Printer Friendly

Fertilisers and phosphorus loss from productive grazing systems.

Abstract

This paper reviews phosphorus loss from productive high rainfall grazing systems. In particular it describes the processes occurring when phosphatic fertilisers are added to soil, the different pathways through which fertiliser and other nutrient sources may contribute to phosphorus losses, and an evaluation of the management strategies aimed at minimising phosphorus loss.

It is now generally accepted that soil is not an endless sink for phosphorus uptake and that at the landscape scale the highest concentrations of phosphorus loss occur in surface runoff, followed by macropore flow and vertical matrix flow. However, loads of phosphorus lost through these pathways are unknown. The development of an understanding of the transport mechanisms and phosphorus species being transported is fundamental to developing management strategies that are effective in decreasing phosphorus losses from grazing systems.

Introduction

Phosphorus supplements in the form of manures, vegetable material, and bones have been used to enhance productivity since the beginning of agriculture. The first manufactured phosphatic fertiliser can probably be attributed to the German chemist Liebig, who in 1840 suggested that, by dissolving bones in sulfuric acid, phosphorus was made more available to plants (Cathcart 1980). This product, later known as superphosphate, was so popular that bones were soon in short supply. Mining of phosphate rock for fertiliser manufacture started in 1847 when 500 t was mined in Suffolk, England. By the mid-1860s, when the first mines opened in the United States (Cathcart 1980), world production was over 100000 t of phosphate rock annually (Van Kauwenbergh 1992).

Today the world's fertiliser industry mines approximately 600 million t of phosphate ore, producing over 160 million t of rock concentrate and 40 million t of [P.sub.2][O.sub.5] annually. This [P.sub.2][O.sub.5] is contained in approximately 131 million t of fertiliser products (Gregory 1992).

The most direct benefit from the use of fertiliser phosphorus is increased agricultural production. Additional benefits include decreased erosion due to greater groundcover, increased soil water-holding capacity, and decreased modulus of rupture, which may be related to increased soil organic matter, especially in grazed pastures (Lutz and Haque 1975; Barrow 1980). Unfortunately, in making additional phosphorus available to plants, stimulating plant growth and increasing the turnover of phosphorus, fertilisers can also have detrimental environmental impacts.

Phosphorus and other nutrients in water stimulate the growth of plants, including undesirable algae (Pilgram 1986; Sharpley et al. 1987; Sharpley and Smith 1992). This nutrient enrichment and stimulation of plant growth is commonly referred to as eutrophication and it limits the potential uses of the affected water. Nutrients lost from agricultural systems may well come directly from fertiliser, fertilisers applied some years previously, unfertilised soils, and a range of other sources including plants and grazing animals.

In the context of increased risk of algal bloom occurrence, waterbodies are considered at greater risk if the phosphorus concentration exceeds 0.05 mg/L (Cottingham et al. 1995). This is significantly lower than grazing systems, where the total phosphorus concentrations in the soil solution can exceed 1 mg/L (Sanyal and DeDatta 1991; Slattery et al. 1998).

The cost of eutrophication is not just ecological. In Australia, the financial cost of eutrophication is rapidly escalating as the incidence of algal blooms increases. In Victoria during 1996-97 there were over 101 blooms reported in 91 water storages (Anon. 1996). The cost of the capital infrastructure necessary to treat these blooms and loss of amenities are difficult to quantify. One now infamous algal bloom of over 1000 km of the Barwon-Darling River (Australia) in 1991-92 is reported to have cost the local tourist industry alone $AU9 million and has the dubious honour of being the world's largest recorded riverine algal bloom (Cottingham et al. 1995).

This paper reviews phosphorus loss from productive grazing systems. In particular it describes (i) mechanisms and sources of phosphorus lost from high rainfall, pasture-based grazing systems; (ii) processes occurring when phosphatic fertilisers are added to soil; (iii) different ways in which fertiliser phosphorus may contribute to phosphorus losses; and (iv) evaluation of the management strategies aimed at minimising phosphorus loss.

Phosphorus-cycle

Most Australian soils have inadequate supplies of phosphorus and nitrogen for optimum plant growth (Sale 1992). Off-site inputs of phosphorus are required to develop a balanced pasture sward in which legumes, predominantly clovers, fix atmospheric nitrogen. The application of phosphatic fertilisers therefore rectifies both nutrient limitations. Phosphorus also enters the farm in feed, concentrates, and, to a lesser extent, in rainfall.

In grazing systems, the plants, animals, and soil can all store phosphorus. Plants incorporate inorganic phosphorus from the soil solution into their structure. Animals eat the plants, incorporating the plant phosphorus in their own biomass. Ultimately some of this phosphorus is returned to the soil when the plants and animals and their wastes are decomposed. Decomposing plant and animal products, along with the soil microflora and fauna that undertake such decomposition, provide a significant store of phosphorus. Importantly, if the conversion of organic phosphorus to inorganic forms, termed `mineralisation', is less than the rate at which phosphorus in the soil water is incorporated into microbial biomass, there is less phosphorus available for plant growth.

Phosphorus mineralised from organic matter and inorganic fertilisers contribute to the inorganic phosphorus in soil water, predominantly orthophosphate. In a process termed `fixation' such compounds attach to insoluble materials and so are stored in a form that is again not immediately available to plants (Engelstad and Terman 1980).

The distinction between phosphorus stored in its inorganic or organic form is important. Microorganisms largely control phosphorus release from organic sources and mineralisation is favoured by conditions that favour microbial growth. The rate of phosphorus release from inorganic sources is largely a chemical process dependent on `equilibrium' concentrations of reactants such as iron and aluminium, and the chemical conditions that influence them such as pH and oxygen potential.

Dairy farming is one of the most intensive pasture-based grazing systems. A simple nutrient balance suggests that phosphorus is accumulating in these systems. The phosphorus content of milk is approximately 0.95 g/L (Grace 1983). For a dairy farm producing 500000 L of milk from 100 cows annually, approximately 475 kg of phosphorus is needed to replace that leaving the farm in milk. Assuming 25% herd replacement and 3.2 kg of phosphorus per cow (Grace 1983), a further 80 kg of phosphorus is needed to compensate for the cows leaving the farm. Assuming no other losses, [is less than] 600 kg of phosphorus is needed to replace phosphorus leaving the farm in produce.

The average dairy stocking rate for Gippsland (Victoria, Australia) is 1.84 cows/ha and the average phosphorus application is approximately 44 kg/ha annually (Drysdale 1998). These figures suggest that, for dairy systems, significantly more phosphorus is applied than is being recovered in produce. While marginal changes in phosphorus stored in the standing crop may occur, this excess phosphorus must either be stored in the soil or lost.

Pathways of phosphorus loss

Both wind and water transport nutrients from agricultural land into aquatic systems. The dominant transport mechanism depends on the nutrient in question and a range of site-specific factors. However, wind erosion is not generally considered an important mechanism of phosphorus loss from well-managed grazing systems in high rainfall areas (Anon 1996).

Historically, the soil fabric has been assumed to act as an almost endless sink for applied phosphorus, and most phosphorus that was lost was believed to be lost in surface runoff attached to eroded sediments (Russell 1957). However, in the last 40 years a number of factors have changed, along with our views on phosphorus. Both dissolved and particulate phosphorus have been measured in runoff from grazing systems (Cullen 1991; Hazel 1991; Nash and Murdoch 1996a; Nelson et al. 1996). Fertiliser has been applied to pastures, increasing the rate at which phosphorus is cycled, productivity, and the amount of phosphorus stored in soil. Soil has been shown to have a finite capacity to hold phosphorus and when this limit is reached the concentration of phosphorus in soil water increases (Barrow 1989; Holford 1989).

Australian soils in their native state were usually highly weathered and nutrient-depleted (Sale et al. 1997). Presumably due to repeated fertiliser use, many of these soils are now much closer to phosphorus saturation, and the subsequent loss of solution phosphorus, than they were in the 1950s. It follows that, in addition to erosion, phosphorus may be lost as water moves through soil in processes variously described as matrix flow, macropore flow, or interflow, or over soil in infiltration excess overland flow or surface runoff. No doubt all these processes contribute to phosphorus loss in varying degrees.

The processes by which water and associated phosphorus move through or over soil are presented in Fig. 1. While the starting point is always at the soil surface where rain falls, the paths from there are very different and rarely as clearly defined as the diagram suggests.

[Figure 1 ILLUSTRATION OMITTED]

Vertical matrix flow will be defined as water moving slowly through soil to lower strata from where it eventually moves laterally to a stream. It may take days or longer for matrix flow to reach a stream and the intimate contact with soil during that time allows for the physical and chemical removal of particulate material and solutes from the water. Matrix flow is the main pathway by which unsaturated flow contributes to nutrient loss.

In macropore or by-pass flow, water moves quickly through large soil pores (Kirkby et al. 1997) to a porous region of the profile from where it is conveyed laterally to a stream. Artificial subsurface drainage can readily perform the latter function (Armstrong and Harris 1996; Dils and Heathwaite 1996). Compared with matrix flow, in macropore flow the opportunity for water and soil to interact is limited by the speed of water movement and the large volume of water relative to the surface area of soil over which it moves.

Interflow is a term commonly used to describe water moving vertically into surface soil, usually the first 20 mm, and then laterally along the surface, for example at the interface of the A and B horizons. Where the soil surface is sloping or the soil profile has been disrupted by clearing and ploughing, interflow may come to the surface and combine with infiltration excess overland flow. For the purposes of this review, the term surface runoff will be used to describe collectively interflow and infiltration excess overland flow, which is simply the excess water that the soil could not take in. The slow movement of surface flow through and over the nutrient-rich surface soil, especially on upper slopes, provides an ideal opportunity for the mobilisation of nutrients either attached to small particles or as solutes.

It should be noted that as surface runoff concentrates, especially in the lower reaches of a catchment, the energy of the water per unit soil area tends to increase. As a result, opportunities for physical and chemical removal of nutrients from the water decrease and the opportunities for mass movement of soil through soil erosion increase.

At a landscape scale, the highest concentrations of phosphorus occur in surface runoff, followed by macropore flow and matrix flow. This ranking is consistent with the level of interaction between water and the soil matrix occurring in each. For example, Nash and Murdoch (1996a) measured total phosphorus concentrations in surface runoff of up to 14.7 mg/L, while subsurface water yielded [is less than] 0.05 mg/L (D. Nash unpublished data), and Fleming et al. (1997) measured total phosphorus in surface runoff and interflow, finding almost twice the phosphorus concentration in surface runoff.

Unfortunately there is less information available relating to the loads of nutrients lost via the different processes. While measurement of the concentrations of nutrients in water is comparatively easy (Rhoades and Oster 1986; Sharma et al. 1996), the same is not true for the volumes of water movement. Often by necessity the scale and design of studies pre-disposes them to particular results. For example, plot studies ([is less than] 1 ha) have been used extensively to study nutrient loss. While these studies are extremely useful for investigating specific processes such as erosion, their applicability at a landscape scale is questionable (Gburek et al. 1996). Many processes, including sediment deposition, are scale-dependent and the residence time of surface runoff in most plot studies would be very low compared with that in a catchment, allowing less time for dissolution and desorption processes to occur. The methods used for water measurement are also likely to alter the balance between nutrient loss pathways. For example, in some studies of water movement, subsurface drains are used both to prevent the movement of water into and onto the trial area and to take water away for measurement (Heng et al. 1991; Magesan et al. 1996). Such a system clearly alters the hydrology of the site, increasing the likelihood of infiltration at the expense of surface runoff. These are not fatal flaws, provided the limitations are recognised and the results are not applied to the entire toposequence; unfortunately they often are.

There is insufficient information to conclude that any one pathway is primarily responsible for phosphorus losses from grazing systems at a catchment scale (Table 1). Generally surface runoff contains higher phosphorus concentrations than matrix flow or macropore flow, but is only a small portion of the rainfall applied to pastures. Considerable opportunity exists for the loss of phosphorus in larger volumes of less concentrated solutions through matrix flow and possibly macropore flow.

Table 1. Phosphorus losses from different land uses

Flow weighted mean for year. See also Dillon and Kirchner (1974); Sonzogoni et al. (1980); Kofoed (1984); Gregg et al. (1993). TP, total phosphorus; DRP, dissolved reactive phosphorus
Year Country Pathways Phosphorus
 loss

1983 Australia Surface runoff 0.22
1983 Australia Surface runoff 0.02-0.45
1985 Australia Surface runoff 4
1985 Ireland Surface runoff 0.4-2
1991 Australia Stream 1.0
 monitoring 1.1
1991 Australia Stream 1.3-1.8
 monitoring
1993 England Surface runoff 1.35
 simulations
1996 England Surface runoff/ 3
 lysimeters
1996 Australia Surface runoff 1.9-5.5
1996 England Undrained
 surface runoff
 Drained
 surface runoff
 Drainage

Year Production Scale Range TP
 (kg P/ha year) Type (mg/L)

1983 Sheep grazing 15 ha 0.45-2.0
1983 Mown pasture Plots 10 by 4 m 0.36-6.75
1985 Grazed, irrigated Bay 1.3-21.2
 pasture
1985 Intensive 6 ha 0.095-0.295
 grassland
1991 Sheep grazing 130 ha
 Cattle grazing 300 ha
1991 Dairy grazing 124 and 79 ha 0.5-95
 predominantly 1.0-3.8
1993 Heavily grazed Hillslope 0.4-9.4
 permanent
 pasture
1996 Intensive 1-ha
 grassland lysimeters
1996 Grazed, fertile 1.8 ha 3.8-22.3
 pasture
1996 Intensive 1-ha 0.026-1.773
 grassland lysimeters
 0.010-0.892
 0.010-0.605

Year DRP/TP Author

1983 <70% Tham (1983)
1983 >50% Greenhill et al.
 (1983b)
 Greenhill et al.
 (1983a)
1985 >50% Small (1985)
1985 36-45% Jordan and Smith
 (1985)
1991 <14% Nelson et al. (1991)
 69% Nelson et al. (1996)
1991 >50% Hazel (1991)
1993 <20% DRP Heathwaite (1993)
1996 30% Haygarth and
 Jarvis (1996a)
1996 >90% Nash and
 Murdoch (1996a)
1996 40% Haygarth and
 Jarvis (1996b)
 50%
 31%


Phosphorus sources

When rain falls on a pasture it washes over the plants onto the soil surface. The rainwater, plants, added fertilisers, animal waste products, and the soil all contribute to the nutrients contained in the water. It is generally accepted that rainfall makes little direct contribution of phosphorus to runoff (Tabatabai and Laflen 1976; Mays et al. 1980; Greenhill et al. 1983a). Rather, the soil-plant system, which includes plants, animals, added fertilisers, and soil, contributes to the nutrients found in runoff water.

Plants

A number of authors have studied the contribution of plants to nutrient loss, almost exclusively using model studies and decomposing material (Jones and Bromfield 1969; Timmons et al. 1970; Bromfield and Jones 1972; Schreiber 1985; Schreiber and McDowell 1985; Havis and Alberts 1993). Bromfield and Jones (1972) and Jones and Bromfield (1969) studied nutrient losses from `hayed-off' phalaris and clover plants under a wide range of laboratory conditions. Of the total phosphorus in plant material, 60-83% was water-soluble and up to 62% of this phosphorus (37-51% of the total) was leached by 125 mm of rainfall over 96 h. Environmental conditions clearly played a large role in the results, but concentrations of phosphorus varied from 2 to 150 mg/L, suggesting that plants may be an important source of phosphorus. Interestingly, plants grown with higher levels of soil solution phosphorus contained more total phosphorus, with a higher percentage of that phosphorus in water-soluble form, than those grown in less fertile media and lost about the same percentage of water-soluble phosphorus. The increased loss of phosphorus at higher levels of available phosphorus suggests that the losses of phosphorus from plants may be more important in highly fertile pastures (Bromfield and Jones 1972).

A simple model of potential nutrient losses from plants based on the figures of Bromfield and Jones (1972) is presented in Table 2. Plants are potentially an important source of the inorganic phosphorus lost from grazing systems, particularly in surface runoff. Nexhip et al. (1997) compared the phosphorus lost from flood-irrigated bays after mowing and grazing. Their results suggest that, after mechanical damage, growing plants may also contribute directly to phosphorus loss. Phosphorus concentrations from the mown control were statistically greater than losses from the two lowest grazing pressures of 100 and 200 cows/ha immediately after defoliation (Nexhip et al. 1997).
Table 2. A model of water-soluble phosphorus (WSP) in pasture plants

Total dry matter Phosphorus WSP WSP
 (kg/year) concentration (%)(B) in pasture
 (%)(A) (kg/ha.year)

 15 000 0.25 60 23
 15 000 0.25 83 31
 15 000 0.45 60 41
 15 000 0.45 83 56
 10 000 0.25 60 15
 10 000 0.25 83 21
 10 000 0.45 60 27
 10 000 0.45 83 37


(A) Anon. (1997a).

(B) Bromfield and Jones (1972).

Soils

The surface is the most phosphorus-rich zone for all but a few soils. It is here that broadcast fertilisers are applied, animals defecate, and plants, acting as biological pumps, deposit phosphorus extracted from lower in the profile. It is here too that rainfall and soil interact, usually to a depth of only a few millimetres (Sharpley et al. 1981; Ahuja 1986).

Soil is an extremely complex material and surface soils in particular contain a vast array of inorganic and organic compounds, detrital material, and flora and fauna. In reviewing the physico-chemical processes controlling phosphorus in soil, Holford (1989) listed the forms of phosphorus in order of increasing stability as:

* phosphorus in solution ([is less than] 1% of total)

* inorganic phosphorus in plant residues

* inorganic phosphorus adsorbed on surfaces of clay and organic matter

* inorganic phosphorus in various compounds, ranging from sparingly soluble to extremely soluble

* inorganic phosphorus occluded or absorbed in P-reactive minerals

* very stable organic phosphorus in plant, animal, and microbial material.

Such lists tend to underestimate the role of organic phosphorus, which is highly variable but generally accounts for 50% of total soil phosphorus (Richardson 1994), especially near the surface (Perrott and Sarathchandra 1989; Perrott et al. 1990).

The distinction between inorganic and organic forms of phosphorus is important. The two forms of phosphorus react differently in soil and this affects their transport in water.

Inorganic phosphorus

Plants absorb phosphorus as inorganic ions from the soil solution (Holford 1997). Consequently the reactions of inorganic phosphorus in soil have been extensively studied (Smith 1965; Sample et al. 1980; Barrow 1989; Holford 1989) and are well reviewed elsewhere (Wild 1949; Norrish and Rosser 1983; Haynes 1984; Sanyal and DeDatta 1991; Schulthess and Sparks 1991).

The concentration of phosphorus in soil solution is maintained by precipitation/dissolution and adsorption/desorption processes. The phosphorus buffering capacity is a commonly used measure of a soil's ability to maintain the phosphorus concentration in the soil solution and, presumably, other water such as surface runoff. It is most often computed from an isotherm comparing the amounts of phosphorus adsorbed by soil to various equilibrium phosphorus concentrations under standard conditions. The slope of the isotherm at an arbitrary solution phosphorus concentration is referred to as the buffering capacity or sorptivity (Holford 1989). It is of note that the slope of the straight-line graph comparing phosphorus sorbed against the log of the equilibrium phosphorus concentration is also referred to as phosphorus buffering capacity (Rayment and Higginson 1992), but this later definition will not be used in this paper. The higher the soil buffering capacity, the higher the proportion of phosphorus in the solid phase compared with the solution phase, the lower the rate of phosphorus mobilisation from the solid phase, and the lower the rate of diffusion through the solution phase (Holford 1989).

The buffering capacity is affected by the number of sites where phosphorus can be held, be these adsorption sites or precipitation sites, and the affinity of these sites for phosphorus (Holford 1989). As phosphorus is added to soil, the number of available sites and their affinity for phosphorus progressively decrease, as indicated by a decreasing slope of the isotherm. It follows that as phosphorus is added to soil, for example through fertiliser additions, the buffering capacity is decreased. Further, since sorption isotherms are curvilinear, the higher the soil fertility the greater the change in buffering capacity from fertiliser additions.

Buffering capacity is a useful way of comparing how soils respond to increased phosphorus in the soil solution. However, when phosphorus concentrations decrease, for example after rain, both the buffering capacity and the amount of phosphorus that can be re-mobilised, labile phosphorus (Schofield 1955), are needed in order to estimate a soil's `replenishment capacity' (Holford 1989). The soil's replenishment capacity would appear to be the single most important factor affecting the mobilisation of phosphorus in water after dilution. As might be expected, it is negatively related to soil buffering capacity and positively related to the amount of labile phosphorus in soil. However, the loss of phosphorus in flowing water will depend on both the replenishment capacity of the soil and the rate at which phosphorus can be mobilised.

The reactions of inorganic phosphorus in soil imply that soil tests aimed at quantifying phosphorus availability for plants in the top 75-150 mm of the profile, such as Olsen P, Bray P, or Morgan P, are of limited value in determining nutrient loss. The dilution of the phosphorus-rich surface soil with the phosphorus-deficient subsoil will grossly underestimate phosphorus availability to surface water. As the buffering capacities of the surface and subsurface soils are not linearly related to their respective phosphorus contents, the Olsen, Bray, or Morgan phosphorus test value will be lower than would otherwise be the case. Mixing will also increase the sampling and measurement errors expressed as a percentage of the mean. Further, the use of salt solutions in the extractants would be expected to increase phosphorus availability, inadequately reflecting the situation when the soil solution is diluted (Barrow 1989).

The following hypothetical example is considered. Two adjacent paddocks on a dairy farm are treated identically with the exception that one has undergone pasture renovation in the previous year. Pasture renovation includes cultivation of the soil to at least 10 cm, resulting in mixing of the high-phosphorus surface soil with phosphorus-deficient subsoil. While both profiles contain the same total phosphorus concentration, the renovated soil will lose less phosphorus since it will have less labile phosphorus and a higher buffering capacity at the surface. Such a simplistic example emphasises that traditional measurements of soil phosphorus status that include the root-zone reflect what roots are exposed to, but provide a poor indication of potential phosphorus movement at the soil surface (Sharpley 1985).

Agronomic soil tests are environmentally important as they help determine how best to achieve optimum productivity with minimum phosphorus additions. They may also be useful in predicting leaching of phosphorus in heavily fertilised cropping soils (Heckrath et al. 1995). However, other tests such as phosphorus extraction by water are likely to be more useful for determining potential phosphorus losses (Sharpley et al. 1982; Sharpley and Smith 1989; Haygarth 1997).

The reactions of phosphorus in soil also help explain the concentrations of inorganic phosphorus lost by the different pathways. It is likely that surface runoff will move slowly over and through poorly buffered surface soil with high labile phosphorus stores. Both soil properties and the time available for soil and water interaction favour high concentrations of phosphorus. The concentration of phosphorus in macropore flow may be only marginally less than that found in surface flow. The rapid transmission of water through macropores and the formation of poorly buffered coatings on their internal surfaces would tend to minimise changes in phosphorus concentrations as the water moves through the otherwise highly buffered, low in labile phosphorus, subsoil. However, the concentration of phosphorus in matrix flow would be dramatically decreased as it moves slowly through the highly buffered subsoil matrix.

Organic phosphorus

Organic components of soil can affect phosphorus mobility in 3 ways: (i) as the compound transporting phosphorus, (ii) by complexing inorganic phosphorus in water thereby stabilising its concentration in solution, and (iii) by blocking adsorption/precipitation sites, decreasing the capacity of soil to remove phosphorus from water.

Organic forms of phosphorus have received far less study than inorganic forms. Most organic phosphorus in soil originates from animal and plant wastes, and the decomposition of soil biota. Only small amounts of phosphorus are usually found in the urine of grazing animals (Braithwaite 1976), suggesting that faeces is the dominant form by which phosphorus is returned to the soil. The phosphorus concentration of cattle dung is usually around 0.5% on a dry weight basis (Thompson 1989). While the total phosphorus concentration of dung can vary depending on the phosphorus concentration in pasture (Rowarth et al. 1988), it would appear that the organic phosphorus component remains unchanged (Rowarth 1987). The predominant inorganic form of phosphorus in sheep dung has been identified as dicalcium phosphate (Barrow 1975). Unfortunately there is less information available on phosphorus contained in cattle faeces or the dominant organic phosphorus components. It is as yet unclear whether the organic matter which influences phosphorus mobility is actually contained in organic materials such as dung applied to soil, or whether it is derived from organisms using these materials as substrate, or both (Dickinson and Craig 1990).

The analysis of organic phosphorus lost via different processes can best be described as rudimentary. The best evidence for the mobility of organic materials contributing to phosphorus mobility comes from numerous studies in which leaching of organic and inorganic phosphorus have been compared (Reddy et al. 1978; Chardon and Oenema 1995; Edwards et al. 1996; Chardon et al. 1997). Nash and Murdoch (1996b), for example, investigated phosphorus losses from 300-mm-deep soil cores treated with 50 kg P/ha in the form of single superphosphate (mono-calcium phosphate) and cattle faeces. In the case of the fertiliser, a single peak of phosphorus was leached from the core the second week after treatment. However, the leachate from the faeces-treated cores had significantly higher phosphorus concentrations than the controls for the 5 weeks following treatment. The predominant forms of phosphorus in the leachate were also different, dissolved reactive phosphorus and dissolved non-reactive phosphorus for the fertiliser and faeces treatments, respectively. The study concluded that, while additions of phosphorus in fertiliser did not statistically increase losses of phosphorus in leachate, the application of phosphorus in faeces did. These results are consistent with the long-term Rothamsted studies where the application of manures, as compared with inorganic fertilisers, has increased phosphorus at depth in grazed pastures (Johnston and Poulton 1992).

Organic matter may influence the concentration of inorganic phosphorus in solution. The ability of anionic organic matter to compete with phosphorus for adsorption sites in ligand-exchange is well documented but has not been quantitatively described (Holford 1989; Schulthess and Sparks 1991). A second mechanism by which organic matter may increase the mobility of inorganic phosphorus, particularly in acidic soils, is by decreasing the chemical activity of iron or aluminium, which would otherwise precipitate phosphorus from solution (Thomas 1975; Bloom et al. 1979). In the highly organic environment of the soil surface one might expect both mechanisms to be important. But there is little quantitative information available on either mechanism.

Fertiliser

The type of fertiliser applied to pasture and the reaction products that form are potentially important in determining the quantities of phosphorus lost in runoff. Some of the more important compounds used in phosphatic fertilisers are presented in Table 3.

Table 3. Some common phosphate compounds in fertiliser and soil (adapted from Chien et al. 1989)
Compound Formula Common name
 (acronym)

Monocalcium Ca[([H.sub.2][PO.sub.4]).sub.2]. Superphosphate
 phosphate [H.sub.2]O (SSP/DSP/TSP)
Monoammonium [NH.sub.4][H.sub.2][PO.sub.4] MAP
 phosphate
Diammonium [([NH.sub.4]).sub.2][HP0.sub.4] DAP
 phosphate
Dicalcium Ca[HPO.sub.4] DCP (anhydrous)
 phosphate Ca[HPO.sub.4].2[H.sub.2]O DCP
Hydroxyapatite [Ca.sub.10][([PO.sub.4]).sub.6] Rock phosphate
 [(OH).sub.2] (RP)

Compound Phosphorus Water
 conc. (%) solubility

Monocalcium Single 9 High
 phosphate Double 18
 Triple 21
Monoammonium 23 High
 phosphate
Diammonium 20 High
 phosphate
Dicalcium Variable Low
 phosphate
Hydroxyapatite Variable Low


Phosphatic fertilisers can be broadly classified as being (i) water-soluble, (ii) partially water-soluble, and (iii) water-insoluble (Chien et al. 1989). When rainfall occurs immediately after application and surface runoff results, fertiliser solubility is important in determining phosphorus losses (Fig. 2). Phosphorus forms that are water-soluble release a large proportion of their phosphorus immediately upon wetting, although a significant portion of that phosphorus may subsequently revert to less soluble forms (Lindsay and Stephenson 1959a, 1959b). Insoluble fertilisers and reaction products may still be transported in surface runoff if erosion occurs or if they have reacted with other soil constituents to form water-soluble compounds.

[Figure 2 ILLUSTRATION OMITTED]

The most common water-soluble phosphatic fertilisers are single-, double-, and triple-superphosphate and mono- and di-ammonium phosphates. Superphosphates are formed by the reaction of an acid with rock phosphate. Where sulfuric acid is used the resulting fertiliser is a combination of monocalcium phosphate and gypsum (Ca[SO.sub.4]), commonly referred to as single-superphosphate (SSP). The gypsum in SSP affects the physical properties of the granules and provides sulfur, another essential nutrient for plants. If phosphoric acid rather than sulfuric acid is reacted with the rock phosphate, a more concentrated monocalcium phosphate fertiliser, commonly called triple-superphosphate, is produced.

Ammonium phosphates are produced by reacting ammonia with phosphoric acid. The products formed depend on the molar ratios of reactants but are commonly mono- and di-ammonium phosphate. Ammonium phosphates are almost completely water-soluble. However, to decrease dust formation the compounds are commonly coated with oils or waxes (I. Grant pers. comm.). The effect of such coatings on phosphorus loss has not been investigated.

There are a number of partially water-soluble phosphorus fertilisers and their chemical properties vary considerably (Chien et al. 1989). Some in this group of fertilisers are made by reacting anhydrous or aqueous ammonia with single- or triple-superphosphate, nitric acid with rock phosphate, or rock phosphate with insufficient sulfuric acid to form SSP. By mixing water-soluble compounds such as SSP with insoluble compounds such as rock phosphate, partially water-soluble compounds are formed. Water-insoluble phosphatic fertilisers include unreacted rock phosphate, heat treated (calcined) rock phosphate, and dicalcium phosphate formed by reacting hydrochloric acid with rock phosphate.

Single superphosphate is the most common phosphatic fertiliser applied to pasture soils in Australia (Anon 1997b). In recent years high analysis fertilisers, particularly di-ammonium phosphate, mono-ammonium phosphate, and double-and triple-superphosphate, have comprised a larger percentage of phosphatic fertilisers used (Anon 1997b). This review will focus mainly on the reactions of SSP and di-ammonium phosphates (DAP), which represent the extremes of mono-calcium phosphate and ammonium orthophosphate based fertilisers.

A conceptual model of fertiliser assimilation into soils is provided in Fig. 3. Phosphorus behaviour in soils may be explained by a series of precipitation and adsorption reactions (Holford 1989), which are affected by a range of properties including salt concentration and pH (Barrow 1989). Precipitation/dissolution reactions dominate when there is a large change in phosphorus concentration, when the cation concentration is high, and when soil pH is low or high, for example in the immediate vicinity of a fertiliser granule. Adsorption/desorption processes dominate when phosphorus concentration changes are small, solution cation concentrations are low, and where micro-surfaces are large, for example in clay soil. With minor exceptions (Schreiber and McDowell 1985), few authors have considered the potential effects of these reactions in terms of nutrient losses from fertilisers or their products.

[Figure 3 ILLUSTRATION OMITTED]

When fertilisers are first applied to soil they take up water. Surface-applied SSP is initially wetted by direct rainfall, by capillary uptake of water from the soil into the fertiliser granule, and by vapour transfer from the soil or atmosphere due to the hygroscopic nature of the mono-calcium phosphate (Williams 1969). Similar processes would be expected to occur for DAP except that the absence of the calcium sulfate carrier present in SSP would restrict the opportunities for moisture uptake by capillarity. The high solubility of DAP would tend to enhance water movement to the DAP granule once the wetting process has commenced, by establishing a high osmotic gradient in the soil. The wetting of granules in irrigated systems is probably similar to that occurring with rainfall.

Factors likely to affect water uptake from the atmosphere by fertiliser granules include relative humidity, temperature, and physical properties of the particles themselves such as size, shape, and porosity. For hygroscopic uptake of water by SSP, a relative humidity [is greater than] 90% is necessary (Williams 1969). DAP may require a similar humidity which is exceeded at even low soil moisture contents (Payne 1988).

Coatings applied to fertiliser granules may also affect moisture uptake and phosphorus dissolution. One reason for adding coatings is to slow the release of phosphorus, thereby increasing its availability for crop growth (Engelstad and Terman 1980). For example, Terman et al. (1970) found that a sulfur-coated phosphatic fertiliser was unavailable to a first crop of rice, but after degradation, it became available for a second crop. Similarly, Allen and Mays (1971) observed that insufficient P was released from sulfur-coated DAP for early growth of forage sorghum and resulted in lower total yields than uncoated DAP. It would appear likely that most, if not all, fertiliser coatings would impede moisture uptake, thereby delaying the release of phosphorus.

The main components of superphosphate are mono-calcium phosphate (MCP) and calcium sulfate. MCP dissolution occurs as water is absorbed into the fertiliser granule (Lehr et al. 1959). The incongruent dissolution of MCP is shown by the following partially balanced equation (Eqn 1) (Lindsay and Stephenson 1959a, 1959b; Lindsay et al. 1962b) and Eqn 2 and 3, where MTPS is metastable triple-point solution:

(1) Ca[([H.sub.2][PO.sub.4].sub.2] [multiplied by][H.sub.2]O ?? Ca[HPO.sub.4] + [H.sub.3][PO.sub.4] + [H.sub2]O

(2) (Ca[H.sub.2][PO.sub.4].sub.2) [multiplied by] [H.sub.2]O ?? [CaHPO.sub.4] [multiplied by] [2H.sub.2]O + MTPS

(3) (Ca[H.sub.2][PO.sub.4].sub.2) [multiplied by] [H.sub.2]O ?? [xH.sub.4]O ?? Ca[HPO.sub.4] + TPS

The solution that forms within a superphosphate granule is supersaturated, leading to the precipitation of sparingly soluble di-calcium phosphate dihydrate (DCPD). Depending on the calcium activity, 20-34% of the total phosphorus may be precipitated as DCPD at the granule site (Lehr et al. 1959). The movement of water-soluble phosphorus from SSP granules is virtually complete within 24 h (Lawton and Vomocil 1954).

Surface deposits of DCPD at the granule site have been decreased by 90% through the incorporation of other compounds, such as ammonium sulfate, with MCP (Bouldin et al. 1960). How this may occur is not clear, but perhaps part of the answer lies in the use of reagent-grade MCP in the experiments. Calcium from the soil would be needed for precipitation of DCPD. The sulfate component of the mixture would tend to precipitate calcium sulfate in the soil, establishing a competing reaction with the DCPD and encouraging DCPD precipitation away from the granule site. It is unlikely that commercial SSP with a calcium sulfate carrier would have responded the same. However, it would be interesting to know, for example, how the precipitation of DCPD at the soil surface would have been affected if organic matter had decreased the calcium activity in solution by complexation.

The highly acidic solution (pH 1.5) diffusing from a superphosphate granule dissolves soil minerals, creating a concentrated solution of phosphate, calcium, sulfate, iron, aluminium, and other ions (Lindsay and Stephenson 1959a, 1959b). DCPD initially forms, but as vapour transfer and mass flow of water extend the wetted zone, a series of adsorption and precipitation reactions decrease the phosphorus concentration. The importance of these individual fixation reactions depends on, amongst other things, the relative proportions of the cations in the soil (Sample et al. 1980). It is generally accepted that in acidic soils, such as those prevalent in the grazing regions of Australia, the formation of aluminium and iron phosphates are important fixation reactions (Engelstad and Hellums 1992).

The fixation of ammonium phosphates in the soil is likely to be affected by the solution pH of these compounds, 3 [multiplied by] 5 for MAP and 8.0 for DAP (Lindsay et al. 1962a). MAP forms reaction products that are initially similar to MCP in soils, DCPD unless exchangeable magnesium is high. In contrast, DAP forms a range of compounds under the same conditions, including calcium diammonium diphosphate monohydrate (Ca[([NH.sub.4]).sub.2][([HPO.sub.4]).sub.2] [multiplied by] [H.sub.2]O), octacalcium hexaphosphate pentahydrate ([Ca.sub.8][H.sub.2][([PO.sub.4]).sub.6] [multiplied by] 5[H.sub.2]O), calcium phosphate dihydrate (Ca[HPO.sub.4] [multiplied by] 2[H.sub.2]0), and magnesium ammonium phosphate hexahydrate (Mg[NH.sub.4][PO.sub.4] [multiplied by] 6[H.sub.2]O) (Bell and Black 1970).

The reaction products from the application of DAP to soil have more recently been investigated by Sample et al. (1979) and Moody et al. (1995). Sample et al. (1979) showed that while DAP increased the pH near the application site, it decreased soil pH at the wetting front. This low pH zone, attributed to the hydrolysis of water and the precipitation of more acidic compounds than DAP such as MAP, appeared to dissolve soil components resulting in the precipitation of aluminium phosphates. In contrast Moody et al. (1995) showed precipitation reactions involving calcium and magnesium were more important in the fixation of fertiliser phosphorus.

These different results may in part reflect the experimental conditions under which the measurements were taken. In one case measurements were made 4 weeks after fertiliser application on soil from an unspecified depth, at 19% soil water (Sample et al. 1979), while in the other case, measurements were made on predominantly topsoils 5 days after fertiliser application at 10 kPa matric suction (Moody et al. 1995). In the latter experiment, DAP was shown to increase dissolved organic carbon in soil water. The effects, if any, of these compounds on the activity and consequently the solubility of phosphate compounds are unknown.

Under field conditions, the prediction of how much phosphorus will be lost from fertiliser in runoff is extremely difficult. The rate of phosphorus release from a fertiliser granule and its subsequent reactions depends on both fertiliser and soil properties. There appear to be only a few chemical pathways by which fertiliser phosphorus is assimilated so we can, in a general sense, predict the types of reaction products. But a prediction of the relative proportions of these products and their subsequent reactions is extremely difficult. Many reaction products are metastable and gradually transform into more stable and insoluble compounds. Others dissolve incongruently with the release of soluble phosphorus, concomitant with the formation of a less soluble phase.

In most acidic soils, DCP slowly decomposes. Taranakites and amorphous iron and aluminium phosphates change to increasingly insoluble strengite-like and variscite-like crystalline compounds (Holford 1989). Phosphate, once adsorbed to the surface of crystals, may penetrate the lattice from where it cannot be easily remobilised by water (Barrow 1989).

In neutral to alkaline soils, DCP dissolution is much slower, allowing formation of octacalcium phosphate (OCP) and possibly colloidal hydroxyapatite. In some very calcareous soils, OCP has been observed to form directly from MCP, while in other soils varying in pH, DCP has persisted unchanged for 26 months (Holford 1989).

Liming of soil often increases phosphorus fixation, and may result in an improved residual value of phosphorus (Barrow 1980). The lower solubility of DCP under more alkaline conditions may result in its transformation to octacalcium phosphate (OCP) and possibly to colloidal hydroxyapatite (Holford 1989).

Soil is a complex medium where competing equilibria control inorganic chemical and adsorption reactions. When the effects of biological organisms and organic compounds and the constantly changing physical environment (Robinson 1942) are also included, we have a system that we are only starting to understand.

Phosphorus in surface runoff

The forms of phosphorus in surface runoff have received little examination in the literature but are important for two reasons. Firstly they provide an indication of the phosphorus availability in receiving waters (Sharpley and Syers 1979), and secondly they indicate the likely sources of phosphorus and the mechanisms by which it may be mobilised. Phosphorus in runoff is commonly analysed on the basis of filtration characteristics and reactivity in an acid-molybdate solution (Fig. 4). Phosphorus that passes through a 0.45-[micro]m filter and is reactive in an ascorbic acid-molybdate medium, dissolved reactive phosphorus (DRP), is the most bio-available form (Bostrom et al. 1988; Jansson 1988). However, total phosphorus (TP) is probably the most common measure of phosphorus in runoff and could be considered the total amount of phosphorus available to algae over the long term (Ryden et al. 1973; Bostrom et al. 1988).

[Figure 4 ILLUSTRATION OMITTED]

In DRP measurement the colour formation relies on the presence of orthophosphate, hence DRP is often equated to inorganic phosphorus or orthophosphate. However, the DRP test has been shown to overestimate orthophosphate concentrations due to the hydrolysis of acid-labile phosphorus compounds (Burton 1973; Broberg and Pettersson 1988). Further, it fails to differentiate between orthophosphate attached to a carrier, be that a small particle (colloid) or a ligand, and orthophosphate in true solution, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] or its associated form relative to the solution pH.

The term `colloid' is a vague description, defined here as any particle that passes through a 0.45-/[micro]m filter. In the current context, of primary interest are those entities which either chemically or physically sorb orthophosphate. They could include clays, metals, organometallic complexes, and organics. Orthophosphate associated with any of these moieties may well be detected by the phosphomolybdate method (Burton 1973). However, the short-term bioavailability of these types of associated phosphate is unknown.

Hydrolysis of labile organic phosphorus compounds has long been known to cause overestimation of the orthophosphate concentration of waters (Broberg and Pettersson 1988). The formation and stability of the phosphomolybdate complex requires a highly acidic solution, pH approximately 0.4. At this pH an unknown portion of organic phosphate, which is known to contribute a large fraction of the phosphorus pool in soils (Chardon et al. 1997), may be hydrolysed and measured as DRP. The importance of this analytical problem cannot be over emphasised. The DRP concentration of runoff after DAP application may suggest large losses of orthophosphate from the fertiliser. In fact the DAP may have mobilised organic phosphorus, which was then hydrolysed to orthophosphate during analyses (Moody et al. 1995).

The proportion of TP as DRP is commonly measured in runoff studies (Sharpley et al. 1991; Fleming et al. 1997; Nash and Murdoch 1997) and indicates whether phosphorus may be lost through erosion or dissolution processes. Knowledge of the mechanism of phosphorus loss may enable better management decisions. For example, vegetative filter strips, commonly termed buffer strips (Hairsine and Grayson 1993; Hairsine 1996), are predominantly designed to physically remove particulate phosphorus from runoff and are of marginal value in removing phosphorus present as TDP/DRP (Dillaha et al. 1988; Grayson et al. 1994). It follows that since phosphorus in runoff from pasture is usually dominated by soluble forms (Sharpley et al. 1994; Nash and Murdoch 1997), buffer strips and similar mechanisms, while potentially useful for a range of other reasons, will not significantly reduce phosphorus losses from grazing systems.

Management to minimise phosphorus loss

The most important factor determining the load of phosphorus lost in surface runoff is the volume of water leaving the pasture, rather than the concentration of nutrients in that water (Burwell et al. 1975; Haygarth 1997). Losses may be exacerbated by high soil water potentials prior to rainfall, the short-term management history of the pasture (Haygarth and Jarvis 1996a), soil type, fertiliser formulation and time of application, pasture cover, temperature, and humidity, but in most studies, water volume is the most important factor (Nash et al. 1998).

In New Zealand, phosphorus lost in runoff from a hill pasture has been compared with that from a forested catchment (McColl et al. 1977). The pasture tended to lose more phosphorus than the forested catchment in low flow conditions, but differences were small and not always significant. However, during large storms, the hill pasture lost approximately 3 times as much reactive phosphorus and 2-5 times as much total phosphorus. Over the 2-year study an estimated 47-70% of phosphorus losses occurred in infrequent, large events, with 31% of phosphorus being lost in a single storm. Less than 20% of phosphorus losses occurred at low flows. Similar results have been obtained from Australian dairy pastures where a single storm system accounted for 69% of the annual phosphorus loss (Nash and Murdoch 1996a), and elsewhere (Syers 1974; McColl 1979; Lowrance et al. 1984).

Unfortunately, it is difficult to decrease greatly the amount of surface runoff from grazing systems, especially in the larger storms. And even if it was possible to decrease surface runoff there would probably be undesirable consequences such as increased pugging of wet soils, rising watertables, and changed environmental flows in the very streams we aim to protect. Consequently a more suitable approach could be to decrease the concentration of phosphorus in runoff and thereby reduce its impact on receiving waters.

If phosphorus cannot be physically removed from surface runoff, the best way to decrease the phosphorus concentration is by minimising its availability at the source. Phosphorus is mobilised from the top few millimetres of soil. It is tempting to opt for a simple decrease in the phosphorus concentrations in that region of the profile as being the solution to our problem. We could do this by using less phosphate fertiliser, but if we are correctly managing the application rates this would result in less pasture production. Alternatives include cultivation of the soil yearly and application of nutrients through subsurface injection. But such simplistic notions fail to recognise that we are dealing with a complex system. It is likely that cultivation would decrease soil structural stability, decreasing infiltration, increasing pugging, and increasing runoff volumes. Further, our inability to maintain groundcover at crucial times during the year would predispose the paddocks to erosion and increased phosphorus loss by that process and we may have to apply additional fertiliser to compensate for the higher buffering in the cultivated zone. In attempting to decrease phosphorus losses we may inadvertently increase them. Clearly we need to develop management practices which consider the grazing system, soils, plants, and animals together, rather than attempting to decrease phosphorus losses in isolation.

Perhaps one of the easiest ways of decreasing the concentration of phosphorus in runoff from a grazing system is to modify the timing of fertiliser applications. In simple terms, this means the application of fertiliser when the soil conditions are best for assimilation and as long before runoff occurs as possible. Currently many graziers in southern Australia apply fertiliser in late summer-early autumn when runoff is rare. Whether this practice decreases agronomic productivity compared with split applications is an important issue yet to be resolved. Related issues also yet to be resolved include nutrient imbalances occurring when a capital application of fertiliser aimed at increasing soil phosphorus reserves is applied in a single dose, and the relative effectiveness of nitrogen-based fertilisers applied with additional phosphorus as opposed to those applied without.

If there is a requirement to add phosphorus fertilisers when runoff might reasonably be expected, for example in late winter, we should optimise the conditions for phosphorus assimilation, remembering that the aim is to reduce phosphorus availability to water and not to plants. Field measurements suggest phosphorus fertilisers have a `half life' of approximately 10 days with respect to fertiliser loss and that timing of fertiliser applications is the most important factor influencing phosphorus losses in high-input grazing systems (Nash et al. 1998). As outlined previously, there are sound theoretical reasons to suspect that not all phosphorus fertilisers behave the same when applied to soil. The environmental attributes of two commercial phosphatic fertilisers have been compared in model studies (Barlow et al. 1998). While the variation between granules was relatively high, statistically significant differences between fertilisers were observed (see Fig. 2). These results, which have yet to be verified in field trials, suggest that by altering formulations significant decreases in phosphorus losses will occur. Further research is needed, not only to compare different formulations but also to find the optimum soil conditions for decreasing phosphorus losses. With reliable meteorological forecasts now provided 7 days in advance, by decreasing the half-life of fertiliser to 4 days and applying fertiliser at least 5 days before runoff, phosphorus losses could be more than halved, especially in high-input systems. Clearly more research is required into the environmental aspects of fertiliser management.

However, fertiliser management alone is unlikely to decrease phosphorus losses to acceptable levels. In southern Australian studies where fertilisers had been applied in autumn and would not have been expected to contribute directly to nutrient losses, total phosphorus concentrations in runoff were still 3-5 mg/L, well in excess of water quality targets (Environment Protection Authority 1995). The determination of management strategies to deal with phosphorus lost from the soil/plant system rather than directly from fertiliser is more difficult, primarily because we do not know the true source and mechanism by which it is mobilised. Stocking intensity may be altered to decrease losses if significant phosphorus is being mobilised from faeces or directly from grass. Were this to be used, the effects of animal traffic on soil physical properties, especially soil infiltration rate and the formation of hard pans, would need to be considered. We can only speculate on the effects of current management practices, such as pasture harrowing, that spread dung and loosen the soil surface, presumably increasing the decomposition of organic phosphorus sources.

Phosphorus lost directly from grass could also be addressed by breeding plants that are productive at lower soil phosphorus levels and by developing plants that lose less phosphorus to surface runoff. Plant breeding is clearly a long-term option that should be considered, as plants are currently bred for genetic potential rather than nutrient efficiency. A more efficient plant would require less fertiliser and result in an overall reduction in phosphorus cycling and consequent phosphorus loss. Rather than breeding new plants, perhaps there is a need to consider other pasture species and fodder production systems which use or require less phosphorus than our current ryegrass-clover-based systems. For example, we could use a pure grass sward as in Europe and supply nitrogen fertiliser as it is needed rather than depend on clover, which is known to be more reliant on higher phosphorus concentrations, due to its lower root mass, compared with ryegrass, and its different root morphology (Frame 1992).

Phosphorus mobilised from the soil surface may come from organic or inorganic sources. In the case of inorganic phosphorus, an appropriate management system may include a cropping rotation and pasture renovation every 10 years to decrease surface deposits of phosphorus. Such rotations using a summer cover crop of turnips and re-sowing permanent pasture are used on many farms at present. Organic sources of phosphorus at the surface can be decreased by less intrusive management practices. If the phosphorus that we are measuring as DRP in runoff requires a carrier, be that a colloid or a ligand, phosphorus losses could potentially be reduced by managing the carrier rather than the phosphorus concentration per se.

Before we set about the expensive process of developing different production systems, we clearly need to identify the relative contributions of different phosphorus sources on nutrient loss. Chemical tracing techniques may well be the tools which can achieve this within a tolerable time-frame.

Concluding discussion

It is unlikely any one management strategy will decrease phosphorus losses from grazing systems to an acceptable level, especially since the ultimate target should be zero loss (Environment Protection Authority 1995). Modifications to fertiliser formulations and management may be a good start. For example, by assuming fertiliser directly accounts for 40% of the phosphorus lost from grazing systems, it is possible that improved fertiliser management could decrease phosphorus losses by 20% or more. But what quantitative information do we really have on which to base such assessments?

By monitoring, it is shown that nutrient levels are rising in many streams draining agricultural catchments (Anon 1995). Unfortunately the spatial and temporal variation and in-stream processes make it difficult to attribute changes in nutrient loads to particular practices. It is easy to forget that while production processes have been studied since European colonisation, only in recent years has any real attention been focused on investigating nutrient loss processes at a farm scale. Clearly, as our fundamental understanding of phosphorus loss improves so will our ability to predict quantitatively the effects of management changes. Importantly, the use of such process-based models and monitoring management practices at a farm scale will indicate the success or otherwise of nutrient reduction strategies long before it is apparent through stream monitoring.

Perhaps it is also time to question the efficiency of protecting water resources by setting water quality targets. It is unlikely that phosphorus losses increase linearly with productivity per unit area. On the contrary, as higher production requires more intrusive and intensive management systems, the opportunities for decreasing phosphorus losses increase. So, by doubling production there may be a slight overall increase in phosphorus loss per unit area but a decrease in phosphorus loss per unit of production. It follows that by setting water quality targets for streams in a catchment, the most productive enterprises may be penalised. A better system may be to set production targets for catchments and maximise environmental productive efficiency.

Acknowledgments

The authors would like to thank Mr Peter Moate and Dr Nick Uren for reviewing manuscript drafts, and Pivot Limited and the Victorian Sustainable Agricultural Strategy for financial support.

References

Ahuja, L. R. (1986). Characterization and modeling of chemical transfer to runoff. Advances in Soil Science 4, 149-88.

Allen, S. E., and Mays, D. A. (1971). Sulfur-coated fertilisers for controlled release: agronomic evaluation. Journal of Agricultural Food and Chemistry 19, 809-12.

Anon. (1995). Algal bloom & nutrient status of Victorian inland waters. Status. National Library of Australia, Melbourne.

Anon. (1996). Blue-green algae and nutrients in Victoria. A resource handbook. National Library of Australia, Melbourne.

Anon. (1997a). Phosphorus for dairy farms: Project results year 2. Agriculture Victoria Ellinbank, Department of Natural Resources and Environment, Melbourne.

Anon. (1997b). Pivot Limited 78th Annual Report and Statement of Accounts 1996-97. Pivot Limited, Melbourne.

Armstrong, A. C., and Harris, G. L. (1996). Movement of water and solutes from agricultural land: The effects of artificial drainage. In `Advances in Hillslope Processes'. (Eds A. C. Armstrong and G. L. Harris.) pp. 187-211. (John Wiley: Chichester, UK.)

Barlow, K., Halliwell, D., and Nash, D. (1998). Control of relative humidity and soil water content, for phosphorus fertiliser runoff simulations. In `National Soils Conference, Environmental Benefits of Soil Management'. (Ed. P. Mulvey.) pp. 439-42. (Australian Society of Soil Science Inc.: Brisbane.)

Barrow, J. (1989). Surface reactions of phosphate in soil. Agricultural Science 2, 33-8.

Barrow, N. J. (1975). Chemical forms of inorganic phosphate in sheep faeces. Australian Journal of Soil Research 13, 63-7.

Barrow, N. J. (1980). Evaluation and utilisation of residual phosphorus in soils. In `The Role of Phosphorus in Agriculture'. (Eds F. E. Khasawneh, E. C. Sample, and E. J. Kamprath.) pp. 333-59. (American Society of Agronomy: Madison, WI.)

Bell, L. C., and Black, C. A. (1970). Crystalline phosphates produced by interaction of orthophosphate fertilisers with slightly acid and alkaline soils. Soil Science Society of America Journal 34, 735-40.

Bloom, P. R., McBride, M. B., and Weaver, R. M. (1979). Aluminium organic matter in acid soils: buffering and solution aluminium activity. Soil Science Society of America Journal 43, 488-93.

Bostrom, B., Persson, G., and Broberg, B. (1988). Bioavailability of different phosphorus forms in freshwater systems. Hydrobiologia 170, 133-55.

Bouldin, D. R., Lehr, J. R., and Sample, E. C. (1960). The effect of associated salts on transformations of monocalcium phosphate monohydrate at the site of application. Soil Science Society of America Journal 24, 464-8.

Braithwaite, G. D. (1976). Calcium and phosphorus metabolism in ruminants with special reference to parturient paresis. Journal of Dairy Research 43, 501-20.

Broberg, O., and Pettersson, K. (1988). Analytical determination of orthophosphate in water. Hydrobiologia 170, 45-59.

Bromfield, S. M., and Jones, O. L. (1972). The initial leaching of hayed-off pasture plants in relation to the recycling of phosphorus. Australian Journal of Agriculture 23, 811-24. Burton, J. D. (1973). Problems in the analysis of phosphorus compounds. Water Research 7, 291-307.

Burwell, R. E., Timmons, D. R., and Holt, R. F. (1975). Nutrient transport in surface runoff as influenced by soil cover and seasonal periods. Soil Science Society of America Journal 39, 523-28.

Cathcart, J. B. (1980). The phosphate industry in the United States. In `The Role of Phosphorus in Agriculture'. (Eds F. E. Khasawneh, E. C. Sample, and E. J. Kamprath.) pp. 19-41. (American Society of Agronomy: Madison, WI.)

Chardon, W. J., and Oenema, O. (1995). Leaching of organically bound phosphorus. In `Annual Report 1995'. pp. 39-42. (DLO Research Institute for Agrobiology and Soil Fertility (AB-DLO): Wageningen.)

Chardon, W. J., Oenema, O., del Castilho, P., Vriesema, R., Japenga, J., and Blaauw, D. (1997). Organic phosphorus in solutions and leachates from soils treated with animal slurries. Journal of Environmental Quality 26, 372-8.

Chien, S. H., Sale, P. W. G., and Hammond, L. L. (1989). Comparison of the effectiveness of phosphorus fertiliser products. In `Phosphorus Requirements for Sustainable Agriculture in Asia and Oceana'. pp. 143-56. (International Rice Research Institute: Manila.) Cottingham, P., Bennison, G., Dunn, R., Lidston, J., and Robinson, D. (1995). `Algal Bloom and Nutrient Status of Victorian Inland Waters.' (Government of Victoria: Melbourne.) Cullen, P. (1991). Land use and declining water quality. Australian Journal of Soil and Water Conservation 4, 4-8.

Dickinson, C. H., and Craig, G. (1990). Effects of water on the decomposition and release of nutrients from cow pats. New Phytology 115, 139-47.

Dillaha, T. A., Sherrard, J. H., Lee, D., Mostaghimi, S., and Shanholtz, V. O. (1988). Evaluation of vegetative filter strips as a best management practice for feed lots. Journal of the Water Pollution Control Federation 60, 1231-8.

Dillon, P. J., and Kirchner, W. B. (1974). The effects of geology and land use on the export of phosphorus from watersheds. Water Research 9, 135-48.

Dils, R., and Heathwaite, A. L. (1996). Phosphorus fractionation in hillslope hydrological pathways contributing to agricultural runoff. In `Advances in Hillslope Processes'. (Eds M. G. Anderson and S. M. Brookes.) pp. 229-51. (John Wiley: Chichester, UK.)

Drysdale, G. (1998). Dairy farm performance analysis 1996-97. Annual Report, DNRE, Melbourne.

Edwards, D. R., Daniel, T. C., and Moore, P. A. Jr (1996). Quality of runoff from pasture fields treated with organic and inorganic fertilisers. In `The International Conference on Agricultural Engineering'. (Ed. U. P. D. Madrid.) (European Society of Agricultural Engineers: Madrid.)

Engelstad, O. P., and Hellums, D. T. (1992). Water solubility aspects of phosphate fertilisers: agronomic aspects--A literature review. International Fertiliser Development Center (IFDC), Alabama.

Engelstad, O. P., and Terman, G. L. (1980). Agronomic effectiveness of phosphate fertilisers. In `The Role of Phosphorus in Agriculture'. (Eds F. E. Khasawneh, E. C. Sample, and E. J. Kamprath.) pp. 311-32. (American Society of Agronomy: Madison, WI.)

Environment Protection Authority. (1995). Protecting water quality in Central Gippsland. Schedule F5--Waters of the Latrobe and Thomson River Basins and Merriman Creek, Catchment and Draft Policy Impact. Environment Protection Authority, Melbourne.

Fleming, N. K., Cox, J. W., and Chittleborough, D. J. (1997). Pathways, loads, and forms of phosphorus in runoff from adjacent subcatchments on a dairy farm at Flaxley, South Australia. CRC for Soil and Land Management, Adelaide.

Frame, J. (1992). `Improved Grasslands Management.' (Framing Press: UK.)

Gburek, W. J., Sharpley, A. N., and Pionke, H. B. (1996). Identification of critical source areas for phosphorus export from agricultural catchments. In `Identification of Critical Source Areas for Phosphorus Export from Agricultural Catchments'. (Eds M. G. Anderson and S. M. Brookes.) pp. 263-82. (John Wiley: Chichester, UK.)

Grace, N. D. (1983). The mineral requirements of grazing ruminants. New Zealand Society of Animal Production, Publication No. 9. pp. 106-11.

Grayson, R., Gippel, C. J., Finlayson, B. L., Hart, B. T., Hawken, R., McMahon, T. A., and Tillead, J. (1994). Mitigation of sediment and nutrient movement from sources in the Latrobe River catchment into the Gippsland Lakes. Report No. UME 8. Land and Water Resources Research and Development Corporation, Canberra.

Greenhill, N. B., Peverill, K. I., and Douglas, L. A. (1983a). Nutrient concentrations in runoff from pasture in Westernport, Victoria. Australian Journal of Soil Research 21, 139-45. Greenhill, N. B., Peverill, K. I., and Douglas, L. A. (1983b). Nutrient loads in surface runoff from sloping perenial pastures in Victoria, Australia. Australian Journal of Soil Research 26, 503-6.

Gregg, P. E. H., Hedley, M. J., Cooper, A. B., and Cooke, J. (1993). Transfer of nutrients from grazed pastures to aquatic surface waters and their impact on water quality. In `Proceedings of the New Zealand Fertiliser Manufacturers Association Conference'. pp. 64-75.

Gregory, D. I. (1992). Global structure of the phosphate fertiliser industry. In `Phosphate Fertilisers and the Environment'. (Ed. J. J. Schultz.) pp. 13-26. (International Fertiliser Development Center: Tampa, FL.)

Hairsine, P. (1996). Comparing grass filter strips and near natural riparian forests for buffering intense hillslope sediment sources. Conference on Stream Management in Australia, Merrijig, 19-23 February.

Hairsine, P. B., and Grayson, R. B. (1993). Further perspectives on buffer strips: surface transport and deposition processes. Land and Water Resources Research and Development Corporation, Australia, Occasional Paper 01/93, pp. 45-52.

Havis, R. N., and Alberts, E. E. (1993). Nutrient leaching from field-decomposed corn and soybean residue under simulated rainfall. Soil Science Society of America Journal 57, 211-18.

Haygarth, P. (1997). Agriculture as a source of phosphorus transfer to water: sources and pathways. Scientific Committee on Phosphates in Europe Newsletter 21, 1-15.

Haygarth, P. M., and Jarvis, S. C. (1996a). Soil derived phosphorus in surface runoff from grazed grassland lysimeters. Water Research 31, 140-8.

Haygarth, P. M., and Jarvis, S. C. (1996b). Pathways and forms of phosphorus losses from grazed grassland hillslopes. In `Advances in Hillslope Processes'. (Eds M. G. Anderson and S. M. Brookes.) pp. 283-94. (John Wiley: Chichester, UK.)

Haynes, R. J. (1984). Lime and phosphorus in the soil-plant system. Advances in Agronomy 37, 249-315.

Hazel, P. (1991). Draft report--Mt Lofty Ranges dairy farm water quality monitoring program 1987-1989. Engineering and Water Supply Department of South Australia, Adelaide.

Heathwaite, A.L. (1993). The impact of agriculture on dissolved nitrogen and phosphorus cycling in temperate ecosystems. Chemistry and Ecology 8, 217-31.

Heckrath, G., Brookes, P. C., Poulton, P. R., and Goulding, K. W. T. (1995). Phosphorus leaching from soils containing different phosphorus concentrations in the Broadbalk Experiment. Journal of Environmental Quality 24, 904-10.

Heng, L.K., White, R.E., Bolan, N.S., and Scotter, D.R. (1991). Leaching losses of major nutrients from a mole-drained soil under pasture. New Zealand Journal of Agricultural Research 34, 325-34.

Holford, I. (1989). Phosphate behaviour in soils. Agricultural Science 2, 15-21. Holford, I. C. R. (1997). Soil phosphorus: its measurement and its uptake by plants. Australian Journal of Soil Research 35, 227-39.

Jansson, M. (1988). Phosphate uptake and utilisation by bacteria and algae. Hydrobiologia 179, 177-89.

Johnston, A. E., and Poulton, P. R. (1992). The role of phosphorus in crop production and soil fertility: 150 years of field experiments at Rothamsted, United Kingdom. In `Phosphate Fertilizer and the Environment'. (Ed. J. J. Schultz.) pp. 45-63. (International Fertiliser Development Center, Phosphate Fertiliser and the Environment: Tampa, FL.)

Jones, O. L., and Bromfield, S. M. (1969). Phosphorus changes during leaching and decomposition of hayed off pasture plants. Australian Journal of Agricultural Research 29, 653-63.

Jordan, C., and Smith, R.V. (1985). Factors affecting leaching of nutrients from an intensively managed grassland in county Antrim, Northern Ireland. Journal of Environmental Management 29, 1-15.

Kirkby, C. A., Smythe, L. J., Cox, J. W., and Chittleborough, D. J. (1997). Phosphorus movement down a toposequence from a landscape with texture contrast soils. Australian Journal of Soil Research 35, 399-417.

Kofoed, A. D. (1984). Pathways of nitrate and phosphate to ground and surface waters. In `Environment and Chemicals in Agriculture'. (Ed. F. P. W. Winteringham.) pp. 27-69. (Elsevier: London.)

Lawton, K., and Vomocil, J. A. (1954). The dissolution and migration of phosphorus from granular superphosphate in some Michigan soils. American Soil Science Society Journal 23, 26-32.

Lehr, J. R., Brown, W. E., and Brown, E. H. (1959). Chemical behaviour of monocalcium phosphate monohydrate in soils. American Soil Science Society Journal 23, 3-7.

Lindsay, W. L., Frazier, A. W., and Stephenson, H. F. (1962a). Identification of reaction products from phosphate fertilisers in soils. Soil Science Society of America Journal 26, 446-52.

Lindsay, W. L., Frazier, A. W., and Stephenson, H. F. (1962b). Nature and reactions of mono-calcium phosphate monohydrate in soils: III Studies with metastable triple point solutions. American Soil Science Society Journal 23, 342-45.

Lindsay, W. L., and Stephenson, H. F. (1959a). Nature of reactions of monocalcium phosphate monohydrate in soils. I. The solution that reacts with the soil. Soil Science Society of America Journal 23, 12-7.

Lindsay, W. L., and Stephenson, H. F. (1959b). Nature of the reactions of monocalcium phosphate monohydrate in soils: II. Dissolution and precipitation reactions involving iron, aluminium, manganese, and calcium. Soil Science Society of America Journal 23, 18-22.

Lowrance, R. R., Todd, R. L., and Asmussen, L. E. (1984). Nutrient cycling in an agricultural watershed: II. Streamflow and artificial drainage. Journal of Environmental Quality 13, 27-32.

Lutz, J. F., and Haque, I. (1975). Effects of phosphorus on some physical and chemical properties of clays. American Soil Science Society Proceedings 39, 33-6.

McColl, R. H. S. (1979). Factors affecting downslope movement of nutrients in hill pasture. Progress in Water Technology 11, 271-85.

McColl, R. H. S., White, E., and Gibson, A. R. (1977). Phosphorus and nitrate run-off in hill pasture and forest catchments, Taita, New Zealand. New Zealand Journal of Marine and Freshwater Research 11, 729-44.

Magesan, G. N., White, R. E., and Scotter, D. R. (1996). Nitrate leaching from a drained, sheep-grazed pasture. I. Experimental results and environmental implications. Australian Journal of Soil Research 34, 55-67.

Mays, D. A., Wilkinson, S. R., and Cole, C. V. (1980). Phosphorus nutrition of forages. In `The Role of Phosphorus in Agriculture'. (Eds F. E. Khasawneh, E. C. Sample, and E. J. Kamprath.) pp. 805-46. (American Society of Agronomy: Madison, WI.)

Moody, P. W., Yo, S. A., Edwards, D. G., and Bell, L. C. (1995). Effect of banded fertilisers on soil solution composition and short-term root growth. III. Monocalcium phosphate without gypsum. Australian Journal of Soil Research 33, 899-914.

Nash, D., and Murdoch, C. (1996a). Phosphorus in runoff from a highly fertile dairy pasture. In `Australian and New Zealand National Soils Conference 1996. Soil Science--Raising the Profile'. pp. 189-90. (Australian Society of Soil Science Inc.: Parkville, Vic.)

Nash, D., and Murdoch, C. (1996b). Phosphorus leaching from cattle dung and fertiliser in a krasnozem. In `Australian and New Zealand National Soil Conference 1996. Soil Science--Raising the Profile'. pp. 187-8. (Australian Society of Soil Science Inc.: Parkville, Vic.)

Nash, D., and Murdoch, C. (1997). Phosphorus in runoff from a fertile dairy pasture. Australian Journal of Soil Research 35, 419-29.

Nash, D., Hannah, M., Halliwell, D., and Murdoch, C. (1998). Phosphorus in runoff from a pasture based grazing system. In `National Soils Conference, Environmental Benefits of Soil Management'. (Ed. P. Mulvey.) pp. 447-9. (Australian Society for Soil Science Inc.: Brisbane.)

Nelson, P. N., Cotsaris, E., and Oades, M. J. (1996). Nitrogen, phosphorus, and organic carbon in streams draining two grazed catchments. Journal of Environmental Quality 25, 1221-9.

Nelson, P. N., Cotsaris, E., Oades, J. M., and Bursill, D. B. (1991). Organic carbon in water, its sources and role in nutrient transport. AWRAC Report No. 86/66, Australian Centre for Water Treatment and Water Quality Research, Glen Osmond, South Australia.

Nexhip, K. J., Mundy, G. N., Collins, M. D., and Austin, N. R. (1997). Development of nutrient water quality targets for irrigated pasture sub-catchments. Rep. No. 15037. Institute of Sustainable Irrigated Agriculture, Tatura.

Norrish, K., and Rosser, H. (1983). Mineral phosphate. In `Soils: An Australian Viewpoint'. pp. 335-61. (CSIRO Division of Soils: Melbourne.)

Payne, D. (1988). The behaviour of water in soil. In `Russell's Soil Conditions and Plant Growth'. (Ed. A. Wild.) pp. 315-37. (Longman Scientific and Technical: Avon.)

Perrott, K. W., and Sarathchandra, S. U. (1989). Phosphorus in the microbial biomass of New Zealand soils under established pasture. New Zealand Journal of Agricultural Research 32, 409-13.

Perrott, K. W., Sarathchandra, S. U., and Waller, J. E. (1990). Seasonal storage and release of phosphorus and potassium by organic matter and the microbial biomass in a high producing pastoral soil. Australian Journal of Soil Research 28, 593-608.

Pilgram, J. J. (1986). `Issues in the Management of Australia's Water Resources.' (Cheshire: Melbourne.)

Rayment, G. E., and Higginson, F. R. (1992). `Australian Laboratory Handbook of Soil and Water Chemical Methods.' (Inkata Press: Melbourne.)

Reddy, G. Y., McLean, E. O., Hoyt, G. D., and Logan, T. L. (1978). Effects of soil, cover crop, and nutrient source on amounts and forms of phosphorus movement under simulated rainfall conditions. Journal of Environmental Quality 7, 50-4.

Rhoades, J. D., and Oster, J. D. (1986). Solute content. In `Methods of Soil Analysis: Part 1, Physical and Mineralogical Methods'. (Ed A. Klute.) pp. 985- 1006. (American Society of Agronomy: Madison, WI.)

Richardson, A. E. (1994). `Soil Microorganisms and Phosphorus Availability.' (CSIRO Publications: Melbourne.)

Robinson, R. R. (1942). Phosphorus fixation as effected by soil temperature. Agronomy Journal 34, 301-6.

Rowarth, J. S. (1987). Phosphate cycling in grazed hill-country pasture. PhD Thesis, Massey University, Palmerston North.

Rowarth, J. S., Gillingham, A. G., Tillman, R. W., and Syers, J. K. (1988). Effects of season and fertiliser rate on phosphorus concentrations in pasture and sheep faeces in hill country. New Zealand Journal of Agricultural Research 31, 187-93.

Russell, E. J. (1957). `The World of Soil.' (Collins: London.)

Ryden, J. C., Syers, J. K., and Harris, R. F. (1973). Phosphorus in runoff and streams. Advances in Agronomy 25, 1-45.

Sale, P. (1992). Plant nutrition and nutrient deficiencies in grazed pasture. In `Facts on Fertiliser'. (Ed. G. S. James.) (Grasslands Society of Victoria.)

Sale, P. W. G., Simpson, P. G., Lewis, D. C., Gilkes, R. J., Bolland, M. A., Ratkowsky, D. A., Gilbert, M. A., Garden, D. L., Cayley, J. W. D., and Johnson, D. (1997). The agronomic effectiveness of reactive phosphate rocks. 1. Effect of the pasture environment. Australian Journal of Experimental Agriculture 37, 921-36.

Sample, E. C., Khasawneh, F. E., and Hashimoto, I. (1979). Reactions of ammonium ortho- and polyphosphate fertilisers in soil: III. Effects of associated cations. Soil Science Society of America Journal 43, 58-65.

Sample, E. C., Soper, R. J., and Racz, G. J. (1980). Reactions of phosphate fertilizers in soils. In `The Role of Phosphorus in Agriculture'. (Eds F. E. Khasawneh, E. C. Sample, and E. J. Kamprath.) pp. 263-310. (American Society of Agronomy: Madison, WI.)

Sanyal, S. K., and DeDatta, S. K. (1991). Chemistry of phosphorus transformations in soil. Advances in Soil Science 16, 1-120.

Schofield, R. K. (1955). Can a precise meaning be given to `available' soil phosphorus. Soils and Fertilisers 18, 373-5.

Schreiber, J. D. (1985). Leaching of nitrogen, phosphorus, and organic carbon from wheat straw residues: II. Loading rates. Journal of Environmental Quality 14, 256-60.

Schreiber, J. D., and McDowell, L. L. (1985). Leaching of nitrogen, phosphorus, and organic carbon from wheat straw residues. I. Rainfall intensity. Journal of Environmental Quality 14, 251-6.

Schulthess, C. P., and Sparks, D. L. (1991). Equilibrium-based modeling of chemical sorption on soils and soil constituents. Advances in Soil Science 16, 121-63.

Sharma, M. L., Ahmed, M., and Richards, Q. D. (1996). Reliability of suction cups for measuring soil water quality. In `Australian and New Zealand National Soils Conference 1996. Soil Science--Raising the Profile'. p. 239. (Australian Society of Soil Science Inc.: Parkville, Vic.)

Sharpley, A. N. (1985). Depth of surface soil-runoff interaction as affected by rainfall, soil slope, and management. Soil Science Society of America Journal 49, 1010-15.

Sharpley, A. N., Ahuja, L. R., and Menzel, R. G. (1981). The release of soil phosphorus in runoff in relation to the kinetics of desorption. Journal of Environmental Quality 10, 386-91.

Sharpley, A. N., Chapra, S. C., Wedepohl, R., Sims, J. T., Daniel, T. C., and Reddy, K. R. (1994). Managing agricultural phosphorus for protection of surface waters: Issues and options. Journal of Environmental Quality 23, 437-51.

Sharpley, A. N., and Smith, S. J. (1989). Mineralisation and leaching of phosphorus from soil incubated with surface-applied and incorporated crop residue. Journal of Environment Quality 18, 101-5.

Sharpley, A. N., and Smith, S. J. (1992). Prediction of bio-available phosphorus loss in agricultural runoff. Journal of Environmental Quality 21, 32-7.

Sharpley, A. N., Smith, S. J., and Menzel, R. G. (1982). Prediction of phosphorus losses in runoff from southern plains watersheds. Journal of Environmental Quality 11, 247-50.

Sharpley, A. N., Smith, S. J., and Naney, J. W. (1987). Environmental impact of agricultural nitrogen and phosphorus use. Journal of Agriculture and Food Chemistry 35, 812-17.

Sharpley, A. N., and Syers, J. K. (1979). Phosphorus inputs into a stream draining an agricultural watershed. Water, Air and Soil Pollution 11, 417-28.

Sharpley, A. N., Troeger, W. W., and Smith, S. J. (1991). The measurement of bio-available phosphorus in agricultural runoff. Journal of Environmental Quality 20, 235-38.

Slattery, W. J., Edwards, D. G., Bell, L. C., Coventry, D. R., and Helyar, K. (1998). Soil acidification and the carbon cycle in a cropping soil of north-eastern Victoria. Australian Journal of Soil Research 36, 273-90.

Small, D.R. (1985). Effects of irrigation land use in nutrients in runoff. Report No.78/99, Department of Resources and Energy, Canberra.

Smith (1965). Aluminium and iron phosphates in soil. Journal of the Australian Institute of Agricultural Science 31, 110-26.

Sonzogoni, W. C., Chesters, G., Jeffs, D. N., Konrad, J. C., Ostry, R. C., and Robinson, J. B. (1980). Pollution from land runoff. Environmental Science and Technology 14, 148-53.

Syers, J. K. (1974). Effect of phosphate fertilisers on agriculture and the environment. New Zealand Agricultural Science 8, 149-64.

Tabatabai, M. A., and Laflen, J. M. (1976). Nitrogen and sulfur content and pH of precipitation in Iowa. Journal of Environmental Quality 5, 108-12.

Terman, G. L., Allen, S. E., and Engelstad, O. P. (1970). Response by paddy rice to rates and sources of applied phosphorus. Agronomy Journal 62, 390-94.

Tham, S. H. (1983). Nutrient runoff from agricultural land. Report No. 159, Agriculture Victoria.

Thomas, G. W. (1975). The relationship between organic matter content and exchangeable aluminium in acid soil. American Soil Science Society Proceedings 35, 591.

Thompson, R. (1989). Manure--an asset or liability. In `Proceedings: Livestock Officers (Dairying) Conference, North Coast Agricultural Institute, Wollongbar, NSW. 12-14 September'. (Ed. D. Heptonstall.) pp. 79-85 (North Coast Agricultural Institute: Wollongbar, NSW.)

Timmons, D. R., Holt, R. F., and Latterall, J. J. (1970). Leaching of crop residues as a source of nutrients in surface runoff water. Water Resources Research 6, 1367-75.

Van Kauwenbergh, S. J. (1992). The global phosphate rock resource base--Technical and economic considerations. In `Phosphate Fertilisers and the Environment'. (Ed. J. J. Schultz.) pp. 27-39. (International Fertiliser Development Center: Tampa, FL.)

Wild, A. (1949). The retention of phosphate by soil: A review. Journal of Soil Science 1, 221-38.

Williams, C. H. (1969). Moisture uptake by surface-applied superphosphate and movement of the phosphate and sulphate into the soil. Australian Journal of Soil Research 7, 307-16.

Manuscript received 8 September 1998, accepted 11 January 1999

David M. Nash and David J. Halliwell

Agriculture Victoria Ellinbank, RMB 2460, Hazeldean Rd, Ellinbank, Vic. 3821, Australia.
COPYRIGHT 1999 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Nash, David M.; Halliwell, David J.
Publication:Australian Journal of Soil Research
Date:May 1, 1999
Words:12844
Previous Article:Hydraulic conductivity in soils irrigated with wastewaters of differing strengths: Field and laboratory studies.
Next Article:Metal-induced chromium(VI) sorption by two calcareous soils.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters