Long-term effects of wheat production management practices on some carbon fractions of a semiarid Plinthustalfs.
Declining soil fertility and increasing atmospheric carbon (C) can be a signal of escalating pressure on arable lands and lack of sustainable management practices to rehabilitate soils of prolonged cultivation. Continuous cultivation following increased food demand for the growing population has resulted in the breakdown of traditional farming practices such as recycling of crop residues, fallowing, crop-livestock farming and agroforestry that were capable of restoring C and soil productive capacity (McNeill and Winiwarter 2004). Incorporation of crop residues by tillage has proved on many occasions to be one unsustainable residue management system commonly practiced across the globe. This system accelerates decomposition and mineralisation of crop residues, causing great emissions of greenhouse gases into the atmosphere. Tillage also disrupts soil aggregates and subjects protected C to biological oxidation and erosion (Chen et al. 2009). Consequently, all the processes governed by soil C deteriorate and pose a long-term threat to soil quality and food security.
Disposal of crop residues is another method used to manage crop residues after harvesting. In many countries, including South Africa (Kotze and Du Preez 2007; Loke et al. 2012), these organic materials are still regarded as waste and are disposed of by burning to ease tillage operations and destroy habitat for pests and diseases (Singh and Rengel 2007). Indeed the use of fire is a quick way to release nutrients and increase, although only temporarily, soil pH. Nevertheless, burning crop residues can also have some negative effects on soil C, nutrient loss by volatilisation and several soil biological properties (Singh and Rengel 2007). The use of chemical inputs to improve soil fertility and crop production has also been a challenge in most developing countries due to high costs, dwindling purchasing power and limited credit facilities to farmers (Bakht et al. 2009). Previous studies show that crop production can still decline even in well-fertilised soils if their C content is low (Lobe et al. 2001; Kotze and Du Preez 2007). Therefore, restoration of soil C is essential to regenerate degraded soils and improve ecosystem functioning.
The only way to restore or improve soil C in arable land is to increase C inputs and reduce its losses (Janzen 2005; Chen et al. 2009). These attributes are associated with conservation tillage, which has been touted as a perfect alternative to mouldboard ploughing. In conservation tillage systems, at least 30% of crop residues are retained at or near the soil surface after harvesting to provide cover against erosion and improve water infiltration (Singh and Rengel 2007). Surface-retained residues and minimum soil disturbance also reduce C losses through decomposition or erosion and improve nutrient use efficiency (Singh and Rengel 2007).
Wiltshire and Du Preez (1993), Kotze (2004), Kotze and Du Preez (2007) and Loke et al. (2012) investigated the effect of residue management under conservation (no- and mulch-tillage) and conventional tillage systems on soil organic C (SOC) content of a semiarid Plinthustalf in a wheat trial located at Bethlehem, South Africa, and found that surface-retained residues in no-tillage treatments increased SOC in the upper 0-50 mm layer compared with incorporated burned or unburned wheat straw in the ploughed plots. Similar results were reported elsewhere under similar and different agro-ecological settings (Bakht et al. 2009; Carvalho et al. 2009; Chen et al. 2009; Sun et al. 2012). Nevertheless, adoption rates of conservation tillage systems in most developing countries, including South Africa are still lagging behind (Derpsch and Friedrich 2009; Friedrich et al. 2012), probably because the benefits become evident only after a couple of decades and the implements (e.g. planters) adapted to conservation tillage systems are expensive (Singh and Rengel 2007). Friedrich et al. (2012) are of the opinion that limited research and practical results, together with poor information dissemination to farmers, are the major factors that crippled expansion and spread of conservation agriculture in most countries. As such, little is known regarding the sustainability of conservation tillage practices on soil quality, especially under semiarid climates.
To assess the sustainability of conservation tillage systems on soil C storage, it is crucial to measure different C pools because they differ in their response timeframes to changes in land use and management systems as well as their roles in soil (Gregorich et al. 1994; Janzen 2005). Soil C exists in organic and inorganic forms, and is estimated to be the largest C pool in terrestrial ecosystems (Batjes 1996; Shi et al. 2012), storing 1550 Pg of SOC and 950 Pg of soil inorganic C (SIC) (Batjes 1996). Although both SOC and SIC have been shown to play a critical role in global C cycling, SIC has received little attention (Wang et al. 2010; Lai et al. 2015). Many researchers disregarded SIC citing its negligible contribution towards C sequestration (Batjes 1996) as well as its restricted distribution within arid to semiarid regions (Wu et al. 2009; Wang et al. 2010; Shi et al. 2012), which occupy at least one-third of the earth's land surface (Batjes 1996; Shi et al. 2012). In fact, SIC dynamics are not fully understood because some researchers believe SIC to be less sensitive to soil management changes (Batjes 1996), but new research developments revealed that production management practices can alter SIC (Mikhailova and Post 2006).
Based on its recalcitrance to microbial decomposition, SOC can be classified into labile, slow and passive pools. Labile C serves as energy source for the decomposer community and is a sensitive indicator of changes in land use and soil management; whereas slow to passive C pools are more relevant to long-term soil structural formation and stability as well as C sequestration (Von Lutzow et al. 2006). The SIC exists as carbonates either derived from parent material (primary or lithogenic) or formed through soil processes (secondary or pedogenic) (Wu et al. 2009; Shi et al. 2012; Jin et al. 2014; Lal et al. 2015). The latter can further be subdivided into pedo-lithogenic carbonate due to calcium ([Ca.sup.2+]) or magnesium ([Mg.sup.2+]) inherited from parent material and pedo-atmogenic carbonate due to [Ca.sup.2+] or [Mg.sup.2+] originating from external sources (Sanderman 2012; Ahmad et al. 2015). Inorganic C is also important for soil structural stability and nutrient cycling (Bronick and Lal 2005). Because SOC is a reservoir of [Ca.sup.2+] and [Mg.sup.2+] and a source of soil carbon dioxide (C[O.sub.2]), its presence in sufficient levels can stimulate dissolution or precipitation of SIC when it decays (Bronick and Lal 2005). The C[O.sub.2] produced during SOC decomposition can be incorporated to form SIC, and thus C sequestration (Sanderman 2012).
Traditional wet chemical extraction methods have been used to characterise these soil C pools. Several modern analytical tools are also available, including solid state [sup.13]C Nuclear Magnetic Resonance (NMR) spectroscopy. Non-destructive NMR spectroscopy is considered one useful technique to monitor transitional changes in the chemical composition of C contained in crop residues during decomposition and humification (Baldock et al. 1992, 1997). However, there is limited research regarding the combined use of wet chemical extraction and solid state [sup.13]C NMR spectroscopic analytical procedures to characterise soil C. Although wet chemical extraction methods may give inconsistent results compared with those obtained by [sup.13]C NMR spectroscopy due to their efficiency differences (Baldock et al. 1992; Kogel-Knabner 1997), characterisation of soil C with chemical extraction methods in combination with solid state [sup.13]C NMR spectroscopy can improve our understanding of the relationship between SOC quality and quantity of the semiarid Plinthustalfs (Vazquez et al. 2016).
Plinthustalfs are the dominant soil types in the Bethlehem agro-ecosystem, and because of their sandy texture they tend to have low C stabilisation capacity. Therefore, understanding C dynamics of the semiarid sandy Plinthustalfs is vital for developing a sustainable management system. The objective of this study was to evaluate the impact of two residue management (unburned and burned) under three tillage (mouldboard ploughing, stubble mulch tillage and no-tillage) systems and two weed control methods (chemical and mechanical) on soil C pools of a semiarid Plinthustalf in a long-term wheat trial near Bethlehem, using both wet chemical extraction and solid state [sup.13]C NMR spectroscopic methods. We hypothesised that surfaceretained wheat residues in conservation tillage (no- and mulchtillage) systems will restore soil C pools and SOC quality in the upper 0-50 mm soil layer.
Material and methods
A trial was established in 1979 at the Agricultural Research Council (ARC) Small Grain Institute (28[degrees]9'S, 28[degrees]17'E; 1680m above sea level) near Bethlehem in the Eastern Free State of South Africa to study the effects of some wheat management practices on soil fertility and crop productivity. Prior to acquisition of the site by the institute it was conventionally tilled for at least 20 years, but other management details are unknown. The mean annual rainfall in this area is 695 mm and the mean annual class-A pan evaporation is 1883 mm, resulting in a mean annual aridity index of 0.37. Most of the rain (79%) falls during October-March, with mean daily temperatures ranging from 6.7[degrees]C in July to 20.1[degrees]C in January.
According to the Land Type Survey Staff (2001), the trial was laid on a land type defined as a plinthic catena, which in upland positions has margalitic or duplex soils derived from Beaufort mudstone, shale, sandstone and grit with dolerite sills in places. The land type covers a substantial 420 000 ha. According to the USDA system, the soil would fall under the Great Group Plinthustalfs (Soil Survey Staff 1998). This Plinthustalf consists of three diagnostic horizons: an orthic Ap (0-300 mm), yellow-brown apedal B1 (300-650 mm) and soft plinthic B2 (>650 mm), containing 18, 23 and 36% clay respectively. The parent material comprises a mixed deposit of aeolian and colluvial origin on shale that increases with depth from 750 to 900 mm.
The trial was laid out on a 2-3% north-facing slope using a randomised complete block design with three blocks serving as replicates. Each block comprises 36 field treatments: two straw management treatments (unburned and burned), three tillage systems (no-tillage, stubble mulch and ploughing), two weed control methods (chemical and mechanical) and three levels of nitrogen (N) fertilisation in a factorial arrangement. The N levels were 20, 30 and 40 kg N [ha.sup.-1] and soil samples were taken only from plots that received the intermediate N level (30 kg N [ha.sup.-1]), because this rate reflects farmers' common practice in the region. Plots were 6 m x 30 m with 10-m borders between blocks and were cropped annually with winter wheat (Triticum aestivum L.) without any rotation or replacement with a summer crop. However, in 1990 and 1991 oats (Avena sativa L.) was used as a substitute crop, as a way to reduce soil-borne disease (take-all, Gaeumannomyces graminis) that occurred in some treatments. A fallow period of five months was maintained in this trial to restore soil water between harvesting and seeding, during which most of the rainfall events were expected.
Immediately after harvesting in December, wheat straw in the relevant no-tilled, stubble mulched and ploughed treatments was burned or left unburned. In the ploughed treatments, just after burning, a two-way offset disc was used to incorporate either unburned wheat straw or wheat straw ash to 150-mm depth, followed by mouldboard ploughing to a depth of 250 mm in February, when the soil was sufficiently moist and easy to work. The stubble mulch treatments were not disked, but cut at 100-150 mm using a v-blade and then ripped with a 50-mm width chisel plough at 300-mm spacing to the same depth and at the same time as ploughed treatments. The no-tilled plots were not ploughed. Stubble in no-tilled plots was neither burned nor cut. In fact, no-tilled treatments were slightly disturbed only during planting.
Weeding was done once in the relevant treatments when needed the first time, generally in March. This was done either by a mechanical cultivator (rodweeder or v-blade depending on the soil water level) or by spraying non-selective herbicides at recommended rates - initially only glyphosate (as Roundup, Monsanto Co.; 369 g a.i. [L.sup.-1]) was used, later on it was alternated with Paraquat at 200 g a.i. [L.sup.-1] to reduce chances of herbicide resistance developing.
All treatment plots were slightly disturbed with a combined seeder-fertiliser drill used for sowing the wheat seed together with the premixed fertiliser. A fertiliser blend of 3 : 2 :0 NPK (25%) + 0.75% zinc (Zn) was applied at a rate of 133 kg [ha.sup.-1] that resulted in N, P (phosphorus) and Zn applications of 20, 13 and 1 kg [ha.sup.-1] respectively. Limestone ammonium nitrate (28% N) was thoroughly mixed with this fertiliser blend and applied at rates of 36 and 71 kg [ha.sup.-1] to supplement N levels to 30 and 40 kg [ha.sup.-1] in the relevant treatments respectively. From the start of the experiment, the wheat cultivar Betta was planted.
Soil sampling and selection
To allow for maximum soil settling after the last cultivation, sampling was done after the rainy season just before planting in June. Soil samples were collected in June 1999, from the depth of 0-50 mm because the effects of the applied treatment combinations were concentrated in the upper 50 mm (see also Wiltshire and Du Preez 1993; Kotze 2004; Kotze and Du Preez 2007). The applied treatment combinations included the following, which represent common management practices in the Bethlehem agro-ecosystem: unburned (UB)-no-tillage (NT)-chemical weeding (CW); burned (B)-mouldboard plough (MP)-CW; UB-MP-CW; UB-stubble mulch (SM)-CW; B-MP-mechanical weeding (MW); and UB-MP-MW. The experiment was a completely randomised block design in a factorial arrangement. However, because some of the treatments that involved no-tillage and stubble mulch tillage were not representative (e.g. residue was burned and mechanical weed control method was applied) of conservation agriculture, such treatments were excluded during sampling selection.
Seven auger cores (70-mm diameter) were taken from the centre-line of each treatment plot and mixed thoroughly. Samples were dried at room temperature, sieved (<2mm) and stored in paper bags in a cool, dry and dark room free from contaminants in a basement to minimise soil reactions. Soil bulk density was measured in neither previous nor the current study, and so only concentration values of SOM indices were considered.
Analyses were carried out to determine soil C, SOC, SIC, permanganate oxidisable C (POXC), cold water extractable C (CWEC), hot water extractable C (HWEC), extractable humic substances (CEX), C in humic acids (CHa)> C in fulvic acids (CFA) and organic C functional groups. Plant biomass was not sampled in this study; however, grain yield and the harvest index, which was assumed to be 50% for rainfed wheat (G. M. Ceronio, University of the Free State, Bloemfontein, South Africa, pers. comm.), were used to estimate the amount of plant residue returned to soil after harvesting (Plant biomass = Grain yield/0.5).
Total C was analysed by dry combustion (Nelson and Sommers 1982) with a TruSpec Leco CN-2000 analyser (LECO Corp., St Joseph, MI, USA). Approximately 0.43 g of sieved air-dried encapsulated soil samples were placed in a loading head, and one-by-one dropped into a 950[degrees]C furnace and flushed with oxygen for rapid and complete combustion. Combustion gases were then passed through a secondary furnace (850[degrees]C) for further oxidation before being collected in a vessel where oxygen was injected and mixed with combustion gases. These gases were purged through a C[O.sub.2] infrared detector, which measured C as C[O.sub.2].
SOC and SIC
The SOC was measured with a modified Mebius procedure (Nelson and Sommers 1982). Of sieved air-dried soil, 0.5 g was weighed in a 150-mL glass beaker and reacted with 10 mL of 0.5 N potassium dichromate ([K.sub.2][Cr.sub.2][O.sup.7]) and 15mL of concentrated sulfuric acid (H2S04). Samples were placed on a preheated sand bath at a temperature of 130[degrees]C for 10 min. Samples were then removed from the sand bath and 35 mL of deionised water was added to each. Excess [K.sub.2][Cr.sub.2][O.sup.7] was titrated with 0.2 N ferro-ammonia sulfate until the end point was reached, which was detected by a millivoltmeter with a platinum electrode. Omission of phosphoric acid in this procedure was not considered to cause errors because samples air-dried for more than 2 days are said to contain insignificant amounts of soluble ferrous iron that can interfere with the oxidation process (Nelson and Sommers 1982). The SIC was calculated as the difference between total C and SOC.
The POXC was analysed according to Culman et al. (2012). Briefly, 2.5 g of sieved air-dried soil was weighed into 50-mL Falcon tubes, wherein 18mL of deionised water and 2mL of 0.02 M potassium permanganate solution were added. The tubes were shaken for 2 min at 240 oscillations [min.sup.-1] on an oscillating shaker and then centrifuged for 5 min at 906g for the soil to settle down. Immediately thereafter, 0.5 mL of the supernatants were transferred to new 50-mL Falcon tubes and diluted with 49.5 mL of deionised water before reading the sample absorbance on a spectrophotometer at wavelength 550 nm to obtain POXC.
CWEC and HWEC
The CWEC and HWEC were determined according to Ghani et al. (2003). In short, 3 g of air-dried soil samples were transferred to 50-mL Falcon tubes and extracted with 30 mL of deionised water. Samples were shaken for 30 min on an end-over-end shaker and centrifuged for 20 min at 1233g-, The supernatants were filtered through a 0.45-[micro]m cellulose membrane filter into separate vials for C analysis. A 30-mL aliquot of deionised water was again added to the sediments in the same tubes and shaken on a vortex shaker to suspend the sediments. The capped tubes were then left in a hot-water bath at 80[degrees]C for 16 h. Samples were again shaken on a vortex shaker and centrifuged for 20 min at 1233g. The supernatants were filtered through a 0.45-[micro]m cellulose membrane filter. The C in both extracts was determined according to the modified Mebius procedure, and 5 mL of the extracts was used instead of soil. The C obtained from the first extraction was CWEC, and that from the second was HWEC.
The sequential procedure of Schnitzer (1982) was slightly modified to extract and fractionate humic substances. Of sieved air-dried soil, 5 g was weighed into 50-mL Falcon tubes and reacted with 30 mL of extraction solution (0.1 N sodium hydroxide and 0.1 M sodium pyrophosphate decahydrate). The contents were shaken on an oscillating shaker for 1 h and centrifuged at 906g for 15 min. Insoluble material contained in the supernatant was isolated from soluble alkaline material (extractable humic substances). Soluble alkaline material was then precipitated with 0.05 N [H.sub.2]S[O.sub.4] to fractionate humic and fulvic acids. The [C.sub.EX] and [C.sub.HA] were determined with the Mebius procedure (Nelson and Sommers 1982). The [C.sub.FA] was calculated as the difference between [C.sub.EX] and [C.sub.HA] x Humification index (HI = [C.sub.HA]/SOC) and polymerisation index (PI = [C.sub.HA]/[C.sub.FA]) were calculated according to Abril et al. (2013).
Bulk soil samples were pre-treated with hydrofluoric acid (HF) to remove magnetic materials, concentrate organic C and increase the signal-to-noise ratio of the resultant NMR spectra as recommended by Skjemstad et al. (2001) and Mathers et al. (2002). Briefly, 5 g of ground and sieved air-dried soil was weighed into 50-mL Falcon tubes and 45 mL of 2% HF was added. The tubes were shaken on end-over-end shaker for 8 (4 x 1 h, 3 x 16 h and 1 x 64 h) successive times. Samples were then centrifuged after every extraction at 402g for 20 min and the supernatant filtered through a 5-[micro]m Millipore Durapore membrane filter to recover the light fraction. After the final extraction, residues together with the light fraction trapped on the membrane filter were washed five times with deionised water, oven-dried at 75[degrees]C and ground to powder using a mortar and pestle for NMR analysis.
Whole HF-treated soil samples were packed in cylindrical zirconia rotors before analysis. The NMR analysis was done on a 400-MHz Bruker AVANCE III spectrometer equipped with a 4-mm VTN multinuclear double resonance magic angle spinning probe, operating at room temperature. The [sup.13]C NMR spectra were recorded at 100.6 MHz, using the cross polarisation magic angle spinning (CPMAS) technique. A rotating speed of 14000 Hz was used with a contact time of 1 min, a recycle delay of 1 s and an acquisition time of 12.8 min. All spectra were recorded with 86016 scans. The following ranges were integrated: 0-50 ppm (alkyl C), 50-1 10ppm (O-alkyl C), 110-160 ppm (aromatic C) and 160-180 ppm (carbonyl C).
Statistical analyses were performed with IBM SPSS Statistics version 24 software package (IBM Corp., Armonk, New York). A one-way analysis of variance (ANOVA) was used and means compared with Tukey's honestly significant difference post-hoc test (HSDX). Before ANOVAs were performed, all data were tested for normality and homogeneity using Shapiro-Wilk and Levene tests respectively. Statistical analyses were performed at 95% confidence level. Pearson's correlation coefficients were calculated to assess relationships among the measured variables. For [sup.13]C NMR spectra, three replicate soil samples per treatment combination were mixed thoroughly and subjected to NMR spectroscopy as composite samples due to high costs and time needed to obtain spectra, and no statistical analyses were performed.
SOC and SIC
Applied treatment combinations had significant effects on SOC and SIC (Fig. 1). The SOC and SIC concentrations were higher (,P<0.05) in plots that received a combination of unburned straw, no-tillage and chemical weeding compared with treatments that involved mouldboard ploughing, regardless of the method of straw management and weed control. A significant (P<0.05) increase in SIC was also detected in the UB-NT-CW relative to UB-SM-CW plots. The SOC and SIC values respectively ranged from 9.68 and 0.30g [kg.sup.-1] in UB-MP-CW to 13.86 and 3.99g [kg.sup.-1] in UB-NT-CW plots.
Labile carbon fractions
Not all the three labile C fractions (e.g. HWEC) were significantly (P<0.05) affected by the applied treatment combinations (Fig. 2). The CWEC concentrations were only affected by the B-MP-CW combination, which reduced (P< 0.05) CWEC levels (616.49 vs 1081.18 mg [kg.sup.-1]) relative to UB-NT-CW. Treatment combinations that involve mouldboard ploughing whether straw was burned or left unburned and weeds controlled by herbicides or mechanically resulted in lower (P<0.05) levels of POXC compared with no-tillage combined with unburned wheat straw and chemical weeding. Relative to the treatments involving no-tillage, treatments involving mouldboard ploughing resulted in a loss of 27% POXC on average.
The [C.sub.EX] and [C.sub.FA] concentrations showed significant changes as a result of the applied treatments but CHA did not (Fig. 3). The [C.sub.EX] and [C.sub.FA] concentrations increased significantly (P<0.05) as tillage intensity decreased, with the highest values in the UB-NT-CW plots. However, the [C.sub.EX] values did not differ significantly between no-tilled, ploughed and stubble mulched plots where wheat straw was left unburned and weeds were treated with herbicides. Similarly, unburned straw and chemical weeding also did not change [C.sub.FA] significantly in no-tilled compared with stubble mulched plots. The [C.sub.EX] and [C.sub.FA] concentrations respectively varied from 8.41 and 5.83 g [kg.sup.-1] in B-MP-CW to 10.86 and 8.78 g [kg.sup.-1] in UB-NT-CW plots.
The HI and PI values did not vary much among treatments except in no-tillage combinations where both were significantly (P<0.05) lower compared with those in UB-MP-CW and B-MP-MW plots, respectively (Table 1). The HI values ranged from 0.15 in no-tilled treatments to 0.28 in the unburned ploughed plots where mechanical weeding was employed. The PI values were in the range of 0.24-0.44, with the lowest values for no-tillage combinations.
Molecular structural composition of SOC
Fig. 4 shows a solid state CPMAS [sup.13]C NMR spectra of a soil cropped annually with winter wheat; interpretations were according to Baldock et al. (1992, 1997) and Kogel-Knabner (1997, 2002). Within 0-50 ppm, resonances at 28-33 ppm are indicative of the presence of aliphatic C in long-chain polymethylene structures. Distinctive shoulders around 25 ppm in some spectra (UB-MP-MW, B-MP-MW and BMP-CW) arise from short-chain methyl C. In the 50-110ppm region, signals in the vicinity of 50-61 ppm originate from methoxyl C structures of lignin, but can also be assigned to amine C of proteins. Dominating signals at 61-75 ppm are often ascribed to oxygenated C of carbohydrate structures. Peaks at 103-105 ppm (di-O-alkyl C) are characteristic of anomeric C of polysaccharides. Broad bands within 125-130ppm confirm the presence of Cand hydrogen (H)-substituted aromatic C predominantly of lignin origin, and peaks at 169-173 ppm are typical of carboxylic, ester and amide C groups.
Distributions of [sup.13]C over chemical shift regions in the spectra are presented in Table 2. The highest alkyl C was recorded in the UB-MP-MW plots (23%), followed in a descending order by B-MP-MW (22%), UB-SM-CW (22%), UB-MP-CW (20%), UB-NT-CW (18%) and B-MP-CW plots (17%). Oxygenated carbohydrates were higher in no-tilled (55%) and stubble mulched (50%) plots due to unburned straw and chemical weeding compared with the other treatments. Aromatic C ranged from 21% in UB-NT-CW to 31% in B-MP-MW plots, and carbonyl C varied from 4% in B MP-MW to 8% in UB-MP-MW plots. The alkyl C/O-alkyl C varied slightly between treatment combinations, within 0.34-0.52; the lowest values were for UB-NT-CW and the highest for UB-MP-MW plots.
SOC and SIC
Surface accumulation of wheat straw is of much significance in no-tillage treatments as it slows the rate of residue decomposition and mineralisation processes compared with when crop residues are incorporated in the soil (Bakht et al. 2009; Chen et al. 2009). Therefore, higher SOC in the UB-NTCW plots could be attributed to a constant placement of crop residues at or near the soil surface after harvesting, and limited soil disturbance compared with mouldboard ploughed plots where the burned or unburned wheat straw was incorporated in the soil. In the tropical central region of Brazil, incorporation of crop residues by tillage was also found to accelerate residue decomposition compared with no-tillage (Carvalho et al. 2009). Intimate contact between crop residues and the soil mineral phase in the ploughed plots was also cited as the main cause.
Soil and crop residue mixing favours biological activity and diversity, and consequently increases decomposition rates and SOC losses - many studies, including the current results, have substantiated this (Carvalho et al. 2009; Clayton 2012; Moussadek et al. 2014). In contrast, results for the arid region of Cordoba Province, Argentina, showed that the effects of tillage were site-specific and the rate of decomposition depended not only on tillage intensity, but also on other management factors (Vazquez et al. 2016). In their study, SOC in cultivated soils resembled that in the virgin soils. Such unusual SOC responses in cultivated soils were associated with crop irrigation and fertilisation, which improved the phytomass returned to soil and thus masked the losses effected by continuous cultivation. Fertiliser applications also probably regulated microbial activity and slowed SOC decomposition (Gregorich et al. 1994).
Plant biomass turnover rates are usually slower in arid to semiarid environments due to warm-dry climatic conditions, with detrimental effects on SOC particularly in the absence of irrigation (Vazquez et al. 2016). As indicated in the Material and methods section, plant biomass was not sampled in this study. However, calculations from grain yield and harvest index showed that, on average, 4.00-4.36 t [ha.sup.-1] plant biomass was returned to the ploughed treatments since the inception of this trial in 1979 compared with no-tilled plots in which 3.64 t [ha.sup.-1] of wheat straw was recycled (Table 3). Lower plant biomass production in no-tillage treatments was probably due to surface nutrient stratification, which has been reported to be among the major limiting factors for nutrient uptake, especially during dry periods in a growing season (Loke 2012; Loke et al. 2013, 2014). Unlike no-tillage, ploughing distributes nutrients within the rhizosphere (beyond 0-50 mm), including less mobile P, thereby improving their accessibility to plants (Du Preez et al. 2001) and consequently biomass production (Loke 2012). In other words, data from the 0-50 mm soil layer cannot explain the overall soil fertility in this trial. Despite that, higher SOC mineralisation overrode higher SOC inputs in the ploughed plots, suggesting that long-term cultivation of arid to semiarid soils under dryland crop production had detrimental effects on SOC in the 0-50 mm soil layer (Lobe et al. 2001 ; Kotze and Du Preez 2007; Chen et al. 2009; Loke et al. 2012).
Rapid SOC decomposition increases the partial pressure of soil C[O.sub.2], which subsequently accelerates dissolution of carbonates (Eqn 1), restricting SIC re-precipitation and accrual (Wu et al. 2009; Shi et al. 2012), and this was probably the case in this study. This is confirmed by the results obtained in the eastern part of Northern China where intensive soil cultivation resulted in a substantial loss (around 51 %) of SIC compared with native grasslands (Wu et al. 2009). Exposure of calciferous layers by tillage to erosive forces was also attributed to such losses. Lobe et al. (2001, 2002) reported incidences of wind eroded silt-associated C in the cultivated soils of the Harrismith agro-ecosystem, not far from our experimental site. As such, another possible reason for lower SOC and SIC in the ploughed plots could be displacement of bare C-rich surface soil by erosion (Janzen 2006).
CaC[O.sub.3] + [H.sub.2]O + [down arrow]C[O.sub.2][right arrow][Ca.sup.2+] + 2HC[O.sup.-.sub.3] (dissolution) (1)
Results of this study are, however, inconsistent with those obtained from Russian Chernozem (Mikhailova and Post 2006). Higher SIC in the cultivated soils was presumed to be due to fertiliser and organic manure applications and released base cations ([Ca.sup.2+] and [Mg.sup.2+]) during SOC decomposition that resulted in SIC formation and accumulation as indicated by Eqns 2 and 3 (Mikhailova and Post 2006; Wu et al. 2009; Shi et al. 2012; Lal et al. 2015). In our case, there were no appreciable differences between treatments in relation to their pH values and the amounts of [Ca.sup.2+] and [Mg.sup.2+] or percentage base saturation (Table 4), and therefore did not differentially contribute to changes in SIC. However, strong positive relationships of SIC with POXC (r=0.70, P<0.01), [C.sub.EX] (r=0.63, P<0.01) and [C.sub.FA] (r= 0.72, P<0.01) indicate that these SOC fractions probably stimulated SIC re-precipitation and protected SIC through soil aggregation in no-tillage combinations (Bronick and Lal 2005; Lal et al. 2015). However, a negative correlation with PI (r = -0.52, P<0.05) (Table 5) could imply that increased decomposition in the ploughed and to some extent stubble mulched plots increased C[O.sub.2] partial pressure, and therefore constrained SIC accrual (e.g. Eqn 1, Shi et al. 2012). Regardless, dissolution (Eqn 1) or precipitation (Eqn 2) of SIC could be regarded as C sequestration mechanisms, most importantly if the consumed C[O.sub.2] is the product of decomposition-humification processes (Sanderman 2012; Ahmad et al. 2015).
[H.sub.2]O + [down arrow]C[O.sub.2][right arrow][H.sub.2]C[O.sub.3][right arrow][H.sup.+] + HC[O.sup.-.sub.3] (precipitation) (2)
2HC[O.sup.-.sub.3] + [Ca.sup.2+][right arrow]CaC[O.sub.3] + [H.sub.2]O + |C[O.sub.2] (precipitation) (3)
Labile C fractions
Although the applied treatment combinations induced changes in SOC, this property is often slow to change due to a myriad of C compounds comprising it (Janzen 2005). Vazquez et al. (2016) also indicated that soil quality depends mostly on the quantities of individual fractions other than on total SOC. Some of these C fractions are sensitive enough to monitor short- and medium-term SOC changes as a result of soil management (Gregorich et al. 1994; Janzen 2005; Chen et al. 2009; Vazquez et al. 2016). Results of this study revealed that POXC and CWEC were sensitive labile fractions that distinguished changes in SOC as a result of the applied treatment combinations.
Concentrations of POXC and CWEC are lower but consistent with those reported in other studies conducted under similar climatic conditions (Sequeira and Alley 2011; Vazquez et al. 2016). The lower quantities are probably due to site-specific conditions and the applied management. For example, in this trial, wheat has been produced under rain-fed and monoculture. but in other areas irrigation is included as part of management, fallow periods are replaced with cover crops and crop rotation is practiced, where high biomass-producing crops are rotated with legumes, which recharged the labile pool upon litter or plant fall (Sequeira and Alley 2011; Vazquez et al. 2016). Unlike legume residues, wheat straw has a high C : N ratio close to 80:1, which can translate into lower labile C fractions when added to the soil. It is also possible that 17 years of sample storage contributed to lower concentrations of labile C fractions, although biochemical reactions in air-dried soil samples have been shown to be negligible (Blake et al. 2000).
Despite their low concentrations, higher POXC and CWEC in no-tillage combinations indicate restricted intimate interactions between crop residues and mineral particles, thereby limiting accessibility of labile C compounds to the decomposer community (Chen et al. 2009). Additionally, POXC and CWEC as reservoirs of plant nutrients and energy sources for soil organisms were perhaps preserved biologically through microbial immobilisation in no-tilled treatment combinations (Baldock et al. 1992, 1997). Immobilisation of these labile fractions can lead to medium- or long-term C storage in the soil.
Increased concentrations of POXC and CWEC in the treatment combinations that included no-tillage could suggest that SOC therein was less humified and susceptible to loss if the soil was cultivated. The positive correlation between SOC and POXC (r = 0.59, P<0.05) and CWEC (r = 0.71, P<0.01) indicated linear increase in these labile fractions as SOC increased (Table 5). These correlations suggested the labile nature of SOC due to regular surface placement of crop residues in no-tillage treatments as opposed to their incorporation in the ploughed plots (Chen et al. 2009). Our results further revealed that even if equal amounts of residues were added in no-tilled and ploughed plots (Table 3), regular cultivation would always provide ideal conditions for their rapid decomposition and loss of labile C fractions until an equilibrium state was reached.
Equilibrium state of SOC is often reflected by lower concentrations of labile C fractions and accumulation of recalcitrant fractions. Partial burning of crop residues also removed most of the fragile and flammable plant parts consisting of labile organic compounds, leaving behind plant parts more resistant to microbial decomposition. An increase in [C.sub.EX] and [C.sub.FA] in no-tilled plots where wheat straw was left unburned and weeds treated with herbicides, particularly compared with ploughed plots where straw was burned and weeds controlled mechanically, indicates that exclusion of cultivation stimulated [C.sub.EX] and [C.sub.FA] formation and resulted in less humified SOC (Guimaraes et al. 2013; Seddaiu et al. 2013). The positive correlations (Table 5) between labile C fractions (POXC, CWEC and HWEC) and [C.sub.EX] and [C.sub.FA] also demonstrated that these fractions were similar in regard to their chemical composition (Sequeira and Alley 2011). The distribution of [sup.13]C in the chemical shift intervals of the NMR spectra reported by Spaccini et al. (2006) revealed that [C.sub.FA] contained higher carbohydrates relative to CHA. Moral et al. (2005) also reported strong positive relationships between [C.sub.FA] and carbohydrates, indicative that [C.sub.FA] is highly labile and a readily available energy source for soil organisms compared with [C.sub.HA] (Guimaraes et al. 2013).
On the one hand, it could be inferred that non-significant changes in CHA implied that decomposition in the ploughed combinations was not severe. If that was the case, no-tillage treatment combinations only reduced fuel-emitted C by excluding some of the tillage operations and the emptying of filled labile C reserves by tillage and less returned crop residues, particularly in the burned treatments (Janzen 2005, 2006). On the other hand, [C.sub.EX] and [C.sub.FA] were enriched with labile compounds, hence oxidative losses of humic substances began with the two before [C.sub.HA] was affected (Vazquez et al. 2016). Thus, prolonged soil cultivation destabilised the balance between humification and mineralisation of SOC by accelerating mineralisation, resulting in lower [C.sub.EX] and [C.sub.FA] (Sun et al. 2012; Guimaraes et al. 2013; Kotze et al. 2016). Although more C inputs were added in the unburned ploughed plots (Table 3), it seems C fluxes also increased relative to no-tilled treatments. In other words, equilibrium between C inputs and losses with combinations that involved no-tillage was possible, and hence could be recommended as a replacement for mouldboard ploughing in a semiarid soil that has low protection capacity for C.
In general, residue incorporation by tillage not only subjected freshly added crop remnants to rapid microbial decomposition, but also affected stable C ([C.sub.EX] and [C.sub.FA]) fractions. Non-significant changes in [C.sub.HA] across the sampled treatment combinations again denoted that there was no priming. In fact, a decline in [C.sub.EX] and [C.sub.FA] in the cultivated treatment combinations could indicate the initial stage of nutrient mining by soil microbes, which may lead to priming (mineralisation of [C.sub.HA]) over time. However, this study revealed that mean residence times or accumulation of recalcitrant C fractions were not exclusively dependent on their selective preservation, but also on microbial availability and soil management (Kogel-Knabner 2002; Kotze et al. 2016). Thus, in contrast to mouldboard ploughing, higher [C.sub.EX] and [C.sub.FA] in notillage treatments could be due to slower decomposition rates of wheat straw and higher labile C fractions (Sun et al. 2012; Guimaraes et al. 2013). Turnover rates of [C.sub.EX] and [C.sub.FA] are also more rapid than those of CHA, as such frequent supply of organic materials to soil and slow mineralisation constantly recharges the labile C of humic substances ([C.sub.EX] and [C.sub.FA]) (Moral et al. 2005; Kotze et al. 2016; Vazquez et al. 2016).
The HI and PI values, which reflected the degree of SOC decomposition and polymerisation in the sampled plots (Baldock et al. 1992; Guimaraes et al. 2013), also confirmed minimum SOC decomposition in no-tillage treatment combinations. The lower levels of HI and PI in no-tilled treatment combinations are indicative of mobile and less humified SOC compared with ploughed plots (Mathers et al. 2003; Helfrich et al. 2006; Sun et al. 2012; Guimaraes et al. 2013). In other words, constant supply and surface retention of wheat straw with no-tillage constrained rapid decomposition and humification processes, resulting in higher accrual of labile C fractions including [C.sub.EX] and [C.sub.FA] (Sun et al. 2012; Vazquez et al. 2016). Thus, levels of recalcitrant C fractions ([C.sub.HA]) in the cultivated landscapes were relative, implying that they increased as the extent of humification-polymerisation processes increased. This statement is supported by the positive correlations of [C.sub.HA] with HI (r=0.82, P< 0.01) and PI (r=0.90, P<0.01), which also indicate that [C.sub.HA] was a marker of SOC stabilisation (Table 5). The negative correlations between [C.sub.EX] and PI (r=-0.49, P<0.05) as well as between [C.sub.FA] and HI (r = -0.82, P<0.01) and PI (r = -0.90, P<0.01) further demonstrated that increased humification and polymerisation processes deprived cultivated treatments of [C.sub.EX] and [C.sub.FA] (Table 5). Despite that, all HI and PI values were less than one across the sampled treatment combinations (Table 1), implying lower aromaticity of [C.sub.HA]. higher labile C fractions and good quality SOC that can improve soil biological fertility and nutrient cycling for plant nutrition (Guimaraes et al. 2013).
Molecular structural composition of SOC
Across the sampled treatment combinations, the four SOC functional groups were in the following order: O-alkyl C>aromatic C>alkyl C>carbonyl C (Fig. 4 and Table 2). Oxygenated C and anomeric CI of cellulose and hemicellulose were the major contributors to O-alkyl C, which decreased from 55 to 43% as tillage intensity increased. This indicates that minimum soil disturbance accompanied by surface retention of crop residues for no-tillage and stubble mulch tillage increased the labile C fractions unlike for ploughed plots where unburned or burned straw was incorporated in the soil, thereby stimulating rapid crop residue decomposition and loss of O-alkyl C.
Although it is difficult to separate effects of crop residue management from tillage effects, unburned straw combined with chemical weeding seemed to have a slight influence even in the ploughed plots as they slightly increased O-alkyl C (by 1-6%) compared with unburned straw combined with mechanical weeding or burned straw combined with either chemical or mechanical weeding. This is not surprising because burning removed most combustible plant parts, which are rich in labile and easily digestible organic compounds like carbohydrates and amino acids (Kavdir et al. 2005; Knicker 2007), but mechanical weeding accelerated the fate of already decomposing residues due to mouldboard ploughing. Burning might have also reduced soil microbes and their source of energy, and therefore limited re-synthesis of O-alkyl C (Baldock et al. 1992, 1997; Knicker 2007).
Many reports have shown that during the initial phase of decomposition, O-alkyl C is the first SOC component to decrease, as it is dominated by easily oxidisable organic compounds (Mathers et al. 2003; Helfrich et al. 2006; Spaccini et al. 2006; Carvalho et al. 2009). A decrease in Oalkyl C due to SOC decomposition is often accompanied by an increase in alkyl, aromatic and carbonyl C (Baldock et al. 1992; Mathers and Xu 2003). However, the results of this study exhibited a slightly different pattern in some treatments because the lowest alkyl and carbonyl C (17 and 4% respectively) were observed in the B-MP-MW plots (Table 2). This was probably due to differences in fire intensity or the weed control method used, because in the other B-MP-CW plots, the lowest alkyl and carbonyl C accounted for 22 and 7%, respectively. Higher aromatic C in the burned treatments and lower alkyl and carbonyl C indicates that SOC was dominated by black C (Kavdir et al. 2005; Knicker 2007), but could imply in plots where wheat straw was left unburned that SOC therein was dominated by lignin units, which are also resistant to microbial degradation (KogelKnabner 1997, 2002).
Despite inconsistent trends displayed by SOC functional groups, the present results still showed that the degree of decomposition (alkyl C/0-alkyl C ratio) was lowest in the UB-NT-CW and highest in the UB-MP-MW plots (Table 2). This suggests that SOC in the latter was more humified and richer in recalcitrant aliphatic C derived from lipids, fatty acids and waxes and lower in carbohydrates (Baldock et al. 1992; Helfrich et al. 2006; Hilscher et al. 2009). Thus, soil organisms utilised carbohydrates, resulting in higher polymethylene C due to its selective preservation or in situ synthesis (Baldock et al. 1992; Kogel-Knabner 2002).
The soil samples used in this study were archived for 17 years, which may create uncertainties regarding the results. However, the SOC concentrations of the selected soil samples for this study were tested against those obtained in 1999 immediately after sampling (Kotze 2004; Kotze and Du Preez 2007) in an attempt to establish whether sample storage duration affected these soil samples. Although there were no substantial changes in SOC as a result of sample storage duration, it is possible that labile C fractions, which were not tested due to lack of such data in the previous studies, could be affected. Blake et al. (2000) used soil samples collected in 1883, 1904 and 1964 and found non-significant changes in pH, SOC and total N, which implied that once samples were air-dried, soil reactions stopped or became negligible.
Concerns regarding aging of charred C derived from partially burned crop residues remain unresolved. Recent reports have shown that fire (Knicker 2007) and aging of charred C (Rechberger et al. 2017; de la Rosa et al. 2018) altered the chemical composition of SOC. These studies show aging of charred C under incubation experiments and field conditions, and we are not aware of studies that evaluated the influence of storage duration of air-dried soil samples on the aging of charred C and associated changes. Our opinion is that changes in charred C in the archived soil samples could be minimal, unlike under field conditions where soil reactions are also governed by other factors including climatic conditions. Above all, with an increase in or presence of carboxyl C of the charred crop residues, several studies reviewed by Knicker (2007) indicate that charred C is not as microbially and chemically resistant as previously believed. Therefore, if accepted, it means the modified Mebius procedure used, solid state 13C NMR spectroscopy and subtraction of SOC from total C to determine SOC and SIC respectively, should be of less concern. However, differences in fire intensity can result in conflicting results and conclusions on charred C (Knicker 2007).
Results showed that straw management and weed control methods in the ploughed plots did not induce significant changes to the measured soil C fractions. Consequently, it is obvious that the degree of tillage was the major determining factor of the directional change (accumulation or loss) for the measured C fractions. Incorporation of unburned or burned straw by mouldboard ploughing increased decomposition rates following residue-soil contact and exposure of physically protected C fractions to microbial decomposers compared with treatment combinations that included no-tillage. An increase in [C.sub.EX] and [C.sub.FA] in no-tillage combinations implied that decomposition occurred but at a very slow rate, which could improve nutrient use efficiency. Prolonged cultivation disrupted formation of SIC and humic substances. Non-significant responses of [C.sub.HA] and lower [C.sub.EX] and [C.sub.FA] in the cultivated plots indicate an initial stage of nutrient mining, which can result in [C.sub.HA] priming over time. Lower HI and PI in the UBNT-CW plots also suggested a lower degree of decomposition. Although the functional groups of SOC were not statistically analysed, their responses were consistent with chemically fractionated SOC fractions such as O-alkyl C increasing with concomitant decreases in aromatic, alkyl and carboxyl C. Slightly lower alkyl C and O-alkyl C could also be indicative of less humified SOC in no-tillage combinations. No-tillage demonstrated potential to restore historic C and SOC quality in the 0-50 mm soil layer of this drought-prone sandy soil with lower C stabilisation capacity, and so could be an alternative to mouldboard ploughing.
Conflicts of interest
The authors declare no conflicts of interest.
This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors. However, the authors of this manuscript would like to thank the ARC-Small Grain Institute Bethlehem for the use of the site for sampling and yield data. Special thanks are also extended to ARC personnel for giving a hand during soil sampling.
Abril A, Merlo C, Noe L (2013) Realistic soil C sink estimate in dry forests of western Argentina based on humic substance content. Journal of Arid Environments 91. 113-118. doi:10.1016/j.jaridenv.2012.12.009
Ahmad W, Singh B. Dalai RC, Dijkstra FA (2015) Carbon dynamics from carbonate dissolution in Australian agricultural soils. Soil Research 53, 144-153.
Bakht J, Shafi M, Jan MT, Shah Z (2009) Influence of crop residue management, cropping and N fertilizer on soil N and C dynamics and sustainable wheat (Triticum aestivum L.) production. Soil & Tillage Research 104, 233-240. doi:10.1016/j.still.2009.02.006
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 ,3C NMR spectroscopy. Biogeochemistry 16. 1-42. doi: 10.1007/BF02402261
Baldock JA, Oades JM, Nelson PN, Skene TM, Golchin A, Clarke P (1997) Assessing the extent of decomposition of natural organic materials using solid-state 13C NMR spectroscopy. Australian Journal of Soil Research 35, 1061-1083. doi: 10.1071/S97004
Batjes NH (1996) Total carbon and nitrogen in the soils of the world. European Journal of Soil Science 47. 151-163. doi: 10.1111/j.13652389.1996.tb01386.x
Blake L, Goulding KWT. Mott CJB, Poulton PR (2000) Temporal changes in chemical properties of air-dried stored soils and their interpretation for long-term experiments. European Journal of Soil Science 51, 345-353. doi: 10.1046/j. 1365-2389.2000.00307.X
Bronick CJ, Lai R (2005) Soil structure and management: a review. Geoderma 124, 3-22. doi:10.1016/j.geoderma.2004.03.005
Carvalho AM, Bustamante MMC, Alcantara FA, Resck IS, Lemos SS (2009) Characterization by solid-state CPMAS 13C NMR spectroscopy of decomposing plant residues in conventional and no-tillage systems in Central Brazil. Soil & Tillage Research 102. 144-150. doi:10.1016/ j.still.2008.08.006
Chen H. Hou R, Gong Y, Li H, Fan M, Kuzyakov Y (2009) Effects of 11 years of conservation tillage on soil organic matter fractions in wheat monoculture in Loess Plateau of China. Soil & Tillage Research 106. 85-94. doi: 10.1016/j.still.2009.09.009
Clayton HG (2012) Long-term effects of tillage practices on biological indicators of a soil cropped annually to wheat. MSc Agrie, dissertation. University of the Free State, Bloemfontein, South Africa.
Culman SW, Snapp SS, Freeman MA, Schipasnski ME, Beniston J, Lai R, Drinkwater LE, Franzluebbers AJ, Glover JD. Grandy AS, Lee J, Six J, Maul JE, Mirksy SB, Spargo JT, Wander MM (2012) Permanganate oxidizable carbon reflects a processed soil fraction that is sensitive to management. Soil Science Society of America Journal 76, 494-504. doi: 10.2136/sssaj2011.0286
de la Rosa JM, Rosado M, Paneque M. Miller AZ, Knicker H (2018) Effects of aging under field conditions on biochar structure and composition: Implications for biochar stability in soils. The Science of the Total Environment 613-614, 969-976. doi:10.1016/j.scitotenv.2017.09.124
Derpsch R, Friedrich T (2009) Development and current status of no-till adoption in the world. In "Proceedings 18th triennial conference of the International Soil Tillage Research Organization (ISTRO)'. June 15-19, 2009. Izmir, Turkey. (CD-ROM)
Du Preez CC, Steyn JT, Kotze E (2001) Long-term effects of wheat residue management on some fertility indicators of a semi-arid Plinthosol. Soil & Tillage Research 63, 25-33. doi : 10.1016/S0167-1987(01 J00227-6
Friedrich T, Derpsch R. Kassam A (2012) Overview of the global spread of conservation agriculture. Field Actions Science Reports, Special Issue 6. http://factsreports.Revues.org /1941 (accessed 03/02/2017).
Ghani A, Dexter M, Perrott KW (2003) Hot-water extractable carbon in soils: a sensitive measurement for determining impacts of fertilization, grazing and cultivation. Soil Biology & Biochemistry 35, 1231-1243. doi: 10.1016/S0038-0717(03)00186-X
Gregorich EG, Carter MR, Angers DA, Monreal CM, Ellert BH (1994) Towards a minimum data set to assess soil organic matter quality in agricultural soils. Canadian Journal of Soil Science 74, 367-385. doi:10.4141/cjss94-051
Guimaraes DV, Gonzaga MIS. Da Silva TO, Da Silva TL. Da Silva-Dias N. Matias MIS (2013) Soil organic matter pools and carbon fractions in soil under different land uses. Soil & Tillage Research 126, 177-182. doi: 10.1016/j.still.2012.07.010
Helfrich M, Ludwig B, Buurman P. Flessa H (2006) Effect of land use on the composition of soil organic matter in density and aggregate fraction as revealed by solid-state 13C NMR spectroscopy. Geoderma 136, 331 -341. doi: 10.1016/j.geoderma.2006.03.048
Hilscher A, Heister K, Siewert C, Knicker H (2009) Mineralization and structural changes during the initial phase of microbial degradation of pyrogenic plant residues in soil. Organic Geochemistry 40, 332-342. doi: 10.1016/j.orggeochem.2008.12.004
Janzen HH (2005) Soil carbon: a measure of ecosystem response in a changing world. Canadian Journal of Soil Science 85, 467-480. doi: 10.4141/S04-081
Janzen HH (2006) The soil carbon dilemma: shall we hoard it or use it? Soil Biology & Biochemistry 38, 419-424. doi: 10.1016/j.soilbio.2005. 10.008
Jin Z, Dong Y, Wang Y, Wei X, Wang Y, Cui B. Zhou W (2014) Natural vegetation restoration is more beneficial to soil surface organic and inorganic carbon sequestration than tree plantation on the Loess Plateau of China. The Science of the Total Environment 485-486, 615-623. doi: 10.1016/j.scitotenv.2014.03.105
Kavdir Y, Ekinci H, Yiiksel O. Mermut AR (2005) Soil aggregate stability and [sup.13]C CP/MAS-NMR assessment of organic matter in soils influenced by forest wildfires in Canakkale, Turkey. Geoderma 129. 219-229. doi: 10.1016/j .geoderma.2005.01.013
Knicker H (2007) How does fire affect the nature and stability of soil organic nitrogen and carbon? A review. Biogeochemistry 85. 91-118. doi: 10.1007/sl 0533-007-9104-4
Kogel-Knabner I (1997) [sup.13]C and [sup.15]N NMR spectroscopy as a tool in soil organic matter studies. Geoderma 80. 243-270. doi: 10.1016/ S0016-7061(97 (00055-4
Kogel-Knabner I (2002) The macromolecular organic composition of plant and microbial resides as inputs to soil organic matter. Soil Biology & Biochemistry 34. 139-162. doi: 10.1016/S0038-0717(01) 00158-4
Kotze E 2004. Influence of long-term wheat residue management on some fertility indicators of an Avalon soil at Bethlehem. MSc dissertation, University of the Free State, Bloemfontein, South Africa.
Kotze E, Du Preez CC (2007) Influence of long-term wheat residue management on organic matter in an Avalon soil. South African Journal of Plant and Soil 24, 114-119. doi: 10.1080/02571862.2008. 10639890
Kotze E, Loke PF, Akhosi-Setaka MC, Du Preez CC (2016) Land use change affecting soil humic substances in three semi-arid agroecosystems in South Africa. Agriculture, Ecosystems & Environment 216, 194-202. doi: 10.1016/j.agee.2015.10.007
Lal R, Negassa W, Lorenz K (2015) Carbon sequestration in soil. Current Opinion in Environmental Sustainability 15, 79-86. doi: 10.1016/ j.eosust.2015.09.002
Land Type Survey Staff (2001) 'Land types of South Africa.' (ARC Institute for Soil Climate and Water: Pretoria)
Lobe I, Amelung W, Du Preez CC (2001) Losses of carbon and nitrogen with prolonged arable cropping from sandy soils of the South African Highveld. European Journal of Soil Science 52, 93-101. doi: 10.1046/ j. 1365-2389.2001 ,t01 -1 -00362.x
Lobe I, Du Preez CC, Amelung W (2002) Influence of prolonged arable cropping on lignin compounds in sandy soils of the South African Highveld. European Journal of Soil Science 53, 553-562. doi: 10.1046/ j. 1365-2389.2002.00469.x
Loke PF (2012) Long-term effects of residue management on soil fertility indicators, nutrient uptake and wheat grain yield. MSc dissertation, University of the Free State, Bloemfontein, South Africa.
Loke PF, Kotze E, Du Preez CC (2012) Changes in soil organic matter indices following 32 years of different wheat production management practices in semi-arid South Africa. Nutrient Cycling in Agroecosystems 94, 97-109. doi : 10.1007/s10705-012-9529-6
Loke PF, Kotze E, Du Preez CC (2013) Impact of long-term wheat production management practices on soil acidity, phosphorus and some micronutrients in a semi-arid Plinthosol. Soil Research 51, 415-426. doi: 10.1071/SR 12359
Loke PF, Kotze E, Du Preez CC (2014) Long-term effects of wheat production management practices on exchangeable base cations and cation exchange capacity of a Plinthosol in semi-arid South Africa. Communications in Soil Science and Plant Analysis 45, 1083-1105. doi: 10.1080/00103624.2013.867060
Mathers NJ, Xu Z (2003) Solid-state [sup.13]C NMR spectroscopy: characterization of soil organic matter under two contrasting residue management regimes in a 2-year-old pine plantation of subtropical Australia. Geoderma 114, 19-31. doi: 10.1016/S0016-7061 (02)00339-7
Mathers NJ, Xu Z, Berners-Price SJ, Perera MCS, Saffigna PG (2002) Hydrofluoric acid pre-treatment for improving [sup.13]C CPMAS NMR spectral quality of forest soils in south-east Queensland, Australia. Australian Journal of Soil Research 40,665-674. doi: 10.1071/SR01073
Mathers NJ, Mendham DS, O'Connell AM, Grove TS, Xu Z, Saffigna PG (2003) How does residue management impact soil organic matter composition and quality under Eucalyptus globulus plantations in Southwestern Australia? Forest Ecology and Management 179, 253-267. doi: 10.1016/S0378-1127(02)00527-3
McNeill JR, Winiwarter V (2004) Breaking the sod: Humankind, history and soil. Science 304, 1627-1629. doi: 10.1126/science. 1099893
Mikhailova EA, Post CJ (2006) Effects of land use on soil inorganic carbon stocks in the Russian Chernozem. Journal of Environmental Quality 35, 1384-1388. doi:10.2134/jeq2005.0151
Moral R, Moreno-Caselles J, Perez-Murcia MD, Perez-Espinosa A, Rufete B. Paredes C (2005) Characterisation of the organic matter pool in manures. Bioresource Technology 96, 153-158. doi:10.1016/j.biortech. 2004.05.003
Moussadek R, Mrabet R, Dahan R, Zouahri A, El Mourid M, Van Ranst E (2014) Tillage system affects soil organic carbon storage and quality in central Morocco. Applied and Environmental Soil Science 2014, doi : 10.1155/2014/654796
Nelson DW, Sommers LE (1982) Total carbon, organic carbon and organic matter. In 'Methods of soil analysis, Part 2. Chemical and microbiological properties'. (Ed. AL Page) pp. 539-579. (American Society of Agronomy: Madison. WI)
Rechberger MV, Kloss S, Rennhofer H, Tintner J, Watzinger A, Soja G, Lichtenegger H, Zehetner F (2017) Changes in biochar physical and chemical properties: accelerated biochar aging in an acidic soil. Carbon 115, 209-219. doi: 10.1016/j.carbon.2016.12.096
Sanderman J (2012) Can management induced changes in the carbonate system drive soil carbon sequestration? A review with particular focus on Australia. Agriculture, Ecosystems & Environment 155, 70-77. doi:10.1016/j.agee.2012.04.015
Schnitzer M (1982) Organic matter characterization. In 'Methods of soil analysis. Part 2. Chemical and microbiological properties'. (Ed. AL Page) pp. 581 -593. (American Society of Agronomy: Madison, WI)
Seddaiu G, Porcu G, Ledda L, Roggero PP, Agnelli A, Corti G (2013) Soil organic matter content and composition as influenced by soil management in a semi-arid Mediterranean agro-silvo-pastoral system. Agriculture, Ecosystems & Environment 167, 1-11. doi: 10.1016/j.agee. 2013.01.002
Sequeira CH. Alley MM (2011) Soil organic matter fraction as indices of soil quality changes. Soil Science Society of America Journal 75. 1766-1773. doi:10.2136/sssaj2011.0067
Shi Y, Baumann F, Ma Y, Song C, Kuhn P, Scholten T, He JS (2012) Organic and inorganic carbon in the topsoil of the Mongolian and Tibetan grasslands: pattern, control and implications. Biogeosciences 9, 2287-2299. doi:10.5194/bg-9-2287-2012
Singh B. Rengel Z (2007) The role of crop residues in improving soil fertility. In 'Soil biology: Nutrient cycling in terrestrial ecosystems'. 10th edn. (Eds P Marschner, Z Rengel) pp. 183-214. (Springer: Berlin)
Skjemstad JO, Dalai RC, Janik LJ, McGowan JA (2001) Changes in chemical nature of soil organic carbon in Vertisols under wheat in south-eastern Queensland. Australian Journal of Soil Research 39, 343-359. doi:10.1071/SR99138
Soil Survey Staff (1998) 'Key to Soil Taxonomy.' 8th edn. (Pocahontas Press: Blacksburg, VA)
Spaccini R, Mbagwu JSC, Conte P, Piccolo A (2006) Changes of humic substances characteristics from forested to cultivated soils in Ethiopia. Geoderma 132, 9-19. doi:10.l016/j.geoderma.2005.04.015
Sun CY, Liu JS, Wang Y, Zheng N, Wu XQ, Liu Q (2012) Effect of long-term cultivation on soil organic carbon fractions and metal distribution in humic and fulvic acid in black soil. Northeast China. Soil Research 50. 562-569. doi: 10.1071/SR12100
Vazquez C, Iriarte AG, Merlo C, Abril A, Kowaljow E, Meriles JM (2016) Land use impact on chemical and spectroscopical characteristics of soil organic matter in an arid ecosystem. Environmental Earth Sciences 75, 883. doi: 10.1007/s 12665-016-5655-9
Von Liitzow M, Kogel-Knabner I, Ekschmitt K. Matzner E, Guggenberger G, Marschner B. Flessa H (2006) Stabilization of organic matter in temperate soils: mechanisms and their relevance under different soil conditions. European Journal of Soil Science 57,426-445. doi: 10.1111/ j. 1365-2389.2006.00809.x
Wang Y, Li Y, Ye X, Chu Y, Wang X (2010) Profile storage of organic/ inorganic carbon in soil: From forest to desert. The Science of the Total Environment 408. 1925-1931. doi:10.1016/j.scitotenv.20I0.01.015
Wiltshire GH, Du Preez CC (1993) Long-term effects of conservation practices on the nitrogen fertility of a soil cropped annually to wheat. South African Journal of Plant and Soil 10. 70-76.
Wu H, Guo Z, Gao Q, Pen C (2009) Distribution of soil inorganic carbon storage and its changes due to agricultural land use activity in China. Agriculture, Ecosystems & Environment 129. 413-421. doi: 10.1016/ j.agee.2008.10.020
P. F. Loke (iD) (A,C), E. Kotze (A), C. C. Du Preez (A), and L. Twigge (B)
(A) Department of Soil, Crop and Climate Sciences, University of the Free State, PO Box 339, Bloemfontein 9300, South Africa,
(B) Department of Chemistry, University of the Free State, PO Box 339, Bloemfontein 9300, South Africa.
(C) Corresponding author. Email: firstname.lastname@example.org
Received 15 February 2018, accepted 21 May 2018, published online 20 August 2018
Caption: Fig. 1. Concentrations of soil organic (SOC) and inorganic (SIC) carbon as influenced by long-term wheat production management practices in the upper 0-50mm layer. Significant differences (P<0.05) are indicated by different letters for each fraction. Vertical bars with horizontal caps indicate standard deviation, unburned (UB)-notillage (NT)-chemical weeding (CW); burned (B(-mouldboard plough (MP)-CW; UB-MP-CW; UB-stubble mulch (SM)-CW; B-MP-mechanical weeding (MW); UB-MP-MW.
Caption: Fig. 2. Effects of wheat production management practices on labile carbon (C) fractions in the upper 0-50 mm soil layer. Significant differences (P< 0.05) are indicated by different letters for each fraction. Vertical bars with horizontal caps indicate standard deviation. HWEC, hot water extractable carbon; CWEC, cold water extractable carbon; POXC, permanganate oxidisable carbon; unburned (UB)-no-tillage (NT)-chemical weeding (CW); burned (B)-mouldboard plough (MP)-CW; UB-MP-CW; UB-stubble mulch (SM)-CW; B-MP-mechanical weeding (MW); UB-MP-MW.
Caption: Fig. 3. Interactive effects of wheat straw management, tillage and weed control methods on soil humic substances; [C.sub.EX], extractable humic substances, CHA, humic acids; [C.sub.FA], fulvic acids. Significant differences (P<0.05) are indicated by different letters for each fraction. Vertical bars with horizontal caps indicate standard deviation, unburned (UB)-notillage (NT)-chemical weeding (CW); burned (B)-mouldboard plough (MP)-CW; UB-MP-CW; UB-stubble mulch (SM)-CW; B-MP-mechanical weeding (MW); UB-MP-MW.
Caption: Fig. 4. Solid state CPMAS [sup.13]C NMR spectra of surface soil (0-50 mm) cropped annually with winter wheat, unburned (UB)-no-tillage (NT)-chemical weeding (CW); UB-mouldboard plough (MP)-CW; burned (B)-MP-CW; UB-stubble mulch (SM)-CW; UB-MP-mechanical weeding (MW); B-MP-MW. (1) alkyl C; (2) O-alkyl C; (3) di O-alkyl C; (4) aromatic C; (5) carbonyl C.
Table 1. Response of indices of soil organic carbon decomposition to different wheat production management systems Significant differences (P<0.05) are indicated by different letters for each index. Unburned (UB)-no-tillage (NT)-chemical weeding (CW); burned (B)-mouldboard plough (MP)-CW; UB-MP-CW; UB-stubble mulch (SM)-CW; B-MP-mechanical weeding (MW); UB-MP-MW Treatment Degree of Polymerisation combinations decomposition index Humification index UB-NT-CW 0.15 [+ or -] 0.01a 0.24 [+ or -] 0.02a B-MP-CW 0.20 [+ or -] 0.04ab 0.28 [+ or -] 0.09ab UB-MP-CW 0.28 [+ or -] 0.06b 0.39 [+ or -] 0.11ab UB-SM-CW 0.21 [+ or -] 0.05ab 0.34 [+ or -] 0.07ab B-MP-MW 0.27 [+ or -] 0.01ab 0.44 [+ or -] 0.03b UB-MP-MW 0.22 [+ or -] 0.04ab 0.31 [+ or -] 0.07ab Table 2. Relative concentrations of alkyl C, O-alkyl C, aromatic C and carbonyl C as affected by wheat production management practices in the 0-50 mm layer Unburned (UB)-no-tillage (NT)-chemical weeding (CW); burned (B)-mouldboard plough (MP)-CW; UB-MP-CW; UB-stubble mulch (SM)-CW; B-MP-mechanical weeding (MW); UB-MP-MW Treatment combinations Functional groups UB-NT-CW B-MP-CW UB-MP-CW Alkyl C (%) 18 22 20 O-alkyl C (%) 55 45 49 Aromatic C (%) C (%) 21 26 24 Carbonyl C (%) 6 7 7 Alkyl C/O-alkyl C 0.34 0.49 0.41 Treatment combinations Functional groups UB-SM-CW B-MP-MW UB-MP-MW Alkyl C (%) 22 17 23 O-alkyl C (%) 50 48 43 Aromatic C (%) C (%) 22 31 26 Carbonyl C (%) 6 4 8 Alkyl C/O-alkyl C 0.44 0.36 0.52 Table 3. Estimated quantities of plant biomass (t [ha.sup.-1]) returned to the soil after harvesting since establishment of the wheat trial near Bethlehem Unburned (UB)-no-tillage (NT)-chemical weeding (CW); burned (B)-mouldboard plough (MP)-CW; UB-MP-CW; UB-stubble mulch (SM)-CW; B-MP-mechanical weeding (MW); UB-MP-MW Treatment combinations Year UB-NT-CW B-MP-CW UB-MP-CW 1979 2.89 3.66 3.33 1981 3.93 4.29 3.57 1982 3.64 2.27 2.77 1983 2.13 2.39 4.00 1984 5.55 5.61 5.42 1985 1.55 2.10 2.12 1986 6.12 7.88 7.02 1987 3.38 5.53 4.12 1988 3.71 5.85 4.87 1991 3.39 3.70 4.49 1993 8.45 9.96 9.04 1994 0.41 0.53 0.73 1995 3.42 5.37 4.52 1996 5.09 5.32 6.40 1997 3.64 4.10 3.87 1998 0.85 1.10 0.95 1999 3.79 4.39 3.38 Mean 3.64 4.36 4.15 Treatment combinations Year UB-SM-CW B-MP-MW UB-MP-MW 1979 3.05 3.12 2.92 1981 4.23 4.58 4.82 1982 2.52 3.26 2.49 1983 4.16 2.80 3.36 1984 5.67 6.00 5.54 1985 1.95 2.91 2.50 1986 6.90 7.67 6.83 1987 4.45 5.35 4.83 1988 4.83 5.50 5.09 1991 3.10 3.30 3.65 1993 8.97 10.37 8.48 1994 0.53 0.53 0.73 1995 3.61 2.56 2.91 1996 5.44 5.88 5.70 1997 3.65 3.56 3.62 1998 1.47 1.11 1.11 1999 3.53 4.15 3.55 Mean 4.00 4.27 4.01 Table 4. Chemical soil properties of the selected samples (Kotze 2004) K, potassium; Ca, calcium; Mg, magnesium; Na, sodium; BS, base saturation; unburned (UB)-no-tillage (NT)-chemical weeding (CW); burned (B)-mouldboard plough (MP)-CW; UB-MP-CW; UB-stubble mulch (SM)-CW; B-MP-mechanical weeding (MW); UB-MP-MW Chemical properties Treatments pH K ([H.sub.2]O) (mg [kg.sup.-1]) UB-NT-CW 5.37 419 B-MP-CW 5.39 310 UB-MP-CW 5.56 267 UB-SM-CW 5.12 400 B-MP-MW 5.55 424 UB-MP-MW 5.29 360 Chemical properties Treatments Ca Mg (mg [kg.sup.-1]) (mg [kg.sup.-1]) UB-NT-CW 660 106 B-MP-CW 694 79 UB-MP-CW 709 87 UB-SM-CW 590 84 B-MP-MW 661 83 UB-MP-MW 615 88 Chemical properties Treatments Na BS (%) (mg [kg.sup.-1]) UB-NT-CW 13 84 B-MP-CW 11 93 UB-MP-CW 12 82 UB-SM-CW 11 81 B-MP-MW 12 84 UB-MP-MW 11 90 Table 5. Pearson's correlation coefficients (r) estimated among different soil C fractions ** Significant at P < 0.01 (two-tailed); * significant at P< 0.05 (two-tailed); ns, non-significant. SOC, soil organic carbon; SIC, soil inorganic carbon, [C.sub.EX], extractable humic substances; [C.sub.HA], humic acids; [C.sub.FA], fulvic acids; POXC, permanganate oxidisable carbon; CWEC, cold water extractable carbon; HWEC, hot water extractable carbon; HI, humification index; PI, polymerisation index SOC SIC [C.sub.EX] [C.sub.HA] SOC -- SIC ns -- [C.sub.EX] 0.63 ** 0.65 ** -- [C.sub.HA] ns ns ns -- [C.sub.CA] ns 0.72 ** 0.90 ** -0.50 * POXC 0.59 * 0.70 ** 0.84 ** ns CWEC ns 0.51 * 0.50 * ns HWEC 0.71 ** ns 0.48 * ns HI ns ns ns 0.82 ** PI ns -0.52 * -0.49 * 0.90 ** [C.sub.FA] POXC CWEC HWEC SOC SIC [C.sub.EX] [C.sub.HA] [C.sub.CA] -- POXC 0.87 ** -- CWEC ns 0.68 ** -- HWEC ns 0.48 * ns -- HI -0.75 ** -0.63 ** ns ns PI -0.81 ** -0.63 ** ns ns
|Printer friendly Cite/link Email Feedback|
|Author:||Loke, P.F.; Kotze, E.; Preez, C.C. Du; Twigge, L.|
|Date:||Sep 1, 2018|
|Previous Article:||Characterising preferential flow and its interaction with the soil matrix using dye tracing in the Three Gorges Reservoir Area of China.|
|Next Article:||Impact of postburn jhum agriculture on soil carbon pools in the north-eastern Himalayan region of India.|