure that O2 was not becoming depleted. Then the flasks were opened, weighed, and left to dry in a fume hood at low flow for 72 h. By the end of this period, the flasks were weighed again and the moisture level calculated using separate determinations of moisture content (Table 1). The moisture content after drying corresponded to 72 ± 1.6% (mean ± standard error) of that in the fresh material. After flushing the headspace with compressed air, fresh CH4 was added and the incubation resumed. Again headspace CH4 concentrations were monitored until removal rates were linear. Finally, the samples were vented, rewetted to the original moisture contents, and incubated with fresh CH4, and CH4 removal rates were determined again. In this experiment, CH4 was analyzed using a Model GC-14 (Shimadzu, Kyoto, Japan) equipped with a Porapak Q column (50°C) and flame ionization detector (150°C). The carrier gas was He at a flow rate of 60 mL min–1.
|
Statistical Analyses
Surface crust characteristics were compared between subsamples by a general linear model using SAS 8.1 (SAS Institute, 2000); normal distribution of the different variables was first evaluated by the method of Shapiro and Wilk (1965). Statistical differences were determined for each variable using a Tukey–Kramer adjustment for multiple comparisons. In the second incubation experiment CH4 removal was, once started, approximately linear despite the relatively low headspace concentrations, and CH4 oxidation rates were therefore estimated using linear regression.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
o = –ECe(
s/)0.036, where ECe is the electrical conductivity of a saturated extract (dS m–1), s and are the volumetric water contents of the saturated extract and the fresh material, respectively, and 0.036 is an empirical conversion factor (MPa dS–1 m–1) (Rawlins and Campbell, 1986). Assuming this expression is valid for the surface crust slurries prepared for EC measurement, the level of EC in UCS subsamples corresponded to around –0.2 MPa, the EC of DCS0 to –0.35 MPa, and the EC of DCSC to –0.7 MPa. Treatments effects on mineral N were very similar to the effects observed for EC, with the highest concentrations recorded in the DCSC treatment for total ammoniacal N, NO–2, and NO–3 (Table 1). The presence of NO–2 and NO–3 provided evidence for nitrification activity in the surface crust materials, although in DCSC, NH+4 and NO–2 oxidation activity was clearly unbalanced, as revealed by the accumulation of NO–2. The differences in total N concentrations between treatments were less pronounced, with the highest concentrations being observed in DCSC and UCS0 (0- to 2-cm depth).
The CFU numbers decreased 10-fold with depth in UCS0, but not in UCSC (Table 1). These measurements were not replicated and can therefore not support any conclusions about the conditions for aerobic microbial activity in the different layers. However, in view of the lower moisture content at the 0- to 2-cm depth in UCS0, a larger air-filled pore space and thus O2 availability in the upper parts of UCS0 can be assumed.
Incubation of Surface Crust Extracts
Figure 2
shows concentrations of CH4 (top) and O2 (bottom) during incubation of extracts of surface crust material. Following a lag phase of 4 d (UCS) or 7 to 10 d (DCS), headspace CH4 was consumed in all treatments. Methane oxidation was approximately linear, once initiated. Oxygen exhibited the same temporal trends and distribution among treatments as CH4. There was a strong relationship (r2 = 0.93, P < 0.001) between headspace CH4 and O2 concentrations. During the linear phase of CH4 oxidation, O2 consumption ranged from 1.1 to 1.8 mol O2 mol–1 CH4. Methane oxidation proceeded without a lag phase after venting and addition of new CH4 (data not shown).
|
|
Incubation of Autoclaved Surface Crust Material
During the 14-d incubation of autoclaved samples there was no indication of CH4 oxidation. After 14 d, the recovery of CH4 was 95 to 98% in the six samples. In comparison, up to 90% of the added CH4 was removed during incubation of fresh material from the same samples (data not shown).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
In the present study, extracts of surface crust material from all four storage treatments showed methanotrophic activity following lag phases of 4 to 10 d in the different treatments (Fig. 2). It is not clear whether the lag phase reflected a need to overcome disturbances associated with sampling and mixing of the surface crust materials, such as an increase in NO–2 availability (see below), or if it was due to methanotrophs being inactive; many of the organisms extracted for this assay may not have been exposed to combinations of CH4 and O2 that would support the process at the time of sampling (King, 1992).
Methanotrophic activity in field-moist surface crust materials was stimulated by partial drying (Fig. 3), in accordance with previous findings on the importance of air-filled pore space (Schnell and King, 1996). Rewetting of the materials reduced net methanotrophic activity in some treatments only, but this response may have been obscured by bacterial growth during incubation, or by aeration of hitherto anaerobic microsites in the surface crust that were initially a source of CH4. The CH4 uptake rates of up to 4.2 µmol CH4 m–3 s–1 were comparable with, or higher, than the activities of 0.48 to 0.91 µmol CH4 m–3 s–1 quoted by Segers (1998) for wetlands and rice field soil. In these systems, up to 90% of the CH4 produced at greater depths can be reoxidized, which indicates that reoxidation of CH4 could also be significant in slurry storages.
The DCSC treatment had the lowest CH4 oxidation activity during incubation of fresh material (Fig. 3), and was characterized by an osmotic potential of –0.7 MPa. Polonenko et al. (1986) observed a general decrease in viability of soil microorganisms when exposed to –0.5 MPa. Methane oxidation is also adversely affected by water potential; a 50% reduction of methanotrophic activity was observed when the water potential of a Methylosinos trichosporium culture was reduced from around –0.2 to –0.7 MPa by addition of NaCl or sucrose (Schnell and King, 1996). Hence, osmotic stress may have inhibited CH4 oxidation more in DCSC than in the other surface crust materials.
The comparatively low methanotrophic activity in DCSC coincided with the highest level of mineral N (see Table 1). Mineral N has been shown to inhibit CH4 oxidation to varying extents, depending on substrate availabilities and habitat (Hütsch, 1998; Kravchenko, 2002; Wang and Ineson, 2003). For NH+4 and NO–3, effects appear at relatively high concentrations and are probably related to a nonspecific salt effect (osmotic stress), whereas inhibition by NO–2 is strong (but reversible) even at low concentrations (Whalen, 2000). The concentration of NO–2 in the treatment DCSC corresponded to approximately 100 mM in the aqueous phase and would, if evenly distributed, probably have resulted in a complete inhibition of CH4 oxidation (Whalen, 2000). As it were, the heterogeneous surface crust material probably contained niches where mineral N availability was compatible with CH4 oxidation, although some level of inhibition due to mineral N cannot be excluded.
At present no information is available about the organisms responsible for methanotrophic activity in surface crusts. The accumulation of NO–2 and NO–3 was greatest in the treatment DCSC, which had the lowest CH4 oxidation activity, indicating that CH4 oxidation by NH3 oxidizing bacteria was not important. The Type II methanotrophs M. trichosporium and Methylocystis spp. have been isolated from livestock manure (Heyer et al., 2002), and a phospholipid fatty acid biomarker specific for Type II methanotrophs has been isolated from a cattle manure pile (A. Gattinger, personal communication, 2004). Methylosinos trichosporium has been shown to survive well under prolonged anaerobic starvation, and to recover rapidly when exposed to conditions suitable for growth (Reichardt et al., 1997; Roslev and King, 1994, 1995). It is therefore likely that the CH4 oxidation observed in this study was due to methanotrophs originating from the slurry.
Methanotrophic activity depends not only on (micro)climatic conditions, but also on the availability of substrates. Based on theoretical considerations and experiments with Methylocystis sp. and Pseudomonas sp., van Bodegom et al. (2001) concluded that methanotrophs will be out-competed by heterotrophic activity except under micro-aerophilic conditions and with a shortage of carbon substrates. This implies that methanotrophic activity in slurry storages will be restricted by transport of O2 into the matrix of the surface crust from above, but also by the supply of acetate and other substrates for heterotrophic growth from the slurry below. The potential inhibition of methanotrophs by mineral N species and hyper-osmotic conditions was already mentioned. Diffusional constraints may be less for NH3, which can be transported in the gas phase, a potential advantage for methanotrophs, which could otherwise become N limited (Bodelier and Laanbroek, 2004).
Few studies have examined the effect of slurry storage coverage on CH4 emissions. Sommer et al. (2000) found that CH4 emissions from experimental cattle slurry storages with a natural surface crust, or with an artificial crust of leca pebbles or straw, were on average 38% less than emissions from uncovered slurry. Husted (1994) conducted a seasonal study of CH4 emissions from full-scale storages of pig and cattle slurry and found a significant interaction between crust formation and CH4 emissions, especially at lower temperatures. The data indicated a reduction of around 90% for both slurry types.
The effect of a solid cover was quantified in the storage experiment from which surface crust material was sampled for the present study; here, a 15% reduction of CH4 emissions from both untreated and digested cattle slurry was obtained by addition of a simple wooden lid (Clemens et al., 2005). Two different mechanisms can be proposed to explain the effect of a solid cover on methanotrophic activity. Any type of cover may stimulate CH4 oxidation by reducing moisture fluctuations due to wetting and drying, since this will enable populations of methanotrophic bacteria to develop in zones where substrate concentrations support the process. A solid cover could further stimulate CH4 oxidation by increasing the retention time and thus steady state concentration of CH4 in the air above the surface crust. The apparent half-saturation constant (Kapp) of CH4 is typically in the range 2 to 10 µM (King, 1992), corresponding to gas phase concentrations of 1500 to 7000 µL L–1 CH4 (Wiesenburg and Guinasso, 1979), that is, far above the atmospheric level of 1.8 µL L–1. It suggests that an increase in CH4 concentration above the surface crust will enhance methanotrophic activity. Williams and Nigro (1997) conducted a laboratory study in which the air exchange rate above stored cattle and pig slurry was varied between 0.5 and 0.0025 m s–1 to simulate increasing cover tightness. The release of CH4 was reduced by up to 90% at the lowest flow rate, which the authors explained by entrapment of gas bubbles by particles in the slurry. However, recent data from experiments with surface crust materials exposed to a range of headspace CH4 concentrations did show a stimulation of CH4 oxidation with increasing CH4 availability (S.O. Petersen, unpublished data, 2004).
Methane oxidation in slurry storage surface crusts could represent a cost-effective greenhouse gas mitigation option. The extra labor costs for establishment and maintainance of a surface crust are moderate, although a surface crust is obviously less practicable with lagoons as compared with systems handling the manure in more concentrated form. If a crust does not form naturally, straw is available to most farmers. Disturbance of the crust during the regular transfer of slurry from livestock buildings can be minimized by inserting the nozzle below the surface crust, and homogenization before emptying can be achieved by mixing with or without destroying the surface crust. In 2001, 85 and 96%, respectively, of pig and cattle slurry storages in Denmark had >80% coverage (Anonymous, 2003), almost exclusively in the form of a surface crust. These percentages are still increasing due to recently imposed regulations intended to mitigate NH3 volatilization. Hence, practical problems associated with handling of manure in storages with a surface crust can be overcome.
A preliminary assessment of slurry storage coverage as a greenhouse gas mitigation option was recently made that assumed a reduction in CH4 emissions of 20% with a surface crust, and a reduction of 25% if a solid cover was included (Hansen et al., 2004). The added N fertilizer value of the slurry due to a reduction of NH3 volatilization was also accounted for. This assessment indicated that the costs of establishing a surface crust would to be $1.2 Mg–1 carbon equivalents, while the costs of combining a surface crust with a solid cover would be around $5.3 Mg–1 carbon equivalents. These estimates based on a limited set of experimental data thus indicate that greenhouse gas mitigation via CH4 oxidation is cost effective (Hyman et al., 2002; Schneider, 2002). It should be stressed, however, that CH4 fluxes must be evaluated concurrently with N2O and NH3 fluxes to produce an overall assessment of the greenhouse gas balance as determined by storage conditions.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in JEQ:
This article has been cited by other articles:
|
|
 |
|
  T. L. Hansen, S. G. Sommer, S. Gabriel, and T. H. Christensen Methane Production during Storage of Anaerobically Digested Municipal Organic Waste J. Environ. Qual., April 3, 2006; 35(3): 830 - 836. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal |
