Printer Friendly

Changes in chemical nature of soil organic carbon in Vertisols under wheat in south-eastern Queensland.


Cultivation and crop production can have profound effects on the C levels of soils. Prolonged cultivation invariably leads to a decline in C and N levels in soil, particularly when virgin land is first brought under cultivation (Dick 1983; Dormaar 1983; Dalal and Mayer 1986b). In Australia, these C and N losses from the surface horizon of soils have been highly variable (Russell and Williams 1982; Dalal and Mayer 1986b; Russell and Jones 1996). In addition, long-term cultivation also leads to degradation of other soil properties associated with soil organic matter including soil fertility, leading to decline in crop yields (Dalal and Mayer 1986a).

Decline in total organic C (TOC) can be attributed to a number of factors including reduced input from the crop compared with the native vegetation, changes in soil water status, and the physical effects of tillage which increase soil aeration as well as ensuring that soil decomposer organisms are constantly brought into contact with crop residues and soil C fractions. Although changes in TOC were often large following prolonged cropping, these changes were found to be even more pronounced in organic C content of different particle size fractions of these soils, with the organic C in the sand-size fractions ([is greater than] 20 [micro]m) declining the most and in the clay-size fractions ([is less than] 2 m[micro]) the least (Dalal and Mayer 1986c). Similar observations were reported for fractions separated on the basis of particle density (Dalal and Mayer 1986d). These observations indicated that some form of physical protection at the clay/microaggregate level was important in dictating the relative decomposability of a range of soil organic C fractions or pools. These studies were confirmed by Skjemstad et al. (1986) and Skjemstad and Dalal (1987) using a combination of solid-state [sup.13]C nuclear magnetic resonance and infrared spectroscopy. Significant differences in the chemistry of TOC and various soil organic C fractions were also observed. For example, the organic C from a Waco soil was shown to be more aromatic than that from a Langlands-Logie soil, indicating that differences in the chemical structure of the soil organic C in different soils and even in fractions or pools from the same soil may also be important in moderating decomposition.

Skjemstad et al. (1996) recently demonstrated that char (charcoal) could contribute significantly to the organic C content of some soils and was unevenly distributed through particle size fractions, being concentrated in the clay and silt fractions and usually absent in the sand fractions of soils. Char is determined along with other soil organic C fractions (Skjemstad and Taylor 1999) but is unlikely to play an active role in soil organic C dynamics, probably persisting in the soil environment for centuries if not millennia. Recently, Skjemstad et al. (1999b) showed that a Waco soil (Pellustert) contained a significant proportion (30%) of its soil organic C as char C, while a grey clay (Chromustert) from the same region contained much less char C (4%). The presence of variable quantities of char C could, therefore, also impact significantly on the C dynamics in the soils reported by Dalal and Mayer (1986a).

The aim of the present study was to determine the major mechanisms of soil organic C protection by comparing the chemistry of the organic C from a number of fractions isolated from 2 contrasting soil types after different times of cultivation. The high energy UV photo-oxidation technique described by Skjemstad et al. (1993) was used to isolate and estimate soil char content and the impact of physical protection at the microaggregate level. The isolated fractions were also used as input parameters for the RothC soil C turnover model (Jenkinson 1990) to test if fractions separated by a combination of physical and chemical techniques could be used to initialise and parameterise such models and to study their simulation of soil C dynamics in cultivated soils.

Materials and methods


Two soil types (Waco and Langlands-Logie) were selected for this study. Land use and detailed sampling procedures are described by Dalal and Mayer (1986a). Sample locations, years under cultivation, sampling depths, and some physical and chemical properties are given in Table 1.

Table 1. Soil type, taxonomy, duration under cultivation, and depth, and some physical and chemical properties of the two soils
Years Depth pH C N C/N Clay Ca[CO
 (B) (cm) ([H.sub (mg/g soil) .sub.3]
 .2]O) %

 Waco Pellustert(A)

 0 0-10 8.0 18.2 1.34 13.6 69 0.1
 0 10-20 8.4 10.8 0.93 11.6 78 0.5
 0 20-30 8.9 9.9 0.71 13.9 77 1.1
 0 90-120 9.2 4.4 0.38 11.6 71 2.9
 19 0-10 8.2 12.5 0.89 14.0 78 0.2
 35 0-10 8.0 12.1 0.83 14.6 78 0.1
 50 0-10 8.4 11.4 0.74 15.4 80 0.1
 50 10-20 8.6 8.7 0.67 13.0 82 0.5
 50 20-30 8.7 8.1 0.59 13.7 80 0.6

 Langlands-Logie Chromustert(A)

 0 0-10 7.1 27.0 2.41 11.2 50 1.3
 0 10-20 8.0 11.0 1.21 9.1 55 2.8
 0 20-30 8.4 8.7 0.94 9.3 53 3.4
 0 90-120 5.7 2.8 0.12 23.3 62 n.d.
 20 0-10 7.8 11.9 1.12 10.6 53 1.9
 35 0-10 8.2 9.4 0.89 10.6 50 0.9
 45 0-10 8.2 7.5 0.77 9.7 52 1.9
 45 10-20 8.4 5.0 0.57 8.8 57 2.4
 45 20-30 8.8 4.7 0.47 10.0 57 2.1

(A) Soil Survey Staff (1989).

(B) Years under cultivation.

Carbon analyses

Soil samples and particle-size fractions [is greater than] 53 [micro]m were analysed on a LECO CR12 carbon analyser as described by Merry and Spouncer (1988). Corrections for carbonate C, determined by Collins calcimeter (Loveday and Reeve 1974), were made where appropriate. Fractions [is less than] 53 [micro]m were analysed by a modified Heanes (1984) method as described by Skjemstad et al. (1996).

Separation of fractions

Soil samples (10 g), ground to pass through a 0.25-mm sieve, were shaken overnight with 40 mL of 0.5% sodium hexametaphosphate solution. Calcareous soils were first treated with [SO.sub.2] solution to remove Ca[CO.sub.3], centrifuged, and washed with water. Sodium-saturated soil samples were given a short ultrasonic dispersion (20 s) with a Branson B30 ultrasonic probe fitted with a 13-mm horn and set to 70% power and 50% pulse. After allowing the sediment to settle for about 1 min, the suspended material was passed through a 200-[micro]m sieve over a 53-[micro]m sieve. The process was repeated twice, and during the final treatment, all of the sediment was washed into the sieves. The sediment, which consisted of sand and particulate organic matter, was gently worked with a rubber policeman to ensure clay/silt material was passed through the sieves and that no soil aggregates were retained. The suspensions were collected in measuring cylinders and made to 500 mL, and thoroughly mixed before carbon analysis and photo-oxidation. The organic material retained on the sieves is collectively defined as particulate organic carbon (POC; Camberdella and Elliott 1992).

Photo-oxidation of fractions

The photo-oxidation procedure used was essentially the same as that described by Skjemstad et al. (1996). Aliquots of the [is less than] 53 [micro] fractions, containing 2-2.5 mg of organic carbon, were placed in quartz tubes and exposed to a 2 kW Hg vapor lamp for 0.5, 1, 2, and 4 h. Air at a rate of 50 mL/min was injected into each tube to provide a source of oxygen and to ensure that the soil particles remained in suspension during the treatment. After photo-oxidation, the suspensions were transferred to 50 mL centrifuge tubes and washed 3 times with water before analysis for carbon. All photo-oxidations were carried out in duplicate with further replication where samples were required for N and [Delta] [sup.13]C measurements.

[sup.13]C nuclear magnetic resonance spectroscopy

The [sup.13]C cross polarisation with magic angle spinning (CP/MAS) nuclear magnetic resonance (NMR) spectra were obtained at 50.309 MHz on a Varian Unity 200 spectrometer with a 4.7 T wide-bore Oxford superconducting magnet using a 7 mm Doty Scientific MAS probe. Spectra were accumulated using a 1 ms contact time and a 300 ms recycle time, and with other parameters as described by Skjemstad et al. (1994a) after pretreatment with 2% hydrofluoric acid (HF) to remove paramagnetic materials and concentrate the organic fractions. Calcareous soils were treated with [SO.sub.2] solution and washed prior to HF treatment to prevent the potential precipitation of large amounts of calcium fluoride.

Mass spectroscopy

Total N and [Delta] [sup.13]C analyses were carried out by automated nitrogen carbon analysis-mass spectroscopy (ANCA-MS) on a 20-20 Europa Scientific mass spectrometer.

Scanning electron microscopy

Scanning electron microscopy (SEM) was carried out using a Cambridge Stereoscan S250 on samples coated with 20 nm of carbon. Elementary characterisation was performed using a Link AN1000 EDX analyser.

Results and discussion

Soil C and N

The data in Table 1 show that, as reported by Dalal and Mayer (1986b), organic C and N declined rapidly over an initial period, after cultivation, followed by a slower decline with continuing cultivation. Some small changes in C/N ratios are also apparent with continuing cultivation, and again as reported by Dalal and Mayer (1986e), C/N increased in the Waco soil while it decreased in the Langlands-Logie soil. In the virgin sites, the decrease in organic C with depth was much more pronounced in the Langlands-Logie soil than the Waco soil.

[sup.13]C NMR spectra of the soil samples

The chemical nature of the soil samples was assessed using solid-state [sup.13]C CP/MAS NMR spectroscopy following HF treatment. The spectra from each of the surface horizons (0-10 cm) from the virgin sites and from the sites under the longest periods of cultivation are given in Fig. 1. The samples give spectra typical of soil organic matter with contributions from alkyl C (0-45 ppm), O-alkyl C (45-105 ppm), aryl C (105-145 ppm), O-aryl C (145165 ppm), carbonyl C (165-185 ppm), and aldehyde/ketonic C (185-210 ppm). Peaks at 230 and 275 ppm are due to spinning side bands from the aryl and carbonyl peaks, respectively. Cultivation of the Waco site appears to have greatly changed the nature of the soil organic matter resulting in a large increase in the proportion of aryl C with a concomitant decrease in O-alkyl C. At the Langlands-Logie site, little change in the relative distribution is apparent following cultivation.


Solid state [sup.13]C CP/MAS NMR spectra were also determined on the surface samples from the sites of intermediate periods of cultivation. The relative contributions of each of the spectral regions, corrected for spinning side bands (Skjemstad et al. 1994a), and the absolute amounts of each of these functional groups in g C/kg soil are given in Fig. 2. The Waco soil shows a general decrease in the proportion of O-alkyl C between 0 and 20 years with a concomitant increase in aryl and the [f.sub.a] (the fraction of C that is aromatic i.e. aryl C + O-aryl C) but with few other obvious changes (Fig. 2a). Correcting for changes in total soil C with time, however (Fig. 2b), shows that the aryl C content of the soil has remained relatively unchanged with only the O-alkyl C showing any large changes. The O-alkyl C represents polysaccharide structures such as cellulose in various stages of decomposition that would be expected to be susceptible to further decomposition. It is of interest that after the initial significant changes in functional groups, very little further change occurs with the possible exception of a small decrease in O-alkyl C between 35 and 50 years of cultivation. These data suggest, therefore, that at least in the Waco soil the aryl C and its associated functional groups are protected against microbial decomposition.


In the Langlands-Logie soil, there appears to be little change in the chemical composition of the organic matter (Fig. 2c), irrespective of the duration under cultivation, with the possible exception of an increase in alkyl C. Correcting the data for changes in the organic C content clearly shows that as organic C decreases with increasing time of cultivation, these decreases, unlike the situation in the Waco soil, are reflected equally in all functional groups (Fig. 2d).

Baldock et al. (1992) showed that the ratios of O-alkyl C to alkyl C decreased with increased degree of decomposition and suggested that this measurement could be used to ascertain the relative degree of decomposition of organic C within different fractions of a soil. To test if this approach was useful in this study, the O-alkyl/alkyl C ratios were calculated for the two soil types. Table 2 shows that there is a gradual decrease in this ratio with increasing time under cultivation at both sites. This change indicates that the degree of decomposition of the organic C appears to increase with increase in time under cultivation but this increase is small, particularly after the first 20 years.

Table 2. Changes in the O-aikyl C/alkyl C ratios of the two soils (0-10 cm) with increasing time of cultivation
Years under cultivation O-alkyl C/alkyl C

 0 2.0
 19 1.1
 35 1.0
 50 0.8

 0 1.1
 20 0.9
 35 0.9
 45 0.8

Although some clear changes in the chemical composition of the organic matter are evident, these changes could result from movement of material down the profiles, such as in the case of soils under sugar cane (Skjemstad et al. 1999a), rather than from changes resulting in situ through microbial activity. To ascertain any changes in the chemical composition of the organic matter with increased depth in the profiles of the 2 soil types, solid state [sup.13]C CP/MAS NMR spectra were also determined on the 10-20, 20-30, and 90120 cm horizons of the virgin sites and the 10-20 and 20-30 cm horizons of the sites under the longest periods of cultivation.

The relative contributions of each of the spectral regions and the absolute amounts of each of these functional groups in g C/kg soil for the virgin sites are given in Fig. 3. At the Waco site, a large increase in aryl C with concomitant decreases in O-alkyl C and alkyl C are evident (Fig. 3a). Even at a depth of 90-120 cm, aryl C is high but here alkyl C represents a higher proportion of the functionality. When the data are corrected for changes in organic C content (Fig. 3b), the aryl C remains relatively constant to 30 cm with the other functional groups showing similar decreasing trends. At the Langlands-Logie site, less change in functionality is evident in the top 30 cm, although there appears to be an increase in aryl C below 10 cm with a corresponding decrease in alkyl C (Fig. 3c). At 90-120 cm, the organic matter is dominated by alkyl C. Because there is little change in functionality evident with increasing depth, the corrected data (Fig. 3d) show a similar decrease in all functional groups.


After 50 years of cultivation, the Waco profile remains highly aryl in nature with the [f.sub.a] in the 40-50% range with other major functional groups generally near or below 20% (Fig. 4a). The profile now shows a consistency in functional groups in the top 30 cm compared with the virgin site. Correcting for changes in organic C content shows a similar trend but accentuates the decrease in alkyl C between the surface 10 cm and the underlying horizons. A comparison of Figs 4b and 3b shows that the aryl C has been maintained at about 4.0 g C/kg soil but that the other major functional groups have decreased. The Langlands-Logie profile appears to have increased in aryl C relative to other functional groups but generally reflects a similar functionality to that of the virgin site. A comparison of Figs 4d and 3d, however shows that, after 45 years of cultivation, the contents of all functional groups have decreased substantially.


The O-alkyl C to alkyl C ratios of the profiles from the virgin sites show little change with depth within the top 30 cm, decreasing slightly in the Waco soil and increasing slightly in the Langlands-Logie soil (Table 3). At greater depth, the ratios are less than half those of the surface horizons, indicating the more decomposed nature of the organic C in the subsoils. After some decades of cultivation, however, the ratios of the surface 10 cm have decreased to 0.8 with less change or no change apparent in the deeper horizons. Dalal and Mayer (1986b) demonstrated that the rate of loss of carbon from these profiles decreased with increasing depth, although as indicated in Table 1, even in the 20-30 cm layers significant losses can be measured. The larger changes in the O-alkyl to alkyl ratios in the surface horizons therefore probably reflect the much greater losses of C in these surface horizons through microbial activity, which decreases with depth.

Table 3. Changes in the O-alkyl C/alkyl C ratios of the two soils with increasing depth in the profiles, before and after cultivation
Depth Virgin site Cultivated site
(cm) Waco Langlands- Waco Langlands-
 Logie (50 years) Logie
 (45 years)

0-10 2.0 1.1 0.8 0.8
10-20 1.8 1.2 1.3 1.2
20-30 1.7 1.3 1.4 1.3
90-120 0.7 0.5 n.a. n.a.

n.a., not available.

The NMR data are consistent with the loss of C from the profiles through microbial oxidation and provide no evidence for C movement down the profiles. Surface erosion cannot be ruled out on the presented evidence but it would seem unlikely that this was substantial based on the evidence from the [sup.13]C NMR measurements.

Fractionation of the soils

The surface (0-10 cm) samples were first physically fractionated to separate the POC from the [is less than] 53 [micro]m C fraction. The latter fraction was further fractionated using high energy UV photo-oxidation to separate the protected C fraction (Skjemstad et al. 1996). The protected fraction was then analysed for aryl C content using solid-state [sup.13]C CP/MAS NMR spectroscopy following HF treatment and the char C content calculated. This fractionation scheme resulted in 4 soil C fractions ranging from a relatively biologically labile fraction (POC) to a highly recalcitrant fraction (char C). Two intermediate fractions were also measured consisting of the [is less than] 53 [micro]m C removed by photo-oxidation (termed here the humic C fraction) and the [is less than] 53 [micro]m C remaining after photo-oxidation which was not aryl and therefore not char C (termed here the physically protected C).

Changes in these fractions for the surface horizons with increasing time under cultivation are given in Fig. 5. In the Waco soil (Fig. 5a), all of the pools, with the exception of the char C pool, decreased with increased time of cultivation. The POC pool begins at about 3.5 g C/kg soil and reaches an equilibrium value of about 1.5 g C/kg soil by 19 years. Similarly, both the humic and physically protected carbon (PPC) pools reach equilibrium values of about 3.5 and 1.0 g C/kg soil, respectively, also by 19 years. The char C pool, however, is maintained at about 6.0 g C/kg soil throughout the 50 years of cultivation. It must be remembered that these samples were not taken from the same site at different times but rather were collected from different sites with the same soil type but which had been cultivated for different periods of time. Some variation in the different pools might therefore be expected, reflecting natural variation in sites rather than changes imposed by cultivation. Variations in the char C content therefore probably reflect site variation. Under the conditions imposed by the particular cultivation system used here (input, fallow period, etc.), the dominant soil C pool in the Waco soil, after about 20 years of cultivation, is char C. Despite the relatively high clay content of this soil type (70-80%, Table 1), the PPC pool is relatively small, although it is likely that clay content will also retard the decomposition rate of the humic pool.


The Langlands-Logie virgin site contains substantially more organic C and this is reflected in substantially larger POC and humic pools. The PPC pool is similar but the char C pool is substantially lower (3 v. 6 g C/kg soil in the Waco soil). Again, cultivation leads to substantial decreases in the all but the char C pools. It is of interest that both the POC and PPC attain equilibrium values at the same levels as the Waco soil, 1.5-1.0 g C/kg soil, by the first sampling time of 20 years. The humic pool, however, continues to decline over the entire 45-year period, probably reaching an equilibrium value [is less than] 2 g C/kg soil, which is about half that at the Waco sites. This reinforces the view that the dynamics of the humic pool is moderated by the presence of clay, since the Langlands-Logie soil contains less clay (50-53% v. 69-80%, Table 1). Hassink (1994, 1996), for example, demonstrated that a soil's capacity to protect organic C was directly related to the content of clay and silt. The data also suggest that the char C content decreases with time from 3.0 to 2.0 g C/kg soil over the 45-year study period. This may not be significant, however, reflecting variations between sampling sites and/or the difficulty in measuring small quantities of char against a high background of total organic C.

It is of interest that the PPC pool is small, even in these clay soils, and unlike the humic pool, does not appear to be influenced by clay content. The photo-oxidation technique is highly efficient in oxidising organic matter that is either not protected within an inorganic matrix or is not char. These soils were pretreated with excess [SO.sub.2] solution to remove carbonate and then treated with sodium hexametaphosphate to sodium-saturate the clays and to complex any remaining divalent cations. This treatment effectively reduces all soil aggregates to [is less than] 53 [micro]m, but in these smectitic soils, may effectively disperse much of the clay rendering the organic C less protected than it would be in the untreated soil. Experience with other strongly structured soil types, such as Oxisols and Andosols (Golchin et al. 1997), show that the organic C within aggregates [is less than] 53 [micro]m from these soils can be very difficult to remove by photo-oxidation even when char C content is low. Skjemstad et al. (1993) also showed that flocculating dispersed suspensions with [Al.sup.3+] substantially reduced the ability of the photo-oxidation technique to remove soil organic C from suspensions. It must be concluded, therefore, that the PPC fraction in the 2 soils from the current study must be highly protected within microaggregate structures that are highly stable to dispersion.

Considering the high stability of the PPC fraction to photo-oxidation, it is of interest that this fraction is decreased by about a factor of 2 during the initial 20 years or so of cultivation. Even increasing the time of photo-oxidation treatment from 2 to 4 h did not significantly reduce the contribution of this fraction. Figure 6 shows a comparison of solid-state [sup.13]C CP/MAS NMR spectra between the [is less than] 53 [micro]m fractions from the surface soils uncultivated (2 and 4 h) and from the soils that had been under the longest periods of cultivation (2 h). In the uncultivated Waco soil (Fig. 6a), the spectrum of the sample after 2 h of photo-oxidation shows a significant contribution from alkyl, O-alkyl, and carbonyl C as well as from the expected aryl peak at 130 ppm. Increasing the photo-oxidation time to 4 h resulted in a reduction in the alkyl and O-alkyl C peaks but these still remained significant. After 19 years and 50 years of cultivation, the spectra become dominated by the aryl C peak, indicating that only char C survived the treatment. The alkyl and O-alkyl C that survived the treatment in the soil from the uncultivated site must represent physically protected C that did not survive 19 years of cropping. This observation suggests that mechanisms of physical protection that are apparently effective under virgin conditions are not effective under exploitive cultural practices. In the Waco soil, therefore, it must be concluded that the most effective means of long-term protection of organic C against decomposition during cropping is through the chemical, char mechanism.


Similar results can be seen in the Langlands-Logie soil type (Fig. 6b). This soil contains much less char C (Fig. 5) as reflected in the relatively small contribution from the aryl peak. Increasing the photo-oxidation time did not change the contribution from the various functional groups but a period of cultivation did. Unlike the Waco soil, however, contributions from functional groups other than aryl groups are evident even after 45 years of cultivation. Although the PPC fraction is relatively small (Fig. 5), physical protection appears to be equal to char in importance as a means of protection against decomposition.

To determine whether the aromatic material was in fact char, the material remaining after photo-oxidation and HF treatment for the virgin sites and those under the longest periods of cultivation from each site was examined by SEM. The micrographs from the Waco soils (Fig. 7a, b), show highly angular pieces of material in the size range [is less than] 1 to about 40 [micro]m. The larger pieces show a morphology consistent with fragments of plant material but with a highly aromatic chemistry (Fig. 6) and must therefore be pieces of char. EDX analyses confirmed that these structures were not mineral in origin. The Langlands-Logie soil contained much less char material and even after prolonged cultivation and photooxidation, the organic material remaining contained [is less than] 50% char C. This is also confirmed by the micrograph in Fig. 7c, which shows that in the virgin soil, the majority of the organic matter forms a gelatinous, amorphous-like background imbedded with a small number of char-like pieces with angular but plant-like morphology. As expected after 45 years of cultivation, as the relative contribution from char increases, the angular, plant-like pieces become more obvious and appear to increase in frequency.


Another indication that the material remaining after photo-oxidation is significantly different in chemistry and therefore potentially also different in biological activity is demonstrated by the C/N ratios of this material. Table 4 shows that photo-oxidation results in a substantial increase in the C/N ratio of these samples from about 14 to 40 for the Waco soil and about 10 to 25 in the Langlands-Logie soil. The increase in C/N ratio with increasing time of cultivation in the Waco soil therefore reflects the relative increase in the non-photo-oxidisable char fraction. In the Langlands-Logie soil, this relative increase is small and the change in C/N ratio of the soil is affected more by the large loss in POC fraction, which will also have a C/N ratio higher than the whole soil. Increasing the time of photo-oxidation increased the C/N ratio of the residues in both soils. The residue from the Waco soil (Dalal and Mayer 1986e), after some years of cultivation, consisted almost entirely of char material (Fig. 6) and it can therefore be assumed that the C/N ratio of soil char material is reflected by the C/N ratio of the residues. The char in the Waco soil therefore has a C/N ratio near 40. It is not possible to determine the C/N ratio of the char in the Langlands-Logie soil because only about half of the resistant material is char (Fig. 5). Assuming, however, that the non-char material has a C/N ratio near 10, similar to the whole soil (Dalal and Meyer 1986e), then a C/N ratio near 40 for the char material would result in the C/N ratio near 25 measured (Table 4).

Table 4. The C/N ratios and [[Delta].sup.13]C values of samples from the two soils before and after photo-oxidation
Soil type Years(A) Treat- C N C/N [[Delta].
 ment(B) sup.13]C %

Waco 0 WS 18.2 1.34 13.6 -14.6
 19 WS 12.5 0.89 14.0 -14.6
 35 WS 12.1 0.83 14.6 -15.1
 50 WS 11.4 0.74 15.4 -14.4
 0 2 h UV 7.2 0.24 30.4 -14.4
 19 2 h UV 7.3 0.18 40.4 -14.1
 35 2 h UV 6.9 0.20 34.8 -14.6
 50 2 h UV 6.6 0.16 41.8 -14.4
 0 4 h UV 5.5 0.14 38.4 -14.2
Langlands- 0 WS 27.0 2.41 11.2 -22.4
 Logie 20 WS 11.9 1.12 10.6 -22.1
 35 WS 9.4 0.89 10.6 -22.4
 45 WS 7.5 0.77 9.7 -22.4
 0 2 h UV 5.6 0.25 22.0 -22.9
 20 2 h UV 4.1 0.16 26.1 -22.8
 35 2 h UV 3.2 0.13 24.3 -22.9
 45 2 h UV 3.3 0.17 19.3 -23.0
 0 4 h UV 4.6 0.16 28.4 -23.2

(A) years under cultivation.

(B) WS, whole soil; UV, high energy UV photo-oxidation.

The [[Delta].sup.13]C values of the samples were also measured (Table 4). The native vegetation of the Waco soil was largely [C.sub.4] pasture (Dalal and Mayer 1986a) with a [[Delta].sup.13]C value near -12%, while the crop, largely C3 wheat, has a [[Delta].sup.13]C value near -26% (Park and Epstein 1960; LeFroy et al. 1993). The [[Delta].sup.13]C values consistently near -14.5% for the Waco soil therefore reflect the abundance of the original vegetation and indicate that inputs from crops to the soil C pools have been very small, even over a period of 50 years. The char fraction also reflects the native vegetation and even residue burning does not seem to have affected the char signal significantly. The Langlands-Logie soil shows a [[Delta].sup.13]C signal closer to [C.sup.3] vegetation with no change with increasing time of cultivation. Again, the protected material is similar to the soil and demonstrates that the char in this fraction was derived from a [C.sup.3] source.

Simulation of carbon decline using the RothC model

Since char C appears to represent a highly protected form of soil organic C, it should be a useful fraction for initialising soil C turnover models. In particular, the RothC model (Jenkinson 1990) includes an inert pool (IOM) which remains unchanged for millennia. This model was therefore tested using the data generated from this study. The carbon contribution from 4 pools other than the inert pool, (a) easily decomposable plant material (DPM, 10.0/ year), (b) resistant plant material (RPM, 0.3/year), (c) microbial biomass (BIO, 0.66/year), and (d) humified organic matter (HUM, 0.02/year), are also required. For this simple experiment, initial BIO was taken as a small fraction of the HUM (5%). To initialise the model, the POC fraction was used to represent the RPM material, the IOM was taken as the char C, and the HUM fraction was taken as 95% of the difference between the TOC and the POC plus char C fractions, the remaining 5% being BIO. Since the PPC fraction appeared to change during cultivation at about the same rate as the HUM fraction (Fig. 5a), these 2 pools were combined. Total soil C and the C pools were calculated as t C/ha to 23 cm, assuming an average bulk density of 1.1 for Waco and 1.3 for Langlands-Logie. Climatic data, as long-term monthly averages, were obtained from local weather data. C input data were taken from average crop yields (5 t dry matter per ha) assuming continuous wheat rotations with 1 year fallow in 5 years as summarised by Dalal and Mayer (1986a).

Simulations modelling the C changes using these assumptions were run for both sites. Fig. 8a, b shows that the model, as parameterised with the set values, coped reasonably well with the observed decreases in TOC. In particular, the model simulated the TOC and the measured pool structure for the Waco soil with the measured POC pool being higher than the simulated RPM pool and the measured HUM pool being slightly lower than the corresponding simulated pool (Fig. 8a). For the Langlands-Logie soil, the POC pool is similarly larger than the simulated RPM pool but the measured and simulated HUM pools are very different. It must to be remembered, however, that the model pools are conceptual in nature and were not based on any particular set of chemically or physically identifiable soil fractions. As such, it should not be surprising to find that the measured and modelled pools are not identical. The data do indicate, however, that an inert pool of about the magnitude of the measured char C pools is required. In fact, the measured and modelled pool structures for the Langlands-Logie soil would be in better agreement if there was no inert/char pool included. This adds weight to the argument that the PPC fraction should not be included with the IOM pool.


To determine if the relationship between measured and modelled data could be improved, the decomposition rate constant of the RPM pool was reduced to slow down the decomposition of this pool. For both sites, the modelled RPM pool was much lower than the measured POC fraction. By changing the decomposition rate constant to 0.1/year, the simulated and measured pools more closely coincided (Fig. 8c, d). The result of slowing the decomposition rate of this pool was to increase the TOC by the increased amounts of RPM remaining in the soil. There was also a very small decrease in the HUM pool due to less cycling of C from the RPM to the HUM pool.

The impact of changing the rate constant for the HUM pool was also tested by increasing its rate constant from 0.02/year to 0.03/year. The measured TOC, HUM, and POC pools and their corresponding modelled pools now agreed well (Fig. 8e, f) considering that the climate data, residue inputs, and rotations have been averaged over the period of 45-50 years. Although it is difficult to draw too many conclusions from just 2 sites under similar climates and with similar clay contents and mineralogy, it would appear that the 3 soil fractions separated by the reported technique provide pools that are of the same order of magnitude and similar rate constants as the 3 major pools in the RothC model. Since soil organic matter in reality consists of a continuum of materials decomposing at a wide range of rates, the definition of individual pools must be somewhat arbitrary and it is unlikely that any fractionation scheme will exactly mimic the pools required for any relatively complex soil carbon turnover model.

The POC fraction, for example, consists of plant residues in various stages of decomposition. Some of these residues may well be physically protected by the soil matrix in such a way that they behave more like the HUM pool, even though some plant-like characteristics still remain. Camberdella and Elliott (1992) also reported that a portion of the POC (POM) fraction appeared to persist for some time. From the Langlands-Logie soil, however, it is clear that the RPM pool must be of comparable size to the POC fraction in order to allow for the observed rapid initial decline in TOC to occur. The HUM pool of the model and that measured by the fractionation scheme also appear to be comparable in size and rate constant. The more rapid rate constant implied by the data may be real and may imply that the smectitic clays of these soils cannot protect the HUM pool as effectively as other clay types. This would imply that in an undisturbed state, some form of organic carbon/clay interaction protects the HUM pool but that this association is relatively weak, and once disturbed by cultivation, decomposition of the HUM pool is more rapid than expected. Using natural abundance [sup.13]C, Skjemstad et al. (1994b) were able to demonstrate a similar process in a Chromustert from Narayen in south-eastern Australia. In this study, a significant pool of organic C which had been stable for at least 40 years declined rapidly once the soil was subjected to cultivation and cropping. Again, the soil contained a significant amount of smectitic clay.

The organic matter in these soil types appeared to be largely proteinaceous in nature (Skjemstad and Dalal 1987). [sup.13]C NMR spectra of the HUM pool obtained by subtraction of the spectra from the IOM and POC pools from that of the whole soil (spectra not shown) are consistent with proteinaceous material along with some polysaccharide material. Similarly, correcting the C and N content of the whole soil with that from the POC and IOM pools gave C/N ratios for the HUM pools from all of the surface soils near 7.5. This demonstrates the biologically labile nature of the HUM pool once available to the biosphere.

Despite reasonable agreement between the model pool structure and the measured fractions, before any soil fractionation scheme could be adopted to initialise a particular model, model simulations would need to be compared with a large number of field situations where climate and plant residue inputs were well known.


The nature of the organic C in the 2 soil types was shown by solid-state [sup.13]C NMR to be very different. The organic matter in the Waco soil was dominated by aryl C whereas that of the Langlands-Logie soil was high in O-alkyl and alkyl C. Following cultivation, the aryl C content of the Waco soil remained constant with other functional groups declining somewhat equally. In contrast, the Langlands-Logie soil showed little change in functionality with time of cultivation.

The aryl C was shown to be present largely in the form of char C, which appeared to be highly resistant to decomposition following cultivation compared with other forms of organic C. The highly physically protected C fraction was small in both soils but appeared to be subject to decline in the early period of cultivation. The humic C pool in both soils declined rapidly after cultivation. Simulations of soil organic C decomposition using the pools defined by the fractionation scheme with the RothC turnover model indicated that the humic pool was more labile than predicted by the model. This increased lability appears to be a function of the highly proteinaceous nature of this pool, which also appears to be less protected by organic C/clay interactions than expected from the high clay content of these soils.


Baldock JA, Oades JM, Waters AG, Peng X, Vassallo AM, Wilson MA (1992) Aspects of the chemical structure of soil organic materials as revealed by solid-state [sup.13]C NMR spectroscopy. Biogeochemistry 16, 1-42.

Camberdella CA, Elliott ET (1992) Particulate soil organic matter changes across a grassland cultivation sequence. Soil Science Society of America Journal 56, 777-783.

Dalal RC, Mayer RJ (1986a) Long-term trends in fertility of soils under continuous cultivation and cereal cropping in southern Queensland. I Overall changes in soil properties and trends in winter cereal yields. Australian Journal of Agricultural Research 24, 265-279.

Dalal RC, Mayer RJ (1986b) Long-term trends in fertility of soils under continuous cultivation and cereal cropping in southern Queensland. II Total organic carbon and its rate of loss from the soil profile. Australian Journal of Agricultural Research 24, 281-292.

Dalal RC, Mayer RJ (1986c) Long-term trends in fertility of soils under continuous cultivation and cereal cropping in southern Queensland. III Distribution and kinetics of soil organic carbon in particle-size fractions. Australian Journal of Agricultural Research 24, 293-300.

Dalal RC, Mayer RJ (1986d) Long-term trends in fertility of soils under continuous cultivation and cereal cropping in southern Queensland. IV Loss of organic carbon from different density functions. Australian Journal of Agricultural Research 24, 301-309.

Dalal RC, Mayer RJ (1986e) Long-term trends in fertility of soils under continuous cultivation and cereal cropping in southern Queensland. V Rate of loss of total nitrogen from the soil profile and changes in carbon:nitrogen ratios. Australian Journal of Agricultural Research 25, 83-93.

Dick WA (1983) Organic carbon, nitrogen and phosphorus concentrations and pH in soil profiles as affected by tillage intensity. Soil Science Society of America Journal 47, 102-107.

Dormaar JF (1983) Chemical properties of soil aggregates after sixty-seven years of cropping to spring wheat. Plant and Soil 75, 51-61.

Golchin A, Clarke P, Baldock JA, Higashi T, Skjemstad JO, Oades JM (1997) The effects of vegetation and burning on the chemical composition of soil organic matter in a volcanic ash soil as shown by [sup.13]C NMR spectroscopy. I. Whole soil and humic acid fraction. Geoderma 76, 155-174.

Hassink J (1994) Effects of soil texture and grassland management on soil organic C and N and rates of C and N mineralisation. Soil Biology and Biochemistry 26, 1221-1231.

Hassink J (1996) Preservation of plant residues in soils differing in unsaturated protective capacity. Soil Science Society of America Journal 60, 487-491.

Heanes DL (1984) Determination of total organic-C in soils by an improved chromic acid digestion and spectroscopic procedure. Communications in Soil Science and Plant Analysis 15, 1191-1213.

Jenkinson DS (1990) The turnover of organic carbon and nitrogen in soil. Philosophical Transactions of the Royal Society of London B 329, 361-368.

LeFroy RDB, Blair G J, Strong WM (1993) Changes in soil organic matter with cropping as measured by organic carbon fractions and [sup.13]C natural isotope abundance. In `Plant nutrition from genetic engineering to field practice'. (Ed. NJ Barrow) pp. 551-554. (Kluwer Academic Publishers: Dordrecht, Netherlands)

Loveday J, Reeve R (1974) Carbonate determination. In `Methods of analysis of irrigated soils'. (Ed. J Loveday) pp. 131-134. Technical Communication of the Commonwealth Agricultural Bureaux. No. 54.

Merry RH, Spouncer LR (1988) The measurement of carbon in soils using a microprocessor-controlled resistance furnace. Communications in Soil Science and Plant Analysis 19, 707-720.

Park R, Epstein S (1960) Carbon isotope fractionation during photosynthesis. Geochimica et Cosmochimica Acta 21, 110-126.

Russell JS, Williams CH (1982) Biogeochemical interactions of carbon, nitrogen, sulfur and phosphorus in Australian agroecosystems. In `The cycling of carbon, nitrogen, sulfur and phosphorus in terrestrial ecosystems'. (Eds IE Galbally, JR. Freeny) pp. 61-75. (Australian Academy of Science: Canberra)

Russell JS, Jones PN (1996) Continuous, alternate and double crop systems on a Vertisol in subtropical Australia. Australian Journal of Experimental agriculture 36, 823-830.

Skjemstad JO, Clarke P, Taylor JA, Oades JM., McClure SG (1996) The chemistry and nature of protected carbon in soil. Australian Journal of Soil Research 34, 251-271.

Skjemstad JO, Catchpoole VR, Le Feuvre RP (1994b) Carbon dynamics in Vertisols under several crops as assessed by natural abundance [sup.13]C. Australian Journal of Soil Research 32, 311-321.

Skjemstad JO, Clarke P, Taylor JA, Oades JM, Newman RH (1994a) The removal of magnetic materials from surface soils. A solid state [sup.13]C CP/MAS n.m.r, study. Australian Journal of Soil Research 32, 1215-1229.

Skjemstad JO, Dalal RC (1987) Spectroscopic and chemical differences in organic matter of two Vertisols subjected to long periods of cultivation. Australian Journal of Soil Research 25, 323-335.

Skjemstad JO, Dalal RC, Barren PF (1986) Spectroscopic investigations of cultivation effects on organic matter of Vertisols. Soil Science Society of America Journal 50, 354-359.

Skjemstad JO, Janik LJ, Head MJ, McClure SG (1993) High energy ultraviolet photo-oxidation: a novel technique for studying physically protected organic matter in clay- and silt-sized aggregates. Journal of Soil Science 44, 485-499.

Skjemstad JO, Taylor JA (1999) Does the Walkley and Black method determine soil charcoal? Communications in Soil Science and Plant Analysis 30, 2299-2310.

Skjemstad JO, Taylor JA, Janik L J, Marvanek SP (1999) Soil organic carbon dynamics under long-term sugarcane monoculture. Australian Journal of Soil Research 37, 151-164.

Skjemstad JO, Taylor JA, Smernik RJ (1999) Estimation of charcoal (char) in soils. Communications in Soil Science and Plant Analysis 30, 2283-2298.

Soil Survey Staff (1989) `Keys to soil taxonomy.' USDA-USAID-SMSS Tech. Monograph 6. (Cornell University: Ithaca, NY)

Manuscript received 20 December 1999, accepted 16 June 2000

J. O. Skjemstad(A), R. C. Dalal(B), L. J. Janik(A), and J.. A. McGowan(A)

A CSIRO Land and Water, PMB 2, Glen Osmond, SA 5064 Australia. B Department of Natural Resources, RSK, Natural Sciences Precinct, 80 Meiers Road, Indooroopilly, Qld 4068, Australia.
COPYRIGHT 2001 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Skjemstad, J. O.; Dalal, R. C.; Janik, L. J.; McGowan, J. A.
Publication:Australian Journal of Soil Research
Geographic Code:8AUQU
Date:Mar 1, 2001
Previous Article:Nitrate accumulation under cropping in the Ferrosols of Far North Queensland wet tropics.
Next Article:Carbon and nitrogen mineralisation in sand, silt, and clay fractions of soils under maize and pasture.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters