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


     

Published online 1 January 2007
Published in Soil Sci Soc Am J 71:163-170 (2007)
DOI: 10.2136/sssaj2006.0242
© 2007 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 Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yang, Z.
Right arrow Articles by Riley, H.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Yang, Z.
Right arrow Articles by Riley, H.
Agricola
Right arrow Articles by Yang, Z.
Right arrow Articles by Riley, H.
Related Collections
Right arrow Soil Fertility and Productivity
Right arrow Nutrient Management

SOIL FERTILITY & PLANT NUTRITION

Aggregate Associated Sulfur Fractions in Long-Term (>80 Years) Fertilized Soils

Zhihui Yang and Bal Ram Singh*

Dep. of Plant and Environmental Sciences, Norwegian Univ. of Life Sciences, P.O. Box 5003, N-1432 Aas, Norway

Sissel Hansen

Bioforsk-Norwegian Agric. & Environ.Res. Inst., Organic Food and Farming Division, N-6630 Tingvoll, Norway

Zhengyi Hu

Institute of Soil Science, Chinese Academy of Sciences, No. 71, E. Beijing Rd., P.O. Box 821, Nanjing, 210008, P.R.China

Hugh Riley

Bioforsk-Norwegian Agric. & Environ. Res. Inst., Arable Crops Division, N-2350 Kisevegen 337, Nes På Hedmark, Norway

* Corresponding author (balram.singh{at}umb.no).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Understanding soil sulfur pools and associated aggregates S fractions can provide a platform for monitoring S dynamics in soils. A long-term experiment established in 1922 on an Aquic Eutrocryepts in South-eastern Norway was chosen to investigate the effects of long-term fertilization on S fractions in bulk soil and those associated with aggregates. Chloroform fumigation-extraction was used to determine Microbial biomass S (MBS) and the wet chemical analysis method was used to fractionate soil S into ester S (hydriodic acid reducible S), carbon-bonded S (Raney nickel reducible S) and residual S (Raney nickel non-reducible S). High farmyard manure (FYM) application resulted in higher MBS in bulk soil than nitrogen + potassium (NK) application, but it did not differ significantly from the control. Application of FYM at 60 Mg ha–1 resulted into accumulation of total S, total organic S and carbon-bonded S in bulk soils, while mineral fertilizer (nitrogen+phosphorus+potassium+sulfur [NPKS] and NK) and the medium rate of FYM did not increase the accumulation of total S and organic S fractions. The macroaggregate sizes (>2 and 1–2 mm) and the finest aggregate size (<0.106 mm) showed significantly greater total S concentration than other aggregate sizes. Ester S and residual S were predominant organic S fractions and they accounted for 39 to 52% and 38 to 51% of the organic S, respectively. The macroaggregate sizes (>2 and 1–2 mm) contained the highest ester S, but microaggregates (<0.106 mm) exhibited higher carbon-bonded S and residual S than other aggregates. In conclusion, the accumulation of S was dependent on fertilizer type, the rate of FYM application and aggregate sizes.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The main source of S for plants originates from soil. Plants can only take up sulfate or a small amount of organic S from soils. Most S is accumulated as organic S. Previous studies have shown that long-term cropping with mineral fertilizer and FYM increased soil organic and inorganic S and S mineralization in soils (Eriksen and Mortensen, 1999; Reddy et al., 2002). Other studies have shown that continuous application of superphosphate enhanced soil organic matter and total or organic S accumulations in soils under grazed pasture (Nguyen and Goh, 1990). However, Knights et al. (2000) reported that large S inputs from atmospheric deposition and from sulfate fertilizer did not accumulate as organic and inorganic S in arable soils, but organic S obviously increased after FYM supply. Most organic S accumulated from fertilizer or other input sources were present in carbon-bonded S fractions, while ester S accounted for 18 to 41% of organic S (Strickland et al., 1987; Knights et al., 2000). A small amount of sulfate existed in soil, indicating that other S sources must be involved to meet crop S requirements. Although both carbon-bonded S and ester S can release sulfate through biological and biochemical processes involving microbial and enzymatic activity (McGill and Cole, 1981), pot experiments demonstrated that ester S depleted more than carbon-bonded S (Shan et al., 1997; Hu et al., 2002) and they suggested that ester S had higher bioavailability. Thus, the composition of soil S pools is important to assess the S supply capacity to crops under field conditions.

A long-term fertilization trial in southeastern Norway was established in 1922 to investigate the effects of FYM and mineral fertilizer on crop yield, nutrient supply and changes in soil properties over long periods of time. The effect of fertilization on crop yield, soil fertility, nutrient balances (N, P, and K) and soil organic matter has been well documented by Riley (2006). However, no information is available from this experiment on soil S pools and S fractions, which can provide a platform for monitoring S dynamics in soils.

Sulfur, like N, is one of the components in soil organic matter. Soil organic matter plays an important role in nutrient availability and soil aggregate formation. An aggregation model has been proposed that primary mineral particles coagulate with plant roots, polysaccharides and humic substances to form microaggregates and that, in turn, microoaggregates coagulate to form macroaggregates (Tisdall and Oades, 1982). Previous studies demonstrated that soil aggregation was closely related to soil organic matter since macroaggregates contained more organic C, N, and P than microaggregates (Elliott,1986). Elliott (1986) suggested that the loss of organic C and N resulting from cultivation comes mainly from the organic material that binds individual microaggregtaes into macroaggregates, not from organic matter within microaggregates. Manure application has been reported to increase the proportion of large-size aggregates (Mazurak et al., 1977) and water-stable aggregates (Sun et al., 1995), but only a short-term effect was observed by Debosz et al. (2002).

The formation of stable macroaggregates is strongly linked to soil organic matter dynamics as well as nutrient supply. Elliott (1986) reported that more labile and readily mineralized soil organic matter was associated with macroaggregates than with microaggregates and was the dominant source of nutrients lost during cultivation. From the viewpoint of nutrient supply, macroaggregates may be the primary sulfur source for crops. Therefore, knowledge of the association of S in organic materials and in different sizes of soil aggregates may provide a better understanding of the amount, forms and biochemical transformation of S in the ecosystem. However, information on S fractions associated with soil aggregates and organic matter is rather scanty. The objectives of this study were to investigate: (i) the effects of long-term application of mineral fertilizer and organic manure on the soil aggregate associated S fractions; (ii) the amount and composition of different S pools in soils.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Description of the Experimental Site
A long-term fertilizer experiment was established in 1922 on an Aquic Eutrocryepts at Møystad, Southeast Norway (60° 47' N 11° 11E') (Soil Survey Staff, 2003). It contained 52% sand, 34% silt, and 14% clay. The mean rainfalls were 330 mm from May to September and 270 mm from October to April over the period of 1983–2003. The mean air temperatures were 12.6 and –1.3°C in the above two seasons. Among a large number of experimental treatments, only five contrasting fertilizer treatments were selected for the present study (Table 1) and they are: control (without fertilizer), NPKS, NK, medium amount of FYM (FYMmed) and high amount of FYM (FYMhigh). The experimental design was randomized complete block design. Each treatment was replicated three times. The plot size was 7.7 x 3.9 m (30 m2). Plowing has been kept to a minimum to reduce carry-over of soil from plot to plot. A seven-course rotation (oats [Avena sativa L.], potatoes [Solonum tubersom L.], wheat [Triticum aestivum L.), under sown barley [Hordeum vulgare L.], 3-yr ley) was employed. From the beginning of the experiment (1922) to 1982, calcium nitrate, superphosphate, and potassium chloride have been used as N, P, and K sources in mineral fertilizer treatments (NPKS and NK). From 1983, the levels of FYM and NPKS fertilizer were increased so as to reflect modern practice (Table 1). In the NPKS treatment, compound fertilizer (N/P/K/S 14:6:16:2.7) was used from 1983 to 1992, followed by compound fertilizer 11:5:17:9.0 until 2000 and compound fertilizer 11:5:18:9.5 thereafter. All these fertilizers were from Norsk Hydro (Norsk Hydro As, Oslo, Norway). The FYM was applied mostly in the solid form (not slurry) in spring, and composted cattle manure was applied in all years since 1983. Application rates for medium and high amounts of FYM were 40 and 60 Mg manure ha–1. The composted cattle manure contained 0.49% of total S. No information on S content of FYM applied before 1982 is available.


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

  		Table 1. Fertilizer rates under the long-term fertilizer experiment.{dagger}

 
Soil Sampling
Soil samples were collected from plow depth (0–20 cm) in autumn 2003, when the rotation was ley. Ten soil cores were taken randomly in each plot and mixed thoroughly to make a bulk soil sample. At the same time, 1 L of undisturbed soil core was taken from all the selected treatments and air-dried for aggregate separation. Bulk soil samples for chemical analysis were air-dried and passed through a 2-mm sieve for inorganic S determination. Ten grams of sieved bulk soil was ground again with a grill machine before determining total S.

Dry Aggregate Separation
Undisturbed soil cores were air-dried and then used for dry aggregate analysis. The apparatus consisted of a set of four coarse sieves with a wooden frame having the length of 444 breadth of 247 mm and height of 60 mm. The set of sieves had 20-, 6-, 2-, and 0.6-mm diameter openings and thus yielded five aggregate size fractions of >20, 6 to 20, 2 to 6, 0.6 to 2, and <0.6 mm. The sieves were stacked with largest opening (20 mm) on the top and the smallest (0.6 mm) in the bottom. The air dried undisturbed soil was placed on the uppermost sieve and the set of stacked sieves was mounted on a motorized sieving device machine, where the set moved back and forth with a speed of 240 rpm for 12 min. After this time aggregate fraction retained on each sieve was collected and weighed. Each aggregate fraction was calculated as percent of the total soil used.

Wet Aggregate Separation
Water-stable aggregate separation was conduced with a sieving apparatus. Undisturbed soil cores were passed through a 6-mm sieve with gentle crushing and shaking. Three hundred grams of soil aggregates of 6 mm were passed through a nest of sieves with 2-, 1-, 0.5-, 0.25-, 0.106-mm openings. The stacked sieves were placed in a bucket and connected with a motor. The water level was brought to the base of the top sieve. The stacked sieves moved up and down (30 times min–1) for 30 min. During the sieving procedure, the water level was adjusted to cover all soil aggregates. After wet sieving, the nest of sieves was removed and different sizes of aggregates (>2, 1–2, 0.5–1, 0.25–0.5, 0.106–0.25, and <0.106 mm) were collected. The floating organic material, including crop roots and residues (>2 mm), was decanted as it was not considered as soil organic matter. The wet aggregates were dried at 40°C in an oven and then weighed. Finally, all samples of different aggregates were ground through a 0.106-mm sieve before chemical analysis.

Sulfur Fractionation
Total S content in bulk soils and aggregates was determined by dry combustion (HORIBA Carbon/Sulfur Analyzer; EMAL Technology, Kyoto, Japan). Inorganic sulfate was extracted by 10 mM Ca (H2PO4)2 in a soil to solution ratio of 1:5. Samples were shaken on an end-over-end shaker at 400 rev min–1 for 60 min at room temperature. The suspension was filtered through a Schleicher & Schuell filter paper, )GMBH Dassel Germany). Sulfate content in extracts was determined with ICP–AES (Thermo Jarrell Ash Corp. a subsidiary of Thermo Instrument Systems, Inc., Franklin, MA). Hydriodic reducible S was determined by reduction with a mixture of hydriodic acid, hypophosphoric acid, and formic acid (Johnson and Nishita, 1952). Hydrogen sulfide evolved was absorbed in NaOAc-Zn (OAc)2 solution, and then sulfide was measured as the methylene blue complex with a spectrophotometer at 670 nm. Carbon-bonded S was reduced to H2S with Raney Ni and the sulfide formed was measured colorimetrically as described above for Hydroiodic reducible S (Lowe and DeLong, 1963). Ester S was calculated from the differences between HI-reducible S and inorganic S. The difference between total S and HI-reducible sulfur + carbon-bonded sulfur was regarded as residual S.

Total S amount in bulk soils was calculated according to following equation:

Formula

Bulk density of different treatments was measured using core method (Blake and Hartge, 1986).

Microbial Biomass Sulfur
Microbial biomass S content was measured with a fumigation-extraction method in the moist bulk soil samples and rewetted (60% water content) dry aggregates (Wu et al., 1994). Five grams of soil (dry basis) were put into a 50-mL centrifuge tube and 1 mL of CHCl3 was added. The closed tubes were agitated to spread the soil over the inner surface of the tube, to expose the maximum surface area to chloroform vapor. Soil samples were kept in a fume hood for 24 h in the dark at room temperature before opening the lids to allow evaporation under the fume hood overnight. Another 5 g of soil was used for controls (non-fumigated). The fumigated and non-fumigated soils (control) were extracted using 25 mL of 10 mM L–1 Ca (H2PO4)2. Suspensions were shaken on an end-over-end shaker at 400 rev min–1 for 60 min. Extracts were centrifuged (12100 x g; 10000 rev min–1) for 5 min and then filtered using Schleicher & Schuell filter paper. The S content in filtrated solution was determined with ICP–AES. The biomass S was calculated from the following equation:

Formula
where Fs is the sulfate difference between fumigated soil and non-fumigated soil; Ks is the conversion factor; and Bs is biomass S.

Since only part of the biomass S is recovered as sulfate with the fumigation method, conversion factors have been investigated by many workers (Wu et al., 1994; Castellano and Dick, 1991). The values have varied with soil type and extractants. Here we used 0.39 for Ca (H2PO4)2 in biomass calibration (Castellano and Dick, 1991).

Statistical Analysis
An analysis of variance was performed by the General Linear Model (GLM) procedure of the SAS package (SAS Institute, 2001). Significant differences between means for treatments were compared with LSD at P < 0.05. The significance level in tables and figures is shown by lowercase letters for aggregate sizes and by uppercase letters for fertilizer treatments.


   RESULTS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Microbial Biomass Sulfur
Microbial biomass S concentration ranged from 5.1 to 8.6 mg kg–1, averaging 6.8 mg kg–1 and accounting for 1.2% of the total S in bulk soils. This result was consistent with those reported in the literature (Banerjee and Chapman, 1996; Wu et al., 1994). They estimated that soil microbial biomass generally accounted for 1.5 to 5% of the total S. The bulk soils with high amount of FYM (FYMhigh) applied for more than 80 yr showed a significantly (p < 0.05) higher MBS than that applied with NK only (Fig. 1A ). However, the MBS value in FYM treatment did not differ significantly from the control or mineral fertilizer treated soils. Microbial biomass S content also differed among dry aggregates (Fig. 1B). Averaged for all treatments, the aggregate size of 6 to 20 and 2 to 6 mm contained significantly higher amount of MBS than the aggregates of <0.106-mm size.


Figure 1
View larger version (36K):
[in this window]
[in a new window]

  		Fig. 1. Microbial biomass S in bulk soil and in dry aggregates under different fertilizer treatments (FYMmed, medium amount of farmyard manure; FYMhigh, high amount of farmyard manure; MBS, microbial biomass S. Values followed by different letters are significantly different at p < 0.05. Error bars are standard deviations).

 
Total Soil Sulfur
Total Sulfur in Bulk Soils
The total S in bulk soils ranged from 1392 kg ha–1 in the control (without fertilizer) to 1808 kg ha–1 with the high amount of FYM application (Table 2). The plots receiving either mineral fertilizer (NPKS and NK) or FYMmed for more than 80 yr didn't result in significant accumulation of total S compared with the control treatment. Moreover, there was no significant difference of total S between NPKS and NK treatments, although the former treatment received a large amount of S annually (Table 1). In contrast, the total S in bulk soils of the FYMhigh treatment was significantly higher by 402 to 416 kg ha–1 than control and NK treatments.


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

  		Table 2. Total S, S fractions and percentage of total S in bulk soils under different fertilizer treatments.{dagger}

 
Sulfur in Dry Aggregates
The mean S concentration for all treatments in dry aggregate sizes >20, 6 to 20, 2 to 6, 0.6 to 2 and <0.6 mm were 539, 514, 582, 556, and 500 mg kg–1, respectively (Fig. 2A ). A significantly (p < 0.05) higher S concentration was found in aggregates of 2 to 6 mm than aggregates of 6 to 20 and <0.6 mm. The FYMhigh application significantly (p < 0.05) increased the S concentration in most aggregates as compared with control treatment except for 0.6- to 2-mm aggregates. The S concentration in the NKPS treatment was significantly (p < 0.05) higher only in aggregates of <0.6 and 2 to 6 mm as compared with control, but it did not differ significantly in any of the aggregates of the NK treatment.


Figure 2
View larger version (58K):
[in this window]
[in a new window]

  		Fig. 2. Total S concentration and distribution percentage in dry aggregate size groups. (FYMmed, medium amount of farmyard manure; FYMhigh, high amount of farmyard manure. Values followed by different lowercase letters are significantly different for aggregates at p < 0.05. Values followed by different uppercase letters are significantly different for fertilizer treatments in each aggregate size at p < 0.05).

 
Based on the weight and S concentration in dry aggregate fractions, the distribution of total S in different aggregate sizes was calculated. The distribution percentages of total S in aggregate sizes of >20, 6 to 20, 2 to 6, 0.6 to 2, and <0.6 mm among the fertilizer treatments were 4 to 9, 16 to 26, 25 to 31, 29 to 32, and 12 to 20%, respectively (Fig. 2B). The aggregates 0.6 to 2 and 2 to 6 mm in most fertilizer treatments contributed significantly (p < 0.05) larger proportion of total S compared with other aggregate sizes, while >20-mm aggregate size had the smallest proportion of total S. The higher S concentration and relatively larger weight of 2 to 6 and 0.6 to 2 mm aggregates led to more total S being found in these aggregate fractions. Long-term application of FYMhigh and NPKS increased the S accumulation in dry aggregates of <0.6 and 2 to 6 mm by 7 and 5%, respectively, as compared with the control. However, NPKS application correspondingly led to decreased S in larger aggregates of >20 and 6 to 20 mm as compared with control.

Sulfur in Water-Stable Aggregates
On average, the mean S concentrations for all treatments in water-stable aggregates >2, 1 to 2, 0.5 to 1, 0.25 to 0.5, 0.106 to 0.25, and <0.106 mm were 612, 591, 459, 472, 441, and 625 mg kg–1, respectively. The macroaggregates (>2 and 1–2 mm) and the microaggregates (<0.106 mm) contained significantly (p < 0.05) higher S than other aggregate sizes (Fig. 3A ). The differences in S concentration among aggregate sizes, 0.5 to 1, 0.25 to 0.5, and 0.106 to 0.25 mm, were not pronounced. The FYMhigh application significantly (p < 0.05) increased S concentration in aggregates of <0.106, 0.106 to 0.25, 0.25 to 0.5, and >2 mm as compared with control. The influence of NPKS, NK, and FYMmed application on S concentration among different sizes of water-stable aggregates was not found to be significant as compared with control.


Figure 3
View larger version (56K):
[in this window]
[in a new window]

  		Fig. 3. Total S concentration and distribution percentage in water stable aggregate size groups. (FYMmed, medium amount of farmyard manure; FYMhigh, high amount of farmyard manure. Values followed by different lowercase letters are significantly different for aggregates at p < 0.05. Values followed by different uppercase letters are significantly different for fertilizer treatments in each aggregate size at p < 0.05).

 
The distribution of total S in different water-stable aggregates showed that the highest percentage of total S (31%) was found in medium-sized aggregates (0.5–1 mm), while the aggregate size of 0.106 to 0.25 and <0.106 mm showed the lowest total S (Fig. 3B). In aggregates > 2 mm, the total S in FYM treatments was significantly (p < 0.05) higher than in mineral fertilizers or control treatments. However, aggregates of 0.5 to 1 mm in the FYM treatments showed significantly (p < 0.05) lower total S than in other treatments. The control treatment had a significantly (p < 0.05) higher total S in aggregates of 1 to 2 mm than did by fertilizer treatments.

Inorganic Sulfur
The long-term FYM and NK application did not significantly increase inorganic S in the bulk soil as compared with control. It was, however, noticed that the inorganic S in the NPKS treatment was increased by four times compared with the control treatment and it was significantly (p < 0.05) higher than all other treatments (Table 2) Among aggregate fractions, the differences were not pronounced except that the finest aggregate size (<0.106 mm) had a significantly (p < 0.05) higher value of 15 mg kg–1 (Table 3) than the other size fractions.


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

  		Table 3. Inorganic S content in water stable aggregates under different fertilizer treatments.{dagger}

 
Organic Sulfur Fractions
Organic S in bulk soils ranged from 477 mg kg –1 in the control treatment to 651 mg kg–1 in the FYMhigh treatment, which accounted for 96.7 to 99.6% of the total S in the soil (Table 2). Ester S contributed 43 to 52% of total S in bulk soils (Table 2). The effect of long-term applications of FYM and mineral fertilizer on ester S in bulk soil was not significant as compared to control. However, NPKS application increased ester S by 276 kg ha–1 more than did NK application. On average, ester S concentration in all aggregate fractions for all treatments was 232 mg kg–1, accounting for 44% of the total organic S. Ester S concentration in water stable macroaggregates (>2 and 1–2 mm), averaged for all treatments, was significantly (p < 0.05) higher than other aggregates (Table 4). In other aggregate sizes only minor differences in ester S were observed in the present study. The effects of either mineral fertilizers or FYM application on ester S in all water-stable aggregates were not significant.


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

  		Table 4. Organic Sulfur fractions in water stable aggregates size under different fertilizer treatments.{dagger}

 
Carbon-bonded S varied from 81 to 184 kg ha–1 in bulk soils, which accounted for 6 to 10% of the total S in the soil (Table 2). The FYM application (both FYMhigh and FYMmed) significantly (p < 0.05) increased the accumulation of carbon-bonded S in bulk soil by 58 to 103 kg ha–1 compared with control (Table 2). Moreover, the accumulation of carbon-bonded S increased with increasing amount of FYM application because FYMhigh treatment showed higher value than FYMmed. The long-term application of mineral fertilizer (NPKS and NK) didn't result in the accumulation of carbon-bonded S. On average, the significantly (p < 0.05) highest carbon-bonded S, 61.3 mg kg–1, was found in the finest water-stable microaggregates (<0.106 mm), while the lowest value was found in the macroaggregates (>2 mm) (Table 4). Similarly in most fertilizer treatments except FYMmed, microaggregates (<0.106 mm) showed significantly (p < 0.05) higher carbon-bonded S than macroaggregates (>2 mm). The FYMhigh application significantly (p < 0.05) increased carbon-bonded S in most aggregates except for >2 mm as compared with control. The other fertilizer treatments (mineral fertilizer and FYMmed) did not show any significant effect on carbon-bonded S in all water stable aggregates.

Generally, carbon-bonded S is calculated from the difference between total organic S and ester S. However, in the present study, both ester S and carbon-bonded S were determined separately. The sum of the two S fractions was not equal to total organic S. This implied that another S fraction could be included during the S fractionation procedure, and this was defined as residual S (Lowe, 1964). The mean residual S in bulk soil was 710 kg ha–1, accounting for 45.3% of total organic S, but the amount of residual S among fertilizer treatments did not differ significantly (Table 2). The mean concentration of residual S in all water stable aggregate fractions was 243 mg kg–1. Averaged for all treatments, microaggregates (<0.106 mm) showed a significantly (p < 0.05) higher concentration of residual S than other aggregate size (Table 4). Generally, in FYM treatments, macroaggregates (>2 mm) and the finest aggregates (<0.106 mm) contained relatively higher residual S than other fractions. There was no significant difference in residual S among water-stable aggregates in NPKS treatment. However, FYMhigh application resulted in significant (p < 0.05) accumulation of residual S in aggregates of >2, 0.25 to 0.5, and <0.106 mm as compared with the control and NK treatments.


   DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Our study shows that the accumulation of total S and organic S fractions by long-term fertilization practices was dependent on fertilizer types (mineral or FYM) and their amount. The application of NPKS did not result in significant accumulation of either total S or organic S fractions although the S input from this fertilizer was higher than input by FYM application. This could be due to inorganic sulfate from NPKS being susceptible to leaching from the surface soils. These results are in contrast to those from New Zealand where an accumulation of soil total S after 37 yr of superphosphate application to a pasture field was observed (Nguyen and Goh, 1990). Eriksen and Mortensen (1999) also found that the long-term fertilization with animal manure and NPKS fertilizer increased soil organic S in a sandy soil compared with unfertilized soil. However, Knights et al. (2001) reported that there was no increase in total S after 150 yr supply of mineral fertilizer (NPKMgS). They further reported that accumulation of organic S occurred only when organic C accumulated, and particularly in treatments receiving long-term application of organic manure. Probably, a significant increase of total S and organic S caused by the high amount of FYM application in the present study was contributed by the organic C accumulation because a significant increase of soil organic C was observed in the plots receiving high amount of FYM (data not shown here). However the medium amount of FYM application (FYMmed) did not increase total S pools because S input from this treatment was only two thirds of FYMhigh treatment from 1983 to 2003.

Organic S was the main S fraction in the soils of the present study (Table 2), which has also been found in many other studies (Solomon et al., 2001; Chapman, 2001; Reddy et al.,2002). Our study shows that the long-term application of FYM resulted in the accumulation of carbon-bonded S compared with the control (Table 2). Previous studies have shown that sulfate from atmospheric deposition and fertilizers was incorporated into both carbon-bonded S and ester S, resulting in carbon-bonded S being the major form (Strickland et al., 1987; Knights et al., 2000). Farmyard manure in the present study contained about 0.49% of total S. According to Pedersen et al. (1998) total S in organic manure consists of 40% carbon-bonded S, 20% sulphide S, and about 40% other organic and inorganic sulfate S. Leaf litter, root input and microbial protein synthesis are the main sources of carbon-bonded S (Saggar et al., 1998). It was noticed that ester S either in bulk soils or in water-stable aggregates was not influenced by either FYM or mineral fertilizer application in the present study. Unlike carbon-bonded S, ester S is active S form, which is susceptible to conversion into sulfate via biochemical mineralization. This inter-transformation is strongly controlled by the supply of and the need for S released (McGill and Cole, 1981). Under condition of low sulfate, such as in the soils used in this study [2–15 mg kg–1 SO42––S extracted by 10 mM L–1 Ca(H2PO4)2], this biochemical transformation procedure probably did not lead to evident accumulation of ester S.

Generally, organic S is divided into ester S and carbon-bonded S. Ester S mainly consists of sulfate esters (–C–O–S). This compound can be reduced by hydriodic acid, while other S forms including –SO3H (sulfonic acid), –SO2 (sulfone), –SH (sulfhydryl), –C–S–S–C (disulfide), and heterocyclic S cannot be reduced by hydriodic acid and are defined as carbon-bonded S. The traditional fractionation method for soil S was to determine ester S by hydriodic acid reducible S and then carbon-bonded S was calculated from the difference between total organic S and ester S. The assumption that all organic S in soils that is not hydriodic acid reducible consists of carbon-bonded S is, however, probably oversimplified and it is not justified in view of the existence in soils of unidentified S differing from ester S and carbon-bonded S. Freney et al. (1970) determined the carbon-bonded S by Raney nickel regent. They found that there was a discrepancy between the determined and the calculated values (total S-HI reducible S) for carbon-bonded S. The Raney nickel method mainly reduces the S component in amino acids and it would not reduce the carbon-bonded S of aliphatic sulfones and sulfonic acids (Freney et al., 1970). Lowe (1964) first defined the part of organic S not reduced by the Raney nickel method as residual S. Although both Raney nickel reducible S and non-reducible S (residual S) are carbon-bonded S, their bioavaibility differ (Shan et al., 1997; Hu et al., 2002). Therefore, fractionation of soil carbon-bonded S into Raney nickel reducible S and residual S can provide a better understanding of different S pools. Carbon-bonded S in our study accounted for only 6 to 10% of total S (Table 2). This value is lower than that reported by many workers (Solomon et al., 2001; Chapman, 2001) because their results included both Raney nickel reducible S and non-reducible S (residual S) as carbon-bonded S. Therefore, ester S and Raney nickel non-reducible S (residual S) were the dominant S fractions in the present study (Table 2). Recent studies showed that X-ray absorption near-edge structure (XANES) spectroscopy provided better S fractionation (Zhao et al., 2006; Solomon et al., 2005; Prietzel et al., 2003), but, Solomon et al. (2005) did not find a close correlation between the results of wet-chemical procedure and XANES spectroscopy (ester S in the bulk soils and the humic substance extracts from XANES versus hydriodic acid reducible S, r = 0.27 and 0. 39, respectively). They further pointed out that the proportion of ester S measured by wet-chemical analysis was generally less than the proportion of ester S determined by XANES spectroscopy. Moreover, the bioavailability of S forms fractionated by XANES remains obscure.

Microbial biomass S is a main factor controlling S transformation rate. In general, microorganisms incorporate C, N, and P into their cell structure. The magnitude of MBS is influenced by soil organic input as well as by other nutrients. Saggar et al. (1981) found that cellulose addition increased microbial S. Microbial biomass S was also shown to be increased by application of municipal compost (Perucci, 1990). Wu et al. (1993) reported that 33% of S in oil-seed rape residues and 20% of S in barley straw was converted into MBS after 15 d at 25°C. In the present study, the high amount of FYM application showed higher MBS than NK application. Probably, P deficiency or imbalance in NK plots limited microbial growth, and thus S immobilization. However, this trend was not found in aggregates (Fig. 1B). The discrepancy could be explained by the different procedure of sample handling between bulk soil and aggregates. In our study, moist bulk soil samples were kept under 4°C, and MBS was determined using moist soil. As for MBS in dry aggregates, rewetted dry aggregates were used, in which microbial activity could be different from that under field conditions. Generally, the smallest dry aggregates, <0.6 mm, showed a lower MBS than large aggregates of 2 to 6 and 6 to 20 mm (Fig. 1B). In previous studies, macroaggregtes showed higher microbial biomass than microaggregates (Hernández-Hernández and López-Hernández, 2002) because macroaggregate formation was closed to microbial activities (Gupta and Germida, 1988).

An increased total S concentration in water stable aggregates > 2 and 1–2 mm was observed in this study (Fig. 3). Aggregate formation and their stability recently have been closely related to soil organic C content (Green et al., 2005). During the aggregate formation, higher S added through atmospheric deposition, fertilizer and plant residues could be incorporated into soil organic matter along with C, explaining the high S content of these aggregates (Fig. 3). In the present study, the finest water stable aggregates of <0.106 mm also contained higher total S concentration than medium size aggregates (Fig. 3). It was found by other researchers that manure application leads to higher concentrations of organic C, N, and P in microaggregates < 0.1 mm than in macroaggreagtes (He et al., 1995; Maguire et al., 1998). In general, aggregates of <0.106 mm contained a large part of sand. During the procedure of aggregate fractionation in the present study, the sand was not separated from the finest aggregate size (<0.106 mm), probably resulting in high total S content in this fraction.

According to the aggregation model of Tisdall and Oades (1982), the binding of macroaggregates (>0.25 mm) is more susceptible to management practices, but binding of microaggregates (<0.25 mm) seems to be relatively stable. Macroaggregates form also from microaggregates by binding agents such as polysaccharides, hemicellulose, and other polymers. Persistent binding agent, aromatic humic material, is involved in the formation and stability of microaggregates (Tisdall and Oades,1982). Bettany et al. (1979) suggested that carbon-bonded S is primarily associated with the aromatic core of humic acids. The present study also confirmed that the finest microaggregates < 0.106 mm contained higher carbon-bonded S and residual S than other aggregate sizes (Table 4). McGill & Cole (1981) suggested that ester S is a transitory form. The enhanced microbial activities associated with macroaggregates could lead to accelerated transformation of carbon-bonded S to ester S, before S is released as inorganic S, resulting in higher ester S in macroaggregates >2 or 1 to 2 mm (Table 4). These results suggested that macroaggregate associated S could be the predominantly available S pool for plants.


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Long-term fertilization (>80 yr) influenced soil S pools and aggregate associated S fractions in the soil. Organic S accounted for more than 97% of total S and inorganic S only accounted for 0.4 to 3%. In organic S, ester S and residual S were predominant S fractions. Long-term application of high amount of FYM (60 mg ha–1) resulted in high total S, total organic S, and carbon-bonded S accumulation in bulk soils as compared with the control. Similar effects of mineral fertilizer (NPKS and NK) and a medium amount of FYM were not seen. The results implied that the accumulation of total S and organic S fractions in bulk soil was dependent on fertilizer type and the rate of FYM application.

The dry aggregates 6 to 20 and 2 to 6 mm contained the higher MBS than small aggregates < 0.106 mm. Water stable macroaggregates 1 to 2 and > 2mm and microaggregates < 0.106 mm contained higher total S and total organic S concentrations than other aggregate sizes. Generally, high ester S concentration existed in macroaggregates 1 to 2 and > 2 mm, while high carbon-bonded S and residual S concentrations in microaggregates < 0.106 mm. The results implied that formation of macroaggregates from microaggregates resulted in an accumulation of ester S and macroaggregate associated S could be the predominantly available S pool for plants.


   ACKNOWLEDGMENTS
 
The study was supported by the Research Council of Norway through a collaborative project with the Norwegian Centre for Ecological Agriculture and this assistance is gratefully acknowledged. The senior author is also grateful to the Loan Bank of Norway for providing the fellowship during the period of this study.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Abbreviations: FYM, farmyard manure; FYMmed, medium amount of farmyard manure; FYMhigh, high amount of farmyard manure; MBS, microbial biomass S; NK, nitrogen + potassium application; NPKS, nitrogen + phosphorus + potassium + sulfur application; XANES, X-ray adsorption near-edge structure spectroscopy.

Received for publication June 23, 2006.


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





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yang, Z.
Right arrow Articles by Riley, H.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Yang, Z.
Right arrow Articles by Riley, H.
Agricola
Right arrow Articles by Yang, Z.
Right arrow Articles by Riley, H.
Related Collections
Right arrow Soil Fertility and Productivity
Right arrow Nutrient Management

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