Published online 25 August 2005
Published in Soil Sci Soc Am J 69:1639-1643 (2005)
DOI: 10.2136/sssaj2004.0364
© 2005 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
Nutrient Management & Soil & Plant Analysis
Production of Nitrogen-15-Labeled Pig Manure for Nitrogen Cycling Studies
Peter Sørensen* and
Ingrid K. Thomsen
Danish Institute of Agricultural Sciences, Dep. of Agroecology, P.O. Box 50, Reseach Centre Foulum, 8830 Tjele, Denmark
* Corresponding author (peter.sorensen{at}agrsci.dk)
 |
ABSTRACT
|
|---|
Nitrogen-15 labeled pig (Sus scrofa) manure was produced to study the homogeneity of labeled pig manure prepared for soil N cycling studies. Growing pigs were fed with a diet consisting of mainly 15N-labeled barley (Hordeum vulgare L.) and peas (Pisum sativum L.) during a period of 11 d. Labeled feces and urine were collected separately. The 15N enrichment of urine N was lower than that of fecal N except for the first 2 d of the 15N feeding period. After 11 d of 15N feeding, the 15N enrichment of fecal N was 7% lower and urine N 28% lower than the enrichment of feed N. The homogeneity of 15N-labeling of feces N, sampled at different times during the 15N feeding, was tested by an incubation test in quartz sand and soil. After 12 wk in soil at 20°C, 25 to 27% of the feces N was in inorganic form, and the N mineralization rate declined after this period. The test indicated that the 15N-labeled pig feces had a satisfactory labeling homogeneity even though the feed ingredients had variable 15N enrichments. We recommend the collection of urine and feces separately and the use of labeled feces in combination with unlabeled urine and vice versa in studies of the fate of pig manure N in agroecosystems. If feces and urine with a similar 15N enrichment can be obtained the two components can be mixed in the same manure.
Abbreviations: ANI, added-nitrogen interaction • PE, polyethylene
|
INTRODUCTION
|
|---|
ABOUT HALF OF THE N APPLICATION to agricultural land in Denmark is supplied with animal manure. Improved understanding of the cycling of animal manure N in soil is important for optimizing the utilization of N in agriculture and reducing losses to the environment. The fate of manure N can normally only be measured indirectly by use of the conventional difference method where N uptake and losses are measured in plots with and without manure application. Isotopic labeling technique can be used to study the fate of manure N in agroecosystems directly. The inorganic N fraction in manure can be labeled by adding a small amount of 15N-labeled ammonium (Trehan and Wild, 1993). Labeling of the organic N has been made by feeding poultry with 15N-labeled barley (Kirchmann, 1990; Thomsen, 2004), sheep with 15N-labeled hay (Sørensen et al., 1994; Thomsen et al., 1997), dairy cattle with 15N-labeled silage (Powell and Wu, 1999; Langmeier et al., 2002), and pigs with labeled soybean and barley (Chantigny et al., 2004).
In principle all feed ingredients should have the similar 15N enrichment to obtain a uniform labeling of feces. In practice, however, it is difficult to produce a complex diet with a constant enrichment. Even with one homogeneous feed source, the excreta may not be uniformly labeled due to endogenous N with a lower enrichment being excreted to the digestive tract (Sørensen et al., 1994). Chantigny et al. (2004) found some heterogeneity in the 15N enrichment of pig manure (mixture of feces and urine) after feeding with a diet containing ingredients of variable 15N enrichment. We hypothesize that the labeling heterogeneity observed by Chantigny et al. (2004) was mainly due to different enrichments of feces N and urine N, and that a separate collection of urine and feces could overcome the problem of labeling heterogeneity.
The aim of this study was to evaluate the homogeneity of 15N-labeled pig manure produced for N cycling studies. The urine and feces were collected separately during a period of 15N feeding, and the labeling homogeneity of feces N was studied during decomposition in quartz sand and soil.
|
MATERIALS AND METHODS
|
|---|
Production of 15N-labeled Pig Manure
In spring 2000, two plots of 50 m2 were sown to barley (Hordeum vulgare L. cv. Alanis) and two plots of 36 m2 to field peas (Pisum sativum L. cv. Athos). On May 2, K15NO3 (9.631 atom% 15N excess) corresponding to 100 kg N ha–1 was applied to one of the barley plots. The other barley plot received the same amount of N in unlabeled mineral fertilizer. The two pea plots were supplied with K15NO3 or unlabeled mineral fertilizer on May 2, May 18, and June 9, respectively, corresponding a total of 92 kg N ha–1.
A diet containing barley and peas was formulated according to the Danish nutritional recommendations for pigs. The ingredients were ground and mixed to two similar diets containing either labeled or unlabeled N. The composition of the experimental diets is shown in Table 1. The labeled diet was fed to two female pigs (45–50 kg) for 11 d and the unlabeled diet to two other pigs from the same brood. The pigs were placed in metabolism cages and fitted with bladder catheters for separate collection of urine and feces. After an adaptation period of 5 d on similar unlabeled feed, feces, and urine were collected quantitatively for 11 d after the start of 15N feeding. The samples were pooled for each 24-h collection period and then frozen (–18°C). The first sampling was pooled from 0 to 16 h after the first 15N feeding (Day 1). The next pooled sample was from 16 to 40 h after the start of 15N-feeding (Day 2), and so on. One to two days after the start of 15N feeding, one of the pigs (Pig 2) caught an infection with fever and lost its appetite. It was treated with antibiotics ("Potrox" injection) and it recovered after 1-2 d. The labeled and unlabeled excreta have subsequently been used in studies of the fate of pig slurry N (Sørensen and Thomsen, 2005).
Evaluation of 15N Labeling
The uniformity of labeling of fecal N was evaluated by incubation in N-free quartz sand and in soil (Sørensen et al., 1994).
Sand Incubation Test
Portions of feces sampled on Days 3, 5, 8, and 11 (Pig 1 and 2) containing 10 mg N was applied to 20-g quartz sand in 100-mL polyethylene (PE) flasks and 3.38-g demineralized water was added corresponding to 55% of the water-holding capacity. The mixtures were incubated at 20°C. The flasks were covered with Parafilm (American National Can, Chicago, IL) with holes for aeration, and lost water was replaced weekly. Enough samples were prepared for triplicate sampling after 0, 1, 4, and 12 wk. The samples were extracted for analysis of ammonium and nitrate and the 15N enrichment of ammonium N.
Soil Incubation Test
A loamy sand (Typic Hapludult, mixed mesic) was sampled from the plow layer of the arable field where field experiments with the labeled manure were located (Sørensen and Thomsen, 2005). The soil contained 90 g of clay (<2 µm), 120 g of silt (2–20 µm), 760 g of sand (20 µm–2 mm) kg–1 dry soil. Soil C and N contents were 1.65 g N and 17.2 g C kg–1 soil. Moist soil samples (50 g on dry matter basis) were added to 250 mL PE flasks and portions of feces sampled on Days 5 and 11 (Pig 1) corresponding to 10 mg N in feces were mixed with the soil. Water was added equivalent to 55% of the water-holding capacity of the soil. The flasks were covered with Parafilm with holes for aeration, and lost water was replaced weekly. The mixture was incubated in the dark at 20°C. Separate soil samples without feces additions were included. After 0-, 1-, 2-, 4-, 8-, 12-, 16-, and 32-wk triplicate samples of each treatment were extracted for analysis of inorganic N.
Analytical Methods
Total N in feces samples was determined using a Kjeldahl method (Tecator Kjeltec Auto 1030, Tecator, Höganäs, Sweden). Inorganic N in soil, sand and feces was extracted by 2 M KCl for 1 h (2.5:100 for feces and 1:2 for soil/sand), followed by centrifugation and filtration through glass filters. Ammonium (NH4–N) and nitrite + nitrate N (NO2–N + NO3–N) in extracts were measured by flow colorimetry (Autoanalyzer II, Bran + Luebbe GmbH, D-22803 Norderstedt, Germany). Ammonium N in sand and feces extracts was concentrated for 15N analysis using a diffusion procedure where N in KCl extracts after adding MgO was diffused as NH3 to an acidified glass filter enclosed in Teflon tape (Sørensen and Jensen, 1991). Nitrogen-15 in ammonium and nitrate in soil extracts were measured using the same procedure after the addition of Devarda's alloy to the extract. Ammonium N in Kjeldahl extracts was concentrated by diffusion as described by Jensen (1991). The 15N enrichments of concentrated extracts were determined using a mass spectrometer coupled to an elemental analyzer (Europa Scientific Integra, Crewe, UK).
The amount of soil inorganic N derived from feces was estimated by measuring the 15N content of this soil N pool. Inorganic N derived from feces was likewise calculated by the difference method by subtracting inorganic N in unamended soil from inorganic N in feces-amended soil. Assessment of statistical differences was based on analysis of variance using the SAS procedure GLM (SAS Institute Inc., Cary, NC) and least significant differences (LSD) (P < 0.05) were calculated.
|
RESULTS AND DISCUSSION
|
|---|
Feed Composition
It was anticipated that the split application of 15N to peas would produce similar enrichment as in barley where the whole 15N amount was applied in one application. However, the peas grew very well due to the climatic conditions and apparently a significant biological N fixation occurred despite the N fertilization. The 15N was therefore diluted and the 15N enrichment of the peas was <50% of that in barley (Table 1). A small amount of unlabeled N in the form of synthetic methionine and wheat bran was also included in the labeled diet. Thus, the ingredients of the labeled diet were not uniformly labeled.
Composition of Pig Excreta
The average ratio of excreted fecal N/urinary N was 0.35:0.65 over the 11-d feeding period (data not shown), which is within the normal range for growing pigs (Sørensen and Fernandez, 2003). The apparent utilization of feed N calculated from feed N intake minus measured N excretion was 41%.
After the start of 15N feeding, the enrichment of urine N quickly increased while the 15N enrichment of feces was initially lower (Fig. 1)
. However, after 3 d the enrichment of fecal N exceeded that of urine N as also observed for ruminants (Sørensen and Jensen, 1998; Powell and Wu, 1999). At the end of the 15N-feeding period, the 15N enrichment of fecal N was 7% lower than in the feed due to excreted unlabeled endogenous N (Sørensen et al., 1994), while the enrichment of urinary N was 28% lower than in the feed (Fig. 1). The difference in enrichment of feces and urine makes it difficult to label total slurry N uniformly, and urine- and feces-derived N should therefore be followed separately as the lability of urine and feces N is different (Jensen et al., 1999). In accordance with this, Chantigny et al. (2004) found that when 15N labeled pig excreta (urine and feces) were collected without separation of urine and feces, the 15N enrichment of mineralized N was lower than that of the total N pool. The fate of urine N in animal manure may be influenced by the other manure components such as feces and bedding material. Therefore the labeled component should be mixed with the other manure components in unlabeled form, as practiced by for example, Thomsen et al. (1997) and Jensen et al. (1999). This allows following the fate of each component separately and the fate of total manure N can finally be calculated from the weighted recovery of N from each manure component. Alternatively, feces and urine selected for having a similar 15N enrichment may be mixed, and the fate of total N can be followed.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1. The 15N content of pig feces and urine excreted by Pig 1 during 11 d of feeding with a 15N-enriched diet.
|
|
The composition of feces N sampled at different time after the start of 15N-feeding is shown in Table 2. The proportion of fecal N found as ammonium varied from 6 to 16%, and this proportion varied from day to day (Table 2). The 15N enrichment of ammonium N contained in feces was always lower than that for total fecal N (Table 2), probably because a high proportion of ammonium in feces derives from endogenous N with a low 15N enrichment.
Evaluation of Labeling Homogeneity of Fecal Nitrogen
The labeling of the urine N can be assumed to be sufficiently homogeneous (Jensen et al., 1999). Feces are composed of feed residues, microbial products, microorganisms, secretions, and tissues from the digestive tract with different lability and different 15N enrichment after feeding with 15N. Therefore an evaluation of the uniformity of feces labeling was performed by incubation of feces in N-free quartz sand and in soil. In sand the enrichment of inorganic N derived from feces can be measured directly, but the mineralization process is not as fast in sand as in soil with a natural microbial population (Sørensen et al., 1994). Although the ammonium N in feces had lower 15N enrichment than total fecal N (Table 2), there was generally no significant difference between the enrichment of ammonium N in sand and the enrichment of total fecal N after a few weeks of decomposition (Fig. 2)
. An exception to this was feces sampled on Day 5 from Pig 2 where a low enrichment of ammonium N compared with total N was observed after 4 wk of decomposition (Fig. 2). The reason for this was unclear, but it might be related to the previous infection. These results suggest that after a few weeks' decomposition, an uneven enrichment of fecal N will not cause problems in N cycling studies.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 2. Nitrogen-15 enrichment of ammonium N formed during decomposition of pig feces in quartz sand. The feces were sampled on Days 3, 5, and 11 after the start of 15N-feeding. Bars indicate standard errors (n = 3), NS = not significant (p < 0.05).
|
|
After application to soil, the uniformity of labeling can only be evaluated indirectly by comparing the inorganic N derived from feces, estimated with 15N technique, with the net N mineralization from the same material estimated with the difference method. A constant relationship between net mineralization and mineralization of labeled N indicates a uniform labeling. One week after application to soil, 13 to 15% of the fecal N was in inorganic form increasing to 25 to 27% after 12 wk (Fig. 3)
. After this period the N mineralization rate declined. There was a great similarity between inorganic N derived from feces when determined by 15N technique and estimated with the difference method (Fig. 4)
. The correlation coefficient between the two methods was as high as 0.96. The slope of the regression line (Fig. 4) was 0.96 ± 0.03 and not significantly different from 1. The results suggest that the enrichment of mineralized N remained nearly constant over the test period and that the enrichment of mineralized N was equal to that of total fecal N. We therefore conclude that the labeling of the feces used in this study was sufficiently homogeneous, and no corrections for uneven labeling were needed.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 3. Inorganic soil N derived from feces as estimated either by the 15N technique or the difference method, after incubation of 15N-labeled pig feces (Pig 1) in a sandy soil for 32 wk. The feces were sampled on Days 5 and 11 after the start of 15N-feeding. Bars indicate standard errors (n = 3).
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 4. Relationship between inorganic soil N derived from feces as estimated either by the 15N technique or the difference method, after incubation of 15N-labeled pig feces in a sandy soil for 32 wk. The feces were sampled on Days 5 and 11 after start of 15N-feeding (see Fig. 3).
|
|
The correspondence between net release of total and labeled inorganic N in the present study (Fig. 3 and 4) indicates that added N interactions (ANI) due to the mineralization/immobilization turnover, as defined by Jenkinson et al. (1985), were minimal in the soil incubation test. Sørensen (2001) found that when labeled organic manure N was applied to soil together with unlabeled ammonium N or vice versa, there was a significant interaction with manure 15N being mineralized while unlabeled ammonium N was immobilized simultaneously. Thus, a significant ANI occurred. In the present study, about 10% of the fecal 15N was ammonium and apparently the microorganisms involved in the manure decomposition obtained the immobilized N from this labeled ammonium N and from mineralized fecal N, resulting in insignificant ANI in the soil incubation test.
|
CONCLUSIONS
|
|---|
Pig feces with a satisfactory labeling homogeneity were obtained by feeding pigs on 15N-labeled feed, even though the feed ingredients had variable 15N enrichments. The 15N enrichment of urine was lower than of fecal N except the first 2 d of the 15N-feeding period. Therefore it is recommended that urine and feces are collected separately and that labeled feces are used in combination with unlabeled urine and vice versa in studies of the fate of pig manure N in agroecosystems. In soil incubations, inorganic N derived from feces N measured by the isotope technique was similar to the estimate using the difference method. However, the isotope technique has advantages over the difference method in field experiments of longer duration by having a higher precision and enabling the measurement of residual manure N in soil and estimation of residual N effects in the years after manure application.
|
ACKNOWLEDGMENTS
|
|---|
The work was supported by the Ministry of Food, Agriculture and Fisheries (projects: HAR98-DJF-1 and VMPIII aktstykke, slurry technology). We thank the staff in "Research Unit Organic Matter and Microbial Ecology" for skilled technical assistance.
Received for publication November 24, 2004.
|
REFERENCES
|
|---|
- Boisen, S., and J.A. Fernández. 1997. Prediction of the total tract digestibility of energy in feedstuffs and pig diets by in vitro analyses. Anim. Feed Sci. Tech. 68:277–286.[CrossRef]
- Chantigny, M.H., D.A. Angers, C. Pomar, and T. Morvan. 2004. The use of 15N-enriched feed to label pig excreta for N cycling studies. Can. J. Soil Sci. 84:43–48.
- Jenkinson, D.S., R.H. Fox, and J.H. Rayner. 1985. Interactions between fertilizer nitrogen and soil nitrogen–the so-called ‘priming effect’. J. Soil Sci. 36:425–444.[CrossRef]
- Jensen, B., P. Sørensen, I.K. Thomsen, E.S. Jensen, and B.T. Christensen. 1999. Availability of nitrogen in 15N-labeled ruminant manure components to successively grown crops. Soil Sci. Soc. Am. J. 63:416–426.[Abstract/Free Full Text]
- Jensen, E.S. 1991. Evaluation of automated analysis of 15N and total N in plant material and soil. Plant Soil 133:83–92.[CrossRef]
- Kirchmann, H. 1990. Nitrogen interactions and crop uptake from fresh and composted 15N-labelled poultry manure. J. Soil Sci. 41:379–385.
- Langmeier, M., E. Frossard, M. Kreuzer, P. Mäder, D. Dubois, and A. Oberson. 2002. Nitrogen fertilizer value of cattle manure applied on soils originating from organic and conventional farming systems. Agronomie 22:789–800.[CrossRef][ISI]
- Powell, J.M., and Z. Wu. 1999. Nitrogen-15 labeling of dairy feces and urine for nutrient cycling studies. Agron. J. 91:814–818.[Abstract/Free Full Text]
- Sørensen, P. 2001. Short-term nitrogen transformations in soil amended with animal manure. Soil Biol. Biochem. 33:1211–1216.[CrossRef]
- Sørensen, P., and J.A. Fernandez. 2003. Dietary effects on the composition of pig slurry and on the plant utilization of pig slurry nitrogen. J. Agric. Sci. (Cambridge) 140:343–355.[CrossRef]
- Sørensen, P., and E.S. Jensen. 1991. Sequential diffusion of ammonium and nitrate from soil extracts to a polytetrafluoroethylene trap for 15N determination. Anal. Chim. Acta 252:201–203.[CrossRef]
- Sørensen, P., and E.S. Jensen. 1998. The use of 15N labelling to study the turnover and utilization of ruminant manure N. Biol. Fertil. Soils 28:56–63.
- Sørensen, P., and I.K. Thomsen. 2005. Separation of pig slurry and plant utilization and loss of 15N-labeled slurry nitrogen. Soil Sci. Soc. Am. J. 69:1644–1651 (this issue).[Abstract/Free Full Text]
- Sørensen, P., E.S. Jensen, and N.E. Nielsen. 1994. Labelling of animal manure nitrogen with 15N. Plant Soil 162:31–37.[CrossRef]
- Thomsen, I.K. 2004. Nitrogen use efficiency of 15N-labeled poultry manure. Soil Sci. Soc. Am. J. 68:538–544.[Abstract/Free Full Text]
- Thomsen, I.K., V. Kjellerup, and B. Jensen. 1997. Crop uptake and leaching of 15N applied in ruminant slurry with selectively labelled faeces and urine fractions. Plant Soil 197:233–239.[CrossRef]
- Trehan, S.P., and A. Wild. 1993. Effects of an organic manure on the transformations of ammonium nitrogen in planted and unplanted soil. Plant Soil 151:287–294.[CrossRef]
This article has been cited by other articles:
|
|
|
|
 
P. Sorensen and I. K. Thomsen
Separation of Pig Slurry and Plant Utilization and Loss of Nitrogen-15-labeled Slurry Nitrogen
Soil Sci. Soc. Am. J.,
August 25, 2005;
69(5):
1644 - 1651.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
