|dc.description.abstract||Successive biosolids applications increase soil levels of heavy metals which can be toxic to soil organisms and plants, as well as humans and animals in the food chain. Current guidelines for metals permitted in soils subjected to biosolids applications were established on the basis of total soil metal concentrations. But some research suggests that total soil metal concentrations are not a satisfactory index for estimating bioavailability of metals to plants.
A field trial was established at Bromley to study the effects of metal concentration and metal bioavailability on plants grown in soils amended with biosolids. Bromley farmland has had successive biosolids applications for about 30 years resulting in sites with metal concentrations ranging from low to high levels. Five sites were chosen that covered a range from low to high soil metal concentrations. Each site was divided into 8 plots of 0.5 by 2.0 m and sown with three crops: lucerne (Medicago sativa), barley (Hordeum vulgare) and fodder beet (Beta vulgaris). Analysis of soils included total soil metal concentrations (Cd, Cr, Cu, Zn, Ni and Pb) and selected soil properties. Several chemical extractants (AAAc-EDTA, NH₄NO₃, NH₄OAc, EDTA, CaCl₂, CaNO₃ and NH₄Cl) for estimating soil metal bioavailability were compared. Plant root and shoot metal concentrations at the five sites were measured.
Results indicated that there are clear differences in the bioavailability of heavy metals in biosolids-amended soils among different plant species. The field trial confirmed that total soil metal concentrations for the assessment of soil metal bioavailability are of limited value compared with the use of weak chemical extractants such as NH₄NO₃, CaCl₂, CaNO₃, NH₄Cl, and NH₄OAc. But, the number of sites used in this field study was small. It was therefore desirable to expand the study with further trials.
Glasshouse trials were established to further examine the use of weak chemical extractants to estimate soil metal bioavailability and to provide data to improve the recommendations of thresholds for plants grown on biosolids-amended soils. Thirteen sites were chosen including the original five experimental field sites. A pot experiment using soils from the thirteen sites was set up in a glasshouse. Pots were placed in a Yoden incomplete Latin square design, replicated four times (n=52). Each pot was sown with silver beet (Beta vulgaris) seeds and grown until they reached maturity (~60 days). A second glasshouse trial was carried out using the same thirteen sites, but replacing silver beet with lucerne. Analyses performed included total metal concentrations (Cd, Cr, Cu, Zn, Ni and Pb) and selected soil properties. Estimates of soil metal bioavailability were made using extractions with EDTA, NH₄NO₃ and CaCl₂. Plant shoot metal concentrations were measured.
Results indicated that silver beet accumulated higher metal concentrations than lucerne, particularly Cd. This confirmed the difference in bioavailability of heavy metals in biosolids-treated soils between plant species found in the field trial. Plant metal concentrations were significantly correlated with total and EDTA extractable (strong) metals as well as with NH₄NO₃ and CaCl₂ extractable (weak) metals. These positive correlations were not consistent with the field trial results. The field trial and glasshouse trials both had a large range of metal concentrations, but the glasshouse trial had a narrow range of soil properties. Significant correlations of plant metal concentrations with total soil metal concentrations are not unusual in soils with a large range of metal concentrations and similar soil properties as existed in our glasshouse trial.
The field research area was located next to the Estuary of the Heathcote and Avon rivers, which periodically floods saline water over the farmland containing our sites. Soil salinity can increase plant uptake of Cd, possibly by forming complexes that increase the bioavailability of Cd to the plants. An additional glasshouse trial was established to measure metal concentration and uptake of plants grown in soil amended with biosolids containing varying salt concentrations. One site was chosen from the five experimental field sites to keep total soil metal concentrations constant, which then enabled us to use a range of salinity treatments. Two pot experiments were set up in a glasshouse. Pots were placed in a complete Latin square design consisting of four treatments: NaCl concentrations at 0 (control), 1000 (low), 5000 (medium) and 10,000 (high) mg L⁻¹, replicated four times (n=16). Each pot was sown with silver beet seeds and supplied with salinity treatments (100 mL per pot) every second day. Sodium chloride concentrations were chosen based on conductivity levels of 855 µS cm⁻¹ (control), 1700; 3300 and 6600 µS cm⁻¹ , respectively to, which silver beet were found to vary in tolerance. The duplicate pot experiment was used to prevent disturbance to the plants in the main experimental set when removing soil for measuring conductivity levels. Another set of pots containing soil from the same site was placed in an incubator at 20°C and the same salinity treatments and statistical design adopted. These pots were set up to determine the relationship between extractable metal concentrations and soil salinity. Analyses performed included total metal concentrations (Cd, Cr, Cu, Zn, Ni and Pb) and selected soil properties. Estimates of soil metal bioavailability were made using extractions with NH₄NO₃. Plant shoot metal concentrations were measured and plant uptake of metals calculated.
Results indicated that concentrations and uptake of some metals by plants are affected by increasing salt concentrations. Metals in the soil solution as determined by the weak extractant did not appear to be directly affected by increasing salt concentrations. Instead, they appeared to be affected by an increase in soil pH, which increased at the higher salt concentrations. The effect of pH on extractable metals may have indirectly affected metal concentrations in the plants. Our results also suggested that the formation of complexes were occurring between Cl⁻ and Cd²⁺.
Chemical extractants are typically used to estimate soil bioavailability, but such estimates do not take into account soil properties. Simple mathematical models that use easily measured soil parameters may provide more practical estimates of metal bioavailability to plants. McBride et al. (1997) found that metal activity and solubility of metals in soil depend on a few key soil properties, and derived an equation for predicting soil metal solubility that uses easily measured soil parameters (pH, total metal concentration and organic matter). In an attempt to achieve better predictions of plant metal concentrations, we modified models based on McBride's equation.
Instead of estimating metal solubility, we modified the equation to predict metal concentrations in plants directly. Data from Chapter 4 was subjected to step-wise multiple regression using the modified equations for predicting lucerne or silver beet on the basis of total metal or EDT A extractable metals or CaCl₂ extractable metals, soil pH and soil OM content.
Results indicated that total metal concentration in soil and pH are important factors for controlling the bioavailability of metals to plants. Soil OM appeared to be less important. Assessment of bioavailability in plants based on total metal concentrations proved to be more accurate than assessment with either EDTA or CaCl₂ extracts. The models derived in this study were good predictors for some metals, but not for other metals. In most cases, predicted metal concentrations in plants did not show an improvement above the predictions obtained in Chapter 4 using non-logarithmic transformed data.
Further research needs to be carried out to investigate the relationship between plant concentration of metals and the bioavailability of metals. Results from our study are not sufficiently clear to allow us to draw firm conclusions on this relationship. In addition, it was not possible to use our data for recommendation of thresholds for soil metal concentration for selected crops intended for human or animal consumption, respectively. This was generally due to not being able to establish clear relationships between plant metal concentrations and metal bioavailability either established from weak extractants or predicted from the models. We were able to confirm that there are differences in the bioavailability of heavy metals in biosolids-amended soils among different plant species and that the concentration or bioavailability of some metals is affected by soil salinity. Models derived from equations predicting metal concentrations in plants have limitations and need further investigation to be of any use in setting thresholds for soil metal concentrations.||en