Sandy soils were sampled from second rotation sites of Ptnus pznaster -4it. on the Gnangara Mound in Western Australia. Adsorption isotherms were measured for atrazine [6-chloro-~~-eth~l-~~-iso~ro~yl-1.3,5-triazine-2,4-diamine], simazine [6-chloro-N,N1-diethyl-1,3,5-triazine-2,4-diamine], and glyphosate [N-phosphonomethyl-aminoaceticacid]. Adsorption isotherms were also measured for degradation products of atrazine: hydroxyatrazine [6-hydroxy-~2-ethyl-~4-isopropyl-1,3,5-triazine-2,4-diamine] (HA), desethylatrazine [6-chloro-N-isopropyl-l,3,5-triazine-2,4-diamine] (DEA); and of glyphosate: aminomethylphosphonic acid (AMPA).The adsorption of the 2 triazines was proportional to soil organic carbon content and was not affected significantly by other soil parameters. The affinity for soil organic carbon was in the order atrazine = simazine = DEA > HA. Affinity of atrazine for the type of organic matter in the Gnangara Mound soils (expressed as KOc) was significantly greater than is commonly reported for other soils.The adsorption of glyphosate and AMPA increased strongly with iron and aluminium content of soils and decreased with increasing soil organic carbon content. This would indicate that glyphosate and AMPA are mainly adsorbed by clay minerals, while soil organic matter competes for adsorption sites and inhibits adsorption.Contrary to what is usually reported for batch adsorption of pesticides in soils, significant increases in adsorption of the triazines and glyphosate were measured after 1 day of equilibration.

Distribution constants for Cd and Pb were measured for 33 temperate soils. The distribution constants were related to soil organic matter and pH at three different ionic strengths of soil extractants. The soil extractants consisted of solutions of CaCl2, NaCl, and KCl and were taken to be representative of the composition of the soil solution phases. For each ionic strength a significant log‐log correlation was found between the distribution constants, related to organic matter content of the soil and the hydrogen ion concentrations of the soil extracts. Exchangeable or mobile fractions of Cd and Pb in the soil were calculated from the product of distribution constant and concentration in a soil extract. With the aid of adsorption data of previous work with other soils, assuming a similar relationship as found for Cd and Pb, comparable results were obtained for Cu and Zn. Exchangeable fractions were found to lie in a range of about 1 to 5% of total metal in the soils for Pb and about 10 to 50% for Cd, Zn, and Cu.

Adsorption of Be, F, B, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Sr, Mo, Tc, Ag, Cd, Sn, Sb, Cs, Ba, Hg, Pb, Bi, and P was measured in a sandy soil and a sandy loam soil at concentration levels closely corresponding to those to be expected for field conditions. Mobilities of the elements in the soils were calculated from the adsorption data. Of the trace elements studied, F, B, and Tc were found to be very mobile in both the sandy and sandy loam soils. Manganese, Sr, and Sb were highly mobile only in the sandy soil, and Mo in the sandy loam soil. In general, sludge solutions appeared to increase the mobility of elements in a soil. This is due to a combination of complexation by dissolved organic compounds, high background concentrations, and high ionic strengths of the soil solutions. The relative effects of these factors vary strongly among elements.Equations were derived predicting the rates of accumulation in soils and accompanying increases in the soil solutions of trace elements added with sewage sludge.When adsorption was related to soil organic matter content, for many trace elements the strength of adsorption was found to depend only on pH, increasing with increasing pH.

The adsorption-desorption behaviour of four pesticides, in four Western Australian soils differing in their physico-chemical properties, has been studied using a batch technique. Sorption data for Fenamiphos, Linuron and Simazine could be fitted to a Freundlich-type equation, while that for Diquat more closely fitted the Langmuir equation. The value of the exponent in the fitted Freundlich isotherms varied from 0.70 to 1.00 for the various pesticide and soil combinations. Adsorption was found to be better correlated with organic matter than other soil properties, and decreased with increasing pH for all pesticides except Diquat. This study supports the observation of B. T. Bowman that the Freundlich equation in mole fraction form is more useful for comparing relative adsorption. Comparison of the present adsorption data with previous studies showed that the organic matter in these soils behaved differently in terms of pesticide adsorption. In particular, the octanol-water partition approach, for predicting the adsorption of non-ionic organic compounds, gave unsatisfactory predictions of Linuron and Fenamiphos adsorption in these soils. Desorption by a consecutive method showed hysteresis with all pesticide-soil combinations examined. The magnitude of the hysteresis was unaffected by changes in the soil : solution ratio. Using a dilution method, experimental procedures such as centrifugation and shaking have been shown to only slightly affect desorption hysteresis. The suggestion that hysteresis results from organic matter effectively 'locking in' the adsorbed pesticides is supported. The presence of methanol in the desorption solution results in swelling of organic polymers and facilitates the release of the adsorbed pesticides. Consequently, hysteresis was found to be essentially absent from the Linuron isotherm on removal of organic matter from the soil.

Natural levels of inorganic phosphate in soils of Rottnest Island are quite high: about 300 mg/kg as P (or about 4 t of P per ha per meter depth of soil). In comparison, the production of phosphorus in wastewater from sewage, treated on Rottnest Island, amounts to approximately 2 t per year. The phosphate, occurring in the soil naturally, is mainly in mineral form and not very soluble. Solution concentrations in the soils are less than 0.001 mg/L P-PO4 (at pH values of 8.5-8.9). Conditions in the calcareous soils of Rottnest Island are favourable for precipitation of phosphate as calcium phosphates. Theoretically >>99% of phosphate in wastewater from treated sewage can be stored indefinitely as hydroxy-, fluoro- and chloro-apatites. In practice, however, application of wastewater to these soils will result in a (kinetically defined) finite concentration of phosphate to move through the soil slowly as a sharp front. The effective width of the frontal zone was taken to be equal to the longitudinal hydrodynamic dispersivity. Mobilities, relative to water, of fronts resulting from step increases of phosphate in soils were then calculated with an experimentally obtained, time-dependent, adsorption equation and the average residence time of phosphate in the frontal zone. Calculated mobilities were verified experimentally by leaching phosphate through small columns of soil at different concentrations of phosphate and rates of infiltration. For concentrations in wastewater between 10 and 15 mg/L P-PO4, mobilities of phosphate, relative to water, in soils of Rottnest Island are less than 2% for expected infiltration rates of wastewater between 0.5 and 1 cm/day.