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


     

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 Similar articles in ISI Web of Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rochette, P.
Right arrow Articles by Côté, D.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Rochette, P.
Right arrow Articles by Côté, D.
Agricola
Right arrow Articles by Rochette, P.
Right arrow Articles by Côté, D.
Related Collections
Right arrow Agroclimatology
Right arrow Air Pollution
Right arrow Soil Biochemistry
Published in Soil Sci. Soc. Am. J. 68:1410-1420 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

DIVISION S-8—NUTRIENT MANAGEMENT & SOIL & PLANT ANALYSIS

Carbon Dioxide and Nitrous Oxide Emissions following Fall and Spring Applications of Pig Slurry to an Agricultural Soil

Philippe Rochettea,*, Denis A. Angersa, Martin H. Chantignya, Normand Bertranda and Denis Côtéb

a Soils and Crops Research and Development Centre, Agriculture and Agri-Food Canada, 2560 Hochelaga Blvd., Sainte-Foy, QC, Canada, G1V 2J3
b Institut de Recherche et de Développement en Agroenvironnement, Complexe Scientifique, 2700 rue Einstein, Sainte-Foy, QC, Canada, G1P 3W8

* Corresponding author (rochettep{at}agr.gc.ca).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
In Québec, most pig slurry is applied to agricultural soils in the spring and fall. A study was initiated to compare the impact of the contrasting spring and fall weather conditions on CO2 and N2O emissions, and on the transformation of pig slurry C and N in a loamy soil cropped to maize (Zea mays L.). Treatments were approximately 200 kg total N ha–1 either as a spring (SPRING) or fall (FALL) application of pig slurry, and 150 kg N ha–1 as NH4NO3 (control). Fluxes of CO2 and N2O, and soil O2, CO2, N2O, NH4+, NO3–, extractable C and microbial biomass C (MBC) contents were measured 50 times over a 1-yr period. Fluxes of N2O were generally low during the experiment but were greatly increased in recently manured soils when soil O2 concentration fell below 0.20 mol mol–1. Soil was warm and well-aerated following spring slurry application. Under these conditions, slurry NH4–N was rapidly nitrified and high N2O emissions attributed to denitrification occurred when soil was rewetted by abundant rainfall. For the fall applied slurry, wet and cool conditions limited net nitrification and resulted in little accumulation of NO3–N, thus limiting potential for subsequent denitrification and N2O emissions. Cumulated N2O emissions during the experiment represented 1.74, 2.73, and 1.14% of added N in the FALL, SPRING, and NH4NO3 plots, respectively. Fluxes of CO2 and cumulated CO2–C losses were also greater for SPRING than for FALL application. Our results clearly show that the impacts of the timing of animal manure application on N2O emissions cannot be generalized, but will vary between years in response to interactions between crop, climatic, and soil factors.

Abbreviations: FCO2, soil-surface flux of CO2 • FN2O, soil-surface flux of N2O • MBC, microbial biomass C • WFPS, water-filled pore space


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
ATMOSPHERIC N2O contributes to stratospheric ozone depletion and global warming. Estimates of soil–surface N2O emissions following application of animal manure account for approximately 3.5 Pg CO2–equivalent annually in Canada (Desjardins and Riznek, 2000), or 9% of all anthropogenic sources of N2O. However, the accuracy of these estimates is limited by our incomplete understanding of the interactions between manure addition, and the biological and physical parameters controlling N2O production and consumption in soils.

Nitrous oxide is a free intermediate product of respiratory denitrification in agricultural soils and can also be produced during nitrification when low O2 levels limit the complete oxidation of intermediate products to NO–3 (Firestone and Davidson, 1989). Farm animal slurries contain large amounts of NH4–N and easily decomposable organic C that can stimulate N2O production in soils (Christensen, 1983; Eggington and Smith, 1986; Sharpe and Harper, 1997; Rochette et al., 2000b; Whalen et al., 2000). The impact of pig slurry application on N2O production is strongly influenced by crop and soil conditions. Crop uptake has a direct effect on soil mineral N levels and therefore on the amounts of substrates available for nitrification and denitrification (Chantigny et al., 1998). For example, lowest denitrification rates and N2O losses in grasslands occurred when slurry was applied during periods of rapid plant growth (Thompson and Pain, 1989; Allen et al., 1996; Chadwick, 1997). Aeration and temperature are the main environmental factors controlling nitrification and denitrification in soils. The rate of these reactions increases with temperature but responds differently to soil aeration. Nitrification is a process that requires well-aerated conditions while denitrification is inhibited by O2 and occurs when aeration is restricted, such as in wet soils (Firestone and Davidson, 1989).

Soil temperature, soil water content, and crop growth rate are related to climatic conditions and to astronomical rhythms. Thus, the time of year when animal slurry is applied strongly influences the conditions under which the transformations of the added C and N substrates will occur and the amounts of N2O produced. Several studies have aimed at quantifying the effect of time of application on denitrification and N2O emission rates on grasslands. While slurry application at spring time has often resulted in rates lower than at other times of the year because of the higher crop N uptake (Allen et al., 1996; Chadwick, 1997; Thompson and Pain, 1989), higher rates have also been observed in spring (Chadwick et al., 2000), fall (Weslien et al., 1998), and winter (Allen et al., 1996) in response to high soil water contents. Few studies have addressed the effects of time of slurry application on N2O emission rates in annual crops. In Québec, eastern Canada, pig slurry cannot be applied during winter. Therefore, it is a common practice for farmers to apply slurry in the fall and in the spring to make optimum use of their manure storage capacity. Our objective was to compare the impact of the contrasting spring and fall weather conditions on CO2 and N2O emissions, and on the transformation of pig slurry C and N in a loamy soil cropped to maize.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The study was conducted on a Le Bras loam (frigid, fine loamy, mixed Aeric Haplaquept) (0.28 g g–1 sand, 0.48 g g–1 silt, and 0.24 g g–1 clay) near Québec City, Canada (latitude 46° 05', longitude 71° 02', altitude 110 m) from October 1998 to October 1999. On average, Québec City receives 1174 mm of annual precipitations while mean daily air temperature is 19.1°C in July and –12.1°C in January. Soil pH was 5.9, and soil C and N contents averaged across treatments were 5.9, 21.4, and 1.42 g kg–1, respectively, before the 1998 fall slurry application. Treatments consisted of application of pig slurry in the fall (13 Oct. 1998) (FALL) and in the spring (17 May 1999) (SPRING) at a rate corresponding to approximately 200 kg total N ha–1 (Table 1). Control plots (NH4NO3) receiving mineral fertilizer broadcasted before planting (17 May 1999; 150 kg N ha–1 as NH4NO3, 150 kg P2O5 ha–1, 150 kg K2O ha–1) were also included in the experimental design. Treatments were replicated three times and plots (3 by 9 m) were arranged in a randomized complete block design. The slurry was from a commercial hog and sow operation and was applied using a trail hose applicator equipped with harrow teeth for immediate shallow (0–10 cm) incorporation. In 1998, the slurry had a pH of 8.0 and contained 3.1 kg m–3 total N, 2.2 kg m–3 NH4–N, and 0.41 kg m–3 P. In 1999, the slurry had a pH of 7.8 and contained 1.8 kg m–3 total N, 1.1 kg m–3 NH4–N, and 0.56 kg m–3 P. The plots were moldboard plowed and disked once before fall application. Plots were kept bare in autumn 1998 and maize was planted on 17 May 1999 at 76000 seeds ha–1 and 0.75-m interrows, and harvested on 15 Oct. 1999.


View this table:
[in this window]
[in a new window]
  		Table 1. Summary of C and N inputs and gaseous outputs during the experiment.

 
Soil-Surface Carbon Dioxide and Nitrous Oxide Fluxes
In situ FCO2 and FN2O were measured 50 times from 13 Oct. 1998 to 15 Oct. 1999 using nonsteady-state chambers. Two square acrylic frames (0.55 by 0.55 m, 0.14 m height, 6.35 mm wall thickness) were inserted to a depth of 0.10 m in each plot on 10 Oct. 1998 and were left at the same locations for the duration of the experiment. The height of the frames was measured (20 measuring points per frame) at regular intervals during the experiment to account for variations in headspace due to soil settling. Fluxes of N2O were measured using nonflow-through chambers detailed by Lessard et al. (1994). At sampling time, the frames were tightly covered with a vented (Hutchinson and Mosier, 1981) and insulated square plexiglass lid, and air samples were collected at regular intervals (0, 10, 20, and 30 min). Samples were taken through a rubber septum into 7.5-mL pre-evacuated glass vials using a two-way needle. Losses of N2O during storage were estimated to <5% after 120 d (E. van Bochove, personal communication, 2003). The gas samples were analyzed for N2O concentrations within 2 wk by means of a gas chromatograph (Model 5890 Series II, Hewlett-Packard, North Holliwood, CA) fitted to an electron capture detector. Fluxes of CO2 were measured by flow-through chambers detailed by Rochette et al. (1997). The CO2 flux measuring system was equipped with a LI-6200 CO2 analyzer (LI-COR Inc, Lincoln, NE) and a vented and insulated 0.14-m high square plexiglass chamber covering the same area as the frames. During each flux measurement, the chamber was fixed to a frame and the CO2 concentration inside the chamber was measured once every second during 2-min deployments. Flux measurements were always made between 0900 and 1200 h (eastern standard time).

Soil-surface CO2 and N2O fluxes (FG) were calculated using the following equation (Hutchinson and Livingston, 1993):

[1]
where {partial}C/{partial}t is the rate of increase in gas concentration inside the chamber at deployment time = 0, A is the surface area covered by the chamber (0.3025 m2), V is the total chamber volume (approximately 0.012 and 0.072 m3 depending on frame and chamber height), Mmol is the molar mass of the measured gas (44 g mol–1 for both gases) and Vmol is the volume of a mole of gas at the air temperature inside chamber at deployment time = 0 (0.022 to 0.025 m3 mol–1). When gas concentration in chambers varied linearly with time, {partial}C/{partial}t was obtained from simple linear regression. The relationship between concentration and time was tested with the Student's t test for n = 4 and {alpha} = 0.10 (Hutchinson and Livingston, 1993). Second-order polynomial equations were fitted to concentrations when {partial}C/{partial}t decreased with time, and the Student's t test was applied to the slope of the curve at time = 0. In both cases, if the calculated t was greater than the critical value of t (2.78), the rate of change in concentration was considered significantly different from zero. Cumulative N2O-N losses were calculated by linearly interpolating N2O emissions between sampling dates, assuming that measurements made between 0900 and 1200 h were a good estimator of average daily FN2O.

Concentrations of O2, CO2, and N2O in the soil profile were monitored within 2 m of the chamber frames. Soil air samples were collected in each plot every day when flux measurements were made. The soil gas probes were made of plastic mesh cylinders (10 cm length; 1.5 cm diameter) containing glass beads (3 mm diameter). The cylinders were inserted horizontally at depths of 7.5 and 15 cm on 10 Oct. 1998 and connected to the surface using plastic tubes (Bev-a-line IV, Ryan Herco Industrial Plastics, Seattle, WA) (0.60 m length; 6.35 mm o.d.; 3.18 mm i.d.) with a two-way Luer-type stopcock valve (Cole Parmer, Vernon Hills, IL) at the surface end. Air samples were collected through a polyisoprene septum (male luer-lock stopper with injectable membrane, Vygon, Ecouen, FR) using a 10-mL Plastipak syringe after expelling 20 mL of air from the tubes to account for the dead volume of the tubes. The gas samples were analyzed for N2O, O2, and CO2 concentrations within 2 d of sampling by means of gas chromatograph fitted to electron capture and thermal conductivity detectors (Model 3800, Varian, Walnut Creek, CA). No O2 and CO2 measurements were made in September 1999 due to technical problems with the analytical instruments. Also, at several dates in the fall of 1998 and early spring 1999, water-saturated conditions prevented soil air sampling in some plots.

Soil temperature at a depth of 7.5 cm inside each frame and volumetric water content at depths of 7.5 and 15 cm (one profile per plot) were measured at the time of gas flux measurements. Soil temperature near the frames was also monitored at 7.5 and 15 cm and hourly averages were stored in a datalogger (CR10, Campbell Scientific, Logan, UT). Soil temperature was measured using copper-constantan thermocouples while soil moisture was determined by three-rod time-domain reflectometry probes read using a cable tester (Model 1502C, Tektronix Inc., Beaverton, OR). Water-filled pore space (WFPS) was calculated using bulk densities measured in May 1999 (1.1 g cm–3 in all treatments) and assuming a mineral particle density of 2.65 g cm–3. Rainfall was measured daily with standard rain gauges located 200 m from the experimental plots. Statistical analysis was performed on log-transformed flux values. Treatment effects were examined for each sampling date using the General Linear Models procedure of SAS (SAS Institute, 1989).

Soil Analyses
Composite (≥3) soil samples (0–12 cm) were collected outside chamber frames in each plot 43 times during the experiment. Samples were returned to the laboratory and extracted within 4 h with 1 M KCl (10 g field-moist soil in 50 mL). The extracts were frozen until analysis for NO3–N using an ion chromatograph (Model 4000i, Dionex, Sunnyvale, CA) and NH4–N by colorimetry (N'konge and Ballance, 1982). Additional soil samples were sieved in the field at 6 mm and stored immediately at 4°C. Soil MBC measurements were performed within 24 h of sampling using the chloroform-fumigation extraction technique (Wu et al., 1990). Two 50-g subsamples of field-moist soils were placed in 100-mL beakers. One subsample was fumigated for 24 h at room temperature in a vacuum desiccator containing 25 mL of CHCl3. The other subsample was kept in the dark at 4°C for 24 h. Both fumigated and nonfumigated soils were extracted with 100 mL of 0.25 M K2SO4. After shaking for 1 h on a reciprocal shaker, the suspensions were centrifuged at 1000 x g and filtered (Whatman 934-AH). The organic C content of the extracts was determined by UV-persulfate oxidation on a DC-180 Carbon Analyzer (Dorhman Co., Santa Clara, CA). An extraction efficiency (Kec factor) of 0.45 was used to calculate the MBC (Wu et al., 1990). The K2SO4–extractable C (EC) from the nonfumigated soil was used as an estimate of soil available C. A fraction of the soil samples was air-dried and sieved at 2 mm for the determination of soil pH (CaCl2, 0.01 M). Total C and N contents were determined on a Leco CNS 1000 (LECO Inc., St. Joseph, MI) on finely ground (<0.05 mm) air-dried soil samples.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Soil Temperature and Water-Filled Pore Space
Pig slurry application had no effect on soil water content and temperature, so values were averaged over the three treatments (Fig. 1a,c). The wet and cool soil at the time of fall slurry application contrasted with the dry and warm conditions at the time of spring application. Soil WFPS at 15 cm was near 70% at the onset of the experiment (13 Oct. 1998) and low evaporation maintained WFPS high during the entire fall of 1998 despite little rainfall (Fig. 1a). In 1999, water from snowmelt created wet soil conditions in early spring but the soil dried rapidly during an exceptionally dry spring. Soil water was replenished by abundant rainfall in late May and early June when nearly 100 mm of rain were received (Fig. 1a). The rest of the summer was of a succession of wet/dry cycles in response to rainfall temporal patterns during which WFPS remained mostly between 25 and 50%. In late summer and early fall 1999, abundant rainfall and low evaporation brought the soil water content back to fall 1998 levels (WFPS = 70%). Soil temperature at 15 cm was below 10°C at the time of fall slurry application and decreased until the first snowfall (Fig. 1c). In 1999, soil warmed up rapidly during the dry spring and temperature was near 20°C when slurry was applied to the SPRING plots. Soil temperature reached a maximum of 25°C in July followed by a gradual decrease to 3°C at the end of October.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 1. Temporal variations of (a) soil water-filled pore space at 15 cm and daily rainfall; (b) soil air O2 and CO2 concentration at 15 cm; (c) soil temperature and N2O concentration at 15 cm; (d) soil-surface N2O flux from a loam soil cropped to maize and amended with ammonium nitrate fertilizer (NH4–NO3), and pig slurry applied in the fall (fall slurry) or in the spring (spring slurry). Fall application of slurry was made on 13 Oct. 1998, and ammonium nitrate and spring slurry were applied on 17 May 1999. Error bars represent the LSD (P = 0.05) for treatment comparison at each date.

 
Soil Mineral Nitrogen
Soil ammonium and nitrate contents were low in all plots before fertilizer and slurry applications (Fig. 2a,b). Fertilizer application in spring 1999 immediately increased both soil NO–3 and NH+4 contents, while the slurry only increased soil NH+4. In the FALL plots, soil NH+4 contents were back to background levels 30 d after slurry application. That decrease was not associated with a simultaneous increase in soil NO–3, suggesting other fates than nitrification for the slurry NH+4 (NH3 volatilization and fixation on clays) or a rapid NO–3 loss from the surface soil layer (0–12 cm). Nitrates could be lost from the surface soil layer by leaching in the wet soil, or by a very close coupling between nitrification and denitrification (Petersen et al., 1991). The latter scenario is partly supported by the high soil water content and N2O concentrations during that period (Fig. 1a,c). However, nitrification is usually slow at low temperatures (Cookson et al., 2002) and high water contents (Thompson and Pain, 1989). Also, high levels of soil NO–3 in the FALL plots in spring 1999 (Fig. 2b) indicate a delayed effect of slurry application on NO–3 content. Slurry mineral N that was immobilized in microbial biomass in the fall of 1998 could have been gradually released and nitrified in the spring of 1999. Between 15 and 25% of total slurry NH+4–N" can be immobilized within a few days following pig slurry application (Morvan et al., 1997; Sorensen and Amato, 2002). The significant role of MBC as a transient pool of slurry N over the winter agrees with the large increase in biomass shortly after fall slurry application (Fig. 3a). The delayed effect of fall slurry application on soil NO–3 content could also be explained by the mineralization of slurry organic N and by the triggering of the mineralization of native soil N.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 2. Temporal variations of (a) soil NH+4–N (0–12 cm); (b) soil NO–3–N (0–12 cm), and (c) cumulated N2O emissions for a loam soil cropped to maize and amended with ammonium nitrate fertilizer (NH4–NO3), and pig slurry applied in the fall (fall slurry) or in the spring (spring slurry). Fall application of slurry was made on 13 Oct. 1998, and ammonium nitrate and spring slurry were applied on 17 May 1999. Error bars represent the LSD (P = 0.05) for treatment comparison at each date.

 


View larger version (33K):
[in this window]
[in a new window]
  		Fig. 3. Temporal variations of (a) microbial biomass C (0–12 cm); (b) soil K2SO4–extractable C (0–12 cm), (c) soil-surface CO2 flux, and (d) cumulated CO2 emissions at day of year (DOY) 174 (23 June 1999) for a loam soil cropped to maize and amended with ammonium nitrate fertilizer (NH4–NO3), and pig slurry applied in the fall (fall slurry) or in the spring (spring slurry). Fall application of slurry was made on 13 Oct. 1998, and ammonium nitrate and spring slurry were applied on 17 May 1999. Error bars represent the LSD (P = 0.05) for treatment comparison at each date.

 
Soil conditions in SPRING plots at the time of slurry application were much warmer and drier than in the fall of 1998 (Fig. 1a,c). Under these conditions, most slurry NH+4 was nitrified within 30 d following slurry application as indicated by the simultaneous decrease in soil NH+4 and increase in NO–3 (Fig. 2a,b). In late May 1999, NO–3 accumulated in all plots in absence of leaching and denitrification losses and when N uptake by the establishing maize plants was very small. After 15 June 1999, soil NH+4 and NO–3 contents were generally similar in all treatments.

Microbial Biomass and Extractable Carbon
The MBC and EC contents showed little variation in the NH4NO3 control plots during the experiment (MBC: 20 to 40 g C m–2; EC: 5 to 8 g C m–2) (Fig. 3a,b). Compared with NH4NO3 control plots, MBC and EC were increased by a factor of 2 to 10 following fall and spring slurry applications. Similar rapid increases in MBC following addition of pig slurry have been reported in laboratory (Saviozzi et al., 1997) and in field (Rochette et al., 2000a) studies. Slurries supply large amounts of easily available C such as volatile fatty acids (Kirchmann and Lunvall, 1993), which can sustain considerable microbial activity (Paul and Beauchamp, 1989; Sorensen, 1998). The increase in MBC following slurry application was relatively short-lived since the effects of fall and spring applications were small and most often not significant 25 d after application. The short duration of slurry-induced MBC contrasts with season long increases reported in plots receiving solid dairy cattle manure for the third year (Rochette and Gregorich, 1998) and pig slurry for the19th year (Rochette et al., 2000a). This is in agreement with the observation that background MBC values reflect the cumulative rather than the current-year input of C in the soil (McGill et al., 1986).

Soil Atmosphere
Concentration of a given gas at a given location in the soil air phase is the net result of its production, consumption, and transport rates. Gas transport in soils decreases with increasing soil water content because gas diffusivity is approximately 104 times slower in water than in air, while microbial gas production is affected by substrate availability and environmental factors such as water, temperature, and redox potential. Soil O2 concentrations were decreased and CO2 concentrations were increased at three dates in 1998 and four dates in 1999 in the slurry-treated plots compared with the NH4NO3 control plots (Fig. 1b). Since soil water content was similar across treatments, differences in soil O2 and CO2 are more likely the result of higher heterotrophic respiration in presence of slurry-derived C and N substrates than of differences in gas diffusivity. Oxygen concentration in soil air (15 cm depth) remained >0.20 mol mol–1 during the experiment except for three short post-rainfall periods in June and July 1999 when season low values of 0.16 mol mol–1 in SPRING, 0.17 mol mol–1 in FALL, and 0.185 mol mol–1 in NH4NO3 plots were reached. The modest decreases in O2 concentration in the NH4NO3 control plots are similar to those observed by Jorgensen et al. (1998) shortly after a nonamended dry soil was rewet (0.19 mol mol–1 at 5 cm). Soil O2 concentrations in FALL and SPRING plots are also in agreement with values measured by Stevens and Cornforth (1974) after an application of slurry to a soil column in the laboratory. Much lower values (<0.05 mol mol–1) were reported by Burford (1976) following the application of cow slurry under field conditions. However, surface sealing by the large rate of applied slurry (55 L m–2) in the latter study may have restricted the downward diffusion of oxygen in the soil profile (Stevens and Cornforth, 1974).

Temporal variations in CO2 concentrations essentially mirrored those of O2 but CO2 fluctuations were approximately half those in O2 (Fig. 1b). Assuming a respiratory quotient of 1, the lower CO2 concentration could be explained by the much greater (approximately 30 times) solubility of CO2 than O2 in soil water (Kilham, 1994; Chantigny et al., 2002). Fluctuations of CO2 concentration relative to those of O2 were greatest immediately following fall and spring slurry applications than at other times during the study. This could be explained by the greater contribution to total CO2 production by anaerobic reactions that have a respiratory quotient >1 and by nonbiological CO2 sources such as the dissociation of slurry carbonates (Sommer and Sherlock, 1996; Chantigny et al., 2001).

Soil N2O concentrations in the NH4NO3 plots remained close to atmospheric concentrations (0.3 µmol mol–1) during the study (Fig. 1c). Fall application of slurry resulted in a marked increase in soil N2O during most of the following month (up to 88 µmol mol–1). This effect was still measurable shortly after snowmelt in April 1999 but not beyond as the N2O concentrations in the FALL and NH4NO3 plots were nearly identical during the rest of the 1999 growing season. Spring application of slurry was not immediately followed by an increase in soil N2O. However, soil N2O in SPRING plots reached high levels (up to 89 µmol mol–1) following heavy rainfalls 21 and 43 d after slurry application. Peak concentrations measured in this study were smaller than those observed by Eggington and Smith (1986) but were much greater than those previously reported in soils amended with pig slurry (Rochette et al., 2000b) or solid cattle manure (Lessard et al., 1996) in eastern Canada. Reasons for the higher values in the present study are unclear but are likely related to soil conditions more favorable to microbial N2O production since higher rates of manure were applied in the previous studies. Production of N2O during denitrification depends not only on the amounts of C and N substrates but also on the N2/N2O ratio of the gaseous products. Pig slurry would favor N2 formation by increasing pH and C substrates while supplying little NO3 (Firestone and Davidson, 1989). However, the long period of aerobic conditions that has preceded the N2O bursts in 1999 has allowed for the nitrification of the slurry NH4–N. The dry soil conditions may also have decreased the N2/N2O ratio of denitrification by affecting nitrous oxide reductase more than nitrate reductase (Dendooven and Anderson, 1995). For example, low N2/N2O ratios have been measured in a cropped soil shortly after rewetting (Bergsma et al., 2002).

Periods of increased soil N2O concentration corresponded clearly with periods of restricted aeration under high WFPS or shortly following rainfall (Fig. 1a,c). These periods also coincided with the depletion in soil O2 (Fig. 1b), thereby suggesting the predominant role of denitrification for the N2O production. It is interesting to note how anaerobic activity developed in the soil despite relatively high O2 concentrations (16–19%). This observation could be explained by the concept of "hot spots" where O2 is consumed at a rate greater than that at which O2 can diffuse from the macropores to these biologically active sites. This situation occurs at discrete locations with high water and available C contents (Sexstone et al., 1985; Højberg et al., 1994); conditions that are met in the slurry-amended soil following rainfall. The relatively high O2 concentrations at which high values of soil N2O and FN2O were observed confirm that O2 concentration in soil air samples more likely describe conditions in the larger pores (Rolston, 1978) and that it may not be a good indicator of the aeration status of the soil at the microsite level (Sexstone et al., 1985; Højberg et al., 1994). It could be argued that WFPS is a better index of the availability of O2 for the soil microbes because of its direct measure of the soil O2 storage capacity and its indirect estimate of the resistance to O2 diffusion from the macropores to the biologically active sites. This is strongly supported by the absence of large FN2O at WFPS lower than 50% (Fig. 4), a value similar to that proposed by Linn and Doran (1984) and Chantigny et al. (1998) as the threshold where anaerobic processes such as denitrification become significant in the soil. However, in the present study, episodes of increased FN2O were always associated with relatively low but consistent decreases in soil O2 while high soil water contents often occurred without increased N2O emissions. We suggest that the advantage of O2 concentration in the soil atmosphere over WFPS as an indicator of N2O production episodes in this study may be linked to the fact that O2 measurements also reflect the availability of C substrates for heterotrophic activities, including respiration and denitrification.



View larger version (39K):
[in this window]
[in a new window]
  		Fig. 4. Relationships between soil-surface N2O emissions, soil nitrate (0–12 cm) and soil water-filled pore spaces (0–15 cm) in a loam soil following fall and spring application of pig slurry. Fall application of slurry was made on 13 Oct. 1998, and ammonium nitrate and spring slurry were applied on 17 May 1999.

 
Soil-Surface Carbon Dioxide Fluxes
Soil-surface CO2 fluxes in the NH4NO3 control plots were in the same range as those reported for agricultural soils under annual crops in eastern Canada (Rochette et al., 1991; Rochette and Gregorich, 1998). The effects of slurry application on CO2 fluxes were similar to those on MBC. Stimulating effects of manure application on both EC and FCO2 have been previously reported in laboratory incubations (Focht et al., 1979) and in the field (Gregorich et al., 1998; Rochette and Gregorich, 1998).

A strong CO2 flush often follows application of anaerobically stored slurry when slurry carbonates are dissociated in contact with an acidic soil (Sommer and Sherlock, 1996; Rochette et al., 2000a; Chantigny et al., 2001). Such stimulation of FCO2 occurred in the first 6 h after the spring application (Fig. 3c) but not after the fall application. The absence of a post-application burst in the FALL plots could not be explained by differences in carbonate content between the spring and fall slurries (data not shown). Alternatively, it may be hypothesized that larger amounts of CO2 have solubilized in the soil solution in the fall because of higher water contents and lower temperature as compared with the spring (Kilham, 1994; Chantigny et al., 2002).

The carbonate flush in the spring plots was short-lived and was followed by a period of increased FCO2 compared with NH4NO3 control plots (Fig. 3). During this period, soil heterotrophs were likely actively decomposing easily available organic C such as volatile fatty acids, which are abundant in anaerobically stored slurry (Paul and Beauchamp, 1989; Kirchmann and Lunvall, 1993; Chantigny et al., 2001). Application of slurry had no measurable effect on FCO2 one month after fall and spring applications. Soil-surface CO2 fluxes on all treatments increased in July and August in response to the higher soil temperature and greater contribution of root-rhizosphere respiration of the fully grown maize plants (Rochette and Flanagan, 1997).

Cumulated mineralized slurry C was calculated as the difference between emissions from the slurry and NH4NO3 control plots. Differences were cumulated until 23 June, date corresponding approximately with the period when the contribution of the maize root-rhizosphere respiration becomes significant in eastern Canada (Rochette and Flanagan, 1997). Cumulated mineralized slurry C was 51.1 g C m–2 in the SPRING plots (Fig. 3d) or 63% of the applied slurry C (Table 1). This value is within the range of pig slurry decomposition rates measured in the laboratory (Bernal and Kirchmann, 1992; Dendooven et al., 1998). In the FALL plots, cumulated CO2–C emissions during the period when the soil was not covered by snow were 27.2 g C m–2 or 33% of the slurry C. Of course this value is an underestimate of total slurry C mineralization because of the exclusion of FCO2 during the snow-covered period. Van Bochove et al. (2001) measured an accumulation of CO2 (0.4 mmol mol–1 at 10 cm) in the FALL plots during the 1998–1999 winter period, indicating that CO2 production occurred in the snow-covered soil. Fluxes of CO2 from snow-covered soil are usually small (Sommerfeld et al., 1993; Grant and Pattey, 1999) but could account for significant amounts of C because of the long duration of winter in Québec City (150 d).

Soil-Surface Nitrous Oxide Fluxes
Episodes of N2O emissions coincided with periods of increased N2O concentration in soil (Fig. 1c,d), thereby suggesting a relatively rapid gas diffusion through the top 15 cm of soil. Fluxes of N2O were small during most of the experiment on all plots with background values ≤0.15 mg m–2 h–1 (Fig. 1d) but were strongly stimulated during short episodes in plots receiving pig slurry. This is in agreement with several reports of increased N2O emissions following the application of animal slurries (Chadwick et al., 2000; Rochette et al., 2000b), solid manure (Lessard et al., 1996) or swine effluent (Sharpe and Harper, 1997). Highest emissions (1.40 mg m–2 h–1) were measured 3 d after the fall slurry application and returned to near-background levels at the onset of winter (Fig. 1d). Pig slurry contains large amounts of mineral N and easily available organic C, and we assumed that under the wet conditions that prevailed in the fall of 1998, most N2O production arose from denitrification (Bergstrom et al., 1994; Bremner, 1997).

Nitrous oxide fluxes were low in the FALL plots immediately following the spring 1999 snowmelt (0.08 mg m–2 h–1) but rapidly increased during a 7-d emission episode under relatively wet soil conditions (WFPS = 50–60%). High N2O fluxes have been reported at the time of spring thaw at several locations in Canada (Wagner-Riddle and Thurtell, 1998; Lemke et al., 1998; Grant and Pattey, 1999) and elsewhere (Christensen and Tiedje, 1990). Wagner-Riddle and Thurtell (1998) measured greater spring FN2O in plots where manure had been applied the preceding fall. Similarly, Lemke et al. (1998) reported that spring thaw FN2O were related to fall soil mineral N. In this study, the absence of spring thaw FN2O in plots other than the FALL plots is therefore in agreement with the observation that management practices that increase soil C and N in the fall can enhance N2O emissions the following spring when snowmelt water limits soil aeration. However, emissions were small compared with those observed by Wagner-Riddle and Thurtell (1998) and Lemke et al. (1998) possibly because of the rapid drying of the soil following snow melt (Fig. 1a). Soil N2O concentration measurements in FALL plots have been monitored during the 1998–1999 winter and have been published elsewhere (van Bochove et al., 2001). High soil N2O concentrations at 10-cm depth (40 µmol mol–1) indicated that N2O production (likely denitrification) occurred under snow-covered frozen soil and that part of this N2O may have contributed to the early spring N2O emissions. Following this early spring burst, FN2O in the FALL plots in 1999 remained low and similar to those in the NH4NO3 control (Fig. 1d).

Contrary to values observed in the FALL plots and in several other reports (Rochette et al., 2000b; Whalen et al., 2000), spring slurry application had no immediate impact on FN2O except for a weak (0.37 mg m–2 h–1) and short increase during the day following application (Fig. 1d). A similar delay between application and N2O emissions has been reported by Chadwick et al. (2000) and Chantigny et al. (2001). Emissions of N2O remained low (from 0.017–0.18 mg m–2 h–1) during the 18 d following application when slurry NH+4 was being nitrified in a relatively dry soil (20–25% WFPS). The N2O emissions during this period were only slightly higher than background levels and are in agreement with the low N2O yield of nitrification (Hutchinson and Davidson, 1993). This period was followed by abundant rainfall that triggered two successive episodes in which the highest emissions in this experiment were recorded (3.46 mg m–2 h–1). At that time, conditions were optimum for denitrification in the SPRING plots: (i) WFPS had raised to levels at which soil anaerobic conditions can occur (50– 60%) (Linn and Doran, 1984; Chantigny et al., 1998); (ii) nitrates had accumulated during the previous 3 wk in absence of leaching and low crop uptake; and (iii) available organic C from slurry was still present as indicated by the higher EC and FCO2 measured in the SPRING than in FALL and control plots. The rapid increase in FN2O following soil rewetting was attributed to denitrification (Davidson, 1992). This is in agreement with the capacity of a portion of soil denitrifiers (Rudaz et al., 1991) and denitrification enzymes (Smith and Parsons, 1985) to maintain their potential during long periods of dry soil conditions.

Interestingly, soil NO–3 contents were similar on all treatments at the time of FN2O bursts in 1999, but N2O emissions only occurred in the SPRING plots, likely as a result of greater C availability (Weir et al., 1993). The major role played by the C substrates in determining FN2O has been illustrated in laboratory incubations by correlation between CO2 and N2O emissions (Loro et al., 1997; Gödde and Conrad, 2000; Tenuta et al., 2000; Del Grosso et al., 2000; Azam et al., 2002). Our observation that the ratio of N2O-N losses to slurry CO2-C losses was the same in FALL and SPRING plots (0.012 g N2 O-N/g CO2–C) suggests that a similar relationship may be established under field conditions (Table 1). Following the high post-slurry application bursts in June 1999, FN2O in slurry plots did not rise above those in the NH4NO3 control plots during the rest of the study. During that period, frequent rainfalls occurred and soil NO–3 was above the 10 mg kg–1 threshold below which it limits denitrification (Barton et al., 1999), thereby suggesting that FN2O were probably limited by the lack of available organic C to sustain intense denitrification. Fluxes in all plots were slightly increased in the fall 1999 when soil water content was increasing back to near saturation.

Cumulated N2O-N emissions during the experiment were 599 mg m–2 in the SPRING, 323 mg m–2 in the FALL and 174 mg m–2 in the NH4NO3 control plots (Table 1). As for CO2, winter FN2O emissions in the FALL plots were not measured and the actual total N2O-N losses are likely greater than those reported. In the FALL plots, van Bochove et al. (2001) measured N2O concentration increases in the soil profile but not in the overlaying snow. They concluded that a frozen layer near the soil surface likely prevented the atmospheric release of the gas until snowmelt in late spring. It is impossible to estimate what part of that degassing was not included in our spring thaw FN2O episode but the apparent absence of large emissions during the 1999 winter (van Bochove et al., 2001) decreases the risk for a large underestimation due to the absence of measurements during that period.

The fraction of N applied as mineral fertilizer in the NH4NO3 control plots that was lost as N2O (1.14%) agrees with previously published values (Bouwman, 1996), but those measured in the FALL (1.74%) and SPRING (2.73%) plots amended with pig slurry were greater than the default emission coefficient (1.25%) proposed by the Intergovernmental Panel on Climate Change (IPCC) inventory guidelines (IPCC, 1997). The greater stimulation of FN2O by pig slurry than by mineral fertilizer is in agreement with the observation that most soil N2O arises from denitrification, which is largely controlled by the supply of organic matter and available soluble C (Bremner, 1997; Tenuta et al., 2000). Despite a few reports of lower FN2O on soils receiving manure (Velthof and Oenema, 1997; Li et al., 2002), there is growing evidence that N2O losses from land receiving manure are greater than those receiving mineral N (Christensen, 1983; Lessard et al., 1996; Rochette et al., 2000b). However, the amounts of N2O emitted following slurry application to agricultural soils are strongly dependent on soil and climatic conditions, and we suggest that locally determined emission coefficients could improve inventories of agricultural sources of greenhouse gases.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Fall application of manure in eastern Canada has greater potential for increasing N2O emissions than at other periods of the year for several reasons: (i) mineral N is added to soil at a time when plant uptake is low; (ii) low evaporation usually results in high soil water content which favors denitrification; (iii) denitrification may continue during winter when snow-covered soils are wet, cool, and frequently unfrozen (Chantigny et al., 2002); and (iv) fall supply of N and organic C can trigger important N2O losses during the following spring at the time of snow melt (Wagner-Riddle and Thurtell, 1998; Lemke et al., 1998). While these concerns are valid, our FN2O measurements during the snow-free season were greater following spring than fall application. These differences in N2O emissions were attributed to the contrasting soil environmental conditions at the time of spring and fall slurry applications. Wet and cool fall conditions limited net nitrification and resulted in little accumulation of NO–3, thus limiting potential for subsequent denitrification and N2O emissions. During winter, a frozen soil surface limited gaseous losses (van Bochove et al., 2001) and rapid soil drying following snow melt limited denitrification and N2O emissions in the spring. On the other hand, the soil was warm and well aerated following spring slurry application. Under these conditions, slurry NH4–N was rapidly nitrified and intense denitrification associated with high N2O emissions occurred when soil was rewetted by abundant rainfall. Our results clearly show that the impacts of the timing of animal manure application on N2O emissions cannot be generalized, but will likely vary between years in response to interactions between crop, climatic, and soil factors.


   ACKNOWLEDGMENTS
 
This work was supported by the PERD program of Agriculture and Agri-Food Canada. We thank P. Jolicoeur, N. Bissonnette, F. Ouellet, and J. Lizotte for assistance in soil-surface gas flux measurements, soil analyses, statistical analyses, and plot maintenance.

Received for publication December 31, 2002.


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




This article has been cited by other articles:


Home page
 Agron. J.Home page
E. R. Loria, J. E. Sawyer, D. W. Barker, J. P. Lundvall, and J. C. Lorimor
Use of Anaerobically Digested Swine Manure as a Nitrogen Source in Corn Production
Agron. J., June 26, 2007; 99(4): 1119 - 1129.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
P. Rochette, D. A. Angers, M. H. Chantigny, B. Gagnon, and N. Bertrand
In situ Mineralization of Dairy Cattle Manures as Determined using Soil-Surface Carbon Dioxide Fluxes
Soil Sci. Soc. Am. J., March 29, 2006; 70(3): 744 - 752.
[Abstract] [Full Text] [PDF]

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 Similar articles in ISI Web of Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rochette, P.
Right arrow