Pilot Study of Influences of Atmospheric Pressure Fluctuations on Fluxes of Trace Gases from Soils

_______________________

Eugene S. Takle, PI 3010 Agronomy Hall Iowa State University Ames, IA 50011 Tel: 515-294-9871 Fax: 515-294-2619 gstakle@iastate.edu	James R. Brandle, Co-PI School of Natural Resource Science 101 Plant Industry University of Nebraska -- Lincoln Lincoln, NE 68583-0814 Tel: (402) 472-6626 Fax: (402) 472-2964 fofw084@unlvm.unl.edu
Rick Garcia, Co-PI LI-COR Environmental Division 4421 Superior Street, Lincoln, NE 68504 Tel: 402-467-3576 Fax: 402-467-2819 rgarcia@licor.com	Bill Massman, Co-PI USDA Forest Service Rocky Mountain Research Station 240 West Prospect Ft. Collins, CO 80526 Tel: (970) 498-1296 Fax: (970) 498-1314 Wmassman/rmrs@fs.fed.us
Charles W. Rice, CO-PI Department of Agronomy Throckmorton Hall Kansas State University Manhattan, KS 66506 Tel: (785) 532-7217 cwrice@ksu.edu	Additional Collaborators: Jay Ham, Kansas State University Gerard Kluitenberg, Kansas State University Irina Litvina, Agrophysical Research Inst.
	_____________________________ Richard E. Hasbrook, Contracts and Grants Officer 20 May 1999

I. Introduction

Recent interest in carbon flux budgets in the aftermath of the Kyoto Conference have heightened the need for improved accuracy in measurements of fluxes of carbon dioxide from soils. Diffusional movement has long been considered the dominant process by which trace gases move from the subsurface source to the surface. Although bulk transports resulting from pressure fluctuations has been considered as a possible augmentation to diffusional movement, only within the last 5-10 years has there been sufficient evidence to justify a systematic examination of this possible mechanism. We believe the time has come to address this issue with a combination of field measurements and numerical models.

Although atmospheric pressure fluctuations had been suggested by Buckingham as a possible mechanism for gas movement in soils nearly a century ago, few studies –either theoretical or observational – have sought to evaluate the importance of this process, perhaps in part because Buckingham concluded that barometric pressure changes were unimportant for the upper 15 m of soil. More recent studies have suggested pressure variations in soils may arise from diurnal and semi-diurnal barometric waves, passage of synoptic weather systems, atmospheric "turbulence, "wind speed", and "quasi-static pressure fields induce by wind blowing across irregular topography".

Early discussions of the role of high frequency gas movement in soils pin the cause as atmospheric turbulence without careful consideration of the physical linkage to forcing in the soil environment. In the past decade, however, Clarke and Waddington (1991) point to both spatial and temporal variations in pressure, and Massman et al. (1997) acknowledge the role of irregular topography is setting up lateral pressure gradients. Wang and Takle (1995) calculate the patterns of static atmospheric surface pressure that are established in the vicinity of agricultural shelterbelts, and Takle and Wang (1997) briefly discuss the importance of the horizontal surface pressure gradient established by atmospheric flow through vegetation and its possible role in soil gas exchange.

These studies demonstrate the importance of heterogeneous pressure fields in promoting gas movement in porous solids and suggest that the observed large terrain-induced and vegetation-induced static pressure gradients at the ground likely lead to large spatial inhomogeneities in soil fluxes of trace gases. I am not aware of any systematic measurements of such spatial inhomogeneities. A recent report by Fan et al (1998) asserts that the central US has a large but yet unidentified sink for carbon dioxide of magnitude sufficient to account for all fossil carbon emitted in the US by combustion. The scientific and political implications of their results call for a more intensive survey of carbon fluxes from vegetation and soils in the central US. Before embarking on such a survey, it is important to understand the potential spatial heterogeneity of soil carbon dioxide emissions to ensure that measurements are representative of large areas. Such a study should include measurements of fluxes using chambers in concert with numerical simulations done in advance to identify, if possible, optimal experimental design.

II. Previous workshop

A Workshop on the Influences of Atmospheric Pressure Fluctuations on Fluxes of Trace Gases from Soils was held 20-21 April 1999 at the University of Nebraska – Lincoln. Attendees at the workshop and meeting summary are given in Appendix A. The workshop had the following objectives:

A. Determine whether a project consisting of a field experiment, with supporting theoretical work, is justified to measure the range of spatial and temporal variability of soil fluxes of trace gases introduced by atmospheric pressure fluctuations.

B. If Objective A is affirmed, what theoretical work is needed to establish an experimental design?

C. Given what we now know and acknowledging that the result of Objective B is yet to come, what is the likely scope of the field experiment for such a project?

D. From Items A-C, outline an action plan with participants, tasks, and time-tables.

The group concluded that (Item A) such a project was justified since existing evidence strongly suggest pressure influences are influencing measurements. Based on this conclusion, W. Massman has revisited the theoretical aspects of the issue and has supplied some background calculations to address Item B in section III. The scope of the field project (Item C) is given in section IV.

III. Theoretical background: Enhancement of soil diffusional flux of CO₂ by pressure pumping (W. Massman)

The total soil CO₂ flux is comprised of a diffusional component (Eqn. 1) and an advective component (Eqn. 2). To keep the present discussion simple, I will ignore some terms that are relatively nonessential and approximate as needed. I am also assuming that the pressure field is quasi-static rather than being associated with traveling turbulent structures.

Fd~n(tau)CnD(0,1)d XCO2/dz	(1)
Fp~1.518paXCO2v	(2)

where

n = 1/2 = soil porosity

t = 2/3 = soil tortuosity

C_o = 1.963 kg m^-3 = density of CO₂ at STP

D(0,1) = 0.1381 (10^-4) m² s^-1 = molecular diffusivity of CO₂ in air at STP

z = depth into soil, which by convention z < 0

X_CO2= CO₂ mol fraction (ppmV), here taken to be a function of z

the factor 1.518 is the ratio of the molecular mass of CO₂ to the molecular mass of air

p_a= 1.292 kg m^-3 = density of air at STP

v = pressure-induced advective velocity, here taken to be a function of z

X_CO2= depth average of X_CO2^v

The important assumptions for this exercise are how X_CO2 and v vary with depth. From Chuck's data I will assume that X_CO2= 10,000 ppmV at z = - 1 m and that X_CO2 = 400 ppmV at the soil surface (z = 0 m) and I will assume a LINEAR GRADIENT between the two depths. From my work and the work of others I will assume that v(z) = v_oe^2z, where v_o = 10^-4 m s^-1. Because z < 0, this is equivalent to assuming a dampening depth of about 0.5 m. The depth averaging will take place over the top meter of soil. The results are F_d ª 0.09 mg-CO₂ m^-2 s^-1 and F_v @ 0.17 mg-CO₂ m^-2 s^-1. In other words present results suggest that a quasi-static pressure field induced by an obstruction to air flow could increase the diffusional flux of CO₂ from soils by 200%. Please bear in mind that (I) this is a crude estimate only, (ii) is subject to considerable uncertainty and (iii) can vary considerably with soil characteristics.

IV. Field program

A. Overview

An intensive field project of 1-2 weeks is envisioned, with site preparation well in advance to allow the soil to lie dormant for a few weeks before measurements begin. The site will be at the Mead Agroforestry Research Facility where a wheat planting will be harvested in early July and tilled about 6 weeks prior to the experimental period. The tentative measurement period will be sometime between late August and end of September. Under these conditions the soil will likely be dry to eliminate moisture effects and the lack of vegetation will eliminate root respiration. Fence posts accommodating a 3-4 foot fence will be put in place on two lines allowing erection of a fence with a couple of possible orientations, depending on wind direction. This will allow the fence to be put in place within a couple of hours. Two LiCor 6400 samplers will be deployed, along with equipment for measuring meteorological variables, soil characteristics, soil CO₂ fluxes, and atmospheric pressure fluctuations, to test the experimental hypothesis.

B. Proposed measurements

Soil Sampling, (Charles W. Rice)

Soil samples will be collected after root sampling. The soil samples will be collected at 0-5, 5-15, 15-30, and 30-60 cm depths. Soil samples will be analyzed for inorganic N, total C and N, microbial biomass C and N, and microbial activity. In addition to the soil C and biological measurements, the soil will be characterize chemical and physically including bulk density, texture, pH, electrical conductivity, and nutrient analyses.

Microbial biomass C and N will be determined by the fumigation incubation method (Jenkinson and Powlson, 1976). Soil respiration as an index of microbial activity will be measured by incubating field moist in 160-mL serum bottles. The CO₂-C concentration in the headspace of the serum bottle is measured 4 times during a 48 h incubation. Total C and N are measured by direct combustion.

CO₂ flux measurements, (Charles W. Rice)

Carbon dioxide flux measurements will be taken during the intensive measurement period. Gas samplers (4) are constructed to collect soil air CO₂ from different depths at a single site (Burton and Beauchamp, 1994; Sotomayor and Rice, 1999). The device is constructed from a 0.75 m long, 2.54-cm o.d., polyvinyl chloride (PVC) pipe that contained stainless tubing (0.14-cm i.d., 0.32 cm o.d.) extending from the soil surface to the sampling depths (0.10, 0.25, 0.5, and 0.75 m). A half-moon notch is cut on the pipe at the soil air entry point to prevent tube blockage. Each end of the stainless tube is capped with Swagelok (Solon, OH) copper gas tight fittings with septa to prevent contact of soil air with the atmosphere and permit sampling by syringe. The multilevel gas samplers will be installed by first predrilling a soil core equivalent in depth and diameter to the multilevel sampler using a Giddings hydraulic probe (Ft.Collins, CO). Gas samples are collected routinely by first withdrawing and discarding a volume of gas equivalent to the internal volume of each tube. Second, a 5-mL sample is withdrawn (1 mL sec^-1) with a polypropylene syringe and injected rapidly into a previously evacuated 5-mL Wheaton (Millville, NJ) serum bottle crimped with a butyl rubber stopper. Three replicates of ambient air gas samples (5 mL) are collected simultaneously in the field and 5 mL of CO₂ standard are stored as samples in five separate serum bottles. These serve for calibration purposes and as quality control checks for leakage.

LiCor CO₂ measurements (Rick Garcia, James Brandle)

LiCOR has agreed to provide 1 or 2 LI-6400 portable gas exchange systems which will be setup for soil CO₂ flux measurements. Rick Garcia will devote a couple of days training one or two individuals in the use of the instruments and oversee data collection. A University of Nebraska student will conduct the measurements and participate in the analysis of results.

Meteorological measurements, (James Brandle, Eugene Takle)

Measurements of air temperature, humidity, solar radiation, wind speed, wind direction, soil temperature, soil moisture, and precipitation will be made beginning approximately 2 weeks before CO₂ flux measurements begin and continue through the intensive measurement period, with high sampling rates when fluxes are being measured. Variables will be measured at one location far from the fence, and several locations in the vicinity of the fence, both up and downwind. Standard barometric pressure will also be recorded.

Pressure measurements, (James Brandle, Eugene Takle)

Pressure perturbations introduced by the fence will be measured by a bank of sensitive pressure transducers each connected to lengths of tubing that will have open ends at approximately 10 locations at select distances both up and downwind from the fence. All sites will be referenced to an X-filter located far from the influence of the fence. Measurements will be made at 10 second intervals for at least part of the intensive measurement period to provide the dp/dt information needed for driving the pressure pumping model.

Measurement strategy

When the soil has lain dormant for a few weeks after tillage and is sufficiently dry and free of vegetation, the posts will be installed in two orientations to accommodate the fence. Also the soil sampling tubes will be installed, as will meteorological and pressure sensors on one of the fence potential orientations. An intensive measurement period will be called during a period of low winds and unchanging synoptic meteorological pattern to get background data on flux changes in the absence of pressure variations. When strong winds become imminent, another intensive measurement period will be declared for measuring fluxes under pressure fluctuations induced by the fence. Synoptic meteorological conditions will be monitored to look for a period of frontal passage or other opportunity to take data during a period of large changes in background atmospheric pressure. This will constitute a third measurement period. Nighttime periods of high winds will be particularly optimum, because radiation-induced temperature and evaporation differences will be minimized, allowing aerodynamically induced pressure fluctuations to play a larger relative role in creating fluctuations in soil trace gas concentration differences.

V. Data Analysis and Interpretation

Data from the gas sampling network, pressure sensors and meteorological stations will be combined and made available to all investigators and collaborators. Analysis will proceed in Fall 1999, with a meeting of investigators and collaborators for late 1999 to establish experimental conclusions. Conclusions and recommendations for further research will be completed by spring 2000.

Literature Cited

Buckingham, E., 1904: Contributions to our knowledge of the aeration of soils. USDA, Bureau of Soil Bull. No. 25. U. S. Government Printing Office, Washington, DC.

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.

Clarke, G. K. C., and E. D. Waddington, 1991: A three-dimensional theory of wind pumping J. Glaciol, 37, 89-96.

Clements, W. E., and M. H. Wilkening, 1974: Atmospheric pressure effects on Rn²²² transport across the earth-air interface. J. Geophys. Res. 79, 5025-5029.

Duwe, M. P., 1976: The diurnal variation of radon flux from the soil due to atmospheric pressure change and turbulence. Ph.D. Thesis University of Wisconsin.

Elberling, B., F. Larsen, S. Christensen, and D. Postma, 1998: Gas transport in a confined unsaturated zone during atmospheric pressure cycles. Water Resources Res. 34, 2855-2862

Fan, S. M. Gloor, J. Mahlman, S. Pacala, J. Sarmiento, T. Takahashi, and P. Tans, 1998: A large terrestrial carbon sink in North America implied by atmospheric and oceanic carbon dioxide data and models. Science, 282, 442-446.

Jenkinson, D.S., and D.S. Powlson. 1976. The effects of biocidal treatments on metabolism in soil. V. A method for measuring soil biomass. Soil Biol. Biochem. 8: 209-213.

Owensby, C. E., J. M. Ham, A. K. Knapp, D. Bremer, and L. M. Auen, 1997: Water vapour fluxes and their impact under elevated CO₂ in a C4-tallgreass prairie. Global Change Biology, 3, 189-195.

Kimball, B. A., 1973: Water vapor movement through mulches under field conditions Soil Sci. Soc. Am. Proc. 37, 813-818.

Kimball, B. A., and E. R. Lemon, 1972: Theory of soil air movement due to pressure fluctuations. Agric. Meteorol 9, 163-181.

Massman, W. J., R. A. Sommerfeld, A. R. Mosier, K. F. Zeller, T. J. Hehn, and S. G. Rochelle, 1997: A model investigation of turbulence-driven pressure pumping effects on the rate of diffusion of CO₂, N₂O and CH₄ through layered snowpacks. J. Geophys. Res. 102, 18,851-18,863.

Massmann, J., and D. F. Farrier, 1992: Effects of atmospheric pressures on gas transport in the vadose zone. Water Resources Res. 28, 777-791.

Schmidt, R. A., E. S. Takle, J. R. Brandle, and I. V. Litvina, 1995: Static pressure at the ground under atmospheric flow across a windbreak. Eleventh Symposium on Boundary Layers and Turbulence, Charlotte. Amer. Meteor. Soc., 517-520.

Shurpali, N. J., S. B. Verma, and R. J. Clement, 1993: Seasonal distribution of methane flux in a Minnesota peatland measured by eddy correlation. J. Geophys. Res. 98, 20,649 – 20,655.

Sotomayor, D., and C.W. Rice. 1999. Soil air CO₂ and N₂O concentrations in profiles under tallgrass prairie and cultivation. J. Environ. Qual. (In Press).

Takle, E. S., and H. Wang, 1997: Reply to comments by J. D. Wilson and C. J. Mooney on ’A Numerical Simulation of Boundary-Layer Flows Near Shelterbelts’. Boundary-Layer Meteorology, 85, 151-159.

Wang, H., and E. S. Takle, 1995: A numerical simulation of boundary-layer flows near shelterbelts. Boundary Layer Meteor., 75, 141-173