Review: A bioavailability-based rationale for controlling metal and metalloid contamination of agricultural land in Australia and New Zealand.Introduction
The accumulation of heavy metals and metalloids in soil, especially arsenic (As), cadmium (Cd), chromium (Cr), copper (Cu), lead (Pb), mercury (Hg), nickel (NJ), selenium (Se), and zinc (Zn), is of concern in agricultural production systems due to the potential threat of adversely affecting food quality (safety and marketability), crop growth (through phytotoxicity), or environmental health (soil flora/fauna and terrestrial animals). Half-lives are often quoted for persistence of metal pollution in the environment, but this is inappropriate nomenclature as, unlike organic pollutants (e.g. organic pesticides), metals and metalloids persist indefinitely. Thus, metal pollution of soil at any point in the landscape can only be reduced by transporting the pollutant elsewhere e.g. to areas downslope or downwind by erosion, to surface or groundwaters by leaching, to the atmosphere by volatilisation, to animals or humans by removal in agricultural crops, or to dedicated disposal areas by physical removal of soil or plants (phytoremediation).
In Australia and New Zealand, where agricultural exports are increasingly marketed internationally on the basis of freedom from impurities or contamination, the issue of managing and regulating metal contamination of agricultural soils has gained importance. Metal impurities enter the food chain through accession from the atmosphere, through fertiliser use, and because agricultural land is increasingly seen in many countries as a suitable receptor for wastewater, biosolids, and green wastes from urban areas. Internationally, the regulatory and management philosophies developed to control metal pollution of soil from these sources vary markedly. Recently in Australia and New Zealand, guidelines have been developed to control metal contamination of soil from sewage biosolids (NZDoH 1992; SA EPA 1996; NSW EPA 1997; NZWWA 1999), reviews and changes to maximum permitted concentrations for metals in foodstuffs have occurred (ANZFA 1997), and there has Ken the development of investigation threshold guidelines for concentrations of metals in soil based on human health and plant phytotoxicity criteria (Imray and Langley 1996; NSW EPA 1998; NEPC 1999). Many of these regulations and guidelines have been developed based principally on data from overseas. This review paper summarises the current situation in Australia and New Zealand with regard to management of metal contamination of agricultural soils. It also charts the derivation of the local regulatory framework and identifies where inconsistencies in the framework exist, and where local information is either available or still needed to produce a more meaningful and scientific basis to regulatory issues for metals in our soils. Finally, this paper provides recommendations for the derivation of more appropriate criteria for guidelines of metal contaminants in soil.
Australian and New Zealand climates and soils
In both Australia and New Zealand there are considerable variations in climate and soils, which will influence the ability of the land to cope with metal contamination. For example, even within a relatively small country like New Zealand, mean air temperatures range from 15 [degrees] C to [is less than] 7 [degrees] C and rainfall from [is greater than] 5000 mm on the western Southern Alps to [is less than] 500 mm in the Central Otago basin (Taylor and Pohlen 1968). Soils in New Zealand are likewise diverse, with all of the 11 soil orders of the US Soil Taxonomy represented (Hewitt 1997). Australia also has a considerable variation in soils and a much greater variation in climate than New Zealand, ranging from wet tropical to temperate to dry desert. Mean annual air temperatures range from 29 [degrees] C to 5 [degrees] C, and 50% of the country has a rainfall of [is less than] 300 mm per year and 80% [is less than] 600 mm (Taylor 1983). However, most of the main centres of population and surrounding areas are located in the wetter regions of Australia (rainfall [is greater than] 700 mm).
There are several features of soils in Australia and New Zealand that distinguish them from the bulk of the soils found in Europe and North America. Many of the soils of Australia are pedologically very old, are highly weathered, and have a very low level of soil fertility, particularly phosphorus (P) and micronutrients (Taylor 1983). Surface horizons of Australian soils generally have low organic matter contents and weak structure development, and subsoils often have much higher clay contents than surface horizons (Hubble et al. 1983). Many soils are light-textured, but wherever significant clay fractions are found, they are dominated by clays with low cation exchange capacity (CEC), mainly iron (Fe) and aluminium (Al) oxyhydroxides and 1:1 layer silicates (kaolin). A substantial proportion (25-30%) of Australian soils are sodic in nature, i.e. exchange complex dominated by sodium (Chartres 1995) and significant areas of land are also affected by salinity (Northcote and Skene 1972) and acidity (Chartres 1998).
In contrast, New Zealand soils are relatively young and generally less highly weathered than their Australian counterparts. New Zealand soils also have relatively high organic matter contents, not just in comparison to Australia, but in comparison to similar soils worldwide (Tate et al. 1997). Again in contrast to Australian soils, marked texture contrasts between surface and subsoil horizons are rare in New Zealand (Hewitt 1997), and sodicity and salinity are also rare. In both countries, however, variable charge minerals form an important component of many soils. The short-range order aluminosilicates (allophane and imogolite) and the oxyhydroxides of Fe and Al are particularly important in many New Zealand soils, for example those formed on volcanic parent materials (McLaren and Cameron 1996). The importance of variable charge minerals contrasts with the situation in Europe and North America where many soils are dominated by permanent charge minerals.
The above differences are important when considering the reactions of metals in soils. For example, soil to plant transfer of Cd has been found to be large in certain regions of Australia due to soil salinity (McLaughlin et al. 1994, 1997a, 1997b), and cationic metal mobility has the potential to be high in variable charge soils having net positive charge (e.g. subsoils in Ferrosols) (Naidu et al. 1994). Sandy soils generally allow greater bioavailability of added metal compared with clay-rich soils, and low soil pH also generally increases metal bioavailability (Chancy and Hornick 1978; Chaudri et al. 1999).
Agricultural systems affected or potentially affected
The management of metal contamination of agricultural land requires a good understanding of the nature of the affected or potentially affected soils. Soil properties such as organic matter content, mineralogy, soil pH, and salinity can all have major effects on the potential bioavailability and/or mobility of contaminant metals. To a large extent, the guidelines for metal limits in soils currently in use in Australia and New Zealand have derived from European or North American regulations. For example, the NZ Department of Health (1992) recommended limits for heavy metals in biosolid and biosolid-amended soils closely follow the UK regulations. Similarly in Australia, guidelines produced by State authorities (SA EPA 1996; NSW EPA 1997; Victorian EPA 1997; Tas. DoELM 1999) and by the Agricultural and Resource Management Committee of Australia and New Zealand (ARMCANZ 1995) are based largely on European regulations. Unfortunately, the origin of these European threshold values are obscure, and where recent studies have been carried out, they have been on soils and under climatic conditions substantially different to those existing in Australia and New Zealand. Thus the direct importation of such regulations to be applied in Australia and New Zealand, without adequate consideration of local conditions, must be questioned.
Contamination of soils with metals can arise directly as a result of the active disposal of wastes onto land, or indirectly as a result of some other agricultural, industrial, or human activity (Merry and Tiller 1991). Although the bulk of metal-contaminated or potentially contaminated soils are likely to be in the regions immediately surrounding urban developments and/or industrial sites, truly rural areas can also be affected. For example, the inputs of Cd to agricultural soils through additions of phosphatic fertilisers have been well documented for both Australia and New Zealand (Roberts et al. 1994; McLaughlin et al. 1996). In addition to Cd, application of phosphatic fertilisers also inadvertently adds other potentially toxic elements to soil, including F, Hg, and Pb (McLaughlin et al. 1996). Such contamination, although generally at a relatively low level compared with that in Europe, is widespread wherever phosphatic fertilisers have been used for any length of time. Land used for both cropping (agricultural and horticultural) and grazing is affected, apart from extensive rangelands that do not receive fertilisers.
The use of fertilisers is by no means the only agricultural operation that can result in metal contamination of soils. Although their use has now been mainly phased out, several common pesticides used fairly extensively in agriculture and horticulture in the past contained substantial concentrations of metals. Compared with fertilisers, the use of such materials has been more localised, being restricted to particular crops or sites. However, the degree of residual contamination is often much higher than that caused by fertiliser applications. Examples of such pesticides are copper-containing fungicidal sprays such as `Bordeaux mixture' (copper sulfate) and copper oxychloride. Lead arsenate was used in fruit orchards for many years to control a range of parasitic insects. Merry et al. (1983) reported a considerable residual accumulation of As, Cu, and Pb in orchard soils in South Australia and Tasmania resulting from the use of these pesticides. Concentrations of As, Cu, and Pb in surface soils were 25-35 times greater than background values. Arsenic-containing compounds were also used extensively to control cattle ticks and to control pests in banana plantations in Queensland and north-eastern New South Wales (Vaughan 1993). In the former case, this has resulted in extremely localised high-level As contamination of soils at cattle dip sites (McLaren et al. 1998), and in the latter a more widespread but much lower level of contamination. In New Zealand and Australia, timbers have been preserved with formulations of Cu, Cr, and As (CCA) and there are now many derelict sites where soil concentrations of these elements greatly exceed background concentrations. Such contamination has the potential to cause problems, particularly if sites are redeveloped for other agricultural or non-agricultural purposes. Indeed, runoff from a derelict timber treatment site in New Zealand has contaminated small areas of adjacent pasture land with As, Cr, and Cu to concentrations as high as 1300 ms/kg (Yeates et al. 1994).
Certain animal wastes produced in agriculture also have the potential to cause metal contamination of the soil. Although most manures are seen, quite rightly, as valuable fertilisers, in the pig and poultry industry Cu and Zn may be added to diets as growth promoters (Chancy and Oliver 1996). The manures produced from animals on such diets contain high concentrations of Cu and/or Zn, and if repeatedly applied to restricted areas of land can cause considerable build-up of these metals in the soil (McGrath et al. 1982). Furthermore, these wastes also contain Cd, usually at concentrations [is less than] 10 mg/kg, but high application rates to soil can result in significant accumulation of Cd in soils and horticultural produce (Jinadasa et al. 1997).
However, the wastes produced in urban and industrial areas provide the most likely sources of heavy metal contamination when applied to soils and there is increasing pressure in both Australia and New Zealand for land disposal of sewage effluents and biosolids rather than disposal to the sea (McLaren and Smith 1996). In addition, the increasing pressure to divert green organic and other urban wastes from landfill is resulting in their application to agricultural soils. Land disposal of biosolids has the potential to contaminate large areas of land, and since transportation costs of such bulky material are high, the land immediately surrounding cities is probably most at risk. In Australia, we estimate that in excess of 175 000 t dry biosolids are produced each year by the major metropolitan water authorities, and while some centres are disposing of biosolids to landfill or stockpiling, agricultural reuse is likely to increase in the future. Currently, most biosolids applied to agricultural land are used in arable cropping situations where they can be incorporated into the soil. However, provided the appropriate grazing withholding periods are observed (e.g. NZ Dept. of Health 1992; NSW EPA 1997), or if the material is injected into the subsurface as practiced by Melbourne Water at Werribee in Victoria, biosolids can also be used on pasture land. In Australia, Sydney Water is the major water authority practising reuse of biosolids in agriculture, with the bulk of the annual biosolid production being applied to agricultural land. There is also considerable interest in both Australia and New Zealand in the application of biosolids to plantation forests. This method of biosolids reuse is currently being practiced by authorities in several parts of the world including the USA, UK, and France. Recent studies in New Zealand (Cameron et al. 1994) have demonstrated the potential build-up of metals in biosolids-treated forest soils.
There is also considerable interest in the potential for composting biosolids with other organic materials such as sawdust, straw, or garden waste. In the USA, composting of biosolids has increased substantially over the past 10-15 years (Goldstein and Steuteville 1996), a trend that seems likely to continue. In Australia and New Zealand some composting of biosolids has also been tried. If the trend towards composting of biosolids continues, there will be implications for metal contamination of soils. As a result of the addition of materials containing low concentrations of metals to biosolids, composts generally have much lower metal concentrations than biosolids on their own, but their use is generally less restricted and monitored than that of biosolids (see below). Whereas these materials have lower metal concentrations than biosolids, they are often applied at much higher rates (up to several hundred tonnes per hectare), and therefore have the potential to considerably increase metal contamination of soil, especially in horticulture.
In addition to biosolids, the long-term irrigation of land with sewage effluent can also lead to an increase in metal concentrations. Although the metal concentrations in sewage effluent are usually relatively low (McLaren and Smith 1996), and thus it may take a long time for heavy metals to accumulate in effluent-irrigated soils, the potentially harmful effects that may arise as a result of future land use should not be ignored (Page and Chang 1985).
Although biosolids, in terms of the volumes produced, probably represent the major potential source of soil metal contamination by wastes, other types of wastes, such as tannery effluent, fly ash, and pulp and paper sludges, can also contribute to heavy metal accumulation in soil (Cameron et al. 1997). Such contamination is often relatively localised and restricted to the close proximity of the waste source. However, where large volumes of waste are produced, such as in the booming recycling industry for green and organic wastes, substantial areas of land may be required for disposal and may involve application to agricultural or forestry land.
The examples of soil metal contamination discussed above all result from the deliberate application of materials to the land in the form of either wastes or agricultural chemicals. It is also recognised that human activity within urban areas can result in contamination of surrounding rural areas by means of atmospheric transport and deposition of contaminants, including heavy metals. Clearly such contamination can affect all possible types of land use. Sources of metal contaminants in the atmosphere include industrial and domestic burning of fuels (coal, oil, wood), vehicle emissions, and a whole range of industrial activities including metal smelting, petroleum refining, and other chemical production processes. Atmospheric pollutants are removed from the atmosphere and returned to terrestrial surfaces by both dry and wet deposition, and it has been demonstrated that major sources of pollution, such as metal smelters, can result in substantial enhancement of soil metal concentrations for considerable distances from the source. One of the best-documented examples of this type of contamination in the Southern Hemisphere relates to a lead-zinc smelter at Port Pirie in South Australia. Cartwright et al. (1976) have shown that enhancement of Cd, Cu, Pb, and Zn concentrations in surface soils can be detected at up to 40-65 km from the smelter, depending on the direction of the prevailing wind. Probably the most common contaminant emitted into the urban atmosphere is Pb derived from automotive exhausts, and its deposition back to the terrestrial environment has been more frequently studied than any other contaminant. Although many studies have demonstrated the relatively short-distance ([is less than] 30 m) accumulation of lead in soils alongside major highways (Smith 1976), Tiller et al. (1987) have shown a much wider dispersal of Pb into the countryside from the city of Adelaide in South Australia.
It is clear from the above discussion that substantial areas of agricultural, horticultural, and forestry land have already been contaminated with metals to some degree, and that there is considerable potential for Continued contamination. While levels of soil contamination in Australia and New Zealand are generally less than those reported in northern hemisphere countries, Australia and New Zealand have plant production systems that rely more heavily on plant-microbe symbioses (e.g. Rhizobium, mycorrhizae) which are very sensitive to metal inputs (Chaudri et al. 1993). Soils in Australia and New Zealand may also be more sensitive to metal contamination than those in the northern hemisphere (McLaughlin et al. 1997a, 1997b). It is important that those activities resulting in contamination are identified, so that the risks and extent of soil contamination with metals can be minimised.
Chemical characteristics of fertilisers and wastes in Australia
There are few published reviews of the composition of fertilisers, composts, biosolids, soil amendments, and other wastes with regard to concentrations of contaminants. In general, potassic and nitrogenous fertilisers have been found to have low levels of contaminants (Best 1992; Rayment et al. 1992; Zarcinas and Nable 1992; Corry et al. 1993). Phosphatic and trace element fertilisers appear to have the highest levels of contaminants, reflecting the source of raw material used in their manufacture (Oertel and Stace 1947; Williams 1978). Australia and New Zealand have traditionally sourced phosphate rock for manufacture of phosphatic fertilisers from the oceanic island deposits of Christmas Island, Nauru, and Banaba (formerly Ocean Island). Perhaps the most comprehensive screening study of contaminants in fertilisers was performed by Williams (1978) (Table 1), with later surveys concentrating more on the elements of principal concern, namely Cd, Hg, and Pb (Best 1992; Rayment et al. 1992; Zarcinas and Nable 1992; Corry et al. 1993).
Table 1. Concentrations of metals in phosphatic fertilisers in Australia and North America (from Williams 1978 and NFDC 1980)
SSP, single superphosphate; TSP, triple superphosphate
Australia SSP Others SSP mean range range Pollutant (mg/kg) (mg/kg) (mg/kg) (mg/kg P) Cd 20-49 12-60 40 440 Cr 20-78 32-130 61 670 Cu 10-201 2-58 28(A) 307 F 9600-17 400 400-22 000 14 300 157(B) Pb 2-71 2-27 19 176 Mo 0.1-3.2 0.1-0.8 0.5 6 Ni 1.4-7.5 0.9-14.0 3.9 43 Zn 269-488 182-1100 393 4320 Central North Western Florida Carolina USA TSP TSP TSP Pollutant (mg/kg P) (mg/kg P) (mg/kg P) Cd 39 320 432 Cr 408 1106 2478 Cu 78 58 110 F -- -- -- Pb 34 15 21 Mo -- -- -- Ni 141 345 668 Zn 433 2925 5075
(A) Excludes 2 high values of 155 and 201 mg/kg.
(B) Units are g/kg P.
Phosphatic fertilisers used in Australia have generally had higher concentrations of Cd, Pb, and Zn than the materials manufactured in eastern USA, but similar concentrations to those used in western USA. Over the last 15 years there has been a shift in the source of manufactured fertilisers used in Australia and New Zealand, with fewer products derived locally from island sources of rock and more material being imported from eastern USA and other countries. This is reflected in a fall in Cd concentrations in P fertilisers as noted by Zarcinas and Nable (1992) (Fig. 1).
[Figure 1 ILLUSTRATION OMITTED]
Agricultural limestones, gypsums, composts, and other soil amendments may also contain metal contaminants. Again, few data are available for Australian or New Zealand materials, but some typical analyses are shown in Table 2.
Table 2. Concentrations of metals in selected soil amendments, composts, and other wastes
Pollutant Lime(ABC) Gypsum(ABCD) Phosphogypsum(ACD) Cd <2.8-17.0 <2.8 <2.8-6.4 Cr 6.5-50.3 <2.5-25.0 -- Cu -- -- -- Pb <15.0-549.0 <10.0-32.6 <10.0 Hg -- <0.1 <0.5 Mo -- -- -- Zn -- -- -- Domestic Green waste/ Green waste grease trap Green waste Pollutant compost(E) sludge(E) waste(F) compost(G) Cd 1.3 <0.5 0.4 0.6 Cr -- -- 45.0 26.0 Cu 18.0 5.0 47.0 43.0 Pb 21.0 <10.0 96.0 191.0 Hg -- -- 0.1 -- Mo 85.0 16 -- -- Zn 73.0 46.0 205.0 244.0
(A) Best (1992).
(B) Rayment et al. (1992).
(C) Zarcinas and Nable (1992).
(D) Corry et al. (1993).
(E) Melbourne, K. Wilkinson (pers. comm.).
(F) Auckland, G. Fietjie (pets. comm).
(G) Christchurch, R. McLaren (unpubl. data).
Biosolids are increasingly being re-used on agricultural land in Australia and New Zealand. For example, Sydney Water in NSW produces approximately 200 000 wet tonnes per year of biosolids, 99% of which is put to beneficial re-use (Barry et al. 1998). In Queensland, approximately 250 000 wet tonnes are produced annually (Barry et al. 1998). Biosolids vary markedly in quality, both between treatment works and also temporally for any single treatment works, mainly due to changes in quality of the wastewater stream with time and across locations (Pantsar-Kallio et al. 1999). Since 1983 when de Vries (1983) surveyed the quality of 20 biosolids from around Australia, there has been no published compilation of the typical quality of biosolids in Australia. Recently, Barry et al (1998) investigated the quality of biosolids in coastal Queensland, and observed contaminant concentrations similar to those found by de Vries almost 20 years ago, except for Cd which was considerably lower (Table 3). Trade waste restrictions in the amounts of Cd discharged to sewerage systems is the most likely reason for this improvement in biosolid quality for this element. Table 3 also includes data from two recent surveys of biosolids composition carried out in New Zealand (Ogilvie 1998; McLaren et al. 1999a). From these limited surveys it would appear that Australian biosolids have higher concentrations of Cu and Zn than those in New Zealand, with other metal concentrations being similar between the countries.
Table 3. Typical quality (mg/kg) of biosolids produced at some of the major treatment works in Australia and New Zealand
Australia(A) Pollutant Range Mean Median As -- -- -- Cd 2-285 41 26 Cr -- -- -- Cu 251-2480 856 671 Pb 55-1980 562 424 Hg -- -- -- Mo -- -- -- Ni 20-318 88 60 Se -- -- -- Zn 241-5510 2070 1930 Queensland(B) New Zealand Pollutant Mean Range Median(C) Range(D) As 16 2-50 -- 3-38 Cd 3.6 0.8-8.1 2.5 0.1-7.6 Cr 110 28-1113 50 2-1331 Cu 662 104-2283 311 21-789 Pb 171 19-629 103 15-501 Hg 4.4 0.2-20.0 -- 0.3-15.6 Mo -- -- -- 0.6-17.8 Ni 48 15-320 25 6-254 Se -- -- -- 1.2-5.5 Zn 1305 185-5925 724 140-2142
(A) 20 biosolids from around Australia (de Vries 1983).
(B) Mean and range across 33 sewage treatment works in coastal Queensland (Barry et al. 1998).
(C) Median across 21 biosolids in New Zealand (McLaren et al. 1999a).
(D) Range across 36 biosolids in New Zealand (Ogilvie 1998; McLaren et al. 1999a).
Fertiliser-derived contaminants differ markedly in the form in which they occur compared with contaminants in composts or biosolids. In fertilisers, the contaminants are effectively present as inorganic salts, and in soil their availability is similar to the pure chemical form e.g. Cd in phosphatic fertilisers behaves similarly to Cd[([H.sub.2][PO.sub.4]).sub.2] and [CdHPO.sub.4] (Mortvedt and Osborn 1982). In the short-term at least, contaminants in biosolids are much less available in soil compared with the equivalent contaminant salts (Korcak and Fanning 1985; Bell et al. 1991).
Current regulatory frameworks in Australia and New Zealand
Regulations or guidelines relating to heavy metals in the environment in Australia and New Zealand are generally based on, or modified from, overseas guidelines.
In Australia, metal concentrations in soils and soil amendments (including fertilisers) are regulated at the State level. This has resulted in significant discrepancies between States with regard to maximum permissible concentrations (MPC) for contaminants in fertilisers and soil amendments, and with regard to guidelines for reuse of biosolids on soils. Food quality regulations have been consistent between States, as these were assessed and monitored federally through the then National Food Authority (NFA) and the National Health and Medical Research Council (NH&MRC). The Australian and New Zealand food authorities recently merged to form the Australia and New Zealand Food Authority (ANZFA), so that hannonisation of food MPCs between the two countries is imminent (ANZFA 2000). For some time now, guidelines for assessing soil contamination have been common between Australian States and New Zealand through the guidelines for assessment and management of contaminated sites produced by the Australian and New Zealand Environment and Conservation Council (ANZECC) (ANZECC/NHMRC 1992), and more recently in Australia through the National Environmental Health Forum (Imray and Langley 1996). The recently formed National Environmental Protection Council (NEPC) released guidelines (in December 1999) for assessing risks from metals (and other contaminants) in soils through a draft National Environment Protection Measure (NEPM). This NEPM covers all aspects of site assessment including sampling considerations, analyses, risk assessment, and community consultation but is focussed more on urban and industrial soil contamination. However, risk assessment and remediation of contaminated soils are still the jurisdiction of State environmental and health agencies.
In New Zealand, the New Zealand Resource Management Act (1991) integrated a previously large number of environmental statutes into a single, all encompassing law, the purpose of which is the promotion of sustainable management of natural and physical resources. The legislation focuses on the effects of activities rather than the activities per se and is implemented within a hierarchy of policies, plans, rules, and consents designed to minimise adverse effects on the environment (Roberts et al. 1996). However, there is neither legislative definition of a contaminated site, nor any statutory threshold levels which would define whether a site is contaminated. To date, such matters are addressed by means of government policy including the development of national guidelines and standards, and involve more than one government department (Agriculture, Environment, and Health). At the local government level, regional councils have the responsibility to develop regional policies and plans consistent with environmental statutes, including the regulation of waste discharges to air, water, and land.
Regulations controlling metal concentrations in fertilisers in Australia have focussed mostly on Cd and are not consistent across States (Table 4). Some states also regulate concentrations of other metals (and fluorine); for example, in Queensland, the maximum concentration of F in fertilisers has been set at 2.5% and in rock phosphates at 4%, the maximum concentration of Hg in all fertilisers has been set at 2 mg/kg and the maximum concentrations of Pb in phosphatic and trace element fertilisers are 50 and 500 mg/kg, respectively.
Table 4. Regulations controlling Cd in fertilisers in Australia Queensland(A) Victoria(B) Type of fertiliser/soil (mg/kg (mg/kg amendment product) (mg/kg P) product) (mg/kg P) Non-P fertiliser, 10 other than phosphogypsum, or trace element fertiliser Non-P fertiliser, other than trace element fertiliser Non-P fertiliser 10 P fertilisers 350 350 (contain at least 2% P) P fertilisers (not defined) Phosphogypsum 15 Trace elements 50 Tasmania(C) NSW(D) Type of fertiliser/soil (mg/kg (mg/kg amendment product) (mg/kg P) product) (mg/kg P) Non-P fertiliser, other than phosphogypsum, or trace element fertiliser Non-P fertiliser, 10 10 other than trace element fertiliser Non-P fertiliser P fertilisers 350 (contain at least 2% P) P fertilisers 350 (not defined) Phosphogypsum Trace elements 80 50 WA(E) Type of fertiliser/soil (mg/kg amendment product) (mg/kg P) Non-P fertiliser, other than phosphogypsum, or trace element fertiliser Non-P fertiliser, other than trace element fertiliser Non-P fertiliser 80 P fertilisers (contain at least 2% P) P fertilisers 500 (not defined) Phosphogypsum Trace elements
(A) Agricultural Standards Regulation 1997 (Subordinate Legislation 1997, No. 277 of Agricultural Standards Act 1994) [Sections 20 & 21, effective 1 October 1997], Government of Queensland.
(B) Agricultural and Veterinary Chemicals Act 1992; Agricultural and Veterinary Chemicals (Fertilisers) Regulations 1995. Regulation 35 (1a & b), Government of Victoria.
(C) Statutory Rules 1993, No. 200; Fertiliser Regulations 1993. Regulation 30 (2 a, b & c), Government of Tasmania.
(D) Fertilisers Act 1985; Fertilisers Regulation 1997 [effective 1 September 1997]. Regulations 6 (Schedule 1, Part 1, Sub-part 2) and 9. Government of New South Wales.
(E) Government Gazette, WA, 31 December 1992, Reg. 2 amended.
There are currently no regulations in New Zealand limiting the metal contents of fertilisers. However, superphosphate fertiliser manufacturers have introduced voluntary controls on the Cd content of phosphatic fertilisers, with targeted Cd concentrations falling from 420 mg Cd/kg P in 1995 to 280 mg Cd/kg P in the year 2000 (Roberts et al. 1996).
Biosolids (sewage sludges)
In Australia, guidelines for reuse of biosolids on soils are published for NSW, SA, and Tasmania (SA EPA 1996; NSW EPA 1997; DoELM 1999), and draft guidelines are available in Victoria (Victorian EPA 1997) (Tables 5 and 6).
Table 5. Guidelines for controlling metal concentrations in biosolids for reuse in Australia (Beavers 1993; SA EPA 1996; NSW EPA 1997; Vic. EPA 1997; Tas. DoELM 1999)
In South Australia, Victoria, Queensland, and Tasmania both Grades A and B biosolids may be used on agricultural soils. CACT contaminant acceptance concentration threshold (mg/kg); MPC, maximum permitted concentration (mg/kg); MAL, maximum annual loading (kg/ha.year); MCL, maximum cumulative loading (kg/ha)
NSW Biosolid SA Biosolid CACT MPC MAL Pollutant A B C D A B As 20 20 20 30 20 20 0.07 Cd 3 5 20 32 3 11 0.15 Cr 100 250 500 600 -- Cu 100 375 2000 2000 200 750 12 Pb 150 150 420 500 200 300 15 Hg 1 4 15 19 1 9 0.1 Mo -- -- -- -- -- Ni 60 125 270 300 60 145 3 Se 5 8 50 90 -- -- -- Zn 200 700 2500 3500 250 1400 30 Vic. Biosolid Qld Biosolid MPC MAL MCL MPC (draft) Pollutant A B A B As 20 75 4 20 20 75 Cd 3 85 0.25 2.5 2 85 Cr 400 3000 15 75 200 3000 Cu 200 4300 20 100 140 4300 Pb 200 840 15 75 150 840 Hg 1 75 0.6 3 1.5 57 Mo 4 75 -- -- -- 75 Ni 60 420 4 20 60 420 Se 3 100 5 25 -- 100 Zn 250 7500 50 250 300 7500 Tas. Biosolid CACT Pollutant A B As 20 20 Cd 3 20 Cr 100(A) 500(A) Cu 100 1000 Pb 150 420 Hg 1 15 Mo -- -- Ni 60 270 Se 5 50 Zn 200 2500
(A) If biosolids fail the Cr limits, a combined [Cr.sup.III]/[Cr.sup.VI] analysis may be performed and [Cr.sup.VI] should not exceed 100 mg/kg (Grades A and B) and [Cr.sup.III] should not exceed 400 mg/kg (Grade A) or 3000 mg/kg (Grade B).
Table 6. Guidelines for controlling metal concentrations (mg/kg) in soils for reuse of biosolids in Australia (SA EPA 1996; NSW EPA 1997; Tas. DoELM 1999)
MPC, maximum permitted concentration; MASCC, maximum allowable soil contaminant concentration
NSW Soil MPC SA Agricultural Non-agricultural Soil Tasmania Pollutant land land MPC MASCC As 20 20 20 14 Cd 1 5 3 0.7 Cr 100 250 -- 35(A) Cu 100 375 200 42 Pb 150 150 200 105 Hg 1 9 1 0.7 Mo -- -- -- -- Ni 60 125 60 42(A) Se 5 8 -- 3.5 Zn 200 700 250 140
(A) These limits can be ignored on Red Ferrasols due to naturally high background concentrations of these elements in this soil type.
In New Zealand, recommended limits for heavy metals in biosolids intended for application to arable land (Table 7) are included in the NZ Department of Health (1992) public health guidelines for the safe use of sewage effluent and biosolids on land. In addition, the Agricultural and Resource Management Council of Australia and New Zealand (ARMCANZ) has produced national guidelines which are now in draft form as part of the National Water Quality Management Strategy (NWQMS). There are also new draft guidelines in New Zealand, prepared for the New Zealand Water and Wastes Association (NZWWA 1999). While there is some consistency between these various documents, there are also significant discrepancies. In general the regulations for metals are largely based on European guidelines (CEC 1986), and are significantly more conservative than US EPA regulations (US EPA 1993) (Table 8).
Table 7. Guidelines for metal concentrations in biosolids and for controlling metal concentrations in soils for reuse of biosolids in New Zealand (NZ DoH 1992) and acceptance limits for biosolids suggested by the New Zealand Water and Wastes Association (NZWWA 1999)
MPC, maximum permitted concentration; MAL, maximum annual loading; MCL maximum cumulative loading. NZWWA grades are PU = public use grade, A = agricultural grade, and NA = non-agricultural grade
NZ DoH Biosolid Soil MPC MAL MCL MPC Pollutant (mg/kg) (kg/ha) (kg/ha) (mg/kg) As 10 0.2 2.5 15 Cd 3 0.2 2.5 15 Cr 600 15.0 125.0 1000 Cu 140 12.0 100.0 1000 Pb 300 15.0 125.0 600 Hg 1 0.1 1.0 10 Mo -- -- -- -- Ni 35 3.0 20.0 200 Se -- -- -- -- Zn 300 30.0 250.0 2000 NZWWA Grade PU Grade A Grade NA Soil MPC Pollutant (mg/kg) (mg/kg) (mg/kg) (mg/kg) As -- -- -- -- Cd 1 3 6 1 Cr 600 1800 3600 600 Cu 100 300 600 100 Pb 300 900 1800 300 Hg 1 3 6 1 Mo -- -- -- -- Ni 60 180 360 60 Se -- -- -- -- Zn 300 900 1800 300
Table 8. Guidelines for controlling metal concentrations in soils for reuse of biosolids in the USA (US EPA 1993) and Europe (CEC 1986)
US EPA Limit Cumulative Limit conc. conc. for loading for `safe biosolid(B) limit biosolid'(C) Pollutant (mg/kg) (kg/ha) (mg/kg) As 75 41 41 Cd 85 39 39 Cr 3000 3000 1200 Cu 4300 1500 1500 Pb 840 300 300 Hg 57 17 17 Mo 75 18 18 Ni 420 420 420 Se 100 100 36 Zn 7500 2800 2800 US EPA CEC(A) Annual Annual loading Biosolid loading limit Soil MPC MPC limit(D) Pollutant (kg/ha.year) (mg/kg) (mg/kg) (kg/ha.year) As 2.0 Cd 1.9 1-3 20-40 0.15 Cr 150 Cu 75 50-140 1000-1750 12 Pb 15 50-300 750-1200 15 Hg 0.85 1.0-1.5 16-25 0.1 Mo 0.90 Ni 21 30-75 300-400 3 Se 5.0 Zn 140 150-300 2500-4000 30
(A) Assumes soil pH in range 6-7.
(B) Absolute limit for beneficial use.
(C) Unrestricted use.
(D) Based on 10-year average.
These guidelines are intended to provide advice on the safeguards necessary to protect public health, and in the case of heavy metals are aimed primarily at preventing metal accumulation in the food chain and toxic effects on plants and soil fauna. Limits have been set for MPCs in biosolids and soil. There are also limits for maximum annual loading rates (MALs) and maximum cumulative loading rates (MCLs) for agricultural land, although the time for this limit to be reached is not defined. In New Zealand, the NZDoH (1992) recommendations refer specifically to arable land and the guidelines suggest that there are some situations where there is no pathway or transfer mechanism for a possible public health effect, when the recommended limits may be exceeded. For soils receiving biosolids, in addition to the limits in Tables 5-7, several further controls relating to soil pH, soil slope, soil water regime, and proximity to watercourses, roads, property boundaries, and residences are deemed necessary to protect public health. Minimum soil pH limits vary and often the basis of measurement is not specified, despite the fact that soil pH measured in 0.01 M Ca[Cl.sub.2] solution may be 0.5-1.0 units lower than that in water. Threshold pH values (with the measurement basis in parentheses) vary from 4.5 in Tasmania (unspecified) through 5.0 in New Zealand (water) and ARMCANZ (unspecified), to 5.5 in SA and NSW (0.01 M Ca[Cl.sub.2]). Furthermore, biosolid reuse on irrigated land is discouraged in the SA guidelines due to the risks from salinity in irrigation waters significantly enhancing the soil-plant transfer of Cd (McLaughlin et al. 1994).
In the NZ biosolid guidelines, no rationale is given for the considerable discrepancies between the MCL and the soil MPC values, assuming incorporation of the metal into the top 200 mm soil.
There are no regulations/guidelines in Australia and New Zealand specifically limiting metal additions to land in wastes/manures/composts other than biosolids. The Australian Standard for composts, AS 4454 (Anon. 1997), suggests that contaminant levels should meet relevant biosolid guidelines. Organic wastes (including agricultural wastes) are often limited by regional authorities on the basis of nitrogen loadings but metal contents are not generally considered. Industrial wastes that are known to contain metals (e.g. Cr in tannery waste) are generally considered case by case. In such situations, the approval for land disposal may be based on the Environmental Soil Quality Guidelines (see below) or other overseas guidelines on a case-by-case basis.
The soil guidelines developed by ANZECC and the NH&MRC (ANZECC/NHMRC, 1992) are generally regarded as Government policy in Australia and New Zealand. The guidelines for environmental soil quality are shown in Table 9. These guidelines are being used increasingly in New Zealand in relation to resource consent hearings for disposal of wastes. Unfortunately, the interim nature and limitations of these guidelines are not always appreciated.
Table 9. Environmental Soil Quality Guidelines (mg/kg) (ANZECC/NHMRC 1992)
Environmental Heavy Background investigation metal A B As 0.2-30 20 Ba 20-200 Cd 0.04-2 3 Cr 0.5-110 50 Co 2-170 Cu 1-190 60 Pb <2-200 300 Mn 4-12 600 500 Hg 0.001-0.1 1 Mo <1-20 Ni 2-400 60 Sb 4-44 20 Sn 1-25 50 Zn 2-180 200
In the ANZECC/NH&MRC guidelines, limits to protect human health were also suggested which triggered remediation activity (`C' values), but these have since been superseded by health-based investigation levels (HBILs) (Imray and Langley 1996). These are limits based on a more complex series of scenarios for potential exposure of humans to soil contaminants than the original C values, viz. dermal contact, inhalation, soil-plant transfer, soil-plant-animal-human transfer, soil-human transfer (soil ingestion), These HBILs were developed as a series of graded levels for contaminants dependent on intended use of the contaminated soil (which affects exposure). The HBILs are significantly higher than ANZECC `B' values or the `C' values which they replaced (e.g. HBILs for Zn vary from 7 000 to 35 000 ms/kg depending on intended land use), but are unlikely to be applied in agricultural soils where environmentally based guidelines, protecting plant production, animal health, and food quality, are more applicable.
Assessment of metal contamination to protect the agricultural environment has generally been based on ANZECC B values, although the NSW EPA recently published a set of phytotoxicity-based investigation levels (PILs) based on protection of plants from metal toxicity (NSW EPA 1998) (Table 10). These PILs are based on values derived in the UK (DoE 1996) for light-textured soils with pH in the range 6-8. They are primarily designed to protect native species used for revegetation of contaminated soils in urban areas. As such, they will probably be used as end-points for remediation of urban/industrial soil where plant growth is envisaged in the land use, but the values may also be used in agricultural areas due to the targeted exposure pathway being plant growth. Ecologically based investigation levels (EILs) have also been developed in Victoria by a risk-pathway model for soil fauna (Walker et al. 1999). The exposure scenario developed was for an urban residential environment and the EILs generated are not relevant for agricultural soils. However, the risk pathway approach could be applied to the agricultural environment to develop EILs for agricultural soils.
Table 10. Draft Phytotoxicity Investigation Levels (mg/kg) (NSW EPA 1998)
Element Draft PIL As 20 Cd 3 Cr (III) 400 Cr (VI) 10 Cu 100 Pb 600 Ni 60 Zn 200
Environmental trigger or limit values for metal concentrations in soils in the Netherlands are shown in Table 11. The Netherlands has so-called `reference values' or `target values', which represent soils with negligible risks, while `intervention values' represent soils with the maximum permissible risk levels due to contamination. Reference values were determined as the upper 95th percentile of concentrations found in `unpolluted' sites (Lexmond and Edelman 1987), while intervention values were determined using a semimechanistic risk pathway model, C-SOIL (van den Berg and Roels 1991). While derived by different methods, both sets of values are sliding scale `algorithms' based on clay and organic matter content. The rationale for this approach, at least for the reference values, was that these factors were found to be strongly related to metal concentrations in unpolluted soils, clay being important for metal retention and organic matter to take account of changes in soil bulk density (Lexmond and Edelman 1987).
Table 11. Dutch reference values and intervention values (all mg/kg) for contaminants in soil based on ecotoxicity and human health considerations (from Smit 1998)
Element concentrations are normalised on the basis of clay (L) and organic matter (H) content of soil
Ref. value Reference for standard Element value model soil(A) As 15 + 0.4 (L+H) 29 Cd 0.4 + 0.007 * (L+3H) 0.8 Cr 50 + 2L 100 Cu 15 + 0.6 * (L+H) 36 Pb 50 + (L+H) 85 Hg 0.2 + 0.0017 * (2L+H) 0.3 Ni 10 + L 35 Zn 50 + 1.5 * (2L+H) 140 Ecotoxicity Human health intervention intervention Element value value As 40 680 Cd 12 35 Cr 230 2250 Cu 190 16 000 Pb 290 300 Hg 10 200 Ni 210 6600 Zn 720 56 000
(A) Standard soil of 25% clay, 10% organic matter.
The reference values are regarded as the desirable upper limit for metals in agricultural soils (F. A. M. de Haan, pers. comm.) and these have recently been refined further (Table 12), so that their derivation raises important issues with regard to adoption of this approach in Australia and New Zealand. Firstly, `unpolluted', like `background', is not a well-defined condition due to contaminant movement through erosion and atmospheric accession of contaminants to seemingly `unpolluted' soils. Secondly, by definition, 5% of the natural population of `unpolluted' soils are above the reference value--does this define these soils as `polluted'? Finally, the approach of reference or target values indicates that the philosophical approach in the Netherlands to controlling soil contamination is that any increases in soil metal concentrations above statistically defined background or target values are unacceptable. Hence, significant effort has recently been expended to construct balances for heavy metals in various agricultural systems (Moolenaar et al. 1997), so that acceptable MPCs for contaminants in agricultural inputs can be calculated based on zero net accumulation.
Table 12. Dutch signal values for heavy metals and arsenic in agricultural soils (in mg/kg dry matter) (from Smit 1998)
A, grassland with grazing sheep; B, grassland with grazing livestock; 2, arable land for fodder crops; 3, arable land for food-plants; 4, arable land for growing of ornamental plants
Element Sand and reclaimed peat soils 1A 1B 2 3 4 As 30 30 30 30 30 Cd 2 2 0.5 0.5 5 Cr 200 200 200 200 200 Cu 30 50 50 50 50 Hg 2 2 2 2 2 Pb 150 150 150 150 150 Ni 15 15 15 15 15 Zn 200 200 100 100 100 Element Clay and peat soils 1A 1B 2 3 4 As 50 50 50 50 50 Cd 3 3 1 1 10 Cr 300 300 300 300 300 Cu 30 80 80 200 200 Hg 2 2 2 2 2 Pb 150 150 150 200(A) 800 Ni 50/70(B) Zn 350 350 350 350 350
(A) This value has to be used cautiously, as sufficient data on atmospheric deposition of lead are lacking. Atmospheric deposition may increase the lead content of the plant; an effect not caused by soil processes.
(B) 50 mg/kg for clay soils and 70 mg/kg for peat soils.
Metal concentrations in food in Australia and New Zealand are covered by legislation enforced by the State health departments in Australia and by the Ministry of Health in New Zealand. Food MPCs for Cd were recently reviewed and MPCs for other contaminants are currently under review. Maximum permitted concentrations for metals in food in Australia are shown in Table 13, and for New Zealand in Table 14.
Table 13. Maximum permitted concentrations of inorganic contaminants in foodstuffs in Australia (NFA 1993; ANZFA 1997)
MPC Metal Food (mg/kg FW) Arsenic Beverages 0.15 Foods except chicken livers, 1.0 (total) fish, crustaceans, molluscs, and seaweed Cadmium Chocolate 0.50 Kidney 2.50 Liver 1.25 Leafy vegetables 0.10 Meat 0.05 Peanuts 0.10 Rice 0.10 Root and tuber vegetables 0.10 Wheat 0.10 Copper Beverages 5.0 Cocoa and chocolate 50.0 Edible offal other than 100.0 ovine livers Nuts 50.0 Ovine livers 200.0 Other foods (except molluscs) 10.0 Lead Beverages 0.2 Infant foods 0.3 Offal 1.0 Other foods 0.5 Mercury Other foods (except seafood) 0.03 Selenium Beverages 0.2 Brazil nuts 10.0 Edible offal 2.0 Other foods 1.0 Tin Canned foods (asparagus, fruits, 250.0 fruit juices, tomato products and green beans) Other canned foods 150.0 Foods not in cans 50.0 Zinc Beverages 5.0 Other foods (except oysters) 150.0
Table 14. Maximum permissible concentrations for contaminants in food (New Zealand Food Regulations 1984)
Permissible concentration Element (mg/kg FW) Notes As 2(A) Any food except fish, beverages, and other liquid food Cd 1 Any food other than shellfish Cu 30 Any food except animal offal, tea, all beverages, and other liquid food F 10 Any food except shellfish, all beverages, and other liquid food Pb 2(B) Any food except tea, baking powder, and all beverages Hg 0.03 Any food except fish, fish products, and feral pigmeat Se 2 Any food Sb 1 Any food except beverages and other liquid food Sn 40 Any food except foods packed in metal cans Zn 40 Any food except meat, shellfish, all beverages, and other liquid food
(A) 1 mg/kg for foods that meet the special physiological needs of infants and young children.
(B) 0.3 mg/kg for foods that meet the special physiological needs of infants and young children.
Cadmium regulations are currently being harmonised between the countries, and any revised limits for other contaminants are also likely to be harmonised. It has been proposed (ANZFA 1999) that MPCs for Cu, Sb, Se, and Zn be discontinued due to lack of significant risks to human health from the levels generally found in foods. For other elements a new series of MPCs have been proposed (Table 15).
Table 15. Proposed MPCs for arsenic, lead, and tin in foodstuffs relevant to agriculture in Australia and New Zealand (from ANZFA 1999)
MPC Metal Food (mg/kg FW) Arsenic Cereals 1.0 (total) Lead Brassicas 0.60 Cereal, pulses, and legumes 0.20 Edible offal of cattle, sheep, 0.50 pig, and poultry Fruit 0.10 Infant formulae 0.02 Meat of cattle, sheep, pig, 0.10 and poultry Vegetables (except brassicas) 0.10 Tin All canned foods 250.00
It has also been proposed that a series of values for `generally expected levels' (GELs) be established for As, Cu, Hg, Sb, Se, and Zn in foods (ANZFA 1999). These GELs would not be legally enforceable, but would act as target concentrations for good agricultural practices and encourage minimisation of metal transfer through the food chain. Determination of the values used for GELs is still in process (as at April 2000).
Inconsistencies between soil, biosolid, and food regulations for metals
ANZECC A, B, and C values were derived largely from regulations developed for assessing soil contamination in the Netherlands (Moen 1988), and concerns regarding adoption of these values for Australia were expressed 10 years ago by Tiller (1989). The ANZECC C values have since been replaced by HBILs (Imray and Langley 1996), while refinement of B values has been restricted to NSW with the development of phytotoxicity-based ElLs by NSW EPA (NSW EPA 1998), which are based largely on the ANZECC B values. There appears to be inconsistency between soil metal limits derived from the various biosolid regulations and ANZECC B values or the EILs developed by NSW EPA (Table 10). For example, soil MPCs for non-food chain applications of biosolids are purportedly `... protective of the environment, human health, animal health ...' (NSW EPA 1997), which presumably includes phytotoxicity, yet phytotoxicity-based EILs are significantly lower than the soil MPCs for biosolid reuse in non-agricultural soils for Cd, Cu, Ni, and Zn (Tables 6 and 10). Protection of the food chain is also not assured by the MPCs for agricultural soils, as several crops in various regions of Australia have been found to violate food Cd regulations at (total) soil Cd levels significantly [is less than] 1 mg/kg (McLaughlin et al. 1994, 1997b; Jinadasa et al. 1997). Furthermore, concentrations of metals in agricultural soils that are clearly not phytotoxic often exceed by a considerable margin the ANZECC B values and the NSW EILs, particularly for Cr and Ni, which have background concentrations in Ferrosols of up to 700 and 500 mg/kg, respectively (Barry 1997; Lottermoser 1997; CSIRO unpubl. data). These concerns all arise because the biosolid MPCs and phytotoxicity ElLs are single trigger values based on total concentrations of metal in soils, with no assessment of metal bioavailability. This is discussed further in later sections of this review.
Assessment of the potential adverse effects of selected pollutants in soil
Due to the differing reactions of the various pollutants with soils and the varying uptake/translocation rates by plants, pollutant elements can be grouped into those posing little risk to agriculture, those posing food chain risks, and those posing risks to plant or animal productivity (Chaney and Oliver 1996). Cadmium is an element of major concern with regard to food chain contamination. Compared with As, Cu, Cr, Hg, Ni, Pb, and Zn, Cd is unique in that it reaches concentrations of concern in crops (to human and animal consumers) at levels that are well below those where phytotoxicity is expressed. Chromium, Hg, and Pb are very strongly retained by the solid phase in most soils, so that accumulation of these elements poses small risks compared with As, Cu, Ni, and Zn. These latter elements may adversely affect plant growth or microbial activity in soil. We, therefore, will focus discussion on the adverse effects of the metals As, Cd, Cu, Ni, and Zn.
The UK soil limit value for As is 50 mg/kg (DoE 1989). In contrast, the Australian and New Zealand soil limits are only 20 and 10 mg/kg, respectively, and this causes considerable concern because many apparently healthy agricultural soils have As concentrations near to, or exceeding, these values (Table 16).
Table 16. Concentrations of metals found in Australian and New Zealand agricultural soils, primarily in horticultural regions (Barry 1997; CSIRO unpubl. data)
Range Metal (mg/kg) As 1-20 Cd 0.02-2.0 Cr 2-700 Co 0.5-150 Cu 0.4-200 Pb 5-80 Hg <0.01-0.15 Mo 0.2-5.0 Se 0.5-3.0 Ni 1-500 Zn 2-250
Arsenic and soil biological processes
The over-riding factor controlling the toxicity of the metalloid As to soil processes and to plants is its speciation and/or bioavailability in the soil. Relatively soluble oxides and salts of As have been used as pesticides for many years, although this practice has generally ceased throughout the world. There have been many studies on the effects of inorganic, soluble, and hence extremely bioavailable As compounds on soil microbiota and processes. Maliszewska et al. (1985) found that [As.sub.2][O.sub.3] (As[III]) and [Na.sub.2]HAs[O.sub.4] (As[V]) both stimulated the development of bacteria and actinomycetes in soil, even at extremely high concentrations (10 000 mg/kg), and Lighthart et al. (1983) found that microbial respiration was stimulated in some soils at 3750 mg As[III]/kg (as NaAs[O.sub.2]). Tabatabai and co-workers found inhibition of soil biochemical properties by As[III] and As[V] at 1875 or 375 mg/kg soil (Juma and Tabatabai 1977; Tabatabai 1977; Liang and Tabatabai 1978; AI-Khafaji and Tabatabai 1979). As[III] was always more inhibitory than As[V], except for inhibition of the enzyme phosphatase. As[V] normally predominates in aerobic soils (Lepp 1981; Tamaki and Frankenberger 1992). The most sensitive biochemical property, nitrification, was inhibited by 71% in a sandy soil amended with 375 mg As/kg soil (Liang and Tabatabai 1978). Significant inhibition of other properties occurred only at the higher rate of As[V] amendment.
Bardgett et al. (1994) and Yeates et al. (1994) investigated the effects of spillage of timber treatment chemicals (As, Cr, and Cu) from a timber treatment plant onto adjacent pasture land in New Zealand. Arsenic concentrations of up to 1265 mg/kg were found in the soil. The heavy metals had been present on the site for a number of years, and therefore, unlike the previous studies, any effects measured on soil biological properties were chronic effects and were from metals which were possibly no longer in freely bioavailable forms. Microbial biomass and activity, enzyme activities, and nutrient transformations were strongly inhibited at the higher metalloid concentrations. The most extreme effect was on the enzyme sulphatase, which was more than 60% inhibited at an As concentration of 400 mg/kg (Bardgett et al. 1994). However, like many contaminated soil situations, it was impossible to determine from their results which of the contaminants was primarily responsible for the reduced biological activity (the site with 400 mg As/kg had 250 mg Cr/kg and 390 mg Cu/kg). In contrast, abundance of nematodes was unaffected by the contamination, although there was a significant change in community structure. In the companion study (Yeates et al. 1994), it was concluded that contamination of the pasture soil by 161, 148, and 109 mg/kg of As, Cr, and Cu, respectively, caused no significant depression of biological activity, including numbers of nematode and enchytraeid worms. Some depression of activities was found at concentrations of the 3 elements at around 400 mg/kg, and at 800 mg/kg all processes were strongly inhibited.
The results of the investigations outlined above highlight the inadequacy of toxicity assessments based on total concentrations in soils. They also suggest that in some soils, As will not unduly adversely affect soil biological properties at concentrations more than an order of magnitude greater than the Australian or New Zealand soil investigation values. This appears to be true even when As is added to the soil as a soluble salt.
Arsenic phytotoxicity and uptake by plants
Again, the major factor controlling the toxicity of As is the bioavailability of the form in the soil. Water-soluble As is more phytotoxic than other more firmly bound forms (Lepp 1981). Sheppard (1992) reviewed As phytotoxicity and found reports listing widely different As concentrations which resulted in plant yield reductions. For example, [is less than] 25 mg/kg of As in soil was shown to reduce bean yield by 14%, whereas another study found 414 mg/kg of As reduced bean yield by 30%. Oats at 10 mg As/kg suffered a yield reduction of 22%, but at 540-850 mg As/kg in other soils, minimal (5%) yield reductions were observed. O'Neill (1995) stated that the level of As in edible plants is generally low, often being close to the limit of detection, even when the crops are grown on contaminated land.
In an area of south-west England, 722 [km.sup.2] of land contains naturally elevated levels of As, with concentrations in excess of 110 and up to 892 mg/kg soil (Mitchell and Barr 1995; O'Neill 1995). Vegetables grown in home gardens in this area, with the exception of lettuce, did not exceed the recommended 1 mg/kg limit for As, indicating the low risk to food chain contamination by geogenically derived soil As.
High levels of As have been found associated with gold smelters in Canada. The vegetation that grew in these areas generally contained low concentrations of As, except where soil levels were above 1000 mg/kg, which either resulted in phytotoxicity or growth of tolerant genotypes (O'Neill 1995).
Yeates et al. (1994) measured pasture herbage yield and metal uptake on a silt loam soil with varying degrees of contamination by As, Cr, and Cu. Herbage yield was unaffected at 161 mg As/kg soil, but declined at higher As concentrations. It is, therefore, clear from the studies outlined above that As phytotoxicity is highly dependent on soil properties.
Speir et al. (1992) investigated the feasibility of using sawdust from As-, Cr-, Cu- preserved timber as a soil amendment. Beetroot, clover, and lettuce plants were grown in pots containing a silt loam soil amended with 10% treated sawdust, which resulted in an As concentration of 66 mg/kg in the pots. Levels of As in the plant leafy tissues and the beetroot bulb were no different in any of the treatments and were [is less than] 10 mg/kg; however, the fibrous roots accumulated high levels of As (100-380 mg/kg). The scientific data generated in New Zealand are similar to data elsewhere (Sheppard 1992) in that they indicate that using total soil concentrations of As to generate a single guideline value to protect plants or soil biota is not possible.
Stock and human health effects of arsenic
No data are available in Australia or New Zealand to assess risks to stock or humans from As in soils. In the study in south-west England, Mitchell and Barr (1995) reported that As toxicity in Cornish cattle was positively identified in the form of dysentery and respiratory distress. Uptake by these cattle may have been as high as 50 mg As/day, due to grazing on contaminated pastures, with 60-70% of the As entering inadvertently through soil ingestion.
The US EPA (1993) have set a biosolid and cumulative loading limit for As of 41 mg/kg (kg/ha). The critical pathway that leads to this value is `Pathway 3, Children Ingesting Sewage Sludge', and is derived from a threshold for non-cancer health risk of 0.0008 mg As/kg bodyweight.day and an ingestion rate of 0.2 g dry soil/day. The US EPA have recently published risk-based soil screening levels (SSLs) for contaminants in soils (US EPA 1996). These are described as a tool developed by the EPA to help standardise and accelerate the evaluation and clean up of contaminated soils at sites on the National Priorities List where future residential land use is anticipated. The decision to use the Soil Screening Guidance at a site will be driven by the potential benefits of eliminating areas, exposure pathways, or contaminants from further investigation. The generic SSL for As is 0.4 mg/kg. This figure is calculated from a cancer risk level of 1 in 1 000 000. The As value is by far the most stringent of all of the heavy metal SSLs quoted in the document and appears to be seriously at odds with US EPA biosolids regulations (Rule 503, US EPA 1993). This SSL value is also quite unrealistic as nearly all soils contain more than 0.4 mg/kg from natural sources. Almost equally unrealistic is the Danish soil quality criterion for As of 2 mg/kg, derived from application of an ecotoxicological model to data from amendment of soils with inorganic As salts, mostly under laboratory conditions (Scott-Fordsmand and Pedersen 1995).
Compared with the other metals, Cd is more mobile in soils, in relation to both leaching and availability to plants. Cadmium is less strongly sorbed by soils than Cu, Ni, and Zn (Tiller et al. 1984), but fortunately is usually present in lower concentrations in fertilisers, manures, and biosolids than the other metals (Williams 1978). The chemistry and availability to plants of Cd in agricultural soils have recently been reviewed (McLaughlin et al. 1996).
Cadmium and soil biological processes
Several databases have reviewed the impact of Cd on soil biological processes; however, a number of these have used the addition of soluble Cd salts to simulate pollution sources (Witter 1992; Scott-Fordsmand and Pederson 1995; Will and Suter 1995a, 1995b). A wide range of indices for biological activity in soil have been investigated: population size or growth, population diversity, respiration, enzyme activities, C and N mineralisation, [N.sub.2] fixation, and denitrification. These will not be reviewed here, and the reader is referred to the above reviews for data sources. McGrath (1999) recently reviewed the critical contaminant concentration criteria developed by the above authors and noted the existence of significant differences in critical concentrations, i.e. `lowest observable adverse effect concentrations' (LOAECs). Values for the 10th percentile critical concentration (the 10th percentile of the range of LOAECs) varied between the reviews by 1-2 orders of magnitude (Fig. 2). An important consideration in comparing the LOAECS from a range of different studies is that no single biological criterion is used. The data in Fig. 2 have been collated from studies assessing the effect of Cd on the population response of different microbial species, and a range of microbial enzyme activities and microbial processes (organic matter mineralisation, nitrogen fixation and cycling, microbial respiration, etc.). Different microbial species and microbial metabolic/biochemical processes will not respond in the same way to a given concentration of metal, even if metal concentrations are adjusted for bioavailability in soil. Even within a bacterial genus, sensitivity to metal toxicity can differ. Giller et al. (1993) demonstrated that the species Rhizobium meliloti was less sensitive to heavy metals than other rhizobial species such as R. leguminosarum and R. loti. This general point must be considered when critical concentrations based on different populations and processes are compared, and raises the need for a standard set of protocols to assess the ecotoxicity of metal contaminants to soil microbial populations and microbial functional processes.
[Figure 2 ILLUSTRATION OMITTED]
One area of investigation regarding the impact of Cd on soil biological processes that is the subject of some controversy is the impact of Cd in biosolids applied to land on populations of the symbiotic nitrogen-fixing bacteria R. leguminosarum bv. trifolii (McGrath 1999). Declines in numbers and the survival of an ineffective strain of R. leguminosarum bv. trifoIii on historic biosolid trials at Woburn in the UK (McGrath et al. 1988; Giller et al. 1989; Chaudri et al. 1992) have been considered by some authors to be due to the high concentrations of Cd in the soils (Smith 1996, 1997). Smith (1997) argued that the biosolids used in this historic trial contained high concentrations of Cd that would not be encountered in biosolids produced today (due to strictly enforced trade waste guidelines), and that because of this, Cd was the causative metal at this site. However, evidence of a similar decline in R. leguminosarum bv. trifolii numbers at a field biosolid trial at Braunschweig in Germany (Chaudri et al. 1993; McGrath et al. 1995) indicated that Zn, as opposed to Cd, wag the causative agent (McGrath 1999). The Braunschweig trial had similar elevated soil Zn concentrations to Woburn, but concentrations of Cd were at background values. The authors concluded that Cd was not the main ecotoxic metal at these sites and indicated that Zn was the critical element in causing the decline in rhizobial populations (Chaudri et al. 1993). McGrath and Chaudri (1999) argue that the study of Smith (1997) did not demonstrate an effect of Cd, but indicated a possible effect of Zn on the rhizobial populations.
With reference to the determination of a `threshold' value for Cd in soils receiving biosolids that protects the soil biota, McGrath (1999) suggested a total soil concentration of 10 mg/kg, below which there was little impact on soil microorganisms. However, it should be noted that this threshold has been determined to protect soil biota, and would not necessarily be protective of plant uptake, crop quality, etc.
Cadmium phytotoxicity and uptake by plants
Cadmium pollution without co-contamination by Zn or Cu is rare, as most of the sources of metals added to soils also contain these elements. In fertilisers the Zn: Cd ratio is close to 10, while in biosolids the values are in the region 100-500. Cadmium, therefore, is unlikely to reach phytotoxic levels in agricultural soils before Zn does. However, human and animal health are threatened at Cd concentrations in the plant well below phytotoxicity thresholds (Chaney and Oliver 1996).
Transfer of trace concentrations of Cd to crops and animals has been well documented in Australia and New Zealand (Merry 1988; Oliver et al. 1993a, 1993b, 1994; McLaughlin et al. 1994, 1997b; Roberts et al. 1994; Jinadasa et al. 1997). Transfer of Cd from soils to plants was recently reviewed by McLaughlin et al. (1996) and will not be discussed again here. However, it is worth noting that violations of the food standard occur in soils with total Cd concentrations [is less than] 0.25 mg/kg, and that no (or weak) relationships have been demonstrated in Australia between total Cd concentrations in soil and concentrations in crops (McLaughlin et al. 1994; 1997a).
Stock and human health effects of cadmium
Again, Cd is unlikely to reach zootoxic concentrations in agricultural soils before concerns are raised regarding transfer of Cd to the food chain. The high concentration of Cd in offal from grazing animals was identified as a concern over 10 years ago (Langlands et al. 1988), and since then our understanding of factors leading to Cd accumulation in grazing stock has advanced (Lee et al. 1996). Older animals have higher Cd concentrations than younger animals, and there is currently a voluntary ban in South Australia and eastern Australia on offal for human consumption from animals of age [is greater than or equal to] 4 years. In New Zealand, the meat industry now automatically rejects the kidneys of slaughtered sheep over 2.5 years of age (Roberts et al. 1994).
Risks to human health from food chain Cd in Australia are controlled by the Australian food standards code (ANZFA 1997). Food Cd limits have been developed with regard to the Provisional Tolerable Weekly Intake (PTWI) published by the Food and Agriculture Organisation and the World Health Organisation, with the level currently being 7 [micro]g/kg bodyweight. Through the Australian Market Basket Survey (AMBS) and dietary modelling performed by the ANZFA, Cd intakes for various population groups in Australia have been calculated, with all currently well below the PTWI (Stenhouse 1992). Nevertheless, Cd is continually being added to many soils in excess of the amounts removed in produce, so that long-term control of Cd in agriculture may require controls on inputs, or on production of certain crops on particular soils prone to high soil-plant transfer of this element.
Copper, nickel, and zinc
As noted above, these elements are present in phosphatic fertilisers, but at low concentrations unlikely to raise soil concentrations to phytotoxic or ecotoxic levels. However, higher loadings of these metals to soil may occur in soils receiving biosolids and, along with Cd, are of most concern. Copper, along with Pb and As, may also be present in pesticide formulations, and significant accumulation of these elements has been documented in orchard soils in Australia (Merry et al. 1983).
Copper, nickel, and Zinc, and soil biological processes
Again, only few data are available in Australia or New Zealand to assess risks to biota in agricultural soils from these elements, with most data derived from overseas sources. Scott-Fordsmand (1997) has reviewed the toxicity of Ni to soil organisms and processes and concludes (for Ni) by stating `the majority of research reports used in this review are based on experiments involving the addition of soluble nickel salts to soils, making Ni more available than would be expected in the field situation'. Unfortunately, exactly the same can be said about the other heavy metals, such as Cu and Zn.
Toxicity of Cu and Zn is almost always greater than that of Ni (e.g. Juma and Tabatabai 1977; Tabatabai 1977; Liang and Tabatabai 1978; Al-Khafaji and Tabatabai 1979; Yadav et al. 1986; Doelman and Haanstra 1986, 1989). Scott-Fordsmand (1997) tabulated NOECs and LOECs (no-observed-effect concentrations and lowest-observed-effect concentrations) derived for Ni from a number of studies, all of which have used metal salts as soil amendments. Similar to Sheppard's (1992) findings for As, NOECs and LOECs overlapped considerably across the range of soils and measurement conditions reviewed. For example, NOEC values for Ni ranged from 30 to [is greater than] 1000 mg/kg and LOEC values from 10 to [is greater than] 7000 mg/kg. This again highlights the difficulties in determining a single value trigger or limit values based on different investigations using a range of `target' biological responses to metals and using total metal concentrations, in a range of soils with contrasting physico-chemical characteristics. Due to the much higher solubility, bioavailability, and hence toxicity of heavy metal salts, artificial contamination with salts is now considered inappropriate for assessing the environmental impacts of biosolid application to land (Logan and Chaney 1983; Chaney et al. 1987; Sommers et al. 1987). However, the investigation of toxic effects on soil microbial processes attributable to heavy metals from biosolid application to land is considered by some authors to be complicated by the biosolid being simultaneously contaminated by several metals. This can be overcome when biosolids are selected that have high concentrations of a single metal, i.e. high Cu or high Zn such as in the study of Smith (1997), or in case of the Braunschweig study previously referred to (Chaudri et al. 1993; McGrath et al. 1995), where Cd levels were near background and Zn levels were elevated in soils receiving a high Zn but low Cd biosolid.
McGrath et al. (1995) list LOAECs for individual heavy metal effects on soil microbial biomass and on a Rhizobium species or clover yield (Table 17). However, there is some controversy over the interpretation of the results of the various trials. Smith (1996) states that experimental trials, such as Luddington and Lee Valley (Table 17), have received very high rates of highly metal-contaminated biosolid to increase the soil metal content in a single treatment or in a very limited number of applications. Smith considers that this approach may have overestimated the availability and potential toxicity of heavy metals in soils compared with `normal operational practice' based on assumptions made by Logan and Chancy (1983) that large one-off applications of biosolids can lead to increased metal toxicity. However, Chancy and Ryan (1993) noted that single large applications of biosolids only temporarily overestimate metal bioavailability, and reported the results of studies demonstrating that within 2 years of application there were no increased phytotoxic effects due to increased metal bioavailability. A study by Dowdy et al. (1978) compared the metal content of snap beans (Phaseolus vulgaris L.) in soils receiving biosolid applications spread over a 3-year period with equivalent biosolid loadings in a single application. Their results showed no difference in the metal content in leaf tissue in soils that had received biosolid applications over 3 years or soils that had received one-off applications (3 years after biosolids were applied). It therefore appears that whilst the concerns of Smith (1996) that one-off applications of biosolids overestimate metal bioavailability may be valid in the short term, within 1-2 years there is no difference between `operationally' applied annual biosolid loadings, and large one-off biosolid loadings.
Table 17. Minimum concentrations (mg/kg soil) of metals in soils at which biological properties are affected (adapted from McGrath et al. 1995)
Experimental site Ni Cu Zn Metal concentration affecting clover or Rhizobium leguminosarum biovar. trifolii Woburn, UK 22 70 180 Gleadthorpe, UK 150 281 Braunschweig 1, Germany 15 48 200 Braunschweig 2, Germany 11 27 130 Metal concentration affecting soil microbial biomass Woburn 22 70 180 Luddington, UK 150 281 Lee Valley, UK 384 857 Ultuna, Sweden 35 125 230 Braunschweig 1 23 102 360 Braunschweig 2 24 111 386
The results from the Woburn and Braunschweig biosolid application field studies that strongly indicated the impact of Zn in soils on Rhizobium populations were subject to independent review by the UK Ministry of Agriculture Fisheries and Food (MAFF) as part of a review of the rules for sewage sludge application to agricultural land (MAFF/DoE 1993). The independent review noted that `Since recent research shows that deleterious effects on microorganisms can be observed at Zn concentrations of 250 mg/kg and sometimes lower, a soil limit value for Zn of 200 mg/kg for all soils of pH 5.0-7.0 (reduced from the previous limit of 300 mg/kg) was recommended' (MAFF/DoE 1993).
In comparison to the UK approach, the US EPA (1993) analysis of risks from biosolid-derived metals to soil biota (Sewage Sludge-Soil-Soil Biota pathway) screened out all metals from the evaluation except Cu. The study by Hartenstein et al. (1981) was used to develop a NOEC of 1500 mg Cu/kg soil (a cumulative pollutant loading of 2900 kg Cu/ha) using earthworms (Eisenia foetida) as the receptor species. However, the Technical Support Document for the Land Application of Sewage Sludge (US EPA 1992) states that `there is no evidence that earthworms are the most sensitive species, but due to the lack of data, metal loading criteria have been developed using earthworm data'. In developing Rule 503, US EPA discussed the apparent effects of other heavy metals on soil microbial activity at the Woburn site, noting that similar studies in the US had not been able to repeat the UK results. The US EPA concluded that further research was needed to provide a meaningful interpretation of microbial metal toxicity observations in relation to setting limits for biosolid-treated agricultural land (US EPA 1993).
The LOAECs shown in Table 17 are generally significantly below the current Australian and NZ guideline values for metals in soil, and indeed are well below background values for some elements for Australian soils (Table 16). However, the data in both Tables 16 and 17 are for total metal concentrations in soil, and lack of consideration of metal bioavailability results in the paradox that LOAECs based on total metal concentrations in soil are sometimes below background values. This will be discussed further below.
Copper, nickel, and zinc phytotoxicity and uptake by plants
Copper, Ni, and Zn can be taken up from soil by plants to potentially phytotoxic levels (Chaney and Oliver 1996). There have been several reviews of the literature with regard to phytotoxicity of metals (Chang et al. 1992; Will and Suter 1995a; 1995b; Scott-Fordsmand 1997), but much of the experimentation has been conducted in nutrient solutions or in soils amended with metal salts. Some experimentation investigating phytotoxicity due to application of biosolids has been reported (Johnston et al. 1983; Chang et al. 1992; Berti and Jacobs 1996), but generally only where metal loading rates have been well above the relevant guideline levels. For example, Johnston et al. (1983) observed little or no evidence of Cu or Zn toxicity in a field trial from single applications of up to 213 kg Cu/ha and 1610 kg Zn/ha in biosolids. Berti and Jacobs (1996) reported severe toxicity of Zn and/or Ni to maize (Zea mays), sorghum (Sorghum bicolor), sudangrass (Sorghum sudanese), and soybeans (Glycine max) from cumulative metal loadings of 2100 kg Ni/ha and 11 300 kg Zn/ha, well above US EPA limits. Hyun et al. (1998) reported results from a field trial in California where cumulative loadings of 1370 kg Cu/ha, 10 234 kg Zn/ha, and 720 kg Ni/ha in biosolids had no effect on growth of Swiss chard (Beta vulgaris), lettuce (Lactuca sativa), radish (Raphanus sativus), turnip (Brassica rapa), and carrot (Daucus carota). In this experiment, the initial soil pH was 6.0 and the CEC was 8.5 [cmol.sup.+]/kg (Chang et al. 1982), so that phytotoxicity could have been expected. The US EPA (1993) soil limits for Ni, Cu, and Zn (Table 8) have been derived using phytotoxicity as the most critical risk pathway. The US EPA limits are based on the most protective of two approaches, a probability-based approach and a response/rate approach. The review of Chang et al. (1992) described the former methodology. The latter approach used a critical (phytotoxic) plant metal concentration combined with a plant metal uptake response to increasing biosolid metal loading (derived from field experimentation) to calculate a threshold metal loading for soil. These approaches were recently criticised (Schmidt 1997) because of the lack of recognition of inter-species sensitivity to metals (maize was used as the receptor species for the probability-based threshold), the use of a 50% yield reduction as the threshold, and the use of a single biosolid loading limit for all soil types. The first two of these criticisms only apply to the probability-based approach as the response/rate methodology used lettuce as the (sensitive) receptor species and used a LOAEC rather than a 50% yield reduction.
US EPA (1993) and Smith (1996), after reviewing the literature on the phytotoxicity of biosolid-borne heavy metals, concluded that phytotoxicity has only been observed in glasshouse pot or field trials when high-metal biosolids have been applied at high loadings to soil or when soils are acidic ([pH.sub.(water)] [is less than] 5.5). While these experimental approaches to estimate appropriate soil limits for heavy metals would appear to aggravate potential phytotoxic effects, they provide a highly conservative estimate of the effects of potentially toxic metals on crop yields. It has also been argued that soluble salts should not be used to simulate the effects of biosolid-borne heavy metals in soils because the metals have much higher bioavailability (Logan and Chaney 1983; Chaney et al. 1987; Sommers et al. 1987). However, there is no evidence from long-term trials that soils are able to convert a large portion of the metals added in biosolids to insoluble, unavailable forms (McBride 1995). Nor is there evidence of increased availability of metals with time (McBride 1995; Chang et al. 1997), but this possibility cannot be completely discounted, especially when the requirement is to protect soils for sustainable use in perpetuity. Taking a precautionary approach, it may well be argued that the data from soil amendment studies using soluble salts may indeed afford the best measure of such long-term protection.
However, all the above methods suffer from the common problem of failing to recognise the importance of soil pH, clay content, organic matter content, and mineralogy in controlling metal availability in soil. It is, therefore, unlikely that biosolid-metal loading rate, or rate of soluble metal addition, will have a good relationship to phytotoxic effects, due to the variety of
reactions that metals undergo in soils and the range of soil characteristics which can mitigate plant uptake of metals from soil. For example, in Palmerton, Pennsylvania, USA, emissions from a Zn smelter have greatly elevated the soil concentrations of Zn, Cu, Cd, Pb, and Ni. In an assessment of possible health effects, it was found that, with frequent liming and additions of humus to the soils, vegetables could be grown successfully which were safe for human consumption (Baker and Senft 1995). At Port Pirie in Australia, potential adverse effects from metals dispersed from the smelter to adjacent farmland were not fully realised due to the alkaline nature of the surrounding soils reducing soil-plant transfer of metals (Cartwright et al. 1976; Merry et al. 1981).
Chromium, mercury, and lead
These elements are unlikely to reach phytotoxic concentrations in most soils. Concentrations in fertilisers are generally too low to be of concern, but soils receiving wastes such as biosolids, tannery, and other industrial wastes may accumulate high concentrations of these metals.
Chromium exists in soils in two oxidation states: [Cr.sup.III] as a trivalent cation and [Cr.sup.VI] as a polyvalent anion. The former is strongly retained by soils and is relatively non-toxic, while the latter is mobile and phyto- and zoo-toxic at low concentrations in solution ([is less than] 10 mg/L). Studies relating the impact of Cr on soil biota have almost always used metal salt addition to soil to test toxicity hypotheses (Will and Suter 1995a). The study by Ross et al. (1981) investigating the impact of Cr salts on soil respiration supports the [Cr.sup.III], [Cr.sup.IV] toxicity hypothesis. A concentration of 100 mg/kg of [Cr.sup.III] reduced soil respiration by 41%; however, a 10 mg/kg soil concentration of [Cr.sup.VI] reduced respiration by a similar magnitude. The form in which Cr is added to soil is therefore critical in determining toxicity. Chromium in biosolids is mostly in the [Cr.sup.III] form, and there are no reports of significant oxidation to [Cr.sup.VI] in biosolid-amended soils. Rates of soil-plant transfer of biosolid Cr appear low (US EPA 1993), and this is the reason that the guidelines for biosolid reuse in agriculture in South Australia omitted a guideline value for [Cr.sup.III]. Soils where tannery wastes are applied, or CCA from wood preservation, may have very high concentrations of [Cr.sup.III], but again oxidation to [Cr.sup.VI] appears minimal, except where environmental circumstances fit together in a very narrow and delicately balanced optimum (Bartlett 1997). In their investigation of contamination of agricultural land by CCA from a timber treatment plant, Bardgett et al. (1994) and Yeates et al. (1994) found no evidence of residual [Cr.sup.VI] remaining in the soil (unpubl. data). Chromium concentrations may also be high in Australian soils due to high concentrations in underlying parent materials (Lottermoser 1997), but there are no data to assess if non-indigenous microorganisms or plants are affected by these concentrations. Certainly, Ferrosols are extremely productive agricultural soils in Australia providing sufficient P is added to address the high P retention capacity of these soils.
The behaviour and plant uptake of Hg and Pb in fertilisers and in soils were recently reviewed by McLaughlin et al. (1996). These elements are very strongly retained by soil and are usually present in wastes and fertilisers at concentrations that are unlikely to pose environmental risks. Some industrial wastes may contain high Hg or Pb concentrations, which could significantly elevate soil concentrations, but the major risk from these elements is to human health through soil ingestion (by children) in urban areas (US EPA 1993).
Regulatory philosophies for controlling metals in soils
The primary rationale for establishing regulations for metal levels in soils is to limit or prevent the exposure of organisms to unacceptable hazards. Superficially this may appear to be a straightforward objective. However, there are 3 major complicating factors: (1) organisms differ in their sensitivity to different metals, (2) pathways of exposure to metals vary, and (3) soil properties and properties of the material contaminated with metals (fertilisers, biosolids, etc.), in addition to other environmental factors, strongly influence the degree of exposure (bioavailability) at any given total metal concentration in soil.
Current philosophies for deciding which target organisms should be protected by soil regulations and for determination of pathways of exposure to metals have been reviewed comprehensively (Chaney and Ryan 1993; McGrath et al. 1994; Cook and Hendershot 1996) and will only be discussed briefly here.
The most precautionary approach, which is favoured by some northern European countries including Sweden (Swedish Environmental Protection Agency 1994), is maintenance of the status quo in terms of total metal concentration in soil, whereby any inputs of metals to soils should not exceed outputs, e.g. by removal in crops. Clearly, compliance with such an approach ensures protection of even the most metal-sensitive organisms (provided the soil is not already contaminated). However, its feasibility is restricted in many countries due both to the presence of contamination sources which have yet to be identified and/or minimised, and to economic/cultural considerations. For example, addition of P to the soil is essential for sustainable agriculture in Australia and New Zealand (and many other countries), yet most sources of fertiliser P contain significant amounts of Cd which are not (currently) economical to remove during the manufacturing process (McLaughlin et al. 1996). As plants take up only a small percentage of the Cd added to soil in phosphatic fertilisers and leaching of Cd through the soil profile is minimal (Holm et al. 1998), concentrations of Cd in many fertilised soils are increasing. It has been calculated that the Cd load in agricultural soils in Australia is increasing by 1-5 g/ha.year (McLaughlin et al. 1996) due to additions of phosphatic fertilisers. In New Zealand pasture soils, Cd accumulation rates between 3 and 13 g/ha.year have been calculated (Loganathan et al. 1997; Gray et al. 1999a). Similarly, many Australian and some New Zealand soils are Zn deficient, hence Zn is added to the soils in the course of normal agricultural practice. It is possible that such an action may have a deleterious impact on indigenous soil microorganisms, which have adapted to conditions of low Zn; however, addition of Zn is essential if crops are to be produced on these soils--again the choice of the target indicator for adverse effect is important.
The alternative philosophy, adopted by many European countries and the USA, is to allow concentrations of metals in soils to increase above current levels, but regulate concentrations of metals in soils to levels that will maintain environmental health for agricultural purposes and also avoid any off-site impacts due to contaminant movement. This strategy is often termed `effects-based regulation'. With this strategy it is possible that highly sensitive soil organisms are not protected. However, as long as this is not detrimental to the overall biological function and agricultural productivity of a soil, including the ability to produce crops/livestock containing metal concentrations below those which may affect the health of human or animal consumers, the regulations are deemed effective. Whilst both the European and US regulations are `effects based' there is a major difference in the philosophy of determining an `effect'. The European approach is strongly based on observed impacts of metals, and this is highlighted in the revision of the UK sludge Zn limit based on observed effects on Rhizobium discussed earlier. In contrast, the US EPA approach is very much based on detailed risk pathway modeling, determining 14 potential pathways of exposure to metals to different organisms and establishing limits which would protect the most sensitive organism in the pathway (US EPA 1992). However, it has been argued that a lack of data on organism response to metals and transfer coefficients for metals from soils to organisms resulted in humans being erroneously selected as the most sensitive organism with respect to most metals, and that phytotoxic effects as well as toxicity to microorganisms could be expected at metal concentrations significantly lower than the established metal limits (McGrath et al. 1994; McBride 1995). Other concerns relating to off-site movement of metals in soils amended with biosolids have also been raised recently (McBride 1998; McLaren et al. 1999b). There are, therefore, huge discrepancies between Europe and the USA in terms of maximum concentrations of metals allowed in agricultural soils amended with biosolids, with almost an order of magnitude difference in concentrations of most contaminants, except Pb. In general, European limits have been used as a basis for deriving maximum permissible concentrations (MPC) of metals in biosolid-amended soils in several States in Australia and in New Zealand.
A major consideration for regulatory authorities is that contamination of agricultural soils by metals is effectively irreversible. In most soils, metals are highly immobile and only a small percentage of the total content of metal in a soil is removed through crop uptake, leading to residence times of the order of thousands of years (McGrath 1987). While strategies do exist for remediating metal-contaminated soils (Burns et al. 1996), they are extremely expensive and thus only suited to small volume, high value soils, usually in urban areas. Decontamination of agricultural soils is not feasible given current technology, although there are some management practices (e.g. liming, salinity reduction, see McLaughlin et al. 1996) which can reduce the bioavailability of metals in agricultural soils. Therefore, if the philosophy of allowing total metal concentrations in soils to increase is to be adopted, it is critical that:
(a) regular review of the regulations be undertaken;
(b) research into the long-term cycling and bioavailability of metals in soils be encouraged; and
(c) bench-mark sites or regular monitoring programs be undertaken to assess possible adverse effects.
Metal bioavailability in soils as a principle for regulatory control
The soil quality criteria established by the USA and the European Community are based on assessment of total metal concentrations in soils within a range of soil pH levels, recognising that pH is the major factor affecting metal bioavailability. Whilst some European States set metal limits based on pH criteria (UKDoE 1989), there is no pH rule in the US biosolid guidelines (US EPA 1993). However, it is assumed that agricultural soils where biosolids are applied will be limed to pH 6.5-7.0. A close examination of the US EPA Sludge Rule Technical Support Document (US EPA 1992) reveals that few of the field studies reviewed by the regulatory authorities in determining the extent of potential metal exposure pathways, most notably soil--plant transfer, were carried out on soils outside of this pH range. While the procedure for determining total metal concentrations in soils is relatively simple, this measure does not reflect the fact that many soil properties other than pH, and a variety of other environmental factors unrelated to soil type, greatly influence the bioavailability of metals in any given soil. Furthermore, small changes in soil pH can have a large influence on metal availability.
Metal bioavailability to plants and many soil microorganisms is a function of the bioavailability of dissolved metal species in the soil solution and the ability of the soil to buffer metal concentrations in the soil solution. These 2 components have been described, respectively, as the Intensity (I) and Quantity (Q) factors (Tiller et al. 1972), and the relationship between these, the buffer power (C). Factors affecting metal availability through modifying I and Q are discussed below.
Factors affecting metal intensity (I)
The bulk soil characteristics which influence the distribution of metals between the solid and solution phase include the concentration of metals in the soil (Hendrickson and Corey 1981), soil pH (Dijkshoorn et al. 1981; Anderson and Christensen 1988), CEC (King 1988), and the nature of exchange sites (Zachara et al. 1992; Naidu et al. 1997). The mechanisms by which these different characteristics influence metal concentration in soil solution are reasonably well understood and will not be reiterated here. However, as discussed above, most studies on metal bioavailability have been conducted with soils from temperate zones which tend to be dominated by permanent charge minerals in cooler and wetter environments. It is not appropriate to simply adopt the results of these studies as the basis for determining limit values for Australian and New Zealand soils, which generally contain significant amounts of variable charge components, may be saline, and have soil temperatures that are often higher, and where wetting and drying cycles markedly affect soil chemical, physical, and biological properties. Further studies are required to determine how these factors affect metal bioavailability.
In contrast to the bulk soil characteristics which influence I, much less is known about the relationship between I and biological responses (uptake/toxicity for plants and microorganisms). Many reports have suggested that only the uncomplexed, or `free ion', metal species in solution are the forms which are available for uptake by biota. This implies that the presence of metal binding ligands in the soil solution act to reduce metal bioavailability, which has been demonstrated in several studies (Halvorsen and Lindsay 1977; Checkai et al. 1987). However, more recent studies have suggested that organisms may also be able to absorb some metal complexes such as chloro- and sulfato-complexes from solution (Smolders and McLaughlin 1996; McLaughlin et al. 1998), and metal-organic chelates (Phinney and Bruland 1994; Campbell 1995; Vercauteren and Blust 1996; McLaughlin et al. 1997c). The efficiency of uptake of these metal species may be lower than for the free ion. However, the presence of such ligands in soils increases the total concentration of metals in solution by altering the equilibrium of metal distribution between soil and solution phases. Hence, even though the rate of uptake of the complexed metal species may be reduced compared with the free ion, the increase in total concentration of the metal in soil solution may be such that overall uptake is enhanced compared with soils in which the ligand is not present. Indeed, excessive uptake of Cd by potatoes and oilseeds growing in saline soils which, according to total element analysis, do not contain high concentrations of Cd, has been attributed to the high chloride concentrations in soil solution leading to significant Cd complexation (Li et al. 1994; McLaughlin et al. 1994; 1997b). Similarly, phytoremediation can be enhanced by the application of chelating agents to soil to increase uptake of metals (Blaylock et al. 1997). The ionic strength of the soil solution can affect bioavailability of metals. While increasing ionic strength reduces the solution activity of free metal ions at constant solution concentration, high concentrations of cations such as Ca or Na in solution displace metals from exchange sites on soils (Andersson 1976a; Tiller et al. 1979; Bingham et al. 1983; Christensen 1984) or from complexes with dissolved organic compounds (DeWit 1992; Hamon et al. 1995), leading to increased concentrations of free metal ions in solution and thus potentially enhancing metal availability. Changes in the concentration of major cations in soil solutions occur in agriculture when soils are limed or fertilised (Andersson 1976a; Lorenz et al. 1994), but the significance of these reactions in the field is yet to be demonstrated. Certainly, addition of P fertiliser (without Cd) stimulates Cd uptake, but this effect is more likely related to enhancement of root proliferation in Cd-contaminated soil layers (Williams and David 1977).
High concentrations of cations in solution may also reduce the bioavailability of undesirable metals through competition at the soil solution-(biological) membrane interface. For example, studies in solution culture have shown that Ca strongly inhibits the uptake of some metals by plants with maximum inhibition for Zn occurring at Ca concentrations around 400 mg/L (Chaudhry and Loneragan 1972; Hamon 1995). As reported Ca concentrations in soil solution are variable and often [is less than] 400 mg/L (Kinniburgh and Miles 1983; Lorenz et al. 1997), they may have a large effect on metal bioavailability, but this needs further quantification. At the root surface, inhibition of uptake of one heavy metal by another is also evident, with studies showing that Ni, Cu, and Zn are mutually competitive (Clarkson and Luttge 1989). There is also evidence of an interaction between Zn and Cd, with Zn additions to Zn-deficient soil leading to a reduction in the Cd content of wheat grain (Oliver et al. 1994) and young lettuce and spinach leaves (McKenna et al. 1993), but only, for the latter cases, when the concentration of Cd was relatively low. The mechanisms underlying the Zn/Cd interaction are still unclear.
The availability of soil nutrients can also affect metal concentrations in soil solution and hence bioavailability. For example, some plants release phytochelatins or organic acids into the rhizosphere in response to iron or P deficiency. These organic compounds not only increase the concentrations of Fe or P, but can also solubilise Zn and other metals (Zhang et al. 1989; Chairidchai and Ritchie 1992), thereby enhancing metal bioavailability. Increased uptake of Cd by crops grown in rotation with lupins has been partly attributed to acidification of the soil by lupins (Oliver et al. 1993b), which release large concentrations of citrate from their specialised proteoid root systems in order to enhance P acquisition (Braum and Helmke 1995). Similarly, as there is no literature information on the effects of nutrient deficiencies in biosolid-amended soils, it is interesting to speculate upon the consequences to metal bioavailability when, for example, the P value of a biosolid application has diminished over time. Will metals become more available to subsequent crops due to plants releasing organic exudates in order to sequester additional P? If fertilisers are added to counter P deficiency, will increases in metal availability still occur as a result of the ion exchange processes described above?
Factors affecting metal quantity (Q)
The capacity of the solid phase metal reserves and the rate at which these reserves can replenish metals removed from solution by plants, through gaseous losses (As, Hg) or by leaching, provide further controls on metal bioavailability in soils. In its simplistic form, the capacity factor (Q) is simply the total metal concentration of the soil, so that the only factor affecting the size of Q is the mass balance for metals, i.e. inputs-outputs. However, the rate at which metals can be supplied to the soil solution depends on both the physical and chemical characteristics of metals in the solid phase, so that the effective size of Q may be much smaller than the total metal concentration. For example, if a soil is contaminated with metals in a fused siliceous material that is highly insoluble and does not react to any extent with the soil solution, total metal concentrations may be increased, but the effective size of Q is unchanged.
Metals associated with the exchange complex on the soil solid phase are in direct equilibrium with soluble metals in the soil solution and therefore constitute the most readily bioavailable fraction of Q. Studies on metal sorption kinetics have reported short-term hysteresis effects (i.e. the inability to desorb all of the adsorbed metals), which has been attributed to the existence of specialised sites that form inner-sphere complexes with the sorbed metal (Hendrickson and Corey 1981). This type of sorption appears to be important at low metal loadings of the soil, and to progressively decrease in importance as metal loading (pollution) increases (Nakahone and Young 1993).
However, metal hysteresis effects have been shown to increase as a function of time and temperature (Barrow 1986). This longer term and/or high energy effect may be due to metals moving away from the exchange complex and into other solid phase fractions. Such processes could have a large impact on Q, as significant quantities of metals are frequently found in associations with solid phase components that render them either partially or completely incapable of participating in equilibrium reactions with the soil solution and, thus, are essentially non-bioavailable in the short term. While the nature of such associations is not clearly defined, it is thought that metals may be incorporated within the matrix of soil particles through diffusion (Barrow 1986), occluded in (hydr)oxides, form extremely stable complexes with insoluble organic compounds (Kerndorff and Schnitzer 1980; Beckett 1989; Swirl and McLaren 1991), or be contained within micropores that are not directly accessible to the soil solution (Fischer et al. 1996; Gerth et al. 1992). It should be noted that strong acid dissolution to determine `total' metal concentrations in soil would liberate metals from most of these fractions, thereby overestimating the amount of bioavailable metal.
The impact of time, or ageing processes, needs to be carefully considered when defining metal limits for soils, as the total metal concentration of the soil as determined by a strong acid digest may not change over time, but metal bioavailability can. It is well established that the bioavailability of soil Zn can decrease over time (Barrow 1986; Brennan 1990). More recently it was demonstrated that Cd can exist in non-bioavailable pools in at least some soils and that the increase in this pool size may also be a time-dependent process, but one which occurs at a very slow rate (Hamon et al. 1997a, 1997b). In addition, for some New Zealand soils it has been shown that the ability of sorbed Cd to desorb back into solution decreases with increasing contact time with the soil (Gray et al. 1998). Much more research needs to be conducted into the soil properties which influence temporal decline in the availability of metals. However, if the increase in size of non-bioavailable metal pools is related to the hysteresis effects for sorption that have been observed in laboratory studies, then the rate at which bioavailability declines is undoubtedly different for different metals. For example, Gerth et al. (1992) found that the ability to extract metals adsorbed on goethite and equilibrated for 21 days increased in the order Zn [is less than] Cd [is less than] Ni. This was postulated to be due to metal ions diffusing at different rates into micropores, where they become entrapped. Different rates of decrease in phytoavailability for different metals could have important implications for assertions that the bioavailability (to humans) of plant Cd from plants growing in soils polluted with both Cd and Zn is lower than that from plants grown in soils polluted with the single metal (McKenna et al. 1993). While this may be true in the short term, if soil ageing leads to a rapid decline in Zn phytoavailability with respect to Cd, then clearly any mitigating effect of Zn on Cd availability will be lessened over time.
Soil ageing may also lead to increases in metal bioavailability if metals are originally associated with organic phases that are subject to degradation in the long term. The potential for increasing metal availability over time in biosolid-amended soil has been considered by McBride (1995). However, apart from an initial enhancement in metal availability during the first year or two following application of biosolid, long-term increases in metal availability due to decomposition of biosolid organic matter have not been observed to date (Chang et al. 1997).
Superimposed upon the Q/I relationships are various environmental factors which can also influence metal bioavailability to individual species at any given time. Unexplained annual variations in metal uptake are frequently observed in field trials and may be due to climatic conditions affecting factors such as plant growth rates, root distribution in soil, soil temperature, moisture content, and salinity.
For metals where plant uptake is not limited by diffusion of metal through soil to the root, which may be the case for most metals in highly contaminated soils, it is likely that plant transpiration rates will influence bioavailability. The more a plant transpires, the greater the mass flow of soil solution to the root surface, hence the higher the concentration of associated metals that are transported to the root surface, potentially enhancing plant metal uptake and impacting upon the soil microorganisms which preferentially colonise the rhizosphere. The high rates of evapotranspiration in many regions in Australia may, therefore, provide a higher exposure of metals to plants, compared with cooler climates where mass flow of soil solution to the plant root is less. This hypothesis needs testing experimentally.
The depth of the contaminated soil layer can also influence metal bioavailability to plants. Experiments by Williams and David (1977) illustrated the importance of metal distribution in the soil on Cd availability to plants. Clover grown in pots in which added Cd was evenly distributed throughout the soil accumulated up to an order of magnitude more Cd than plants in which the same total Cd additions were confined to the top 2 cm of the soil. Use of pot experiments for determining metal bioavailability to plants has often been criticised (Bell et al. 1991) as giving unrealistically high estimates of metal bioavailability which appear not to be reflected in field trials (Naylor et al. 1987). It is postulated that the primary reason why metal availability in pot trials is greater than in field experiments is due to all of the plant roots being in the confined space of the pot and hence all are exposed to the metal-rich soil (Naylor et al. 1987; Chaney and Ryan 1993). However, the results of field trials may give equally misleading underestimates of metal bioavailability unless careful assessment has been made of the depth of metal contaminated soils and root penetration (McBride 1995; Hamon et al. 1999). Such values are rarely reported in the literature. To date there has been no attempt to include depth of contamination as a parameter in soil quality criteria regulations.
Methods for estimating metal bioavailability for regulatory purposes
From the above discussion, it is clear that a single measure of metal concentration in soil fails to adequately describe the risk from metal exposure. Metal availability depends on both Q and I, and the relationship between these. Soil pH is a critical factor affecting the relationship between Q and I. Soil salinity is important for Cd partitioning. Many different methods have been proposed for extracting `available' metals from soil, with all extracting a variable proportion of Q and/or I. For example, metals extracted by 0.01 M Ca[Cl.sub.2], 0.1 M Ca[Cl.sub.2], and 1.0 M [NH.sub.4][NO.sub.3] (Symeonides and McRae 1977; Sauerbeck and Styperek 1985; Whitten and Ritchie 1991; He and Singh 1993; Mench et al. 1994; Andrewes et al. 1996; Gray et al. 1999b) probably more closely reflect I, while extractants such as DTPA and EDTA (Lindsay and Norvell 1978; Clayton and Tiller 1979) and concentrated acids may give estimates closer to Q, depending on extraction time, chelate concentration, and soil:solution ratio. Few studies have been undertaken investigating soil tests for metals for a wide range of Australian or New Zealand soil types and/or validated under field/commercial conditions. McLaughlin et al. (2000) have recently reviewed this area of research.
In Switzerland and in Baden Wurtemburgh in Germany, selective extractions of soil have become the regulatory standard for controlling metal pollution (Hani 1996; Prue[Beta] 1997). The German DIN 19730 is based on extraction of soil with 1 M [NH.sub.4][NO.sub.3] solution and determination of As, Cd, Cu, Cr, Hg, Mn, Ni, Pb, and Zn in the extract. Soil limits based on the selective extraction are pH-dependent, recognising the importance of pH on metal bioavailability (Table 18).
Table 18. Background values (90th) percentile for extractable metals ([micro]g/kg) in soils (DIN 19730), as determined by 1.0 M [NH.sub.4][NO.sub.3] (from Prue[Beta] 1997)
Exceedance of these values indicates a potential risk of adverse effects and is used to trigger further investigation of the site
Soil pH Metal <4.0 4.0-4.5 4.5-5.0 5.0-5.5 5.5-6.0 As 60 50 40 40 40 Cd 80 50 20 15 10 Cr 50 40 15 12 10 Cu 300 280 250 250 250 Ni 1000 1000 600 300 250 Pb 3000 2000 150 30 15 Zn 5000 4000 3000 1000 300 soil pH Metal 6.0-6.5 6.5-7.0 7.0-7.5 >7.5 As 40 40 45 50 Cd 5 3 3 3 Cr 10 12 15 15 Cu 250 300 350 400 Ni 200 200 200 200 Pb 10 6 4 3 Zn 200 170 130 100
Clearly, it is important that regulations or guidelines protect potential risks as well as apparent (current) risks. Hence, a simple measure of I is inappropriate and some measure of Q is essential, as practiced in Baden Wurtenburgh (Prue[Beta] 1997). Indeed there are good arguments that extraction of soil with hot aqua regia (a quasi-total value) or exhaustive EDTA extraction (Clayton and Tiller 1979) should be used to determine Q. A technique is required that releases and determines metals from all the pools likely to influence I either presently, or in the future. Indeed, this was the rationale behind the development of a method involving the 7-day extraction of soil with EDTA solution (Clayton and Tiller 1979). This procedure was designed to extract all the `surface' metal in soil and to avoid extraction of metals firmly bound within crystal lattices, the latter being unlikely to be bioavailable even in the long term.
In addition to a determination of Q, we contend that some modifying factor needs to be applied to effectively assess risk. This can be accomplished in one of two ways. Firstly, Q may be modified by the use of adjustment factors, an algorithm approach. For example, ecotoxicity values for metals in the Netherlands are normalised to a standard soil of 25% clay and 10% organic matter content (van den Berg and Roels 1991). Equations are used to modify threshold values for any soil depending on its clay and organic matter content, as these were deemed to be the most important factors controlling metal behaviour in those soils. However, these modifying factors are derived from a study of background concentrations of metals in soils, and not from a database of modifiers of metal toxicity (Lexmond and Edelman 1987). They therefore have a weak scientific basis for modifying metal toxicity thresholds. Alternatively, it is possible to use empirical models of the type proposed by McBride et al. (1997) to estimate metal concentrations or activities in soil solution (I) from properties such as total metal concentrations, pH, and soil organic matter content. Variability in, and widespread applicability of, the relationships need to be resolved before these models could be reliably used for regulatory purposes. Secondly, a separate determination of I may be made, so that current and potential risks are predicted by soil measurements, an approach we favour. We suggest that hazard assessment through soil testing should be based on a 2-pronged approach, one extractant should estimate Q (potentially bioavailable) and the other should estimate I (currently bioavailable). The concentration of metal in the soil solution is too impractical to determine routinely, so a weak unbuffered salt solution is most likely to give the best estimate of I. Whether metal speciation needs to be considered (apart from As, Cr, and Hg) is a point of debate. As discussed by McLaughlin et al. (2000), evidence that consideration of free ion activities actually improve predictions of hazards from metals in soil is equivocal. Certainly speciation analysis is needed for those elements with strong redox couples (As, Cr, and Hg) where different toxicities depend on the form of the metal, which is then dependent on redox potential.
In the USA, total metal concentration data from selected studies on soils amended with different biosolids were used to evaluate metal bioavailability, with no assessment being attempted of the actual proportion of metals in bioavailable pools (US EPA 1993). The limits derived from these studies may be protective with respect to the specific biosolids and site characteristics examined in the studies. However, because percent bioavailable metal was unknown, but probably significantly less than 100% in the studies, the developed limits do not necessarily provide a protective benchmark that is applicable to all sites or biosolids. This is particularly important in light of new research showing that biosolid metal bioavailability can vary markedly depending on biosolid chemistry (Jing and Logan 1992; Rogers 1997; Rogers and McLaughlin 1999). For example, Chu et al. (1998) compared the metal distribution in biosolids from Hong Kong treatment works using a sequential extraction procedure. Biosolids from 2 treatment works employing identical treatment processes, but serving different catchments, had very different patterns of metal distribution in the sequentially extracted pools, especially in pools extracted by the least aggressive extractants (more likely to be the bioavailable pools).
It is also clear that total concentrations of metals in soils are a poor indicator of actual bioavailability and that bioavailability is likely to be greatly overestimated in most soils if total concentrations are assumed to be 100% available. This fact has long been recognised by the scientific community who have investigated alternative methods for assessing metal availability, principally phytoavailability, as follows.
A variety of methods have been used to collect soil solution (see Litaor 1988 for a review). Comparisons of concentrations of metals in the soil solution with metal uptake by plants have in some instances (e.g. Gerritse et al. 1983; Jopony and Young 1993; Hamon et al. 1995) but not always (Lorenz et at. 1997; McLaughlin et al. 1997a) yielded positive correlations. In a study of 10 contaminated soils by Lorenz et al. (1997), reasonable correlations were observed between soil solution Cd and plant uptake of Cd; however, good relationships were not found for Zn. It was hypothesised that the lack of a good relationship between Zn concentration in solution and plant uptake from the soils was due to soil properties leading to different rates of replenishment of Zn in the soil solution, or due to plant uptake of Zn being subject to metabolic controls and not simply responding linearly to Zn concentration in solution (Lorenz et al. 1997). For assessment of food chain (potato) Cd risks due to contamination of soil by Cd in phosphatic fertilisers, McLaughlin et al. (1997a) found a poor correlation between Cd concentrations or activities in soil solution and Cd in potato tubers. A potential problem with the use of soil solution as an index of metal availability is the temporal changes that may occur in field soils (Edmeades et al. 1985), which are likely to affect measured metal concentrations through ion-exchange reactions (Andersson 1976b).
There is, therefore, no compelling evidence that analysis of metal concentrations in soil solutions provides an accurate and precise index for hazard assessment due to metal contamination. Furthermore, the logistical difficulty of collecting soil solution precludes widespread adoption and use by commercial or regulatory agencies.
A variety of chemical extractants has been used to predict metal availability in soils. It should be noted that most studies have examined availability of Cd, Cu, Ni, Pb, and Zn; there is little information on the extractability of less common, but nevertheless potentially significant contaminants such as As, Ag, Cr, and Hg. The 3 major forms of extractants which have been tested are organic chelates, mildly acidic solutions, and neutral salt extractants. These types of extractants have been chosen as it is considered that they release into solution metals which are predominantly associated with the exchange sites on the soil solid-phase, and hence, as discussed above, the most bioavailable fraction of metals. The organic compounds most frequently used to extract metals are EDTA and DTPA. These soluble compounds are strong chelators of metals, and act by desorbing metal ions from exchange sites on the solid phase into solution. However, Lebourg et al. (1996), in a review of the extraction literature, concluded that both EDTA and dilute acid solutions had a tendency to extract metals from non-plant available pools of metals in the soil, hence overestimating phytoavailability. In effect, it appears that these extractants are tending to measure Q rather than I, and therefore are assessing potential toxicity rather than immediate toxicity. This concept is rarely recognised by researchers. O'Connor (1988) clearly indicated that the use of EDTA or DTPA may be limited in acid or polluted soils due to saturation of the complexation capacity of the chelates under these conditions. Nakahone and Young (1993) cautioned against the use of organic complexing agents to determine surface-adsorbed metals by stating that such agents are also likely to disrupt mineral adsorption surfaces, hence liberating unspecified fractions of soil metals.
A diversity of neutral salt extractants at different concentrations has been used in attempts to assess metal bioavailability. Historically, the most popular extractants have been Ca[Cl.sub.2], Ca[([NO.sub.3]).sub.2], Na[NO.sub.3], and [NH.sub.4][NO.sub.3] at concentrations ranging from 0.01 M to 1 M (McLaughlin et al. 2000). The cationic component of the extractant displaces metals from exchange sites on the solid phase and into solution. For chloride extractants, there will also be movement of Cd into solution through the formation of complexes with Cl (Hahne and Kroontje 1973). Researchers who favour the use of neutral salt extractants cite the fact that they do not greatly alter soil pH during extraction and that they closely approximate the natural ionic strength of the soil as good reasons for their use (Lebourg et al. 1996). However, as is the case with the other methods, mixed results have been obtained following the use of neutral salt extractants to predict metal availability (to plants). Lebourg et al. (1996) reported an extensive, but uncritical, list of regression coefficients from numerous studies comparing metals extracted by various neutral salt extractants to plant uptake. The variation in the listed [R.sup.2] values (for log-log plots) covers the range from significant relationships observed between plant uptake and extractable metal, to no relationship being apparent. As reviewed by McLaughlin et al. (2000) good correlations are generally found only in those studies where the number of soils assessed was small, where pollution sources were few, and where experimentation was performed under controlled glasshouse conditions.
Given the number of different factors that determine Q, I, and hence bioavailability, it is not surprising that inconsistent results are achieved with all of these methods, each of which essentially quantify only 1 or 2 of the factors, namely, total metal concentration in solution and/or amount of metals tentatively associated with solid-phase exchange sites. Nevertheless, mild extraction methods do provide a better indication of immediate metal bioavailability than total concentrations as the amounts of metals extracted are significantly influenced by the properties of the individual soils.
However, care must be taken before mild extractant methods can be relied upon as indicators of bioavailability. In contrast to total metal concentrations, there is the potential to underestimate bioavailability with these techniques, as there is no guarantee that they are capable of extracting all forms of bioavailable metals in all soils. Hence, the utility of these methods for assessment purposes would be dependent on a large number of calibration experiments being performed on the representative Australian and New Zealand soil types for a range of organisms. It may be valuable to adopt an approach similar to that of Prue[Beta] (1997) for calibrating extractant metal concentrations to derive limit values. Prue[Beta] (1997) began the calibration process by nominating an `action value', which was defined as the soil quality criterion indicating a site which should be monitored closely as crops grown on this soil may violate MPC of metals and the quality of water draining through such soil may be unacceptable for groundwater standards. A `threshold value' was also defined as the soil quality criterion at which the agriculture use of a soil should be restricted, or severely limited, due to a high certainty of problems associated with metal pollution. Prue[Beta] then compared plant uptake of metals against the concentration of metals extracted by [NH.sub.4][NO.sub.3] (1 M) from 400 paired soil/plant analyses. The action value was quantitatively defined as the [NH.sub.4][NO.sub.3]-extractable metal concentration that corresponded to 95% of the plants containing less than the German MPC of the element of interest. Threshold values were set at the [NH.sub.4][NO.sub.3] extractable metal concentration that corresponded to the MPC in plants being exceeded in 70% of cases.
Conclusions--how to incorporate bioavailability into regulations?
Too often the usefulness of soil extraction techniques are judged by the data they produce in isolation, or how model metal compounds are selectively dissolved. In practice, utility should be measured by how such techniques can improve predictions of plant uptake, adverse effects on human health, or eco- or phyto-toxicity due to metal pollution of soils. Examples of such applications are scarce.
Regulators and landowners need to know two things with regard to metal pollution of soil:
(1) what is the toxicity of the soil now? (more related to I)
(2) will the metals become toxic in the future? (more related to 2).
From the literature, it would appear that for Cd, Zn, and perhaps Cu soil tests based on determination of the most available metal pools, while not perfect, appear to better answer (1) than total metal concentrations or metals removed by strong extractants. The utility of considering free metal ion activities in soil extracts is questionable at this stage. However, metal bioavailability (and associated indices of bioavailability) can change due to changes in pH or decomposition of organic matter, and the second concern above is often relevant for environmental protection policies. Hence the reliance of many regulatory agencies on total concentrations of metal in soil--a conservative approach only when calibrated against toxicity studies using metal salts, which assumes that all of the metal in soil is potentially toxic.
As mentioned earlier, we suggest a 2-pronged approach. One is the the use of a dilute salt extractant to assess I and the factors that control I in the short term. The choice of extractant is a question which further research needs to answer, in a similar way to the research behind the DIN 19730 standard in Germany. The second is an assessment of effective Q, and here a complexing agent such as EDTA, an extraction in dilute acid, or other similar methods seem appropriate. These extractants will not attack a large proportion of metal strongly bound or occluded in mineral forms but will solubilise a large proportion of the surface-bound and moderately soluble solid phases. Again research is needed to refine the answers.
We believe such an approach would obviate the need in many cases to have site-specific risk assessments because a total metal concentration is reported which exceeds highly conservative guideline values based on total concentrations. Furthermore, this approach would help to minimise conflicts in the current regulatory framework for metal limits in agricultural soils based on total concentrations in soil. The approach is also adequately protective as it continues to provide a measure of safety by estimating potentially available metal in soil.
Finally we suggest that bioavailability based regulatory approaches should also address the issue of chemical forms and bioavailability of metals in the material being applied to soils, be it fertilisers, biosolids, or other waste materials. There is evidence to suggest (certainly in the case of biosolids) that the chemistry of metals in the material and the specific binding capacity of the matrix has a major role in controlling metal bioavailability in soils. Again research is required to further test this hypothesis.
The senior author thanks the Grains Research and Development Corporation for support during the period this review was compiled. We thank Phil Mulvey for his comments on a draft of the manuscript.
Al-Khafaji AA, Tabatabai MA (1979) Effects of trace elements on arylsulfatase activity in soils. Soil Science 127, 129-133.
Anderson PR, Christensen TH (1988) Distribution coefficients of Cd, Co, Ni, and Zn in soils. Journal of Science 39, 15-22.
Andersson A (1976a) On the determination of ecologically significant fractions of some heavy metals in soils. Swedish Journal of Agricultural Research 6, 19-25.
Andersson A (1976b) On the influence of manure and fertilisers on the distribution and amounts of plant-available Cd in Soils. Swedish Journal of Agricultural Research 6, 27-36.
Andrewes P, Town RM, Hedley MJ, Loganathan P (1996) Measurement of plant available cadmium in New Zealand soils. Australian Journal of Soil Research 34, 441-452.
Anon. (1997) `Australian standard[R] 4454--Composts, soil conditioners and mulches.' (Standards Australia: Homebush, NSW)
Anon. (1998) Draft guidelines for the assessment and management of contaminated land in Queensland. Queensland Government, Department of Environment, Brisbane.
ANZECC/NHMRC (1992) `Australian and New Zealand guidelines for the assessment and management of contaminated sites.' (Australia and New Zealand Environment and Conservation Council/National Health and Medical Research Council: Canberra)
ANZFA (1997) `Food standards code, Amendment 35.' (Australia and New Zealand Food Authority: Canberra)
ANZFA (1999) Contaminants in food--metals, Proposal P157. Australia and New Zealand Food Authority, Canberra.
ANZFA (2000) Draft Australia New Zealand Food Standards Code released. Australia and New Zealand Food Authority. Available at http://www.anzfa.gov.au/documents/mr07_00.asp (verified 11 Mar. 2000)
ARMCANZ (1995) Guidelines for sewerage systems--biosolids management, Water Technology Committee Occasional Paper No.1/95. Agriculture and Resource Management Council of Australia and New Zealand, Canberra.
Baker DE, Senft JP (1995) Copper. In `Heavy metals in soils.' (Ed. BA Alloway) pp. 179-205. (Blackie Academic and Professional: Glasgow)
Bardgett RD, Speir TW, Ross DJ, Yeates GW, Kettles HA (1994) Impact of pasture contamination by copper, chromium, and arsenic timber preservative on soil microbial properties and nematodes. Biology and Fertility of Soils 18, 71-79.
Barrow NJ (1986) Testing a mechanistic model. II. The effects of time and temperature on the reaction of zinc with a soil. Journal of Science 37, 277-286.
Barry GA (1997) Total heavy metal status of horticultural soils in Queensland. Final Report of Project VG 404. Horticultural Research and Development Corporation, Gordon, NSW.
Barry GA, Gardner EA, Bloesch P (1998) Studies on reuse of biosolids on Queensland agricultural lands. In `WaterTECH, Proceedings of the Australian Water and Wastewater Association Conference'. Brisbane, Qld. (CD ROM)
Bartlett RJ (1997) Chromium redox mechanisms in soils: should we worry about Cr(VI)? In `Chromium environmental issues'. (Eds S Canali et al.) pp. 1-20. (FrancoAngeli: Milan, Italy)
Beavers PD (1993) Guidelines for the use of biosolids--the Queensland scene. Water Journal 20, 23-26.
Beckett PHT (1989) The use of extractants in studies on trace metals in soils, sewage sludges, and sludge-treated soils. Advances in Soil Science 9, 143-176.
Bell RF, James BR, Charley RL (1991) Heavy metal extractability in long-term sewage sludge and metal salt-amended soils. Journal of Environmental Quality 20, 481-486.
Berti WR, Jacobs LW (1996) Chemistry and phytotoxicity of soil trace elements form repeated sewage sludge applications. Journal of Environmental Quality 25, 1025-1032.
Best EK (1992) Cadmium and lead in fertilisers. Unpublished Report ACU 92.09. Queensland Department of Primary Industries, Brisbane.
Bingham FT, Strong JE, Sposito G (1983) Influence of chloride salinity on cadmium uptake by Swiss chard. Soil Science 135, 160-165.
Blaylock MJ, Salt DE, Dushenkov S, Zakharova O, Gussman C, Kapulnik Y, Ensley BD, Raskin I (1997) Enhanced accumulation of Pb in Indian mustard by soil-applied chelating agents. Environmental Science and Technology 31, 860-865.
Braum SM, Helmke PA (1995) White lupin utilizes soil phosphorus that is unavailable to soybean. Plant and Soil 176, 95-100.
Brennan RF (1990) Reaction of zinc with soil affecting its availability to subterranean clover. II. Effect of soil properties on the relative effectiveness of applied zinc. Australian Journal of Soil Research 28, 303-310.
Burns RG, Rogers S, McGhee I (1996) Remediation of inorganics and organics in industrial and urban contaminated soils. In `Contaminants and the soil environment in the Australasia-Pacific region'. (Eds R Naidu et al.) pp. 411-450. (Kluwer Academic Publishers: Dordrecht)
Cameron KC, McLaren RG, Adams JA (1994) Application of municipal sewage sludge to low fertility forest soils: the fate of nitrogen and heavy metals. Transactions of the 15th World Congress of Soil Science 3a, 467-482.
Cameron KC, Di HG, McLaren RG (1997) Is soil an appropriate dumping ground for our wastes. Australian Journal of Soil Research 35, 995-1035.
Campbell PGC (1995) Interactions between trace metals and aquatic organisms: a critique of the free ion activity model. In `Metal speciation and bioavailability in aquatic systems'. (Eds A Tessier and DR Turner) pp. 45-97. (John Wiley and Sons: New York)
Cartwright B, Merry RH, Tiller KG (1976) Heavy metal contamination of soils around a lead smelter at Port Pirie, South Australia. Australian Journal of Soil Research 15, 69-81.
CEC--Commission of the European Communities (1986) Council directive (86/278/EEC) on the protection of the environment, and in particular of the soil, when sewage sludge is used in agriculture. official Journal of the European Communities 181, 6-12.
Chairidchai P, Ritchie GSP (1992) The effect of pH on zinc adsorption by a lateritic soil in the presence of citrate and oxalate. Journal of Soil Science 43, 723-728.
Chaney RL, Hornick SB (1978) Accumulation and effects of cadmium on crops. In `Cadmium 77: Proceedings of the 1st International Cadmium Conference. San Francisco'. pp. 125-140. (Metal Bulletin Ltd: London)
Chaney RL, Ryan JA (1993) Heavy metals and toxic organic pollutants in MSW-composts: research results on phytoavailability, bioavailability, fate etc. In `Science and engineering of composting: Design, environmental, microbiological and utilization aspects'. (Eds HAJ Hoitink and HM Keener) pp. 451-506. (Ohio State University: Ohio)
Chaney RL, Oliver DP (1996) Sources, potential adverse effects and remediation of agricultural soil contaminants. In `Contaminants and the soil environment in the Australasia-Pacific region'. (Eds R Naidu et al.) pp. 323-359. (Kluwer Academic Publishers: Dordrecht)
Chaney RL, Bruins RJF, Baker DE, Korcak RF, Smith, JE, Cole D (1987) Transfer of sludge-applied trace elements to the food chain. In `Land application of sludge: Food chain implications'. (Eds AL Page et al.) pp. 67-99. (Lewis Publishers Inc.: Chelsea)
Chang AC, Granato TC, Page AL (1992) A methodology for establishing phytotoxicity criteria for chromium, copper, nickel, and zinc in agricultural land application of municipal sewage sludges. Journal of Environmental Quality 21, 521-536.
Chang AC, Hyun H, Page AL (1997) Cadmium uptake for Swiss chard grown on composted sewage sludge treated field plots: plateau or time bomb? Journal of Environmental Quality 26, 11-19.
Chang AC, Page AL, Foster KW, Jones TE (1982) A comparison of cadmium and zinc accumulation by four cultivars of barley grown in sludge-amended soils. Journal of Environmental Quality 11, 409-412.
Chartres CJ (1995) Sodic soils: an introduction to their formation and distribution in Australia. In `Australian sodic soils: distribution, properties and management'. (Eds R Naidu et al.) pp. 35-40. (CSIRO: Melbourne)
Chartres CJ (1998) Land degradation processes and water quality effects: soil acidification. In `Farming action--catchment reaction, the effect of dryland farming on the environment'. (Eds J Williams et al.) pp. 158-161. (CSIRO Publishing: Melbourne)
Chaudhry FM, Loneragan JF (1972) Zinc absorption by wheat seedlings and the nature of its inhibition by alkaline earth cations. Journal of Experimental Botany 23, 552-560.
Chaudri AM, McGrath SP, Giller KE (1992) Metal tolerance of isolates of Rhizobium leguminosarum biovar trifolii from soil contaminated by past applications of sewage sludge. Soil Biology and Biochemistry 24, 83-85.
Chaudri AM, McGrath SP, Giller KE, Rietz E, Sauerbeck DR (1993) Enumeration of indigenous Rhizobium leguminosarum biovar trifolii in soils previously treated with metal-contaminated sewage sludge. Soil Biology and Biochemistry 25, 301-309.
Chaudri AM, Knight BP, Barbosa-Jefferson VL, Preston S, Paton GI, Killham K, Coad N, Nicholson FA, Chambers BJ, McGrath SP (1999) Determination of acute Zn toxicity in pore water from soils previously treated with sewage sludge using bioluminescence assays. Environmental Science and Technology 33, 1880-1885.
Checkai RT, Corey RB, Helmke PA (1987) Effects of ionic and complexed metal concentrations on plant uptake of cadmium and micronutrient metals from solution. Plant and Soil 99, 335-345.
Christensen TH (1984) Cadmium soil sorption at low concentrations: I Effect of time, cadmium load, pH, and calcium. Water Air and Soil Pollution 21, 105-114.
Chu CW, Poon CS, Cheung RYH (1998) Characterization of raw sludge, chemically modified sludge and anaerobically digested sludge in Honk Kong. Water Science and Technology 38, 25-32.
Clarkson DT, Luttge U (1989) Mineral nutrition: divalent cations, transport and compartmentation. Progress in Botany 51, 93-112.
Clayton PM, Tiller KG (1979) A chemical method for the determination of the heavy metal content of soils in environmental studies. CSIRO Australian Division of Soils Technical Paper 41, pp. 1-17.
Cook N, Hendershot WH (1996) The problem of establishing ecologically based soil quality criteria: the case of lead. Canadian Journal of Soil Science 76, 335-342.
Corry H, Berg G, Shelley B, Peverill K (1993) Survey of heavy metal contaminants in fertilisers and soil amendments. Research Report Series No. 147, Victorian Department of Agriculture, Australia.
DeWit H (1992) Proton and metal binding to humic substances. PhD thesis, Wageningen Agricultural University, The Netherlands.
Dijkshoorn W, Lampe JEM, van Broekhoven LW (1981) Influence of soil pH on heavy metals in ryegrass from sludge-amended soil. Plant and Soil 61, 277-284.
DoE (1989) `Code of practice for agricultural use of sewage sludge.' Department of the Environment (HMSO: London)
DoE (1996) Code of practice for agricultural use of sewage sludge.' (Department of the Environment: London)
DoELM (1999) `Tasmanian biosolids reuse guidelines.' (Department of Environment and Land Management: Hobart)
Doelman P, Haanstra L (1986) Short- and long-term effects of heavy metals on urease activity in soils. Biology and Fertility of Soils 2, 213-218.
Doelman P, Haanstra L (1989) Short- and long-term effects of heavy metals on phosphatase activity in soils: An ecological dose-response model approach. Biology and Fertility of Soils 8, 235-241.
Dowdy RH, Larson WE, Titrud JM, Latterell JJ (1978) Growth and metal uptake of Snap Beans grown on sewage sludge amended soil: A four year field study. Journal of Environmental Quality 7, 252-257.
Edmeades DC, Wheeler DM, Clinton OE (1985) The chemical composition and ionic strength of top soil solutions from New Zealand top soils. Australian Journal of Soil Research 23, 151-165.
Fischer L, Zur Muhlen E, Brummer GW, Niehus H (1996) Atomic force microscopy (AFM) investigations of the surface topography of a multidomain porous goethite. European Journal of Soil Science 47, 329-334.
Gerritse RG, Van Driel W, Smilde KW, Van Luit B (1983) Uptake of heavy metals by crops in relation to their concentration in the soil solution. Plant and Soil 75, 393-404.
Gerth J, Brummer GW, Tiller KG (1992) Retention of Ni, Zn and Cd bi Si-associated goethite. Zeitschrift fur Pflanzenernahrung und Bodenkunde 156, 123-129.
Giller KE, McGrath SP, Hirsch PR (1989) Absence of nitrogen fixation in clover grown on soil subject to long term contamination with heavy metals is due to survival of only ineffective rhizobium. Soil Biology and Biochemistry 21, 841-848.
Giller KE, Nussbaum R, Chaudri AM, McGrath SP (1993) Rhizobium meliloti is less sensitive to heavy-metal contamination in soil than R. leguminosarum bv Trifolii or R. loti. Soil Biology and Biochemistry 25, 272-278.
Goldstein N, Steuteville R (1996) Steady climb for biosolids composting. Biocycle 37, 68-78.
Gray CW, McLaren RG, Roberts AHC, Condron LM (1998) Sorption and desorption of cadmium from some New Zealand soils: effect of pH and contact time. Australian Journal of Soil Research 36, 199-216.
Gray CW, McLaren RG, Roberts AHC, Condron LM (1999a) The effect of long-term phosphatic fertiliser application on the amounts and forms of cadmium in soils under pasture in New Zealand. Nutrient Cycling in Agroecosystems 47, 169-178.
Gray CW, McLaren RG, Roberts AHC, Condron LM (1999b) Cadmium phytoavailability in some New Zealand soils. Australian Journal of Soil Research 37, 461-477.
Hahne HCH, Kroontje W (1973) Significance of pH and chloride concentration on behaviour of heavy metal pollutants: Mercury (II), cadmium (II), zinc (II), and lead (II). Journal of Environmental Quality 2, 444-450.
Halvorsen AD, Lindsay WL (1977) The critical [Zn.sup.2+] concentration for corn and the nonabsorption of chelated zinc. Soil Science Society of America Journal 41, 531-534.
Hamon R, Lorenz S, Holm P, Christensen T, McGrath S (1995) Temporal variation in trace metal speciation and other soil solution components during growth of radish. Plant, Cell and Environment 18, 749-756.
Hamon RE (1995) Identification of factors governing Cd and Zn bioavailability in polluted soil. PhD thesis, University of Nottingham, England.
Hamon RE, Holm PE, Lorenz SE, McGrath SP, Christensen TH (1999) Metal uptake by plants from sludge-amended soils: caution is required in the plateau interpretation. Plant and Soil 216, 53-64.
Hamon RE, Holm PE, Lorenz SE, McGrath SP, Christensen TH (1999) Metal uptake by plants from sludge-amended soils: caution is required in the plateau interpretation. Plant and Soil 216, 53-64.
Hamon R, McLaughlin M, Naidu R (1997a) Evidence for the immobilisation of Cd in soils. In `Proceedings of the 4th international conference on the biogeochemistry of trace elements'. Berkeley, USA. (Eds IK Iskandar et al.) pp. 425-426. (US Army Cold Regions Research and Engineering Laboratory: Hanover, NH)
Hamon R, Wundke J, McLaughlin M, Naidu R (1997b) Availability of zinc and cadmium to different plant species. Australian Journal of Soil Research 35, 1267-1277.
Hani H (1996) Soil analysis as a tool to predict effects on the environment. Communications in Soil Science and Plant Analysis 27, 289-306.
Hartenstein R, Neuhauser EF, Narahara A (1981) Effects of heavy metal and other elemental additives to activated sludge on growth of Eisenia fetida. Journal of Environmental Quality 10, 372-376.
He QB, Singh BR (1993) Plant availability of cadmium in soils. I. Extractable cadmium in newly and long-term cultivated soils. Acta Agriculture Scandinavica. Section B. Soil and Plant Science 43, 134-141.
Hendrickson LL, Corey RB (1981) Effect of equilibrium metal concentrations on apparent selectivity coefficients of soil complexes. Soil Science 131, 163-171.
Hewitt AE (1997) Are New Zealand soils distinctive?: a subterranean view of NZ ecosystems. New Zealand Soil News 45, 7-16.
Holm PE, Christensen TH, Lorenz SE, Hamon RE, Domingues HC, Sequeira EM, McGrath SP (1998) Measured soil water concentrations of cadmium and zinc in plant pots and estimated leaching outflows from contaminated soils. Water, Air and Soil Pollution 102, 105-115.
Hubble GD, Isbell RF, Northcote KH (1983) Features of Australian soils. In `Soils: an Australian viewpoint', pp. 17-47. (CSIRO Australia: Melbourne)
Hyun H, Chang AC, Parker DR, Page AL (1998) Cadmium solubility and phytoavailability in sludge-treated soil: effects of soil organic carbon. Journal of Environmental Quality 27, 329-334.
Imray PLA, Langley A (1996) Health-based soil investigation levels. National Environmental Health Forum monographs No. 1, South Australian Health Commission, Adelaide.
Jinadasa KBPN, Milham PJ, Hawkins CA, Cornish PS, Williams PA, Kaldor CJ, Conroy JP (1997) Survey of cadmium concentrations in vegetables and soils of Greater Sydney, Australia. Journal of Environmental Quality 26, 924-933.
Jing J, Logan TJ (1992) Effects of sewage sludge cadmium concentration on chemical extractability and plant uptake. Journal of Environmental Quality 21, 73-81.
Johnston NB, Beckett PHT, Waters CJ (1983) Limits of zinc and copper toxicity from digested sludge applied to agricultural land. In `Environmental effects of organic and inorganic contaminants in sewage sludge'. (Eds RD Davis et al.) pp. 75-81. (D. Reidel Publishing Company: Dordrecht)
Jopony M, Young S (1993) Assessment of lead availability in soils contaminated by mine spoil. Plant and Soil 151, 273-278.
Juma NG, Tabatabai MA (1977) Effects of trace elements on phosphatase activity in soils. Soil Science Society of America Journal 41, 343-346.
Kerndorff H, Schnitzer M (1980) Sorption of metals on humic acid. Geochimica et Cosmochimica Acta 44, 1701-1708.
King LD (1988) Effect of selected soil properties on cadmium content of tobacco. Journal of Environmental Quality 17, 251-255.
Kinniburgh DG, Miles DL (1983) Extraction and chemical analysis of interstitial water from soils and rocks. Environmental Science and Technology 17, 362-368.
Korcak RF, Fanning DS (1985) Availability of applied metals as a function of type of soil material and metal source. Soil Science 140, 23-34.
Langlands JP, Donald GE, Bowles JE (1988) Cadmium concentrations in liver, kidney and muscle in Australian sheep and cattle. Australian Journal of Experimental Agriculture 28, 291-307.
Lebourg A, Sterckeman T, Ciesielski H, Proix N (1996) Interet de differents reactifs d'extraction chimique pour l'evaluation de la biodisponibilite des metaux en traces du sol. Agronomie 16, 201-215.
Lee J, Rounce JR, Mackay AD, Grace ND (1996) Accumulation of cadmium with time in Romney sheep grazing ryegrass-white clover pasture: effect of cadmium from pasture and soil intake. Australian Journal of Agricultural Research 47, 877-894.
Lepp NW (1981) `Effect of heavy metal pollution on plants.' pp. 299-322. (Applied Science Publishers: London)
Lexmond ThM, Edelman Th (1987) Huidige achtergrondwaarden van het gehalte aan een aantal zware metalen en arseen in grond. In `Handboek voor milieubeheer, bodembescherming'. (Samson, Alphen ann de Rijn: The Netherlands)
Li U-M, Chaney RL, Schneiter AA (1994) Effect of soil chloride level on cadmium concentration in sunflower kernels. Plant and Soil 167, 275-280.
Liang CN, Tabatabai MA (1978) Effects of trace elements on nitrification in soils. Journal of Environmental Quality 7, 291-293.
Lighthart B, Baham J, Volk VV (1983) Microbial respiration and chemical speciation in metal-amended soils. Journal of Environmental Quality 12, 543-548.
Lindsay WL, Norvell WA (1978) Development of a DTPA soil test for zinc, manganese and copper. Soil Science Society of America Journal 42, 421-428.
Litaor MI (1988) Review of soil solution samplers. Water Resources Research 24, 727-733.
Logan TJ, Chaney RL (1983) Utilization of municipal wastewater and sludge on land--metals. In `Proceedings of the 1983 workshop on utilization of municipal wastewater and sludge on land'. (Eds AL Page et al.) pp. 235-326. (University of California: Riverside, CA)
Loganathan P, Hedley MJ, Gregg PEH, Currie LD (1997) Effect of phosphate fertiliser type on the accumulation and plant availability of cadmium in grassland soils. Nutrient Cycling in Agroecosystems 47, 169-178.
Lorenz SE, Hamon RE, Holm PE, Domingues HC, Sequiera EM, Christensen TH, McGrath SP (1997) Cadmium and zinc in plants and soil solutions from contaminated soils. Plant and Soil 189, 21-31.
Lorenz SE, Hamon RE, McGrath SP, Holm PE, Christensen TH (1994) Applications of fertiliser cations affect cadmium and zinc concentrations in soil solutions and uptake by plants. European Journal of Soil Science 45, 159-165.
Lottermoser BG (1997) Natural enrichment of topsoils with chromium and other heavy metals, Port Macquarie, New South Wales, Australia. Australian Journal of Soil Research 35, 1165-1176.
MAFF/DoE (1993) `Review of the rules for sewage sludge application to agricultural land. Soil fertility aspects of potentially toxic elements'. Ministry of Agriculture Fisheries and Food/Department of the Environment (MAFF Publication: London)
Maliszewska W, Dec S, Wierzbicka H, Wozniakowska A (1985) The influence of various heavy metal compounds on the development and activity of soil micro-organisms. Environmental Pollution (Series A) 37, 195-215.
McBride MB (1995) Toxic metal accumulation from agricultural use of sludge: are USEPA regulations protective? Journal of Environmental Quality 24, 5-18.
McBride MB, Sauve S, Hendershot WH (1997) Solubility control of Cu, Zn, Cd and Pb in contaminated soil. European Journal of Soil Science 48, 337-346.
McBride MB (1998) Soluble trace metals in alkaline-stabilised sludge products. Journal of Environmental Quality 27, 578-584.
McGrath D, Fleming GA, McCormack RF, Poole DBR (1982) Effects of applying copper-rich pig slurry to grassland. 2. Land-spreading trial. Irish Journal of Agricultural Research 21, 49-60.
McGrath SP (1987) Long-term studies of metal transfers following application of sewage sludge. In `Pollutant transport and fate in ecosystems'. (Eds PJ Coughtrey et al.) pp. 301-317. (Blackwell Scientific Publishers: Oxford)
McGrath SP (1999) Adverse effects of cadmium on soil microflora and fauna. In `Cadmium in soils and plants'. (Eds MJ McLaughlin and BR Singh) pp. 199-218. (Kluwer Academic Publishers: Dordrecht)
McGrath SP, Brookes PC, Giller KE (1988) Effects of potentially toxic metals in soil derived from past applications of sewage sludge on nitrogen fixation by Trifolium repens L. Soil Biology and Biochemistry 20, 415-424.
McGrath SP, Chaudri AM (1999) Long term effects of metal contamination on Rhizobium. Soil Biology and Biochemistry 31, 1205-1207.
McGrath SP, Chang AC, Page AL, Witter E (1994) Land application of sewage sludge: scientific perspectives of heavy metal loading limits in Europe and the United States. Environmental Reviews 2, 108-118.
McGrath SP, Chaudri AM, Giller KE (1995) Long-term effects of metals in sewage sludge on soils, microorganisms and plants. Journal of Industrial Microbiology 14, 94-104.
McKenna IM, Chaney RL, Williams FM (1993) The effects of cadmium and zinc interactions on the accumulation and tissue distribution of zinc and cadmium in lettuce and spinach. Environmental Pollution 79, 113-120.
McLaren RG, Cameron KC (1996) `Soil science: sustainable production and environmental protection.' 2nd edn (Oxford University Press: Auckland)
McLaren RG, Clouston SD, Clucas L (1999a) Concentrations and forms of metals in a range of New Zealand biosolids. In `Function and management of plants in land treatment systems. Proceedings of the New Zealand land treatment collective No. 19'. (Eds H Wang and M Tomer) pp. 37-46. (New Zealand Land Treatment Collective: Christchurch)
McLaren RG, Naidu R, Smith J, Tiller KG (1998) Fractionation and distribution of arsenic in soils contaminated by cattle dip. Journal of Environmental Quality 27, 348-354.
McLaren RG, Smith CJ (1996) Issues in the disposal of industrial and urban wastes. In `Contaminants and the soil environment in the Australasia-Pacific region'. (Eds R Naidu et al.) pp. 183-212. (Kluwer Academic Publishers: Dordrecht)
McLaren RG, Taylor M, Hendry T, Clucas L (1999b) Leaching of metals and nutrients from soils treated with metal-amended sewage sludge. In `Best management practices for production'. (Eds LD Currie et al.) pp. 251-260 (Massey University: Palmerston North: New Zealand)
McLaughlin MJ, Andrew SJ, Smart MK, Smolders E (1998) Effects of sulfate on cadmium uptake by Swiss chard. I. Effects on complexation and calcium competition in nutrient solution. Plant and Soil 202, 211-216.
McLaughlin MJ, Tiller KG, Beech TA, Smart MK (1994) Soil salinity causes elevated cadmium concentrations in field-grown potato tubers. Journal of Environmental Quality 23, 1013-1018.
McLaughlin MJ, Tiller KG, Naidu R, Stevens DP (1996) Review: the behaviour and environmental impact of contaminants in fertilisers. Australian Journal of Soil Research 34, 1-56.
McLaughlin MJ, Tiller KG, Smart MK (1997a) Speciation of cadmium in soil solution of saline/sodic soils and relationship with cadmium concentrations in potato tubers. Australian Journal of Soil Research 35, 1-17.
McLaughlin MJ, Maier NA, Rayment GE, Sparrow LABG, McKay A, Milham P, Merry RH, Smart MK (1997b) Cadmium in Australian potato tubers and soils. Journal of Environmental Quality 26, 1644-1649.
McLaughlin MJ, Smolders E, Merckx R, Maes A (1997c) Plant uptake of cadmium and zinc in chelator-buffered nutrient solution depends on ligand type. In `Plant nutrition for sustainable food production and environment'. (Eds T Ando et al.) pp. 113-118. (Kluwer Academic Publishers: Dordrecht)
McLaughlin MJ, Zarcinas BA, Stevens DP, Cook NE (2000) Testing soils and plants for heavy metals. Communications in Soil Science and Plant Analysis 31, 1661-1700.
Mench M, Vangronsveld J, Didier V, Clijsters H (1994) Evaluation of metal mobility, plant availability and immobilization by chemical agents in a limed-silty soil. Environmental Pollution 86, 279-286.
Merry RH (1988) Investigations on cadmium in South Australia: Rainfall, soils, cereals, pastures and soil-plant relations. In `Cadmium accumulations in Australian agriculture' (Eds J Simpson and WJ Curnow) pp. 62-79. (Bureau of Rural Resources: Canberra, ACT)
Merry RH, Tiller KG (1991) Distribution and budget of cadmium and lead in an agricultural region near Adelaide, South Australia. Water, Air, and Soil Pollution 57/58, 171-180.
Merry RH, Tiller KG, De Vries MPC, Cartwright B (1981) Contamination of wheat crops around a lead-zinc smelter. Environment Pollution (Series B) 2, 37-48.
Merry RH, Tiller KG, Alston AM (1983) Accumulation of copper, lead and arsenic in some Australian orchard soils. Australian Journal of Soil Research 21, 549-561.
Mitchell P, Barr D (1995) The nature and significance of public exposure to arsenic: a review of its relevance to south west England. Environmental Geochemistry and Health 17, 57-82.
Moen JET (1988) Soil protection in the Netherlands. In `Contaminated soil `88'. (Eds K Wolf et al.) pp. 1495-1503. (Kluwer Academic Publishers: Dordrecht)
Moolenaar SW, van der Zee SEATM, Lexmond TM (1997) Indicators of the sustainability of heavy-metal management in agro-ecosystems. The Science of the Total Environment 201, 155-169.
Mortvedt JJ, Osborn G (1982) Studies on the chemical form of cadmium contaminants in phosphate fertilisers. Soil Science 134, 185-192.
Naidu R, Bolan NS, Kookana RS, Tiller KG (1994) Ionic-strength and pH effects on the sorption of cadmium and the surface charge of soils. European Journal of Soil Science 45, 419-429.
Naidu R, Kookana RS, Sumner ME, Harter RD, Tiller KG (1997) Cadmium sorption and transport in variable charge soils: a review. Journal of Environmental Quality 26, 602-617.
Nakahone LN, Young SD (1993) The significance of (radio-) labile cadmium pools in soil. Environmental Pollution 82, 73-77.
Naylor LM, Barmasse M, Loehr RC (1987) Uptake of cadmium and zinc by corn on sludge-treated soils. Biocycle 28, 37-41.
NEPC (1999) National environment protection (assessment of site contamination) measure. National Environment Protection Council Service Corporation, Adelaide.
NFA (1993) `Food standards code.' (National Food Authority, Australian Government Printing Service: Canberra)
NFDC (1980) Distribution of cadmium and selected heavy metals in phosphate fertiliser processing. National Fertiliser Development Center Bulletin Y-1569, Tennessee Valley Authority, Alabama.
Northcote KH, Skene JKM (1972) `Australian soils with saline and sodic properties.' Soil Publications No. 27. (CSIRO Australia: Melbourne)
NSW EPA (1997) Environmental guidelines: Use and disposal of biosolids products. New South Wales Environmental Protection Authority, Sydney.
NSW EPA (1998) Draft guidelines for the NSW site auditor scheme (Draft). New South Wales Environmental Protection Authority, Sydney.
New Zealand Ministry of Health (1984) `The New Zealand food regulations'. (GP Publications: Wellington)
New Zealand Resource Management Act (1991) `The statutes of New Zealand 1991.' Vol. 2, pp. 595-976. (New Zealand Government: Wellington)
NZDoH (1992) `Public health guidelines for the safe use of sewage effluent and sewage sludge on land.' (New Zealand Department of Health: Wellington)
NZWWA (1999) New Zealand guidelines for the beneficial use of biosolids. Draft No. 2. New Zealand Water and Waste Association, Woodward-Clyde (NZ) Ltd, Auckland.
O'Neill PA (1995) Arsenic. In `Heavy metals in soils.' (Ed. BJ Alloway) pp. 83-99. (Blackie Academic and Professional: Glasgow)
O'Connor GA (1988) Use and misuse of the DTPA soil test. Journal of Environmental Quality 17, 715-718.
Oertel AC, State HCT (1947) A spectrochemical survey of some phosphate rocks and superphosphates. Journal of the Council for Scientific and Industrial Research 20, 100-113.
Ogilvie D (1998) National study of the composition of sewage sludge. New Zealand Water and Wastes Association, Auckland.
Oliver DP, Gore PJ, Moss HJ, Tiller KG (1993a) Cadmium in wheat-grain and milling products from some Australian flour mills. Australian Journal of Agricultural Research 44, 1-11.
Oliver DP, Hannam R, Tiller KG, Wilhelm NS, Merry RH, Cozens GD (1994) The effects of zinc fertilization on cadmium concentration in wheat grain. Journal of Environmental Quality 23, 705-711.
Oliver DP, Schultz JE, Tiller KG, Merry RH (1993b) The effect of crop rotations and tillage practices on cadmium concentration in wheat grain. Australian Journal of Agricultural Research 44, 1221-1234.
Page AL, Chang AC (1985) Fate of wastewater constituents in soil and groundwater: trace elements. In `Irrigation with reclaimed municipal wastewater--A guidance manual'. (Eds GS Pettygrove and T Asano) pp. 13.1-13.16. (Lewis Publishers Inc.: Chelsea)
Pantsar-Kallio M, Mujunen S-P, Hatzimihalis G, Koutoufides P, Minkkinen P, Wilkie JJ, Connor MA (1999) Multivariate data analysis of key pollutants in sewage samples: a case study. Analytica Chimica Acta 393, 181-191.
Phinney JT, Bruland KW (1994) Uptake of lipophilic organic Cu,Cd and Pb complexes in the coastal diatom Thalassiosira weissflogii. Environmental Science and Technology 28, 1781-1790.
Prue[Beta] A (1997) Action values for mobile ([NH.sub.4][NO.sub.3]-extractable) trace elements in soils based on the German national standard DIN 19730. In `Contaminated Soils. 3rd International conference on the biogeochemistry of trace elements'. (Ed. R Prost) pp. 415-423. (Institut National de la Recherche Agronomique: Paris)
Rayment GE, Best EK, Hamilton DJ (1992) Cadmium in fertilisers and soil amendments. Department of Primary Industries, Queensland.
Roberts AHC, Cameron KC, Bolan NS, Ellis HK, Hunt S (1996) Contaminants and the soil environment in New Zealand. In `Contaminants and the soil environment in the Australasia-Pacific region'. (Eds R Naidu et al.) pp. 579-628. (Kluwer Academic Publishers: Dordrecht)
Roberts AHC, Longhurst RD, Brown MW (1994) Cadmium status of soils, plants and grazing animals in New Zealand. New Zealand Journal of Agricultural Research 37, 119-129.
Rogers S (1997) Measurement of total heavy metals in biosolids and soils receiving biosolids is a poor indicator of metal bioavailability. In `Proceedings of the 4th International Conference on the Biogeochemistry of Trace Elements'. Berkeley, USA, June 1997. (Eds IK Iskandar et al.) pp. 147-148. (US Army Cold Regions Research and Engineering Laboratory: Hanover, NH)
Rogers SL, McLaughlin MJ (1999) Relationship between biosolid metal concentration, metal fractionation and bioavailability. In `Proceedings of the 5th International Conference on the Biogeochemistry of Trace Elements'. Vienna. pp. 280-281.
Ross DS, Sjogren RE, Bartlett RJ (1981) Behavior of chromium in soils: IV. Toxicity to microorganisms. Journal of Environmental Quality 10, 145-148.
SA EPA (1996) South Australian biosolids guidelines for the safe handling, reuse or disposal of biosolids. Department of Environment and Natural Resources, South Australian Environment Protection Authority, Adelaide.
Sauerbeck DR, Styperek P (1985) Significance of ion species and concentration for the determination of `available' cadmium in soils. In `Cooperative network on trace elements, fourth issue'. (Coordinating Centre, State University: Belguim)
Schmidt JP (1997) Understanding phytotoxicity thresholds for trace elements in land-applied sewage sludge. Journal of Environmental Quality 26, 4-10.
Scott-Fordsmand JJ (1997) Toxicity of nickel to soil organisms in Denmark. Reviews in Environmental Contamination and Toxicology 148, 1-34.
Scott-Fordsmand JJ, Pedersen MB (1995) `Soil quality criteria for selected inorganic compounds.' (Danish Environmental Protection Agency: Denmark)
Sheppard SC (1992) Summary of phytotoxic levels of soil arsenic. Water, Air and Soil Pollution 64, 539-550.
Smit H (1998) Pollution prevention policies for agricultural soils in the Netherlands. In `Contaminated Soils. Proceedings of the 3rd International conference on the biogeochemistry of trace elements'. (Ed. R Prost) CD ROM (Institut National de la Recherche Agronomique: Paris)
Smith SR (1996) `Agricultural recycling of sewage sludge and the environment'. (CAB International: UK)
Smith SR (1997) Rhizobium in soils contaminated with copper and zinc following the long term application of sewage sludge and other organic wastes. Soil Biology and Biochemistry 29, 1475-1489.
Smith WH (1976) Lead contamination of the roadside ecosystem. Journal of the Air Pollution Control Federation 26, 753-766.
Smolders E, McLaughlin MJ (1996) Effect of CI and Cd uptake by swiss chard in nutrient solution. Plant and Soil 179, 57-64.
Sommers L, van Volk V, Giordano PM, Sopper WE, Bastian R (1987) Effect of soil properties on accumulation of trace elements by crops. In `Land application of sludge food chain implications'. (Eds AL Page et al.) pp. 5-24. (Lewis Publishers Inc.: Chelsea)
Speir TW, August JA, Feltham CW (1992) Assessment of the feasibility of using CCA (copper, chromium and arsenic)- treated and boric acid-treated sawdust as soil amendments. Plant and Soil 142, 235-248.
Stenhouse F (1992) `The 1992 Australian market basket survey'. National Food Authority. (Australian Government Publishing Service: Canberra)
Swedish Environmental Protection Agency (1994) Kungorelse med foreskrifter om skydd for miljon, sarskilt marken, nar avloppsslam anvands i jordbruket. 30/4/1994. Statens naturvardsverks forfattningssamling.
Swift RS, McLaren RG (1991) Micronutrient adsorption by soils and soil colloids. In `Interactions at the soil colloid-soil solution interface'. (Eds GH Bolt et al.) pp. 257-292. (Kluwer Academic Publishers: Dordrecht)
Symeonides C, McRae SG (1977) The assessment of plant-available cadmium in soils. Journal of Environmental Quality 6, 120-123.
Tabatabai MA (1977) Effects of trace elements on urease activity in soils. Soil Biology and Biochemistry 9, 9-13.
Tamaki S, Frankenberger WT (1992) Environmental biochemistry of arsenic. Reviews in Environmental Contamination and Toxicology 124, 79-110.
Tas. DoELM (1999) Tasmanian biosolid reuse guidelines. Tasmanian Department of Environment and Land Management, Hobart.
Tate KR, Giltrap DJG, Claydon JJ, Newsome PF, Atkinson IAE, Taylor MD, Lee R (1997) Carbon stocks in New Zealand terrestrial ecosystems. Journal of the Royal Society of New Zealand 27, 315-335.
Taylor JK (1993) The Australian environment. In `Soils: an Australian viewpoint', pp. 1-2. (CSIRO Australia: Melbourne)
Taylor NH, Pohlen IJ (1968) The soil environment. In `Soils of New Zealand, Part 10'. Bull. No. 26, pp. 7-14. (NZ Soil Bureau: Wellington)
Tiller KG (1989) Heavy metals in soils and their environmental significance. Advances in Soil Science 9, 114-142.
Tiller KG, Gerth J, Brummer G (1984) The relative affinities of Cd, Ni and Zn for different soil clay fractions and goethite. Geoderma 34, 17-35.
Tiller KG, Honeysett JL, Devries MPC (1972) Soil zinc and its uptake by plants. II. Soil chemistry in relation to prediction of availability. Australian Journal of Soil Research 10, 165-182.
Tiller KG, Nayyar VK, Clayton PM (1979) Specific and non-apecific sorption of cadmium by soil clays as influenced by zinc and calcium. Australian Journal of Soil Research 17, 17-28.
Tiller KG, Smith LH, Merry RH, Clayton PM (1987) The dispersal of automotive lead from metropolitan Adelaide into adjacent rural areas. Australian Journal of Soil Research 25, 155-166.
UKDoE (1989) The Sludge (use in agriculture) regulations, Public Health, England, Wales and Scotland. SI 1263. United Kingdom, Department of Environment, London.
US EPA (United States Environment Protection Agency) (1992) `Technical Support Document for Land Application of Sewage Sludge. Volume 1'. p. 523. (Office of Water, US Environmental Protection Agency: Washington DC)
US EPA-United States Environment Protection Agency (1993) Part 503--Standards for the use and disposal of sewage sludge. Federal Register 58, 9387-9404.
US EPA United States Environment Protection Agency (1996) Soil screening guidance. Available at http://www.epa.gov/oerrpage/superfund/resources/soil/index.htm (verified 11 Mar. 2000)
Van den Berg R, Roels JM (1991) Risk assessment to man and the environment in case of exposure to soil contamination. Integration of different aspects. Rijksinstituut voor Volksgezondheid en Milieu (RIVM) Report No. 725201013, RIVM, Bilthoven, the Netherlands.
Vaughan GT (1993) The environmental chemistry and fate of arsenical pesticides in cattle dip sites and banana land plantations. CSIRO Division of Coal and Energy Technology, Report CET/LHIR148, CSIRO Australia, Melbourne.
Vercauteren K, Blust R (1996) Bioavailability of dissolved zinc to the common mussel Mytilus edulis in complexing environments. Marine Ecology Progress Series 137, 123-132.
Victorian EPA (1997) Environmental guidelines for the management of biosolids (Draft). Environment Protection Authority, Melbourne.
de Vries MPC (1983) Investigations on twenty Australian sewage sludges--their evaluation by means of chemical analysis. Fertiliser Research 4, 75-87.
Walker P, McConnell S, Gibson E, Williams N (1999) A pilot study on developing risk based ecological investigation levels for contaminated sites in Victoria. In `Contaminated site remediation: Challenges posed by urban and industrial contaminants. Proceedings of the 1999 Contaminated Site Remediation Conference'. pp. 781-788. (Center for Groundwater Studies: Wembley, WA)
Whitten MG, Ritchie GSP (1991) Calcium chloride extractable cadmium as an estimate of cadmium uptake by subterranean clover. Australian Journal of Soil Research 29, 215-221.
Will ME, Suter GW (1995a) Toxicological benchmarks for screening potential contaminants of concern for effects on soil and litter invertebrates and heterotrophic process. Report ES/ER/TM-126/R2, Oak Ridge National Laboratory, Oak Ridge, Tennessee.
Will ME, Suter GW (1995b) Toxicological benchmarks for screening potential contaminants of concern for effects on terrestrial plants. Report ES/ER/TM-85/R2, Oak Ridge National Laboratory, Oak Ridge, Tennessee.
Williams CH (1978) Impurities in Australian phosphate fertilisers. Journal of the Australian Institute of Agricultural Science 44, 132-135.
Williams CH, David DJ (1977) Some effects of the distribution of cadmium and phosphate in the root zone on the cadmium content of plants. Australian Journal of Soil Research 15, 59-68.
Witter E (1992) Heavy metal concentrations in agricultural soil critical to microorganisms. Staten Naturvardverket Rapport 4079, Stockholm, Sweden.
Yadav DS, Kumar V, Singh M (1986) Inhibition of soil urease activity and nitrification with some metallic cations. Australian Journal of Soil Research 24, 527-532.
Yeates GW, Orchard VA, Speir TW, Hunt JL, Hermans MCC (1994) Impact of pasture contamination by copper, chromium, arsenic timber preservative on soil biological activity. Biology and Fertility of Soils 18, 200-208.
Zachara JM, Smith SC, Resch CT, Cowan CE (1992) Cadmium sorption to soil separates containing layer silicated and iron and aluminum oxides. Soil Science Society of America Journal 56, 1074-1084.
Zarcinas B, Nable RO (1992) Boron and other impurities in South Australian fertilisers and soil amendments. CSIRO Division of Soils Divisional Report 118, 10 pp.
Zhang F, Romheld V, Marschner H (1989) Effect of zinc deficiency in wheat on the release of zinc and iron mobilizing root exudates. Zeitsehrift fur Pflanzenernahrung und Bodenkunde 152, 205-210.
Manuscript received 8 November 1999, accepted 10 May 2000
M. J. McLaughlin(A), R. E. Hamon(A), R. G. McLaren(B), T. W. Speir(C), and S. L. Rogers(A)
(A) CSIRO Land and Water, PMB 2, Glen Osmond, SA 5064, Australia.
(B) Soil, Plant and Ecological Sciences Division, PO Box 84, Lincoln University, Canterbury, New Zealand.
(C) Institute of Environmental Science and Research Ltd, PO Box 50 348, Porirua, New Zealand.