HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Qualls, R. G. Articles by Wershaw, R. L. Search for Related Content PubMed Articles by Qualls, R. G. Articles by Wershaw, R. L. GeoRef GeoRef Citation Agricola Articles by Qualls, R. G. Articles by Wershaw, R. L. Related Collections Humic Substances Forest Soils

DIVISION S-7—FOREST & RANGE SOILS

Formation and Loss of Humic Substances During Decomposition in a Pine Forest Floor

^a Dep. of Environmental and Resource Sciences, Univ. of Nevada, Reno, NV 89557
^b U.S. Geological Survey, Box 25046, MS 408, Denver, CO 80225

^* Corresponding author (qualls{at}unr.edu)

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
Since twice as much C is sequestered in soils as is contained in the atmosphere, the factors controlling the decomposition rate of soil C are important to the assessment of the effects of climatic change. The formation of chemically resistant humic substances might be an important process controlling recycling of CO₂ to the atmosphere. Our objectives were to measure the rate of formation and loss of humic substances during 13 yr of litter decomposition. We placed nets on the floor of a white pine (Pinus strobus) forest to separate each annual layer of litter for 13 yr and measured humic substance concentration using NaOH extraction followed by chromatographic fractionation. The humic acid fraction increased from 2.1% of the C in litterfall to 15.7% after 1 yr. On a grams per square meter (g m^-2) basis the humic substance fraction increased during the first year and then declined, with a half decay time (t_1/2) of 5.1 yr, which was significantly slower than the bulk litter (t_1/2 = 3.9 yr). The carboxylic C concentration estimated from ¹³C nuclear magnetic resonance (NMR) increased in the litter over time, though total mass of carboxylic acid C in the forest floor also declined over the 13-yr period (t_1/2 = 4.6 yr). While humic substances in the forest floor decomposed at a somewhat slower rate than bulk litter during Years 1 to 13, they decomposed much faster than has been calculated from ¹⁴C dating of the refractory fraction of organic matter in the mineral soil.

Abbreviations: AMU, atomic mass unit • ¹³C NMR, carbon-13 nuclear magnetic resonance • CP/MAS, cross polarization, magic angle spinning • DOC, dissolved organic carbon • DP/MAS, direct polarization, magic angle spinning • t_1/2, half decay time

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
APPROXIMATELY TWICE AS MUCH C is stored in the soil as in the atmosphere (Lal et al., 1995). Consequently, the factors that control the storage of C in the soil are among those that play an important role in regulating the concentration of CO₂ in the atmosphere. The first-order decomposition rate constant of lignin, one of the slowest decomposing substances present in original plant material has been reported as 0.18 yr^-1 a half decay time (t_1/2) of 3.8 yr in a temperate soil (Lynch, 1991). However, the observed residence time of C is on the order of hundreds or even thousands of years in some mineral soils (Campbell et al., 1967). The factors that retard the mineralization of C originating from plant compounds are largely responsible for the storage of C in soil and the chemical transformation of C into humic substances may help explain this decrease in biodegradability. Humic substances are regarded as more resistant to decomposition than compounds directly synthesized by plants (Stevenson, 1994). Other hypotheses that might explain the observed persistence of C in soil include: physical protection from microbes within interstices too small for access by microbial cells, and formation of organomineral complexes that tend to protect against enzymatic attack. These hypotheses are not mutually exclusive. For example, adsorption of humic substances to Fe oxyhydroxides (Stevenson, 1994) and subsequent accretion of layers of the complexes might represent a combination of chemical and physical protection from enzymes.

Humic substances are operationally defined, but have been demonstrated to have characteristics distinct from compounds synthesized directly by plants (Stevenson, 1994). There are several theories of how humic substances are formed, but most studies agree that they are comprised of large (>700 atomic mass units [AMU]) molecules, with carboxylic acid, phenolic hydroxyl, and aromatic functional groups (Stevenson, 1994). Operationally, humic substances have been defined as substances extractable in NaOH. With the recognition that there might be NaOH soluble substances that are not humic, subsequent steps involving precipitation of humic acids at pH 1 and adsorption of fulvic acids to a hydrophobic resin have been added (Swift, 1996). Other steps have been added by Leenheer (1981) to further separate aquatic dissolved organic matter into humic substances, hydrophylic acids, hydrophobic neutral substances, hydrophilic bases, and hydrophilic neutral substances.

It is generally believed that during humification the carboxylic acid functional group concentration of organic matter increases (Stevenson, 1994). The carboxylic acid functional groups in humic substances are particularly important in controlling acidity, cation exchange, and translocation of metals in soils. Several studies have compared the humic substance concentration or functional group concentration (using cross-polarization, magic angle spinning [CP/MAS] ¹³C NMR) of various soil horizons that reflect increasing degrees of decomposition (Kögel-Knabner et al., 1991; Zech et al., 1992; deMontigny et al., 1993). Other studies have reconstructed long-term rates of decomposition from chronosequence studies (Harmon et al., 2000). The actual rate of development and loss of humic substances or functional group characteristics in forest soils have not, to our knowledge, been measured directly over the long term (e.g., 13 yr).

Our objectives were: (i) to measure the rate of formation and loss of humic substances in fallen litter as a function of time, and (ii) to estimate the rate of formation and loss of carboxylic functional groups in a forest soil organic matter as a function of time. To accomplish these objectives, we laid nets on the floor of a Pinus strobus forest for 13 yr to keep annual layers separate, measured total C loss, analyzed the concentration of humic substances by NaOH extraction, and chromatographic fractionation, and estimated the change in concentration of carboxyl C by CP/MAS ¹³C NMR.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
Separation of Annual Litter Layers
The research site was a pine plantation established in 1956 on the north-facing Watershed 17 at the Coweeta Hydrologic Laboratory, in the Nantahala Mountains of North Carolina, a long-term ecological research site of the National Science Foundation. Mean annual precipitation is 194 cm and mean annual air temperature is 12.6°C (Swift et al., 1988). The forest floor was a mor type. The soil was an Evard-Cowee gravely loam, a fine-loamy, parasesquic or oxic, mesic Typic Hapludult. Texture in the 0- to 20-cm depth was 59% sand, 22% silt, and 19% clay (McGinty, 1976). The organic matter concentration was 5.9% in the 0 to 30 cm, and 5.0% in the 30- to 60-cm mineral soil depths (McGinty, 1976). The pH was 4.2 in the Oa horizon (this study) and 4.7 in the 0- to 10-cm depth of the A horizon. In 1970-1971 litterfall was measured as 388 g m^-2 yr^-1, 318 g m^-2 yr^-1 of which was foliar (Cromack and Monk, 1974).

Beginning in August, 1984, we laid a 91 by 91 cm square of nylon mesh, with openings measuring 2 by 4 mm, on five plots each year until 1997 (the technique of Jorgenson et al., 1980). Consequently, the autumn 1984 litterfall was the first layer delineated. A 15-cm high "fence" of screen surrounded each plot to prevent movement of litter by wind. The pliant nylon mesh was chosen because it allowed litter to lie in a more natural manner than stiffer fiberglass window screen. Litterfall was measured every 3 (autumn) to 6 wk beside each of the five plots in the 1985-1986 litterfall period and again in the 1997-1998 period. Litterfall was used as the standing stock for Time 0. Measurement of decomposition rate by this method requires a litterfall rate that is either known for each year or that it remains constant over the period (Jorgenson et al., 1980). Foliar litterfall in 1997-1998 was 218 g m^-2 yr^-1 C ± 25 (sd). In 1985-1986 foliar litterfall was 204 ± 37 g m^-2 yr^-1 C, and was not significantly different than in 1997-1998 (P 0.05). Consequently, it was assumed that litter inputs did not exhibit any long-term trend throughout the 13-yr period, and thus the decline in mass in the annual layers could be used as a measure of decomposition.

All annual layers were separated and collected in August 1998, giving 13 yr of separated, annual layers (the 1984 layer was not used). When litter was harvested, we noted that even in the advanced stages of decay, the particles were bound together by a network of either living or dead mycelia. Litter was carefully separated from nets in the laboratory and air dried at about 24°C to a constant mass and weighed. Woody material and cone debris were separated from foliar litter but in this paper we have only analyzed the foliar component of the litter. Subsamples were ground in a small ball mill for analysis. Subsamples were also dried at 70°C to determine moisture content, and then combusted to determine ash-free dry mass. Carbon and N concentration were measured using a Perkin-Elmer 6400 C-H-N analyzer (Perkin-Elmer, Norwalk, CT).

We calculated the loss of C from the first year of decomposition during the first year as the litterfall (in g m^-2) minus the mass of C (in g m^-2) laying on the 1997 net that was approximately 1 yr old. The decomposition occurring from previous Years t to Year t - 1 was calculated as the C mass in Layer t minus C mass in the Layer t - 1.

Flux of Fine Particulate Carbon
Measurement of decomposition rates using this technique depends on the assumption that the migration of fine particles through the nets and annual layers was small. We measured the flux of fine particulate C through the net by placing 56-mm Whatman glass fiber filter discs (Whatman International Ltd, Maidstone, UK) just below the undersurface of the nets in 1997. The discs were placed under layers of 1-, 2-, 3-, 4-, 5-, 6-, 9-, 11-, and 13 yr-old litter in a nonoverlapping arrangement so that they would represent the cumulative flux from all annual layers above each filter disc. One set of filter discs was removed after 14 d to serve as a control for the chance that the disturbance in placing the filters would cause deposition of particles on the surface. A second set of filters was left in place for 1 yr. For downward-moving fine particles to be deposited in the matrix of the filter it was important that the filters not repel water. In the laboratory, we observed the ability of the moist filters to absorb droplets of water clinging to particles of litter touching the filter and they appeared not to repel droplets of water.

The filters were collected carefully and wrapped in aluminum foil. The filters were dried at 105°C, weighed to the nearest 0.1 mg, combusted at 450°C, and the loss of weight was measured. The organic C deposited on the filter was estimated as 50% of the loss of mass on combustion. The C deposited on the control filters was subtracted. To calculate the annual flux of fine particulate C from Layer x to Layer x - 1, we assumed that most of the flux into Layer x - 1 originated from Layer x, and not those laying several layers above. We based this assumption on the finding that the flux of fine particulate C actually declined in the older layers as a percentage of the mass in the layer above the filter.

Extraction and Fractionation
Two methodological tests were performed using the NaOH extraction procedure. First we extracted three replicate 0.6-g samples of a composite of the 13-yr old litter with seven sequential extractions with 30 mL of 0.1 M NaOH under N₂ for 16 h each. We measured the cumulative mass of C extracted by each sequential NaOH extraction under either oxic or anoxic conditions to determine the optimal number of extractions.

We also determined whether unaltered lignin might be dissolved during the NaOH extraction, thus being misinterpreted as humic substances, and whether it might be removed by precipitation of the NaOH extract at pH 7 and adsorption on XAD-8 resin. We used a commercial lignin preparation (Organosolv Lignin, Aldrich Chemical Co., Milwaukee, WI) with a number-average molecular weight of 800 AMU and a weight-average molecular weight of 3500 AMU (Aldrich Chemical Co., product information circular, Milwaukee, WI). We extracted it using 0.1 M NaOH under either ambient oxic or anoxic, N purged conditions. Then we neutralized the extract rapidly to pH 7 with HCl, allowed it to sit 24 h, centrifuged the extract, and pumped the supernatant through XAD-8 resin and eluted the adsorbed substances with 0.1 M NaOH.

Polyphenol concentrations of the litter samples were determined by the Folin-Ciocalteau method using three sequential extractions with 50% water and 50% methanol (Waterman and Mole, 1994).

Subsamples of the new litterfall, and of 1-, 3-, 5-, 7-, 9-, 11-, and 13-yr-old litter were extracted by the procedure outlined in Fig. 1 . This procedure represents a slightly modified version of the procedure for analysis of humic substances of the International Humic Substances Society (Swift, 1996) supplemented by Leenheer's (1981) more detailed fractionation procedure. Samples consisting of the equivalent of 0.3 g C were extracted with 30 mL of 0.1 M NaOH using three sequential 16-h extractions under anoxic conditions in a N₂ purged bag. After centrifugation, the solutions were quickly neutralized to pH 7, allowed to sit for 24 h and then centrifuged again. The three supernatants were combined, diluted five-fold, and a 250-mL sample was then pumped through 3.75 mL XAD-8 resin. The weak hydrophobic acid fraction (phenolic fraction) was eluted with 0.1 M NaOH. The effluent was acidified to pH 1, allowed to sit 24 h without further pH adjustment, and then centrifuged. Because some particles were not removed by centrifugation, the supernatant was filtered through a 0.45-µm membrane filter to remove the humic acid fraction. The supernatant was again pumped through XAD-8 resin at about pH 1 and the fulvic acid fraction was then eluted with 0.1 M NaOH. The effluent was then pumped through cation- and anion-exchange columns as described by Leenheer (1981). The hydrophilic acid fraction was calculated as the fraction of the total C removed by the anion exchange column. Subsamples were taken for dissolved organic C (DOC) analysis after each step of the procedure. The sum of the humic acid (C precipitated at pH 1) and the fulvic acid fractions were interpreted as the humic substance fraction.

View larger version (42K): [in this window] [in a new window]   Fig. 1. Extraction and fractionation procedure.

¹³C NMR Analysis
The carboxylic acid C concentration has been used as an index of humification of soil organic matter. We used area of the peak in direct polarization, magic angle spinning (DP/MAS) or CP/MAS ¹³C NMR spectra from 165 to 190 ppm as a measure of carboxylic acid and amide functional groups.

Solid-state CP/MAS ¹³C NMR spectra were obtained for the litterfall, 1-, 9-, and 13-yr-old litter. In addition, spectra were obtained for the humic acid and humin fractions of the litterfall and the 9- and 13-yr-old litter. A single spectrum was acquired for each sample, where the ‘sample’ was a composite of 2 g from each of the five replicate plots. Solid-state CP/MAS ¹³C spectra were measured at 50.298 Mhz on a 200 Mhz Chemagnetics CMX spectrometer (Varian Inc., Palo Alto, CA) with a 7.5 mm-diam. probe. The spinning rate was 5000 Hz. The acquisition parameters for the solid samples were a contact time of 1 ms, pulse delay of 1 s, and a pulse width of 4.5 µs for the 90° pulse (Wershaw et al., 2000). The number of free induction decay signal acquisitions ranged from 1552 for the litterfall to 5229 for the 13-yr-old litter. A line broadening of 100 Hz was applied in the Fourier transformation of the free induction decay data. In addition, DP/MAS ¹³C spectra were measured on the new litterfall and 13-yr-old litter samples. The DP/MAS ¹³C spectra were obtained using a 37° pulse and a 48-s pulse delay, based on optimization experiments reported in Wershaw et al. (2000). The number of free induction decay signal acquisitions was 468 for litterfall and 3020 for 13-yr-old litter.

The NMR spectra were divided into chemical shift regions as follows: aliphatic C, 0 to 45 ppm; ether C, 45 to 62 ppm; aliphatic-OH C, 62 to 90 ppm; anomeric C, 90 to 110 ppm; aromatic C, 110 to 165 ppm; carboxyl, amide, and ester C, 165 to 190 ppm; ketone C, 190 to 230 ppm. The total area under the spectrum and areas within each chemical shift region were measured by cutting and weighing. The area between 165 to 190 ppm is attributed to carboxylic acid C, amide C and ester C. We assumed that the ester C contribution was small compared with the carboxyl C. Antweiler (1991) found about 0.6% of the mass of Suwannee fulvic acid was hydrolyzable ester C.

Carboxylic Functional Group Concentration
Because the carbonyl C of the carboxyl and amide functional groups is unprotonated, reduced cross-polarization efficiency causes an underestimation of the signal because of carbonyl C relative to other C atoms (Mao et al., 2000). The DP/MAS ¹³C spectrum is not subject to that underestimation (Kinchesh et al., 1995). We calculated the ratios of the relative peak areas between 165 and 190 ppm for the DP/MAS spectra divided by the corresponding CP/MAS spectra of the new litterfall and the 13-yr-old litter. This ratio was 1.2 for new litterfall and 1.4 for the 13-yr-old litter. To correct the CP/MAS relative peak areas for the other samples, we multiplied by a factor of 1.3 to give a corrected measure of the carboxyl plus amide C concentration.

Using the DP/MAS peak areas between 165 and 190 ppm directly, or using the relative peak area that had been corrected for its underestimation in the CP/MAS spectra, we calculated the relative area due to carboxyl C alone, without the amide C, as follows. We estimated the proportion of amide C independently using elemental analysis of C and N. Using ¹⁵N NMR, most N in humic acids has been attributed to amide N. We assumed 90% of the N was amide N and terminal amine N based on a study of NaOH soil extracts (Knicker et al., 1993). Thus we assumed that:

[1]

where C_amide is the moles of amide C and N_total is the moles of total N. The relative peak area for amide C (A_{amide C}/A_total) can thus be estimated by:

[2]

where N/C is the molar ratio of total N/total C, based on elemental analysis of the sample. Now the relative peak area of the 165- to 190-ppm range can be divided into:

[3]

where A_COOH/A_total is the relative area attributed to carboxyl C.

Substituting Eq. [2] into Eq. [3] and solving for A_COOH/A_total

[4]

Note that applying this calculation to separate the carboxyl and amide relative peak areas without having corrected for an underestimation in the CP/MAS spectra would result in a further underestimation in the carboxyl concentration.

Finally the ratio of carboxyl C/total C was multiplied by the mass of C remaining in each of the five replicate plots of litterfall, (Year 0), and litter of ages 1, 9, and 13 yr to obtain an estimate of the rate of loss of carboxyl C from forest floor during decomposition.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
Methodological Tests
The amount of C extracted anoxically from the 13-yr-old litter as a function of the number of sequential extractions is shown in Table 1. Three sequential extractions removed 90.2% as much C as seven extractions so we chose three extractions as a balance between complete extraction and possible additional hydrolysis of otherwise insoluble lignin or other components.

View larger version (19K): [in this window] [in a new window]   Fig. 2. Loss of C and humic substances from the forest floor as a function of time. Each point represents one of five replicate plots. Data for the first year is connected by a dashed line. Data for Years 1 to 13 are represented by a solid exponential regression line and the regression statistics refer to data from 1 to 13 yr. As a statistical rationale for excluding the first year decomposition from the regression, we found that neither the 95% nor the 99% confidence intervals for the data at x = 0 yr included the intercept predicted from the regressions in Fig. 2.

The changes in the standing stock of humic substances over time were a product of a decline in the mass of each annual layer and (excepting the first year) an increase in the percentage of the total C in the humic substance fraction (Table 4). The humic acid fraction of total C increased significantly (P < 0.01) from 2.1% in new autumn litter to 15.7% in 1-yr-old litter and then increased to 26.3% after 13 yr. The fulvic acid (hydrophobic acid) fraction varied from about 7.5% in new litter to 9.9% in 13-yr-old litter but there was no significant trend with time.

In our procedure, the phenolic fraction was that which failed to precipitate at pH 7, but was hydrophobic enough to adsorb to XAD-8 resin at pH 7 and desorb at alkaline pH (Fig. 1). Polyphenols occur in this fraction (Leenheer, 1981; Qualls and Haines, 1991), but phenolics originating from lignin lysis might also occur in this fraction (see Table 2). The polyphenols in this fraction might be considered as part of the humic substances and they would likely be included in the fulvic or humic acid fraction of the less detailed procedure outlined by Swift (1996). We simply report this fraction separately because of the possibility that it may contain lysed lignin fragments and because polyphenols are produced as part of the original plant tissue. This phenolic fraction (analyzed by chromatography) ranged from 6.0% of the total C in freshly fallen foliar litter to 2.8% in 11-yr-old material, and there was a significant decrease with age of the litter (Table 4). The analysis of the polyphenol concentration of litter samples using the Folin-Ciocalteau method was similar to that of the phenolic fraction isolated by chromatography for new litter. However, in older litter there was little correspondence with the phenolic fraction. The Folin Ciocalteau method measures reducing equivalents (Waterman and Mole, 1994) rather than pH-dependent ionization. The partial oxidation of polyphenols over time might help explain this lack of correspondence.

Carboxyl Carbon Remaining in Litter Versus Age
Carboxyl C remaining in the litter declined exponentially with age of the litter with a half decay time of about 4.6 yr (Fig. 3) , slightly faster than the humic substance fraction. This rate of loss was significantly different than that of the total forest floor C (Fig. 2), as tested by analysis of covariance of the slopes of the ln(y) versus time curves. In contrast to the loss of total C (Fig. 2), the loss of carboxyl C was not disproportionately great during the first year, so we included the data for fresh litterfall in the regression analysis. The regression model using exponential decay (Carboxyl C = 12.6 exp [-0.15t]) was highly significant (P < 0.001) and the predicted value for the 9-yr-old litter was within the 99% confidence interval for the mean of the data at 9 yr. Nevertheless, the model must be used with caution because only four times are represented. The loss of carboxyl C in grams per square meter per year (m^-2 yr^-1) was a product of the loss of mass from the litter and an increase in the carboxyl C fraction from 5.7 to 10.0% over 13 yr (Table 5, Fig. 4) . In fact, the carboxyl C concentration of both the whole litter and the humic acid fractions increased sharply during the first year just as did the percentage of C in the percentage of humic acid fraction using the fractionation procedure (Tables 2 and 5). The increase in carboxyl concentration during the first year of decomposition, from 5.8 to 7.7%, was because of an increase in the carboxyl content of the humic acid fraction from 6.5 to 11.5% during the first year (Table 5).

View larger version (12K): [in this window] [in a new window]   Fig. 3. Loss of carboxylic C functional groups from the forest floor. Each data point represents one of five replicate plots.

View larger version (33K): [in this window] [in a new window]   Fig. 4. (A) Carbon-13 direct polarization magic angle spinning (13C DP/MAS) and (B) 13C cross polarization magic angle spinning (CP/MAS) and spectra for litter samples. Spectra for the humic acid and humin fractions are not shown.

The high concentration of carboxyl functional groups is one of the most important properties of humic substances. They are responsible for many of the important ecological roles of humic substances such as cation exchange, the acidity of organic soil horizons, and the formation of organometallic complexes (Stevenson, 1994). Our data illustrate that while the concentration of carboxyl C increases as part of the humification process, there is a net loss of carboxyl C because of the decreasing mass of organic matter over time. It is likely that this loss represents a combination of the loss of carboxyl C originally present in new litter and production of new carboxyl groups during oxidative degradation or microbial synthesis. Microscopic observation of the various layers revealed abundant remains of fungal mycelia binding particles together.

We examined the correlation between concentration of carboxyl C (Table 5) and four characteristics believed to reflect the extent of decomposition: (i) the percentage C remaining (R² = 0.97, P < 0.05), (ii) the C/N ratio (R² = 0.98, P < 0.05) (iii) the aliphatic C/aliphatic-OH C ratio (Baldock et al., 1997) (R² = 0.74, P < 0.05) and (iv) the percentage of humic substance C concentration (R² = 0.95, P < 0.05). Thus the content of carboxyl C is well correlated with other traditional measures of the extent of decomposition.

Comparison of Carboxyl Carbon Concentrations with Other Studies
The vast majority of solid state ¹³C NMR analyses of litter and soil organic matter have been performed using the CP/MAS technique, particularly because of the very long signal acquisition times necessary for solid state ¹³C DP/MAS measurements. We found that it underestimated the peak area from 165 to 190 ppm by 20 to 40%. A study by Mao et al. (2000) also reported that the CP/MAS ¹³C NMR of peat and humic acid samples significantly underestimated carboxyl C concentration relative to DP/MAS measurements.

Other studies that have measured the relative peak area in the carbonyl region (about 165–190 ppm) of different layers of coniferous forest floors have found results very similar to our CP/MAS data. Baldock and Preston (1995), using CP/MAS ¹³C NMR to examine forest floor layers in a Red pine (Pinus resinosa) plantation, found carbonyl C increased from 5% in fresh litter to 8% in the Oa horizon. In a comparison of the forest floor of six different forests with Spruce, Beech, and Pine vegetation, Zech et al. (1992) found an average of 7% carbonyl C in the Oi horizon and 9.7% in the Oa horizon. Again, these values are similar to those found in our study. In three of the same sites, Kögel-Knabner et al. (1991) found that the relative peak area from 160 to 220 ppm in the humic acid fractions increased from 6% in the Oi horizon, to 11% in the Oe horizon, and 16% in the Oa horizon. These results were consistent with our conclusion that most of the increase in carboxyl C with age occurred in the NaOH-extractable fraction. In coniferous forests, deMontigny et al. (1993) found that the relative peak area between 159 and 185 ppm increased from 6% in litter to 8% in the Oa horizon in one forest type, and from 4 to 9% in another forest type. Lorenz et al. (2000) found only a small and inconsistent difference in relative peak areas in the 165- to 190-ppm range for Black spruce and Norway spruce needles during 10 to 12 mo. of decomposition. However, only 20 to 25% of the mass was lost over this relatively short period.

Changes in the Concentration of Other Functional Groups with Age
Comparing the DP/MAS with the CP/MAS spectra for the 0- and 13-yr-old litter, the percentage peak area for aliphatic C, (0–45 ppm), ether plus secondary amide C (45–62 ppm), aromatic C (110–165 ppm), carboxyl plus amide C (165–190), and ketone C (190–230 ppm) were higher, while the percentages for aliphatic-OH C (62–90 ppm) and anomeric C, (90–110 ppm) were correspondingly lower.

The percentage concentration of aromatic C changed little as the litter aged (Table 5, Fig. 4), suggesting that the aromatic C was being decomposed at about the same rate as the bulk litter. The aromatic C present at any time likely represents: (i) plant derived lignin and polyphenols, (ii) aromatic C in humic substances modified from lignin or polyphenols, or (iii) aromatic C synthesized during decomposition and stabilized in the humic fraction. It is possible that the lack of a large change in aromatic C percentage represents a balance between loss of lignin C and formation of humic-associated aromatic C. Some studies have noted an increase of aromatic C in residues of increasing age, presumably as carbohydrates are lost faster than aromatic compounds (deMontigny et al., 1993). Baldock et al. (1997) found an increase in aromatic C percentage from 20% in Red Pine fresh litter to 25% in the Oa and A horizons. In contrast, Zech et al. (1992) found little change in aromatic C in forest floor layers of increasing degree of decomposition.

In our study, the anomeric C percentage, associated with carbohydrates, decreased with age of the litter, but the decrease was small, meaning that the loss of anomeric C was only slightly faster than the loss of C in the litter as it decomposed (Table 5). Pinus strobus litter contains 31% lignin (Cromack and Monk, 1974), a relatively high percentage that may physically protect cellulose from decomposition.

In comparing the DP/MAS and CP/MAS relative peak areas for aliphatic C (0–45 ppm), the peak area for DP/MAS is considerably larger in the new litterfall and slightly larger in the 13-yr-old litter (Fig. 4). This suggests that even in the air-dried litter, some of the aliphatic C, such as waxes, resins, and lipids, may be in a liquid-like state, or exhibited a high degree of molecular motion, because the CP/MAS technique would not detect liquid-state C. The percentage of aliphatic C increased with age in litter from 15.6 to 23.9% using CP/MAS and from 22.7% to 25.7% using DP/MAS (Table 5). Every study we reviewed has observed a comparable increase in the percentage of alkyl C in forest floor layers of increasing degree of decomposition (Kögel-Knabner et al., 1991; Zech et al., 1992; deMontigny et al.; 1993; Baldock et al., 1997). For example, using CP/MAS, Baldock and Preston (1995) found an increase of from 16% alkyl C in Red pine fresh litter to 19% in the Oa horizon and 21% in the A horizon.

Another commonly observed characteristic of litter decomposition is a decrease in the C/N ratio (Jorgensen et al., 1980). In our study, the C/N ratio (mass ratio) decreased from 63 in litterfall to about 25 after both 9 and 13 yr of decomposition (Table 5). The C/N ratio for the entire O horizon was 42 (this study). The C/N ratio of the 13-yr-old litter was similar to that of the 0- to 10-cm depth of the A horizon (McGinty, 1976). By 13 yr, the C/N ratio was about the same in the bulk litter as in the humin fraction, meaning that N was not more concentrated in NaOH-soluble fractions. The incorporation of N into humic substances is another characteristic of the humification process (Stevenson, 1994).

Fate of Carbon Lost from the Forest Floor Layers
Carbon lost from the forest floor layers delineated by nets could occur by: (i) mineralization, (ii) leaching of DOC, (iii) downward movement of fine particles, or (iv) export by invertebrates. The downward movement of particles has been shown to be a minor form of export. We also assume the export by invertebrates was minor because earthworm (Oligochaeta) casts were rarely found and only one earthworm was found in 48 soil pits dug in a neighboring watershed. Furthermore, downward movement of frass and invertebrate remains would have been measured as fine particulate matter export. Respiration of assimilated C by microbes and invertebrates would have been included as "decomposition" or mineralization. Invertebrates were not excluded from the layers since they were exposed from the surface during the first year, the mesh was 2 by 4 mm which is larger than most litter invertebrates, and millipedes (Diplopoda) were frequently found in the internal litter layers.

The leaching of DOC from the forest floor was measured as 40.5 g m^-2 yr^-1 on a nearby watershed dominated by deciduous species in 1986-1987 (Qualls et al., 1991). This flux is equivalent to 23.7% of foliar litterfall. Essentially all of this leaching originated from the Oi horizon of the forest floor. We made an estimate of the leaching of DOC from the white pine forest floor by assuming that the leaching of DOC from both the pine and deciduous forest floors was proportional to the amount of water soluble C in litterfall. Our measures of the water soluble C in new pine and deciduous litter were 20.4 and 21.1%, respectively, of the total C in litter.

[5]

Using this equation, we estimated a net flux of 49.7 g m² yr^-1 DOC from the pine forest floor. This annual flux is equivalent to 22.8% of foliar litterfall. Our water extractions indicated that 10.4% of the C in 1-yr-old litter was water-soluble. Although some of this water-soluble C may have been produced during decomposition, it appears that not all water-soluble C was lost during the first year of decomposition, probably because of desorption equilibria (Qualls, 2000). Consequently, a portion of the large amount of C lost in the first year (Fig. 2) was probably because of leaching in addition to mineralization.

Using three assumptions: (i) that the forest floor was in steady state, (ii) that fine particulate export was minor, and (iii) that net export by invertebrates was minor; then we estimate that 22.8% of the C in the 13-yr-old litter was leached, 69.7% was mineralized, and 7.5% remained after 13 yr.

A fractionation of the water-soluble extracts of litterfall indicated that about 38% was humic substance C, mostly fulvic acid, and another 5% was the phenolic fraction. In the NaOH extracts, 7.5% of the C in the freshly fallen litter was in the fulvic acid fraction. The fact that the fulvic acid fraction in the NaOH extracts was more or less constant over time (Table 4) meant that the mass of fulvic acid (in g m^-2) declined at about the same rate as litter C, in contrast to the humic acid fraction. The leaching of the fulvic acid fraction probably contributed to the loss of the mass of carboxyl C, particularly during the first year, because fulvic acids in forest floor solution are much higher in carboxyl C concentration than litter (Guggenburger et al., 1994). In another study, freshly fallen litter was found to have a substantial concentration of fulvic acid and the authors concluded a portion of the process of humification began during leaf senescence (Qualls et al., 1991).

Formation and Loss of Humic Substances and the Long-Term Storage of C
In this study, much of the humic substance fraction, especially the humic acid fraction, was formed in the first year of decomposition. In contrast, the fulvic acid fraction was already present in newly senesced litter. Thus, most of the process occurred during the first year of decomposition. Our measurements of the net process of accumulation and loss were likely the result of a continuing process of formation and the mineralization of humic substances, and leaching of fulvic acids.

After the first year, humic substances decomposed with a half-decay time of 5.1 yr, somewhat slower than the bulk litter (t_1/2 = 3.9 yr). Humic substances are relatively resistant to decomposition (Stevenson, 1994). Evidence for their resistance to decomposition includes their tendency to become more concentrated in soil, and the observation that ¹⁴C dates of humic substances in soils are older than bulk soil organic matter (Campbell et al., 1967). Campbell et al. (1967) found a mean C residence time of 250 yr in a forest Alfisol. Raich and Schlesinger (1992) used the ratio of soil respiration to soil C pools to estimate a considerably lower mean residence time of 29 yr for temperate forests. The order of magnitude differences in the residence time of soil C suggested by these estimates, of course, may reflect site-specific differences but also method related differences. Short-term decomposition studies may not detect very small fractions that decompose at very slow rates. In our unusually long decomposition study, only about 7.5% of the C remained after 13 yr in the oldest layer, and the rate of loss after the first year fit an exponential model relatively well. Nevertheless, some very small fraction could have been far more resistant to decomposition, eventually yielding a residue with a much longer-term residence time.

	CONCLUSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 
Our study of pine litter decomposition extended over 13 yr, over which time about 93% of the C was lost. The mass per unit area of humic substances increased during the first year, because of the formation of the humic acid fraction, and then decreased from Years 1 to 13. From Years 1 to 13, the loss of humic substances corresponded to a t_1/2 of 5.1 yr, slower than that of the total C with a t_1/2 of 3.9 yr. The concentration of humic substances and carboxyl C in litter increased from 0 to 13 yr but the mass of carboxyl C decreased. The loss of mass of carboxyl C corresponded to a t_1/2 of 4.6 yr. The lower rates of loss of humic substances and carboxyl C than total C correspond with what might be expected during humification, but the differences are not so great as to explain the long residence times of soil measured by ¹⁴C dating and C budgets. Continued long-term studies of decomposition processes may shed light on these differences.

	ACKNOWLEDGMENTS

 
Research supported in part by Nevada Agricultural Experiment Station, publication #5202369. We greatly appreciate the cooperation of the personnel of the Coweeta Hydrologic Laboratory, support from the NSF Long-term Ecological Research program, and field assistance from Todd Ackerman.

Received for publication June 3, 2002.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSION REFERENCES

 

• Antweiler, R.C. 1991. The hydrolysis of Suwanee fulvic acid. p. 163–177. In R.A. Baker (ed.) Organic substances and sediments in water. Vol. 1. Humics and soils. Lewis Pub., Boca Raton, FL.
• Baldock, J.A., J.M. Oades, P.N. Nelson, T.M. Skene, A Golchin, and P. Clarke. 1997. Assessing the extent of decomposition of natural organic materials using solid-state ¹³C NMR spectroscopy. Aust. J. Soil Res. 35: 1061–1083.
• Baldock, J.A., and C.M. Preston. 1995. Chemistry of carbon decomposition processes in forests as revealed by solid-state carbon-13 nuclear magnetic resonance. p. 89–117. In W.W. McFee and J.M. Kelly (ed.) Carbon forms and functions in forest soils. SSSA, Madison WI.
• Campbell, C.A., E.A. Paul, D.A. Rennie, and K.J. McCallum. 1967. Factors affecting the accuracy of the carbon-dating method in soil humus studies. Soil Sci.104:81–85.
• Cromack, K., Jr., and C.D. Monk. 1974. Litter production, decomposition, and nutrient cycling in a mixed hardwood watershed and a white pine watershed. p. 609–624. In F.G. Howell et al. (ed.) Mineral cycling in southeastern ecosystems. U.S. Energy Res. and Dev. Admin., Washington, DC.
• deMontigny, L.E., C.M. Preston, P.G. Hatcher, and I. Kögel-Knabner. 1993. Comparison of humus horizons from two ecosystem phases on northern Vancouver Island using ¹³C CP/MAS NMR spectroscopy and CuO oxidation. Can. J. Soil Sci. 73:9–25.
• Guggenburger, G., W. Zech, and H.R. Schulten. 1994. Formation and mobilization pathways of dissolved organic-matter—Evidence from chemical structural studies of organic matter fractions in acid floor solutions. Org. Geochem. 21:51–66.
• Harmon, M.E., O.N. Krankina, and J. Sexton. 2000. Decomposition vectors: A new approach to estimating woody detritus decomposition dynamics. Can. J. For. Res. 30:76–84.
• Jorgensen, J.R., C.G. Wells, and L.J. Metz. 1980. Nutrient changes in decomposing loblolly pine forest floor. Soil Sci. Soc. Am. J. 44:1307–1314.[Abstract/Free Full Text]
• Kinchesh, D.S., D.S. Powlson, and E.W. Randall. 1995. ¹³C NMR studies of organic matter in whole soils: I. Quantitation possibilities. Eur. J. Soil Sci. 46:125–138.
• Knicker, H., R. Frund, and H.-D. Ludemann. 1993. The chemical nature of nitrogen in soil organic matter. Naturwissenschaften 80:219–221.
• Kögel-Knabner, I., P.G. Hatcher, and W. Zech. 1991. Chemical structural studies of forest soil humic acids: Aromatic carbon fractions. Soil Sci. Soc. Am. J. 55:241–247.[Abstract/Free Full Text]
• Lal, R., J. Kimble, E. Levine, and C. Whitman. 1995. World soils and greenhouse effect: An overview. p. 1–7. In Lal et al. (ed.) Soils and global change. Lewis Pub. Boca Raton, FL.
• Leenheer, J. 1981. Comprehensive approach to preparative isolation and fractionation of dissolved organic carbon from natural waters and waste waters. Environ. Sci. Technol. 15:578–587.
• Lorenz, K., C.M. Preston, S. Raspe, I.K. Morrison, and K.H. Feger. 2000. Litter decomposition and humus characteristics in Canadian and German spruce ecosystems: information from tannin analysis and ¹³C CPMAS NMR. Soil Biol. Biochem. 32:779–792.
• Lynch, J.M. 1991. Sources and fate of soil organic matter. In E.S. Wilson (ed.) Advances in soil organic matter research, the impact on agriculture and the environment. Royal Soc. of Chem., Cambridge.
• Mao, J.-D., W.-G. Hu, K. Schmidt-Rohr, G. Davies, E.A. Ghabbour, and B. Xing. 2000. Quantitative characterization of humic substances by solid-state carbon-13 nuclear magnetic resonance. Soil Sci. Soc. Am. J. 64:873–884.[Abstract/Free Full Text]
• McGinty, D.T. 1976. Comparative root and soil dynamics on a white pine watershed in the hardwood forest in the Coweeta Basin. Ph.D. Dissertation. University of Georgia, Athens, GA.
• Qualls, R.G. 2000. Comparison of the behavior of soluble organic and inorganic nutrients in forest soils. Forest Ecol. Manage. 138:29–50.
• Qualls, R.G., and B.L. Haines. 1991. Geochemistry of dissolved organic nutrients in water percolating through a forest ecosystem. Soil Sci. Soc. Am. J. 55:1112–1123.[Abstract/Free Full Text]
• Qualls, R.G., B.L Haines, and W.T. Swank. 1991. Fluxes of dissolved organic nutrients and humic substances in a deciduous forest ecosystem. Ecology 72:254–266.
• Raich, J.W., and W.H. Schlesinger. 1992. The global carbon dioxide in soil respiration and its relationship to vegetation and climate. Tellus 44B 81–99.
• Stevenson, F.J. 1994. Humus Chemistry—Genesis, Composition, Reactions. 2nd ed., Wiley, New York.
• Swift, L.W., G.B Cunningham, and J.E. Douglass. 1988. Climatology and hydrology. p. 35–56. In W.T Swank and D.A. Crossley, Jr (ed.). Forest Hydrology and Ecology at Coweeta. Springer-Verlag, New York.
• Swift, R.S. 1996. Organic matter characterization. p. 1011–1069. In D.L. Sparks et al. (ed.) Methods of soil analysis. Part 3. Chemical methods. SSSA Book Ser. 5. SSSA, Madison WI.
• Waterman, P.G., and S. Mole. 1994. Analysis of Phenolic Plant Metabolites. Blackwell Scientific, London.
• Wershaw, R.L., G.R. Aiken, J.A. Leenheer, and J.R. Tregellas. 2000. Structural-group quantitation by CP/MAS ¹³C NMR measurements of dissolved organic matter from natural surface waters. p. 63–81. In E.A Ghabbour and G Davies (ed.). Humic Substances, Versatile Components of Plants, Soil and Water. Royal Society of Chemistry, Cambridge, UK.
• Zech, W., F. Zeigler, I. Kögel-Knabner, and L. Haumaier. 1992. Humic substances distribution and transformation in forest soil. Sci. Total Environ. 117/118:155–174.

This article has been cited by other articles:

				R. G. Qualls Biodegradability of Humic Substances and Other Fractions of Decomposing Leaf Litter Soil Sci. Soc. Am. J., September 1, 2004; 68(5): 1705 - 1712. [Abstract] [Full Text] [PDF]

				D. A. N. Ussiri and C. E. Johnson Sorption of Organic Carbon Fractions by Spodosol Mineral Horizons Soil Sci. Soc. Am. J., January 1, 2004; 68(1): 253 - 262. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Qualls, R. G. Articles by Wershaw, R. L. Search for Related Content PubMed Articles by Qualls, R. G. Articles by Wershaw, R. L. GeoRef GeoRef Citation Agricola Articles by Qualls, R. G. Articles by Wershaw, R. L. Related Collections Humic Substances Forest Soils