BSES Limited FINAL REPORT – SRDC PROJECT BSS199 IMPROVING THE MANAGEMENT OF ACID AND SODIC SOILS WITH GREEN TRASH RETENTION USING CALCIUM-BASED AMELIORANTS/PRODUCTS by B.L. SCHROEDER1,4, A.D. NOBLE2,4, F.A. ROBERTSON 1,4 , P.N. NELSON2,4 and A.W. WOOD3,4 SD03019 1BSES Limited, Ashfield Rd, Bundaberg, Qld 4670. 2Formerly CSIRO Land and Water, Davies Laboratory, PMB Aitkenvale, Qld 4814. 3CSR Sugar, PMB 4, Ingham, Qld 4850. 4CRC for Sustainable Sugar Production, James Cook University, Townsville, Qld 4870. Contact: Dr Bernard Schroeder BSES Limited Private Bag 4 Bundaberg DC Q 4670 Phone: 61 7 4132 5200 Fax: 61 7 4132 5253 Email: bschroeder@bses.org.au The research organisations are not partners, joint venturers, employees or agents of SRDC and have no authority to legally bind SRDC, in any publication of substantive details or results of this Project. BSES Limited Publication SRDC Final Report SD03019 November 2003 Copyright © 2003 by BSES Limited All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of BSES Limited. Warning: Our tests, inspections and recommendations should not be relied on without further, independent inquiries. They may not be accurate, complete or applicable for your particular needs for many reasons, including (for example) BSES Limited being unaware of other matters relevant to individual crops, the analysis of unrepresentative samples or the influence of environmental, managerial or other factors on production. Disclaimer: Except as required by law and only to the extent so required, none of BSES Limited, its directors, officers or agents makes any representation or warranty, express or implied, as to, or shall in any way be liable (including liability in negligence) directly or indirectly for any loss, damages, costs, expenses or reliance arising out of or in connection with, the accuracy, currency, completeness or balance of (or otherwise), or any errors in or omissions from, any test results, recommendations statements or other information provided to you. CONTENTS Page No SUMMARY......................................................................................................................... i 1.0 BACKGROUND..................................................................................................... 1 2.0 OBJECTIVES......................................................................................................... 1 3.0 PARTIES INVOLVED AND LINKAGES .......................................................... 3 4.0 METHODOLOGY................................................................................................. 3 5.0 EFFECTS OF TRASH RETENTION ON SOIL ACIDITY.............................. 5 5.1 Introduction ................................................................................................ 5 5.2 Net H+ input from existing trash retention trials .................................... 6 5.2.1 Methods and materials................................................................... 6 5.2.2 Results and discussion.................................................................... 7 5.3 Influence of changed management of sugarcane on chemical properties .................................................................................................. 13 5.3.1 Methods and materials................................................................. 13 5.3.2 Results and discussion.................................................................. 15 5.4 Assessment of the effect of water-soluble organic components leached from plant materials on the cation mobilisation in acid light-textured soils .................................................................................... 22 5.4.1 Methods and materials................................................................. 22 5.4.2 Results and discussion.................................................................. 24 5.5 Conclusions ............................................................................................... 31 6.0 TRASH DECOMPOSITION, SOIL SODICITY AND THE EFFICACY OF CALCIUM-BASED AMENDMENTS FOR AMELIORATING SODIC SOILS .................................................................................................................... 32 6.1 Introduction .............................................................................................. 32 6.2 Trash decomposition and soil sodicity.................................................... 33 6.2.1 Methods ......................................................................................... 33 6.2.2 Results ........................................................................................... 36 6.2.3 Discussion...................................................................................... 39 6.3 Effects of lime, gypsum, trash and their interactions on sodic soils and sugarcane growth.............................................................................. 41 6.3.1 Methods and material .................................................................. 41 6.3.2 Results and discussion.................................................................. 42 6.4 Structural stability of sodic soils as influenced by gypsum/molasses .. 44 6.4.1 Materials and methods................................................................. 44 6.4.2 Results ........................................................................................... 46 6.4.3 Discussion...................................................................................... 55 6.5 Structural stability of sodic soils as influenced by molasses and lime. 56 6.5.1 Materials and methods................................................................. 56 6.5.2 Results ........................................................................................... 57 6.5.3 Discussion...................................................................................... 65 6.6 Effect of trash management on susceptibility of soils to becoming or remaining sodic......................................................................................... 66 6.6.1 Materials and method .................................................................. 66 6.6.2 Results and discussion.................................................................. 68 6.7 Conclusions ............................................................................................... 70 7.0 EFFECT OF CALCIUM-BASED AMENDMENTS ON TRASH DECOMPOSITION ............................................................................................. 71 7.1 Introduction .............................................................................................. 71 7.2 Effect of various calcium amendments on trash decomposition and soil amelioration ....................................................................................... 72 7.2.1 Method and materials .................................................................. 72 7.2.2 Results and discussion.................................................................. 75 7.3 Combined effect of calcium-based amendments and trash retention on nutrient availability............................................................................. 89 7.3.1 Methods and materials................................................................. 89 7.3.2 Results and discussion.................................................................. 90 7.4 Conclusions ............................................................................................... 98 8.0 TRASH DECOMPOSITION AND C AND N RELATIONSHIP ................... 99 8.1 Introduction .............................................................................................. 99 8.2 Measuring trash decomposition in controlled conditions..................... 99 8.2.1 8.2.2 8.2.3 Methods ....................................................................................... 100 Results ......................................................................................... 102 Discussion.................................................................................... 106 8.3 Trash decomposition and effects of trash, mill mud, mill ash and urea on soil C and N............................................................................... 108 8.3.1 Methods ....................................................................................... 108 8.3.2 Results ......................................................................................... 111 8.3.3 Discussion.................................................................................... 117 8.4 Conclusions ............................................................................................. 119 9.0 OUTPUTS ........................................................................................................... 120 9.1 Effects of trash retention on soil acidity............................................... 120 9.2 Trash decomposition, soil sodicity and the efficacy of calciumcontaining amendments and trash for ameliorating sodic soils......... 121 9.3 Effect of calcium-based amendments on trash decomposition .......... 122 9.4 Trash decomposition and C and N relationships ................................ 124 9.5 General .................................................................................................... 125 10.0 EXPECTED OUTCOMES................................................................................ 125 11.0 FUTURE RESEARCH NEEDS AND RECOMMENDATIONS .................. 126 12.0 PUBLICATIONS ARISING FROM THE PROJECT ................................... 127 13.0 ACKNOWLEDGEMENTS............................................................................... 127 14.0 REFERENCES ................................................................................................... 128 SUMMARY Although the advantages of leaving cane residues on the harvested land have been well documented, and nutrient cycling in relation to green-cane trash blanketing has been well examined within a Cooperative Research Centre for Sustainable Sugar Production (CRC Sugar) program activity, little work has been done to investigate the effects of trash management on soil chemical properties and the ameliorative advantages of applying Cacontaining amendments to crop residues. As lime, lime/gypsum applications are generally used to ameliorate acid soil conditions, a Ca-source followed by leaching is the principal strategy for managing sodic soils, and mill by-products (mill mud and mill ash) are often applied to plant cane, it was considered pertinent to investigate the effects of these and other amendments on the decomposition of trash on sugar industry soils. This project aimed at quantifying the effects of trash retention on soils acidity, examining the efficacy of Ca-containing amendments and trash for ameliorating sodic soils, assessing the effect of Ca-based amendments on trash decomposition and nutrient availability, and quantifying trash decomposition and C and N relationships. To do this, the project comprised four distinct but interlinked facets and included a series of laboratory investigations, glasshouse experiments and a field trial. Green-cane trash retention has an effect on soil properties. Increases in organic carbon due to trash retention result in improved charge characteristics of the soil, but with acidic cations (H+ and Al3+) predominantly retained on the exchange complex. Lime application would, however, neutralise this effect and result in an increased capacity of the soil to retain other nutrients, such as K+, Ca2+ and Mg2+. As the ash alkalinity of sugarcane is small compared to that of other litter materials, trash does not contribute significantly to neutralising acidity generated by nitrogen mineralisation. Decomposition of sugarcane is generally unaffected by soil sodicity. Cane trash and ameliorants such as molasses and soil organic matter are all important in the management of sodic soils. All forms of organic additions examined improved soil structural properties and reduced sodicity or the tendency of soil to become or remain sodic. There were no major interactions between organic matter and Ca-containing amendments on ameliorating sodic conditions; they all had positive effects irrespective of each other. Calcium-based amendments such as lime and mill mud have the potential to accelerate the decomposition of sugarcane trash when applied to the surface of the trash layer. When comparing the effectiveness of the amendments in assisting decomposition, millmud repeatedly outperformed the other amendments. Other amendments, such as Ca(OH)2 and KOH, did have the ability to enhance the decomposition process to some extent, but they are not considered suitable for widespread use in the industry due to their corrosive nature. Importantly, the enhanced trash decomposition that occurred due to lime millmud, mill ash and lime applied to trash in pots under semi-controlled conditions was also found to be applicable in the field. The differences in the rates of decomposition in the simulated subtropical and moist tropical conditions suggest that environmental conditions involving temperature and the amount of water (rain and/or irrigation) are important factors within this trash/ameliorant system. ii Lime applied to the surface of the trash layer did have the ability to ameliorate acidic conditions in the underlying soil to some extent. This has important implications for certain soils when minimum tillage is practiced. Although millmud and mill ash applications to trash resulted in significantly higher amounts of N, P and K removed by the harvested biomass, these elevated values could either have been the result of the nutrients within the amendments themselves, or due to enhanced decomposition of the trash, or a combination of both of these effects. The effect on nutrient uptake by sugarcane when lime was applied to trash was less apparent. Environmental effects (water, temperature and soil properties) are again considered important factors in relation to the overall nutrient availability in these amended trash systems The small-scale decomposition experiments showed that decomposition of trash used in such trials is unaffected by factors such as drying, rough grinding or whether the trash is green or senesced. However, as decomposition is affected by soil moisture and placement, such factors need to be carefully controlled. Importantly, measurement of decomposition as CO2 loss appears to be a more reliable and sensitive method than measuring dry matter remaining in such small-scale investigations. Although millmud can contribute relatively large amounts of N, applications in the presence of trash may reduce N availability initially due to microbial immobilisation. Generally, the amendments, such as lime, millmud and millmud/mill ash mixtures applied to the surface of a trash layer have positive results for sugarcane production. Lime applied to trash has the ability to counter acidity generated by the mineralisation of N from the organic matter. Sources of organic matter (trash, millmud, molasses, etc), whether present individually or in combination with each other, have the ability to ameliorate sodic soil conditions. Ameliorants such as lime and millmud have the potential to accelerate trash decomposition when applied to the surface of the trash layer. Ultimately more efficient return of organic matter to the soil will improve amelioration of soil acidity and sodicity, and result in improved nutrient- use efficiency. Although some technology transfer ha occurred in terms of papers, presentations and posters, further extension of the concept is proposed. In particular, a paper entitled ‘Cane trash and Ca-based soil amendments: A combination too good to miss’ will be presented at the 2004 ASSCT conference. Further extension will make use of grower meetings and workshops. 1.0 BACKGROUND With the move away from burning sugarcane to the retention of crop residue in much of the Australian sugar industry, major implications for water, carbon and nutrient dynamics have been identified and investigated in various research programs. However, little attention has been directed towards investigating the short and longer-term impacts of trash layers on changes in chemical properties. In particular, there have been reports in literature of both increases and decreases in soil pH with the addition of organic matter to agricultural soils. In this light, local evidence suggested that decreases in soil pH have occurred in at least some of the long-term trash-retention trials conducted at various sites in the industry. Such increases in soil acidity could have important implications for both the acid and sodic sugarcane-producing soils. A decrease in soil pH associated with trash could have negative implications for already acid soils, but could be advantageous to alkaline sodic soils. In both cases it would be pertinent to consider the use of Cacontaining amendments (lime and/or gypsum) in association with trash for ameliorative purposes. In acid soils, the surface applications of lime (or lime/gypsum) to trash would be considered important to counteract any increases in soil acidity. However, lime could also increase the decomposition rate of trash and possibly contribute to improved nutrient use efficiency in such systems. In sodic soils, the organic acids and carbon dioxide produced during decomposition of trash may improve the efficacy of various Ca-containing amendments for improving soil physical conditions. 2.0 OBJECTIVES This project aimed to develop a set of recommendations for cane growers to ensure full utilisation of the interactive processes between trash and Ca-containing amendments. The specific objectives of the project were to: 1. Quantify the effects of trash retention on soil acidity and sodicity; 2. Determine the influence of Ca-containing amendments on trash decomposition and nutrient availability under various environmental conditions; 3. Determine the influence of trash decomposition on the efficacy of Ca-containing amendments in improving acid and sodic soils; 4. Assess strategies for maximising the beneficial interactions between trash and Ca- containing amendments. All the objectives of the project have been achieved. The underlying theme of the project was to ensure the optimum utilisation of the interaction between trash and Ca-containing amendments to enable sustainability, which encompasses both productivity and environmental issues. This implied that strategies would be sought to counteract any negative effects on soils properties due to green-cane trash blanketing (GCTB) and to translate any positive interactive effects into increased productivity. 2 The methodology used in this project consisted of four separate, yet interlinked facets (Figure 1) that aimed to answer the following questions: • Does the presence of trash affect soil pH? • If so, why? • What process is involved? • What liming strategies are applicable for countering any negative effects? • What strategies could be used to harness any positive effects? • Would these strategies have an effect on nutrient availability? Objective 1 – Quantify the effects of trash retention on soil acidity. Facet (i) of the project was specifically aimed at determining the effect of trash retention on the chemical properties of the underlying soil. This section of the project was achieved by quantifying the net proton input associated with long-term burn/trash trials, determining changes in soil properties of the soils associated with the two systems, and assessing the effect on soil of the water-soluble organic and inorganic components derived from plant litter. Objective 2 – Determining the influence of Ca-containing amendments on trash decomposition and nutrient availability in improving acid and sodic soils. Facets (iii) and (iv) of this project were aimed at determining the influence of Cacontaining amendments on trash decomposition. A range of amendments was assessed and included lime, gypsum and the important sugar industry by-products – millmud and mill ash. Other possible amendments were also included in the studies. The assessments involved pot experiments and laboratory investigations and a field trial. Importantly, lime and millmud applied to the surface of the trash layer were found to be useful in ameliorating the underlying soil and capable of improving the nutrition of sugarcane grown within such trash amended systems. Objective 3 – Determining the influence of trash on the efficacy of Ca-containing amendments in improving acid and sodic soils. Long-term trash/burn trials provided the opportunity to study the effects of trash retention on soil chemical properties. Increases in organic C with trash retention and corresponding increase in charge characteristics inherently improves nutrient retention, especially in lowCEC tropical soils. Laboratory based experiments were important in assessing the quality of leachates originating from litter materials. Facets (ii), (iii) and (iv) dealt with various aspects of the interaction between trash and Ca-containing amendments. It appears that millmud applied to the surface of a trash blanket has a number of positive results (accelerated decomposition, ameliorative effects and possible improvements in nutrientuse efficiency). Objective 4 – Assessing strategies for maximising the beneficial interactions between trash and Ca-containing amendments. Facet (iii) in particular investigated practical means of maximizing the benefits of applying Ca-containing amendments to trash. Amendments such as lime, millmud and mill ash (routinely available in the sugar industry) were assessed. Each of these in 3 combination with trash was found to have beneficial effects. These amendments applied to the surfaces of trash blankets have shown to have ameliorative, nutritional and environmental benefits. 3.0 PARTIES INVOLVED AND LINKAGES This project was established as a joint initiative involving scientists from BSES, CSIRO Land and Water, and CSR Sugar Ltd. The common association of the participating scientists was the CRC for Sustainable Sugar Production (CRC Sugar). Importantly, the scientists that were involved had experience in aspects of trash management, trash decomposition, soil science, soil fertility and crop nutrition. The project had important linkages to several other projects, funded either by CRC Sugar or SRDC. These included: • CRC Sugar funded: 2.3.8 – Opportunities for better nutrient supply and uptake; • CRC Sugar funded: 2.2.2.1 – Nutrient cycling in relation to trash-management systems; • CRC Sugar funded: 2.3.3 – Different aspects of soil acidity; • CRC Sugar funded: 2.3.2 – A tool kit for improved diagnosis and management of sodic soils; • SRDC funded: BSS197 – Products and mechanisms for amelioration of sodic soils; • SRDC funded: CSR24 – Improving the environment for sugarcane growth through the amelioration of soil acidity. 4.0 METHODOLOGY As indicated in the schematic representation of the project (Figure 1) the project was divided into four facets within two main sections. The first section related to the effect of trash retention (Facet i.). The second section related to aspects of Ca-containing amendments applied to trash (Facets ii, iii and iv). The four facets aimed at: i. Quantifying the effects of trash retention on soil acidity; ii. Examining the efficacy of Ca-containing amendments and trash for ameliorating sodic soils; iii. Assessing the effect of Ca-based amendments on trash decomposition; iv. Quantifying trash decomposition and C and N relationships. The responsibility for the various facets lay with different scientists working within the project. The names of the appropriate persons are shown in Figure 1. The methodology associated with each of these facets (i, ii, iii and iv) is detailed within relevant sections (Sections 5, 6, 7 and 8, respectively). 4 Section 1: Trash retention AD Noble and FA Robertson i. Quantify effects of trash retention on soil acidity by: a) Determining the H+ input from existing trash retention trials b) Evaluating the influence of changed management of sugarcane on soil chemical properties c) Assessing the effect of watersoluble organic components leached from plant material on cation mobilisation Section 2: Application of Cacontaining amendments Sodic soils PN Nelson ii. Examining the efficacy of Cacontaining amendments and trash for ameliorating sodic soils by: a) Determining the effect of soil sodicity on trash decomposition b) Determining the effects of lime, gypsum, trash and their interactions on sodic soils and sugarcane growth c) Evaluating the influence of gypsum/molasses on the structural stability of sodic soils d) Determining the effect of trash management on susceptibility of soils to becoming or remaining sodic Acid soils BL Schroeder and AW Wood iii. Assessing the effect of Ca-based amendments on trash decomposition by: a) Determining the rate of decomposition of trash in the presence of such amendments b) Assessing the combined effect of Ca-based amendments and trash retention on nutrient availability Figure 1 Acid and sodic soils RA Robertson, AD Noble and PN Nelson iv. Quantifying trash decomposition and C and N relationships by: a) Measuring trash decomposition in controlled conditions b) Determining trash decomposition and effects of trash, millmud, mill ash and urea on soil C and N Schematic representation of the various facets of the project 5 5.0 EFFECTS OF TRASH RETENTION ON SOIL ACIDITY 5.1 Introduction Under intensive sugarcane production systems, continual acidification of the soil resource base is likely to occur through the use of ammoniacal nitrogen fertilisers inputs, base removal as a result of crop export, and the general high rainfall environment under which sugarcane is grown that facilitates the losses of basic cations through leaching (Moody and Aitken 1995). Whilst Australian sugarcane cultivars are relatively tolerant of acidity (i.e. low pH and Al phytotoxicity), they are highly intolerant of low levels of soluble Ca2+ (Hurney 1971; Ridge et al. 1980; Haysom et al. 1986; Hetherington 1986). Reghenzani (1993) in a survey of soil nutrient levels estimated that 67% of the soils in north Queensland had suboptimal levels of exchangeable Ca2+, and the remediation of a Ca2+ deficiency can result in significant increases in yield (Hurney 1971; Ridge et al. 1980; Haysom et al. 1986; Quinan and Wood 1989, 1993; Noble et al. 2000). It has been clearly shown that long-term sugarcane monoculture results in a gradual decline in soil fertility that is associated with soil acidification (Bramley et al. 1996; Garside et al. 1997; Moody and Aitken 1997). Whilst the alleviation of acid soil infertility of surface horizons by regular additions of lime effectively neutralises a high proportion of acidity generated through anthropogenic activities, the remediation of subsoil acidity is more problematic. Considerable work has been conducted in which lime has been mechanically mixed into the soil to depth (0.5-1 m) in order to neutralise acidity (McKenzie and Nyborg 1984; Hammel et al. 1985). Such treatments have included: chisel ploughing in combination with rotary tillage (Doss et al. 1979); subsoil lime injectors both mechanical (Anderson and Hendrick 1983) and pneumatic (Saayman and van Huyssteen 1981); mouldboard ploughing (Lutz and Jones 1971); and various mechanical procedures in combination with lime and gypsum (Farina and Channon 1988). These methods have all shown that the neutralisation of subsoil acidity has been effective in establishing a more hospitable rooting environment, which has usually translated into increased yields particularly in years of below average rainfall. The economic efficacy of such treatments is determined by commodity price structures, soil type and tillage costs and may not be a feasible option for all situations (Sumner 1995). However, in view of the residual value of lime, the economics of lime application, whether for top or subsoil amelioration, should be considered over more than one crop year (Mallawaarachchi et al. 1998). The adverse effects of long-term sugarcane monocultures have resulted in a decline in the productive capacity of soils and, hence, productivity (Garside et al. 1999). In 1993, a series of rotation break trials was established on long-term sugarcane monoculture fields to determine if rotations caused changes in the chemical, physical and biological properties of the soil that would reverse the negative effects of declining yield (Garside et al. 1997). Significant increases in cane and sugar yield have been observed following the implementation of a break crop management strategy (Garside et al. 1999). With respect to aspects of changes in soil fertility associated with GCTB or the influence of crop rotations under Australian production systems, the focus has been on rates of decomposition, the carbon and nitrogen cycles, and changes in the chemical status of soils (Spain and Hodgen 1994; Robertson and Thorburn 2000). Of importance in assessing the 6 impact of changed management practices on the soil resource base is quantification of possible changes in exchange properties of soils associated with different management regimes. As part of this study, changes in chemical attributes and surface charge characteristics of soils associated with two forms of changed management in a sugar production system were quantified. Three interlinked but separately initiated studies were undertaken to: 1. Quantify the net proton input associated with long-term trash retention systems through the evaluation of changes in soil-chemical attributes from long-term burn/trash retention trials from a range of sites within the eastern sugar-growing belt. 2. One of the dominant impacts of trash retention on soil chemical properties is the enhanced organic matter content of soils. Changes in soil-chemical properties were quantified using two contrasting production systems that are currently being promoted in the Australian sugar industry; long-term trash retention versus burnt system, and a grass-ley production system versus continuous monoculture of cane. In addition to the standard chemical attributes measured, surface charge was characterised, as this is one of the dominant soil attributes affected by organic matter conservation. 3. Assess the effect of water-soluble organic and inorganic components emanating from different surface-applied plant materials (intact or ashed) on the mobility of cations and the remediation of acidity. In addition, the chemical composition of the watersoluble leachate prior to and after leaching through a soil column was compared. 5.2 Net H+ input from existing trash retention trials 5.2.1 Methods and materials Site characteristics and treatment Burnt and GCTB systems were compared in three field sites at three locations along the eastern sugar belt; these have contrasting climatic conditions (Table 1). The Tully and Mackay sites were established in 1990 and 1992, respectively, whilst the Harwood site was established in 1995. Robertson and Thorburn (2000) give further details of their management. All three trials were planted to cultivar Q124. Table 1 Selected site characteristics of trials used in the assessment of net proton input under trash retention and burn management systems Location Year established Date sampled Soil texture Mean annual rainfall (mm/yr) Previous cropping history Tully 17.56 S; 145.56 E 1990 November 1997 Clay 4067 Sugarcane Mackay 21.10 S; 149.07 E 1992 August 1998 Loam 1668 Sugarcane Harwood 29.50 S; 153.20 E 1995 January 1998 Clay/loam 1021 Vegetables, sugarcane 7 Soil samples Soil samples were collected by taking 12 cores (3.0-3.4 cm internal diameter) in two regular transects from the top of each row and across the inter-row, with each core being sectioned into the depth layers 0-2, 2-5, 5-10 and 10-25 cm. Soils were air dried and passed through a 2-mm sieve. Ash alkalinity determinations Composite sugarcane trash samples were collected from each of the three trial sites and submitted for chemical analysis and ash alkalinity. Samples were dried to a constant weight at 60°C in a forced-draft oven before being ground to pass through a 2-mm mesh and then Tema milled to a powder. The elemental composition of the leaf material was determined by X-ray florescence spectrometry (XRF) on four replicated pressed pellets of the milled trash (Norrish and Hutton 1977). The ash alkalinity was determined by the difference between total cations and anions: Σ (Ca2+ + Mg2+ + K+ + Na+) - Σ (SO42- + H2PO4- + Cl-). pH buffer capacity determination, organic carbon and particle size analysis We used a modified discontinuous batch method of James and Riha (1986) to determine the acid and base neutralizing capacity (ANC and BNC, respectively) of soils collected from each of the trial sites and depth intervals. The determination of pH buffer capacity was undertaken on all samples collected. In brief, 2 g of air-dried soil that had been sieved to pass through a 2-mm mesh were shaken in polypropylene tubes containing a total liquid volume of 25 mL and included 0, 0.5, 2.0, 4.0, 10.0 or 20.0 mL of 0.001M HNO3 or 0.001M KOH, 1.0 mL KNO3 and 250 µL chloroform. After equilibrating on an end-over-end shaker for 22 h at 25°C, the suspensions were centrifuged and the supernatant pH measured. pH buffer curves were constructed for both the ANC and BNC by taking the inverse of the linear section of plots of acid/base addition versus pH. Organic carbon (OC) content and particle size distribution were determined on composite samples from each depth interval for each treatment. Organic carbon was determined by wet oxidation using the Walkley and Black (1934) method as described in Method 6A1 in Rayment and Higginson (1992). Particle size was determined using the methodology of Coventry and Fett (1979). 5.2.2 Results and discussion Ash alkalinity determination The simplest method of determining the ash alkalinity in plant material is by titration of an acidified solution containing the ashed sample with standardised base (Jarvis and Robson 1983). Whilst this method is sufficient in most cases to give an accurate assessment of this parameter, it has the inherent potential error associated with the volatilisation of S and Cl during the ashing process (Jungk 1968; Kennedy 1992). Under such circumstances, losses of these anions would result in an over-estimation of the ash alkalinity. A far more accurate assessment of ash alkalinity can be undertaken by subtracting the total cations from the total anion complement of the plant material as determined by XRF. The percent concentrations of both cations and anions are converted to their respective equivalent concentrations of the species presented in equation 1. Ash alkalinity (cmolc/kg) = Σ(Ca2+ + Mg2+ + K+ + Na+) - Σ(SO42- + H2PO4- + Cl-) (1) 8 The elemental composition and the ash alkalinity of the three trash samples of Q124 collected from the three trial sites are presented in Table 2. The Ca concentration from the trash samples collected from the Mackay site were higher than the other two sites, whilst the K concentration was higher in the Tully samples. The Si concentration in the Tully sample was considerably lower than those sampled at the Mackay and Harwood sites, possibly reflecting the higher leaching environment under which cane is produced (Table 2). There was a highly significant simple linear relationship between the Ca concentration in the stalk and ash alkalinity over all sites (Figure 2). Table 2 Elemental composition and ash alkalinities of cane trash samples collected from each of three trial sites. Each value is the mean of four replicates Site Ca Mg K Na Al Si P (mg/g) Cl SO4 S Total S Mackay 3.06 1.31 3.49 0.48 1.99 25.9 0.44 1.31 0.33 0.34 0.64 Harwood 1.95 1.34 2.50 0.01 2.01 26.3 0.34 0.81 0.28 0.34 0.59 Tully 1.56 1.08 6.28 0.01 3.17 14.5 0.49 2.66 0.35 0.28 0.60 Site Mackay Zn Cu Fe Mn Ash alkalinity ppm (cmolc/kg) 32 24 2124 236 28.12 Harwood 27 14 1760 183 20.80 Tully 24 15 1174 183 20.46 Figure 2 Ash alkalinity (cmol/kg) 32 30 28 26 24 22 20 18 0.10 0.15 0.20 0.25 0.30 Ca concentation (%) 0.35 Relationship between Ca concentration in the stalk and ash alkalinity of samples collected from Harwood, Mackay and Tully. Equation: Ash alkalinity (cmol/kg) = 11.16 (± 1.46) + 54.32 (± 6.24)Ca conc. (%): r2=0.894. 9 Ash alkalinity has been used extensively to describe the acid-base balance in foods, grains and fodder crops and is a measure of the organic-anions that are produced internally by the plant in order to balance excess cation uptake. It is the decomposition of these organic anions that influence the proton pool. The ash alkalinity of the trash material can be expressed as CaCO3 equivalents if it is assumed that 1 mol H+ requires 50 g CaCO3 to be neutralised. Consequently, the net amount of alkalinity in the form of CaCO3 equivalents associated with the trash ranges from 10.23 to 14.6 kg CaCO3/tonne dry matter. These rates of alkalinity addition on an annual basis through trash retention, are small compared to the litters collected from some plant species (i.e. Melia azedarach 126 kg CaCO3/tonne dry matter) (Noble et al. 1996) and, therefore, do not make a significant contribution to the net acid input on an annual basis. Robertson (1998) reported net annual trash additions to the Harwood, Mackay and Tully sites of 8.76, 11.50 and 6.93 t/ha, respectively. Based on a mean CaCO3 equivalent of 12.41 kg/t of dry matter, the total potential alkalinity release from trash applied would be equivalent to 108, 142 and 86 kg/ha/yr for the three sites. Table 3 pH buffer capacity values for soils collected from the three trash management trials. Location Mackay Mackay Mackay Mackay Mackay Mackay Mackay Mackay Harwood Harwood Harwood Harwood Harwood Harwood Harwood Harwood Tully Tully Tully Tully Tully Tully Tully Tully Treatment GCTB GCTB GCTB GCTB Burnt Burnt Burnt Burnt GCTB GCTB GCTB GCTB Burnt Burnt Burnt Burnt GCTB GCTB GCTB GCTB Burnt Burnt Burnt Burnt Depth (cm) 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 Acid Base Mean pHBC pHBC pHBC (mmol H+ OH+/kg.unit pH) 25.4 32.8 29.1 22.8 28.0 25.4 26.3 26.5 26.3 28.2 26.3 27.2 22.0 25.6 23.8 22.8 24.2 23.5 24.5 24.2 24.3 27.2 24.6 25.9 45.6 55.8 50.7 49.6 50.0 49.8 51.0 49.7 50.3 49.7 49.2 49.4 44.8 50.2 47.5 52.1 48.7 50.4 50.2 48.7 49.4 52.1 49.0 50.5 34.6 28.5 31.5 40.2 22.3 31.2 41.5 20.2 30.8 41.7 19.6 30.6 35.5 19.9 27.7 38.3 19.6 28.9 37.9 20.6 29.2 40.8 19.5 30.1 10 pH buffer capacity determination Soil pH buffer capacity (pHBC) values for each of the depth intervals collected at the three sites are presented in Table 3. pHBC is calculated from the inverse of the slope of the linear region of curve of the plot between pH (y) and acid/base added (x). This linearity occurs between pH 4 and 6. An example of the ANC and BNC curves for the 02 cm depth interval of the Mackay soil is presented in Figure 3. pH pH 6.6 6.4 6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 4.6 0 4.8 (a) 10 20 30 40 50 60 OH added (mmol/kg) 4.6 4.4 4.2 4.0 (b) Figure 3 3.8 0 5 10 15 20 H+ added (mmol/kg) 25 pH buffer curves for (a) base and (b) acid additions for the Mackay 0-2 cm under a green cane trash blanket treatment pHBC values in the surface 0-2 cm were generally slightly higher under the GCTB treatments when compared to the burnt. These differences diminished with depth. The 11 pHBC values as determined by either acid or base addition were similar in the case of the Mackay and Harwood sites, but noticeably different in the case of the Tully site (Table 3). This could not be explained on the basis of clay content, as both the Tully and Harwood samples had similar clay contents (Table 4). Whilst no mineralogical characterisation was undertaken of samples, it is plausible that this disparity may in part be associated with the presence of sesquioxidic surfaces that are common to highly weathered tropical soils. As the pH of these soils declines, ANC increases through the consumption of protons with a concomitant increase in positive surface charge. This would result in a higher capacity to buffer shifts in pH associated with acid addition. Table 4 Soil organic carbon (OC) and particle size distribution of soils collected for pH buffer capacity determination from Mackay, Harwood and Tully Location Mackay Mackay Mackay Mackay Mackay Mackay Mackay Mackay Harwood Harwood Harwood Harwood Harwood Harwood Harwood Harwood Tully Tully Tully Tully Tully Tully Tully Tully Treatment GCTB GCTB GCTB GCTB Burnt Burnt Burnt Burnt GCTB GCTB GCTB GCTB Burnt Burnt Burnt Burnt GCTB GCTB GCTB GCTB Burnt Burnt Burnt Burnt Depth (cm) 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 0-2 2-5 5-10 10-25 Mean pHBC (mmol/kg.unit pH) 29.1 25.4 26.3 27.2 23.8 23.5 24.3 25.9 50.7 49.8 50.3 49.4 47.5 50.4 49.4 50.5 31.5 31.2 30.8 30.6 27.7 28.9 29.2 30.1 OC 2.5 1.5 1.3 1.1 1.5 1.3 1.2 1.2 3.2 2.7 2.6 2.4 3.0 2.6 2.5 2.7 3.1 1.7 1.4 1.3 1.6 1.4 1.3 1.3 Clay 15.9 17.7 18.5 18.1 15.6 17.4 17.9 18.1 35.6 37.6 37.1 38.1 34.3 35.0 37.0 36.9 33.0 37.3 40.7 39.3 33.3 36.6 36.7 37.8 Silt C Sand (%) 27.0 12.6 26.7 10.9 24.6 11.1 24.8 11.6 27.2 13.8 25.4 13.1 25.4 12.9 26.3 11.5 46.7 4.4 46.9 3.7 47.3 3.7 46.9 3.5 44.5 4.8 45.3 5.4 43.1 6.1 44.1 5.7 38.8 7.0 36.8 6.2 32.0 6.3 35.9 5.6 32.9 6.1 33.2 5.4 32.7 5.4 31.9 5.3 F Sand 44.4 44.6 45.8 45.5 43.5 44.1 43.8 44.1 13.3 11.8 11.9 11.5 16.4 14.4 13.8 13.4 21.2 19.7 21.0 19.4 27.6 24.7 25.1 25.0 It is generally agreed that pHBC is a function of soil pH and clay and organic matter content. To determine whether these routinely measured soil attributes could explain measured mean, acid and base pHBC, we undertook simple linear regression analysis. Based on the above difference in the buffer capacity associated with the addition acid or base between the Tully soil and both the Harwood and Mackay soils, the latter two soils were grouped as a single data set and Tully data treated separately. Regression and correlation coefficients are presented in Table 5. It is clearly evident that in the combined Harwood/Mackay data set, regardless of whether mean, acid or base pHBC was used, OC 12 and clay content were highly significantly (P < 0.001) correlated (Table 5). Clay content tended to explain a greater degree of the variation than OC. In the case of the Tully data set, acid pHBC was best described by clay content. In contrast, base pHBC was best described by OC (Table 5). This would tend to support the above argument based on differences in mineralogy between the soils from the wet tropics and those from regions outside this area. In the case of the Tully site, the soil appears to be exhibiting ‘variable charge’ characteristics. The term ‘variable charge’ is used to describe organic and inorganic soil constituents with reactive surface groups whose charge varies with pH, ionic concentration and composition of the soil solution. Such groups are the surface carboxyl, phenolic and amino functional groups of organic materials in soils, and surface hydroxyl groups of Fe and Al oxides, allophane and imogolite. The hydroxyl surface groups are also present on edges of some phyllosilicate minerals such as kaolinite, mica, and hydroxyl-interlayered vermiculite. The variable charge is developed on the surface groups as a result of adsorption or desorption of ions that are constituents of the solid phase, i.e. H+, and the adsorption or desorption of solid-unlike ions that are not constituents of the solid. This explains the strong relationships between acid pHBC and clay (i.e. sesquioxide surfaces) and the base pHBC and OC, as organic matter has only ability to buffer increases in OH-. Table 5 Regression coefficients for selected linear regressions for mean, acid and base pHBC for the Tully (n=8) and combined Harwood/Mackay (n=16) data sets. Values in parenthesis are the standard errors Variance Site pHBC Constant OC (%) Clay (%) accounted for by model (%) Harwood/ Mean 6.31 (4.41) 15.10 (2.00) 78.8*** Mackay Mean 4.14 (1.51) 1.25 (0.05) 97.4*** Mean 2.61 (0.97) 3.91 (0.78) 1.00 (0.06) 99.0*** Acid 7.29 (5.75) 14.34 (2.61) 66.0*** Acid 2.60 (1.63) 1.28 (0.06) 97.1*** Acid 2.57 (1.79) 0.07 (1.44) 1.28 (0.11) 96.9*** Base 5.33 (3.25) 15.86 (1.47) 88.4*** Base 5.69 (2.56) 1.21 (0.08) 92.4*** Base 2.65 (0.90) 7.74 (0.73) 0.73 (0.05) 99.2*** Tully Mean 28.49 (1.37) 0.95 (0.79) 5.9 ns Mean 23.96 (7.06) 0.16 (0.19) - Mean 6.11 (4.48) 2.48 (0.46) 0.54 (0.11) 81.4** Acid 43.73 (2.32) -3.00 (1.34) 36.7 ns Acid 2.79 (4.52) 0.98 (0.12) 90.0*** Acid 5.59 (7.24) -0.39 (0.75) 0.92 (0.17) 88.6** Base 13.26 (0.78) 4.89 (0.45) 94.3*** Base 45.10 (14.40) -0.65 (0.39) 19.9 ns Base 6.63 (5.56) 5.35 (0.57) 0.16 (0.13) 94.7** ** significant at the p < 0.01 level; *** significant at the p < 0.001 level; ns non-significant. 13 Consequently, it could be argued that deriving a pedotransfer function to estimate pHBC from routinely determined soil attributes (i.e. clay and organic carbon content) over a wide range of agroecological zones would be problematic where there are soils of mixed mineralogy. Gillman and Sumpter (1986) observed that in soils of the wet tropics exhibiting variable charge characteristics, approximately 50% of the lime added to raise the pH of a soil suspensions from 4 to 6 was used in generating cation exchange capacity and not in directly increasing soil pH. Consequently, the variable charge nature of certain soils (i.e. ferrosols) will have a significant effect on pHBC. There are several potential errors that are inherent in the estimation of pHBC capacity that have been outlined by Helyar and Porter (1989). For example, the measurement of pH buffer capacity is in general derived from short-term laboratory incubation studies that do not take into account the slow pH buffering reactions (i.e. the dissolution of solid phases; biological processes) (Ridley et al. 1990). To overcome problems associated with short-term incubation studies, pH buffering capacity has been determined through field trials where pH changes have been monitored over a long period of time (Helyar et al. 1990). However, errors associated with incomplete dissolution or uneven mixing of lime has been suggested. In addition, acidification may occur on the unlimed treatments at different rates to that of limed treatments (Porter et al. 1995). Clearly the method used to determine pHBC will directly influence the ability to predict the quantity of lime required to neutralize acidity or to achieve a pre-determined pH level. It is generally agreed that short-term equilibration methods may underestimate buffering capacity and, therefore, lime requirement. However, pHBC is a dynamic characteristic and, therefore, any estimated measure is subject to ongoing change. 5.3 Influence of changed management of sugarcane on chemical properties 5.3.1 Methods and materials Green Cane Trash Blanketing / Burnt trial A long-term GCTB/burn trial was established in July 1990 at BSES Tully, north Queensland (17º59'S; 145º56' E). The climate of this region is tropical with a mean annual rainfall of approximately 4300 mm, concentrated in October-March. The treatments of interest in this evaluation were (a) GCTB: trash after mechanical harvesting retained on the soil surface with no soil cultivation and (b) Burnt: trash burnt before and after harvest and soil cultivated to a depth of 15 cm after harvest. The soil was classified as an acidic dystrophic yellow dermosol (Isbell 1996) or an oxic humitropept (Soil Survey Staff 1990). The trial was a completely randomized block design with three replications and was planted to Q124 with an annual application of 180 kg N/ha as urea. Seven 7 crops were harvested from the site before the trial was ploughed out and replanted. Soil samples were collected in October 1997, prior to ploughout, from 10 random points within a plot at depth intervals of 0-2, 2-5 and 5-10 cm and bulked to form composite depth samples. Rotation trial The second trial, also in the Tully area, was a rotation experiment established to investigate crop-management strategies to reduce the effects of yield decline in long-term sugarcane monocultures. Only three treatments of interest in this study are discussed in detail; Garside et al. (1999) give a comprehensive discussion of the trial. The three 14 treatments that were selected for this study were: (a) the long-term continuous GCTB; (b) pasture ley; and (c) a bare-fallow treatment. The site had been under continuous cane with minimal fallow breaks for the 20 years prior to the imposition of the trial. In November 1993, cane was removed from plots to be established to the pasture/legume ley consisting of a mixture of signal grass (Brachiaria decumbens) and pinto peanut (Arachis pintoi). No fertilizer was applied to the pasture during the study. The pasture was managed by periodic slashing, leaving the residue on the plots. Prior to each slashing, pasture growth was measured by quadrat sampling with dry matter production of tops being about 20 t/ha/yr. The bare fallow plot was maintained vegetation-free through periodic application of herbicide during the study. The continuous-cane plot was maintained over six consecutive crops. The soil was a yellow kandosol (Isbell 1996). The trial was a completely randomized block design with three replications. Prior to the implementation of cane treatments in May 1994, soil samples (0-10 cm depth interval) were collected from all treatments at 30 individual points to form a composite sample for each plot. In August 2000, prior to the plough-out of all treatments, three random soil samples were collected from the 0-10 cm depth interval and bulked to form a composite sample for each plot. Chemical analysis Prior to chemical analysis, soils were air-dried and ground to pass through a 2-mm sieve before pH (1:5 water) was measured. Basic exchangeable cations were determined by atomic absorption spectrometry after replacement with 0.1M BaCl2-NH4Cl (Gillman and Sumpter 1986). Acidic cations were extracted with 1M KCl and the extractant titrated to pH 8.0 as described in Method 15G1 (Rayment and Higginson 1992). The effective cation exchange capacity (ECEC) was calculated as the sum of basic and acidic cations (Ca2++Mg2++K++Na++Al3++H+). Organic carbon was determined by wet oxidation using the Walkley and Black (1934) method as described in Method 6A1 in Rayment and Higginson (1992). Composite samples were produced for each of the depth intervals and treatments from both studies. Charge fingerprints, which are curves describing the total cation exchange capacity (CECT), base cation exchange capacity (CECB) and anion exchange capacity (AEC) across a range of pH values were determined on samples using the methodology described by Gillman and Sumpter (1986). In brief, soils were Ca saturated and brought to equilibrium in a 0.002M CaCl2 matrix. Suspension pH was adjusted to six values ranging from approximately 4.0 to 7.5, and exchangeable Ca and Al displaced with NH4NO3. The Al content in solution was determined using the pyrocatechol-violet method (Bartlett et al. 1987) and Ca by atomic absorption spectrometry. Amounts adsorbed were calculated taking into account the amounts present in the entrained solutes. The CECB is operationally defined as the Ca adsorbed, and CECT as the Ca and Al adsorbed. Modifications to the method of Gillman and Sumpter (1986) were made so that the amount of acid or base added to achieve the desired pH range was quantified. This effectively allowed an estimation of the pH buffering capacity (pHBC) to be made on the composite samples. Dr M. Braunack, formerly of BSES, undertook soil bulk density measurements at each of the sites independently using the standard methodology of Blake and Hartge (1986). 15 To assess the possible contribution of particular functional organic groups to charge generation, we carried out carbon fractionation on the above composite samples using 33mM and 333mM KMnO4 to discriminate between the oxidisability of the soil organic C (Blair et al. 1995). Conteh et al. (1999) showed that differential ease of oxidation was correlated with the proportional amounts of various functional organic groups in the soil organic matter. Total organic C was determined by dry combustion at 1300°C in a Leco carbon analyzer. Table 6 Selected soil chemical properties after 7 years of treatments from a long-term trash retention/burnt cane trial established at Tully, north Queensland. Values in parentheses are the SE of the mean Parameter pHw Ca (cmolc/kg) Mg (cmolc/kg) K (cmolc/kg) Na (cmolc/kg) Exchangeable acidity (cmolc/kg) W-B organic C (g/kg) Total organic C (g/kg) 33mM KMnO4 oxidisable C (C1) (g/kg) 333mM KMnO4 oxidisable C (C3) (g/kg) pH buffer capacity (cmol H+/kg.pH unit) ECEC (cmolc/kg) Bulk density (g/cm3) Treatment GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt GCTB Burnt 0-2 5.11 (0.01) 5.33 (0.01) 2.49 (0.31) 1.86 (0.40) 1.03 (0.08) 0.76 (0.08) 0.39 (0.02) 0.30 (0.01) 0.11 (0.01) 0.12 (0.001) 1.31 (0.23) 0.91 (0.06) 25.0 (1.6) 13.6 (2.1) 30.0 14.7 3.04 0.93 6.74 2.43 2.79 1.92 5.32 (0.30) 3.95 (0.54) 1.13 1.17 Depth (cm) 2-5 5.13 (0.02) 5.33 (0.01) 1.96 (0.26) 1.77 (0.24) 0.77 (0.02) 0.66 (0.00) 0.19 (0.01) 0.20 (0.05) 0.09 (0.01) 0.09 (0.01) 1.41 (0.25) 1.01 (0.01) 15.6 (0.4) 12.8 (0.6) 16.4 13.2 1.21 0.80 3.95 2.02 2.22 1.97 4.42 (0.14) 3.74 (0.20) 1.28 1.24 5-10 5.26 (0.03) 5.41(0.07) 1.88 (0.26) 1.70 (0.49) 0.60 (0.02) 0.51 (0.06) 0.10 (0.01) 0.14 (0.03) 0.09 (0.01) 0.07 (0.01) 1.44 (0.28) 1.06 (0.16) 12.2 (0.4) 11.6 (1.1) 12.5 12.3 0.71 0.67 1.86 2.28 2.03 2.01 4.11 (0.06) 3.48 (0.38) 1.25 1.26 5.3.2 Results and discussion Changes in soil organic carbon, pH buffer capacity and exchangeable cations GCTB/Burnt trial: Significant decreases in soil pH were recorded over the depth interval 0-5 cm under the trash retention system (Table 6). The largest difference in soil pH between the two management systems occurred in the 0-2 cm depth interval where a difference in pH of 0.22 units was recorded over the period of the trial (7 years). This pH decrease is to be expected, since this is the zone of greatest biological activity where the processes of mineralization and, hence, acid generation are at their greatest. There was a trend for increases in exchangeable Ca2+, Mg2+, K+ and ECEC in the surface 0-2 cm depth under the GCTB treatments, significant (P < 0.05) only in the case of K (Table 6). 16 However, these differences diminished with depth. Associated with the decrease in pH, there was a corresponding increase in exchangeable acidity under the GCTB system when compared to the burnt (Table 6). Soil organic carbon levels in the top 0-2 cm increased substantially under the GCTB system, but this increase did not persist with depth. The amount of total soil organic carbon in the 10-cm depth interval of the GCTB system was equivalent to 20.9 t/ha whilst that under the burnt regime amounted to 16.1 t/ha. Consequently, a total of 4.8 t/ha of total organic carbon accumulated under the GCTB system over 7 years. The pH buffering capacity of soils was highest in the 0-2 cm depth interval under the GCTB system and declined with depth until it was the same as that for the burnt treatment at 5-10 cm (Table 6). This increase in buffering capacity can be attributed to elevated soil organic carbon levels in the GCTB system, as there was a significant (P < 0.05) positive correlation between total organic C and pH buffer capacity (Figure 4). Using the pH buffer capacity of the GCTB treatment for each of the depth intervals, the net proton input in the GCTB treatment relative to the burnt was estimated to be 0.71 kmol/ha/yr using equation (2) (Helyar and Porter 1989). NAAR = [[(pHB x pHBCB) – (pHT x pHBCT)] x BD x V]/T (2) where the subscripts B and T refer to burn and trash retained production systems; pHBC is the pH buffering capacity for each of the production systems (kmol H+/kg.pH unit); BD is soil bulk density (kg/m3); V is the soil volume in the depth interval under consideration (m3/ha); and T is time (years) since the establishment of the trial. pH buffer capacity (mmol H+/kg.pH unit ) 3 y = 0.0466x + 1.3882 R2 = 0.9343 2.5 2 Figure 4 1.5 y = 0.0306x + 0.9949 R2 = 0.459 1 10.0 15.0 20.0 25.0 30.0 Total organic C (mg/kg) 35.0 Relationship between pH buffer capacity and total organic C for the GCTB/Burnt Trial (▪) and the Rotation Trial (♦) 17 Table 7 Changes in soil chemical properties (0-10 cm) associated with the conversion from continuous cane to a grass ley system 6 years after imposition of treatments at Feluga, north Queensland. Values in parentheses are the SE of the mean Parameter pHw Ca (cmolc/kg) Mg (cmolc/kg) K (cmolc/kg) Na (cmolc/kg) Exchangeable acidity (cmolc/kg) W-B organic C (g/kg)) Total organic C (g/kg) 33mM KMnO4 oxidisable C (C1) (g/kg) 333mM KMnO4 oxidisable C (C3) (g/kg) pH buffer capacity (cmol H+/kg.pH unit) ECEC (cmolc/kg) Bulk density (g/cm3) Treatment Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Cont. cane Ley Fallow Year 1994 2000 6.55 (0.19) 6.03 (0.21) 6.40 (0.41) 5.46 (0.23) 6.03 (0.13) 5.76 (0.09) 3.53 (0.54) 1.55 (0.14) 3.26 (0.72) 1.83 (0.30) 2.15 (0.11) 1.77 (0.37) 0.54 (0.06) 0.34 (0.01) 0.43 (0.05) 0.37 (0.01) 0.34 (0.01) 0.15 (0.01) 0.25 (0.02) 0.21 (0.02) 0.22 (0.01) 0.24 (0.01) 0.17 (0.02) 0.14 (0.01) 0.03 0.08 (0.02) 0.04 (0.04) 0.10 (0.01) 0.04 0.07 (0.01) 0.07 (0.01 0.09 (0.01) 0.13 (0.05) 0.24 (0.04) 0.20 (0.05) 0.26 (0.20) 12.6 (0.04) 7.7 (0.5) 12.6 (0.12) 14.3 (0.4) 12.3 (0.06) 7.9 (0.1) nd 14.2 nd 25.1 nd 11.5 nd 1.07 nd 2.12 nd 0.68 nd 2.21 nd 5.56 nd 1.41 nd 1.14 nd 1.76 nd 1.39 4.43 (0.61) 2.27 (0.05) 4.10 (0.68) 2.79 (0.20) 2.89 (0.16) 2.57 (0.22) nd 1.40 (0.04) nd 1.38 (0.04) nd 1.39 (0.03) 18 Rotation trial: Table 7 gives changes in selected soil chemical properties for the 0-10 cm depth intervals at the start and end of the study. At the end of the study (2000), soil pH was significantly higher (P < 0.05) under the continuous cane treatment at 0-10 cm when compared to the grass ley (Table 7). Differences in pH were not evident at the 10-20 cm depth interval (data not presented). In exchangeable basic cations there were no significant differences between treatments with depth (data not presented). The continuous cane treatment significantly (P < 0.05) decreased exchangeable Ca and Mg over the 6-year period, whereas exchangeable K levels did not change (Table 7). Pasture ley for 6 years (with return of all harvested material to the plots) did not cause any changes in the levels of exchangeable cations. After 6 years, exchangeable acidity was significantly (P < 0.05) higher under the pasture ley when compared to the continuous cane. Total soil organic carbon increased from 19.9 t/ha under the continuous cane to 34.6 t/ha under the pasture ley over the 0-10 cm depth interval, indicating a 14.7 t/ha difference in total organic C between the systems over the 6-year period. Associated with the increase in organic C under the grass ley system there was a corresponding increase in the pH buffer capacity (Table 7). Assuming that the pH buffer capacity over the intervening 6-year period did not change under a particular management system, the net acid inputs between 1994 and 2000 for the continuous cane, grass ley and fallow systems were 1.38, 3.81 and 0.87 kmol/ha/yr, respectively. Clearly, the greater net proton input under the grass ley system can in part be explained by the increase in the organic carbon content of the soil (Helyar and Porter 1989). Surface charge characteristics The concept of surface charge fingerprinting as a means of describing the electrochemical characteristics of a soil was first described by Gillman and Sumpter (1986). The approach distinguishes that portion of the cation exchange capacity (CEC) that retains basic cations, and predicts changes in CEC as soil pH and ionic strength are varied. To assess this, we determined charge fingerprints on a composite sample for treatments collected from the 0-2, 2-5 and 5-10 cm depth intervals in the case of the GCTB/Burnt Trial and the 0-10 and 10-20 cm depth intervals for the Rotation Trial (Figures 5 and 6, respectively). The CECB, as depicted in Figures 5 and 6, is a measure of the maximum amount of exchangeable basic cations that can be retained in exchangeable form, and is, therefore, of agricultural significance. In contrast, the total cation exchange capacity (CECT) is an estimate of the charge associated with both acidic and basic cations, and only becomes significant at pH values of < 5.5. It is clearly evident that there are substantial differences in the charge characteristics associated with different management systems. In both of the trials under review, there is a vertical upward shift in the charge curve that is more pronounced in the surface horizons but diminishes with depth. The impact of GCTB on charge characteristics is confined to the 0-5 cm depth interval and can be directly attributed to the increased soil organic matter content (Figure 7). Similarly, the effects of the grass ley on charge characteristics were confined to the 0-10 cm depth. At pH 5.5 there is a convergence in the two charge curves, indicating that the CEC is base saturated. This is of agronomic significance, since it clearly shows that by increasing the 19 pH to this value any potential problems associated with acidic cations (i.e. Al3+ and H+) would be eliminated and the base cation complement on the exchange complex maximised. 7 6 GCTB CECB GCTB CECT 5 Burnt CECB Burnt CECT 4 Exchange capcity (cmolc/kg) 3 2 (a) 1 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7 6 GCTB CECB GCTB CECT 5 Burnt CECB Burnt CEC T 4 Exchange capacity (cmolc/kg) 3 2 1 45 7 6 GCTB CECB GCTB CECT 5 Burnt CECB Burnt CECT 4 (b) 67 Exchange capacity (cmolc/kg) 3 Figure 5 2 (c) 1 4.0 4.5 5.0 5.5 6.0 6.5 7.0 pH0.002 Surface charge fingerprints of the (a) 0-2 cm, (b) 2-5 cm and (c) 5-10 cm depth intervals under a long-term GCTB and burnt system. CECT is represented by dotted lines and CECB by solid lines 20 Figure 6 Charge capacity (cmolc/kg) 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 4.0 4.0 3.5 3.0 Continous CECB Continous CECT Pasture ley CECB Pasture ley CECT 4.5 5.0 Continuous CECB Continuous CECT Pasture ley CECB Pasture ley CECT (a) 5.5 6.0 6.5 Charge capacity (cmolc/kg) 2.5 2.0 1.5 (b) 1.0 4.5 5.0 5.5 6.0 6.5 pH0.002 Changes in surface charge characteristics associated with the establishment of a pasture ley system in the (a) 0-10 cm and (b) 10-20 cm depth intervals 5 y = 0.0818x + 2.2069 R2 = 0.806 4 CECB pH 5.5 (c molc /kg) 3 y = 0.0507x + 1.5381 2 R2 = 0.7862 Figure 7 1 10.0 15.0 20.0 25.0 30.0 Total Organic C (g/kg) 35.0 Relationships between CECB at pH 5.5 and total organic C in the 0-10 cm layer for the GCTB/Burnt Trial ( ■ ) and the Rotation Trial ( ♦ ) 21 To quantify the degree of charge generation associated with changed management, CEC values should be compared at the same pH (i.e. pH 5.5). For the 0-2, and 2 - 5 cm depth intervals the CEC increases between the burnt and GCTB systems were 1.50 and 1.11 cmolc/kg, respectively (Figure 5). However, the increase in CEC between the continuous cane and ley pasture system over the 0-10 cm was 0.75 cmolc/kg (Figure 6). By taking a weighted average of the CEC over a 10 cm depth interval, the increase in CEC between the burnt and GCTB treatments reduces to 0.67 cmolc/kg. The increase in CEC associated with the changed management can be attributed to changes in soil organic carbon, as there was a significant (P < 0.05) positive correlation between total organic C and the interpolated CECB at pH 5.5 (Figure 7). Data from the two sites lie on separate regression lines, with the GCTB/Burnt Trial having a higher CECB at a given total organic C content than the Rotation Trial. Extrapolation of the regression lines to zero total organic C indicates that the charge associated with clay minerals in the two soils equates to 2.2 cmolc/kg and 1.5 cmolc/kg for the GCTB/Burnt Trial and the Rotation Trial, respectively. This difference is attributed to the different clay contents of the 0-10 cm layers of the two sites, 36.8 % and 20.2 %, respectively. Figure 8 ECEC (cmolc/kg) 6 5 4 1:1 3 2 1 123456 CECT (cmolc/kg) Relationship between effective cation exchange capacity (ECEC) and the total cation exchange capacity (CECT) as determined from the surface charge fingerprint for all sites and depth intervals at soil pH An important characteristic of the charge fingerprint is that it allows an estimation of CECB and CECT at soil pH. Consequently, if the basic and acidic cations removed by the BaCl2/NH4Cl and KCl extractants, respectively, are all exchangeable cations, then their sum (the ECEC) should be equal to CECT at soil pH, within the limits of the experimental error associated with these estimations. A comparison of these two parameters for all data from both trial sites clearly shows that in all cases the ECEC is larger than CECT at soil pH, indicating that greater amounts of cations were extracted than could be accounted for by the CECT (Figure 8). The magnitude of these differences ranged from 9-31% of the actual measured CEC. This would suggest that a component of the cation complement that is extracted by 0.1M BaCl2/NH4Cl is not held on the exchange complex and is, 22 therefore, potentially subject to leaching. A similar discrepancy between CECT and ECEC was observed by Noble et al. (2000) on soils sampled during the dry season. We speculate that, since these samples were taken during the drier part of the year, mineralisation of organic matter may have occurred, resulting in an accumulation of soluble cations in the surface horizons. This clearly shows one of the advantages of constructing charge fingerprints since it highlights that a component of the extractable cations are in a soluble form and, therefore, vulnerable to leaching. Organic carbon fractionation Organic C oxidisable by 33mM KMnO4 (C1) or 333mM KMnO4 (C3) was highly linearly correlated with total organic C across both sites and all treatments [r2 = 0.98 and 0.95, respectively (P < 0.001)]. This suggests that the proportions of different functional organic groups were not greatly affected by burning, trash retention or pasture. The similarity in composition of the soil organic C between sites and across treatments is also evident in that the slopes of the linear regressions between CECB and total organic C (TOC), and between pH buffer capacity (pHBC) and TOC for the two sites are not significantly (P < 0.05) different (Table 8). Table 8 Linear regression parameters for relationships between CECB at pH 5.5 and total organic C and between pH buffer capacity and total organic C. Standard errors are in parentheses Relationship CECB versus total organic C GCTB/Burnt Trial Rotation Trial pHBC versus total organic C GCTB/Burnt Trial Rotation Trial Slope 0.082 (0.020) 0.059 (0.019) 0.047 (0.006) 0.033 (0.026) Intercept 2.21 (0.35) 1.54 (0.32) 1.388 (0.109) 0.948 (0.446) r2 0.81 0.79 0.93 0.46 5.4 Assessment of the effect of water-soluble organic components leached from plant materials on the cation mobilisation in acid light-textured soils 5.4.1 Methods and materials Plant residues Plant residues used in this study and region from where they were collected are given in Table 9. The species were diverse, in that they represent different plant orders. In this respect the grasses (monocots) are represented by sugarcane and Urochloa mosambicensis; leguminous dicots represented by Stylosanthes collected from the semiarid tropics; and freshly fallen leaf-litter material collected from the tree species Melia azedarach. Melia azedarach was used as a benchmark species, since it has been recognised as having a high ash alkalinity (Noble et al. 1996). Samples were dried at 65°C for 48 h in a forced-draft oven and ground to pass through a 2-mm sieve. 23 Table 9 Type and collection location of plant residues used in leaching studies Plant material Sugarcane trash White cedar (Melia azedarach) senesced leaf material Stylosanthes (Seca) whole-plant material Urochloa mosambicensis Collection location Herbert River District, north Queensland Canberra, ACT Lansdown, north Queenland Lansdown, north Queensland Preparation of leaching columns The surface 0-15 cm of an acid light-textured sand collected was collected from the Herbert River district, north Queensland. Particle size analysis showed that soil was dominated by sand and was classified as having a sandy loam texture. The site had supported sugarcane for the past 30 years. Soil was air dried and sieved to pass through a 2-mm mesh. The leaching columns were 15-cm long perspex tubes with an internal diameter of 2.5 cm. Columns were packed to a depth of 12 cm at a uniform bulk density of 1.48 g/cm3. Each column contained 87 g of soil and a total pore volume of 18.61 cm3. Prior to packing the column, a 2.5 cm Millipore GF filter pad was placed at the bottom of the column to prevent soil from falling out of the column. Once the column was packed, a Millipore GF filter was placed on the soil surface. Preparation of the litter material A mixture of litter and acid washed sand was placed on the top of the Millipore GF filter. This mix consisted of 1.227 g (equivalent to 50 t/ha dry matter) of litter and 5 g of acid washed sands thoroughly mixed. A third Millipore GF filter was placed on top of the surface of the litter/sand mix to distribute the distilled water evenly over the surface of the soil within the column and prevent surface impact. Columns were leached with distilled water over an 8-h period with a total of 5.4 pore volumes. After this leaching phase, the surface placed trash/sand mixture was carefully removed from the surface of the column to expose the GF filter separating the soil from the litter/sand mixture. A further 20 mL of distilled water was passed through the column. The effluent emanating from the bottom of the column was collected and the pH and EC measured at the completion of the leaching phase. The above procedure was undertaken for each of the litter materials and was replicated three times. In addition, a sample of the trash/sand mixture was leached with 5.4 pore volumes of distilled water and the leachate collected. This represented the composition of the leaching emanating from the trash/sand mixture prior to passing through the soil column. An additional litter treatment was prepared with cane trash, where the sand-trash mixture was treated with KOH. This treatment was undertaken to assess the effect of applying alkaline materials on the composition of the leachate emanating from the trash. The trash/sand mixture was prepared by taking 6.14 g of sugarcane trash and mixing it with 25 g acid washed sand. 0.30 g KOH was mixed with 20 mL of distilled water, added to the trash/sand, thoroughly mixed and allowed to equilibrate for 16 h. The mixture was then oven dried at 40°C over 24 h and the above leaching procedure undertaken. As a control treatment, columns were prepared with only soil and leached as described previous. 24 Soil and leachate analysis At the completion of the leaching phase, the columns were allowed to drain free and the soil gently blown out of the columns. The columns were sectioned into 2-cm sections from the top to the bottom of the column. Soils were dried at 40°C in a forced-draft oven for 48 h and passed through a 2 mm sieve. Soil pH and EC were measured in a 1:5 soil:water solution matrix. Exchangeable cations (Ca2+, Mg2+, K+) were extracted using 0.1M BaCl2/NH4Cl (Gillman and Sumpter 1986) and determined by atomic absorption spectroscopy. Exchangeable Al3+ was determined on the leachates using the pyrocatechol-violet method (Bartlett et al. 1987). The leachate extract was stored at 4°C until analysis. The dissolved organic carbon was determined using the modified Mebius method of Yeomans and Bremner (1988). The cation concentration was determined by atomic absorption spectroscopy. 5.4.2 Results and discussion Current practices of GCTB management dictate that a large volume of trash is returned to the soil surface after harvest. This trash blanket, which is allowed to breakdown over the course of the subsequent growing season, releases significant amounts of nutrients and organic carbon back into the soil. Contrasting this, in burnt-cane production systems there is a significant quantity of nutrients release on combustion that is effectively readily available for possible uptake by the crop or lost through leaching and runoff. It is important to quantifying the impact of these different management strategies on soil chemical properties. In undertaking this assessment, sugarcane trash should be compared with other litter materials that differ in chemical composition. Hence, litter materials that have contrasting chemical composition were evaluated. Effect of litter/trash material on soil properties and leachate composition Soil pH: As expected, there was no significant change in soil pH with depth in the Control treatment with leaching (Table 10). However, with the addition of litter material to the surface of columns and subsequent leaching, significant (P < 0.05) increases in pH were evident with depth down the column, although the largest increases were observed in the surface 0-6 cm (Table 10). The unamended cane litter had a significant effect on soil pH when compared with the Control, with an effect on pH to a depth of 6 cm. However, the greatest increase in pH with depth was achieved where KOH was mixed into the sugarcane trash, emulating the potential role of surface treating trash blankets with alkaline materials to change the characteristics of the litter. Soil pH was increased from 5.50 to 6.54 in the surface 0-2 cm in the cane trash and cane trash + KOH treatments respectively (Table 10). Contrastingly, Melia, Stylosanthes and Urochloa all resulted in significant increases in soil pH in the surface 0-2 cm, but these effects diminished with depth, with Melia showing the highest degree of persistence (Table 10). There appeared to be little difference in soil pH between the Stylosanthes litter collected from the soil surface and freshly harvested Stylosanthes plant material (Table 10). By taking the mean pH of the litter treatments over the entire depth of the column and subtracting it from the mean of the Control, an estimate of the degree of shift in pH associated with the treatments can be achieved and used to compare between ashed and non-ashed treatments (see below). These differences in pH were 0.29, 0.81, 0.66, 0.38, 25 0.38 and 0.47 for the cane trash, cane trash +KOH, Melia, Stylosanthes litter, Stylosanthes fresh and Urochloa, respectively. Table 10 Changes in soil pHw with depth after leaching with selected plant litter materials Treatment Depth (cm) 0-2 2-4 4-6 6-8 8-10 10-12 Control 4.96 5.01 5.01 5.02 4.95 4.99 Cane trash 5.50 5.34 5.26 5.22 5.21 5.19 Cane trash + KOH 6.54 6.12 5.97 5.67 5.35 5.16 Melia azedarach leaf litter 5.61 5.62 5.72 5.50 5.28 5.21 Stylosanthes litter collected from soil surface 5.62 5.43 5.32 5.32 5.28 5.27 Stylosanthes fresh material 5.68 5.45 5.29 5.22 5.17 5.13 Urochloa mosambicensis fresh material 5.97 5.72 5.49 5.29 5.15 5.15 LSD (0.05) 0.13 Exchangeable cations: Mean exchangeable Ca2+ content in the Control treatment was 0.36 cmolc/kg (Table 11). Contrastingly, the greatest increase in Ca2+ on the exchange complex was observed in the treatment leached with extracts emanating from the Melia litter, where there was an average of 1.30 cmolc/kg of Ca2+ down the entire profile. In previous leaching column studies using Melia azedarach litter material, similar changes in exchangeable Ca2+ levels have been observed, this being attributed to the high Ca content in that litter (Noble et al. 1996). There were little, if any, significant changes in Ca2+ content on the exchange complex with leaching in the other litters. Table 11 Changes in exchangeable Ca2+ with depth after leaching with selected plant litter materials Treatment Depth (cm) 0-2 2-4 4-6 6-8 8-10 10-12 Control 0.36 0.36 0.34 0.36 0.35 0.38 Cane trash 0.32 0.18 0.19 0.21 0.22 0.22 Cane trash + KOH 0.25 0.24 0.27 0.32 0.36 0.39 Melia azedarach leaf litter 1.81 1.49 1.38 1.25 0.98 0.90 Stylosanthes litter collected from soil surface 0.47 0.24 0.25 0.25 0.26 0.29 Stylosanthes fresh material 0.42 0.23 0.29 0.27 0.26 0.25 Urochloa mosambicensis fresh material 0.36 0.39 0.52 0.64 0.63 0.71 LSD (0.05) 0.25 With the exception of the cane trash + KOH treatment, exchangeable Mg2+ in the 0-2 cm depth significantly (P < 0.05) increased with leaching (Table 12). The largest increase occurred in the Urochloa, followed by the cane-trash treatments. In the Melia treatment, significant changes in Mg2+ were evident down the entire depth of the leaching column; this was probably associated with the elevated Mg2+ concentration in leachate (see 26 discussion below). Significant declines in the Mg2+ were observed in the cane trash + KOH treatment associated with elevated K+ concentrations in the leachate. Table 12 Changes in exchangeable Mg2+ with depth after leaching with selected plant litter materials Treatment Depth (cm) 0-2 2-4 4-6 6-8 8-10 10-12 Control 0.14 0.13 0.13 0.13 0.14 0.15 Cane trash 0.43 0.15 0.12 0.13 0.13 0.14 Cane trash + KOH 0.10 0.05 0.05 0.07 0.09 0.11 Melia azedarach leaf litter 0.23 0.21 0.23 0.24 0.22 0.23 Stylosanthes litter collected from soil surface 0.21 0.11 0.13 0.13 0.14 0.16 Stylosanthes fresh material 0.22 0.11 0.13 0.14 0.14 0.15 Urochloa mosambicensis fresh material 0.63 0.13 0.10 0.14 0.15 0.18 LSD (0.05) 0.04 The addition of the litter to the soil surface in the columns resulted in significant increases in exchangeable K+ to depth (Table 13). As expected, the greatest increase was observed in those treatments receiving cane trash + KOH, where additional K+ was supplied to the leachate (Table 13). Table 13 Changes in exchangeable K+ with depth after leaching with selected plant litter materials Treatment Depth (cm) 0-2 2-4 4-6 6-8 8-10 10-12 Control 0.13 0.13 0.14 0.14 0.14 0.16 Cane trash 0.37 0.39 0.33 0.26 0.19 0.18 Cane trash + KOH 1.82 1.21 1.06 0.77 0.54 0.39 Melia azedarach leaf litter 0.34 0.46 0.50 0.43 0.44 0.41 Stylosanthes litter collected from soil surface 0.36 0.33 0.26 0.19 0.18 0.20 Stylosanthes fresh material 0.56 0.36 0.26 0.21 0.19 0.18 Urochloa mosambicensis fresh material 0.86 0.75 0.47 0.29 0.18 0.17 LSD (0.05) 0.12 The sum of exchangeable bases is presented in Table 14; there has been a significant increase in the sum of basic cations over the control after leaching. Whilst exchangeable Al3+ was not determined on samples, part of the increase can be attributed to the displacement of Al3+ from the exchange complex through complexation with metaloorganic ligands and through precipitation within the column due to an increase in pH. In addition, soluble carbon that is retained in the column could potentially contribute to increased CEC, although as discussed below there was little retention within the column. 27 Table 14 Changes in sum of bases (sum Ca2+, Mg2+, K+) with depth after leaching with selected plant litter materials Treatment Depth (cm) 0-2 2-4 4-6 6-8 8-10 10-12 Control 0.65 0.66 0.64 0.68 0.66 0.73 Cane trash 1.13 0.73 0.65 0.61 0.55 0.56 Cane trash + KOH 2.20 1.52 1.41 1.19 1.02 0.92 Melia azedarach leaf litter 2.40 2.18 2.12 1.96 1.65 1.57 Stylosanthes litter collected from soil surface 1.08 0.73 0.69 0.61 0.64 0.71 Stylosanthes fresh material 1.25 0.76 0.75 0.69 0.65 0.63 Urochloa mosambicensis fresh material 1.89 1.31 1.15 1.11 1.00 1.12 LSD (0.05) 0.31 Leachate characteristics: EC, Ca, Mg and K levels were all significantly reduced in the leachate after passing through the column in all treatments other than the cane trash + KOH treatment (Table 15), suggesting that retention of these cations within the column had occurred. In contrast, in the treatment receiving KOH, the concentrations of Ca2+ and Mg2+ increased in the leachate as evidenced by the decrease on the exchange complex (Tables 11 and 12), whilst the K+ concentration declined. pH declined in all leachates after passing through the column with a concomitant increase in Al (Table 15). Table 15 Leachate and extract composition before and after passing through a soil column. Values in parenthesis represent the SE of the mean Treatment Control Cane trash Cane trash + KOH Melia azedarach leaf litter Stylosanthes litter collected from soil surface Stylosanthes fresh material Urochloa mosambicensis fresh material EC Extract Leachate (mS/cm) na 0.12 (0.02) 0.39 0.33 (0.01) 1.43 0.60 (0.05) 1.25 0.75 (0.02) 0.21 0.21 (0.01) na 0.19 (0.01) 0.76 0.57 (0.01) pH Extract Leachate na 5.17 (0.26) 5.71 4.69 (0.04) 9.63 4.40 (0.04) 5.52 4.43 (0.06) 6.68 4.68 (0.02) Na 4.75 (0.02) 6.06 4.79 (0.01) OC Extract Leachate (%) na 0.02 (0.01) 0.24 0.18 (0.01) 0.34 0.25 (0.02) 0.67 0.56 (0.01) 0.08 0.06 (0.01) 0.06 0.12 (0.01) 0.24 0.18 (0.00) Al Extract Leachate (mg/L) Na 0.11 (0.01) 0.16 0.30 (0.01) 0.10 0.72 (0.15) 0.15 6.33 (0.69) 0.09 0.21 (0.01) 0.09 0.23 (0.02) 0.16 0.65 (0.07) Treatment Control Cane trash Cane trash + KOH Melia azedarach leaf litter Stylosanthes litter collected from soil surface Stylosanthes fresh material Urochloa mosambicensis fresh material Ca Extract Leachate (mg/L) na 0.12 (0.05) 0.36 0.42 (0.02) 0.06 1.14 (0.46) 7.23 1.42 (0.24) 0.52 0.27 (0.03) 0.12 0.23 (0.02) 0.03 1.17(0.09) Mg Extract Leachate (mg/L) na 0.07 (0.01) 1.09 0.59 (0.01) 0.10 0.94 (0.05) 2.56 1.69 (0.05) 0.38 0.30 (0.01) 0.10 0.25 (0.01) 0.59 0.58(0.02) K Extract Leachate (mg/L) na 0.13 (0.02) 1.79 0.45 (0.01) 8.02 1.49 (0.12) 4.23 1.83 (0.27) 0.92 0.33 (0.01) 0.84 0.31 (0.00) 3.21 0.84(0.56) 28 There were relatively little if any changes in OC concentrations in the leachate after passing through the soil column (Table 15), suggesting little retention within the column. However, a significant relationship between the OC content in the extract prior to passing through the column and the amount of Al leached from the soil (Figure 9) clearly indicated the role of dissolved organic carbon in the formation of organo-metal complexes. It is also evident that the litter extracts differ in their ability to complex Al, since the OC concentrations in the Urochloa and cane-trash treatments were the same in the extract, but the concentrations of Al in the leachate were 0.65 and 0.30 mg/L, respectively (Table 15). Al in leachate (cmolc/L) 7 6 y = 5.91x - 26.82x2 + 48.15x3 - 0.07 R2 = 0.998 5 4 3 2 1 0 Figure 9 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 OC in extract (%) Relationship between OC in the leaching extract of litter materials prior to passing through a soil column and the Al3+ concentration after passing through the column Effect of ashed litter/trash on soil properties and leachate composition Soil pH: Soil pH was significantly increased in the surface 0-2 cm after leaching with the ashed extract (Table 16). The greatest increase in soil pH was in the ashed Melia, followed by the ashed cane-trash treatments (Table 16). An evaluation of the ∆pH resulted in changes in pH from the control of 0.38, 0.37, 0.22, 0.33 and 0.42 for the cane trash, Melia, Stylosanthes litter, Stylosanthes fresh and Urochloa, respectively. When these values are compared with the unashed treatments, the Melia treatment has undergone a dramatic decline in effectiveness i.e. 0.66 to 0.37. This suggests that with the combustion of Melia litter significant alkalinity is lost. The lack of movement of alkalinity (pH increases) to depth within the column with ashing suggests that much of the inherent alkalinity that was in the litter is either lost on combustion and /or used to generate surface charge in the 0-2 cm depth interval (see below for discussion). 29 Table 16 Changes in soil pH with depth after leaching with selected plant ashed litter materials Treatment Depth (cm) 0-2 2-4 4-6 6-8 8-10 10-12 Control 5.24 5.24 5.24 5.24 5.24 5.24 Cane trash 6.42 5.73 5.52 5.45 5.31 5.34 Melia azedarach leaf litter 6.61 5.75 5.36 5.31 5.31 5.36 Stylosanthes litter collected from soil surface 5.62 5.57 5.40 5.39 5.38 5.40 Stylosanthes fresh material 6.14 5.59 5.43 5.41 5.40 5.43 Urochloa mosambicensis fresh material 6.41 5.81 5.55 5.47 5.38 5.35 LSD (0.05) 0.12 Exchangeable cations: Changes in exchangeable Ca2+ and Mg2+ were confined to the surface 0-2 cm depth interval and, thereafter, there was little change (Tables 17 and 18). The greatest increases in these two elements were observed in those treatments under the Melia and Stylosanthes litter treatments (Table 17 and 18). In the unashed Melia treatment exchangeable Ca2+ and Mg2+ increased down the entire column, suggesting that the extract form this litter was more effectively in distributing cations down the soil column when compared to combusted litter material. Exchangeable K+ followed the same trends Ca2+ and Mg2+ with changes being confined to the surface 0-2 cm and ashed cane trash having the greatest impact (Table 19). Associated with the increase in soil pH in the 0-2 cm depth interval, there was a corresponding increase in the sum of bases on the exchange complex (Table 20). Table 17 Changes in exchangeable Ca2+ with depth after leaching with selected plant ashed litter materials Treatment Depth (cm) 0-2 2-4 4-6 6-8 8-10 10-12 Control 0.76 0.76 0.76 0.76 0.76 0.76 Cane trash 0.53 0.52 0.61 0.68 0.70 0.66 Melia azedarach leaf litter 1.20 0.73 0.69 0.67 0.76 0.67 Stylosanthes litter collected from soil surface 1.17 0.63 0.71 0.71 0.73 0.71 Stylosanthes fresh material 0.74 0.60 0.63 0.63 0.61 0.65 Urochloa mosambicensis fresh material 0.60 0.51 0.61 0.65 0.61 0.65 LSD (0.05) 0.12 30 Table 18 Changes in exchangeable Mg2+ with depth after leaching with selected plant ashed litter materials Treatment Depth (cm) 0-2 2-4 4-6 6-8 8-10 10-12 Control 0.16 0.16 0.16 0.16 0.16 0.16 Cane trash 0.19 0.09 0.12 0.15 0.15 0.15 Melia azedarach leaf litter 0.63 0.12 0.13 0.13 0.14 0.14 Stylosanthes litter collected from soil surface 0.32 0.12 0.14 0.15 0.15 0.14 Stylosanthes fresh material 0.26 0.13 0.15 0.14 0.15 0.15 Urochloa mosambicensis fresh material 0.25 0.10 0.10 0.12 0.11 0.15 LSD (0.05) 0.01 Table 19 Changes in exchangeable K+ with depth after leaching with selected plant ashed litter materials Treatment Depth (cm) 0-2 2-4 4-6 6-8 8-10 10-12 Control 0.13 0.13 0.13 0.13 0.13 0.13 Cane trash 1.16 0.31 0.19 0.15 0.14 0.15 Melia azedarach leaf litter 0.99 0.39 0.18 0.13 0.15 0.13 Stylosanthes litter collected from soil surface 0.21 0.15 0.14 0.14 0.14 0.14 Stylosanthes fresh material 0.73 0.16 0.13 0.13 0.13 0.13 Urochloa mosambicensis fresh material 0.96 0.35 0.29 0.21 0.15 0.14 LSD (0.05) 0.13 Table 20 Changes in the sum of bases (Ca2+, Mg2+ and K+) with depth after leaching with selected plant ashed litter materials Treatment Control Cane trash Melia azedarach leaf litter Stylosanthes litter collected from soil surface Stylosanthes fresh material Urochloa mosambicensis fresh material LSD (0.05) 0-2 1.65 1.96 2.87 1.78 1.82 1.98 2-4 1.65 1.15 1.49 1.11 1.11 1.21 Depth (cm) 4-6 6-8 1.65 1.65 1.25 1.35 1.34 1.24 1.29 1.33 1.16 1.13 1.32 1.25 0.25 8-10 10-12 1.65 1.65 1.34 1.26 1.39 1.25 1.34 1.32 1.11 1.18 1.08 1.18 Leachate characteristics: In general the concentration of Ca, Mg and K in the extracts was lower (Table 21) than that form the un-ashed samples (Table 15). For example, the concentration of Ca in the extract from the ashed and unashed Melia litter was 0.24 and 7.23 mg/L, respectively (Tables 15 and 21), suggesting that on combustion there is the potential for elements to be volatized. However, it is more plausible that with combustion sparingly soluble oxides of these metals are produced that restrict the rate of dissolution. The OC content in the extracts was undetectable, suggesting OC was lost on combustion. Similarly, there were trace concentrations of Al3+ in the leachates clearly showing the efficacy of organo-metalic complexation in removing Al from the exchange complex. 31 Table 21 Leachate and extract composition from ashed litter/trash materials before and after passing through a soil column EC pH Treatment Ext Lch Ext Lch (µS/cm) Control Cane trash 273 258 9.54 5.09 Melia azedarach leaf 341 245 10.34 5.19 litter Stylosanthes litter collected from soil 139 179 9.22 5.41 surface Stylosanthes fresh material 209 209 9.59 5.21 Urochloa mosambicensis 424 387 9.73 5.01 fresh material LSD (0.05) Ext = Extract; Lch = Leachate ns = not significant at P = 0.05. OC Ext Lch (%) 0.024 0.00 0.010 0.00 0.002 Al Ext Lch (mg/L) 0.11 0.00 0.01 0.00 0.01 0.00 0.012 0.01 0.01 0.01 0.012 0.01 0.01 0.00 0.008 0.04 ns 0.01 0.01 Ca Ext Lch (mg/L) 0.12 0.05 0.56 0.24 0.43 0.61 0.28 0.31 0.34 0.25 0.85 0.08 Mg Ext Lch (mg/L) 0.07 0.06 0.16 0.22 0.12 0.19 0.07 0.14 0.09 0.15 0.36 0.02 K Ext Lch (mg/L) 0.13 1.53 0.11 1.38 0.09 0.13 0.07 0.76 0.07 2.14 0.14 0.03 5.5 Conclusions Estimation of net proton addition rates under a continuous cane trash blanketed production system is achieved through an assessment of temporal charges in pHBC and pH. This implies measurements of these two soil attributes at a point prior to the imposition of treatments and, subsequently, at some point in the future. Within the current suite of longterm trash-retention versus burnt trials established at Harwood, Mackay and Tully this could not be undertaken as the soil analytical data were not available at the initiation of the study. Hence, we had to resort to an alternative set of studies within the YDJV where soil samples were collected prior to the establishment of treatments. The increased net proton addition in the GCTB system relative to the burnt system was estimated to be 0.71 kmol H+/ha.year and the net proton additions for the continuous cane, grass ley and fallow systems were 1.38, 3.81, and 0.87 kmol H+/ha.year, respectively. Moody and Aitken (1997) estimated rates of acid addition under continuous sugarcane production systems to be about 3.2 kmol H+/ha/yr, which is about 2.5 times higher than the estimates that we made. These differences in net acidification rates may in part be attributed to methodology of measuring pHBC and the time over which the assessment was made. To put these rates of acidification in the context of other production systems in the tropics, a banana production system produces 34.2 kmol H+/ha/yr, whilst a nitrogenfertilized tropical pasture produces 6.4 kmol H+/ha/yr (Moody and Aitken 1997). Clearly, sugarcane-production systems are not major generators of acid when compared to these other production systems. However, if acid accumulation associated with these sugarcane production systems are not addressed, potential long-term damage to the soil resource base will occur that will impact on the long-term sustainability and productivity of these production systems. In moving from a burnt-cane production system to GCTB there appeared to be a slightly greater degree of acidification under the latter system. However, this statement should be treated with caution in that this effect was only observed at a single site (i.e. Tully trial). 32 In fact, the results from the leaching column study indicate that the water-soluble fraction that emanate from unashed cane trash are more effective in transferring alkalinity to depth than ashed trash. Moreover, treating the cane trash with KOH proved to be more effective in transferring alkalinity to depth, when compared to cane trash on its own. This outcome is significant in that it suggests that introducing a highly alkaline compound to the trash significantly alters the characteristics of the leachate, thereby having a positive impact on alkalinity transfer. This potentially could be a means of neutralizing annual acid generation and supplying K to the crop. The long-term potential benefit associated with GCTB is the greater nutrient availability, as it is speculated that burning results in either the volatilization of elements and/or their immobilization through the formation of basic oxides. Organic carbon levels increased by 4 t/ha under the GCTB treatments compared with the burnt over a 9-year period. In contrast, under a grass ley system, organic carbon increased by 9 t/ha over a 6-year period compared with continuous cane, clearly demonstrating the superior nature of grass leys in sequestering carbon in this tropical environment. Charge fingerprints were derived from bulk samples collected from each of the trials. The divergence between the charge fingerprints due to treatment was greatest in the surface 10 cm, but this diminished with depth indicating that the influence of trash retention or grass leys on soil charge characteristics was confined to the surface layer. The increases in CEC at pH 5.5 of the GCTB treatment compared with the burnt, and of the grass ley treatment compared with the continuous cane were estimated to be 0.67 and 0.75 cmolc/kg, respectively, over the 0-10 cm depth interval. Plots of ECEC versus CEC clearly indicated that 9-31% of the cations measured were not associated with the exchange complex and would, therefore, be subject to leaching loss. Whilst there was a higher net acidification rate under trash retention and grass ley production systems when compared to burnt or continuous cane systems, this must be viewed in the context of an overall enhancement in the surface charge characteristics of the soil through soil organic matter increase. Through a routine liming program instituted at plough out, acid generated over the rotation can be neutralized and, therefore, should not be viewed as a potential problem. In contrast, the generation of increased surface charge can only be viewed as beneficial. However, the increase in charge characteristics associated with elevated soil organic matter is transient. When the system is perturbed through cultivation, a significant proportion of this increased charge will be lost through organic matter mineralization. 6.0 TRASH DECOMPOSITION, SOIL SODICITY AND THE EFFICACY OF CALCIUM-BASED AMENDMENTS FOR AMELIORATING SODIC SOILS 6.1 Introduction The suite of exchangeable cations has a marked influence on the physical properties of soil. Soils are termed ‘sodic’ if more than 6% of the cation exchange capacity is taken up by sodium. The higher the proportion of sodium, the more sodic the soil, and the worse its physical properties from a cropping point of view. Sugarcane yield reaches zero when surface soil sodicity or exchangeable sodium percentage (ESP) is greater than about 30% (Nelson and Ham 2000). Sodic soils cover a considerable proportion of all sugarcane 33 growing areas in Australia, except for the wet tropics (Nelson et al. 2001). In addition to causing crop management problems, cultivation of sodic soils can lead to excessive turbidity in runoff, with consequent detrimental impacts on downstream ecosystems. Ca-containing ameliorants are used to reduce sodicity. The Ca replaces exchangeable Na, which must be leached from the soil. Trash is also known to improve crop performance on sodic soils, largely due to its ability to retain water and reduce evaporation losses. The aim of this section was to examine interactions between Ca-containing ameliorants (lime and gypsum) and trash when applied to sodic soils. Interactions between Cacontaining ameliorants and molasses were also examined, as molasses has shown promise as an ameliorant, and was expected to interact with gypsum and lime. Soil organic matter has an important influence on the cation exchange properties of soil, and, therefore, influences the tendency of a soil to become or remain sodic (Nelson and Oades 1998). The effect of trash management on cation exchange properties of soil was examined in some detail. Sections of this work were carried out by Ahmad Suriadi in the University of Adelaide as part of his Masters-degree program. 6.2 Trash decomposition and soil sodicity In Australia, significant areas of sodic soils (exchangeable sodium percentage, ESP, > 6) are used for cropping. Sugarcane is grown on soils with various degrees of sodicity, including those where sodicity is a serious limit to crop production (Nelson and Ham 2000; Nelson et al. 2001). The main reason for this limitation of crop growth is the adverse effect of sodicity on soil structure and the movement of water, air and roots through the soil. Management options aim at minimising these negative effects (Nelson et al. 2001). Mechanical harvesting of unburnt sugarcane leaves large amounts of trash on the soil surface, e.g. 6-25 t dry matter/ha, most of which decomposes during the 12 months after harvest (Robertson 2003). Decomposition of sugarcane trash has not, however, been measured in sodic soils. It is important to know whether soil sodicity affects the rate of trash decomposition because there may be consequences for nutrient (particularly nitrogen) availability, and also the ease with which farming operations such as irrigation or cultivation can be performed. The effect of sodicity on decomposition also needs to be known to improve confidence in the use of simulation models, such as APSIM (Thorburn et al. 2001), to predict the effects of trash management on sodic soils. In studies with other crop residues, sodicity has been shown to increase, decrease, or have no effect on decomposition (Nelson and Oades 1998). The objective of this study was to measure the effect of soil sodicity on decomposition of sugarcane trash. 6.2.1 Methods Experimental details Sugarcane (cultivar Q124) was grown in 40-cm diameter pots in a glasshouse for 1 year, fertilised with weekly applications of a mixed fertiliser including 15N-labelled ammonium sulfate. At maturity, the green and senesced leaves, which normally become trash during mechanical harvesting of the cane, were collected. They were dried in an oven at 65ºC for 34 2 d, shredded through a cutter-grinder (2-10 cm long and 0.2-1.0 cm wide), mixed thoroughly, and subsampled. The fragment size of the subsample was then reduced to < 5 cm by cutting with scissors. Analysis of the trash gave the following results: total C, 36.3%; total N, 0.57%; C:N ratio, 64; water-soluble C, 3.9%; water-soluble N, 0.2%; and 15N atom excess, 6.8%. Soil was collected from the 0-15 cm depth of cane fields at three locations in northern Queensland (Table 22). It was dried in an oven at 45ºC for 6 d, then crushed to pass through a 2-mm screen, and sampled for analyses. The soils were then weighed into 13 cm-diameter plastic pots with holes on the bottom (1 kg /pot). The following treatments were then imposed on an equal number of pots of each soil: (1) Soil left untreated (‘Original’, TO); (2) Soil made non-sodic, target ESP < 3 (‘Non-sodic’, TNS); and (3) Soil made sodic, target ESP 20 (‘Sodic’, TS). Treatments TNS and TS were achieved through a three-stage leaching process: Stage 1 - the pots of soil were placed in a 30 L tub containing a 4 cm depth of the first solution (SAR 0, 0.5 molc/L calcium chloride for TNS treatment; SAR 20, 0.5 molc/L calcium chloride and sodium chloride for the TS treatment). More solution was gradually added to the tub over 1 d to bring it close to the height of the soil surface, and the pots left to soak for another day. The pots were then removed from the tub and one saturation volume of the solution slowly poured over the soil, whilst maintaining saturation. Stage 2 - the above procedure was repeated with the second solution (SAR 0, 0.1 molc/L calcium chloride for TNS; SAR 20, 0.1 molc/L calcium chloride and sodium chloride for TS). Stage 3 - Over 4 d, deionised water (equivalent of 8 soil saturation volumes) was poured on to the soil in each pot and allowed to drain. The pots were kept covered to minimise evaporation. Table 22 Soils used in the experiment (0-15 cm depth) Location Texture Particle size (%sand, %silt, %clay) Total N (%) Total organic C (%) C:N ratio CEC: clay ratio Burdekin 147.23 E, 19.50 S sandy clay loam 53, 25, 22 0.057 0.68 12 0.53 Mackay 149.07 E, 21.10 S sandy clay loam 56, 28, 16 0.058 1.08 19 0.15 Proserpine 148.55 E, 29.39 S loamy sand 82, 13, 5 0.056 0.71 13 0.20 The pots were then weighed to estimate soil water holding capacity (WHC). The soils were again dried at 45ºC for 6 d, crushed to < 2 mm, and thoroughly mixed within treatments. All treatment x soil type combinations were sampled for analyses. Sugarcane trash (10 g) and soil (900 g) were mixed and added to 13-cm-diameter plastic pots with holes on the bottom, to provide a trash:soil ratio similar to field conditions (10 t dry matter /ha). The pots were shaken gently to settle the contents to a bulk density of 1.1 g/cm3. Water was applied to around 80% of field capacity. Enough pots were prepared for four replicates of each treatment x soil combination, plus four blanks (empty pots). 35 The pots were arranged in a glasshouse under shade cloth, in a randomised complete block design. Every 7-12 d, they were watered (usually to around 80% of saturation, but to >100% of saturation on several occasions) using filtered rainwater with electrical conductivity of 0.01 dS/m. The soils dried out considerably between waterings. Temperature was not controlled and varied between 17 and 43ºC. The TS treatment was re-imposed on day 41 because the soil was found to be less sodic than expected. This time, one soil saturation volume of SAR 50, 0.5 molc/L sodium/calcium chloride solution was leached through the soil, followed by one soil saturation volume of SAR 50, 0.1 molc/L sodium/calcium chloride solution, then 2-3 saturation volumes of deionised water. Drainage ceased in the Burdekin soils, and became very slow in the Mackay soils, soon after the second volume of water was applied. Several days later, small samples of soil (5-7 g) were taken from each of the TS pots (replicates were pooled) for measurement of exchangeable cations. Carbon dioxide (CO2) being released from the pots was trapped on four occasions (days 3, 20, 56 and 77) by placing vials containing 20 mL of 1M sodium hydroxide on the soil surface and sealing the pots in heavy-duty zip-seal plastic bags for 2-3 d. At the end of the experiment (day 136), the contents of the pots were cut in half vertically: one half was retained for soil analyses, and the other half was washed under running water and the remaining trash caught on a 2-mm sieve. Analytical methods Carbon dioxide in the sodium hydroxide traps was determined by titration against dilute hydrochloric acid (Zibilske 1994). Total C and N analyses were done on dried subsamples of soil ground to < 250 µm and trash ground to < 500 µm. Total N and C in soil and trash were determined using a Leco CNS 2000 combustion analyser. Soil inorganic N was determined by extracting 15-g portions of moist soil in 2M potassium chloride, and measuring ammonium and nitrate in the filtered extracts by automated colorimetric analysis (Rayment and Higginson 1992, method 7C2). Soil pH and electrical conductivity (EC) were measured in 1:5 soil-water extracts (Rayment and Higginson 1992, method 3A1). Cation exchange capacity (CEC) and exchangeable calcium, magnesium, potassium and sodium (CaE, MgE, KE, and NaE) were determined by compulsive exchange with silver-thiourea (Rayment and Higginson 1992, method 15F1). Clay content was measured by the pipette method. Spontaneously dispersible clay was measured using turbidimetry according to the method of Rengasamy et al. (1984). Calibration curves of turbidity vs. clay concentration were prepared for each soil. Calculations and statistical analyses Cumulative CO2 evolved during the experiment was estimated from the area under the curve of CO2 accumulation. The main effects and interactions of leaching treatment and soil type were tested for each sampling time separately by two-way analysis of variance for a randomised complete block design using the Systat program (SPSS Inc.). Relationships between selected variables were investigated using Pearson’s correlation. Unless stated otherwise, effects were taken as significant where Bonferroni probabilities were ≤ 0.05. 36 Table 23 Soil properties before trash addition (day 0), after sodic treatments reimposed (day 41), and at the end (day 136) of the experiment. Subscript E denotes exchangeable cations. TO, TS, TNS denote unamended, sodic, and non-sodic treatments, respectively. Least significant differences (LSD) are for soil x treatment interaction Treatment Burdekin TO TS TNS Mackay TO TS Day 0 TNS CEC (cmol(c)/kg) 11.7 13.2 11.7 2.5 3.6 4.1 CaE (cmol(c)/kg) MgE (cmol(c)/kg) KE (cmol(c)/kg) NaE (cmol(c)/kg) ESP 5.6 11.0 11.0 1.4 3.3 3.8 4.5 0.6 0.3 0.7 0.0 0.1 0.2 0.2 0.2 0.2 0.1 0.1 1.4 1.4 0.2 0.2 0.2 0.1 12.0 10.6 1.7 8.0 5.5 2.4 pH 6.51 7.12 6.68 4.45 5.07 4.83 EC (dS/m) 0.18 0.13 0.09 0.04 0.04 0.03 Soil inorganic N (µg/g) 38.0 6.9 5.5 13.7 9.4 7.5 Saturated WHC (% w/w) 43 39 40 38 33 35 Total pore volume (% v/v) 66 62 63 61 56 58 Day 41 CEC (cmol(c)/kg) 20 5 CaE (cmol(c)/kg) 13.2 3.7 MgE (cmol(c)/kg) 0.9 0.2 KE (cmol(c)/kg) 0.5 0.3 NaE (cmol(c)/kg) 5.3 1.0 ESP 26.6 19.2 Day 136 CEC (cmol(c)/kg) 15.5 17.5 15.4 9.6 10.0 10.2 CaE (cmol(c)/kg) MgE (cmol(c)/kg) KE (cmol(c)/kg) NaE (cmol(c)/kg) ESP 7.2 11.9 11.2 2.0 3.4 3.4 4.1 0.8 0.7 0.6 0.1 0.2 0.4 0.4 0.4 0.2 0.2 0.2 0.5 1.3 0.2 0.2 0.6 0.2 3.2 7.4 1.3 2.1 6.0 2.0 pH 6.43 7.05 6.51 5.00 5.34 5.20 EC (dS/m) 0.09 0.11 0.08 0.02 0.06 0.03 Soil inorganic N (µg/g) 3.5 1.3 0.9 1.8 1.5 1.6 Clay (%) 21.6 26.3 16.8 14.8 14.6 14.4 Dispersible clay (% total) 0.44 0.96 0.23 1.52 1.32 1.34 Trash retrieved (%) 48 57 37 46 50 35 Recovery of 15N (%) 75 85 101 83 107 75 Proserpine TO TS TNS 1.1 1.5 1.7 0.8 1.4 1.7 0.2 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.0 6.7 0.0 4.35 5.61 5.24 0.20 0.03 0.03 99.0 4.4 3.1 26 26 26 48 48 48 3 1.7 0.2 0.3 0.5 18.5 7.7 8.3 7.2 1.4 1.6 1.7 0.2 0.1 0.2 0.1 0.1 0.1 0.1 0.3 0.1 1.3 3.6 1.4 5.47 5.76 5.67 0.03 0.06 0.04 1.0 1.3 1.6 4.7 4.6 4.7 1.34 1.92 1.65 64 70 59 81 74 64 LSD(0.05) 2.9 0.4 0.04 NS 0.22 0.18 0.03 1.4 5.5 1.3 NS NS 6.2.2 Results Soil type effects In the original soils, clay content, CEC, ESP, and WHC decreased in the order Burdekin > Mackay > Proserpine (TO, Table 23). Soil pH was greater in the Burdekin soil than in the Mackay and Proserpine soils. The Burdekin soil also had a higher ratio of CEC to clay content than the Mackay and Proserpine soils (Table 22). The values of the ratio indicated mixed mineralogy in the Burdekin soil and kaolinite-dominated mineralogy in the other two (Shaw et al. 1998). Total soil N was similar in the three soils, but total organic C and C:N ratio were greater in the Mackay soil (Table 22). Inorganic N concentration in the Proserpine soil was more than twice that in the Burdekin and Mackay soils, reflecting 37 recent fertiliser application. Inorganic N in all soils was dominated by nitrate (data not shown). Leaching treatment effects The day 0 soil analysis showed that TNS had effectively reduced the original soil sodicity levels, rendering the TNS soils non-sodic (ESP 0-2) (Figure 10). At day 0, the TS treatment made the Proserpine soil mildly sodic, but had little effect on sodicity in the Burdekin and Mackay soils. Re-imposition of TS on day 41 markedly increased soil ESP, making all the TS soils highly sodic (ESP 18-27) (Figure 10). Following both leaching treatments, the Burdekin soil attained a higher ESP than the Mackay or Proserpine soils for the same level of solution SAR, indicating a higher Gapon coefficient for the Burdekin soil (Table 24). The Gapon coefficients were higher following the second leaching due to the higher concentration of the solution used. The original soils were of intermediate sodicity (ESP 0-12). Soil pH was increased by both TS and TNS in the Burdekin and Mackay soils, and by TS only in the Proserpine soil (Table 23). Soil inorganic N was reduced by the leaching treatments, due to loss of nitrate. Total soil N and organic C were not significantly affected by the leaching treatments (data not shown). ESP (%) 30 20 10 0 Figure 10 Burdekin Mackay Proserpine 0 41 136 0 41 Day 136 0 41 136 Untreated, TO Non-sodic, TNS Sodic, TS Exchangeable sodium percentage (ESP) before trash addition (day 0), after sodic treatments re-imposed (day 41), and at the end of the experiment (day 136) Table 24 Gapon coefficients for the TS soils following the first (day 0) and second (day 41) leaching Time Burdekin Mackay Proserpine Day 0 (SAR 20) 0.0064 0.0031 0.0036 Day 41 (SAR 50) 0.0080 0.0054 0.0059 38 Trash decomposition In the TO treatments, CO2 evolution was greater in the Burdekin soil than the Mackay and Proserpine soils. In the Burdekin soil, CO2 evolution was reduced in the TS relative to TNS and TO treatments. The leaching treatments had no significant effect on CO2 evolution from the Mackay and Proserpine soils (Figure 11). Cumulative CO2 Evolution (µg C / g soil) 3000 2000 1000 0 3000 2000 1000 0 3000 2000 1000 0 0 Proserpine Mackay 50 100 Days Burdekin 150 Non-sodic, TNS Untreated, TO Sodic, TS Figure 11 Cumulative CO2 evolution as affected by soil and treatment. Bars on Burdekin graph are LSD (P=0.05) for treatment x soil interaction. Treatments were not significantly different in Proserpine and Mackay soils The trash retrieved from the soil at the end of the experiment was very variable among replicates, but tended to be greatest in the TS and least in the TNS treatment, though differences were not statistically significant (Table 23). Average dry matter recovery was 52%. Recovery of trash-15N at the end (mean 85%) was similarly variable among replicates and did not differ significantly among treatments. Trash retrieved was weakly negatively correlated with cumulative CO2 released during the experiment (r = -0.43, P < 0.05 for all data; r = -0.62, P < 0.07 for replicate means). Neither CO2 evolved nor trash dry matter retrieved were related to trash 15N retrieved. Treatment effects on day 136 The treatment effects on day 136 generally reflected those on day 0 (Table 23). Total N and organic C showed no significant change in time (data not shown). Soil pH increased slightly in the Mackay and Proserpine soils, but was unchanged in the Burdekin soil 39 (Table 23). The Burdekin and Proserpine soils originally had relatively high EC, and this was significantly reduced by day 136 (P < 0.05). Soil CEC was universally greater on day 136 than on day 0 (P < 0.05). Soil CEC was correlated with clay content (r = 0.91, P < 0.05). In all soils, ESP in the TS treatment was reduced to < 8 on day 136 (P < 0.05). The ESP in TNS and TO soils generally declined with time, except in the Proserpine soil (Figure 10). The percentage of this clay that was spontaneously dispersible was < 2% in all cases, and decreased in the order: Proserpine > Mackay > Burdekin. Spontaneously dispersible clay was not significantly affected by the treatments. The TS treatment increased total clay content in the Burdekin soil (P < 0.05), but not in the Mackay and Proserpine soils. The effect in the Burdekin soil probably occurred because the standard dispersion treatment for measuring particle size distribution was not adequate to fully disperse the soil in its non-sodic state. In the Burdekin soil only, final soil NaE , ESP, clay content and dispersible clay were negatively related with final cumulative CO2 released (r = -0.74, -0.75, -0.72, -0.80, respectively, P < 0.05). By day 136, soil inorganic N was reduced to a very low level in all pots (Table 23). Soil-water content measured at the end of the experiment and estimated by weighing the pots during the experiment followed the same trends as WHC. Generally, Burdekin > Mackay > Proserpine, and TNS > TS (effects did not always attain significance at P < 0.05, but trends were consistent, data not shown). Final soil water content was correlated with clay content (r = 0.89, P < 0.05). 6.2.3 Discussion The experimental approach used in this study allowed the effects of soil sodicity on trash decomposition to be measured without the confounding influences of soil, trash, and environment factors, which would have been unavoidable in a field study. The TS treatment decreased trash decomposition in the Burdekin soil by about 18% (cf. CO2 release from TNS). The effect was related to high NaE and high ESP in the Ts soil. In contrast, TS had no effect on decomposition in the Mackay and Proserpine soils. This may have been partly due to lower levels of sodicity being achieved in these soils, and partly due to other soil factors. For example, the amount and type of clay has been shown to influence the effects of sodicity (Churchman et al. 1993; Nelson et al. 1997). The Burdekin soil, which had the greatest clay content and the highest pH, sealed up when the TS was imposed, and thereafter water penetration was very much slower than in the Mackay and Proserpine soils (data not shown). This response in the Burdekin soil can be attributed to blocking of the soil pores because of dispersion and swelling of the clay at high ESP and pH (Churchman et al. 1993). Thus, it is likely that much of the sodicity effect on decomposition was due to impaired aeration. 40 The TS treatment was strongly sodic according to ESP (18-27). Milder sodicity (ESP 812) had no effect on trash decomposition (CO2 production) in any of the soils, as can be seen by comparing the TO and TNS treatments (Figure 8). This experiment was watered using rainwater of low electrical conductivity, therefore the effect of high ESP on soil hydraulic properties would have been maximised (Quirk and Schofield 1955). The effects of soil sodicity on decomposition would probably be even less in the field than in this study, because trash is usually left on the surface rather than being mixed with the soil. Effects of sodicity on organic matter decomposition and soil microbial populations have been little studied. However, Nelson et al. (1996, 1997) found that the size of the soil microbial biomass was unaffected, and decomposition of added legume material was either unaffected, slightly increased, or slightly decreased by increasing soil sodicity. The maximum ESP achieved by TS in the Burdekin soil (27) was very high. Under field conditions, such a soil would present serious difficulties for crop production (Nelson and Ham 2000). An 18% reduction in trash decomposition, as found in this study, may therefore be of relatively little consequence to the farmer. The TS treatment resulted in a higher ESP in the Burdekin soil than the other soils because of its higher Gapon coefficient (Table 24). The difference in coefficients may have been partly due to differences in mineralogy, but also to the different ratios of clay to organic matter. Cation exchange sites on organic matter tend to have a lower affinity for Na than exchange sites on clay minerals (Nelson and Oades 1998). Recent work has shown that retention of sugarcane trash in long-term field trials reduces the tendency of soil to become sodic (Suriadi et al. unpubl. data). Trash retrieved at the end of the experiment was very variable among replicates because the pot contents were divided without mixing, and because of the inconsistent ease of trash removal from the different soil types. This resulted in the poor correlation between trash retrieved and cumulative CO2 evolution. We recommend using a larger trash fragment size than used in this study when decomposition is to be measured by similar retrieval techniques. Mineralisation of native soil C can be affected by sodicity (Nelson et al. 1996) and may have differed among soils and leaching treatments in this experiment. However, any significant effect on the experimental results would have been unlikely, because mineralisation of soil C is much lower than the mineralisation of C in trash (see Section 8). The large decrease in soil inorganic N during the experiment probably resulted from a combination of microbial immobilisation during decomposition of the trash, leaching (Proserpine soil), and denitrification (Mackay and Burdekin). In related studies, sugarcane trash was shown to promote microbial immobilisation of N, but applications of up to 400 kg N/ha as urea did not increase the rate of trash decomposition (Fiona Robertson, unpubl. data). Thus, the large concentration of inorganic N in the TO Proserpine soil is not likely to have influenced trash decomposition in this experiment. Decomposition of the trash changed the cation exchange properties of the soils in several ways. The increase in CEC from day 0 to day 136 must have been due to the production 41 of organic materials with negative charge. In the Mackay and Proserpine soils there may also have been a contribution from the effect of increased pH on variable charge components of the CEC. Increases in CaE, especially in the Burdekin soil, must also have originated from the trash. The trash contained approximately 0.28% Ca, which, at the trash addition rate used, corresponds to a Ca addition of 1.5 cmolc/kg soil. The increase in CaE occurred within the first 40 d. The amount of K added in trash was 0.74 cmolc/kg soil. Increases in soil contents of exchangeable cations were greatest in the Burdekin soil, probably due to less leaching loss than in the other two soils. Substantial decreases in ESP of the Burdekin soil from day 0 to day 41 were due mostly to the increase in CaE rather than loss of Na. 6.3 Effects of lime, gypsum, trash and their interactions on sodic soils and sugarcane growth The aim of this investigation was to determine whether interactions between lime, gypsum and trash influence the amelioration of alkaline and acid sodic soils. 6.3.1 Methods and material The trial was set up as a twin factorial design, with two soils (Burdekin or Proserpine), and nine treatments involving ameliorant (nil, gypsum or lime) and trash (nil, surface or incorporated). There were four replicates, resulting in a total of 72 pots. Each pot was planted with a one-eye sett of sugarcane (Q117) on 5 July 1999. The pots were placed on Besser blocks on a gravel bed at a rectangular spacing of 1 x 0.7 m, and were surrounded by a row of guard pots. Wire fences were erected to prevent the pots from falling over. Many did fall over during Cyclone Tessi, but virtually no soil was lost and the pots were subsequently righted. The pots were made from PVC pipe 0.3 m in diameter and 0.6 m long, with a perforated base. Subsoil and then topsoil from each of the two soils was placed in the pots in layers, each 0.25 m deep. Three doughnut-shaped plastic baffles were taped to the inside of the pot at evenly spaced depths as the soil was placed in the pots. These were to prevent roots moving down the side of the pots. By the end of the experiment the soil had slumped by an average of 80 mm. The gypsum (Winton-mined gypsum, with 13.5% S and 17.4% Ca) application rate was equivalent to 10 t/ha (71 g/pot). The lime (Zinaback lime, 37% Ca) was added at an equivalent rate of Ca to the gypsum (4.7 t/ha, 33.4 g/pot). Trash was chopped into pieces 50-70 mm long and added at 10 t/ha (DM). In the incorporated treatments it was incorporated into the top 100 mm. Urea was added initially at 478 kg/ha (3.4 g/pot) and a slow release fertilizer was added in the second year. The pots were irrigated daily or twice daily using drippers and were hand weeded from time to time. T he pots were initially set up at BSES Burdekin, but were shifted to CSIRO Davies lab on 26 October 1999. Cane length was measured on 7 February 2000. It varied from 0-700 mm, with no significant effect of treatments. On 14 February 2000, following a rainfall event, leachate was collected. The first harvest was conducted on 2 October 2000. Trash was removed and the green leaves plus cabbage and stalks were weighed separately. The green leaves plus cabbage was mulched and occasional samples taken for water content measurement. 42 The stalks were analysed for CCS using a small mill and assuming constant fibre content. Trash was added again to appropriate pots, to the surface in both ‘surface’ and ‘incorporated’ treatments. The second harvest was conducted on 18 April 2001, in the same way as the first. The soil was then sampled, taking 3 x 70 mm cores per plot and bulking them for four depths. 6.3.2 Results and discussion The treatments had little or no significant effect on cane yield or CCS during the experiment, due largely to high variability (Figures 12 and 13). The high variability appeared to be due to cracks forming in some pots; in those pots irrigation water and ameliorants flowed preferentially down the cracks. Growing sugarcane in sodic soils these large pots was extremely difficult to manage. Leachate properties reflected soil type but not treatment (Figure 14). The final soil sampling indicated effects of gypsum and lime on the Burdekin soil, but no interactions between trash treatment and gypsum or lime. In the Burdekin soil, lime generally decreased ESP at the surface and gypsum generally decreased ESP at depth (Figure 15). Leaching alone had considerably reduced ESP in all treatments. The experiment did not result in clear results, because of the difficulty in maintaining uniform water content and flow through the soils. Subsequent pot trials were carried out in smaller pots without plants. Total dry matter (g/pot) CCS 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0 CSICSICSICSICSI CSI Control Gypsum Lime Control Gypsum Lime Burdekin Proserpine 15 14 13 12 11 Figure 12 Total above-ground dry matter (bars) and CCS (points) of cane at first harvest (October 2000). Trash treatments are indicated by C (control), S (surface-applied) and I (incorporated). Error bars show standard deviations 43 Figure 13 Cane dry matter (g/pot) CCS 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0 CSICSICSICSICSI CSI Control Gypsum Lime Control Gypsum Lime Burdekin Proserpine 15 14 13 12 11 10 9 8 7 6 5 Cane dry matter (bars) and CCS (points) at the second harvest (April 2001). Trash treatments are indicated by C (control), S (surfaceapplied) and I (incorporated). Error bars show standard deviations Leachate Na (meq/L) 7 Burdekin 6 Proserpine 5 4 3 2 1 0 C S I C S IC S IC S I C S I C S I Control Gypsum Lime Control Gypsum Lime Figure 14 Sodium concentration of pot leachate collected on 15 February 2000. Letters indicate the trash treatments: C, control; S, surface applied; I, incorporated 44 Depth (cm) 1 0 10 20 30 40 50 ESP 234 Burdekin 5 Zero, Zero Trash Zero, Surf Trash Zero, Inco Trash Gyps, Zero Trash Gyps, Surf Trash Gyps, Inco Trash Lime, Zero Trash Lime, Surf Trash Lime, Inco Trash Figure 15 Exchangeable sodium percentage (ESP) in the Burdekin soil at the end of the experiment 6.4 Structural stability of sodic soils as influenced by gypsum/molasses Stable structure and structural stability are essential for a fertile soil. One of the measures of stable structure is stability of soil aggregates in water. Such changes in water-stable aggregation have generally been correlated with the quality and quantity of organic matter in the soil (Tisdall and Oades 1982; Piccolo 1996; Puget et al. 2000). Gypsum (CaSO4.2H2O) and organic materials are commonly used as ameliorants to overcome sodicity and related problems. In the sugarcane-growing districts of Queensland, gypsum is available at about $90 t-1, and a substantial source of organic material is available as molasses at about $45 t-1. Molasses, a by-product of sugar production, contains carbohydrates (62%), protein (3-4%) and ash (8.6%) (Weber and Van Rooyen 1971), as well as cations (calcium, 0.07%; magnesium, 0.08%; potassium, 0.5%; sodium, 0.03%), nitrogen (0.39%) and water (20% w/w), and has a pH of about 5.5 (data from this work). The aim of this work was to determine whether molasses alone or combined with gypsum, could improve the structural stability of two sodic soils used for sugarcane production. 6.4.1 Materials and methods Soils The soils used were the top layer (0-15 cm depth) of sodic soils from Proserpine and the Burdekin. The Burdekin soil is a mesonatric brown sodosol and the Proserpine soil a natric brown kurosol (Isbell 1996). Table 25 shows some of the physical and chemical properties of the soils. Both soils contained a range of clay minerals; the Burdekin soil is dominated by illite and illite-smectite, and the Proserpine soil by illite and kaolinite. 45 Table 25 Soil pHA Burdekin Proserpine 8.0 4.9 ESP Burdekin 7.9 Proserpine 18.8 A1:5 soil:water extracts. Selected properties of soil used in this study ECA (dS/m) 0.2 0.1 Org C (%) 0.8 0.7 Cations (cmol(+)/kg) Ca Mg K 6.5 3.9 0.5 0.9 0.4 0.4 Particle size (%) Clay Silt Sand 21.7 25.4 52.8 5.0 12.8 82.2 CEC Na (cmol(+)/kg) 0.9 10.9 0.4 2.3 Texture Sandy clay loam Loamy sand Experimental design and treatments The experiment had a factorial design with three factors (gypsum, molasses, and soil), each at two levels, and four replicates. Each experimental unit was a pot of soil, kept at 25°C for a period of 12 weeks. Data were analysed using the Genstat V Statistical Package (Genstat 5 Committee 1987). The gypsum and molasses factors consisted of an application of 0 or an application equivalent of 10 t ha-1 on an oven-dry basis (6.08 g pot-1). For the molasses treatments, molasses was dissolved in water and sprayed uniformly onto a thin layer of soil, which was then dried and mixed before being placed in pots. Pots (8.8 cm inner diameter, 10 cm long) contained 500 g treated soil, which produced bulk densities in the range of 1.4-1.5 g/mL. Fine mesh was placed over a funnel at the base of each pot to retain soil but to allow leachate to pass. All the pots were wetted to 85% of field capacity (FC) and maintained in this state for 7 d to provide time for dissolution, diffusion and microbial activity. Pots were leached with 0.5 pore-volumes of distilled water (88 mL or equal to 14.5 mm rainfall for the Burdekin soil and 72 mL or equal to 11.8 mm rainfall for the Proserpine soil). This water was applied as a spray over a period of 2.5-3 h. Leachate was collected and analysed for dissolved organic carbon (DOC), cations (Ca2+, K+, Mg2+ and Na+), pH and EC. The sodium adsorption ratio (SAR) of solutions was calculated from: SAR = [Na+]/([Ca2+] + [Mg2+])0.5, where cation concentrations are expressed in mmol/L. Between leaching events, pots were allowed to dry for 7-10 d, which was long enough to bring them to their original air-dry water contents. At this point, they were again leached with water, during which leachate was collected for analysis. These wetting and drying cycles were conducted to simulate the changes in soil solution concentration that occur in the field and to promote the migration of solutes and colloids that may impact on soil structural stability. Leaching events were conducted a total of fivr times and all pots were then air-dried for soil analysis after the last leachate had been collected. Soil analyses The pH and EC were determined using 1:5 soil:water extracts. Exchangeable cations and CEC were determined using methods described by Rayment and Higginson (1992). Exchangeable cations (Na+, K+, Ca2+, and Mg2+) were extracted with 1 M NH4Cl at pH 7 and concentrations were measured using Inductively Coupled Plasma Atomic Emission Spectrometry (ICPAES). The CEC of the leached samples was then measured by 46 displacement of NH4+ with K-Ca displacing solution (15% KNO3 and 6% Ca(NO3)2.4H2O) and subsequent measurement of NH4+ by auto-analyser. The organic carbon content was measured in a Leco furnace and the ESP was calculated from exchangeable sodium concentrations and the measured CEC. To measure wet aggregate stability, 25 g of soil aggregates (5-10 mm) were wet-sieved for 10 min on a set of two sieves (0.25, 0.125 mm), following the method described by Kemper and Rosenau (1986). The fraction remaining on each sieve was collected, oven dried at 105°C and weighed. The fraction <0.125 mm was calculated by difference. To measure spontaneous dispersion, 20 g of sieved soil (< 2 mm) was placed in a 150 mL vial and 100 mL of deionised water was poured gently down the side of a vial, which was then allowed to stand for 24 h. The vial was then slowly inverted once and returned upright, giving just enough agitation so that the aggregates simply slid down the side of the vial. The supernatant was then stirred slowly and gently to allow the suspended matter to mix without further disturbance to the soil sediment in the container. After allowing an appropriate sedimentation time for particles > 2 µm, the supernatant containing dispersed clay was drawn off and its concentration measured by turbidimetry. An individual calibration curve for both soils was used to convert turbidity to clay concentration. A similar procedure was used to measure mechanical dispersion. The 1:5 soil:water suspension was agitated on an end-over-end shaker for one hour at 60 rpm. The top 2 cm of suspension containing < 2 µm material was drawn off after allowing an appropriate sedimentation time and measured by turbidimetry. 6.4.2 Results Leachate properties Figures 16 and 17 show the effects of molasses and gypsum on the properties of successive leachates from the Burdekin and Proserpine soils, respectively. The first leaching was conducted on soil that had been slowly wetted to 85% of field capacity; subsequent leachings were conducted on air-dried soil. After five leachings, there were significant changes in leachate properties imposed by molasses and gypsum treatments for both soils. As the fourth leachate was only analysed for EC and pH, data for this leaching are not presented. However, as the pH and EC values in the fourth leachate were intermediate between those of leachates 3 and 5, it is probable that SAR and DOC values were also intermediate between those of leachates 3 and 5. In the Burdekin soil (Figure 16), pH and SAR were generally reduced when molasses and gypsum were added. However, in the case of the lighter-textured, more acid Proserpine soil (Figure 17), molasses tended to increase pH and SAR, whereas gypsum did the opposite. Molasses and gypsum increased the EC of leachate in both soils. For the Burdekin leachate, the EC was generally greatest in the gypsum-only treatment compared with either molasses-only or molasses+gypsum. The highest EC was found in the second leachate of the gypsum-only treatment and the lowest EC was found in the fifth leachate of the control. For the Proserpine soil, the EC was greatest for the molasses+gypsum treatment. The lowest EC was found in the first leachate of the control and the highest EC was found in the third leachate of the molasses+gypsum treatment. The average EC of Burdekin leachates ranged between 0.5 and 6.5 dS/m for all treatments and for the Proserpine soil, between 0.3 and 5.5 dS/m. pH EC 47 8 lsd (p = 0.01) 6 4 2 0 C G 8.5 lsd (p = 0.01) 8.0 M 7.5 7.0 6.5 CGM 10 lsd (p=0.01) LI L II 8 L III LV 5 M+ M+ SAR 3 0 500 375 C lsd (p=0.01) G M M+ DOC 250 125 0 C G M M+ Figure 16 pH, EC, SAR, DOC in successive leachates of the Burdekin soil after molasses and gypsum treatments. C, M, G, M+G refer to control, molasses, gypsum and molasses+gypsum, respectively. L I-V are leachates from leaching events I-V 48 6 lsd (p = 0.01) 4 pH EC 2 0 C G 7.5 lsd (p = 0.01) LI L II 6.5 L III LV 5.5 M M+ 4.5 C G 2.0 lsd (p = 0.05) 1.5 M M+ SAR 1.0 0.5 0.0 CG 500 lsd (p = 0.01) 375 M M+ DOC 250 125 0 C G M M+ Figure 17 pH, EC, SAR, DOC in successive leachates of the Proserpine soil after molasses and gypsum treatments. C, M, G, M+G refer to control, molasses, gypsum and molasses+gypsum respectively. L I-V are leachates from leaching events I-V 49 Dissolved organic carbon (DOC) concentrations were high in molasses treatments. For the Burdekin soil, the highest DOC was found in the second leachate of the molasses-only treatment and the lowest DOC was found in the fifth leachate of the gypsum-only treatment. DOC generally decreased as leaching proceeded, except in the control. For the Proserpine soil, the lowest DOC was found in the first leachate of the gypsum-only treatment and the highest was found in the second leachate of the molasses-only treatment. In general, DOC was less in molasses+gypsum than in molasses-only treatments in both soils as calcium presumably immobilised some organic carbon. Clay dispersibility Both gypsum and molasses reduced spontaneous and mechanical dispersion substantially in both soils (Figure 18 and 19). In the Burdekin soil, the reduction in spontaneously dispersed clay compared to the control was dramatic when gypsum or molasses was added alone or in combination. The lowest spontaneously dispersed clay was found in the molasses+gypsum treatment, although this was not significantly different from the gypsum-only and molasses-only treatments, and the highest one was found in the control. When the soil was mechanically dispersed, gypsum, alone or combined with molasses, dramatically reduced clay dispersion, whereas the effect of molasses alone was much less pronounced. The highest mechanically dispersed clay was found in the control and the lowest one was found in the gypsum-only treatment. 4 Spontaneousl dispersed clay 3b 2 1 aaa 0 22.0 c Mechanicall dispersed clay 16.5 b 11.0 Figure 18 5.5 a 0.0 a C G M M+G Spontaneous dispersion and mechanical dispersion of the Burdekin soil after gypsum and molasses treatments. C, M, G, M+G refer to control, molasses, gypsum and molasses+gypsum (each 10 t/ha), respectively. Letters show which results are significantly different (P = 0.001) 50 In the coarser-textured Proserpine soil (Figure 19), the amount of spontaneously and mechanically dispersed clay was generally much less than in the Burdekin soil. Spontaneous dispersion was reduced by all treatments, with molasses+gypsum being the most effective. The lowest spontaneously dispersed clay was found in the molasses+gypsum treatment and the highest was found in the control. When soil was mechanically disturbed, gypsum treatments produced a dramatic reduction in dispersed clay compared with the molasses-only treatment. The lowest mechanically dispersed clay was found in the gypsum-only treatment and the highest one was found in the control. Spontaneous dispersed clay 0.15 d 0.10 c 0.05 b a 0.00 Figure 19 Mechanicall dispersed clay 5d 4 c 3 2 1b a 0 C G M M+G Spontaneous dispersion and mechanical dispersion of the Proserpine soil after gypsum and molasses treatments. C, M, G, M+G refer to control, molasses, gypsum and molasses+gypsum (each 10 t/ha), respectively. Letters show which results are significantly different (P = 0.001) Wet aggregate stability The effects of molasses and gypsum on wet aggregate stability of the soils are shown in Figure 20. In the Burdekin soil, molasses and gypsum treatments either alone or combined, provided significant increases in aggregate stability. Molasses and gypsum increased the proportion of macro-aggregates (> 250 µm) and decreased the proportion of aggregates < 125 µm, with the combined effects being greatest; aggregates in the range of 125-250 µm were relatively unaffected except by the combined treatment. In the Proserpine soil, gypsum had no significant effect on aggregate stability, either alone or 51 combined with molasses. Molasses significantly increased the proportion of macroaggregates, while decreasing the proportion of both aggregate fractions < 250 µm. % Aggregate Burdekin 100 75 50 25 0 CGM <0.125 mm 0.125-0.25 mm >0.25 mm 100 Proserpine M+ % Aggregate 75 50 25 0 C G M M+ Figure 20 Wet aggregate stability of the Burdekin and Proserpine soils after molasses and gypsum treatments. C, M, G, M+G refer to control, molasses, gypsum and molasses+gypsum, respectively Soil chemical properties Table 26 shows the chemical properties of the two soils at the end of the incubation. Leaching alone decreased pH for the control treatment in the Burdekin soil, but the effect was reversed in the Proserpine soil. In the Burdekin soil, the decrease in pH was enhanced by the addition of both treatments with the gypsum-only treatment being greatest. However, molasses increased pH in the Proserpine soil, while gypsum did the opposite. The lowest pH was observed in the gypsum-only treatment for both soils and the highest pH was observed in the control and in the molasses-only treatment for the Burdekin and Proserpine soils, respectively. 52 The EC of both soils was significantly increased by gypsum but not by molasses (Table 26). In the Burdekin soil, the EC of the soil treated with molasses+gypsum was slightly greater than that treated with gypsum alone. The EC values of the gypsum-treated soils indicated that gypsum remained in the soil after five leaching events. The highest EC was observed in the molasses+gypsum treatment, and this was significantly different for all other treatments; the lowest EC was observed in the molasses-only treatment, although this was not significantly different from the control. In the Proserpine soil, the highest EC was observed in the gypsum-only treatment, although this was not significantly different from the molasses+gypsum treatment. The lowest EC was observed in the control, although this was not significantly from the molasses-only treatment. Table 26 pH and electrical conductivity (EC) in 1:5 soil:water extracts, exchangeable sodium percentage (ESP), cation exchange capacity (CEC) and organic carbon content of the Burdekin and the Proserpine soils after molasses and gypsum treatments and five leaching events. C, G, M and M+G refer to control, gypsum (10 t/ha), molasses (10 t/ha) and molasses+gypsum, respectively Treatment pH EC (dS/M) ESP CEC (cmol(+)/kg) Org. C (%) Burdekin soil Initial 8.0 0.2 7.9 10.9 0.8 C 7.7c 0.1a 4.1c 10.6a 0.8a G 7.1a 1.6b 0.4a 10.4a 0.7a M 7.6b 0.2a 2.4b 11.6b 1.0b M+G 7.1a 1.9c 0.2a 11.2b 1.0b Significance * * ** ** ** Proserpine soil Initial 4.9 0.1 18.8 2.3 0.7 C 5.4b 0.1a 3.4c 2.1a 0.6a G 4.6a 2.2b 0.3a 2.2a 0.7a M 6.4d 0.1a 2.1b 2.5b 0.8b M+G 6.1c 2.1b 0.2a 2.5b 0.9b Significance ** ** ** ** ** Letters (a, b, c, d) show which results are significantly different at P = 0.05 (*) and P = 0.01 (**) Exchangeable sodium percentage (ESP) was determined to investigate the changes in sodicity after addition of molasses and gypsum; both treatments reduced the ESP substantially. Leaching alone decreased the ESP for the control in both soil and its effect was more pronounced in the Proserpine soil. In both soils, gypsum reduced ESP more than molasses because of the high concentration of calcium in solution. The lowest ESP was observed in the molasses+gypsum treatment although this not significantly different from those of gypsum-only. Cation exchange capacity (CEC) and organic carbon were significantly increased by molasses treatment in both soils, whereas the gypsum had no effect. The concentrations of organic carbon in leachate and in soil were used to calculate the percentage of carbon lost in respiration, presumably released as CO2 or leached as DOC during experiments (Table 27). n general, a substantial proportion of the molasses added 53 was lost in respiration (CO2) but a small proportion was leached as DOC. In the molasses-only and molasses+gypsum treatments of both soils, the percentages of carbon released as CO2 and leached as DOC were similar at 31-36% and 1-2%, respectively. Table 27 Percentage of carbon lost in respiration (CO2) and dissolved organic carbon (DOC) leached in the control (C), gypsum (G), molasses (M) and molasses+gypsum (M+G) treatments in the Burdekin and Proserpine soils Treatment Initial Burdekin soil C 8000 G 8000 M 14222 M+G 14222 Proserpine soil C 7000 G 7000 M 13222 M+G 13222 Organic carbon (mg/kg) Leachate (DOC) Soil (final) Leachate + soil % released as CO2 % leached as DOC 73 8050 8123 -1 0.9 33 7225 7258 9 0.5 205 9525 9730 32 2.1 164 9700 9864 31 1.7 23 6200 6223 11 0.4 19 6700 6719 4 0.3 175 8300 8475 36 2.1 123 8625 8748 34 1.4 The concentrations of cations in the leachate and on the exchange sites of soil were used to calculate the abundances of cations in the system at the beginning and at the end of experiments (Table 28). Molasses and gypsum affected the proportion of cations in the leachate and on the exchange sites of both soils. In the Burdekin soil, when gypsum was added, Ca in the leachate and in the soil increased as expected. The displacement of other exchangeable cations by Ca can be seen in Table 27. In general, the proportions of Na, Mg and K cations in the leachate were higher than in the soil when gypsum was added. Molasses also facilitated removal of Na, but not to the extent of the gypsum treatment. The proportions of Ca, Mg and K were generally higher in soil than in leachate for the molasses-only treatment in both soils; molasses contains considerable amounts of these cations. In the Burdekin soil, when molasses+gypsum was added, 95% of Na was removed in the leachate; in the Proserpine soil, a lower proportion of Na was found in the leachate. Adding molasses+gypsum partially compensated for the loss of K through leaching; this loss of plant available-K might be a negative effect of adding gypsum alone. 54 Table 28 Cation contents in soil (cmol(+)/kg) in control (C), gypsum (G), molasses (M) and molasses+gypsum (M+G) treatments Treatment C G M M+G C G M M+G C G M M+G C G M M+G C G M M+G C G M M+G C G M M+G C G M M+G Initial Soil Addition 6.5 0.0 14.4 0.04 14.4 0.9 0.0 14.4 0.0 14.4 3.9 0.0 0.0 0.1 0.1 0.4 0.0 0.0 0.1 0.1 0.9 0.0 0.0 0.01 0.01 0.4 0.0 0.0 0.01 0.01 0.5 0.0 0.0 0.2 0.2 0.4 0.0 0.0 0.2 0.2 Total Leachate Ca, Burdekin 6.5 0.1 20.9 1.9 6.6 0.2 20.9 1.9 Ca, Proserpine 0.9 0.1 15.3 1.8 0.9 0.2 15.3 2.0 Mg, Burdekin 3.9 0.1 3.9 3.0 4.0 0.3 4.0 3.4 Mg, Proserpine 0.4 0.1 0.4 0.2 0.5 0.1 0.5 0.3 Na, Burdekin 0.9 0.51 0.9 1.01 1.0 0.66 1.0 0.99 Na, Proserpine 0.4 0.26 0.4 0.35 0.4 0.31 0.4 0.33 K, Burdekin 0.5 0.02 0.5 0.3 0.7 0.1 0.7 0.7 K, Proserpine 0.4 0.03 0.4 0.4 0.7 0.3 0.7 0.4 Soil 7.1 17.9 7.2 19.9 0.8 13.7 1.4 17.3 4.8 1.3 4.7 1.6 0.3 0.2 0.4 0.2 0.53 0.09 0.33 0.05 0.16 0.05 0.14 0.10 0.5 0.01 0.8 0.1 0.4 0.0 0.4 0.4 Final Total Leachate (%) 7.2 1 19.8 9 7.4 3 21.7 9 0.8 10 15.5 11 1.6 11 19.4 10 4.9 3 4.3 69 5.0 6 4.9 68 0.4 19 0.4 52 0.5 24 0.5 64 1.0 49 1.1 92 1.0 67 1.0 95 0.4 62 0.4 87 0.5 69 0.4 76 0.5 5 0.3 96 0.9 7 0.8 89 0.4 7 0.4 93 0.7 46 0.8 54 Soil (%) 99 91 97 91 90 89 89 90 97 31 94 32 81 48 76 36 51 8 33 5 38 13 31 24 95 4 93 11 93 7 54 46 55 6.4.3 Discussion The questions addressed in this study were whether molasses alone or combined with gypsum could improve the structural stability of sodic soils. The results of leachate analysis show that additions of molasses and gypsum significantly changed the properties of sodic topsoils. For the soil leachates, gypsum increased EC and decreased pH immediately. The effect of the gypsum on pH was due to the displacement of adsorbed protons by calcium. Similar results were found by Sekhon and Bajwa (1993) and by Chorom and Rengasamy (1997). However, this effect may be temporary and progressively lost as the gypsum is leached. In the Burdekin soil, the EC of leachate in the molasses+gypsum treatment was generally lower than of the treatment with gypsum alone (Figure 16). This may indicate that some soluble calcium is complexed by organic matter and immobilized in the soil. Conversely, organic matter may have been protected from leaching by the presence of soluble Ca in the soil. This can be seen in Figure 17 and Table 27, where DOC was generally lower with molasses+gypsum than with molasses alone. Muneer and Oades (1989) also found that DOC decreased when gypsum and glucose were added to soil. When gypsum was applied, the concentration of DOC decreased; soluble calcium from gypsum inhibited decomposition of organic matter by microorganisms and reduced the release of the DOC by leaching. Calcium tends to block and protect functional groups in organic matter that represent sites of initial decomposition and also cross-links flexible polymers to create more dense, rigid molecules that are more stable to both chemical and biological degradation (Oades 1989). However, in the Proserpine soil the EC after molasses+gypsum treatment was higher than after gypsum alone (Figure 17). This may be because the clay content of the soil was too low to protect cations in the soil solution from leaching so that the effects of molasses and gypsum on EC were simply additive in this case. The small decrease in pH of the Burdekin soil brought about by addition of molasses is most likely due to the pH of the molasses (5.5), and to the CO2 and organic acids produced during incubation. However, in the less buffered, more acidic Proserpine soil, addition of molasses increased soil pH to a higher value than that of either the soil or the molasses. This may have been due to reactions such as the mineralisation of cations or nitrogen in the molasses, or to the uptake of nitrate by microorganisms. Cation exchange capacity (CEC) in both soils increased significantly with molasses but not with gypsum treatment. This was presumably due to the decomposition products of the carbohydrates in molasses having high CEC. The data in Table 26 suggest that this additional organic matter in the Burdekin soil had a CEC of approximately 300 cmol(+)/kg C, whereas that in the Proserpine soil had a CEC of about 160 cmol(+)/kg C; these values were calculated from the increases in soil CEC in relation to increases in organic carbon (Table 26). Organic exchange sites may have a higher selectivity for Ca than mineral exchange sites (Nelson and Oades 1998). The water stability of aggregates and the clay dispersibility of the soils were significantly affected by molasses and gypsum. In the Burdekin soil, both molasses and gypsum treatments increased macro-aggregation at the expense of micro-aggregates. Organic matter from molasses, probably mostly carbohydrate, was able to stabilise micro- 56 aggregates into macro-aggregates, while the gypsum effect was mainly due to the flocculation of clay resulting in the aggregation of clay particles into domains and reduced dispersion (Chorom and Rengasamy 1997; Nelson and Oades 1998). In the Proserpine soil, molasses increased wet aggregate stability to an even greater degree than in the Burdekin soil, but, as with the Burdekin soil, had a far less dramatic effect on mechanical dispersion, indicating that molasses was unable to stabilise aggregates under these more destructive conditions. The resistance to slaking afforded by increased organic matter is probably insufficient to withstand the rigours of mechanical dispersion in which the preliminary destruction of macro-aggregates provides opportunities for extensive dispersion. On the other hand, gypsum was able to prevent clay from dispersing but unable to maintain macro-aggregates. The extents of both spontaneous and mechanical dispersion in both soils generally reflect the ESP and EC data in Table 26 with higher ESP and lower EC promoting dispersion. It is significant that treatment of both soils with molasses alone resulted in substantial reductions in ESP and spontaneous dispersion. 6.5 Structural stability of sodic soils as influenced by molasses and lime The addition of gypsum to soil can increase the stability of aggregates by lowering the exchangeable sodium percentage (ESP), increasing the electrolyte concentration and, thereby, reducing dispersion. However, gypsum additions are lost relatively quickly when water infiltrates through soil because gypsum is quite soluble. Lime is a less soluble source of Ca, and also tends to be cheaper than gypsum in the Queensland sugarcanegrowing districts. Lime may in fact be too insoluble to be of benefit, particularly in soils with high pH. However, when considering the joint use of lime and molasses, there are several mechanisms whereby the addition of molasses may help to dissolve the lime. Organic acids and CO2 are produced by microbial decomposition of organic matter, and under certain circumstances may reduce soil pH (Baldock et al. 1994). Therefore, in soils of high pH, the addition of large quantities of organic matter in the form of molasses might be expected to lower pH out of the alkaline range. The increased partial pressure of CO2 and any organic acid might then aid the dissolution of the lime, freeing Ca2+ ions (Robbins 1986) and lowering the soil pH. The Ca2+ ions may then exchange with Na+ ions adsorbed to clay particles, allowing Na+ ions to be leached, thus eventually decreasing the ESP (Lehrsch et al. 1993). In this trial, we assess the influence of molasses and lime on the structural stability of soils. 6.5.1 Materials and methods The materials and methods for this experiment are virtually identical to those described in Section 6.4.1, except that the treatments in this experiment were molasses and lime rather than molasses and gypsum. The amount of calcium added to the soils as lime was similar to that added as gypsum described in the previous experiment (Section 6.4.1), and the application rate of molasses was also the same. Table 1 describes some of the properties of the soils we used. 57 6.5.2 Results Leachate properties of treated soils Figures 21 and 22 show the effects of molasses and lime on the properties of five successive leachates from the Burdekin and Proserpine soils. The first leaching was conducted on soil that had been slowly wetted to 85% of field capacity and kept for 7 d.; subsequent leaching events were conducted after soil had been air-dried from the previous leaching event. There were significant changes in leachate properties related to molasses and lime treatments. In general, pH, electrical conductivity (EC), sodium adsorption ratio (SAR) and dissolved organic carbon (DOC) decreased as leaching proceeded, although the magnitude of reductions fluctuated. In the Proserpine soil, the pH of the first leachate was lower than that of the second leachate in all treatments, even though these were not significantly different for the lime-only and the molasses+lime treatments. In the Burdekin soil, by contrast, the pH was lower in the first leachate of the control and the molasses-only treatments. This may simply be an experimental artefact caused by the first leaching being conducted on moist soil, while subsequence leachings were conducted on air-dried soil. Similar results were found by Nelson et al. (1998), where the pH of soil is lower under moist conditions than dry. The average pH of the Burdekin leachates ranged between 7.2 and 8.2 in all treatments, while for the Proserpine soil the range was 4.8 to 8.0. In the Burdekin soil, the highest and the lowest pH were observed in the first leachate of the lime-only and the control treatments respectively and in the Proserpine, the highest and the lowest pH were observed in the second leachate of the molasses+lime and the first leachate of the control, respectively. Electrical conductivity (EC) of leachates decreased from the first to the fifth leachate in virtually all treatments for both soils. This indicates that cation dissolution was greater in the early stages and diminished as successive leaching proceeded. The EC of each leachate in the molasses+lime treatment was usually higher than for the leachates of all other treatments in the same leaching events. Sodium adsorption ratio (SAR) of leachates decreased as leaching proceeded in all treatments for both soils. In the Burdekin soil, the treatments had little effect on leachate SAR, whereas in the Proserpine soil SAR values were highest in the molasses and lime+molasses treatments. In all leaching events, dissolved organic carbon (DOC) concentration was highest in the molasses-only treatment and second highest in the lime+molasses treatment. It was highest in the first or second leaching event in all cases, and decreased thereafter. In the Burdekin soil, lime alone decreased DOC, whereas in the Proserpine soil lime alone had no effect. EC (dS/m) pH 8.5 lsd (p=0.001) 8.0 7.5 7.0 2.0 lsd (p=0.001) 1.0 58 0.0 12 lsd (p=0.001) 8 SAR 4 DOC (mg/L) 0 500 375 250 125 0 Figure 21 lsd (p=0.001) LI LII LIII LIV LV C L M L+M Burdekin soil: pH, electrical conductivity (EC), sodium adsorption ratio (SAR) and dissolved organic carbon (DOC) of five successive leaching events (LI- LV) after molasses and lime treatments. C, L, M and M+L refer to control, lime, molasses, molasses + lime. Error bars represent one standard deviation 59 pH 8 lsd (p=0.001) 6 4 3 lsd (p=0.001) 2 1 0 2 lsd (p=0.001) 1 EC (dS/m) SAR 0 500 375 lsd (p=0.001) 250 LI LII LIII LIV LV DOC (mg/kg) 125 0 C L M L+M Figure 22 Proserpine soil: pH, electrical conductivity (EC), sodium adsorption ratio (SAR) and dissolved organic carbon (DOC) of five successive leaching events (LI- LV) after molasses and lime treatments. C, L, M and M+L refer to control, lime, molasses, molasses + lime. Error bars represent one standard deviation 60 Clay dispersibility Figure 23 and 24 present the change in spontaneous and mechanical dispersion in the soils. In the Burdekin soil, molasses and lime, alone or together, decreased spontaneous dispersion dramatically and to an equal extent (Figure 23). The amount of dispersed clay for each treatment was markedly greater when the soil was mechanically dispersed. Lime, molasses and lime+molasses treatments significantly reduced (P = 0.01) the dispersed clay compared with the control. 2.0 b Spontaneously dispersed clay (g/kg) 1.0 0.0 20 b 10 aaa a aa Mechanically dispersed clay (g/kg) Figure 23 0 C L M M+L Spontaneous and mechanical dispersion of the Burdekin soil as influenced by molasses and lime. C, L, M, M+L represent control, lime-only, molasses-only and molasses+lime treatments, respectively. Letters show results that are significantly different (P = 0.01). Error bars represent one standard deviation In the coarser-textured Proserpine soil, as was observed in the previous experiment (Section 6.4), the amount of dispersed clay in spontaneous and mechanical dispersion was much less than that in the Burdekin soil (Figure 24). The spontaneously dispersed clay was significantly higher in the control than in the other treatments. However, when the soil was mechanically dispersed, the amounts of the dispersed clay in the molasses-only and lime-only treatments were not significantly different from the control. 61 Spontaneously dispersed clay (g/kg) b 0.10 a aa 0.05 0.00 5.0 b ab b 2.5 a Mechanically dispersed clay (g/kg) 0.0 C L M M+L Figure 24 Spontaneous and mechanical dispersion of the Proserpine soil as influenced by molasses and lime. C, L, M, M+L represent control, lime-only, molasses-only and molasses+lime treatments, respectively. Letters show results that are significantly different (P = 0.01). Error bars represent one standard deviation Wet aggregate stability Figure 25 shows the influence of molasses and lime application on wet aggregate stability of soils. In the Burdekin soil, molasses-only, lime-only and molasses+lime treatments all increased the proportion of macroaggregates (> 0.250 mm) largely at the expense of aggregates fraction < 0.125 mm. This indicates an increase in water stability of aggregates. The greatest proportion of macroaggregates (or the smallest proportion of aggregates < 0.125 mm) occurred in the molasses+lime treatment, although this was not significantly different (P = 0.01) from the molasses-only treatment, yet significantly different from the lime-only treatment. The proportion of 0.125-0.250 mm aggregates was relatively constant in all treatments. In the Proserpine soil, the proportion of macroaggregates increased at the expense of both smaller aggregate fractions in both molasses-only and molasses+lime treatments. In the lime-only treatment, the proportion of aggregates < 0.125 mm appears to have increased marginally with a complementary decrease in aggregates 0.125-0.250 mm, while the proportion of macroaggregates remained unchanged. 62 Burdekin 100 75 % aggregates 50 25 0 <0.125 mm 0.125-0.25 mm Proserpine >0.25 mm 100 75 % aggregates 50 25 Figure 25 0 C L M M+L Wet aggregate stability of soils as influenced by molasses and lime treatments. C, L, M, M+L represent control, lime-only, molasses-only and molasses+lime treatments, respectively Chemical properties of treated soils Table 29 presents the change in chemical properties of the Burdekin and Proserpine soils after molasses and lime treatments. In the Burdekin soil, pH decreased in the control and with molasses alone, increased with lime alone, and stayed the same with lime+molasses. In Proserpine soil, pH increased in all treatments. The EC of both soils was generally increased by molasses and lime applications, with their effects being additive. Leaching alone reduced the EC for both soils with the effect being more pronounced in the Proserpine soil. The EC for both soils increased in the order of treatments control molasses >lime =molasses+ lime. The cation exchange capacity (CEC) and organic carbon content were significantly increased for both soils after addition of molasses but not lime. The abundance of cations in the system at the beginning and at the end of the experiment is shown in Table 30. In general, molasses and lime increased the proportion of cations in the leachate and on exchange sites of both soils. The cations in the leachates for each treatment were higher in the Proserpine soil than that in the Burdekin soil except for sodium. This reflects the coarser texture of Proserpine soil relative to that of the Burdekin soil. It should be noted, however, that the volume of water leached during each event was greater in the Burdekin soil than in the Proserpine soil (88 mL versus 72 mL). The proportion of cations in leachate was small compared with those on the exchange phases of soil in all treatments except sodium in the Burdekin soil. In the Burdekin soil, it appeared that Ca from lime displaces exchangeable Na but not exchangeable Mg and K, which is fortunate for plant nutrition. 64 Table 30 Cation abundance (cmol(+)/kg) in control (C), lime (L), molasses (M) and molasses+lime (M+L) treatments Treatment Initial Soil Addition Total Leachate Soil C L M L+M C L M L+M C L M L+M C L M L+M C L M L+M C L M L+M C L M L+M C L M L+M Ca, Burdekin 7.9 0.0 7.9 0.1 7.4 17.9 25.8 0.1 23.3 0.04 7.9 0.2 7.5 17.9 25.8 0.2 14.5 Ca, Proserpine 0.9 0.0 0.9 0.1 0.8 17.9 18.7 0.5 15.8 0.04 0.9 0.1 0.9 17.9 18.8 0.5 15.0 Mg, Burdekin 4.7 0.0 4.7 0.1 4.8 4.7 0.0 4.7 0.1 4.3 0.1 4.8 0.2 4.7 0.1 4.8 0.2 4.2 Mg, Proserpine 0.4 0.0 0.4 0.1 0.3 0.4 0.0 0.4 0.1 0.3 0.1 0.4 0.1 0.6 0.1 0.4 0.2 0.6 Na, Burdekin 1.4 0.0 1.4 0.5 1.2 1.4 0.0 1.4 0.6 0.6 0.01 1.4 0.6 0.5 0.01 1.4 0.6 0.4 Na, Proserpine 0.7 0.0 0.7 0.1 0.7 0.7 0.0 0.7 0.1 0.7 0.01 0.7 0.1 0.7 0.01 0.7 0.1 0.7 K, Burdekin 0.5 0.0 0.5 0.01 0.5 0.5 0.0 0.5 0.01 0.5 0.2 0.8 0.05 0.9 0.2 0.8 0.05 0.8 K, Proserpine 0.4 0.0 0.4 0.03 0.4 0.4 0.0 0.4 0.04 0.4 0.2 0.7 0.3 0.6 0.2 0.7 0.4 0.5 Final Total Leachate (%) Soil (%) 7.5 1.2 23.4 0.6 7.7 2.4 14.7 1.5 98.8 99.4 97.6 98.5 0.9 9.1 16.3 3.3 1.0 11.2 15.5 3.2 90.9 96.7 88.8 96.8 4.9 1.9 4.4 2.4 4.9 4.3 4.4 5.0 98.1 97.6 95.7 95.0 0.4 28.3 0.4 26.0 0.7 18.1 0.8 25.6 71.7 74.0 81.9 74.4 1.7 30.1 1.2 49.6 1.2 54.4 1.0 61.2 69.9 50.4 45.6 38.8 0.7 8.0 0.8 11.4 0.8 11.6 0.8 16.6 92.0 88.6 88.4 83.4 0.5 1.3 0.5 1.1 0.9 5.1 0.9 5.3 98.7 98.9 94.9 94.7 0.4 6.7 0.4 9.3 0.9 31.8 0.9 42.2 93.3 90.7 68.2 57.8 65 The proportion of carbon lost in respiration, presumably released as CO2, and leached as DOC was calculated to show the balance of organic carbon in the system during the period of experiment (Table 31). The different amounts of carbon in the soil (before and after incubation) and in the leachate were assumed to be due to carbon lost in respiration. In general, a substantial proportion of molasses added was lost in respiration (CO2) and only a small proportion was leached as DOC. In the molasses-only treatment, the proportion of CO2 was higher in the Proserpine soil than that in the Burdekin soil. In both soils, the proportions of carbon lost in respiration and leached as DOC in the molassesonly and molasses+lime treatments were in range of 30-36% and 1-2%, respectively. DOC was higher in the molasses-only than in the molasses+lime treatments, which is consistent with what was observed in the previous pot trial (Section 6.4). In the lime-only and control treatments, the proportion of carbon lost in respiration was very small compared with molasses-only and molasses+lime treatments. Table 31 Percentage of carbon lost in respiration (CO2) and dissolved organic carbon (DOC) leached in the control (C), lime (L), molasses (M) and molasses+lime (M+L) treatments in the Burdekin and Proserpine soils Treatment Burdekin soil C L M M+L Proserpine soil C L M M+L Organic carbon (mg/kg soil) Initial Leachate (DOC) Soil (final) Leachate + soil 7800 7800 14022 14022 96 48 230 141 8050 8303 9525 9146 8146 8351 9755 9287 7000 7000 13222 13222 35 33 227 140 6200 7136 8300 8621 6235 7169 8527 8761 %% released leached as CO2 as DOC -4 1.2 -7 0.6 30 1.6 34 1.0 11 0.5 -2 0.5 36 1.7 34 1.1 6.5.3 Discussion Lime and molasses both had significant effects on soil properties. The addition of molasses did not increase the solubility of lime as was expected, despite a significant amount of CO2 being produced. Good aeration may have prevented a build-up of CO2 concentration in the soil. DOC of the leachates increased markedly in the molasses treatments. However, when lime was also added, the DOC was lower than that of the molasses-only treatment. This may indicate that some dissolved organic matter was protected from leaching by the presence of calcium in the soil, as was found and discussed in the previous pot trial (Section 6.4). Further experiments are needed to determine the best combination rates of molasses and lime to overcome problems associated with sodicity. In the Proserpine soil, the pH was increased when molasses was added as was observed and discussed in pot trial 2. 66 Wet aggregate stability and clay dispersibility of soils were influenced by molasses and lime. An increase in macroaggregation occurred at the expense of microaggregation, which may have been due to carbohydrate production after molasses addition plus fungal hyphae produced during incubation. Observations made during the experiments indicate that fungal hyphae were abundant in the soils to which molasses was added. However, macroaggregates are easily disrupted and broken into microaggregates when increasing dispersive energy was applied to soil. In the Burdekin soil, which had a clay content 4 times higher than the Proserpine soil, the amount of spontaneously and mechanically dispersed clay brought about by the molasses and lime treatments were significantly lower than that of the control. This probably results from strong binding of mucilages and colloidal organo-mineral complexes, which may have induced less clay to disperse even though the higher dispersive energy was applied to soil (Tisdall and Oades 1982). 6.6 Effect of trash management on susceptibility of soils to becoming or remaining sodic Depending on management, soils can become more or less sodic. For example, irrigation with saline water increases soils sodicity, as does rising groundwater. On the other hand, application of gypsum or lime, together with irrigation with good quality water and adequate drainage, can decrease sodicity. For any given management regime, the extent to which soil becomes more or less sodic depends on the ‘cation exchange selectivity’ of the soil. Cation exchange selectivity refers to the fact that some cations are held more strongly than others. For example, the cations Ca2+ and Mg2+, which have two charges (divalent), are held more strongly than Na+ and K+, which have one charge (monovalent). However, the difference between binding strength of divalent and monovalent cations depends on the nature of the cation exchange sites. Exchange sites on organic matter generally have higher relative selectivity for divalent cations than exchange sites on clays. Therefore, soils with high organic matter content are generally less susceptible to becoming sodic than soils with low organic matter content (Nelson and Oades 1998). The hypothesis tested in this experiment was that sugarcane trash management, through its effect on soil organic matter, alters the cation exchange selectivity of soil. Therefore, GCTB, as opposed to burning, may make soils less susceptible to becoming sodic, and easier to ameliorate if already sodic. 6.6.1 Materials and method Soils were sampled from the long-term GCTB/burnt trials at Ayr, Mackay and Tully (0-5 cm depth) and from a long-term pasture/wheat-fallow trial in Adelaide (0-10 cm depth). Soil properties are shown in Table 32. 67 Table 32 Selected properties of soil used in this experiment Soil sites and OC CEC pH Particle size (%) Mineralogy management (g/kg) (cmol(+)/kg) Clay Silt Sand Ayr (B) 14.4 8.0 7.04 10 11 79 I#, K#, Q*, F* Ayr (G) 17.5 9.5 6.91 Tully (B) 12.3 7.3 5.53 34 35 31 K#, S#, I*, Q*, F* Tully (G) 18.1 11.0 5.23 Mackay (B) 13.1 7.7 5.49 18 25 57 K#, F#, Q*, I* Mackay (G) 14.3 7.7 5.35 Urrbrae (WF) 11.4 9.9 5.85 11 27 62 RIM#, I#, Q*, K* Urrbrae (PP) 31.1 10.3 5.65 OC = organic C content, CEC = cation exchange capacity, pH is in 1:5 soil:water extract, K = kaolinite, RIM = randomly interstratified minerals, I = illite, Q = quartz, S = smectite, F = feldspar, B = burnt, G = green, WF = wheat/fallow and PP = permanent pasture. * and # = minor and major clay mineral Cation exchange selectivity experiments are usually conducted by mixing the soils with solutions containing various salt solutions, shaking them, and then removing the solution by centrifugation. That technique, while rapid and simple, destroys soil structure and does not mimic the percolation of water that occurs in the field. In this experiment, soil aggregates were placed on sintered glass funnels, and solutions were introduced and removed by slowly raising or lowering the pressure head. The technique, while slow and laborious, is a much better representation of the conditions under which cation exchange occurs in the field. Mercury (1 mg/L) was added to all salt solutions to prevent microbial activity during the experiment. The soils (three replicates of each) were initially saturated with CaCl2, and excess salt was removed with water washes. They were then equilibrated, by repeated saturation and drainage, with solutions containing different amounts and proportions of NaCl and CaCl2. In the solutions, total amount of salt is expressed as total cation concentration (TCC), and the amount of sodium as a proportion of the total salt is expressed as sodium adsorption ratio (SAR). Two levels of TCC (2 and 10mmol(+)L-1) and two levels of SAR (3 and 10 mmol(+)L-1) were used. Higher values of TCC and SAR could not be used because they led to aggregate collapse and clogging of the sintered funnel with dispersed clay. After equilibration, solutions were analysed for Na and Ca using flame photometry and atomic absorption spectrophotometry. Cation exchange selectivity, expressed as the Gapon coefficient (kG, see equation below), was calculated using mass balances. This method was validated by sampling the soils and measuring exchangeable cation contents. The higher the Gapon coefficient, the higher the selectivity for Na, and the more likely the soil is to become or remain sodic. Gapon coefficient (kG) = ESR / SAR Where ESR or exchangeable sodium ratio is a measure of the sodicity of the soil (exchangeable Na dived by exchangeable Ca), similar to ESP, and where SAR or sodium adsorption ratio is the sodicity of the water passed through the soil. 68 6.6.2 Results and discussion Figure 26 shows that the method developed for measuring kG, using solution properties and a mass balance approach, gave virtually identical results to the conventional method, using soil and solution properties. 0.012 0.008 y = 1.015x R2 = 0.98 kG from solution properties alone 0.004 Figure 26 0.000 0.000 0.004 0.008 0.012 kG from soil and solution properties (ESR/SAR) Relationship between kG from soil and solution properties (ESR/SAR) and from solution properties alone 0.0100 0.0075 0.0050 d c c bc c a c a kG (mmol L-1)-0.5 0.0025 0.0000 BG Ayr BG Mackay BG Tully WF PP Urrbrae Figure 27 Effects of soil type and trash management practices (B burnt, G GCT, WF wheat/fallow, PP permanent pasture) on the average value of kG (irrespective of TCC and SAR treatments) Figure 27 shows the effect of soil type and sugarcane trash management practices (organic matter content) on the average value of kG (mean for all levels of TCC and SAR). Analysis of variance showed that there was a highly significant effect (P < 0.001) of soil 69 type and trash management practices on values of kG (Table 33). There was also a significant interaction effect (P = 0.05) between soil type and trash management on the value of kG. Table 33 Significance of the treatments as indicated by analysis of variance Effect kG Trash management (C) ** SAR (N) * TCC (T) ns Soil (S) ** CxN ns CxT ns NxT ** CxS * NxS * TxS ns CxNxT * CxNxS ns CxTxS ns NxTxS ns CxNxTxS * * and ** indicate significant at P = 0.05 and P = 0.001, respectively ns = non significant Soil type significantly (P = 0.001) influenced the value of kG, probably due to clay content and type. The average values of kG decreased in the order Ayr (0.0084), Mackay (0.0076), Urrbrae (0.0070) and Tully (0.0064 (mmol L-1)-0.5). Trash management practices significantly (P = 0.001) influenced the value of kG. Figure 27 shows that the green treatment generally resulted in lower kG values than the burnt treatment, even though the organic matter contents resulting from these management practices (Table 8) were not significantly different in Mackay. SAR significantly (P = 0.05) affected the value of kG while TCC did not (Table 33). However, the interaction between SAR and TCC significantly influenced (P = 0.001) the value of kG (Figure 28). In general, the value of kG decreased as SAR increased. These trends were generally the same for all soils within each trash management practice. 70 kG (mmol L-1)-0.5 0.015 0.010 burnt Ayr soil green 0.005 0.000 0.012 0.008 Mackay soil burnt green kG (mmol L-1)-0.5 0.004 kG (mmol L-1)-0.5 0.000 0.012 0.009 0.006 0.003 0.000 0.012 0.008 burnt Tully soil green Urrbrae soil wheat fallow permanent pasture kG (mmol L-1)-0.5 0.004 Figure 28 0.000 T2S3 T10S3 T2S10 T10S10 T2S3 T10S3 T2S10 T10S10 Effects of total cation concentration (TCC) and sodium adsorption ratio (SAR) on the value of kG for different trash management practices in each soil. T2, T10 & S3, S10 denote TCC 2, 10 mmol(+)L-1 and SAR 3, 10, respectively 6.7 Conclusions Decomposition of sugarcane trash was unaffected by high levels of soil sodicity (ESP 1827) in a light- and a medium-textured soil, and was reduced by 18% in a heavier-textured soil. We attribute the latter effect to impaired soil aeration due to the combination of high sodicity, high clay content, high pH, and relatively low organic matter content. Decomposition was unaffected by milder sodicity (ESP 8-12). On the basis of these findings, it is unlikely that soil sodicity will be an important limit to trash decomposition in sugarcane fields where viable crops can be grown. 71 Molasses and gypsum both increased the structural stability of sodic soils used in this study, but the effects depended to some extent on clay content. The effects of gypsum were more pronounced in the Burdekin soil (higher clay content), particularly in relation to reducing clay dispersion. On the other hand, molasses alone increased macroaggregation in the both soils regardless of soil texture. On the basis of this laboratory work, it is, therefore, likely that the addition of molasses alone or in combination with gypsum may beneficial for overcoming problems with sodic soils, particularly those of higher clay content used for either sugarcane production or in regions where molasses is available relatively cheaply. Molasses and lime improved the structural stability and reduced the ESP of both sodic soils and both show promise as ameliorants either alone or in combination. In our experiment, the rate of 10 t/ha molasses did not lower the pH of the soil sufficiently to dissolve the lime substantially. Further experiments are needed to determine the best combination rates of molasses and lime to overcome problems associated with sodicity. The method used for measuring cation exchange selectivity, using minimal disturbance to soil aggregates, proved successful. Cation exchange selectivity was significantly influenced by trash management in all soils due to changes in the content and nature of soil organic matter. GCTB can significantly reduce the tendency of sugarcane-growing topsoils to become sodic or remain sodic compared to a burnt system. Cane trash, other organic ameliorants such as molasses, and soil organic matter all have an important role to play in the management of sodic soils. All forms of organic additions examined improved the structural properties of sodic soils, and reduced sodicity or the tendency of soil to become or remain sodic. There were no major interactions between organic matter and Ca-containing ameliorants; they all had positive effects irrespective of each other. 7.0 EFFECT OF CALCIUM-BASED AMENDMENTS ON TRASH DECOMPOSITION 7.1 Introduction Although the advantages of retaining sugarcane crop residues as a trash blanket have been well documented (Wood 1991; Thorburn et al. 1999), nutrient cycling in relation to GCTB is not fully understood. As part of a Cooperative Research Centre for Sustainable Sugar Production (CRC Sugar) subprogram, researchers have examined the return of nutrients to soil and the factors that affect the breakdown of organic matter associated with trash management systems (Thorburn et al. 2001; Robertson and Thorburn 2002). Despite the many advantages of nutrient cycling and release (Mitchell, and Larsen 2000; Robertson and Thorburn 2000; Klok and Kingston 2002) and soil ameliorative effects of trash retention (Moody 2002), advantages of enhanced decomposition, slow trash decomposition rates in some circumstances (Kingston et al. 1998), and reports of possibly acidification with trash retention (AD Noble, pers. comm.) have indicated that there is a need to further develop the trash retention system. 72 Although lime and gypsum are routinely applied to sugarcane soils to supply calcium (Calcino 1994; Kingston and Aitken 1996), and lime is used to ameliorate acidic conditions (Noble et al. 1997, Aitken et al. 1998), little is known about the effects of applying calcium-containing amendments to sugarcane trash, both in relation to trash decomposition and properties of the underlying soils. This section reports on the results of an investigation aimed at determining the effects of applying calcium-based amendments on trash decomposition. The investigation consisted of two phases. Phase one dealt with the effect of various amendments on the rate of trash decomposition and the subsequent ameliorative effect on the underlying soil. In phase two, we investigated the nutrient release and uptake associated with this interaction. 7.2 Effect of various calcium amendments on trash decomposition and soil amelioration 7.2.1 Method and materials This investigation consisted of four separate but interlinked experiments. Experiment 1 This uncropped pot experiment, conducted in the Queensland Department of Natural Resources and Mines (NRM) glasshouse at Indooroopilly, was used to determine the effect of applying lime to trash. Two acidic topsoils (hydrosol and kurosol) with differing chemical properties (Table 34) were used in a glasshouse experiment. Soil (50 mm deep), either unamended or amended with lime, was placed in trays (300 mm x 400 mm) and wet to field capacity. A 50-mmthick blanket of cane trash was added to each tray. Lime application rates of 0, 3 and 6 t/ha were applied on the surface of the trash or incorporated into the soil prior to spreading the trash. An additional treatment was established involving unamended soil, trash addition and urea (160 kg N/ha) spread on the surface of the trash. Field moisture capacity was maintained by evenly spraying water onto the surface of the trash in each tray to predetermined tray masses. Table 34 Initial chemical properties of soils used in Experiment 1 Soil Hydrosol Kurosol pH (water) 3.85 4.95 pH (CaCl2) 3.61 3.87 Org C (%) 1.01 1.05 Exch Ca Exch K ECEC cmol(+)/kg 2.02 0.31 12.5 0.31 0.10 3.10 Al Mn mg/kg 110.6 7.3 46.5 28.9 Trays were destructively sampled after 4 and 7 months. The undecomposed trash, the 010 mm soil/decomposed trash layer and the 10-50 mm soil layer were sampled separately. The extent of decomposition of trash was evaluated by measuring the dry mass of 73 remaining trash. The chemical characteristics of the residual trash and both soil layers were determined, as was the microbial biomass in the 0-10 mm soil layer (at 4 months). Experiments 2 and 3 Two fully randomised pot trials were established, one at the NRM glasshouse at Indooroopilly (subtropical conditions) and the other in the glasshouse at CSR Sugar Macknade Mill at Ingham (in the moist tropics). The experiment conducted at Indooroopilly included five different soil types: a hydrosol from Coolum, a kurosol from Yandina, a humic podosol from Nambour, a dermosol from Ingham, and a vertosol from Ayr. The chemical and physical properties of these soils are presented in Table 35. The experiment at Macknade included a rudosol (river levee soil) and a tenosol (flood plain alluvial soil), both from the Herbert River sugarcane area, as well as the kurosol and dermosol soils used in the Indooroopilly experiment for cross referencing purposes (Table 35). Table 35 Initial chemical properties of soils used in Experiments 2 and 3 Soil Local soil identifier Hydrosol Kurosol Podosol Vertosol Kandosol Garrett McCord Cook Barratta Red loam Kurosol Kandosol Tenosol Dermosol McCord Red loam River Bank River Overflow pH water 4.5 4.7 5.2 7.0 5.5 4.7 5.5 5.0 5.3 Org C K Ca (%) Experiment 2 0.63 6.34 1.0 0.18 0.36 3.3 0.21 1.41 1.4 0.26 11.78 1.3 0.49 1.83 Experiment 3 1.0 0.18 0.36 1.3 0.49 1.83 1.0 0.18 1.45 1.6 0.32 6.98 Mg Na Al ECEC (cmol(+)/kg) 3.59 1.52 4.34 16.42 0.38 0.06 2.31 3.29 1.16 0.21 1.54 4.63 6.51 0.61 - 19.16 0.59 0.03 0.38 3.32 0.38 0.06 2.31 3.29 0.59 0.03 0.38 3.32 0.94 0.06 1.44 4.07 2.54 0.08 0.07 9.99 In both trials, soil (80 mm deep) was placed in square containers (160 mm x 160 mm) and partitioned into different depth intervals (0-10, 10-20, 20-50 and 50-88 mm below the soil surface) using plastic coarse-meshed grids. With the exception of the treatments without trash, effectively 40 t ha-1 of air-dried trash was applied to the soil surface in each container. Various amendments were then spread on the surface of the trash (Table 36). As the experiment conducted at Indooroopilly was aimed at simulating weather conditions in the subtropics, temperatures in the glasshouse ranged from a minimum of 10°C to a maximum of 30°C. The pots were watered regularly at average rates of 300 mL twice weekly. In the Macknade trial, moist tropical conditions were simulated. Glasshouse temperatures ranged from a minimum of 15°C to a maximum of 40°C and the pots were watered regularly at average rates of 500 mL twice weekly. The pots were destructively sampled after 7 months by retrieving the residual trash in each pot. The residual trash was determined by the combining the obviously undecomposed trash layer with the organic material remaining after the soil/trash interlayer was washed with deionised water to eliminate any soil particles and/or inorganic residues. This 74 residual trash was oven dried at 70°C and weighed. Soil pH (water) values were determined for each soil sample after they had been air-dried and ground to pass through a 2-mm screen. Table 36 List of amendments included in the Indooroopilly (Experiment 2) and Macknade (Experiment 3) pot trials Treatment Amendment and equivalent rate* 1 Lime (calcium carbonate) at 3.0 t ha-1) 2 Gypsum (calcium sulphate) at 4.10 t ha-1 3 Calcium hydroxide at 2.22 t ha-1 4 Lime (1.5t ha-1) + gypsum (2.05 t ha-1) 5 Mill mud at 30t dry mud ha-1) 6 Potassium hydroxide 1.7 t ha-1) 7 Magnesium oxide at 1.2t ha-1 8 Urea at 160 kg N ha-1 9 Lime at 3.0 t ha-1, No trash 10 No amendment, Trash only (40 t ha-1) 11** No trash, no amendment * Rates of amendments, other than that of millmud, were calculated to give Ca supply capacities and/or acid neutralising capabilities similar to 3 t/ha of calcium carbonate. * Indooroopilly only Experiment 4 This field trial was established at Macknade to evaluate the interaction between trash and lime/millmud under field conditions. Details are given in Table 37. Composite soil samples were collected from beneath the GCTB at CSR Technical Field Department: Block B14 after harvest of a first-ratoon crop, prior to the layout of the trial. The amount of trash originally present was determined by collecting trash from random areas of 1.5 m2 between adjacent rows (row spacing of 1.5 m) along the full length of the block. At the same time, quadrats were established in each plot using coarse plastic mesh (covering a length of 3 m between rows) to facilitate measurement of trash decomposition. Treatments (Table 37) were applied on the surface of the trash. Immediately prior to harvest of the next crop, residual trash samples were collected from each quadrat. Composite soil samples were also collected from beneath each quadrat (0-5 cm and 5-10 cm depths). Organic C (0-50 mm soil depth), soil pH (0-50 mm and 50-100 mm soil depth) and exchangable Al) were used to determined to assess the ameliorative effect of the various amendments applied to the surface of the trash blanket. 75 Table 37 Details of the field trial conducted at Macknade (Experiment 4) Experiment 4 Macknade Design details: Randomised block Site: CSR Technical Field Department: Block 14 Soil type: River Bank (Tenosol) Variety: Q164 After 1st Ratoon Crop: Date established: June 2000 Date harvested: July 2001 Mean soil values Treatment applied (pre-treatment) Topsoil (0 – 200 mm) Lime (2.5 tonnes ha-1) 5.73 Lime (5 tonnes ha-1) pH(water) Org C (%) 0.7 Millmud (200 tonnes ha-1) ex Victoria Mill BSES P (mg kg-1) 123 Millmud/millash (200 tonnes ha-1) ex Macknade Mill Exch. K (cmol(+) kg-1) 0.09 Control (unamended trash) Exch. Ca (cmol(+) kg-1) 7.55 Exch. Mg (cmol(+) kg-1) 1.49 Exch. Na (cmol(+) kg-1) 0.15 Extr. Al (cmol(+) kg-1) - Sulphate S (mg kg-1) 5 Clay (%) 12 ECEC (cmol(+) kg-1) 9.28 Subsoil pH (water) Exch. K (cmol(+) kg-1) Exch. Ca (cmol(+) kg-1) Exch. Mg (cmol(+) kg-1) Exch. Na (cmol(+) kg-1) Extr. Al (cmol(+) kg-1) ECEC (cmol(+) kg-1) 6.72 0.06 8.53 2.27 0.18 11.04 7.2.2 Results and discussion Experiment 1 Samples taken at 4 months after establishment indicated that lime had a significant effect on trash decomposition, but the extent of decomposition was dependent on the method of lime application and soil type (Table 38), although no interactive effects were apparent (Figure 29). Spreading lime on the surface of the trash layer enhanced decomposition. This was attributed to possible increased solubility of organic matter and improved microbial activity with increased pH within the decomposing trash blanket. Although there was no significant treatment effect on microbial biomass with increased pH, it was highest in the surface soil (0-10 mm) where lime (6 t/ha) was spread onto the trash (data not shown here). Although urea significantly increased trash decomposition across soils, it was slightly less effective than lime (3 t/ha). 76 Table 38 Amounts of residual trash after 4 months in Experiment 1 Soil Hydrosol Kurosol Significance Lsd(p<0.05) Lsd(p<0.01) Residual trash (t/ha) 20.45 19.61 ** 0.23 0.31 Lime applied (t/ha) 0 3 6 Residual trash (t/ha) 20.57 19.83 19.70 ** 0.28 0.38 Application Spread on trash Incorporated in soil Residual trash (t/ha) 20.36 19.70 ** 0.23 0.31 21.5 21.0 Spread on trash Incorporated in soil 20.5 Residual dry trash (tonn 20.0 19.5 19.0 18.5 Lime: 0 t/ha 3 t/ha Hydrosol 6 t/ha Lime: 0 t/ha 3 t/ha Kurosol 6 t/ha Figure 29 Residual trash as influenced by lime rate and method of application. Interactive effect (lime x method of application) was not statistically significant The 7-month sampling showed that the differences in residual trash had markedly narrowed for both soils and no significant differences existed between treatments (Figure 30). The overall slow decomposition rate and loss in treatment differences have been attributed to the low volume of water applied. This resulted directly from the fact that the pots used had sealed bottoms and it was necessary is restrict water application to maintain the soil moisture content at field capacity. As expected, incorporating lime into the soil increased soil pH in both the 0-10 mm (Figure 31) and 10-50 mm layers (Figure 32). Spreading lime onto the trash layer also resulted in marked increases in soil pH values in the surface soil (0-10 mm). Although there was evidence of small pH increases in the 10-50 mm soil layer from the application 77 of lime onto the surface of the trash after 7 months, these differences were not significant. However, the substantially lower exchangeable Al values associated with the 10-50 mm soil layer at 4 and 7 months (Figure 33) was evidence that some amelioration had occurred with the application of lime to the surface of the trash. These data offer encouraging evidence that the interaction between lime application and trash may have important implications for the amelioration of sub-surface soil acidity in minimum tillage systems. Figure 30 Trash (tonnes/ha) Trash (tonnes/ha) 26 25 24 23 22 21 20 19 18 17 0 24 6 Sampling time (months) 8 Unlimed Lime (3t/ha) spread on trash Lime (3t/ha) incorporated in soil Lime (6t/ha) spread on trash Lime (6t/ha) incorporated in soil Hydrosol 26 25 24 23 22 21 20 19 18 17 0 24 6 Sampling time (months) Kurosol 8 Unlimed Lime (3t/ha) spread on trash Lime (3t/ha) incorporated in soil Lime (6t/ha) spread on trash Lime (6t/ha) incorporated in soil Trash decomposition as influenced by time and lime application 78 pH(w) 7.5 7 6.5 6 5.5 5 4.5 4 3.5 0 2 Hydr4osol 6 8 Sampling times (months) Hydrosol Unlimed Lime (3t/ha) spread on trash Lime (3t/ha) incorportaed in soil Lime (6t/ha) spread on trash Lime (6t/ha) incorportaed in soil pH(w ) 9 8.5 8 7.5 7 6.5 6 5.5 5 4.5 4 0 Kurosol 2 46 Sampling time (months) Kurosol 8 Figure 31 pH of the 0-10mm soil depth (immediately below the trash layer) as influenced by time and lime application pH(w) 6.5 6 5.5 5 4.5 4 3.5 02 468 Sampling times (months) Unlimed Lime (3t/ha) spread on trash Lime (3t/ha) incorportaed in soil Lime (6t/ha) spread on trash Lime (6t/ha) incorportaed in soil pH(w) 9 8.5 8 7.5 7 6.5 6 5.5 5 4.5 4 02468 Sampling times (months) Figure 32 Hydrosol Kurosol pH values of the 10-50 mm soil depth as influenced by time and lime application 79 Al (mg/kg) Al (mg/kg) 110 90 70 50 30 10 -10 0 2 4 6 8 Sampling times (months) Hydrosol Unlimed Lime (3t/ha) spread on trash Lime (3t/ha) incorportaed in soil Lime (6t/ha) spread on trash Lime (6t/ha) incorportaed in soil 60 50 40 30 20 10 0 -10 0 2 4 6 8 Sampling times (months) Kurosol Figure 33 Exchangeable Al values of the 10-50 mm soil depth as influenced by lime application and time Experiments 2 and 3 Trash decomposition in the Indooroopilly trial (Experiment 2) was affected by both soil type (Figure 34) and amendment applied (Figure 35). Residual trash (mass of remaining trash expressed as a percentage of the original mass of trash and amendment applied) indicated that although decomposition was similar over the 7-month period on the kurosol, podosol and kandosol, the amount of residual trash on the hydrosol was significantly lower than that on the vertosol. This difference could be related to the origin and pH of the two soils. The vertosol is from an area of relatively low rainfall (Ayr) and has a pH value of 7 (highest in the range of soils considered – Table 35). The hydrosol is a strongly acid soil from Coolum. It is found in an area characterised by moderately high rainfall and subject to periodic water-logged conditions. It is possible that anaerobic micro-organisms would naturally be associated with soils in such bottom-land positions. As the pots received about 600 mL of water each week, the interface between the trash and soil would have been moist for much of the 7-month period. Although the application of lime and a combination of lime and gypsum resulted in residual trash levels significantly lower (P = 0.05) than that of unamended trash, mill mud (applied at a rate of 30 t of dry material) resulted in residual trash levels markedly lower (about 30% units) than any of the other treatments (Figure 35). The application of KOH also resulted in residual trash levels significantly lower than the unamended trash and the majority of the other treatments. This may have been the result of alkaline hydrolysis of the organic matter. In Experiment 3 at Macknade, the application of millmud again resulted in the most marked decomposition of the trash layer (Figure 36). In this case, however, application of the other amendments (including lime and the lime/gypsum mixture) did not give rise to significantly lower levels of residual trash. Trash decomposition was not affected by soil type (Figure 37). 80 55 Residual trash (% of original trash and amendment added) LSD(0.05) = 2.63 54 53 52 51 50 49 48 47 46 45 Figure 34 Hyrdosol Kurosol Podosol Vertosol Kandosol SOIL Residual trash (expressed as a percentage of the mass of original trash and amendment applied) associated with each of the soil types after a 7-month period in Experiment 2 Residual trash (% of origina trash and amendment added) 60 LSD(0.05) = 3.53 LSD(0.01) = 4.72 55 50 45 40 35 30 25 20 15 Lime (3t/ha) Ca(OH)2(2.2t/ha) Millmud(30t dry /ha) MgO (1.2 t/ha) Trash only Gypsum (4.1t/ha) *Lime and gypsum KOH (1.69 t/ha) Urea (160 kg/ha) Amendment applied to trash Figure 35 Residual trash (expressed as a percentage of the mass of original trash and amendment applied) as influenced by the type of amendment applied to the trash blanket in Experiment 2 81 60 Residual trash (% of the original trash and amendment added) LSD (0.05) = 5.81 LSD (0.01) = 7.79 55 50 45 40 35 30 25 20 15 Lime (3 t/ha) Ca(OH)2(2.2t/ha) Millmud (30t dry/ha) MgO (1,2 t/ha) Trash only Gypsum (4.1 t/ha) *Lime and gypsum KOH (1.7 t/ha) Urea (160 kg/ha) Amendment applied to trash Figure 36 Residual trash (expressed as a percentage of the mass of original trash and amendment applied) as influenced by the type of amendment applied to the trash blanket in Experiment 3 50 Residual trash (% of original trash and amendment applied) 48 46 44 42 40 38 36 34 32 Figure 37 30 Kurosol Kandosol Tenosol Dermosol Soil Residual trash (expressed as a percentage of the mass of original trash and amendment applied) associated with each of the soil types after a 7-month period in Experiment 3. Differences were not statistically significant When the residual trash data relating to the soils common to the two experiments (kurosol and kandosol) were pooled, it was found that similar trends to the applied amendments 82 were apparent (Table 39). Although the amount (dry mass) of residual trash on the kandosol was significantly lower than that on the kurosol (probably due to better aeration), the most marked difference in residual trash was due to the different environments, with the mean amount of residual trash in the moist tropical environment significantly lower (P < 0.01) than the mean mount of trash remaining in the simulated subtropical environment. When considering the interaction between the simulated environment and amendment applied (Figure 38), we found that millmud was the only amendment that did not result in a significant difference in residual trash between the subtropical and moist tropical conditions. This means that millmud is effective in enhancing trash decomposition across the two simulated environments. This result is also well illustrated when relative residual trash (the mass of residual trash associated with each amendment expressed as a percentage of the mass of unamended trash remaining after the 7-month period) was plotted against the type of amendment applied in the subtropical (Figure 39) and moist tropical (Figure 40) environments. In the simulated subtropical conditions (Figure 39) the application of lime, KOH and millmud resulted in significantly lower relative residual trash values. In contrast, only the millmud application resulted in a significantly lower relative residual trash value in the simulated moist tropical environment (Figure 40). It would appear that the more severe conditions (larger amounts of water applied and/or higher temperatures) in the tropical environment ultimately negated any enhanced decomposition effect due to lime (and KOH) application as observed in the subtropical environment. This supports the results of Experiment 1, which showed that the significant differences in residual trash values due to lime application (at 4 months) was eliminated after 7 months. Table 39 Trash (expressed as a percentage of the original trash and amendment applied) remaining above the two common soils (Kurosol and Kandosol) included in the simulated subtropical and moist tropical environments after 7 months Simulated environment Subtropics Moist tropics Residual trash (%) 47.65 33.86 Soil Kurosol Kandosol Residual trash (%) 41.99 39.52 Significance ** * LSD(p=0.05) 2.27 2.27 LSD(p=0.01) 3.05 Interaction Simulated environment x amendment Other interactions Amendment added to trash Lime (3 t/ha) Gypsum (4.1 t/ha) Ca(OH)2 (2.2 t/ha) Lime (1.5 t/ha) + gypsum (2.05 t/ha) Millmud (30 dry t/ha) KOH (1.7 t/ha) MgO (1.2 t/ha) Urea (160 kg/ha) Trash only Significance * NS Residual trash (%) 40.19 43.38 45.53 43.13 20.95 37.61 45.10 43.81 47.10 ** 0.23 0.31 83 Residual trash (% of original trash and amendment added) LSD(0.05) = 6.82 65 60 Subtropical Indooroopilly 55 Moist tropics Macknade 50 45 40 35 30 25 20 15 10 Lime (3 t/ha) Ca(OH)2(2.2 t/ha Millmud (30t dry/ha) MgO (1.2 t/ha) Trash only Gypsum (4.1 t/ha) *Lime and gypsum KOH (1.7 t/ha) Urea (160 kg/ha) Amendment added to trash Figure 38 Residual trash as influenced by environment (subtropical versus moist tropical) and amendment applied 110 LSD(0.05) = 7.5 LSD(0.01) = 9.9 100 90 Relative residual trash (%) 80 70 60 50 40 30 Lime (3 t/ha) Ca(OH)2(2.2t/ha) Millmud (30dry t/ha) MgO (1.7 t/ha) Gypsum (4.1 t/ha) *Lime + gypsum KOH (1.7 t/ha) Urea (160 kg/ha) Amendment added to trash Figure 39 Relative residual trash values (expressed as a percentage of the unamended trash remaining after the 7-month period) in the simulated subtropical environment 84 110 Relative residual trash (%) LSD(0.05) = 18.2 LSD(0.01) = 24.3 100 90 80 70 60 50 40 Lime (3 t/ha) Ca(OH)2(2.2t/h) Millmud (30dry t/ha) MgO (1.2 t/ha) Gypsum (4.1 t/ha) *Lime + gypsum KOH (1.7 t/ha) Urea (160 kg/ha) Amendment added to trash Figure 40 Relative residual trash values (expressed as a percentage of the unamended trash remaining after the 7-month period) in the simulated moist tropical environment In assessing the ameliorative effect of applying an amendment to the trash blanket, only selected treatments were considered: lime on trash, millmud on trash, surface applied lime (no trash), trash (no amendment), and no trash or amendment (Experiment 3 only). In Experiment 2 (subtropical environment) soil type, treatment and depth below the trash layer had a significant interactive effect on soil pH (Figure 41). As expected, the surface applied lime increased soil pH across the five soil types. This was most pronounced in the surface layers and generally declined with depth. The presence of a trash blanket (in the absence of any amendments) had variable effects on soil pH. While surface applied trash appeared to increase the pH of the kurosol, humic podosol and kandosol, the effect was less marked in the hydrosol possibly due to a higher buffering capacity. In contrast the presence of trash resulted in a decrease in soil pH in the vertosol. The addition of lime to the surface of the trash layer also resulted in variable effects. In the hydrosol, podosol and kandosol, the effect on soil pH was either similar or less than that of surface applied lime (no trash). However, in kurosol, lime application to the trash blanket resulted in higher pH values in the sub-surface layers than when lime was applied in the absence of trash, thus supporting earlier findings. In experiment 3 (moist tropical environment), pH trends were similar to those the occurred in Experiment 2 in the acid soils, particulary in the kurosol and kandosol, which were common to both experiments (Figure 42). The general increase in soil pH with trash retention was less marked in the dermosol than in the other soils. Once again, this is thought to be due to a higher buffering capacity. Lime application to trash on the kurosol again resulted in higher pH values in the subsurface layers than when lime was applied in 85 the absence of trash. This has important implications for certain acid soils where amelioration of sub-surface layers can be enhanced by this interaction of lime and trash, possibly due to more soluble organo-calcium complexes. 8.5 8.0 7.5 7.0 pH (water) 6.5 6.0 Lime on trash 5.5 Milmud on trash Surface applied lime (no trash) Trash (no amendment) 5.0 No trash and no amendment 4.5 Initial soil pH: 4.5 Initial soil pH: 4.7 Initial soil pH: 5.2 Initial soil pH: 7.0 Initial soil pH: 5.5 4.0 Depth 10-20 mm 50-80 mm Depth 10-20 mm 50-80 mm Depth 10-20 mm 50-80 mm Depth 10-20 mm 50-80 mm Depth 10-20 mm 50-80 mm 0-10 mm 20-50 mm 0-10 mm 20-50 mm 0-10 mm 20-50 mm 0-10 mm 20-50 mm 0-10 mm 20-50 mm Hydrosol Kurosol Podosol Vertosol Kandosol Figure 41 Interactive effect of soil type, treatment and depth on soil pH in simulated subtropical conditions. LSD(0.01) = 0.46 8.5 8.0 7.5 7.0 Soil pH (water) 6.5 6.0 5.5 5.0 Lime on trash Millmud on trash Surface applied lime (no trash) 4.5 Trash (no amendment) Initial soil pH: 4.7 Initial soil pH: 5.5 Initial soil pH: 5.0 Initial soil pH: 5.3 4.0 Depth 10 - 20 mm 50 - 80 mm Depth 10 - 20 mm 50 - 80 mm Depth 10 - 20 mm 50 - 80 mm Depth 10 - 20 mm 50 - 80 mm 0-10 mm 20 - 50 mm 0-10 mm 20 - 50 mm 0-10 mm 20 - 50 mm 0-10 mm 20 - 50 mm Kurosol Kandosol Tenosol Dermosol Figure 42 Interactive effect of soil type, treatment and depth on soil pH in simulated tropical conditions. LSD(0.05) = 0.77; LSD(0.01) = 0.84 86 Experiment 4 In the field trial at Macknade, all but the lime (applied at 2.5 t/ha) resulted in significantly less trash than the control (trash with no amendment added) after 12 months (Figure 43). Millmud (200 t wet material /ha (equivalent of 60 dry t/ha)), millmud/ash combination (200 t wet matter/ha (equivalent of 60 dry t/ha) and lime (5 t/ha) were equally effective in their ability to decompose trash. 18 16 Residual trash (% of initial trash) LSD(0.05) = 6.0 14 12 10 8 6 Lime (2.5 t/ha) Millmud (60t dry/ha) Trash only Lime (5 t/ha) *Mud/ash(60 dry t/ha Amendment applied to trash Figure 43 Residual trash (% of initial trash present) associated with each of the amendments applied to the surface of the trash blanket in the field trail at Macknade (Experiment 4) A highly significant (P < 0.01) increase in soil organic C occurred in the soil layer 0-50 mm below the soil-trash interface when millmud and the millmud/mill ash mixture were applied (Figure 44). These increases were presumably, to a large extent, the result of the substantial amounts of organic matter added, rather than due to direct inputs from the trash, despite enhanced decomposition (Figure 43). Although lime applied at 5 t/ha resulted in a significantly lower amount of residual trash, no increase in soil organic C was noted (Figure 44). Although the initial pH value and exch. Ca value of the soil indicated that amelioration of acidic conditions was not required, soil pH values were affected by the application of the amendments to the trash blanket, with lime elevating the mean soil pH to 6.23 and 6.58 (over the full depth interval of 0-100 mm) when lime was applied at rates of 2.5 and 5 t/ha, respectively (Table 40). Although the millmud/mill ash mixture resulted in the largest increased in soil pH to 6.89, no shift in soil pH occurred with the millmud 87 application. The absence of change in soil pH was also apparent when the interactive effect of amendment applied to the trash blanket and depth below the trash-soil interface was considered (Figure 45). This is probably due to the increased buffer capacity of the soil due to increases in organic C (Figure 17). This soil inherently has a relatively low buffer capacity with a clay content of 12% and org C = 0.7% (Table 37). In general millmud/mill ash mixtures are generally lower in organic C than plain millmud and have added abilities to increase soil pH values due to the nature of the ash material. The absence of a change in soil pH when millmud was applied to trash has important implications for certain soils, particularly where enhanced trash decomposition is desirable without further elevation of soil pH above 5.5. 1.6 1.5 Org C (%) LSD(0.05) = 0.21 LSD(0.01) = 0.30 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 0.5 Figure 44 Lime (2.5 t/ha) Millmud (60t dry/ha) Trash only Lime (5 t/ha) *Mud/ash (60t dry/ha Amendment applied to trash Organic C in the 0-50 mm soil depth immediately below the trash layer in the field trial at Macknade Table 40 Soil pH values as affected by amendments applied to the surface of trash. Two soil depth classes were considered (0-50 mm and 50-100 mm below soil trash interface) Soil depth (mm) 0-50 50-100 Soil pH(water) 6.16 5.80 Significance LSD (< 0.05) LSD(<0.01) Significance ** 0.11 0.15 ** Amendment added to trash Soil pH(water) Lime (2.5 t/ha) 6.23 Lime (5 t/ha) 6.58 Millmud (60 dry t/ha) 5.60 Millmud/mill ash (60 dry t/ha) 6.89 Trash only 5.59 ** 0.18 0.24 Interaction LSD(< 0.05) = 0.26, LSD(< 0.01 = 0.34 88 7.4 7.2 7.0 6.8 6.6 0-50 mm soil depth 50-100 mm soil depth 6.4 Soil pH(water) LSD(0.05) = 0.26 LSD(0.01) = 0.34 6.2 6.0 5.8 5.6 5.4 5.2 Lime (2.5 t/ha) Millmud (60dry t/ha) Trash only Lime (5 t/ha) *Mud/ash(60 dry t/ha Amendment applied to trash Figure 45 pH of the soil below the trash blanket as influenced by the application of various amendments Despite already low exchangeable Al values in the soil prior to the application of various amendments to the trash (Table 37), lime treatments reduced the values to zero (Table 41), with millmud and millmud/mill ash mixture delivering an intermediate result. Table 41 Exchangeable Al values of soil as affected by amendments applied to the surface of trash. Two soil depth classes were considered (0-5 mm and 5-10 mm below soil trash interface) Soil depth (mm) 0-5 5-10 Exch. Al (cmol(+)/kg) 0.035 0.024 Significance LSD (< 0.05) * 0.011 Significance NS Amendment added to trash Lime (2.5 t/ha) Lime (5 t/ha) Millmud (60 dry t/ha) Millmud/mill ash (60 dry t/ha) Trash only Significance LSD (< 0.05) LSD (< 0.01) Interaction Exch. Al (cmol(+)/kg) 0.000 0.000 0.044 0.033 0.071 * 0.011 0.023 89 7.3 Combined effect of calcium-based amendments and trash retention on nutrient availability This section of the investigation was made up of two separate but linked pot experiments. Both were aimed at determining whether the application of calcium-based amendments onto trash would deliver benefits in terms of nutrient availability to sugarcane. One trial (Experiment 5) was conducted at Macknade and included soils found in the Herbert River district. The other trial was conducted in Bundaberg and included three local (Bundaberg) soil and one soil from the Herbert region to enable cross-referencing. Table 42 Selected chemical properties and physical attributes associated with the soils used in the pot experiments at Macknade (Experiment 5) and Bundaberg (Experiment 6) Exp No Soil type pH (water) 5 River sand 5.1 Red loam 5.6 River Overflow 5.3 Fine Black Sand 5.5 6 Woongarra 6.3 River sand 5.2 Oakwood 5.8 Kepnock 5.4 Accepted critical value* * Calcino, 1994; Wood et al. 2003 Org C (%) 0.4 0.9 1.6 2.6 1.1 0.4 1.2 0.8 BSES P (mg/kg) 137 38 102 61 65 176 10 39 40 BSES Si (mg/kg) 420 140 1200 510 540 390 250 115 70 Nitric K (cmol(+)/kg) 1.46 1.79 3.25 1.08 0.73 1.68 0.66 0.42 Cu Zn (mg/kg) 0.8 2.1 0.2 0.6 1.2 3.0 0.5 0.8 4.7 8.8 0.7 2.2 0.6 1.0 1.0 1.5 0.2 0.6 Exp No Soil type Exchangeable actions K Ca Mg Na Al (cmol(+)/kg) 5 River sand 0.13 1.27 0.52 0.07 0.48 Red loam 0.19 3.04 0.44 0.02 0.13 River Overflow 0.32 6.98 2.54 0.08 0.07 Fine Black Sand 0.10 3.27 0.39 0.09 0.23 6 Woongarra 0.25 4.21 3.32 0.18 - River sand 0.13 1.29 0.51 0.08 0.50 Oakwood 0.38 3.56 2.40 0.09 0.07 Kepnock 0.30 3.86 1.07 0.16 0.07 Accepted critical value* 0.24 1.2 0.25 * Calcino, 1994; Wood et al., 2003 ECEC Textural class 2.47 Loamy sand 3.82 Sandy clay loam 7.69 Silty clay loam 4.08 Loamy sand 8.69 Clay loam 2.51 Loamy sand 6.50 Clay loam 5.46 Loam 7.3.1 Methods and materials Experiments 5 and 6 Sufficient quantities of soils ranging widely in chemical and physical properties (Table 42) were collected from the Herbert and Bundaberg districts, for two fully randomised pot experiments (Table 43). These soils (mixed with basal nutrient applications – Table 43) were placed in 0.25 m diameter freely draining plastic pots. Three previously germinated single-eyed setts of sugarcane (cultivars Q158 at Macknade, and Q141 and Q155 at 90 Bundaberg) were planted in each pot. Air-dried trash, cut to 50 mm lengths, was applied to each pot (equivalent of 15 t/ha). The amendments (Table 43) were spread on the surface of the trash layer in each pot. The chemical analyses of the millmud and mill ash are shown in Table 44. The pots were placed on exterior seedling benches and watered regularly with automatic overhead water sprinkling systems. The sugarcane plants were fully harvested at 6 months of age. The harvested plants were weighed and composite samples formed by combining the plant material from replicates 1 and 3, and 2 and 4. These samples were shredded, dried and re-weighed. The dried biomass was sent to a laboratory for chemical analysis. Table 43 Details of pot experiments conducted at Macknade and Bundaberg Pot experiment Type Soils Variety Amendment applied to trash Basal nutrient applications Replicates Planting date Harvest date Age of cane at harvest 56 Soil x amendment Soil x amendment Fully randomised Fully randomised River Sand (Rudosol) Woongarra (Ferrosol) Red Loam (Kandosol) River Sand (Rudosol) ex Macknade River Overflow (Dermosol) Oakwood (Kandosol) Fine Black Sand (Hydrosol) Kepnock (Yellow Dermosol) Q158 Q155 Trash only Millmud (80 wet t/ha) Millash (80 wet t/ha) Lime (5 t/ha) 100 kg N/ha 50 kg K/ha Trace element mix 4 (64 pots) 4 (64 pots) October 2000 Feb 2001 December 2000 June 2001 6 months 6 months Table 44 Chemical analysis of amendments (dried) used in the pot experiments at Macknade and Bundaberg Exp No Amendment Org C N P K Ca Mg S (%) 5 Millmud 29.8 2.34 1.03 0.34 1.68 0.40 0.21 Mill ash 31.1 0.61 0.38 0.84 0.72 0.40 0.22 7.3.2 Results and discussion The mean dry-matter yield values associated with each of the soil types included in Experiment 5 (Figure 46) reflected the perceived potential of the soils to produce sugarcane crops. While the dermosol (River Overflow) is considered one of the most productive and fertile soils found in the Herbert River district (Wood et al. 2003), the other three are thought of as being moderately productive. 91 30 28 26 Dry matter yield (tonnes/ha) LSD(0.05) = 3.17 LSD(0.01) = 4.37 24 22 20 18 16 14 Rudosol1 Kandosol2 Soil type Dermosol3 Hydrosol4 Figure 46 Mean dry matter yield values associated with each of the soil types included in Experiment 5 24 23 Dr matter yield (tonnes/ha) LSD(0.05) = 3.17 LSD(0.01) = 4.37 22 21 20 19 18 17 16 15 Figure 47 Trash only Millmud (80wet t/ha) Mill ash (80wet t/ha Lime (5 t/ha) Amendment added to trash Mean dry matter yield values associated with each of the amendments applied to the trash in Experiment 5 The mean dry-matter yield values associated with the amendments applied to the trash in Experiment 5 (Figure 47) indicated that the largest response occurred due to the application of millmud (80 wet t/ha), but that the application of mill ash (80 wet t/ha) and 92 lime (5 t/ha) also resulted in significantly higher yields than the trash-only treatment. Although, both the mill mud and mill ash could have directly and indirectly improved the moisture status of the soil (compared with the trash only treatments), this would not have been the case when lime was added. In terms of the soil test values (Table 42) responses to applied nutrients other than N and K were unlikely (Wood et al. 2003), despite the large amounts of nutrients present in the millmud and mill ash. The responses to the amendments applied to the trash layer were therefore attributed, at least in part, to the enhanced decomposition of the trash and presumably the release of N from increased organic carbon. This effect occurred across all four soil types within the experiment, with the interactive effects (soil x amendments applied) being not statistically significant. The amount of nutrients (N, P and K) removed by the harvested biomass expressed as kg/ha (Figure 48) were used to assess the nutrient availability resulting from the interaction between trash and the amendments applied. When considering the individual soils, the dermosol, as expected, supplied the largest amount of nutrients (257 kg N/ha, 36 kg P/ha and 237 kg K/ha). The rudosol supplied the smallest mean amounts of N and K, 122 kg N/ha and 124 kg K/ha, respectively, and the hydrosol supplied the least amount of P (17 kg/ha). In terms of amendments applied to the trash, the millmud and mill ash treatments resulted in significantly higher amounts of N, P and K removed by the harvested biomass (Figure 48). These elevated values could either have been the result of the nutrients within the amendments themselves, or due to enhanced decomposition of the trash, or a combination of both of these effects. Although the mean amounts of N and K removed by the harvested biomass when lime was applied to the trash were larger than that associated with the trash only treatment, these differences were not statistically significant. Phosphorus was the only nutrient to be influenced by a significant interactive effect between soil type and amendments applied (Figure 49). In the rudosol, kandosol and dermosol, significantly more P was removed by the harvested biomass when millmud and mill ash were applied to the trash than with unamended trash. In contrast, P removal on the hydrosol remained almost constant across all the trash/ amendment treatments. Possible reasons for this include the strongly P-sorbing character of this soil (Wood et al. 2003). 93 K removed by harvested biomass (kg K/ha) LSD(0.05) = 21.09 LSD(0.01) = 29.06 P removed by harvested biomass (kg P/ha) LSD(0.05) = 3.48 LSD(0.01) = 4.79 N removed by harvested biomass (kg N/ha) LSD(0.05) = 31.37 LSD(0.01) = 43.22 280 260 240 220 200 180 160 140 120 100 Rudosol1 40 35 30 25 20 15 Rudosol1 260 240 220 200 180 160 140 120 100 Rudosol1 Figure 48 210 200 N removed by harvested biomass (kg N/ha) LSD(0.05) = 31.37 LSD(0.01) = 43.22 190 180 170 160 150 140 Kandosol2 Soil type Dermosol3 Hydrosol4 130 Trash only 34 Millmud (80wet t/ha) Mill ash (80wet t/ha Amendment applied to trash Lime (5 t/ha) 32 P removed by harvested biomass (kg P/ha) LSD(0.05) = 3.48 LSD(0.01) = 4.79 30 28 26 24 22 20 18 Kandosol2 Soil type Dermosol3 Hydrosol4 16 Trash only 200 Millmud (80wet t/ha) Mill ash (80wet t/ha Amendment applied to trash Lime (5 t/ha) 190 K removed by harvested biomass (kg K/ha) LSD(0.05) = 21.09 LSD(0.01) = 29.06 180 170 160 150 140 130 120 Kandosol2 Soil Type Dermosol3 Hydrosol4 110 Trash only Millmud (80wet t/ha) Mill ash (80wet t/ha Amendment applied to trash Nutrients (N, P and K) removed plotted against soil amendments applied to the trash in Experiment 5 Lime (5 t/ha) type and Figure 49 P removed by harvested biomass LSD(0.05) = 6.96 LSD(0.01) = 9.59 55 50 45 40 35 30 25 20 15 10 5 Trash only Rudosol (River Sand) Kandosol (Red Loam) Dermosol (River Overflow) Hydrosol (Fine Black Sand) Millmud (80wet t/ha) Mill ash (80wet t/ha Amendment applied to trash Lime (5 t/ha) Interactive effect of soil type and amendments applied to the trash on P removed by the harvested biomass in Experiment 5 94 In Experiment 6, the highest mean dry matter yield was attained on the ferrosol (Figure 50). This was expected, as it is naturally a high yield potential soil. The yields obtained on the other two local Bundaberg soils (kandosol (Oakwood) and dermosol (Kepnock)), were in line with the general yield potential ranking. Dry matter yield on the rudosol was almost identical to that obtained on this soil under tropical conditions (Experiment 5). Figure 50 20 19 18 17 Dry matter yield (tonnes/ha) 16 15 LSD(0.05) = 1.96 LSD(0.01) = 2.70 14 13 12 11 10 9 Ferrosol1 Rudosol2 Soil type Kandosol3 Dermosol4 Mean dry matter yield values associated with each of the soil types included in Experiment 6 30 25 Dry matter yield (tonnes/ha) LSD(0.05) = 1.96 LSD(0.01) = 2.70 20 15 10 Figure 51 5 Trash only Millmud (80wet t/ha) Mill ash (80wet t/ha Lime (5 t/ha) Amendment applied to trash Mean dry matter yield values associated with the amendments applied to the trash in Experiment 6 The mean dry matter yield values associated with the amendments applied to the trash in Experiment 6 (Figure 51) indicated that, as in the case of the Macknade experiment, the largest response was due to the application of millmud (80 wet t/ha). Although the application of mill ash (80 wet t/ha) also resulted in a significant increase (P < 0.05) in dry matter production compared to the unamended trash treatment, lime (5 t/ha) applied to the trash blanket did not result in significantly different yields. This was surprising, as lime 95 applied to trash had increased decomposition of trash layer in all previous experiments and had resulted in increased yields in Experiment 5. Figure 52 Dry matter yield (tonnes/ha) LSD(0.05) = 3.93 LSD(0.01) = 5.41 40 35 30 25 20 15 10 5 0 Trash only Ferrosol (Woongarra) Rudosol (River Sand ex Macknade) Kandosol (Oakwood) Dermosol (Kepnock) Millmud (80wet t/ha) Mill ash (80wet t/ha Amendment applied to trash Lime (5 t/ha) Interactive effect of soil type and amendment applied to the trash on dry matter yield in Experiment 6 Table 45 Dry matter yield data relating to environment (subtropical versus moist tropical) and amendment applied to trash on a rudosol (common soil in Experiments 5 and 6) Environment Moist tropical Subtropical Dry matter yield (t/ha) 16.48 15.81 Significance NS Significance NS Amendment added to trash Trash only Millmud (80 wet t/ha) Mill ash (80 wet t/ha) Lime (5 t/ha) Significance LSD (< 0.05) Interaction Dry matter yield (t/ha) 12.83 20.80 15.35 15.63 * 3.98 Unlike the Macknade experiment (Experiment 5), soil type and amendment applied to the trash resulted in a significant interactive effect (P < 0.01) on dry matter yield (Figure 52). In particular, millmud applied to the trash layer on the rudosol (River Sand) and the dermosol (Kepnock) resulted in significantly lower dry matter yields than those achieved on the ferrosol (Woongarra) and kandosol (Oakwood). However, the yields (rudosol and dermosol) were not substantially different to the mean dry matter yield value (22.5 t/ha) associated with the millmud application to trash in Experiment 5. Comparisons of yields obtained in the two experiments was possible due to the inclusion of the rudosol (River 96 Sand ex Macknade) in both trials (Table 45). Yields associated with the River Sand were unaffected by the environment (moist tropical climate at Macknade versus subtropical conditions at Bundaberg), but affected markedly by millmud applications. No interactive effects were apparent. Nitrogen removal by the harvested biomass, in terms of soil type and amendment applied to the trash (Figure 53), and the interaction between these two factors (Figure 54) reflected the trends that occurred in dry matter yields (Figures 50-52). These general trends also occurred in relation to K removal by the harvested biomass (Figures 57 and 58). Although P removal in relation to the amendments applied (Figure 55) also reflected the general dry matter yields (Figure 51), the P removal in the millmud treatment differed according to soil type. That P removed by the harvested biomass from the ferrosol was lower than that removed from the rudosol (Figure 56) is probably due to differences in Psorbing characteristics. N removed by harvested biomass (kg N/ha) LSD(0.05) = 19.14 LSD(0.05) = 27.37 N removed by harvested biomass (tonnes/ha) LSD(0.05) = 19.14 LSD(0.01) = 26.37 190 180 170 160 150 140 130 120 110 100 90 Ferrosol1 Rudosol2 Soil type Kandosol3 Dermosol4 300 250 200 150 100 50 G_1:1 G_2:2 G_3:3 Amendment applied to trash G_4:4 Figure 53 N removed by the harvested biomass plotted against soil type and amendments applied to the trash in Experiment 6 Figure 54 N removed by harvested biomass (kg N/ha) LSD(0.05) = 38.28 LSD(0.01) = 52.74 400 350 300 250 200 150 100 50 0 G_1:1 Ferrosol (Woongarra) Rudosol (River Sand ex Macknade) Kandosol (Oakwood) Dermosol (Kepnock) G_2:2 G_3:3 Amendment applied to trash G_4:4 Interactive effect of soil type and amendments applied to the trash on N removed by the harvested biomass in Experiment 6 97 P removed by harvested biomass (kg P/ha) LSD(0.05) = 2.16 LSD(0.01) = 2.98 P removed by the harvested biomass (kg P/ha) LSD(0.05) = 2.16 LSD(0.01) = 2.98 29 28 27 26 25 24 23 22 21 20 19 18 Ferrosol1 Figure 55 Rudosol2 Soil type Kandosol3 Dermosol4 55 50 45 40 35 30 25 20 15 10 5 Trash only Millmud (80wet t/ha) Mill ash (80wet t/ha Amendment applied to trash Lime (5 t/ha) P removed by the harvested biomass plotted against soil type and amendments applied to the trash in Experiment 6 Figure 56 P removed by the harvested biomass (Kg P/ha) LSD(0.05) = 4.32 LSD(0.01) = 5.95 70 60 50 40 30 20 10 0 Trash only Ferrosol (Woongarra) Rudosol (River Sand ex Macknade) Kandosol (Oakwood) Dermosol (Kepnock) Millmud (80wet t/ha) Mill ash (80wet t/ha Amendment applied to trash Lime (5 t/ha) Interactive effect of soil type and amendments applied to the trash on P removed by the harvested biomass in Experiment 6 K removed by harvested biomass (kg K/ha) LSD(0.05) = 14.11 LSD(0.01) = 19.44 K removed by harvested biomass (kg K/ha) LSD(0.05) = 14.11 LSD(0.01) = 19.44 190 180 170 160 150 140 130 120 110 100 Ferrosol1 Rudodsol2 Soil type Kandosol3 Dermosol4 240 220 200 180 160 140 120 100 80 Trash only Millmud (80wet t/ha) Mill ash (80wet t/ha Amendment applied to trash Lime (5 t/ha) Figure 57 K removed by the harvested biomass plotted against soil type and amendments applied to the trash in Experiment 6 98 Figure 58 K removed by harvested biomass (kg K/ha) LSD(0.05) = 28.21 LSD(0.01) = 38.87 300 Ferrosol (Woongarra) Rudosol (River Sand ex macknade) 250 Kandosol (Oakwood) Dermosol (Kepnock) 200 150 100 50 0 Trash only Millmud (80 wet t/ha Mill ash (80 wet t/h Amendment applied to trash Lime (5 t/ha) Interactive effect of soil type and amendments applied to the trash on K removed by the harvested biomass in Experiment 6 The differences noted between the two trials are best explained by the differences in the amount of water that moved through the pots. As both experiments were conducted on external seedling benches, the Macknade experiment would have benefited from both the irrigation and the relatively high rainfall in the moist tropical conditions (596 mm of rainfall was recorded at Macknade over the duration of the experiment). At Bundaberg, only 203 mm of rainfall was recorded during the trial period and, hence, much of the water was from the automatic irrigation system. 7.4 Conclusions Calcium-based amendments such as lime and mill mud have the potential to accelerate the decomposition of sugarcane trash. Importantly this decomposition appears to be most effective when the amendments, particularly lime, millmud and mill ash, are applied to the surface of the trash layer. Lime applied beneath the trash and incorporated in the soil did not enhance the decomposition process. When comparing the effectiveness of the amendments in assisting decomposition, millmud repeatedly outperformed the other amendments. Whereas other amendments such as Ca(OH)2 and KOH did have the ability to enhance the decomposition process to some extent, they are not considered suitable for widespread use in the industry due to their corrosive nature. Amendments such as gypsum and MgO did not contribute to increased trash decomposition. Importantly, the enhanced trash decomposition that occurred due to lime millmud, mill ash and lime applied to trash in pots under semi-controlled conditions was also found to be applicable in the field. The differences in the rates of decomposition in the simulated subtropical and moist tropical conditions suggest that environmental conditions involving temperature and the amount of water (rain and/or irrigation) are important factors within this trash/ameliorant system. Lime applied to the surface of the trash layer did have the ability to ameliorate acid soil conditions. In some soils lime applied to the trash blanket resulted in higher pH values in 99 the subsurface soil layers than when lime was applied in the absence of trash. This apparent synergistic effect has important implications for certain soils, especially when minimum tillage is practiced. Importantly, millmud applications to the surface of the trash layer also resulted in some amelioration of the subsurface layers. Although millmud and mill ash applications to trash resulted in significantly higher amounts of N, P and K removed by the harvested biomass, these elevated values could either have been the result of the nutrients within the amendments themselves, or due to enhanced decomposition of the trash, or a combination of both of these effects. The effect on nutrient uptake by sugarcane when lime was applied to trash was less apparent. Environmental effects (water, temperature and soil properties) are considered important factors in relation to the overall nutrient availability in these amended trash systems 8.0 TRASH DECOMPOSITION AND C AND N RELATIONSHIP 8.1 Introduction In GCTB sugarcane systems, as in any cropping system involving retention of harvest residues, it is important to know not only the inputs of organic matter and nutrients in the residues, but how these residues interact with the soil-plant environment. Although this has been dealt with to some extent in Section 7, a more targeted approach within a ‘micro’ investigation system under more controlled conditions was deemed useful. Knowledge of the decomposition characteristics of the residues is fundamental to understanding their effects in the field. It is also pertinent to know what factors affect trash decomposition, and how trash decomposition affects other soil properties and processes. Apart from the results of experiments reported earlier in this report, these subjects that have received little research attention, and much remains to be understood. This section describes two experiments undertaken to address some of these uncertainties and determine whether small-scale pot experiments (using pots with < 20% of the surface area of those used in the experiments reported in Section 7) can be used to investigate some of these. The first experiment was done to gain basic information on decomposition of sugarcane trash in controlled conditions, and to help choose experimental conditions for future decomposition studies. The second experiment was done to measure the effects of mill mud, mill ash, urea, and sugarcane trash on soil C and N mineralisation using this type of experimental methodology. 8.2 Measuring trash decomposition in controlled conditions Measurements of crop residue decomposition made under field conditions can sometimes be difficult to interpret due to variations in climate, soil, crop, and management factors. As such, experimentation under standardised (laboratory or glasshouse) conditions is often a practicable alternative way to measure decomposition and determine controlling factors. 100 Ideally, the conditions under which crop residues decompose in the field should be simulated in controlled environment studies. This may, however, be difficult to achieve, because experimentation on a small scale with limited resources requires the researcher to make compromises. It is usually necessary to dry the residue and to reduce the size of residue fragments (by chopping or grinding) to facilitate subsampling and handling. In many decomposition experiments, factors such as the size to which the residue is cut, whether the residue is fresh or dried, the placement of the residue in or on the soil, and the soil water content, vary from the field situation. The extent to which these factors affect decomposition of sugarcane trash is unknown, but studies with residues of other crops suggest that all these factors may be important (Parr and Papendick 1978; Christensen 1986; Bremer et al. 1991; Reddy et al. 1994; Andrén et al. 1995; Sørensen et al. 1996; Devêvre and Horwáth 2000). This study was done to gain basic information on decomposition of sugarcane trash in controlled conditions using small-scale investigative units, and to help choose experimental conditions for future decomposition studies. The objective of the work was to determine how decomposition of sugarcane trash is affected by: (1) whether the trash is green or senesced; (2) whether the trash is fresh or oven-dried; (3) whether the trash is chopped or ground; (4) whether the trash is on the soil surface or incorporated into the soil; and (5) whether the soil is wet or dry. 8.2.1 Methods Experimental Soil for the experiment was collected from the 0-10 cm depth of a sugarcane field at the Central Sugar Experiment Station, Mackay, Queensland (21.10 S, 149.07 E). The soil had contents of total C, total N, and inorganic N of 1.08%, 0.05%, and 7 µg/g, respectively. Particle size distribution was 56, 27, 17% (sand, silt, clay). The soil was dried, crushed and sieved (< 2 mm) before use. Trash was collected from mature sugarcane (cultivar Q124) growing in the field at the same location. Within 2 h of collection, the trash was separated into green leaves and senesced leaves, and was either oven-dried (65°C) or refrigerated (6°C) for 28 h. The trash was then either chopped with scissors (< 40 mm long) or passed through a cutter grinder (20-60 mm long x 3-6 mm wide). Soil and trash were sampled for analyses. Sufficient plastic pots (100 mm diameter, sealed at the base) were prepared containing 250 g dry soil. Trash (equivalent to 4 g oven-dry) was either mixed with the soil or placed on the surface. Masses of soil and trash used in this experiment were about 4% of that used in the investigations reported in Section 7. Pots that contained unamended soil were also included. The soils were made up to the either 40% or 80% of saturation (i.e. 13% or 26% water content) by addition of deionised water. Table 46 shows the experimental treatments. The pots were placed inside heavy-duty zip-seal plastic bags (left open). Wire frames were placed in the pots, so that vials could be inserted for collection of CO2. The pots were placed in the dark at a constant temperature of 25°C. Water was added weekly to maintain the original soil water contents. 101 Table 46 Experimental treatments Factor Trash fraction Trash water content Trash fragment size Trash placement Soil water content Level 1 Green Fresh Chopped Surface Wet (80-90% saturation) Level 2 Senesced Dried Ground Incorporated Dry (40-50% saturation) Carbon dioxide was collected from the pots on 16 occasions over the next 8 months, using alkali traps (vials containing 20 mL of 1M sodium hydroxide were placed in the frames, and the plastic bags sealed for 2-3 d). At the end of the experiment, the contents of the pot were divided vertically, one half for retrieval of remaining trash, the other half for chemical analyses. The experiment was designed as a (½ x 25) fractional factorial (Box and Hunter 1961) with five replicates, i.e. not all combinations of factors and levels were present. This design enabled the main effects and 2-way interactions to be measured, while limiting the samples to a manageable number. Analytical methods Soil inorganic N (ammonium + nitrate) was determined by extracting fresh soil in 2M potassium chloride, followed by automated colorimetric analysis of the extracts (Rayment and Higginson 1992, method 7C2). Total N and C were determined in dried soil (< 2 mm) and trash (< 1 mm) using a Leco combustion analyser. Carbon dioxide in the sodium hydroxide traps was measured by titration against dilute hydrochloric acid (Zibilske 1994). Soil pH and electrical conductivity were measured in 1:5 soil-water extracts (Rayment and Higginson 1992, method 3A1). Trash soluble C was extracted by shaking 1 g dry trash with 40 mL of deionised water (at room temperature) for 1 h, then filtering the extract. The total C concentration of the extract was determined by ICP. Residual trash was retrieved from the soil by washing the sample in approximately 3 L water containing 2 mL of Decon detergent then rinsing on a 2 mm sieve under running water. The trash was then dried at 65°C. Calculations and statistical analysis Cumulative CO2-C evolved during the experiment (∆C) was estimated from the area under the curve of CO2 accumulation against time. Mineralisation of C due to trash (∆CT) was calculated as ∆C from trash-amended soil, less ∆C from unamended soil. Loss of trash-C was calculated as ∆CT expressed as a percentage of trash-C applied. Apparent net mineralisation or immobilisation of N (∆N) was calculated as soil inorganic N at the end, less soil inorganic N at the start of the experiment. Apparent net mineralisation or immobilisation of N due to trash (∆NT) was calculated as ∆N from trash-amended soil, less ∆N from unamended soil. Apparent net mineralisation or immobilisation of N per unit of C mineralised was calculated as ∆N / ∆C for unamended soil (∆NC), and ∆NT / ∆CT for trash-amended soil (∆NCT). 102 The significance of treatment effects (main effects and 2-way interactions) was measured by Analysis of Variance (ANOVA) for the appropriate fractional factorial design. Differences between some individual treatment combinations were also assessed by simple one-way ANOVA. Least significant differences (LSD) were calculated when ANOVA indicated significant effects (P < 0.05). 8.2.2 Results Trash quality Total C did not differ significantly among trash fractions or trash treatments (mean 36%). Green leaves contained 1.3% N and had a C:N ratio of 29; senesced leaves contained 0.4% N and had a C:N ratio of 93. This was not affected by trash drying or cutting treatments (Table 47). Soluble C in the trash was 3-10%, and was greater in green than in senesced leaves. The grinding treatment increased soluble C in both green and senesced leaves (Table 47). Trash dry matter and C loss Evolution of CO2 from unamended soil was small compared with that from trash-amended soils. Cumulative CO2 (∆C) at the end of the experiment averaged 175 mg C/pot in the unamended soils, and 700-1200 mg C/pot in the trash-amended soils. In unamended soils, there was a consistent, but usually non-significant, trend for ∆C to be slightly greater in wet soil than in dry soil (Figure 59). Table 47 Trash C and N at the start of the experiment (presented as % of dry matter) Trash Green, fresh, chopped Green, fresh, ground Green, dried, chopped Green, dried, ground Senesced, fresh, chopped Senesced, fresh, ground Senesced, dried, chopped Senesced, dried, ground LSD (P < 0.05) Total N (%) 1.32 1.25 1.34 1.26 0.44 0.36 0.45 0.39 0.11 Total C:N ratio 27 27 30 31 87 88 91 91 6 Soluble C (%) 9.3 10.3 8.9 10.0 3.6 4.6 3.1 4.0 0.8 103 ∆C (mg C / pot) 200 100 0 0 100 200 Dry soil Wet soil Figure 59 Cumulative CO2 evolution from unamended soils (∆C). Bars are LSD (P = 0.05) for unamended soil only Trash decomposition measured as CO2 evolution (∆CT), was initially faster in green leaves than in senesced leaves (until day 20), but this trend was reversed from day 94 onwards, and by the end of the experiment (day 244) ∆CT was not significantly different between trash fractions (Figure 60). 1000 ∆CT (mg C / pot) 500 0 0 50 100 150 200 250 Green trash Senesced trash Figure 60 Day Cumulative CO2 evolution from green and senesced trash (∆CT). Bars are LSD (P = 0.05) for main effect Trash water content had no effect on ∆CT (data not shown). Trash placement and trash particle size also had some early transient effects on ∆CT. During the first 20-38 d, senesced leaves decomposed more rapidly when incorporated 104 with the soil than when placed on the surface; and green leaves decomposed more rapidly on the surface than when incorporated (data not shown). Ground trash decomposed faster than chopped trash only when placed on the surface and only for the first 13 d (data not shown). Trash particle size did not affect final ∆CT. At most sampling times and at the end of the experiment, ∆CT showed a highly significant interaction between trash placement and soil water. In wet soil, ∆CT was greater with trash on the surface; in dry soil, ∆CT was greater with trash incorporated (Figure 61). 1000 ∆CT (mg C / pot) 500 0 0 50 100 150 200 250 Days Surface, Wet soil Surface, Dry soil Incorporated, Wet soil Incorporated, Dry soil Figure 61 Interacting effects of trash placement and soil water content on cumulative CO2 evolution from trash (∆CT). Bars are LSD (P = 0.05) for interaction At the beginning of the experiment only, ∆C per day was positively correlated with both the original C:N ratio and soluble C content of the trash (r = 0.8, 0.6, 0.6 for days 6, 13, 20, respectively, P < 0.01, n = 16). Trash decomposition measured as dry matter (DM) loss was more variable among replicates than ∆C, but generally showed similar trends (Figure 62). Trash fraction and trash water content had no effect on DM loss. There was a significant interaction between trash placement and soil water content. In wet soil DM loss was greater with surfaceapplied trash, and in dry soil DM loss was greater with incorporated trash. DM loss tended to be greater when trash was ground than when it was chopped, but this is not considered to be a true treatment effect, but rather a reflection of the relative ease with which trash could be retrieved from the chopped treatments compared to the ground treatments. Percentage decomposition of trash during the experiment was 32-89% (mean 60%) when measured as DM loss, and 35-66% (mean 50%) when measured as C loss from CO2 data (Figure 62). Correlation between DM loss and C loss gave r = 0.51 (P < 0.001, n = 80). 105 Trash C loss (%) Trash DM loss (%) (e) Surface, Chopped 80 60 40 20 0 (a) Surface, Chopped 80 60 40 20 0 (f) Surface, Ground (g) Incorporated, Chopped (h) Incorporated, Ground (b) Surface, Ground (c) Incorporated, Chopped (d) Incorporated, Ground Green, Dry soil Green, Wet soil Senesced, Dry soil Senesced, Wet soil Figure 62 Dry matter (DM) and C loss from trash at the end of the experiment. Bars are LSD (P = 0.05) from one-way ANOVA N mineralisation and immobilisation Apparent net N mineralisation in unamended soil (∆N) averaged 16 mg N/pot. Apparent net mineralisation (positive value) or immobilisation (negative value) of N due to trash (∆NT) ranged from -11 to 9 mg N/pot. It was significantly affected by trash fraction (green, mean 5.3 mg N/pot > senesced, mean –9.1 mg N/pot) (Figure 63a-c). Generally, ∆NT was greater for surface than for incorporated trash (except for chopped trash in dry soil). Neither trash water content (data not shown) nor trash particle size, had a consistent effect on ∆NT. There was a trend for ∆NT to be lower in wet soil than in dry soil (Figure 63 a-c). Apparent net N mineralisation expressed on the basis of C mineralisation (∆NC) was 12.6 mg N/g C for unamended soil. The corresponding values for trash (∆NCT) were 8 and -17 mg N/g C for green trash and senesced trash, respectively. Again, ∆NCT was generally greater for surface than for incorporated trash, except for chopped trash in dry soil. Trash water content (data not shown) and trash particle size, had no consistent effect on ∆NCT (Figure 63 d-f). With senesced trash that was incorporated, or ground and surfaceapplied, ∆NCT was lower in wet soil than in dry soil. With green trash, ∆NCT tended to be lower in wet soil than in dry soil when the trash was incorporated, but soil-water content had no effect when the trash was surface-applied (Figure 63 d-f). Correlation analysis showed no relationship between ∆NT and ∆NCT (with data separated by trash fraction, or with pooled data). 106 ∆NCT (mg N/ g C) 20 10 0 -10 -20 -30 10 5 0 -5 -10 (e) Surface, Chopped (a) Surface, Chopped (f) Surface, Ground (g) Incorporated, Chopped (h) Incorporated, Ground (b) Surface, Ground (c) Incorporated, Chopped (d) Incorporated, Ground ∆NT (mg N / pot) Green, Dry soil Green, Wet soil Senesced, Dry soil Senesced, Wet soil Figure 63 Apparent net N mineralisation from trash (∆NT), and apparent net N mineralisation per unit of C mineralisation from trash (∆NCT), at the end of the experiment. Bars are LSD (P = 0.05) from one-way ANOVA 8.2.3 Discussion The aim of this work was to measure the effects of the experimental treatments on decomposition of trash. The conclusions to be drawn from the results depend on whether ‘decomposition’ is being measured as DM loss, C mineralisation, or apparent net N mineralisation / immobilisation. Measuring decomposition as DM loss may appear simple and inexpensive in comparison with C and N measurements, but it was somewhat unreliable in this small-scale experiment. Variability among replicates was large, because of the difficulty in retrieving trash from the soil at the end of the experiment, particularly when the trash had been ground. Division of the pot contents before trash retrieval would have contributed to the variability. Decomposition was probably overestimated because particles smaller than 2 mm were not retrieved. It could be expected that soils of higher clay content than that used in this experiment would exacerbate these problems. Notwithstanding these deficiencies, the DM measurements generally agreed with the CO2 measurements (except with ground trash). DM loss suggested that trash fraction and trash water content had no effect on decomposition. The effect of trash placement depended on soil water content; in wet soil DM loss was greater with surface-applied trash, and in dry soil DM loss was faster with incorporated trash. To improve the usefulness of DM loss as an indicator of decomposition in similar controlled environment studies, the trash should not be ground or shredded, the trash fragment size should be > 4 cm, and the pot contents should not be divided. 107 Measurements of decomposition as C mineralisation (CO2) were less variable within treatments than the DM loss measurements. Thus, they can be considered more reliable, and had the advantage of allowing decomposition to be measured non-destructively over time. Decomposition may have been underestimated, since the alkali absorption method used in this experiment has been shown to slightly underestimate CO2 evolution from the soil surface (Freijer and Bouten 1991). The relative differences among treatments, however, would not have been affected. The weakness of the correlation between CO2-C loss and DM loss can be attributed mainly to the experimental error associated with the latter. Decomposition as measured by ∆CT was temporarily influenced (for up to 5 weeks) by whether the trash was green or senesced, and whether the trash was chopped or ground, but neither trash fraction nor trash particle size affected the final decomposition at day 244. The early increase in ∆CT from green trash could be attributed to its greater soluble C content and smaller lignin content (Thorburn, unpubl. data). The temporary increase in ∆CT from ground trash was probably due to liberation of soluble C by grinding, and improved soil contact because of the smaller particle size. Similar trends have been reported by Bremer et al. (1991) for decomposition of straw and green material from lentil. Decomposition as measured by C mineralisation was not affected by whether the trash was fresh or oven-dried. Decomposition was affected by both trash placement and soil water content and, more importantly, there was a strong interaction between these factors; in wet soil trash decomposed more rapidly on the surface (because of improved aeration), and in dry soil trash decomposed more rapidly when incorporated (because of improved water availability and soil contact). So, in decomposition experiments involving CO2 measurements, soil water content and trash placement need to be carefully controlled. In short-term experiments of several weeks, the proportion of green and senesced material, and its particle size needs to be controlled, but in long-term experiments, variations in trash fraction and particle size would be less important. Ovendrying the trash is unlikely to affect CO2 measurements, and allows greater flexibility in sample handling. Decomposition as measured by apparent net N mineralisation or immobilisation was most strongly influenced by trash fraction, with net mineralisation predominating in the green trash and net immobilisation predominating in the senesced trash, in accordance with the initial C:N ratio of the trash. Apparent N mineralisation or immobilisation was not affected by whether the trash was fresh or oven-dried, or whether it was chopped or ground. The ratio of net N mineralised or immobilised per unit of C mineralised (∆NC) from a residue gives an indication of the balance between mineralisation and immobilisation of N, assuming no loss of N, and reflects the C and N composition of the original residue. Thus, ∆NCT would be expected to vary little within green trash and within senesced trash treatments. This was the case in dry soil, indicating that trash placement did not affect the balance between N mineralisation and immobilisation. In wet soil, however, ∆NCT was lower when trash was incorporated than when it was surface-applied, indicating either (a) increased immobilisation with incorporated trash, or (b) N loss with incorporated trash. Option (a) is unlikely as the decomposition substrate did not change. Option (b), on the other hand, is highly probable: incorporation of labile organic matter into wet soil can lower oxygen tension and stimulate denitrification (Alexander 1977). 108 Thus, in decomposition experiments where apparent net mineralisation / immobilisation of N is measured, it would be very important to control the proportion of green and senesced material, as well as and soil water content and trash placement. In particular, N mineralisation may be underestimated (or immobilisation overestimated) when trash is incorporated into wet soil. Oven-drying, grinding, or chopping the trash are not likely to affect N mineralisation / immobilisation. This study illustrates the usefulness of combining measurements of C and N mineralisation. The lack of correlation between the two show that they measure different aspects of decomposition, and the C measurements are a valuable aid to interpretation of the N measurements. The interacting effects of trash placement and soil water content shown by DM loss and C and N mineralisation have practical implications for trash management in the field. If the soil is predominantly fairly dry, incorporation of the trash is likely to slightly increase DM loss without affecting net N mineralisation/immobilisation; if the soil is wet for extended periods, incorporation of the trash is likely to slow DM loss, and increase N loss from the field through denitrification. 8.3 Trash decomposition and effects of trash, mill mud, mill ash and urea on soil C and N As indicated in Section 7, sugarmill by-products such as mill mud and mill ash have long been recognised as potentially valuable sources of soil nutrients. General application guidelines have been developed (Calcino 1994) based on nutritional considerations, but in practice mill mud and ash are applied to canefields at various rates, according to factors such as cost, distance from the sugar mill, and convenience. Field trials suggest that a typical application of 150 t/ha of mill mud or mill mud-ash mixture can provide around 100 kg /ha of plant-available N (around 20% of the total N content) in the year of application (Calcino 1994). How nutrient availability may be affected by the presence of cane trash, or other fertilisers such as urea, is not known. Nutrient release from mill mud or ash has not been measured directly. The objective of this work was to measure, under standardised small-scale conditions (similar to those described in Section 8.2), the individual and interacting effects of mill mud, mill ash, urea, and sugarcane trash on soil C and N mineralisation. 8.3.1 Methods Experimental Soil was collected from the 0-15 cm depth of sugarcane fields at three locations: Herbert, Burdekin, and Mackay (Table 48). The Burdekin and Mackay soils were also used in the experiment described in section 4.3, and the Mackay soil was also used the experiment described in section 4.2. The soil was dried in an oven at 45ºC for 6 d, then crushed to pass through a 2 mm screen, and sampled for analyses. Sugarcane (cultivar Q124) was grown in 40-cm-diameter pots in a glasshouse for 1 year, fertilised with weekly applications of a mixed fertiliser including 15N-labelled ammonium sulfate. At maturity, the green and senesced leaves, were collected. They were dried in 109 an oven at 65ºC for 2 d, shredded through a cutter-grinder (2-10 cm long and 0.2-1.0 cm wide), mixed thoroughly, and subsampled. The fragment size of the subsample was then reduced to < 4 cm by cutting with scissors. Table 48 Soils used in the experiment (0-15 cm depth) Parameter Location Texture Particle size (%sand, %silt, %clay) Total N (%) Total organic C (%) C:N ratio Inorganic N (µg/g) Saturation water content (%) Herbert Unknown Organic sand Not determined 0.14 2.76 20 26 26 Burdekin 147.23 E, 19.50 S Sandy clay loam 53, 25, 22 0.04 0.53 13 11 42 Mackay 149.07 E, 21.10 S sandy loam 56, 28, 16 0.05 0.95 19 8 40 Fresh mill mud and mill (bagasse) ash were obtained from Racecourse sugar mill, Mackay. It was dried in an oven at 45ºC for 6 d. Clumps of dry mill mud were crushed to < 10 mm for use in the experiment. The ash was free-flowing with particles < 5 mm. Samples of ash and mud (500 g) were ground to pass through a 2 mm screen before being subsampled for analyses (Table 49). Table 49 C and N in amendments used in the experiment. Results are % of dry matter Measurement Total C (%) Total N (%) Total C:N ratio Soluble C (%) Soluble N (%) 15N atom % Mill ash 8.8 0.06 142 0.38 0.000 - Mill mud 26.7 1.22 22 2.87 0.003 - Trash 39.9 0.68 60 1.94 0.012 7.04 Experimental units were prepared by mixing 300 g dry soil with the appropriate amendments and placing the mixtures in plastic pots (9-cm diameter) lined with plastic bags. Water was added to bring the mixtures to 70% of saturation. The equivalent application rates of the amendments were: trash 10 t DM/ha; mill mud 40 t DM/ha (equivalent to 160 t/ha wet); mill ash 40 t DM/ha; urea 180 kg N /ha (applied as 5 mL of 18.1 mg N /mL solution). The scale of the pots and mass of soils trash and amendment were about 5% of those used in the investigations reported in Section 7 of this report. The amendments were applied alone to the Herbert and Burdekin soils, and alone and in combinations to the Mackay soil, giving 20 treatments (Table 50). Four replicates of the treatments were prepared. The pots were placed inside heavy-duty zip-seal plastic bags 110 (left open). Wire frames were placed in the pots so that vials could be inserted for collection of CO2. The pots were placed in the dark at a constant temperature of 25°C. Water was added weekly to maintain the original soil water contents. The pots were arranged as randomised complete blocks, and positions within blocks were rerandomised on three occasions. Table 50 Experimental treatments Treatment Soil Amendments 1 Herbert Unamended soil 2 Herbert Ash 3 Herbert Mud 4 Herbert Trash 5 Burdekin Unamended soil 6 Burdekin Ash 7 Burdekin Mud 8 Burdekin Trash 9 Mackay Unamended soil 10 Mackay Ash 11 Mackay Mud 12 Mackay Trash 13 Mackay Urea 14 Mackay Ash + Urea 15 Mackay Mud + Urea 16 Mackay Trash + Ash 17 Mackay Trash + Mud 18 Mackay Trash + Urea 19 Mackay Trash + Ash + Urea 20 Mackay Trash + Mud + Urea Total C 1760 5336 1995 1760 5336 1995 1760 5336 1995 0 1760 5336 3755 7331 1995 3755 7331 Total applied (mg/pot) Total N Soluble C Soluble N -- - 12 76 0.00 244 573 0.68 34 97 0.59 --- 12 76 0.00 244 573 0.68 34 97 0.59 --- 12 76 0.00 244 573 0.68 34 97 0.59 90 0 90 102 76 90 334 573 90.68 46 173 0.59 277 670 1.26 124 97 90.59 136 173 90.59 367 670 91.26 Carbon dioxide was collected from the pots on 10 occasions over the next 6 months, using alkali traps (vials containing 20 mL of 1M sodium hydroxide were placed in the frames, and the plastic bags sealed for 2-3 d). At the end of the experiment, the contents of each pot were mixed and sampled for analyses. Analytical methods Soil inorganic N (ammonium + nitrate) was determined by extracting 15 g of fresh soil in 2M potassium chloride, followed by automated colorimetric analysis of the extracts ((Rayment and Higginson 1992, method 7C2). Total N and C were determined in dried soil (< 2 mm) and trash (< 1 mm) using a Leco combustion analyser. Carbon dioxide in the sodium hydroxide traps was measured by titration against dilute hydrochloric acid (Zibilske 1994). Soil pH and electrical conductivity were measured in 1:5 soil-water extracts (Rayment and Higginson 1992, method 3A1). Trash-soluble C was extracted by shaking 1 g dry trash with 40 mL of deionised water (25oC) for 1 h, then filtering the extract. The total C concentration of the extract was determined by ICP-OES. 111 Calculations and statistical analysis Cumulative CO2-C evolved during the experiment (∆C) was estimated from the area under the curve of CO2 accumulation against time. Mineralisation of C due to the amendment (∆CA) was calculated as ∆C from amended soil, less ∆C from unamended soil. Apparent net mineralisation or immobilisation of N (∆N) was calculated as soil inorganic N at the end, less soil inorganic N at the start of the experiment. Apparent net mineralisation or immobilisation of N due to the amendment (∆NA) was calculated as ∆N from amended soil, less ∆N from unamended soil. Apparent net mineralisation or immobilisation of N per unit of C mineralised was calculated as ∆N / ∆C for unamended soil (∆NC), and ∆NA / ∆CA for amended soil (∆NCA). The soil and single-amendment effects were tested by analysis of variance (ANOVA) for a two-way factorial, using treatments 1-12. Single and mixed-amendment effects in the Mackay soil were also tested by one-way ANOVA, using treatments 9-20. Least significant differences (LSD) were calculated when ANOVA indicated significant effects (P < 0.05). 8.3.2 Results C mineralisation In unamended soil, mineralisation of C (∆C) was similar in the Burdekin and Herbert soils, and greater than in the Mackay soil (Figure 64). 300 Unamended soil 250 ∆C (mg C / pot) 200 150 Figure 64 100 50 0 0 50 100 150 Days Mackay Burdekin Herbert 200 Cumulative CO2 evolution (∆C) in unamended soils. Bars are LSD (P = 0.05) for unamended soils only Mill ash had no effect on C mineralisation in any of the soils (∆C in ash-amended soils was not significantly different from unamended soils, Figure 65). Trash and mill mud produced large increases in ∆C, with mill mud mineralising around twice as much C as trash per pot (Figure 65). When C mineralisation was expressed per unit of amendment 112 dry matter applied ((∆CA), however, C mineralisation decreased in the order trash > mill mud > mill ash (Figure 66). Most of the difference in ∆CA between trash and mill mud was due to CO2 evolved during the first 88 days (data not shown). 3000 (a) Herbert 2000 (b) Burdekin (c) Mackay ∆CA (mg C / pot) 1000 0 0 50 100 150 0 50 100 150 0 50 100 150 Days Ash Mud Soil Trash Figure 65 Cumulative CO2 evolution from soils (∆C) and amendments(∆CA). Bars are LSD (P = 0.05) for soil type x amendment interaction. 250 ∆CA (mg C / pot / g DM applied) 200 150 100 Figure 66 50 0 0 50 100 150 Days 200 Ash Mud Trash Cumulative CO2 evolution from amendments (∆CA) per g dry matter. Bars are LSD (P = 0.05) for main effect) In the Mackay soil, urea applied alone had no effect on C mineralisation (Figure 67). Urea applied with trash caused a small transient increase in total ∆CA (∆CA from trash + urea was significantly greater than ∆CA from trash until day 24). Ash applied with trash slightly increased total ∆CA (to day 108), the effect of which was evident till the end of the experiment. Urea and ash applied together with trash had a similar effect on ∆CA as their combined effects when applied alone with trash (Figure 67). The greatest ∆CA was 113 from trash + mud, and trash + mud + urea (Figure 67). The mixtures containing mill mud gave a slightly smaller ∆CA at the end of the experiment than the total ∆CA of the same amendments alone (Figure 70a). The other amendments gave similar ∆CA values, whether mixed or alone. Throughout the experiment, ∆C (per day, or cumulative) was positively correlated with total C applied (r ≥ 0.82, P ≤ 0.001, n = 80, data not shown). Figure 67 ∆CA (mg C / pot) 3000 2000 1000 0 0 50 100 150 Days Ash Mud Urea Trash Ash + Urea Mud + Urea Trash + Ash Trash + Mud Trash + Urea Trash + Ash + Urea Trash + Mud + Urea Cumulative CO2 evolution (∆CA) in the Mackay soils. Bars are LSD (P = 0.05) from one-way ANOVA N mineralisation Apparent net mineralisation of N (∆N) in unamended soil was greatest in the Herbert soil (14 mg/pot), and similar in the Burdekin and Mackay soils (6 and 7 mg/pot, respectively (Figure 68 a-c). Mill ash had no effect on N mineralisation (∆N) in the Herbert and Mackay soils, and a very small positive effect in the Burdekin soil (Figure 68 a-c). Trash alone immobilised approximately the same amount of N as was mineralised by the unamended soils. Mill mud alone mineralised similar amounts of N in all soils (mean 8 mg/pot) (Figure 68 a-c). The net immobilising effect of trash, net mineralising effect of mill mud, and net neutral effect of mill ash were more evident when ∆NA was expressed per unit of amendment dry matter applied (Figure 68 d-f). 114 ∆N or ∆ NA (mg N / g DM / pot) 1 0 -1 -2 -3 10 5 0 -5 -10 -15 (d) Herbert (a) Herbert (e) Burdekin (b) Burdekin (f) Mackay (c) Mackay Soil Ash Mud Trash ∆N or ∆NA (mg N / pot) Figure 68 Apparent net N mineralisation / immobilisation from soils (∆N) and amendments (∆NA), per pot and per unit dry matter (DM) applied. Bars are LSD (P = 0.05) for soil type x amendment interaction Figure 69 ∆NA (mg N / pot) S A M T U AU MU TA TM TU TAU TMU 70 60 50 Soil Ash 40 Mud Urea 30 Trash Ash + Urea 20 Mud + Urea Trash + Ash 10 Trash + Mud Trash + Urea 0 Trash + Ash + Urea Trash + Mud + Urea -10 Amendment Apparent net N mineralisation / immobilisation from soil (∆N) and amendments (∆NA) in the Mackay soil. Bars are LSD (P = 0.05) for soil type x amendment interaction (short bar) and one-way ANOVA (long bar) 115 In the Mackay soil, urea applied alone produced the greatest ∆NA (74 mg N/pot). The ∆NA was not affected when ash was applied with urea, but was reduced when mill mud and/or trash were applied with urea (Figure 69). Ash applied with trash resulted in ∆NA similar to that from trash applied alone (Figure 69). Mill mud applied with trash tended to increase ∆NA compared with trash alone (but the effect was not significant, one-way ANOVA). The ∆NA for amendment mixtures was less than the total ∆NA for the same amendments alone, particularly for mixtures containing mill mud (Figure 70). The ∆NA was not correlated with total N applied, the overall C:N ratio of the amendments, or with ∆CA. (a) 3000 2000 1000 0 (b) 80 60 40 20 0 Mixed Alone ∆CA (mg C / pot) Ash + Urea Mud + Urea Trash + Ash Trash + Mud Trash + Urea Trash + Ash + Urea Trash + Mud + Urea Ash + Urea Mud + Urea Trash + Ash Trash + Mud Trash + Urea Trash + Ash + Urea Trash + Mud + Urea ∆NA (mg N / pot) Figure 70 Cumulative CO2 evolution (∆CA) and apparent net N mineralisation /immobilisation (∆NA) from amendments applied alone or in mixtures Apparent net N mineralisation or immobilisation expressed on the basis of C mineralisation (∆NC) was relatively high for unamended soils, with Herbert > Mackay > Burdekin (Figure 71). The corresponding measure for mill mud alone (∆NCA) was positive but very much smaller, and similar in the three soils. The ∆NCA for trash alone was negative, and lower in the Herbert than in the Burdekin and Mackay soils (Figure 71). The trash + mud mixture gave a ∆NCA close to zero. The trash + mud + urea mixture gave a considerably smaller ∆NCA than the trash + urea, and the trash + urea + ash mixtures (Figure 71). Total C and N In all soils, total soil N was significantly increased by mill mud, but not by ash or trash (Figure 72 a-c). Urea did not significantly affect total N in the Mackay soil (data not shown). Total soil C was increased by mill mud in all soils, and by ash in the Burdekin and Mackay soils (Figure 72 d-f). For both total N and C, the mixtures of amendments in the Mackay soil conformed to the trends shown by the individual amendments (data not shown). ∆NCA (mg N / g C) Soil (H) Soil (B) Soil (M) Mud (H) Mud (B) Mud (M) Trash (H) Trash (B) Trash (M) Trash + Ash Trash + Mud Trash + Urea Trash + Ash + Urea Trash + Mud + Urea Mud + Urea 116 80 60 40 20 0 -20 Figure 71 Amendment Apparent net N mineralisation or immobilisation per unit of C mineralisation (∆CNA) 3.0 (d) Herbert (e) Burdekin (f) Mackay Total soil C (%) 2.0 1.0 Total soil N (%) 0.0 0.20 (a) Herbert 0.15 0.10 0.05 0.00 0 (b) Burdekin (c) Mackay 172 0 Day 172 0 Ash Mud Soil Trash 172 Figure 72 Effects of amendments on total soil C and N. Bars are LSD (P = 0.05) from soil type x amendment interaction 117 Trash 15N Recovery of trash-15N at the end of the experiment averaged 87%. Recovery from trash alone was slightly less in the Herbert soil than in the other soils (Figure 73). In the Mackay soils, 15N recovery was generally not affected by the amendments, except for a small reduction where trash, mill mud and urea were applied together (Figure 73). 100 90 80 70 60 50 40 Trash 30 Trash + Ash 20 Trash + Mud Trash + Urea 10 Trash + Ash + Urea Trash + Mud + Urea 0 Recovery of Trash-15N (%) T(Herbert) T(Burdekin) T(Mackay) TA TM TU TAU TMU Figure 73 Amendment Recovery of 15N from labelled trash. Bars are LSD (P = 0.05) from one-way ANOVA) 8.3.3 Discussion The soil amendments investigated in the study - mill mud, mill ash, urea, and sugarcane trash - all had different effects on soil C and N mineralisation. Mill ash was essentially inert in terms of C mineralisation. This is not unexpected, considering that it had been burnt in the sugar mill furnace, where any labile organic matter would have been destroyed. For this reason too, ash had almost no effect on N mineralisation. In the Burdekin soil, ash incorporation greatly improved soil friability, and this may have resulted in the very small increase in ∆NA observed. Mill mud and trash both provided large amounts of decomposable C. On the basis of DM or total C content, mill mud contributed less C (113 mg /g DM, and 420 mg /g C applied) than trash (219 mg /g DM, and 550 mg /g C applied). However, the rates at which these amendments are usually applied in the field means that a typical application of mill mud (150 t/ha moist, 40 t DM/ha) will add about twice as much labile C to the soil as a typical trash blanket (10 t DM/ha). 118 Mill mud caused net mineralisation of soil N, whereas trash caused net N immobilisation. This is in keeping with the original C:N ratios of these amendments. Net mineralisation of N from mill mud was consistent in the three soils, and was approximately the same as net N mineralisation from the unamended Burdekin and Mackay soils, and rather less than that from the Herbert soil. Trash immobilised approximately the same amount of N as was mineralised from the unamended soils, immobilisation being greater in the Herbert than the other soils. The largest ∆NA was produced by urea (almost 10 times greater than the ∆NA from mill mud). Of the 90 mg N /pot applied, 74 mg N /pot was recovered in the urea-only treatments at the end of the experiment. This discrepancy may have resulted from microbial immobilisation, physical stabilisation, or denitrification of N. As expected, urea had no effect on C mineralisation as the latter was not limited by N availability. Mixtures of trash, ash, and urea showed C and N mineralisation trends that agreed with their effects when applied alone. With mixtures containing mill mud, however, ∆CA and ∆NA were reduced in comparison with mineralisation from individual amendments (Figure 70). The relatively reduced ∆CA in the mixtures may have been due to a reduction in the efficiency of C use by the microbial biomass, because of over-supply of labile C. The decrease in ∆NA in mixtures, however, is likely to be due to increased loss of N by denitrification in the presence of mill mud, also caused by the large supply of labile C. That ∆NCA was also smaller in mixtures containing mill mud than would be expected from the ∆NCA of individual amendments substantiates the suggestion of denitrification (Figure 71). Leaching was not a consideration in this experiment, as the pots were sealed at the base. Although the results presented in Section 7 indicated that there was enhanced decomposition of trash due to the surface application of millmud, results of the micro investigations indicate that ∆CA was smaller when mud and trash decomposed together than when they decomposed alone. As individual millmud only treatment was not included in the investigations reported in Section 7, a comparison of this kind was not possible. However, it is valid to conclude that both investigations show that decomposition associated with the combined trash/millmud treatment was better than trash alone. If millmud applied to the surface of trash causes better decomposition than trash and millmud individually, then a possible explanation is that the soil microbial population preferentially decomposed trash-C in the presence of mill mud. Despite the positive effects of application of millmud to trash, further work will be required to adequately understand the interactions. Differences in methodology between the experiments reported in this section and those described in Section 7 will need to be taken into account. Here, the mill mud was from a different source, was dried, was applied at a smaller rate, and was incorporated into the soil. Importantly the scale of investigation used here was much smaller than that used in Section 7 and did not link to a field investigation The ∆NA from urea alone minus the ∆NA from trash alone was considerably greater than the ∆NA from the trash + urea treatment (Figure 69). This suggests that: (i) N immobilisation was promoted in the presence of urea; (ii) denitrification was promoted in the presence of trash; or (iii) both processes were promoted simultaneously. 119 The ∆NA from mill mud alone represented 3.2% of the total N applied. This is an unexpectedly small proportion for almost 6 months of decomposition of a labile substance with such a low C:N ratio (22). It contrasts markedly with the approximately 20% of total N considered to be mineralisable in the first year following mill mud application to cane crops in the field (Calcino 1994). In the present experiment, there is a strong possibility that significant denitrification also occurred when mill mud was applied alone, and that ∆NA was therefore underestimated. It is possible, however, that the low ∆NA was partly due to the particular mill mud used in this experiment. The apparent denitrification losses in this experiment have important implications for future studies involving mill mud and / urea and trash. The risk of N loss could be controlled by including a shorter sampling period than was used here, or by periodic leaching the soils with water to avoid accumulation of inorganic N, or by incubating at a lower soil water content. The concomitant measurement of C mineralisation can provide a check on the reliability of apparent N mineralisation measurements. The potential for mineralisation and loss of N from mill mud clearly requires further study. Recovery of trash-15N at the end of the experiment (mean 87%) indicated that ash, mud and urea had little effect on N loss from trash. Any differences in microbiallyimmobilised 15N could not, however, be determined. The significant increase in total soil N caused by mill mud application suggests that mud will stimulate net N mineralisation into the longer term. 8.4 Conclusions Green trash initially caused more C mineralisation than senesced trash, but whether the trash was green or senesced made no difference to DM or C loss at the end of the experiment. Net N mineralisation occurred in green trash and net N immobilisation occurred in senesced trash. Oven-drying the fresh trash had no effect on DM loss, C or N mineralisation. Ground trash temporarily mineralised more C than chopped trash, but there was no difference in final DM loss, C or N mineralisation between ground and chopped trash. There was a strong interaction between trash placement and soil water content: in dry soil, DM loss and C mineralisation were greater when trash was incorporated with the soil; in wet soil, DM loss, C mineralisation, and apparent N mineralisation were greater when trash was surface-applied. There was evidence of N loss through denitrification when trash was incorporated into wet soil. In such controlled-environment decomposition experiments, it is most important to control the proportion of green and senesced trash, the placement of the trash, and the soil water content. Measurement of decomposition as DM loss is not recommended where the trash fragment size has been greatly reduced. Mill ash had a negligible effect on soil C and N mineralisation. Trash greatly increased C mineralisation and resulted in strong net N immobilisation. Mill mud greatly increased C mineralisation, and resulted in a small amount of net N mineralisation. Urea provided ten times more mineral N than mill mud, and did not affect C mineralisation. These trends were the same in three contrasting soil types. Mixtures containing combinations of ash, trash, and urea showed C and N mineralisation that agreed with mineralisation patterns 120 when these amendments were applied alone. Mixtures containing mill mud, however, apparently mineralised slightly less C and much less N than was mineralised from the individual amendments. Significant loss of N through denitrification was suspected in all treatments containing mill mud. 9.0 OUTPUTS This project has had a number of useful outputs. These include: • Data and knowledge relating to the effects of trash on soil acidity; • Information regarding the influence of Ca-containing amendments on trash decomposition; • Additional information on the management of sodic and acid soils; • Further knowledge regarding N transformations with trash retention; • Information about nutrient availability due to trash/Ca-containing amendment interactions; • Recommendations for growers to maximise the interaction between trash and Ca- containing amendments in acid and sodic soils. • Advice on liming strategies for trash retention systems. Following the previous order of reporting, important developments, results and conclusions are summarised below. 9.1 Effects of trash retention on soil acidity Here we determined: a. The net proton (H+) input resulting from long-term trash retention from existing field trials. b. Influence of changed management of sugarcane on chemical properties. c. Effect of water-soluble organic components leached from plant material. • Trash retention in the long-term trash retention trials resulted in increases in soil organic carbon at each of the sites, particularly in the surface soil layer. The extent of this increase appears to be a function of period under GCTB. • The increase in organic carbon, particularly at the Tully site (the longest under GCTB), has in turn resulted in a shift in the buffering capacity and an increase in proton retention. The difference in net proton input (H+) between the GCTB and burnt systems was calculated as 4.14 kmol H+/ha over an 8-year period. • Ash alkalinity has been used extensively to describe the acid-base balance in foods, grains and fodder crops and is a measure of the organic anion production that is produced internally by the plant in order to balance excess cation uptake. It is the decomposition of these anions of organic origin that influence the proton pool. Ash alkalinity of the trash material can be expressed as CaCO3 equivalents if it is assumed that 1 mol H+ requires 50 g CaCO3 to be neutralised. • The net amount of alkalinity in the form of CaCO3 equivalents associated with the trash ranges from 10.23 to 14.6 kg CaCO3/tonne dry matter. These rates of alkalinity being returned on an annual basis are small compared to the litters collected from 121 some plant species (i.e. Melia azedarach 126 kg CaCO3/tonne dry matter) and therefore do not contribute significantly to acid neutralisation associated with nitrogen mineralisation. • It is suggested that in the case of the burnt treatment, the increase in pH is associated with the release of this alkalinity. The potential generation of acid through the mineralisation of nitrogen contained in the trash is negligible since this source is lost on combustion. • Although the retention of trash resulted in an increase in the charge characteristics through the accretion of soil organic matter, the release of alkalinity from this process was not sufficient to neutralise the acidity generated by the mineralisation of nitrogen in the trash. • Organic carbon and charge characteristics of soil will increase with time as a consequence of trash retention. This positive consequence on soil chemical properties is important when assessing the long-term impact of changed management systems. The use of charge fingerprints is useful as an assessment of this. • Charge fingerprints were determined on samples collected from the 0-2 cm depth interval from the Tully site. The difference in charge fingerprinting between the burnt and trash retention system and convergence of the charge curves at approximately pH 5.5 is of significant agronomic importance as it clearly shows that by increasing the pH to this value one would eliminate any potential problems associated with acidic cations (ie Al3+). At pH 5.5 the amount of charge generated by the trash retained and burnt treatments was 4.6 and 3.3 cmolc/kg, respectively. This suggests that the positive impact of trash retention has been the increased ability of this surface horizon to retain nutrients. • Whilst it is evident that acidic cations have been retained on the exchange complex under the trash retention treatment, the application of lime would effectively neutralise this potential constraint and result in a substantial capacity to retain nutrients. It is suggested that this technique of charge fingerprinting is an effective means of assessing the impact of changed management on the soil resource base. 9.2 Trash decomposition, soil sodicity and the efficacy of calciumcontaining amendments and trash for ameliorating sodic soils Here we determined: a. Trash decomposition and soil sodicity. b. Effects of lime, gypsum, trash and their interactions on sodic soils and sugarcane growth. c. Structural stability of sodic soils as influenced by gypsum/molasses. d. Structural stability of sodic soils as influenced by molasses and lime. e. Effect of trash management on susceptibility of soils to becoming remaining sodic. • Decomposition of sugarcane trash was unaffected by high levels of soil sodicity (SEP 18 – 27) in a light- and a medium-textured soil, and was reduced by 18% in a heavier-textured soil. The latter effect was attributed to impaired soil aeration due to a combination of high sodicity, high clay content, high pH and relatively low organic matter content. Decomposition was unaffected by milder sodicity (ESP 8-12) across all soils 122 • It is unlikely that soil sodicity will be an important limit to trash decomposition in sugarcane fields where viable crops can be grown. • In the uncropped pot trial (using sodic soils from the Burdekin and Proserpine), lime and molasses, singly and in combination, increased electrical conductivity (EC) and reduced exchangeable sodium percentage (ESP). For EC, the effects of the two treatments were additive. In the Burdekin soil, lime increased pH, molasses decreased pH, and in combination there was a slight increase in pH. In the Proserpine soil, lime increased pH substantially, whereas molasses and molasses + lime increased pH slightly. • In both soils, lime and molasses, singly or in combination, decreased the amount of spontaneously and mechanically dispersible clay. The effect was most marked for spontaneously dispersible clay in the Burdekin soil and least marked for mechanically dispersible clay in the Proserpine soil. • In the Burdekin soil, wet aggregate stability was increased by lime and molasses, with their combined effect being additive. In the Proserpine soil, wet aggregate stability was increased by molasses but was not affected by lime. • Lime and molasses, singly and in combination, show promise as ameliorants for improving the physical properties of sodic soils. • The Gapon constant was significantly influenced by trash management in the Tully trial (the longest of the three GCTB trials considered), and by cropping history in the South Australian trial. The difference between Gapon constant between treatments at each site were consistent with differences in organic carbon concentrations and associated differences in cation exchange capacity. (Explanation: Cation exchange selectivity refers to the fact that some cations are held more strongly than others. For example, the cations Ca2+ and Mg2+, which have two charges (divalent), are held more strongly than Na+ and K+, which have one charge (monovalent). However, the difference between binding strength of divalent and monovalent cations depends on the nature of the cation exchange sites. Exchange sites on organic matter generally have higher relative selectivity for divalent cations than exchange sites on clays. Therefore, soils with high organic matter content are generally less susceptible to becoming sodic than soils with low organic matter content. The relative preference of a material for these two cations is described by the Gapon constant, KG. The higher the value of KG, the lower the relative preference of the material for Ca2+ relative to Na+. A soil with a high value of KG is more likely to become sodic under the influence of saline water, and more difficult to ameliorate, or make less sodic, than a soil with a low value of KG. Therefore, management practices that lower KG values, are beneficial from a sodicity management point of view. Cation exchange sites in soil occur on clay or organic matter. Exchange sites on organic matter tend to have lower values of KG than exchange sites on clay particles, so increasing soil organic matter by trash retention may lower the overall KG of the soil). 9.3 Effect of calcium-based amendments on trash decomposition Here we determined: a. The effect of various calcium amendments on trash decomposition and soil amelioration. b. The combined effect of calcium-based amendments and trash retention on nutrient availability. 123 • Calcium-based amendments, such as lime and millmud, have the potential to accelerate the decomposition of sugarcane trash. • This decomposition appears to be most effective when the amendments, particularly lime, millmud and mill ash, are applied to the surface of the trash layer. • Lime had a significant effect on trash decomposition, but the extent of decomposition was dependent on the method of lime application and soil type. Spreading lime on the surface of the trash layer enhanced decomposition. • As expected, incorporating lime into the soil beneath trash increased soil pH. Spreading lime onto the trash layer also resulted in marked increases in soil pH values in the surface soil (0-10 mm), but although there was evidence of small pH increases after seven months in the 10-50 mm soil layer (from the application of lime onto the surface of the trash), these differences were not significant. However, the substantially lower exchangeable Al values associated with the 10-50 mm soil layer at 4 and 7 months was evidence that some amelioration had occurred due to the application of lime to the surface of the trash. It appeared that the interaction between lime application and trash may possibly have important implications for the amelioration of sub-soil acidity with minimum tillage systems. • Data from two glasshouse pot trials conducted on five different soil types in different environments (drier subtropical versus moist tropical) indicated that application of certain amendments onto the surface of sugarcane trash had a significant effect on trash decomposition. Relative to unamended trash, the application of filtermud most successfully enhanced the rate of trash decomposition. Of the other amendments considered, significant differences were also applicable when lime (3 t/ha), potassium hydroxide (1.7 t/ha) and urea (160 kg/ha) were applied to the surface of the trash. • Soil type, treatment and depth below the trash layer had significant interactive effects on soil pH. As expected, the surface applied lime increased soil pH across the five soil types. This was most pronounced in the surface layers and generally declined with depth. The presence of a trash blanket (in the absence of any amendments) had variable effects on soil pH. In some soils the effect on soil pH was either similar or less than that of surface applied lime (no trash). However, in the case of the kurosol, lime application to the trash blanket resulted in higher pH values in the subsurface layers than when lime was applied in the absence of trash. This has important implications for certain acid soils where amelioration of subsurface layers can be enhanced by this interaction of lime and trash, possibly due to more soluble organocalcium complexes. • Mill mud application to the surface of the trash layer resulted in some amelioration of the subsurface layers, although the effect was not as pronounced as that with lime. • The simulated environmental differences between the drier subtropics and moist tropics (drier and cooler versus wetter and warmer) were reflected in the rate of decomposition of the trash (the amount of residual trash expressed as a percentage of the initial amount of trash in each pot). Significantly less trash remained at Macknade (higher temperature and moisture) than at Indooroopilly after the 7-month period in all treatments (including unamended trash) except millmud. As a result, we conclude that millmud appears to consistently enhance trash decomposition over a fairly broad range of moisture and temperature conditions. 124 • Enhanced trash decomposition that occurred due to millmud and millmud/mill ash, and lime applied to trash in pots under semi-controlled conditions was also found to be applicable in the field. • Although millmud and mill ash applications to trash resulted in significantly higher amounts of N, P and K removed by the harvested sugarcane biomass, these elevated values could either have been the result of the nutrients within the amendments themselves, from the trash, or a combination of both. • Millmud or millmud/mill ash mixtures both have positive effects on trash decomposition and ameliorative effects. • Environmental effects (water, temperature and soil properties) are considered important factors in determining nutrient availability in the amendment/trash systems. 9.4 Trash decomposition and C and N relationships Here we determined: a. Trash decomposition in controlled conditions. b. Trash decomposition and effects of trash, millmud, mill ash and urea on soil C and N. • When measuring decomposition of crop residues under glasshouse or laboratory conditions, it is desirable to approximate the conditions under which the residues decompose in the field. Close simulation of field conditions is, however, difficult to achieve because experimentation on a small scale (10-cm-diameter pots) with limited resources requires the researcher to make compromises. • In small pot experiments, factors such as the size to which the residue is cut; whether the residue is fresh or dried; the placement of the residue in or on the soil; and the soil water content vary from the field situation. The extent to which these factors affect decomposition of sugarcane trash is unknown, but studies with residues of other crops suggest that all these factors may be important. • To help choose experimental conditions for future decomposition studies, the effects of these factors on decomposition of sugarcane trash were determined. • Trash may be safely oven-dried and roughly ground before application to the soil, which greatly simplifies handling and subsampling of the trash. • The strong interaction between trash placement and soil water content means that these factors must be carefully controlled in decomposition experiments. • Measurement of trash decomposition as DM remaining at the end of the experiment is not an adequate method for small-scale studies where the trash fragment size had been greatly reduced. • Measurement of decomposition as CO2 loss is a more reliable and sensitive method, and allows decomposition to be measured non-destructively over time (it is also much easier). • Although it has been suggested, by growers and scientists, that trash decomposition may be increased by incorporating the trash into the soil by cultivation, the results of the small-scale pot experiments indicate that incorporation of trash may possibly hinder decomposition in situations where the soil is wet for long periods. • Application of straight mill ash does not appear to improve soil C or N availability (though it may improve the availability of some other nutrients and have nonnutritional benefits). 125 • When millmud and/or urea are applied in the presence of trash, N availability may initially be reduced due to microbial immobilisation. 9.5 General The advantages of retaining trash in the field can be further enhanced by the application of ameliorants to the surface of the trash blanket. Generally, the amendments, such as lime, millmud and millmud/mill ash mixtures applied to the surface of a trash layer have positive results for sugarcane cane production. Lime applied to trash has the ability to counter acidity generated by the mineralisation of N from the organic matter. Sources of organic matter (trash, millmud, molasses, etc), whether present individually or in combination with each other, have the ability to ameliorate sodic soil conditions. The potential of ameliorants such as lime and millmud to accelerate trash decomposition when applied to the surface of the trash layer has important implications for improving the return of organic matter to the soil, for ameliorating of soil acidity, and ultimately improving nutrient-use efficiency. Rates of up to 5 t/ha of lime (as determined by soil tests) and between 100 and 200 t/ha of wet millmud or millmud/mill ash mixtures applied to the surface of trash layers during the crop cycle would be appropriate to enable growers to reap the benefit of using Ca-based amendments in combination with trash retention. This practice and these rates are achievable in the sugar industry, as lime and millmud are often applied to sugarcane prior to planting. Although some technology transfer ha occurred in terms of papers, presentations and posters, further extension of the concept is proposed. In particular, a paper entitled ‘Cane trash and Ca-based soil amendments: A combination too good to miss’ will be presented at the 2004 ASSCT conference. Further extension will make use of grower meetings and workshops. 10.0 EXPECTED OUTCOMES • A clearer understanding of the effect of GCTB on soil organic matter and its ability to increase thew buffer capacity of soil, particularly where trash retention has been practiced for an extended period. • Greater awareness of the H+ inputs from trash and the influence of trash retention systems on soil chemical properties. • Increased use of GCTB on sodic soils, with the knowledge that trash decomposition is generally not affected by soil sodicity. • Increased use of millmud and mill ash as amendments applied to the surface of the trash blanket for amelioration of acid soils. • Increased use of millmud and millash for improving nutrient-use efficiency. • Increased use of lime applied to the surface of the trash layer for ameliorating acid soil conditions. • Use of Ca-based amendments to accelerate the decomposition of trash in circumstances where a trash blanket has detrimental effects on the growing crop (cool, wet climates, water-logged conditions, tropical areas, etc). 126 • Better use of combinations of lime, gypsum and molasses in combination with trash blanketing for ameliorating the physical effects of sodic soils. • Minimal wastage of nutrients for sugar industry by products and reduced off-farm impacts. • Greater awareness amongst growers, extension staff and agribusiness representatives for the need to use any synergies of GCTB and amendments usually applied to cane and/or those available in the industry as by-products. • Greater adoption of practices that lead to best practice management on-farm. 11.0 FUTURE RESEARCH NEEDS AND RECOMMENDATIONS A large amount of research and development that has been undertaken in relation to GCTB, especially within CRC Sugar projects, and has provided much information about the advantages and limitations of trash retention. This project has added to the pool of knowledge by evaluating the effects of applying commonly used ameliorants to the surface of trash layers. Although additional information is still required to validate the results from the small-scale pot experiments (Section 8) under field conditions, much of the information is ready for extension to the industry. Some technology transfer has occurred (Section 12) and in particular, a paper titled ‘Cane trash and Ca-based soil amendments: A combination too good to miss’ (BL Schroeder and AW Wood) is planned for presentation at the conference of the Australian Society of Sugar Cane Technologists in Brisbane in 2004. Apart from this, a concerted effort is needed to communicate many of the outputs listed in Section 9 to the industry via a range of extension methods. A booklet or information sheet summarising the important aspects would be an advantage. To enable more widespread use of the positive interaction between trash retention and Cabased amendments, research is needed to enable millmud and mill ash to be more widely used in canegrowing districts. At present, the relatively high water content makes transport of these products over longer distances uneconomical. Further field trials are needed to validate whether the range of outputs are appropriate in districts other than the dry subtropics (Bundaberg) and moist tropics (Herbert). Although the investigations relating to sodic soils and the C and N relationships were conducted on soils from the Burdekin, Mackay and Proserpine, larger pot experiments and/or field trials are need to evaluate the effect of other environments (especially rainfall and temperature) on trash decomposition in the presences of lime, millmud, mill ash, etc. An important by-product, dunder, was not included in this investigation. In the event of increased ethanol production in the Australian sugar industry, the implications of applying combinations of Ca-based products and dunder to trash layers will be important. Research is also needed to evaluate the availability of micro-nutrients within the amended trash systems. This could have important implications for soils with marginal to low trace element reserves. 127 12.0 PUBLICATIONS ARISING FROM THE PROJECT Presentations and published papers Marschner B & Noble AD. 2000. Chemical and biological processes leading to the neutralisation of soil acidity after incubation with different litter materials. Journal of Soil Biology and Biochemistry 32, 805-813. Schroeder BL & Aitken RL. 1998. Effect of lime on green cane trash decomposition. National Soil Acidification Conference, 15-17 July 1998, Noosa, Australia. Schroeder BL & Wood AW. 2004. Cane trash and Ca-based soil amendments: A combination too good to miss. Proceedings of the Australian Society of Sugar Cane Technologists in prep. Suriadi A, Murray RS, Grant CD & Nelson PN. 2002. Structural stability of sodic soils in sugarcane production as influenced by gypsum and molasses. Australian Journal of Experimental Agriculture 42, 315-322. Suriadi A, Murray RS, Grant CD & Nelson PN. 2002. Influence of crop management and water quality on the exchange selectivity of soils. ASSSI Future Soils Conference, 2-6 December 2002, Perth. Suriadi A, Murray RS, Grant CG & Nelson PN. 2000. Structural stability of sodic soils as influenced by gypsum and molasses. Sodicity Conference, 28 February – 1 March 2000, Tatura. Suriadi A, Murray RS, Grant CD & Nelson PN. in prep. Sodium-calcium exchange selectivity of soils under different trash management practices. Wood AW, Schroeder BL & Aitken RL. 2001. The effect of various calcium amendments applied to sugarcane trash. 10th Australian Agronomy Conference, Hobart, Tasmania, 28 January – 1 February 2001. Poster papers Noble AD & Berthelsen S. 2000. A new approach to understanding cation chemistry. Presented at CRC Sugar Annual Review and Planning Meeting, Townsville, 6 – 8 March 2000. Schroeder BL & Aitken RL. 1999. Effect of lime on green cane trash decomposition. Presented at CRC Sugar Annual Review and Planning Meeting, Townsville, 23 25 February 1999. Schroeder BL, Wood AW & Aitken RL. 2000. Assessing the effect of various Caamendments on the decomposition of trash. Presented at CRC Sugar Annual Review and Planning Meeting, Townsville, 6 - 8 March 2000. 13.0 ACKNOWLEDGEMENTS The Sugar Research and Development Corporation, Cooperative Research Centre for Sustainable Sugar Production, BSES, CSIRO Land and Water, and CSR Sugar contributed financially to the work in this report. Dr Robert Aitken was initially involved in the project. His involvement ceased after his resignation from the Queensland Department of Natural Resources. 128 Jason Hillyard, Tim Shapter, Lee Aitken, Ruth Mitchell, Kylee Sankowsky and Patricia Nelson (BSES), Lanelle Martin, Naomi Mackee and Sandra Coco (CSR Sugar), Bruce Crompton and Pat Victor (QDNRM) are thanked for their valuable technical assistance during the various facets of the investigation. Much of the experimental work associated with Paul Nelson’s assessment of the efficacy of calcium-based amendments for ameliorating sodic soils was carried out by Ahmad Suriadi in the University of Adelaide as part of his Masters project. His University of Adelaide supervisors Dr Robert Murray and Dr Cameron Grant are gratefully acknowledged for their contributions. 14.0 REFERENCES Aitken RL, Moody PW & Dickson T. 1998. Field amelioration of acidic soils in south east Queensland. I. Effect of amendments on soil properties. Australian Journal of Agricultural Research 49, 627-637. Alexander M. 1997. Introduction to Soil Microbiology. John Wiley and Sons, New York. Anderson DL & Hendrick JG. 1983. Subsoil lime injector. Soil Science Society of America Journal 47, 337-339. Andrén O, Rajkai K & Kätterer T. 1995. Water and temperature dynamics in a clay soil under winter wheat: influence on straw decomposition and N immobilization. Biology and Fertility of Soils 20, 1-8. Baldock JA, Aoyama M, Oades JM, Susanto & Grant CD. 1994. Structural amelioration of a South Australian red brown earth using calcium and organic amendments. Australian Journal of Soil Research 32, 571-594. Bartlett RJ, Ross DS & Magdoff FR. 1987. Simple kinetic fractionation of reactive aluminium in soil ‘solution’. Soil Science Society of America Journal 51, 14701482. 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. Blake GR & Hartge KH. 1986. Bulk density. In: A Klute (ed.) Methods of Soil Analysis. Part 1. Physical and Mineralogical Methods. 2nd Edn. pp. 363-375. Soil Science Society of America, Madison, WI, USA. Box GEP & Hunter JS. 1961. The 2k-p fractional factorial designs. Technometrics 3, 449. Bramley RGV, Ellis N, Nable RO & Garside AL. 1996. Changes in soil chemical properties under long term sugar cane monoculture and their possible role in sugar yield decline. Australian Journal of Agricultural Research 34, 967-984. Bremer E, van Houtum W & van Kessel C. 1991. Carbon dioxide evolution from wheat and lentil residues as affected by grinding, added nitrogen and the absence of soil. Biology and Fertility of Soils 11, 221-227. Calcino DV. 1994. Australian sugarcane nutrition manual. SRDC/BSES, Brisbane, Australia. Chorom,M & Rengasamy P. 1997. Carbonate chemistry, pH, and physical properties of an alkaline sodic soil as affected by various amendments. Australian Journal of Soil Research 35, 49-61. 129 Christensen BT. 1986. Barley straw decomposition under field conditions: effects of placement and initial nitrogen content on weight loss and nitrogen dynamics. Soil Biology and Biochemistry 18, 523-529. Churchman GJ, Skjemstad JO & Oades JM. 1993. Influence of clay minerals and organic matter on effects of sodicity on soils. Australian Journal of Soil Research 31, 779800. Conteh A, Blair GJ, Lefroy R & Whitbread A. 1999. Labile organic carbon determined by permanganate oxidation and its relationships to other measurements of soil organic carbon. Humic Substances and the Environment 1, 3-15. Coventry RJ & Fett DER. 1979. A pipette and sieve method of particle-size analysis and some observations on its efficacy. CSIRO Division of Soils Divisional Report No. 38, Townsville, Australia. Devêvre OC & Horwáth WR. 2000. Decomposition of rice straw and microbial carbon use efficiency under different soil temperatures and moistures. Soil Biology and Biochemistry 32, 1773-1785. Doss BD, Dumas WT & Lund ZF. 1979. Depth of lime incorporation for correction of subsoil acidity. Agronomy Journal 41, 541-544. Farina MPW & Channon P. 1988. Acid-subsoil amelioration: I. A comparison of several mechanical procedures. Soil Science Society of America Journal 52, 169-175. Freijer JI & Bouten W. 1991. A comparison of field methods for measuring soil carbon dioxide evolution: experiments and simulation. Plant and Soil 135, 133-142. Garside AL, Bell MJ, Cunningham G, Berthelsen J & Halpin N. 1999. Fumigation and rotation effects on the growth and yield of sugarcane. Proceedings of the Australian Society of Sugar Cane Technologists 21, 69-78. Garside AL, Smith MA, Chapman LS, Hurney AP & Magarey RC. 1997. The yield plateau in the Australian sugar industry: 1970-1990. In: BA Keating & JR Wilson (eds.). Intensive Sugarcane Production: Meeting the Challenges Beyond 2000, pp. 103-124. CAB International, Wallington, UK. Genstat 5 Committee. 1987. Genstat 5 Reference Manual. Oxford University Press, London. Gillman GP & Sumpter EA. 1986. Modification to the compulsive exchange method for measuring exchange characteristics of soils. Australian Journal of Soil Research 24, 61-66. Hammel JE, Sumner ME & Shahandeh H. 1985. Effect of physical and chemical profile modification on soybean and corn production. Soil Science Society of America Journal 49, 1508-1512. Haysom MB, Hurney AP & Neilsen PJ. 1986. The influence of lime on the acid soils of Ingham. Proceedings of the Australian Society of Sugar Cane Technologists 8, 5562. Helyar KR & Porter WM. 1989. Soil acidification, its measurement and the processes involved. In: AD Robson (ed.). Soil Acidity and Plant Growth, pp 61-101. Academic Press, Sydney. Helyar KR, Cregan PD & Godyn DL. 1990. Soil acidity in New South Wales – current pH values and estimates of acidification rates. Australian Journal of Soil Research 28, 523-537. Hetherington SJ. 1986. Effects of aluminium toxicity on the growth of sugarcane. MSc thesis, Department of Agriculture, University of Queensland, Brisbane, Australia. Hurney AP. 1971. An evaluation of liming in the Cairns-Babinda area. Proceedings of the Queensland Society of Sugar Cane Technologists 38, 109-112. 130 Isbell RF. 1996. The Australian Soil Classification. CSIRO Publishing, Melbourne. James BR & Riha SJ. 1986. pH buffering in forest soil organic horizons: relevance to acidic deposition. Journal of Environmental Quality 15, 458-462. Jarvis SC & Robson AD. 1983. The effects of nitrogen nutrition of plants on the development of acidity in Western Australian soils. II Effects of differences in cation/anion balance between plant species grown under non-leaching conditions. Australian Journal of Agricultural Research 34, 355-365. Jungk A. 1968. Die Alkalinität der Pflanzenasche als Maß für den Kationenüberschuß in der Pflanze. Z. Pflanzenernähr. Bodenkd. 120, 99-105. Kemper WD & Rosenau RC. 1986. Aggregate stability and size distribution. In: A. Klute (ed.). Methods of Soil Analysis. Part 1. Physical and Mineralogical Methods. 2nd edition. pp. 425-442. Soil Science Society of America, Madison, WI, USA. Kennedy IR. 1992. Acid Soils and Acid Rain, pp. 254. John Wiley & Sons Inc, New York. Kingston G & Aitken RL. 1996. Final Report on SRDC Project BS98S, Factors affecting the residual value of lime. BSES Publication SRDC Final Report SD97002. Kingston G, Davis RJ, Chapman FL & Nielsen PJ. 1998. Progress with green cane systems research in cool wet environments. Proceedings of the Australian Society of Sugar Cane Technologists 20, 552. Klok JA & Kingston G. 2002. Nutrient release from trash blankets in the wet tropics. Proceedings of the Australian Society of Sugar Cane Technologists 24, 480. Lehrsch GA, Robbins CW & Hansen CL. 1993. Cottage cheese (acid) whey effects on sodic soil aggregate stability. Arid Soil Research and Rehabilitation. 8, 19-13. Lutz JA & Jones GD. 1971. Yield of wheat and soybeans and oil and protein content of soybeans as affected by fertility treatments and deep placement of limestone. Agronomy Journal 63, 931-934. Mallawaarachchi T, Hurney AP & Noble AD. 1998. Long-term profitability of liming to manage soil acidity in sugarcane: Tully, north Queensland. Proceedings of the Australian Society of Sugar Cane Technologists 20, 257-263. McKenzie RC & Nyborg M. 1984. Influence of subsoil acidity on root development and crop growth in soils of Alberta and northeastern British Columbia. Canadian Journal of Soil Science 64, 681-697. Mitchell RDJ & Larsen P. 2000. A simple method for estimating the return of nutrients in sugarcane trash. Proceedings of the Australian Society of Sugar Cane Technologists 22, 212-216. Moody PW. 2002. Maintenance of soil organic matter. In: R.C. Bruce (ed.). Managing Soils, Nutrients and the Environment for Sustainable Sugar Production, pp. 99101. CRC for Sustainable Sugar Production, Townsville. Moody PW & Aitken RL. 1995. Soil acidification and lime use in some agricultural systems in Queensland, Australia. In: Plant-Soil Interactions at Low pH, pp. 749752 (eds RA Date, NJ Grundon, GE Rayment and ME Probert.). Kluwer Academic Publishers. Moody PW & Aitken RL. 1997. Soil acidification under some tropical agricultural systems: I. Rates of acidification and contributing factors. Australian Journal of Soil Research 35, 163-173. Muneer M & Oades JM. 1989. The role of Ca-organic interactions in soil aggregate stability. II. Field studies with 14C-labelled straw, CaCO3 and CaSO4.2H2O. Australian Journal of Soil Research 27, 401-409. 131 Nelson PN, Baldock JA, Clarke P, Oades JM & Churchman GJ. 1999. Dispersed clay and organic mater in soil: their nature and associations. Australian Journal of Soil Research 37, 289-315. Nelson PN, Baldock JA & Oades JM. 1998. Changes in dispersible clay content, organic carbon content, and electrolyte composition following incubation. Australian Journal of Soil Research 36, 883-897. Nelson PN, Barzegar AR & Oades JM. 1997. Sodicity and clay type: influence on decomposition of added organic matter. Soil Science Society of America Journal 61, 1052-1057. Nelson PN & Ham GJ. 2000. Exploring the response of sugar cane to sodic and saline conditions through natural variation in the field. Field Crops Research 66, 245255. Nelson PN, Ham GJ, Kingston G, Burgess D, Wood A, Christianos N, Hardy S, Wilson P & Lawer A. 2001. Diagnosis and Management of Sodic Soils under Sugarcane. CRC for Sustainable Sugar Production, Townsville. Nelson PN, Ladd JN & Oades JM. 1996. Decomposition of 14C-labelled plant material in a salt-affected soil. Soil Biology and Biochemistry 28, 433-441. Nelson PN & Oades JM. 1998. Organic matter, sodicity and soil structure. In: ME Sumner and R Naidu (eds.). Sodic soils- distribution, properties, management, and environmental consequences, pp. 51-75. Oxford University Press, New York. Noble AD, Bramley RGB, Wood AW & Hurney AP. 1997. Sugarcane and soil acidity – Why should we be worried? Proceedings of the Australian Society of Sugar Cane Technologists 19, 187-199. Noble AD, Gillman GP & Ruaysoongnern S. 2000. A cation exchange index for assessing degradation of acid soil by further acidification under permanent agriculture in the tropics. European Journal of Soil Science 51, 233-243. Noble AD, Zenneck I & Randall PJ. 1996. Leaf litter ash alkalinity and neutalisation of soil acidity. Plant and Soil 179, 293-302. Norrish K & Hutton JT. 1977. Plant analysis by X-ray spectrometry. 1. Low atomic number elements, sodium to calcium. X-Ray Spectrometry 6, 6-11. Oades JM 1989. An introduction to organic matter in soils. In: JB Dixon & SB Weed (eds.). Minerals in Soil Environments, pp. 89-159. Soil Science Society of America, Madison, WI, USA. Parr JF & Papendick RI. 1978. Factors affecting the decomposition of crop residues by microorganisms. In: Crop Residue Management Systems. pp. 101-129, American Society of Agronomy, Madison, WI, USA. Piccolo A. 1996. Humus and soil conservation. In: A Piccolo (ed) Humus Substances in Terrestrial Ecosystems, pp. 225-264. Elsevier, Amsterdam, The Netherlands. Porter WM, McLay CDA Dolling PJ. 1995. Rates and sources of acidification in agricultural systems of southern Australia. In: RA Date et al. (eds). Plant-Soil Interactions at Low pH: Principles and Management, pp. 75-83. Kluwer Academic Publishers, The Netherlands. Puget P, Chenu C & Balesdent J. 2000. Dynamics of soil organic matter associated with particle-size fractions of water-stable aggregates. European Journal of Soil Science 51, 595-605. Quinan DJ & Wood AW. 1989. The effects of lime applications on acid soils in the Hambledon Mill area. Proceedings of the Australian Society of Sugar Cane Technologists 11, 70-75. 132 Quinan DJ & Wood AW. 1993. Long term effects of lime application on some soils in the Cairns area. Proceedings of the Australian Society of Sugar Cane Technologists 15, 368. Quirk JP & Schofield RK. 1955. The effect of electrolyte concentration on soil permeability. Journal of Soil Science 6, 163-178. Rayment GE & Higginson FR. 1992. Australian laboratory handbook of soil and water chemical methods. Inkata Press, Melbourne. Reddy MV, Reddy VR, Yule DF, Cogle AL & George PJ. 1994. Decomposition of straw in relation to tillage, moisture, and arthropod abundance in a semi-arid tropical Alfisol. Biology and Fertility of Soils 17, 45-50. Reghenzani JR. 1993. A survey of the nutritional ststus of north Queensland sugarcane soils with particular reference to zinc. Proceedings of the Australian Society of Sugar Cane Technologists 15, 298-304. Rengasamy P, Greene RSB, Ford GW & Mehanni AH. 1984. Identification of dispersive behaviour and the management of red-brown earths. Australian Journal of Soil Research 22, 413-431. Ridge DR, Hurney AP & Haysom MBC. 1980. Some aspects of calcium and magnesium nutrition in north Queensland. Proceedings of the Australian Society of Sugar Cane Technologists 2, 55-61. Ridley AM, Helyar KR & Slattery WJ. 1990. Soil acidification under subterranean clover (Trifolium subterraneum) pastures in north-eastern Victoria. Australian Journal of Experimental Agriculture 30, 195-201. Robbins CW. 1986. Sodic calcareous soil reclamation as affected by different amendments and crops. Journal of Agronomy 78, 916-920. Robertson FA. 1998. Effects of trash management on nitrogen and carbon cycling in sugacane soils. Progress Report on CRC Activity 2.2.2.1. CRC for Sustainable Sugar Production. Robertson FA. 2003. Sugarcane Trash Management: Consequences for Soil Carbon and Nitrogen. Final Report: CRC Sugar, Townsville, Australia. Robertson FA & Thorburn PJ. 2000. Trash management - consequences for soil carbon and nitrogen. Proceedings of the Australian Society of Sugar Cane Technologists 22, 225-229. Robertson FA & Thorburn PJ. 2002. Factors affecting trash decomposition and nitrogen release. Proceedings of the Australian Society of Sugar Cane Technologists 24, 487. Saayman D & van Huyssteen L. 1981. The efficiency of different methods of lime application during soil preparation. South African Journal of Oenology and Viticulture 2, 26-36. Sekhon BS & Bajwa MS. 1993. Effect of organic matter and gypsum in controlling soil sodicity in rice-wheat-maize system irrigated with saline water. Agricultural Water Management 24, 15-25. Shaw RJ, Coughlan KJ & Bell LC. 1998. Root zone sodicity. In: M.E. Sumner and R. Naidu (eds.). Sodic soils: distribution, properties, management and environmental consequences, pp. 95-106. Oxford University Press, New York. Soil Survey Staff. 1992. Keys to Soil Taxonomy. Soil Management Support Services, Technical Monograph No. 6. 4th Edn. Virginia Polytechnic Institute and State University: Blackburg, VA. 133 Sørensen P, Ladd JN & Amato M. 1996. Microbial assimilation of 14C of ground and unground plant materials decomposing in a loamy sand and a clay soil. Soil Biology and Biochemistry 28, 1425-1434. Spain AV & Hodgen MJ. 1994. Changes in the composition of sugarcane harvest residues during decomposition as a surface mulch. Biology and Fertility of Soils 17, 225231. Sumner ME. 1995. Amelioration of subsoil acidity with minimum disturbance. In Subsoil management techniques, pp. 147-185 (eds NS Jayawardane and BA Stewart). Lewis Publishers: Boca Raton, USA. Thorburn PJ, Lisson SN, Probert ME, Keating BA & Wood AW. 1999. Plant residue retention in sugarcane systems: Modelling impacts on sustainable production. p. 70. In: Food and Forestry: Global changes and global challenges. The Global Change and Terrestrial Ecosystems Focus 3 Conference Delegates manual. Elsevier Science, Oxford. Thorburn PJ, Probert ME & Robertson FA. 2001. Modelling decomposition of sugar cane surface residues with APSIM-Residue. Field Crops Research 70, 223-232. Tisdall JM & Oades JM. 1982. Organic matter and water-stable aggregates in soils. Journal of Soil Science 33, 141-163. Walkley A & Black IA. 1934. An examination of the Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid titration method. Soil Science 37, 29-38. Weber HW & Van Rooyen PC. 1971. Polysaccharides in molasses meal as an ameliorant for saline-sodic soils compared to other reclamation agents. Geoderma 6, 233-253. Wood AW. 1991. Management of crop residues following green harvesting of sugar cane in north Queensland. Soil Tillage Research 20, 69-85. Wood AW, Schroeder BL & Stewart RL. 2003. Development of site specific nutrient management guidelines for sustainable sugarcane production. Proceedings of the Australian Sugar Cane Technologists 25, pp. 14. Yeomans JC & Bremner JM. 1988. A rapid and precise method for routine determination of organic carbon in soil. Communications in Soil Science and Plant Analysis 19, 1467-1476. Zibilske LM. 1994. Carbon mineralisation. In: Weaver RW, Angle S, Bottomley P, Bezdicek D, Smith S, Tabatabai A & Wollum A. (eds.). Methods of Soil Analysis. Part 2, Microbiological and Biochemical Properties, pp. 835-863. Soil Science Society of America Inc, Madison, USA.