In the past, contaminated soil as a source for water contamination has been largely
neglected from the South African legislation. Inconsistent evaluation and remediation
of contaminated sites have resulted in many sources of water contamination not
being sufficiently addressed. The Draft National Norms and Standards for the
Remediation of Contaminated Land and Soil Quality (GN 233 of 2012) (henceforth
Norms and Standards) was published for comments and suggestions in August
2012. A number of uncertainties have been identified that may impact on the
successful implementation of the Framework and the Norms and Standards.
Some of the issues are related specifically to the setting of soil screening values
(SSV) for protection of water resources. There is no particular method specified to
determine the soluble fraction of contaminants in soil. In a phase 1 assessment
SSVs are used to judge whether constituents present in the soils are at
concentrations high enough to pose a potential risk to the receiving environment.
With the determination of the SSV a known water quality standard is converted to a
total concentration by making use of a dilution factor and partitioning coefficient (Kd).
The proposed Kd values in the Framework are surrounded by uncertainties and
information regarding Kd values for South African soils are limited. In addition, the
Framework does not take into account the natural background concentration of soils
to differentiate between anthropogenic and natural contamination.
Setting appropriate extraction method plays the key role for an objective and
standardised initial assessment of soluble concentrations in the soil. Similarly, the
selection of appropriate Kd values based on soil properties minimizes the
uncertainties during the estimation of SSV. Appropriate screening of contaminated
land is imperative to the registration of contaminated land and has significant
implications for industry, government and the environment. The aims of this study were: a) to assess fast and simple analytical methodologies
which can be performed by a commercial laboratory to mimic the standard saturated
paste extraction method to determine soluble metal concentration in soils, b) to
determine Kd values for selected South African soils in 10 soil horizons, and c) to determine baseline concentration ranges for selected soluble contaminants to assist
in the setting of appropriate soil screening values for the protection of water
resources. The study will test the hypothesis that a 1:2.5 (soil:water ratio) water
extraction could better mimic the standard saturated paste extraction than the 1:20
ratio, commonly used in USA. The 1:2.5 ratio can be done by most commercial
laboratories and gives a better indication of the pore water quality.
To assess analytical methods, four extraction methods (1:2.5, 1:5, 1:10 and 1:20
soil:water) were tested against the standard saturated paste extraction using three
trace metals (Cu, Pb and V). These three metals were selected based on reactivity
and environmental abundance. Lead and Cu are B type metals which complex
readily with organic material and are generally more environmentally toxic.
Anthropogenic activities are increasing enrichment of B type metals in the
environment. Vanadate is an A type metal and is redox sensitive. Its association with
the Bushveld complex was the reason for its inclusion in this study. The baseline
soluble concentrations of Cu, Pb, and V were determined for selected 100 South
African soils using the extraction method selected above. The Kd values of Cu, Pb,
and V were determined for selected 10 soil horizons (1:1 clay dominated A horizon,
Vertic soil dominated by 2:1 clay minerals, Yellow oxidic / Plinthic (Soft plinthic B
horizon), low clay red oxidic B horizon, red oxidic B horizon, plinthic B horizon,
gleyed horizon (G horizon), melanic A horizon, orthic A horizon with high organic
content (OC) and E-horizon) using a batch method. For each constituent three
different metal concentrations were added to each of the soil horizons. After
extraction the concentration in the solution was used to determine the amount of
metal sorbed to the soil.
Considering the standard saturated paste extract as a bench mark, the 1:2.5
soil:water ratio extraction gave more representative soil pore water quality for Cu, Pb
and V in the low to medium concentrations. Therefore, the hypothesis was accepted.
It was also apparent that the Kd values selected for the Framework are not
representative of typical South African soils. The Kd values for Cu range between
12.7 and 19044 L kg-1. These values exceeded the value of 10 L kg-1 provided by the
Framework in all soils. For V the Kd values (10.5 – 865 L kg-1) in all soils were lower than the value of 1000 L kg-1 specified in the Framework. For Pb, the Kd of 100 L kg-1
as indicated in the Framework is not representative of soils found in South Africa.
For Pb, low clay content, weathered soils have lower Kd values, whereas higher clay
content soils have Kd values up to 4 orders of magnitude higher than the Kd in the
Framework. Therefore, due to the large variability in Kd values, a single value cannot
be used for all soil types.
To conclude, the 1:2.5 soil:water ratio was found to be more representative to the
soil pore content especially in the low to medium solute concentrations. The study
also showed that a single Kd value cannot be used across soil types. The Kd values
currently used in the Framework have low representativity of the South African soils.
In the meantime, the Kd values for Cu, Pb, and V generated from this study for
selected South African soil horizons could be used by the framework. There is,
however, a need to develop local Kd values of contaminants across soil types.