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 (22) Citing Articles via Google Scholar Google Scholar Articles by McLauchlan, K. K. Articles by Hobbie, S. E. Search for Related Content PubMed Articles by McLauchlan, K. K. Articles by Hobbie, S. E. GeoRef GeoRef Citation Agricola Articles by McLauchlan, K. K. Articles by Hobbie, S. E. Related Collections Ecosystem Restoration Soil Organic Matter Soil Analysis

DIVISION S-3—SOIL BIOLOGY & BIOCHEMISTRY

Comparison of Labile Soil Organic Matter Fractionation Techniques

^a Environmental Studies Program, Dartmouth College, 6182 Steele Hall, Hanover, NH 03755
^b Dep. of Ecology, Evolution, and Behavior, Univ. of Minnesota, 1987 Upper Buford Circle, St. Paul, MN 55108

^* Corresponding author (kendra.k.mclauchlan{at}dartmouth.edu)

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Labile soil organic carbon (SOC_L), soil organic carbon with a relatively short turnover time, is an important source of energy for the belowground portion of ecosystems and is sensitive to land management changes. Many techniques exist to differentiate and quantify labile SOC, but rarely have these been directly compared. Here we compare the results of four common chemical, physical, and biological methods of empirically measuring labile SOC with soils taken from 33 restored grasslands that differ in length of time since cessation of agriculture. Among sites, microbial biomass C, acid-hydrolyzable C, the amount of C respired after 12 d of a laboratory incubation, and light fraction carbon (LFC) were all positively correlated with one another, although there were large differences in the sizes of the pools estimated with each method. Acid-hydrolyzable C consistently provided the largest estimate and 12-d incubations the smallest estimate of labile SOC. The quantity of labile SOC obtained by fitting respiration data from a laboratory incubation with a two-pool model with separate decay constants for each pool was also positively correlated with the three measures of labile soil C not derived from respiration data, although this technique was sensitive to whether the decay constant of the recalcitrant pool was constrained or not. All methods showed increases in labile SOC pools with increases in total soil organic carbon (SOC_T) pools, although the rate of change varied between techniques. The size of stable aggregates correlated positively with hydrolyzable C and SOC_T, supporting the idea that aggregates may physically protect soil C from decomposition, although the degree to which this C is labile is unclear.

Abbreviations: GMD, geometric mean diameter • LFC, light fraction carbon • POM, particulate organic matter • SOC, soil organic carbon • SOC_L, labile soil organic carbon • SOC_T, total soil organic carbon • SOM, soil organic matter

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
SOIL ORGANIC MATTER (SOM) is an important ecosystem property, regulating nutrient supply to plants and microbes, soil moisture, and long-term C storage. Soil organic matter is a heterogeneous, dynamic substance that varies in C and N content, molecular structure, decomposition rate, and turnover time (Oades, 1988). It is considered to be composed of several discrete pools with a negative relationship between pool size and decomposition rate—the smallest pools decompose most rapidly (Jenkinson, 1990; Parton et al., 1987; Smith et al., 1997).

In two-pool models, the pool with the smallest size and most rapid turnover is termed labile and the larger pool with slow turnover is termed recalcitrant. The lability of SOM is defined as the ease and speed with which it is decomposed by microbes and depends on both chemical recalcitrance and physical protection from microbes. Chemical recalcitrance is conferred by high molecular weight, irregular structure, and/or aromatic structures (Krull et al., 2003). Soil organic matter can be vulnerable to microbial degradation because of low chemical recalcitrance of its components, lack of stabilization onto clay mineral surfaces, or lack of physical protection inside soil aggregates (Krull et al., 2003).

The quantities of labile and recalcitrant SOM are sensitive to land management, especially agriculture, which reduces inputs to SOM through removal of plant biomass. Tillage also increases outputs of SOM through physical disturbance of soil structure which exposes organic matter to oxidation by microbes and promotes erosion. To more fully understand the response of SOM dynamics to changes in tillage and other soil management actions, the sizes and turnover times of different SOM pools must be measured accurately and consistently. Tillage reduces labile SOM and cessation of tillage causes labile SOM to increase (Six et al., 2000). Recalcitrant SOM has been considered more resistant to tillage, although it is also thought to be sensitive to soil management (Ding et al., 2002). The formation of soil aggregates that physically protect otherwise labile SOM from microbial decomposition may be at least partially responsible for the increase in labile SOM with reduction in tillage (Six et al., 2000). Although SOM inside aggregates is theoretically chemically equivalent to unprotected SOM, physical protection of SOM confers a longer residence time.

There are many techniques that measure the size and turnover time of SOM pools, and many of these techniques are used sequentially in analyses, but few have been directly compared. At least nine methods have been used to separate SOM into labile and recalcitrant pools, and these methods rely on chemical, physical, or biological separation (Doran et al., 1999; Karlen et al., 1998). Most methods quantify either a labile pool or a recalcitrant pool, and calculate the other pool by difference from SOC_T.

Chemical fractionation methods include hydrolyzing labile C with acid (Paul et al., 2001), digesting it with permanganate (Weil et al., 2003), or extracting it with hot water (Gregorich et al., 2003). Each method assumes that the same properties that make SOM degradable by microbial enzymes make it less resistant to chemical attack or more soluble in hot water.

Physical fractionation relies on differences in either particle density or size of labile and recalcitrant fractions. A light fraction can be separated from a heavy fraction by flotation with a dense liquid (Gregorich and Janzen, 1996). The light fraction is considered to be labile whereas the heavy fraction is assumed to be stabilized onto surfaces of clay particles, making it more resistant to microbial degradation. Sieving soil into different size classes separates small aggregates or particles from larger particles (Kemper and Chepil, 1965; Six et al., 1998), which contain SOC that is partially physically protected from microbial degradation, although not chemically recalcitrant. Dispersion and sieving of the sand-sized organic matter can be used to isolate the particulate organic matter (POM) fraction that is considered labile (Cambardella and Elliott, 1992).

Biological separation empirically separates labile SOC from recalcitrant SOC by allowing microbes to mineralize SOC under controlled temperature and moisture conditions and in the absence of new organic inputs. This method assumes that microbes will mineralize the most labile C first, with recalcitrant C being mineralized later. The technique involves measuring CO₂ produced by mineralization of SOC during the course of a laboratory incubation of soil (Alvarez and Alvarez, 2000; Pastor et al., 1993). Although some aggregate structure is destroyed during sieving that occurs before the incubation, some aggregates survive this process and are intact during the incubation. Thus, some of the labile C that was protected by aggregate structure in the field becomes available for microbial mineralization during the incubation (Kristensen et al., 2003). Additionally, a direct measure of the pool size of living soil organisms, microbial biomass, can be quantified with either chloroform fumigation incubation or chloroform fumigation extraction (Beck et al., 1997; Paul et al., 1999). Microbial biomass is more readily decomposed than SOC_T, with turnover times of days to weeks, and is therefore considered a labile SOC pool, although it does not encompass all labile C (Parton et al., 1987).

Although most of these techniques have been shown to be useful in fractionating SOC, it is unknown whether there are systematic differences among techniques in the way SOC is partitioned. Different techniques to measure the labile portion of the SOC pool, SOC_L, may differ in inclusivity or identity of the labile pool. That is, different techniques sample different portions of the SOC pool (Paul et al., 2003). The consistency of the differences between techniques across multiple quantities and types of SOC_L is important for understanding C and N cycling in soils and for managing levels of SOC_L in agricultural or other soils. The amount of variation between different techniques should be considered when interpreting differences in measured quantities of SOC_L between studies, treatments, vegetation types, or across large spatial extents.

The primary objective of this study was to compare the results of four common chemical, physical, and biological methods of estimating SOC_L: LFC, hydrolyzable C, respiration from soils incubated in the laboratory for 12 d, and microbial biomass C (Table 1). Although acid hydrolysis has been used primarily in a three-pool context for identifying recalcitrant C that is then used to constrain respiration data (Paul et al., 2001), here we wanted to investigate its usefulness as an independent measure of the more labile portion of soil C in a two-pool context. Similarly, C respired from early in an incubation is used as an indicator of labile C (Mahmood et al., 1997), and we wanted to test the C respired by Day 12 as an example of this approach.

The second objective was to examine changes in SOC_L pool size, as measured by the five different techniques, as SOC_T pool size changes. We predicted that while the quantity of SOC_L measured by the different techniques would be different, SOC_L as measured by each technique would positively correlate with SOC_T. The final objective was to evaluate the relationship of the quantity of SOC_L with macroaggregate size to understand the role of physical protection in determining levels of SOC_L. Because we hypothesized that aggregates protect labile C from decomposition, we predicted that aggregate size would be positively correlated with SOC_L measured with techniques that destroy aggregates and determine the amount of C inside, such as LFC, hydrolyzable C, incubation labile C, and respired soil C from 12 d of laboratory incubation.

To provide a range of SOM characteristics, we chose soils from 33 grasslands that had similar parent material, topography, climate, and vegetation. All but two sites had been cultivated, but differed in their time since the cessation of agriculture. Thus, the main factor causing differences in SOC_T and SOC_L levels between sites was time since cessation of agriculture, which produced a wide range of SOC_L values without variation in other soil-forming factors that may have necessitated altering the analytical methods.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Study Area and Soil Sampling
Soil samples were collected from 33 sites in northwestern Minnesota. Thirty-one sites are former agricultural fields that had been converted to perennial grassland at different times in the past 40 yr. Sites are either owned by private landowners and enrolled in the Conservation Reserve Program (CRP) or owned by the Fish and Wildlife Service. Sites were located in adjoining Ottertail and Grant Counties, encompassing an area approximately 50 km in radius. The three most common types of seed mix used for grassland establishment on all sites were (i) monospecific stands of Bromus inermis Leyss., a perennial, nonnative C₃ grass; (ii) C₃ perennial grasses (generally B. inermis) planted with legumes (generally Medicago sativa L.); and (iii) perennial native C₄ grasses, generally Andropogon gerardii Vitman, Sorghastrum nutans (L.) Nash, and Panicum virgatum L. All of the sites therefore contained perennial grassland vegetation that had not changed significantly in composition since agricultural abandonment.

Two sites have native prairie soils that have never been plowed. All sites share soil parent material of calcareous glacial till from the Laurentide Ice Sheet, deposited approximately 12000 yr before present. There are slight differences in characteristics such as soil texture and landscape relief among the sites; the soils are from three different soil series complexes: Formdale (fine-loamy, mixed, superactive, frigid Calcic Hapludolls), Sisseton (coarse-loamy, mixed, superactive, frigid Typic Eutrudepts), and Barnes (fine-loamy, mixed, superactive, frigid Calcic Hapludolls) (Soil Survey Staff, 1997).

On the shoulder slope position at each site, five 1.9-cm-diam. cores were taken to the 10-cm depth within each of three 2- by 2-m plots and composited. Soil samples were kept at 4°C for a maximum of 3 d until they could be processed. Some of the fresh soil was passed through a 2-mm sieve to remove rocks and plant roots. Subsamples of this soil were taken for analysis, including microbial biomass and the laboratory incubation, and the remainder was air-dried and ground. The rest of the fresh soil sample was passed through a 4-mm sieve and immediately air-dried for later analysis of aggregate size structure.

Soil Carbon Methods
Total organic C and total N for soil samples were determined by combustion with an elemental analyzer (Model ECS 4010, COSTECH Analytical, Valencia, CA). The SOC_T was measured after pretreatment with phosphoric acid to remove carbonates.

Measurements of respiration from laboratory-incubated soils were used to determine labile and recalcitrant C pools. Twenty grams of fresh soil were adjusted to a common moisture (30%) and incubated aerobically in glass jars for 360 d at 25°C in the dark. At each sampling period (1, 3, 6, 12, 28, 45, 63, 95, 119, 169, 224, and 360 d), jars were capped and the headspace sampled through a septum in the jar lid after 0 and 24 h (Robertson et al., 1999). The headspace gas was analyzed for CO₂ concentration on a Shimadzu gas chromatograph using a thermal conductivity detector and a Poropak N column. Rates of C mineralization were determined from the difference between final and initial CO₂ concentrations. Jars were covered with polyethylene film between sampling dates to allow for O₂ exchange, and samples were periodically adjusted to 30% gravimetric moisture with deionized water.

The C that is mineralized initially is considered to be labile, while the remaining fraction is considered recalcitrant (Townsend et al., 1997). The quantities and rates of labile and recalcitrant C mineralized during the course of the incubation were calculated by parameterizing two-pool models (Alvarez and Alvarez, 2000) with nonlinear curvefitting in JMP 5.0 (SAS Institute, Cary, NC) to the equation

where C_t is the known cumulative amount of C respired at sampling period t, C_l is the size of the labile C pool, k_l is the decay constant for the labile pool, and k_r is the mineralization rate for the recalcitrant pool which was held constant to constrain estimates of C_l (Wedin and Pastor, 1993). During the curve fitting, C_t and C_l were expressed as micrograms of C per gram of soil, k_l was expressed per day, and k_r was expressed as micrograms of C per gram of soil per day. These values were then converted to an areal basis using bulk density. The C_l was initially estimated without constraining k_r. Then, the average value of k_r for all samples obtained with this approach, 3 µg g^–1 soil d^–1, was used to set the k_r constant during the constrained curvefitting. The cumulative amount of C that had been mineralized 12 d after the incubation began was calculated separately using mineralization rates from the first four sampling periods. Day 12 was chosen as an estimate of SOC_L because it represents the initial period of mineralization, when the majority of the material being mineralized is labile C.

Microbial biomass C and N were determined by chloroform fumigation and direct extraction on fresh soils (Anderson and Joergensen, 1997). One 10-g sample of fresh soil was extracted with 0.5 M potassium sulfate solution. A second 10-g sample was placed in a vacuum desiccator in a chloroform atmosphere for 5 d, after which it was extracted with 0.5 M potassium sulfate solution. Filtered extracts were analyzed for total organic C and total N using a Shimadzu total organic C and total N analyzer (TC VCPM analyzer, Kyoto, Japan). Chloroform-labile microbial biomass N and C were calculated as the difference in extractable N and C before and after fumigation, with correction factors of K_en = 0.54 (Brookes et al., 1985) and K_ec = 0.45 (Beck et al., 1997), respectively.

Air-dried soil samples were separated into light fraction and heavy fraction material by flotation with a heavy liquid (aqueous sodium iodide) adjusted to a density of 1.7 g cm^–3. The light fraction represents poorly decomposed, relatively labile SOM, and the heavy fraction corresponds to mineral-associated, more recalcitrant SOM (Gregorich et al., 1989; Jastrow, 1996). Sodium iodide solution (70 mL) was added to soil samples (20 g) which were then sonicated to disrupt soil structure with an energy of 225 J mL^–1 applied across 12 min. The soil material was allowed to separate for 48 h. The light fraction material was isolated, washed with deionized water, dried overnight in a 65°C oven, weighed, and analyzed for total C and N by combustion.

Soil C was separated into two chemical fractions, labile and resistant, using an acid digest (Sollins et al., 1999). Identifiable plant materials were removed from air-dried, ground soil, and inorganic carbonates were removed with hydrochloric acid pretreatment. Soil samples (1 g) were refluxed for 16 h in digestion tubes with 10 mL of 6 M hydrochloric acid solution. The residue was isolated, washed with 100 mL of deionized water, dried overnight in an 80°C oven, weighed, and analyzed for total C by combustion. The chemically-resistant residue, or unhydrolyzable fraction, represents recalcitrant soil C pools, and has yielded ¹⁴C dates much older than bulk soil (Paul et al., 1997). The acid digest hydrolyzes polysaccharides and nitrogenous material, leaving the polyaromatic humics and lignin (Martel and Paul, 1974). The hydrolyzable fraction, which when considering three soil C pools represents a measure of slow and labile C, was obtained by subtracting unhydrolyzable C from SOC_T. Here, because we are considering two pools, the hydrolyzable fraction can be considered more labile than unhydrolyzable C.

Aggregate Size
Aggregate size distribution was measured by physical separation of soil fractions through wet sieving. An air-dried soil sample of 50 g that had been passed through a 4-mm screen was wetted for 10 min, and soil was sieved mechanically for 10 min into five water-stable aggregate size classes: >2000 µm, 1000 to 2000 µm, 500 to 1000 µm, 250 to 500 µm, and <250 µm (Cambardella and Elliott, 1992; Six et al., 1998). An index of aggregate size, geometric mean diameter (GMD), was calculated with the formula

where w_i is the weight of aggregates in a size class with average diameter x_i and w_s is the weight of the sample (Kemper and Chepil, 1965).

Data Analysis
Because of both the lack of an independent variable and measurement error in all variables, linear relationships between different methods of measuring labile soil C were analyzed with Model II orthogonal regression, also called errors-in-variables regression (Sokal and Rohlf, 1994). In contrast to least squares regression, Model II regression adjusts for variance in both variables, similar to correlations (Warton and Weber, 2002). Statistical significance indicates rejection of the null hypothesis that the slope of the regression line equals one. Model II regressions were performed between all five labile soil C measurements, SOC_T, and aggregate size using JMP 5.0.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The SOC_T among sites varied threefold, ranging from 18 to 70 Mg ha^–1. There were positive relationships between SOC_T and microbial C, LFC, hydrolyzable C, Day 12 labile C, and C_l (Table 2, Fig. 1). These positive relationships indicate that labile C increased with SOC_T. On average, microbial biomass C was 2% of SOC_T, LFC was 5% of SOC_T, C_l was 4% of SOC_T, and hydrolyzable C was 55% of SOC_T.

View larger version (24K): [in this window] [in a new window]   Fig. 1. Model II regressions between total soil organic carbon (SOCT) and five different methods for measuring labile soil organic carbon (SOCL): (a) light fraction C (P < 0.0005), (b) microbial C (P < 0.00001), (c) hydrolyzable C (P < 0.00001), (d) incubation labile C (P < 0.0005), and (e) Day 12 labile C (P < 0.005). All measurements were made on samples from the 0- to 10-cm soil depth.

View this table: [in this window] [in a new window]   Table 3. Estimates of slope and confidence limits ( = 0.05) for slope estimates obtained with Model II regression between total soil organic carbon (SOC, Mg ha–1) and different methods of determining labile SOC (kg ha–1).

Chemical fractionation provided much higher estimates of SOC_L than either light fraction or microbial C, and the difference between techniques was greater when labile C content was higher. Acid hydrolysis provided estimates of SOC_L that were consistently greater than SOC_L obtained with the other three methods (Fig. 2). The y-intercept for the relationships between SOC_L measured with acid hydrolysis and microbial C was 2183 kg ha^–1, while that between hydrolyzable C and LFC was 2988 kg ha^–1, indicating that acid hydrolysis produces higher estimates of SOC_L than either microbial or LFC at low SOC_L (Fig. 2). Additionally, because the slopes of these relationships were both greater than one (21.6 and 8.9, respectively), the differences between measurement techniques increased as the labile pool increased. Similarly, the slopes of the relationships between Day 12 labile C and the other three measurements were all greater than one—15 for LFC, 142 for hydrolyzable C, and 6.5 for microbial C—indicating that SOC_L obtained from those three techniques increased at a greater rate than did Day 12 labile C.

View larger version (34K): [in this window] [in a new window]   Fig. 2. Model II regressions between different techniques to measure labile soil C pools: (a) microbial C and light fraction C (P < 0.005), (b) microbial C and hydrolyzable C (P < 0.00001), (c) light fraction C and hydrolyzable C (P < 0.001), (d) Day 12 labile C and light fraction C (P < 0.001), (e) Day 12 labile C and hydrolyzable C (P < 0.0005), and (f) Day 12 labile C and microbial C (P < 0.005). All measurements were made on samples from the 0- to 10-cm soil depth.

View larger version (33K): [in this window] [in a new window]   Fig. 3. Model II regressions between incubation labile carbon (Cl) and other techniques for measuring labile soil organic C pool sizes: (a) light fraction C (P < 0.2), (b) hydrolyzable C (P < 0.001), (c) microbial biomass C (P < 0.1), and (d) Day 12 labile C (P < 0.00001). The Cl is determined with a two-pool model using constrained nonlinear curve fitting to respiration data from a 360-d laboratory incubation. All measurements were made on samples from the 0- to 10-cm soil depth.

View larger version (17K): [in this window] [in a new window]   Fig. 4. Cumulative CO2–C respired during the course of a long-term laboratory incubation of soils at 25°C. An equation is fit to the data Ct = Cl (1 – e–klt) + krt, where Ct is the known cumulative amount of C respired at sampling period t, Cl is the size of the labile C pool, kl is the decay constant for the labile pool, and kr is the mineralization rate for the recalcitrant pool, which is set at 3 µg g–1 soil d–1. For this sample, Cl = 1544 kg ha–1, kl = 0.012 d–1.

View larger version (17K): [in this window] [in a new window]   Fig. 5. Relationship between the size of the labile carbon pool (Cl) and the decay constant for the labile pool (kl). These two parameters are simultaneously determined by fitting the curve Ct = Cl (1 – e–klt) + krt, where Ct is the known cumulative amount of C respired at sampling period t, and kr is the mineralization rate for the recalcitrant pool which is set at 3 µg g–1 soil d–1 (r2 = 0.55, P < 0.0001, y = 0.025 – 0.0000065x).

View larger version (21K): [in this window] [in a new window]   Fig. 6. Relationship between the size of the labile C pool (Cl) determined without constraining kr and Cl determined with kr set at 3 µg g–1 soil d–1 (P < 0.9).

View larger version (17K): [in this window] [in a new window]   Fig. 7. Model II regressions between an index of soil aggregate size, geometric mean diameter (GMD), compared with two different soil C measurements: (a) hydrolyzable C (P < 0.01) and (b) total SOC (P < 0.05). Higher values of GMD indicate larger aggregate size. All measurements were made on samples from the 0- to 10-cm soil depth.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The four techniques that were positively correlated—microbial biomass C, LFC, hydrolyzable C, and Day 12 labile C—are all appropriate to use as an index of SOC_L across a range of SOC_L levels, although they are closer to each other in absolute amount at low SOC_L. Collins et al. (2000) found that labile C comprised 3 to 8% of SOC_T, similar to the range found in this study. Differences in SOC_L pool size among these techniques become important when the absolute value of SOC_L is of interest, but each individual technique seems to provide an accurate index of relative changes in SOC_L.

The hypothesis that hydrolyzable C would be greater than SOC_L measured with other techniques was supported, probably because hydrolyzable C includes more slow-turnover material than the other techniques. The hypothesis that microbial biomass C would be the smallest SOC_L pool was not supported because Day 12 labile C included a smaller portion of SOC_L, but both microbial biomass C and Day 12 labile C could be considered indicators of SOC_L.

As predicted, SOC_L as measured by each technique was positively correlated with SOC_T, although microbial biomass C had a lower rate of increase and hydrolyzable C had a higher rate of increase as SOC_T increased compared with other techniques. Many studies have shown positive correlations between microbial biomass and SOC_T (Powlson et al., 1987), and it is not surprising that other techniques to measure SOC_L also show this pattern. However, it had not been demonstrated that SOC_L, as measured by different techniques, increases in different ways as SOC_T increases. This pattern warrants further investigation, as it implies that detecting differences in labile C pool sizes between treatments with different amounts of SOC_T may be difficult with a technique such as microbial biomass C that does not increase greatly with increases in SOC_T (Hargreaves et al., 2003).

Because hydrolyzable C, microbial biomass C, Day 12 labile C, and LFC correlated well with each other, at a minimum they all provide an index of SOC_L. However, no single technique should be considered a direct measurement of SOC_L. For example, the size of the more labile pool as measured by acid hydrolysis, which is regularly >50% of SOC_T, is not a small pool with short turnover time. As such, hydrolyzable C includes more C than just biologically labile C, as has been found in other studies (Paul et al., 2001). Conversely, microbial biomass does not encompass all SOC_L in a soil sample; there is some SOC_L that is not contained within microbial bodies. The C respired by Day 12 of the laboratory incubation does not include the entire labile C pool. Nor does light fraction material, as it is mostly derived from identifiable plant material and may not include soluble labile C.

Although there were good correlations among these four techniques, the differences in slope and intercept in the linear relationships indicate that the techniques are fractionating the soil C pool differently and that the differences among techniques are dependent on the amount of SOC_L. Because of systematic variation in the absolute size of pools measured with each technique, they could be measuring different C pools or just different-sized portions of the same labile pool. It is impossible to distinguish these alternatives with direct comparisons of the SOC_L pool size measured with different techniques. The best correlation was between hydrolyzable C and microbial biomass C, but the closest absolute values were between microbial biomass C and LFC. The differences between techniques were higher when labile C was greater, making comparisons between treatments in systems where labile C is increasing or high (such as in natural systems) more difficult than comparisons where labile C is decreasing or depleted, such as in agricultural systems.

Estimating the size of SOC_L with the curve-fitting technique may be the most accurate technique because the laboratory incubation most closely approximates natural decomposition in the soil, and mathematical estimation of SOC_L avoids error associated with choosing a length of incubation. However, this technique must be applied carefully. There are several equations to choose from that may be biologically relevant (Alvarez and Alvarez, 2000), and sometimes parameters must be constrained to give accurate results. For example, in this study, C_l estimated without constraining k_r did not correspond with C_l estimated while constraining k_r.

The curve-fitting technique provides information about the decay constant of the labile pool in addition to the pool size of SOC_L. If information about k_l is desirable, the curve-fitting technique must be used, which requires relatively long incubation times. When comparing C_l between samples, variation in k_l must be considered. C_l correlates with other measurements of SOC_L, especially hydrolyzable C (Collins et al., 2000).

Positive relationships between aggregate size, acid-hydrolyzable C, and SOC_T support the idea that as aggregates form, they physically protect C (Jastrow, 1996), but the characteristics of this physically-protected C are not clear. Furthermore, these results indicate that aggregate size, rather than mass or amount of aggregates, is important for influencing the amount of soil C since larger aggregates were associated with higher levels of total and unhydrolyzable C. These data support an accretive model of aggregate development where microaggregates and small macroaggregates join together to form large macroaggregates (Six et al., 2000).

The prediction that aggregate size would positively correlate with all measurements of SOC_L was not supported. There are several different possible reasons that the relationships between aggregate size and microbial biomass, LFC, Day 12 labile C, and C_l were not very strong. It may be that the C inside soil aggregates is more biochemically recalcitrant than other studies have suggested (Six et al., 2001). Alternatively, we separated aggregates into larger size classes than those that have been useful for estimating labile C dynamics in agricultural systems, and perhaps the dynamics of microaggregates show different patterns than those of macroaggregates (Oades, 1988; Six et al., 1998). Second, aggregate structure may be preserved, at least partially, in the soil incubation, preventing some SOC_L from being decomposed. In this case, no relationship would be expected between aggregate size and Day 12 labile C and C_l. Third, we did not measure the C content of the different aggregate size classes, which may have correlated better with measures of labile C than aggregate size (Jastrow, 1996). Techniques that separate inter- and intraaggregate SOC_L may provide additional information that makes explicit the role of physical protection in labile C dynamics.

All four empirical methods used here assume two pools of SOM, but considering more pools may help clarify SOM dynamics. Currently, more than two pools can be measured with multiple density fractionations (Sohi et al., 2001) or fractionation schemes with multiple techniques can be used in conjunction (Paul et al., 2001). We wanted to investigate and compare simple techniques for fractionating SOC. There are several other well-established techniques to measure labile C pool sizes or turnover times that were not used in this study, including POM (Cambardella and Elliott, 1992) and permanganate digest (Weil et al., 2003), and it would be useful to compare the results of these methods against the four considered here. Some techniques with intriguing possibilities to understand labile C composition are nuclear magnetic resonance spectrometry (Six et al., 2001) and pyrolysis (Neff et al., 2002), although interpretation and biological meaning are more difficult than the analyses.

	ACKNOWLEDGMENTS

 
This work was conducted by K. McLauchlan at the University of Minnesota. We thank Jennifer King, Sherri Morris, and Cathleen McFadden for analytical assistance. Joe Craine, Dorian Hasselmann, and Bryan Holtz provided field and lab assistance. The Land Institute, the Andrew Mellon Foundation, and Dayton-Wilkie Natural History Funds provided financial support. We extend thanks to Deborah Allan, Sherri Morris, Michael Russelle, Joe Craine, and three anonymous reviewers for helpful discussion and comments on an earlier version of the manuscript.

Received for publication March 7, 2004.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

• Alvarez, R., and C.R. Alvarez. 2000. Soil organic matter pools and their associations with carbon mineralization kinetics. Soil Sci. Soc. Am. J. 64:184–189.[Abstract/Free Full Text]
• Anderson, T.H., and R.G. Joergensen. 1997. Relationship between SIR and FE estimates of microbial biomass C in deciduous forest soils at different pH. Soil Biol. Biochem. 29:1033–1042.
• Beck, T., R.G. Joergensen, E. Kandeler, F. Makeschin, E. Nuss, H.R. Oberholzer, and S. Scheu. 1997. An inter-laboratory comparison of ten different ways of measuring soil microbial biomass carbon. Soil Biol. Biochem. 29:1023–1032.
• Brookes, P.C., A. Landman, G. Pruden, and D.S. Jenkinson. 1985. Chloroform fumigation and the release of soil nitrogen: A rapid direct extraction method to measure microbial biomass nitrogen in soil. Soil Biol. Biochem. 17:837–842.
• Cambardella, C.A., and E.T. Elliott. 1992. Particulate soil organic-matter changes across a grassland cultivation sequence. Soil Sci. Soc. Am. J. 56:777–783.[Abstract/Free Full Text]
• Collins, H.P., E.T. Elliott, K. Paustian, L.G. Bundy, W.A. Dick, D.R. Huggins, A.J.M. Smucker, and E.A. Paul. 2000. Soil carbon pools and fluxes in long-term corn belt agroecosystems. Soil Biol. Biochem. 32:157–168.
• Ding, G., J.M. Novak, D. Amarasiriwardena, P.G. Hunt, and B. Xing. 2002. Soil organic matter characteristics as affected by tillage management. Soil Sci. Soc. Am. J. 66:421–429.[Abstract/Free Full Text]
• Doran, J.W., A.J. Jones, M.A. Arshad, and J.E. Gilley. 1999. Determinants of soil quality and health. p. 17–36. In R. Lal (ed.) Soil quality and soil erosion. Lewis Publ., Boca Raton, FL.
• Gregorich, E.G., and H.H. Janzen. 1996. Storage of soil carbon in the light fraction and macroorganic matter. p. 167–190. In M.A. Carter and B.A. Stewart (ed.) Structure and organic matter storage in agricultural soils. Lewis Publ., Boca Raton, FL.
• Gregorich, E.G., M.H. Beare, U. Stoklas, and P. St-Georges. 2003. Biodegradability of soluble organic matter in maize-cropped soils. Geoderma 113:237–252.[ISI]
• Gregorich, E.G., R.G. Kachanoski, and R.P. Voroney. 1989. Carbon mineralization in soil size fractions after various amounts of aggregate disruption. J. Soil Sci. 40:649–660.
• Hargreaves, P.R., P.C. Brookes, G.J. Ross, and P.R. Poulton. 2003. Evaluating soil microbial biomass carbon as an indicator of long-term environmental change. Soil Biol. Biochem. 35:401–407.
• Jastrow, J.D. 1996. Soil aggregate formation and the accrual of particulate and mineral-associated organic matter. Soil Biol. Biochem. 28:665–676.
• Jenkinson, D.S. 1990. The turnover of organic carbon and nitrogen in soil. Philos. Trans.: Biol. Sci. 329:361–367.
• Karlen, D.L., J.C. Gardner, and M.J. Rosek. 1998. A soil quality framework for evaluating the impact of CRP. J. Prod. Agric. 11:56–60.
• Kemper, W.D., and W.S. Chepil. 1965. Size distribution of aggregates. p. 499–510. In C.A. Black et al. (ed.) Methods of soil analysis. Part I—Physical and mineralogical properties, including statistics of measurement and sampling. Agron. Monogr. 9. ASA, Madison, WI.
• Kristensen, H.L., K. Debosz, and G.W. McCarty. 2003. Short-term effects of tillage on mineralization of nitrogen and carbon in soil. Soil Biol. Biochem. 35:979–986.
• Krull, E.S., J.A. Baldock, and J.O. Skjemstad. 2003. Importance of mechanisms and processes of the stabilisation of soil organic matter for modelling carbon turnover. Funct. Plant Biol. 30:207–222.
• Mahmood, T., F. Azam, F. Hussain, and K.A. Malik. 1997. Carbon availability and microbial biomass under an irrigated wheat maize cropping system receiving different fertilizer treatments. Biol. Fertil. Soils 25:63–68.
• Martel, Y.A., and E.A. Paul. 1974. Effects of cultivation on the organic matter of grassland soils as determined by the fractionation and radiocarbon dating. Can. J. Soil Sci. 54:419–426.
• Neff, J.C., A.R. Townsend, G. Gleixner, S.J. Lehman, J. Turnbull, and W.D. Bowman. 2002. Variable effects of nitrogen additions on the stability and turnover of soil carbon. Nature (London) 419:915–917.[Medline]
• Oades, J.M. 1988. The retention of organic matter in soils. Biogeochemistry 5:35–70.
• Parton, W.J., D.S. Schimel, C.V. Cole, and D.S. Ojima. 1987. Analysis of factors controlling soil organic matter levels on Great Plains grasslands. Soil Sci. Soc. Am. J. 51:1173–1179.[Abstract/Free Full Text]
• Pastor, J., B. Dewey, R.J. Naiman, P.F. McInnes, and Y. Cohen. 1993. Moose browsing and soil fertility in the boreal forests of Isle Royale National Park. Ecology 74:467–480.[ISI]
• Paul, E.A., R.F. Follett, S.W. Leavitt, A. Halvorson, G.A. Peterson, and D.J. Lyon. 1997. Radiocarbon dating for determination of soil organic matter pool sizes and dynamics. Soil Sci. Soc. Am. J. 61:1058–1067.[Abstract/Free Full Text]
• Paul, E.A., D. Harris, M.J. Klug, and R.W. Ruess. 1999. The determination of microbial biomass. p. 291–317. In G.P. Robertson (ed.) Standard soil methods for long-term ecological research. Oxford Univ. Press, New York.
• Paul, E.A., S.J. Morris, and S. Bohm. 2001. Determination of soil C pool sizes and turnover rates: Biophysical fractionation and tracers. p. 193–206. In R. Lal (ed.) Assessment methods for soil carbon. Lewis Publ., Boca Raton, FL.
• Paul, E.A., S.J. Morris, J. Six, K. Paustian, and E.G. Gregorich. 2003. Interpretation of soil carbon and nitrogen dynamics in agricultural and afforested soils. Soil Sci. Soc. Am. J. 67:1620–1628.[Abstract/Free Full Text]
• Powlson, D.S., P.C. Brookes, and B.T. Christensen. 1987. Measurements of soil microbial biomass provide an early indication of changes in total soil organic matter due to straw incorporation. Soil Biol. Biochem. 19:159–164.
• Robertson, G.P., D. Wedin, P.M. Groffman, J.M. Blair, E.A. Holland, K.J. Nadelhoffer, and D. Harris. 1999. Soil carbon and nitrogen availability: Nitrogen mineralization, nitrification, and soil respiration potentials. p. 258–271. In G.P. Robertson (ed.) Standard soil methods for long-term ecological research. Oxford Univ. Press, New York.
• Six, J., E.T. Elliott, and K. Paustian. 2000. Soil macroaggregate turnover and microaggregate formation: A mechanism for C sequestration under no-tillage agriculture. Soil Biol. Biochem. 32:2099–2103.
• Six, J., E.T. Elliott, K. Paustian, and J.W. Doran. 1998. Aggregation and soil organic matter accumulation in cultivated and native grassland soils. Soil Sci. Soc. Am. J. 62:1367–1377.[Abstract/Free Full Text]
• Six, J., G. Guggenberger, K. Paustian, L. Haumaier, E.T. Elliott, and W. Zech. 2001. Sources and composition of soil organic matter fractions between and within soil aggregates. Eur. J. Soil Sci. 52:607–618.
• Smith, P., J.U. Smith, D.S. Powlson, W.B. McGill, J.R.M. Arah, O.G. Chertov, K. Coleman, U. Franko, S. Frolking, D.S. Jenkinson, L.S. Jensen, R.H. Kelly, H. Klein-Gunnewiek, A.S. Komarov, C. Li, J.A.E. Molina, T. Mueller, W.J. Parton, J.H.M. Thornley, and A.P. Whitmore. 1997. A comparison of the performance of nine soil organic matter models using datasets from seven long-term experiments. Geoderma 81:153–225.[ISI]
• Sohi, S.P., N. Mahieu, J.R.M. Arah, D.S. Powlson, B. Madari, and J.L. Gaunt. 2001. A procedure for isolating soil organic matter fractions suitable for modeling. Soil Sci. Soc. Am. J. 65:1121–1128.[Abstract/Free Full Text]
• Soil Survey Staff. 1997. Soil Survey of Ottertail County, Minnesota, USA. USDA-NRCS, U.S. Gov. Print. Office, Washington, DC.
• Sokal, R., and J. Rohlf. 1994. Biometry: The principles and practice of statistics in biological research. W.H. Freeman, New York.
• Sollins, P., C. Glassman, E.A. Paul, C. Swanston, K. Lajtha, J.W. Heil, and E.T. Elliott. 1999. Soil carbon and nitrogen: Pools and fractions. p. 89–105. In G.P. Robertson (ed.) Standard soil methods for long-term ecological research. Oxford Univ. Press, New York.
• Townsend, A.R., P.M. Vitousek, D.J. Desmarais, and A. Tharpe. 1997. Soil carbon pool structure and temperature sensitivity inferred using CO₂ and ¹³CO₂ incubation fluxes from five Hawaiian soils. Biogeochemistry 38:1–17.
• Warton, D.I., and N.C. Weber. 2002. Common slopes test for bivariate errors-in-variables models. Biometrical J. 44:161–174.
• Wedin, D.A., and J. Pastor. 1993. Nitrogen mineralization dynamics in grass monocultures. Oecologia 96:186–192.
• Weil, R.R., K.R. Islam, M.A. Stine, J.B. Gruver, and S.E. Samson-Liebig. 2003. Estimating active carbon for soil quality assessment: A simplified method for laboratory and field use. Am. J. Alternative Agric. 18:3–17.

This article has been cited by other articles:

				B. Majumder, B. Mandal, P. K. Bandyopadhyay, A. Gangopadhyay, P. K. Mani, A. L. Kundu, and D. Mazumdar Organic Amendments Influence Soil Organic Carbon Pools and Rice-Wheat Productivity Soil Sci. Soc. Am. J., May 1, 2008; 72(3): 775 - 785. [Abstract] [Full Text] [PDF]

				L. E. Kellogg, S. D. Bridgham, and D. Lopez-Hernandez A Comparison of Four Methods of Measuring Gross Phosphorus Mineralization Soil Sci. Soc. Am. J., June 21, 2006; 70(4): 1349 - 1358. [Abstract] [Full Text] [PDF]

				E. A. Paul, S. J. Morris, R. T. Conant, and A. F. Plante Does the Acid Hydrolysis-Incubation Method Measure Meaningful Soil Organic Carbon Pools? Soil Sci. Soc. Am. J., April 19, 2006; 70(3): 1023 - 1035. [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 (22) Citing Articles via Google Scholar Google Scholar Articles by McLauchlan, K. K. Articles by Hobbie, S. E. Search for Related Content PubMed Articles by McLauchlan, K. K. Articles by Hobbie, S. E. GeoRef GeoRef Citation Agricola Articles by McLauchlan, K. K. Articles by Hobbie, S. E. Related Collections Ecosystem Restoration Soil Organic Matter Soil Analysis