A Subsurface, Closed-Loop System for Soil Carbon ... - PubAg - USDA

SOIL BIOLOGY & BIOCHEMISTRY

A Subsurface, Closed-Loop System for Soil Carbon Dioxide and Its Application to the Gradient Efflux Approach T. M. DeSutter* Dep. of Soil Science Walster Hall North Dakota State Univ. Fargo, ND 58105 formerly with USDA-ARS National Soil Tilth Lab. Ames, IA, 50011

T. J. Sauer T. B. Parkin USDA-ARS National Soil Tilth Lab. Ames, IA, 50011

J. L. Heitman Soil Science Dep. Williams Hall North Carolina State Univ. Raleigh, NC, 27695

Carbon dioxide concentrations in the soil can vary both temporally and spatially. Methodology was developed to semicontinuously measure subsurface concentrations of CO2 using expanded, porous Teflon (ePTFE) tubing. Lengths of ePTFE tubing (7.6 m) were buried at 0.02, 0.1, and 0.18 m below the soil surface in a Harps loam soil (fine-loamy, mixed, superactive, mesic Typic Calciaquoll) in central Iowa, and also positioned directly on the soil surface (0 m). Soil atmospheric gases that diffused through the walls of the tubing were circulated in a closed-loop design through solid-state CO2 sensors to determine the concentration of CO2 at each depth. Independent measures of CO2 concentrations were also determined by sampling the in-line gas stream of the ePTFE system and from samples extracted from gas wells positioned near the buried tubing. Good agreement (r2 > 0.95) was observed between the ePTFE system and the independent measures, with the ePTFE having biases of 1.2 and 1.37 times greater than the in-line and gas well samples, respectively. The soil-gas diffusion coefficient of CO2 (Ds) was determined using intact soil cores and values were about 2.5 times less than two popular models used to predict Ds in soil. Estimates of CO2 flux using Fick’s Law, six approaches to determine the vertical CO2 concentration gradient, and three methods to determine Ds ranged from >800 to 0.99. Values were derived by taking the 2 provides evidence of surface sealing or first derivative at z = 0.02. increased heterotrophic and autotrophic ¶ Vertical concentration gradient as determined by using linear regression and computing the slope of the line for CO2 concentration with depth (Tang et al., 2003). respiration. This approach has an advantage in that only two depths are needed to # Vertical concentration gradient. SSSAJ: Volume 72: Number 1 • January–February 2008

131

Table 2. Estimated CO2 efflux from a Harps loam soil using six approaches to determine the vertical CO2 gradient (Table 1) and three methods to estimate the soil-gas diffusion coefficient of CO2 in this soil (Fig. 7) and measured CO2 efflux determined by an automated sampling chamber. DOY

CO2 concentration gradient approach

Estimated CO2 efflux Using measured Ds Using Moldrup et al. (2000) Ds Using Millington and Quirk (1961) Ds

CO2 efflux, automated chamber

———————————— μmol m−2 s−1 ————————————— 237.5

4.5

11.9

10.3

3.6

239.5

274.8

873.8

692.1

1.4

241.5

100.2

293.3

242.1

3.9

243.5

10.1

27.3

23.3

2.9

1.0

2.7

2.3

3.6

239.5

0.8

2.4

1.9

1.4

241.5

1.4

4.0

3.3

3.9

243.5

1.1

3.0

2.6

2.9

237.5

0.7

2.0

1.7

3.6

239.5

0.7

2.2

1.7

1.4

241.5

0.9

2.5

2.1

3.9

243.5

0.7

2.0

1.7

2.9

237.5

1.3

3.3

2.9

3.6

239.5

5.6

17.9

14.2

1.4

241.5

4.8

14.0

11.6

3.9

243.5

1.9

5.0

4.3

2.9

1.0

2.6

2.2

3.6

239.5

1.4

4.5

3.6

1.4

241.5

1.5

4.5

3.7

3.9

237.5

237.5

A1—dC/dz, z = 0 m

dC/dz, z = 0.02 m

A2—slope

A3—ΔC/Δz, 0.02–0 m

ΔC/Δz, 0.1–0 m

243.5

1.0

2.7

2.3

2.9

0.9

2.3

2.0

3.6

239.5

0.4

1.1

0.9

1.4

241.5

0.7

2.1

1.8

3.9

243.5

0.8

2.2

1.9

2.9

237.5

ΔC/Δz, 0.1–0.02 m

are out of scale with many reported values (Ham and Knapp, 1998; Jassal et al., 2005; Parkin and Kaspar, 2003; Tang et al., 2003). Approach A1 evaluated at z = 0.02 m, when combined with the two modeled Ds approaches, yielded efflux values more consistent with the automated chambers (Table 2). In contrast, efflux estimates using the measured Ds values were nearly three times lower than the effluxes determined using the automated chamber at z = 0.02 m. The A2 approach underestimated efflux on all days compared with the automated chamber when using the measured Ds values, even on DOY 237.5 when the concentration profile was most linear (Table 2). Although greater estimates were observed when using the modeled Ds values, values were still less than the automated chamber except for DOY 239.5 using Moldrup et al. (2000) (Table 2). The potential error of using the A2 approach when estimating efflux follows from the fact that the assumption of a linear vertical concentration gradient is not always valid. After rainfall events, such as DOY 239.5, the concentration gradient becomes very nonlinear and this approach may greatly underestimate the concentration gradient (Fig. 8, Table 2). Of the three finite-difference approaches for determining vertical CO2 gradients (A3), gradients evaluated between 0.02 and 0 m yielded the greatest efflux values (Table 2). The combination of this CO2 gradient and the measured Ds model gave estimates of efflux that were similar to the automated chamber except 1 d after rainfall (DOY 239.5). Here, as observed when dC/dz was evaluated at z = 0 m, the estimated CO2 efflux sharply increased 132

when the chamber efflux decreased. Using the gradient between 0.02 and 0 m and between 0.1 and 0 m, the two modeled Ds approaches tended to overestimate efflux compared with the automated chambers on the 2 d following the rain event (Table 2). Efflux was underestimated in all cases compared with the automated chamber using all Ds approaches and evaluating the gradient between 0.1 and 0.02 m (Table 2). Using the chamber efflux data and the A3 approach between 0.02 and 0 m, the Ds values would have needed to be 1.3 × 10−6, 7.3 × 10−8, 3.0 × 10−7, and 6.7 × 10−7 m2 s−1 for DOY 237.5, 239.5, 241.5, and 243.5, respectively, to correctly estimate the chamber efflux. Using the modeled Ds formula (Fig. 6) and the above information, estimates of the volumetric water content between 0.02 and 0 m would be 0.22, 0.44, 0.35, and 0.29 m3 m−3 for DOY 237.5, 239.5, 241.5, and 243.5, respectively; however, the 0.06-m volumetric water content data, which were used for all Ds estimates in Fig. 7, were 0.28, 0.32, 0.30, and 0.29 m3 m−3 for DOY 237.5, 239.5, 241.5, and 243.5, respectively. Clearly, the 0.06-m data were not able to accurately describe changes in surface (0–0.02-m) soil water content in this example. Surface soil water content was overestimated during periods of drying and underestimated during periods of wetting. A critical problem of evaluating gas fluxes from the soil surface using (Eq. [1]) is getting accurate estimates of volumetric water content and AFP near the soil surface to use in the Ds models. Currently, only two methods known to us have been used to determine volumetric water content just below the soil surface SSSAJ: Volume 72: Number 1 • January–February 2008

(0.01 m). These are hand sampling and the dual-probe heat capacity sensor (Heitman et al., 2003; Ochsner et al., 2003; Tarara and Ham, 1997). Hand sampling can become tedious and unrealistic if hourly estimates of efflux are to be determined. Use of dualprobe heat capacity sensors, however, can provide hourly estimates of volumetric soil water content to within 0.01 m3 m−3 (Heitman et al., 2003). Assessment of the feasibility of such approaches to determine near-surface water content in combination with gradient approaches warrants further study.

CONCLUSIONS Estimates of CO2 concentrations in the soil can be achieved using the porous tubing (ePTFE) design outlined here. The advantages of this approach are: semicontinuous, in situ measurements of CO2 concentrations with depth; use of the same sensor(s) in determining CO2 concentrations at all depths; the ability to integrate over a large area; fast equilibration between soil atmosphere and inner tubing volume; and the ability to sample the air stream for other gases. There are also numerous disadvantages, however, including: initial disturbance of the soil at installation; difficulty with size and placement of tubing near the soil surface; the need to remove water or water vapor before CO2 analysis; and the cost of the ePTFE used here, although there are less expensive alternatives (DeSutter et al., 2006). A point of interest that should also be considered is the outside diameter of the tubing and how this may affect overall tubing depth. In this study, the center of the ePTFE tubing was positioned at the target depths of 0.02, 0.1, and 0.18 m. However, using the 0.02-m position as an example, the tubing depth ranged from 0.015 to 0.025 m below the surface, leaving 50% of the tubing above and below the target depth. Surface roughness at the time of installation may also complicate the determination of the exact tubing depth at installation. This uncertainty may affect the calculated concentration gradient and the estimate of efflux. Although not observed in this study, trenching may alter the soil physical structure. Following heavy rains, the trenched soil may settle, have a reduction in macroporosity, and hardsetting may occur after drying (Karunatilake and van Es, 2002). These results may affect the estimates of Ds and ultimately the depth of the tubing. The soil used in this study was tilled before installation of the tubing and there was no visual evidence of soil settling. The likelihood of soil settling after trenching in undisturbed soils (i.e., prairie, forest, no-till agriculture) is great, however, and thus increased care should be taken when assessing both Ds and tubing depth in the trench compared with the surrounding soil. Using CO2 concentration data with depth with the objective of determining CO2 efflux using Fick’s Law (Eq. [1]) can be challenging. The ability to accurately estimate CO2 efflux hinges on both accurate Ds values and accurate vertical concentration gradients. Not only must gradient and Ds be estimated accurately, they must be estimated for the soil depth for which the CO2 efflux is to be calculated. For example, if CO2 efflux from the soil surface is of interest, then the Ds and dC/dz must be estimated for the surface or near-surface soil. Use of average soil profile Ds or dC/dz may not yield representative estimates of surface CO2 emissions. Soil-specific Ds values determined for our soil were about 2.5 times less than those predicted using two popular Ds models. Thus, caution in data interpretation should be used when values of Ds are not determined independently, as done here and elsewhere SSSAJ: Volume 72: Number 1 • January–February 2008

using intact cores (Jassal et al., 2005) or in situ field methods (Lai et al., 1976; McIntyre and Philip, 1964; Rolston et al., 1991). Depending on the approach used and the depth of interest, gradients may vary widely (Table 1). Determining vertical concentration gradients using linear regression was not useful after a rainfall event but may be more applicable during dry soil conditions. The finite-difference approach is the most straightforward but has the potential to misrepresent curvilinear concentration patterns with depth. Overall, this experiment points to many questions that remain about estimating CO2 flux within and from the soil using the gradient approach. Future research is needed to improve the reliability of the gradient flux approach, including new methods for determining the Ds of surface soils (0–0.02 m), quantifying the number and placement of tubing and sensors for determining CO2 concentrations with depth, the use of automated flux chambers to help define the proper gradient approach and Ds determination, and application of sensors to determine the water content in surface soils (0–0.02 m). Even with the above concerns, the gradient approach can be a tool to estimate gas efflux or, although not a focus of this research, the most useful application of this approach may be to determine gas production zones within the soil. When using the trenching method in areas where disturbance of the soil is not possible or is impractical, aboveground chamber methods should strongly be considered. ACKNOWLEDGMENTS Technical assistance was provided by Tim Hart, David Meek, Forrest Goodman, Paul Doi, and Jadon Kool. REFERENCES Burton, D.L., and E.G. Beauchamp. 1994. Profile nitrous oxide and carbon dioxide concentrations in a soil subject to freezing. Soil Sci. Soc. Am. J. 58:115–122. Chen, D., J.A.E. Molina, C.E. Clapp, R.T. Venterea, and A.J. Palazzo. 2005. Corn root influence on automated measurement of soil carbon dioxide concentrations. Soil Sci. 170:779–787. Dane, J.H., and J.W. Hopmans. 2002. Pressure cell. p. 694–688. In J.H. Dane and G.C. Topp (ed.) Methods of soil analysis. Part 4. Physical methods. SSSA Book Ser. 5. SSSA, Madison, WI. Davidson, E.A., and S.E. Trumbore. 1995. Gas diffusivity and production of CO2 in deep soils of the eastern Amazon. Tellus Ser. B 47:550–565. Davidson, E.A., K.E. Savage, S.E. Trumbore, and W. Borken. 2006. Vertical partitioning of CO2 production within a temperate forest soil. Global Change Biol. 12:944–956. DeSutter, T.M., T.J. Sauer, and T.B. Parkin. 2006. Porous tubing for use in monitoring soil CO2 concentrations. Soil Biol. Biochem. 38:2676–2681. Drewitt, G.B., T.A. Black, and R.S. Jassal. 2005. Using measurements of soil CO2 efflux and concentrations to infer the depth distribution of CO2 production in a forest soil. Can. J. Soil Sci. 85:213–221. Evans, D. 1965. Gas movement. p. 319–330. In C.A. Black et al. (ed.) Methods of soil analysis. Part 1. Agron. Monogr. 9. ASA and SSSA, Madison, WI. Fierer, N., O.A. Chadwick, and S.E. Trumbore. 2005. Production of CO2 in soil profiles of a California annual grassland. Ecosystems 8:412:429. Fierer, N., and J.P. Schimel. 2003. A proposed mechanism for the pulse in carbon dioxide production commonly observed following the rapid rewetting of a dry soil. Soil Sci. Soc. Am. J. 67:798–805. Fuller, E.N., P.D. Schettler, and J.C. Giddings. 1966. A new method for prediction of binary gas-phase diffusion coefficient. Ind. Eng. Chem. 58:19–27. Glinski, J., and W. Stepniewski. 1985. Soil aeration and its role for plants. CRC Press, Boca Raton, FL. Gut, A., A. Blatter, M. Fahrni, B.E. Lehmann, and T. Staffelbach. 1998. A new membrane tube technique (METT) for continuous gas measurements in soil. Plant Soil 198:79–88. Gut, A., A. Neftel, T. Staffelbach, M. Riedo, and B.E. Lehmann. 1999. Nitric

133

oxide flux from soil during the growing season of wheat by continuous measurements of the NO soil-atmosphere concentration gradient: A process study. Plant Soil 216:165–180. Gut, A., S.M. van Dijk, M. Scheibe, U. Rummel, M. Welling, C. Ammann, F.X. Meixner, G.A. Kirkman, M.O. Andreae, and B.E. Lehmann. 2002. NO emission from an Amazon rain forest soil: Continuous measurements of NO flux and soil concentration. J. Geophys. Res. 107:1–10. Ham, J.M., and A.K. Knapp. 1998. Fluxes of CO2, water vapor, and energy from a prairie ecosystem during the seasonal transition from carbon sink to carbon source. Agric. For. Meteorol. 89:1–14. Hashimoto, S., and M. Suzuki. 2002. Vertical distributions of carbon dioxide diffusion coefficients and production rates in forest soils. Soil Sci. Soc. Am. J. 66:1151–1158. Heitman, J.L., J.M. Basinger, G.J. Kluitenberg, J.M. Ham, J.M. Frank, and P.L. Barnes. 2003. Field evaluation of the dual-probe heat-pulse method for measuring soil water content. Vadose Zone J. 2:552–560. Hirano, T., H. Kim, and Y. Tanaka. 2003. Long-term half-hourly measurement of soil CO2 concentration and soil respiration in a temperate deciduous forest. J. Geophys. Res. 108:1–13. Jassal, R.S., T.A. Black, G.B. Drewitt, M.D. Novak, D. Gaumont-Guay, and Z. Nesic. 2004. A model of the production and transport of CO2 in soil: Predicting soil CO2 concentrations and CO2 efflux from a forest floor. Agric. For. Meteorol. 124:219–236. Jassal, R., A. Black, M. Novak, D., K. Morgenstern, Z. Nesic, and D. GaumontGuay. 2005. Relationship between soil CO2 concentrations and forestfloor CO2 effluxes. Agric. For. Meteorol. 130:176–192. Kaleita, A.L., J.L. Heitman, and S.D. Logsdon. 2005. Field calibration of the Theta probe for Des Moines Lobe soils. Appl. Eng. Agric. 21:865–870. Karunatilake, U.P., and H.M. van Es. 2002. Rainfall and tillage effects on soil structure after alfalfa conversion to maize on a clay loam soil in New York. Soil Tillage Res. 67:135–146. Lai, S.H., J.M. Tiedje, and A.E. Erickson. 1976. In situ measurement of gas diffusion coefficient in soils. Soil Sci. Soc. Am. J. 40:3–6. Liang, N., T. Nakadi, T. Hirano, L. Qu, T. Koike, Y. Fujinuma, and G. Inoue. 2004. In situ comparison of four approaches to estimating soil CO2 efflux in a northern larch (Larix kaempferi Sarg.) forest. Agric. For. Met. 123:97–117. McIntyre, D.S., and J.R. Philip. 1964. A field method for measurement of gas diffusion into soils. Aust. J. Soil Res. 2:133–145. Millington, R.J., and J.M. Quirk. 1961. Permeability of porous solids. Trans.

134

Faraday Soc. 57:1200–1207. Moldrup, P., T. Olesen, P. Schjønning, T. Yamaguchi, and D.E. Rolston. 2000. Predicting the gas diffusion coefficient in undisturbed soil from soil water characteristics. Soil Sci. Soc. Am. J. 64:94–100. Ochsner, T.E., R. Horton, and T. Ren. 2001. A new perspective on soil thermal properties. Soil Sci. Soc. Am. J. 65:1641–1647. Ochsner, T.E., R. Horton, and T. Ren. 2003. Use of the dual-probe heat-pulse technique to monitor soil water content in the vadose zone. Vadose Zone J. 2:572–579. Parkin, T.B., and T.C. Kaspar. 2003. Temperature controls on diurnal carbon dioxide flux: Implications for estimating soil carbon loss. Soil Sci. Soc. Am. J. 67:1763–1772. Pumpanen, J., H. Ilvesniemi, and P. Hari. 2003. A process-based model for predicting soil carbon dioxide efflux and concentration. Soil Sci. Soc. Am. J. 67:402–413. Risk, D., L. Kellman, and H. Beltrami. 2002. Carbon dioxide in soil profiles: Production and temperature dependence. Geophys. Res. Lett. 29:1–4. Rolston, D.E. 2002. The soil gas phase. p. 1113–1139. In J.H. Dane and G.C. Topp (ed.) Methods of soil analysis. Part 4. Physical methods. SSSA Book Ser. 5. SSSA, Madison, WI. Rolston, D.E., R.D. Glauz, G.L. Grundmann, and D.T. Louie. 1991. Evaluation of an in situ method for measurement of gas diffusivity in surface soils. Soil Sci. Soc. Am. J. 55:1536–1542. Schwendenmann, L., and E. Veldkamp. 2006. Long-term CO2 production from deeply weathered soils of a tropical rain forest: Evidence for a potential positive feedback to climate warming. Global Change Biol. 12:1878–1893. Tang, J., D.D. Baldocchi, Y. Qi, and L. Xu. 2003. Assessing soil CO2 efflux using continuous measurements of CO2 profiles in soils with small solidstate sensors. Agric. For. Meteorol. 118:207–220. Tang, J., L. Misson, A. Gershenson, W. Cheng, and A.H. Goldstein. 2005. Continuous measurements of soil respiration with and without roots in a ponderosa pine plantation in the Sierra Nevada mountains. Agric. For. Meteorol. 132:212–227. Tarara, J.M., and J.M. Ham. 1997. Measuring soil water content in the laboratory and field with dual-probe heat-capacity sensors. Agron. J. 89:535–542. Turcu, V.E., S.B. Jones, and D. Or. 2005. Continuous soil carbon dioxide and oxygen measurements and estimation of gradient-based gaseous flux. Vadose Zone J. 4:1161–1169.

SSSAJ: Volume 72: Number 1 • January–February 2008