Tillage-induced changes to soil structure and organic carbon fractions in New Zealand soils.
The form and stability of soil structure are important for soil aeration, storage and transmission of water and nutrients, and root penetration and development. The interrelationships between soil structure and soil organic matter (SOM) are dynamic where the decomposition products of SOM affect soil aggregation and aggregate stability, while soil structure influences biological activity and therefore organic carbon decomposition (Golchin et al. 1997).
It is widely reported that when grazed pasture soils are brought into cultivation, soil organic carbon levels and the stability of soil aggregates are reduced (Low 1972; Golchin et al. 1995a; Haynes and Beare 1995). Conversely, the size of soil aggregates increases under intensive cultivation as a result of increased clodding (Voorhees and Lindstrom 1984; Hakansson et al. 1988). Aggregate stability and aggregate size are state indicators of soil structure, and as such, their mean weight percent and mean weight diameters are useful to characterise the effects of increasing duration of conventional cultivation on the quality of the soil tilth. An earlier study indicates that a decline in SOM and deterioration in soil structure following intensive cropping in the southern North Island, New Zealand, gave rise to a number of deleterious effects that greatly influenced the sustainability of grain production systems (Shepherd 1992).
Labile fractions of soil C such as carbohydrates (polysaccharides) were reported to be involved in the aggregation of soil particles by their binding and cementing actions (Tisdall and Oades 1982; Oades 1984; Haynes and Beare 1995). Golchin et al. (1997) suggested that microbial decomposition of carbohydrates associated with aggregates and the resultant production of binding agents of mucilages and metabolites would promote the stability of soil aggregates. With the decline of carbohydrate levels as a result of continued microbial decomposition, the production of mucilage and metabolite glues would reduce, and the stability of aggregates decrease. Soil aggregates are also stabilised by the physical enmeshing of soil particles and microaggregates by mycorrhizal fungal hyphae, fine roots, and root-derived particulate organic carbon (Tisdall and Oades 1982; Golchin et al. 1997). While water-extractable and acid-extractable carbohydrate C have been more strongly correlated with the stability of macro-aggregates than total organic carbon (TOC) or total extractable carbohydrate C (Angers and Mehuys 1989; Haynes et al. 1991), other investigators have found a poor correlation between water-extractable and acid-extractable carbohydrate C and aggregate stability (Angers et al. 1993; Carter et al. 1994). Golchin et al. (1995a, 1995b) suggested that only a part of soil carbon or carbohydrate may be involved in aggregate stability. They noted that occluded fractions of carbon and O-alkyl C were highly correlated with aggregate stability, while total organic C and total O-alkyl C were not closely correlated. Elliott (1986) hypothesized that the loss of organic carbon caused by cultivation is considered chiefly a loss of organic material that binds individual micro-aggregates, and not a loss of organic carbon within micro-aggregates. Soil microbial biomass has also been found to be significantly related to macro-aggregate stability (Sparling et al. 1992) in clay and loam soils. However, neither microbial biomass nor water-extractable carbohydrate C appears to be related to aggregate stability in sandy soils (Degens et al. 1994).
The use of [sup.13]C nuclear magnetic resonance (NMR) spectroscopy in recent studies of SOM has also revealed the importance of the aliphatic fraction in influencing aggregate stability (Capriel et al. 1990; Dinel et al. 1990). Lipids containing hydrophobic aliphatic C-H units were suggested to form a water-repellent lattice around soil aggregates, thus increasing their water stability. The strong bonding effects of humic polymers have been also proposed (Haynes and Beare 1995) to contribute to the stability of soil aggregates. Our knowledge concerning the effect of cultivation on the chemical composition of SOM and the role played by the aliphatic fraction, in particular, in influencing soil structure and soil aggregate stability is still very limited.
Scanning electron microscopy (SEM) provides a means of observing intact soil aggregate microstructures at a level that has been termed sub-microscopic, where microaggregation mechanisms can be seen (Smart and Tovey 1981; Bisdom 1983; Sullivan and Koppi 1987). The strong casual relationship between organic matter and soil microstructural development has been illustrated using SEM in a number of studies (Powers and Skidmore 1984; Metzger and Robert 1985; Sullivan and Koppi 1987; Monreal and Kodama 1997). SEM has also been used to show the correlation between soil microstructure and aggregate stability (Churchman and Tate 1986; Metzger and Yaron 1987).
This study examines the relationships between aggregate stability and dry aggregate-size, total organic carbon (TOC), hot-water-soluble carbohydrate (WSC) and acid-hydrolysable carbohydrate (AHC) C, the SOM functional groups measured by solid state [sup.13]C-NMR, and the scanning electron micrographs of 5 soils of contrasting mineralogy under pasture and cropping. We also wished to determine if the changes in the mildly extracted labile fraction were attributable to changes in the structural composition of soil organic C determined by [sup.13]C-NMR and SEM.
Materials and methods
Soils, location, and land use
A well-drained Manawatu silt loam, poorly drained Kairanga silty clay loam, and very poorly drained Moutoa humic clay, all formed from quartzo-feldspathic alluvium with a predominant mica-vermiculite mineralogy, were sampled under long-term pasture and under increasing duration of maize and/or barley cropping from the Manawatu region of New Zealand (Table 1). A well-drained allophane-rich soil (Egmont black loam) formed from andesitic tephra was sampled under long-term pasture and 20 years of barley in the Taranaki region. The Egmont soil was double-cropped each year with spring-sown barley and a winter green-feed brassica. Well-drained Patumahoe clay soils with a kandite/crystalline Fe and Al oxide and short-range order Si, Al, and Fe sesquioxide mineralogy were sampled in the Pukekohe district under long-term pasture and 40 years vegetable production. The Patumahoe soil is formed from strongly weathered basalt and rhyolitic tephra. Like the Egmont soil, the Patumahoe soil was double-cropped each year. The study areas have a temperate humid climate with mean annual temperatures of 12.9, 13.6 and 15.3 [degrees] C, and mean annual rainfalls of 936, 1062, and 1185 mm for the Manawatu, Taranaki, and Pukekohe sites, respectively (New Zealand Meteorological Service 1983).
Table 1. Soil types and their classification, land use, site symbols, particle-size distribution, and mineralogy Interstrat. mica-vermic, interstratified mica vermiculite; Vol. glass, volcanic glass; H.I.V, hydrated interlayered vermiculite Soil type and Land use Site soil classification (years) symbol Manawatu silt loam Pasture MP (Dystric Fluventic Eutrochrept(A)) Maize (4) M4M [Weathered Fluvial Recent Soil(B)] Maize (11) M11M Maize (20) M20M Maize(11)+ M11M10P pasture (10) Kairanga silty clay Pasture KP loam (Typic Endoaquept(A)) Maize (4) K4M [Typic Orthic Gley Soil(B)] Maize (11) K11M Barley (23) K23B Barley (30) K30B Maize(11)+ K11M10P pasture (10) Moutoa humic clay Pasture MoP (Typic Endoaquoll(A))[Acidic Maize (4) Mo4M Recent Gley Soil(B)] Maize (11) Mo11M Maize (11) + Mo11M11P pasture (11) Patumahoe clay Pasture PP (Typic Haplohumult(A)) Cropping (40) P40C [Allophanic Oxidic Granular Soil(B)] Egmont black loam Pasture EP (Typic Hapludand(A)) [Typic Orthic Barley (20) E20B Allophanic Soil(B)] Soil type and Land use Particle-size soil classification (years) distribution(C) % sand % silt % clay Manawatu silt loam Pasture 17 61 22 (Dystric Fluventic Eutrochrept(A)) Maize (4) 16 64 20 [Weathered Fluvial Recent Soil(B)] Maize (11) 6 72 22 Maize (20) 7 71 22 Maize(11)+ 4 72 24 pasture (10) Kairanga silty clay Pasture 10 53 37 loam (Typic Endoaquept(A)) Maize (4) 9 49 42 [Typic Orthic Gley Soil(B)] Maize (11) 5 54 41 Barley (23) 12 49 39 Barley (30) 4 58 38 Maize(11)+ 9 52 39 pasture (10) Moutoa humic clay Pasture 4 37 59 (Typic Endoaquoll(A))[Acidic Maize (4) 8 32 60 Recent Gley Soil(B)] Maize (11) 1 44 55 Maize (11) + 14 19 67 pasture (11) Patumahoe clay Pasture 6 31 63 (Typic Haplohumult(A)) Cropping (40) 7 26 67 [Allophanic Oxidic Granular Soil(B)] Egmont black loam Pasture 32 39 29 (Typic Hapludand(A)) [Typic Orthic Barley (20) 33 41 26 Allophanic Soil(B)] Soil type and Land use Minerals in clay soil classification (years) fraction Manawatu silt loam Pasture Mica:33-48% (Dystric Fluventic Vermic.: 10-20% Eutrochrept(A)) Maize (4) Interstrat. mica- [Weathered Fluvial vermic: 3-4% Recent Soil(B)] Maize (11) Chlorite: 6-7% Smectite: 0-3% Maize (20) Kandite: 7-8% Feldspar: 11-15% Maize(11)+ Quartz: 12-13% pasture (10) Kairanga silty clay Pasture Mica:38-52% loam Vermic.: 10-15% (Typic Endoaquept(A)) Maize (4) Interstrat. mica- [Typic Orthic Gley vermic: 0-3% Soil(B)] Maize (11) Chlorite: 2-3% Smectite: 0-10% Barley (23) Kandite: 5-8% Feldspar: 14-17% Barley (30) Quartz: 12-20% Maize(11)+ pasture (10) Moutoa humic clay Pasture Mica: 25-42% (Typic Vermic: 8-14% Endoaquoll(A))[Acidic Maize (4) Interstrat. mica- Recent Gley Soil(B)] vermic: 0-7% Maize (11) Chlorite: 4-5% Kandite: 8-24% Maize (11) + Feldspar: 13-18% pasture (11) Quartz: 15-23% Patumahoe clay Pasture Kandite: 31-45% (Typic H.I.V: 16-34% Haplohumult(A)) Cropping (40) Mica-vermic: 0-10% [Allophanic Oxidic Gibbsite: 9% Granular Soil(B)] Geothite: 8-9% Allophane: 1-2% Ferrihydrite: 1% Quartz: 8 14% Cristobalite:7-12% Egmont black loam Pasture Vol. glass: 36-43% (Typic Hapludand(A)) Allophane:24-28% [Typic Orthic Barley (20) Ferrihydrite: 3% Allophanic Soil(B)] Vermic: 11-20% Kandite: 9-10% Chlorite: 2-5% Feldspar: 4-7% Quartz: 2%
All sites were located on commercially operated rain-fed farms. The long-term pasture sites had been under pasture for more than 80 years. All cropping sites were conventionally cultivated using a mould-board plough followed by levelling and rolling, discing, and power harrowing. Replicated samples were taken from each of the paddock treatments in early spring from compacted wheel-traffic areas after harvest and before cultivation. The experiments were pseudo-replicated with respect to the treatments only. The number of replicates varied according to the measurement, e.g. 9 replicates for bulk density, 4 for dry aggregate-size distribution and aggregate stability. For TOC, carbohydrate, [sup.13]C NMR, SEM, and particle-size analysis, 3 replicates were pooled. The bulk density data were subjected to an analysis of variance to determine the statistical significance of treatment effects using SYSTAT for Windows software package and the general linear model procedure (Wilkinson 1996). The treatment means were used to study the correlations among soil parameters.
Aggregate stability, aggregate-size distribution, bulk density, and particle-size analysis
The aggregate stability of air-dried 2-4 mm size soil aggregates was determined on 4 replicate samples from 0-200 mm depth by the method of Gradwell and Birrell (1979) based on Yoder (1936). The 2-4-mmdiameter aggregates were wet-sieved on a stack of sieves with openings 2.0, 1.0, and 0.5 mm, and the aggregates remaining on each sieve oven dried and weighed. Soil aggregates were then dispersed in water and passed through the sieves again. The sand fraction [is greater than] 0.5 mm retained on the sieves was dried and weighed, and the amount subtracted from the aggregate weight. The amount subtracted was negligible because all soils contained [is less than or equal to] 3% sand [is greater than] 0.6 mm. The results are expressed as a mean weight diameter (MWD) (using the formula below), where the upper and lower limits are 3.00 and 0.25 mm, respectively:
MWD = [Sigma] (Wt% sample on sieve x mean inter-sieve size/100)
Dry aggregate-size distribution of the topsoil was measured using a drop-shatter procedure. Four large replicate cores (200 mm diam. by 200 mm deep) were taken with a hydraulic ram, extruded, and dropped (up to a maximum of 3 times) from a height of 1.0 m onto a firm surface to break the soil down into its primary structural units. If large clods broke away after the first or second drop, they were individually dropped once or twice again. If a clod shattered into primary structural units after the first or second drop, it was not necessary to drop it again. No part of the soil sample, or the clods breaking away from the sample, was dropped more than 3 times. The soil was dry sieved through a stack of sieves with mesh openings of 2, 5, 10, 20, 50, 100, and 150 mm. The weight percent of soil retained on each sieve was calculated, and the overall aggregate MWD calculated using the above formula. All soils were sampled and `dry' sieved in the field when soil-water matrix potential was approximately -100 kPa.
Soil bulk density was determined on 9 undisturbed (100-mm-diam.) cores sampled with a hydraulic ram at 0-100 mm and 100-200 mm depths. Particle-size distribution was determined by the pipette method of Claydon (1989).
Total organic C
A representative soil sample comprising 10 cores from each of 3 replicates was sampled with a 25-mm-diameter tube corer at 0-100 mm and 100-200 mm depths over a 50 by 50 m area. Samples were bulked and subsampled for TOC, carbohydrate, and [sup.13]C NMR analysis. A subsample was passed through a 2-mm sieve, air-dried at 28 [degrees] C, and ground in a Tema mill to 0.20 mm for total carbon analysis using a Leco induction furnace (Blakemore et al. 1987). All soils sampled in this study had a pH [is less than] 6.0 and no Ca[CO.sub.3] was present; therefore, total carbon values represented TOC. Total organic carbon is expressed as a sum of the 2 sample depths (in t/ha).
Carbohydrates were determined on representative soil samples from 0-10 cm and 10-20 cm depths in 2 types of soil extracts, i.e. hot water and dilute [H.sub.2][SO.sub.4]. Initially, both moist and dry soil samples were extracted with hot-water. It was observed that air-drying had a slight tendency to lower the amount of extractable carbohydrate-C. The amounts of carbohydrate-C of air-dry soils were linearly correlated (r = 0.958, n = 30, P [is less than] 0.01) with those of moist soils. Therefore, only the air-dry samples were used for [H.sub.2][SO.sub.4] extraction.
The equivalent weight of 2 g soil was extracted with 20 mL de-ionised water at 80 [degrees] C (Haynes and Swift 1990) or 20 mL of 1.5 M [H.sub.2][SO.sub.4] (Angers and Mehuys 1989) for 24 h with regular shaking, and filtered through glass fibre filter papers (Whatman GF/C). The carbohydrate content of the extracts was determined using the anthrone method (Brink et al. 1960). The carbohydrates extracted with hot water are designated as water-soluble carbohydrates (WSC), and those extracted with acid as acid-hydrolysable carbohydrates (AHC). Both WSC and AHC are expressed in t/ha to a depth of 20 cm by taking the sum of the 0-10 cm and 10-20 cm depths.
[sup.13]C-NMR analysis of SOM
The major functional groups of the organic carbon fraction in the Kairanga and Moutoa soils at 0-10 cm depth were characterised using solid state [sup.13]C-NMR on samples that had received a minimum amount of pre-treatment. Samples were prepared by removing all visible roots with tweezers. The soil was dispersed in reverse osmotic water at room temperature with the aid of a glass rod, and the suspension passed through a 63-[micro]m sieve to remove the sand fraction and any fine roots. The sand fraction was removed to reduce the paramagnetic interference from magnetite and other fine minerals present in the soil. The remaining clay and silt suspension was stirred with a magnetic `flea' to remove any further magnetite, and then air-dried at 26 [degrees] C. Negligible amounts of C may have been lost during this process. A 0.38-g portion of each sample was packed in a zirconia 7-mm-diameter cylindrical rotor, retained with Vespel end caps, and spun at 5 kHz in a magic-angle spinning probe for [sup.13]C-NMR experiments at 50.3 MHz in a Varian Inova-200 NMR spectrometer. Each 5-[micro]s proton preparation pulse was followed by a l-ms cross-polarisation contact time, 16 ms of data acquisition and a 400-ms recovery delay. Between 116 000 and 134 000 pulses were acquired over periods of 13.4-15.5 h for each sample.
The absolute contribution of the major functional groups of SOM was calculated from their relative contribution using the very high correlation between the area of the curve and TOC (where area = 39.9 + 33.0xTOC; r = 0.991, P [is less than] 0.001). Given the excellent correlation between the total NMR signal area and TOC, it was assumed that either all the C in the sample is quantitatively identified by the spectra, or the C that is not identified is similar in nature to the one that is identified. The 4 major functional groups accounted for 91-95% of the total C because of signals outside the chemical shift ranges reported. These signals ([is less than] 10 and [is greater than] 190 ppm) are assigned primarily to magic-angle-spinning sidebands and the outer edges of a very broad background signal, possibly associated with C in proximity to paramagnetic or ferromagnetic carbon.
Scanning electron microscopy
Field-moist, sieved 2-4-mm aggregates of topsoils were collected for micro-structural comparisons between pasture, long-term cropping, and after conversion back to pasture. The aggregates were freeze-dried, mounted onto aluminium stubs, sputter-coated with gold, and their surface microstructures examined in a Cambridge Stereoscan 250 Mark 3 scanning electron microscope.
All soils showed a significant decline in aggregate stability under cropping, with the allophanic Egmont soil showing the least change (Figs 1, 2). The MWD of water-stable aggregates in the mica-rich coarse-textured Manawatu and finer textured Kairanga soils declined by 64-71% from moderate levels under pasture to low to very low levels in just 4 years of cropping (Fig. 1). MWDs declined further with continued cropping, except under medium-term (11 years maize) cropping on Kairanga soils where it increased slightly. Under medium- to long-term cropping on both Manawatu and Kairanga soils, 85-90% of aggregates were [is less than] 0.5 mm in diameter. Aggregate stabilities were very high in the humic Moutoa soil, and while remaining high after short-term (4 years maize) cropping, declined to low levels after 11 years maize (Fig. 2). After 10 years of pasture following 11 years of maize on the Manawatu and Kairanga soils, the MWD of stable aggregates recovered to levels between short-term cropping and pasture, with the [is less than] 0.5 mm fraction accounting for 45% of aggregates, and the remaining size fractions occurring in approximately equal proportions. The MWD of the humic Moutoa soil recovered to high levels after 11 years of pasture following 11 years of maize, levels that were similar to those under short-term cropping where only 8% of the aggregates were [is less than] 0.5 mm. While the MWD of the oxide-rich Patumahoe soil declined from high to very low levels under 40 years of cropping, unlike the micaceous mineral soils, only 44% of the aggregates were [is less than] 0.5 mm. Like the humic Moutoa soils, the MWD of stable aggregates in allophanic Egmont soils under pasture is very high with 90% of the aggregates between 2 and 4 mm. Unlike the Moutoa soil, however, the MWD was still moderate to high after long-term (20 years) barley cropping, with 50% aggregates 2-4 mm and only 20% [is less than] 0.5 mm.
Dry aggregate-size distribution
The MWD of the dry aggregate-size distribution in mica-rich soils increased under cropping, with the largest increases occurring in the Kairanga soil, and the smallest increases occurring in the Manawatu soil under medium- and long-term cropping (Figs 3, 4). Aggregates [is less than] 10 mm accounted for 60-93% of the all of the 3 mica-rich soils under pasture, while clods [is greater than] 100 mm accounted for 60-83% of the Kairanga soil under medium- and long-term cropping. In the Kairanga soils, clod development was greater under maize than barley, with 83% of the soil comprising clods [is greater than] 150 mm after 11 years maize. The MWD and proportion of dry aggregates after 10 years of pasture following 11 years of maize returned to levels similar to those under short-term cropping in the Manawatu, Kairanga, and Moutoa soils (Figs 3, 4). The proportion of the finer aggregates [is less than] 10 mm recovered markedly, while the coarser aggregates decreased. By comparison, the MWD of the oxide-rich Patumahoe and particularly the allophanic Egmont soil increased only slightly under long-term cropping, with aggregates [is less than] 10 mm accounting for 42% and 66% of the soil, respectively (Fig. 4). Only the Egmont soil did not show a decrease in the [is less than] 2 mm aggregates under long-term cropping.
Changes in total organic C
In all soils, except the allophanic Egmont soil, TOC declined markedly with cropping (Fig. 5). The rate of decline varied with the length of cultivation and soil type. In the coarse-textured Manawatu soils, the rate of decline averaged 2.4 t C/ha.year during medium-term (11 years) maize cropping, and slowed to 0.8 t C/ha.year during 11 and 20 years of maize. In the finer textured Kairanga soil, TOC declined rapidly at an average rate of 3.3 t/ha.year during short-term (4 years of maize) cropping, and slowed to 0.9 t/ha. year during medium-term (11 years of maize) and long-term (23 and 30 years of barley) cropping. The initial C loss in the fine-textured humic Moutoa soil was negligible during short-term (4 years of maize) cropping, but increased to 1.1 t/ha.year between 4 and 11 years of maize. In fine-textured, oxide-rich Patumahoe soils, TOC declined at an average rate of 1.2 t/ha.year during long-term (40 years) cropping. By comparison, TOC declined at an average rate of only 0.20 t/ha.year in the allophanic-rich Egmont soil during 20 years of barley. On conversion back to (10-11 years) pasture, the recovery of TOC in medium-term cultivated soils was 1.1 t/ha.year in the Manawatu, 0.9 t/ha.year in the Moutoa, and only 0.05 t/ha.year in the Kairanga soil.
The WSC fraction (Fig. 6) ranged from 0.25 to 2.2 t C/ha and AHC from 5.3 to 14.7 t C/ha. The amounts of WSC and AHC were greater in the pasture soils than the cultivated soils, with the exception of Moutoa humic clay soil cultivated for 4 years. In all soils, the WSC and AHC fractions showed a consistent decrease with continuous cropping with the possible exception of Moutoa soil. The samples from the 3 sites that were restored to pasture (for 10-11 years) had values of WSC and AHC similar to the long-term pasture soils. The WSC and AHC fractions represented on average 1.4% and 9.4%, respectively, of total C. Both of these fractions were linearly correlated (r -- 0.663, P [is less than] 0.01; and 0.728, P [is less than] 0.001) with TOC (Table 2). The WSC fraction constituted 5-22% of the AHC.
Table 2. Simple correlations between biochemical and soil structural properties Soil property Total Hot-water soluble Acid- carbon carbohydrate hydrolysable (WSC) carbohydrate (AHC) Total carbon -- 0.663(**) 0.728(***) WSC -- 0.795(***) AHC -- Aggregate stability Soil property Aggregate Aggregate-size stability distribution Total carbon 0.814(***) 0.446 n.s. WSC 0.703(***) 0.369 n.s. AHC 0.801(***) 0.488(*) Aggregate stability -- 0.591(**) (*) P < 0.05; (**) P < 0.01; (***) P < 0.001; n.s., not significant.
Major functional groups of soil organic carbon
Solid state [sup.13]C-NMR spectra of the Kairanga soils (0-10 cm) were dominated by O-alkyl C at a chemical shift of 73 ppm and alkyl C at 30 ppm (Fig. 7). There were lesser amounts of acetal C (105 ppm) and carboxyl C (173 ppm). Aromatic C (130 ppm) and phenolic C (149 ppm) occurred in exceptionally low amounts. The amount of C representing each of the 4 main functional groups is given in Table 3. O-alkyl C fraction was the dominant form (47%) under permanent pasture, while 29% comprised alkyl C, 11% aromatic C, and 8% carboxyl C. After medium-term (11 years of maize) cropping, all the functional groups declined. The O-alkyl C declined by 45%, alkyl C by 20%, and both aromatic C and carboxyl C by 32% (Table 3). Under long-term (30 years of barley) cropping, all the functional groups were reduced to half the pasture levels (Table 3). However, the major decline in these functional groups occurred during the first 11 years of maize cropping. After 10 years of pasture following 11 years of maize, O-alkyl C recovered to 40% of the amount present under long-term pasture but alkyl C, aromatic C, and carboxyl C did not recover. From the [sup.13]C-NMR spectra (Fig. 7) and the absolute contribution of each functional group expressed as a percent of total C (Shepherd et al. 2001), alkyl C was seen to recover slightly after 10 years of pasture following 11 years of maize.
Table 3. Absolute contribution (t/ha) of major functional groups of SOM for the Kairanga and Moutoa soils (0-10 cm) as determined by solid state [sup.13]C-NMR and chemical analysis Land use Total [sup.13]C-NM R chemical organic shift range (ppm) C Alkyl C O-Alkyl C 10-48 48- 110 Kairanga silty clay loam KP 54 15.7 25.4 K11M 36.8 12.5 14 K30B 25.3 7.6 10.4 K11M10P 39.1 12.1 18.4 Moutoa humic clay MoP 86.2 33.6 33.6 Mo11M 78.1 29.7 28.9 Mo11M11P 97.2 40.8 33 Land use [sup.13]C-NM R chemical shift range (ppm) Chemical analysis Aromatic C Carboxyl Acid-hydrolysable 110- 160 C160- 190 carbohydrate Kairanga silty clay loam KP 5.9 4.3 4.3 K11M 4.1 3 3.6 K30B 3 2.5 2.8 K11M10P 3.5 2.7 4.5 Moutoa humic clay MoP 7.8 6.9 5.5 Mo11M 8.6 6.2 4.5 Mo11M11P 9.7 7.8 6
The SOM fraction of the humic Moutoa clay was also dominated by alkyl C and O-alkyl C, with moderate carboxyl C, and low acetal C, aromatic C, and phenolic C (Fig. 8). Under permanent pasture alkyl C and O-alkyl C each accounted for 39%, with 9% aromatic C and 8% carboxyl C (Table 3). Like the Kairanga soils, all the organic functional groups, except aromatic C, declined during medium-term (11 years of maize) cropping (Table 3). However, the decline in the Moutoa soil was markedly less than for Kairanga soils, and recovery was much faster (Table 3).
Scanning electron microscopy
SEM showed marked microstructural differences between aggregates from pasture and cropped sites (Figs 9-13). Generally, the surface coatings of aggregates under long-term pasture comprised aligned mineral particles glued by bridges or web-like structures of organic carbon. In contrast, aggregates under long-term cropping were devoid of the organo-mineral coatings but had aligned clays and silts only. On conversion back to pasture, aggregates showed the re-establishment of organo-mineral coatings and a mesh of web-like strands of organic carbon bridges, but are less well developed than under long-term pasture (Figs 9-11). While the oxide-rich Patumahoe and allophanic Egmont soils had lost their organo-mineral coating under long-term cropping, both soils show strong mineral microstructural development (Figs 12, 13). Silt-size vermiculitic minerals, silica polymorphs, and aggregated halloysite and kaolinite are enmeshed in crystalline Fe and Al oxides and short-range order Fe and alumino-silicates in the Patumahoe soil, while porous, semi-spheroidal allophane dominates the surface of the Egmont soil.
Changes in dry aggregate-size distribution and aggregate stability of 5 soils with different drainage properties, textures, and mineralogical compositions demonstrate that a significant decline in soil structure occurs on conversion from pasture to short-term conventional cropping, especially after medium- and long-term cropping (Figs 1-4). Similar structural declines have been reported in a large number of studies reviewed by Dexter (1988) and Kay (1997).
The MWD of water-stable aggregates decreased, while dry aggregate-size increased with cropping, with significant recovery after 10-11 years pasture following cropping. The abundance of [is less than] 0.5 mm water-stable aggregates increased with cropping at the expense of larger aggregates (Figs 1, 2). The trend was most pronounced in the micaceous Manawatu and Kairanga soils, moderately pronounced in the micaceous humic Moutoa and oxide-rich Patumahoe soils, and least pronounced in the allophanic Egmont soil. The anomalous increase in the MWD of water-stable aggregates in the Kairanga soil after 11 years of (winter-wet harvested) maize is considered to be due to the formation of compressed and remoulded secondary fragments that are more water stable than those produced under long-term (summer-dry harvested) cereal cropping. The structural condition, based on the dry aggregate-size distribution and aggregate stability, indicated that the soils could be ranked for their resistance (least susceptibility) to structural degradation under cropping in the following order: Egmont [is greater than] Patumahoe [is greater than] Moutoa [is greater than] Manawatu [is greater than] Kairanga.
We attribute the increase in dry aggregate-size under medium- and long-term cropping to an increase in soil strength and tensile strength (Rogowski et al. 1968; Perfect et al. 1993), and to a decline in the number and spatial distribution of failure zones (Kay 1997). This would result from the combined effects of SOM loss (Perfect et al. 1993), aggregate stability decline, the increase in soil bulk density and decline in soil porosity (Shepherd et al. 2001), the compactive forces of wheel traffic, and the mechanical disruption of primary aggregates under cultivation and their subsequent reformation as secondary clods.
The distribution of the size of soil aggregates indicates the susceptibility of soils not only to structural breakdown but also to clodding. The ranking of the soils included in this study in terms of their susceptibility to clodding was: Kairanga [is greater than] Moutoa [is greater than] Manawatu [is greater than] Patumahoe [is greater than] Egmont (Figs 3, 4). Oxide-rich Patumahoe soils, and particularly the allophanic Egmont soils, showed the greatest resistance to clod development, a function of their mineralogy and low tensile and shear strengths (data not shown). Dry aggregate-size distribution is also strongly influenced by soil texture (Perfect et al. 1993) with coarser textured soils having a greater resistance to clodding. The lower susceptibility of the micaceous coarse-textured Manawatu soil to clodding compared with the more organic-rich finer textured Moutoa soil is considered to be a function of its coarser texture and lower tensile and shear strength. It is also thought to be due to the fact that wheel traffic during cultivation, sowing, and sidedressing on the Manawatu soil occurred when the water content was below the critical water content (CWC) for compaction, where the CWC is the amount of water in the soil at which it is most prone to compaction. The Kairanga soil has the greatest susceptibility to clodding due to its poor drainage, micaceous mineralogy, and relatively high clay and low organic carbon content.
The relative difference in dry aggregate-size distribution between the pasture and recovery sites for the 3 micaceous soils suggests that the humic Moutoa soil showed the greatest resilience, followed by the coarse-textured Manawatu soil and fine-textured Kairanga soil. The MWD of water-stable aggregates was similar for Manawatu and Kairanga soils after 10 years of conversion back to pasture (Fig. 1). Recovery sites were not investigated on the Patumahoe and Egmont soils. However, the comparative effect of long-term cropping on the dry aggregate-size distribution and aggregate stability indicates that the oxide-rich Patumahoe and the allophanic Egmont soils are more resilient than the micaceous soils. The ranking of the soils for their resilience was Egmont [is greater than] Patu-mahoe [is greater than] Moutoa [is greater than] Manawatu [is greater than] Kairanga.
We attribute the loss of organic C under conventional continuous cultivation to the repeated exposure and subsequent aeration and oxidation of labile and older humified organic C associated with macro-aggregates and macropores. We also attribute the loss of organic C to the mineralisation by decomposer micro-organisms of previously occluded C in soil micropores (Shepherd et al. 2001), exposed as a result of the decline in aggregate stability and breakdown of soil aggregates. Total organic C declined at different rates for each soil under short- and medium-term cropping, but the rate of decline was similar under long-term cropping for the mica-rich and oxide-rich soils (Fig. 5). The little change in TOC in the allophanic Egmont soil under long-term cropping is considered to be due to the strong complexation of organic C with amorphous forms of Fe and Al. The greater loss of organic C in the Fe oxide-rich Patumahoe soil compared with the allophanic Egmont soil suggests that stabilisation of organic C against microbial attack is greater in the presence of complexation with Al hydroxides than with Fe hydroxides. The stability of organic C in the allophanic Egmont soil was also proposed by Parfitt et al. (1997) to be a function of the small particle size and large specific surface of the soil, which was capable of adsorbing organic molecules. The rates of recovery of TOC in the coarse-textured Manawatu and fine-textured humic Moutoa soils were similar on conversion back to pasture. Total organic C recovery rates in the fine-textured Kairanga soil were very much slower.
While TOC can change markedly under cultivation, [sup.13]C NMR evidence suggests that the proportional distribution of functional groups does not change (Skjemstad et al. 1986; Preston et al. 1994). Preston et al. (1994) noted only minor differences in [sup.13]C NMR spectra and the proportion of the C fractions after 65 years of cultivation of a Mollisol from British Columbia. Conversely, Golchin et al. (1995a) reported significant changes in both the amount and proportions of the different types of carbon in 5 uncultivated and cultivated Australian soils. In our study, all major functional groups of SOM declined after medium- and long-term cropping, with O-alkyl C showing the fastest rate of decline and alkyl C showing the slowest rate of decline (Table 3). The O-alkyl C group is derived from a combination of residual plant and microbial decomposition of plant residues (Cheshire 1979; Burns and Davies 1984), and is widely reported to play a major role in governing the water stability of soil aggregates by its binding and cementing action and strong absorbance by mineral surfaces. This relationship is consistent with the results of our study where the humic Moutoa soil has a higher aggregate stability and higher amounts of O-alkyl C than the Kairanga soil under pasture and cropping. However, the correlation coefficient of O-alkyl C with aggregate stability is weak (r = 0.852; P [is less than] 0.05), suggesting that other factors must be contributing to the stability of soil aggregates. Further, O-alkyl C is also weakly correlated to the MWD of the dry aggregate-size distribution (r = 0.862; P [is less than] 0.05).
Acid-hydrolysable carbon accounted for about 15-26% of the signal intensity in the O-alkyl region of the corresponding solid state [sup.13]C-NMR (Table 3). The higher content of O-alkyl C determined by NMR spectroscopy relative to AHC obtained by a chemical approach may be explained by the fact that signals of ether and OH-substituted C compounds other than carbohydrates can contribute to the intensity and chemical shift region between 48-110 ppm. But their contribution would be small. It means that the discrepencies between AHC and O-alkyl C are due to those carbohydrates that were not detected by chemical methods due to incomplete hydrolysis.
Several studies emphasise the contribution of WSC and AHC fractions in the formation of soil aggregates (Angers and Mehuys 1989; Angers et al. 1993; Ball et al. 1996). The WSC fraction comprises mainly free sugars and exocellular polysaccharides of microbial origin, which are involved in stabilising soil aggregates (Haynes and Swift 1990). Microbially derived polysaccharides act as a transient between microaggregates and hence are important factors in aggregation (Tisdall and Oades 1982). A decline in this fraction with cultivation is consistent with a decline in soil structure, suggesting that this fraction is involved in soil aggregate formation through physical binding and chemical cementation. Loss of WSC fraction with continued cultivation is well known to decrease the degree of aggregation, more so in some soils than others (Sollins et al. 1996), the variation reflecting differences in the stability of aggregates. When 3 of our sites were restored to pastures, a build-up in WSC was accompanied by an improvement in soil structure, as indicated by an increase in aggregate stability and a decrease in the MWD of dry aggregates (Figs 1-4). In grazed pastures, 70-80% of the above-ground herbage is grazed and excreta are returned as organic input. Pasture roots also contribute an additional 4-5 t C/ha annually (Saggar et al. 1997, 1999a). In cropping, most of the organic input is from roots (~1.0 t C/ha for maize roots). Kinchesh et al. (1995) showed that amounts of carbohydrates were higher where inputs of organic carbon were largest. Aggregate stability would, therefore, be expected to be higher under a pasture than a cropping regime.
Our results show how the greater stability of the soil under pasture was reflected in its larger content of TOC, WSC, and AHC. More WSC was extracted from humic Moutoa soil with the highest organic carbon levels. The fine-textured (Kairanga) soil contained more WSC than the coarse-textured (Manawatu) soil. However, WSC fraction declined faster in the fine-textured soil, as did the structural breakdown. The proportion of WSC as AHC in cultivated allophanic Egmont soil and oxide-rich Patumahoe soil was considerably lower (5.5-7.6%) than other soils (Fig. 6). This may be due to greater chemical bonding of the WSC fraction to mineral surfaces in these high surface-area soils. Saggar et al. (1994, 1996) showed that turn-over and stabilisation of microbial metabolites produced during the decomposition of labelled substrates were dependent on clay surface area. Higher surface area made metabolites less accessible to decomposers, particularly in soils enriched in short-range-order (allophanic) minerals and in expansible layer silicates (Saggar et al. 1999b), suggesting their reduced extractability.
Total organic C, WSC, and AHC fractions were linearly correlated with aggregate stability (Table 2). The WSC fraction did not show any significant correlations with dry aggregate-size distribution, while the AHC fraction was weakly correlated with dry aggregate-size distribution. Considering the linear correlations between aggregate stability and dry aggregate size, and between AHC and dry aggregate size, and that the 3 soils with the highest aggregate stabilities also have the highest amounts of AHC, the more complex polysaccharides of the AHC fraction could play a greater role in aggregate stability than the free sugars and exocellular carbohydrates of the WSC fraction. This is supported by the observation that hot water does not appear to extract a fraction enriched in microbially synthesised carbohydrates (Ball et al. 1996). Moreover, the more complex polysaccharides of both plant and microbial origin, in addition to free sugars and exocellular polysaccharides of microbial origin, are included in the AHC fraction (Cheshire 1979).
Alkyl C comprises a significant proportion of the SOM in our study, accounting for 29-34% of TOC in the fine-textured Kairanga soil and 38-42% of TOC in the humic Moutoa soil (Table 3). Its accumulation is attributed to the formation of polymethylene end products by the microbial mineralisation of SOM under aerobic conditions (Oades 1995), and to the selective utilisation of other labile forms of plant C and preservation of plant- derived polymers (such as cutin, suberin, etc.). The accumulation of alkyl-C is also attributed to its relative recalcitrance (Skjemstad et al. 1983,1986; K6gel-Knabner et al. 1991; Baldock et al. 1992), with the degree of recalcitrance being dependent on the structure of the -CH2-chain. Skjemstad and Dalai (1987) suggested that long-chain alkyl C structures are less recalcitrant than the more branched alkyl C structure. The relatively sharp band shape of the alkyl C peak at 30 ppm under pasture (Fig. 7) suggests a predominance of the less recalcitrant long-chain structures, while the broader band shapes under 11 years of maize and 30 years of barley indicate a predominance of the more recalcitrant, branched alkyl group. The reduction in the intensity and change in band shape of the alkyl C peak from pasture to 11 years of maize suggests that much of the less recalcitrant long-chain alkyl C had been mineralised under medium-term cropping. The further reduction in alkyl C after long-term (30 years) cropping, as indicated by the reduction in both peak intensity (Fig. 7) and its absolute contribution (Table 3), indicates that the more recalcitrant branched alkyl C continues to decline with increasing cropping.
The very high wet aggregate stability of the humic Moutoa soil under pasture compared with the Kairanga soil is attributed to the higher amount of organic carbon and the amount and type of alkyl C present in the Moutoa soil. Ma'shum et al. (1988) noted that long-chain polymethylene compounds were responsible for water repellency in some Australian soils. The humic Moutoa pasture soil, with its alkyl C fraction dominated by long-chain polymethylene structures, is severely water repellent, with a molarity of ethanol droplets (MED) of 3.6. After medium-term (11 years) cropping, the surviving alkyl C fraction is predominantly in the branched form and the soil is non-repellent (MED = 0.1). While the humic Moutoa soil contains more than twice the amount of alkyl C of the Kairanga soil under pasture, the severely water-repellent Moutoa pasture soil contains only slightly more alkyl C than the non-repellent Moutoa soil after an 11-year cropping (Table 3). This suggests that the amount and particularly the form of alkyl C is important in governing the degree of water repellency and, indirectly, aggregate stability. The formation of a protective, stabilising, water-repellent lattice of long-chain polymethylene compounds around the Moutoa soil aggregates under pasture is consistent with the work of Capriel et al. (1990) and Dinel et al. (1990), who found high correlation coefficients between the aliphatic water-repellent fraction and the water stability of soil aggregates. While the allophanic Egmont soil was not analysed using [sup.13]C NMR because of the potentially high paramagnetic effect, infrared evidence suggests that the soil is enriched in alkyl C in the [is less than] 2 [micro]m fraction (Parfitt et al. 1997). The physical occlusion of comparatively large amounts of alkyl C in fine and very fine micropores, and its subsequent protection from decomposer microorganisms, could explain, in part, the high aggregate stability of the Egmont soil under pasture and after long-term cropping (Fig. 2).
The apparent stability of the aromatic C peak at 130 ppm (Fig. 7) under increasing cropping of the Kairanga soil could be a function of its inability to degrade appreciably under anaerobic conditions (Sommers et al. 1981)--conditions that would occur for a significant part of the year in the degraded medium- and long-term cropping sites. Alternatively, the stability of aromatic C could be due to its reported recalcitrant nature (Frund et al. 1994). The decline in aromatic C shown in Table 3 may be caused by the loss of the phenolic subgroup of aromatic C at 149 ppm (Fig. 7), which can be degraded during oxidation. Unlike the fine-textured Kairanga soils, the aromatic C peak at 130 ppm in the humic Moutoa soil declined slightly under cropping, suggesting possible degradation of the non-phenolic components of aromatic C under the better aerated conditions that would prevail in the Moutoa soil. The decline of the aromatic C is also seen from the absolute contribution of percent C (data not shown). Given the small amounts of aromatic C and carboxyl C present in these soils and the absence of a corresponding increase with aggregate stability in the Kairanga soil on recovery after 10 years of pasture, aromatic and carboxyl C are not considered to contribute significantly to the stability of soil aggregates.
Given the moderately strong correlation (P [is less than] 0.01) between aggregate stability and dry aggregate size, both the lack of a significant correlation between WSC and dry aggregate size, and the weak correlation (P [is less than] 0.05) between AHC and dry aggregate size, could be due to other factors besides the chemical structure of the organic carbon present. The coarse-textured mica-rich Manawatu soil, the oxide-rich Patumahoe soil, and particularly the allophanic Egmont soil have lower tensile and shear strengths (data not shown) than the heavier textured mica-rich Kairanga and humic Moutoa soils. While the type of organic carbon plays a significant role in determining soil structural characteristics, the difference in tensile and shear strength, due to differences in soil mineralogy and texture, is considered to be a major contributing factor governing dry aggregate-size distribution in these soils.
SEM of soil microstructures generally relates to the aggregate water-stabilities. The more stable aggregates under pasture (Figs 1, 2) exhibit surface coatings with visible organo-mineral interactions (Figs 9-13). Contrasting with this, less stable aggregates after cropping show surface skins of predominantly orientated mineral particles, with very little visible organic bridging. On conversion back to pasture, the proportion of water-stable aggregates increased with the re-establishment of organic bridging between silt and clay particles. Metzger and Robert (1985) similarly found that increasing levels of organic carbon (through the addition of sludge extracts) promoted micro-aggregation through organo-mineral interactions, as seen in SEMs. Powers and Skidmore (1984) also demonstrated that undisturbed soils had more bridging (speculated to be organic bonding) between primary particles and micro-aggregates than disturbed soils subjected to simulated tillage. These differences were commensurate with aggregate stability differences. Blank and Fosberg (1989) illustrated similar differences in SEM micro-aggregation comparing virgin and cultivated soils in South Dakota.
Our relationships between soil microstructural development, aggregate stability, and soil organic carbon for the mica-rich soils under cropping and after conversion back to pasture are in accord with mechanisms proposed by Tisdall and Oades (1982), as updated and expanded by Jastrow and Miller (1997). The exceptions were the oxide-rich Patumahoe soil and the allophane-rich Egmont soil. While the aggregate stability of the Patumahoe soil declined to very low levels under long-term cropping, partly as a result of the loss of SOM and its associated binding, cementing, and enmeshing effects, the soil still maintained a strongly developed microstructure (Fig. 12b). This is probably a function of the crystalline Fe and Al oxides and short-range order Fe and alumino-silicates present (Table 1). The allophanic Egmont soil maintained not only a strongly developed microstructure under long-term cropping, but also a relatively high aggregate stability, despite the apparent loss of the surface coating of SOM (Fig. 13b). Organic bridging for developing water-stabilising aggregate coatings was less important for this soil compared with the oxide-rich and particularly with the micaceous mineral soils. Whereas SEM demonstrates the effect of organic bridging on aggregate stability, it also highlights the importance of other contributing factors such as soil mineralogy, and the amount, type, and location of the organic carbon present. While the aggregate stability of the mica-rich humic Moutoa soil declined with the loss of the organic bridging, it did not decline to the same level of the mica-rich mineral soils, suggesting that the amount and type, or position, of SOM are also important determinants. Further, the physical occlusion of comparatively large amounts of alkyl C in the fine and very fine micropores of the Egmont soil could help explain its continued strong microstructure after long-term cropping (Fig. 13).
We were unable to make any connections between SEM microstructures and aggregate-size distributions. Although there have been reported relationships between the extent and type of organo-mineral bridging and dry aggregate sizes, we did not examine them. Also, we were unable to distinguish visually the various components of organic carbon, as recorded by the chemical and NMR fractionation, in the scanning electron micrographs.
The conversion of pasture soils to continuous cropping using conventional cultivation markedly decreased the water stability of soil aggregates and increased dry aggregate size in soils with a mica-dominant mineralogy. Soils containing appreciable amounts of Fe and Al oxides were more resistant to structural degradation than the micaceous soils, while allophanic soils showed the greatest resistance and resilience. Reductions in total C, extractable carbohydrates, alkyl C, and O-alkyl C, and the loss of the protective coating of glue and web-like structures of organic carbon around the outer surface of soil aggregates under increasing cultivation, were related to the reduction in aggregate stability and increase in dry aggregate size in the mineral micaceous soils. While the above factors influence the aggregate stability and dry aggregate-size distribution of the mica-rich humic Moutoa soil, the very high aggregate stability of the Moutoa soil under pasture is attributed to the presence of a protective, stabilising water-repellent lattice of long-chain polymethylene compounds around the soil aggregates. While the loss of total C, organic bridging, and extractable carbohydrates in the oxide-rich Patumahoe soil described the decline in aggregate stability after long-term cropping, the absence of a significant increase in dry aggregate size is considered to be a function of the lower tensile and shear strength of these soils, arising from the presence of crystalline Fe and Al oxides and amorphous Fe and alumino-silicate minerals. Although the allophanic Egmont soil lost its surface coating of SOM around aggregates under long-term cropping, it did not show an appreciable decline in total C, AHC, and aggregate stability, or an increase in dry aggregate size. This is attributed to the strong complexation of organic carbon with amorphous forms of Fe and particularly Al, the occlusion of alkyl C in protected fine and very fine micropores, the strong microstructure, and the inherent low tensile and shear strength of a soil dominated by allophane.
The authors extend their grateful thanks to Gerrard Lilley for aggregate stability measurements, Joe Whitton for mineralogical analyses, and to Roger Parfitt and Nanthi Bolan for their useful comments and review of the manuscript. Douglas Hopcroft, Hort+Research, assisted with electron microscopy. We are also indebted to the farmers for allowing such ready access to their fields. Financial support for this study was provided by the Foundation for Research, Science and Technology under Contract CO9804.
Angers DA, Mehuys GR (1989) Effect of cropping on carbohydrate content and water-stable aggregation of a clay soil. Canadian Journal of Soil Science 69, 373-380.
Angers DA, Samson N, Legere A (1993) Early changes in water-stable aggregation induced by rotation and tillage in a soil under barley production. Canadian Journal of Soil Science 73, 51-59.
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.
Ball BC, Cheshire MV, Robertson EAG, Hunter EA (1996) Carbohydrate composition in relation to structural stability, compactibility and plasticity of two soils in a long-term experiment. Soil and Tillage Research 39, 143-160.
Bisdom EBA (1983) Submicroscopic examination of soils. Advances in Agronomy 36, 55-96.
Blakemore LC, Searle PL, Daly BK (1987) Methods for chemical analysis of soils. New Zealand Soil Bureau Scientific Report 80.
Blank RR, Fosberg MA (1989) Cultivated and adjacent virgin soils in north-central South Dakota: II Mineralogy and micromorphological comparisons. Soil Science Society of America Journal 53, 1490-1499.
Brink RH, Dubak P, Lynch DL (1960) Measurements of carbohydrates in soil hydrolyzates with anthrone. Soil Science 89, 157-166.
Burns RG, Davies JA (1984) The microbiology of soil structure. Biology Agriculture and Horticulture 3, 95-113.
Capriel P, Beck T, Borchert H, Harter P (1990) Relationship between soil aliphatic fraction extracted with supercritical hexane, and soil microbial biomass and aggregate stability. Soil Science Society of America Journal 54, 415-420.
Carter MR, Angers DA, Kunelius HT (1994) Soil structural form and stability, and organic matter under cool-season perennial grasses. Soil Science Society of America Journal 58, 1194-1199.
Cheshire MV (1979) `Nature and origin of carbohydrates in soil.' (Academic Press: London)
Churchman GJ, Tate KR (1986) Effect of slaughterhouse effluent and water irrigation upon aggregation in seasonally dry New Zealand soil under pasture. Australian Journal Soil Research 24, 505-516.
Claydon JJ (1989) Determination of particle size distribution of fine grained soils--Pipette method. DSIR Division of Land and Soil Sciences Technical Record LH5.
Degens BP, Sparling GP, Abbott LK (1994) The contribution from hyphae, roots and organic C involved in the macro-aggregation of a sandy loam soil under long-term clover-based or grass pastures. European Journal of Soil Science 45, 459-468.
Dexter AR (1988) Advances in the characterization of soil structure. Soil and Tillage Research 11, 199-238.
Dinel H, Schnitzer M, Mehuys GR (1990) Soil lipids: Origin, nature, content, decomposition and effect on soil physical properties. In `Soil biochemistry'. (Eds JM Bollag, G Stotsky) Vol. 6, pp. 397-429. (Marcel Dekker: New York)
Elliott ET (1986) Aggregate structure and carbon, nitrogen and phosphorus in native and cultivated soils. Soil Science Society of America Journal 50, 627-633.
Frund R, Haider K, Ludemann HD (1994) Impacts of soil management practices on the organic carbon structure-investigations by CPMAS [sup.13]C-NMR spectroscopy. Zeitschrift fur Pflanzenernahrung und Bodenkunde 157, 29-35.
Golchin A, Baldock JA, Oades JM (1997) A model linking organic carbon decomposition, chemistry, and aggregate dynamics. In `Soil processes and the carbon cycle'. (Eds R Lal, JM Kimble, RF Follet, BA Stewart) Ch. 17, pp. 245-266. Advances in Soil Science. (CRC Press: Boca Raton, FL)
Golchin A, Clarke P, Oades JM, Skjemstad JO (1995a) The effects of cultivation on the composition of organic carbon and structural stability of soils. Australian Journal of Soil Research 33, 975-993.
Golchin A, Oades JM, Skjemstad JO, Clarke P (1995b) Structural and dynamic properties of soil organic carbon as reflected by [sup.13]C natural abundance, pyrolysis mass spectrometry and solid state [sup.13]C-NMR spectroscopy in density fractions of an Oxisol under forest and pasture. Australian Journal of Soil Research 33, 59-76.
Gradwell MW, Birrell K (1979) Methods for physical analysis of soils. New Zealand Soil Bureau Scientific Report 10C. Department of Scientific and Industrial Research, Wellington.
Hakansson I, Voorhees WB, Riley H (1988) Vehicle and wheel factors influencing soil compaction and crop response in different traffic regimes. Soil and Tillage Research 11, 239-282.
Haynes RJ, Beare MH (1995) Aggregation and soil organic carbon storage in meso-thermal, humid soils. In `Structure and organic carbon storage in agricultural soils'. (Eds MR Carter, BA Stewart) pp. 213-262. Advances in Soil Science. (CRC Lewis Publishers: Boca Raton, FL)
Haynes RJ, Swift RS (1990) Stability of soil aggregates in relation to organic constituents and soil water content. Journal of Soil Science 41, 73-83.
Haynes RJ, Swift RS, Stephen RC (1991) Influence of mixed cropping rotations (pasture-arable) on organic carbon content, water stable aggregation and clod porosity in a group of soils. Soil and Tillage Research 19, 77-87.
Hewitt AE (1998) New Zealand soil classification. 2nd edn. Landcare Research Science Series No. 1. Lincoln, NZ: Manaaki Whenua -- Landcare Research New Zealand.
Jastrow JD, Miller RM (1997) Soil aggregate stabilization and carbon sequestration: Feedbacks through organo-mineral associations. In `Soil processes and the carbon cycle'. (Eds R Lal, JM Kimble, RF Follet, BA Stewart) Ch. 15, pp. 207-223. Advances in Soil Science. (CRC Press: Boca Raton, FL)
Kay BD (1997) Soil structure and organic carbon: A review. In `Soil processes and the carbon cycle'. (Eds R Lal, JM Kimble, RF Follet, BA Stewart) Ch. 13, pp. 169-197. Advances in Soil Science. (CRC Press: Boca Raton, FL)
Kinchesh DS, Powlson DS, Randall EW (1995) [sup.13]C NMR studies of organic carbon in whole soils: II. A case study of some Rothamsted soils. European Journal of Soil Science 46, 13%146.
Kogel-Knabner I, Hatcher PG, Zech W (1991) Chemical structural studies of forest soil humic acids: aromatic carbon fraction. Soil Science Society of America Journal 55, 241-247.
Low AJ (1972) The effect of cultivation on the structure and other physical characteristics of grassland and other arable soils. Journal of Soil Science 24, 249-259.
Ma'shum M, Tate ME, Jones GE Oades JM (1988) Extraction and characterization of water-repellent materials from Australian soils. Journal of Soil Science 39, 99-109.
Metzger L, Robert M (1985) A scanning electron microscopy study of the interactions between sludge organic components and clay particles. Geoderma 35, 159-167.
Metzger L, Yaron B (1987) Influence of sludge organic carbon on soil physical properties. In `Advances in soil science'. (Ed. BA Stewart) Vol. 7, pp. 141-163. (Springer-Verlag: New York).
Monreal CM, Kodama H (1997) Influence of aggregate architecture and minerals on living habitats and soil organic matter. Canadian Journal of Soil Science 77, 367-377.
New Zealand Meteorological Service (1983) Summaries of climatological observations to 1980. NZ Meteorological Service Miscellaneous Publication 177, New Zealand Ministry of Transport, Wellington.
Oades JM (1984) Soil organic carbon and structural stability: mechanisms and implications or management. Plant and Soil 76, 319-337.
Oades JM (1995) Recent advances in organo-mineral interactions: Implications for carbon cycling and soil structure. In `Environmental impact of soil component interactions'. (Eds PM Huang, J Berthelin, J-M Bollag, BW McGill, AL Page) pp.119-134. (Lewis Publishers: London)
Parfitt RL, Theng BKG, Whitton JS, Shepherd TG (1997) Effects of clay minerals and land use on organic carbon pools. Geoderma 75, 1-12.
Perfect E, Kay BD, Ferguson JA, da Silva AP, Denholm KA (1993) Comparison of functions for characterizing the dry aggregate size distribution of tilled soil. Soil and Tillage Research 28, 123-139.
Powers DH, Skidmore EL (1984) Soil structure as influenced by simulated tillage. Soil Science Society of America Journal 48, 879-884.
Preston CM, Newman RH, Rother P (1994) Using [sup.13]C CPMAS NMR to assess effects of cultivation on the organic carbon of particle size fractions in a grassland soil. Soil Science 157, 26-35.
Rogowski AS, Moldenhauer WC, Kirkham D (1968) Rupture parameters of soil aggregates. Soil Science Society of America Journal 32, 720-724.
Saggar S, Hedley C, Mackay AD (1997) Partitioning and translocation of photosynthetically fixed [sup.14]C in grazed pastures. Biology and Fertility of Soils 25, 152-158.
Saggar S, Mackay AD, Hedley CB (1999a) Hill slope effects on the vertical fluxes of photosynthetically fixed [sup.14]C in a grazed pasture. Australian Journal of Soil Research 37, 655-666.
Saggar S, Parshotam A, Hedley C, Salt G (1999b) [sup.14]C-labelled glucose turnover in New Zealand soils. Soil Biology and Biochemistry 31, 2025-2037.
Saggar S, Parshotam A, Sparling GP, Feltham CW, Hart PBS (1996) [sup.14]C-labelled ryegrass turnover and residence times in soils varying in clay content and mineralogy. Soil Biology and Biochemistry 28, 1677-1686.
Saggar S, Tate KR, Feltham CW, Childs CW, Parshotam A (1994) Carbon turnover in a range of allophanic soils amended with [sup.14]C-labelled glucose. Soil Biology and Biochemistry 26, 1263-1271.
Shepherd TG (1992) Sustainable soil-crop management and its economic implications for grain growers. In `Sustainable land management: Proceedings of the International Conference on Sustainable Land Management'. Napier, 1991. (Ed. PR Henriques)pp. 141-152. (Hawke's Bay Regional Council: Napier, NZ)
Shepherd TG, Parshotam A, Newman RH (2001) Effects of cultivation and conversion back to pasture on the porosity and organic carbon of a clayey soil. European Journal of Soil Science (In press).
Skjemstad JO, Dalai RC (1987) Spectroscopic and chemical differences in organic carbon of two Vertisols subjected to long periods of cultivation. Australian Journal of Soil Research 25, 323-335.
Skjemstad JO, Dalal RC, Barron PF (1986) Spectroscopic investigation of cultivation effects on organic carbon of Vertisols. Soil Science Society of America Journal 50, 354-359.
Skjemstad JO, Frost RL, Barron PF (1983) Structural units in humic acids from south-eastern Queensland soils as derived by [sup.13]C NMR spectroscopy. Australian Journal of Soil Research 21, 539-547.
Smart P, Tovey NK (1981) `Electron microscopy of soils and sediments: examples.' (Clarendon Press: Oxford)
Soil Survey Staff (1996) `Keys to soil taxonomy.' 7th edn. (United States Department of Agriculture, Natural Resources Conservation Service: Washington, DC)
Sollins P, Homann P, Caldwell BA (1996) Stabilization and destabilization of soil organic carbon: mechanisms and controls. Geoderma 74, 65-105.
Sommers LE, Gilmour CM, Wildung RE, Beck SM (1981) The effect of water potential on decomposition processes in soils. In `Water potential relations in soil microbiology'. (Eds JF Parr, WR Gardner, LF Elliot) Ch. 3, pp. 97-117. SSSA Special Publication No. 9. (Soil Science Society of America: Madison, WI)
Sparling GP, Shepherd TG, Kettles HA (1992) Changes in soil organic C, microbial biomass and aggregate stability under continuous maize and cereal cropping, and after restoration to pasture in soils from the Manawatu region, New Zealand. Soil and Tillage Research 24, 225-241.
Sullivan LA, Koppi AJ (1987) In situ soil organic carbon studies using scanning electron microscopy and low temperature ashing. Geoderma 40, 317-332.
Tisdall JM, Oades JM (1982) Organic carbon and water stable aggregates in soils. Journal of Soil Science 33, 141-163.
Voorhees WB, Lindstrom MJ (1984) Long-term effects of tillage method on soil tilth independent of wheel traffic compaction. Soil Science Society of America Journal 48, 152-156.
Wilkinson L (1996) `SYSTAT 6.0 for windows: statistics.' (SPSS Inc.: Chicago)
Yoder RE (1936) A direct method of aggregate analysis of soils and a study of the physical nature of erosion losses. Journal of American Society of Agronomy 28, 337-351.
Manuscript received 7 April 2000, accepted 18 October 2000
T. G. Sheperd(B), S. Saggar, R. H. Newman(A), C. W. Ross, and J. L. Dando
Landcare Research New Zealand, Private Bag 11052, Palmerston North, New Zealand.
(A) Industrial Research Limited, PO Box 31 310, Lower Hutt, New Zealand.
(B) Corresponding author; e-mail: Shepherdg@landcare.cri.nz
|Printer friendly Cite/link Email Feedback|
|Author:||Shepherd, T. G.; Saggar, S.; Newman, R. H.; Ross, C. W.; Dando, J. L.|
|Publication:||Australian Journal of Soil Research|
|Article Type:||Statistical Data Included|
|Date:||May 1, 2001|
|Previous Article:||Soil organic matter in rainfed cropping systems of the Australian cereal belt(*).|
|Next Article:||Soil microbial biomass, metabolic quotient, and carbon and nitrogen mineralisation in 25-year-old Pinus radiata agroforestry regimes.|