SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 25 January 2008
Published in Soil Sci Soc Am J 72:393-401 (2008)
DOI: 10.2136/sssaj2007.0033
© 2008 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Welzmiller, J. T.
Right arrow Articles by Thompson, T. L.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Welzmiller, J. T.
Right arrow Articles by Thompson, T. L.
Agricola
Right arrow Articles by Welzmiller, J. T.
Right arrow Articles by Thompson, T. L.
Related Collections
Right arrow Soil Kinetics
Right arrow Redox Processes

SOIL FERTILITY & PLANT NUTRITION

Elevated Carbon Dioxide and Irrigation Effects on Soil Nitrogen Gas Exchange in Irrigated Sorghum

Jaydene T. Welzmillera, Allan D. Matthiasa, Scott Whitea and Thomas L. Thompsonb,*

a Dep. of Soil, Water and Environmental Science, Univ. of Arizona, 1177E. 4th St., 429 Shantz Building, Tucson, AZ 85721
b Dep. of Plant and Soil Science, Texas Tech Univ., 15th and Detroit, Lubbock, TX 79409-3121

* Corresponding author (thomas.thompson{at}ttu.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The impacts of increasing atmospheric CO2, an important greenhouse gas, on soil microbial production and consumption of other greenhouse gases such as N2O are uncertain. This study was conducted during the 1998 and 1999 summer growing seasons at the Free-Air CO2 Enrichment (FACE) site in Maricopa, AZ. The objective was to measure N2O and denitrification emission rates in a C4 sorghum [Sorghum bicolor (L.) Moench] production system with ample and limited flood irrigation rates under FACE (seasonal mean = 579 µmol mol–1) and control (seasonal mean = 396 µmol mol–1) CO2. Plots were sampled for N2O flux using both chamber and intact incubated soil core techniques. Nitrogen gas (N2O plus N2) emissions were measured using intact incubated soil cores with C2H2 inhibition. Nitrous oxide emissions measured with chambers increased markedly after irrigation and fertilization following prolonged periods without water under both elevated and control CO2 conditions. Within 5 d of fertilization and irrigation, N2O emissions measured with chambers were <250 g N2O-N ha–1 d–1 until subsequent irrigations. Emissions measured from cores ranged from –0.11 to >250 g N2O-N ha–1 d–1. Seasonal cumulative N2O-N emissions measured using chambers were <1.5 kg N ha–1. Seasonal N-gas losses measured during 1999 were as high as 3.7 kg N ha–1, and were highest with elevated CO2 and the high irrigation treatment. During periods when significant emissions were recorded, the primary end product of denitrification was N2 rather than N2O. Water-filled pore space (WFPS) was the most important single factor controlling N-gas emissions, with the largest emissions (>500 g N2O-N ha–1 d–1) coming with >55% WFPS. Neither soil NO3– nor soil organic C alone limited N gas emissions. Elevated CO2 did not result in increased N2O or N-gas emissions with either ample or limited irrigation.

Abbreviations: DEA, denitrifying enzyme activity • FACE, Free-Air Carbon Dioxide Enrichment • GC, gas chromatography • SOC, soluble organic carbon • WFPS, water-filled pore space


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The atmospheric concentration of CO2 in 2005 was about 379 µmol mol–1, an increase of ~31% since the beginning of the Industrial Revolution (Intergovernmental Panel on Climate Change, 2007). Increased atmospheric CO2 may change N cycling in soils, including N-gas efflux. Nitrous oxide is another important radiatively active greenhouse gas whose concentration in the atmosphere is predicted to increase from the current ~320 nmol mol–1 to 350 to 460 nmol mol–1 by 2100 (Intergovernmental Panel on Climate Change, 2007). Nitrous oxide is of growing concern because of its relatively long atmospheric residence time (114–120 yr) and increasing radiative forcing due to increasing concentration (radiative forcing is currently ~0.15 W m–2 compared with 1.51 W m–2 for CO2) (Intergovernmental Panel on Climate Change, 2007). Nitrous oxide is ~300 times more radiatively active than CO2 (Rodhe, 1990). Furthermore, N2O catalyzes stratospheric O3 loss (Crutzen, 1971).

Nitrous oxide originates mainly within terrestrial ecosystems via soil microbial denitrification and nitrification (Frenchel et al., 1998; Smith et al., 2003). Denitrifying microbes are greatly influenced by the availability of soil C as an energy source, which in turn may be directly or indirectly affected by the concentration of atmospheric CO2. Carbon is especially a limiting factor for denitrification in many systems (Bremner and Shaw, 1958; McCarty and Bremner, 1992). In addition to the availability of C sources, the denitrification rate in soils is dependent on several other factors including temperature, pH, redox status, and the concentration of NO3– (Westerman and Tucker, 1978; Velthof and Oenema, 1995).

There is a need for improved understanding of the relationships between increasing atmospheric CO2 concentrations, soil C, and N2O emissions from soils. Elevated atmospheric CO2 has increased aboveground sorghum biomass (Ottman et al., 2001) and the above- and belowground biomass of other C4 plants (Maroco et al., 1999). Organic C concentrations in soil have increased under elevated atmospheric CO2 (Ghannoum and Conroy, 1998). Zak et al. (1993) found that microbial N transformations that depend on C supply, such as mineralization, immobilization, N2 fixation, and denitrification, could be accelerated under elevated atmospheric CO2. The acceleration is believed to be due to an increase in C supply through root exudation. De Luis et al. (1999) found that both water stress and elevated atmospheric CO2 concentrations resulted in increased belowground biomass. Smart et al. (1997) found that when wheat (Triticum aestivum L.) was grown hydroponically with elevated atmospheric CO2, the denitrification potential was much higher than the denitrification potential under ambient CO2. They concluded that the higher potential resulted from increased root nonstructural carbohydrate accumulation and consequent exudation under elevated CO2. In the Swiss FACE experiment, after 2 yr of elevated atmospheric CO2, Marilley et al. (1999) found that the total number of bacteria in the bulk soil remained unchanged but that bacteria in the rhizosphere of ryegrass (Lolium perenne L.) increased, presumably from increased C through root exudation. In the same FACE experiment, Ineson et al. (1998) found a 27% increase in denitrification rates under elevated atmospheric CO2. They hypothesized that the increase in denitrification emissions was due to increased C input into the soil, although no tests were performed to confirm their hypothesis.

In further research at the Swiss FACE experiment, Baggs et al. (2003a) concluded that the effects of increases in CO2 may depend on plant composition and fertilizer management. They found that N2O emissions were significantly increased under elevated CO2 and within highly N-fertilized Lolium perenne monoculture and mixed L. perenne/Trifolium repens L. swards. They also found that elevated CO2 had no effect on emissions from T. repens monoculture. They attributed the increased N2O emissions from L. perenne monoculture and mixed swards to increased belowground C allocation, which provided energy for denitrification. Baggs et al. (2003b) used 15N applications on plots of L. perenne swards at the Swiss FACE experiment to determine that total denitrification (N2O + N2) was increased under elevated CO2. This determination supported their hypothesis that increased belowground C allocation under elevated CO2 provided the energy for denitrification. They found nitrification and denitrification to be the main N2O production processes under ambient and elevated CO2, respectively. They also reported that the ratio of N2 to N2O was often higher under elevated CO2.

In contrast to the findings at the Swiss FACE experiment, Billings et al. (2002) found that there was no significant effect on N2O emissions or N mineralization rates on Mojave Desert soils exposed to elevated CO2 levels. Billings et al. (2002) also found that the potential for high denitrifying enzyme activity (DEA) did not necessarily cause high N2O fluxes from denitrification—especially under moist, cool winter conditions. Similarly, Mosier et al. (2002) found no statistically significant effect of elevated CO2 on N2O or other trace gas exchanges in Colorado shortgrass steppe during 43 mo of observation.

Root exudation increases under water-stressed conditions (Barber and Gunn, 1974; Smucker, 1982; Haller and Stolp, 1985). With increases in root exudation from water stress and elevated atmospheric CO2, it is possible that microbial communities will undergo more intense periods of denitrification and produce more N2O under water-stressed and elevated CO2 conditions. The objective of this study was to determine the effects of atmospheric CO2 concentrations and irrigation management on N-gas emissions in an irrigated sorghum production system in the desert southwestern United States.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Setup of FACE Study
The experiment was conducted at the University of Arizona Maricopa Agricultural Center Free-Air CO2 Enrichment (FACE) site during the 1998 and 1999 summer growing seasons. The soil at the site is a Trix clay loam (fine-loamy, mixed, calcareous, hyperthermic Typic Torrifluvent). DeKalb DK 54 sorghum was planted in rows spaced 76 cm apart and seeds were placed 4 cm deep. Planting occurred on 15 July 1998 and 14 June 1999. The experiment was a split-plot design with factorial combinations of CO2 (ambient or elevated) and irrigation treatment (Dry or Wet). The CO2 treatment was the main-plot factor and irrigation was the subplot factor. Each treatment was replicated four times (see Fig. 1 ).


Figure 1
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 1. Plot diagram (not drawn to scale) of field experiment showing the locations of the four elevated CO2 rings (dark circles) and the four ambient CO2 rings (light circles). Also shown are the dry and wet irrigation zones within each ring.

 
Carbon Dioxide Enrichment
The FACE technology was used to achieve elevated atmospheric CO2 concentrations inside plot areas in a manner similar to previous Maricopa FACE experiments (Hendrey, 1993; Kimball et al., 1995, 1999; Pinter et al., 1996). Eight rings, 25 m in diameter, were constructed in a 12-ha field as shown in Fig. 1. The centers of the rings were separated by 98 m within a replicate. Around each ring were 32, 2.5-m-high, vertical stand pipes constructed of polyvinyl chloride pipe placed approximately every 2.4 m around the circumference of the rings. Each stand pipe was equipped with a solenoid valve that allowed CO2 injection. Air enriched with CO2 was blown into the rings near the top of the crop canopy through tridirectional jets in the vertical stand pipes. Concentrations of CO2 were measured within the center of each ring using an infrared gas analyzer approximately 10 cm above the crop canopy. Wind speed and direction were measured north of each FACE ring. These values were computer controlled to maintain appropriate CO2 concentrations within the centers of the rings. Carbon dioxide flow rates were updated every second, while the choice of which vertical pipes released CO2 was updated every 4 s. In plots with ambient atmospheric CO2 concentrations, air blowers similar to those for the FACE rings provided similar air movement. In 1998, the FACE treatment began on 31 July when slightly <50% plant emergence had occurred. In 1999 it began on 2 July at about 50% plant emergence.

Carbon dioxide concentrations in the FACE plots in 1998 were 556 µmol mol–1 during the day and 603 µmol mol–1 at night. During 1999, CO2 concentrations in the FACE plots were 566 and 607 µmol mol–1 during day and night, respectively. Previous FACE experiments with wheat indicated that FACE CO2 concentrations were within 10% of the average 87% of the time (Hendrey, 1993; Nagy et al., 1994). Control-plot CO2 concentrations were 363 µmol mol–1 during the day and 428 µmol mol–1 at night in 1998 and 373 µmol mol–1 during the day and 433 µmol mol–1 at night in 1999.

Irrigation and Fertilization
Half of the area within each ring received ample water (Wet), while the other half received a limited amount of water (Dry). Water was applied using flood irrigation. Wet plots were irrigated when 30% of plant-available water was depleted. They were irrigated with an amount calculated to replace 100% of potential evapotranspiration since the previous irrigation, minus rainfall (Fox et al., 1992). To ensure that irrigation water distribution was acceptably uniform, a minimum of 100 mm of water was added at each irrigation. Dry plots received approximately one-third the amount of water as Wet plots to severely stress plants. Thus, the Dry treatment plots received only two irrigation events each season (on 27 July and 11 Sept. 1998, 28 June and 6 Aug. 1999), compared with seven irrigations in 1998 and six irrigations in 1999 applied to the Wet treatments. In 1998, 1218 mm of irrigation plus rainfall (1198 + 20 mm) was applied to the Wet treatment and 474 mm (454 + 20 mm) to the Dry treatment. In 1999, 1047 mm (894 + 153 mm) of irrigation and rainfall was applied to the Wet treatment, and 491 mm (338 + 153 mm) to the Dry treatment. Seasonal rainfall totals were measured at an automated weather station 1 km from the study site. In both 1998 and 1999, preplant fertilizer (93 kg N ha–1 and 41 kg P ha–1) was applied and incorporated by harrow in the form of urea (46–0–0) and monoammonium phosphate (11–52–0). Nitrogen was applied in the irrigation water as urea–NH4NO3 (32–0–0) on 11 Sept. 1998 (57 d after planting) and 6 Aug. 1999 (54 d after planting). In 1998, the Wet treatment received an additional N application of 62 kg N ha–1 on 26 September (72 d after planting). This additional application was necessary because the Dry treatment received a large amount of irrigation water (and N) on 11 September to uniformly irrigate the dry, cracked soil. Total N applications (preplant + season) to both Dry and Wet treatments were 279 kg N ha–1 in 1998 and 265 kg N ha–1 in 1999.

Nitrous Oxide and Nitrogen Gas Measurement Methods
Two methods, closed chamber and intact core, were used for measuring N2O emissions within a 4-m2 area of each half ring. The use of closed chambers allowed measurements under waterlogged soil conditions in which soil cores were difficult to collect. Chamber measurements also allowed integration of gas fluxes over a larger surface area than permitted by core sampling. In addition, the use of chambers allowed collection of N2O that may have been dissolved in soil water and transported through the sorghum plants via transpiration to the atmosphere (e.g., Muller, 2003). Soil cores were used to determine rate-limiting factors for N2O and N-gas (N2O plus N2) emissions. Soil cores were used exclusively when the sorghum became too tall for the chambers (by about midseason). Nitrogen-gas measurements were performed exclusively on soil cores.

Closed Chamber Measurements
A closed (non-flow-through, non-steady-state) chamber method, as described by Matthias et al. (1980), was used to collect N2O emitted from the soil surface. Metal chamber bases (50 by 50 cm) were permanently placed 10 cm into the soil before plant emergence in 1998 and slightly after emergence in 1999. One base was installed within the designated 4-m2 study area in each FACE ring–irrigation treatment combination. The two bases per ring were separated by about 8 m. Corrugated, insulated white plastic was used to construct chamber (enclosure) walls. The enclosure was replaced periodically to allow chamber heights to increase as plants grew. Chamber heights were 10.8, 48.3, and 88.9 cm during both seasons. A small battery-operated fan was used to mix air within the chamber. A vent tube permitted equalization of air pressure inside and outside of the chamber. Chamber air temperature was monitored to within ±0.5°C. During 1998, sampling was 30 min after chambers were placed on bases. On each sampling date during 1998 (Table 1 ), one of four replicates of each treatment was sampled at 0, 10, 20, 30, and 60 min. Samples from chambers in all replicates were taken at 0, 15, and 30 min after chamber placement during 1999. Forty-milliter samples were extracted using a syringe and immediately injected into evacuated, aluminum sealed, air-tight serum vials. Air was pressurized inside the vials to minimize sample contamination during storage. Samples were analyzed using gas chromatography (GC) within 2 to 3 d following collection in the field.


View this table:
[in this window]
[in a new window]

  		Table 1. Dates and days after planting (DAP) in 1998 and 1999 when emissions measurements were made using chamber, core, and acetylene inhibition methods. Planting dates were 15 July 1998 and 14 June 1999.

 
The N2O emission rate was calculated from the equation

Formula
where F is the flux density (g N2O-N ha–1 d–1), k is a units conversion factor (0.00018), T is the temperature of the air within the chamber (K), V is the volume of the air within the chamber (cm3), A is the area of the soil within the chamber (cm2), and {Delta}C/{Delta}t is the rate of change in the concentration of N2O in the air within the chamber (µmol mol–1 N2O min–1).

The value of {Delta}C/{Delta}t was estimated by linear regression analysis of N2O concentration vs. time. The curvilinear model of Hutchinson and Mosier (1981, especially Eq. [6]) was also used to estimate {Delta}C/{Delta}t at time t = 0 min for comparison with {Delta}C/{Delta}t estimated by linear regression. Estimation of {Delta}C/{Delta}t by the curvilinear model was subject to the mathematical criterion: (C1 – C0)/(C2 – C1) > 1 where C0, C1, and C2 were the N2O concentrations measured at times 0, t1, and 2t1, respectively. For 1998, the N2O concentrations measured at t = 0, 30, and 60 min were used in the curvilinear model. For 1999, concentrations measured at t = 0, 15, and 30 min were used.

Statistical comparison (Neter and Wasserman, 1974) was done to determine if {Delta}C/{Delta}t != 0 (P < 0.05) from the linear regression analysis. The comparison also permitted estimation of the lower detection limits of N2O emissions. For 1998, the limits were about 6.3, 6.6, and 14 g N2O-N ha–1 d–1 for the 10.8-, 48.3-, and 88.9-cm-tall chambers, respectively. For 1999, the limits were about 6.6, 6.0, and 10 g N2O-N ha–1 d–1, respectively. These relatively low limits permitted detection of some statistically significant negative emission rates.

Core Measurements
Nitrous oxide flux was also measured using an intact core technique (Ryden et al.1987; Weier et al., 1993). Soil cores were collected in 1998 using a split core sampler (Art's Manufacturing and Supply, American Falls, ID). Inside the sampler was a perforated plastic sleeve 19 cm long and 7 cm in diameter. The sampler was pounded into the ground and removed. Severe soil compaction occurred using this method on moist soils. Cores were taken in 1999 by pressing plastic sleeves into the soil with a rubber mallet. Pressing the plastic sleeves into the soil reduced compaction within the soil cores. Three soil cores from the top 23 cm of soil were taken from each of the 16 plots in 1998. Two cores per plot were taken during the 1999 season (Table 1). Each core sample was placed in a 1-L glass jar and sealed with a lid containing two rubber septa. One septum was used for injection of acetylene (C2H2) for measurement of N-gas emissions (see below), and a Tensicorder (Soil Measurement Systems, Tucson, AZ) was applied to the second septum to ensure that the jar did not leak. Jars were incubated outdoors under shade. Forty milliliters of gas was extracted from the septum after 24 h of incubation and placed in evacuated, aluminum sealed, air-tight serum vials for N2O analysis using GC.

Nitrogen Gas Measurements
The C2H2 inhibition technique described by Ryden et al. (1987) and Mosier and Klemedtson (1994) was used to measure N-gas (N2O plus N2) emissions due to denitrification plus nitrification. Intact cores were collected as described above. Two cores from each of the 16 plots were taken from the top 23 cm of soil and placed inside 1-L jars for incubation. Forty milliliters of air was removed from each jar and replaced with 40 mL of C2H2 for a final C2H2 concentration of 7 to 10% (v/v). A 40-mL sample was withdrawn at the end of 24 h and analyzed using GC.

Denitrifying Enzyme Activity
Denitrifying enzyme activity was analyzed on field-moist samples collected at the beginning and end of the 1999 season, using the method of Luo et al. (1996). Twenty grams of soil and 20 mL of a solution containing glucose and KNO3 were placed in 125-mL Erlenmeyer flasks. Final concentrations were 50 mg NO3––N kg–1 soil and 300 mg glucose-C kg–1 soil. The flasks were then capped with rubber stoppers equipped with two glass tubes. The tubes were used to remove all air in the flask and replace it with N2 and 5% (v/v) C2H2. Flasks were placed on a rotary shaker for 2 h and a 5-mL gas sample was collected from each flask for analysis of N2O by GC.

Gas Chromatography
Gas samples were analyzed using a Shimadzu 14A GC (Shimadzu Corp., Tokyo) equipped with a 63Ni electron capture detector source (370 MBq) heated to 300°C. A pure N2 carrier gas was used at a rate of 40 mL min–1. The GC was fitted with two stainless steel columns (2 mm i.d. by 3.05 m) filled with Porapak Q, 80/100 mesh (Supelco, Bellefonte, PA), and maintained at 100°C. The retention time for the N2O was 3.4 min.

Concentrations of N2O were calculated from a standard curve using certified standards containing concentrations of 500 and 1000 nmol mol–1 N2O in N2 (Scotty Specialty Gasses, Plumbsteadville, PA). Samples that contained concentrations >>1000 nmol mol–1 required a separate standard curve to be established by diluting 1000 µmol mol–1 of N2O in N2 to 340, 102, and 51 µmol mol–1.

Soil Analysis
After incubation and gas sampling, soils were removed from the plastic sleeves, oven dried at 65°C, and ground and sieved to <2 mm. Nitrate concentrations were determined using 2 mol L–1 KCl extraction and steam distillation (Bremner and Mulvaney, 1982). Soluble organic C (SOC) concentrations were determined in 1:1 water extracts. Soil extracts were treated with two drops of 3 mol L–1 H3PO4 to remove any carbonate in solution. Samples were then processed with a total organic C analyzer (Beckman 915A, Beckman Instruments, Fullerton, CA). The SOC standards were made using potassium biphthalate in CO2–free water. The water-filled pore space (WFPS) percentage was calculated by measuring the volume of soil within a core and determining the mass/volume of water removed through oven drying the soil cores.

Data Analysis
The mean seasonal cumulative N2O-N and N-gas values reported in Table 2 were calculated in a similar manner for both chamber and core methods. First, we assumed that nonzero fluxes occurred only on the days when measurements were made. Next, we summed the daily N flux values to obtain the seasonal total N for all days from each of the 16 replicate measurement locations in the field. Third, we averaged the four replicate totals for each of the four treatments.


View this table:
[in this window]
[in a new window]

  		Table 2. Treatment means (and SE) of seasonal cumulative N2O-N emissions (measured by chamber and core methods) and seasonal cumulative N-gas emissions (measured by acetylene inhibition method).

 
Gas fluxes for each plot and sampling method were arithmetically meaned and then averaged across treatments to calculate means and standard errors for each treatment. An ANOVA for a split-plot design was used to evaluate the influence of elevated CO2 and water treatment on N2O, N gas, SOC, and DEA. Normal and lognormal distributions of N2O and N-gas emissions were determined using the Wilk–Shapiro method (Shapiro and Wilk, 1965). Coefficients of variation were determined for normal and lognormal data for comparison.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Carbon Dioxide and Irrigation Effects on Nitrous Oxide and Nitrogen Gas Emissions
Elevated atmospheric CO2 significantly affected N2O fluxes measured in chambers (Fig. 2 ) or soil cores (Fig. 3 ) on only one of the 17 sampling days (31 July 1998, 15 d after planting), suggesting that elevated CO2 will probably not increase N2O emissions from southwestern U.S. irrigated sorghum production systems similar to our study site. This is consistent with findings for unirrigated western U.S. ecosystems reported by Billings et al. (2002) for Mojave Desert soils and by Mosier et al. (2002) for Colorado shortgrass steppe. It is contrary to Ineson et al. (1998), who found a 27% increase in N2O emissions from elevated atmospheric CO2 at the Swiss FACE experiment. The N2O emission data presented by Ineson et al. (1998), however, were collected after 2 yr of continuous exposure to elevated CO2, while our measurements were made during single growing seasons of exposure to elevated CO2. Ineson et al. (1998) also used a C3 plant species (Lolium perenne L.), which are known for having greater physical responses than C4 species to elevated atmospheric CO2 (Poorter, 1993; Davey et al., 1999; De Luis et al., 1999). Therefore, the probability that differences in N2O emissions would be detected in their system should be greater.


Figure 2
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 2. Average N2O emissions measured using the chamber method during 1998 and 1999 (error bars indicate the standard error of the mean). Symbols in legend denote results from the ambient (Amb.) and elevated (Elev.) CO2 and irrigation (Irr) treatments both years.

 

Figure 3
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 3. Average N2O emissions measured using the core method during 1998 and 1999 (error bars indicate the standard error of the mean). Symbols in legend denote results from the ambient (Amb.) and elevated (Elev.) CO2 and irrigation (Irr) treatments both years.

 
On two dates during each year, irrigation treatment significantly affected N2O emissions measured using chambers (12 and 17 Sept. 1998, 24 Sept. and 8 Oct. 1999; Fig. 2). During 1998, the highest N2O flux on a single date was observed in the FACE, Dry treatment; in 1999, the highest values was observed in the FACE, Wet treatment. Westerman and Tucker (1978) reported increased short-term N2O fluxes from soils after water addition, following long periods without water.

The rates of N2O emissions illustrated in Fig. 2 were calculated using linear regression analysis of concentration vs. time to estimate {Delta}C/{Delta}t. Rates were also estimated using the curvilinear model of Hutchinson and Mosier (1981), but relatively few data met the criteria (C1 – C0)/(C2 – C1) > 1. In fact, N2O concentration data satisfied the criteria for only 28.5% of the flux measurements (8 of 28) in 1998. Similarly, in 1999, 29.6% (19 of 64 measurements) met the criteria. For data that satisfied the criteria, the ratio ({Delta}C/{Delta}t from linear regression)/({Delta}C/{Delta}t from curvilinear model) was, on average (± standard deviation), 1.25 ± 0.45 and 1.06 ± 0.59 in 1998 and 1999, respectively. Mean ratios >1 were not expected because {Delta}C/{Delta}t is typically largest when the enclosure is installed on the base at time t = 0 min (Hutchinson and Mosier, 1981).

Nitrous oxide emissions measured using the chamber method were, on average, uniformly low (<500 g N2O-N ha–1 d–1) both years, except for 12 Sept. 1998 (Day 58 after planting), when average fluxes >1000 were recorded (Fig. 2). In 1998, statistical analysis showed that 46.4% of the measurements had {Delta}C/{Delta}t = 0 (P < 0.05). In 1999, 34.5% had {Delta}C/{Delta}t = 0.

The N2O fluxes measured with cores (Fig. 3) were usually lower than those measured with chambers (Fig. 2). The difference in the scale of measurement between the two methods should not have resulted in higher fluxes measured with chambers. It is possible, however, that chamber sampling better captured rhizosphere N2O emissions and N2O release from plant leaves.

To our knowledge, there are few comparisons between the chamber and core methods in the literature (Aulakh et al., 1991; Bronson et al., 1997). Our objective was not specifically to compare the two methods. Rather, we chose to use both methods due to the inherently high variability of N2O and N-gas fluxes from soils. Chamber and core measurements of N2O emissions were performed simultaneously on a total of six dates during the experiment. In 1998, both methods showed relatively low emissions on the two similar sampling dates. In 1999, three of the four congruent sampling dates showed the same trend of low N2O emissions (Fig. 2 and 3). At 58 d after planting in 1999, the chamber method produced higher N2O fluxes than observed with cores for two of the treatments (elevated CO2 with Wet and Dry irrigation treatments). Considering the high variability inherent in measuring N-gas emissions, it appears that the two methods exhibited similar trends.

Similar to N2O emissions, elevated atmospheric CO2 did not significantly affect N-gas (N2 plus N2O) emissions on any dates in 1998, and affected only one sampling date in 1999. Irrigation treatment did not affect N-gas emissions in 1998, but did affect emissions on four sampling days in 1999. Measured N-gas emissions were much higher during 1999 than 1998 (Fig. 4 ). This may be due to the higher WFPS on the sampling dates in 1999. During 1998, WFPS was <55% when samples were collected, while values up to 85% were measured during 1999.


Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 4. Average N-gas (N2O + N2) emissions measured using the acetylene inhibition method during 1998 and 1999 (error bars indicate the standard error of the mean). Symbols in legend denote results from the ambient (Amb.) and elevated (Elev.) CO2 and irrigation (Irr) treatments both years.

 
Seasonal cumulative N2O-N emissions measured using chambers (Table 2) were <1.5 kg N ha–1 in all cases. Thus these losses were not of agronomic significance. Losses measured using chambers were similar between the seasons, with somewhat higher losses measured during 1998. Losses of N2O-N measured using cores were much lower during 1998 than 1999; the reason for this is not known. Losses of N2O-N measured using cores were of similar magnitude as those measured using chambers in 1999. Nitrogen-gas losses measured using C2H2 inhibition in 1998 were very low and similar in magnitude to N2O-N losses measured using cores. Nitrogen-gas losses measured during 1999 were as high as 3.7 kg N ha–1, and were highest with elevated CO2 and the Wet irrigation treatment. None of these differences were statistically significant, however. The ratio N2O-N/N-gas loss measured using cores was calculated for both seasons (Table 3 ). This ratio should be <1; however, during 1998 some ratios were >1. This may be an artifact of the very low fluxes measured using both core methods during 1998. During 1999, however, the ratios were <<1, suggesting that the main N loss product was N2. Soil water contents were, on average, much higher during 1999. Aulakh et al. (1992) found that when WFPS was >70%, 80 to 90% of N-gas emissions from denitrification were N2 rather than N2O.


View this table:
[in this window]
[in a new window]

  		Table 3. Treatment means (n = 4 replicates) and SE of the ratios of seasonal cumulative N2O-N emissions (measured by core method) to seasonal cumulative N-gas emissions (measured by acetylene inhibition method).

 
Carbon Dioxide and Irrigation Effects on Denitrifying Enzyme Activity and Soluble Organic Carbon
Neither DEA nor SOC (data not shown) were affected by elevated atmospheric CO2 or by irrigation treatment. The lack of response in SOC may be because we did not measure SOC in rhizosphere soil. The DEA, however, was significantly higher at the end of the season in all treatments (Fig. 5 ). Using wheat, Smart et al. (1997) found that the denitrification potential under elevated atmospheric CO2 was three to 24 times higher than the denitrification potential with ambient CO2. Their conclusion was that the effect of elevated CO2 on the denitrification potential was due to increased root nonstructural carbohydrate accumulation and consequent root exudation with elevated atmospheric CO2. Zak et al. (1993) found that respired C was significantly increased under elevated atmospheric CO2 (692 µmol mol–1), compared with ambient CO2, in the rhizosphere. Similar to our findings, however, they did not find significantly increased C in the bulk soil with Populus grandidentata Michx. Similarly, Marilley et al. (1999) reported that the total number of bacteria in bulk soil remained unchanged under elevated atmospheric CO2 but that bacteria in the rhizosphere of Lolium perenne increased. Carnol et al. (2002) found that 4 yr of elevated CO2 with Scots pine (Pinus sylvestris L.) had direct effects on nitrification and denitrification. They hypothesized that the effect resulted from microbial physiological adaptation or selection of nitrifiers under elevated CO2.


Figure 5
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 5. Average denitrifying enzyme activity at 26 and 102 d after planting in 1999 (error bars indicate the standard error of the mean). Amb. = ambient, Elev. = elevated.

 
Barnard et al. (2004a) found that several years of elevated CO2 in four European grasslands had little effect on microbial biomass N, nitrifying enzyme activity (NEA), DEA, or extractable soil NH4+ and NO3–. Exceptions to this were DEA and NO3–, which decreased by 22 and 45% at a French grassland site under elevated CO2. They concluded that elevated CO2 may increase microbial immobilization of N in highly disturbed ecosystems, with the effect weakening as the system approaches equilibrium.

Rate-Limiting Factors for Nitrous Oxide Production
Nitrate (Westerman and Tucker, 1978), SOC (Burford and Bremner, 1975), DEA (Luo et al., 1996), and WFPS (Linn and Doran, 1984; Davidson, 1991; Jorgensen and Jorgensen, 1997) have all been found to limit denitrification. We found that NO3– was weakly correlated with N2O emissions (Fig. 6 ), which suggests that NO3– alone was not limiting in our system. The SOC (data not shown) was also weakly correlated with N2O emissions (R2 = 0.126), suggesting that SOC alone was also not a rate-limiting factor for denitrification. Fluxes of >500 N2O-N g ha–1 d–1 occurred only when WFPS (Fig. 7 ) was ≥55%. Linn and Doran (1984), Davidson (1991), and Jorgensen and Jorgensen (1997) reported a 60% critical WFPS for high N2O emissions. Higher N-gas production than N2O production also occurred at >55% WFPS. This, and the low N2O-N/N-gas ratios measured during 1999 (Table 3), suggests that the end product for denitrification in this system was N2 rather than N2O at >55% WFPS. Aulakh et al. (1992) and Quian et al. (1997) reported that at >70% WFPS, 80 to 90% of N-gas emissions were N2 rather than N2O for denitrification, consistent with our data from 1999, reported in Table 3. Ineson et al. (1998) found that N2O dominated denitrification losses; however, they used C2H2 inhibition 7 d after a simulated rainfall. Therefore, it is unlikely that WFPS was >60% when they determined the end product of denitrification for their system. In our experiments, when WFPS was <55%, N2O fluxes were at times negative, indicating that denitrification could have acted as a sink for N2O at low WFPS, similar to findings from Robertson and Tiedje (1987).


Figure 6
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 6. Nitrous oxide emissions measured using the core method vs. the concentration of NO3– on selected days during 1998 and 1999.

 

Figure 7
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 7. Nitrous oxide emissions measured using the core method and N-gas (N2O + N2) emissions measured using the acetylene inhibition method vs. water-filled pore space on selected days during 1998 and 1999.

 

   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Nitrous oxide and N-gas emissions were not significantly affected by elevated CO2 concentrations measured in either chambers or core incubation. These results, while limited in scope because of the short duration of elevated atmospheric CO2, suggest that as CO2 concentrations increase, there will not be major increases in denitrification in C4 cropping environments such as irrigated sorghum in the desert southwestern United States. Seasonal cumulative N2O-N emissions measured using chambers were <1.5 kg N ha–1. Seasonal N-gas losses measured during 1999 were as high as 3.7 kg N ha–1, and were highest with elevated CO2 and the high-irrigation treatment. Elevated CO2 is unlikely to increase emissions of N2O from irrigated C4 systems such as sorghum. Neither NO3– nor SOC alone were limiting factors for N2O emissions. High N2O and total N-gas emissions occurred only at >55% WFPS, and under such conditions the major N gas lost was N2.The DEA increased during each season, but not due to elevated CO2 or irrigation treatment.


   ACKNOWLEDGMENTS
 
We thank the USDA Arid Lands Research Center (formerly the USDA Water Conservation Laboratory) for setting up and maintaining the sampling sites and equipment used in this project, and the members of the Soil Fertility Lab at the University of Arizona (1998–2001) for their participation in field work during this study. This research was supported by Interagency Agreement no. DE-AI03-97ER62461 between the Department of Energy Office of Biological and Environmental Research, Environmental Sciences Division and the USDA, Agricultural Research Service (Bruce A. Kimball, PI); by Grant no. 97-35109-5065 from the USDA, Competitive Grants Program to the University of Arizona (Steven W. Leavitt, PI); and by the USDA, Agricultural Research Service. It is part of the DOE/NSF/NASA/EPA Joint Program on Terrestrial Ecology and Global Change (TECO III). This work contributes to the Global Change in Terrestrial Ecosystems (GCTE) Core Research Programme.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication January 22, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Welzmiller, J. T.
Right arrow Articles by Thompson, T. L.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Welzmiller, J. T.
Right arrow Articles by Thompson, T. L.
Agricola
Right arrow Articles by Welzmiller, J. T.
Right arrow Articles by Thompson, T. L.
Related Collections
Right arrow Soil Kinetics
Right arrow Redox Processes

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome