The development and assessment of sustainable agricultural systems is a complex process. Before attempts are made towards developing sustainable systems, a primary concern is having a workable definition of sustainability. Definitions of sustainability vary according to the perspective of the observer, and whether this is based on considerations of agricultural production, social and political stability, or conservation of the environment and biodiversity. A very general statement of sustainable development was provided in 1987 by the Brundtland Report (WCED 1987), which stated that "Sustainable development ensures the satisfaction of economic, social and cultural needs of the present generation without detriment to the needs of future generations." A more specific concern affecting both current and future generations is the question of health, which must be factored into the sustainability equation.
A more practical definition for agricultural systems is encompassed in the Framework for Evaluating Sustainable Land Management (FESLM) developed by Smyth and Dumanski (1993) with the involvement of the International Board of Soil Research and Management (IBSRAM). Under the FESLM, a system is considered to be sustainable if it combines technologies, policies and activities aimed at integrating socioeconomic principles with environmental concerns so as to simultaneously maintain or enhance production and services, reduce production risks, protect the potential of natural resources and prevent degradation of soil and water quality; it must also be economically viable and socially acceptable. That is, the system must embrace the five pillars of sustainability: productivity, stability, protection, viability, and acceptability. The issue of sustainability can be considered at the farm scale, the regional scale, or the global scale.
At the global level, an important outcome from the development of sustainable production systems will be improved food security and poverty reduction, achieved through a balance between population and food supply. At the local level, it relates to the welfare of individuals, of family groups, and communities, but still relies on balancing the pressure on the resource base with the demand for output from the system.
Sustainability in Intensive Vegetable Production
Intensive vegetable production, like intensive agriculture in general, is very productive in terms of yield per unit of land, compared to less intensive agriculture. The high values of the vegetable products provide a strong incentive to maximize yield. These systems, therefore, generally require much greater use of external inputs, to the point that productivity on the basis of some inputs might be fairly low.
Generally, as a result of the significant level of inputs, the stability of production of these systems is fairly high.
These systems are economically viable as long as the cost of the excess inputs remain low relative to the value of production. Experience shows that as prices for inputs such as fertilizers, pesticides, and irrigation increase, farmers quickly improve the efficiency with which the inputs are used, or they become unviable.
The social acceptability of intensive vegetable production by the population at large is frequently fairly high. However, the proximity to population centers means that acute sensitivities to off-site effects can arise and seriously decrease acceptability. Generally, the acceptability is fairly high within the farming community, until the health of farmers is seriously affected or perceived to be threatened.
The final pillar of sustainability, the protection of the resource base and the prevention of degradation of the soil and water quality, is the pillar that is frequently given least attention in many intensive agricultural systems, including intensive vegetable production. As education levels rise, environmental awareness increases, so that a community recognizes the failure to satisfy the protection and the acceptability pillars concurrently. Unfortunately, in many cases recognition occurs only after productivity, stability, and viability, or at least long-term sustainability, have been affected.
The location of intensive vegetable production near areas of high population density results from the demand for food which is perishable, or not as easily stored or transported as are staple food products. This high demand for vegetables results in a price structure which allows for relatively low efficiency of utilization of some of the less expensive inputs.
Increasing population puts pressure on land for increased production and through competition for other uses. Increasing urbanization means that cities are expanding rapidly, which provides increased demand for vegetables and pressure on agricultural land for housing, industry and infrastructure. There is very little concern for the quality of the land that is taken over by expanding cities, which highlights the need for improved land use policies.
Land for intensive vegetable production is affected in other ways by its closeness to areas with high populations. Firstly, there are significant man-made effects on the environment, which are greater near population centers, and which affect the air, water, and soil. Secondly, inappropriate management of intensive vegetable production systems can have significant off-site effects on the groundwater and surface water in terms of both their quality and their quantity, both of which are already under direct threat from the increasing urban populations.
An assessment of sustainability requires consideration of all five pillars of the FESLM. The development and implementation of sustainable agricultural systems requires a multidisciplinary approach, with significant participation by farmers, or at least consideration of their needs and perceptions. The functions of many parts of intensive vegetable production systems and the immediate economic impacts of current production technologies is understood reasonably well, but other parts need further research, so that we have a better understanding of their biophysical functions.
The requirements for and responses to many of the inputs, particularly the more expensive ones, are known in some detail. The impact on the soil of high rates of inputs is less well understood. The problem for biophysical scientists in developing sustainable systems, rather than assessing them, is to balance the factors by which a system is assessed in terms of productivity, stability, viability, and acceptability, with factors by which it is assessed in terms of protection. Thus, the challenge is to develop systems which protect the potential of the resource base whilst being acceptable in terms of the other four pillars.
The title of this paper, "Soil as a filter for nutrients and chemicals: sustainability aspects", might suggest that soil is commonly considered rather passively, its role being simply to hold up the plants while water, nutrients and other agrochemicals are applied. In intensive agricultural systems in general, this is frequently the manner in which soil is considered. However, it is important that the soil is considered in a much more active sense as playing a role in the supply of water and nutrients: as an active, living filter, with separate but closely related chemical, physical, and biological dimensions to its filtering role. These dimensions are dependent on the inherent properties of the soil and on management. Soils must be managed to be filters which are sufficiently porous for infiltration, in order to reduce runoff and erosion, but sufficiently non-porous so as to reduce leaching and increase the efficiency of nutrient use. Appropriate management requires that the cycling of carbon, nutrients, water, and contaminants in the system be understood. The development of systems which preserve the resource base, and avoid degradation of soil and water, requires a detailed understanding and monitoring of the flows and balances of carbon, nutrients, and water.
The Carbon, Nutrient, and Water Cycles
The cycles of carbon, nutrients, and water are intimately interrelated. However, the different components and their roles need to be identified before more sustainable management systems can be devised. Carbon is added to the system by CO 2 fixation, and removed in products and residues. Some or all of the residues may be retained within the system. Other flows of C into the system are as organic amendments from off-site, such as crop residues, and animal and other wastes. Losses of C, other than by crop removal, occur during the decomposition of crop residues and soil organic matter (SOM), removal of residues and SOM by erosion, and leaching of soluble organic compounds.
The major flows for most nutrients are: additions to the system in the form of fertilizers or organic amendments, recycling through crop residues, and losses through product and residue removal, erosion, runoff, and leaching. In addition, N is added to the system from the atmosphere through biological nitrogen fixation. S, N, and other nutrients can be added through rainfall, and soluble nutrients can be added in irrigation water. Some nutrients, particularly N, can be lost directly to the atmosphere.
The major flows of water, which have significant effects on the movement and transformation of soil components, are rain, irrigation, evapotranspiration, runoff and leaching.
Most pesticides, herbicides, and other agrochemicals are applied to protect crops, but some arrive as irrigation water contaminants, coming from runoff or erosion from other areas. Contaminants other than water-transported pesticides and herbicides, or their breakdown products, can be deposited from the atmosphere or as contaminants such as cadmium and mercury, in fertilizers.
The highly interrelated nature of these cycles means that different management systems can have synergistic and antagonistic effects on other parts of the cycle or cycles, or have no impact at all.
The Cycling and Management of Carbon
While plants require only atmospheric CO 2 as a C source, soil C has great significance for the growth and nutrition of crops. Plant and animal residues, the precursors for soil C, are an important source of nutrients. Plants do not partition nutrients between harvested product and residue uniformly. Likewise, animals do not assimilate nutrients in proportion to the concentrations in feed. Thus the balance of nutrients in these residues is not in the exact proportions that plants require. Nevertheless, these organic residues constitute an important source of macro- and micronutrients for intensive vegetable production systems.
While the carbon in the organic amendments is not of direct use to plants (with the possible exception of some exogenous plant hormones), it is a major energy source for microorganisms. Microorganisms use the residues as energy and nutrient sources during the conversion of residues into forms of SOM, with the concomitant widening of the C-to-nutrient ratios as nutrients are used by microorganisms or released for plant uptake.
The management of residue additions and decomposition enables adjustment of the release of nutrients for plant growth. Rates of residue decomposition can be increased or decreased through the timing, placement, preparation and incorporation of the residues, the control of soil moisture and, to a certain extent, temperature, the concurrent application of nutrients in inorganic fertilizers to prime decomposition, and the selection of appropriate residues.
Residues need to be chosen for an environment so that residue decomposition results in nutrient releases that match crop demands or achieve the desired changes in SOM. If release is not in synchrony with demand, surplus nutrients are free to leach below the root zone or become unavailable through chemical or biochemical processes. If nutrients are leached below the rooting zone, groundwater quality may be jeopardized, as well as nutrient recovery reduced.
The major determinants of the rate of organic residue decomposition are the forms of C compounds, particularly the proportion of lignin and polyphenols compared to more labile compounds, the content of nutrients, particularly (but not only) nitrogen, and the physical characteristics. Nutrient release can be measured as plant response to nutrient release in the field. Alternatively, the potential value of different residues can be assessed in the laboratory, by measurement of nutrients, lignin and polyphenols to estimate breakdown (Tian et al. 1995, Vanlauwe et al. 1997), or by a simple method for measuring the relative breakdown rate of residues under controlled conditions (Lefroy et al. 1995).
Beside directly contributing to the chemical fertility of soil through additions of nutrients, SOM contributes to the anion and cation adsorption capacity of the soil. This is of particular importance in light textured soils which have very little adsorption capacity. Increasing the adsorption capacity of the surface layers of the soil profile can reduce, or even prevent, the leaching of nutrients. Leached nutrients are, of course, unavailable to the plants, and furthermore, are potential contaminants of groundwater and surface waters. Additionally, in acidic soils with low base saturation, SOM may reduce toxic effects of aluminum by complexation with organic compounds (Myers et al. 1997).
SOM can have significant effects on the fate of pesticides, as microorganisms mediate much of the breakdown. By increasing soil microbial activity through residue additions, the breakdown rate of pesticides increases. In addition, the adsorption capacity of SOM helps retain pesticides and their breakdown products in the surface layers of the soil, thus reducing runoff and leaching and the resultant contamination of ground water. Similarly, adsorption of those biocides which reduce soil biological activity can mitigate the negative effect on the beneficial activities of soil microbes and fauna.
In addition to the effects of residues on soil chemical fertility, physical fertility is significantly affected. By increasing the quantity and, to a certain extent, changing the type of SOM, the degree and stability of soil aggregation can be improved. Better aggregation improves aeration, reduces soil strength, and increases infiltration of water and the quantity of plant available water. Most of these changes are beneficial to the resource base and therefore to the production system, and are a buffer against poor management, such as inappropriate tillage or irrigation scheduling. Although increased infiltration may promote leaching, the substantial increase in the chemical adsorption capacity of the surface layers as SOM increases will impede downward movement of nutrients and pesticides, and improved soil physical conditions, resulting in deeper root penetration, will allow greater capture of the nutrients which moved down the profile.
There is evidence that significant quantities of soluble carbon move down the soil profile, thus reducing the carbon content of the surface soil layers and increasing the content of the deeper layers. The movement of soluble carbon compounds almost certainly means that there is a concomitant movement, and potential loss, of nutrients. Carbon movement can increase the chemical and physical fertility of these deeper layers, but it can affect nutrient dynamics, most particularly for N. Whether from mineralized organic sources or fertilizers, nitrates and nitrites which have moved down the profile can be recovered if there is sufficient root activity at depth. However, if root activity is insufficient and soluble carbon compounds have moved down the profile with, or to, the nitrates and nitrites, the carbon compounds can provide a substrate for denitrifying bacteria, resulting in a loss of N to the atmosphere. Although this process will decrease N recovery by the crop, it might be preferable to nitrate contamination of groundwater.
The surface application of residues as mulch, as well as bringing the effects which result from organic matter incorporation, can impede erosion directly. Mulch reduces the impact of rain droplets, and this reduces the detachment of soil particles. In addition, mulch reduces the volume and velocity of surface water flow, and this in turn reduces soil particle entrainment and transport from the field (Rose 1987). Together, these effects reduce soil erosion.
Residue mulch will reduce moisture lost to evaporation, and affect soil temperature through insulation and altered albedo. These can have positive and negative effects on production and the resource base through direct affects on plant growth, on microbial activity, and on potential leaching. In addition, mulch can reduce the potential for leaching of agrochemicals by direct adsorption of the chemicals in the mulch layer, thus preventing, or slowing, entry to the soil.
Where buildup or maintenance of SOM is of greater importance than nutrient release, the addition of low-quality residues that decompose slowly might be desirable. The buildup or maintenance of SOM is important on both a local and global scale. On a local scale, SOM buildup improves the general chemical, physical, and biological fertility of the soil and preserves the resource base. On a global scale, the buildup sequesters C, which reduces atmospheric CO 2 and combats global warming.
The balance between the various flows of carbon can be monitored by measurements of products and residues, and changes in soil carbon pools. The effects of residues on nutrient release, adsorption of nutrients and agrochemicals, infiltration of water, soil strength, etc., result in a complex set of potentially positive and negative effects on the resource base. The management of carbon in the various ways outlined above is a powerful tool for managing the short- and long-term productive capacity of soils.
The complex interactions between various management technologies and SOM highlight the need for appropriate tools to monitor SOM, at least during the development and assessment of sustainable management systems, if not in routine soil analyses. Total carbon is not an appropriate measure, as it is not sensitive to short-term changes in the quality of the resource base. Most attempts to develop models of SOM turnover, and to relate SOM dynamics to soil fertility, have involved the separation of carbon into a number of pools on the basis of their rate of turnover (Parton et al. 1987, McCaskill and Blair 1988). Only when a number of pools with very different turnover rates are incorporated into these models can the modeled release of nutrients and changes in SOM approximate the observed variations.
To be appropriate for measuring changes in SOM, a method must measure the rapidly cycling labile carbon pool. Blair et al. (1995) measured a labile carbon pool (C L) by oxidation with potassium permanganate. C L is used, in combination with similar data from a soil of a reference area, to calculate a Carbon Management Index (CMI). The CMI measures the relative sustainability of different agricultural systems by comparing the changes that occur in total carbon (C T) and C L as a result of a particular agricultural practice. More importance is attached to changes in C L, than in the non-labile (C NL) component of the SOM, since C L is most active in nutrient dynamics and in improving soil structure.
Wonprasaid et al. (in press) collected soil samples down to 40 cm from a rice field on a Paleaquult in northeast Thailand, and compared these with samples from an adjacent forest ( Table 1). A large decline in C T, and an even greater decline in C L, were observed down the profile. The carbon in the top 20 cm was massively reduced by an extended period of rice cultivation. These changes are reflected in low CMI values. The high CMI in the 20 to 40 cm layer of the cultivated soil results from an increase in C T and C L relative to the forest soil, indicating movement of C down the profile. In more extensively cropped soils in this region, the lability of carbon (L), the ratio of C L to C NL, has been shown to stay constant or even increase down the profile, indicating even greater movement of labile C. These changes in soil carbon and CMI are associated with reduced fertility and low rice yields.
In contrast, some agricultural systems result in an increase in C L and CMI relative to the reference soil. For instance, data from a soil used for sugarcane production in Queensland, Australia, demonstrate the benefits of careful residue management (Blair et al. 1995). In this case, where all cane residue was returned rather than burned, the cropped site had higher C L, C T, and CMI than the adjacent non-cropped reference soil ( Table 2).
Changes in C L and the CMI have been related to changes in the physical and chemical fertility of soil. Whitbread et al. (1996) showed that the significant declines in water stable aggregates and infiltration that occurred with extensive cropping were correlated with changes in C L and CMI.
The Cycling and Management of Nutrients
While farmers appreciate the positive effects of good nutrition on intensive vegetable production, the relative cost of nutrient inputs is such that they have a limited understanding of, or perceived need for, exact responses to nutrient management. As the cost of nutrients is low relative to the cost of all inputs, there is little incentive to maintain a good balance between nutrient addition and crop demand. The result is that fertilizer inputs are often much greater than required. A major incentive for fine-tuning nutrient additions is not improved nutrient use efficiency, but rather reduced negative effects from the unrecovered nutrients that result from excessively high applications. The main negative effects of high fertilizer applications are occurrences of serious nutrient imbalances and toxicities that affect yields or crop quality, and off-site effects from leaching and erosion of nutrients, particularly N and P.
Accurate measurement of nutrient off-take in product and return in residues is an important first step toward appreciation of the flows and balances of nutrients. Although good data are rarely available, it should be possible to obtain reasonable estimates of this part of the nutrient balance through good estimates of yields and careful use of mean nutrient content data for the species and, ideally, for the particular agroecosystem and set of management practices.
The nutrient inputs from fertilizers are probably the component of nutrient balances which can be most easily and accurately estimated from average application rates and compositions of fertilizers, at least for the major nutrients. The situation for organic amendments is rather different. While it should be possible to gather information on, or reasonable estimates of, the quantity of organic amendments applied, it is difficult to estimate nutrient additions accurately. Frequently, estimates of rates of addition do not consider the moisture content of the amendments. In addition, the nutrient contents, either on a wet- or dry-weight basis, are not known. Mean values must be used with great care, even when corrected for moisture content. The variation in nutrient contents of organic amendments is enormous, even when they consist of the same type of raw material.
In many cases, inorganic fertilizers have replaced organic amendments to satisfy the large nutrient supplies required for intensive vegetable production. However, because of the twin pressures to reduce fertilizer applications, and to recycle nutrients in municipal wastes more effectively, it is likely that organic amendments will remain a significant part of the nutrient cycle of many intensive vegetable production systems. To this end, more accurate monitoring of nutrient cycling requires improved methods for estimating inputs from inorganic and, more particularly, organic sources.
Biological N fixation can contribute significant amounts of N to the system. The value of N in above-ground residues and in roots depends on how much residue is retained and how the system is managed. The breakdown rate of annual legume residues is generally high because N concentrations are high and C is mostly in easily decomposed forms. Management practices which further hasten breakdown, such as chopping and incorporation, and good management of soil moisture, increase N availability for the subsequent crop, but also increase leaching potential if root activity in the soil is low or the potential for N retention is low.
Inputs of nutrients from atmospheric sources can be significant. The major inputs are of N and S, and the major anthropogenic sources are from fossil fuel combustion (S and N) and metal smelting (S). As crops require more N than S, the relative importance of atmospheric sources of nutrients is greatest for S. Measurement of atmospheric inputs of S in Peninsular Malaysia indicated inputs of between 1 and 30 kg S/ha per year, depending on proximity to major cities and proximity to the coast. These have a significant implication for fertilizer applications (Lefroy and Hussin 1989). Annual atmospheric accessions of S were up to 100 kg S/ha near major cities in industrialized North America and Europe before serious steps were taken to reduce emissions in the late 1980s (Whelpdale 1992).
Precise monitoring of atmospheric inputs of nutrients is not required. However, reasonable estimates of accessions, particularly of S, are needed to make important adjustments to fertilizer inputs.
The contribution of nutrients in irrigation can be significant. Nutrients in the water come from natural sources, the atmosphere, and runoff, leaching and erosion from agricultural land. The nutrient contributions from irrigation depend on the concentration, the amount of irrigation water applied, and the hydraulic conductivity and sorption capacity of the soil. Precise monitoring of nutrient inputs will only be achieved if the concentration and quantities of water applied are regularly monitored. However, good estimates can be obtained from estimates of water applied and average nutrient concentrations.
In addition to nutrient removal in harvested products, major losses can occur through runoff, erosion, and leaching. Loss rates are a combination of the rate of application of fertilizers and organic amendments, residue management, the SOM decomposition rate, and, most importantly, water management. In an analysis of the threats to groundwater quality from different production systems, Aminuddin et al. (1996) concluded that the risks were much greater from vegetable and sugarcane production than from lowland rice systems. Compared to rice, vegetables and sugarcane are more frequently grown on lighter textured soils and receive heavier and more frequent additions of fertilizer. Moreover, lowland rice soils are predominantly under reduced conditions, so that the N in the soil solution is largely in the form of NH 4 + rather than NO 3 -. Nitrate is more easily leached than NH 4, and is of much greater concern as a health hazard when it contaminates groundwater.
Nutrients are also lost directly to the atmosphere. Nitrogen is by far the major nutrient lost to the atmosphere, and microbial denitrification is the main pathway. Reducing N losses to the atmosphere requires improving recovery by adjusting the timing and placement of inorganic and organic amendments, managing crop residue and SOM decomposition rates, managing the movement of N and C in the soil profile by managing the water cycle, and encouraging deeper root penetration.
Improved interpretation of flows and cycles of the nutrient balances for the purpose of assessing sustainability or improving fertilizer practices, requires better measurements of inputs and losses. The consequences of negative and positive balances can be measured by soil nutrient analyses, particularly trends over time and space, including down the soil profile. Better nutrient management requires knowledge of the nutrient requirements of crops and the likely responses to nutrient amendments. This includes improved knowledge of the time course of nutrient demand by different crops. The consequence will be the development of improved ways to manipulate the timing of release of organic and inorganic nutrients to match demand â€” in other words achieving synchrony. Synchrony can be achieved through the development and use of slow-release fertilizers, or by controlling nutrient availability through the timing and placement of fertilizers, through varying the fertilizer particle and granule size, and through management of the breakdown of organic residues and SOM.
An important part of improving nutrient management will be the development and improvement of in-season nutrient monitoring tools. This might involve increased use of leaf analyses, particularly for N. Alternatively, it might involve a more farmer-friendly approach, such as correlating leaf color with N content, as is being attempted for rice.
The Cycling and Management of Water
The key to managing the cycling of water is to satisfy crop demand while being mindful that water can transport nutrients and agrochemicals in runoff and leachate. While very high accuracy might be difficult, estimation of the inputs are relatively easy through measurement of rainfall and irrigation. Loss measurements are more difficult. Evapo-transpiration can be estimated from information on crop growth and climatic conditions. Runoff can also be estimated, using appropriate sampling techniques and by calibrating rainfall, infiltration, and runoff. Leaching is more difficult to estimate, although the consequences of leaching can be measured by the presence of nutrients and agrochemicals deep in the profile or in groundwater.
Rather than attempting to measure all of the components of the water cycle in detail, measurements can be made of the consequences of the water cycle i.e. changes in soil water content. A number of tools and techniques exist, varying in sophistication, to monitor soil water accurately. These include neutron moisture meters, time domain reflectometers, tensiometers, and arrays of gypsum blocks. A more farmer-friendly, but highly site-specific approach, might be to calibrate soil penetrometer resistance with a relevant soil water content range and a farmer estimate of soil strength. Various soil water models which require only simple inputs have been developed. It is probable that these models will be advanced to the point that a set of reference tables can be produced for different soil types, climates, crops, and stages of crop development to improve soil water management practices.
There are very few specific management tools which can be used to alter the water cycle. Where irrigation is used, the scheduling of irrigation and the type of irrigation â€” flood, sprinkler, drip â€” can be selected, depending on the leaching potential. Other management tools are less specific, in that they affect infiltration and water movement in the soil through management of tillage and bed preparation, surface cover, and soil physical properties.
Crop management has a major effect on the water cycle. Adjustments of row and plant spacings, intercrop combinations and crop rotations can be used to improve the capture of nutrients and water. The key point is to maintain active roots in the profile, particularly in times of high risk for movement of water and nutrients. Deep rooted intercrop species can be used to recover nutrients and water from depth. This principle was clearly demonstrated by Stirzaker (1996) in growing vegetables between rows of lucerne ( Medicago sativa). By allowing some deep movement of water, particularly during the period between the vegetable crops, major competition for water between the shallower-rooted vegetable crops and the deep-rooted lucerne was avoided. Nutrients which leached beyond the vegetable root systems were used by the lucerne and returned to the surface when the lucerne tops were used as a mulch. The lucerne mulch provided benefits through better water use and as a nutrient source, while its roots greatly reduced the deep percolation and nutrient loss. The yield reduction per area as a result of intercropping with a species such as lucerne, might be partially off-set by better use of water and nutrients, by reduction of off-site effects, by preservation of the resource base, and by reducing the need to rotate land. These systems require careful development and very careful management to avoid serious competition resulting in reduced productivity.
The Cycling and Management of Agrochemicals
Agrochemicals are a very important part of intensive vegetable production systems. Better understanding of predator-prey relations and pest and disease ecology, host plant resistance and agrochemical applications has led to improved management of pests and diseases. Effective control of insect pests and diseases has contributed to increased and more stable production, and to the viability of current intensive vegetable systems. Social acceptability and natural resource protection, including prevention of degradation of soil and water quality, have not always been the foremost considerations of agrochemical users. More recently, prevalent misuse of agrochemicals has caused alarm.
The movement of agrochemicals primarily depends on transport by water, but also it depends on the reactivity of the particular chemicals. For highly soluble chemicals, runoff and leaching are major loss pathways. For chemicals of low solubility, erosion is more important, either through direct movement or movement of soil particles on which the chemicals are adorbed. The potential for environmental damage by chemicals depends on rates of breakdown which differ significantly among chemicals. Moreover, chemical half-lives vary with temperature, soil moisture, and soil type, through the effects of these factors on microbial and chemical activity. Volatilization is an important loss pathway for some chemicals. The environmental consequences of these chemicals depend on a combination of the quantity applied, the rate of breakdown, the rate of loss, the mobility, and the toxicities of these chemicals and their breakdown products.
Some pesticides move very slowly. For instance, simple models suggest that the time taken to leach through a 1 m soil profile is 5.5 years for Lindane, and 40 years for Dieldrin. This can be compared to approximately 0.3 years for chloride (Sharma et al. 1996). Despite these differences, chemicals of very different mobility are sometimes detected at the same profile position having after been applied at much the same time. This shows the spatial variability of soils that results in preferential flow, and means that the movement of agrochemicals, and plant nutrients for that matter, is very event driven (Flury 1996).
Methods of reducing the on- and off-site effects of agrochemicals are logical: reduce the quantities applied, choose chemicals on the basis of expected half-lives, manage water carefully, and use appropriate application methods. Because many soil processes are linked, positive and negative effects can result from particular management options. For instance, residue incorporation can increase SOM, which has the positive effect of increasing adsorption on soils. However, the additional SOM, particularly in combination with reduced tillage, can increase the preferential flow of water and thus increase leaching (Flury 1996). The balance between positive and negative outcomes must be carefully assessed.
Application practices can affect the movement of agrochemicals. Increasing the period between an application and rain or irrigation can significantly decrease leaching, as a result of increased adsorption and breakdown. Similarly, residue incorporation can increase adsorption and microbial breakdown, and thus decrease the potential for leaching.
It is clear that very complex sets of variables must be considered in the development of sustainable management systems. This complexity evolves from the interaction of and between the carbon, nutrient, and water cycles. There is much to be learned about each of these cycles, but the greater problems are in appreciating how alternative management options interact and then in selecting the more sustainable set of options.
Tools need to be developed to handle these complex interactions. One option is to use appropriate carbon, nutrient and water models combined with decision support systems (DSS). Ultimately, such DSS or model combinations need to be converted into packages which can be used easily by extensionists and farmers.
It is important that, as far as possible, major decisions have a sound biophysical and economic basis. Furthermore, it is essential that farmers' perceptions and attitudes are built into the decision-making process at an early stage. Rather than waiting for the dissemination stage, farmers must participate in deciding which of the myriad of management possibilities should be evaluated in the development stage.
The ultimate aim is to develop systems that are of similar or superior productivity to present systems, require less inputs, preserve the resource base, and create low or no off-site effects and no ill effects on farmers' health. Such management systems must be developed in cognizance of carbon, nutrient, and water cycles, and they must simultaneously satisfy the five pillars of sustainability.
- Aminuddin, B.Y., M.L. Sharma and I.R. Willett. 1996. (Eds.). Agricultural Impacts on Groundwater Quality. ACIAR Proceedings No. 61. 97 pp.
- Blair, G.J., R.D.B. Lefroy and L. Lisle. 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.
- Flury, M. 1996. Experimental evidence of transport of pesticides through field soils â€” a review. Journal of Environmental Quality 25: 25-45.
- Lefroy, R.D.B. and A. Hussin. 1991. Sulfur in rainfall and its relevance to Malaysian agriculture. In: Developments in Soil Research in Malaysia, Y.M. Khanif, S.R. Syed Omar and J. Shamshuddin (eds.). Malaysian Society of Soil Science, Kuala Lumpur, pp. 14-22.
- Lefroy, R.D.B., Y. Konboon and G.J. Blair. 1995. An in vitro perfusion method to estimate rates of plant residue breakdown and associated nutrient release. Australian Journal of Agricultural Research 46: 1467-1476.
- McCaskill, M.R. and G.J. Blair. 1988. Development of a simulation of sulfur cycling in grazed pasture. Biogeochemistry 5: 165-181.
- Myers, R.J.K., M. van Noordwijk, and Patma Vityakon. 1997. Synchrony of nutrient release and plant demand: plant litter quality, soil environment and farmer management options. In: Driven by Nature: Plant Litter Quality and Decomposition, G. Cadisch and K.E. Giller (eds.). CAB International, UK, pp. 215-238.
- Parton, W.J., D.S. Schimel, C.V. Cole and D.S. Ojima. 1987. Analysis of factors controlling soil organic matter levels in Great Plains Grasslands. Soil Science Society of America Journal 51: 1173-1179.
- Rose, C.W. 1987. Controlling erosion for cropping system experiments. In: Soil Management under Humid Conditions in Asia and the Pacific, M. Latham (Ed.). IBSRAM Proceedings No. 5, pp. 297-316.
- Sharma, M.L., C.H. Mohammud, M. Mohd Zain, M. Ahmed and Q. Richards. 1996. Piezometer grid studies to access pollution sources at tobacco and paddy farms, Kelantan, Malaysia. In: Agricultural Impacts on Groundwater Quality, B.Y. Aminuddin, M.L. Sharma and I.R. Willett (Eds.). ACIAR Proceedings No. 61, pp. 53-61.
- Smyth, A.J. and J. Dumanski. 1993. FESLM: An International Framework for Evaluating Sustainable Land Management. FAO World Soil Resources Reports No. 73. FAO, Rome, 74pp.
- Stirzaker, R.J. 1996. Lucerne/vegetable alley cropping to control leaching. In: Proceedings of the 8 th Australian Agronomy Conference, M. Ashgar (Ed.). The Australian Society of Agronomy, Carlton, Victoria, pp. 522-525.
- Tian, G., L. Brussaard, and B.T. Kang. 1995. An index for assessing the quality of plant residues and evaluating their effects on soil and crop in the (Sub-) humid tropics. Applied Soil Ecology 2: 25-32.
- Vanlauwe, B., J. Diels, N. Sanginga, and R. Merckx. 1997. Residue quality and decomposition: An unsteady relationship? In: Driven by Nature: Plant Litter Quality and Decomposition, G. Cadisch and K.E. Giller (eds.). CAB International, UK, pp. 157-166.
- Whelpdale, D.M. 1992. An overview of the atmospheric sulphur cycle. In: Sulphur Cycling on the Continents: Wetlands, Terrestrial Ecosystems and Associated Water Bodies, R.W. Howarth, J.W.B. Stewart and M.V. Ivanov (Eds.). SCOPE report No. 48, John Wiley and Sons, Chichester, UK, pp. 5-26.
- Whitbread, A.M., R.D.B. Lefroy and G.J. Blair. 1996. Changes in soil physical properties and soil organic carbon fractions with cropping on a red brown earth soil. In: Procee dings of the 8th Australian Agronomy Conference, M. Ashgar (Ed.). The Australian Society of Agronomy, Carlton, Victoria, pp. 582-585.
- Wonprasaid, S., K. Naklang, S. Tippayaruk, R.D.B. Lefroy, A.M. Whitbread and G.J. Blair. Legume residue and rice stubble effects on subsequent rice yields and soil carbon dynamics at Ubon, Thailand. Proceedings of the Conference on Maximising Sustainable Rice Yields Through Improved Soil and Environmental Management. (November 1996, Khon Kaen, Thailand). DOA/SFST/ISSS. (In press).
- WCED (World Commission on Environment and Development). 1987. Our Common Future. Oxford University Press, New York.
Index of Images
Table 1 Soil Carbon Fractions of a Cultivated and Forested Paleaquult from Ubon, Thailand
Table 2 Soil Carbon Fractions of a Pelloxerert Used for Sugarcane Production in Queensland, Australia, Compared to an Adjacent Non-Cropped Area
Download the PDF. of this document, 188,277 bytes (184 KB).