Tracing the nitrogen, sulfur, and carbon released from plant residues in a soil/plant system.
Abstract: Matching plant residue mineralisation rate to plant nutrient demand is one way of increasing the efficiency of nutrient cycling. A glasshouse experiment was conducted in a Soloth soil with a [C.sub.4] [Delta][sup.13]C signature using drained pots to examine the effect on the yield of Japanese millet (Echinochloa frumentocea) and the fate of [sup.15]N, [sup.35]S, and C (using [Delta][sup.13]C shift) from the [C.sub.3] plants Flemingia macrophylla, Medicago truncatula hay, and wheat (Triticum aestivum) straw applied at 3 t/ha in the presence of N and NPKS fertiliser. The yield of Japanese millet at 91 days was highest where medic hay had been added (13.7 g/pot) and lowest where wheat straw was used (11.5 g/pot). Recovery of [sup.35]S by the millet was highest in the wheat straw and medic hay treatments (mean 11.5%), whilst recovery of [sup.15]N was highest from medic hay (15.8%). Leaching losses of [sup.35]S were highest in the Flemingia and medic treatments (mean 8.1%), and [sup.15]N loss in leachate was highest in the medic hay treatment (6.6%). A maximum of 1.5% of the C added in residues was recovered in the leachate of the medic hay treatment.
Article Type: Statistical Data Included
Subject: Plant-soil relationships (Research)
Crop residues (Environmental aspects)
Nitrogen (Environmental aspects)
Sulfur (Environmental aspects)
Carbon (Environmental aspects)
Authors: Konboon, Yothin
Blair, Graeme
Lefroy, Rod
Whitbread, Anthony
Pub Date: 05/01/2000
Publication: Name: Australian Journal of Soil Research Publisher: CSIRO Publishing Audience: Academic Format: Magazine/Journal Subject: Agricultural industry; Earth sciences Copyright: COPYRIGHT 2000 CSIRO Publishing ISSN: 0004-9573
Issue: Date: May, 2000 Source Volume: 38 Source Issue: 3
Geographic: Geographic Scope: Australia Geographic Code: 8AUST Australia
Accession Number: 62707811
Full Text: Introduction

There is a need to study and monitor the fate of carbon and nutrients released from crop residues to enable them to be managed effectively, and to improve and sustain soil organic matter and crop productivity. Many studies have utilised [sup.15]N (Catchpoole and Blair 1990) and [sup.35]S (Blair et al. 1994) labelled plants to study the dynamics of nitrogen (N) and sulfur (S) from plant residues However, monitoring changes in organic matter is constrained by the difficulty of detecting changes in carbon (C) against a large and variable background of existing soil organic matter.

Nutrient dynamics have often been monitored by directly measuring the concentration of soil organic matter in long-term experiments or by the use of radioisotopes. O'Brien (1984) used the `bomb carbon' [sup.14]C radioisotope technique to study the effects of pasture improvement and earthworms on carbon input rate and decomposition time and diffusivity down the profile. Recent workers have adopted the use of the natural abundance of [sup.13]C to investigate the turnover of organic matter in soils receiving additions of crop residues (Martin et al. 1990; Cerri et al. 1991). When [C.sub.4] plants fix C by photosynthesis they discriminate between [sup.12]C and [sup.13]C to a different degree than [C.sub.3] plants, which results in different ratios of these two isotopes of C. This is calculated as the [Delta][sup.13]C shift. For example, the use of [Delta][sup.13]C shift has been used in studies of soil organic matter dynamics following deforestation and long-term cultivation and these have shown that it is possible to quantify the losses of C derived from native vegetation and the carbon input from crop residues (Cerri et al. 1991).

A glasshouse experiment, using drained pots, was undertaken to examine the fate of N, S, and C applied in residues of the [C.sub.3] plants flemingia (Flemingia macrophylla) leaf, medic (Medicago truncatula) hay, and wheat (Triticum aestivum) straw, which were labelled with [sup.15]N and [sup.35]S. These residues were added to a soil with a predominately [C.sub.4] [Delta][sup.13]C signature and the [Delta][sup.13]C technique was used to monitor changes in soil carbon.

Materials and methods

The experiment consisted of 2 phases, a plant residue production phase in which [sup.35]S and [sup.15]N labelled plant residues were produced (April-October 1994) and an experimental phase from November 1994 to February 1995 where the [sup.35]S and [sup.15]N released from incorporated residues was traced when Japanese millet was grown in a pot experiment.

Plant residue production phase

Plastic pots (30 cm diameter for Flemingia and 20 cm diameter for wheat and medic), lined with a plastic bag to prevent contamination of the pot with radioisotope and prevent drainage, were filled with a 1:1 mixture by volume of vermiculite and sand. The pots were maintained near field capacity with distilled water, and a complete nutrient solution was applied at intervals.

Once the plants were well established, the Flemingia was heavily defoliated and the wheat and medic thinned to 10 and 5 plants per pot, respectively. Applications of [sup.15]N and [sup.35]S solution were then commenced. The [sup.15]N was applied as 98.94% enriched [sup.15][NH.sub.4]Cl in solution with a concentration of 0.242 mg [sup.15]N/mL and 5 mL of this solution was applied by syringe to each pot every 2 days. At the same time a syringe was used to apply 5 mL of carrier-free [Ca.sup.35][SO.sub.4] containing approximately 0.068 MBq/mL over the pot surface, making sure that no solution contacted the plant.

Soil and pot preparation

A Soloth soil from Mackay, Queensland, was collected from an area which has been under [C.sub.4] vegetation (sugarcane) for more than 50 years. This soil had a [sup.13][Delta]C value of -14.35 [salinity], which was considered sufficiently different from the added residues to be able to separate the contribution to soil C pools from native and added C. The soil was air-dried, cleaned, and passed through a 2-mm sieve before being used in the experiment. Key chemical properties of the soil are presented in Table 1.

Table 1. Selected soil characteristics of the Soloth soil used in the experiment

(A) Determination by oxidation with KMn[O.sub.4 ] (Blair et al. 1995).

Pots, 30 cm deep, were made from 15-cm-internal diameter polyvinyl chloride (PVC) pipe fitted and silicone sealed with PVC end caps which had several holes (2-mm diameter) to allow drainage. A fine (1-mm mesh) nylon screen was placed above the holes to prevent the loss of soil particles. The pots were then placed on top of a modified end cap, to enable the collection of leachate. Before the soil was added to the pots they were marked into depths of 0-8, 8-16, and 16-24 cm (hereafter referred to as top, middle, and bottom parts, respectively). Each part was filled with 2.033 kg of air-dried soil and a plastic mesh (5 mm gutter guard) was used to separate the top and middle soil layers.

Experimental design and layout

Treatments were laid out according to a split-plot design with 3 replicates. The main-plot factor was harvest time and the subplot factors consisted of plant residue and fertiliser. Japanese millet (Echinochloa frumentocea), a [C.sub.4] crop, was selected as an indicator crop because its [Delta][sup.13]C value of -13.40 [salinity] was similar to that of the soils. In summary the design was: main-plot factor, 3 harvests [27 and 48 days after planting, and at maturity (91 days)]; subplot factors, 4 residues [non-residue control, Flemingia leaf, barrel medic hay, and wheat straw] x 2 fertilisers [N, NPKS] x 3 replicates.

The experiment was conducted over 12 weeks in a controlled temperature (20 [degrees] C night-30 [degrees] C day) glasshouse at the Department of Agronomy and Soil Science, University of New England, Australia.

Experimental treatments

The 4 residue treatments evaluated were a non-residue control, Flemingia macrophylla leaf litter, barrel medic (Medicago truncatula) hay, and wheat (Triticum aestivum) straw. Hereafter referred to as nonresidue, Flemingia leaf, medic hay, and wheat straw, respectively. The carbon, nutrient, and isotopic characteristics of the applied plant residues are presented in Table 2.

Table 2. Carbon, nutrient, and isotopic characteristics of the three plant residues used in the glasshouse experiment

Before application the plant residues were cut into 2-3 cm pieces and dried in an oven at 65 [degrees] C. The residue was applied at 5.4 g/pot, which is equivalent to approximately 3 t/ha on the basis of pot surface area. The plant residues, and fertiliser applications consisting of either N or N, P, K, and S, were incorporated into the top 8-cm soil layer before potting. The two fertiliser treatments applied were equivalent to 100 kg/ha of 10:0:0:0 and 30:25:20:8 (N:P:K:S). These are hereafter referred to as N and NPKS treatments.

The millet showed symptoms of nitrogen deficiency, so nitrogen, as urea solution, was applied at 10 and 30 kg/ha to the N and NPKS fertiliser treatments, respectively, 49 and 63 days after planting.

Management of the Japanese millet crop

Approximately 10 seeds of millet were planted into each pot and these were thinned to 4 healthy seedlings. The pots were then placed in a glasshouse in which the temperature was maintained between 20 [degrees] C and 30 [degrees] C. The pots were weighed periodically to maintain the soil moisture near field capacity, except during the weekly period of leachate collection when each pot was watered to 25% above field capacity. Leachate was collected 1 day after excess watering, weighed to record the volume, and stored in plastic bottles in a freezer prior to analysis.

At each harvest the millet plants were cut at the soil surface and at the third harvest, the tops were separated into grain and stem plus leaf. The millet roots were removed from each soil layer and cleaned. The plant shoot, grain, and root samples were then oven-dried at 80 [degrees] C, weighed, and ground to pass a 1-mm sieve and stored for subsequent chemical analyses.

Sample collection and measurement

The soil from the pots, which were destructively harvested at each harvest time, was pushed out of the pot, laid on trays and separated into the top, middle, and bottom sections. The soil within each section was mixed thoroughly and visible roots were removed from each soil layer. A 500-g subsample of each soil layer was then taken, air-dried, ground through a 1-mm sieve, and stored in a plastic bottle for chemical analyses.

Leachate was subsampled for nitrate and ammonium determination using an autoanalyser (Technicon 1977). Nitrate was reduced to nitrite (Adamsen et al. 1985) using a Cd reduction column. Ammonium in the leachates was analysed using a variation of the indophenol blue method for the autoanalyser (Technicon 1977).

A diffusion method, similar to that described by Brooks et al. (1989) and Jensen (1991), was used to prepare leachate samples for nitrate and ammonium [sup.15]N analyses. Whatman GF/D filter papers were pretreated with 2.5 M [KHSO.sub.4] to absorb the N and placed inside glass vials containing the leachate, which were then sealed After absorption the filter papers were placed into tin capsules before being analysed for [sup.15]N by ANCA-MS.

The procedure used to determine the C concentration in the leachate was similar to that described by Oweczkin et al. (1995). The leachate samples (weeks 1-4, weeks 5-7, and weeks 8-12) were pooled together to reduce the number of samples. The leachate samples were analysed for carbon by ICP-AES, with a carbon line profiled at 193.09 nm on the polychromator, and calibrated against acid-treated standards.

The [sup.35]S activity of the leachate was determined on a 3-mL sample which was placed into a vial and mixed with 17 mL of scintillation fluid consisting of toluene, p-terphenyl, POPOP, and teric. The [sup.35]S activity in the sample was then determined in a liquid scintillation counter.

Total C, labile C, total N, [Delta][sup.13]C, and atom% [sup.15]N were determined on a finely ground soil sample ([is less than] 0.5 mm) containing approximately 350 [micro]g C, which was weighed into tin cups for analyses using an ANCA-MS calibrated against an appropriate soil standard. Labile carbon was determined by the method described by Blair et al. (1995). In this method, the soil is reacted with an excess of 333 mM KMn[O.sub.4] and the reduced permanganate, which is proportional to the amount of carbon oxidised, is measured by spectrophotometry.

Extractable S in soil samples was determined using the KCl-40 soil test developed by Blair et al. (1991). A 3-mL subsample of extract was taken for determination of [sup.35]S activity by liquid scintillation counter.

The ground millet plant parts and plant residues were subsampled (approximately 0.2 g) for digestion in a sealed container (Anderson and Henderson 1986) and the digest was measured for cations, P, and S by ICP-AES. A 3-mL subsample of the plant extract was also taken for measurement of [sup.35]S.

Finely ground plant and residue samples, containing approximately 350 [micro]g C, were weighed into tin cups for analyses of C and N concentrations, [Delta][sup.13]C abundance, and atom% [sup.15]N in an ANCA-MS system which was calibrated against an appropriate flour standard.

The proportion of added residue present in the soil organic fraction was calculated from the following equation:

[Chi] = ([Delta]f - [Delta]s)/([Delta]r - [Delta]s)

where [Delta]f is the [Delta][sup.13]C value of the soil after residue incorporation, [Delta]s is the [Delta][sup.13]C of the original soil or soil of the control treatment, and [Delta]r is the [Delta][sup.13]C value of the plant material (Cadisch and Giller 1996).

From the total weights (w) of carbon from each sample, the amount of residue present (R) was calculated from the equation:

R = [Chi] x w

The proportion of [sup.15]N recovered which was derived from plant residue was calculated using:

Proportion of residue [sup.15]N recovered = [atom%(f) - atom%(c)] x Nt/[atom%(r) - atom%(c)] x Nr

where atom%(f) is the atom% [sup.15]N value of each component (e.g. soil, plant shoot, leachate, etc.) alter residue incorporation; atom%(c) is the atom% [sup.15]N value of the same component in the control treatment, and atom%(r) is the atom% [sup.15]N value of the plant material added; Nt is the total N content of each component (e.g. soil, plant shoot, leachate, etc.); and Nr is the total N content of plant residue (Catchpoole 1988).

The proportion of [sup.35]S derived from a plant residue was calculated using:

Proportion of residue [sup.35]S = (radioactivity of [sup.35]S of each component)/(radioactivity of added [sup.35]S from residue)

In order to calculate the amount of residue S in each component using [sup.35]S data, the following formula was used:

Specific activity of residue (SA) = radioactivity of [sup.35]S in residue/total S in residue

Amount of residue S in each component = SA x radioactivity of the component

Data were subjected to a factorial analysis of variance and differences between means were determined using Duncan's Multiple Range Test (DMRT). Means were separated at P = 0.05 and means in tables sharing a common letter are not significantly different.

Results

Millet tops yields

At the Day 27 harvest, there were no significant differences in plant top yield between the residue treatments in the N-alone treatment (Table 3). Application of Flemingia leaf and wheat straw in the NPKS treatment produced the highest plant top yields. The plant tops yields in the Flemingia leaf and wheat straw treatments was significantly higher in the NPKS than the N treatment but there was no difference in the medic hay and control treatments (Table 3).

Table 3. Effect of plant residue and fertiliser rate on plant top yield (g/pot) of millet

Within each harvest, means followed by a common letter are not significantly different at P = 0.05 by DMRT

At the 48-day harvest, application of medic hay, Flemingia leaf, and wheat straw in the presence of N alone produced increased plant top yield compared with the nonresidue control. In the NPKS treatment the application of Flemingia leaf or wheat straw produced a higher plant top yield than the control (Table 3).

The response of plant top yield (including grain) to the treatments at the 91-day harvest was similar to that at the second harvest. When only N was applied, the highest plant top yield of 13.40 g/pot was measured in the medic hay treatment followed by Flemingia leaf, wheat straw, and the control treatments (Table 3). An interaction between plant residue and fertiliser treatment was again found, with NPKS enhancing the crop yield in the Flemingia leaf, wheat straw, and control treatments compared with the N-alone treatment but there was no significant yield increase in the medic hay treatment.

Uptake of [sup.35]S by Japanese millet

There was no significant difference between the residue treatments in the percentage of residue S contained in the millet when the crop was harvested at Day 27 and Day 48. At the Day 91 harvest the application of wheat straw and medic hay resulted in greater uptake of residue S by the millet and these were significantly higher than that observed in the Flemingia leaf treatment (Table 4).

Table 4. Recovery of [sup.35]S in millet tops through time (harvest Days 27, 48, 91), percentage of applied [sup.35]S recovered in KCl-40 extract in the soil layers (top, middle, bottom), and loss through leaching as affected by harvest times and plant residues (as % of [sup.35]S added in residue)

Within a column, means followed by a common letter are not significantly at P = 0.05 by DMRT

[sup.35]S concentration in the KCl-40 extract

At the Day 91 harvest the highest KCl-40 extractable S concentrations were found in the bottom soil layer. The highest concentration of residue [sup.35]S in the KCl-40 extract, in each soil layer, was found in the medic and wheat straw treatments. The highest recovery of [sup.35]S in the KCl-40 extract in the middle and bottom soil layers was also in these two treatments.

Losses of sulfur in leachate

The greatest loss of S by leaching over the 12-week period, 21.4 mg, was observed from the wheat straw treatment followed by 18.7, 18.2, and 16.9 mg from medic hay, Flemingia leaf, and the control treatments, respectively.

Leaching of residue S, as measured by [sup.35]S, gradually increased from the beginning of the experiment and reached a maximum at Week 5 in all residues. There was a significantly higher (29.7%) loss of residue S from the NPKS treatment than the N-alone treatment when averaged over residue additions. After 12 weeks the greatest loss of residue S by leaching occurred in the wheat straw and medic hay treatments, 8.7% and 7.4%, respectively, which were significantly different from the 2.3% of residue S that was leached from the Flemingia leaf treatment (Table 4). These higher leaching losses were associated with a higher recovery of [sup.35]S from the residues present in the bottom soil layers of these treatments.

Recovery of [sup.15]N by Japanese millet

At the Day 27 harvest there was no effect of residue on the uptake of [sup.15]N by millet (Table 5). At the Day 48 harvest the highest uptake of [sup.15]N was recorded in the medic hay treatment and this was significantly higher than in the wheat straw and Flemingia leaf treatments. A similar result was recorded at the Day 91 harvest.

Table 5. Percentage recovery in millet tops of [sup.15]N added in residues in millet tops as affected by harvest time and residue treatment

Within a column, means followed by a common letter are not significantly different at P = 0.05 by DMRT

Between the Day 48 and 91 harvests there was an increase in the [sup.15]N content of the millet plants in all residue treatments, with the rate of increase differing between residues.

Recovery of residue N ([sup.15]N) in soil

More than 70% of the [sup.15]N applied in the residue was recovered in the top soil layer throughout the experiment (Table 6), with the highest proportion in the Flemingia treatment. In each residue treatment there was a significant decline in the proportion of [sup.15]N recovered in the top soil layer over time. However, there were no increases in the N concentration in the deeper soil layers. More than twice the proportion of added [sup.15]N was recovered in the middle soil layer in the medic hay and wheat straw treatments than in the Flemingia treatment.

Table 6. Percentage of added [sup.15]N recovered in each soil layer as affected by residue and harvest time and total leaching loss at 91 days (% of [sup.15]N input in residue)

Means followed by the same letter within the [sup.15]N recovery data within a harvest date and within the leaching data are not significantly different at P = 0.05 by DMRT

Losses of nitrogen in leachate

The amount of [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] leached was very low in all treatments and no [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] was leached after week 5; thus, the leaching data presented are the total of [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. Leaching loss of mineral N from all residue treatments was highest in Week 1, with the highest amount leached at this time from the non-residue treatment followed by the medic hay, Flemingia leaf, and wheat straw treatments (data not presented). After week 4 the rate of leaching of mineral N declined gradually through to Week 12. The amount of N leached from the non-residue treatment was generally greater than from the residue treatments at most times except Week 4, and this resulted in the greatest amount of mineral N being lost from this treatment. The smallest loss of [sup.15]N was from the wheat straw treatment. Fertiliser rate had no effect on the leaching of mineral N.

In all 3 residue treatments, the leaching loss of [sup.15]N gradually increased from the beginning of the experiment and treatments reached their maxima at different times. After 12 weeks the highest cumulative [sup.15]N leaching loss of 6.7% was from the medic hay treatment, and this was significantly higher than that in the wheat straw and Flemingia leaf treatments (Fig. 1).

[Figure 1 ILLUSTRATION OMITTED]

Carbon in soil and leachate

Generally, the total C concentration ([C.sub.T]) was higher in the top soil layer, where the residue had been incorporated, than in the middle and bottom soil layers (data not presented). Within the top soil layer, application of plant residues significantly increased soil [C.sub.T] by [is greater than] 1.0 mg/g compared with the non-residue treatment.

There were significant differences between residues in the amount of C added in them that was lost in leachate throughout the experiment. The greatest loss of residue C in the leachate was in the medic and wheat straw treatments (Table 7).

Table 7. Percentage of residue C in each soil layer at 91 days as affected by residue and soil depth and percentage of residue C leached

Means within the soil layer data and within the leaching data followed by a common letter are not significantly different at P = 0.05 by DMRT

Discussion

Crop yields

The differences in decomposition rate (medic [is greater than] Flemingia [is greater than] wheat) of the 3 residues used in this study as measured by Lefroy et al. (1995) were not reflected in the millet yields measured here. At the early stage of growth, the poor millet yields from the treatment receiving medic hay compared with Flemingia leaf and wheat straw was most likely the result of higher organic acid production during the rapid decomposition period of the medic hay. Moreover, during the leachate collection period the pots were temporarily subjected to waterlogging (approximately 5-6 h) and this may have stimulated the accumulation of organic acids from rapidly decomposing residue (Cannell and Lynch 1984).

Despite the low contribution of residue N to the millet in the Flemingia leaf and wheat straw treatments (Table 5), millet yields were quite high in these treatments. This suggests that the N from the applied inorganic fertiliser had compensated for the lower N supplied from residues, especially during the early stage of crop growth. By maturity there were no differences in crop yields among the treatments receiving the 3 residues, but they produced a higher yield than that of the control treatment. Generally, application of residues improved crop yields over the control by approximately 44-70%. Results from this study support the findings of many workers that application of plant residues improves crop yield (Broadbent 1984; Ta and Fads 1990; Bremer and van Kessel 1992; Wonprasaid et al. 1995).

Movement of nutrients and C

The recovery of [sup.35]S in the plant, soil, and leachate was near 100%, while the recovery of [sup.15]N ranged from 85% (medic hay) to 90% (wheat straw). The loss of residue N (10-15%) was most likely due to ammonia ([NH.sub.3]) volatilisation and/or denitrification occurring during the decomposition process (Floate 1970; De Datta 1995). The amounts of N unaccounted for in the present study are in agreement with the findings of Broadbent and Nakashima (1974), who reported 15-17% loss of residue N.

A higher percentage of residue N and S was found in the leachate and in plant uptake from treatments receiving medic hay. Compared with the Flemingia leaf treatment, leachate N and S from the medic hay treatment was 3- and 6-fold higher, respectively, while plant uptake of N and S from the medic hay treatment was 3- and 2-fold higher than uptake from the Flemingia leaf treatments. The high N and S concentration of medic hay (2.76% and 0.29%, respectively), in addition to its rapid decomposition rate, resulted in greater amounts of N and S and the higher percentages of total N and S released during the decomposition process. Once released the nutrients were readily available for plant utilisation. However, the high level of nutrient release led to a nutrient supply greater than plant demand and this resulted in the movement of excess nutrients down the soil column with subsequent loss through leaching, as found by Becker et al. (1994). As a consequence of greater plant uptake and N and S losses through leachate, the total soil N and S was lower with the application of medic hay than that with the Flemingia leaf treatment. This suggests a greater residual value with the slower breakdown material.

Application of wheat straw led to a high residue S but low residue N recovery in leachate, with a significant amount of both residue N and S utilised by the millet. This suggests that most of the N released from wheat straw was utilised by the millet crop, and this resulted in a low loss of residue N through leaching. By contrast, the high S concentration in the wheat straw (0.23%) resulted in a high loss of residue S through leaching, which was similar to that observed from medic hay. This confirms the results from the perfusion study of Lefroy et al. (1995) which showed that the release of S was high from wheat straw.

In contrast to medic hay and wheat straw, application of Flemingia leaf resulted in the highest percentage of residue S and N remaining in the soil, with lower amounts of nutrients found in the leachate and in the plant (Table 4). The residue N and residue S in soil resulting from Flemingia leaf application were both 1.25 times higher than that observed from the medic hay treatment. This resulted from the slow decomposition of this residue as shown in the perfusion study of Lefroy et al. (1995). The slow decomposition rate resulted in a slower release of nutrients, and as a result, a significant amount of nutrients remained in undecomposed residue and soil microbial biomass and less-extractable soil S. Subsequently, the loss of nutrients through leaching was reduced significantly compared with the rapid breakdown residue.

Considering the N concentration in the two legume residues, the N concentration in Flemingia leaf (3.70%) was higher than that of medic hay (2.76%), but more N was leached and taken up by the crop from medic hay. This indicates that a high residue-N concentration does not always result in increased N mineralisation of plant residues. In addition, the difference in N concentration between medic hay and Flemingia leaf indicated that the loss of fertiliser N through leaching was not proportional to the amount applied, but was dependent on the degree of decomposition of the residue, and the amounts of nutrient released in relation to plant demand. This supports similar findings by Stevenson (1986).

The rapid release of plant residue-S from both medic hay and wheat straw is presumably because of major S-containing components of these residues, such as protein and amino acids, are very labile to soil microorganisms and are decomposed prior to other more resistant fractions, such as cellulose (Stevenson 1986).

Application of the 3 plant residues had similar effects on the leaching of C to those on the leaching of nutrients. Application of medic hay and wheat straw lead to greater amounts of C loss through leaching at 91 days. These losses were approximately 21% higher than losses from the Flemingia leaf treatment. However, the direct impact of leaching of soluble C from these 3 residues was small, with maximum losses of 1.5% of the total C added to the soil. These results are higher than that in the study of Miller et al. (1994), who measured leaching of C from residue using simulated rainfall and reported that approximately 0.45-0.83% of organic C was leached from added plant residues.

In the Flemingia leaf treatment, a higher percentage of residue N and S remained in the soil than in the other treatments. The soil in Flemingia leaf treatment also had a higher total soil C concentration than the other treatments, and there was a lower leaching loss of C from this soil. This indicates that residues and green manures with slow breakdown rates, such as Flemingia leaf, are a better choice to use as soil amendments to rehabilitate soil fertility and C in the longer term. Residues which breakdown more slowly result in a build-up in C and nutrients which will benefit succeeding crops. This suggestion is confirmed by this study and by the field study in Northeast Thailand by Wonprasaid et al. (1995), who reported the longer term benefits of using slow breakdown residues in the rehabilitation of soil C and nutrients.

It is concluded that the [Delta] [sup.13]C technique used in this study was barely sensitive enough to monitor changes in the sources of C in the various soil pools or in leachate. A number of options exist to improve measurement of C dynamics, particularly C movement in the soil profile and in leachate. The use of [sup.13]C and [sup.14]C labelling of residues would likely provide greater accuracy than the natural abundance technique used here.

Acknowledgments

The work presented in this paper has been made possible by the financial support of the Australian Centre for International Agricultural Research (ACIAR) through Projects 9102 and 9448. Technical assistance in taking soil samples, sample analyses, and performing experiments from Mrs Leanne Lisle, Mrs Judi Kenny, and Mr Michael Faint is also appreciated.

References

Adamsen FJ, Bigelow DS, Scott GR (1985) Automated methods for ammonium, nitrate and nitrite in 2 M KCl-phenylmercuric acetate extracts of soil. Communications in Soil Science and Plant Analysis 15, 175-182.

Anderson DL, Henderson L (1986) Sealed chamber digestion for plant nutrient analyses. Agronomy Journal 78, 937-939.

Becker M, Ladha JK, Ottow JCG (1994) Nitrogen losses and lowland rice yield as affected by residue nitrogen release. Soil Science Society of America Journal 58, 1660-1665.

Blair GJ, Chinoim N, Lefroy RDB, Anderson GC, Crocker GJ (1991) A sulfur soil test for pastures and crops. Australian Journal of Soil Research 29, 619-626.

Blair GJ, Lefroy RDB, Lisle L (1995) Soil carbon fractions based on their degree of oxidation, and the development of a carbon management index for agricultural systems. Australian Journal of Agricultural Research 46, 1459-1466.

Blair G J, Till AR, Boswell CC (1994) Rate of recycling of sulfur from urine, faeces and litter applied to the soil surface. Australian Journal of Soil Research 32, 543-554.

Bremer E, van Kessel C (1992) Plant-available nitrogen from lentil and wheat residues during a subsequent growing season. Soil Science Society of America Journal 56, 1155-1160.

Broadbent FE (1984) Nitrogen release and carbon loss from soil organic matter during decomposition of added plant residues. Soil Science Society of America Proceedings 12, 246-249.

Broadbent FE, Nakashima T (1974) Mineralization of carbon and nitrogen in soil amended with carbon-13 and nitrogen-15 labelled plant material. Soil Science Society of America Proceedings 38, 313-315.

Brooks PD, Stark JM, McInteer BB, Preston T (1989) Diffusion method to prepare soil extracts for automated nitrogen-15 analysis. Soil Science Society of America Journal 53, 1707-1711.

Cadisch G, Giller KE (1996) Short Communication: Estimating the contribution of legumes to soil organic matter build up in mixed communities of C3/C4 plants. Soil Biology and Biochemistry 28, 823-825.

Cannell RQ, Lynch JM (1984) Possible adverse effects of decomposing crop residues on plant growth. In `Organic matter and rice'. pp. 455-475. (International Rice Research Institute: Los Banos, Philippines)

Catchpoole DW (1988) The contribution of tree legumes to the nitrogen economy and forage production in the humid topics. PhD Thesis. The University of New England, Armidale, NSW.

Catchpoole DW, Blair GJ (1990). Forage tree legumes. III. Release of nitrogen from leaf, faeces and urine derived from leucaena and gliricidia leaf. Australian Journal of Agricultural Research 41, 539-547.

Cerri CC, Volkoff B, Andreaux F (1991) Nature and behaviour of organic matter in soils under natural forest, and after deforestation, burning and cultivation, near Manaus. Forest Ecology and Management 38, 247-257.

De Datta SK (1995) Nitrogen transformations in wetland rice ecosystems. Fertilizer Research 42, 193-203.

Floate MJS (1970) Decomposition of organic materials from hill soils and pastures. II. Comparative studies on the mineralization of carbon, nitrogen and phosphorus from plant materials and sheep faeces. Soil Biology & Biochemistry 2, 173-185.

Jensen ES (1991) Evaluation of automated analysis of [sup.15]N and total N in plant material and soil. Plant and Soil 133, 83-92.

Lefroy RDB, Konboon Y, Blair GJ (1995) An in vitro perfusion method to estimate residue plant residue breakdown and associated nutrient release. Australian Journal of Agricultural Research 46, 1467-1476.

Martin A, Mariotti A, Balesdent J, Lavelle P, Vuattoux R (1990) Estimate of organic matter turnover rate in a savanna soil by [sup.13]C natural abundance measurements. Soil Biology & Biochemistry 22, 517-523.

Miller MH, Beauchamp EG, Lauzon JD (1994) Leaching of nitrogen and phosphorus from the biomass of three cover crop species. Journal of Environmental Quality 23, 267-272.

O'Brien BJ (1984) Soil organic carbon fluxes and turnover rates estimated from radiocarbon enrichments. Soil Biology & Biochemistry 16, 115-120.

Oweczkin IJ, Kerven GL, Ostatek-Boczynski Z (1995) Determination of dissolved organic carbon by inductively coupled plasma atomic emission spectrometry. Communications in Soil Science and Plant Analysis 26, 2739-2747.

Stevenson FJ (1986) `Cycles of soil: Carbon, nitrogen, phosphorus, sulfur, micronutrients.' (John Wiley and Sons: New York)

Ta TC, Faris MA (1990) Availability of N from [sup.15]N-labelled alfalfa residues to three succeeding barley crops under field conditions. Soil Biology & Biochemistry 22, 835-838.

Technicon (1977) Nitrate and nitrite in soil extracts. Industrial Method No. 487-77A.

Wonprasaid S, Naklang K, Khonthasuvon S, Mepetch S, Hemthanon B, Tippayaruk S, Lefroy RDB (1995) Organic matter residue management in lowland rice in Northeast Thailand. In `Soil organic matter management for sustainable agriculture'. (Eds RDB Lefroy, GJ Blair, ET Craswell) Ubon, Thailand, 24-26 August 1994. pp. 98-103. (ACIAR: Canberra)

Manuscript received 30 June 1999, accepted 13 December 1999

Yothin Konboon(A), Graeme Blair(B), Rod Lefroy(C), and Anthony Whitbread(B)

(A) Ubon Rice Research Center, PO Box 65, Ubon Ratchathani, 34000 Thailand.

(B) Agronomy and Soil Science, University of New England, Armidale, NSW, 2351 Australia.

(C) IBSRAM, PO Box 9-109, Jatujak, Bangkok, 10900 Thailand.
Soil characteristics                Value

Carbon (%)                            1.09
Nitrogen (%)                          0.07
[Delta][sup.13]C ([salinity])       -14.35
[sup.15]N (atom%)                     0.3680
Labile C (mg/g)(A)                    1.04
Non-labile C (mg/g)                   9.83
Extractable S KCl-40 ([micro]g/g)     6.1
Total Soil S ([micro]g/g)            93.7
Colwell P ([micro]g/g)               38.6


Plant residues

Properties                      Flemingia leaf   Medic hay

Carbon (%)                           47.25         44.04
Nitrogen (%)                          3.70          2.76
Phosphorus (%)                        0.25          0.20
Potassium (%)                         1.24          3.29
Sulfur (%)                            0.23          0.29
C : N ratio                          12.8          16.0
[sup.15)N (atom%)                     2.47          3.46
[Delta][sup.13]C ([salinity])       -24.91        -26.32

                                  Plant
                                 residues

Properties                      Wheat straw

Carbon (%)                         41.52
Nitrogen (%)                        0.36
Phosphorus (%)                      0.25
Potassium (%)                       1.24
Sulfur (%)                          0.23
C : N ratio                       116.6
[sup.15)N (atom%)                   8.17
[Delta][sup.13]C ([salinity])     -26.14


Fertiliser
                              application

Harvest       Residue         N       NPKS

Day 27    Non-residue       0.19b     0.25b
          Flemingia leaf    0.22b     0.42a
          Medic hay         0.19b     0.24b
          Wheat straw       0.21b     0.41a
Day 48    Non-residue       0.94d     1.48cd
          Flemingia leaf    2.35b     3.40a
          Medic hay         2.51b     1.96bc
          Wheat straw       1.91bc    2.55a
Day 91    Non-residue       5.54e    10.05cd
          Flemingia leaf   10.16cd   17.16a
          Medic hay        13.40bc   13.92ab
          Wheat straw       7.68de   15.30ab


% Recovery of [sup.35]S
                        in millet

Plant residue    Day 27   Day 48   Day 91

Flemingia leaf    0.7a     5.0a    10.0b
Medic hay         0.9a     7.4a    21.6a
Wheat straw       1.5a     8.1a    21.4a

                 % Recovery of [sup.35]S   % [sup.35]S

Plant residue    Top     Middle   Bottom     leached

Flemingia leaf    5.2b    1.6b     0.7b       2.3b
Medic hay        11.6a    4.4a     2.0a       7.4a
Wheat straw      11.3a    4.4a     2.4a       8.7a


Harvest time

Plant residue    Day 27   Day 48   Day 91

Flemingia leaf    0.5a     3.2b     4.5c
Medic hay         1.8a    11.4a    15.8a
Wheat straw       1.1a     4.9b     9.8b


Harvest

Residue          Soil layer   Day 27   Day 48   Day 91

Flemingia leaf   Top          86.2a    80.5a    79.2a
                 Middle        1.7b     2.0b     1.6b
                 Bottom        0.7b     1.7b     0.7b
                 Leached                         1.0b

Medic hay        Top          80.8a    62.7a    56.9a
                 Middle        4.8b     4.9b     5.2b
                 Bottom        1.3b     1.9b     1.1b
                 Leached                         6.6a

Wheat straw      Top          79.1a    75.6a    72.0a
                 Middle        5.0b     4.3b     5.0b
                 Bottom        2.7b     1.9b     2.1b
                 Leached                         1.6b


Soil layer

Residue           Top    Middle   Bottom   Leached

Flemingia leaf   7.4a     1.2b     3.5ab    0.4b
Medic hay        2.9ab    4.5a     0.0b     1.5a
Wheat straw      4.7a     2.1b     1.5b     1.3a
Gale Copyright: Copyright 2000 Gale, Cengage Learning. All rights reserved.