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

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Petersen, S.O. Articles by Eriksen, J. Search for Related Content PubMed PubMed Citation Articles by Petersen, S.O. Articles by Eriksen, J. Agricola Articles by Petersen, S.O. Articles by Eriksen, J. Related Collections Nutrient Management Water Pollution Animal Waste

TECHNICAL REPORT
Waste Management

Denitrification Losses from Outdoor Piglet Production

Spatial and Temporal Variability

^a Dep. of Crop Physiology and Soil Science, Danish Institute of Agricultural Sciences, P.O. Box 50, DK-8830 Tjele, Denmark
^b Dep. of Biometry, Danish Institute of Agricultural Sciences, P.O. Box 50, DK-8830 Tjele, Denmark

Corresponding author (Soren.O.Petersen{at}agrsci.dk)

Received for publication March 13, 2000.

	ABSTRACT

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

 
Animal welfare considerations have stimulated the development of outdoor piglet (Sus scrofa) production systems, but the high levels of nutrients excreted suggest that nutrient losses from this system may be high. This study first described the spatial distribution of denitrification activity in a 5- x 5-m grid within and outside a paddock immediately after the sows (32 sows ha^-1 for 6 mo) were removed in October 1997, and again the following March. Denitrification rates averaged 0.01 kg N ha^-1 d^-1 outside, and 0.5 kg N ha^-1 d^-1 inside the paddock in October, while the corresponding figures in March were 0.01 and 0.1 kg N ha^-1 d^-1. The highest denitrification rates were observed around the feeder, and this was also the case for concentrations of dissolved organic C and inorganic N in the soil. A statistical model that included both soil parameters and distance to feeder and huts gave the best description of the variability, but there was no significant autocorrelation between sampling points. In a second phase, seasonal variation of denitrification activity within a paddock (12 sows ha^-1 yr^-1) was quantified; 10 soil cores were sampled along a transect 11 times between March 1998 and February 1999. There was a significant positive effect of dissolved organic carbon (DOC) on denitrification at <25% gravimetric soil moisture (i.e., to November in this study). Both climate and management (position of huts and feeder) appeared to influence denitrification, which was estimated to be 69 kg N ha^-1 yr^-1, or 11% of the N surplus of this production system.

Abbreviations: DOC, dissolved organic carbon

	INTRODUCTION

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

 
IN recent years the concept of food quality has evolved to include aspects of the production process itself, such as animal welfare. Piglet production in open pastures may better comply with behavioral needs of the animals than most indoor housing systems, but the intake of feeds is high and there is a correspondingly high deposition of nutrients in excreta (Larsen and Kongsted, 2000). This has raised concerns about the nutrient efficiency of such a production system.

The fate of surplus nutrients with outdoor piglet production is largely unknown (Worthington and Danks, 1992). The potential for losses is high since excreta are left at the soil surface and deposited under a wide range of climatic conditions, and the distribution of nutrients across the field is probably less uniform than if excreta were collected in manure storages and spread mechanically. Nitrogen may be leached to below the root zone as NO^-₃ or dissolved organic N in periods with net infiltration, or it may be lost to the atmosphere via NH₃ volatilization or denitrification.

The present report describes measurements to characterize and quantify denitrification activity associated with piglet production. Denitrification (i.e., the production of N₂ and N₂O) is carried out by facultatively anaerobic heterotrophic bacteria (although NH⁺₄ oxidation may also be a source of N₂O). Denitrifying bacteria require metabolizable C and NO^-₃ and the absence of O₂, and faecal materials are known to stimulate the process (Elliott et al., 1990; Petersen et al., 1991, 1996). The spatial variability of denitrification activity is known to be extreme, both on a macroscopic (e.g., Robertson et al., 1988) and on a microscopic scale (Parkin, 1987; Nielsen and Revsbech, 1994), and this has made intensive sampling and statistical evaluation of measurement data indispensible tools in field studies of denitrification. In the present study, the main objectives were to quantify denitrification associated with outdoor piglet production, and to describe the field-scale distribution and temporal dynamics of denitrification activity, as influenced by bulk soil characteristics and management practices.

	MATERIALS AND METHODS

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

 
The study consisted of two parts, (i) a description of spatial variability at two times following a period of piglet production and (ii) a seasonal study with samplings along a transect across a paddock.

Part I: Spatial Variability
This part was carried out in southwestern Denmark (55.5°N, 8.2°E) on a loamy sand (6% clay, 17% silt, and 73% sand) with 2.5% C. The soil was freely drained prior to the introduction of pigs. Annual precipitation is 860 mm, and the mean temperature 7.6°C. Sampling took place in early October 1997 within and outside a 3350-m² paddock shortly after the sows and piglets were removed, and again around 1 Mar. 1998. Three successive groups of farrowing sows (32 sows ha^-1) had been kept in the paddock from one week before until seven weeks after farrowing in the period between March and early October 1997, corresponding to a stock density of 14 sows ha^-1 yr^-1. The field was then left undisturbed until the following spring. Sows had individual huts until after farrowing, when huts were shared. The precise location and orientation of huts, which changed between each group of sows and after farrowing, was recorded (data not shown). A feeder was located at two different positions along one edge of the paddock in April through July and August through October, respectively.

At each sampling, a total of ca. 145 soil cores were sampled to approximately 18 cm depth from the intersections of a 5- x 5-m (in some locations 2.5- x 2.5-m; Fig. 2) grid, including 15 samples from outside the paddock. The two soil cores removed from each sampling point in October 1997 and March 1998, respectively, were taken on opposite sides of the actual coordinates (i.e., 5 to 10 cm apart). Acrylic cylinders (i.d. 44 mm, height 200 mm) were inserted in a stainless steel corer, resting on a flange to enable sampling of noncompacted soil cores. The inserts were perforated (four 2-mm-diam. holes per 20 mm of length) to facilitate gas exchange between soil and air. Soil cores were transported to the laboratory in insulated boxes and stored at 2°C until analysis (within 4 d).

View larger version (60K): [in this window] [in a new window]   Fig. 2. The contour plots show the distribution of several parameters in October 1997, shortly after the sows were removed from the paddock (left-hand column), and around 1 Mar. 1998 (right-hand column). The tilted frame delineates the paddock, the individual sampling positions are shown as black dots, and the two positions of the feeder are indicated by white squares. (A) and (B), denitrification activity (kg N ha-1 d-1); (C) and (D), dissolved organic carbon (DOC, mg C kg-1 dry wt. soil); (E) and (F), NO-3 (mg N kg-1 dry wt. soil).

Part II: Temporal Variability
For practical reasons the seasonal study was carried out at a different location (56.3°N, 9.5°E). The soil type was again a loamy sand, freely drained prior to introduction of pigs, with 6% clay, 10% silt, 78% sand, and 3.6% C. The annual precipitation averages 630 mm, and the annual temperature 7.3°C. Farrowing sows (12 sows ha^-1 yr^-1) were first introduced to the paddock in early April 1998, shortly after the first sampling, and generally stayed there from one week before until three to four weeks after farrowing. A transect was defined from the mid-point of one side of the paddock toward an opposite corner, and 10 points were sampled at 5-m intervals along this transect. Figure 1 shows a frame that was used to identify the actual (pre-selected) sampling position relative to each of the 10 sampling points. Management of huts was as described above for Part I of this study, while the feeder was repositioned a few times over the course of the study. The precise location of huts and feeder was only registered if they were within 1 to 2 m distance from sampling positions.

View larger version (20K): [in this window] [in a new window]   Fig. 1. In the seasonal study (Part II), the sampling positions at 5-m spacing along a transect were located using a measuring tape, and the actual sampling then took place at a 20-cm distance from the sampling position in a pre-selected direction, using a frame as shown. The direction chosen for the individual sampling was randomly selected before the seasonal study was initiated.

Release of Nitrous Oxide
Compaction of the soil surface due to animal traffic could have lead to entrapment of gases. Thus, in Part II, where measurements were performed immediately after sampling, the perforation of the acrylic cylinders introduced the possibility that a pool of trapped N₂O would be released, interfering with the N₂O accumulation derived from denitrification activity. To investigate the temporal dynamics of N₂O release, six soil samples were obtained in December 1998 from three different environments within the paddock: a grass-covered area ca. 2 m from the fence, bare soil ca. 2 m from the feeder, and bare soil ca. 2 m from an area with standing water. The soil cores were autoclaved twice on consecutive days (Wolf et al., 1989). The absence of N₂O production was ascertained, before the soil cores were pre-incubated overnight in the presence of ca. 1.7 µL L^-1 N₂O. After rapid transfer to new gas-tight cylinders, the time course of N₂O release from soil air and soil water was examined using the standard procedure (i.e., with mixing of the headspace prior to sampling). The observed accumulation of N₂O in the headspace was fitted by an exponential equation (Sommer and Jensen, 1994):

[1]

where N is the observed concentration of N₂O (µL L^-1) at time t (min), N_max is the headspace concentration at equilibrium, b is a sample-specific parameter, and 0.3 is the approximate atmospheric background of N₂O.

Analytical Techniques
Nitrous oxide was analyzed on a Varian (Walnut Creek, CA) 3300 gas chromatograph with EC detector and a 1-m column packed with Porapak T (Cobert Resources, St. Louis, MO). The carrier gas was a mix of Ar and CH₄ (95/5%) at a flow rate of 50 mL min^-1, and temperatures of oven and detector were 45 and 400°C, respectively. Gas samples were introduced automatically via a sample loop (Parkin, 1985). Standards in Venojects were included at the beginning and during each run.

After denitrification measurements, soil samples were passed through an 8-mm screen to remove stones and roots. Subsamples were extracted in 1 M KCl (1:2, w/v) for 30 min and centrifuged. The supernatant was filtered (0.7 µm) through pre-washed glass fiber filters. The filtrate was used for determination of NH⁺₄ and NO^-₃, as well as for DOC. For Part I, NH⁺₄ concentrations were determined with a flow injection analyzer (Lachat Instruments, Milwaukee, WI), NO^-₃ on a Technicon AutoAnalyzer (Bran and Luebbe, Norderstedt, Germany), and DOC on a Model 5000A total organic carbon analyzer (Shimadzu, Kyoto, Japan). Due to various technical problems, alternative analytical techniques had to be adopted for Part II of this study. Inorganic N was analyzed using the microplate methods described by Sims et al. (1995), while DOC was analyzed (after 10-fold dilution to avoid interference from chloride) on a Model 700 total organic carbon analyzer (O.I. Corp., College Station, TX). Separate subsamples of sieved soil were used for determination of soil moisture (24 h at 105°C) and, in Part II only, loss-on-ignition (an additional 3 h at 450°C) (Combs and Nathan, 1998).

Statistical Analyses
Denitrification rates were determined by linear regression of N₂O accumulation over time; nonsignificant rates are presented as zero. Total denitrification losses for the paddock (Part I) or experimental period (Part II) were estimated using arithmetic means (Parkin et al., 1988; Velthof et al., 1996).

The data set was also explored for relationships between denitrification activity and potential influencing factors. For Part I, models including only soil characteristics were examined, as well as models that included also distance to huts, feeder, and fence from each sampling point. The effect of hut location was expressed in an aggregated variable reflecting the number and distribution of animals between huts on any specific day, and assuming that deposition of excreta would occur in random direction and decrease proportionally with the squared distance to the hut. Distance to the feeder was calculated separately for the periods before and after relocation of the feeder in late July.

Preliminary inspection of the data had indicated that log-transformed denitrification rates were better explained by the independent variables (in some cases also log transformed) than untransformed denitrification rates, and so the following model was defined:

[2]

where Y_i = denitrification rate at point i; µ_i = exp( ß_j X_ij) = model output for denitrification at point i; ß_j = coefficient for the variable j; X_ij = value of variable j at point i; and E_i = random effect at point i, {E_i} N(0,V), where V is a covariance matrix that takes into account that logarithmic values of Y_i are assumed to have a constant variance and possibly a correlation between observations. The covariance matrix was estimated from the predicted mean and random effects. Model parameters were determined by an iterative procedure with linearization of nonlinear effects (Prentice and Zhao, 1991). Contour plots were created from the actual measurement data using Surfer Version 6 (Golden Software, 1997).

In the seasonal study (Part II) only soil characteristics (i.e., NH⁺₄, NO^-₃, DOC, and soil temperature) were available as independent variables for the statistical analysis. The best description of the variability was obtained by the following model:

[3]

where Y_ij = denitrification rate (g N ha^-1 d^-1) at point i to time j; D_ij = log(DOC) (µg C g^-1 dry wt. soil) at point i to time j; W_ij = soil moisture at point i to time j; E_i = random effect of point i, E_i N; and F_ij = random effect of point i to time j, F_ij N(0,²). The terms a, b, c, and d are fixed coefficients for the intercept and the respective variables to be estimated, and ²_E and ² are variance components to also be estimated. The term dD_ijW_ij represents a significant interaction between DOC and soil moisture, which was revealed by the analysis.

	RESULTS AND DISCUSSION

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

 
In 1997, the stock of outdoor sows represented 3% of the total sow population in Denmark, a proportion that is expected to increase further, since this production system is in accordance with consumers' demands for increased animal welfare in pig production. Controlling environmental effects of such an intensive production requires an understanding of the links between management practices, soil conditions, and nutrient losses. The present study was carried out to provide quantitative information about denitrification losses.

Part I: Spatial Distribution
In October 1997 (Fig. 2A) denitrification rates ranged from 0 to 4 kg N ha^-1 d^-1 (although an extreme of 16.5 kg N ha^-1 d^-1 was recorded in one single grid point). The arithmetic mean inside the paddock was 0.5 kg N ha^-1 d^-1 (coefficient of variation [CV], 314%) and outside the paddock was 0.01 kg N ha^-1 d^-1 (CV, 176%). At the sampling in March 1998 (Fig. 2B) the level of denitrification activity inside the paddock was much lower, 0.1 kg N ha^-1 d^-1 (CV, 334%), but outside the paddock the level was similar to the first sampling, 0.01 kg N ha^-1 d^-1 (CV, 207%). Hence, the presence of animals stimulated subsequent denitrification and its spatial variability, even after ca. 5 mo without pigs. At both sampling times the highest rates were recorded near the second position of the feeder (the left-hand position in Fig. 2). The level of variability is in accordance with previous studies on the distribution of denitrification activity (Folorunso and Rolston, 1984; Robertson et al., 1988), and a similar increase in variability due to animals was observed for N₂O emissions by Clayton et al. (1994). Ambient soil temperatures around the time of sampling were slightly above (October) or below (March) the temperature of 10°C at which denitrification rates were determined, which implies that levels of in situ denitrification activity at the two samplings were even more different than indicated by these measurements.

The distribution of DOC in October is shown in Fig. 2C; except for an outlier (top left) the highest concentrations of this parameter were also observed around the feeding area. The level of DOC dropped between October and March, although some spatial heterogeneity was still observed (Fig. 2D). Similarly, NO^-₃ concentrations were elevated around the feeding area in October (Fig. 2E), but this pattern had completely disappeared by March (Fig. 2F), suggesting large losses of NO^-₃.

Different models were tested for their ability to explain the distribution of denitrification acitivity. They were not restricted by preconceptions about the relationship (i.e., linear or logarithmic) between dependent and independent variables. Both models with only soil parameters as explaining variables, and models with soil parameters as well as distance to feeder, huts, and fence as explaining variables, were investigated. For the October 1997 data set, the extreme 16.5 kg N ha^-1 d^-1 recorded near the feeder had to be disregarded in the model exercise. Including distance to feeder, huts, and fence in the model significantly improved the description of the observed variability over a model based solely on soil parameters, and for the March 1998 data set only a model with distance to feeders and DOC could be estimated. There was a tendency for spatial correlation between denitrification rates at adjacent sampling points, but the model was not significantly improved by taking this into account.

In October 1997, denitrification rate decreased with distance to the last position of the feeder, and it increased with distance to the fence and with the concentration of DOC, NO^-₃, and total mineral N (Table 1); the amount of variability explained by the best model was 47%. In March 1998 there was also a negative relationship between denitrification and distance to the feeder, and a positive effect of DOC concentration (Table 1), probably because both denitrification and DOC were elevated close to the feeder. At this sampling time denitrification rates varied much less; the model was able to account for 80% of the variability. Even though distance to feeder and fence were not statistically independent of excretory deposits, as reflected in the distribution of DOC and inorganic N, their contribution to explain denitrification variability suggests that the behavior of animals within the paddock may have stimulated denitrification, conceivably by frequent traffic around the feeder and less intense traffic along the fence.

Part II: Seasonal Study
The seasonal study included 11 samplings covering a period of 336 d. Soil temperature at the 10-cm depth ranged from 16.7°C in July to 1.6°C in December, and gravimetric soil moisture from 16 to almost 30%. The temporal variability of these parameters, as well as of DOC, NH⁺₄, and NO^-₃, were averaged across the 10 samples obtained along the transect (Fig. 3). Except for the summer period (July–September), DOC concentrations inside the paddock were elevated compared with the level at the first sampling, before the sows were introduced. A few extreme rates were recorded during spring, but these were not linked to extreme rates of denitrification (data not shown). Inorganic N started to accumulate within the paddock during the spring of 1998, then dropped to a lower level between June and September. During fall and winter, inorganic N accumulated in the soil (Fig. 3). Nitrification activity was indicated by NO^-₃ accumulation during summer and again during winter, and so the intermediate drop in NO^-₃ concentrations in the fall most likely reflected a changed balance between nitrification activity and loss mechanisms. In Part I of this study there was an almost complete disappearance of NO^-₃ between October and March in the absence of animals (Fig. 2E,F), which also suggests that the accumulation recorded during winter in the seasonal study represented a dynamic balance between N deposition and losses.

View larger version (25K): [in this window] [in a new window]   Fig. 3. Average concentrations of dissolved organic carbon (DOC, top) and inorganic N (middle), and soil temperature and gravimetric moisture content (bottom) throughout the seasonal study. Error bars represent standard errors (SE).

View larger version (44K): [in this window] [in a new window]   Fig. 4. Seasonal variations in denitrification activity (kg N ha-1 d-1) at 10 sampling positions along a transect across a paddock (not all sampling dates are shown).

Denitrification activity was positively related to both DOC and soil moisture (Table 1). However, there was also a negative interaction between these two variables, and the relationship between denitrification, DOC, and soil moisture could be expressed as:

[4]

where denitrification is given as kg N ha^-1 d^-1, W is gravimetric soil moisture (% of dry wt. soil), and DOC is given as mg C kg^-1 dry wt. soil. In Eq. [4], the term (0.0128 - 0.00054 x W) will be positive at gravimetric water contents of up to ca. 25%. Hence, in the experimental period through the sampling in November (cf. Fig. 4), a positive effect of DOC on denitrification activity is predicted by Eq. [4], and during winter an apparent negative effect is predicted. The temporal dynamics of (0.0128 - 0.00054 x W) is shown in Fig. 5. The pattern indicated in Fig. 5 is in line with the understanding that DOC is an important regulator of denitrification activity at lower moisture contents, because it induces the oxygen consumption needed to maintain anaerobic volumes (Petersen et al., 1996). In contrast, an inhibition of denitrification by DOC at high soil moisture levels is not likely to occur. Instead, we suggest that the negative relationship observed during winter was caused by a partly anaerobic soil environment, where DOC accumulated because the mineralization of organic compounds to CO₂ was delayed (Gale and Gilmour, 1988), and where denitrification was limited by low temperatures and NO^-₃ availability. According to Fig. 3, average NO^-₃ concentrations increased during winter, but NO^-₃ diffusion to the sites of denitrification may still have restricted the process, as previously observed in a model system with a discrete layer of manure (Petersen et al., 1996).

View larger version (18K): [in this window] [in a new window]   Fig. 5. A model on the relationship between denitrification and soil parameters (Eq. [4], see text) revealed a negative interaction between dissolved organic carbon (DOC) and soil moisture (W). The term (0.0128 - 0.00054 x W) x DOC, which is plotted against sampling time, represents the combined effect of DOC and this interaction.

	GENERAL DISCUSSION

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

 
The spatial and temporal trends in denitrification activity reported here document that this process can be a substantial but highly variable sink for N in outdoor piglet production systems. The present study is, to our knowledge, the first published report of denitrification from outdoor pig production, but the results may be compared with gaseous losses from pastures grazed by cattle or sheep. Ball and Ryden (1984) reported that 5 to 9% of 250 to 500 kg NO^-₃ N ha^-1 applied as fertilizer to mowed grassland was lost via denitrification. Nitrogen inputs derived from grazing animals stimulate gaseous N losses above what is derived from fertilizer alone; Clayton et al. (1994) found a three- to fourfold increase in N₂O emissions in a grazed pasture compared with an ungrazed plot both receiving 185 kg N ha^-1; the net effect of grazing was 0.4 kg N₂O–N ha^-1 d^-1 during three weeks after fertilization. Ruz-Jerez et al. (1994) estimated an annual denitrification of 19.3 kg N ha^-1 yr^-1 in a pasture that received 400 kg N ha^-1 in urea and was rotationally grazed by sheep, while denitrification from grazed pastures with N supplied via biological N₂ fixation was much lower. Nitrogen losses as N₂ and N₂O were up to 5% of the N inputs. Losses of N₂ are probably at least of the same magnitude as N₂O (Ruz-Jerez et al., 1994), and so all three studies indicate annual losses of 5 to 10% of N inputs from grazed pastures with a high N status. In contrast, Luo et al. (1999) found that denitrification losses from short-term (12–24 h) cattle grazing in paddocks with white clover–ryegrass, during the subsequent 14-d period, were <1% of N deposited. It is possible that animal treading, by damaging the grass cover and compacting the soil, will increase the potential for denitrification from excretal returns and, accordingly, that the duration of the grazing period (like the stock density) may have a bearing on the proportion of N lost via denitrification.

The denitrification loss of 69 kg N ha^-1 yr^-1 estimated for the seasonal study exceeded those reported above for grazed systems. The piglet production system had N surpluses, calculated as the difference between N in feed supplied and N in piglets produced, of 490 kg N ha^-1 yr^-1 in 1997 (Part I) and 620 kg N ha^-1 in 1998 (Part II). When expressed as a fraction of the net N input (i.e., excretal returns), the denitrification loss was 11% and therefore not very different from those observed for other grazing systems. It should be emphasized, though, that N₂O emissions, which may be substantial (Ruz-Jerez et al., 1994), were not fully accounted for in this study.

The losses appeared to be linked to management practices, as well as to climatic conditions. While climate is beyond control, management could be adjusted to minimize risks for environmental losses. First, the surplus of N deposited at any given point within the paddock may be reduced by systematic relocation of huts and feeder (already subject to regulation in Denmark), by dividing the paddock into minor units with only one sow, or perhaps by a rotational system. It should also be possible to reduce the N intake by optimized feeding strategies. The grass cover may help to prevent soil compaction. In the production systems studied there was a clear seasonal variation in the proportion of the paddock with an intact grass cover, which ranged from <50% in January to ca. 80% during summer. Rooting out of grass by pigs may be regulated by insertion of a ring in the pig's nose, but in the present study this behavior was not a major cause of damage to the grass, which was mainly due to treading.

In conclusion, this study has demonstrated that denitrification contributes significantly to N losses in outdoor piglet production systems. Some N losses are probably inevitable but, especially for organic farming practices that rely heavily on within-farm nutrient cycling, such losses should be minimized.¹

	ACKNOWLEDGMENTS

 
The technical assistance of Anker Giversen, Tina Trankjær Olsen, and Gitte Hastrup Andersen is gratefully acknowledged. We also wish to thank Vivi Aarestrup Larsen for valuable information about the farms studied.

	NOTES

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

 
¹ In an article published after preparation of this manuscript, Williams et al. (Soil Use Manage. 16:237–243, 2000) estimated nitrous oxide losses from outdoor pig farming systems at <1% of total N excreted.

	REFERENCES

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

 

• Ball, P.R., and J.C. Ryden. 1984. Nitrogen relationships in intensively managed temperate grasslands. Plant Soil 76:23–33.
• Clayton, H., J.R.M. Arah, and K.A. Smith. 1994. Measurement of nitrous oxide emissions from fertilized grassland using closed chambers. J. Geophys. Res. 99 D8:16599–16607.
• Combs, S.M., and M.V. Nathan. 1998. Soil organic matter. p. 53–58. In Recommended chemical soil test procedure for the North Central Region. North Central Reg. Publ. no. 221 (revised). Stn. Bull. 1001. Missouri Agric. Exp. Stn., Univ. of Missouri, Columbia.
• Elliott, P.W., D. Knight, and J.M. Anderson. 1990. Denitrification in earthworm casts and soil from pastures under different fertilizer and drainage regimes. Soil Biol. Biochem. 22:601–605.
• Folorunso, O.A., and D.S. Rolston. 1984. Spatial variability of field-measured denitrification gas fluxes. Soil Sci. Soc. Am. J. 48: 1214–1219.[Abstract/Free Full Text]
• Golden Software. 1997. Surfer Version 6. Golden Software, Golden, CO.
• Gale, P.M., and J.T. Gilmour. 1988. Net mineralization of carbon and nitrogen under aerobic and anaerobic conditions. Soil Sci. Soc. Am. J. 52:1006–1010.[Abstract/Free Full Text]
• Larsen, V.Aa., and A.G. Kongsted. 2000. Sows on pasture. p. 99–105. In J.E. Hermansen, V. Lund, and E. Thuen (ed.) Ecological animal husbandry in the Nordic Countries. Rep. no. 2/2000. Danish Res. Centre for Organic Farming, Tjele, Denmark.
• Luo, J., R.W. Tillman, and P.R. Ball. 1999. Factors regulating denitrification in a soil under pasture. Soil Biol. Biochem. 31:913–927.
• Nielsen, T.H., and N.P. Revsbech. 1994. Diffusion chamber for nitrogen-15 determination of coupled nitrification–denitrification around soil–manure interfaces. Soil Sci. Soc. Am. J. 58:795–800.[Abstract/Free Full Text]
• Parkin, T.B. 1985. Automated analysis of nitrous oxide. Soil Sci. Soc. Am. J. 49:273–276.[Abstract/Free Full Text]
• Parkin, T.B. 1987. Soil microsites as a source of denitrification variability. Soil Sci. Soc. Am. J. 51:1194–1199.[Abstract/Free Full Text]
• Parkin, T.B., J.J. Meisinger, S.T. Chester, J.L. Starr, and J.A. Robinson. 1988. Evaluation of statistical estimation methods for lognormally distributed variables. Soil Sci. Soc. Am. J. 52:323–329.[Abstract/Free Full Text]
• Petersen, S.O., K. Henriksen, and T.H. Blackburn. 1991. Coupled nitrification–denitrification associated with liquid manure in a gel-stabilized model system. Biol. Fertil. Soils 12:19–27.
• Petersen, S.O., T.H. Nielsen, Å. Frostegård, and T. Olesen. 1996. O₂ uptake, C metabolism and denitrification associated with manure hot-spots. Soil Biol. Biochem. 28:341–349.
• Prentice, R.L., and L.P. Zhao. 1991. Estimating equations for parameters in means and covariance of multivariate discrete and continuous responses. Biometrics 47:825–839.[Web of Science][Medline]
• Robertson, G.P., M.A. Huston, F.C. Evans, and J.M. Tiedje. 1988. Spatial variability in a successional plant community: Patterns of nitrogen availability. Ecology 69:1517–1524.[Web of Science]
• Ruz-Jerez, B.E., R.E. White, and P.R. Ball. 1994. Long-term measurement of denitrification in three contrasting pastures grazed by sheep. Soil Biol. Biochem. 26:29–39.
• Sexstone, A.J., T.B. Parkin, and J.M. Tiedje. 1985. Temporal response of soil denitrification rates to rainfall and irrigation. Soil Sci. Soc. Am. J. 49:99–103.[Abstract/Free Full Text]
• Sims, G.K., T.R. Ellsworth, and R.L. Mulvaney. 1995. Microscale determination of inorganic nitrogen in water and soil extracts. Commun. Soil Sci. Plant Anal. 26:303–316.
• Smith, M.S., and L.L. Parsons. 1985. Persistence of denitrifying enzyme activity in dried soils. Appl. Environ. Microbiol. 49:316–320.[Abstract/Free Full Text]
• Sommer, S.G., and C. Jensen. 1994. Ammonia volatilization from urea and ammoniacal fertilizers surface applied to winter wheat and grassland. Fert. Res. 37:85–92.
• Tiedje, J.M., A.J. Sexstone, D.D. Myrold, and J.A. Robinson. 1982. Denitrification: Ecological niches, competition and survival. Antonie van Leuwenhoek 48:569–583.
• Velthof, G.L., S.C. Jarvis, A. Stein, and O. Oenema. 1996. Spatial variability of nitrous oxide fluxes in mown and grazed grasslands on a poorly drained clay soil. Soil Biol. Biochem. 28:1215–1225.
• Wolf, D.C., T.H. Dao, H.D. Scott, and T.L. Lavy. 1989. Influence of sterilization methods on selected soil microbiological, physical, and chemical properties. J. Environ. Qual. 18:39–44.[Abstract/Free Full Text]
• Worthington, T.R., and P.W. Danks. 1992. Nitrate leaching and intensive outdoor pig production. Soil Use Manage. 8:56–60.
• Yoshinari, T., and R. Knowles. 1976. Acetylene inhibition of nitrous oxide reduction by denitrifying bacteria. Biochem. Biophys. Res. Comm. 69:705–710.[Web of Science][Medline]

This article has been cited by other articles:

				J. M. Siegford, W. Powers, and H. G. Grimes-Casey Environmental Aspects of Ethical Animal Production Poult. Sci., February 1, 2008; 87(2): 380 - 386. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Petersen, S.O. Articles by Eriksen, J. Search for Related Content PubMed PubMed Citation Articles by Petersen, S.O. Articles by Eriksen, J. Agricola Articles by Petersen, S.O. Articles by Eriksen, J. Related Collections Nutrient Management Water Pollution Animal Waste