Comparison of soil carbon dioxide flux measurements by static and portable chambers in various management practices

Soil & Tillage Research 118 (2012) 123–131

Contents lists available at SciVerse ScienceDirect

Soil & Tillage Research journal homepage: www.elsevier.com/locate/still

Comparison of soil carbon dioxide flux measurements by static and portable chambers in various management practices Upendra M. Sainju *, Thecan Caesar-TonThat, Anthony Caesar USDA-ARS, Northern Plains Agricultural Research Laboratory, 1500 North Central Avenue, Sidney, MT 59270, United States

A R T I C L E I N F O

A B S T R A C T

Article history: Received 3 August 2011 Received in revised form 30 October 2011 Accepted 31 October 2011 Available online 26 November 2011

Portable chamber provides simple, rapid, and inexpensive measurement of soil CO2 flux but its effectiveness and precision compared with the static chamber in various soil and management practices is little known. Soil CO2 flux measured by a portable chamber using infrared analyzer was compared with a static chamber using gas chromatograph in various management practices from May to October 2008 in loam soil (Luvisols) in eastern Montana and in sandy loam soil (Kastanozems) in western North Dakota, USA. Management practices include combinations of tillage, cropping sequence, and N fertilization in loam and irrigation, tillage, crop rotation, and N fertilization in sandy loam. It was hypothesized that the portable chamber would measure CO2 flux similar to that measured by the static chamber, regardless of soil types and management practices. In both soils, CO2 flux peaked during the summer following substantial precipitation and/or irrigation (>15 mm), regardless of treatments and measurement methods. The flux varied with measurement dates more in the portable than in the static chamber. In loam, CO2 flux was 14–87% greater in the portable than in the static chamber from July to mid-August but 15–68% greater in the static than in the portable chamber from late August to October in all management practices. In sandy loam, CO2 flux was 10–229% greater in the portable than in the static chamber at all measurement dates in all treatments. Average CO2 flux across treatments and measurement dates was 9% lower in loam but 84% greater in sandy loam in the portable than in the static chamber. The CO2 fluxes in the portable and static chambers were linearly to exponentially related (R2 = 0.68–0.70, P  0.01, n = 40–56). Although the trends of CO2 fluxes with treatments and measurement dates were similar in both methods, the flux varied with the methods in various soil types. Measurement of soil CO2 flux by the portable chamber agreed more closely with the static chamber within 0–10 kg C ha 1 d 1 in loam soil under dryland than in sandy loam soil under irrigated and non-irrigated cropping systems. Published by Elsevier B.V.

Keywords: Soil respiration Chamber CO2 measurements Agricultural practices Soil types

1. Introduction Agricultural practices contribute about 25% of the total anthropogenic source of CO2, a greenhouse gas responsible for global warming (Post et al., 1990; Duxbury, 1994). Management practices, such as crop residue input to the soil, tillage, and cropping sequence, can emit CO2 as a result of soil organic matter and crop residue mineralization and root and microbial respiration (Curtin et al., 2000; Sainju et al., 2008, 2010). In contrast, atmospheric CO2 absorbed by plants during photosynthesis is stored in the soil as organic matter after crop residues are returned to the soil, a process known as C sequestration (Lal et al., 1995; Paustian et al., 1995). In the terrestrial ecosystem, soils are important reservoir of C containing about 1500 Pg C, which is three

* Corresponding author. Tel.: +1 406 433 9408; fax: +1 406 433 5038. E-mail addresses: [email protected], [email protected] (U.M. Sainju). 0167-1987/$ – see front matter . Published by Elsevier B.V. doi:10.1016/j.still.2011.10.020

times greater than that stored in the vegetation (Schlesinger, 1997). Agricultural soils contain around 170 Pg C to a depth of 1 m (Cole et al., 1996), out of which 54 Pg C has been estimated to be lost through CO2 emissions in the last two centuries (Paustian et al., 1995). Carbon storage in the soil is determined by the balance between the amount of plant residue C added to the soil and rate of C mineralized as CO2 emission in unmanured soil (Rasmussen et al., 1980; Peterson et al., 1998). Soil and crop management practices, such as irrigation, tillage, cropping sequence, and N fertilization can influence soil surface CO2 emissions (Curtin et al., 2000; Sainju et al., 2008). Irrigation can increase CO2 emissions compared with no irrigation by increasing soil water availability (Sainju et al., 2008), microbial activity, C mineralization, and respiration (Calderon and Jackson, 2002). Decreased tillage intensity reduces soil disturbance and microbial activity, which in turn, lowers CO2 emissions (Curtin et al., 2000). In contrast, increased tillage intensity increases CO2 emissions by increasing aeration due to greater soil disturbance (Roberts and Chan, 1990), and by physical degassing of dissolved

124

U.M. Sainju et al. / Soil & Tillage Research 118 (2012) 123–131

CO2 from the soil solution (Jackson et al., 2003). Cropping can increase CO2 emissions compared with fallow by increasing root respiration and the amount of crop residue returned to the soil (Curtin et al., 2000; Amos et al., 2005; Sainju et al., 2007, 2008). Similarly, residue quality, such as C/N ratio, can alter the decomposition rate of residue (Kuo et al., 1997), thereby influencing CO2 emissions (Al-Kaisi and Yin, 2005). Nitrogen fertilization, however, has variable effect on CO2 emissions (Mosier et al., 2006; Al-Kaisi et al., 2008). Management practices can also indirectly influence CO2 emissions by altering soil temperature and water content (Parkin and Kaspar, 2003; Amos et al., 2005; Sainju et al., 2008). Coarse-textured soil can emit greater CO2 flux than fine-textured soil (Sainju et al., 2008). Measurement of CO2 flux with the static chamber using gas chromatograph is a standard method where gas samples are collected over a certain intervals of time and flux is calculated as a result of concentration gradient over time (Hutchinson and Mosier, 1981; Liebig et al., 2010). The benefits of this method are (1) continuous measurement of CO2 flux at the same place without soil disturbance for a long time, resulting in more accurate determination of the flux and (2) measurement of all greenhouse gas (CO2, N2O, and CH4) fluxes in one gas sample at the same time (Hutchinson and Mosier, 1981; Liebig et al., 2010). Such measurements are, however, tedious, complex, and expensive. Other disadvantages of this method are the underestimation of CO2 flux due to suppression of the gas concentration gradient at the soil surface following chamber deployment and the microclimate effect inside the chamber that alter the flux (Healy et al., 1996; Rochette and Bertrand, 2007; Venterea, 2010). The infrared CO2 analyzer attached to the data logger in the portable chamber can immediately analyze CO2 flux and therefore provides a simple, rapid, and inexpensive method of measuring the flux (Sainju et al., 2008, 2010). The disadvantages of this method are (1) measurement within a short equilibration period (2 min), resulting in potential error due to flushing in of atmospheric CO2 inside the chamber, (2) determination of CO2 flux only as opposed to determination of three greenhouse gases at the same time in the static chamber method, and (3) high spatial and temporal variability (Sainju et al., 2008, 2010). Although CO2 flux measurements have been compared using static and dynamic chamber methods (Rochette et al., 1992; Nay et al., 1994; Jensen et al., 1996), little is known about their comparison of measurements in various soil types and management practices in dryland and irrigated cropping systems. We hypothesized that CO2 flux measured by the portable chamber method would be similar to that measured by the static chamber method, regardless of management practices and soil and climatic conditions. Our objective was to compare and relate CO2 flux measured by the static chamber using gas chromatograph and the portable chamber using infrared analyzer in various irrigation, tillage, cropping sequence, and N fertilization practices in sandy loam and loam soils under dryland and irrigated cropping systems in U.S. northern Great Plains. 2. Materials and methods 2.1. Experimental sites and treatments Soil CO2 fluxes were measured in plots established in 2006 in a dryland farm site 11 km west of Sidney, eastern Montana, and in 2005 in an irrigated site in Nesson Valley, western North Dakota, USA. In Sidney, the soil was Williams loam (fine-loamy, mixed, frigid, Typic Argiborolls [International classification: Luvisols]) with 350 g kg 1 sand, 325 g kg 1 silt, 325 g kg 1 clay, 1.42 Mg m 3 bulk density, and 7.2 pH at the 0–15 cm depth. Previous cropping system for the last 6 yrs was spring wheat (Triticum aestivum L.)fallow. In Nesson Valley, the soil was Lihen sandy loam (sandy,

mixed, frigid, Entic Haplustolls [International classification: Kastanozems]) with 720 g kg 1 sand, 120 g kg 1 silt, 160 g kg 1 clay, 1.51 Mg m 3 bulk density, and 7.7 pH at the 0–15 cm depth. Previous cropping history for the last 20 yrs was dominated by alfalfa (Medicago sativa L.), crested wheatgrass (Agropyron cristatum [L.] Gaertn), and western wheatgrass (Pascopyrum smithii [Rydb.] A. Love). Soil organic C concentrations at 0–5 and 5–15 cm depths before the initiation of the experiment were 13.3 and 10.6 g kg 1, respectively, in Sidney and 13.7 and 9.9 g kg 1, respectively, in Nesson Valley. In Sidney, main-plot treatments were three cropping sequences {(no-tilled continuous malt barley (Hordeum vulgaris L.) [NTCB], notilled malt barley-pea (Pisum sativum L.) [NTB-P], and conventionaltilled malt barley-fallow [CTB-F])}, each with two split-plot N fertilization rates of 0 and 80 kg N ha 1. While NTCB had only one cropping phase (malt barley), other cropping sequences had two phases in the rotation. For example, NTB-P had malt barley and pea phases and CTB-F had malt barley and fallow phases. Malt barley was planted annually in NTCB, in rotation with pea in NTB-P, and in rotation with fallow in CTB-F. Each phase of the cropping sequence occurred in every year. The 80 kg N ha 1 was the recommended rate of N fertilization to malt barley in dryland cropping systems at the experimental site. In NTCB and NTB-P, plots were left undisturbed, except for fertilizer application and planting crops in rows. The CTBF was the conventional farming system where plots were tilled with field cultivator equipped with C-shanks and 45-cm wide sweeps and coiled-toothed spring harrows with 60 cm rods. Plots were tilled to a depth of 10 cm during planting and fallow periods two to three times a year for seedbed preparation and weed control. Nitrogen fertilizer was applied at 0 or 80 kg N ha 1 to malt barley. Before applying N fertilizer, soil samples to a depth of 60 cm were tested for NO3–N content and N fertilization rates were adjusted. For pea, N fertilizer was not applied. Weeds in no-tilled treatments were controlled by applying preplant and postharvest herbicides and in conventionaltilled treatments by a combination of herbicides and conventional tillage to a depth of 10 cm as needed. Treatments were laid out in split-plot arrangement in a randomized complete block with three replications. The split plot size was 12.0 m  6.0 m. In Nesson Valley, main-plot treatment consisted of two irrigation systems (irrigated vs. non-irrigated) and split-plot treatment of five management practices (conventional-tilled malt barley with 67– 134 kg N ha 1 [CTBFN], conventional-tilled malt barley with 0 kg N ha 1 [CTBON], no-tilled malt barley-pea with 67– 134 kg N ha 1 [NTB-PN], no-tilled malt barley with 67– 134 kg N ha 1 [NTBFN], and no-tilled malt barley with 0 kg N ha 1 [NTBON]). In NTB-PN, both malt barley and pea phases were present in every year. The recommended N fertilization rates for irrigated and non-irrigated malt barley at the site were 134 and 67 kg N ha 1, respectively. The variation in N rates between irrigated and nonirrigated malt barley was due to the differences in grain yields and N uptake between irrigated and non-irrigated conditions. Soil NO3–N test to a depth of 60 cm was used to adjust N rate before applying N fertilizer. No N fertilizer was applied to pea. While plots in no-tilled treatments were left undisturbed, except for planting and applying fertilizers, plots in tilled-treatments were plowed with a rototiller and a single-pass field cultivator to a depth of 10 cm at planting. Weeds were controlled with herbicides in no-tilled plots and a combination of herbicides and tillage in tilled plots, similar to Sidney. Treatments were laid out in split-plot arrangement in a randomized complete block with three replications. The size of each experimental unit was 10.6 m  3.0 m. 2.2. Crop management In Sidney, six-row malt barley (cultivar Certified Tradition, Busch Agricultural Resources, Fargo, North Dakota) was planted to

U.M. Sainju et al. / Soil & Tillage Research 118 (2012) 123–131

a depth of 3.8 cm at 45 kg ha 1 and pea (cultivar Majoret, Macintosh Seed, Havre, Montana) at 101 kg ha 1 with a no-till drill equipped with double-shoot BartonTM (http://www.flexicoil.com/barton.asp) disk openers in April, 2006–2008. Pea seeds were inoculated with proper Rhizobium sp. At the same time, P fertilizer as triple superphosphate (45% P) at 29 kg P ha 1 and K fertilizer as muriate of potash (60% K) at 27 kg K ha 1 were banded to malt barley and pea. Nitrogen fertilizer as urea (45% N) was broadcast to malt barley a week after planting. No fertilizers were applied to fallow phase. Similarly, no irrigation was applied. In August, 2006– 2008, malt barley and pea grain yields were determined from a swath of 12.0 m  1.5 m using a combine harvester and biomass (stems + leaves) yields from an area of 1.0 m  1.0 m outside yield rows. After grain harvest, crop biomass residues were returned to the soil. In Nesson Valley, malt barley (cultivar Certified Tradition, Busch Agricultural Resources, Fargo, North Dakota) was planted to 3.8 cm depth at 90 kg ha 1 in the irrigated treatment and at 67 kg ha 1 in the non-irrigated treatment with a no-till drill in late April, 2005– 2008. Similarly, pea (cultivar Majorete, Macintosh Seed, Havre, Montana) was planted at 200 kg ha 1 in irrigated and nonirrigated treatments. In irrigated malt barley, half of the N fertilizer as urea (or 67 kg N ha 1) was banded at planting and the other half was broadcast at 4 wks after planting. In nonirrigated malt barley, all N fertilizer was banded at planting. Phosphorus fertilizer (as triple super phosphate at 25 kg P ha 1) and K fertilizer (as muriate of potash at 21 kg K ha 1) were banded to both malt barley and pea at planting. All banded fertilizers were applied to a depth of 5 cm, 2.5 cm away from the seed row. No N fertilizer was applied to pea. In the irrigated treatment, water was applied with a self-propelled electric linear move sprinkler from 10 to 25 mm per application for a total of 87 mm in 2005, from 10 to 34 mm per application for a total of 236 mm in 2006, from 13 to 31 mm for a total of 56 mm in 2007, and from 6 to 25 mm for a total of 47 mm in 2008. At the time of CO2 flux measurement in 2008, water was applied at 25 mm in late May, 6 mm in mid-June, and 16 mm in early July. Each treatment received water as needed based on soil water content and crop demand. In late July and early August, 2005–2008, malt barley and pea grain yields were determined from an area of 10.6 m  1.5 m using a combine harvester and biomass (leaves + stems) from two 0.5 m2 areas per plot outside yield rows. After grain harvest, crop biomass residues were returned to the soil. 2.3. Carbon dioxide flux measurements Immediately after planting, soil surface CO2 flux was measured weekly to biweekly in all treatments from May to October 2008 with closed portable and static chambers in Sidney and Nesson Valley until the ground froze. All measurements were taken between 9 A.M. and 12 A.M. of the day to reduce the variability in CO2 flux due to diurnal changes in temperature (Parkin and Kaspar, 2003). In the portable chamber method, CO2 flux was measured with an Environmental Gas Monitor chamber with an infrared CO2 analyzer attached to a data logger (model EGM-4, PP System, Haverhill, MA). The chamber was 15 cm tall, 10 cm in diameter, and had capacity to measure CO2 flux from 0 to 9999 mg CO2C m 2 h 1 (or 0–654 kg C ha 1 d 1). As described in the instruction manual, CO2 flux was measured by placing the chamber at the soil surface for 2 min in each plot until the flux was recorded in the data logger. Measurements were made from two places, in and between crop rows, 8 m apart, in each plot by placing the portable chamber near the static chamber throughout the study period. In the static chamber method, CO2 flux was measured with a chamber made from polyvinyl chloride pipe, 20 cm diameter by 15 cm tall (Hutchinson and Mosier, 1981; Liebig et al., 2010). The chamber

125

covered areas in and between crop rows and was installed to a depth of 7.5 cm in each plot. Two chambers per plot, 8 m apart, were deployed in each treatment. At the time of gas sampling, a cover, 20 cm diameter by 10 cm tall, was placed at the top of chamber and tightened to the chamber using soft tire attached to the cover (Liebig et al., 2010). The cover contained ports for gas sampling and ventilation. Gas samples were collected at 0, 20, and 40 min intervals by injecting a needle attached to a syringe (30 mL) in the port and transferring them to extainers (Hutchinson and Mosier, 1981; Liebig et al., 2010). As with the portable chamber, gas samples were collected from two chambers per plot. The CO2 concentration in the gas sample was measured with a gas chromatograph with a thermal conductivity detector (Varian Inc., Walnut Creek, CA) in the laboratory. The CO2 flux was calculated by using a linear or curvi-linear regression (Hutchinson and Mosier, 1981) as the difference in concentration gradient over time. The underestimation of CO2 flux due to suppression of the gas concentration gradient in the static chamber method was adjusted during calculation of the flux as shown by Venterea (2010). The bulk density of soil core (2.5 cm diameter), which is measured by dividing the weight of oven-dried soil at 110 8C by the volume of the core, and soil water content at the time of gas sampling were used for the correction of CO2 flux in the static chamber. Plants that grew above the height of the chamber were chopped off before collecting gas samples. At the time of CO2 measurement, soil temperature near the chamber was measured to a depth of 15 cm using a temperature probe. Similarly, gravimetric soil water content was measured near the chamber by collecting soil sample to a depth of 15 cm with a hand probe (2.5 cm diameter) every time CO2 flux was measured. As with CO2 flux, measurements were made from two places per plot near static chambers. The moist soil was oven-dried at 110 8C and water content was determined. Daily average air temperature and total precipitation during the study were collected from a meteorological station located within 1 km of each study site. 2.4. Data analysis Data for CO2 flux was analyzed by using the Analysis of Covariance and soil temperature and water content by the Analysis of Variance in the MIXED model of SAS (Littell et al., 1996). For eliminating the effect of water vapor on CO2 flux inside the chamber, soil water content was used as a covariable during the analysis of CO2 flux. For the data collected for loam soil in Sidney, cropping sequence was considered as the main plot, N fertilization as the split plot, CO2 measurement method as the split–split plot, and date of measurement as the repeated measure variable for analysis. For sandy loam soil in Nesson Valley, irrigation system was considered as the main plot, management practice as the split plot, CO2 measurement method as the split–split plot, and date of measurement as the repeated measure variable for analysis. For calculating CO2 flux and soil temperature and water content for a treatment, values for two chambers within a plot were averaged. Similarly, for a cropping sequence, values were averaged across phases. Means were separated by using the least square means test when treatments and interactions were significant (Littell et al., 1996). Regression analysis was performed to relate CO2 flux measured by the static and portable chamber methods. Statistical significance was evaluated at P  0.05, unless otherwise stated. 3. Results and discussion 3.1. Soil temperature and water content Soil temperature increased from May to July and then declined in the loamy site in Sidney and in the sandy loam site in Nesson

126

U.M. Sainju et al. / Soil & Tillage Research 118 (2012) 123–131

Fig. 1. Effect of management practice on soil temperature (A and B) and water content (C and D) from May to October 2008 in loamy and sandy loam sites. In the loamy site, CTB-F denotes conventional-tilled malt barley-fallow; NTB-P, no-tilled malt barley-pea; and NTCB, no-tilled continuous malt barley. In sandy loam site, CTBFN denotes conventional-tilled malt barley with 67–134 kg N ha 1; CTBON, conventional-tilled malt barley with 0 kg N ha 1; NTB-PN, no-tilled malt barley-pea with 67–134 kg N ha 1; NTBFN, no-tilled malt barley with 67–134 kg N ha 1; and NTBON, no-tilled malt barley with 0 kg N ha 1. LSD (0.05) is the least significant difference between treatments at P = 0.05. Double line arrows in (C) and (D) indicate dates of tillage, sowing (April–May), and harvest (August) and single line arrows show dates of irrigation.

Valley (Fig. 1A and B). While soil temperature did not vary among treatments in the loamy site, it was greater in CTBON and NTBON than in NTB-PN and NTBFN in May and July in the sandy loam site. Reduced shade intensity due to lower biomass yield in N unfertilized treatments increases soil temperature (Sainju et al., 2008, 2010). Soil water content varied among treatments and measurement dates, with greater levels following substantial precipitation events and/or irrigation in May, July, and October (Figs. 1C and D and 2A and B). The absence of crops during fallow probably increased water content in CTB-F compared with other treatments in July and September in the loamy site (Fig. 1C). In the sandy loam site, increased water conservation due to the presence

Fig. 2. Daily total precipitation from May to October 2008 in (A) Sidney, Montana (loam soil) and (B) Nesson Valley, North Dakota (sandy loam soil).

of surface residue may have increased water content in NTBFN compared with other treatments in August. 3.2. Soil surface carbon dioxide flux 3.2.1. Loamy site In the loamy site, the coefficient of variation in the measurement of CO2 flux within two chambers per plot in the static chamber method varied from 5 to 20%. Similarly, the coefficient of variation for CO2 flux within a treatment in a replication ranged from 8 to 102%. In the portable chamber method, the coefficient of variation for CO2 flux within two measurements per plot ranged from 8 to 25% and within a treatment in a replication ranged from 7 to 155%. It is not uncommon to observe spatial variability exceeding 100% in the measurement of greenhouse gas fluxes using the static chamber method (Parkin, 1985; Parkin et al., 1987). In the portable chamber method, the variability was even higher. The differences in variability between the portable and static chamber method were probably resulted from differences in the chamber size and placement of chamber at the soil surface. While there was difference in the internal volume of the chamber between the two methods (1178 cm3 in the portable vs. 4712 cm3 in the static chamber), the chamber was placed at the soil surface in the portable chamber method according to the instruction manual but inserted to a depth of 7.5 cm in the static chamber method. The variability in CO2 flux measurement results from variations in chamber size, chamber placement (in the crop row vs. between rows), soil properties (e.g. soil temperature, water content, organic matter, and texture), vegetation, and landscape position (Parkin, 1985; Parkin et al., 1987; Livingston and Hutchinson, 1995). Jensen et al. (1996) found that spatial variability in the CO2 flux measurement was large in both static and dynamic (portable) chamber methods, sometime exceeding 100%, and the variability was 1.5–2.0 times larger in the dynamic than in the static chamber

U.M. Sainju et al. / Soil & Tillage Research 118 (2012) 123–131

method. The variability can be reduced by taking as many measurements as possible by installing large number of chambers within a treatment and taking measurements under similar soil type, vegetation, and landscape units (Gilbert, 1987; Livingston and Hutchinson, 1995). Because of the small plot sizes (3.0–6.0 m wide  10.6–12.0 m long) at both study sites, it was decided to take only two measurements per plot, as deployment of too many chambers in small plots hindered field operations, such as tillage, fertilizer and herbicide applications, and harvest. Although variability was high, significant differences in CO2 flux among treatments and measurement dates still occurred (Table 1). The CO2 flux was significantly influenced by cropping sequence, measurement method, and date of measurement in the loamy site (Table 1). Interactions were significant for cropping sequence  measurement method, N fertilization  measurement method, cropping sequence  date of measurement, cropping sequence  N fertilization  date of measurement, cropping sequence  measuremeasurement method  date of measurement, and N fertilization  measurement method  date of measurement. Soil water content as a covariable did not influence CO2 flux. The CO2 flux increased with increased soil temperature and water content or precipitation in July and August and then declined in all treatments (Figs. 1–4). The CO2 flux was more variable with date of measurement in all treatments in the portable than in the static chamber. In the tillage and cropping sequence treatments, CO2 flux, averaged across N fertilization rates, was greater from 14% in NTB-P to 87% in NTCB in the portable than in the static chamber in late July and early August (Fig. 3). In contrast, from late August to late September, CO2 flux was lower from 15% in CTB-F to 61% in NTCB in the portable than in the static chamber. In the N fertilization treatments, CO2 flux, averaged across tillage and cropping sequences, was greater from 19% in 0 kg N ha 1 to 74% in 80 kg N ha 1 in the portable than in the static chamber in late July and early August (Fig. 4). From late August to late September, CO2 flux, however, was lower from 28% in 80 kg N ha 1 to 68% in 0 kg N ha 1 in the portable than in the static chamber. Averaged across N fertilization rates and measurement dates, CO2 flux was 14% lower in the portable than in the static chamber in NTCB (Table 2). Averaged across tillage, cropping sequences, and measurement dates, CO2 flux was 12% lower in the portable than in

127

the static chamber in 0 kg N ha 1. Averaged across treatments and measurement dates, CO2 flux was 9% lower in the portable than in the static chamber. The increased CO2 flux immediately following substantial precipitation (>15 mm) during increased soil temperature and water content in July and August in both static and portable chambers as a result of enhanced microbial activity is well known (Curtin et al., 2000; Sainju et al., 2008, 2010). During this period, CO2 flux was greater in the portable than in the static chamber and the result was more pronounced in NTCB than in CTB-F and NTB-P (Fig. 3). Similarly, the result was more pronounced in 80 than in 0 kg N ha 1 (Fig. 4). In contrast, CO2 flux was lower in the portable than in the static chamber from late August to late September. The limit at which the portable chamber produced greater or lower CO2 flux than the static chamber in all management practices was 13 kg C ha 1 d 1. Jensen et al. (1996) reported that the dynamic chamber produced as much as 700% greater CO2 flux than the static chamber above 6.5 kg C ha 1 d 1 (or 100 mg CO2-C m 2 h 1), but

Table 1 Analysis of variance for soil CO2 flux measurement in various management practices in loamy and sandy loam sites. Loam

Sandy loam

Source

Significance Source

Significance

Cropping sequence (C) N fertilization (N)

*

NS NS

CN CO2 method (M) CM NM CNM Date of measurement (D) CD ND CND MD CMD NMD CNMD Soil water (W)

NS

Irrigation (I) Management practice (P) IP CO2 method (M) IM PM IPM Date of measurement (D) ID PD IPD MD IMD PMD IPMD Soil water (W)

NS NS NS

a

NSa

** * *

NS ***

***

NS ***

NS * *

NS NS

Not significant. Significant at P  0.05. Significant at P  0.01. *** Significant at P  0.01. *

**

NS *** ** *

NS ***

*** ** *

NS NS

Fig. 3. Effect of measurement method on soil surface CO2 flux from July to October 2008 in various tillage and cropping sequences in the loamy site. CO2 measurement methods are portable, CO2 flux measured with a portable chamber using infrared analyzer; and static, CO2 flux measured with a static chamber using gas chromatograph. CTB-F denotes conventional-tilled malt barley-fallow; NTB-P, no-tilled malt barley-pea; and NTCB, no-tilled continuous malt barley. LSD (0.05) is the least significant difference between treatments at P = 0.05.

U.M. Sainju et al. / Soil & Tillage Research 118 (2012) 123–131

128

12% lower below this value. Similarly, Nay et al. (1994) in a laboratory experiment with controlled CO2 fluxes, observed that, while the dynamic chamber provided accurate values, the static chamber underestimated fluxes above 6.5 kg C ha 1 d 1 but overestimated below this value. The reasons for greater CO2 flux values in the portable than in the static chamber at high fluxes and lower values at low fluxes may be a result of differences in CO2 concentration gradient in the static chamber at various fluxes. Jensen et al. (1996) suggested that, at high fluxes, CO2 concentration gradient from the soil to the chamber atmosphere decreases due to increased chamber headspace CO2 concentration above the ambient level, thereby underestimating the flux by the static chamber. At low fluxes, the chamber headspace CO2 concentration decreases below the ambient, thereby increasing the gradient from soil to the chamber atmosphere and overestimating the flux. Since underestimated values of CO2 fluxes measured by the static chamber were corrected for the suppression of the gas due to concentration gradient as suggested by Venterea (2010), the average CO2 flux measured by the portable chamber was still 9% lower than that measured by the static chamber in the loamy soil under dryland cropping system. Probably differences in the chamber size and placement in the soil and analysis of CO2 by infrared vs. gas chromatograph influenced CO2 flux measurements by the portable and static chamber methods.

Fig. 4. Effect of measurement method on soil surface CO2 flux from July to October 2008 in various N fertilization rates in the loamy site. CO2 measurement methods are portable, CO2 flux measured with a portable chamber using infrared analyzer; and static, CO2 flux measured with a static chamber using gas chromatograph. LSD (0.05) is the least significant difference between treatments at P = 0.05.

Table 2 Interaction of measurement method with tillage, cropping sequence, N fertilization on average soil CO2 flux across measurement dates in the loamy site. Tillage and cropping sequencea

N fertilization rate (kg N ha 1)

CO2 measurement methodb

CO2 flux 1

(kg C ha CTB-F

Portable Static

8.18bc 8.12bc

NTB-P

Portable Static

6.92c 7.90bc

NTCB

Portable Static

8.72b 10.11a

0

Portable Static

7.64b 8.69a

80

Portable Static

8.24ab 8.73a

Means CTB-F NTB-P NTCB

d

1

)

c

8.15ab 7.41b 9.42a Portable Static

7.94b 8.71a

a Tillage and cropping sequences are CTB-F, conventional-tilled malt barleyfallow; NTB-P, no-tilled malt barley-pea; and NTCB, no-tilled continuous malt barley. b CO2 measurement methods are portable, CO2 flux measured with a portable chamber using infrared analyzer; and static, CO2 flux measured with a static chamber using gas chromatograph. c Numbers followed by different letters within a column in a set are significantly different at P = 0.05 by the least square means test.

3.2.2. Sandy loam site The coefficient of variation in CO2 flux between two chambers per plot varied from 3 to 22% and within a treatment in a replication varied from 5 to 129% in the static chamber method in the sandy loam site in Nesson Valley. In the portable chamber method, the coefficient of variation between two measurements within a plot ranged from 6 to 30% and within a treatment in a replication ranged from 2 to 93%. As in the loamy site in Sidney, although the variability within a plot was lower, high spatial variability in CO2 flux sometime exceeding 100% occurred within a treatment. The CO2 flux was significantly influenced by measurement method and date of measurement (Table 1). Interactions were significant for irrigation  measurement method, management practice  measurement method, measurement method  date of measurement, irrigation  measurement method  date of measurement, and management practice  measurement method  date of measurement. As in the loamy site, soil water as a covariable did not influence CO2 flux. Similar to the loamy site, CO2 flux increased from May to July during increased soil temperature and water content due to substantial precipitation and/or irrigation (>15 mm) and then declined in all treatments (Figs. 1, 2 and 5–7). The CO2 flux in the portable chamber was either greater than or equal to that in the static chamber at most measurement dates from May to October (Figs. 5–7). In the irrigated and non-irrigated treatments, CO2 flux, averaged across management practices, was 20% lower to 229% greater in the portable than in the static chamber (Fig. 5). In the tilled cropping treatment, CO2 flux, averaged across irrigation systems, was 9–204% greater in the portable than in the static chamber (Fig. 6). In the no-tilled cropping treatment, CO2 flux was 10% lower to 152% greater in the portable than in the static chamber (Fig. 7). Averaged across management practices and measurement dates, CO2 flux was greater from 60% in the non-irrigated to 110% in the irrigated treatment in the portable than in the static chamber (Table 3). Averaged across irrigation systems and measurement dates, CO2 flux was greater from 59% in NTBON to 96% in NTBFN in the portable than in the static chamber. Averaged across treatments and measurement dates, CO2 flux was 84% greater in the portable than in the static chamber.

U.M. Sainju et al. / Soil & Tillage Research 118 (2012) 123–131

129

Fig. 5. Effect of measurement method on soil surface CO2 flux from May to October 2008 in various irrigation systems in the sandy loam site. CO2 measurement methods are portable, CO2 flux measured with a portable chamber using infrared analyzer; and static, CO2 flux measured with a static chamber using gas chromatograph. LSD (0.05) is the least significant difference between treatments at P = 0.05.

Fig. 7. Effect of measurement method on soil surface CO2 flux from May to October 2008 in various no-tilled cropping practices in the sandy loam site. CO2 measurement methods are portable, CO2 flux measured with a portable chamber using infrared analyzer; and static, CO2 flux measured with a static chamber using gas chromatograph. NTB-PN denotes no-tilled malt barley-pea with 67–134 kg N ha 1; NTBFN, no-tilled malt barley with 67–134 kg N ha 1; and NTBON, no-tilled malt barley with 0 kg N ha 1. LSD (0.05) is the least significant difference between treatments at P = 0.05.

Fig. 6. Effect of measurement method on soil surface CO2 flux from May to October 2008 in various tilled cropping practices in the sandy loam site. CO2 measurement methods are portable, CO2 flux measured with a portable chamber using infrared analyzer; and static, CO2 flux measured with a static chamber using gas chromatograph. CTBFN denotes conventional-tilled malt barley with 67– 134 kg N ha 1; and CTBON, conventional-tilled malt barley with 0 kg N ha 1. LSD (0.05) is the least significant difference between treatments at P = 0.05.

As in the loamy site, greater positive differences in CO2 flux between the portable and static chamber in all treatments were noted during increased soil temperature and water content from May to August (Figs. 1 and 5–7) when soil microbial activity was higher. The differences, however, narrowed thereafter as soil temperature and water content declined, although the flux in the portable chamber was equal to or greater than that in the static chamber at most measurement dates. Greater CO2 flux in the portable than in the static chamber, regardless of treatments (Table 3), suggests that the portable chamber method provides greater CO2 flux values than the static chamber method in the sandy loam soil under irrigated and non-irrigated cropping systems. The results were similar to that reported by several researchers (Ewel et al., 1987; Rochette et al., 1992), who reported that the dynamic chamber consistently produced greater CO2 fluxes than the static chamber, with larger differences at higher fluxes in sandy loam soil. Since the chamber was placed at the soil surface for 2 min at the time of CO2 flux measurement in the

U.M. Sainju et al. / Soil & Tillage Research 118 (2012) 123–131

130

Table 3 Interaction of measurement method with irrigation and management practice on average soil CO2 flux across measurement dates in the sandy loam site. CO2 measurement methodb

CO2 flux (kg C ha

Irrigated

Portable Static

37.4ac 17.8c

Non-irrigated

Portable Static

32.9b 20.6c

CTBFN

Portable Static

36.8a 19.9bc

CTBON

Portable Static

33.2a 17.1c

NTB-PN

Portable Static

35.8a 19.3bc

NTBFN

Portable Static

36.6a 18.7bc

NTBON

Portable Static

33.4a 21.0b

Portable Static

35.3a 19.2b

Irrigation

Management practicea

1

d

1

)

Means

a Management practices are CTBFN, conventional-tilled malt barley with 67– 134 kg N ha 1; CTBON, conventional-tilled malt barley with 0 kg N ha 1; NTB-PN, no-tilled malt barley-pea with 67–134 kg N ha 1; NTBFN, no-tilled malt barley with 67–134 kg N ha 1; and NTBON, no-tilled malt barley with 0 kg N ha 1. b CO2 measurement methods are portable, CO2 flux measured with a portable chamber using infrared analyzer; and static, CO2 flux measured with a static chamber using gas chromatograph. c Numbers followed by different letters within a column in a set are significantly different at P = 0.05 by the least square means test.

portable chamber without the presence of collar according to the instruction manual, it could be possible that higher CO2 flux in the portable than in the static chamber result from increased soil disturbance, air leak from outside to inside the chamber, and differences in chamber size and chamber placement in the soil between the two methods. Although the underestimated values of CO2 fluxes due to suppression of gas from concentration gradient were adjusted in the static chamber as suggested by Venterea (2010) as in the loamy site, the portable chamber still showed higher CO2 fluxes than the static chamber. A comparison of averaged CO2 fluxes across treatments and measurement dates under dryland cropping system in the loamy site and under non-irrigated cropping system in the sandy loam site (Tables 2 and 3) indicated that CO2 flux was 4.1 times greater in the portable chamber method and 2.4 times greater in the static chamber method in the sandy loam than in loam. This shows that coarse-textured soil emitted greater CO2 fluxes than finetextured soil, with greater values in the portable than in the static chamber method, since average monthly air temperature and total precipitation were similar in the loamy (13.5 8C and 350 mm) and the sandy loam (12.5 8C and 373 mm) sites. Probably soil texture influenced root and microbial respiration and organic matter mineralization, thereby affecting CO2 fluxes between the sites. 3.3. Relationship between static and portable chamber methods The CO2 flux measured by the static and the portable chamber methods in various management practices was related linearly in the sandy loam site to exponentially in the loamy site (R2 = 0.68– 0.70, P  0.01, n = 40–56) (Fig. 8). An increase in CO2 flux by 1 kg C ha 1 d 1 measured by the static chamber method increased the flux by 1.7 kg C ha 1 d 1 measured by the portable chamber method in the sandy loam site. In the loamy site, an increase in CO2 flux measured by the static chamber method slightly increased the

Fig. 8. Relationship between the static and portable chamber methods in soil surface CO2 flux measurements in (A) sandy loam and (B) loamy sites. CO2 measurement methods are portable, CO2 flux measured with a portable chamber using infrared analyzer; and static, CO2 flux measured with a static chamber using gas chromatograph.

flux measured by the portable chamber method within the range of 0–10 kg C ha 1 d 1, after which the flux measured by the portable chamber increased rapidly compared with that measured by the static chamber. The static chamber method explained 68–70% of variability in CO2 flux measured by the portable chamber method. Contrary to our hypothesis, the results indicate that the portable chamber method measured greater CO2 flux than the static chamber method in sandy loam site under irrigated and nonirrigated cropping systems but only slightly greater among treatments in the range of 0–10 kg C ha 1 d 1 in the loamy site under dryland cropping systems, regardless of management practices. Probably variations in soil and environmental conditions between locations resulted in different mineralization rates of soil and crop residue C and root and microbial respiration, which were measured differently by the static and portable chamber methods as various CO2 fluxes. It has been known that perturbations caused by microclimate (e.g. soil temperature, water content, humidity, etc.) and gas pressure inside the static chamber can influence CO2 flux measurements (Parkin and Venterea, 2010). For example, chambers can be flooded during high precipitation and/or irrigation events, during which either excess water should be pumped out from the chambers or chambers need to taken out and reinstalled (Parkin and Venterea, 2010). Similarly, high humidity inside the chamber may facilitate algal growth at the soil surface. Such conditions were more likely to occur in the irrigated than in the non-irrigated treatments in the sandy loam site where irrigation was applied. We, however, did not observe any such incidents during CO2 flux measurements at both sites. Although CO2 flux measured by the portable and the static chamber was linear in the sandy loam site, an exponential relationship occurred at the loamy site, similar to that reported by several researchers (Ewel et al., 1987; Rochette et al., 1992; Jensen et al., 1996).

U.M. Sainju et al. / Soil & Tillage Research 118 (2012) 123–131

Our CO2 fluxes of 2–15 kg C ha 1 d 1 measured by the static chamber in the loamy site were within the range of 2– 20 kg C ha 1 d 1 as reported by various researchers (Beyer, 1991; Norman et al., 1992; Jensen et al., 1996). Similarly, our fluxes of 2– 25 kg C ha 1 d 1 measured by the portable chamber were within the range of 2–165 kg C ha 1 d 1 measured by the dynamic chamber (Jensen et al., 1996). In the sandy loam site, our fluxes of 2– 40 kg C ha 1 d 1 measured by the static chamber exceeded the reported values of 2–20 kg C ha 1 d 1 (Beyer, 1991; Norman et al., 1992; Jensen et al., 1996) but the values of 10–80 kg C ha 1 d 1 measured by the dynamic chamber were within the range of 2– 165 kg C ha 1 d 1 (Jensen et al., 1996). Variations in soil and climatic conditions and management practices among locations probably influenced CO2 fluxes measured by the static and dynamic chambers. 4. Conclusions The CO2 flux measured by the static and portable chamber methods showed variable results in various soil types and management practices, in contrast to our hypothesis. In the loamy site under dryland cropping systems, the portable chamber measured CO2 flux greater than that measured by the static chamber above 13.0 kg C ha 1 d 1 but lower flux below this value in management practices that contained tillage, cropping sequence, and N fertilization. In the sandy loam site under irrigated and non-irrigated cropping systems, the portable chamber measured mostly greater CO2 flux than that measured by the static chamber throughout the growing season in management practices that contained irrigation, tillage, crop rotation, and N fertilization. Under similar management practices, sandy loam produced 2.4–4.1 times greater CO2 flux as measured by the portable and static chamber than loam. Spatial and temporal variability in CO2 flux were slightly greater with the portable than with the static chamber, regardless of soil types and management practices, although both methods had higher variability. Because of the close values of the average CO2 flux measured by the portable and static chamber across treatments and measurement dates, portable chamber method may measure similar fluxes compared with the static chamber method within the range of 0– 10 kg C ha 1 d 1 in the loam soil under dryland cropping systems than in the sandy loam soil under irrigated and non-irrigated cropping systems. Acknowledgements We acknowledge the help provided by Christopher Russell and Joy Barsotti for installing chambers and collecting gas samples and data in the field and analyzing them in the laboratory. We also acknowledge USDA-ARS-GRACEnet project for providing part of the funding for this project. References Al-Kaisi, M.M., Kruse, M.L., Sawyer, J.E., 2008. Effect of nitrogen fertilizer application on growing season carbon dioxide emission in a corn–soybean rotation. J. Environ. Qual. 37, 325–332. Al-Kaisi, M.M., Yin, X., 2005. Tillage and crop residue effects on soil carbon and carbon dioxide emission in corn–soybean rotation. J. Environ. Qual. 34, 437–445. Amos, B., Arkebauer, T.J., Doran, J.W., 2005. Soil surface fluxes of greenhouse gases in an irrigated maize-based agroecosystem. Soil Sci. Soc. Am. J. 69, 387–395. Beyer, L., 1991. Intersite characterization and variability of soil respiration in different arable and forest soils. Biol. Fertil. Soils 12, 122–126. Calderon, F.J., Jackson, L., 2002. Rototillage, disking, and subsequent irrigation: effects on soil nitrogen dynamics, microbial biomass, and carbon dioxide efflux. J. Environ. Qual. 31, 752–758. Cole, V.C., Cerri, T., Minami, K., Mosier, A., Rosenberg, N., Sauerbeck, D., Dumanski, D.J., Duxbury, J., Freney, J., Gupta, R., Heinemeyer, O., Kolchugina, T., Lee, J., Paustian, K., Powlson, D., Sampson, N., Tiessen, H., Van Noordwijk, M., Zhao, Q., 1996. Agricultural options for mitigation of greenhouse gases. In: Watson, R.T. (Ed.), Climate Change 1995. Impacts, Adaptations, and Mitigation of Climate

131

Change: Scientific-Technical Analysis. IPCC Working Group II. Cambridge Univ. Press, Cambridge, MA, pp. 745–771. Curtin, D., Wang, H., Selles, F., McConkey, B.G., Campbell, C.A., 2000. Tillage effects on carbon dioxide fluxes in continuous wheat and fallow-wheat rotations. Soil Sci. Soc. Am. J. 64, 2080–2086. Duxbury, J.M., 1994. The significance of agricultural sources of greenhouse gases. Fertil. Res. 38, 151–163. Ewel, K.C., Cropper Jr., W.P., Gholz, H.L., 1987. Soil CO2 evolution in Florida slash pine plantations. I. Changes through time. Can. J. For. Res. 17, 325–329. Gilbert, R.O., 1987. Statistical Methods for Environmental Pollution Monitoring. Van Nostrand Reinhold Inc., New York, NY. Healy, R.W., Striegl, R.G., Russell, T.F., Hutchinson, G.L., Livingston, G.P., 1996. Numerical evaluation of static chamber measurements of soil–atmosphere gas exchange. Identification of physical processes. Soil Sci. Soc. Am. J. 60, 740–747. Hutchinson, G.L., Mosier, A.R., 1981. Improved soil cover method for field measurement of nitrous oxide fluxes. Soil Sci. Soc. Am. J. 45, 311–316. Jackson, L.E., Calderon, F.J., Steenwerth, K.F., Scow, K.M., Rolston, D.E., 2003. Responses of soil microbial processes and community structure to tillage events and implications for soil quality. Geoderma 114, 305–317. Jensen, L.S., Mueller, T., Tate, K.R., Ross, D.J., Magid, J., Nielsen, N.E., 1996. Soil surface CO2 flux as an index of soil respiration in situ: a comparison of two chamber methods. Soil Biol. Biochem. 28, 1297–1306. Kuo, S., Sainju, U.M., Jellum, E.J., 1997. Winter cover crop effects on soil organic carbon and carbohydrate. Soil Sci. Soc. Am. J. 61, 145–152. Lal, R., Kimble, J.M., Stewart, B.A., 1995. World soils as a source or sink for radiatively-active gases. In: Lal, R. (Ed.), Soil Management and Greenhouse Effect. Advances in Soil Science. CRC Press, Boca Raton, FL, pp. 1–8. Liebig, M.A., Tanaka, D.L., Gross, J.R., 2010. Fallow effects on soil carbon and greenhouse gas flux in central North Dakota. Soil Sci. Soc. Am. J. 74, 358–365. Littell, R.C., Milliken, G.A., Stroup, W.W., Wolfinger, R.D., 1996. SAS System for Mixed Models. SAS Inst. Inc., Cary, NC. Livingston, G.P., Hutchinson, G.L., 1995. Enclosure-based measurement of trace gas exchange: applications and sources of error. In: Matson, P.A., Harriss, R.C. (Eds.), Biogenic Trace Gases: Measuring Emissions from Soil and Water. Methods in Ecology. Blackwell Scientific Cambridge University Press, London, pp. 14–51. Mosier, A.R., Halvorson, A.D., Reule, C.A., Liu, X.J., 2006. Net global warming potential and greenhouse gas intensity in irrigated cropping systems in northeastern Colorado. J. Environ. Qual. 35, 1584–1598. Nay, S.M., Mattson, K.G., Bormann, B.T., 1994. Biases of chamber methods for measuring soil CO2 efflux demonstrated with a laboratory apparatus. Ecology 75, 2460–2463. Norman, J.M., Gracia, R., Verma, S.B., 1992. Soil surface CO2 fluxes and the carbon budget of a grassland. J. Geophys. Res. 97, 18845–18853. Parkin, T.B., 1985. Automated analysis of nitrous oxide. Soil Sci. Soc. Am. J. 49, 273–276. Parkin, T.B., Kaspar, T.C., 2003. Temperature controls on diurnal carbon dioxide flux: implications for estimating soil carbon loss. Soil Sci. Soc. Am. J. 67, 1763–1772. Parkin, T.B., Starr, J.L., Meisinger, J.J., 1987. Influence of sample size on measurement of soil denitrification. Soil Sci. Soc. Am. J. 51, 1492–1501. Parkin, T.B., Venterea, R.T., 2010. USDA-ARS GRACEnet project protocols, chapter 3, chamber-based trace gas flux measurements. In: Follett, R.F. (Ed.), Sampling Protocols. USDA-ARS, Fort Collins, CO, pp. 3-1–3-39. Paustian, K., Robertson, G.P., Elliott, E.T., 1995. Management impacts on carbon storage and gas fluxes in mid-latitudes cropland. In: Lal, R. (Ed.), Soils and Global Climate Change. Advances in Soil Science. CRC Press, Boca Raton, FL, pp. 69–83. Peterson, G.A., Halvorson, A.D., Havlin, J.L., Jones, O.R., Lyon, D.G., Tanaka, D.L., 1998. Reduced tillage and increasing cropping intensity in the Great Plains conserve soil carbon. Soil Till. Res. 47, 207–218. Post, W.M., Peng, T.H., Emanuel, W.R., King, A.W., Dale, V.H., DeAngelis, D.L., 1990. The global carbon cycle. Am. Sci. 78, 310–326. Rasmussen, P.E., Allmaras, A.R., Rhoade, C.R., Roager Jr., N.C., 1980. Crop residue influences on soil carbon and nitrogen in a wheat-fallow system. Soil Sci. Soc. Am. J. 44, 596–600. Roberts, W.P., Chan, K.Y., 1990. Tillage-induced increases in carbon dioxide loss from soil. Soil Tillage Res. 17, 143–151. Rochette, P., Bertrand, N., 2007. Soil surface gas emissions. In: Carter, M., Gregorich, E.G. (Eds.), Soil Sampling and Methods of Analysis. CRC Press, Boca Raton, FL, pp. 851–861. Rochette, P., Gregorich, E.G., Desjardins, R.L., 1992. Comparison of static and dynamic closed chambers for measurement of soil respiration under field conditions. Can. J. Soil Sci. 72, 605–609. Sainju, U.M., Caesar-TonThat, T., Lenssen, A.W., Evans, R.G., Kolberg, R., 2007. Longterm tillage and cropping sequence effects on dryland residue and soil carbon fractions. Soil Sci. Soc. Am. J. 71, 1730–1739. Sainju, U.M., Jabro, J.D., Caesar-TonThat, T., 2010. Tillage, cropping sequence, and nitrogen fertilization effects on dryland soil carbon dioxide emission and carbon content. J. Environ. Qual. 37, 98–106. Sainju, U.M., Jabro, J.D., Stevens, W.B., 2008. Soil carbon dioxide emission and carbon sequestration as influenced by irrigation, tillage, cropping system, and nitrogen fertilization. J. Environ. Qual. 39, 935–945. Schlesinger, W.H., 1997. Biogeochemistry. An Analysis of Global Change. Academic Press, San Diego, CA. Venterea, R.T., 2010. Simplified method for quantifying theoretical underestimation of chamber-based trace gas fluxes. J. Environ. Qual. 39, 126–135.