- Title
- Predicting phytotoxicity of metal(loid)s and their mixtures in soil using pore-water based transfer functions
- Creator
- Kader, Mohammed Abdul
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2017
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- The terrestrial environment is exposed to mixtures of contaminants simultaneously through various sources and exposure routes. Soil contamination by a single xenobiotic is an exception rather than the rule. However, environmental toxicity studies and environmental regulations primarily deal with individual toxicants without an informed method for dealing with multiple toxicants. Scientifically, the importance of contaminant mixtures in environmental media is recognised, but to date the field of environmental chemistry and toxicology has been unable to adequately resolve this issue. As a result, no guidelines have been established to adequately deal with mixtures in the terrestrial environment. The problem of contaminant mixtures in soil is exacerbated by incomplete knowledge of single metal(loid) contaminant bioavailability to a range of receptor groups, such as plants, soil invertebrates, microbial communities and mammals. The aim of this study was to further enhance an understanding of phytotoxicity to individual metal(loid)s (arsenic (As), cadmium (Cd), copper (Cu), lead (Pb) and zinc (Zn)) and their mixtures in different soils using both total and bioavailable fractions. In individual studies, toxicity was related to soil pore-water parameters (total dissolved and free ion activities). Datasets generated were used to derive transfer functions for metal(loid)s solubility, uptake and plant growth (toxicity) across the full dataset (10 soils). Soils were sampled from across the Eastern Australian mainland, including New South Wales, Queensland, South Australia and Victoria. The methodology for each experiment was largely consistent throughout this document. Cucumis sativa L. (cucumber) was used as a model plant in four week greenhouse pot studies. Cucumber was chosen based on a preliminary screening study which indicated cucumber had good sensitivity and growth rates compared to other agricultural plant species tested. Shoot growth was chosen because we established root elongation was significantly less sensitive (producing higher toxicity values) than the four week pot studies. A Rhizon sampler as well as a double chamber centrifugation method were used to extract soil pore-water from water saturated soils at the end of the experiments. Geochemical modelling and aqueous phase speciation was conducted using both WHAM 7 and PHREEQC where appropriate. Arsenic phytotoxicity expressed as total concentrations in soil was strongly related to the Freundlich partitioning constant (Kf) for both root elongation and shoot dry weight data. Since Kf values were related to soil pH and EC₅₀ (effective concentration at 50% growth inhibition) data, As phytotoxicity was controlled by sorption reactions. Effective median concentration (EC₅₀) based on total As in soil was 1.6 times higher for root elongation studies than four week pot studies. Thus, root elongation in cucumber was a less sensitive end point parameter for As to cucumber. In the whole dataset (n=178, ten soils) the EC₅₀ for As based on total concentration in soil was 72 mg kg-1 (range of 55- 95 mg kg-1). However, for individual soils it was as low as 17 mg kg-1, which was less than both threshold values in Australia and USA regulations. As a result, root elongation data for ecotoxicological assessment should be treated with precautions and the threshold values for As in sandy alkaline soils should be reviewed. In soil pore-water studies of As, soluble carbonate and phosphate had a strong relationship to As phytotoxicity and data indicated that dissolved phosphate and carbonate resulted an antagonistic effects on arsenate toxicity. Based on the relationship of these parameters, a terrestrial biotic ligand-like model (BLM) was applied to As phytotoxicity using measured arsenate concertation including carbonate and phosphate concentrations in soil pore-water. Although phosphate effects on As toxicity have been established in the literature before, it was the first time an established relationship with carbonate has been shown. As a result, data revealed that an ion interaction model similar to the BLM for arsenate is possible which improves the current risk assessments at both individual As and As co-contaminated soils. Cadmium accumulation in plants is a great concern for human health and even higher organisms. Cd partitioning constants (Kf) were strongly correlated to both Cd partitioning and plant uptake. The derived transfer functions for Kf value parameterised with pH, CEC and OC resulted a good relation (R² ≥0.91, n=41 soil). The transfer functions for dissolved and free Cd concentrations were also successfully developed using Kf values and utilised for Cd transfer in plant shoots in independently contaminated soils (15). The series of equations predicting Cd accumulation from pore-water Cd (dissolved and free) were able to predict the measured uptake data. The good relationship indicated that Cd uptake to cucumber shoots could be predicted with dissolved Cd and Cd2+ without other pore-water parameters such as pH or cations. The free ion activity model has been established for better prediction of toxicity than total concentrations, however, it’s application in terrestrial systems is limited. Pore-water chemistry and measured Cu2+ was successfully described Cu phytotoxicity in soil with different Cu loading. Dissolved Cu and Cu2+ in soil pore-water fitted well with growth response for all 10 soils with R² values of 0.73 and 0.66, respectively. Separation of soils as acid and alkaline and fitting separately showed that there was an strongly significant fit for both log Cu2+ and log Cupw in acidic soils (R² =0.92 and 0.86, respectively) but weakly significant fit for alkaline soils. Data suggests alkaline soils should be treated on a case - by - case in the absence of a suitable model accounting for pH differences in soil pore-water. The pCu (measured Cu2+ in soil pore-water) was predicted by developing transfer function using pore-water pH and total concentration in soil with higher R² value (0.91, n =82). Despite only 12 weeks ‘ageing’ there was quantitative agreement between pCu model from our study and predicted pCu from Sauvé et al (1998). [More detail in thesis abstract]. Similarly, when a solution-only model was used to predict Zn in shoot, DOC was negatively related to Zn in shoot, indicating a reduction in uptake/ translocation of Zn from solution with increasing DOC. Phytotoxicity expressed in terms of Pb2+ was observed to occur in the nanomolar range in neutral to alkaline soils (EC₅₀ values 90 to 853 nM) and micromolar levels for acidic soils (EC₅₀ values 7.35 to 9.66 μM). Internal Pb concentrations relating to toxicity (PT50) in roots and shoots also decreased with increasing pore water pH (R²=0.52 to 0.53). From a series of dose response studies we developed transfer functions predicting Pb uptake in cucumber and we validated these functions with long term Pb contaminated soils. The significant independent parameters were pore-water Pb2+ and dissolved Pb plus dissolved organic carbon (DOC). The observed RMSE for the Pb-DOC model and Pb2+ were 2.6 and 8.8, respectively. The Pb-DOC model tended to under-predict Pb whilst Pb2+ was over-predicted despite reasonable RMSE values. In binary mixtures of As and Cu in soil, As pore-water concentrations were significantly reduced in the presence of Cu, with higher doses of Cu exhibiting a greater effect. Pore-water pH played a significant role in the interaction, the effect of Cu on As was higher in alkaline soils (pH >7). This could be due to the adsorption/surface precipitation or tertiary bridging complexation though no precipitation was established in soil soluble phases using equilibrium concentration estimated by PHREEQC. Dissolved pore water Cu was significantly reduced in the presence of As (p>0.05) only in alkaline soils. Similarly, pore-water pH reduced As uptake and was significantly reduced in the presence of Cu and a significant modulation effect was observed in our whole dataset. Since the antagonistic impact on As solubility and uptake was caused by Cu in As-Cu mixtures, a predominantly additive response was established using response additive model. In As-Zn mixtures, the similar effect of Zn was also observed on As solubility. In case the of Zn and Zn2+ activity in pore-water, a significant effect of As was observed only in alkaline soils. As a result, a significant interaction effect of As and Zn in shoot uptake was observed in the whole dataset and mostly additive to antagonistic (lower than additive) effect was observed. Growth response varied with both concentration range and soil type. In the case of metal(loid) mixtures at low toxic concentrations, chemodynamics of Cd and Cu (total dissolved and free) were significantly changed in the presence of Zn and the most prominent effect was observed in alkaline soils and Ferrosol. A positive effect was also established for Zn2+ with Cd in the whole dataset. Although a negative correlation of cations was mostly observed with As, the significant effect was established on Cd2+ only in a few soils. Arsenic pore-water concentration varied in binary and tertiary mixtures but was significantly reduced (p>0.05) in quaternary mixtures at both EC₁₀ and EC₂₀ levels in each soil. Based on predicting combined toxicity the concentration additive (CA) model mostly underestimated plant response compared to the response additive (RA) model. Except in the alkaline sandy PB soil, an antagonistic response was established in binary As-Cu, As-Zn and Cd-Zn mixtures in each soil whereas, additive to antagonistic response was observed in all tertiary and quaternary mixtures fitting with the RA model.
- Subject
- phytotoxicity; pore-water; mixture; bioavailability; logistic model; concentration additive; response additive
- Identifier
- http://hdl.handle.net/1959.13/1342491
- Identifier
- uon:28969
- Rights
- Copyright 2017 Mohammed Abdul Kader
- Language
- eng
- Hits: 1996
- Visitors: 2375
- Downloads: 444
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Thesis | 9 MB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Abstract | 409 KB | Adobe Acrobat PDF | View Details Download |