G Model AGEE-3390; No of Pages 8 Agriculture, Ecosystems and Environment xxx (2009) xxx–xxx Contents lists available at ScienceDirect Agriculture, Ecosystems and Environment journal homepage: www.elsevier.com/locate/agee Nitrous oxide emission at low temperatures from manure-amended soils under corn (Zea mays L.) Olga Singurindy 1, Marina Molodovskaya, Brian K. Richards *, Tammo S. Steenhuis Department of Biological and Environmental Engineering, Riley-Robb Hall, Cornell University, Ithaca, NY 14853-5701, USA A R T I C L E I N F O A B S T R A C T Article history: Received 21 October 2008 Received in revised form 10 March 2009 Accepted 10 March 2009 Available online xxx Manure fertilization of soil significantly impacts the level of nitrous oxide (N2O) emission. Despite their short duration, periods of significant N2O emissions during soil thaws in winter and spring are an important portion of the total annual emissions from agricultural lands. The goal of this study was to understand the effects of tillage, moisture content and manure application on N2O emissions from agricultural soils at low temperatures. We summarize here both field and laboratory experiments. The field chamber study was focused on quantification of N2O flux from a field (Hudson clay loam: fine, illitic, mesic Glossaquic Hapludalf) growing corn (Zea mays L.) located in upstate New York. The field was moldboard plowed in the fall and then fertilized with liquid dairy manure. Intact soil cores were collected from the site both before and after field treatments for subsequent laboratory incubation that included freeze–thaw cycles. The results demonstrated that tillage reduced N2O emissions in nonmanured soils by 20–30% in the 35–50 days following the tillage event, attributed to improved aeration resulting from reduced bulk densities and pore space saturation. The maximal emission of 200 mg N m2 h1 was found at soil temperatures greater than 5 8C and at WFPS between 40 and 70%. Subsequent application of liquid manure caused an increase in the total intensity of N2O emission. The emission of N2O from manure-amended soils was not limited to thawing events: emissions began at soil temperatures below 0 8C and continued even after complete soil freezing. The tillage history prior to manure application was found to have a significant influence on N2O emission during freezing/thawing cycles following manure application. The subsequent total winter N2O emissions were greater from the field areas that were tilled earlier in the fall, particularly in the first few freeze–thaw cycles, during which the maximum N2O fluxes occurred.Increasing soil saturation in a wet area formed during a spring thaw caused increasing N2O emissions up to a maximum of 200 mg N m2 h1 at 60–70% saturation. However, emissions dropped dramatically with further increases in soil moisture, decreasing to 50 mg N m2 h1 in the most saturated areas (90% saturated). Overall, maximal emissions were found at temperatures greater than 5 8C and at water filled porosities between 40 and 70%. ß 2009 Elsevier B.V. All rights reserved. Keywords: Nitrous oxide emission Denitrification Tillage effects Soil freeze–thaw Soil moisture content 1. Introduction A primary source of gaseous N losses to the atmosphere comes from the spreading of animal waste on agricultural fields, amounting to 35% of the global annual emission (Kroeze et al., 1999). Key agricultural management practices regulating N2O formation and release from agricultural fields include use of manure as a fertilizer, crop cultivation and land treatments. For a given amount of N applied to soils, manure application typically * Corresponding author. Tel.: +1 607 255 2463. E-mail address: bkr2@cornell.edu (B.K. Richards). 1 Present address: Department of Earth and Ocean Sciences, University of British Columbia, BC, Canada. results in greater N2O emissions than does synthetic N fertilization (Clayton et al., 1997). Nitrous oxide is produced in soil by microbial nitrification and denitrification processes. The most important factors controlling these processes are soil mineral N (NH4+ and NO3) concentrations, oxygen partial pressure and, in the case of denitrification, available carbon to fuel the heterotrophic processes (e.g. Clough et al., 2003). Soil water content influences diffusion conditions in the soil and thus impacts the supply of oxygen (Robertson and Tiedje, 1987) which in turn controls the amount of N2O emitted. When the water filled pore space (WFPS) exceeds 60%, denitrification becomes the dominant process producing N2O (e.g. Lemke et al., 1999), however production of N2O declines when the WFPS exceeds 80% because N2O is reduced to N2 (Veldkamp et al., 1998). In contrast, any N2O production at WFPS below 60% is 0167-8809/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.agee.2009.03.001 Please cite this article in press as: Singurindy, O., et al., Nitrous oxide emission at low temperatures from manure-amended soils under corn (Zea mays L.). Agric. Ecosyst. Environ. (2009), doi:10.1016/j.agee.2009.03.001 G Model AGEE-3390; No of Pages 8 2 O. Singurindy et al. / Agriculture, Ecosystems and Environment xxx (2009) xxx–xxx typically attributed to nitrification, because increased diffusion of O2 limits denitrification (Robertson and Tiedje, 1987). Significant N2O emissions occur during winter and spring soil thaw events, characterized by N2O peaks which usually last for several days (Christensen and Tiedje, 1990). Despite its limited duration, the phenomenon represents a significant fraction of the total annual emissions from agricultural lands (e.g. Wagner-Riddle and Thurtell, 1998) and may exceed 50% of the annual emissions (Flessa et al., 1995). There are two possible explanations for N2O emission during freezing–thawing cycles: (1) gradual accumulation of N2O produced in the unfrozen subsoil that is unable to diffuse through the frozen soil surface until a thaw takes place (e.g. Goodroad and Keeney, 1984) and (2) favorable denitrification conditions at the time of soil thawing result in a surge of N2O production due to greater carbon and nitrogen availability for microbial activity and high degree of soil saturation (Edwards et al., 1986; Nyborg et al., 1997). A number of studies have measured N2O emissions from soils exposed to repeated freeze–thaw cycles. The results are interesting but also indicate large variability (and hence uncertainty) arising from soil heterogeneity and the complex interactions between chemical, physical and biological factors. Nitrous oxide emissions have been reported to decrease with repeated freeze–thaw cycles (Schimel and Clein, 1996; Prieme and Christensen, 2001). The decrease in gas production suggests either depletion in microbial nutrient availability or damage to soil microbes. Some studies reported significant N2O losses from cultivated soils following freeze–thaw cycles in spring (Nyborg et al., 1997; Wagner-Riddle and Thurtell, 1998). Thawing causes the disruption of soil structure (primarily macroaggregates) and enhances microbial activity due to release of organic C from plant and microbial detritus. The combination of these two factors can change the denitrification potential of soil and therefore influence N2O production during a freeze–thaw event (van Bochove et al., 2000). Consequently, tillage practices have significant effects on the water-stable aggregate distribution. Reduced tillage also reduces total soil porosity and increases soil water content, a factor known to restrict oxygen diffusion through soil (Rice and Smith, 1982). More research is needed to obtain a quantitative understanding of how farm management practices can affect and ideally reduce N2O emissions. Of special interest is understanding some of the apparent complexities involving the interactions between tillage practices and soil conditions and how these affect N2O emissions soon after manure application when potential losses are the greatest. A good example of the complexities involved in designing manure management strategies is the practice of manure injection into soil (for liquid slurries) or rapid plow down of surface spread manures to reduce odor emissions (Webb et al., 2004). Although this is known to reduce ammonia emissions, there are concerns that these direct applications into soil may significantly increase N2O emissions by increasing the pool of mineral N in soil (Bouwman, 1996). We focused our study on understanding the interactions between tillage effects, moisture content, manure application and N2O emissions from corn field soils at low temperatures. 2. Materials and methods Field monitoring using chambers was carried out in order to examine the evolution of nitrous oxide flux as affected by mild winter temperature fluctuations (freeze–thaw cycles), soil tillage, and winter liquid manure application. To further investigate these phenomena under more controlled conditions, soil columns were extracted from the field and monitored during a series of laboratory experiments. 2.1. Field monitoring experiments The experimental site was planted with corn on a clay loam soil (Hudson series—fine, illitic, mesic Glossaquic Hapludalf) located on a large dairy farm in central New York (Fig. 1). The mean annual precipitation is 850 mm and the mean annual temperature is 10 8C. The site was historically fertilized with dairy manure once a year, either spring or fall. On October 10th, 2004, polyvinylchloride chambers (each covering a rectangular area of 0.89 m2) were installed in the field after corn harvesting and about 6 months after the last fertilization. On November 4, 2004 one-half of the field (designated here as area #1) was moldboard plowed. The remaining half (designated as area #2) was plowed on December 16, 2004. On December 17, 2004 liquid dairy manure (5.4% dry matter, total C 4.6–11.3% and total N contents ranging between 0.3 and 0.5%, wet weight basis) was injected into the field with a tractor and a draghose at a liquid loading rate of 75,000 L ha1. The experiment ended on 15 April 2005. The number of chambers installed in the field varied, with 31 chambers permanently installed and 18 additional chambers installed for specific campaigns. The distance between chambers was 25 m, with a total grid area of 30625 m2 (3.06 ha). The additional chambers were installed in the field during five distinct 3-day campaigns, four during winter thawing events and the fifth when spring thawing started at the end of the experiment. Each chamber consisted of a frame permanently located in the ground which could be sealed for 1–3 h with removable plastic lid and additional plastic seal to prevent air exfiltration from the chambers. Each chamber lid was equipped with rubber septa to allow gas sampling with a syringe. Triplicate gas samples were collected in evacuated crimp-sealed minivials at the start and termination of each testing interval. Twenty vertical metal tubes were installed at a depth of 5 cm at 10 locations in the field in order to collect gas samples during winter when the soil surface was frozen. During the experiment, gas samples were collected at intervals ranging from 1 to 4 days. Soil temperature measurements were made using digital thermometers. Soil samples were collected at 10 cm depth near each chamber whenever air samples were taken, and were transported to the laboratory for NO3, NH4+, pH, and gravimetric moisture content analysis. 2.2. Laboratory experiments For laboratory experiments, 16 intact soil cores (10 cm dia., 20 cm deep) were collected from field areas #1 and #2 in December 2004. Ten additional cores (5 each from manured and non-manured areas) were collected on December 22, 2004, which was 5 days after manure application. Cores were obtained by hammering aluminum tubes (10 cm dia., 40 cm long) into the soil and then carefully excavating them. No compression of the soil was observed. The cores were transported to the laboratory and stored outdoors (temperatures thus close to field conditions) for 3 days prior to experimentation. Each soil column microcosm system was built from the aluminum column in which the soil core was extracted, which had a head space of 20 cm. The top of each column was sealed and a rubber septum was installed for sampling. Air samples (6 mL) were taken with syringes every 6 h. After each sampling, the columns were briefly opened to atmosphere and then resealed. Soil column temperatures were monitored by two sensors installed in the middle of six of the columns. During the experiment, the soil cores were subjected to four freeze/thaw cycles by shifting them between a freezer (set at 5.5 8C) and an incubator (25 8C) equipped with a circulation Please cite this article in press as: Singurindy, O., et al., Nitrous oxide emission at low temperatures from manure-amended soils under corn (Zea mays L.). Agric. Ecosyst. Environ. (2009), doi:10.1016/j.agee.2009.03.001 G Model AGEE-3390; No of Pages 8 O. Singurindy et al. / Agriculture, Ecosystems and Environment xxx (2009) xxx–xxx 3 Fig. 1. The topographic map of the field site area (428350 N, 768310 W) where field monitoring experiments were carried out and gas/soil samples were collected. Field areas #1 and #2 correspond to areas that were tilled on November 4th and December 16th, respectively. fan. The duration of the experiment was 55 days. The experiment was started at 22 8C. The columns were then placed in the freezer. Soil temperatures were monitored every 2 h; circa 15 h were required for the columns to freeze and equilibrate. At the end of the experiment soil samples were extracted from the columns, sectioned at 5 cm depth intervals, and analyzed for gravimetric moisture content, NO3, NH4+, and total carbon content. 2.3. Soil and gas analysis Air sample N2O concentrations were measured with a Varian 3700 gas chromatograph (GC) with a Ni63 electron capture detector operated at 350 8C and with Ar:CH4 (95:5) carrier gas (30 mL min1). Solution analysis for NO3 and NH4+ were carried out according to APHA (1985). Soil pH, gravimetric moisture content, and soil organic matter content were determined using standard methods described in ASA (1965). Total carbon content was determined via persulfate oxidation with an OIAnalytical Model 1010 total organic carbon analyzer (O-I-Analytical). Soil bulk density was determined using undisturbed soil cores (Birkeland, 1984). The water filled pore space, defined as the fraction of total pore space filled with water (expressed as percent) was calculated as ½bulk density  gravimetric water content=water density porosity (1)   bulk density (2) porosity ¼ 1  particle density WFPS ¼ The standard particle density used (2.65 g cm3) was modified by a reduction of 0.02 g cm3 per percent of soil organic matter. 2.4. Statistical analysis The error in measured concentration values within each experiment was in all cases less than 0.5%. All concentration measurements in both laboratory and field experiments were repeated in triplicate therefore all reported measurements represent averages of three samples each. Standard error bars (Fig. 2) were included on the graphs to show the variation of the data. A probability level of 5% (P = 0.05) was used to test the statistical significance of all treatments. 3. Results Field data collected before tillage and manure application indicate that initial soil properties were similar in the areas with subsequently differing tillage treatments with respect to soil texture, pH, bulk density, soil C, and total N. There were no significant differences found in average N2O flux or soil moisture content at the beginning of experiment (Table 1). 3.1. Field measurements Fig. 2 presents the N2O emission and gravimetric soil moisture content measured from areas #1 and #2 in the 35 days prior to manure application. Before tillage, the average N2O flux from the field was 147 mg N m2 h1 (Table 1), and tillage resulted in an immediate 40% reduction in area #1 (Fig. 2b). N2O emission remained significantly greater in area #2 (Table 2) for the period from 4 November to 2 December 2004, but during the final 2 days a reduction to 77.5 mg N m2 h1 was observed. The opposite was observed in area #1 soil, where the flux gradually increased and reached a maximum of 166 mg N m2 h1 on 4 December. N2O emission in both areas #1 and #2 generally increased with Please cite this article in press as: Singurindy, O., et al., Nitrous oxide emission at low temperatures from manure-amended soils under corn (Zea mays L.). Agric. Ecosyst. Environ. (2009), doi:10.1016/j.agee.2009.03.001 G Model AGEE-3390; No of Pages 8 4 O. Singurindy et al. / Agriculture, Ecosystems and Environment xxx (2009) xxx–xxx Fig. 3. Field monitoring results from field areas #1 and #2 following manure application: (a) Soil temperature—numbers above the temperature peaks represent the thaw cycle number; (b) N2O emission—numbers above N2O peaks represent the cumulative N2O emission (mg N m2 period1) during each thawing event, the top number corresponding to area #1, the bottom to area #2. Fig. 2. Field monitoring results from field areas #1 (tilled) and #2 (not yet tilled) before manure application: (a) N2O emissions and (b) soil moisture content. The black arrows indicate the first date and amount (total of 5 days) of precipitation that caused increased in N2O emissions. increasing soil moisture content (Fig. 2b) following three indicated precipitation events. In area #2, the maximal emission of 186 mg N m2 h1 was observed on November 9th, 19th, and December 2nd (Fig. 2a). For area #1, the first two peaks (149 and 133 mg N m2 h1) were concurrent with those in area #2, whereas the third peak was delayed by several days relative to area #2. Fig. 3 shows the course of the soil temperatures and N2O emissions following tillage of area 2 and manure application of the whole field on December 17. Soil temperature fluctuations indicate nine cycles of freezing and thawing (Fig. 3a), with seven cycles (lasting from 4 to 12 days) during which N2O emissions occurred. Low levels of emission were found during the thaw cycles between days 41 and 46. The thaw cycle starting at day 51 was apparently too brief to allow emission to occur. The numbers above each peak in Fig. 3b represent the cumulative N2O emissions (mg N m2 period1) during that thawing event (the top number corresponds to area #1, the bottom to area #2). Despite the fact that both field areas had been tilled by the time of manure application, the results shown in Fig. 3b and Table 2 demonstrate continuing differences in N2O emissions between two areas tilled at different times. The differences were greatest during the first three thaw cycles (Table 2), where emissions from earlier-tilled area #1 were much greater than from area #2: e.g. 700 mg N m2 vs. 257 mg N m2 during the first cycle. The greatest emission rate of 289 mg N m2 h1 was observed from area #1 soil during the first thaw-related peak. However, this difference in tillage history effects was no longer observed after the long freezing period (with two short thaws) preceding thawing cycle 4: subsequent N2O peaks were more consistent and were both greater and longer. In some cases a low level of N2O emission continued for a time after complete soil freezing, continuing between first and second thawing cycles and persisting for 3–4 days after the second and third cycles (Fig. 3a). Emission of N2O also occurred just prior to the onset of thawing cycle 6. The cumulative emissions of N2O for all freeze/thaw cycles were 2860 and 2120 mg N m2 for areas #1 and #2 soils, respectively. The pH values were 7.3–7.7 and total N varied from 0.5 to 0.7%. Total mineral nitrogen (NO3 + NH4) was between 19 and 28 mg N kg1 of dry soil. The results of a 3-day campaign during spring thawing are shown in terms of soil saturation and N2O emission (Fig. 4). The results show that the sample grid contained a large wet area where the pore space reached a maximum saturation of 90%, compared to 30–50% saturation in the rest of the field (Fig. 4a). Corresponding to this is the pattern of N2O emissions (Fig. 4b), which increased with increasing soil saturation up to about 60–70% to a maximum of 200 mg N m2 h1. However, at greater 70–90% saturation in the middle of the wet area, N2O emissions dropped dramatically to 50 mg N m2 h1. 3.2. Laboratory experiments The freeze/thaw cycles resulting from the freezer/incubator cycling are evident in the soil core temperatures in Fig. 5a. The Table 1 Soil characteristics from two parts of the field (#1 and #2) at the beginning of the field monitoring experiments, before tillage and manure application. Area Sand (%) Silt (%) Clay (%) pH (in water) Total C (%) Total N (%) Bulk density (g cm3) Soil moisture (%) Nitrous oxide flux (mg N m2 h1) #1 #2 15 16 53 54 32 30 7.2 7.1 1.8 1.7 0.2 0.2 1.25 1. 21 22 148 146 Please cite this article in press as: Singurindy, O., et al., Nitrous oxide emission at low temperatures from manure-amended soils under corn (Zea mays L.). Agric. Ecosyst. Environ. (2009), doi:10.1016/j.agee.2009.03.001 G Model AGEE-3390; No of Pages 8 5 O. Singurindy et al. / Agriculture, Ecosystems and Environment xxx (2009) xxx–xxx Table 2 Summary of cumulative N2O emissions (mean  standard deviation) from field chamber and soil column experiments. Statistical comparisons made between areas 1 and 2: values followed by asterisks are statistically different (P = 0.05). Experiment and time frame Field chambers After tilling, before manure application After tilling and manure application (first three freeze–thaw cycles) After tilling and manure application (entire period) Soil columns Collected after tilling, before manure application (four freeze–thaw cycles) Collected after tilling, after manure application (first freeze–thaw cycle) Collected after tilling, after manure application (four freeze–thaw cycles) laboratory cores extracted from field areas #1 and #2 had differing bulk densities and therefore differing initial WFPS levels of 35% (area #1) and 70% (area #2). During these four freeze/thaw cycles, the N2O emissions from the soil cores collected before manure application were greater for area #2 in terms of peak rates (Fig. 5b) as well as cumulative emissions (1631 mg N m2 vs. 1992 mg N m2 for areas #1 and #2, respectively; Table 2). The greater N2O emission corresponded to the higher soil water content. In area #2 soils the emission peaked immediately after thawing (approximately 300 mg N m2 h1 maximum during the first cycle) and declined markedly thereafter (Fig. 5b). The general trend in emissions over time was similar for both treatments, with the greatest emissions occurring in cycle 1, cycles 2 and 3 having similar emissions, and lower emissions during the last thawing cycle. Period (days) Area #1 (tilled November 4) (mg N m2) Area #2 (tilled December 17) (mg N m2) 35 41 121 2425  120* 1043  34* 2860  115* 3302  159* 588  19* 2120  88* 1631  24* 1425  27* 6128  96 1992  37* 1189  23* 6332  89 50 10 50 Unsurprisingly, the soil cores collected after manure application had greater N2O emission rates (Fig. 5c) than those collected before application, with a maximal emission rate of approximately 396 mg N m2 h1 observed from area #1 soil cores during the first thaw-related peak. The area #1 cores had a significantly greater cumulative emission during the first thaw cycle (Table 2). However, the cumulative emissions of N2O from the four freeze/ thaw cycles did not differ significantly during the four cycles 6128 and 6332 mg N m2 for areas #1 and #2 soils, respectively. In accordance with the field measurements, the most intensive release of N2O was found immediately after thawing started (with temperatures climbing over 0 8C at the point of measurement) in all four freeze/thaw cycles. However, emissions were not limited to the thaw periods: low levels of emission were recorded 2 days before first and third thaws, continued between the first and Fig. 4. Field monitoring results from 3-day chamber array campaign during spring thaw event. (a) Degree of soil saturation and (b) emission of N2O. Chambers were located at each grid point (25 m grid spacing), with a total grid area of 30,625 m2 (3.06 ha). Please cite this article in press as: Singurindy, O., et al., Nitrous oxide emission at low temperatures from manure-amended soils under corn (Zea mays L.). Agric. Ecosyst. Environ. (2009), doi:10.1016/j.agee.2009.03.001 G Model AGEE-3390; No of Pages 8 6 O. Singurindy et al. / Agriculture, Ecosystems and Environment xxx (2009) xxx–xxx Fig. 5. Result of laboratory experiment with undisturbed soil cores collected from field areas #1 and #2: (a) soil temperature, (b) N2O emission from columns collected before manure application, and (c) N2O emission from columns collected after manure application. second thaw events, and also continued 2 days after soil freezing in the second cycle. Finally, unlike the non-manured soil cores, there was no reduction of N2O emission during the fourth thaw, presumably due to the high availability of C (and mineral N) in the manured soils. 4. Discussion The soil water filled pore space strongly affected N2O production, likely via effects on oxygen availability. Fig. 6 integrates the combined effects of WFPS and soil temperature on measured N2O flux. Generally, the optimal conditions for N2O production were at temperatures greater than 5 8C and at soil WFPS levels between 40 and 70%. Temperatures lower than 5 8C reduced soil microbial activity and consequently reduced emissions. The maximal flux of N2O was found at moisture levels of approximately 55% WFPS, which corresponds well to the results obtained by Davidson et al. (2000) for different soil types. Oxygen deficiency and an associated increase in N2O production could even occur at lower saturation levels in soils with high microbial activity. With good soil aeration (WFPS of 40–60%), both organic and clay soils had increased N2O production at a lower temperature than did silt or loam soils, which could be associated with greater organic matter degradation rates in the organic and clay soils (Koponen et al., 2004). High water content favors denitrification associated with the limitation of oxygen diffusion. We could not perform an isotopic analysis needed to show the possible changes of N2O/N2 ratio in our specific experimental conditions. Maag and Vinther (1996) demonstrated that N2O/N2 ratio in denitrification increases with a decrease of temperature, and thus enhance N2O production. Though the gravimetric water contents were sometimes similar (especially after intense precipitation; Fig. 2b) in the tilled and non-tilled parts of the field, the WFPS was different because of the differing bulk densities (1.06 g cm3 tilled vs. 1.25 g cm3 nontilled). Subsequent tillage of area #2 followed by manure application homogenized the two areas of the field with respect to soil bulk density and WFPS. Nevertheless, significant differences in N2O emissions between areas #1 and #2 were still identified during the first three freezing–thawing cycles (Fig. 3b), the total N2O emission being greater in the better pre-aerated soils of area #1. A possible explanation for the high peaks of N2O during thawing was that greater microbial activity resulted in greater mineralization/nitrification rates as well as greater oxygen consumption. Please cite this article in press as: Singurindy, O., et al., Nitrous oxide emission at low temperatures from manure-amended soils under corn (Zea mays L.). Agric. Ecosyst. Environ. (2009), doi:10.1016/j.agee.2009.03.001 G Model AGEE-3390; No of Pages 8 O. Singurindy et al. / Agriculture, Ecosystems and Environment xxx (2009) xxx–xxx Fig. 6. Generalized relationship of N2O emission with temperature and WFPS: field data before manure application. Emission of N2O was always observed during thawing in both laboratory and field experiments. Surprisingly little spatial variability was observed in the field measurements except during the intensive spring thawing such as the 3-day campaign in the early spring, when more ice and snow started to melt at the soil surface. The sample grid encompassed a wet area that was substantially wetter than the rest of the field (Fig. 4a). Corresponding to this is the pattern of N2O emissions, which increased with increasing soil saturation up to about 60–70% WFPS. However, at higher WFPS levels (with the wet spot nearing saturation), N2O emissions dropped dramatically: under saturation, anaerobic conditions would dominate and, consistent with expected trends, favor N2 production during extended denitrifying conditions. A similar phenomenon was observed by Nyborg et al. (1997). In addition, drainage problems are well-known in this specific soil type, especially in spring. Our field measurements demonstrate that N2O emission continued even after complete soil freezing (Fig. 3). Teepe et al. (2001) observed constant N2O emission for several days in freezing periods as evidence of microbial activity in the frozen soil. Goodroad and Keeney (1984) reported that N2O emission was due to the temperature fluctuations that cause microbial activity at the soil surface. Kaiser et al. (1998) suggested that N2O emissions during the time of deepest soil freezing occurred as a result of N2O production in deeper soil horizons, with the gas escaping through frost-induced cracks. In our study, during the period from 41 to 65 days after manure application, considerable snow precipitation caused the formation of the deep snow and ice layer that prevented the escape of nitrous oxide. During the subsequent thaw, the trapped N2O was released within few days, resulting in a high emission peak. The fact that these peaks were wide may be due to the melting snow causing elevated soil moisture conditions that favored denitrifier activity. Edwards and Cresser (1992) showed that 8–20% of the soil water remained unfrozen for several days although the soil temperature was 5 8C. There are significant amounts of unfrozen soil water down to 20 8C (Rivkina et al., 2000), mostly because of 7 salting-out effects. The unfrozen water covers the soil matrix with the thin water film which in turn is covered with ice, creating conditions are favorable for denitrification, because microorganisms are isolated from oxygen by the ice barrier (Teepe et al., 2000). In addition, field and laboratory data demonstrated that, in some cases, thawing begins at the temperatures below 0 8C followed by production of N2O. As a result a small peak of N2O emission was observed before thawing cycle number 6 (Fig. 3b). This phenomenon occurred as a result of the presence of ions in the soil solution. Gas samples taken at 5 cm depth in the field and from soil cores during freezing (soil temperatures 5 8C) demonstrated accumulation of N2O. A portion of the large flux of N2O that was released just after thawing (temperatures close to 0 8C) might have originated from the liberation of N2O stored in the frozen soils, an effect much more pronounced with the manure-applied soils. These results are in agreement with the laboratory observations of Koponen et al. (2004) who suggested that there was N2O production in soils at least down to 6 8C. In the laboratory experiments with cores collected before manure application (Fig. 5b), the maximum thaw-related N2O emission occurred during the first thaw cycle. The first cycle peak was also higher in field #1 soil (Fig. 3b). Koponen and Martikainen (2004) found a similar difference for N2O emission between the two cycles. Schimel and Clein (1996) reported a similar phenomenon for CO2 emissions from tundra and taiga soils. They concluded that freeze–thaw cycles caused a flush of microbial C and N during the first cycle, but after repeated cycles the ability of microbial communities to decompose soil organic matter falls. An additional explanation given by Koponen and Martikainen (2004) is that amount of organic and inorganic substances declines from cycle to cycle, leading to lower thaw-related N2O emissions in subsequent cycles. Moreover, it can be due to the high availability of C as a result of film microorganisms killed during freeze/thaw cycles (Christensen and Tiedje, 1990) coupled with high concentrations of both inorganic and organic solutes (Edwards and Cresser, 1992) may cause favorable conditions for N2O formation in water films. In our previous study (Singurindy et al., 2006) we observed that this specific soil has a high capacity for NH4+ fixation by clay minerals. Moreover, nitrate and ammonium immobilization by soil microbial biomass shortly after field N application (Müller et al., 2002) would increase the potential for N2O emission during freezing–thawing periods. Manured soils (high content of active biomass and organic material) and clay dominated aggregates therefore have an elevated potential for N2O emission during freeze/thaw events. This conclusion is in line with observations that the greatest thaw-related emissions were measured from organic soils and soils with a high content of clay-associated aggregates (Christensen and Tiedje, 1990; van Bochove et al., 2000; Singurindy et al., 2008). 5. Conclusions In this study, we have characterized the N2O emission effects of late fall tillage and liquid dairy manure application to soil. Tillage reduced N2O emission in non-manured soils for the 35–50 days of our observation periods. The differences in the emissions were attributed to the reductions in bulk density and water filled pore space (and thus improved aeration) in the tilled area. Application of liquid manure to the soil increased the total intensity of nitrous oxide emission, and the majority of winter emissions occurred during periodic soil thaw events. In both field and laboratory studies, the timing of tillage prior to manure application was found to have a significant influence on N2O Please cite this article in press as: Singurindy, O., et al., Nitrous oxide emission at low temperatures from manure-amended soils under corn (Zea mays L.). Agric. Ecosyst. Environ. (2009), doi:10.1016/j.agee.2009.03.001 G Model AGEE-3390; No of Pages 8 8 O. Singurindy et al. / Agriculture, Ecosystems and Environment xxx (2009) xxx–xxx emissions, which were greatest from the earlier-tilled field area during the initial freezing/thawing cycles following manure application. In addition, in the field study, the cumulative winter emissions following manure application were greater from the earlier-tilled area. The emission of nitrous oxide from manureamended soils started at temperatures below 0 8C and continued after complete soil freezing. A large wet area formed during a substantial spring thaw had a maximum pore space saturation of 90%, compared to 30–50% saturation in the rest of the field. Corresponding N2O emissions increased with soil moisture to a maximum of 200 mg N m2 h1 at 60–70% saturation. However, emissions dropped dramatically with further increases in soil moisture, decreasing to 50 mg N m2 h1 in the most saturated areas. Overall, maximal emissions were found at temperatures greater than 5 8C and at water filled porosities between 40 and 70%. Acknowledgments This research was supported by USDA-NRI project no. 123527 and Vaadia-BARD Postdoctoral Award No. F1-357-04 from BARD, The United States - Israel Binational Agricultural Research and Development Fund. We thank Hardie Dairy Farms for their cooperation and technical assistance. References APHA, 1985. Standard Methods for the Examination of Water and Wastewater, 16th ed. Port City Press, Baltimore, MD. ASA, 1965. Methods of Soil Analysis: Part 2. Chemical and Microbiological Properties. American Society of Agronomy, Madison, WI. Birkeland, P.W., 1984. Soils and Geomorphology. Oxford University Press, New York. Bouwman, A.F., 1996. Direct emissions of nitrous oxide from agricultural soils. Nutr. Cycl. Agroecosyst. 46, 53–70. Davidson, E.A., Keller, M., Erickson, H.E., Verchot, L.V., Veldkamp, E., 2000. Testing a conceptual model of soil emissions of nitrous and nitric oxides. BioScience 50 (8), 667–680. Flessa, H., Dorsch, P., Beese, F., 1995. Seasonal variation of N2O and CH2 fluxes in differently managed arable soils in southern Germany. J. Geophys. Res. -Atmos. 100, 23115–23124. Goodroad, L.L., Keeney, D.R., 1984. Nitrous oxide emissions from soils during thawing. Can. J. Soil Sci. 64, 187–194. Clayton, H., McTaggart, I.P., Parker, J., Swan, L., Smith, K.A., 1997. Nitrous oxide emissions fro fertilized grassland: a two-year study of the effects of N fertilizer from the environmental conditions. Biol. Fertil. Soils 25, 252–260. Clough, T.J., Sherlock, R.R., Kelliher, F.M., 2003. Can liming mitigate N2O fluxes from a urine-amended soil? Aust. J. Soil Res. 41, 439–457. Christensen, S., Tiedje, J.M., 1990. Brief and vigorous N2O production by soil at spring thaw. J. Soil Sci. 41, 1–4. Edwards, A.C., Cresser, M.S., 1992. Freezing and its effect on chemical and biological properties of soil. Adv. Soil Sci. 18, 59–79. Edwards, A.C., Creasey, J., Cresser, M.S., 1986. Soil freezing effects on upland. Water Res. 20 (7), 831–834. Kaiser, E.A., Kohres, K., Kucke, M., Schnug, E., Heinemeyer, O., Munch, 1998. Nitrous oxide release from arable soil: importance of N fertilization, crop and temporal variation. Soil Biol. Biochem. 30, 1553–1563. Koponen, H.T., Martikainen, P.J., 2004. Soil water content and freezing temperature affect freeze–thaw related N2O production in organic soil. Nutr. Cycl. Agroecosyst. 69, 213–219. Koponen, T.H., Flojt, L., Martikainen, P.J., 2004. Nitrous oxide emissions from agricultural soils at low temperature: a laboratory microcosm study. Soil Biol. Biochem. 36, 757–766. Kroeze, C., Mosier, A., Bouwman, L., 1999. Closing the global N2O budget: a retrospective analysis. Global Biogeochem. Cycles 13, 1–8. Lemke, R.L., Izaurralde, R.C., Nyborg, M., Solberg, E.D., 1999. Tillage and N source influence soil-emitted nitrous oxide in the Alberta Parkland region. Can. J. Soil Sci. 79, 15–24. Maag, M., Vinther, F.P., 1996. Nitrous oxide emissions by nitrification and denitrification in different soil types and at different soil moisture contents and temperatures. Appl. Soil Ecol. 4, 5–14. Müller, C., Martin, M., Stivens, R.J., Laughlin, R.J., Kammann, C., Ottow, J.C.G., Jager, H.-J., 2002. Processes leading to N2O emissions in grassland soil during freezing and thawing. Soil Biol. Biochem. 34, 1325–1331. Nyborg, M., Laidlaw, J.W., Solberg, E.D., Malhi, S.S., 1997. Denitrificationand nitrous oxide emissions from black chernozemic soil during spring thaw in Alberta. Can. J. Soil Sci. 77, 153–160. Prieme, A., Christensen, S., 2001. Natural perturbations, drying–wetting and freezing–thawing cycles, and the emissions of nitrous oxide, carbon dioxide and methane from farmed organic soils. Soil Biol. Biochem. 33, 2083–2091. Rice, C.W., Smith, M.S., 1982. Denitrification in non-till and plowed soils. Soil Sci. Soc. Am. J. 46, 1168–1173. Rivkina, E.M., Friedmann, E.I., McKay, C.P., Gilichinsky, D.A., 2000. Metabolic activity permafrost bacteria below the freezing point. Appl. Environ. Microbiol. 66, 3230–3233. Robertson, G.P., Tiedje, J.M., 1987. Nitrous oxide sources in aerobic soils: nitrification, denitrification and other biological processes. Soil Biol. Biochem. 19, 187–193. Schimel, J.P., Clein, J.S., 1996. Microbial response to freeze–thaw cycles in tundra and taiga cycles soils. Soil Biol. Biochem. 28, 1061–1066. Singurindy, O., Molodovskaya, M., Richards, B.K., Steenhous, T.S., 2006. Nitrous oxide and ammonia emissions from urine applied to soils: texture effect. Vadose Zone J. 5 (4), 1236–1245. Singurindy, O., Molodovskaya, M., Richards, B.K., Steenhous, T.S., 2008. Gaseous nitrogen emission from soil aggregates as affected by clay mineralogy and repeated urine applications. Water Air Soil Pollut. 10.1007/s11270-008-9746-4. Teepe, R., Brumme, R., Breese, F., 2000. Nitrous oxide emissions from frozen soils under agricultural, fallow and forest land. Soil Biol. Biochem. 32, 1807–1810. Teepe, R., Brumme, R., Breese, F., 2001. Nitrous oxide emissions from soil during freezing and thawing periods. Soil Biol. Biochem. 33, 1269–1275. van Bochove, E., Prevost, D., Pelletier, F., 2000. Effects of freeze–thaw and soil structure on nitrous oxide produced in a clay soil. Soil Sci. Soc. Am. J. 64, 1638–1643. Veldkamp, E., Keller, M., NuNez., 1998. Effects of pasture management on N2O and NO emissions from soils in the humid tropics of Costa Rica. Global Biogeochem. Cycl. 12, 71–79. Webb, J., Chadwick, D., Ellis, S., 2004. Emission of ammonia and nitrous oxide following incorporation into the soil of farmyard manures stored at different densities. Nutr. Cycl. Agroecosyst. 70, 67–76. Wagner-Riddle, C., Thurtell, G.W., 1998. Nitrous oxide emissions from agricultural fields during winter and spring thaw as affected by management practices. Nutr. Cycl. Agroecosyst. 52, 151–163. Please cite this article in press as: Singurindy, O., et al., Nitrous oxide emission at low temperatures from manure-amended soils under corn (Zea mays L.). Agric. Ecosyst. Environ. (2009), doi:10.1016/j.agee.2009.03.001