HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (12) Citing Articles via Google Scholar Google Scholar Articles by Widén, B. Articles by Lindroth, A. Search for Related Content PubMed Articles by Widén, B. Articles by Lindroth, A. Agricola Articles by Widén, B. Articles by Lindroth, A. Related Collections Soil Methods/Instrumentation Flow Soil Biochemistry

DIVISION S-7—FOREST & RANGE SOILS

A Calibration System for Soil Carbon Dioxide-Efflux Measurement Chambers

Description and Application

^a Department for Production Ecology, Faculty of Forestry, P.O. Box 7042, Swedish University of Agricultural Sciences, 750 07, Uppsala, Sweden
^b Department of Physical Geography and Ecosystems Analysis, P.O. Box 118, Lund University, 221 00, Lund, Sweden

* Corresponding author (Anders.Lindroth{at}nateko.lu.se)

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Comparisons have revealed large discrepancies among the many methods for measuring soil CO₂ efflux indicating the need for an absolute calibration of methods. This study presents a calibration system, constructed to imitate an area of soil, and its application to two different chamber systems for the measurement of soil CO₂ efflux: one open and one closed dynamic. Air rich in CO₂ was allowed to diffuse through a layer of sand on top of a box of known volume. By measuring the decrease in CO₂ concentration inside the box, the exact CO₂ efflux could be calculated. The CO₂–efflux rates measured by the chambers could then be compared with the efflux rates calculated from the box. The error of the closed-chamber system ranged from an underestimate of 19% to an overestimate of 21%. The errors were most likely caused by a combination of underestimated chamber volume, causing an underestimation of CO₂ efflux, and turbulence within the chamber, which increased the flux by disturbing the boundary layer above the surface. The open-chamber system always overestimated the CO₂ efflux. Disturbing the boundary layer alone was believed to cause a 17% increase in efflux. Increasing negative pressure difference caused a mass flow of CO₂–rich air into the chamber. At a pressure difference of -0.15 Pa, the error was 11 to 40%, depending on air-filled soil volume. Accordingly, soil-water content, a parameter to which soil CO₂ efflux is often related, was found to substantially affect the measurements made by both tested systems. These results point to the need of calibrating systems used for measuring soil CO₂ efflux is measured against a known flux, to elucidate the limits and applicability of each system.

Abbreviations: DP34, dry sand of 34% porosity • DP60, dry sand of 60% porosity • IRGA, infrared-gas analyzer • F_b, CO₂ flux from the calibration box with a chamber on top • F_br, CO₂ flux from the calibration box with no chamber on top—reference flux • F_b-Co, CO₂ flux from the calibration box outside the chamber • F_Cc, CO₂ efflux measured by a closed-dynamic chamber system • F_Co, CO₂ efflux measured by an open chamber system • LOI, loss on ignition • WP34, wet sand of 34% porosity

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
SOIL CO2 EFFLUX is one of the largest fluxes within the global C cycle. It is of the same order of magnitude as global gross primary production (Raich and Schlesinger, 1992). Accurate measurement of soil CO₂ efflux is therefore essential when making C budgets. Besides the large spatial variability common to soil CO₂ efflux and the associated problems of scaling (Janssens and Ceulemans, 1998), it is difficult to obtain a reliable and correct value of the flux at the point of measurement (Lund et al., 1999).

Soil CO₂ efflux is the result of two processes: production, mainly via root and microbial respiration; and transport from the source into the atmosphere. Transport of CO₂ takes place both under the influence of concentration gradients—diffusion flow—and under the influences of pressure gradients—mass flow (Glinski and Stepniewski, 1985). Thus, soil CO₂ efflux can only be measured accurately by a system that does not affect CO₂ production and either does not affect transport, or allows for estimation of the flux before transport was affected. The latter has, however, been shown to depend largely on the mathematical method used (Pedersen et al., 2001).

Several techniques have been used to measure soil CO₂ efflux, including soil CO₂ concentration profiles, eddy-covariance, and chamber methods. The former two methods avoid confounding chamber effects, but their applicability has generally been limited by their methodological requirements (e.g., de Jong and Shappert, 1972; Baldocchi and Meyers, 1991). The many different chamber methods for measuring soil CO₂ efflux exhibit large differences in applicability, spatial and temporal resolution, and to a large extent, accuracy. Several comparisons between methods have been made (Rochette et al., 1992; Norman et al., 1997; Rochette et al., 1997; Janssens et al., 2000; Longdoz et al., 2000). However, no method has yet been recognized as standard. All in situ comparisons will only give relative answers, such as which system measures higher or lower rates than the others. Several studies have also been made on how fluxes measured with a certain chamber change with changes, for example, in airflow rate through the chamber or details in chamber design (Longdoz et al., 2000; Fang and Moncrieff, 1998; Rayment and Jarvis, 1997; Hutchinson and Mosier, 1981; Welles et al., 2001). Nevertheless, these studies do not provide an answer regarding how accurate the measurements are compared with the true efflux.

The large discrepancy between methods calls for an absolute calibration so that accuracy, limitations, design, and applicability can be evaluated. Bekku et al. (1997) compared CO₂ fluxes measured by four different methods to a known CO₂ flux. Unfortunately, the chambers of the systems tested covered the whole surface from which CO₂ was emitted, and several chamber effects have thus been avoided; the conclusions to be drawn from the study are therefore limited. Up to now, only Nay et al. (1994) have compared fluxes measured by chamber systems with a known flux from a larger surface, that is, a proper, absolute calibration.

The aim of the present study was to construct a calibration system imitating the process of CO₂–diffusion through the soil. We used the calibration system to test two different methods for measuring soil CO₂ efflux: one commercial closed-chamber system (Li-Cor 6200-09, Li-Cor Inc., Lincoln, NE) and one open system (Iritz et al., 1997). The study was performed in the laboratory at the Department for Production Ecology, Faculty of Forestry, Swedish University of Agricultural Sciences, in 1998. Several experiments were conducted with the two chamber systems to elucidate the importance of airflow rate, and of the underlying material and boundary layer. Suggestions for improvement of the calibration system are also given.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The calibration system
The calibration system consisted of a diffusion box and a gas analyzer (Fig. 1) . The box, 198 by 114 by 28 cm (length by height by width), was made of 0.7-cm thick polyvinyl chloride (PVC) plastic plates sealed with silicone rubber. The lower part of the box was empty, except for three fans to mix the air, and an air sampling tube. At the top of the box was a 5-cm thick layer of quartz sand, supported by an aluminum grating covered with a stretched aluminum mesh and cotton gauze. The bars of the grating were 20 by 2 mm with the short end toward the sand layer. Two types of sand were used. The first was a well-sorted sand, referred to as DP60, with grains 1 to 2 mm in diameter, a porosity of 60%, and a loss-on-ignition (LOI) of <4 g kg^-1. The other sand, referred to as DP34, or if wet WP34, was unsorted with grain diameters of up to 5 mm. The porosity was 34% and the LOI <6 g kg^-1. A differential pressure transmitter with a resolution of 0.03 Pa (PX653, Omega Engineering Inc., Stamford, CT) was used to measure the pressure difference between the inside and the outside of the chambers. Connected to the pressure gauge and buried in the sand, were two tubes that had outlets exactly at the surface of the sand, one inside the chamber and one outside. In this paper, a lower pressure within the chamber than outside the chamber is denoted a negative pressure difference. At the place where the experiments were performed air movement over the surface of the box was close to nil. In some experiments the sand was wetted. Four Theta Probes (type ML1, Delta-T Devices Ltd, Cambridge, England) monitored the sand-water content. Pressure difference, and sand-water content data were stored on a data logger (CR10, Campbell Scientific Ltd., Shepshed, England), as 5-min averages.

View larger version (30K): [in this window] [in a new window]   Fig. 1. The calibration system with parts of the closed-dynamic system (Li-Cor 6200-09), and open-dynamic system.

To measure the decrease in CO₂ concentration within the box, air was sampled with a tube that extended diagonally downwards across the box. To maintain equal suction along the tube, the total area of the perforations was equal to the cross-sectional area of the tube. Air drawn into the tube from the box was pumped to an infrared-gas analyzer (IRGA) (Li-6262, Li-Cor Inc., Lincoln, NE), and then returned to the box. The CO₂ concentration of sampled air was measured against pure N₂ gas every third second and data was stored on a data logger (CR10, Campbell Scientific Ltd., Shepshed, England) as 5-min averages.

The flux per unit area out of the box was calculated as a discrete function:

[1]

where C_t is the concentration inside the box at time t, V is the volume, and A is the horizontal area of the box. To check the validity of the function, the initial total amount of CO₂ was calculated (C₀V) and compared with the accumulated sum, C_T, of the discrete function F_b:

[2]

where t_n denotes the time at which the CO₂ concentration inside the box has reached ambient level. The difference was typically <0.45%.

As the flux rate is a function of the concentration gradient, according to Fics law of diffusion, the CO₂–efflux rate varied with the CO₂ concentration in the box. Each experiment therefore provided a range of fluxes for testing the chamber methods, from a minimum flux of 0 to a maximum flux of 4 to 12 µmol m^-2 s^-1, depending on the type of sand used. The CO₂–efflux rates estimated by the two different chamber methods were then compared with the CO₂ efflux from the box, F_b.

Closed-Dynamic System
The calibration of the closed system was straightforward; only the type of sand was changed between calibration runs. The closed-dynamic system tested was a Li-Cor soil respiration system, Li-6200-09 (Li-Cor Inc., Lincoln, NE). This system has been used in several studies, and is thoroughly described in Norman et al. (1992). The system consists of an IRGA and control unit, and a chamber. The chamber had a cross-sectional area of 71.6 cm², and the total volume of the system, including the chamber, the tubing and IRGA volume, was 1116 cm³. Air was pumped from the top of the chamber to the IRGA at a rate of approximately 0.7 L s^-1, and then returned at the bottom of the chamber. Carbon-dioxide efflux, F_Cc, was calculated from the increase in CO₂ concentration over time.

[3]

where dC/dt is the change in CO₂ concentration over time (mol mol^-1 s^-1), V is the volume of the system (m³), is the air pressure (atm), R is the molar gas constant (m³ atm mol^-1 K^-1), T is the temperature (K), and A is the soil surface area (m²). Before measurements, the CO₂ concentration was scrubbed down below ambient. Measurements were made from about 15 mol mol^-1 below to 15 mol mol^-1 above ambient. In this way, biases because of altered CO₂ concentration gradients above the surface are minimized (compare Nay et al., 1994). Measurements were made on three previously installed collars with an area of 85 cm². When using the coarse, well-sorted sand, the collars were inserted to a 5-cm depth, but when using the unsorted sand, they were inserted only to a 2-cm depth. In both cases the chamber, when placed on a collar, was 0.5 cm above the sand surface, and the outlet of air 1.5 cm above the sand surface. One measurement was made on every collar at increasing time intervals to cover the range of efflux rates. To test whether the flux was significantly different from the flux from the calibration box, parameters of a regression-line fitted through the measurement points were tested against the 1:1 line.

Open-Chamber System
This chamber was of the open type and designed to maintain near-ambient conditions within the chamber. The system has been described earlier in Iritz et al. (1997) and Morén and Lindroth (2000) and in technical detail in Morén and Lindroth (1999). The chamber was 30 cm wide and 200 cm long. In some of the experiments in this study, only half the chamber was used, (i.e., the chamber was 100 cm long). One end of the chamber was open, whereas the other was connected to an aluminum tube (21 cm in diam.). A fan in the tube drew air through the chamber. Baffles were placed behind the fan to mix the air before sampling, and a propeller anemometer was inserted to measure the airflow. The fan could be regulated to create velocities in the range of about 0.44 to 0.61 m s^-1 through the chamber. Air was sampled through perforated tubes, which extended across the inlet to the chamber and the outlet of the aluminum tube. The CO₂ efflux, F_Co, was calculated from the difference in CO₂ concentration between the incoming and outgoing air, measured with an IRGA (Li-6262, Li-Cor Inc., Lincoln, NE), the volumetric flow and cover soil area.

As open systems are more liable to cause pressure differences between the inside and the outside of the chamber, several experiments were made to elucidate the importance of airflow rate, and of the underlying material and boundary layer. The experiments made with the open chamber are summarized in Table 1.

Experiment 2—Effect of Air-Filled Volume
The effect of pressure differences on the CO₂ efflux may also be influenced by the characteristics of the underlying material. In the second experiment, both types of sand were used, and the unsorted sand (DP34) was wetted to different levels of water-content to create underlying materials of different air-filled volume (WP34). As it was not possible to wet the sand absolutely evenly, the chamber was moved from one half of the calibration box to the other every second hour, to cover the whole surface and to minimize biases caused by uneven water content. Because it was not feasible to achieve exactly equal soil water content twice, there was no possibility of obtaining a proper reference flux. Therefore, the flux measured by the chamber, F_Co, was compared with the flux from the calibration box when the chamber was on top and running, F_b. If the chamber, which covered half of the surface of the calibration box, did not affect the flux, half of the CO₂ leaving the calibration box should enter into the chamber. This is referred to as the expected flux.

Experiment 3—Disturbance of Boundary Layer
The effect of moving air on the CO₂ efflux can be ascribed partly to a negative pressure difference, drawing CO₂–rich air from surrounding area, and partly to the disturbance of the boundary layer. First, a reference flux, F_br, was estimated by running the calibration system with no chamber on top of the calibration box. The 200 cm long chamber, which covered the entire calibration chamber, was then put on top and left running while the flux from the calibration box, F_b, was measured. The flux from the calibration box when the chamber was on top and running, F_b, was then compared with flux from the calibration box with no chamber on top, F_br, at the same CO₂ concentration inside the calibration box. As there was then no surrounding area from which to draw CO₂–rich air, a major part of the difference was assumed to be caused by disturbance of the boundary layer.

Experiment 4—Accuracy of Sampling
Finally, if not captured accurately by the sampling, the boundary layer may cause an error in the measurements. If the inlet of the chamber is placed at the edge of the calibration box, only well-mixed air with an ambient CO₂ concentration will enter and it is less important that air be sampled across the whole inlet. However, if the inlet to the chamber is placed further away from the edge, CO₂–rich air closest to the surface may enter the chamber without being sampled and the CO₂–efflux rates will be overestimated. To discover whether this was the case, the chamber was moved to different positions so that its inlet was 0, 50, or 100 cm from the edge of the calibration box. The flux measured by the chamber when placed at one spot was then compared with the measured flux when the chamber was placed at another spot.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Closed-Dynamic System
Neither of the intercepts of the regressions, fitted through measured fluxes versus flux from the calibration box (Fig. 2) , was significantly different from 0 on a 5% level, which implies that any errors were proportional to the flux. On the coarse well-sorted sand, the closed-dynamic system underestimated the CO₂ efflux by 19% (p < 0.05, Fig. 2). On the unsorted sand, F_Cc was not significantly different from F_b when the sand was dry. When the sand was wetted to a water content of approximately 0.06 m³ m^-3, the closed-dynamic system overestimated the CO₂ efflux by 21% (p < 0.05). No pressure difference was seen between the inside and the outside of the chamber.

View larger version (22K): [in this window] [in a new window]   Fig. 2. The CO2 flux measured by the closed-dynamic system compared with the reference flux for three different types of soil.

View larger version (21K): [in this window] [in a new window]   Fig. 3. The CO2 flux measured by the open system and the flux from the calibration box outside the chamber, relative to the reference flux at increasing pressure difference between the inside and the outside of the chamber. Also included is the relative flux at different pressure differences as measured by Fang and Moncrieff (1998).

View larger version (21K): [in this window] [in a new window]   Fig. 4. The CO2 flux measured by the open chamber relative to the expected CO2 flux as the soil air-filled volume changes, using different types of soil.

View larger version (22K): [in this window] [in a new window]   Fig. 5. The CO2 efflux from the calibration box at different inside CO2 concentrations. First, when the open chamber, which covered the whole surface, was on top and running and secondly when no chamber was present.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

Air is circulated through a closed-dynamic chamber, and in the present study it entered the chamber 1.5 cm above the sand surface. Turbulence inside chambers, induced by the circulating air, influences CO₂ flux directly by the pumping action of pressure fluctuations and indirectly by altering the thickness of the viscous boundary layer and thus the concentration gradient (Hanson et al., 1992). When the thickness of the boundary layer increases, the CO₂ concentration closest to the surface and in the upper soil air increases. If the thickness of the boundary layer differs between the inside and outside of the chamber, CO₂ will diffuse laterally, and either increase or decrease the measured CO₂ flux (Le Dantec et al., 1999). Excessive turbulence within a chamber may thus increase CO₂ efflux. The circulating air within the closed-dynamic chamber tested in the present study may have counteracted the error caused by underestimated volume, which then should have been even larger. However, the effect of turbulence increases with increasing soil porosity. To hide the effect of turbulence, the overestimation should have been even larger and the affected soil depth deeper than the depth of the sand layer, which is what is believed to have occurred.

Since CO₂ is as soluble in water as in air, some CO₂ will dissolve in the soil water during the equilibration phase. The diffusivity in water is, however, about 1/1000 times of that in air, which means that there will be a time-lag between the flow of CO₂ from the air in the calibration box and the flow of CO₂ from the soil water. However, in this study the total water volume is 0.5% of the total volume of the box, and the error is therefore believed to have been negligible. Also, the effect of soil water acted in different directions for the closed and open systems. Another potential error, connected to the soil-water content, is that increasing water vapor within the chamber may dilute the increase in CO₂ concentration such that soil CO₂ efflux is underestimated (Welles et al., 2001). Increasing water vapor within the chamber was not accounted for when calculating fluxes, but using the equation of Welles et al. (2001), the error was found to be <0.6%.

Although the sand we used was more homogeneous than many naturally occurring soils, the flux may not be exactly the same all over the surface of the calibration box. Grace, performing a similar test of systems for measurement of soil CO₂ efflux, found the spatial variation in CO₂ efflux from the calibration box to be larger than any expected errors of the systems tested (J. Grace, University of Edinburgh, UK, personal communication, 2001). This was not the case with the calibration system presented in this study as can be seen in Fig. 2. Each point represents an average of three measurements at different collars and the standard deviation is small enough to exclude the 1:1 line for both DP60 and WP34. The edge effects were, however, not satisfactorily tested. Some extra CO₂ may have emerged around the edges of the calibration box, a feature which is both difficult to prevent and difficult to measure since no collar can be placed on the very edge.

When the open chamber was placed on top of the whole calibration box in Exp. 3, in such a way that no mass flow of CO₂–rich air from surrounding areas was possible, F_b increased by 17% (Fig. 5). This suggests that the turbulent air alone affected and diminished the boundary layer in the open chamber, such that efflux increased.

In Exp. 1, using the 100-cm long chamber and an airflow rate of 0.44 m s^-1, the efflux was 40% larger than expected. Some of this error can be ascribed to excessive turbulence (Fig. 3), but most of the remaining error was probably caused by the -0.15 Pa pressure difference, which induces a mass flow of CO₂ into the chamber. This error increases with increasing negative pressure difference. As noted by Fang and Moncrieff (1996), the evolution of CO₂ is relatively less sensitive to a positive pressure difference than to a negative pressure difference. At a pressure difference of 0.2 Pa, Fang and Moncrieff (1998) measured a CO₂ efflux 0.55 times that under no pressure difference, whereas a pressure difference of -0.2 Pa caused an observed CO₂ efflux 2.5 times that found under no pressure difference. This is much larger than the errors we observed in the present study (Fig. 3).

The different results between our study and Fang and Moncrieff (1998) could be caused by a difference in soil characteristics. For example, the increase in measured CO₂ efflux caused by a negative pressure difference from a soil of large air-filled volume will be much larger than that from a soil of low air-filled volume (Fang and Moncrieff, 1998). Consequently, soil-water content would influence the effect of pressure difference, which was shown in the present study in Exp. 2. The measured CO₂ efflux decreased with an increase in soil-water content as long as the same type of sand was used (Fig. 4). Perhaps not only the porosity per se, but also the tortuosity is important. Lund et al. (1999) found that increased soil-water content diminished the effects a positive pressure had on soil CO₂ efflux within a chamber. Increased soil-water content will also diminish the soil air that adds to the true volume of a closed-dynamic system, causing an increase in estimated CO₂ efflux.

An implication of the effect of soil-water content on the measured CO₂ efflux is that relationships between soil-water content and soil CO₂ efflux previously reported (e.g., Davidson et al., 1998; Epron et al., 1999; Morén and Lindroth, 2000) must be treated with caution. As soil-water content decreases, the CO₂ efflux measured by an open system is overestimated more (provided there is a pressure difference between the inside and the outside of the chamber), whereas the CO₂ efflux measured by a closed-dynamic system is underestimated more. Ventilation within chambers will increase evaporation and dry the soil, increasing the error. When making point measurements this is not a problem, but for automatic systems where measurements are made continuously, it should be considered. Perhaps the error caused by decreasing soil-water content could also explain some of the large differences in the relationship between soil moisture and soil CO₂ efflux reported by others (compare Schlentner and van Cleve, 1985; Davidson et al., 1998; Epron et al., 1999; Morén and Lindroth, 2000).

The magnitude of the impact of the pressure differences and turbulence, discussed above, on estimated CO₂ efflux in a natural environment, is difficult to determine. Above the soil surface within a forest canopy, wind speed is seldom higher than the wind speed through the open chamber. Thus the boundary layer within the chamber, when placed in a forest, would be thinner than that outside and the concentration gradient steeper, leading to increased soil CO₂ efflux. The difference in wind speed between the inside and the outside of the chamber would be at its greatest at night, when conditions normally are calmer. The conditions prevailing in the laboratory in the present study can be compared with a forest at night, when wind is almost absent. Le Dantec et al. (1999), using the same type of closed-dynamic chamber as in the present study, measured a wind speed of 0.4 m s^-1 at an airflow of 1.6 L min^-1. In the present study the airflow was kept at 0.7 L min^-1, which should then correspond to a wind speed of approximately 0.2 m s^-1 through the chamber. This is also higher than normal wind speeds within a canopy. Ideally, conditions within a chamber for measuring soil CO₂ efflux should mimic ambient conditions perfectly. Thus the speed of air through the chamber should vary in time as wind condition changes.

Despite a long history of soil CO₂ efflux measurements, this process remains one of the most difficult to measure in an accurate and appropriate manner (e.g., Lund et al., 1999; Longdoz et al., 2000). No system, not even the eddy-correlation and soil CO₂–profile methods that avoid chamber effects, are without objections. To be able to know what we are measuring, all systems used should be properly tested. This means that they should be calibrated in an absolute manner. The calibration system presented in this study provides such an absolute calibration of systems for measurement of soil CO₂ efflux. Some improvements can, however, be made. It would, for example, be good to make the box much larger so as to minimize the edge effects and to improve the calibration of large chamber systems. Further, the use of water should perhaps be avoided since this may cause errors, although in this case they were believed to be negligible. Another improvement would be to make the soil layer thicker, so that the effect of chamber placement on the soil CO₂ profile could be better studied. The benefit of a thin soil layer is that the decrease in CO₂ concentration within the box can be assumed to be the same as the flux from the surface of the soil layer. One solution would be to keep the CO₂ concentration within the box constant and estimate the flux from the necessary amount of CO₂ added to keep it constant. This would require a gas analyzer that can measure much higher concentrations than the Li-Cor 6262 and a rather constant environment around the box. If there is a large variation in advective transport of air within the soil, the flux from the surface of the box may no longer be the same as the CO₂ added. On the other hand, to what extent chamber methods are able to measure fluxes in an environment where wind and pressure fluctuations causes large advective transport, is of great interest. We have, at this time, no solution to this problem, but the calibration box in the present study is a step toward the development of a reliable system for soil CO₂ efflux measurement.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 

• Turbulence within a chamber may disturb the boundary layer such that soil CO₂ efflux is increased. In the very calm environment prevailing where the study was performed, it is believed that disturbance of the boundary layer increased the measured CO₂ efflux in both the closed-dynamic and open-chamber system, in the latter with 17%. This implies that during calm conditions, chambers with little air movement may be the best alternative.
  
• Soil-water content seriously affected the measurements made by both the open and the closed-dynamic chamber systems tested. Fluxes measured by the closed-dynamic system were higher on wet sand than on dry sand. As fluxes further decreased when sand of higher porosity was used it is believed that some soil air, which is affected by both porosity and soil-water content, added to the volume of the system. The flux measured by the open-chamber system decreased with increasing soil water content, as a result of decreasing effect of the negative pressure difference. The results indicate that both systems works best at high soil-water content; the open-chamber system should provide reliable measurements at an air-filled soil volume below 0.22 m³ m^-3. The results also imply that previously reported relationships between soil-water content and soil CO₂ efflux must be treated with caution.
  
• No perfect chamber system is likely to exist. Each system used ought therefore to be calibrated against some known flux to elucidate the limits and applicability of the system. The calibration system presented in this study provides such an absolute calibration of systems for measurement of soil CO₂ efflux, although some improvements can be made.
  

	ACKNOWLEDGMENTS

 
This project was largely funded by the Swedish National Energy Administration, which is gratefully acknowledged. Contributions were also obtained from the EU within the EUROFLUX and CARBOEUROFLUX projects. Thanks to Mattias Lindberg for building the calibration box, as well as for advice on its design.

Received for publication May 9, 2001.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 

• Baldocchi, D.D., and T.P. Meyers. 1991. Trace gas exchange above the floor of a deciduous forest 1. Evaporation and CO₂ flux. J. Geophys. Res. 96:7271–7285.
• Bekku, Y., H. Koizumi, T. Oikawa, and I. Iwaki. 1997. Examination of four methods for measuring soil respiration. Appl. Soil Ecol. 5:247–254.
• Davidson, E.A., E. Belk, and R.D. Boone. 1998. Soil water content and temperature as independent or confounded factors controlling soil respiration in a temperate mixed hardwood forest. Global Change Biol. 4:217–227.
• de Jong, E., and J.V. Shappert. 1972. Calculation of soil respiration and activity from CO₂ profiles in the soil. Soil Sci. 113:328–333.
• Epron, D., L. Fraque, E. Lucot, and P.-M. Badot. 1999. Soil CO₂ efflux in a beech forest: Dependence on soil temperature and soil water content. Ann. For. Sci. 56:289–295.
• Fang, C., and J.B. Moncrieff. 1996. An improved dynamic chamber technique for measuring CO₂ efflux from the surface of soil. Funct. Ecol. 10:297–305.
• Fang, C., and J.B. Moncrieff. 1998. An open-top chamber for measuring soil respiration and the influence of pressure difference on CO₂ efflux measurement. Functional Ecol. 12:319–325.
• Glinski, J., and W. Stepniewski. 1985. Soil aeration and its role for plants. CRC Press, Boca Raton, FL.
• Hanson, P.J., S.D. Wullschleger, S.A. Bohlman, and D.E. Todd. 1992. Seasonal and topographic patterns of forest floor CO₂ efflux from an upland oak forest. Tree Physiol. 13:1–15.
• Hutchinson, G.L., and A.R. Mosier. 1981. Improved soil cover method for field measurement of nitrous oxide fluxes. Soil Sci. Soc. Am. J. 45:311–316.[Abstract/Free Full Text]
• Iritz, Z., A. Lindroth, and A. Gärdenäs. 1997. Open ventilated chamber for measurements of H₂O and CO₂ fluxes from the soil surface. Soil Technol. 10:169–184.
• Janssens, I.A., and R. Ceulemans. 1998. Spatial variability in forest soil CO₂ efflux assessed with a calibrated soda lime technique. Ecol. Lett. 1:95–98.
• Janssens, I.A., A.S. Kowalski, B. Longdoz, and R. Ceulemans. 2000. Assessing forest soil CO₂ efflux: An in situ comparison of four techniques. Tree Physiol. 20:23–32.[Medline]
• Le Dantec, V., D. Epron, and E. Dufrêne. 1999. Soil CO₂ efflux in a beech forest: Comparison of two closed dynamic systems. Plant Soil 214:125–132.
• Longdoz, B., M. Yearnaux, and M. Aubinet. 2000. Soil CO₂ efflux measurements in a mixed forest: Impact of chamber disturbances, spatial variability and seasonal evolution. Global Change Biol. 6: 907–917.
• Lund, C.P., W.J. Riley, L.L. Pierce, and C.B. Field. 1999. The effects of chamber pressurization on soil-surface CO₂ flux, and the implications for NEE measurements under elevated CO₂. Global Change Biol. 5:269–281.
• Morén, A.-S., and A. Lindroth. 1999. Field measurements of water vapour and carbon dioxide fluxes- chamber system and climatic monitoring by an automatic station. Report No. 4, Department for Production Ecology, Faculty of Forestry, Swedish University of Agricultural Sciences, Uppsala, Sweden.
• Morén, A.-S., and A. Lindroth. 2000. CO₂ exchange at the floor of a boreal forest. Agric. For. Meteorol. 101:1–14.
• Nay, S.M., K.G. Mattson, and B.T. Bormann. 1994. Biases of chamber methods for measuring soil CO₂ efflux demonstrated with a laboratory apparatus. Ecology 75:2460–2463.[ISI]
• Norman, J.M., Garcia, R., and S.B. Verma. 1992. Soil surface CO₂ fluxes and the carbon budget of a grassland. J. Geophys. Res. 97:18845–18853.
• Norman, J.M., Kucharik, C.J., Gower, S.T., Baldocchi, D.D., Crill, P.M., Rayment, M., Savage, K., and R.G. Striegl. 1997. A comparison of six methods for measuring soil-surface carbon dioxide fluxes. J. Geophys. Res. 102:28771–28777.
• Pedersen, A.R., S.O. Petersen, and F.P. Vinther. 2001. Stochastic diffusion model for estimating trace gas emissions with static chambers. Soil Sci. Soc. Am. J. 65:49–58.[Abstract/Free Full Text]
• Raich, J.W., and W.H. Schlesinger. 1992. The global carbon dioxide flux in soil respiration and its relationship to vegetation and climate. Tellus 44B:81–99.
• Rayment, M.B., and P.G. Jarvis. 1997. An improved open chamber system for measuring soil CO₂ effluxes in the field. J. Geophys. Res. 102:28,779–28,784.
• Rochette, P., B. Ellert, E.G. Gregorich, R.L., Desjardins, E. Pattey, R. Lessard, and B.G. Johnson. 1997. Description of a dynamic closed chamber for measuring soil respiration and its comparison with other techniques. Can. J. Soil Sci. 77:195–203.
• Rochette, P., E.G. Gregorich, and R.L. Desjardins. 1992. Comparison of static and dynamic closed chambers for measurement of soil respiration under field conditions. Can. J. Soil Sci. 72:605–609.
• Schlentner, R.E., and K. van Cleve. 1985. Relationships between CO₂ evolution from soil, substrate temperature, and substrate moisture in four mature forest types in interior Alaska. Can. J. For. Res. 15:97–106.
• Welles, J.M., T.H. Demetriades-Shah, and D.K. McDermitt. 2001. Considerations for measuring ground CO₂ effluxes with chambers. Chem. Geol. 177:3–13.

This article has been cited by other articles:

				J. G. Martin, P. V. Bolstad, and J. M. Norman A CARBON DIOXIDE FLUX GENERATOR FOR TESTING INFRARED GAS ANALYZER-BASED SOIL RESPIRATION SYSTEMS Soil Sci. Soc. Am. J., March 1, 2004; 68(2): 514 - 518. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (12) Citing Articles via Google Scholar Google Scholar Articles by Widén, B. Articles by Lindroth, A. Search for Related Content PubMed Articles by Widén, B. Articles by Lindroth, A. Agricola Articles by Widén, B. Articles by Lindroth, A. Related Collections Soil Methods/Instrumentation Flow Soil Biochemistry