Leaching of macronutrients and metals from undisturbed soils treated with metal-spiked sewage sludge. 3. Distribution of residual metals.
In New Zealand, protection of groundwater is a major issue, and the increasing pressure for land treatment of sewage sludge has raised the question of potential nutrient and metal contamination of groundwater from this source. In the first 2 parts of this series of papers (McLaren et al. 2003, 2004), we have described the leaching of macronutrients and metals during a 3-year lysimeter experiment involving a range of different New Zealand soils to which metal-spiked sewage sludge had been applied. In the past, the generally held view by many researchers was that the presence of metals in sewage sludge-treated soil is confined to the cultivation zone, with very little movement below the layer of incorporation (e.g. Chang et al. 1984; Williams et al. 1985; Smith 1996). However, the results of our study showed increased concentrations of cadmium (Cd), nickel (Ni), and zinc (Zn) in drainage leachates from some of the soils that we treated with sewage sludge (McLaren et al. 2004). Similar results have been reported in other recent studies. For example, Keller et al. (2002) noted increased concentrations of metals, particularly copper (Cu) and Ni, in drainage water from undisturbed soil lysimeters treated with sewage sludge, and Speir et al. (2003) observed elevated levels of Cu, Ni, and Zn in groundwater at a pasture site treated with sewage sludge. Similarly, Richards et al. (1998) and McBride et al. (1999) observed considerable evidence of metal mobility at a heavily loaded sludge-amended site, and Roy and Couillard (1998) reported significant metal leaching following sludge application to a deciduous forest soil.
The evidence would now seem incontrovertible that, under some circumstances, metals added to soils in applications of sewage sludge can be leached downwards through the soil profile and can have the potential to contaminate groundwater. However, in the few studies where direct measurements of metal leaching from undisturbed soils have been made following sludge application (e.g. Keller et al. 2002; McLaren et al. 2004), the amounts leached have represented very small proportions of the metals added in the sludge, and the effects have often been somewhat transitory. For example, in part 2 of this series (McLaren et al. 2004), we reported that Zn and Ni concentrations in drainage leachates from sludged lysimeters returned to background levels within 3 years of the sludge application. The metal concentration breakthough curves for the lysimeters also indicated that metal movement was facilitated by macropore flow, i.e. movement through relatively large pores and bypassing interaction with the soil. Such a process has also been suggested by Dowdy et al. (1991) and Camobreco el al. (1996).
However, the possibility of continuing long-term metal mobility, following the macropore flow-facilitated metal leaching observed in the period immediately following sludge application, requires examination. Such a possibility is certainly supported by the observations by McBride et al. (1999) of continuing mobility of metals in a soil l 5 years after sludge application. In the long-term, instead of rapid transport of metals by macropore flow, movement by normal convective and diffusive transport is likely to be more important. On this basis, in addition to the short-term detection of increased metal concentrations in lysimeter drainage leachates, we might also expect to find evidence of metal movement within the lysimeter soil column itself. This paper describes the distribution of residual metals in the lysimeters of 5 sludge-treated New Zealand soils following completion of the 3 years of nutrient and metal leaching monitoring reported previously (McLaren et al. 2003, 2004).
Materials and methods
Initial leaching phase
Full details of the experimental set-up and leaching phase of this study are described by McLaren el al. (2003, 2004). Essentially. 5 contrasting soils were used for this study--3 from forest (Pinus rudiata) plantations and two from pasture sites. The forest soils were a Waikuku loamy sand, a Lismore stony silt loam, and an Eyre shallow silt loam soil; and the pasture sites, a Templeton fine sandy loam and a Horotiu sandy loam. Some important properties of the soils were reported by McLaren el al. (2003). Five undisturbed soil monolith lysimeters were taken from each site, apart from the Horotiu site, from which only 3 lysimeters were collected. Each lysimeter was 0.50 m in diameter and 0.70 m deep. The design of the lysimeter casings and the method of sampling the soils have been described in detail by Cameron et al. (1992).
Sewage sludge obtained from the Christchurch City Council treatment works at Bromley was 'spiked' with additional metals (Cd, Cr. Cu, Ni, Pb, Zn) in the form of metal sulfates (McLaren et al. 2004), and 3 replicate lysimeters of each soil (2 for the Horotiu soil) were treated with the spiked sludge. The sludge was surface-applied to the forest soils and incorporated into the top 10 cm in the pasture soils. The pasture soils were sown with a grass/clover mixture following the sludge application. Two lysimeters of each soil (one for the Horotiu soil) were left as controls with no sludge applied. The lysimeters were exposed to natural rainfall and a limited amount of irrigation during an extended period of dry weather, and leaching of macronutrients and metals was monitored over a 3-year period. The Horotui soil lysimeters were located at Hamilton in the North Island of New Zealand and experienced a total of 3800 mm of rainfall, and an average 1279 mm of drainage, during the 3 years of the experiment. The other soil lysimeters were located at Lincoln in the South Island of New Zealand and experienced only 1825 mm of rainfall during the experiment. Average drainage ranged from 1086 mm for the Waikuku lysimeters to 363 rain for the Templeton lysimeters (McLaren el al. 2003). At Lincoln, the years 1997 and 1998 were relatively dry with <500 mm of rainfall per annum, compared with the long-term average of approximately 660 ram.
On completion of the leaching phase of the study, the lysimeters were destructively sampled in the following way. For the forest soils (Waikuku, Lismore. Lyre), it was still possible to identify some residual sludge material on the soil surface. This was first removed from the lysimeters, being careful not to disturb the underlying forest litter. The forest litter layer was then sampled separately from the underlying mineral soil. The total weights of sludge residue and litter removed from each lysimeter were recorded after drying at 60[degrees]C. The samples were then finely ground using a Tema ring mill.
Following removal of the sludge residue and litter from the forest soils, each lysimeter (including the Templeton and Horotiu pasture soils) was transferred carefully to the laboratory where its base plate and drainage system were removed before placement on a circular table designed to fit just inside the lysimeter casing. By forcing the lysimeter casing downwards, the soil was exposed at tile top of the lysimeter and could then be sampled at the required depth increments. For the forest soils, sampling was carried out in 0.02-m increments down to 0.10m, followed by several 0.05-m increments before increasing the final increments to 0.10 m. In the case of the Lismore soil, sampling was not possible below 0.35 m because of the coarse gravely nature of the material. For the pasture soils, sampling was carried out in 0.05-m increments from the surface, increasing to 0.10-m increments after 0.30m for the Templeton soil.
Before sampling the soil layers, vaseline (petroleum jelly), used to prevent edge flow during leaching (Cameron et al. 1992), was removed from the edge of the exposed soil. In the case of 0.02- and 0.05-m soil increments, the complete soil layer was then carefully sliced off, placed in a polypropylene container, and thoroughly mixed before taking a subsample. For 0.10-m increments, the soil layer was quartered and 2 opposite quarters removed for mixing together prior to subsampling. All sub-samples were air-dried and ground to pass a stainless steel 2-mm sieve. A subsample of the < 2 mm material was then finely ground using a pestle and mortar.
Total metal analysis
Total metal concentrations in tile soil, litter and sludge residue samples were determined using finely ground subsamples. The samples were digested in nitric acid and hydrogen peroxide as described by Kovacs et al. (2000), and metal concentrations (Cd, Cr, Cu, Ni, Pb, Zn) in the digests determined by flame or graphite furnace atomic absorption spectrophotometry (FAAS/GFAAS).
EDTA-extructable soil metals
EDTA-extractable metals were determined as described by McLaren et al. (1984). Samples (10g) of <2mm soil were shaken for 2 h with 25 mL 0.04 M EDTA (disodium salt adjusted to pH 6.0 with NaOH) and the resulting suspensions were centrifuged filtered and metals determined in the filtrates by FAAS or GFAAS.
Calcium nitrate-extractable metals
Samples (5g) of <2 mm soil were shaken for 2h with 30 mL 0.01M Ca[([NO.sub.3]).sub.2] solution and the resulting suspensions were centrifuged, filtered, and metals determined in the filtrates by FAAS or GFAAS.
General soil properties
In addition to the metal analyses, samples from one of the control lysimeters for each soil were analysed for properties that might be expected to influence metal leaching, i.e. soil pH, organic carbon, cation exchange capacity (CEC), and oxalate-extractable iron (Fe) and aluminium (Al). The results of these analyses are shown in Table 1.
Calculation of standard errors and tests of statistical difference (t-tests) were carried out using Minitab[R] Release 14 statistical software.
Results and discussion
Total metal concentrations in sludge residues and litter samples from forest soils
All the identifiable sludge residues recovered from the forest soil lysimeters, irrespective of soil type, had remarkably similar metal concentrations (coefficients of variation, 2.7-5.7%). Concentrations of Cr, Cu, and Pb in the sludge residues were 20-23% higher (P<0.001) than in the original sludge applied 3 years earlier (Table 2), and Cd concentrations were approximately 5% higher (P < 0.05). In contrast, Ni and Zn concentrations were 48% and 18% lower (P< 0.001) than in the original sludge (Table 2). Clearly, during the 3 years of the leaching study there had been substantial differences between metals in their solubilisation and release from the sludge. Increases in concentration, as in the case of Cr, Cu, Pb, and Cd, result from a relatively low solubilisation compared with the rate of sludge organic matter decomposition and loss of mass, mainly as [CO.sub.2]. In the case of Cr, Cu, and Pb, this can be explained by the relatively small proportions (0.1-4%) of these metals present in the original spiked sludge in soluble and exchangeable forms (McLaren et al. 2004). In comparison, the relatively high losses of Ni and Zn from the sludge residue are no doubt related to the higher proportions (>20%) of these metals in soluble and exchangeable forms in the spiked sludge. However, the apparent lack of mobility of Cd was somewhat surprising, in that 19% of this metal occurred in soluble and exchangeable forms in the original sludge (McLaren et al. 2004). However, compared to the other metals, Cd was present in much smaller concentrations and may well have been retained strongly by the remaining sludge organic matter. Certainly, organic matter has been shown to be the most important soil component controlling Cd sorption and desorption in soils (Gray et al. 1998).
Irrespective of their relative mobility as indicated by their concentrations in the sludge residue, concentrations of all 6 metals were significantly (P < 0.001) increased in the underlying forest litter layer compared with the untreated controls (Table 2). This accumulation could be a result of the transport of metals in solution or associated with fine colloidal particles no longer identifiable as 'sludge'. The transport and mixing of sludge particles might be expected to result in the movement of metals in similar proportions to those in which they occur in the sludge. If metal transport takes place predominantly in solution, increasing differential mobility between metals might be expected. McBride et al. (1997, 1999) have investigated the differential mobility of metals in sludge-treated soil using individual element/Cr ratios. This analysis is based on the assumption that Cr (as [Cr.sub.3+]), because of its strong binding to soil particles, is one of the least mobile metals.
Figure 1 shows the metal/Cr ratios for Cd, Cu, Ni, Pb, and Zn in the original sludge, sludge residue, and forest litter layer. For Cd, Ni, and Zn there are large increases in the metal/Or ratio in the litter layer compared with the original sludge or sludge residue, suggesting that transport in solution is very important for these metals. Even for Cu and Pb there is some evidence of differential mobility compared with Cr (Fig. 1). However, it should be noted that Cr itself was not totally immobile, and that there was a substantial increase in Cr concentration in the litter layer (Table 2). The extent to which this Cr, and indeed proportions of the other metals, might have been transported in association with finely divided sludge material is impossible to determine from our data.
[FIGURE 1 OMITTED]
The extent to which the metals added in the sludge were recovered in the sludge residue and litter layer can be seen from the data in Table 3. It is clear that, even for Cr, Cu, and Pb, approximately 35% of the metals applied in the sludge had apparently moved below the litter layer into the underlying mineral soil. In the case of Ni and Zn, 57% and 50%, respectively, appeared to have been transported into the mineral soil during the 3 years of the leaching study. Somewhat surprisingly, Cd showed the lowest estimated movement below the litter layer (10%). However, as discussed above, this may be due to the relatively small amount of Cd applied and its strong retention by organic components of the sludge residue and litter layer. It should be noted that the movements of metals into the mineral soil shown in Table 3 are only estimates, based on their recoveries in the sludge residue and litter. Unfortunately, because of the variation in background metal concentrations between lysimeters, and the lack of accurate bulk density values, it was not possible to directly determine meaningful metal recoveries for the mineral soil layers, or, therefore, calculate overall mass balances for the metals.
Total metal concentrations in mineral soil layers of forest soils
As indicated by the data in Table 3, beneath the forest litter layer, for some of the metals there were substantial increases in concentration in the first few 0.02-m increments of mineral soil due to sludge treatment (Table 4). However, relatively few of these increases in metal concentration were statistically significant, even where the increases were relatively large. This is due predominantly to the large variability that occurred between replicate lysimeters. In the case of Cr, Cu, and Pb, increases in concentration appeared to be restricted to the top 0.04-0.06 m of soil. Below this, differences in concentration of these metals between control and sludge-treated lysimeters were generally of the same order of magnitude as their analytical detection limits. For Cd and Ni, differences between control and sludge-treated lysimeters were observed clearly down to depths of 0.06-0.10m (Table 4) before the differences approached analytical detection limits. Zinc was the only metal showing substantial increases in concentration for all forest soils down to 0.10 m as a result of sludge treatment.
Differences in total soil metal concentrations between sludge-treated and control lysimeters were used to calculate mean metal/Cr ratios of the sludge-derived metals in the uppermost mineral soil layers (Table 5). This was only possible for the top 0.04-0.06-m layers where there were marked increases in Cr concentration. In the case of Cu, the Cu/Cr ratios were very similar to those in the original sludge, sludge residue, and litter, indicating very little difference in mobility compared with Cr. The same was true for Pb in the Waikuku soil, although Pb/Cr ratios did show an increase with depth in the Lismore and Eyre soils (Table 5). However, this trend was even more pronounced for Cd, Ni, and Zn in all 3 soils. The metal/Cr ratios for these elements were generally much higher than in the overlying litter layer, indicating substantial differential mobility compared with Cr.
Extractable metal concentrations in mineral soil layers of forest soils'
The detection of metal movement down the soil profile solely by analysis of total soil metal concentrations has considerable limitations. For some metals in particular, the sensitivity for this type of analysis is not high, and it is difficult, if not impossible, to detect the movement of low concentrations of metals against relatively high background concentrations. The problem is compounded by the natural variations in total background metal concentrations between individual soil lysimeters. An alternative approach is to determine only those metals that occur in relatively labile or soluble forms. As a result of using larger amounts of soil for these types of analysis, detection limits and sensitivities are often much lower than for total analysis. In the present study we have used EDTA extraction to estimate labile pools of metals, and calcium nitrate extraction to estimate soluble metals.
The soil profile distributions of sludge-derived EDTA-and Ca([NO.sub.3)2]-extractable metals for the 3 forest soils are shown in Figs 2 and 3. The data shown in these figures were calculated by difference between metal concentrations in the control and sludge-treated lysimeters. No data are shown for Cr since concentrations extracted by both EDTA and Ca([NO.sub.3)2] were very low and generally close to or below detection limits. Data for other metals were also omitted where concentrations in control and/or sludge-treated samples were close to detection limits.
[FIGURES 2-3 OMITTED]
For all metals, the EDTA-extractable concentrations demonstrated a much deeper movement down the soil profile (Figs 2 and 3) than that suggested by the total metal data shown in Table 4. Sludge-derived extractable Zn was observed down to 0.50-0.60m, Ni down to 0.40-0.50 m, Cd down to 0.20-0.30 m, and Cu and Pb down to 0.10-0.20 m. Somewhat surprisingly, there was little difference in distribution between the 3 soils, in spite of substantial differences in soil properties (Table 1). It might have been expected that movement would have been greatest in the Waikuku soil, since, of the 3 forest soils, this had the coarsest texture, lowest organic C, lowest CEC, and lowest oxalate Fe and Al. There was also more drainage from this soil (McLaren et al. 2003). However, the Waikuku soil did have generally higher pH values than the Lismore or Eyre soils and this may have mitigated metal leaching to some extent. In addition, the Lismore and Eyre soil profiles had more structural development than the Waikuku soil, which may have enhanced leaching. It has been shown that macropore flow is an important process for metal leaching in these soils (McLaren et al. 2004).
For Ni and Zn, and to lesser extent for Cd, Ca([NO.sub.3)2]-extractable metal concentrations as a result of sludge treatment also showed substantial increases to depths similar to those observed with the EDTA-extractable metals (Fig. 2). In fact, substantial proportions of the increased labile metal pools, as estimated with EDTA extraction, were also extractable with Ca([NO.sub.3)2]; on average approximately 50%, 56%, and 74% for Cd, Ni, and Zn, respectively. Thus, substantial concentrations of these 3 metals would appear to remain in readily soluble forms within the soil profile. In contrast, for Cu and Pb, relatively small proportions (1-5%) of their EDTA-extractable pools were in readily soluble forms (Fig. 3). Both these metals are known to bind very strongly with organic matter.
Residual metal concentrations in pasture soils
Total metal concentrations in the Templeton and Horotiu soil lysimeters are shown in Figs 4 and 5. Although the intention had been to incorporate the sludge as uniformly as possible into the top 0.10m of soil, it is clear from the data in Figs 4 and 5 that this was not achieved. For both soils, metal concentrations in the top 0-0.05 m layer of soil were much higher than for the 0.05-0.10 m layer. It is also clear that although the same amount of sludge was applied to both soils, expressed on a dry soil weight basis, metal concentrations in the Horotiu soil were much higher than for the Templeton soil. This is due to the considerable difference in dry bulk densities between the 2 soils. The dry bulk density of the top 0-0.20 m of the Templeton soil is approximately 1.3 g/[mL.sup.3], whereas that of the Horotiu soil ranges from 0.82 to 0.89 g/[mL.sup.3].
[FIGURES 4-5 OMITTED]
For both soils, there was some evidence of small increases in metal concentration (mainly Ni and Zn) in the 0.10-0.15 m layer of soil below the layer of incorporation, but this could have occurred during incorporation rather than by subsequent leaching. Certainly, the data in Figs 4 and 5 provide no evidence of metal movement below 0.15m. In contrast to the forest soils, this was also true for the EDTA- and Ca([NO.sub.3)2]-extractable metal data (not shown).
However, examination of metal/Cr ratios of the sludge-derived metals (sludge treated-control concentrations) did provide some evidence of metal mobilisation in the pasture soils (Fig. 6). For the Horotiu soil, metal/Cr ratios for Cd, Ni, and Zn in the 0-0.05 m layer of soil were lower than in the original sludge indicating some downward movement of these 3 metals compared with Cr. This is confirmed by the greatly increased metal/Cr ratios for these metals in the underlying 0.05-0.10 and 0.10-0.15m layers (Fig. 6). In contrast to Cd, Ni, and Zn, there was little evidence of differential transport of Cu and Pb. A similar pattern was observed for the Templeton soil, with one difference. For Cd, Ni, and Zn, although the differences were not statistically significant, metal/Cr ratios in the 0-0.05 m layer of soil were actually higher than for the original sludge. This could only have resulted from the differential movement of these 3 metals (relative to Cr) upwards into this layer. The most likely explanation for these phenomena is the upward movement of these metals in soil moisture in response to evapo-transpiration forces. Although we have no direct evidence that this occurred, it should be noted that the total drainage from the Templeton soil during the 3 years of the experiment was only one-third of that from the Horotui soil (McLaren et al. 2003). In particular, during the first 2 years (1997 and 1998), the Templeton soil was subject to an extended period of very dry conditions during which there was minimal leaching. Lack of a similar net upward movement of metals in the Horotui soil could be due to a combination of higher total drainage of water through the soil, and stronger metal binding properties of the Horotui soil (see General discussion below).
[FIGURE 6 OMITTED]
As discussed previously, with regard to metals, the lysimeter study described in this series of papers can be regarded as a worst-case scenario (McLaren et al. 2004). However, there is no doubt from the results described above that, in the forest soils, considerable movement of metals down the soil profile has taken place during the 3 years of the experiment. The distribution with depth of extractable metal concentrations (EDTA or Ca([NO.sub.3)2]-extractable; Figs 2 and 3) show patterns that are consistent with movement by convective and diffusive solute transport processes. Maximum depths of metal penetration varied between metals, but were similar for all 3 forest soils, averaging 0.15, 0.20, 0.25, 0.35, and 0.45 in for Pb, Cu, Cd, Ni, and Zn, respectively. These depths of penetration were far greater than determined from examination of total soil metal concentrations alone. Baveye et al. (1999) have also observed higher extractable (DTPA) metal concentrations in the subsoils of sludge-treated soils, despite the absence of detectable elevated total metal concentrations. In our study, the relative depths of penetration of the metals were also consistent with the relative increases in metal/Cr ratios in the litter and underlying mineral soil layers (Fig. 1 and Table 5). The use of increasing metal/Cr ratios in this way, to indicate relative metal mobility from surface-applied sludge, is somewhat different from their use by McBride et al. (1999). Those authors used decreased metal/Cr ratios in sludge-amended surface soil to estimate movement of metals into the subsoil and beyond.
In the present study, in no case were increased extractable metal concentrations detected right to the bottom of the lysimeters. This lends support to the contention that the small amounts of metals leached from the lysimeters over the 3 years prior to soil sampling resulted from macropore flow. However, the metal distribution patterns suggest that with continued drainage, the leaching fronts would have eventually reached the base of the lysimeters, resulting in further, and possibly more substantial and longer, metal leaching pulses. It should be noted that high proportions of the transported metals appeared to be present in relatively weakly bound forms extractable with Ca([NO.sub.3)2]. However, it should also be emphasised that the mineral soil layers of all 3 forest soils used in this study had properties that might be expected to be conducive to metal leaching, i.e. relatively low pH, low organic C, and low CEC (Table 1).
In contrast to the forest soils, evidence of metal transport in lysimeters of the 2 pasture soils was far less obvious. Total, EDTA-extractable, or Ca([NO.sub.3)2]-extractable metal concentration did not provide any convincing evidence of movement of metals below the level of sludge incorporation. However, in spite of this, the variation in metal/Cr ratios did suggest that some differential movement of Cd, Ni, and Zn had taken place in the lysimeters of both soils. This certainly agrees with the observation that sludge application dick in fact, increase Zn concentrations in leachates from both pasture soils, and Ni concentrations in leachates from the Templeton soil (McLaren et al. 2004). The relatively low degree of metal mobility in the 2 pasture soils compared with the forest soils most likely results from a combination of factors. In particular, the sludge was mixed into the top 0.10m of the pasture soils, a process likely to speed tip the binding of metals in the sludge to soil surfaces. The Horotiu soil in particular has relatively high organic carbon content and high levels of iron and allophanic minerals (Table 1), soil constituents known to have high metal-binding capacities (McLaren et al. 2003).
The results from this study indicate that there are large differences in the extent to which metals are likely to be transported through soils following applications of sewage sludge. This clearly has implications for selection of the most appropriate soils and conditions for sludge application to the land. The results also raise concerns regarding the potential long-term leaching of metals from soils. Short-term monitoring of soils following sludge application may measure transitory metal leaching resulting from macropore flow, but may fail to determine the much slower movement of metals by convective and diffusive transport processes. Compared with the poor sensitivity of total soil metal determinations, the calculation of metal/Cr ratios appears to provide a good indication of metal movement, even in the early stages following sludge application.
Table 1. Some properties of the experimental soils (mineral layers) Depth Texture pH (A) Soil C (B) CEC (A) (m) ([H.sub.2]O) (%) (cmol(+)/kg) Waikuku 0-0.10 Loamy sand 5.4 0.57 3.4 0.10-0.20 over sand 5.5 0.38 3.3 0.20-0.30 5.8 0.19 2.0 0.30-0.50 5.9 0.13 1.7 0.50-0.70 6.0 0.08 5.0 Lismore 0-0.10 Silt loam 5.1 3.71 14.6 0.10-0.20 over 5.3 2.29 11.5 0.20-0.30 gravels 5.4 1.98 11.2 Eyre 0-0.10 Silt loam 4.5 4.03 16.8 0.10-0.20 over 4.8 2.60 14.4 0.20-0.30 gravels 5.2 1.22 11.1 0.30-0.40 5.4 0.89 12.2 0.40-0.60 5.7 0.73 10.2 Templeton 0-0.10 Fine sandy 5.6 3.13 13.3 0.10-0.20 loam 5.7 2.45 11.5 0.20-0.30 5.7 1.44 9.3 0.30-0.50 6.1 0.43 6.5 0.50-0.70 6.4 0.18 5.7 Horotiu 0-0.10 Sandy loam 5.2 7.29 29.9 0.10-0.20 5.2 5.39 23.6 0.20-0.30 5.2 2.46 12.4 0.30-0.40 5.3 0.55 5.2 0.40-0.70 5.5 0.41 4.6 Depth Oxalate-extractable (A) (m) Fe (%) A1 (%) Waikuku 0-0.10 0.26 0.02 0.10-0.20 0.20 0.02 0.20-0.30 0.23 0.01 0.30-0.50 0.21 0.03 0.50-0.70 0.01 0.00 Lismore 0-0.10 0.33 0.21 0.10-0.20 0.42 0.23 0.20-0.30 0.14 0.20 Eyre 0-0.10 0.36 0.20 0.10-0.20 0.37 0.21 0.20-0.30 0.45 0.19 0.30-0.40 0.83 0.22 0.40-0.60 0.71 0.40 Templeton 0-0.10 0.48 0.22 0.10-0.20 0.40 0.21 0.20-0.30 0.78 0.25 0.30-0.50 0.51 0.15 0.50-0.70 0.41 0.11 Horotiu 0-0.10 0.58 2.01 0.10-0.20 0.68 2.49 0.20-0.30 0.65 2.32 0.30-0.40 0.46 1.50 0.40-0.70 0.52 1.43 (A) Methods as described by Blakemore et al. (1987). (B) LECO CNS 2000 Analyser. Table 2. Metal concentrations in the original sludge, and in sludge residue and forest litter at the end of the study (mg/kg [+ or -] s.e.) Metal Original sludge (A) Sludge residue (B) Cd 149 [+ or -] 1.6 156 [+ or -] 2.5 Cr 36 257 [+ or -] 50.2 44 522 [+ or -] 699.9 Cu 7440 [+ or -] 115.5 8929 [+ or -] 136.8 Ni 1777 [+ or -] 0.8 931 [+ or -] 8.5 Pb 16 085 [+ or -] 91.6 19 689 [+ or -] 372.2 Zn 17 288 [+ or -] 34.5 14 203 [+ or -] 220.0 Metal Forest Litter Controls (C) Sludge-treated (B) Cd 0.44 [+ or -] 0.09 15.3 [+ or -] 2.0 Cr 71.1 [+ or -] 10.4 2625 [+ or -] 361 Cu 22.3 [+ or -] 6.5 628 [+ or -] 31 Ni 10.7 [+ or -] 0.9 227 [+ or -] 31 Pb 43.0 [+ or -] 3.7 1294 [+ or -] 154 Zn 78.43 [+ or -] 5.5 1682 [+ or -] 210 (A) s.e. based on replicate analyses of single sample. (B) n=9. (C) n=6. Table 3. Recoveries (% [+ or -] s.e.) of applied metals in residual sludge and litter of forest soils Recovery in residual sludge (%) = [M.sub.RS] x 100/[M.sub.S], where: [M.sub.RS] is metal concentration in residual sludge from treated lysimeters (mg/kg) x weight of residual sludge from treated lysimeter (kg/lysimeter), and [M.sub.S] is metal added to lysimeter in sludge application (mg/lysimeter) Recovery in litter layer (%) = ([M.sub.LT]-[M.sub.LC]) x 100/Ms, where: [M.sub.LT] is metal concentration in litter from treated lysimeter (mg/kg) x weight of litter from treated lysimeter (kg), and [M.sub.LC] is metal concentration in litter from control lysimeter (mg/kg) x weight of litter from control lysimeter (kg) (mean control values for each soil used in recovery calculation for individual treated lysimeters) Estimated movement into mineral soil (%) = 100 - recovery in residual sludge-recovery in litter layer Metal Residual sludge Litter layer Cd 66.3 [+ or -] 4.0 23.1 [+ or -] 2.8 Cr 52.5 [+ or -] 3.0 10.7 [+ or -] 1.0 Cu 51.5 [+ or -] 3.2 14.2 [+ or -] 1.8 Pb 52.3 [+ or -] 3.0 12.1 [+ or -] 1.2 Ni 22.8 [+ or -] 1.6 19.8 [+ or -] 3.2 Zn 35.3 [+ or -] 2.4 14.7 [+ or -] 1.9 Metal Estimated movement into mineral soil Cd 10.6 [+ or -] 6.2 Cr 36.7 [+ or -] 3.8 Cu 34.2 [+ or -] 3.5 Pb 57.4 [+ or -] 4.4 Ni 35.6 [+ or -] 3.8 Zn 50.0 [+ or -] 4.0 Table 4. Mean total metal concentrations (mg/kg) in top 0.10m of mineral soil in control and sludge-treated forest soil lysimeters Depth Waikuku soil Lismore soil Eyre soil (m) Control +Sludge Control +Sludge Control +Sludge Cadmium 0-0.02 0.17 1.45 * 0.04 1.51 ** 0.10 1.48 * 0.02-0.04 0.16 0.37 0.02 0.56 0.06 0.37 * 0.04-0.06 0.15 0.22 0.01 0.22 * 0.08 0.11 0.06-0.08 0.08 0.19 0.02 0.06 * 0.06 0.06 0.08-0.10 0.08 0.18 0.01 0.05 0.04 0.04 Chromium 0-0.02 14.7 197.0 22.9 429.0 22.7 285.0 0.02-0.04 13.5 30.9 18.6 175.0 17.7 38.7 0.04-0.06 8.4 23.3 16.3 39.7 16.7 20.1 0.06-0.08 8.3 10.5 16.1 23.6 17.2 17.8 0.08-0.10 7.9 10.8 15.6 21.9 17.5 17.9 Copper 0-0.02 3.8 41.4 5.2 89.8 6.5 71.3 * 0.02-0.04 2.0 7.0 3.8 32.9 4.4 12.9 * 0.04-0.06 2.1 4.1 3.5 8.7 4.9 5.4 0.06-0.08 2.2 5.0 2.9 5.2 3.7 4.3 0.08-0.10 2.7 3.0 3.0 4.0 3.6 4.0 Nickel 0-0.02 7.7 30.6 6.4 60.2 * 15.4 58.7 * 0.02-0.04 8.15 13.8 6.5 35.7 12.0 27.2 * 0.04-0.06 7.5 10.3 8.8 17.1 12.9 17.2 * 0.06-0.08 8.2 11.0 7.2 10.6 13.1 17.2 0.08-0.10 7.0 8.3 6.6 11.0 11.7 13.5 Lead 0-0.02 8.9 78.6 18.0 202.0 45.5 140.9 0.02-0.04 6.0 13.1 12.8 90.9 47.5 52.6 0.04-0.06 5.0 7.2 10.5 23.0 43.5 38.7 0.06-0.08 4.9 8.9 9.9 14.2 24.2 36.4 0.08-0.10 5.5 8.4 9.1 12.3 29.6 32.1 Zinc 0-0.02 29.1 188.0 58.9 344.0 85.7 359.6 * 0.02-0.04 20.5 63.7 53.3 194.2 79.8 157.8 * 0.04-0.06 21.1 37.8 52.3 115.9 82.8 116.2 0.06-0.08 18.9 31.6 * 52.2 85.2 * 79.6 104.3 0.08-0.10 19.4 30.8 * 51.7 73.2 * 79.9 95.4 * P < 0.05; ** P < 0.01 for difference between control and sludge-treated lysimeters. Table 5. Mean metal/Cr ratios ([+ or -] s.e.) of sludge-derived metals in upper mineral soil layers of Waikuku, Lismore, and Eyre forest soils Depth (m) Cd/Cr Cu/Cr Waikuku soil 0-0.02 0.010 [+ or -] 0.0046 0.2 [+ or -] 0.01 0.02-0.04 0.011 [+ or -] 0.0003 0.2 [+ or -] 0.03 0.04-0.06 0.012 [+ or -] 0.0114 0.2 [+ or -] 0.07 Lismore soil 0-0.02 0.004 [+ or -] 0.0009 0.2 [+ or -] 0.01 0.02-0.04 0.008 [+ or -] 0.0029 0.3 [+ or -] 0.06 0.04-0.06 0.017 [+ or -] 0.0064 0.2 [+ or -] 0.03 Eyre soil 0-0.02 0.006 [+ or -] 0.0008 0.3 [+ or -] 0.01 0.02-0.04 0.017 [+ or -] 0.0034 0.4 [+ or -] 0.03 Depth (m) Pb/Cr Ni/Cr Waikuku soil 0-0.02 0.4 [+ or -] 0.02 0.12 [+ or -] 0.018 0.02-0.04 0.4 [+ or -] 0.05 0.32 [+ or -] 0.155 0.04-0.06 0.3 [+ or -] 0.13 0.29 [+ or -] 0.000 Lismore soil 0-0.02 0.5 [+ or -] 0.01 0.15 [+ or -] 0.027 0.02-0.04 0.9 [+ or -] 0.49 0.46 [+ or -] 0.194 0.04-0.06 0.9 [+ or -] 0.46 0.45 [+ or -] 0.067 Eyre soil 0-0.02 0.4 [+ or -] 0.02 0.17 [+ or -] 0.015 0.02-0.04 1.8 [+ or -] 0.00 0.81 [+ or -] 0.197 Depth (m) Zn/Cr Waikuku soil 0-0.02 0.8 [+ or -] 0.11 0.02-0.04 2.0 [+ or -] 0.15 0.04-0.06 1.5 [+ or -] 0.42 Lismore soil 0-0.02 0.8 [+ or -] 0.08 0.02-0.04 2.3 [+ or -] 0.98 0.04-0.06 4.8 [+ or -] 1.60 Eyre soil 0-0.02 1.1 [+ or -] 0.12 0.02-0.04 4.1 [+ or -] 0.78
This work is funded by the Foundation for Research, Science and Technology (contract C03X0205) through the Institute of Environmental Science and Research, Ltd (ESR). The authors acknowledge the support and encouragement of the research program leader Dr Tom Speir of ESR.
Baveye P, McBride MB, Bouldin D, Hinesly TD, Dahdoh MSA, Abdel-Sabour MF (1999) Mass balance and distribution of sludge-borne trace elements in a silt loam soil following long-term applications of sewage sludge. The Science of the Total Environment 227, 13-28. doi: 10.1016/S0048-9697(98)00396-9
Blakemore LC, Searle PL, Daly BK (1987) Methods for chemical analysis of soils. New Zealand Soil Bureau Scientific Report No. 80, NZ Soil Bureau, Lower Hutt, New Zealand.
Cameron KC, Smith NR McLay CDA, Fraser PA, McPherson RJ, Harrison DE Harbottle P (1992) Lysimeters without edge flow: an improved design and sampling procedure. Soil Science Society of America Journal 56, 1625-1628.
Camobreco VJ, Richards BK, Steenhuis TS, Peverly JH, McBride MB (1996) Movement of heavy metals through undisturbed and homogenised soil columns. Soil Science 161, 740-750. doi: 10.1097/00010694-199611000-00003
Chang AC, Warneke JA, Page AL, Lund LJ (1984) Accumulation of heavy metals in sewage sludge-treated soils. Journal of Environmental Quality 13, 87-91.
Dowdy RH, Latterell JJ, Hinesly TD, Grussman RB, Sullivan DL (1991) Trace metal movement in an Aeric Ochraqualf following 14 years of annual sludge applications. Journal of Environmental Quality 20, 119-123.
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. doi: 10.1071/$97085
Keller C, McGrath SP, Dunham SJ (2002) Trace metal leaching through a soil--grassland system after sewage sludge application. Journal of Environmental Quality 31, 1550-1560.
Kovacs B, Prokisch J, Gyori Z, Kovacs AB, Palencsar AJ (2000) Studies on soil sample preparation for inductively coupled plasma atomic emission spectrometry analysis. Communications in Soil Science and Plant Analysis 31, 1949-1963.
McBride MB, Richards BK, Steenhuis TS, Russo JJ, Sauve S (1997) Mobility and solubility of toxic metals and nutrients in soil fifteen years after sludge application. Soil Science 162, 487-500. doi: 10.1097/00010694-199707000-00004
McBride MB, Richards BK, Steenhuis TS, Spiers G (1999) Long-term leaching of trace elements in a heavily sludge-amended silty clay loam soil. Soil Science 164. 613-623. doi: 10.1097/00010694-199909000-00001
McLaren RG, Clucas LM, Taylor MD, Hendry T (2003) Leaching of macronutrients and metals from undisturbed soils treated with metal-spiked sewage sludge. 1. Leaching of macronutrients. Australian Journal of Soil Research 41, 571-588. doi: 10.1071/SR02092
McLaren RG, Clucas LM, Taylor MD, Hendry T (2004) Leaching of macronutrients and metals from undisturbed soils treated with metal-spiked sewage sludge. 2. Leaching of metals. Australian Journal of Soil Research 42, 459-471.
McLaren RG, Swift RS, Quin BF (1984) EDTA-extractable copper, zinc, and manganese in soils of the Canterbury Plains. New Zealand Journal of Agricultural Research 27, 207-217.
Richards BK, Steenhuis TS, Peverly JH. McBride MB (1998) Metal mobility at an old, heavily-loaded sludge application site. Evironmental Pollution 99, 365-377. doi: 10.1016/S02697491 (98)00011-6
Roy M, Couillard D (1998) Metal leaching following sludge application to a deciduous forest soil. Water Research 32. 1642-1652. doi: 10.1016/S0043-1354(97)00375-8
Smith SR (1996) 'Agricultural recycling of sewage sludge and the environment.' (CAB International: Wallingford, UK)
Speir TW, Van Schaik AP, Percival HJ, Close ME, Pang L (2003) Heavy metals in soil, plants and groundwater following high-rate sewage sludge application to land. Water, Air, and Soil Pollution 150, 319-358. doi: 10.1023/A:1026101419961
Williams DE, Vlamis J, Pukite AH, Corey JE (1985) Metal movement in sludge-treated soils alter six years of sludge addition. 2: Nickel, cobalt, iron, manganese, chromium and mercury. Soil Science 140, 120-125.
Manuscript received 9 July 2004, accepted 26 November 2004
R. G. McLaren(A,C), L. M. Clucas(A), and M. D. Taylor(B)
(A) Centre for Soil and Environmental Quality, Soil, Plant and Ecological Sciences Division, PO Box 84, Lincoln University, Canterbury, New Zealand.
(B) Landcare Research, Private Bag 3127, Hamilton, New Zealand.
(C) Corresponding author. Email: email@example.com
|Printer friendly Cite/link Email Feedback|
|Author:||McLaren, R.G.; Clucas, L.M.; Taylor, M.D.|
|Publication:||Australian Journal of Soil Research|
|Date:||Mar 1, 2005|
|Previous Article:||Mineral and elemental distribution in soils formed on the River Niger floodplain, eastern Nigeria.|
|Next Article:||Enumeration of wax-degrading microorganisms in water repellent soils using a miniaturised Most-Probable-Number method.|