Leaching is the downward movement of fertilizer or waste in soil with the drainage water and is a critical process in global nutrient cycling. To be available for plant uptake, fertilizer/waste (solute) must dissolve in the soil water. The combination of soil water and dissolved solute is the soil solution. The soil solution and soil air are present in the pores between soil particles. If water fills all the pores and no air is present, the soil is saturated and leaching is rapid. If both water and air are present, the soil is unsaturated and leaching is slower.

Sorption is expected to be an important process in nutrient removal from stormwater by bioretention systems. The sorption capacity of loamy-sand media from experimental bioretention mesocosms with and without vegetation and prior nutrient enrichment was tested, using two incubation times (24 hours or 72 hours) and nutrient solutions comprising ammonium, organic nitrogen, phosphate, organic phosphorus, and organic carbon, at concentrations representative of stormwater and higher. At stormwater concentrations, enriched media sorbed none or very little phosphate. When the equilibrium concentrations were calculated for all compounds tested, they revealed that at the concentrations expected in stormwater, nutrients could be expected to desorb from the media to solution rather than be removed from solution. No nitrate was extractable from the vegetated media, whereas considerable quantities were extracted from enriched, non-vegetated media. The presence of vegetation appears to reduce the risk of nitrate leaching from filter media. Sorption is not likely to be a long-term sink for nutrients in stormwater bioretention systems. Its role may be to extend the residence time of nutrients in the media so that plants and microbes have the opportunity to mineralise and assimilate these compounds.

The ability of HYDRUS-2D (HYDRUS) to simulate water and chemical transport in large, undisturbed cores of a Vertosol and a Podosol soil was investigated. Parameters required by HYDRUS for simulating water and chemical transport, and nitrogen transformation, were obtained from previously published laboratory studies. HYDRUS simulated the measured cumulative drainage and cumulative chloride (Cl–) leaching behaviour very closely for both soil types, and also provided a very good description of coupled nitrogen transformation (conversion of ammonium to nitrate) and leaching (coefficient of model efficiency ∼1). There was little correlation between measured and predicted potassium (K+) leaching from the Podosol, suggesting that the mathematical equations governing the transport of reactive chemicals did not adequately reflect K+ behaviour in this coarse-textured soil. The reason for this discrepancy is unclear but may have been related to the use of sorption parameters obtained from batch rather than miscible displacement techniques, or mechanisms controlling K+ sorption were not well represented by the general non-linear sorption equation used by HYDRUS. The ability of HYDRUS to accurately simulate water and non-reactive chemical transport agrees with previous studies; however, more investigation into its suitability for predicting the movement reactive chemicals in soil is warranted.

The geochemical partitioning of trace metals in sediments is of great importance in risk assessment and remedial investigation. Selected factors that may control the partitioning behavior of Cu, Pb and Zn in non-sulfidic, estuarine sediments were examined with the use of combined sorption curve-sequential extraction analysis. This approach, which has not been previously used to examine estuarine sediments, allowed determination of sorption parameters for Cu, Pb and Zn partitioning to individual geochemical fractions. Partitioning behavior in sulfidic sediments was also determined by sequentially extracting Cu, Pb, and Zn from synthetic sulfide minerals and from natural sediment and pure quartz sand after spiking with acid-volatile sulfide (AVS). Trace metal sorption to the "carbonate "fraction (pH5, NaOAc extraction) increased with metal loading due to saturation of sorption sites associated with the "Fe-oxide" (NH(2) OH center dot HCl extraction) and "organic" (H(2)O(2) extraction) fractions in non-sulfidic sediments. Freundlich parameters describing sorption to the "Fe-oxide" and "organic" fractions were controlled by the sediment Fe-oxide and organic carbon content, respectively. Sequential extraction of Cu from pure CuS, AVS-spiked sediment and AVS-spiked quartz sand showed that AVS-bound Cu was quantitatively recovered in association with the "organic" fraction. However some AVS-bound Pb and Zn were recovered by the NH(2)OH center dot HCl step (which has been previously interpreted as "Fe-oxide" bound metals) in the sequential extraction procedure used in this study. This indicates that the sequential extraction of Job and Zn in sulfidic sediments may lead to AVS-bound metals being mistaken as Fe-oxide bound species. Caution should therefore be exercised when interpreting sequential extraction results for Pb and Zn in anoxic sediments.

Land application represents a viable option in the management of wastewater such as landfill leachate. The surface charge and ion adsorption characteristics of the receiving soil will play an important role in the success of this disposal strategy. The surface charge characteristics of 6 soils with contrasting chemical and physical properties were studied using charge balance principles. The point of zero net proton charge density, point of zero salt effect and point of zero net charge, and the concentrations of permanent and variable charge as a function of pH, were determined for each soil. The presence and magnitude of permanent negative surface charge was largely dependant on clay mineralogy. All soils displayed increasing negative surface charge density with increasing pH. The magnitude of increase for any particular soil was dependent on the type and amount of colloids exhibiting variable surface charge. The increased negative surface charge was attributed to the ionisation of functional groups on organic matter, hydrous Fe and Al oxides, and edge sites of kaolinite and smectite. The variation in surface charge with pH conformed to charge balance principles for soils containing a mixture of both permanent and variable surface charge, but not for soils containing predominantly permanent surface charge. Proton-consuming processes (e.g. adsorption on permanent charge sites, neutralisation of inherent alkalinity) influenced the overall charge balance, particularly for soils with predominantly permanent surface charge. Ammonium adsorption from landfill leachate was measured as a function of pH for each soil type. Increased negative surface charge resulted in increased ammonium adsorption, and for most soils the preference of the newly created adsorption sites for ammonium remained relatively constant. Since many soils contain a mixture of colloids with permanent and variable surface charge, these charge characteristics and their effect on individual soil properties, contaminant availability, and contaminant mobility must be taken into account when evaluating the suitability of land for wastewater re-use.

The sorption–desorption and leaching behaviour of Cu in a Podosol from south-east Queensland, Australia, was examined. Copper sorption was described by a linear distribution coefficient at low sorption levels (KDCa→0) of 481 L/kg and a sorption capacity (CS,Max) of 382 mg/kg. Selective removal of soil organic matter reduced these values by approximately 95%, indicating that Cu was sorbed predominantly to soil organic matter. The KDCa→0 and CS,Max values from Cu desorption experiments were 934 L/kg and 516 mg/kg, respectively, which indicates that sorption was not fully reversible. This irreversibility was related to aqueous Cu speciation (modelled with MINTEQA2), showing that aqueous complexes between Cu and dissolved organic carbon (DOC) comprised 28.3–72.8% and 21.3–45.4% of aqueous Cu in the sorption and desorption experiment, respectively. Sorption irreversibility was not evident when the corresponding data was presented as free Cu2+ isotherms. Both sorption and desorption experiments with free Cu2+ <0.2 mg/L were described by a KDCa→0 value of approximately 3000 L/kg. Sequential extraction of sorbed Cu indicated that at low concentrations, sorption occurred primarily via specific interactions, with non-specific sorption becoming increasing important at higher concentrations. Desorption of Cu in a column leaching experiment was attributable to exchange of sorbed Cu2+ with Na+. Leaching with a DOC solution of pH 7 and 135 mg/L greatly enhanced Cu mobility due to the formation of aqueous Cu–DOC complexes.

The distribution and enrichment of selected trace metals (Cd, Cr, Cu, Ni, Pb, Sn, Zn) in benthic sediments of the Southport Broadwater, a semi-enclosed coastal body of water adjacent to the Gold Coast city, south-eastern Queensland, Australia, was studied with the objective of assessing the extent and degree of sediment contamination. Sediment samples from the 0-10 cm and 10-20 cm depth intervals of 32 sites within the Southport Broadwater and surrounding residential canals were analysed for particle size distribution, pH, organic C and 'near-total' major (Al, Ca, Fe, Mn) and trace (Cd, Cr, Cu, Ni, Pb, Sn, Zn) metal contents. Sediment contamination for each trace metal was assessed by (1) comparison with Australian sediment quality guidelines, (2) calculation of the index of geoaccumulation based on regional background values, and (3) geochemical normalisation against Al (i.e. the abundance of alumino-silicate clay minerals). Based on this approach, the results indicate that submerged sediments in the study area are not presently enriched with Cd, Cr or Ni, with the spatial distribution of these metals being very well explained by the abundance of alumino-silicate clay minerals. However, several sites were strongly enriched with Cu, Pb, Sn and Zn, arising from sources related to either urban runoff or vessel maintenance activities. The study indicates that several varying approaches are needed for a satisfactory assessment of contaminant enrichment in estuarine sediments.

The geochemical partitioning of copper (Cu), lead (Pb), and zinc (Zn) was examined in benthic sediment profiles (0- to 20-cm depth interval) composed of relatively coarse (65-90% sand-sized particles), noncohesive, suboxic material (Eh +120 to +260 mV). Total Cu, Pb, and Zn concentrations ranged from 8.3 to 194, 16.3 to 74.8, and 30.1 to 220 mg/kg, respectively, and were related to vertical trends in sediment texture. The observed distribution coefficients describing solid-solution partitioning were in the range of 100 to 1000 L/kg. The geochemical partitioning of solid-phase Cu, Pb, and Zn between six operationally defined fractions was examined with a sequential extraction scheme. The association of Cu, Pb, and Zn with amorphous oxides, crystalline oxides, and organic matter was linearly dependent on the abundance of each respective phase. For retention by amorphous oxide minerals, the observed stoichiometry ranged from 5.2 to 23.7 mg/g for Cu, 12.8 to 21.5 mg/g for Pb, and 23.1 to 85.7 mg/g for Zn. Corresponding values for association with crystalline oxides were an order of magnitude less than those for amorphous oxides, indicating a lesser affinity of trace metals for crystalline oxides. The stoichiometric relationships describing association with organic matter ranged from 17.6 to 54.0 mg/g for Cu, 6.1 to 9.6 mg/g for Pb, and 6.4 to 16.4 mg/g for Zn. The results from this study provide an insight into processes controlling trace metal partitioning in coarse-textured, suboxic, estuarine sediments.

The effect of aging on the solid/pore-water partitioning and desorption behaviour of tributyltin (TBT) in sediments was examined. Three sediment samples with contrasting physical and chemical properties were spiked with 10 mg/kg TBT and aged under sterile conditions for periods of time ranging from 1 to 84 days. Aging had a negligible effect on partitioning and desorption behaviour in a sandy sample with very low organic carbon content (0.2% w/w). In contrast, for samples with larger amounts of organic carbon (2.6% and 4.8% w/w), aging caused substantial increases in TBT sorption. For these samples, apparent distribution coefficients (KD,app) obtained from sequential 2 h desorption experiments also exhibited a twofold increase between spiked sediments subjected to aging for 1 day and 84 days. This study demonstrates that aging effects may be an important aspect of TBT fate in contaminated sediments.

Anthropogenic contaminants introduced into the aquatic environment can eventually accumulate in benthic sediments. Sediment quality assessment is therefore an important consideration in ensuring the ecological sustainability of coastal regions. An important research focus associated with sediment quality assessment is determining whether sediments enriched with trace metals actually exert an adverse effect on aquatic ecosystems (Ankley et al., 1994). This issue is complicated because trace metal bioavailability in sediments is controlled by sorption to several solid-phases (such as iron oxyhydroxides, organic matter and reactive sulfide species; Chapman et al., 1998). Furthermore, total metal analysis often provides a poor indication of bioavailability and potential mobility (ANZECC/ARMCANZ, 2000). As such, the use of partial extractions potentially provides a more accurate representation of the concentration of reactive, bioavailable metal species (Burton et al., 2005).

Nutrient release from a coarse-sand (CS) and a fine-sand (FS) estuarine sediment into the overlying water column was studied under static (diffusion-controlled) and dynamic (resuspension) conditions. Resuspension increased the amounts of and P released from the sediment cores compared to those released by diffusion alone due to direct release of these nutrients from the readily available pore-water and ion exchange fractions. The average effective and P diffusion coefficients (D e) were found to be similar for both sediments, and did not vary significantly between static and dynamic conditions. Values of D e were typically of the order of 10−5 to 10−6 cm2 s−1. The average daily and P diffusive flux density from CS and FS into the overlying water column of the study area was estimated to be 24.6, 0.02 and 0.39 mg m−2 d−1, and 66.9, 0.017 and 0.002 mg m−2 d−1, respectively. These fluxes suggest that nutrient release from the sediments via diffusion and small-scale resuspension events may represent only a minor source of and P to the water column of the study area.

The effects of pH and Cu loading on the solid/solution partitioning of Cu in a Podosol from south-east Queensland, Australia was examined. Sorption-desorption of Cu exhibited maximum linear distribution coefficients (KD) at approximately pH 5. Observed decrease in KD values at pH >5 was attributed to increased solubility of native dissolved organic carbon (DOC) at higher pH and subsequent formation of non-sorbing Cu-DOC complexes. Speciation modelling with the MINTEQA2 code indicated that >90% of aqueous Cu was present as Cu-DOC complexes at pH >5.5. The effect of Cu loading was examined with sorption isotherm analysis at pH 5 using solid : solution ratio approaches that were both constant (1 : 2 and 1 : 10) and variable. As the solid : solution ratio increased, the proportion of Cu sorbed decreased due to the formation of Cu-DOC complexes. However, this effect was negligible once these Cu-DOC complexes were accounted for via free Cu 2+ sorption isotherms. This indicated that Cu 2+ sorption at concentrations <0.08 mg/L was described by a KD value of approximately 3000 L/kg. Despite this relatively high KD value for Cu 2+ sorption, the results indicate that Cu-DOC complexes significantly enhance Cu solubility in soils high in DOC. Additional keywords: trace metals, partitioning, solubility, MINTEQA2.

The efficiency of the solute transport models PEAK and CHEMFLO to predict water and chloride (Cl) movement in large undisturbed soil cores from two wastewater reuse areas (core1A and core1B: Vertisol; core2A and core2B: Natrustalf) was investigated. The models predicted cumulative drainage and cumulative leachate Cl from the well-structured Vertisol very closely (CME and r(2)congruent to0.9), but they failed to adequately predict Cl losses from the Natrustalf. Both models estimated nearly twice as much Cl leached from Core2A and Core2B compared with the experimental data. This discrepancy between measured and predicted losses was attributed to Cl diffusion into immobile soil-water regions from which Cl was bypassed during leaching events. Because CHEMFLO or PEAK do not account for immobile soil-water zones and assume a uniform Cl distribution throughout the soil, both models would overestimate Cl leaching losses. Despite this finding, both models provided insight into key mechanisms controlling wastewater transport in two contrasting soil types and encouraged more detailed interpretation of laboratory data.

The Southport Broadwater (Fig. 1) is a semi-enclosed, coastal body of water situated adjacent to the Gold Coast city in south-east Queensland, Australia (153.40ୱ53.42Ŭ 27.80୲7.98ө. The catchment for the Southport Broadwater is intensely urbanised, and as such, the Broadwater is likely to receive significant trace metal and nutrient loads from urban sources. The southern end of the Southport Broadwater also supports a relatively large marina complex, and past marina activities may have contributed to contaminant loads to the aquatic system. This paper provides initial baseline data on trace metals (Cd, Cr, Cu, Ni, Pb, Sn and Zn) and nutrients (N and P) in bottom sediments of the Southport Broadwater.

Tributyltin (TBT) sorption to four natural sediment samples in artificial seawater was examined under a range of modified pH and salinity conditions. Three of the sediment samples were relatively pristine with regard to TBT contamination, but the fourth was a TBT-contaminated sediment from a commercial marina. Sorption of TBT was described well by linear sorption isotherms,with distribution coefficients ranging from 6.1 to 5210 L/kg depending on the pH and salinity. The sediment organic C content and particle size distribution were important determinants of sorption behavior. The presence of resident TBT in the contaminated marina sediment caused a substantial reduction in further TBT sorption. Desorption of TBT from the marina sediment was described by relatively large observed distribution coefficients ranging from 5100 to 9400 L/kg, suggesting that aging effects may reduce sorption reversibility. Increased artificial seawater salinity generally reduced TBT sorption at pH 4 and 6, but enhanced TBT sorption at pH 8. Regardless of salinity, maximum sorption of TBT was observed at pH 6, which is attributed to an optimal balance between the abundance of cationic TBT+ species and deprotonated surface ligands. Consideration of aqueous TBT speciation along with octanol-water partitioning behavior suggests that hydrophobic partitioning of TBTCl0 to nonpolar organic matter was important for pH < 6, while partitioning of TBTOH0 was important at higher pH.

In south-east Queensland, Australia, extensive areas of sandy soils (Podosols) with shallow (<1 m) watertables are used for exotic pine tree production. Leaching of surface-applied fertilizers (di-ammonium phosphate and potassium chloride) in these soils has been suggested as a contributing factor in a decline in local groundwater quality. This study investigated the movement of potassium chloride (KCl) in horizontal columns of sand under unsaturated flow conditions. A pulse of KCl (equivalent on a surface area basis of 300 kg K/ha) was applied to one end of the sand column, and immediately followed by deionized water at a constant flux of 0.398 cm/h for 12h. The position of the maximum chloride (Cl) concentration lagged behind the "piston-front" of water, and potassium (K) moved at about half the rate of Cl. Despite having a very low effective cation exchange capacity (ECEC <2 cmol/kg), the sand retained about 71% of the applied K in the 0-10 cm interval through predominantly K-magnesium (Mg) and K-sodium (Na) exchange. The nonspecific mechanisms responsible for K retention suggests that this chemical could be readily displaced from the ECEC sites through subsequent cation exchange reactions, and eventually move into the groundwater. Resident ions were pushed ahead of the advancing wetting front, and this contribution should be taken into account when evaluating factors responsible for groundwater contamination. The models CHEMFLO and SWIMv2 were used to predict KCl leaching, and although both models predicted the distributions of water content and water-soluble Cl very closely, both over-estimated soluble K concentrations.