Download - Gas Diffusion Coefficient of Undisturbed Peat Soils

Short Paper Soil Sci. Plant Nut!:, 61 (3). 43 1-435.2005 43 1

Gas Diffusion Coefficient of Undisturbed Peat Soils

Ippei Iiyama' and Shuichi Hasegawa

Soil Amelioration Laboratory. Research Group of Regional Environment, Graduate School of Agriculture, Hokkaido University, Sapporo, Hokkaido. 060-8589 Japan

Received November 4,2004; accepted in revised form March 3,2005

Determination of the gas diffusion coefficient D, of peat soils is essential to understand the mechanisms of soil gas transport in peatlands, which have been one of major potential sources of gaseous carbons. In the present study, we aimed at determining the D, of peat soils for various values of the ahfilled porosity a and we tested the validity of the Three- Porosity Model (Moldrup et al. 2004) and the Millington-Quirk model (1961) for predicting the relative gas diffusivity, the ratio of D, to Do, the gas diffusion coefficient in free air. Undisturbed peat soil cores were sampled from aerobic layers in the Bibai mire, Hokkaido, Japan. The MQ model reproduced the measured D,/D, curves better than the TPM. The TPM, a predictive model for undisturbed mineral soils, overestimated the D,/Do values for peat soils, implying that in the peat soils the pore pathways were more tortuous than those in the mineral soils. Since the changes in the D,/D, ratios with the a values of a well-decomposed black peat soil tended to be more remarkable than those of other high-moor peat soils, the existence of a positive feedback mechanism was assumed, such that peat soil decomposition itself would increase the soil gas diffusivity and promote soil respiration.

Key Words: &filled porosity, gas diffusion coefficient, peat soils, relative gas diffusivity.

Peatlands in the northern hemisphere have been considered to be large carbon pools amounting to 455 Pg-C (Gorham 199 l), and accounting for approximately one- third of the total world pool of soil carbon (1,395 Pg-C) estimated by Post et al. (1982). Peatlands are ecosys- tems vulnerable to influences from human activities or climatic changes and can become one of the major sources of carbon in global carbon cycling.

Since soil carbon is emitted from peatlands mainly through the gaseous phase as methane and carbon diox- ide, understanding of the transport of these gases in peat soils is essential for predicting the amount and the rate of carbon losses from peatlands.

The movement of gases in soils occurs mainly by diffusion, and modeling of the soil gas movement requires information about the soil gas diffusion coefficient D,, which depends on the soil air-filled porosity a. In early studies, the shape of the curves of the D,(a) functions was examined using simplifying assumptions for describing the structure of porous media (Millington 1959; Millington and Quirk 1961). These D,(a) models

I Present address: Water Quality and Solute Dynamics Group, Department of Environmental Chemistry and Biochemistry, National Institute for Agro-Environmental Sciences, Kannondai 3-1-3, Tsukuba, Tbaraki, 305-8604 Japan

have been tested mainly using repacked soils. And in recent studies, predictive models of D,(a) functions for undisturbed soils, including Japanese volcanic ash soils (Moldrup et al. 2003), have been improved by considering the soil water characteristic curves (Moldrup et al. 1999,2000,2004). However, presently, the gas diffusion coefficient for peat soils has not been fully documented, which makes it difficult to predict the amount and the rate of gas transport in peat soils.

In the present study, we determined the soil gas diffusivity DJD, (Do = gas diffusion coefficient in free air) of undisturbed peat soil samples, and characterized the changes in the D,/D, with a values for peat soils by comparing the measured data with existing D,(a) models.

Materials and methods Study field and soil profile at the sampling

point. The Bibai mire (43"19'N,14lo48'E), Hokkaido, Japan was used as a study field. This mire, originally a typical ombrotrophic bog, had been preserved by the National Agricultural Research Center for Hokkaido Region with a size of approximately 50 ha. The sur- rounding area of the mire had been completely reclaimed as crop fields for more than 40 years and a decrease in the groundwater table level took place in the

432 I. IIYAMA and S. HASEGAWA

Bulk density(0) (Mg m-3) Carbon content(A) (kg kg-I)

0 0.1 0.2 0.3 0.4 0 0.1 0.2 0.3 0.4 0.5 0

10

20

.$30 5 3 40

50

60

70

h

n

1 1 1 ,

0 0.5 1.0 1.5 2.0 2.5 0 10 20 30 40 50 60 Particle density(.) (Mg ) C N A )

peripheral area of the mire, threatening indigenous plant species like sphagnum and sedge. We collected undisturbed peat soil samples from the area within a distance of 10 m from a drainage ditch. Peat layers at the sampling point had been subjected to aerobic conditions, resulting in the decomposition of peat soil near the sur- face. Figure l(a) shows the profiles of the bulk density and soil particle density, while Fig. l(b) shows the profiles of the soil carbon content and C/N ratio. Soil carbon content and C/N ratio were determined using an automatic C-N analyzer (SUMIGRAPH NC- 1O00, Sumika Chemical Analysis Service, Ltd., Tokyo, Japan) and a gas chromatograph (GC-gA, Shimadzu Corp., Kyoto, Japan).

Undisturbed peat soil samples. Duplicate undisturbed peat cores for determining the soil gas diffusivity D,/D, were sampled from each layer using a sharpened knife and 100cm3 samplers with an inner diameter and height of 50 and 51 mm, respectively. Samples were taken at 12.5, 22.5, 32.5, 42.5, 52.5 and 62.5 cm depths.

To change the air-filled porosity of the peat samples, we applied the hanging water column method (Dane and Hopmans 2002). Firstly, the undisturbed samples were saturated for 24 h in a constant temperature room at 25°C. Thereafter, the samples were dehydrated by low- ering the matric potential step-by-step from - 5 to - 100 cm. For each moisture equilibrium, the samples were weighed to determine the moisture content and the air-filled porosity a, and the DJDO values were determined.

Soil gas dif€usivity determination. We determined DJD,, values based on N2-air binary diffusion phenomena through a soil sample in a single chamber apparatus, according to the methods of Taylor (1949),

Fig. 1. Soil profile at the sampling point in the Bibai mire showing (a) bulk density ( n = 3), soil particle density (n = 2), (b) carbon content ( n = 2) and C/ N ratio ( n = 2 ) . Solid lines denote the average values.

Currie (1960) and Osozawa (1987). The tracer gas was 0, and we measured the changes in its concentration inside the diffusion chamber with a Galvanic cell sensor (OS-3S-D, New Cosmos Electric Co., Ltd., Osaka, Japan). When we found that the space between a peat soil sample and a core wall started to increase due to drying and shrinking processes of the sample at low matric potentials, we filled the space with a sealing compound (NEO SEAL B-3, Nitto Chemical Industry Co., Ltd., Osaka, Japan) to prevent the gas from bypass- ing the soil matrix during the measurements of the breakthrough curves of 0,.

For the determination of D, in the single chamber method, it is preferable to consider the diffusion behavior of the tracer gas inside the chamber. For example, El-Farhan et al. (1996) reported that the determination of D, from the solution of Currie (1960) may overestimate the true value if the air inside the chamber is not mixed well. When the diffusion behavior inside the chamber is considered, the governing equation describing the measurement system becomes

a2c ax2 (0 S x S L , ) ac = D

I

where C is the concentration of the tracer gas in the gaseous phase, t is the elapsed time, x is the distance at the flow direction of the gas, D, is the N,-air binary diffusion coefficient in free air, L, is the length of the soil core (=5.1 cm) and La is the length of the diffusion chamber (= 10.3 cm). Do (cm2 s-I) at an absolute temperature T, Do., , was calculated by the following equation (Satou 1970):

Gas Diffusion Coefficient of Peat Soils 433

where Do.,., = 0.178 (Satou 1970). In the present study, the atmospheric pressure p was assumed to be the same as pwc. The initial and the boundary conditions of the measurement system are

C = C i , O s x ~ L , , t = O

c = c,, L, s x s L, + La, t = 0

c = ci, x = 0, t>O

(3)

~- - 0, x = L, + La, t>O ac ax

(4)

where Ci is the gas concentration in free air and C, is the gas concentration inside the chamber at r = O . We determined the D, values by fitting finite element solu- tions of the initial-boundary problem in Eqs. (l), (3) and (4) to the measured breakthrough curves.

Comparison of the DB(a)/.. , functions of peat soils with existing models. In order to characterize the changes in the DJD, with the a values for peat soils, we used the Three-Porosity Model (TPM) proposed by Moldrup et al. (2004). The TPM is a predictive model that showed reliable predictions of the D,(a)/D, in undisturbed soils for a wide range of soil types and total porosities (Moldrup et al. 2004).

The TPM is described by the following monomial power law function:

D,/Do = @' (a/@)' for O< a 5 @ (5) where @ is the soil total porosity (m3 m-3) and X is a tortuosity-connectivity parameter. To determine X, the

o.20 -1 decomposed (22.5cm)

TPM adopted the empirical relationship given by (Mold- rup et al. 2000)

D,,,/D, = 2aIw3 + 0 . 0 4 ~ ~ ~ (6)

where DFl, is D, at a matric potential of - 100 cm and a, , is the corresponding air-filled porosity. Eq. (6) described the measured data sets well (coefficient of regression R2 = 0.97) for 126 undisturbed mineral soils with 752 measurements, regardless of the soil types or total porosities (Moldrup et al. 2000). Substituting Eq. (6) into ( 5 ) gives

(7)

From Eqs. (5) and (7), it is obvious that the TPM requires only values of alOO and @ to predict a D,(a)/D, curve.

The other model that we tested was the Millington- Quirk model (MQ) (Millington and Quirk 1961)

x = log[(2a,,3 + 0.04a,,)/@l/log(a,~@)

DJD, = (8) The MQ model requires only @ to predict a D,(a)/D, curve and has been so widely used that it can be a good reference for characterizing the measured D,(a)/Do curves for the peat soils.

Results and discussion Figure 2 shows the changes in the DJD, with the a

values for the peat soil samples. The solid line and the dotted line in each graph represent the curves derived from the TPM and the MQ model, respectively. The largest a value of each sample corresponded to uIW. Parameters used in these two models were based on the

(c) High-moor peal (32 5cm)

(42.5cm)

TPM 0.15 -

1

(e) High-moor peat (52 5cm)

(0 High-moor peat (62.5cm)

MQ

0 0.1 0.2 0.3 0.4 0.5 0 . 6 0 0.1 0.2 0.3 0.4 0.5 0 . 6 0 0.1 0.2 0.3 0.4 0.5 0.6 Air-filled porosity a (m3 m 3 ) Air-filled porosity a (m3 m 3 ) Air-filled porosity a (m3 m3)

Fig. 2. Changes in the diffusivity DJD, with the values of the air-filled porosity a for peat soil samples taken from different depths. Solid line and dotted line represent the curves derived from the Three-Porosity Model (TPM) and the Millington-Quirk model (MQ), respectively.

434 I. IIYAMA and S. HASEGAWA

0 BP 12.5cm A H P 22.5cm W HP 32.5cm O H P 42.5cm A H P 52.5cm

h 1 .o 0.9 5 0.8

0.7

0.6

0.5 .g 0.4 3 0.3

% 2

)3

s al

P -120 -100 -80 -60 -40 -20 0

Matric potential (cm)

Fig. 3. Soil water characteristic curves of peat soils at different depths. BP and HP stand for Black Peat and High-moor Peat, respectively.

measurements of duplicate samples for each layer. In the low air-filled porosity region, mainly a S 0.1, the DJD, values of the peat soils remained near zero, suggesting the presence of non-effective air-filled pores that cannot act as air pathways. And as the a value increased, the DJD, of the black peat sample (Fig. 2(a)) increased more remarkably than that of other high-moor peat samples.

The TPM overestimated the measured DJD, curves in all the samples and only the prediction for the black peat sample was relatively close to the measured data set (Fig. 2(a)). Although the accuracy of Eq. (6) was supported by a very high regression coefficient (Moldrup et al. 2000), it was essentially based on D,,,/D, measurements in European mineral soils. For example, Moldrup et al. (2003) recognized that Eq. (6) did not describe the D,,,/D, of volcanic ash soils in Japan, presumably due to the higher macropore continuity of these soils, com- pared with European mineral soils. We consider that the inadequate performance of the TPM in the present study was due to a failure to account for the pore structure inherent to peat soils. Since a porous medium shows a lower DJDn value for a given air-filled porosity in the presence of more tortuous pore pathways, it is expected that peat soils may display a more complex configura- tion of air-filled pores in the direction of gas diffusion than mineral soils.

The MQ model reproduced the measured curves better than the TPM as shown in Fig. 2 (c), (d), (e) and (f), although it tended to underestimate slightly the measured DJD, values for a values smaller than 0.4. As shown in Fig. 2 (c) and (f), on the other hand, the MQ model tended to overestimate the measured curves for a values larger than 0.4. Millington (1959) also showed that the MQ model significantly underestimated the

measured DJD, data from Taylor (1949) for a values smaller than 0.4 and overestimated them for a values around 0.5. Jin and Jury (1996) pointed out the same kind of erroneous estimation when the MQ model was used.

Figure 3 shows the soil water characteristic curves of the peat soils. Three high-moor peat samples at 42.5, 52.5 and 62.5 cm depths displayed almost the same curves (Fig. 3), and the DJD, curves of these three samples shown in Fig. 2 were indistinguishable from each other in their shape. Therefore, we considered that these three layers were a uniform layer in terms of soil gas diffusivity. The high-moor peat sample at 22.5 cm depth showed a higher water retention than other samples, so that the DJD, values remained small even at a matric potential of - 100 cm (Fig. 2(b)).

Despite the similarity of the soil water characteristic curves between the black peat sample and the high-moor peat sample at 32.5 cm depth, the DJD, curves of these two samples (Fig. 2 (a) and (c)) were remarkably different from each other, implying that the pore geometry of the black peat was less tortuous than that of the high- moor peat. Considering that the black peat layer has been more decomposed than other layers, as demon- strated by its lower carbon content (Fig. l(b)), we sug- gest that the soil gas diffusivity in peat soils increases with peat soil decomposition. This implies the existence of a positive feedback mechanism whereby peat soil decomposition itself will result in promoting soil aeration and respiration.

Acknowledgments. We thank Dr. Osamu Nagata and the staff members of the Bibai Research Station, the National Agricul- tural Research Center for Hokkaido Region, for their assistance in the work conducted in the Bibai mire. We also thank Dr. Hideyuki Saito and Mr. Fumiaki Tabo, Graduate School of Agri- culture, Hokkaido University, for allowing us to use the C-N analyzer at their laboratory. This research was supported by “Research Fellowships of the Japan Society for the Promotion of Science for Young Scientists”-granted to the senior author (1. 1.).

REFERENCES

Currie JA 1960: Gaseous diffusion in porous media Part I-A non-steady state method. British Journal of Applied Physics,

Dane, JH and Hoprnans JW 2002: Water retention and storage. In Methods of Soil Analysis: Part 4, Physical Methods. SSSA Book Series 5 , Ed. JH Dane and GC Topp, p. 680- 683, Soil Science Society of America, Inc., Madison, Wis- consin

El-Farhan YH, Petersen LW, Rolston DE, and Glauz RD 1996: Analytical solution for two-region diffusion with two well- mixed end chambers. Soil Sci. Soc. Am. .I., 60, 1697-1704

11,314-317

Gas Diffusion Coefficient of Peat Soils 435

Gorham E 1991: Northern peatlands: Role in the carbon cycle and probable responses to climatic warming. Ecological Applications, 1, 182- 195

Jin Y and Jury WA 1996: Characterizing the dependence of gas diffusion coefficient on soil properties. Soil Sci. SOC. Am. J . ,

Millington RJ 1959: Gas diffusion in porous media. Science,

Millington RJ and Quirk JP 1961: Permeability of porous solids. Trans. Faraday SOC., 57, 1200-1207

Moldrup P, Olesen T, Schjonning P, Yamaguchi T, and Rolston DE 2000: Predicting the gas diffusion coefficient in undisturbed soil from soil water characteristics. Soil Sci. Soc. Am.

Moldrup P, Olesen T, Yamaguchi T, Schjonning P, and Rolston DE 1999: Modeling diffusion and reaction in soils: VIII. Gas diffusion predicted from single-potential diffusivity or permeability measurements. Soil Sci., 164, 75-8 1

Moldrup P, Olesen T, Yoshikawa S, Komatsu T, and Rolston DE

60.66-71

130,100-102

J., 64, 94-100

2004: Three-Porosity Model for predicting the gas diffusion coefficient in undisturbed soil. Soil Sci. Soc. Am. J. , 68,

Moldrup P, Yoshikawa S, Olesen T, Komatsu T, and Rolston DE 2003: Gas diffusivity in undisturbed volcanic ash soils: Test of soil-water-characteristic-based prediction models. Soil Sci. Soc. Am. J . , 67, 41-51

Osozawa S 1987: Measurement of soil-gas diffusion coefficient for soil diagnosis. Soil Phys. Cond. Plant Growth Jpn., 66, 53-60 (in Japanese with English summary)

Post WM, Emanuel WR, Zinke PJ, and Stangenberger AG 1982: Soil carbon pools and world life zones. Nature, 298, 156- 159

Satou K 1970: Diffusion coefficient. In Bussei Teisuu Part 8, Ed. The Society of Chemical Engineers, Japan, p. 223-235, Maruzen, Tokyo (in Japanese)

Taylor SA 1949: Oxygen diffusion in porous media as a measure of soil aeration. Soil Sci. Soc. Am. Proc., 14, 55-61

750-759

Download - Gas Diffusion Coefficient of Undisturbed Peat Soils

Top Related