Sulfur mineralisation in a coarse-textured soil after different sulfate fertilisation histories, and yield responses of wheat and lupin.
The use on crops and pastures of high analysis phosphorus and nitrogenous fertilisers in place of low analysis fertilisers containing sulfate (S[O.sub.4.sup.2-]) has substantially changed the amounts of sulfur (S) added to agricultural soils in southern Australia. Current S[O.sub.4.sup.2-] fertiliser inputs (1-10 kg S/ha), chiefly at seeding, are believed to have a neutral to a moderately positive effect on S balance (National Land and Water Resources Audit 2001). However, S deficiency is widespread in some regions of the Western Australian wheatbelt where sandy soils dominate (Robson et al. 1995). It is possible that simple nutrient budgeting methods underestimate S needs, warranting a closer examination of the capacity of sandy soils to sustain S supply with reduced inputs of S[O.sub.4.sup.2-] fertiliser.
The bulk of S in soil is present in organic forms that cannot be utilised by plants. Scherer (2001) provides an overview of soil S transformations, describes the diverse forms of organic S in soil and outlines the processes linked to sulfur mineralisation that release S[O.sub.4.sup.2-] from soil organic matter (SOM). Mineralisation processes are either enzymatic or biochemical, or microbially mediated depending on the form of organic S. Rates of S mineralisation are thought to be dependent on SOM content and on the C : S ratio of this material (Tabatabai and Al-Khafaji 1980), with net S[O.sub.4.sup.2-] mineralisation apparent when the C : S ratio is <200, and immobilisation apparent for C : S ratios >400 (Scherer 2001). Temperature and soil water content are other key environmental factors affecting S mineralisation.
Most studies on S mineralisation have involved laboratory experimentation. There is limited information on the rates of S mineralisation in field soils. Eriksen et al. (1995) report net S[O.sub.4.sup.2-] mineralisation rates in sandy soils in Denmark to range between 3.3 and 6.7 [micro]g S/g soil.year. These rates of S mineralisation were unable to maintain S[O.sub.4.sup.2-] supply at levels needed to prevent S deficiency in rye grass, barley, and canola. The difficulties encountered in measuring net S[O.sub.4.sup.2-] mineralisation in soil have encouraged the testing of extractants that aim to release S[O.sub.4.sup.2-] from labile soil organic S pools considered to be mineralised during crop growth. The KCl-40 extraction technique introduced by Blair et al. (1991) gives satisfactory predictions of net S[O.sub.4.sup.2-] mineralisation in pasture systems but has not been tested for crop systems grown on sandy soils.
Leaching of S[O.sub.4.sup.2-] below the rooting depth of crops can be an important S loss mechanism. The amount of S[O.sub.4.sup.2-] leached is affected by S[O.sub.4.sup.2-] fertiliser application (Heng et al. 1991), S[O.sub.4.sup.2-] immobilisation (Goh and Gregg 1982), amount of drainage of soil water drainage (Heng et al. 1991), the S[O.sub.4.sup.2-] status of soil (Sakadevan et al. 1994), and the capacity of the soil to adsorb S[O.sub.4.sup.2-] (Bolan et al. 1986; Lefroy et al. 1995). The amounts of S[O.sub.4.sup.2-] leached from soils of the Western Australian wheatbelt are not known.
The aims of this paper are (1) to determine the effect of S application on grain yield in wheat (Triticum aestivum), narrow-leafed lupin (Lupinus angustifolius), and canola (Brassica napus L.); (2) to examine the residual value of applied S[O.sub.4.sup.2-] fertiliser in a sandy soil by determining the amounts of S[O.sub.4.sup.2-] in soil and rates of net S[O.sub.4.sup.2-] mineralisation; (3) to ascertain potential for S[O.sub.4.sup.2-] leaching; and (4) to examine the likely occurrence of S deficiency by determining the relationship between soil S[O.sub.4.sup.2-] supply and plant S[O.sub.4.sup.2-] demand.
Material and methods
The experimental site was located near Konnongorring, 20 km south of Wongan Hills (30[degrees]50'S, 116[degrees]38'E), in Western Australia. Annual rainfall for the area ranges between 264 mm (decile 1) and 521 mm (decile 9), with a long-term average of 386 mm. The soil was classified as grey shallow sandy duplex or Sodosol by Sehokneeht (1997). The site was cropped to wheat in 1989, lupin in 1990, and wheat in 1991. Fertiliser use over this period included 50 kg/ha of diammonium phosphate (DAP) and 60 kg urea/ha in 1989, 75 kg/ha of triple super phosphate (TSP) in 1990, and 50 kg DAP/ha and 50 kg/ha of(N[H.sub.4])[sub.2] S[O.sub.4] in 1991. From 1992 to 1995 the site was used by Wesfarmers CSBP to investigate crop responses to rates and forms of S fertiliser. Sulfur was applied each year as gypsum at 0 ([S.sub.0]), 5 ([S.sub.5]), 15 ([S.sub.15]), and 30 ([S.sub.30]) kg S/ha. In order for the residual value of the S[O.sub.4.sup.2-]fertiliser treatments to be studied, no additional S[O.sub.4.sup.2-] fertiliser inputs were made to these treatments when wheat was grown in 1996, or when lupin was grown in 1997. Gypsum was applied in both years at 15 kg S/ha ([SR.sub.15]) to plots that had not received S[O.sub.4.sup.2-] fertiliser in 1993 and 1994 to evaluate the grain yield response to current season applied S.
Each experimental plot was 2.1 m wide and 40 m long. Treatments were originally replicated 3 times in a randomised block design. However, in 1996 one replicate was damaged by spray drift from an adjacent paddock and the plots in the 2 unaffected replicates were divided in half to give 4 replicates.
Soil bulk density was measured in layers to a depth of 0.5 m using pits dug at 4 locations across the experimental site. Soil recovered in stainless steel rings, each 50 mm in diameter and 100 mm long, was weighed then oven-dried at 105[degrees]C and reweighed. Particle size analysis was conducted by the procedure of Day (1965). The methods used
for measurements of N[O.sub.3.sup.-] and N[H.sub.4.sup.+], bicarbonate-extractable P and K, KCl-40 extractable S, total organic carbon, 0.01 M Ca[Cl.sub.2] soil pH, and oxalate-extractable iron are outlined in Rayment and Higginson (1992). Total P and S were measured by the procedures outlined by Chen et al. (1999).
Sulfate adsorption by soil was determined by weighing 3 g of air-dried soil into a 50-mL centrifuge tube to which was added 15 mL of 0.02 M Ca[Cl.sub.2] plus 2 drops of chloroform (Lisle et al. 1991). Different amounts (0-15 mL) of 50 [micro]g S/mL [K.sub.2]S[O.sub.4] solution were added to give a range of added S[O.sub.4.sup.2-] between 0 and 250 [micro]g S/g soil. The concentration of K was balanced by the addition of 0-15 mL of 0.003 M KCl. The final volume of the extracting solution was 30 mL. Centrifuge tubes were then tumble-shaken for 48 h at 25[degrees]C. The pH of the soil solution was checked at 6, 24, and 30 h, and if necessary, adjusted to the original soil pH (0.01 M Ca[Cl.sub.2]). Each tube was centrifuged at 3000 r.p.m, for 5 min and the supernatant filtered using Whatman No 42 filter paper. The SS[O.sub.4.sup.2-] concentration in the supernatant was measured using a Waters high-pressure liquid chromatograph (HPLC). Sulfate adsorption was determined as the difference between S[O.sub.4.sup.2-] removed from solution following the addition of 0.01 M Ca[Cl.sub.2] containing [K.sub.2]S[O.sub.4] at 50, 100, 150, 200, and 250 [micro]g S/g soil, and S[O.sub.4.sup.2-] in solution when soil was extracted with 0.01 M Ca[Cl.sub.2] without added S[O.sub.4.sup.2-].
Rainfall data for the period 1 January 1992-28 May 1996 were from daily readings taken at the farmhouse located about 500 m from the research site. After 28 May 1996, weather data including rainfall, maximum and minimum air temperature, and soil temperature at 10, 50, and 100 mm were collected using an automatic meteorological station placed at the site.
The ASPIM model configured with the SOILWAT2 module (Probert et al. 1998) was used to predict drainage from 1992 to 1997. Soil water parameters used to initialise the model were taken from Asseng et al. (2001) for a sand over clay or duplex soil, and were started on 1 April of each year using soil water contents set at the plant-available lower limit.
Sulfur input in rainfall was estimated from data published by Hingston and Gailitis (1976) for sites in the central wheatbelt of Western Australia.
The cropping sequence was lupin in 1992, wheat in 1993, canola in 1994, wheat in 1995, wheat in 1996, and lupin in 1997. The cultivars used, seeding dates, seeding rates, and N inputs are given in Table 1. All crops were sown using a minimum tillage (narrow points) cultivation system. Weed management was achieved by the use of herbicides. The herbicides used included Roundup[R] (glyphosate) at 1 L/ha or Sprayseed[R] (paraquat and diquat) at 1.5 L/ha applied before sowing. Ryegrass control was achieved by using triasulfuron (35 g/ha) at seeding of wheat in 1993, fluazifop-P (0.4 L/ha) on 20 July 1994 on canola, and triflumlin (1 L/ha) at seeding of wheat in 1996. Wheat seed was treated with Baytan[R] (triadimenol and cypermethrin) and lupin was treated with Royal[R] (iprodione). Control of redlegged earth mites in canola was achieved by the application of omethoate (0.06 L/ha) at seeding and on 20 July 1994. Phosphate was applied at the rate of 13 kg P/ha in the form of TSP in all years. This rate of application resulted in an input of 0.6 kg S/ha across all treatments. Potassium was applied as muriate of potash at 50 kg K/ha on 25 August 1994 and on 7 June 1996.
Machine harvest of grain was undertaken after maturity between 1992 and 1995. In 1996 and 1997 crop biomass (t/ha) was manually collected in July, August, September, and December in all treatments except the re-applied S[O.sub.4.sup.2-] fertiliser treatment, which was only measured at the grain harvest. At each manual sampling, plants were collected from a 0.5-m length of row at 4 randomly chosen points in each plot. Plant material was bulked to obtain a single plot weight. Lupin and wheat material was separated into stem and grain at final harvest. Wheat and lupin grain yield (t/ha) was also obtained from a machine harvest of each plot in 1996 and 1997.
Plant material was dried at 70[degrees]C, weighed, ground, digested using nitric/perchloric acid mixture (Johnson and Ulrich 1959), and analysed for S using an inductively coupled plasma-atomic emission spectroscopy (ICP-AES, Matilainen and Tummavuori 1996).
Soil profile S[O.sub.4.sup.2-] content
Soil was collected from all plots in layers to 0.5 m in 1996 and 1997 to determine the S[O.sub.4.sup.2-] content within the rooting zone of lupin and wheat. The top 0.1 m of soil was collected in 0.05-m increments from within a 110-mm-diameter stainless steel ring. A 100-mm-diameter sand auger was then inserted through the stainless steel ring to collect soil in 0.1-m increments to 0.5 m. A further 0.2-m depth of soil sample was collected at some plots but the presence of a silcrete layer between 0.6 and 0.7 m in sections of the trial did not permit uniform sampling to 0.7 m.
Soil samples to 0.2 m were used in the assessment of net S[O.sub.4.sup.2-] mineralisation. These were placed in chilled polystyrene containers immediately after collection then transported to Perth within a day of collection where they were stored at 4[degrees]C for a maximum of 2 days. All soil samples were weighed and thoroughly mixed before a subsample was taken. Soil subsamples were then stored at -18[degrees]C until analysed. Soil samples collected below 0.2 m were air-dried in a glasshouse, then mixed thoroughly, sieved (2 mm), and subsampled.
Soil S[O.sub.4.sup.2-] in the top 0.2 m was extracted by shaking 50 g of moist soil with 100 mL of deionised [H.sub.2]O on a rotating shaker for 2 h at 25[degrees]C. Gravimetric water content was determined on separate subsamples. Water extracts were filtered through Whatman No. 1 paper and the filtrate was frozen until analysed for S[O.sub.4.sup.2-] using a Waters HPLC. Soil in layers below 0.2 m was analysed for S using the KCl-40 method of Blair et al. (1991). A comparison was made between water extraction and the KCl-40 method using selected samples from the 0.2-0.5 m soil layers to confirm that both methods extracted the same amount of S[O.sub.4.sup.2-] (data not presented). The amount of S[O.sub.4.sup.2-] in each soil layer (kg S/ha) was calculated from the measured concentration of S[O.sub.4.sup.2-] ([micro]g S/g soil) in the layer, the bulk density of soil in the layer (g soil/[cm.sub.3]), and the depth (cm) of each soil sampled with units expressed as kg/ha. Profile S[O.sub.4.sup.2-] content is the sum of S[O.sub.4.sup.2-] (kg S/ha) calculated for each soil layer sampled.
Net S[O.sub.4.sup.2-] mineralisation
Measurements of net S[O.sub.4.sup.2-] mineralisation were undertaken between 17 June 1996 and 24 November 1997 using the 5 residual S[O.sub.4.sup.2-] fertiliser treatments. Net S[O.sub.4.sup.2-] mineralisation during the crop growing season
was determined from the buildup of S[O.sub.4.sup.2-] in soil contained in PVC tubes (80 mm diameter by 200 mm long) and S[O.sub.4.sup.2-] that was retained in an anion exchange resin (Amberlite IRA) contained in a plastic insert that was sealed into the lower end of each soil core. A detailed description of the technique is given in Anderson et al. (1998b).
Four soil cores per plot were collected at monthly intervals from the same locations used to sample soil for profile S[O.sub.4.sup.2-]. These cores were then incubated at the site for 1 month. Soil within the PVC tubes was sectioned into 0-0.05, 0.05-0.1, and 0.1-0.2 m layers. Soil was handled in the field and prepared for analysis using procedures outlined earlier for the 0-0.2 m profile samples. Sulfate was extracted from anion exchange resin by shaking 5 g of wet resin with 100 mL of 1 M HCl for 1 h at 25[degrees]C on an end-over-end shaker. The extractant was filtered and then measured for S[O.sub.4.sup.2-] using an ICP-AES. Net S[O.sub.4.sup.2-] mineralisation was calculated from the change in soil S[O.sub.4.sup.2-] status (kg S/ha) in the top 0.2 m of soil over the incubation period plus the amount S[O.sub.4.sup.2-] (kg S/ha) retained in the resin trap.
The amount of S[O.sub.4.sup.2-] leached (SL) during the 1996 and 1997 growing seasons for each S[O.sub.4.sup.2-] fertiliser treatment was determined by difference:
SL = soil profile S[O.sub.4.sup.2-] content (0-0.5 m) at sowing
- soil profile content S[O.sub.4.sup.2-] (0-0.5 m) at harvest
+ net S[O.sub.4.sup.2-] mineralisation - plant S content
Plant S includes an estimate of root S (25% of S in plant tops) assuming the ratio of S in shoots and roots is similar to that found for N in wheat grown on similar soil (Atwell et al. 2002).
Treatment differences at each sampling time for DM yield (t/ha), S content (kg S/ha), S[O.sub.4.sup.2-] mineralisation (kg S/ha), and soil S[O.sub.4.sup.2-] content (kg S/ha) within the 0-0.2 m soil layer were subjected to analysis of variance using a package described by Burr (1982). Least significant differences were computed where treatment differences were significant at P = 0.05). Treatment differences in soil profile S[O.sub.4.sup.2-] content (kg S/ha) were analysed using the repeat measures routine in GENSTAT Version 6.
Key chemical and physical properties for the soil are summarised in Table 2. The S[O.sub.4.sup.2-] adsorption capacity of the soil increased from -2 [micro]g S/g soil for the 0-0.1 m soil layer to 8.9 [micro]g S/g soil in the 0.3-0.4 m soil layer from a solution containing an initial concentration of 100 [micro]g S/g soil.
Monthly growing season rainfall and annual rainfall from 1992 to 1997 is shown in Table 3. Annual rainfall was well below the long-term average in 1994, close to the long-term average in 1993, 1996, and 1997, and above the long-term average in 1992 and 1995. The 1996 growing season was characterised by a late break on 16 June followed by 85 mm over June and 65 mm over July 1996. As a result there was 35 mm of drainage below 0.7 m from 20 June to 31 July 1996 (Table 3). In contrast, the 1997 growing season was characterised by a summer rainfall event of 32 mm on 20 February 1997, and an early start to the growing season on 5 April 1997. Subsequent individual rain events were typically < 10 mm and no drainage occurred below 0.7 m (Table 3).
Daily soil temperatures, measured at a depth of 50 mm, typically ranged between 4 and 19[degrees]C over June-August and 7 and 36[degrees]C during September and October. Outside these months, soil minimum temperatures ranged between 8 and 30[degrees]C and maximum soil temperatures ranged between 14 and 53[degrees]C.
Crop biomass and sulfur content
Crop grain yields obtained from 1992 to 1995 after the application of different rates of gypsum are shown in Table 4. Sulfur application did not affect grain yield in lupin in 1992 and wheat in 1993 but there were significant increases in grain yield following the application of S to canola in 1994 and to wheat in 1995. The amounts of sulfur exported in grain are reported in Table 5.
Previous S fertiliser inputs did not affect the early growth of wheat in 1996, but significantly larger biomass was produced in the [S.sub.30] and [S.sub.15] treatments than either the [S.sub.5] or the [S.sub.0] treatment when wheat was sampled in August and September 1996, and at grain harvest. Wheat grain yield in 1996 was sharply reduced in the absence of any external sulfur input (Table 4).
The total S content of wheat grown in 1996 ranged from 4.8 to 8.3 kg S/ha with 1.5-2.6 kg S/ha removed in the grain and 3.3-5.7 kg S/ha recycled via the wheat stubble. The S content of wheat was greater in the Sis and [S.sub.30] treatments than the [S.sub.0] treatment in August 1996 and in the [S.sub.0] and [S.sub.5] treatments for samples collected between September and November 1996 (Fig. 1b). The N/S ratio of wheat grain in 1996 was >17 : 1 for all S treatments, whereas in 1995 the N/S ratio was > 17 : 1 for the [S.sub.0] and [S.sub.5] treatments only. The N/S ratio in wheat grain in 1992 was 15 : 1 to 16 : 1 across all treatments.
[FIGURE 1 OMITTED]
Except for the final harvest, the S content in lupin biomass in 1997 was not significantly affected by previous S[O.sub.4.sup.2-] fertiliser history (Fig. 1d). Lupin biomass (Fig. 1c) and grain yield (Table 4) were also unaffected by S fertilisation. Unlike the response in wheat, re-application of S[O.sub.4.sup.2-] fertiliser (15 kg S/ha) did not affect grain yield compared to the residual S[O.sub.4.sup.2-] fertiliser treatments. Total S content of lupin biomass at harvest in 1997 ranged from 7.7 to 10.0 kg S/ha with 3.8-4.4 kg S/ha recycled via the lupin stubble and 2.8-2.9 kg S/ha removed in the lupin grain. A much higher S content (6.4-6.5 kg S/ha) was found in lupin grain in 1992 (Table 5).
Soil sulfur status
Water-extractable soil S content in the top 0.1 m of soil in June 1996 ranged from 1.8 [micro]g S/g soil in the [S.sub.0] treatment to 7.9 [micro]g S/g soil in the [S.sub.30] (Table 6). By 1997, water-extractable soil S in the [S.sub.5] and [S.sub.15] treatments had declined to the quantities found in the [S.sub.0] treatment. More S was extracted with the KCl-40 technique but the effects of prior S fertilisation on KCl-40 extractable S, including changes over time, were similar to those found for water-extractable S (Table 6). The quantities of labile organic S, determined by difference between water and KCl-40 extraction, increased in response to previous S fertilisation (Table 6).
Sulfate in soil
The quantities of soil S[O.sub.4.sup.2-] in the 0-0.2 m soil layer for all treatments are shown in Fig. 2. Significantly less S[O.sub.4.sup.2-] was found in the [S.sub.0], and [S.sub.5] treatments (6-8 kg S/ha) than the [S.sub.15] and [S.sub.30] treatments (14-16 kg S/ha) on 17 June 1996 when wheat was sown (Fig. 2a). By August 1996, S[O.sub.4.sup.2-] in the 0-0.2 m soil had declined to about 4 kg S/ha in the [S.sub.15] and [S.sub.30] treatments, to 3 kg S/ha in the [S.sub.5], and <2 kg S/ha in the [S.sub.0] treatment. Little S[O.sub.4.sup.2-] remained in the top 0.2 m of soil in the [S.sub.5] and [S.sub.0] treatments from September 1996, with marginally larger quantities of S[O.sub.4.sup.2-] in the [S.sub.15] and [S.sub.30] than the [S.sub.0] and [S.sub.5] treatments. Less than 1 kg S/ha of S[O.sub.4.sup.2-] was detected in the top 0.2 m of soil in all treatments late in the wheat-growing season (Fig. 2a).
[FIGURE 2 OMITTED]
The quantities of S[O.sub.4.sup.2-] in the top 0.2 m of soil at the start of the 1997 growing season (Fig. 2b) were lower than those detected at the start of the 1996 growing season, highlighting the decline in soil S[O.sub.4.sup.2-] supplying capacity without new S inputs. There was a similar quantity (4 kg S/ha) of soil S[O.sub.4.sup.2-] in the [S.sub.0], [S.sub.5], and [S.sub.15] treatments in June and July 1997; only the [S.sub.30] treatment had significantly greater capacity to supply S[O.sub.4.sup.2-]. This difference in S[O.sub.4.sup.2-] in soil between the [S.sub.30] treatment and the other S treatments was not observed after July 1997 because of a decline in S[O.sub.4.sup.2-] in the [S.sub.30] treatment. Only ~2 kg S/ha was detected in soil to 0.2 m in August 1997 for all treatments and < 1 kg S/ha in October 1997, also for all treatments (Fig. 3b). A small increase in soil S[O.sub.4.sup.2-] was detected from October to December 1997.
[FIGURE 3 OMITTED]
The contents of S[O.sub.4.sup.2-] in soil to 0.5 m for all S treatments in June 1996, February 1997, May 1997, and December 1997 are shown in Table 7. The distribution of S[O.sub.4.sup.2-] within the 0.5 m soil profile is only shown for the [S.sub.0] and [S.sub.30] treatments for June 1996, February 1997, May 1997, and December 1997 (Fig. 3). Significant increases in S[O.sub.4.sup.2-] content between [S.sub.0] and [S.sub.30] were apparent in many but not all layers. It is notable that between 17 June 1996 and 27 February 1997, the largest change in soil S[O.sub.4.sup.2-] content occurred above 0.2 m, with a decline of 3-5 kg S/ha for the [S.sub.0] treatment and a decrease of 11-12 kg S/ha for the [S.sub.30] treatment. The S[O.sub.4.sup.2-] content of the 0.3-0.5 m soil layer decreased by 2-5 kg S/ha across all treatments in the same period. In contrast, from May to December 1997 the soil S[O.sub.4.sup.2-] profile content to 0.5m stayed at 18-21 kg S/ha for the [S.sub.0] treatment and 26-33 kg S/ha for the [S.sub.30] treatment.
Net S[O.sub.4.sup.2-] mineralisation
Net S[O.sub.4.sup.2-] mineralisation measured over the 17-month period June 1996 to November 1997 was significantly affected by previous S fertilisation history, with a larger amount mineralised (19 kg S/ha) in the [S.sub.30] treatment than the control treatment (8 kg S/ha) (Table 8). These differences in net S[O.sub.4.sup.2-] mineralisation arose largely because of a significant difference in S mineralisation between treatments in the period February-May 1997. Growing season net S[O.sub.4.sup.2-] mineralisation was not significantly different between S treatments either in 1996 or 1997. However, net S[O.sub.4.sup.2-] mineralisation did exhibit seasonal variation (Table 8), with S immobilisation apparent soon after seeding. Rates of net S[O.sub.4.sup.2-] mineralisation were typically <0.5 kg S/ha/month over the period June-August when soil temperatures ranged between 4 and 19[degrees]C. In contrast, net S[O.sub.4.sup.2-] mineralisation was of the order of 1.6-3.6 kg S/ha.month during September-October when the soil temperatures were between 7 and 36[degrees]C. Net S[O.sub.4.sup.2-] mineralisation decreased when the surface soil dried in late spring 1996. The relationship between growing season net S[O.sub.4.sup.2-] mineralisation and labile organic S extracted by KCl-40 is shown in Fig. 4.
[FIGURE 4 OMITTED]
The neutral to positive values obtained for the unaccounted S[O.sub.4.sup.2-] pool indicate that wheat and lupin derived S[O.sub.4.sup.2-] from soil layers below 0.5 m (Table 7), precluding estimates of S[O.sub.4.sup.2-] leaching using the budgeting methods employed.
Sulfur status in crops
Findings obtained here highlight the importance of maintaining S inputs on sandy soils for wheat production and wheat grain quality. Grain yield responses occurred in treatments where the KCl-40 soil test value was below the critical value (Blair et al. 1991), the S concentration in leaf tissue was below the critical value (Robson et al. 1995), and the N:S ratio in wheat grain was >17:1 (Randall et al. 1981). The low S content of the wheat grain found in this study (<0.12% S) for all residual fertiliser treatments except [S.sub.30], has important implications for grain quality (Moss et al. 1981; Zhao et al. 1999). In contrast, neither previous S[O.sub.4.sup.2-] fertilisation nor reapplication of S[O.sub.4.sup.2-] affected lupin growth. Robson et al. (1995) observed that wheat grown under glasshouse conditions gave an earlier and greater expression of S response than lupin. They also observed that the incidence of S deficiency in lupin crops in the northern agricultural region of Western Australia was low with most tissue samples close to the critical plant concentration whereas 36% of wheat samples were S-deficient or showed marginal S deficiency.
The importance of previous S[O.sub.4.sup.2-] fertiliser inputs in sustaining net S[O.sub.4.sup.2-] mineralisation in the sandy soil when S[O.sub.4.sup.2-] fertilisers are not reapplied is evident in this study as noted previously by Sakadevan et al. (1993). Mineralisation of organic S was seasonally dependent, with very low rates of net mineralisation, and in some cases S immobilisation, apparent soon after seeding. This pattern of mineralisation of organic matter is common in dryland soils where stubble retention and minimum tillage are practiced (Thomas et al. 1992; Gupta et al. 1994). Rates of net S[O.sub.4.sup.2-] mineralisation increased in spring, commensurate with rising soil temperatures, but these rates of mineralisation were unable to meet wheat demand for S. Barrow (1969) observed similar seasonal variation in rates of mineralisation in sandy soils under pasture on the Swan coastal plain in southwestern Western Australia. The net S[O.sub.4.sup.2-] mineralisation of >5 kg S/ha following a rainfall event in February 1997 highlights the likely contribution of mineralisation in years with summer-autumn rain. Summer rainfall also promotes net N mineralisation in soils similar to the one studied here (Murphy et al. 1998).
Accumulation of S[O.sub.4.sup.2-] in soil below 0.3 m is evidence of the historic displacement of S[O.sub.4.sup.2-] to depth. However, S budgets determined in this study suggested little S[O.sub.4.sup.2-] had leached into the clay below 0.5 m. This redistribution and accumulation of S[O.sub.4.sup.2-] at depth has been observed elsewhere (Chen et al. 1999). The higher S[O.sub.4.sup.2-] adsorption capacities in clay should have assisted the retention of S[O.sub.4.sup.2-] (Bolan et al. 1986; Lefroy et al. 1995) and minimised S[O.sub.4.sup.2-] movement from the rooting zone of the crops studied here. The substantial downward movement of S[O.sub.4.sup.2-] within the top 0.3 m of soil was not unexpected given the low S[O.sub.4.sup.2-] adsorption capacity in surface soil. This rapid movement of S[O.sub.4.sup.2-] from surface soil appeared to affect wheat growth more than lupin growth.
Residual value of S[O.sub.4.sup.2-] fertiliser and long-term S[O.sub.4.sup.2-] balance
In general, the residual value of previous S[O.sub.4.sup.2-] fertilisation was short-lived with as little as 15% of the applied S[O.sub.4.sup.2-] retained in the soil as S[O.sub.4.sup.2-] at the commencement of the next growing season after S application. Even lower residual values of S[O.sub.4.sup.2-] fertiliser have been found in sandy soils in the high rainfall zone of Western Australia (>750 mm) with the current season's S[O.sub.4.sup.2-] fertiliser application leached by August (Barrow 1966). These poor residual values for S[O.sub.4.sup.2-] fertiliser contrast with the 31% retention of applied S[O.sub.4.sup.2-] fertiliser found in clay loam soils (McCaskill and Cayley 2000).
The residual value of S[O.sub.4.sup.2-] fertiliser is likely to be affected by seasonal conditions, when S[O.sub.4.sup.2-] fertiliser was applied (seeding v. post seeding) and the rate of application. For example, wheat grown in 1993 was not responsive to current season S[O.sub.4.sup.2-] fertiliser application (Table 4), most likely due to the residual effect of the post seeding application of 12 kg S/ha the previous year. In contrast, the residue value of 15 kg S/ha applied in 1995 was short-lived with wheat grown in 1996 responding to S application (Table 4) after 96 mm of drainage over 2 growing seasons (Table 3).
Nutrient balances are widely used to determine the long-term sustainability of farming systems (Goh and Nguyen 1997). A common practice in the region is to use compound fertilisers at seeding that supply 1-10 kg S/ha, depending on whether fertiliser has a low or high S content. The National Land and Water Resources Audit (2001) has determined that this farming practice is likely to result in a neutral (-2-0 kg S/ha) to moderately positive (5-10 kg S/ha) S balance for the wheatbelt of Western Australia. Our findings suggest that simple budgeting techniques of the type used in the National Land and Water Resources Audit will give a good guide to the long-term S balance when S[O.sub.4.sup.2-] leaching does not occur.
Applicability of the KCl-40 soil S test
Blair et al. (1991) suggested that the extraction of soil with 0.25 M KCl at 40[degrees]C was a more accurate measure of plant-available S than the 0.01 M [Ca([H.sub.2]P[O.sub.4]).sub.2] method because the KCl-40 method extracts a pool of labile organic S that is likely to mineralise over the growing season, as well as S[O.sub.4.sup.2-]. In this study, the quantities of KCl-40 extractable labile organic S were not correlated with growing season net S[O.sub.4.sup.2-] mineralisation in soil. In particular, the KCl-40 method overestimated S mineralisation in soil that received S[O.sub.4.sup.2-] fertiliser the previous season, most probably because of the immobilisation of residual S[O.sub.4.sup.2-] fertiliser.
The nutrient status of soil is typically assessed using only the 0-0.1 m layer. In this study, wheat and particularly lupin, derived large amounts of S[O.sub.4.sup.2-] from the soil layers below 0.1 m. Accurate assessments of the S supplying capacity of soils therefore require either a deep soil S[O.sub.4.sup.2-] test or a combination of measures of net S[O.sub.4.sup.2-] mineralisation coupled with estimates of S[O.sub.4.sup.2-] leaching as suggested by Anderson et al. (1998a). Alternatively, another practical method is to assess the S status of susceptible crops, such as wheat and canola.
The trend to use high analysis N and P fertilisers in crops grown on sandy-textured soils with low S mineralisation and poor retention of S[O.sub.4.sup.2-] will reduce wheat and canola grain yield and degrade grain quality without supplementary S inputs. Development of soil S testing procedures that characterise net S mineralisation as well determine the residual amounts of S[O.sub.4.sup.2-] within root-zones are needed. Simple procedures for assessing the S[O.sub.4.sup.2-] retention in subsoil will assist the prediction of S[O.sub.4.sup.2-] movement from soil profiles and thus the likely fate of residual S[O.sub.4.sup.2-].
The Grains Research and Development Corporation (GRDC) and CSBP Limited provided funding for the project. The assistance of CSBP Ltd field staff in seeding and managing the trial site and technical assistance of Ms Megan Bytheway (CSIRO) during soil and plant sampling is gratefully acknowledged. The experiment was conducted on land generously made available by Ben and Wendy Davey, Konnongorring. Dr Senthold Asseng provided ASPIM predictions of drainage.
Manuscript received 24 June 2005, accepted 4 January 2006
Anderson GC, Blair GJ, Lefroy RDB (1998a) Soil-extractable sulfur and pasture response to applied sulfur 2. Seasonal variation in soil sulfur and sulfur response by pastures under field conditions. Australian Journal of Experimental Agriculture 38, 575-582. doi: 10.1071/EA97136
Anderson GC, Fillery IRP, Dolling PJ, Asseng S (1998b) Nitrogen and water flows under pasture-wheat and lupin-wheat rotations in deep sands in Western Australia. 1. Nitrogen fixation in legumes, net mineralisation, and utilisation of soil-derived nitrogen. Australian Journal of Agricultural Research 49, 329-343. doi: 10.1071/A97141
Asseng S, Fillery IRP, Dunin FX, Keating BA, Meinke H (2001) Potential deep drainage under wheat crops in a Mediterranean climate. 1. Temporal and spatial variability. Australian Journal of Agricultural Research 52, 45-56. doi: 10.1071/AR99186
Atwell BA, Fillery IRP, McInnes KJ, Smucker AJM (2002) The fate of carbon and fertilizer nitrogen when dryland wheat is grown in monoliths of duplex soil. Plant and Soil 241, 259-269. doi: 10.1023/A: 1016141800033
Barrow NJ (1966) The residual value of the phosphorus and sulfur components of superphosphate on some Western Australian soils. Australian Journal of Experimental Agriculture and Animal Husbandry 6, 9-15. doi: 10.1071/EA9660009
Barrow NJ (1969) The accumulation of soil organic matter under pasture and its effect on soil properties. Australian Journal of Experimental Agriculture and Animal Husbandry 9, 437-444. doi: 10.1071/EA9690437
Blair GJ, Chinoim N, Lefroy RDB, Anderson GC, Crocker GJ (1991) A soil sulfur test for pasture and crops. Australian Journal of Soil Research 29, 619-29. doi: 10.1071/SR9910619
Bolan NS, Scotter DR, Syers JK, Tillman RW (1986) The effect of adsorption on sulfate leaching. Soil Science Society of America Journal 50, 1419-1424.
Burr EJ (1982) 'NEVA user's manual.' (University of New England Computer Centre: Armidale, NSW)
Chen W, Blair G, Scott J, Lefroy R (1999) Nitrogen and sulfur dynamic of contrasting grazed pastures. Australian Journal of Agricultural Research 50, 1381-1392. doi: 10.1071/AR98104
Day PR (1965) Particle fractionation and particle-size analysis. In 'Methods of soil analysis. Part 1'. (Eds CA Black, DD Evans, JL White, LE Ensminger, EE Clark) pp. 545-566. (SSSA Madison: WI)
Eriksen J, Mortensen JV, Nielsen JD, Nielsen NE (1995) Sulphur mineralisation in five Danish soils as measured by plant uptake in a pot experiment. Agriculture Ecosystems and Environment 56, 43-51. doi: 10.1016/0167-8809(95)00632-X
Goh KM, Gregg PEH (1982) Field studies on the fate of radioactive sulphur fertiliser applied to pasture. Fertilizer Research 3, 337-351. doi: 10.1007/BF01048938
Goh KM, Nguyen ML (1997) Estimating net annual soil sulfur mineralisation in New Zealand grazed pastures using mass balance models. Australian Journal of Agricultural Research 48, 477-484. doi: 10.1071/A95123
Gupta VVSR, Roper MM, Kirkegaard JA, Angus JF (1994) Changes in microbial biomass and organic matter levels during the first year of modified tillage and stubble management practices on a red earth. Australian Journal of Soil Research 32, 1339-1354. doi: 10.1071/SR9941339
Hingston FJ, Gailitis V (1976) The geographic variation of salt precipitated over Western Australia. Australian Journal of Soil Research 14, 319-335. doi: 10.1071/SR9760319
Heng LK, White RE, Bolan NS, Scotter DR (1991) Leaching losses of major nutrients from mole-drained soil under pasture. New Zealand Journal of Agricultural Research 34, 325-334.
Johnson CM, Ulrich A (1959) Analytical methods for use in plant analysis. California Agricultural Experimental Station Bulletin 766, 26-77.
Lefroy RDB, Santoso D, Blair GJ (1995) Fate of applied phosphate and sulfate in weathered acid soils under leaching conditions. Australian Journal of Soil Research 33, 135-151. doi: 10.1071/ SR9950135
Lisle L, Gaudron J, Lefory R (1991) Laboratory techniques for plant and soil analysis. Department of Agronomy and Soil Science, University of New England, Armidale, NSW, Australia.
Matilainen R, Tummavuori J (1996) Determination of sulfur in fertilisers by inductively coupled plasma-atomic emission spectrometry--spectral and inter-element effects at various wavelengths. Journal--Association of Official Analytical Chemists 79, 1026-1035.
McCaskill MR, Cayley JWD (2000) Soil audit of a long-term phosphate experiment in south-western Victoria: total phosphorus, sulfur, nitrogen and major cations. Australian Journal of Agricultural Research 51,737-748. doi: 10.1071/AR99091
Moss HJ, Wrigley CW, MacRitchie F, Randall PJ (1981) Sulfur and nitrogen effects on wheat. II. Influence on grain quality. Australian Journal of Agricultural Research 32, 213-226. doi: 10.1071/AR9810213
Murphy CV, Spading GP, Fillery IRP, McNeill AM, Braunberger P (1998) Mineralisation of soil organic nitrogen and microbial respiration after simulated summer rainfall events in an agricultural soil. Australian Journal of Soil Research 36, 231-246. doi: 10.1071/S97043
National Land and Water Resources Audit (2001) Australian Agriculture Assessment 2001--Nutrient Management in Australian Agriculture. National Land and Water Resource Audit, Canberra. http://audit.ea.gov.au/anra/agriculture/
Probert ME, Dimes JP, Keating BA, Dalal RC, Strong WM (1998) APSIM's water and nitrogen modules and simulation of the dynamics of water and nitrogen in fallow systems. Agricultural Systems 56, 1-28. doi: 10.1016/S0308-521X(97)00028-0
Randall PJ, Spencer K, Freney JR (1981) Sulfur and nitrogen effects on wheat. I. Concentrations of sulfur and nitrogen and the nitrogen to sulfur ratio in grain, in relation to the yield response. Australian Journal of Agricultural Research 32, 203-212. doi: 10.1071/AR9810203
Rayment GE, Higginson FR (1992) 'Australian laboratory handbook of soil and water chemical methods.' (Inkata Press: Melbourne)
Robson AD, Osborne LD, Snowball K, Simmons WJ (1995) Assessing sulfur status in lupins and wheat. Australian Journal of Experimental Agriculture 35, 79-86. doi: 10.1071/EA9950079
Sakadevan K, Hedley MJ, Mackay AD (1993) Sulphur cycling in New Zealand hill country pastures. I. Laboratory sulphur, nitrogen and carbon mineralization studies. Journal of Soil Science 44, 73-83.
Sakadevan K, Hedley M J, Mackay AD (1994) An in situ lysimeter with a removable ion exchange resin trap for measuring nutrient losses by leaching from grazed pastures. Australian Journal of Soil Research 32, 1389-1400. doi: 10.107|/SR9941389
Scherer HW (2001) Sulphur in crop production---invited paper. European Journal of Agronomy 14, 81-111. doi: 10.1016/S11610301 (00)00082-4
Schoknecht N (1997) Soil groups of Western Australia: a simple guide to the main soils of Western Australia. Resource Management Technical Report No. 171, Agriculture Western Australia.
Tabatabai MA, Al-Khafaji AA (1980) Comparison of nitrogen and sulfur mineralization in soils. Soil Science Society of America Journal 44, 1000-1006.
Thomas GA, Standley J, Hunter HM, Webb AA, Blight GW, Sinclair DP (1992) The effect of fallow management practices on grain sorghum in central Queensland. AlAS Occasional Publication No. 2, pp. 408-413.
Zhao FJ, Salmon SE, Withers PJA, Evans EJ, Monaghan JM, Shewry PR, McGrath SP (1999) Responses of breadmaking quality to sulphur in three wheat varieties. Journal of the Science of Food and Agriculture 79, 1865-1874. doi: 10.1002/(SICI)10970010(199910)79:13 < 1865::A1D-JSFA449 > 3.0.CO;2-R
G. C. Anderson (A), I. R. P. Fillery, (A), D, F. H. Ripper (B), and B. J Leach (C)
(A) CSIRO, Private Bag, PO Wembley, WA 6014, Australia.
(B) CSBP Ltd, PO Box 345, Kwinana, WA 6966, Australia.
(C) Formerly CSBP Ltd, now at John Duff and Associates, Suite 5, 110 Robinson Avenue, Belmont, WA 6104, Australia.
(D) Corresponding author. Email: Ian.Fillery@csiro.au
Table 1. Crop cultivar, time of sowing, seeding rate, and rate and date N fertiliser was applied in each year of the study Year Crop Variety Sowing Seeding rate date (kg/ha) 1992 Lupin Gungurru 12 May 108 1993 Wheat Spear 11 May 75 1994 Canola Narendra 06 June 8 1995 Wheat Spear 07 June 75 1996 Wheat Amery 06 June 75 1997 Lupin Merrit 24 May 80 Year N applied (kg N/ha) Date Rate Date Rate 1992 -- -- -- -- 1993 11 May 20 17 June 20 1994 11 May 92 25 Aug. 50 1995 14 July 46 18 Aug. 46 1996 31 July 46 28 Aug. 46 1997 -- -- -- -- Table 2. Particle size analysis, chemical composition, and S sorption from solution with various S042- concentrations for samples of the soil profile collected on 17 June 1996 Measurement Depth (m): 0-0.05 0.05-0.1 Bulk density (Mg/[m.sup.3]) 1.7 1.7 Gravel (%) 0 0 Sand (%) 89 92 Silt (%) 0 6 Clay (%) 3 2 Fe ([micro]g/g soil) 88 94 pH(Ca[C1.sub.2]) 4.9 4.6 C (%) 0.9 0.5 N[O.sub.3] ([micro]g/g soil) 20 9 N[H.sub.4] ([micro]g/g soil) 16 13 P ([micro]g/g soil) 21 17 K ([micro]g/g soil) 69 35 S ([micro]g/g soil) 4.7 3.9 Total P ([micro]g/g soil) 58.5 (A) Total S ([micro]g/g soil) 59.1 (A) S adsorption (50 (B)) ([micro]g/g soil) n.d. S adsorption (100 (B)) ([micro]g/g soil) -2 (A) S adsorption (150 (B)) ([micro]g/g soil) -0.3 (A) S adsorption (200 (B)) ([micro]g/g soil) 3.6 (A) S adsorption (250 (B)) ([micro]g/g soil) 2.9 (A) Measurement Depth (m): 0.1-0.2 0.2-0.3 Bulk density (Mg/[m.sup.3]) 1.8 1.7 Gravel (%) 0 0 Sand (%) 95 86 Silt (%) 3 2 Clay (%) 2 10 Fe ([micro]g/g soil) 149 209 pH(Ca[C1.sub.2]) 4.6 4.3 C (%) 0.2 0.1 N[O.sub.3] ([micro]g/g soil) 8 8 N[H.sub.4] ([micro]g/g soil) 8 5 P ([micro]g/g soil) 20 25 K ([micro]g/g soil) 38 49 S ([micro]g/g soil) 2.8 2 Total P ([micro]g/g soil) 84.1 21.1 Total S ([micro]g/g soil) 20.8 19.6 S adsorption (50 (B)) ([micro]g/g soil) -0.4 -1.3 S adsorption (100 (B)) ([micro]g/g soil) 6.6 8.2 S adsorption (150 (B)) ([micro]g/g soil) 1.8 n.d. S adsorption (200 (B)) ([micro]g/g soil) 15.7 18.2 S adsorption (250 (B)) ([micro]g/g soil) 24.5 25 Measurement Depth (m): 0.3-0.4 0.4-0.5 Bulk density (Mg/[m.sup.3]) 1.8 1.8 Gravel (%) 2 18 Sand (%) 81 84 Silt (%) 2 2 Clay (%) 13 12 Fe ([micro]g/g soil) 209 206 pH(Ca[C1.sub.2]) 4.3 4.4 C (%) 0.1 0.1 N[O.sub.3] ([micro]g/g soil) 6 6 N[H.sub.4] ([micro]g/g soil) 2 1 P ([micro]g/g soil) 12 5 K ([micro]g/g soil) 44 41 S ([micro]g/g soil) 3 4 Total P ([micro]g/g soil) 20.7 n.d. Total S ([micro]g/g soil) 21.8 n.d. S adsorption (50 (B)) ([micro]g/g soil) 2.2 n.d. S adsorption (100 (B)) ([micro]g/g soil) 8.9 n.d. S adsorption (150 (B)) ([micro]g/g soil) 17.6 n.d. S adsorption (200 (B)) ([micro]g/g soil) n.d. n.d. S adsorption (250 (B)) ([micro]g/g soil) 32.6 n.d. n.d., Not determined. (A) Soil analysis on 0-0.1 m samples. (B) Concentration of added solution ([micro]g/g soil). Table 3. Monthly growing season rainfall and annual rainfall (mm) at the site, and computer-simulated drainage (mm) at 0.7m, from 1992 to 1997 Year April May June July 1992 101 15 84 54 1993 11 114 76 57 1994 -- 80 56 33 1995 16 60 81 145 1996 13 15 86 65 1997 42 27 28 38 Year Aug. Sept. Oct. Nov. 1992 107 53 18 17 1993 59 31 17 21 1994 41 26 3 -- 1995 34 33 16 2 1996 48 51 13 18 1997 54 20 17 12 Year Annual Drainage rainfall below 0.7 m 1992 543 47 1993 395 45 1994 277 0 1995 406 56 1996 335 35 1997 307 0 Table 4. Grain yields (t/ha) obtained after different rates of S fertiliser application to specified crops Year Crop Sulfate treatment: [S.sub.0] [S.sub.5] [S.sub.15] 1992 Lupin 2.21 2.29 2.24 1993 Wheat 3.39 3.51 3.73 1994 Canola 0.24 0.32 0.45 1995 Wheat 0.90 1.35 1.97 1996 Wheat 1.30 1.90 2.00 1997 Lupin 1.66 1.65 1.61 Year Sulfate treatment: I.s.d. [S.sub.30] [S.sub.R15] (P = 0.05) 1992 2.19 2.20 n.s. 1993 3.78 3.67 n.s. 1994 0.46 0.33 0.14 1995 1.86 1.82 0.45 1996 2.30 2.90 0.39 1997 1.48 1.68 0.19 n.s., Not significant. Table 5. Differences between S inputs (kg S/ha) and S removed in grain (kg S/ha) for the period 1992-97 S in rainfall was derived from measurements of S in rainfall for the central wheatbelt of Western Australia reported by Hingston and Gailitis (1976), with input corrected for annual rainfall Year S in S applied in fertiliser rainfall [S.sub.0] [S.sub.5] [S.sub.15] [S.sub.30] 1992 1.5 0.6 6 16 31 1993 1.1 0.6 6 16 31 1994 0.6 0.6 6 16 31 1995 1.4 0.6 6 16 31 1996 1 0.6 1 1 1 1997 0.9 0.6 1 1 1 Year S remove in grain [S.sub.0] [S.sub.5] [S.sub.15] [S.sub.30] 1992 6.4 6.5 6.6 6.5 1993 3.6 3.7 4.1 4.2 1994 1.1 1.4 1.9 2.2 1995 1.2 1.6 2.7 2.6 1996 1.5 2.1 2.4 2.6 1997 2.8 2.9 2.8 2.8 Year S input-output balance [S.sub.0] [S.sub.5] [S.sub.15] [S.sub.30] 1992 -4.3 0.9 11.0 25.9 1993 -1.9 3.4 13.0 27.9 1994 0.1 5.2 14.7 29.4 1995 0.8 5.8 14.7 29.8 1996 0.1 -0.1 -0.4 -0.6 1997 -1.3 -1.0 -0.9 -0.9 Table 6. Quantities of soil S extracted by [H.sub.2]O and KCl-40 expressed as [micro]g S/g soil and kg S/ha Extraction S fractions S treatments: [S.sub.0] [S.sub.5] 1994 KCl-40 ([micro]g S/g soil) 4.3 5.5 1996 [H.sub.2]O ([micro]g S/g soil) 1.8 3.0 KCl-40 (total) ([micro]g S/g soil) 4.3 5.8 [H.sub.2]O (kg S/ha) 2.8 4.8 KCl-40 (total) (kg S/ha) 7.1 9.4 KCl-40 (labile organic S) (kg 4.3 4.6 1997 [H.sub.2]O ([micro]g S/g soil) 1.5 1.7 KCl-40 (total) ([micro]g S/g soil) 3.2 3.1 [H.sub.2]O (kg S/ha) 2.4 2.7 KCl-40 (total) (kg S/ha) 5.2 5.1 KCl-40 (labile organic S) (kg 2.8 2.5 Extraction S fractions S treatments: [S.sub.15] [S.sub.30] 1994 KCl-40 ([micro]g S/g soil) 11.6 13.4 1996 [H.sub.2]O ([micro]g S/g soil) 6.1 7.9 KCl-40 (total) ([micro]g S/g soil) 10.6 11.5 [H.sub.2]O (kg S/ha) 10.0 11.6 KCl-40 (total) (kg S/ha) 17.1 18.4 KCl-40 (labile organic S) (kg 7.1 6.8 1997 [H.sub.2]O ([micro]g S/g soil) 1.6 2.2 KCl-40 (total) ([micro]g S/g soil) 4.1 5.7 [H.sub.2]O (kg S/ha) 2.4 3.6 KCl-40 (total) (kg S/ha) 6.7 9.3 KCl-40 (labile organic S) (kg 4.3 5.7 Table 7. Changes in amounts of soil profile S[O.sub.4.sup.2-](kg S/ha) over the growing seasons, plant S[O.sub.4.sup.2-] uptake (kg S/ha), and by difference either S[O.sub.4.sup.2-] uptake (kg S/ha) from below 0.5 m or the amounts of S[O.sub.4.sup.2-] leached (kg S/ha) estimated to move below 0.5 m in 1996 and 1997 Year Amount Amount of S[O.sub.4.sup.2-] fertiliser in soil to 0.5 m S applied 1996 17.vi.96 27.ii.97 [S.sub.0] 21.8 21.2 [S.sub.5] 25.3 24.4 [S.sub.15] 32.3 28.4 [S.sub.30] 40.1 34.3 1997 25.v.97 4.xii.97 [S.sub.0] 18.7 18.2 [S.sub.5] 21.0 19.7 [S.sub.15] 26.0 25.7 [S.sub.30] 32.9 33.5 Year Amount Growing season Plant fertiliser net S uptake S applied mineralisation of [S.sup.A] 1996 [S.sub.0] 3.8 6.0 [S.sub.5] 2.2 8.2 [S.sub.15] 3.3 9.5 [S.sub.30] 5.0 10.4 1997 [S.sub.0] 3.5 9.6 [S.sub.5] 5.1 11.0 [S.sub.15] 8.0 10.0 [S.sub.30] 6.7 12.5 Year Amount Plant uptake fertiliser S[O.sub.4.sup.2-] from S applied below 0.5 m or S[O.sub.4.sup.2-] leached below 0.5 [m.sup.B] 1996 [S.sub.0] 1.6 [S.sub.5] 5.1 [S.sub.15] 2.3 [S.sub.30] -0.4 1997 [S.sub.0] 5.6 [S.sub.5] 4.6 [S.sub.15] 1.7 [S.sub.30] 6.4 (A) Content of S in root was assumed to 25% of top S based on similar ratio for N in wheat on a similar soil (Atwell et al. 2002). (B) Negative value indicates leaching of S[O.sub.4.sup.2-] below the soil depth of 0.5 m. Table 8. Net S mineralisation (kg S/ha) measured for the fertiliser treatments in 1996 and 1997 Treatment 17.vi.-30.vii 30.vii.-28.viii [S.sub.0] 0.3 -1.0 [S.sub.5] -0.9 -1.0 [S.sub.5] -1.5 0.2 [S.sub.30] 1.0 -0.4 l.s.d. (P=0.05) n.s. n.s. Treatment Growing season 1996 28.viii.-24.ix 24.ix.-22.x [S.sub.0] 2.8 2.3 [S.sub.5] 2.7 2.4 [S.sub.5] 2.4 2.4 [S.sub.30] 2.7 2.0 l.s.d. (P=0.05) n.s. n.s. Treatment Fallow 22.x.-26.xi Total 27.ii.-27.v.1997 [S.sub.0] -0.6 3.8 -0.2 [S.sub.5] -1 2.2 0.2 [S.sub.5] -0.2 3.3 5.8 [S.sub.30] -0.3 5.0 5.5 l.s.d. (P=0.05) 0.60 n.s. 1.4 Treatment 27.v.-27.vi 27.vi.-23.vii [S.sub.0] 0 0.4 [S.sub.5] -0.2 0.8 [S.sub.5] 0.6 0.6 [S.sub.30] -0.3 0.7 l.s.d. (P=0.05) n.s. n.s. Growing season 1997 23.vii.-20.viii 20.viii.-18.ix [S.sub.0] -0.3 1.6 [S.sub.5] 0.0 2.2 [S.sub.5] 1.6 2.4 [S.sub.30] 0.9 2.1 l.s.d. (P=0.05) n.s. n.s. Total 18.ix.-24.xi Total 17.vi.96-24.xi.97 [S.sub.0] 1.8 3.5 7.1 [S.sub.5] 2.3 5.1 7.5 [S.sub.5] 2.8 8.0 17.1 [S.sub.30] 3.3 6.7 17.2 l.s.d. (P=0.05) 1.3 n.s. 3.2 n.s., Not significant
|Printer friendly Cite/link Email Feedback|
|Author:||Anderson, G.C.; Fillery, I.R.P.; Ripper, F.H.; Leach, B.J.|
|Publication:||Australian Journal of Soil Research|
|Date:||Mar 15, 2006|
|Previous Article:||Spatial and time-dependent patterns of selenium (Se) release from selected Se fertiliser granules.|
|Next Article:||Analysis of partial breakthrough data by a transfer-function method.|