SSSAJ
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 29 May 2008
Published in Soil Sci Soc Am J 72:949-959 (2008)
DOI: 10.2136/sssaj2006.0376
© 2008 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bosshard, C.
Right arrow Articles by Oberson, A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Bosshard, C.
Right arrow Articles by Oberson, A.
Agricola
Right arrow Articles by Bosshard, C.
Right arrow Articles by Oberson, A.
Related Collections
Right arrow Soil Methods/Instrumentation
Right arrow Structure and Properties
Right arrow Agroforestry

SOIL FERTILITY & PLANT NUTRITION

Incorporation of Nitrogen-15-Labeled Amendments into Physically Separated Soil Organic Matter Fractions

C. Bossharda, E. Frossarda, D. Duboisb, P. Mäderc, I. Manolovd and A. Obersona,*

a Institute of Plant Sciences, Group of Plant Nutrition, ETH Zurich Research Station Eschikon, Lindau Switzerland
b Acroscope Reckenholz-Tänikon (ART), Reckenholz, Zurich, Switzerland
c Research Institute of Organic Farming (FiBL), Frick, Switzerland
d Agricultural Univ., Dep. of Agrochemistry and Soil Science, Plovdiv Bulgaria

* Corresponding author (astrid.oberson{at}ipw.agrl.ethz.ch).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Physically separated soil organic matter (SOM) fractions may take different functions in soil N dynamics. We studied the effect of long-term organic matter (OM) management and different soil biological activity on the incorporation of N added with organic and mineral amendments into aggregate fractions and size density fractions. We applied 15N-labeled sheep feces, urine, and mineral fertilizer to microplots installed in plots of conventional (CONMIN) and bio-organic (BIOORG) cropping systems. Soil sampled 112 d after amendment was separated into macro-, microaggregates, and microstructures. Aggregates were then fractionated into free light fraction (LF), intra-aggregate particulate organic matter (iPOM), and the mineral-associated organic matter fraction (MF). Of total soil N, 67% was contained in macroaggregates. Size density fractionation of aggregates revealed that about 60% of soil N was stored in MF while LF and iPOM contained together <3% of soil N. Despite long-term OM input and higher soil biological activity in BIOORG than CONMIN the two soils did not differ in the distribution and content of N in aggregate and size density fractions. Recovery of 15N in nonfractionated soil ranged from 20% (SlurryF) to 25% (SlurryU) of originally applied 15N. The small macroaggregates were for each amendment the major sink (7–12% of applied 15N). In all aggregates and for all amendments, MF was the most important 15N sink, totally containing between 6.6% (SlurryF) to 11.6% (SlurryU) of applied 15N. Less than 1% of applied 15N was recovered in LF, and even less (<0.5%) in iPOM. The proportion of amendment-derived N in aggregate fractions and in several size density fractions (LF, fine iPOM, MF) was higher for urine than for feces and mineral fertilizer. Recovery of urine-derived 15N was greater in aggregate fractions of BIOORG than CONMIN soil. During dispersion of aggregates to obtain iPOM and MF, about 27% of total soil N and between 37 and 55% of 15N contained in non-fractionated soil was lost, showing the importance of aggregation to protect N.

Abbreviations: AF, aggregate fraction • Amd, amendment • BIOORG, bio-organic cropping system • CONMIN, conventional cropping system • CS, cropping system • DM, dry matter • HF, heavy fraction • iPOM, intra-aggregate particulate organic matter • LF, light fraction • MF, mineral-associated organic matter fraction • MineralN, mineral fertilizer N (15NH415NO3) • Ndflc, N derived from the labeled component of the amendment, OM, organic matter • POM, particulate organic matter • SDF, size density fraction • SlurryF, sheep slurry (unlabeled urine + 15N-labeled feces) • SlurryU, sheep slurry (15N-labeled urine + unlabeled feces) • SOM, soil organic matter • 0N, unfertilized control


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil organic matter is the most important N reservoir in soils. The amount and quality of SOM present and the rate of SOM turnover are influenced by agricultural management practices, which, in turn, may also affect amounts and forms of N retained in the soil. Repeated manure application and incorporation of animal and crop residues usually increase SOM content and affect its quality (Aoyama et al., 1999; Kong et al., 2007). Soil organic matter and associated N is very heterogeneous and consists of fractions differing in composition, biological function, and stabilization by chemical and physical mechanisms. Different SOM fractions can be obtained by aggregate separation and size density fractionation (Six et al., 2000; von Lützow et al., 2007). Aggregate separation assesses the distribution of organic matter (OM) among large macroaggregates (>2000 µm), small macroaggregates (250–2000 µm), microaggregates (50–250 µm) and microstructures (<50 µm) (Six et al., 1998). Macroaggregates consist of microaggregates, primary organo–mineral complexes and uncomplexed particulate OM (POM). The POM is neither present as readily recognizable litter components (typically > 2 mm) nor incorporated into primary organo–mineral complexes. Microaggregates are composed of primary organo-mineral complexes and clay microstructures (Christensen, 2001; Tisdall and Oades, 1982). In combined aggregate and size density separation schemes (Six et al., 1998), size density fractionation is then used to divide the aggregates into the free light fraction (LF; interaggregate OM) and the heavy fraction (HF). After dispersion the HF is separated into coarse and fine iPOM and the MF. Free LF consists mainly of partially decomposed plant material, animal residues or manure, microbial debris that are not associated to the mineral soil particles and of older uncomplexed material previously occluded within aggregates (Christensen, 2001; Fliessbach and Mäder, 2000). Free LF is more closely related to plant residues and often responds more sensitively toward changes in soil and/or fertility management than iPOM, but both free LF and iPOM fractions are known to be sensitive indicators of the effects of agricultural management practices on SOM (Gregorich et al., 2006).

Crop N supply in organic farming relies on the mineralization of SOM and organic N sources such as animal manure. Therefore, organic farming aims at maintaining or increasing SOM content. Manure N not taken up by the crop shall be retained in the soil to avoid N losses and conserve it as N source for subsequent crops. Few studies are concerned with the long-term effect of animal manure and mineral fertilizer on SOM fractions (Christensen, 1988; Aoyama et al., 1999; Wander et al., 2007). Only few of them used the stable isotope 15N as tracer to follow the incorporation of 15N-labeled mineral fertilizer (Compton and Boone, 2002; Ladd et al., 1977; Monaghan and Barraclough, 1995) or 15N-labeled plant residues (Kölbl et al., 2006; Moran et al., 2005; Vanlauwe et al., 1998; Kong et al., 2007) into different SOM fractions. However, incorporation of 15N-labeled animal manure has not been investigated. Also, we know no study where incorporation of 15N from same amendments applied either to organically or conventionally cropped soils has been followed.

We applied sheep slurries (urine-feces mixture) of which either urine or feces were labeled with 15N, and 15N-labeled mineral fertilizer to microplots installed in plots of a field experiment. These plots were either managed according to conventional cropping practices and had received only mineral fertilizer since 1985 (CONMIN), or were managed according to bio-organic cropping practices and had received only farmyard manure and slurry since 1978 (BIOORG). Four months after application of the amendments, at harvest of mature wheat grown in the microplots, we studied the recovery of 15N in physically separated SOM fractions in soil sampled from the 0- to 18-cm layer in the microplots. Soil microbial biomass and activity are higher in the BIOORG than the CONMIN soil (Fliessbach and Mäder, 2000; Mäder et al., 2006). As OM management, for example, through manure amendment, and soil biological activity may affect the incorporation of OM into different SOM fractions we hypothesized that (i) the distribution of N among the SOM fractions differs between CONMIN and BIOORG soils due to the long-term manure input in BIOORG and (ii) the incorporation patterns of freshly applied 15N-labeled animal manure N and mineral fertilizer N differ because of different forms of N applied with the amendments and because of OM input with animal manure.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Field Experiment and Soil Sampling
Microplots were installed in December 2002 in plots of a long-term field experiment located in Therwil (7°33' E, 47°30' N) near Basel (Switzerland) managed by Acroscope Reckenholz-Tänikon (ART), Zurich, and the Research Institute of Organic Farming (FiBL), Frick, Switzerland. The soil is a loamy silt Typic Hapludalf (Soil Survey Staff, 1999) developed on loess in a temperate climate. Selected soil properties are given in Table 1 . The conception and experimental design of the field experiment have been described in detail by Mäder et al. (2006). Briefly, two conventional and two organic cropping systems are being compared since 1978 in a split-split plot latin square design with four replicates. For our study, we selected the four replicate plots of a conventional (CONMIN) and the bio-organic (BIOORG) cropping system. CONMIN receives exclusively water-soluble mineral fertilizers since 1985 according to official Swiss fertilization guidelines (Walther et al., 2001) (for N input see Table 1) and is otherwise managed according to the rules of integrated plant production (KIP, 1999). BIOORG is managed according to bio-organic guidelines since 1978 (VSBLO, 2003) and gets exclusively organic fertilizers in form of farmyard-manure and slurry (for N input see Table 1) with an average annual organic C input of 2240 kg ha–1 yr–1 (Fliessbach et al., 2007). The two cropping systems mainly differ in fertilization and plant protection, including more frequent mechanical weeding in BIOORG than CONMIN (Table 1). Ploughing (frequency, depth), crop rotation (duration, sequence), and residue management (e.g., removal of winter wheat straw) are the same for BIOORG and CONMIN. A crop rotation period lasts 7 yr. The fourth rotation period (1999–2005) included winter wheat (Triticium aestivum L.), 2 yr grass-clover mixture (composed by Lolium perenne L., Dactylis glomerata L., Festuca pratensis L., Phleum pratense L., Trifolium repens L., Trifolium pratense L.), potatoes (Solanum tuberosum L.), winter wheat, soybean (Glycine max L.), and maize (Zea mays L.). After winter wheat and soybean, green manure intercrops mainly consisting of Phacelia tanacetifolia and Secale cereale, respectively, were sown. Also crop varieties are usually the same in both cropping systems. The plots chosen for our study had in October 2002 been sown with winter wheat (Triticium aestivum, var. Titlis).


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

  		Table 1. Selected soil properties of the topsoil (0–18 cm), type of fertilizers applied and plant protection of the conventional (CONMIN) and the bio-organic (BIOORG) cropping system (n = 4).

 
The microplots were defined by frames with a length of 33 cm, a width of 14 cm, and a height of 23 cm, and were open at the top and the bottom. In December 2002 they were driven into the soil of CONMIN and BIOORG plots to a depth of 18 cm. In March 2003, at beginning of tillering of winter wheat, two 15N-labeled animal manures and a 15N-labeled mineral fertilizer were deployed as a one-time application. The two animal manures were feces-urine mixtures (slurries) just differing in the labeled component. One contained 15N-labeled sheep urine (SlurryU) while the other contained 15N-labeled sheep feces (SlurryF). Mineral fertilizer-N (MineralN) was applied in form of 15NH415NO3 as aqueous solution (109.7 mmol N L–1). A nonfertilized treatment was included as control (0N). Each of the four amendment treatments (SlurryU, SlurryF, MineralN, 0N) was applied to microplots installed in CONMIN and BIOORG, resulting in four replicates per cropping system–amendment combination.

The 15N-labeled urine and feces were obtained by feeding a sheep with 15N-labeled ryegrass hay for 9 d and collecting urine and feces separately. For the study, urine and feces with the highest enrichment excreted on the ninth day were used. Non-labeled urine and feces were collected from the same sheep at Day 6 of the initial 7-d lasting feeding period with non labeled ryegrass hay. The non-labeled ryegrass hay was obtained under identical conditions as the labeled hay (Bosshard, 2007).

To reduce gaseous N losses and for easier application both slurries were diluted 1:1 with water. To minimize disturbance of young winter wheat plants, slurries and mineral fertilizer were distributed into three about 5-cm deep and 14 cm long narrow channels located between the wheat plants. Channels were covered with soil immediately after application of the amendments to minimize gaseous N losses, simulating direct injection. The applied rates and characteristics of the slurries and mineral fertilizer are shown in Table 2 . Slurries contained similar amounts of mineral N (NH4 and NO3) as the mineral fertilizer.


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

  		Table 2. Characteristics and applied rates of the different amendments.

 
For aggregate separation and size density fractionation six soil cores were randomly collected from each microplot to a depth of 18 cm with an auger (Ø 2.5 cm, Eijkelkamp, Netherlands) immediately after harvest of mature winter wheat in July 2003. At harvest also the wheat straw had completely been removed. The mature shoots of wheat contained 37, 10, and 47% of 15N applied with urine, feces, and mineral fertilizer, respectively (Bosshard, 2007). The six soil cores from the same microplot were pooled and air dried. Soil samples were not sieved before drying.

Aggregate Separation
One hundred grams of air-dried soil was capillary rewetted allowing trapped air to escape with minimal disruption of soil structure (Cambardella and Elliott, 1993). Subsequently, soil was wet sieved through a series of three sieves (2000, 250, and 50 µm), exactly as described by Six et al. (1998). Briefly, the soil was submerged for 5 min in deionized water. Aggregate separation was then achieved by manually moving the 2000-µm sieve up and down. The stable aggregate fraction remaining on the sieve was gently washed off the sieve into a pan. Water plus soil that went through the sieve was poured onto the next sieve and the sieving was repeated. The method results in four aggregate fractions: large macroaggregates (>2000 µm), small macroaggregates (250–2000 µm), microaggregates (50–250 µm), and microstructures (<50 µm) (Fig. 1 ). After aggregate separation all aggregate samples were air dried.


Figure 1
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 1. Aggregate separation and size density fractionation (Six et al., 1998) {dagger} using sodium polytungstate, {ddagger} free light fraction, § heavy fraction, hexametaphosphate, # mineral-associated organic matter fraction, {dagger}{dagger} intra-aggregate particulate organic matter.

 
Size Density Fractionation
Size density fractionation was used to separate each of the aggregate fractions > 50 µm into three POM fractions (free LF, coarse and fine iPOM) and the MF (Fig. 1; Six et al., 1998). Size density fractionation was conducted with 10-g aggregate subsamples. First free LF was separated from the HF by density flotation in 1.85 g cm–3 sodium polytungstate. After dispersion with hexametaphosphate the HF was wet sieved through a series of sieves (2000 µm and/or 250 µm and 50 µm depending on the aggregate fraction) resulting in the following iPOM fractions: coarse iPOM (250–2000 µm) and/or fine iPOM (50–250 µm). Material passing the 50-µm sieve was assigned to MF. All size density fractions were air dried after separation.

Nitrogen and Carbon Analyses
Nitrogen, Nitrogen-15, and Carbon Analyses
Air-dried soil, aggregate fractions and size density fractions were finely ground using a ball mill (Retsch, Haan, Germany) before total N and C and 15N abundance analysis on a continuous flow Roboprep CN Biological Sample Converter coupled to a Tracermass Mass Spectrometer (Europa Scientific, Crewe, England).

Atom% Nitrogen–15 Excess
The atom% 15N excess of each sample denotes the 15N abundance of the sample minus the natural abundance of its reference sample. The natural abundance of reference soil samples taken from the 0N microplots where no 15N-labeled amendment was applied was 0.3684 atom% 15N for CONMIN and 0.3693 atom% 15N for BIOORG.

Nitrogen Derived from the Labeled Component of the Amendment (Ndflc)
The proportion of N derived from the 15N-labeled component of the amendments–urine for SlurryU, feces for SlurryF, and 15NH415NO3 for MineralN–in the aggregate fractions or the size density fractions, was calculated according to Eq. [1] using isotope pool dilution principles (Hauck and Bremner, 1976):

Formula 1[1]
where 15Nexfractions denotes the atom% 15N excess of the aggregate fractions or size density fractions and 15Nexlc denotes the atom% 15N excess of the labeled component of the amendment (Table 2). Analysis of the homogeneity of 15N labeling of fecal N by physico- and biochemical techniques revealed that the excess of total fecal 15N can be used for calculations in spite of small deviations in excess among fecal N fractions (Bosshard, 2007).

Nitrogen-15 Recovery
For 15N recovery in the aggregate fractions or the size density fractions the amount of soil in a microplot was calculated by multiplying the volume of the microplot with the soil bulk density. Soil bulk density for the 0- to 18-cm soil layer was 1.3 kg dm–3 and was not significantly different between CONMIN and BIOORG. It resulted in 10.8 kg soil dry matter per microplot. For the aggregate fractions their proportion on dry matter in the soil, and for the size density fractions their proportion on dry matter in the aggregate fractions and subsequently in the soil were used to calculate their respective dry weight per microplot. From the dry weight, the N concentration and the 15N excess of aggregate fractions and size density fractions, respectively, the amount of 15N recovered in each of these fractions was calculated. The amount divided through the amount of 15N applied to the microplots multiplied by 100 is the recovery in each fraction as percentage of applied 15N.

Statistical Analysis
Analysis of variance was performed by using the statistical analyses package SYSTAT 11 (Systat Software Inc., USA). Effects of the main factors cropping system and amendment and of the secondary factors aggregate fraction and size density fraction were tested using a split-plot design. For analysis of variance percentage data was transformed using arcsin-transformation. In case of significant effects separation of means was tested using Tukey's HSD (honestly significant difference) test with a significance level of P ≤ 0.05.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Soil Characteristics
Soil microbial biomass and activity were higher in BIOORG than CONMIN in March 2003 (Table 1). This confirms several previous measurements on microbial biomass and activity in the same soils (Fliessbach and Mäder, 2000; Mäder et al., 2002). Soil pH was lower in CONMIN than BIOORG, probably due to the acidifying effect of mineral fertilizers (Mäder et al., 2006). In spite of long term organic fertilization in BIOORG, total C and N concentrations in soils were not significantly different between CONMIN and BIOORG (Table 1). This agrees with Fliessbach et al. (2007) who did an extended study on soil organic C in the same field experiment and who report a decreasing trend in soil organic C concentrations in BIOORG and CONMIN since starting the field experiment in 1977 (–15% in CONMIN and –9% in BIOORG). Also Wander et al. (2007) found no accumulation of soil organic C under organic cropping. In contrast, several studies suggest that the repeated application of animal manure increases soil N content (Glendining et al., 1997; Kong et al., 2007).

Aggregate Fractions
Distribution and Total Nitrogen Concentration
Distribution of the aggregate fractions was not affected by the cropping system. The sum of the different aggregate fractions represented 98.5% of the non-fractionated soil dry matter (DM). The macroaggregates (>250 µm) represented 68.6% of total soil DM, with the greatest contribution from the small macroaggregates (Table 3 ).


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

  		Table 3. Distribution of aggregate fractions in the soil, total N concentration in the aggregate fractions, and contribution of aggregate fraction N to total soil N. Standard deviation is shown in brackets. Because of no significant differences between CONMIN and BIOORG and between amendments, mean values are shown (n = 24).

 
Total N concentration (g kg–1 aggregate) in the macro- and microaggregates was higher than in the microstructures, and was also not significantly affected by the cropping system (Table 3). Calculated on a non-fractionated soil basis, the small macroaggregates accounted for 40.0%, the large macroaggregates for 26.7%, the microaggregates for 20.0% and the microstructures for 8.1% of total soil N (Table 3). This confirms the importance of the macroaggregates for N storage (Green et al., 2005). Total recovery of soil N in all aggregate fractions was 94.8%. This suggests that some N was lost during aggregate separation, probably by leaching of soluble N compounds during wet sieving and/or gaseous losses during air drying of the aggregate fractions.

Nitrogen Derived from the Labeled Component of the Amendments (Ndflc)
At 112 d after application of the amendments, 15N applied with the amendments was detected in all aggregate fractions (Table 4 ). Between 0.09 and 1.5% of N contained in the aggregates derived from the labeled components of the amendments. The proportion of N derived from the labeled amendments tended to increase with decreasing aggregate size (Table 4). The proportion of N derived from 15N-labeled urine was higher than N derived from labeled feces and mineral fertilizer. For the 15N-labeled urine, the Ndflc was for all aggregate fractions greater in BIOORG than CONMIN soils. The cropping system had no significant effect on N derived from labeled feces or labeled mineral fertilizer. The availability of urinary N to soil microorganisms and crops largely depends on the hydrolysis of urea to ammonium, which was shown to be completed within a few days after addition of urine to soil (Whitehead and Bristow, 1990). This observation was confirmed by a companion study where the 15N enrichment of nitrate showed that a large proportion of soil nitrate derived from urine at 14 d after slurry application (Bosshard, 2007). The higher proportion of urine-derived N in all aggregate fractions compared with the proportion of N derived from the mineral fertilizer, which contained directly available N, may result from the higher amount of total N applied with urine. In addition, urine N was coupled with organic C contained in feces and urine (Ditter et al., 1998). This C probably stimulated microbial activity and might have accelerated microbial immobilization and incorporation of urine N into the aggregates. Thus, the higher proportion derived from urine N in aggregates of BIOORG than CONMIN soils could result from the higher microbial activity of the BIOORG soil. At similar amount of total N applied with feces and mineral fertilizer (Table 2), the proportion of feces-derived N was similar to the proportion of mineral fertilizer-derived N, the only exception being the large macroaggregates of the CONMIN soil where the Ndflc is lower for SlurryF than for MineralN. Of total fecal N, organic N forms in feces were the largest fraction (Table 2) and thereof about 30% were undigested feed N (Bosshard, 2007). Organic N compounds in feces that remained undigested after having passed the animal are supposed to mineralize very slowly in soil (Muñoz et al., 2003; Sørensen and Jensen, 1998). Our results obtained 4 mo after amendment application suggest that the presence of organic compounds does not affect the incorporation of 15N into the aggregates if the 15N is contained in recalcitrant forms.


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

  		Table 4. Proportion of N derived from the labeled component of the amendment (Ndflc) in the aggregate fractions and the non-fractionated soil of the conventional (CONMIN) and the bio-organic (BIOORG) cropping system (CS). Standard deviation is shown in brackets (n = 4).

 
Recovery of Nitrogen-15 Derived from the Labeled Amendments
At 112 d after application, the recovery of urine-derived 15N was in the non-fractionated soil and in aggregate fractions higher for BIOORG than CONMIN soil (Table 5 ). Nearly three times more of urine-derived 15 N was recovered in the large macroaggregates of BIOORG than CONMIN soils (7.6 vs. 2.7%). Also in the small macroaggregates, the recovery was higher for BIOORG than CONMIN. In contrast, the recovery of feces- and mineral fertilizer-derived 15N was not affected by the cropping system.


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

  		Table 5. Recovery of 15N in the aggregate fractions and the non-fractionated soil of the conventional (CONMIN) and the bio-organic (BIOORG) cropping system (CS). Standard deviation is shown in brackets (n = 4).

 
Most of 15N applied with the labeled amendments was recovered in the small macroaggregates (Table 5), confirming their importance as N sink. In spite of their highest Ndflc, the proportion of fertilizer 15N recovered in the microstructures < 50 µm was lowest because they contain only 8% of total soil N (Table 3).

The 15N recovered in the non-fractionated soil accounted on average for 25.1, 20.1, and 21.8% of originally applied 15N in the SlurryU, SlurryF, and MineralN treatment, respectively (Table 5). While recovery of mineral fertilizer N is in the range usually reported for topsoil layers, the recovery of feces is lower (Jensen et al., 1999; Thomsen et al., 1997). The sum of 15N recovered in all aggregate fractions in the SlurryU treatment amounted to 22.4%, in the SlurryF treatment to 15.0% and in the MineralN treatment to 14.8% (Table 5). Hence, between 11 (SlurryU) and 32% (MineralN) of 15N contained in non-fractionated soil was lost during aggregate separation. These losses are higher than losses of about 5% derived from the total N recovery in the aggregates. The greatest proportion of 15N was lost from the MineralN treatment, suggesting that at 4 mo after applications, part of the applied 15N was still present in water soluble form and was leached out of the soil during aggregate separation. In contrast, losses were lowest for the SlurryU treatment of BIOORG. For this soil–amendment combination, microbial immobilization may have increased N retention.

Size Density Fractions
Distribution and Total Nitrogen Concentration
Neither the distribution nor the N and C concentrations and resulting contents in the size density fractions were affected by the cropping system (Tables 6 and 7 ). With a proportion of more than 90% of aggregate dry weight the MF was the dominant fraction in all aggregate fractions (Table 6). Fine and coarse iPOM as well as free LF accounted for <1% of aggregate dry weight of the respective aggregate fractions. However, total N and C concentration (g kg–1 size density fraction dry weight) was for all aggregate fractions higher in free LF and iPOM than in the MF (Table 7), as shown in other studies (Besnard et al., 1996; Pulleman et al., 2005; Six et al., 2001). Coarse iPOM of the large macroaggregates had significantly higher total N and C concentrations than coarse iPOM of the small macroaggregates (Table 7).


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

  		Table 6. Distribution of size density fractions (LF = light fraction; iPOM = intra-aggregate particulate organic matter, MF = mineral-associated organic matter fraction) in soil of the conventional (CONMIN) and the bio-organic (BIOORG) cropping system. Standard deviation is shown in brackets. Because of no significant differences between CONMIN and BIOORG and between amendments, mean values are shown (n = 24).

 

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

  		Table 7. Total N and total C concentration in the size density fractions (LF = light fraction; iPOM = intra-aggregate particulate organic matter, MF = mineral-associated organic matter fraction), and contribution of size density fraction N and C to total soil N and C of the soil of the conventional (CONMIN) and the bio-organic (BIOORG) cropping system. Standard deviation is shown in brackets. Mean values of CONMIN and BIOORG and fertilizer treatments are shown (n = 24).

 
Nitrogen and C contained in all LF and iPOM fractions accounted for 2.7% of total soil N and 4% of total soil C (Table 7). This is relatively low when compared with a number of studies of agricultural soils where on average 14% of total soil N and 19% of total soil C were found in POM fractions (Gregorich et al., 2006). It is, however, comparable with POM recovered from the same soils by Fliessbach and Mäder (2000) although they used a different size density fractionation method. Low contents of LF and iPOM suggest a rapid decomposition of these fractions, thus preventing their accumulation (Wander et al., 2007). Despite having the lowest N and C concentration the MF accounted for about 60% of total soil N due to its big proportion in all aggregate fractions and thus plays an important role for N and C storage. Comparison of aggregate fractions shows that the greatest proportion of N was found in the LF, fine iPOM, and MF of the small macroaggregates.

Only 63.2% of total soil N was recovered in the size density fractions resulting in an overall N recovery of 71.3% as 8.1% of total soil N was recovered in the microstructures. About 10% of total soil N was lost from the small macroaggregates and about 7% each from the large macroaggregates and microaggregates. This suggests that soluble N must have been protected in the aggregates and released during dispersion of the HF and its separation into iPOM and MF. Also only 60% of total soil C was recovered in the size density fractions, suggesting that during dispersion of the soil also significant amount of C was lost.

Carbon/Nitrogen Ratios
The C/N ratios decreased in the order free LF ≥ coarse iPOM > fine iPOM > MF (Table 8 ) suggesting an increased degree of decomposition from free LF to the MF. Decreasing C/N ratios with diminishing particle size were also found by Gregorich et al. (2006), Pulleman et al. (2005) and Six et al. (1998) and might be attributed to a reduction of C content due to exhaustion of easily decomposable OM (Kanazawa and Filip, 1986). The wider C/N ratios of free LF and iPOM compared with the MF suggest that these fractions are richer in recent plant residues (Jastrow, 1996; Willson et al., 2001) whereas the MF is dominated by microbial products (Christensen, 2001). Wheat roots and senescent leaves which might have fallen from wheat plants before harvest and which may subsequently have been incorporated into SOM fractions (straw was removed at harvest) had a much larger C/N ratio (senescent leaves: 60, roots: 47) than the density size fractions indicating that this material has been microbially degraded (Six et al., 2001). The C/N ratio of free LF decreases with decreasing aggregate size confirming that LF related to smaller aggregates is more decomposed (Gregorich et al., 2006).


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

  		Table 8. C/N ratios of the size density fractions (SDF) (LF = light fraction; iPOM = intra-aggregate particulate organic matter, MF = mineral-associated organic matter fraction) of the soil of the conventional (CONMIN) and the bio-organic (BIOORG) cropping system (CS). Standard deviation is shown in brackets. Because of no significant differences between CONMIN and BIOORG and between amendments, mean values are shown (n = 24).

 
Nitrogen Derived from the Labeled Component of the Amendments (Ndflc)
The long-term cropping system did not significantly affect the incorporation of the amendments into the size density fractions. However, the Ndflc was significantly higher for urine than feces or mineral fertilizer. The Ndflc shows that a 15N inflow occurred into each size density fraction during the 4 mo after amendment application. A higher proportion of amendment-derived N was associated with free LF and coarse iPOM than fine iPOM and MF, except for SlurryF where the greatest proportion of amendment-derived N was found in coarse iPOM (Table 9 ). As a part of ruminant feces enters soil as particulate material, and as stabilization of iPOM within aggregates requires time, this comparatively high Ndflc in iPOM suggests that particulate feces N compounds may have been incorporated into aggregates (Aoyama et al., 1999). In contrast, urine- and mineral fertilizer-derived N may have entered the LF through plant debris. Besides these indications, we cannot clearly separate the different pathways such as 15N incorporation through microbial or abiotic processes, or via exudates from wheat roots or wheat residues. Soil bacteria were found to be mostly located in the silt- and clay-fraction where also enzyme activity was higher than in the sand-sized fraction (Blackwood and Paul, 2003; Kandeler et al., 2000; Kirchmann et al., 2004). Therefore, the incorporation of 15N into MF may mainly have been mediated through microbial processes (Christensen, 2001). The importance of abiotic processes, in our soils, was demonstrated by 15N isotopic pool dilution experiments where within the first 30 min after addition of highly enriched (15NH4)2SO4 to sterilized soil more than 30% of the 15NH4+ underwent physicochemical binding processes (data not shown). Therefore, part of NH4 derived from mineral fertilizer and urine may undergo fixation to clay and might through this process end up in the MF.


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

  		Table 9. Proportion of N derived from the labeled component of the amendment (Ndflc) in the size density fractions (SDF) (LF = light fraction; iPOM = intra-aggregate particulate organic matter, MF = mineral-associated organic matter fraction) of the soil of the conventional (CONMIN) and the bio-organic (BIOORG) cropping system (CS). Standard deviation is shown in brackets. Because of no significant differences between CONMIN and BIOORG and between aggregate fractions (AF), mean values are shown (n = 24).

 
Recovery of Nitrogen-15 Derived from the Labeled Amendments
The recovery of amendment 15N in the size density fractions did not differ significantly between CONMIN and BIOORG but was significantly affected by the amendment and differed between aggregate fractions and size density fractions (Table 10 ). Recovery was higher for urine- than feces- or mineral fertilizer-derived 15N, mostly because of greater recovery of urine-derived 15N in the MF. For all amendments, significantly more 15N was recovered in the MF of small macroaggregates than in MF of large macroaggregates and microaggregates (Table 10) because of the large size of this fraction (Table 6). More amendment 15N was recovered in MF than in free LF and iPOM together, showing that MF acted as major N sink for all amendments. This agrees with Gerzabek et al. (2001) who deduced from 15N natural abundance from soils fertilized differently since 1956 that most of applied fertilizer N was stored in the silt-sized fraction.


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

  		Table 10. Recovery of 15N in the size density fractions (SDF) (LF = light fraction; iPOM = intra-aggregate particulate organic matter, MF = mineral-associated organic matter fraction) of the soil of the conventional (CONMIN) and the bio-organic (BIOORG) cropping system (CS). Standard deviation is shown in brackets. Because of no significant difference between CONMIN and BIOORG mean values are shown (n = 8).

 
For SlurryU and MineralN, 15N recovered in free LF generally decreased with decreasing aggregate size while for SlurryF recovery in free LF was highest in the small macroaggregates (Table 10). Provided that free LF associated with large macroaggregates is less protected and has a shorter turnover time than LF associated with small macroaggregates (von Lützow et al., 2007), this observation also supports that urine- and mineral fertilizer-derived 15N may have entered the free LF more recently through plant debris while 15N in LF in small macroaggregates derived from feces may have been deposited earlier, that is, through particulate feces material at amendment application. Recovery was generally low in iPOM fractions, but there was a trend for greater recovery of feces N than urine- and mineral fertilizer- derived N in iPOM.

In total, between 7.4% (SlurryF) and 12.7% (SlurryU) of originally added 15N was recovered in the size density fractions from all aggregate fractions (Table 10). When relating these recoveries to recoveries in aggregate fractions (Table 5), except the microstructures which were not used for size density fractionation, then the estimated 15N losses during size density fractionation ranged between 33 and 47% of 15N in aggregates or between 37 and 55% of 15N contained in non-fractionated soil (Table 5). Estimated 15N losses were higher for SlurryF (47%) than SlurryU (35%) and MineralN (33%). Thus dispersion of aggregates seems to release high amounts of soluble mineral and/or organic N which was previously protected in the aggregate structure and then washed out during wet sieving of the HF. While also mineral 15N may have been lost from MineralN and SlurryU treatments, we assume that mostly organically bound 15N was lost from the SlurryF treatment. Siemens and Kaupenjohann (2002) showed that soluble organic N contributes significantly to N leaching from arable soils. In their study, organic N leaching was higher in manured plots than in plots that received mineral fertilizer N.


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Our first hypothesis that the distribution of N among SOM fractions differs between CONMIN and BIOORG soils due to long-term manure input in BIOORG had to be rejected. At the experimental site of this study, 25 yr of organic farming induced no significant difference in total soil N or in N contained in any aggregate or size density fraction between BIOORG and CONMIN soils. Differences in fertilization strategy are probably overridden by crop rotation (including leys and green manures), residue management and ploughing, which are identical in BIOORG and CONMIN.

In contrast, incorporation of fresh amendment-derived N into different SOM fractions was affected by the amendment, confirming our second hypothesis. A higher proportion of urine-derived N than feces- or mineral fertilizer-derived N was found in the aggregate fractions as well as in free LF, fine iPOM, and MF. These higher proportions of urine-derived N could be attributed to the amount and form of N added with urine and the coupling with C applied with the slurry.

The incorporation of urine-derived N was higher in aggregate fractions of BIOORG than CONMIN, suggesting that higher microbial activity of BIOORG than CONMIN may increase the potential of the BIOORG soil to retain N.

During aggregate separation between 20 and 30% of amendment 15N contained in the non-fractionated soil was lost. During dispersion of aggregates for size density fractionation, another 37 to 55% of 15N contained in non-fractionated soil was lost. This shows the importance of soil aggregation to protect N. As ploughing disrupts soil aggregates, regular ploughing may explain why the suggested potential of BIOORG soil to retain more N cannot be translated into a significant long term effect.


   ACKNOWLEDGMENTS
 
We gratefully acknowledge W. Jossi (ART) and R. Frei (FiBL) for their help in the field work. We warmly thank M. Stocki (University of Saskatchewan, Saskatoon) for providing mass spectrometral analyses and R. Ruh, T. Flura, T. Rösch (all from Group of Plant Nutrition, ETH) for technical assistance in the field and the laboratory. We thank H.-R. Roth (Seminar for statistics, ETH) for advice in statistics and the anonymous reviewers for their helpful comments on our manuscript.


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

Received for publication November 1, 2006.


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




This article has been cited by other articles:


Home page
 Agron. J.Home page
J. Leifeld, R. Reiser, and H.-R. Oberholzer
Consequences of Conventional versus Organic farming on Soil Carbon: Results from a 27-Year Field Experiment
Agron. J., August 31, 2009; 101(5): 1204 - 1218.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bosshard, C.
Right arrow Articles by Oberson, A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Bosshard, C.
Right arrow Articles by Oberson, A.
Agricola
Right arrow Articles by Bosshard, C.
Right arrow Articles by Oberson, A.
Related Collections
Right arrow Soil Methods/Instrumentation
Right arrow Structure and Properties
Right arrow Agroforestry

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