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


     

Published online 19 April 2006
Published in Soil Sci Soc Am J 70:975-985 (2006)
DOI: 10.2136/sssaj2005.0116
© 2006 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gregorich, E. G.
Right arrow Articles by Skjemstad, J. O.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Gregorich, E. G.
Right arrow Articles by Skjemstad, J. O.
Agricola
Right arrow Articles by Gregorich, E. G.
Right arrow Articles by Skjemstad, J. O.
Related Collections
Right arrow Soil Biochemistry
Right arrow Soil Biology

Symposium: Meaningful Pools in Determining Soil C and N Dynamics

Chemical and Biological Characteristics of Physically Uncomplexed Organic Matter

E. G. Gregoricha,*, M. H. Beareb, U. F. McKima and J. O. Skjemstadc

a Agriculture Canada, Central Experimental Farm, Ottawa, ON, Canada, K1A 0C6
b New Zealand Institute for Crop & Food Research, Christchurch, New Zealand
c CSIRO Land and Water, Adelaide Lab., PMB 2 Glen Osmond SA 5064 Australia

* Corresponding author (gregoriche{at}agr.gc.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Physical fractionation methods are based on the premise that soil organic matter (SOM) can be divided into pools of functional relevance. Physically uncomplexed organic matter (OM) is isolated on the basis of particle size and/or density. Our objective here is to review research on the biological and chemical characteristics of physically uncomplexed OM that demonstrates its value (or otherwise) as a meaningful pool of SOM. Chemical characterization indicates that fractions isolated by size are not identical to those separated by density; even materials separated using variations of a particular fractionation method (i.e., different sizes or different densities) have different chemical and biological properties. Physically uncomplexed OM often contains a substantial portion of whole soil carbon (C) and nitrogen (N) and, compared with the whole soil or heavy fraction, has a wide C/N ratio and high O-alkyl (i.e., carbohydrates) and low carbonyl (i.e., proteins) C contents. The response of physically uncomplexed OM to changes in land use and management practices is greater than that of other labile OM fractions or the whole soil C and N. Studies to elucidate the nutrient availability of physically uncomplexed OM suggest that it is not an immediate source of nutrients. That the quantity of physically uncomplexed OM is not always related to the amount of plant residue inputs suggests that other factors may control its accumulation in soil. Thus the quantity and the biological and chemical properties of physically uncomplexed OM are affected by the amount, composition, and accessibility of plant residues entering the soil; environmental conditions that may enhance or constrain decomposition; and the fractionation technique used.

Abbreviations: CT, conventional tillage • LF, light fraction • NMR, nuclear magnetic resonance • NT, no-till • OM, organic matter • POM, particulate organic matter • SOM, soil organic matter


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
SOIL ORGANIC MATTER is studied in various ways, and different fractionation techniques have been developed to characterize its composition and biological availability. Chemical fractionation techniques are commonly used to determine the types and quantity of elements and molecular compounds in SOM, as well as the interactions among these constituents and other organic compounds added to soil. Biological techniques are often used to characterize the bioavailability of SOM and to estimate the size and turnover rate of different SOM pools. Physical fractionation methods separate soil on the basis of size and/or density. These methods are based on the premise that the association of the primary soil particles and their spatial arrangement play a key role in the function of SOM and recognize that the extent and degree to which SOM is bound to inorganic mineral particles regulate its dynamics. Fractions isolated by physical methods are usually characterized by their chemical and biological properties to elucidate their functional relevance. Because of the current interest in characterizing, predicting, and potentially managing soil C dynamics and sequestration, greater attention is currently being given to the research and development of soil physical fractionation methods.

Physically uncomplexed OM is composed of particles of OM that are not bound to mineral particles and are separated by density (using heavy liquids) and/or size (by sieving). Researchers have isolated and characterized these fractions to assess the impacts of land use (Young and Spycher, 1979; Vanlauwe et al., 2000; Carter, 2002), management (Beare et al., 1994a), and vegetation type (Vanlauwe et al., 1998; Russell et al., 2004) on C and N turnover and storage, nutrient availability (Compton and Boone, 2000; Campbell et al., 2001; Salas et al., 2003), decomposition (Magid and Kjærgaard, 2001), physical protection (Beare et al., 1994b), and aggregation processes (Golchin et al., 1994a; Beare et al., 1997; Six et al., 1998; Puget et al., 2000). More specifically, this approach to OM fractionation has been used to investigate and characterize a wide range of ecosystem processes, including its effects on adsorption of chemicals (e.g., Trinitrotoluene [TNT]; Eriksson and Skyllberg, 2001), soil faunal activity (e.g., earthworms; Van Delft et al., 1999), the incidence of disease (e.g., Pythium; Stone et al., 2001), the persistence of insecticidal endotoxins (e.g., Bt {delta}-endotoxins; Hopkins and Gregorich, 2003), and the control of biogeochemistry by topography and texture (Hook and Burke, 2000). Our objective here is to describe research on the biological and chemical characteristics of physically uncomplexed OM that demonstrates its value (or otherwise) as a meaningful pool of SOM. We also discuss apparent inconsistencies in the characteristics of these fractions and identify important gaps in our understanding of the function and value of physically uncomplexed OM (or these fractions). Our observations are underpinned by a review of published data from more than 110 scientific papers.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Nomenclature
Physically uncomplexed OM is isolated using physical fractionation techniques that separate soil material on the basis of particle size and density. As such, it is an operationally defined fraction of SOM. Light fraction (LF) and particulate organic matter (POM) are the two most commonly isolated forms of physically uncomplexed OM. The LF is isolated using liquids of a specific density, most often in the range of 1.6 to 2.0 g cm–3. Particulate OM has been isolated by size alone (e.g., >53 µm), or by a combination of size and density fractionation procedures. Many terms have been used to identify and describe physically uncomplexed OM. Organic matter isolated by density has been described as the "non-combined clay fraction" (Ford and Greenland, 1968) and "mineral-free debris" (Sollins et al., 1984). That which is isolated by sieving (i.e., size fractionation) has been called "macroorganic matter" (Gregorich and Ellert, 1993), "sand-size or coarse particulate organic matter" (Sollins et al., 1999) and "coarse fraction organic matter" (Wander, 2004). In this paper, we define LF as that fraction of physically uncomplexed OM separated by density alone, and POM as that fraction isolated by size.

Chemical Characteristics
Physically uncomplexed OM is a mixture of the remains of plants, animals, and microorganisms at various stages of decomposition, including pollen, spores, and seeds, as well as phytoliths, and charcoal (Greenland and Ford, 1964; Molloy and Speir, 1977; Spycher et al., 1983; Waters and Oades, 1991; Baisden et al., 2002). The proportion of total soil C and N accounted for in physically uncomplexed OM can be substantial, and the amount of soil C and N accounted for in POM is often greater than that in LF (Table 1). Our review of more than 65 studies of agricultural soils showed that POM (isolated by size, between 53 and 2000 µm) had median values of 19% of soil C (up to 65%) and 14% of total soil N (up to 61%), whereas the LF (isolated by density, ≤1.9 g cm–3) had median values of 6% of soil C (up to 36%) and 3% of total soil N (up to 31%). In 13 studies of forest soils, POM-C accounted for 27% of soil C (up to 48%) and 19% of total soil N (up to 35%), whereas the LF accounted for 8% of soil C (up to 74%) and 18% of total soil N (up to 54%). Thus, the C/N ratio of physically uncomplexed OM (Table 1) is usually wider than that of whole soil, which is usually about 10:1 (Paul and Clark, 1996). The few studies that have examined the phosphorus (P) content of LF OM show that <5% of soil organic P resides within physically uncomplexed OM (Curtin et al., 2003; Salas et al., 2003).


View this table:
[in this window]
[in a new window]
  		Table 1. The C/N ratio and proportion of whole soil organic C and total N accounted for in the LF and POM in agricultural, forest, and grassland ecosystems.

 
Distinct differences in chemical structure have been observed between physically uncomplexed OM and whole soil or other physical fractions. Compared with the whole soil or the fine or heavy fraction, LF and POM have higher O-alkyl (i.e., carbohydrates) and lower carbonyl (i.e., proteins) C contents (Golchin et al., 1995; Skjemstad et al., 1999). The content of O-aryl C may also be higher in the uncomplexed OM fraction due to the presence of higher amounts of lignin in the partially decomposed plant residues dominating these fractions.

Physically uncomplexed OM is considered an intermediate pool of OM between fresh plant residues and stabilized SOM (Fig. 1 ), confirmed by chemical analysis of this material. Gregorich et al. (1996b) compared cropped and forest soils and examined the differences in the chemistry of OM in plant residue, whole soil, and the two fractions of physically uncomplexed OM – LF (<1.7 g cm–3) and POM (>53 µm). The C/N ratio, amino acid N content, and proportion of O-alkyl C and alkyl C in these two fractions were intermediate between those of plant residues and whole soil. Differences in the chemical composition of these two uncomplexed OM fractions indicated that they were not identical. Analysis by solid-state 13C nuclear magnetic resonance (NMR) showed that the LF contained greater amounts of carbohydrates and aliphatic compounds than the POM. Isotopic data indicated that there was more recently deposited residues (maize-derived) in the LF than in the POM, suggesting that the LF is more closely related to plant residues than the POM. This is consistent with data obtained from our review of the published literature (Table 1), which shows that the median C/N ratio for POM (53–2000 µm) is lower than that for LF (density < 1.9 g cm–3) within the same land use (except for forest systems, for which there are too few data to draw definitive conclusions). Differences observed between these two fractions of physically uncomplexed OM highlight the fact that size and density methods may isolate OM with different chemical characteristics.


Figure 1
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 1. Physically uncomplexed organic matter is an intermediate between plant residues and soil organic matter. Its quantity and composition in soil are affected by factors that regulate: (1) quantity and type of residue (e.g., amount, composition, and accessibility) and/or (2) microbial activity (e.g., environmental conditions).

 
Since physically uncomplexed OM comprises recent plant residues, it serves as a readily decomposable (i.e., energy-rich, high resource quality) substrate for soil organisms. Hence it is not surprising that uncomplexed OM is inhabited by a substantial portion of the soil microbial biomass (Kanazawa, 1979; Kanazawa and Filip, 1986), and is the site of high biological activity (Alvarez et al., 1998) and a locus of high enzyme activity (Stemmer et al., 1997). The close relationships between changes in POM-C content and CO2 respired and changes in POM-N and net N mineralization led Magid and Kjægaard (2001) to conclude that soil biological activity was localized in the POM. Also, the size and structure of microbial communities associated with the LF are more similar to those of the rhizosphere (a site in the soil matrix containing energetically and nutritionally rich resources) than to those of the heavy fraction (Blackwood and Paul, 2003).

Variation within a particular physical fractionation method can result in differences in the amount and chemical composition of the OM recovered. For example, the differences observed in the isotopic signature and chemical composition of OM isolated when sediment or soil is fractionated into multiple density fractions has been exploited to evaluate OM loadings on minerals (Arnarson and Keil, 2001), effects of decomposition and C turnover (Baisden et al., 2002), and organic mineral interactions (Turchenek and Oades, 1979; Golchin et al., 1994a). Of the published studies we reviewed, 43 used density fractionation methods on soils under a range of land uses (agriculture including arable crops, forage crops, and crop-fallow rotations; forestry; and native grasslands) to isolate physically uncomplexed OM and assessed the effects of solution density on the C/N ratio of the isolated fraction (Fig. 2a ). The results showed that the average C/N ratio ranged between 17 and 22 at solution densities of 1.0 to 1.8 g cm–3. The ratios narrowed with increasing density, to between 10 and 17 at densities of 1.8 to 2.2 g cm–3. The wider C/N ratios at low solution densities (<1.8 g cm–3) reflect the dominant influence of plant constituents (e.g., lignin), whereas at higher densities the isolated fraction contains more mineral particles with adsorbed OM and therefore has a narrower C/N ratio. This trend also reflects the dominance of more decomposed, humified OM (Baisden et al., 2002). Many soils have a bimodal density distribution, and Sollins et al. (1999) suggested a convenient cut-off density for separating light and heavy fractions 1.6 to 1.7 g cm–3. The observed narrowing in the C/N ratio above a density of 1.8 g cm–3 is consistent with this recommendation.


Figure 2
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 2. Carbon to N ratio of physically uncomplexed organic matter isolated at different (a) solution densities (n = 535 fractions) and (b) particle sizes (n = 363 fractions). Values on X-axis are (a) the upper end of the solution density range and (b) the midpoint of the size range. Error bars represent standard deviations. Data are from published studies of agricultural, forest, and grassland ecosystems; 43 studies using density fractionation; and 24 studies using particle size fractionation methods.

 
Of the studies we reviewed, 24 used size fractionation methods on soils under cropping, forestry, and grasslands; they indicated some trends in the size and chemical characteristics of POM fractions (Fig. 2b). Overall, there is a log-linear relationship between the mean size of POM fractions and their C/N ratio. There was little difference in C/N ratios among land uses, and the variation in C/N was low in the smaller size fractions compared with the variation in larger size fractions. The C/N ratios of the different size classes of POM showed that, on average, larger size fractions had wider ratios than smaller size fractions. This is consistent with the findings of Magid and Kjærgaard (2001), who studied changes in POM fractions during plant residue decomposition. The narrowing of the C/N ratio with declining size reflects a gradient in the decomposition of OM. Magid and Kjærgaard (2001) observed that large particles consist mostly of recent plant residue inputs and that changes in POM-C content were closely related to C mineralization. They also observed that large (i.e., POM > 400 µm) and light (<1.4 g cm–3) particles were similar in appearance and in their C and N contents, which is consistent with our findings (Fig. 2a and 2b).

Golchin et al. (1994a, 1994b) separated the LF located between soil aggregates (i.e., free LF) from that released from inside aggregates disrupted by sonication (i.e., occluded LF) in soils from native forest and grassland sites in Australia. The free LF consisted of recognizable intact and partially decomposed plant debris with a high C to N ratio and a high proportion of carbohydrates. Their work also indicated that many soil aggregates have a nucleus of plant debris, which is consistent with the findings of Waters and Oades (1991) and Beare et al. (1994a). The proportion of C in the occluded LF was greater where the clay content was higher. Analysis by 13C-NMR indicated that occluded LF had undergone greater decomposition (i.e., had a lower proportion of O-alkyl C and a higher proportion of alkyl C) than the free LF. In general, the higher the proportion of alkyl C and the lower the O-alkyl C in the occluded LF compared with higher density fractions reflected the greater breakdown of carbohydrates and preservation of the more recalcitrant OM associated with plant residues, such as cutins, suberins, fatty acids, and waxes.

Physically uncomplexed OM can also contain substantial amounts of charcoal, a significant factor in determining the chemical composition and turnover of this OM. The presence of charcoal has been detected, using microscopic techniques, in the LF from forest soils (Spycher et al., 1983), alluvial soils (Baisden et al., 2002), Oxisols (Skjemstad et al., 1990), and volcanic ash soils (Shindo et al., 2004). Skjemstad et al. (1990) observed fragments of charcoal in the LF in soil layers down to 80 cm, and Shindo et al. (2004) estimated that charcoal pieces comprised between 0.3 to 5.4% of the organic C in the LF of volcanic ash soils. These data indicate that charcoal is not uncommon in LF and POM fractions, but generally makes a small contribution to their organic C content.

Skjemstad et al. (1999) used solid-state 13C NMR to determine the contribution of charcoal to the POM fractions from eight different soils of different clay content and mineralogy collected from a range of environments. With one exception, charcoal in the POM fraction contributed 7% or less of the charcoal C present in the whole soil and <6% of the organic C present in the POM fraction. In the exception (an Oxisol) charcoal in the POM fraction contributed 62% of the charcoal C present in the whole soil and 16% of the POM-C. Figure 3 shows the effect that significant amounts of charcoal can have on the solid-state 13C NMR spectra of POM fractions. Sample (a) contains no charcoal and shows characteristic peaks from lipids (30 ppm), carbohydrates (73 and 105 ppm), lignins (130 and 153 ppm), and proteins (56 and 173 ppm). These peaks are also evident in spectrum (b), but the presence of charcoal is indicated by the very large peak at 128 ppm. Since only about 30% of C nuclei in charcoal structures are measured with the cross polarization technique used in solid-state NMR (Skjemstad et al., 1999), the large aryl C peak at 128 ppm is under-represented, and sample (b) probably contains <70% of its organic C content in the form of charcoal. Although charcoal is not commonly abundant in POM fractions its presence could have profound implications for the availability of the C and N in these normally labile pools.


Figure 3
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 3. Solid-state 13C NMR spectra of a particulate organic matter fraction containing no charcoal (a) and containing charcoal (b). (J. O. Skjemstad, unpublished data, 2004).

 
Changes during Decomposition
During decomposition plant residues are physically broken down into smaller pieces. The density of these OM fragments also increases during decomposition due to the breakdown of the vascular structure and the subsequent loss of entrapped air. Thus the density of the solution used to isolate LF OM may help to distinguish the nature and extent of OM decomposition. Magid et al. (1996) conducted a study to trace the fate and distribution of freshly added plant residues in soil. They incubated 14C-labeled plant residues in soil for 200 d, fractionating them at intervals throughout this period, first by size (i.e., >100 and <100 µm) and then by density, dividing each size fraction into three density fractions (i.e., <1.4, 1.4–1.85, and >1.85). Soil could be separated into distinct fractions that reflected the breakdown of freshly added plant residues: one with large particles having lower densities (i.e., <1.4 and 1.4–1.85 g cm–3) and another fraction composed of small particles with higher densities that reflected the build-up of microbial metabolites during the incubation. These results also demonstrate that physically uncomplexed OM is a primary repository of fresh plant residues.

The chemical composition of physically uncomplexed OM can give some clues to changes that occur during plant residue decomposition. Leifeld and Kögel-Knabner (2004) used 13C NMR to characterize the chemical composition of plant litter, uncomplexed OM (free and physically occluded LF), and the heavy fraction (that associated with silt and clay). They used the ratio of alkyl/O-alkyl C as a measure of the extent of decomposition and compared it to the C/N ratio for each isolated fraction (Fig. 4 ). The uncomplexed OM had a lower C/N ratio than plant residue but a similar alkyl/O-alkyl C ratio. These data suggest that during the early stages of decomposition there are large losses of C relative to N losses (i.e., the C/N ratio narrows in going from plant residue to uncomplexed OM), but only small changes occur in the alkyl/O-alkyl C ratio. In contrast, OM associated with mineral particles has a narrower C/N ratio than does uncomplexed OM but larger proportions of alkyl C, presumably due to a loss of O-alkyl C during decomposition and the increase in recalcitrant alkyl C products.


Figure 4
View larger version (17K):
[in this window]
[in a new window]
  		Fig. 4. Relationship between the C/N ratio and alkyl/O-alkyl C ratios of plant tissue, particulate organic matter, and particle-size fractions. (Adapted from Leifeld and Kögel-Knabner, 2004).

 
Indicators of Soil Organic Matter Change
Studies have been conducted to evaluate the relative sensitivity of various OM fractions as indicators of soil quality. Bolinder et al. (1999) evaluated the relative size of different labile OM fractions in soil under conservation and conventional management systems for 16 paired studies in eastern Canada (Table 2). Particulate organic matter accounted for the largest amount of soil C (about 20–25%). Acid-hydrolyzable carbohydrates comprised 10 to 15% and the LF about 5 to 10% of the whole soil C. Even though the LF comprised the smallest proportion of soil C, it was the most sensitive to conservation management. Biederbeck et al. (1994) assessed soils from long-term field trials to determine which labile fractions of OM were influenced most by cropping practices. They found that LF-C and LF-N were the most sensitive of the labile OM fractions measured, about 2.5 to 3 times more responsive to differences in crop management practices than total C or N. In a comparison of tillage systems in Georgia, Beare et al. (1994a) found that there was 18% more organic C in the top 15 cm of soil under no-till treatments (30.7 Mg C ha–1) compared with that under conventional tillage (26.1 Mg C ha–1). In contrast to other studies (e.g., Cambardella and Elliott, 1992), POM accounted for a similar proportion (35%) of the whole soil C and N in both tillage systems. Because POM was not sensitive to tillage-related changes in SOM suggests there is a need for further research to evaluate the LF and POM as indicators of SOM changes under a wide range of management practices.


View this table:
[in this window]
[in a new window]
  		Table 2. Quantities of C in whole soil and SOM fractions{dagger} in soils under conservation and conventional management practices in eastern Canadian soils. (Adapted from Bolinder et al., 1999).

 
Physically uncomplexed OM often shows a much greater response to fertility management practices than total SOM. In a study of soil under bromegrass (Bromus inermis Leyss) to determine the response of LF-N and total N to different rates of fertilizer N (0–336 kg N ha–1), Malhi et al. (2003) found a 10-fold increase in LF-N compared with an increase of about 30% in total soil N (i.e., mineral + organic N) (Fig. 5 ). There was an asymptotic response of LF-N to fertilizer addition, and it was estimated that as much as 17% of the added fertilizer N was stored in the LF. This large response of the LF to fertilizer additions reflects the larger amount of above- and below-ground residue inputs in a well-fertilized soil and suggests that nutrient limitations may constrain the accumulation of uncomplexed OM (Fig. 1). Dalal et al. (2005a) also showed that the amount of live root biomass correlated well with LF-C, but the extent to which below-ground plant residues contribute to the accumulation of uncomplexed OM has yet to be elucidated.


Figure 5
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 5. The quantity of light fraction-N (left axis–diamond symbols and dashed line) and the quantity of total N (right axis–square symbols and solid line) in the surface 30 cm of soil as a function of the amount of N fertilizer applied in a Saskatchewan grassland. (Drawn from data presented by Malhi et al., 2003).

 
Among the 434 size fraction analyses included in our literature review, the most common POM size fraction isolated was 50- to 2000-µm (which includes 53–2000 µm). These data were used to explore differences in the amount and chemical composition of OM in this size fraction under long-term conventional tillage (CT, usually plowing followed by secondary cultivation) and no-till (NT) management. The results for the 50- to 2000-µm POM fraction were also compared with the same measurements for the 0- to 50-µm fraction. On average, a high percentage (86–89%) of total soil organic C was associated with the 0- to 50-µm fraction (i.e., mineral associated), and there were no consistent differences in the relative size of this pool under long-term CT or NT (Table 3). The C/N ratio of the mineral associated 0- to 50-µm fraction was very near to 10, with relatively little variation in this value across studies. On average, C in the 50- to 2000-µm POM fraction accounted for about 20% of the total soil organic C in both tillage systems, despite large differences in total C across the studies. The most notable difference between tillage systems was the C/N ratio of the 50- to 2000-µm POM fraction. Although these ratios vary widely, the average C/N ratio of this fraction was about 22% higher in NT than CT, indicating greater storage of C than N in the POM fraction under NT. The reason the two fractions account for more than 100% (i.e., 106–109%) of the total soil organic C was probably due to contamination of the 50- to 2000-µm POM fraction with mineral soil, particularly where the fractions were not isolated sequentially. Published data on the C and N content of POM are available for other long-term tillage systems (e.g., chisel plow, stubble mulch, ridge tillage, shallow tillage); however, where consistent methods were used, the size of these datasets was insufficient to draw any major conclusions.


View this table:
[in this window]
[in a new window]
  		Table 3. The mean and standard deviation (SD) C content (% total soil C) and C/N ratio of size fractions from soil under long-term conventional tillage (CT) and no-till (NT).

 
Kinetics and Turnover
Physically uncomplexed OM is highly susceptible to breakdown when ecosystems are disturbed and it turns over much more quickly than OM bound to soil minerals (Henderson et al., 2004; Baisden et al., 2002). These labile fractions often undergo relatively large losses when native grasslands or forests are converted to agricultural land uses (Cambardella and Elliott, 1992; Dalal and Mayer, 1986, 1987). Table 4 summarizes some studies that used 13C at natural abundance to distinguish the source of C in either POM or the LF. The losses ranged from 15% (of POM-C) 12 yr after converting forest to sorghum production in Brazil (Shang and Tiessen, 2000) to 84% (of POM-C) after 50 yr under sugarcane in Brazil (Bonde et al., 1992). Much of this loss occurs soon after the change in land use, often within the first two decades; these values are within the range of values reported recently by Dalal et al. (2005a) and Skjemstad et al. (2004). These studies confirm that the loss of physically uncomplexed OM usually follows first-order kinetics (Gregorich and Janzen, 1996) and that its C turnover is faster than that of the whole soil. Dalal and Mayer (1986, 1987) used a first-order equation to estimate that the turnover of LF-C in a range of Australian soils was 3 to 11 times faster than that of the whole soil C. Dalal et al. (2005a) used 13C at natural abundance to estimate that LF-C turned over two to six times faster than whole soil C.


View this table:
[in this window]
[in a new window]
  		Table 4. Retention of original C and loss of C from the LF or POM as shown by 13C natural abundance.

 
Research using the 13C natural abundance technique has also shown that substantial amounts of the original C can remain in the physically uncomplexed OM. For example, studies on soils cropped to maize for 25 to 33 yr found that 27 to 35% of the original C remained in the LF (Table 4). In some systems this retention of C in physically uncomplexed OM can be even greater. In Australia, Dalal et al. (2005a) found that about 60% of the original C in the surface 10 cm remained in the LF 20 yr after the native ecosystem was converted to pasture. Balesdent et al. (1988) reported that about 40% of C from the original prairie grass remained in the POM after almost 100 yr of cropping to timothy (Phleum pratense L.).

Carbon Storage
Since the proportion of total soil C in physically uncomplexed OM may be substantial and is often the largest labile fraction of OM in soil, the storage and turnover of C in this fraction may be important. The LF-C content of soils has been strongly correlated to the rate of whole soil respiration (Janzen et al., 1992; Alvarez and Alvarez, 2000), suggesting that the LF-C is a driving variable in soil respiration (Alvarez et al., 1998).

Biederbeck et al. (1994) analyzed soils from a long-term study in southwestern Saskatchewan, Canada, to investigate the effects of cropping practices on OM content and composition in an arid environment. Their research showed that N fertilization and the substitution of winter wheat for spring wheat systems tended to increase labile OM, including LF-C and LF-N. Although residue C inputs in continuous wheat systems were only about 1.4 times higher than those in wheat systems containing a fallow season, the quantity of LF-C under continuous wheat was almost three times higher than that in the wheat–fallow systems. This finding suggests that factors other than the amount of residues returned to the soil regulate the accumulation of LF. Since the study was conducted in a semi-arid environment, the amount of physically uncomplexed OM may be controlled by the degree to which temperature and moisture constrain residue decomposition. Therefore, systems that have environmental conditions that constrain decomposition may also have more physically uncomplexed OM accumulating than can be explained by residue C inputs alone (Fig. 1). The findings of Biederbeck et al. (1994) suggest that the LF could be managed by using specific agronomic practices. By modifying the timing of residue inputs and soil moisture patterns it may be possible to increase the storage of physically uncomplexed OM and improve the synchrony of nutrient mineralization with crop demands.

Paul et al. (2004) used common sampling and analytical methods to measure controls on SOM dynamics in a multi-site comparison of long-term crop rotations in western Canada. These rotations included various wheat–fallow rotations receiving different organic and inorganic fertilizers. Differences in POM-C between treatments were significantly related to differences in soil C, and the slope of the relationship suggests that approximately one-half of the difference in soil C between treatments could be related to changes in POM-C.

In another study that investigated the potential for using measured SOM fractions in conceptual soil organic C turnover models, Skjemstad et al. (2004) found that POM was a good measure of the resistant plant material pool in the RothC model (Jenkinson, 1990). The measured and modeled pools were in best agreement when a decomposition rate of 0.15 yr–1 was used in the model for the resistant C pool, giving the POM a turnover time of 6.7 yr. This study also demonstrated the rapid decline of POM in the surface 30 cm of three soils in semi-arid, subtropical Australia following clearing of the native vegetation for wheat cropping. After 18 yr of agriculture, 94% of the decline in total organic C could be accounted for by losses from the POM-C pool.

Environmental conditions, particularly soil moisture and temperature, exert a strong influence on the rate of plant residue decomposition. Therefore physically uncomplexed OM may accumulate due to restricted decomposition in environments that are very hot and dry or very cold and wet. Even though it is a labile fraction, physically uncomplexed OM may comprise a large proportion of total soil organic C and N and may therefore play an important role in the short-term storage of soil C and N.

Nutrient Availability
The breakdown of physically uncomplexed OM, ultimately into inorganic constituents, is mediated mainly by soil microorganisms, which derive energy and nutrients from the diverse range of molecules in the SOM. If the nutrients are not taken up by the microorganisms they are available for plant uptake. Thus LF-N and POM-N are both sensitive indicators of the effects of agricultural management practices on SOM (Biederbeck et al., 1994; Bolinder et al., 1999). Results of some studies have led researchers to suggest that physically uncomplexed OM could be used to help develop fertilizer management recommendations (e.g., Willson et al., 2001). Other studies have indicated that this fraction of OM is the primary active sink and source of N affecting both short- and long-term soil fertilities (Bird et al., 2002). In a laboratory incubation study, Magid and Kjærgaard (2001) observed a close relationship between changes in POM-N and net-N mineralization. However, it is still not clear how much physically uncomplexed OM contributes to nutrient availability. Dalal et al. (2005b) reported that LF-N declined by 60 to 70% in soils under both pasture and cropping following the clearing of native vegetation (mulga; Acacia aneaura). Although the LF-N comprised only 2 to 6% of total N in the pasture and cropped soils, they surmised that there would be a lower N supply to plants following land-use change from the native vegetation to pasture and cropping. Results from one of the earliest studies on physically uncomplexed OM suggested that LF was a major source of mineralizable N, accounting for as much as 26 to 60% of net N mineralized in soils from under wheat–fallow and pasture–fallow–wheat rotations in southern Australia (Ford and Greenland, 1968). These estimates were based on the premise that changes in mineralizable N could be directly attributed to changes in LF-N. The LF may be converted to heavy-fraction OM, and LF decline would then be associated with LF turnover but not with LF mineralization (Boone, 1994).

Results of several studies suggest that physically uncomplexed OM may immobilize mineralized nutrients because its C/nutrient ratio is usually wider than that of the whole soil. Janzen (1987) reported that LF-N was significantly correlated with net N mineralization in a study involving long-term wheat cropping. In a subsequent study Janzen et al. (1992) reported that the correlation between LF and N mineralization was not as strong nor as consistent as that with soil respiration. This was presumably due to temporary N immobilization by the LF, which had a wide C/N ratio. In studies of forest soils, Sollins et al. (1984) found that the LF had lower N mineralization potential than the heavy fraction. Similarly Boone (1994) concluded that the heavy fraction is the primary source of N in coarse-textured mineral soils and that the direct contribution of the LF to net N mineralization was relatively small (<13%) in forest and cropped soils. Other research has shown that LF OM can be a rapid short-term sink for inorganic N. For example, a laboratory incubation study showed that the LF immobilized more 15N per unit of C than the heavy fraction, indicating that the composition of SOM controls its function as a site of N incorporation (Compton and Boone, 2002). The presence of microbes in the LF (Kanazawa, 1979; Kanazawa and Filip, 1986) may account for the high rate of N immobilization and provide an alternative to the idea that the LF's chemical composition controls the LF's role in N cycling.

Direct investigations of nutrient mineralization in size and density fractions are limited. Results from an 8-wk laboratory study in which POM was added to whole soil and incubated showed that only 5 to 6% of the POM-N was mineralized; this contribution was about 12% of the total N mineralized from soil OM (Yakovchenko et al., 1998). Whalen et al. (2000) observed that the addition of LF to soil caused N immobilization in short-term incubations. They concluded that the heavy fraction was the main source of potentially mineralizable N, whereas the LF could be considered a potential sink for mineral N.

Marko et al. (1999) found that although the P content of LF and POM was relatively low, it was significantly influenced by cropping rotations in unfertilized maize systems on tropical soils. In an incubation study, Salas et al. (2003) observed that most of the P in POM was released as a result of decomposition in the early stages of incubation, but significant increases in POM-P occurred in later stages, suggesting greater microbial immobilization of P. In a study of volcanic ash soils, Phiri et al. (2001) determined that fallow and maize–bean rotation systems significantly affected the C/P and N/P ratios in the LF and that the amount of LF-P was significantly correlated with the amount of soil available P. Curtin et al. (2003) also observed that the free OM in LF was very low in organic P (Po) and had C/Po ratios greater than 500:1 that were much wider than those of the whole soil OM (~80:1). This led them to conclude that the decomposition the LF would result in net immobilization of P and would unlikely contribute a significant amount of P for plant growth.

Results from these earlier studies indicate that significant differences in the nutrient (i.e., N and P) composition of physically uncomplexed OM can be detected as a result of changes in management practices. It appears, however, that physically uncomplexed OM is not a significant or an immediate source of nutrients and the qualitative nature of C compounds (and N and P contents) determine its role as a net sink or source of nutrients.

Methodological Considerations
The premise of size fractionation is that smaller mineral particles are more chemically and biologically reactive than larger particles because of their larger specific surface area (Sollins et al., 1999). A disadvantage of size fractionation methods is that the organic material is fragmented and comminuted by sample handling and processing such as drying, wetting or rewetting (e.g., during wet sieving), grinding, sieving, and freezing. Also, the amount of energy used to disperse the soil (e.g., sonication) affects the quantity of POM recovered (Oorts et al., 2005), although negligible amounts of POM-C and -N were redistributed to smaller size fractions during ultrasonic dispersion at 1500 J g–1. Size fractionation methods are relatively easy and inexpensive (in terms of materials and labor), and for these reasons Magid and Kjærgaard (2001) advocate their use for separating POM in studies of residue decomposition.

Density fractionation is based on the premise that lighter soil particles (comprising mainly freshly added, partially decomposed, and humified OM) are more labile and reactive than heavier particles, which have variable amounts of adsorbed humified OM. Because soil particles span a range of densities, it is difficult to isolate a homogeneous fraction of either organic material completely free of any mineral particles or soil minerals devoid of organic material. Thus the density of soil particles reflects the ratio of organic materials to mineral particles (Sollins et al., 1999) and plays a crucial role in determining the chemical nature of the physically uncomplexed OM fractions. Even small variations in the density of the heavy liquid used to fractionate the sample can result in large differences in the quantity of C isolated due to contamination by minerals with adsorbed C. Richter et al. (1975) separated the LF using increasingly higher densities and observed that the proportion of C recovered in the LF was similar, with densities between 1.6 and 1.9 g cm–3. But it more than doubled when the density was raised from 1.9 to 2.0 g cm–3, and it increased exponentially at densities > 2 g cm–3.

Sohi et al. (2001) analyzed and compared fractions isolated by density with those isolated by size. They concluded that the size fractions were chemically similar but that the density fractions displayed distinct differences in biological reactivity that reflected their chemical properties. They contended that these fractions would provide a sound basis for developing a model of SOM turnover based on measurable pools.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Two factors affect the quantity of physically uncomplexed OM and its biological and chemical properties: those that control residue input to the soil (e.g., anything that affects the amount, composition, and accessibility of residues) and those that control microbial activity (e.g., any environmental variable that enhances or constrains decomposition). From the standpoint of ecosystem function, physically uncomplexed OM is a meaningful pool of soil OM because it serves as a primary energy source for heterotrophic soil organisms and a reservoir of labile C. Chemical characterization indicates that physically uncomplexed OM has not undergone significant physical and chemical transformations but is at an intermediate stage of decomposition between fresh plant residue and stabilized SOM. Biological characterization indicates that physically uncomplexed OM has a higher turnover rate than that of whole soil OM and other labile fractions. Therefore, its response to changes in land use and/or management practices is quite rapid and often greater than for other labile fractions or whole soil C and N.

Our review of published data on physically uncomplexed OM has identified a number of key areas where further research is needed. Although the importance of plant residues as a source of the material that forms uncomplexed OM is well accepted, the relative contributions of above- and below-ground plant residues to uncomplexed OM remains poorly known. Furthermore, there is very little information concerning the effects that different plant species have on the quantity and biochemical composition of uncomplexed OM. Whereas the C and N contents of LF and POM are well-defined for a wide range of soils and land uses, the quantity of other key nutrients in uncomplexed OM isolated from different sources must still be quantified. Further research is also needed to resolve the conflicting evidence concerning the role of uncomplexed OM in nutrient supply. In particular, it would be useful to identify key environmental and biological factors that regulate the balance between mineralization and immobilization of nutrients from uncomplexed OM.

Received for publication April 8, 2005.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
H. Chung, J. H. Grove, and J. Six
Indications for Soil Carbon Saturation in a Temperate Agroecosystem
Soil Sci. Soc. Am. J., June 18, 2008; 72(4): 1132 - 1139.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
C. Bosshard, E. Frossard, D. Dubois, P. Mader, I. Manolov, and A. Oberson
Incorporation of Nitrogen-15-Labeled Amendments into Physically Separated Soil Organic Matter Fractions
Soil Sci. Soc. Am. J., May 29, 2008; 72(4): 949 - 959.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
D. V. Sarkhot, N. B. Comerford, E. J. Jokela, J. B. Reeves III, and W. G. Harris
Aggregation and Aggregate Carbon in a Forested Southeastern Coastal Plain Spodosol
Soil Sci. Soc. Am. J., September 28, 2007; 71(6): 1779 - 1787.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
D. C. Olk and E. G. Gregorich
Overview of the Symposium Proceedings, "Meaningful Pools in Determining Soil Carbon and Nitrogen Dynamics"
Soil Sci. Soc. Am. J., April 19, 2006; 70(3): 967 - 974.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow