Residual effects of cotton-based crop rotations on soil properties of irrigated Vertosols in central-western and north-western New South Wales.
Crop rotations in combination with either reduced or minimum tillage have become a feature of cotton-based (Gossypium hirsutum L.) farming systems of New South Wales (NSW) over the past 15 years (Cooper 1999; Hulugalle and Daniells 2005). Wheat (Triticum aestivum L.) is the most commonly sown rotation crop in cotton-based farming systems, although various leguminous crops such as faba bean (Vicia faba L.) and woolly pod vetch (Vicia villosa Roth.) have gained popularity in recent years (Rochester et al. 1998; Cooper 1999; Hulugalle and Daniells 2005; Rochester and Peoples 2005). The beneficial effects of sowing rotation crops are claimed to include improved soil physical, chemical, and biological quality; improved energy conservation and timeliness of land preparation; and better water conservation (Hulugalle et al. 1997, 2002; Rochester et al. 1998, 2001; Cooper 1999; Hulugalle and Daniells 2005; Rochester and Peoples 2005). Structure-related soil improvement, N recycling, and reduction in cotton seedling diseases are more likely with wheat, whereas the major benefit associated with the legumes is N fixation and, hence, lower rates of N fertiliser application rates and N losses through volatilisation and leaching (Rochester et al. 1998, 2001; Hulugalle 2005; Hulugalle and Daniells 2005; Rochester and Peoples 2005). The strong interactions which occur between tillage systems and crop rotations in irrigated cotton-based farming systems have also been noted by several authors (McKenzie et al. 1990; Constable et al. 1992; Hulugalle et al. 1997; Hulugalle and Daniells 2005; Sayere et al. 2005). However, these reported changes in soil properties were observed during periods when the cotton-rotation crop sequences were in place. Studies on the residual effects of rotation crops, that is, several years after the cotton-rotation crop sequence was discontinued, are sparse. Hulugalle et al. (1997) reported that improvements in subsoil porosity, water extraction, aggregate stability, and soil organic carbon, and reductions in sodicity, brought about by sowing a minimum-tilled cotton-wheat sequence over a 10-year period were detectible even after 5 continuous cotton crops in an irrigated Vertosol. This same crop sequence also had higher cotton lint yield and numbers of ants. Weaver et al. (2004) observed that deep drainage in a sodic Vertosol reflected past rotation history 3 years after individual rotation treatments were discontinued. These differences were thought to be due to the changes in subsoil structure brought about by the different rotation crops.
Research on the residual effects of crop rotations in other soil types, cropping systems, and geographical regions is also uncommon. Under cool, temperate conditions, residual effects of crop rotations in place for a period of 21 years were evident with respect to soil organic carbon, and wheat and potato (Solanum tuberosum L.) yields 6-7 years later (Ridgman and Wedgwood 1987). Under rainfed, semi-arid conditions, however, residual effects of crop rotations which were in place for a single cycle in a black Vertosol were reflected in sorghum [Sorghum bicolor (L.) Moench.] yield only in the year following the rotation crop (Holland and Herridge 1992). In the same region and soil type, when rotations were imposed for a 22-year period, residual effects of a lucerne (Medicago sativa L.)-cereal rotation were reflected in wheat yields for 3 years at the conclusion of the experiment, whereas residual effects of annual grain legumes such as chickpea (Cicer arietinum L.) were short-lived (Holford et al. 1998). These differences were due to differences in N accumulation under the various rotations. In a moist, subtropical environment, large yield increases in soybean [Glycine max (L.) Merr.] grain yield (68% more than continuous soybean) due to a 3-year pasture of tall fescue (Festuca arundinacea Schreb.) occurred only in the first year after the pasture phase (Edwards et al. 1990). Thereafter, soybean grain yield declined sharply such that in comparison with continuous soybean, tall fescue-soybean rotations yielded only 14% more in the second year after the pasture. It appears, therefore, that the longer a crop rotation is in place, the more likely its residual effects will be reflected in yields of subsequent crops.
In addition to yield, studies on tong-term residual effects of crop rotations most commonly report soil properties such as aggregate stability, quality, and quantity of soil organic matter and its various fractions, nitrogen balance, and ecological and crop health issues such as weed seed banks, nematode and disease incidence, and herbicide impacts (Granatstein et al. 1987; Ridgman and Wedgwood 1987; Pare et al. 1993; Holford et al. 1998; Martens 2000; Legere and Stevenson 2002; Kelley et al. 2003; Doyle et al. 2004). These studies suggest that residual effects on soil properties were apparent only when the crop rotations had been in place for many years, and the rotation crops produced large amounts of residues with moderate to high C : N ratios and phenolic acid contents (Granatstein et al. 1987; Ridgman and Wedgwood 1987; Pare et al. 1993; Holford et al. 1998; Martens 2000; Kelley et al. 2003; Doyle et al. 2004).
The literature surveyed further indicated the following. (a) Research in Vertosols was sparse with most studies being limited to the surface regions of lighter, non-expansive soils, and as noted earlier, focusing on soil organic matter, nitrogen and aggregate stability. Except for 2 reports (Hulugalle et al. 1997; Holford et al. 1998), measurements of subsoil structure, salinity, sodicity, and nutrient status were lacking from most studies. (b) Most research had been conducted under rainfed conditions, with only 2 studies (Hulugalle et al. 1997; Weaver et al. 2004) reporting results from irrigated farming systems. (c) There were no on-farm studies on the long-term residual effects of crop rotations on soil properties. The sole on-farm study on any aspect of the residual effects of crop rotations was that of Weaver et al. (2004), who measured deep drainage and nutrient and salt leaching. At the same time, while research station experiments suggest that residual effects of cotton-based crop rotations in minimum-tilled Vertosols can be apparent even after 5 continuous cotton crops (Hulugalle et al. 1997), it is uncertain whether similar results can be obtained in on-farm experiments under traffic loads, tillage systems, and irrigation management typical of commercial cotton production systems. The objective of this study, therefore, was to quantify the residual effects of cotton-based crop rotations on physical and chemical properties of Vertosols in 2 irrigated cotton farms in central and north-western NSW.
Materials and methods
Experimental sites and treatments
The experiments were located on 2 commercial cotton lamas near Warren in central-western New South Wales (147[degrees]46'E, 31[degrees]47'S) and Merah North in north-western NSW (149[degrees]18'E, 30[degrees]1'S), Australia. Both sites have a semi-arid climate and experience 4 distinct seasons with a mild winter and a hot summer. The hottest month is January (mean daily maxima and minima 33-35[degrees]C and 18-19[degrees]C, respectively) and July the coldest (mean daily maxima and minima 15-18[degrees]C and 2-3[degrees]C, respectively). Mean annual rainfall is 467mm at Warren and 595 mm at Merah North. The soils at both sites were deep, uniform grey clays. The soil at Warren was classified as a grey, self-mulching, calcareous Vertosol, medium-fine (Isbell 1996), and at Merah North an episodic-epicalcareous, self-mulching, grey Vertosol, very fine (Isbell 1996). More detailed descriptions of the soils are given in Hulugalle et al. (1998, 1999, 2002). Soil physical and chemical properties in the 0.0-0.6 m depth of both experimental sites are given in Table 1.
Crop and soil management at Warren
The rotations sown at Warren from 1993 to 1998 were: (1) continuous cotton (cotton sown every summer); (2) long-fallow cotton (cotton sown every other summer with the field maintained as a bare fallow in alternate years); (3) cotton-high input wheat, in which wheat was sown at a rate of 100kg/ha and fertilised at sowing with di-ammonium phosphate, DAP 120% P, 18% N), at a rate of 85 kg/ha and urea (43% N) at a rate of 180 kg/ha; (4) cotton-low input wheat, in which was sown at a rate of 40 kg/ha and fertilised at sowing with DAP at rate of 85 kg/ha; (5) cotton-green manured field pea (Pisum sativum L.). The rotation treatments were terminated in 1998, and a wheat (1998)-cotton (1998-99)-wheat (1999)-summer/winter fallow (2000)-cotton (2000-01) sequence was sown in all plots.
The wheat was sown at a rate of 45 kg/ha and DAP applied at sowing at a rate of 45 kg/ha. As growth of the wheat crop sown during 1998 was poor due to flooding during August, it was sprayed-out and cotton planted in October. The rotation crops were managed as rainfed crops except in 1993 when the high-input wheat received a single irrigation in September. The field peas were sprayed with glyphosate ('Roundup[TM]') in October and incorporated into the beds, and the wheat was harvested in December. The plots were fallowed during the subsequent summer and winter (December-September) and sown with cotton in October and picked in May. Prior to sowing cotton, fertiliser was applied as anhydrous ammonia (82% N) at a rate of 142 kg/ha and mono-ammonium phosphate (22% P, 10% N) at a rate of 90 kg/ha, and after sowing as water-run urea (43% N) at a rate of 195 kg/ha during December.
From 1994 onwards, cotton was sown after minimum tillage which consisted of re-forming old beds (1 m spacing, 0.15m high) and deepening furrows by direct listing with disc-hillers without any prior tillage immediately after the rotation crop phase, followed by bed cultivation to a depth of 75 mm with a Lilliston cultivator. A secondary working of the beds occurred about 6 months later and involved stalk-pulling and mulching of crop residues, middle-busting of the ridges, and re-formation of beds. The practice of renovating beds on the same spot for many years with no deep tillage is referred to as 'permanent beds' (Hulugalle and Daniells 2005). Traffic was controlled by restricting it to specific furrows and avoiding trafficking on ridges. However, some trafficking did occur on ridges during cotton-picking as axle widths of cotton-pickers (hired from external sources) differed from other farm machinery which was provided by the co-operating farmer. The plots, when sown with cotton, were furrow-irrigated with approximately 100 mm of water during each irrigation event. A comprehensive summary of crop and soil management at this site between 1993 and 1998 has been reported by Hulugalle et al. (1998, 1999). The experimental design used was a randomised complete block with 3 replications. Individual plots consisted of 40 rows (sown on beds 0.15 m high, 1 m wide, and 700 m long).
Crop and soil management at Merah North
The rotations sown after minimum tillage (permanent beds) between 1993 and 2000 at Merah North were: (1) continuous cotton; (2) long-fallow cotton; (3) cotton-green manured faba bean (Vicia faba L.) until 1999, when a sorghum crop was sown during the 1999-2000 growing season; (4) cotton-dolichos (Lablab purpureus L.)-green manured faba bean from 1993 to 1994 followed by cotton-unfertilised wheat thereafter. Sowing rates for wheat were 50 kg/ha during 1995 and 1997, and 70 kg/ha during 1999; (5) cotton-dolichos; and (6) cotton-fertilised dolichos with P and K removed by cotton replaced as fertiliser. Cropping practices (mechanised farm operations, frequent herbicide and pesticide application, N fertiliser application to cotton as anhydrous ammonia at rates between 120 and 160 kg N/ha, etc.) used in local cotton production systems were followed. In addition zinc sulfate heptahydrate and zinc oxide were applied to cotton in August 1995 and 1996 at a rate of 6 kg Zn/ha, and a commercially produced soluble Zn/P mixture at a rate of 3 L/ha in November 2004. Gypsum was applied to all plots at a rate of 2.5t/ha in July 1997 and December 1998, 7.5t/ha in September 2000, and 2 t/ha in July 2004.
From October 2000 until April 2005, a cotton (2000-01)-wheat (2001)-sorghum (2001-02)-winter fallow (2002)-cotton (2002-03)-wheat (2003)-summer and winter fallow (2003-04)-cotton (2004-05) sequence was sown in all plots to evaluate the residual effects of the previous rotations. All cereal crops in this sequence were fertilised with urea (43% N) at a rate of 151 kg/ha.
From 1993 until 1997, minimum tillage consisted of slashing of cotton followed by shallow incorporation of cotton stubble with disc-plough into the beds. The beds were renovated on the same spots shortly thereafter. Bed renovation consisted of reforming adjacent furrows and restoring the original bed shape with a disc-hiller. Additional bed cultivation to a depth of 0.05-0.07 m with a Lilliston cultivator followed by reforming took place before sowing cotton. All mechanised traffic was restricted to the furrows. From 1998 onwards, due to increasing weed numbers, the cotton was slashed and root-cut, and the beds renovated. In addition the beds were cultivated with a disc-plough to a depth of 0.10m and reformed before sowing cotton. Rotation crops and cotton were furrow-irrigated with about 100mm of water during each irrigation event. Between 2001 and 2005, both bore water and river water were used for irrigation. This resulted in occasional incursions of salt from bore water into the field. The experimental design was a randomised complete block with 3 replications. Individual plots consisted of 24 rows (sown on beds 0.15 m high and 1 m wide until May 1995, and 0.15 m high and 2 m wide thereafter) which were 431 m long.
Soil sampling and analyses
At Warren, 4 soil pits were dug in each plot with a spade, whereas at Merah North four 100-mm-diameter soil cores were extracted from each plot with a tractor-mounted soil corer in rows adjacent to non-wheel-tracked furrows using a stratified random sampling pattern (Webster and Oliver 1990). Four soil clods were taken along their natural cleavage planes from the 0-0. 15, 0.15-0.30, 0.30-0.45, and 0.45-0.60 m depths in each location to evaluate clod specific volume, an index of soil structure. Additional samples were taken at the same time from these layers for other physical and chemical analyses (see below). Soil from the 4 sampling sites was then bulked to give a composite sample for each depth in every plot. Warren was sampled during April 1999, 2000, and 2001, and Merah North during April 2001, 2002, 2003, and 2005. At Warren, the sampling zone was restricted to a 100-m-wide zone centred on the spot which was 300 m from the head-ditch. Particle size distribution and soil strength measurements prior to the experiment in 1993 had indicated that soil variability was least in this zone (Hulugalle et al. 1999).
Bulk density of oven-dry soil clods was determined after coating air-dried clods with 'saran' resin dissolved in ethyl-methyl ketone followed by oven-drying for 72 h at 110[degrees]C (Blake and Hartge 1986). Bulk density in beds (0-0.15m depth) was measured on air-dry soil aggregates (1-10 mm diameter) with the kerosene saturation method (McIntyre and Stirk 1954). As shrinkage curves had indicated that residual shrinkage of surface aggregates between air-dried and oven-dried water contents was negligible, air-dry soil water content was used to convert these values to an oven-dried equivalent. Bulk density in the surface 0.15 m was expressed as a weighted mean of bulk density evaluated from clods and aggregates in beds, with one-third weighting allocated to values derived from the clods and two-thirds to those from the aggregates. This was done on the basis of a visual assessment of the distribution of clods and loose aggregates in the 0.00-0.15 m depth. The bulk densities in all depths were used to calculate their oven-dried specific volumes (1/bulk density). At both sites, as specific volume at the 0.30-0.45 and 0.45-0.60 m depths did not differ significantly, the results were pooled. Dispersion (after immersion in water of EC 0.4 dS/m) was determined at Warren only during 2000 and at Merah North during 2000 and 2002 with a sediment density-specific gravity meter on air-dried soil aggregates of 1-4 mm diameter (Entwistle et al. 1997). Dispersion index (McKenzie and Austin 1989) was calculated as:
Dispersion index = [(wt of soil particles with ESD <20 [micro]m released due to aggregate breakdown in water)/(total wt of soil particles with ESD <20 [micro]m in sample)] x 100
where ESD is effective spherical diameter.
Air-dried soil (<2 mm) was used to determine plastic limit using a drop-cone penetrometer (Weaver and Hulugalle 2001); pH in 1:5 soil:0.01 M Ca[Cl.sub.2] suspension (Rayment and Higginson 1992); salinity as electrolytic conductivity in a 1 : 5 soil: water suspension (Rayment and Higginson 1992); nitrate-N (measured only in soil sampled from Warren) with a nitrate electrode pre-calibrated with the Kjehldal method after extraction with 0.02 M [K.sub.2]S[O.sub.4] (Keeney and Nelson 1982); and exchangeable Ca, Mg, K, and Na with an inductively coupled plasmaatomic emission spectrometer after washing with aqueous alcohol and aqueous glycerol to remove soluble salts followed by extraction with alcoholic [1.sub.M] N[H.sub.4]Cl at a pH 8.5 (Tucker 1985). Spectral lines used were 422.673 nm for [Ca.sup.2+], 285.213 nm for [Mg.sup.2+], 588.995 nm for [Na.sup.2+], and 766.491 nm for [K.sup.+]. The exchangeable cation concentrations were used to derive an index of sodicity, exchangeable sodium percentage [ESP = (exchangeable Na/[SIGMA]exchangeable cations) x 100]. Total SOC was determined by the wet oxidation method of Walkley and Black on air-dried soil <0.5 mm diameter (Rayment and Higginson 1992). SOC concentration was expressed on a volumetric basis by multiplying the values for each depth interval by its bulk density and the depth increment, followed by summing up all depth intervals. Results are presented in kg/[m.sup.2]. As the bulk density values in the 0.15-0.60 m depth were those derived from clod measurements, which are higher than those obtained using cores, values of SOC sequestered are correspondingly higher. The weighted mean of bulk density derived from aggregate and clod measurements in the 0-0.15 m depth results in values identical to those derived from core sampling (N. Hulugalle, unpublished data). All data were analysed with analysis of variance for a randomised complete block design and standard errors of the means estimated using STATISTIX 8.0 (Analytical Software 2003).
Results and discussion
Soil structural indices at Warren and Merah North
Soil specific volume in the 0-0.15 m depth at Warren did not differ significantly between ex-rotation treatments but did so between years. Mean soil specific volume ([m.sup.3]/100 Mg) in the surface 0.15m was 66.0 in 1999, 61.7 in 2000, and 67.1 in 2001 (P<0.001, s.e.m. 0.79). In the 0.15-0.30m depth at Warren, the ex-continuous cotton had the highest specific volume in 1999 and ex-cotton-LI (low input) wheat the lowest (Fig. 1). In subsequent years, soil specific volume was highest in the ex-cotton-H1 (high input) wheat rotation plots, followed by the ex-cotton-LI wheat. Lowest specific volume during 2000 and 2001 occurred in the ex-continuous cotton plots. Dispersion index at Warren during 2000 was not affected by either rotation history or depth. Mean dispersion index in the 0-0.6 m depth was 14.5 g/100 g.
[FIGURE 1 OMITTED]
Similar to Warren, rotation history did not have any effect on soil specific volume in the 0-0.15m depth at Merah North but did fluctuate between years. Mean specific volume ([m.sup.3]/100Mg) values were 62.0 in 2000, 61.7 in 2001, 56.9 in 2002, 63.0 in 2003, and 49.9 in 2005 (P < 0.001, s.e.m. 0.56). During 2000 in both the 0.15-0.30 and 0.30-0.60 m depths highest specific volume occurred where the cotton-faba bean rotation had been followed by sorghum in 1999 (Fig. 2). Thereafter, in the 0.15-0.30m depth, significant differences (P < 0.05) were observed only during 2002 and 2005 when highest values occurred in ex-cotton wheat and ex-cotton-unfertilised dolichos plots. The ex-long-fallow cotton plots had values similar to these treatments during 2002. In the 0.30-0.60 m depth, either one or both of ex-cotton wheat and ex-long-fallow cotton had higher values (P < 0.01) of soil specific volume than the other treatments (Fig. 2).
[FIGURE 2 OMITTED]
The better subsoil structure (higher soil specific volume) under the ex-cotton-wheat and ex-long-fallow cotton plots at Merah North and both ex-wheat treatments at Warren, and poorer subsoil structure (lower soil specific volume) under the ex-continuous cotton plots at Warren, reflect the differences observed during the rotation experiments (Hulugalle et al. 1998, 1999, 2002). The better subsoil structure under ex-long-fallow cotton plots may due to less intensive cropping and, hence, less intense trafficking, and less frequent irrigation and tillage, as these plots were cropped only every other year. Conversely, poorer structure under ex-continuous cotton may be due to higher cropping intensity. Improvements in subsoil structure due to sowing wheat rotation crops after cotton (Hodgson and Chan 1984; Daniells 1989: Hulme et al. 1991; Constable et al. 1992; Hulugalle et al. 2001 a; Hulugalle and Daniells 2005) and the long-term stability of such structural changes under minimum or reduced tillage (Hulugalle et al. 1997, 2001b) have been reported for a range of Australian Vertosols. The ability of wheat roots to penetrate into sodic subsoils, thereby facilitating the intensity of wetting/drying cycles, has been suggested as the cause of subsoil structural amelioration by wheat rotation crops (Hulugalle et al. 1999, 2002). Subsequent stabilisation of subsoil structure is thought to be primarily due to (a) maintenance of stable biopores by the slow-decaying wheat root systems which have relatively low C/N ratios, and (b) better subsoil structure resulting in faster leaching of excess sodium and, hence, reduction of ESP (Hulugalle et al. 1997, 2001b). Enhanced microbial activity in crack lines (macropores) created by root systems drying out a soil profile has also been suggested by Hulugalle et al. (2001b) to further stabilise subsoil structure.
In addition to the residual effects of the rotation crops, there was a sharp decrease in specific volume in all depths at Merah North during 2002, followed by a large increase in 2003. These changes parallel the fluctuations in ESP (see later discussion). Dispersion index at Merah North, measured in samples taken during 2000 and 2002, was not affected by rotation history but was least in the 0.15-0.30m depth and increased with time in all depths. Mean dispersion index in the 0-0.15 m depth was 8g/100g during 2000 and 18.5g/100g during 2002 (P<0.001, s.e.m. 0.23); 3.7 g/100 g during 2000 and 14.4 g/100 g during 2002 in the 0.15-0.30m depth (P<0.001, s.e.m. 0.61); 8.8 g/100 g during 2000 and 21.7 g/100 g during 2002 in the 0.30-0.45 m depth (P < 0.001, s.e.m. 0.42): and 8.2 g/100 g during 2000 and 20.0g/100g during 2002 in the 0.45-0.60m depth (P < 0.001, s.e.m. 0.49). These changes were significantly (P < 0.001) correlated to variation in ESP with depth and time. Correlation coefficients were 0.74 at 0-0.15m depth, 0.74 at 0.15-0.30m depth, 0.76 at 0.30-0.45 m depth, and 0.84 at 0.45-0.60 m depth.
Nitrate-N at Warren
Nitrate-N concentration at 0-0.15 and 0.15-0.30 m showed a significant interaction (P < 0.05) between rotation history and year (Fig. 3) but treatment effects were absent from other depths. In both 0-0.15 and 0.15-0.30 m depths, lowest values occurred in ex-cotton low input wheat plots. Highest values occurred in the 0.15-0.30m depth during 1999 in ex-continuous cotton and ex-cotton-field pea plots. The low input wheat crop was the only rotation crop at this site that did not receive any nutrient as either mineral fertiliser or through fixation of atmospheric nitrogen. In other treatments, nitrate-N in the ex-long-fallow cotton may have been higher than that in ex-low input cotton wheat because of soil organic matter mineralisation in fallow years, while that under continuous cotton probably remained higher due to reduced N uptake. Hulugalle et al. (1998)and Nehl et al. (2000) reported that black root rot of cotton reduced water extraction and growth of continuous cotton and they surmised that it was due to a reduction in cotton root activity. Presumably this also reduced N uptake. As cotton was sown and N fertiliser applied at rates >200kgN/ha x year (Hulugalle et al. 1998) in continuous cotton plots, reduction in N uptake may have resulted in some accumulation of nitrate-N relative to the other treatments in which cotton was not sown every year. In the 0.30-0.60 m depth, highest values were present with ex-continuous cotton and ex-cotton-high input wheat, although by 2001 differences between treatments were negligible. The possible reason for high N values in the ex-continuous cotton plots was discussed above, while N accumulation in cotton-fertilised wheat rotations has been observed at this site (Hulugalle et al. 1999) and in other studies on irrigated Vertosols (Constable et al. 1992; Hulugalle et al. 2001 a; Hulugalle 2005).
[FIGURE 3 OMITTED]
Nitrate-N concentrations generally decreased with time (Fig. 3). Mean values (In-transformed values in parentheses) in the 0.30-4).45 m depth were 10.4 (2.345)mg/kg in 1999, 13.9mg/kg (2.631) in 2000, and 5.9 (1.776)mg/kg in 2001 (P<0.001, s.e.m. of In-transformed values 0.0595); and in the 0.45-0.60m depth were 9.7 (2.276)mg/kg in 1999, 12.0mg/kg (2.488) in 2000, and 6.5 (1.869)mg/kg in 2001 (P < 0.001, s.e.m, of In-transformed values 0.0783). The high concentrations in 1999, particularly in the surface regions, may have been due to spraying out the fertilised wheat sown in June 1998 and sowing cotton in October 1998. As there was no wheat harvest in 1998, much of the applied fertiliser would have accumulated in the soil.
pH, [EC.sub.1:5], soil organic C, ESP, and plastic limit
There were no residual effects of rotation history on soil pH, [EC.sub.1:5], ESP, and plastic limit in all depths at Warren, although a significant interaction between rotation history and year occurred with respect to SOC sequestration (Fig. 4). Highest values of SOC occurred during 1999 and 2000 in the ex-cotton high input wheat and ex-long-fallow cotton plots. Except for ESP, yearly variations in pH, [EC.sub.1:5], and plastic limit occurred at Warren (Fig. 5). Mean ESP at Warren was 1.3 in the 0-0.15m depth, 2.1 in the 0.15-0.30m depth, 3.6 in the 0.30-0.45 m depth, and 6.1 in the 0.45-0.60 m depth. At the same site, there were increases in pH and [EC.sub.1:5], and decreases in SOC and plastic limit, with time. In all depths, the fall in plastic limit and increase in pH were significantly (P < 0.001) correlated with the fall in SOC. This suggests that the decline in soil organic matter can cause a deterioration of other soil physical and chemical quality indices. The decline in soil organic carbon was observed in previous studies (Hulugalle et al. 1998, 1999) and was thought to be due to the intensive cropping and irrigation practiced in this site. The decline in SOC may also partly be an indirect consequence of black root rot of cotton. By 2001, irrespective of rotation history, black root rot of cotton was widespread at high levels throughout the field at Warren (D. Nehl, unpublished data). Black root rot infestation is known to reduce above- and below-ground growth of cotton (Nehl et al. 2000). Growth reduction in cotton and, consequently, root and dry matter returned to the soil may in turn have affected soil organic matter levels. [EC.sub.1:5] also increased with time at Warren, particularly between 2000 and 2001, and may reflect changes in irrigation water quality.
[FIGURES 4-5 OMITTED]
Similar to Warren, there were no residual effects of rotation history on soil pH, [EC.sub.1:5], ESP, and plastic limit in all depths at Merah North, although significant variations occurred with time in all depths (Fig. 6). SOC sequestration also fluctuated among years, but was not affected by rotation history. Mean SOC sequestration in the 0-0.6 m depth at Merah North was 7.1 kg/[m.sup.2] during 2000, 6.9 kg/[m.sup.2] during 2001, 12.1 kg/[m.sup.2] during 2002, 8.9 kg/[m.sup.2] during 2003, and 7.5 kg/[m.sup.2] during 2005 (P<0.001, s.e.m. 0.15). Key soil chemical changes in this site were sharp increases in [EC.sub.1:5], ESP, and SOC in 2002, followed by decreases in subsequent years. The increases in [EC.sub.1:5] and ESP may have been caused by sowing a wheat crop during winter 2001, which was followed immediately by a sorghum crop during the growing season of 2001-02. Deep drainage out of the 0.6 m depth fell sharply from an average of 152 mm during the 2000-01 cotton season to 90 mm during the 2001 winter (wheat crop) and 89mm during the 2001-02 summer (sorghum crops) (T. Weaver, unpublished data). This may have resulted in salt accumulating and ESP increasing. The sharp increase in soil organic C may have been due to the amount of dry matter that was returned to the soil. Cumulative aboveground dry matter production by the 2 cereal crops was of the order of 25 t/ha, which is about 3-5 times more than that produced by a cotton crop. Felton et al. (2000) also note that C4 plants such as sorghum are more efficient in C accretion than C3 plants such as wheat. By 2005, however, SOC had fallen to values similar to those of 2001. This was probably due to the combined effects of intensive organic matter mineralisation during the irrigated cropping phases in the 2002-03 and 2004-05 summers and the crops sown between 2002 and 2005 (cotton-wheat-summer/winter fallow-cotton) returning insufficient amounts of residues to the soil. The wet conditions in irrigated soils together with the warm to hot climatic conditions which prevail in north-western New South Wales during summer result in greatly accelerated soil organic matter decomposition rates (Felton et al. 2000). Although Hulugalle (2000) concluded that major increases in on-farm carbon sequestration in Vertosols cannot be achieved by a cotton-winter cereal crop rotation, our results suggest that it may be possible to do so with a cotton-winter cereal-ummer cereal sequence where both cereal crops are irrigated and fertilised to maximise crop growth.
[FIGURE 6 OMITTED]
In summary, residual effects of cotton-rotation crops sequences sown between 1993 and 1999 were limited to soil specific volume in both sites, and nitrate-N and SOC at Warren. In general, higher values of specific volume, nitrate-N, and organic C occurred where cotton-wheat rotations and, in particular, fertilised wheat had been sown. In addition, at Merah North subsoil structure in ex-long-fallow cotton was similar to that in the cotton-wheat rotation. Lowest subsoil specific volume was present in ex-continuous cotton plots, and to some extent in ex-cotton-legume plots. These differences mirror those present when the rotation treatments were in place (Hulugalle et al. 1999, 2002). Holford et al. (1998) studied the residual effects of various crop rotations on wheat yields, soil N, and SOC in a rainfed black Vertosol, and similarly reported that the residual effects of grain legume rotation crops on these soil properties and wheat yields were short-lived.
There were no discernible effects of rotation history on the other soil properties measured in this study. This differs from on-station studies, which indicate that there were strong residual effects of past crop rotations on soil chemical properties for several years after cessation of the rotation treatments (Ridgman and Wedgwood 1987; Hulugalle et al. 1997; Holford et al. 1998). Typically, medium-large-scale commercial cotton farms in New South Wales use heavier and larger farm machinery, apply soil amendments such as gypsum more frequently, and irrigate more frequently than government research farms. These differences may have contributed to the limited residual effects of rotation history on many of the soil properties measured in the present study. Our results also suggest that regular sowing of rotation crops is necessary to maintain soil quality in cotton-based farming systems. The commonly used practice of sowing a cotton-cotton-rotation crop sequence (Cooper 1999) may not be sufficient to maintain soil quality in irrigated Vertosols. It is notable that soil organic carbon, a widely accepted measure of soil quality, increased only where a cotton-cereal--cereal sequence was sown.
Residual effects of rotation crops sown after cotton were evident only with respect to soil specific volume, a measure of soil structure, and to a lesser extent nitrate-N and SOC. Best structure occurred where ex-cotton-wheat rotations had been sown. In a heavy clay soil at Merah North (average clay content of 64g/100g in the surface 0.6m) where high ESP values occurred at relatively shallow depths, structure under ex-long-fallow cotton plots was similar to ex-cotton-wheat. At Warren, in a medium clay soil (average clay content of 53 g/100g in the surface 0.6 m) with low to moderate values of ESP, greater SOC sequestration occurred in the ex-cotton-high input wheat rotation and ex-long fallow cotton, and higher values of nitrate-N were present in the ex-cotton-high input wheat rotation. Residual effects of crop rotations are more likely to occur where the residues of the rotation crops are relatively recalcitrant or where cropping intensity is lower.
Funding for this research was provided by the Australian Cotton Co-operative Research Centre and the Australian Cotton Research and Development Corporation. C. Hogendyke ('Auscott', Warren, NSW), and J. and D. Grellman ('Beechworth', Merah North, NSW), their families and staff are thanked for provision of land to conduct the trials, management expertise, and continuing support and interest.
Manuscript received 14 September 2005, accepted 30 March 2006
Analytical Software (2003) 'STATISTIX[R] 8: User's Manual.' (Analytical Software: Tallahassee, FL)
Blake GR, Hartge KH (1986) Bulk density. In 'Methods of soil analysis, Part 1: Physical and mineralogical methods'. (Ed. A Klute) pp. 363-382. (ASA: Madison, WI)
Constable GA, Rochester IJ, Daniells IG (1992) Cotton yield and nitrogen requirement is modified by crop rotation and tillage method. Soil and Tillage Research 23, 41-59. doi: 10.1016/0167-1987(92)90004-U
Cooper JL (1999) A grower survey of rotations used in the New South Wales cotton industry. Australian Journal of Experimental Agriculture 39, 743-755. doi: 10.1071/EA98055
Daniells IG (1989) Degradation and restoration of soil structure in a cracking grey clay used for cotton production. Australian Journal of Soil Research 27, 455-469. doi: 10.1071/SR9890455
Doyle GL, Rice CW, Peterson DE, Steichen J (2004) Biologically defined soil organic matter pools as affected by rotation and tillage. Environmental Management 33(Suppl 1), S528-S538. doi: 10.1007/s00267-003-9160-z
Edwards JH, Thurlow DL, Holliman JL, Pegues MD (1990) Influence of cover crop, perennial sod, and crop rotation on soybean growth and yield. In 'Conservation tillage for agriculture in the 1990s'. NCSU Special Bulletin 90-1. (Eds JP Mueller, MG Wagger) pp. 90-93. (North Carolina State University: Raleigh, NC)
Entwistle PC, Hulugalle NR, Finlay LA (1997) Measuring dispersion with an electronic sediment density meter. In "Moving towards precision with soil and plant analysis'. (Eds LA Sparrow, AJ Fist) pp. 140-142. (Australian Soil and Plant Analysis Council: Launceston, Tas.)
Felton W, Schwenke G, Martin R, Fisher J (2000) Changes in carbon in summer rainfall cropping sytems. In 'Management options for carbon sequestration in forest, agricultural and rangeland ecosystems'. (Eds R Keenan, AL Bugg, H Ainslie) pp. 32-37. (CRC for Greenhouse Accounting: Canberra)
Granatstein DM, Bezdicek DF, Cochran VL, Elliott LF, Hammel J (1987) Long-term tillage and rotation effects on soil microbial biomass, carbon and nitrogen. Biology and Fertility of Soils 5, 265-270. doi: 10.1007/BF00256912
Hodgson AS, Chan KY (1984) Deep moisture extraction and crack formation by wheat and safflower in a Vertisol following irrigated cotton rotations. In 'The properties and utilization of cracking clay soils'. (Eds JW McGarity, EH Hoult, HB So) pp. 299-304. (University of New England Press: Armidale, NSW)
Holford ICR, Schweitzer BE, Crocker GJ (1998) Comparative effects of subterranean clover, medic, lucerne, and chickpea in wheat rotations, on nitrogen, organic carbon, and moisture in two contrasting soils. Australian Journal of Soil Research 36, 57-72. doi: 10.1071/S97036
Holland JF, Herridge DF (1992) Production of summer crops in northern New South Wales. II. Effects of tillage and crop rotation on yields of sorghum. Australian Journal of Agricultural Research 43, 123-134. doi: 10.1071/AR9920123
Hulme P J, McKenzie DC, Abbott TS, MacLeod DA (1991) Changes in the physical properties of a Vertisol following an irrigation of cotton as influenced by the previous crop. Australian Journal of Soil Research 29, 425-442. doi: 10.1071/SR9910425
Hulugalle NR (2000) Carbon sequestration in irrigated vertisols under cotton-based farming systems. Communications in Soil Science and Plant Analysis 31,645-654.
Hulugalle NR (2005) Recovering leached N by sowing wheat after irrigated cotton in a Vertisol. Journal of Sustainable Agriculture 27, (In press).
Hulugalle NR, Daniells IG (2005) Permanent beds in Australian cotton production systems. In 'Evaluation and performance of permanent raised bed cropping systems in Asia, Australia and Mexico. Proceedings of Workshop'. Griffith, NSW, 1-3 March 2005, ACIAR Proceedings No. 121. (Eds CH Roth, RA Fisher, CA Meisner) pp. 162-171. (ACIAR: Canberra)
Hulugalle NR, Entwistle PC, Cooper JL, Allen S J, Nehl DB (1998) Effect of long-fallow on soil quality and cotton lint yield in an irrigated, self-mulching, grey Vertosol in the central-west of New South Wales. Australian Journal of Soil Research 36, 621-639. doi: 10.1071/S97111
Hulugalle NR, Entwistle PC, Cooper JL, Scott F, Nehl DB, Allen S J, Finlay LA (1999) Sowing wheat or field pea as rotation crops after irrigated cotton in a grey Vertosol. Australian Journal of Soil Research 37, 867-889. doi: 10.1071/SR98109
Hulugalle NR, Entwistle PC, Scott F, Kahi J (2001a) Rotation crops for irrigated cotton in a medium-fine, self-mulching, grey Vertosol. Australian Journal of Soil Research 39, 317-328. doi: 10.1071/SR00035
Hulugalle NR, Entwistle PC, Weaver TB, Scott F, Finlay LA (2002) Cotton-based rotation systems on a sodic Vertosol under irrigation: effects on soil quality and profitability. Australian Journal of Experimental Agriculture 42, 341-349. doi: 10.1071/EA00118
Hulugalle NR, Lobry de Bruyn LA, Entwistle P (1997) Residual effects of tillage and crop rotation on soil properties, soil invertebrate numbers and nutrient uptake in an irrigated Vertisol sown to cotton. Applied Soil Ecology 7, 11-30. doi: 10.1016/S0929-1393(97)00027-9
Hulugalle NR, Weaver TB, Finlay LA, Entwistle PC (2001b) Physical and chemical properties of soil neareracks in irrigated Vertisols sown with cotton-wheat rotations. Arid Land Research and Management 15, 13-22. doi: 10.1080/153249801300000789
Isbell RF (1996) 'The Australian Soil Classification.' (CSIRO Publishing: Melbourne)
Keeney DR, Nelson DW (1982) Nitrogen--inorganic forms. In 'Methods of soil analysis, Part 2: Chemical and microbiological properties'. (Eds AL Page, RH Miller, DR Keeney) pp. 643-698. (ASA and SSSA: Madison, WI)
Kelley KW, Long JH Jr, Todd TC (2003) Long-term crop rotations affect soybean yield, seed weight, and soil chemical properties. Field Crops Research 83, 41-50. doi: 10.1016/S0378-4290(03)00055-8
Legere A, Stevenson FC (2002) Residual effects of crop rotation and weed management on a wheat test crop and weeds. Weed Science 50, 101-111.
Martens DA (2000) Management and crop residue influence soil aggregate stability. Journal of Environmental Quality 29, 723-727.
McIntyre DS, Stirk GB (1954) A method for determination of apparent density of soil aggregates. Australian Journal of Agricultural Research 5, 291-296. doi: 10.1071/AR9540291
McKenzie DC, Abbott TS, Anthony DTW, Hulme P J, MacLeod DA, Higginson FR (1990) Management of subsoil degradation in an Australian Vertisol used for irrigated cotton production. In 'Transactions of the 14th International Congress of Soil Science'. Vol. VI, pp. 176-181. (International Society of Soil Science: Kyoto)
McKenzie NJ, Austin MP (1989) Utility of the Factual Key and Soil Taxonomy in the Lower Valley, N.S.W. Australian Journal of Soil Research 27, 289-311. doi: 10.1071/SR9890289
Nehl DB, Mondal AH, Allen SJ (2000) Managing black root. In 'Proceedings of the 10th Australian Cotton Conference'. 16-18 August 2000, Brisbane, Qld. pp. 301-307. (Australian Cotton Grower's Research Association: Orange, NSW)
Pare TC, Chalifour FP, Bourassa J, Antoun H (1993) Residual effects of faba bean and soybean for a second or third succeeding forage-corn production. Canadian Journal of Plant Science 73, 495-507.
Rayment GE, Higginson FR (1992) 'Australian laboratory handbook of soil and water methods.' (Inkata: Melbourne and Sydney)
Ridgman WJ, Wedgwood RB (1987) Some residual effects of crop rotation on light land. Journal of Agricultural Science 108, 189-193.
Rochester I, Peoples M (2005) Growing vetches (Vicia villosa Roth.) in irrigated cotton systems: inputs of fixed N, N fertiliser savings and cotton productivity. Plant and Soil 271, 251-264. doi: 10.1007/s11104-004-2621-1
Rochester IJ, Peoples MB, Constable GA, Gault RR (1998) Faba beans and other legumes add nitrogen to irrigated cotton cropping systems. Australian Journal of Experimental Agriculture 38, 253-260. doi: 10.1071/EA97132
Rochester IJ, Peoples MB, Hulugalle NR, Gault RR, Constable GA (2001) Using legumes to enhance nitrogen fertility and improve soil condition in cotton cropping systems. Field Crops Research 70, 27-41. doi: 10.1016/S0378-4290(00)00151-9
Sayere K, Limon A, Govaerts B (2005) Experiences with permanent bed planting systems CIMMYT/Mexico. In 'Evaluation and performance of permanent raised bed cropping systems in Asia, Australia and Mexico. Proceedings of Workshop'. Griffith, NSW, 1-3 March 2005, ACIAR Proceedings No. 121. (Eds CH Roth, RA Fisher, CA Meisner) pp. 12-25. (ACIAR: Canberra)
Tucker BM (1985) Laboratory procedures for soluble salts and exchangeable cations in soils. CSIRO Australia Division of Soils Technical Paper No. 47.
Weaver TB, Hulugalle NR (2001) Evaluating plastic limit in Vertisols with a drop-cone penetrometer. Communications in Soil Science and Plant Analysis 32, 1457-1464. doi: 10.1081/CSS-100104205
Weaver TB, Hulugalle NR, Ghadiri H (2004) Salt, nutrient and pesticide leaching under a sodic Vertisol irrigated with groundwater in north-west New South Wales. In 'Conserving soil and water for society: sharing solutions. Proceedings of the 13th International Soil Conservation Organisation Conference'. 4-9 July, Brisbane, Qld. (Eds SR Raine, AJW Biggs, NW Menzies, DM Freebairn, PE Tolmie) (ASSSI: Warragul/IECA (Australasia): Picton) [CD-ROM]
Webster R, Oliver MA (1990) 'Statistical methods in soil and land resource survey.' (Oxford University Press: Oxford)
N. R. Hulugalle (A,B), T. B. Weaver (A), and L. A. Finlay (A)
(A) NSW Department of Primary Industries/Australian Cotton Catchment Communities CRC, Australian Cotton Research Institute, Locked Bag 1000, Narrabri, NSW 2390, Australia.
(B) Corresponding author. Email: firstname.lastname@example.org
Table 1. Soil properties in 0.00-15, 0.15-0.30, 0.30-0.45, and 0.45-0.60m depths in June 1993 at Warren and May 1994 at Merah North [EC.sub.1:5], electrolytic conductivity (previously known as electrical conductivity); ESP, exchangeable sodium percentage Warren Soil property 0.00-0.15 0.15-0.30 0.30-0.45 0.45-0.60 Depth (m): Sand (g/100g) 30 34 34 32 Silt (g/100g) 18 14 12 15 Clay (g/100g) 52 52 54 53 Organic C (g/100g) 0.53 0.60 0.57 0.52 pH (0.01 M Ca[Cl.sub.2]) 7.7 7.6 7.6 7.6 [EC.sub.1:5] (dS/m) 0.3 0.3 0.4 0.4 Exchang. cations ([cmol.sub.c]/kg) Ca 21 21 22 19 Mg 13 13 14 14 K 1.0 0.9 1.0 0.7 Na 0.7 1.0 1.5 2.1 ESP 2 3 4 6 Merah North Soil property 0.00-0.15 0.15-0.30 0.30-0.45 0.45-0.60 Depth (m): Sand (g/100g) 22 20 16 16 Silt (g/100g) 17 16 20 18 Clay (g/100g) 61 64 64 66 Organic C (g/100g) 0.78 0.76 0.61 0.47 pH (0.01 M Ca[Cl.sub.2]) 6.8 6.7 7.2 7.2 [EC.sub.1:5] (dS/m) 0.3 0.3 0.4 0.5 Exchang. cations ([cmol.sub.c]/kg) Ca 19 20 21 19 Mg 16 18 18 18 K 1.1 1.1 0.7 0.7 Na 2.9 3.8 6.1 7.3 ESP 8 9 14 16
|Printer friendly Cite/link Email Feedback|
|Author:||Hulugalle, N.R.; Weaver, T.B.; Finlay, L.A.|
|Publication:||Australian Journal of Soil Research|
|Date:||Aug 1, 2006|
|Previous Article:||The influence of season, agricultural management, and soil properties on gross nitrogen transformations and bacterial community structure.|
|Next Article:||The hydrology of Vertosols used for cotton production: II. Pedotransfer functions to predict hydraulic properties.|