| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a IRRI, P.O. Box 3127, Makati Central Post Office (MCPO), 1271 Makati City, Philippines
b Istituto di Chimica Agraria, Università di Bari, Via Amendola, 165/A, 70126 Bari, Italy
d.olk{at}cgiar.org
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and methodsResultsDiscussionREFERENCES |
|---|
Abbreviations: CaHA, calcium humate • ESR, electron spin resonance • FRS, fresh rice straw • FT IR, Fourier-transform infrared • LTCCE, long-term continuous cropping experiment • MHA, mobile humic acid • NMR, nuclear magnetic resonance • PU, prilled urea • RFI, relative fluorescence intensity • SOM, soil organic matter
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionREFERENCES |
|---|
14 Mha. Because floodwaters are maintained on lowland fields during crop growth, these soils are submerged for 8 to 11 months annually. Our knowledge of nutrient cycling under such conditions is inadequate, but SOM appears to play a central role in N cycling under anaerobic conditions (Broadbent, 1979; Cassman et al., 1996b). Recent questions have arisen regarding changes in soil N cycling that possibly develop under long-term intensive lowland rice cropping and their significance to the sustainability of intensive rice cropping (Cassman et al., 1995). Accordingly, this study continues our investigation into the effects of long-term soil submergence on the chemical nature of SOM. Acquired knowledge will support subsequent studies of N cycling in these systems.
In order to study SOM chemistry, two HA fractions were extracted from long-term field trials that varied in the number of irrigated rice crops per year. The MHA fraction is a younger, more labile SOM pool that responds clearly in quality and quantity to crop management, while the response of the more humified, polyvalent cation-bound CaHA fraction to crop management is less pronounced (Olk et al., 1995, 1996, 1998). The MHA accounted for 10 to 13% of total soil organic C in these soils and the CaHA for 8 to 12% (Olk et al., 1996, 1998, and unpublished data). Compared to the CaHA, our previous studies have found the MHA to absorb less visible light and to have higher proportions of C in aliphatic and phenolic forms, P in diester forms, and N in amide forms, while the CaHA had higher proportions of aromatic C and monoester P. Carbon-14 dating established that the CaHA was older than the MHA in a California soil cropped to cotton (Olk et al., 1995) and in two rice soils on the IRRI farm (Olk et al., 1996).
Here both HA fractions were analyzed for elemental concentrations and acid functional groups, and by Fourier-transform infrared (FT IR), fluorescence, and electron spin resonance (ESR) spectroscopies. These analyses follow earlier studies of the same samples by 13C nuclear magnetic resonance (NMR) spectroscopy and CuO oxidation (Olk et al., 1996, 1998), 15N NMR (Mahieu et al., 2000a), and 31P NMR (Mahieu et al., 2000b). In general, these studies associated increasing cropping intensification and soil submergence with a less humified state of both HA fractions; in HA extracted from the submerged soils, 13C NMR and CuO analyses demonstrated notable levels of partially degraded lignin residues represented as phenols and methoxy-C, 15N NMR analysis found slightly higher levels of amide-N, and 31P NMR analysis found higher concentrations of diester P compounds.
A number of earlier studies on the chemistry of SOM formed under submerged conditions also associated incomplete SOM humification with soil submergence. Evidence included less visible light absorption (Mitsuchi, 1974), enhanced signals of lignin residues in 13C NMR (Ye and Wen, 1991), lower concentrations of O-containing functional groups (Tsutsuki and Kuwatsuka, 1978), and higher H concentrations (Yonebayashi and Hattori, 1988). These studies generally selected samples from diverse settings based on soil type or land use, and they did not compare soils varying regularly in rice cropping intensity or fertilizer treatment.
|
Materials and methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionREFERENCES |
|---|
The field without any irrigated rice cropping had been planted to dryland rice for at least the previous 35 years and had no imposed fertilizer treatments. The rice–soybean field supported both irrigated rice and one upland crop annually for 10 years prior to sampling, although the rice crop had not been planted every year. The double-cropped field was the N-source experiment, in which equal amounts of N were added in different organic and inorganic forms as mainplot treatments in two annual rice crops for 11 years prior to sampling. The triple-cropped field was the long-term continuous cropping experiment (LTCCE), being in the 89th consecutive rice crop at time of sampling and with N fertilizer rate as mainplot treatments. Soil organic C ranged from 13 g kg-1 soil in the dryland and rice–soybean fields to 23 g kg-1 in the double-cropped field and up to 25.4 to 28.8 g kg-1 in the triple-cropped soils.
For the N-source experiment, we report results for the minus N fertilizer control treatment and for treatments receiving N (116 kg ha-1 in the dry season, 58 kg ha-1 in the wet season) either as prilled urea (PU), Sesbania rostrata Bremek. & Oberm. green manure, Azolla microphylla Kaulf. green manure, or in a combination treatment with half the N added as fresh rice straw and half as PU (FRS–PU). For the LTCCE, we report results from the treatments receiving either zero or optimal (non-limiting) levels of N fertilizer (200 kg N ha-1 in the dry season and 105 kg N ha-1 in each wet season crop).
Pests, irrigation water, and availability of other nutrients were controlled in the irrigated rice fields to avoid yield loss. These fields were flooded, plowed, and puddled at the beginning of each crop and remained submerged with a floodwater depth of 5 to 10 cm throughout each cropping period of 95 to 100 d. They were drained for harvest and were not irrigated during the subsequent fallow period until time of plowing for the next crop. Consequently, soil was submerged for about 220 days annually for double-cropped fields and about 330 days annually for the triple-cropped field.
Soil was sampled in all fields from the 0- to 15-cm depth in multiple corings per plot, which were then pooled within each plot for subsampling. Immediately after sampling all soils were stored at 4°C at their original field moisture content until HA extraction.
Organic Matter Extraction
As described by Olk et al. (1996, 1998), fresh, undried soil was shaken overnight in a 1:10 (w/v) solution of 0.25 M NaOH under N2 gas. Then, the MHA were isolated by centrifugation and acidification of the supernatant to pH 2, and the soil was washed twice in 0.0025 M CaCl2 to remove solubilized MHA–fine clay complexes that otherwise interfered with subsequent extractions steps. For only the LTCCE soil, this step also required four additional water washes. Then for all fields, the soil was mixed and suspended repeatedly in 1:10 (w/v) washes of 0.1 M HCl until the pH of the supernatant remained below 1.3 in order to exchange for polyvalent cations complexed with SOM. Two subsequent water washes raised the soil pH to at least 2.5. Then the soil was again shaken overnight in 0.25 M NaOH under N2 gas for extraction of the CaHA by centrifugation and acidification. Both the MHA and CaHA were subsequently shaken for three days in 0.5% HF/HCl with daily solution replacement for dissolution of soil particles, dialyzed for three days against 0.01 M HCl, 0.001 M HCl, and water successively with three daily solution replacements, then frozen and lyophilized.
Humic acid fractions were extracted from individual replicate plots, but the HA replicates were combined into treatment composites for the analyses conducted here.
Chemical Analyses
Carbon, H, N, and S concentrations of the HA were determined by a Fisons Instruments EA 1108 elemental analyser (Fisons, Crawley, UK). Oxygen concentration was calculated by difference. Ash contents of all HA samples were <1% as measured by heating to 700°C (data not shown). Atomic ratios were calculated where each element was represented by the quotient of its concentration and atomic weight. Total acidity and carboxylic group concentrations were determined by conventional titration methods (Schnitzer, 1982). Phenolic hydroxyl concentrations were obtained by difference.
Spectroscopic Analyses
The FT IR spectra of HAs were recorded on KBr pellets in the 4000 to 400 cm-1 wavenumber range using a Nicolet 5 PC FT IR spectrophotometer (Nicolet Instr., Madison, WI). Sixty four scans and a peak resolution of 2 cm-1 were used to obtain each spectrum. The KBr pellets were obtained by pressing under reduced pressure a mixture of 1 mg HA and 400 mg KBr, spectrometry grade.
Fluorescence spectra in the emission, excitation, and synchronous-scan modes were obtained on aqueous solutions of HA at a concentration of 100 mg L-1 after overnight equilibration at room temperature and adjustment to pH 8 with 0.05 M NaOH. Spectra were recorded using a Perkin Elmer (Norwalk, CT) LS-5 luminescence spectrophotometer equipped with a Perkin Elmer Data Station 3600 for data generation and processing by PECLS software. Emission and excitation slits were set at 5-nm band width, and a scan speed of 120 nm min-1 was selected for both monochromators. Emission spectra were recorded over the range of 380 to 550 nm at a constant excitation wavelength of 360 nm. Relative fluorescence intensity (RFI) was based on a unitless reciprocal to the gain used to normalize each emission spectrum and was expressed in arbitrary units (Senesi et al., 1991). Excitation spectra were obtained over a scan range of 300 to 500 nm by measuring the emission radiation at a fixed wavelength of 520 nm. Synchronous-scan excitation spectra were measured by scanning simultaneously for both the excitation (varied from 350–550 nm) and emission wavelengths while maintaining a constant, optimized wavelength difference
(Senesi et al., 1991).
The ESR spectra were obtained at room temperature (293 ± 2K) on solid HA samples packed in quartz ESR tubes (4 mm o.d., 3 mm i.d.) using a Bruker AXS (Karlsruhe, Germany) ER-200D SRC ESR spectrophotometer operating at X-band frequency with 100-Kz magnetic field modulation. Field range (H) scanned for the detection of paramagnetic metal ions were 500 and 100 mT using a modulation amplitude of 5 and 1 mT, respectively, a microwave frequency of 9.52 GHz, and a microwave attenuation of 13 dB (corresponding to a microwave power of 10 mW). The physical parameters of paramagnetic metal species giving ESR signals, which are the spectroscopic splitting factor (g-value) and the hyperfine coupling constant |A| (cm-1 x 10-4), were calculated from the experimental ESR spectra using standard equations (Senesi, 1990, 1992). The values of these parameters provide unique chemical and structural information on the oxidation state of paramagnetic metals, binding mechanisms, coordination and symmetry of binding sites, and possibly on the identity of organic ligands involved in ligand complexation (Senesi 1990, 1992).
The ESR spectra of organic free radicals were measured over a small range of field
centered at about the resonance field of the free electron
, using the same microwave frequency and microwave attenuation as above and a modulation amplitude of 0.63 mT (Senesi, 1990). Each spectrum consisted of a single Lorentzian line devoid of any hyperfine structure. The absolute free radical concentration expressed in spins g-1, line widths expressed in mT, and spectroscopic splitting factors (g-values) were calculated using standard procedures and equations (Senesi, 1990).
|
Results |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionREFERENCES |
|---|
|
|
|
Carboxyl group concentration of either HA fraction was smaller in the submerged soils than in the aerated soils. Total acidity and phenolic OH group concentrations of the MHA were generally less in the submerged soils than in the dryland rice and rice–soybean soils.
These results indicate that with increasing intensity of irrigated rice cropping: (a) both HA fractions became especially enriched in S and H, were of higher aliphatic character (smaller C:H ratio), and had lower levels of O and COOH groups; and (b) the MHA generally had lower levels of total acidity and phenolic OH groups in particular than did the corresponding CaHA fractions.
Fertilizer–N Source (Double-Cropped Rice)
Elemental composition and elemental ratios differed only slightly across treatments. The MHA of the treatments receiving organic amendments had generally slightly higher N and C and lower O concentrations than did the MHA of the urea-amended and unfertilized treatments, resulting in slightly lower O:C ratios. No consistent treatment effects on functional groups were clear. The phenolic OH group concentration varied widely among MHA fractions.
Fertilizer N Rates (Triple-Cropped Rice)
Minor differences existed between the minus N fertilizer control and the optimal N fertilizer treatments. The HA of the minus N treatment tended to have slightly higher S and slightly lower H concentrations than the HA of the optimal N treatment. Variations between treatments in N concentrations and C:H, C:N, and O:C ratios were inconsistent. The COOH concentrations were similar, but total acidity and the phenolic group content of both fractions were somewhat lower for the optimal N treatment than for the minus N control.
C:N:S Ratios
The C:S and N:S ratios decreased markedly with increasing soil submergence in both HA fractions (Table 2). The MHA had generally narrower ratios than did the CaHA, although the differences were smaller in the submerged soils. High levels of organic S in the submerged soils were suggested by the fact that the C:N:S ratios of both fractions (Table 2) were much narrower than the atomic ratios derived from data presented by Bettany et al. (1979) for HA extracted from aerated soils along a wide climatic gradient. By comparison, the CaHA fractions of the aerated soils were relatively depleted in S, having comparable atomic ratios as those derived from Bettany et al. (1979).
The C:N:S ratios of the MHA were wider in N fertilizer–amended treatments than in minus N fertilizer treatments for both the double- and triple-cropped fields. The C:N:S ratios differed among the N-source treatments of the double-cropped field in a parallel manner for both HA fractions. Ratios of organic N treatments were not distinct from those of non-organic N treatments.
Fourier-Transform Infared Spectroscopy
To depict the effects of increasing soil submergence, FT IR spectra from the dryland rice, rice–soybean, double-cropped PU, and triple-cropped optimal N treatments are shown for the MHA and CaHA in Fig. 1 and 2
, respectively. The FT IR spectra did not differ noticeably among fertilizer treatments, hence HA spectra from the other four double-cropped treatments and the triple-cropped minus N treatment are not shown.
|
|
2855 cm-1 is ascribed to aliphatic C–H stretching. Third, an absorption at 1719 to 1716 cm-1 was generally strong in CaHA and very weak in MHA and is due to C=O stretching of COOH and other carbonyl groups: e.g., ketones. Fourth, a poorly resolved envelope of at least two distinct absorptions of differing relative intensities was centered between 1660 to 1630 cm-1 and 1620 to 1600 cm-1. The former region is attributable to C=O stretching of amide groups (amide I band) and quinone and/or H-bonded conjugated ketones, while the latter region is attributable to aromatic C=C vibrations, symmetric stretching of COO- groups, and H-bonded C=O of conjugated ketones. Fifth, a peak at 1513 to 1508 cm-1 that is stronger in the MHA than CaHA and absent in the CaHA of the aerated soils can be preferentially ascribed to N–H deformation and C=N stretching of amide groups (amide II band). Sixth, two peaks of the MHA that were generally weak or absent in the CaHA were centered between 1460 to 1450 cm-1 and between 1420 to 1410 cm-1. They are attributed to aliphatic C–H, and OH deformation and C–O stretching of phenolic OH, respectively. Seventh, a weak peak at 1387 to 1381 cm-1 is preferentially assigned to antisymmetric stretching of COO- groups. Eighth, a strong band centered between 1230 to 1210 cm-1 is generally ascribed to C–O stretching, OH deformation of COOH, and C–O stretching of aryl ethers. Ninth, a peak at 1126 to 1125 cm-1 was well pronounced in the MHA of submerged soils but was generally very weak or absent in the CaHA. It may be attributed to C–O of alcohol and ether groups. Finally, a band between 1050 to 1030 cm-1 was generally more intense in MHA than in CaHA and is attributed to C–O stretching of polysaccharide or polysaccharide-like components and/or Si–O of silicate impurities. In comparing fields, the FT IR spectra differed little between the dryland rice soil and the rice–soybean soil for either HA fraction. The HA spectra of these aerated soils differed substantially, however, from the corresponding HA spectra of the double- and triple-cropped rice soils. With increasing cropping intensity, the FT IR spectra of both HA fractions became more structured. The relative intensities of absorption by amide I (1660–1630 cm-1) and amide II (1513–1508 cm-1), aliphatic groups (2928–2922 cm-1, 2855 cm-1, and 1460–1450 cm-1), phenolic OH (1420–1410 cm-1), and alcoholic and ether groups (1126–1125 cm-1) were noticeably enhanced, whereas those of COOH groups (1719–1716 cm-1) and polysaccharides (1050–1030 cm-1) were reduced.
Fluorescence Spectroscopy
The relative fluorescence intensity and the wavelength of fluorescence emission maximum (
em) for HA samples are listed in Table 4
. Each emission mode spectrum (not shown) had a single broad peak centered at the indicated em, with gentle slopes downward on either side.
|
em was 8 to 37 nm shorter for the MHA than for the CaHA (Table 4). The RFI of the CaHA doubled with increasing cropping intensity from the dryland rice soil to the triple-cropped minus N fertilizer soil, whereas the RFI of the MHA did not change consistently. With increasing cropping intensity, em became shorter for the MHA by 30 nm, and the CaHA followed a similar but more muted trend, resulting in large differences between the em of the MHA and CaHA in the most intensively cropped soils. Fertilizer treatment had no visible effect on the spectra or RFI and em values.
High RFI and short em are generally associated with humic substances having low molecular weight, low degree of aromatic condensation and humification, and rich content of electron-donating substituents, such as hydroxyl, methoxyl, and amino groups (Senesi et al., 1991). This description better fits the MHA in general, as well as the HA from submerged soils moreso than HA from aerated soils. In contrast, low RFI and long em are ascribed to high molecular weight, high degree of humification, and rich content of condensed aromatic ring systems bearing electron-withdrawing substituents, such as carbonyl and/or carboxyl groups (Senesi et al., 1991). This description better fits the CaHA than the MHA and also fits HA from aerated soils moreso than HA from submerged soils.
Excitation fluorescence spectra (Fig. 3) are shown for the MHA (left) and CaHA (right) samples from the dryland, rice–soybean, double-cropped PU, and triple-cropped optimal N fertilizer soils. The HA spectra did not noticeably differ by fertilizer treatment. Hence spectra from the other four double-cropped soils and from the triple-cropped minus N fertilizer soil are not shown.
|
The synchronous scan spectra (not shown) of the MHA featured only one peak centered between 468 and 476 nm, whereas the CaHA spectra had a main peak at 473 to 484 nm and also a second intense peak or shoulder at 500 nm. The MHA had more fluorescence at lower wavelengths and the CaHA had more fluorescence at longer wavelengths. With increasing intensity of irrigated rice cropping, the wavelength of the main peak decreased for either HA fraction, especially for the CaHA, and the intensity of the peak at 500 nm in the CaHA spectra also decreased. Fertilizer treatment had no effect on the spectra.
Electron Spin Resonance Spectroscopy
The ESR spectra of all HA samples were qualitatively similar, featuring a sharp and narrow resonance at g 2, which were surrounded on either side by a number of other signals of various complexities and intensities. Two representative ESR spectra for the MHA and CaHA from the double-cropped PU soil are shown in Fig. 4a and 4b , respectively.
|
Among the fertilizer treatments of the double-cropped N-source experiment, free radical concentrations of the CaHA were somewhat lower in the treatments receiving organic fertilizers. This decrease might largely reflect dilution of the accumulating free radicals by formation of new SOM from the additional input of 5 to 11 t ha-1 yr-1 of the organic fertilizers (Cassman et al., 1996a).
Besides the resonance from organic free radicals, an asymmetrical resonance line of variable relative intensity centered at g 4.2 was observed in the ESR spectra of either HA fraction (Fig. 4a,b). This resonance is consistent with high spin Fe+3 cations that are strongly bound to HA-oxygenated functional groups (carboxyl, and possibly phenolic hydroxyl) in inner-sphere complexes in tetrahedral or octahedral sites with rhombic symmetry (Senesi, 1990, 1992). Although no quantitative evaluation is possible from our spectra, the relative intensity of the Fe+3 signal appeared to be higher in CaHA than in the corresponding MHA and to increase with increasing intensity of irrigated rice cropping (data not shown).
The richly structured portion of the ESR spectrum at about g = 2 (center-right side, Fig. 4a,b) can be resolved into two distinct, overlapping rigid-limit (anisotropic) patterns of the axial type. One consists of a major unresolved resonance at higher field associated with a lesser resonance at lower field resolved into a quadruplet. This pattern is typically attributed to Cu+2 cations held in inner-sphere complexes by carboxyl, phenolic hydroxyl, carbonyl, and possibly by amino group ligands arranged in square planar (distorted octahedral) sites (tetragonal symmetry) (Senesi, 1990, 1992). Similar to Fe+3, Cu+2 signals appeared to be relatively more intense in CaHA than in MHA and to increase with soil submergence (data not shown).
Values of the ESR parameters g||, |A|||, and g
for the Cu–HA complexes were restricted within specific ranges for all samples. The parallel spectroscopic splitting factor, g||, ranged from 2.289 to 2.296, the parallel hyperfine splitting constant, |A|||, ranged from 150 to 167, and the perpendicular spectroscopic splitting factor, g, ranged from 2.064 to 2.073. These values are consistent with binding sites involving three O and one N ligand atoms arranged in a square planar (distorted octahedral) coordination around the Cu+2 cation (tetragonal symmetry) (Senesi, 1990, 1992). Further, the binding of Cu+2 cations in these sites would involve a high contribution of covalency. Carboxyl and phenolic hydroxyls are the main oxygenated functional groups involved in Cu+2 binding, but the N group cannot be identified based on this information.
The remainder of the anisotropic pattern around g = 2 was comprised of a superimposition of two hyperfine octuplets consistent with VO+2 ions held in inner-sphere complexes primarily by carboxyl groups in an equatorial plane (Senesi, 1990, 1992). In contrast to Fe+3 and Cu+2, the relative intensity of VO+2 ion signals was greater in the MHA than in the corresponding CaHA, especially in the triple-cropped rice soils (data not shown). As with Fe+3 and Cu+2, the VO+2 signal intensity increased with soil submergence (data not shown).
These results indicate that transition metal ions of nutritional and/or toxicological significance, such as Fe+3, Cu+2, and V+4, can be bound preferentially to MHA (V+4) or CaHA (Fe+3, Cu+2) fractions of SOM, and that their presence in HA-complexed forms generally increases with increasing intensity of irrigated rice cropping.
Correlations between Free Radical Concentrations and Other HA Properties
The concentration of free radicals was highly correlated with other HA properties. Including all MHA and CaHA samples, the correlations were highly significant (P < 0.001) with H concentration
and C:H ratio (
, Fig. 5a,b)
. Comparable correlations were found by Riffaldi and Schnitzer (1972). Free radical concentration was also correlated with visible light absorption at 465 nm (
, P < 0.001) for those samples analyzed by Olk et al. (1996) (Fig. 5c), and with heterocyclic N proportions of total 15N NMR spectral area
for those samples measured by Mahieu et al. (2000a) (data not shown). Schnitzer and Lévesque (1979) also correlated free radical concentration with light absorption, while our correlation with heterocyclic N appears to be original. Thus, the free radical concentration was positively correlated with humification indices and negatively correlated with H concentration. All correlations were stronger for the CaHA than for the MHA (data not shown), probably because the CaHA had a wider range of free radical concentrations. The strongest correlations were for both HA fractions combined, as the younger MHA and the more humified CaHA displayed similar relationships.
|
Martin-Neto et al. (1998) found an inverse association between free radical concentrations and the ratio of light absorption at 465 nm to that at 665 nm for humic acids extracted from grassland soils varying in annual precipitation. They concluded that both parameters were related to humification in their soils. Our results oppose this conclusion: in previous studies of these samples (Olk et al., 1996, 1998), the light absorption ratio did not vary regularly with irrigated rice cropping intensity despite the regular variation in free radical concentration and other humification indices reported here.
|
Discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionREFERENCES |
|---|
More specifically, with increasing soil submergence under intensive cropping, the HA fractions became less polycondensed and oxidized with higher S and H and lower O concentrations, lower levels of free radicals, and fewer COOH groups. The semiquantitative nature of FT IR, fluorescence, and ESR spectroscopies does not allow precise comparisons across samples, but amide or amino, hydroxyl, and methoxyl groups also appeared to be more abundant in submerged soils. Greater H concentrations in HA from submerged soils might reflect an accumulation of proteinaceous, alkyl, and methoxyl groups and a lower abundance of condensed aromatic rings (Yonebayashi and Hattori, 1989).
In a related study using the same analyses as those herein (Olk et al., 1999), the MHA and CaHA extracted from two fertilizer treatments in a double-cropped lowland rice field experiment replicated at three sites in the Philippines also had comparable properties to those noted here for the double- and triple-cropped fields. Thus, MHA and CaHA forming under submerged conditions appear to be incompletely humified across a range of local site and climatic conditions.
Evidence that the MHA was more aliphatic and less humified than the CaHA included its fewer acidic groups, smaller C:N and C:H ratios, richer contents of amides and carbohydrates, and lower content of COOH, high fluorescence intensity and short wavelength of fluorescence emission maximum, and low concentration of organic free radicals. The FT IR spectra of the MHA were visibly more structured than those of the CaHA, suggesting greater chemical diversity.
Comparisons with previously published values for HA properties also highlight the incompletely decomposed nature of the HA of our submerged soils. Elemental concentrations of the MHA from the submerged soils approached the upper limits of values normally found for H and S in HA (32–62 and 1–15 g kg-1, respectively; Stevenson, 1994) and the lower limit of typical O concentrations (328–383 g kg-1), while the O concentration of the CaHA also neared this lower limit. By contrast, the corresponding values for HA from the two aerated soils were well within the normal ranges. Total acidity of the MHA and COOH levels of both HA fractions from the submerged soils were generally well below the normal ranges for HA of tropical soils (6.2–7.5 and 3.8–4.5 mol kg-1, respectively; Stevenson, 1994). However, three out of four functional group concentrations for HA of the dryland soil were also slightly below these ranges. Phenolic OH concentrations of the MHA from the submerged soils were generally below the normal range (2.2–3.0 mol kg-1), while those of the CaHA were above this range.
A comparison of our FT IR spectra with typical IR spectra of HA from soils and other sources as classified by Stevenson and Goh (1971) and reported by Stevenson (1994) shows that (a) the spectra of HA from the aerated soils, especially the CaHA, can be classified as Type I spectra that typify Mollisols and other common aerated soils; and (b) the spectra from the submerged soils, especially the MHA, are more similar to Type III spectra, especially those for HA from lake sediments.
Other IR studies associated increasing humification with effects on HA properties that are comparable to those identified here. Yonebayashi and Hattori (1989) used IR to note greater prominence of amide N signals in HA from wetter soils. Tomikawa and Oba (1991) used IR to identify signals for amides, polysaccharides, aliphatic compounds, and lignin-like substances in less humified HA fractions of larger particle size. Niemeyer et al. (1992) used FT IR to observe increasing carboxyl signals in peats that were increasingly humified.
The increasing prominence of amide signals in the FT IR spectra of the MHA samples with soil submergence opposed the simultaneous trend of decreasing N concentration. Analysis of these MHA samples by 15N NMR also associated increasing soil submergence with greater proportions of N as amide simultaneously with decreasing free amino-N proportions (Mahieu et al., 2000a). Greater proportions of amide-N in HA of the submerged soils might imply less fragmentation of proteinaceous molecules before their incorporation into SOM, resulting in stabilization of longer polypeptide chains. In aerated soils by comparison, vigorous fragmentation of proteinaceous molecules before their incorporation into SOM could result in stabilization of shorter-chained polypeptides or perhaps individual amino acids.
Sulfur accumulation did not have a simple relationship with humification. Evidence for decreasing S levels with humification includes the wider C:S and N:S ratios of the CaHA relative to the MHA and the narrowing ratios with increasing soil submergence. But the lower S levels in N-fertilized treatments than in minus N fertilizer treatments were not matched by any evidence for increased humification from either free radical levels, visible light absorption (Olk et al., 1996, 1998), or ratios of monoester P:diester P (Mahieu et al., 2000b). Application of optimal N fertilizer rates might well affect S concentrations through additional means besides humification.
Previous associations of S concentration with humification were also contradictory. Freney (1986) believed that S is more resistant to mineralization than C or N and will accumulate with humification. However, Bettany et al. (1979) found decreasing levels of organic S in a humified HA fraction with increasing humification and hypothesized that S contained in aliphatic side chains is cleaved during humification.
High correlations between free radical concentrations and other humification indices suggest that free radical formation is slow, associated with humification, and not affected by soil submergence. Decreasing free radical concentrations with increasing rice cropping intensity might be best explained as a dilution of aging SOM that contains the free radicals by increasingly larger amounts of poorly decomposed young SOM. This accumulation would result from decreased mineralization rates under the submerged conditions (Witt et al., 2000) and from increased input rates of crop residues. Despite abundant lignin residues the low free radical concentrations in the triple-cropped soils indicate that partial lignin degradation does occur in submerged rice soils, since fresh lignin is rich in free radicals (Wikander and Nordén, 1988).
Determination of phenolic OH through total acidity suggested that the CaHA of the submerged soils had more phenolic OH than did the corresponding MHA, and that MHA extracted from the aerated soils generally had more phenolic OH than did MHA from submerged soils. However, opposing trends were suggested here by FT IR, previously by 13C NMR spectroscopy (Olk et al., 1996), and by tetramethylammonium hydroxide thermochemolysis (D.C. Olk, unpublished data). The accuracy of phenol measurements is uncertain both for total acidity (Swift, 1996) and 13C NMR (Preston and Schnitzer, 1987), but CuO oxidation of soils from these fields found substantially greater phenol contents in the submerged soils than in aerated soils (Olk et al., 1996).
Assuming that phenols are more prevalent under submerged conditions than aerated conditions, their potential roles in nutrient cycling in submerged rice soils merit further attention. Their capacity to react with N compounds under controlled conditions is well known (e.g., Loomis and Battaile, 1966; Flaig et al., 1975). In ongoing work we are studying the effect of phenols on the availability of native soil N in submerged rice soils.
In addition, phenols are considered important agents for Cu complexing by SOM (Lewis and Broadbent, 1961; Senesi, 1992). Thus, their prevalence in submerged soils may well contribute to the greater abundance of Cu+2 complexes and perhaps also to the greater number of Fe+3 complexes in these HA samples. Further, Cu-specific adsorption can promote aggregation of polyaromatic molecules (Bartoli et al., 1987), which might affect cycling of other SOM-bound nutrients. Carboxyl groups can also complex Cu+2, Fe+3, and VO+2, but their contribution to increased metal complexation in submerged soils is uncertain, given their decreasing relative abundance with increasing soil submergence as found here and previously by 13C NMR (Olk et al., 1996).
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 4, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. F. Pampolino, E. V. Laureles, H. C. Gines, and R. J. Buresh Soil Carbon and Nitrogen Changes in Long-Term Continuous Lowland Rice Cropping Soil Sci. Soc. Am. J., May 1, 2008; 72(3): 798 - 807. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. He, T. Ohno, B. J. Cade-Menun, M. S. Erich, and C. W. Honeycutt Spectral and Chemical Characterization of Phosphates Associated with Humic Substances Soil Sci. Soc. Am. J., August 22, 2006; 70(5): 1741 - 1751. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
 |
|
 F. Fares, A. Albalkhi, J. Dec, M. A. Bruns, and J.-M. Bollag Physicochemical Characteristics of Animal and Municipal Wastes Decomposed in Arid Soils J. Environ. Qual., July 5, 2005; 34(4): 1392 - 1403. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. B. Ve, D. C. Olk, and K. G. Cassman Characterization of Humic Acid Fractions Improves Estimates of Nitrogen Mineralization Kinetics for Lowland Rice Soils Soil Sci. Soc. Am. J., July 1, 2004; 68(4): 1266 - 1277. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Mahieu, D.C. Olk, and E.W. Randall Multinuclear Magnetic Resonance Analysis of Two Humic Acid Fractions from Lowland Rice Soils J. Environ. Qual., March 1, 2002; 31(2): 421 - 430. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
