SEARCH

SEARCH BY CITATION

Keywords:

  • CH4 (methane) emissions;
  • CH4 transport;
  • diffusion limitation;
  • experimental manipulation;
  • greenhouse gases;
  • Juncus effusus (soft rush);
  • morphological barriers;
  • wetlands

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Aerenchymatous plants can transport methane (CH4) from the root zone to the atmosphere, bypassing the surface-oxidizing layers of the soil, yet morphological and anatomical factors that govern the transport of methane have rarely been critically tested in manipulative experiments.
  • Here, we investigated the methane transport capacity of hydroponically grown Juncus effusus, in experiments with roots submerged in nutrient solutions sparged with methane (1.16 mmol CH4 l−1). Through a range of manipulations of the above- and below-ground plant parts, we tested the contradictory claims in the literature regarding which sites provide the greatest resistance to gas transport.
  • Root manipulations had the greatest effect on methane transport. Removing root material reduced methane transport significantly, and especially the lateral roots and the root tips were important. Cutting of the shoots, with or without subsequent sealing, did not alter methane transport significantly.
  • We confirm modelling predictions that the limiting factor for methane transport in the tussock forming wetland graminoid, J. effusus, is the amount of permeable root surface, estimated using the proxy measurement of root length. The aerial tissues do not provide any significant resistance to methane transport, and the methane is emitted from the lower 50 mm of the shoots.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Methane (CH4) production and emission from wetlands are controlled by factors such as water table, temperature, soil redox potential and vegetation (Schimel, 1995; Grünfeld & Brix, 1999; Dinsmore et al., 2009). Plants affect CH4 emissions from wetlands in several ways. They provide litter and root exudates as a carbon source for methanogenic Archaea, and transport of gases via root and shoot aerenchyma frequently dominates wetland O2 and CH4 fluxes (Joabsson et al., 1999; Brix et al., 2001; Ström et al., 2003). The primary function of aerenchyma in wetland plants is to transport oxygen from the atmosphere to the roots (Armstrong, 1979). This process results in radial oxygen loss (ROL) from the roots to the rhizosphere (Colmer, 2003), which can reduce CH4 production and increase CH4 oxidation. Thus, in some cases, the presence of aerenchymatous vegetation reduces CH4 emission from wetlands (Grünfeld & Brix, 1999; Roura-Carol & Freeman, 1999; Dinsmore et al., 2009; Fritz et al., 2011).

Aerenchyma, on the other hand, can also provide a conduit for CH4 from the rhizosphere to the atmosphere, bypassing the oxidizing soil layers (Chanton & Dacey, 1991; Beckett et al., 2001; Colmer, 2003). This pathway can result in greater CH4 emissions from areas inhabited by aerenchymatous plants (Sorrell & Boon, 1994; Schimel, 1995; Shannon et al., 1996; Greenup et al., 2000; Ström et al., 2006). Schimel (1995), in particular, found that the composition of the plant community, based on cover of aerenchymatous plants, is a better predictor of CH4 flux from arctic wet meadow tundra than both water table and CH4 production rates. Greenup et al. (2000) likewise found substantial differences in CH4 emissions within the same peatland, with six times higher emissions from plots with the aerenchymatous plant Eriophorum vaginatum compared with plots without this plant, correlating well with below-ground biomass. Furthermore, they placed glass tubes in the soil to mimic the conduit effect and exclude the organic material supply of aerenchymatous plants, and elevated CH4 emissions were also observed here. Plant-mediated gas transport contributes up to 95% of the total CH4 flux from wetlands (Schütz et al., 1989; Chanton & Dacey, 1991; Grünfeld & Brix, 1999), depending on specific physicochemical conditions and intrinsic differences in species' ability to transport CH4. For example, the presence or absence of pressurized convective gas flow (Brix et al., 1992) can both increase and decrease CH4 emissions, as it can accelerate internal CH4 flux (Sorrell & Boon, 1994), but also increase rhizosphere oxidation and activity of methanotrophic bacteria (Grünfeld & Brix, 1999).

Whilst it seems likely that the large variation between species in plant-mediated CH4 emissions is a function of differences in airspace structure and tissue permeability to gases, understanding of this is hampered by contradictory evidence in the literature regarding the location and significance of the main transport resistances. The high porosities often found in roots of wetland graminoids of 30–50% (Justin & Armstrong, 1987) do not in themselves seem to be rate-limiting for plant gas transport, but it is difficult to generalize about which aspects of root and shoot morphology and physiology predominantly limit fluxes, based on the existing literature. Pioneering work by Sebacher et al. (1985) was able to categorize the CH4 transport capacity of a large number of species based on their growth form and CH4 transport capacity, but limiting factors for CH4 transport were not directly tested. In Carex-dominated wetlands, Morrissey et al. (1993) observed reduced CH4 emissions after induced stomatal closure, and concluded that stomatal conductance is an important control of CH4 flux. Their findings are supported by Schimel (1995), who found elevated CH4 emission from Carex aquatilis after experimentally removing the leaf blade barrier by clipping the leaves. On the other hand, the transfer of gases between the rhizosphere and the root aerenchyma has also often been proposed as a rate-limiting step for CH4 transport in graminoids (Schimel, 1995; Beckett et al., 2001). Many wetland plants have extensive barriers to gas diffusion in the exodermal root tissue (Colmer, 2003; Garthwaite et al., 2008). The barrier is a strategy for conserving oxygen within roots in anoxic soils, but also restricts CH4 flux in the opposite direction (Beckett et al., 2001). Mathematical models on the root-rhizosphere scale have also highlighted root wall permeability as an important factor limiting plant CH4 fluxes (Beckett et al., 2001; Segers & Leffelaar, 2001). Still other studies have identified high resistances to diffusion at the root–shoot junction (Groot et al., 2005).

To be able to quantify the effect of aerenchymatous plants on CH4 emissions, it is important to understand which specific anatomical and morphological factors govern the transport of CH4. Given the divergent literature on factors limiting CH4 fluxes in wetland plants, it is surprising that very few studies have attempted to use manipulative experiments to test predictions regarding rate-limiting resistances to transport. Thus, the objective of this study was to make a detailed empirical investigation of the factors controlling CH4 transport in Juncus effusus. We chose this as a representative, model species as it is now globally widespread in wetlands in all temperate biomes, and is typical of the graminoid taxa that dominate most minerotrophic wetlands. Specifically, we aimed to test the predictions implicit in mathematical modelling (Beckett et al., 2001), that root characteristics such as branching patterns and barriers to diffusion (Colmer, 2003) are the predominant factors controlling diffusive CH4 transport in wetland plants.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Plant material

Small self-sown tussocks of J. effusus L. seedlings were collected in autumn 2010 outside in irrigated beds at the Påskehøjgård growth facility, 15 km north of Aarhus, Denmark. The seedlings, together with the soil surrounding the roots, were brought to the laboratory and the soil was then rinsed from below-ground tissues, taking care to avoid damaging root material, and subsequently grown hydroponically in 30 l plastic tanks containing a nutrient solution consisting of tap water and 0.1 g l−1 ‘Pioneer Macro Green’ (19-2-15 NPK + Mg; Brøste, Kgs. Lyngby, Denmark) and 0.1 ml l−1 ‘Pioneer Micro Plus’ with iron, adjusted to pH 6.5 with H2SO4. The nutrient solution was renewed every week. Air saturation of the solution changed from an initial 100% (0.30 μmol O2 l−1) to c. 50% during the week. The plants were grown in a growth chamber (Weiss Umwelttechnic, Lindenstruth, Germany) with a light : dark cycle of 14 : 10 h at a photosynthetic photon flux density of 150 μmol m−2 s−1, a temperature cycle of 18 : 16°C, and a relative air humidity of 60 : 70%. The plants were grown for at least 3 months before experiments, during which period the tussocks were divided as necessary, and old dead roots were removed. This resulted in a stock of individuals with three to 15 shoots and with new roots formed in laboratory conditions. The plants had two distinct root types, thick white roots with no laterals, hereafter referred to as ‘coarse roots’ and thinner roots with numerous laterals, hereafter referred to as ‘fine roots’. The shoot and root DW, number of shoots and roots, shoot surface area (shoot SA), shoot cross-sectional area (shoot CSA), total length of coarse roots (CRL) and total length of fine roots (FRL) were measured after experiments (Table 1). J. effusus has constitutive aerenchyma formation with high root porosity, irrespective of growth conditions (Justin & Armstrong, 1987), and sections of roots from our plants had fractional root porosities of 37 ± 13% (mean ± 1 SD, = 3), which resembles the porosity of J. effusus roots grown in flooded, reduced soil (Justin & Armstrong, 1987).

Table 1. Size and morphological characteristics of the Juncus effusus tussocks in the experiment
ParameterMean ± 1 SD
  1. = 21; otherwise = 30.

Shoot
Shoot DW (g)0.59 ± 0.17
Number of shoots6.8 ± 2.1
Shoot surface area (mm2)9929 ± 2881
Shoot cross-sectional area (mm2)45 ± 12
Root
Root DW (g)0.15 ± 0.07
Number of fine roots*13 ± 4
Length of fine roots (mm)*3474 ± 1375
Number of coarse roots*9.2 ± 3.0
Length of coarse roots (mm)*1066 ± 607

Methylene blue localization of oxygen release

To determine the location of the gas-permeable surfaces and the development of barriers to radial gas exchange in roots of this species, we prepared methylene blue dye gels as described by Armstrong & Armstrong (1988) and Sorrell (1994), viz. 12 mg l−1 methylene blue dye in a solution of agar (6 g l−1). The solution was poured into 1 l Plexiglas vessels, and after bubbling with N2 gas to remove all oxygen, 130 mg l−1 of the reducing agent sodium dithionite was added. A small tussock of J. effusus was then sealed with coconut oil in the lid of each vessel, and the vessels were placed outside the growth chamber under ambient conditions, to avoid convective mixing of the solution by heat gradients. After c. 24 h, the sites of oxygen release were visually identified based on the development of blue halos around the roots.

Methane flux

Small tussocks of J. effusus, with three to 11 shoots and 11–34 roots, were manipulated to test which parts of the shoot–root system provided the primary resistance to CH4 diffusion. The manipulations included both above- and below-ground manipulations.

The experimental setup for flux measurements consisted of a 30 l plastic tank containing the nutrient solution with a lid fitted with a rubber seal. Holes in the lid had a 10-mm-diameter mounting collar for the plants and on top shoot chambers consisting of 0.9 l transparent acrylic cylinders. The shoot chambers were equipped with water traps to equalize pressure between the chamber and the surrounding atmosphere, and rubber septa for gas sampling.

Before flux measurements, the nutrient solution in the plastic tank was sparged with CH4 to a concentration of c. 1.16 mmol CH4 l−1 (≈ 75% saturation), which is at the upper range of that commonly found in natural wetland soil water (Chanton & Dacey, 1991). The tank was then placed in the growth chamber (same conditions as described earlier), and the plants were fitted to the mounting collar. To prevent gas from the nutrient solution leaching into the shoot chamber without passing through the plant–root system, the tussocks were sealed to the mounting platforms with coconut oil. Preliminary tests showed no diffusion of CH4 through the coconut oil (data not shown). After the mounting process, the platforms with the tussocks were firmly attached to the lid of the tank, with the roots of the plants extending into the nutrient solution.

Shoot chambers were mounted at T0 (the starting time where the first gas sample is taken) over the plants and 0.5 ml gas samples were withdrawn from the shoot chamber over time with a precision glass syringe. A total of five consecutive samples from each plant for each trial were withdrawn at c. 10 min intervals. The gas samples were injected directly into a Shimadzu GC-8A Gas Chromatograph equipped with a Flame Ionization Detector (200°C), using N2 carrier gas (60 ml min−1) and a Porapak Q column (60°C), for analysis of CH4 concentration.

After estimating the CH4 flux for the intact plant, the plants were subjected to a range of manipulations of either roots or shoots (Table 2), after which the flux rate of CH4 through the plant was determined for each manipulation. As manipulations were irreversible, each replicate plant was used only for a single series of manipulations. A total of 30 different tussocks were used in the experiment.

Table 2. Overview of the five experimental manipulation series (A–E) conducted
 Sequential manipulations
123456
  1. Each series included three to six manipulations of the Juncus effusus tussocks. For each manipulation a flux rate was estimated.

Root manipulations
AIntact plantHalf of the roots removedAll roots removed, except oneAll roots removed  
BIntact plantAll fine roots removedDistal 30 mm of remaining roots removed   
CIntact plantAll coarse roots removedDistal 30 mm of remaining roots removedRoots cut at 250 mm lengthRoots cut at 150 mm lengthRoots cut at 50 mm length
Shoot manipulations
DIntact plantAll shoots cut at 150 mm height, except oneAll shoots cut and sealed at 150 mm height, except oneAll shoots cut and sealed at the base, except oneLast shoot cut (no seal) 
EIntact plantAll shoots cut and sealed at 150 mm heightAll shoots cut and sealed at 100 mm heightAll shoots cut and sealed at 50 mm height  

Before each manipulation a 2 ml sample was withdrawn from the nutrient solution with a 10 ml syringe. The syringe was then filled with 8 ml of air, sealed with a rubber stopper and shaken vigorously for 2 min to ensure equilibrium between CH4 in the water and the air. The CH4 concentration in the gas was then analysed by gas chromatography. The concentration of CH4 in the nutrient solution was calculated using Henry's law for phase equilibrium (Lide, 2003).

Data analysis

All data analysis and statistical tests were performed using JMP 9 (SAS Institute Inc., Cary, NC, USA). The transport of CH4 through the plants was calculated from the increase in CH4 concentration in the shoot chamber over time with linear regression analysis. The CH4 concentration in the nutrient solution was plotted against time for each measuring series, to estimate the rate of decrease in CH4 concentration in the nutrient solution. To correct the CH4 flux rates in the plants for the decreasing CH4 concentration in the nutrient solution, all the calculated CH4 fluxes were normalized to a CH4 concentration of 1.16 mM. This normalization also ensured comparability between all the intact plants from the different measuring series with differences in initial CH4 concentrations of the water. Pairwise Pearson correlations were made to clarify which plant characteristics were correlated with plant CH4 transport capacity. Additionally, linear regressions between chosen variables and CH4 transport rate of the intact plants were made.

The mean flux rates for the different manipulations in each series were compared using a repeated-measures ANOVA. Tests of significance were conducted at the 5% level. To ensure comparability between the effect of the different manipulations, plots were also made for each series showing the average normalized transport rate in proportion to that of the intact plant. These data were only used for visualization of the flux rates, and not for statistical calculations.

Finally, based on the Pearson correlations, an empirical model using stepwise multiple regression was made to predict the methane transport through the intact plants. As many of the morphological variables were strongly intercorrelated, we chose only to build the models from variables that had proven to affect transport rates in the manipulation experiments significantly.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Oxidation of methylene blue

Methylene blue oxidation was generally restricted to the apical 20–30 mm of the roots and around the laterals of the fine roots (Fig. 1). A continuous cylinder of oxidized gel was apparent around any fine roots which had extensive lateral formation along their length. Around the tip of the coarse roots a halo of oxidized gel appeared, but little oxidation was observed more distally. The basal parts of the coarse roots showed no oxidation at all.

image

Figure 1. Sites of oxygen release in Juncus effusus as illustrated by the development of blue halos in the methylene blue and agar solution.

Download figure to PowerPoint

Methane flux through J. effusus

Plots of CH4 concentration in the chambers as a function of time were highly linear, with R2 values between 0.95 and 1. The only exceptions were manipulations with very low flux rates, where the variation can be assumed to originate from analytical error. The slopes of these regressions did not differ significantly from zero.

Methane concentration in the water decreased during measurements (Fig. 2). The average rate of decrease in CH4 concentration in the tank was c. 33 μM h−1, varying depending on how much CH4 was transported through the plants and during the manipulations of plants for the following measurement.

image

Figure 2. Example of a complete measuring sequence for a Juncus effusus tussock with four manipulations. The chamber methane (CH4) concentration is in the upper plot and the water CH4 concentration in the lower plot. Arrows indicate flushing of the chamber air between each manipulation, where the CH4 concentration in the chamber is reset to ambient values.

Download figure to PowerPoint

Intact plants

The average CH4 transport rate through the intact plants (± 1 SD) was 8.6 ± 4.6 μmol CH4 h−1 per plant (= 30). The variation in transport rate was mainly caused by the variation in size and anatomy of the plants used, and the correlation coefficients between CH4 transport rate and different plant characteristics can be seen in Table 3, which shows that the length of fine roots and shoot surface area both correlated strongly with CH4 transport rate, whereas the number of shoots did not.

Table 3. Pearson correlation matrix showing the pairwise correlations between plant morphological characteristics of Juncus effusus
 CH4 emission rateNumber of shootsShoot SAShoot CSADW shootsNumber of coarse rootsTotal length of coarse rootsNumber of fine rootsTotal length of fine roots
  1. = 21–30.

  2. SA, surface area; CSA, cross-sectional area.

  3. *, < 0.05; **, < 0.005; ***, < 0.0001.

Number of shoots0.29        
Shoot SA0.64***0.41*       
Shoot CSA0.44**0.160.84***      
DW shoots0.57***0.230.94***0.82***     
Number of coarse roots0.49*0.370.60**0.48*0.64**    
Total length of coarse roots0.39−0.050.340.350.390.66**   
Number of fine roots0.68**0.53*0.66**0.45*0.55*0.22−0.02  
Total length of fine roots0.85***0.260.84***0.68**0.84***0.370.240.75*** 
DW roots0.43***−0.190.66**0.77***0.77***0.43*0.62**0.350.76***

Most of the morphological characteristics were, to some extent, intercorrelated as a function of plant size. The length of fine roots and the shoot surface area, both affecting the CH4 transport rate linearly, are thus highly significantly correlated with each other (Table 3).

Root manipulations

Removal of roots significantly decreased the transport of CH4 through the plants. Removing half the roots resulted in a 39–75% (65% average) reduction in CH4 transport rate, and removing all the roots except one resulted in a further decrease in transport (Fig. 3a).

image

Figure 3. Relative methane (CH4) transport rate in Juncus effusus for three series of root manipulations. Transport rates are presented in proportion to the initial transport rates for intact plants (± 1 SD,= 6). (a) Manipulations: 1, intact plant; 2, half of the roots were cut at the base; 3, the remaining roots were cut at the base, except for one; 4, the final root was cut at the base. (b) Manipulations: 1, intact plant; 2, all fine roots removed, leaving only coarse roots; 3, the distal 30 mm of the remaining coarse roots were cut. (c) Manipulations: 1, intact plant; 2, all coarse roots were removed; 3, the distal 30 mm of the remaining fine roots were removed; 4, all roots were cut to a maximum length of 250 mm; 5, roots were cut to a maximum length of 150 mm; 6, roots were cut to a maximum length of 50 mm. Letters indicate significant difference (< 0.05) between manipulations.

Download figure to PowerPoint

Removing fine roots with laterals had a greater effect on the transport rate than removing coarse roots (Fig. 3b,c). After removing the fine roots, the mean transport rate was reduced by 71 ± 30% (Fig. 3b). Removal of the distal 30 mm of the remaining coarse roots in the same measuring series caused the transport rate to drop to 4% of that of the intact plant, which corresponds to a decline of 86% for this manipulation alone (Fig. 3b), which is disproportionate compared with the amount of root material removed.

Removing coarse roots from intact plants did not change the transport rate significantly, nor did cutting the distal 30 mm of the remaining fine roots (Fig. 3c). Only after cutting the remaining fine roots to a total length of 250 mm was a small but significant decrease in transport rate observed. The subsequent sequential cutting of the roots to 150 and 50 mm in length also caused significant decreases in transport rate.

Shoot manipulations

Manipulations of the shoots had little effect on the CH4 transport rate (Fig. 4). Cutting all the shoots except one at a height of 150 mm did not alter the transport rate, and subsequent sealing of the cut ends showed no effect (Fig. 4a). When cutting and sealing all the shoots, except one, at the base of the mounting platform, a significant decrease (average 45%) was observed. When cutting the last shoot at the height of the mounting platform, droplets of water from the plant conducting tissue formed at the ends, sealing the system, and thus nearly eliminated the flux through the plant. Finally, sequentially cutting and sealing the shoots at increasingly lower heights, to a final height of 50 mm, did not alter the transport rate significantly (Fig. 4b).

image

Figure 4. Relative methane (CH4) transport rate in Juncus effusus for two series of shoot manipulations. Transport rates are presented in proportion to the initial transport rates for intact plants (± 1 SD,= 6). (a) Manipulations: 1, intact plant; 2, all of the shoots, except one, were cut at a height of 150 mm; 3, the ends of the cut shoots were sealed with silicone wax; 4, all the shoots, except one, were cut and sealed at the base; 5, the last shoot was cut at the base. (b) Manipulations: 1, intact plant; 2, all shoots cut and sealed with silicone wax at a height of 150 mm; 3, all shoots cut and sealed at a height of 100 mm; 4, all shoots cut and sealed at a height of 50 mm. Letters indicate significant difference (< 0.05) between manipulations.

Download figure to PowerPoint

Modelling methane transport in intact plants

Based on the results of the manipulation experiments, we excluded shoot variables from the stepwise regression model. Shoot SA and shoot DW were both highly correlated to the CH4 transport rate, but most of the variation explained by these variables was also accounted for by FRL. The results of the simplified least-squares multiple regression model for CH4 transport through intact plants of J. effusus are shown in Eqn 1:

  • display math(Eqn 1)

where inline image is the CH4 transport rate (μmol CH4 h−1 per plant), FRL is the length of the fine roots in mm, and CRN is the number of coarse roots. FRL is the single best predictor of CH4 transport, and explains 72% of the variance alone. Adding CRN to the model as well gives a value of R2 of the full model that is equal to 0.76.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We used plant characteristics and shoot and root manipulations to clarify details about CH4 transport in J. effusus. Unlike some wetland plants, such as Typha sp. and Phragmites australis (Brix et al., 1992), J. effusus has no pressurized gas flow in its aerenchyma (Yavitt & Knapp, 1998; H. Brix & B. K. Sorrell, unpublished). Hence the transport of CH4 in J. effusus can be described with diffusion models similar to those developed for oxygen transport in plants (Armstrong & Beckett, 1987; Sorrell, 1994; Beckett et al., 2001). Some root manipulations significantly affected CH4 transport in this study, whereas shoot manipulations had little effect. Only removing root material with permeable surfaces from the plants reduced the CH4 transport, indicating that the amount of root surface permeable to gases is the most important limiting factor for CH4 transport through J. effusus. We used length of the fine roots rather than their surface area in our model, because of the difficulty in root scans of accurately distinguishing and estimating the permeable surface area of the complex, fine laterals, and length apparently functioned well as a proxy for permeable area, given the high explanatory power of the model.

The majority of the root surface permeable to CH4 consisted of the laterals of the fine roots, with a small contribution of the tips of the coarse roots, leading to strong correlations between fine root development and CH4 transport in our study. The low permeability to gas exchange over the root surfaces occurs because many wetland plants, primarily monocotyledons, develop a barrier to ROL along their roots to avoid oxygen deprivation and also to exclude phytotoxins (Armstrong, 1979). This barrier, consisting of lignin and suberin (Armstrong & Beckett, 1987; Visser et al., 2000; Garthwaite et al., 2008), reduces the loss of oxygen from the internal air spaces to the reducing rhizosphere and enhances the diffusion of oxygen within the root towards the apex (Armstrong & Beckett, 1987; Sorrell, 1994; Colmer, 2003). The early experiments and diffusion models summarized by Armstrong (1979) suggested that the greatest resistance to oxygen transport in plants is in the root/rhizosphere transition zone. More recently, Visser et al. (2000) found that, specifically in J. effusus, roots have a relatively strong constitutive barrier to ROL in their basal parts, but the distal parts just a few cm behind the apex are more permeable. ROL in J. effusus increases with increasing oxygen demand in the root environment, but quickly becomes saturated at higher demands, suggesting that the area permeable to oxygen quickly becomes the limiting factor (Sorrell, 1999). These results agree with the methylene-blue experiment and the flux results presented here, and suggest that the sites of CH4 uptake are the same as for oxygen leakage in reducing environments. Several authors have previously presented the rhizosphere–root interface as the limiting resistance for CH4 transport in some graminoid species (Chanton & Dacey, 1991; Schimel, 1995), as well as model simulations (Beckett et al., 2001), but to our knowledge this is the first time that this concept has been supported by experimental manipulation.

Shoot manipulations, by contrast, had little effect on CH4 transport in our study. If the ability of CH4 to escape from the plant was limited by a high resistance in the shoot walls, cutting the shoots and leaving the cut ends open would result in an increase in CH4 transport as observed for Carex species (Schimel, 1995; Kelker & Chanton, 1997). Likewise, the cutting of the shoots to increasingly lower heights, 150, 100 and 50 mm, with subsequent sealing, did not alter the CH4 transport significantly. From these results we conclude that the shoot tissues are very permeable to CH4, and that all the CH4 escapes from the plant at 50 mm from the base or below. CH4 release is also independent of stomatal conductance in Oryza sativa, where CH4 escapes through micropores in the basal parts (Nouchi et al., 1990). Basal CH4 release also occurs in Pontederia cordata and Sagittaria lancifolia, but through petioles (Harden & Chanton, 1994). In contrast to our study, Sebacher et al. (1985) found very low CH4 release rates in the field from J. effusus, as well as other aerenchymatous plants, and suggested that the relatively small transport rates were restricted by the impermeable outer layers of the shoots, but they also recognized that differences in root structure may be involved. Our study strongly suggests that it is root parameters, including access to sediment methane, not shoot permeability, that control CH4 transport in J. effusus. By inducing stomatal closure or by cutting off the leaf blades with or without subsequent sealing of the cut ends, some studies have found significant resistance in the above-ground parts of Carex sp. (Morrissey et al., 1993; Schimel, 1995; Kelker & Chanton, 1997). In our study, the effect of stomatal closure was not addressed specifically, but since virtually all the CH4 escapes through the base of the shoot, it is unlikely that above-ground parts or stomatal physiology of J. effusus offer any significant resistance to CH4 transport. The high correlation between shoot surface area and CH4 transport was a general effect of plant size on CH4 transport rather than any direct control of CH4 release by shoot morphology, and underlines the importance of combining modelling work with experimental manipulations. The differences in controlling factors for CH4 transport among wetland graminoids might originate from differences in plant architecture as suggested by Schimel (1995). The morphology of J. effusus differs from taxa such as Carex sp., having vestigial leaves and erect leafless shoots that constitute the main photosynthetic organs.

The only shoot manipulations that significantly decreased CH4 transport were when all shoots but one were cut and sealed at the base, allowing CH4 to escape only via a single shoot, after which CH4 transport decreased on average to 56% of that of the intact plant. It seems plausible that the decrease in transport rate from this shoot manipulation was caused by limited connectivity between roots and stems through the rhizome of J. effusus, and that not every root is connected to every shoot. The formation of a new shoot in J. effusus is coupled to the formation of new roots at the elongated rhizome. These newly formed roots and shoots might be interconnected, but connection to the older roots and shoots might be limited. Examples of studies where the rhizome is shown to limit CH4 transport in graminoids are sparse, but Groot et al. (2005) found high resistance in the root–shoot transition of rice tillers when the roots and shoots had been cut. This implies that the connectivity in the rhizome might be limiting for the transport of CH4 if some of the shoots in a tussock are damaged or waterlogged such that gas phase continuity with the rhizome is compromised.

The results presented here will be useful in the evaluation of CH4 emissions in the field. Modelling of CH4 emissions from different land use types requires information of the abiotic and biotic factors controlling the emissions. Walter & Heimann (2000) developed a process-based model to quantify CH4 emissions from natural wetlands. The model estimates the CH4 concentration in a one-dimensional soil profile, and estimates the flux of CH4 from molecular diffusion, ebullition and transport through aerenchymatous plants. For the latter the rooting depth of plants is used as a site-specific model parameter. Locating the specific sites for CH4 uptake in the root system, as was done here for J. effusus, will possibly help to improve these kinds of models. The amount of permeable root surface coinciding with high concentrations of soil water CH4 would then be the sites of particular importance.

We conclude that CH4 in J. effusus is taken up by the laterals of the fine roots and the apical part of the coarse roots, and the same barriers reported for radial oxygen loss going from the atmosphere to the rhizosphere also apply to CH4 diffusion in the opposite direction. Our results empirically confirmed what is predicted by the models of Beckett et al. (2001), that is, that the area of permeable root surface, derived from the proxy of root length of fine roots and number of coarse roots, is the most important controlling factor for CH4 transport in plants. Furthermore, we found evidence of limited connectivity in the rhizome of J. effusus, but further studies are needed to clarify its significance in this species and other similar graminoids. In such plants, we can now confidently confirm that roots provide the primary resistance to diffusion and limit plant CH4 transport, while the shoot resistance is insignificant, with all CH4 escaping from the basal parts of the shoot. Interspecific differences in sites for CH4 uptake by plants, as well as plant-specific controlling anatomical factors for CH4 transport, are important in future studies of aerenchymatous plants in their natural habitat explaining the microsite variability in CH4 emissions.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The study was funded by the Danish Council for Independent Research – Natural Sciences and Danish Ministry of Environment. We thank Søren O. Petersen for additional guidance and support of this work.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Armstrong J, Armstrong W. 1988. Phragmites australis – a preliminary study of soil-oxidizing sites and internal gas transport pathways. New Phytologist 108: 373382.
  • Armstrong W. 1979. Aeration in higher plants. In: Woolhouse HW, ed. Advances in botanical research. London, UK: Academic Press, 225332.
  • Armstrong W, Beckett PM. 1987. Internal aeration and the development of stelar anoxia in submerged roots. A multishelled mathematical model combining axial diffusion of oxygen in the cortex with radial losses to the stele, the wall layers and the rhizosphere. New Phytologist 105: 221245.
  • Beckett PM, Armstrong W, Armstrong J. 2001. Mathematical modelling of methane transport by Phragmites: the potential for diffusion within the roots and rhizosphere. Aquatic Botany 69: 293312.
  • Brix H, Sorrell BK, Lorenzen B. 2001. Are Phragmites-dominated wetlands a net source or net sink of greenhouse gases? Aquatic Botany 69: 313324.
  • Brix H, Sorrell BK, Orr PT. 1992. Internal pressurization and convective gas flow in some emergent freshwater macrophytes. Limnology and Oceanography 37: 14201433.
  • Chanton JP, Dacey JWH. 1991. Effects of vegetation on methane flux, reservoirs, and carbon isotopic composition. In: Sharkey TD, ed. Trace gas emissions by plants. London, UK: Academic Press Limited, 6592.
  • Colmer TD. 2003. Long-distance transport of gases in plants: a perspective on internal aeration and radial oxygen loss from roots. Plant, Cell & Environment 26: 1736.
  • Dinsmore KJ, Skiba UM, Billett MF, Rees RM. 2009. Effect of water table on greenhouse gas emissions from peatland mesocosms. Plant and Soil 318: 229242.
  • Fritz C, Pancotto VA, Elzenga JTM, Visser EJW, Grootjans AP, Pol A, Iturraspe R, Roelofs JGM, Smolders AJP. 2011. Zero methane emission bogs: extreme rhizosphere oxygenation by cushion plants in Patagonia. New Phytologist 190: 398408.
  • Garthwaite AJ, Armstrong W, Colmer TD. 2008. Assessment of O2 diffusivity across the barrier to radial O2 loss in adventitious roots of Hordeum marinum. New Phytologist 179: 405416.
  • Greenup AL, Bradford MA, McNamara NP, Ineson P, Lee JA. 2000. The role of Eriophorum vaginatum in CH4 flux from an ombrotrophic peatland. Plant and Soil 227: 265272.
  • Groot TT, Bodegom PM, Meijer HAJ, Harren FJM. 2005. Gas transport through the root–shoot transition zone of rice tillers. Plant and Soil 277: 107116.
  • Grünfeld S, Brix H. 1999. Methanogenesis and methane emissions: effects of water table, substrate type and presence of Phragmites australis. Aquatic Botany 64: 6375.
  • Harden HS, Chanton JP. 1994. Locus of methane release and mass-dependent fractionation from two wetland macrophytes. Limnology and Oceanography 39: 148154.
  • Joabsson A, Christensen TR, Wallén B. 1999. Vascular plant controls on methane emissions from northern peatforming wetlands. Trends in Ecology and Evolution 14: 385388.
  • Justin SHFW, Armstrong W. 1987. The anatomical characteristics of roots and plant response to soil flooding. New Phytologist 106: 465495.
  • Kelker D, Chanton J. 1997. The effect of clipping on methane emissions from Carex. Biogeochemistry 39: 3744.
  • Lide DR. 2003. Handbook of chemestry and physics. Boca Raton, FL, USA: CRC Press.
  • Morrissey LA, Zobel DB, Livingston GP. 1993. Significance of stomatal control on methane release from Carex-dominated wetlands. Chemosphere 26: 339355.
  • Nouchi I, Mariko S, Aoki K. 1990. Mechanism of methane transport from the rhizosphere to the atmosphere through rice plants. Plant Physiology 94: 5966.
  • Roura-Carol M, Freeman C. 1999. Methane release from peat soils: effects of Sphagnum and Juncus. Soil Biology and Biochemistry 31: 323325.
  • Schimel JP. 1995. Plant transport and methane production as controls on methane flux from arctic wet meadow tundra. Biogeochemistry 28: 183200.
  • Schütz H, Seiler W, Conrad R. 1989. Processes involved in formation and emission of methane in rice paddies. Biogeochemistry 7: 3353.
  • Sebacher DI, Harriss RC, Bartlett KB. 1985. Methane emissions to the atmosphere through aquatic plants. Journal of Environmental Quality 14: 4046.
  • Segers R, Leffelaar PA. 2001. Modeling methane fluxes in wetlands with gas-transporting plants: 1. Single-root scale. Journal of Geophysical Research D: Atmospheres 106: 35113528.
  • Shannon RD, White JR, Lawson JE, Gilmour BS. 1996. Methane efflux from emergent vegetation in peatlands. Journal of Ecology 84: 239246.
  • Sorrell BK. 1994. Airspace structure and mathematical modelling of oxygen diffusion, aeration and anoxia in Eleocharis sphacelata R. Br. Roots. Australian Journal of Marine and Freshwater Research 45: 15291541.
  • Sorrell BK. 1999. Effect of external oxygen demand on radial oxygen loss by Juncus roots in titanium citrate solutions. Plant, Cell & Environment 22: 15871593.
  • Sorrell BK, Boon PI. 1994. Convective gas flow in Eleocharis sphacelata R. Br.: methane transport and release from wetlands. Aquatic Botany 47: 197212.
  • Ström L, Ekberg A, Mastepanov M, Christensen TR. 2003. The effect of vascular plants on carbon turnover and methane emissions from a tundra wetland. Global Change Biology 9: 11851192.
  • Ström L, Lamppa A, Christensen TR. 2006. Greenhouse gas emissions from a constructed wetland in southern Sweden. Wetlands Ecology and Management 15: 4350.
  • Visser EJW, Colmer TD, Blom CWPM, Voesenek LACJ. 2000. Changes in growth, porosity, and radial oxygen loss from adventitious roots of selected mono- and dicotyledonous wetland species with contrasting types of aerenchyma. Plant, Cell & Environment 23: 12371245.
  • Walter BP, Heimann M. 2000. A process-based, climate-sensitive model to derive methane emissions fromnatural wetlands: application to five wetland sites, sensitivity to model parameters, and climate. Global Biogeochemical Cycles 14: 745765.
  • Yavitt JB, Knapp AK. 1998. Aspects of methane flow from sediment through emergent cattail (Typha latifolia) plants. New Phytologist 139: 495503.