SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

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 HighWire
Right arrow Citing Articles via ISI Web of Science (18)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Devêvre, O. C.
Right arrow Articles by Horwáth, W. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Devêvre, O. C.
Right arrow Articles by Horwáth, W. R.
Agricola
Right arrow Articles by Devêvre, O. C.
Right arrow Articles by Horwáth, W. R.
Related Collections
Right arrow Rice
Right arrow Water Stress
Right arrow Other Cropping Systems
Soil Science Society of America Journal 65:499-510 (2001)
© 2001 Soil Science Society of America

DIVISION S-8-NUTRIENT MANAGEMENT & SOIL & PLANT ANALYSIS

Stabilization of Fertilizer Nitrogen-15 into Humic Substances in Aerobic vs. Waterlogged Soil Following Straw Incorporation

Olivier C. Devêvre and William R. Horwáth

University of California, Department of Land, Air and Water Resources, One Shields Ave., Davis, CA 95616-8627

Corresponding author (wrhorwath{at}ucdavis.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
This study was undertaken to investigate and quantify the interactive effects of flooding and straw incorporation on key microbial processes, principally stabilization of fertilizer N into various soil organic matter (SOM) pools. The fate of fertilizer 15N in a paddy soil was examined at 5, 15, and 25°C, with and without rice (Oryza sativa L.) straw added, and under flooded and nonflooded conditions. After a 160-d incubation, three fractions of the SOM were separated and defined as directly alkali-extractable humic substances (DAEHS), reducible metal-bound humic substances (RMBHS), and non-alkali-extractable organic matter (NAEOM). The DAEHS had the highest percentage, up to 50%, of fertilizer 15N recovered at 160 d, indicating that this SOM fraction was the most dynamic fraction of the SOM. On the other hand, the RMBHS is considered the least dynamic pool, containing up to 12% fertilizer 15N after 160 d. The NAEOM was surprisingly highly enriched, up to 28% fertilizer 15N, and showed a significant treatment effect, suggesting that some active components of N cycling were present in this SOM fraction. The addition of rice straw increased the recovery of fertilizer 15N in the above SOM fractions. Flooding significantly reduced the stabilization of fertilizer N compared with the nonflooded treatment. Indices of recalcitrance of the stabilized N confirm that the soil N supply capacity does not decrease with flooding. The total alkali-extractable organic matter (AEOM = DAEHS + RMBHS), as the NAEOM, appears to be a complex and dynamic mixture of potentially mineralizable and recalcitrant forms of N. Our data show that long-term N availability and stabilization into humic fractions is a function of rice residue input and temperature; however, the effects of residue and temperature are inversely related. With increase in temperature of incubation, less fertilizer N becomes stabilized into humic fractions, presumably from increased microbial activity, microbial consumption of potential humic precursors (N-containing precursors of humic substances turned over faster at higher temperatures), and/or formation of different end-products with less humification potential.

Abbreviations: AEOM, alkali-extractable organic matter • ANOVA, analysis of variance • CFI, chloroform-fumigation–incubation • DAEFA, directly alkali-extractable fulvic acids • DAEHA, directly alkali-extractable humic acids • DAEHS, directly alkali-extractable humic substances • IRRI, International Rice Research Institute • NAEOM, non-alkali-extractable organic matter • PLSD, protected least significant difference • PMN, potentially mineralizable N • PRRI, Philippines Rice Research Institute • RMBFA, reducible metal-bound fulvic acids • RMBHA, reducible metal-bound humic acids • RMBHS, reducible metal-bound humic substances • SOM, soil organic matter


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
NITROGEN is the most growth-limiting nutrient in rice cropping systems (Savant and DeDatta, 1982). The fate of N in rice soils is often directly related to management of fertilizers and crop residues. In rice systems, management activities, such as fertilization and residue management, affect losses of N through volatilization and denitrification activity (Aulakh et al., 1992). In addition, N added to the soil as fertilizers or crop residues is immobilized by the soil biomass and subsequently transformed into organic forms that may be relatively resistant to mineralization (Broadbent and Nakashima, 1974; Olson and Swallow, 1984). The availability of stabilized organic N in rice cropping systems is not well known. Better understanding of the behavior of added N fertilizer and the availability of N from the soil organic N is essential for developing efficient N management strategies to enhance and sustain rice production.

Rice is the second most important crop in the world (Food and Agriculture Organization of the United Nations, 1996), and its production yields a large amount of straw residues annually. Management of these residues is required for seedbed preparation, maintaining soil fertility, and weed and pest management. Open-field burning, which has been used traditionally to dispose of residues and sanitize fields against pests and diseases, has become unpopular in many regions of the world because of air pollution concerns. The incorporation of straw residues into the soil combined with winter flooding has been proposed as an alternative to open-field straw burning in California Sacramento Valley and might also become common in other rice systems in the temperate zone. The effect of incorporating straw on long-term fertility in rice cropping systems has not received a great deal of attention to date.

Declining yield in continuously cropped irrigated rice systems where a significant portion of the rice residue is retained was first reported by Flinn et al. (1982). In irrigated rice, grain yield is closely associated with N uptake when the availability of other nutrients is adequate and pest damage does not limit crop growth. Results from the long-term N response experiments at the International Rice Research Institute (IRRI) and at the Philippines Rice Research Institute (PRRI), initiated in 1966 and 1968, respectively, show that soil organic C and total N have remained relatively stable since 1977 at IRRI and have even increased since the initial crop at PRRI where yields have declined (Cassman et al., 1995). Cassman et al. (1995) reported that neither B toxicity nor Zn, P, or K deficiency was responsible for yield reductions. An inadequate N supply in the late season when the crop is dependent on the soil N supply was suggested (Kropff et al., 1994; Cassman et al., 1995). Cassman et al. (1995) reported a significant yield response to N applied at the flowering stage in field experiments, demonstrating that the decrease in N uptake does not result from a reduction in the capacity of the root system to acquire N from soil. Olk et al. (1995)(and 1996) suggested that the accumulation of recalcitrant rice straw products, such as lignin-derived phenols, was responsible for reducing soil N availability. They theorized that stabilization of N into humic fractions might be the dominant process affecting the availability of fertilizer and soil N.

Decomposition of crop residues and SOM formation are sensitive to soil climatic factors, particularly temperature and soil water content. Mineralization and immobilization of N by soil microbes are intimately coupled with decomposition, and it is the balance between these two processes that defines the availability of N for plant uptake. Humic fractions of SOM have a large potential to retain N by means of both biotic (microbially mediated) and abiotic (chemical) reactions. Stevenson (1994) reported that much of the organic N stabilized in soil humic fractions resists attack by soil microbes and thus is not readily available for plant uptake. As a consequence of mineralization–immobilization reactions, up to one-third of the fertilizer N applied to the soil might be stabilized into humic substances, and only a small fraction (<15%) will be available for the next growing season (Kelley and Stevenson, 1996).

The above observations provided a starting point for investigating the relative effects of winter flooding and rice residue incorporation on fertilizer N immobilization in paddy soils. In temperate irrigated rice systems where an incorporation of straw in the fall followed by winter flooding is becoming common, growers might encounter the same problem of yield decline as in the tropical irrigated rice systems. The goal of our project was to enhance our understanding of fertilizer N dynamics in order to improve our ability to accurately predict the factors controlling its availability for rice production as a response to changes in residue management practices. To examine the effects of flooding (simulating a winter flooding) and temperature on fertilizer N stabilization into humic substances during the decomposition of rice straw, we incubated a paddy soil supplemented with 15N-(NH4)2SO4. We describe results used to determine the factors affecting the stabilization of N into humic substances under alternative rice residue management strategies.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil and Rice Straw Sampling
The soil was collected from an ongoing rice straw residue management trial located at Maxwell, Sacramento Valley, CA, on Willows clay (fine, smectitic, thermic Sodic Endoaquert). The soil contains 510 g kg-1 clay, 50 g kg-1 sand, and 440 g kg-1 silt (Gee and Bauder, 1979). It has a pH of 6.6 (U.S. Salinity Laboratory Staff, 1954) a cation-exchange capacity of 37.2 cmol kg-1 (Rhoades, 1982), a total C content of 20.65 g kg-1 (Dumas, 1831), a total N content of 1.96 g kg-1 (Dumas, 1831), and a total Fe content of 19.9 g kg-1 (DTPA-extractable Fe; 162.7 mg kg-1) and a total Mn content of 0.57 g kg-1 (DTPA-extractable Mn; 37.7 mg kg-1) (Deboer and Reisenauer, 1973).

The soil (0–15 cm) from a conventionally managed plot was sampled in the fall after open-field burning of rice residue. The gravimetric soil water content was determined after drying at 105°C for 24 h. The remaining soil was air-dried to 300 g kg-1 moisture and sieved (<0.5 cm) to remove coarse plant debris and to reduce the influence of residual plant material. The soil was stored for <1 wk at 4°C before use. Rice straw was sampled after grain harvest, dried immediately at 70°C, and chopped into fine segments (1–2 cm), similar to field flailing of the straw. Subsamples of dried soil and straw were ball-milled and analyzed for total C and N by the Dumas combustion procedure (Dumas, 1831) (Carlo Erba, Milano, Italy). The C/N ratios of whole soil and the rice straw were 10.6 and 59.2, respectively. The percentage of N of whole soil and the rice straw were 0.19 ± 0.003 and 0.62 ± 0.033, respectively.

Incubation Conditions
Soil (equivalent to 80 g of 105°C-dried soil) was weighed into 120-mL polypropylene specimen containers (Fisher Scientific, Pittsburgh, PA). Nitrogen was mixed into the soil at 24.52 mg N kg-1 dry soil as (15NH4)2SO4 (71.1 atom %), an amount equivalent to available soil N at harvest (data not shown). Rice straw, 0.8 g dry weight (two times average field rate or 62 mg straw N kg-1 dry soil to ensure substrate availability during the 160-d incubation), was added to one-half of the samples and mixed into the soil with a spatula. One-half of the samples (with or without straw added) were brought to 50% of water-holding capacity (WHC = 59.03 ± 1.8%), while the remaining samples were flooded with water to 2 cm above the soil surface. Each container was placed in a sealed 1-L mason jar containing 1 mL of water to prevent soil desiccation. Each mason jar was periodically aerated to maintain a CO2 level below 2%. The soil samples were incubated for 160 d in the dark at temperatures of 5, 15, or 25°C. A temperature of 5°C represents the coldest mean temperature in a nonflooded soil at 5-cm depth in December in the Sacramento Valley. A temperature of 15°C is the average daily maximum reached at 5-cm depth in a nonflooded soil in February and March. A temperature of 25°C is representative of the maximum at 5-cm depth from April through September (G. Fitzgerald, 1997, personal communication). All treatments were replicated four times for a total of 48 samples. The treatments were designated as: no straw, nonflooded (NSNF); no straw, flooded (NSF); straw, nonflooded (SNF); and straw, flooded (SF).

Inorganic Soil Nitrogen Content and Microbial Biomass Nitrogen Determination
During the incubation, soil subsamples were extracted with 2 M KCl (extractant/dry soil, 5:1) for exchangeable inorganic N, determined colorimetrically using a Lachat Quick Chem II Flow Injection Analyzer (Zellweger Analytical, Milwaukee, WI). Soil microbial biomass N was determined by the chloroform-fumigation–incubation (CFI) method as described by Horwáth and Paul (1994). Following Jenkinson and Powlson (1976), microbial biomass N was calculated using a correction factor (KN) of 0.54. The NO3–N and NH4–N in extracts from the soil and the CFI procedure were diffused onto Whatman no. 1 filter paper for 15N yield, after the method described by Brooks et al. (1989) and Khan et al. (1998).

Chemical Separation of Soil Humic Fractions
At the end of the 160-d incubation period, soils from each individual treatment were dried at 35°C, and subsamples were ball-milled prior to analysis. The chemical fractionation procedure for the isolation of soil humic fractions was adapted from McGill and Paul (1976) and Stevenson (1994). To increase the sensitivity for determining the effect of temperature, moisture, and straw incorporation on the dynamics of fertilizer N stabilization, five distinct fractions of the SOM were separated based on their solubility and ionic interactions: (i) directly alkali-extractable humic acids (DAEHA) (adapted from Stevenson, 1994); (ii) directly alkali-extractable fulvic acids (DAEFA) (adapted from Stevenson, 1994), (iii) reducible metal-bound humic acids (RMBHA) (adapted from McGill and Paul, 1976), (iv) reducible metal-bound fulvic acids (RMBFA) (adapted from McGill and Paul, 1976), and (v) non-alkali-extractable organic matter (NAEOM) (adapted from Stevenson, 1994).

The following modifications to the above referenced methods were done. Air-dried soil samples (20 g) were washed with 100 mL of 0.1 M trace metal grade HCl to remove carbonates, and floating debris were siphoned off. Prewashed soils were then extracted five times (for a total of 28 h) with 200 mL each time of 0.4 M NaOH under N2. The DAEHA were separated from the DAEFA by precipitation with HCl at pH 2. Some humic substances in soil are attached to minerals through metal binding, which affects their solubility and other properties (Schulten and Schnitzer, 1995). A significant fraction of the organic matter is retained in the soil in combination with clay minerals after initial extraction with base. These nondirectly alkali-extractable humic substances are coordinated with polyvalent cations (e.g., Mn4+, Fe3+, Al3+) forming macromolecular bridges with clay minerals. The soil used in this study was rich in Fe and Mn (as mentioned above), indicating a high potential to bind humic substances in this way. To reduce and remove Fe and Mn, the soil samples were shaken for 12 h with 200 mL of 1.2 M Na2S2O3 and then rinsed with 200 mL of 0.1 M HCl; the procedure was repeated until no Fe2+ was found in the HCl rinse. Another set of five extractions with 0.4 M NaOH under N2 was then performed to separate the metal-bound humic substances (RMBHA and RMBFA). The alkali-insoluble organic matter remaining with the mineral fraction was designated NAEOM, commonly referred to as humin (Stevenson, 1994). We did not separate the NAEOM from the mineral fraction at the end of the extraction procedure, and therefore it will be referred to as clay-NAEOM. Separated humic fractions and clay-NAEOM were freeze-dried and ground to a fine powder prior to analysis for total N content and 15N enrichment by combustion-GC-IRMS (20-20/ANCA-NT, Europa Scientific, Crewe, UK). Ash content of isolated humic fractions was determined after heating subsamples for 2 h at 550°C.

Nitrogen Availability in Soil and Non-Alkali-Extractable Organic Matter
The extraction procedure with hot 2 M KCl, described by Jalil et al. (1996), was used to measure the potentially mineralizable N (available N for plant uptake) in the whole soil and clay-NAEOM at 160 d. The procedure involved extracting 1 g of sample in 10 mL of 2 M KCl at 100°C on a digestion block for 4 h. The tubes were then removed from the digester, cooled at room temperature, and centrifuged. The supernatant and residues after being rinsed with double-deionized water were dried at 50°C and analyzed for total N content and 15N enrichment.

Hydrolysis with hot 6 M HCl was performed to measure the proportion of recalcitrant N (acid insoluble N) in the whole soil and clay-NAEOM at 160 d. Soil samples (1 g), with 10 mL of 6 M HCl, were heated at 100°C for 4 h on a digestion block; the insoluble residue thus obtained was further rinsed, dried, and ground to a fine powder.

A subsample of the clay-NAEOM (1 g) was also digested with 300 g L-1 H2O2 at 90°C, to estimate the proportion of the 15N-NAEOM fixed into the interlayer space of clay minerals, and therefore protected from attack by microbes. After completion of the digestion the tubes were removed from the digester, cooled at room temperature and centrifuged. Residues in the bottom of each tube, after being rinsed with double-deionized water, were dried at 50°C and analyzed for total N content and 15N enrichment.

Statistical Analysis
The experiment consisted of a factorial arrangement of treatments with 3 factors: straw treatment, flood regime, and temperature. We first performed analysis of variance (ANOVA) to measure the significance among treatments and then run a protected least significant difference (PLSD) test for mean differences. Statistical analyses were performed using StatView Software (StatView 5.0.1, SAS Institute, 1992–1998); when mentioned in the text, significant differences between treatments were measured after Fisher's PLSD at a significance level of 0.05.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
The distribution of fertilizer N in soil humic fractions and other soil N pools was significantly affected by rice straw, soil moisture, and temperature. Among the most significant findings of this experiment is the large recovery of fertilizer N in the clay-NAEOM after 160 d. These results suggest that rice residue and flooding management may have impacts on N availability in rice cropping systems. The following results and discussion define the effect and importance of these alternative rice management practices.

Loss of Fertilizer Nitrogen from the Soil or Immobilization into the Microbial Biomass
In our study, there are two reasons for the loss of fertilizer NH4 from the soil: (i) volatilization as NH3 and (ii) denitrification under anaerobic conditions. Denitrification is known to cause substantial losses of fertilizer NH4 in paddy soils (Mitsui, 1960; Ponnamperuma, 1972; Keeney and Sahrawat, 1986). Though denitrification is an anaerobic process, it has been shown that the reaction occurs under what may appear to be well-aerated conditions (Focht and Verstraete, 1977). Under nonflooded conditions without rice straw added to the soil (NSNF), most of the fertilizer N (95%) was recovered in the soil at the end of the incubation period, with no significant effect of temperature (Fig. 1) . We conclude that the missing fertilizer N in nonflooded systems was lost through denitrification, assuming that volatilization of NH3 was negligible in our experimental conditions at a soil pH of 6.6. As expected under flooded conditions (NSF), a significant amount of fertilizer N was lost through denitrification; however, this loss was significantly less at 5°C (53%) than at higher temperatures, with a maximum loss of 82% of the fertilizer N at 15°C and 73% at 25°C (Fig. 1). These results corroborate the findings of Nommik (1956) and Malhi et al. (1990) who showed that denitrification significantly increases with temperature. According to Patrick and Delaune (1977), a lag time of {approx}2 wk after submerging the soil is necessary for the sequential reduction of all electron acceptors to reach strongly reducing conditions, allowing for some conversion of fertilizer NH4 to NO3. Almost all the NO3 present in the soil disappeared within a few days after submergence (data not shown). However, the magnitude of N loss in the presence of rice plants would have certainly been different. Rice plants transport atmospheric O2 through the stem to the roots, and some of this O2 subsequently diffuses from the root into the adjacent soil layer (Armstrong, 1964), creating a thin oxidized layer in the rhizosphere which can support aerobic microbial populations. Nitrification in the oxidized layer and denitrification in the adjacent reduced soil occur simultaneously in the root vicinity. Plant roots can also accelerate denitrification in the rhizosphere by taking up O2 (Woldendorp, 1963). In paddy soils, the controlling factor of N loss from the rhizosphere is the competition for NO3 uptake between denitrifying bacteria and rice roots (Reddy and Patrick, 1986). In addition, loss of NH4 can also occur in the presence of rice plants. Therefore, the N loss due to denitrification and volatilization in our study without rice plants might have been lower compared to that of systems with plants.



View larger version (36K):
[in this window]
[in a new window]
  		Fig. 1. Fertilizer N lost during the 160-d incubation. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 
The incorporation of rice straw in the flooded treatment led to a significant reduction in the loss of fertilizer N through denitrification: from 53 to 11% at 5°C, from 82 to 11% at 15°C, and from 73 to 7% at 25°C, without straw and with straw added, respectively (Fig. 1). The significant increase in size of the soil microbial biomass after addition of straw (Fig. 2) was most likely responsible for the reduction in fertilizer N loss. The incorporation into the soil of a large amount of C substrate in the form of rice straw (SF and SNF) has been shown to increase microbial activity and size dramatically (Devêvre and Horwáth, 2000) and lead to an immobilization of fertilizer N into the biomass (He et al., 1988). No significant differences in denitrification between the straw flooded treatment (SF) and the straw nonflooded treatment (SNF) were found at 5 and 15°C (Fig. 3) , while at 25°C, the immobilization of fertilizer N in the soil biomass decreased significantly with flooding compared with nonflooded conditions. Figure 2 shows that during the incubation under flooded conditions at 25°C, the biomass N decreased significantly after 60 d to reach a value as low as the non-straw flooded treatment at 160 d. At 25°C, anaerobes, unlike aerobes, might have run out of substrate to support their growth over time since it appears that there was not sufficient available C after 160 d for maintaining the microbial pool size.



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 2. Change in soil microbial biomass N during the incubation period as affected by temperature, moisture status, and straw residue input. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Line bars indicate standard error of the means (n = 4)

 


View larger version (64K):
[in this window]
[in a new window]
  		Fig. 3. Fertilizer N recovered in the microbial biomass at 160 d. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 
Fertilizer Nitrogen Recovered in the Inorganic Pool
Most of the fertilizer N recovered at 160 d in the non-straw nonflooded treatment (NSNF) was found in the inorganic pool (Fig. 4) , with the majority as NO3 (data not shown), indicating that all the fertilizer NH4 had been nitrified. The amount of fertilizer N recovered in the NSNF inorganic pool decreased significantly with increasing temperature (Fig. 4) and was associated with a concurrent increase in soil microbial biomass (Fig. 2). These data show higher use of inorganic N by soil microbes at higher incubation temperatures. Under no straw and flooded conditions (NSF) at 15 and 25°C, <1% of the fertilizer N added to the soil was recovered in the inorganic N pool, as opposed to 20% at 5°C (17% as NO3 and 3% as NH4; data not shown). At 5°C, the rate of denitrification was probably slow enough to explain the high proportion of NO3 remaining after 160 d under flooded conditions. Other explanations for the presence of NH4 in the flooded soil at 160 d would be that anaerobes at 5°C did not use all available NH4 to support their growth; the N requirement of anaerobic metabolism is known to be lower than that of aerobic metabolism (Acharya, 1935).



View larger version (37K):
[in this window]
[in a new window]
  		Fig. 4. Fertilizer N recovered in the inorganic N pool at 160 d. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 
In the straw treatment at 5°C there was almost no fertilizer N recovered in the inorganic pool after 160 d, while 10% was recovered at 25°C with no significant difference between flooded treatments and nonflooded treatments (Fig. 4). All the fertilizer N recovered in the inorganic pool at 25°C under flooded conditions was in the form of NH+4, and in the form of NO-3 under nonflooded conditions (data not shown). At 15°C under flooded conditions, only 3% of the fertilizer N was recovered in the inorganic pool (Fig. 4). The turnover of organic N pools had been accelerated due to an increase in soil microbial activity following an increase in temperature (Devêvre and Horwáth, 2000). At 25°C, part of the fertilizer N previously immobilized in the microbial biomass and subsequently in the SOM pools might have already been remineralized by soil microbes during the 160-d incubation. The following discussion will use additional data to strengthen this hypothesis.

Fate of Fertilizer Nitrogen in the Alkali-Extractable Organic Matter
The DAEHS represent the organic matter extracted with 0.4 M NaOH prior to removal of reducible metals binding clay minerals with humus material. Soil humus is a mixture of polymers of greatly varying sizes and characteristics with no perfect extractant for such heteropolymers (Stevenson, 1994). Many problems associated with extraction procedures have already been reported (Kononova, 1961; Honda, 1983; Stevenson, 1994). When soil samples are extracted with different extractants under various conditions, the ratio of fulvic acids to humic acids might change (Honda, 1996), as can their degrees of polymerization (Kononova, 1961). Fulvic acids, with less aromatic C networks and more aliphatic side chains, can be considered of lower maturity than humic acids; however, they often have the same structural units (Kononova, 1961; Stevenson, 1994). According to our objectives, and for purpose of simplification in the presentation and discussion of results, we considered humic polymers rather than humic acids and fulvic acids separately. We therefore decided to combine DAEHA with DAEFA to form DAEHS (directly alkali-extractable humic substances), and RMBHA with RMBFA to form RMBHS (reducible metal-bound humic substances).

It is commonly assumed that the DAEHS is the most dynamic pool of the SOM, and will be the first to show the effects of treatments (Stevenson, 1994). This pool has been previously identified as "mobile humic substances" by McGill and Paul (1976) and Olk et al. (1995). The RMBHS are defined as the organic matter soluble in 0.4 M NaOH after removal of Na2S2O3–reducible binding metals and is thought to be less dynamic because of the close association of this fraction with metals and clays. The organic constituents of the remaining clay–humus fraction after the second NaOH extraction following removal of binding metals, correspond with a resistant pool (NAEOM). The latter fraction is often referred to as humin (Stevenson, 1994).

Fertilizer 15N was not evenly distributed among treatments between the three SOM fractions (Fig. 5–7) . A large fraction of the added N was removed by the initial NaOH extraction, leading to a more recovery of 15N in the DAEHS for all treatments compared with the RMBHS and clay-NAEOM. In the non-straw treatment, the percentage of fertilizer N recovered in the different soil organic pools was always low compared with the straw treatment, regardless of the temperature of incubation and/or aeration status. In the absence of an exogenous C source, a large proportion of the soil microbial population was probably in a resting state, with very low decay constants that affect neither net mineralization nor immobilization (Paul and Juma, 1981). The lack of a strong significant effect of flooding and temperature in the non-straw treatment compared with the straw treatment supports this observation (Fig. 5–7). The addition of straw significantly increased the incorporation of fertilizer N in the DAEHS regardless of soil aeration status (Fig. 5). The larger amount of N in the DAEHS shows the importance of this pool in sequestering recently added fertilizer N. However, the significant decrease in percentage fertilizer N recovered in DAEHS with rising temperature suggests that this pool had already turned over during the 160-d incubation period. An increase in microbial activity and/or a shift in microbial populations after addition of straw, when temperature of incubation increased was suggested by Devêvre and Horwáth (2000).



View larger version (59K):
[in this window]
[in a new window]
  		Fig. 5. Fertilizer N recovered in the directly alkali-extractable humic substances (DAEHS) at 160 d. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 


View larger version (55K):
[in this window]
[in a new window]
  		Fig. 7. Fertilizer N recovered in the clay-associated non-alkali-extractable organic matter (clay-NAEOM) at 160 d. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 


View larger version (50K):
[in this window]
[in a new window]
  		Fig. 6. Fertilizer N recovered in the reducible metal-bond humic substances (RMBHS) at 160 d. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 
The low amount of 15N in the RMBHS across all the treatments suggests that these fractions represent a more stable humic N pool, with possibly a lower turnover rate than the DAEHS (Fig. 6). The metal-bound humus is probably protected from chemical and biological degradation through development of metal cation bridging bonds to the mineral phase (Turchenek and Oades, 1979; Preston and Schnitzer, 1987). Olk et al. (1995) compared Ca-bound humic acids to mobile humic acids and reported that Ca-humic acids had higher aromaticity, higher C condensation, increasing substitution of carboxyl groups, and decreasing substitution of amino acids, and concluded that the mobile humic acid represented an early stage in SOM formation and the Ca-humic acid a later stage. Our data also suggest that the DAEHS (equivalent to the mobile humus fraction of Olk et al., 1995) have potentially greater N cycling activity than RMBHS. There was, however, a significant decrease in fertilizer N recovered in the RMBHS from the incubation at 5°C compared with that at 25°C in the straw treatments. Furthermore, soil flooding significantly reduced the immobilization of fertilizer N in the RMBHS at 5 and 15°C, while at 25°C there was no significant difference between flooded and nonflooded treatments (Fig. 6). At higher temperatures, microbial activity was higher, and therefore, the turnover rate of the RMBHS was higher. Although considered less dynamic than DAEHS because of its lower extractability, the RMBHS had already turned over during the 160-d incubation.

The effects of temperature and rice residue, therefore, are inversely related. Under higher temperatures, less fertilizer N becomes stabilized in humic fractions, presumably from higher microbial activity and turnover, microbial consumption of potential humic precursors, and/or formation of different end products with less humification potential. These results indicate the importance of microbial activity and production in supplying components to different humic pools.

Stabilization of Fertilizer Nitrogen into the Non-Alkali-Extractable Organic Matter
In general, organic matter is more readily extracted from soils low in clay than from those rich in clay (Schnitzer and Schuppli, 1989). In the Maxwell soil, particularly rich in clay (2:1 smectites, mostly montmorillonites), a significant fraction of the SOM was not extracted but recovered in the clay–humus residue at the end of the extraction procedure. In this high clay system, humic substances can be held within the interlayers of expanding lattice clay minerals like smectites (Sposito, 1989; Stevenson, 1994). Stevenson (1994) mentioned that this occluded OM is not solubilized by conventional extraction procedures, such as NaOH treatment, but can be released by destruction of clays with hydrofluoric acid (HF). The humic fraction that cannot be extracted from soils and sediments by dilute base or acid was reported by Kononova (1961) and Schnitzer and Khan (1978) as humin. The humin or clay-NAEOM is often considered highly refractive and stable (Stevenson, 1994) and can also constitute the most abundant organic fraction in recently formed sediments (Fabbri et al., 1996).

The clay-NAEOM was surprisingly highly enriched in 15N (Fig. 7). In a soil incubated for 7 d at 30°C, He et al. (1988) also found that 22% of the applied 15N was accounted for in the humin fraction in forms that could not be extracted with NaOH. As shown for the extractable OM, a significant increase in stabilization of fertilizer N in the nonextractable OM resulted from the incorporation of straw into the soil. Labeled N could have entered the clay-NAEOM as a result of normal humification processes, clay fixation of 15NH4 and/or through contamination from microbial tissues during the extraction procedures. The high amount of tracer N recovered in this fraction was affected by treatment, suggesting that active components of N cycling were present in the clay-NAEOM. The effects of temperature and flooding suggest that the differences in 15N recovered in the residue after extraction with NaOH did not result exclusively from clay fixation of 15NH4. The comparison between the amount of fertilizer N recovered in the clay-NAEOM (Fig. 7) and the clay-NAEOM after H2O2 oxidation (Fig. 8) gives an estimate of the contribution of abiotic NH4 fixation to the overall enrichment in 15N of the clay-NAEOM. We estimate that between 1 and 5% of the fertilizer N recovered in the clay-NAEOM resulted from abiotic clay fixation.



View larger version (56K):
[in this window]
[in a new window]
  		Fig. 8. Percentage of fertilizer N remaining in the clay-NAEOM fraction at 160 d, after oxidation with hot 30% H2O2. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 
In the straw treatments, when the temperature of incubation increased from 5 to 25°C, the recovery of fertilizer N in clay-NAEOM significantly decreased in both flooded and nonflooded conditions. In the flooded soil compared with the nonflooded soil, significantly less fertilizer N was stabilized. The presence of the rice straw C and fertilizer N led to rapid microbial growth, leading to their conversion into more stable humus forms as a result of microbial turnover (Azam et al., 1989). Field studies on these soils shortly after rice harvest confirm increases in microbial biomass and immobilization of inorganic N (Bird et al., 1998). As suggested by Ladd and Paul (1973) and McGill et al. (1975), after net immobilization of inorganic N, the newly immobilized fertilizer 15N was remineralized and redistributed among more complex heterogeneous and less labile fractions of the SOM. A significant fraction of the stabilized N may also occur as insoluble components of microbial tissues, such as fungal melanins (He et al., 1988). These observations suggest that the clay-NAEOM fraction is a complex and dynamic mixture where added N is turning over and/or where byproducts of the microbial activity are accumulating. The amount of fertilizer 15N stabilized in the clay-NAEOM was dependent on moisture status and temperature, therefore on microbial activity.

Biological Availability of the Stabilized Nitrogen
The procedure of SOM extraction we used in this study did not address biological N availability. To investigate the availability of N in clay-NAEOM, we determined the potentially mineralizable N (PMN) by extraction with hot 2 M KCl, using the protocol described by Jalil et al. (1996). A close relationship has been found between the amount of NH4–N extracted with hot 2 M KCl and the N-supplying capacity of the soil in controlled-environment chambers (Whitehead, 1981; Smith and Li, 1993; Jalil et al., 1996) and field experiments (McTaggart and Smith, 1993; Campbell et al., 1997). According to this definition of PMN, most of the fertilizer N recovered in the clay-NAEOM after 160 d (Fig. 7) is not readily available for plant uptake (Fig. 9) . The comparison between Fig. 7 and Fig. 9 showed that the increase in temperature led to a significant increase in the amount of fertilizer N stabilized (unavailable) in clay-NAEOM, and showed that significantly more fertilizer N was stabilized in this fraction under flooded conditions than under nonflooded conditions at low temperature. At 25°C under both flooded and nonflooded conditions, the fraction of PMN in the clay-NAEOM became negligible.



View larger version (39K):
[in this window]
[in a new window]
  		Fig. 9. Percentage of fertilizer N extractable from the clay-NAEOM fraction at 160 d with hot 2 M KCL. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 
Hydrolysis with hot 6 M HCl and oxidation with hot 30% H2O2 of the clay-NAEOM were drastic treatments undertaken to reveal stabilized (or recalcitrant) pools of N. As mentioned by Stevenson (1994), a large amount of the soil N is recovered in acid-insoluble forms and is thought to occur as structural components of humic substances (Stevenson and He, 1990; Shulten and Schnitzer, 1997). Acid hydrolysis is expected to remove {approx}60 to 80% of the soil N (McGill and Paul, 1976). This treatment results in the release of part of indigenous clay-fixed NH+4, amino acids bound by peptide bonds, and those linked to quinones, but amino acids bonded directly to phenolic rings are probably not released (Stevenson, 1994). We assumed that the most part of fertilizer N remaining in the soil after digestion with hot 6 M HCl is stabilized in a fairly recalcitrant pool, and therefore not readily available for plant uptake. Figure 8 and Fig. 10 illustrate the amount of fertilizer N recovered in the clay-NAEOM after H2O2 oxidation and 6 M HCl hydrolysis, respectively. Not all of the NAEOM or so-called humin fraction should be considered refractive as has been proposed (Stevenson, 1994). Our data suggest that the NAEOM is in fact a complex and dynamic mixture of many forms of N, not all of which are refractive, as would be indicated by resistance to the above treatments. These indices of recalcitrance did not show drastic changes as a function of temperature. The incorporation of straw significantly increased the amount of fertilizer N remaining in recalcitrant forms. Subtraction of the amount of fertilizer N remaining in the clay-NAEOM after 6 M HCl hydrolysis (Fig. 10) from the amount remaining in the whole soil after 6 M HCl hydrolysis (Fig. 11) gives the amount of fertilizer N that was stabilized in the alkali-extractable OM (AEOM = DAEHS + RMBHS) after 160 d (Fig. 12) . The AEOM, like the NAEOM, appears to be a dynamic mixture of potentially mineralizable and recalcitrant forms of N. Moreover, except for the alkali-extractable OM at 5°C, there was always significantly less fertilizer N recovered in the recalcitrant N pools of the flooded soils compared with the nonflooded soils at all three temperatures of incubation for the whole soil, and at 15 and 25°C for the AEOM fraction. These results differ from those reported earlier by Cassman et al. (1995), Gaunt et al. (1995), and Olk et al. (1996). Olk et al. (1996) concluded that abiotic immobilization of available N by phenolic compounds, which accumulated under prolonged submergence, contributed to a decrease in soil N supply and was a probable cause of the yield decline in continuous irrigated rice systems in the tropics. Several earlier studies had provided evidence for the accumulation of lignin-phenols in submerged soils (Tsutsuki and Kuwatsuka, 1979; Tsutsuki and Ponnamperuma, 1987; Ye and Wen, 1991). The scarcity of O2 in flooded soils prevents complete degradation of aromatic compounds (Zeikus, 1981). Consequently, partially decomposed lignins accumulating under waterlogged conditions (Colberg, 1988) can combine with plant and microbial proteins and amino acids to form humic polymers that resist attack by microbes (Haynes, 1986). Mattson and Andersson (1942) and Burge and Broadbent (1961) have also implicated phenolic hydroxyl groups in nonexchangeable fixation of NH3 into SOM. Blocking of phenolic hydroxyl sites diminishes the NH3 fixation capacity of SOM (Burge and Broadbent, 1961). The existence of hydroxyquinone and quinone polymers in SOM could account for the anaerobic fixation of ammonia and the increase in fixation in the presence of O2 (Burge and Broadbent, 1961). According to these hypotheses, we should have measured an increase in fertilizer N stabilization into humus under flooded conditions; however, this was not the case, suggesting the existence of other factors that may control the stabilization and subsequent biological availability of fertilizer N in waterlogged soils. In contrast to our work, Olk et al. (1996) did not directly measure an index of N availability in humic fractions of the SOM. Instead, they correlated field observations with chemical characterization of the SOM constituents.



View larger version (56K):
[in this window]
[in a new window]
  		Fig. 10. Percentage of fertilizer N remaining in the clay-NAEOM fraction at 160 d, after hydrolysis with hot 6 M HCL. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 


View larger version (69K):
[in this window]
[in a new window]
  		Fig. 11. Percentage of fertilizer N remaining in the whole soil at 160 d, after hydrolysis with hot 6 M HCL. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 


View larger version (64K):
[in this window]
[in a new window]
  		Fig. 12. Fertilizer N stabilized in the alkali-extractable organic matter (AEOM) at 160 d. The treatments were designated as: NSNF, no straw nonflooded; NSF, no straw flooded; SNF, straw nonflooded; SF, straw flooded. Within and among bars, different letters show significant differences between treatments (ANOVA, PLSD, significance level 0.05)

 
At PRRI and IRRI long-term experiments, rice yield decline in flooded systems has been associated mainly with N deficiency, despite the use of recommended fertilizer rates (Cassman et al., 1995; Dobermann et al., 1997). Cassman et al. (1997) concluded that N deficiency resulted from a decreasing soil N supply, although total soil N content remained constant or increased with time. Olk et al. (1996) suggested that changes in chemical properties of OM fractions have contributed to N sequestration, resulting in a decrease in soil N supply. However, in our study, less fertilizer N was stabilized in the flooded treatment compared with the nonflooded treatment. As suggested by indices of recalcitrance of the stabilized N, the flooded treatment did not reduce the soil N supply power as much as the nonflooded treatment. An alternative explanation for this discrepancy from Cassman et al. (1997) might be that humic substances with different molecular structures react differently with soil minerals. Conversely, the degree of humic substances attraction will depend on the charge characteristics and density of different clay minerals (Gast, 1977). Since humus bound to clay minerals appears to be relatively stable against biodegradation (Stevenson, 1994), the interaction of organic matter fractions with clay minerals has, therefore, important consequences on the cycling of nutrients (Sposito, 1989). Expanding lattice clays (2:1) offer interlayer spaces for retention of organic substances. This protection reduces the contact of humic material with extracellular enzymes and therefore attack by microbes. The clay type may therefore affect the capacity of soils to physically protect organic C and N (Hassink, 1997); clays with high specific surface area retain more humic substances than clays with low specific surface area (Tate and Theng, 1980). Olk et al. (1996) also mentioned that differences in clay contents might contribute to variable phenol concentrations between soils. Norman and Gilmour (1987) confirmed what was suggested by Allison et al. (1953)-that N fixed by montmorillonitic soils was more available to nitrifiers than N fixed by vermiculitic soils. Since montmorillonites are the dominant clay minerals at Maxwell (R. Southard, 1998, personal communication), not vermiculites like at PRRI (Cassman et al., 1995), this may partly explain observed differences in N-supplying capacity between these two soils.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The addition of straw, regardless of other conditions, had the greatest effect on the stabilization of fertilizer N into SOM; however, this effect was consistently less under flooded conditions. With increasing temperatures of incubation, less fertilizer N was stabilized into SOM. We hypothesize that N-containing precursors of humic substances turned over faster at higher temperatures, giving them less opportunity to stabilize as humic substances. Alternatively, the succession of distinct microbial communities at different temperatures and/or different aeration statuses led to different end-products with various humification potentials. Microbial transformations associated with increases in temperature apparently enhanced the other factors in determining the distribution of immobilized fertilizer N after 160 d. Cell walls of microorganisms may be products that accumulate in SOM, or precursors in the formation of high molecular weight polymers. The incorporation of resistant plant products, such as lignin, may be more important under flooded conditions. On the other hand, the interaction of OM fractions with clay minerals also has important consequences on the cycling of nutrients and on the formation of soil aggregates (Sposito, 1989). More investigations are needed to address the biological availability of N in separate humic pools associated with different type of clay minerals.

The stabilization of N into SOM represents a substantial sink for fertilizer N, although long-term availability of the stabilized N is not easily determined. Two questions remain: (i) How long is the stabilized N tied up before it is remineralized? and (ii) How effectively and efficiently is soil organic N recovered by the rice crop? In rice cropping systems where residue recycling and winter flooding management are becoming common, a more complete understanding of the role of different SOM fractions is essential to determining the nature of N cycling in submerged soils and improving N use efficiency in rice.


   ACKNOWLEDGMENTS
 
We are grateful to the California Energy Commission, California Rice Research Board, and the College of Agricultural Science at UC Davis for financing this research. We would like to thank Susan Lo (UC Davis Rice Project) for assistance setting up and managing this laboratory incubation. We also thank Andrea D. DeLisle, Janette R. Girod, and Elisabeth L. Palmer (undergraduate students at UC Davis) for their valuable technical assistance in SOM extractions and sample preparation. We would like to express our gratitude to Timothy A. Doane for carefully reviewing this article and for his helpful comments. Dr. Dave Harris at the Stable Isotope Facility at UC Davis performed 15N measurements.

Received for publication January 10, 2000.


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




This article has been cited by other articles:


Home page
 Soil Sci.Home page
J. H. Davis, S. M. Griffith, W. R. Horwath, J. J. Steiner, and D. D. Myrold
Fate of Nitrogen-15 in a Perennial Ryegrass Seed Field and Herbaceous Riparian Area
Soil Sci. Soc. Am. J., April 19, 2006; 70(3): 909 - 919.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
D. C. Olk
A Chemical Fractionation for Structure-Function Relations of Soil Organic Matter in Nutrient Cycling
Soil Sci. Soc. Am. J., April 19, 2006; 70(3): 1013 - 1022.
[Abstract] [Full Text] [PDF]

Home page