Quantitative assessment of soil parameter (K D and T C ) estimation using DGT measurements and the 2D DIFS model N.J. Lehto, Ł. Sochaczewski, W. Davison, W. Tych, H. Zhang * Environmental Science Department, Lancaster University, Bailrigg, Lancaster LA1 4YQ, United Kingdom Received 22 May 2007; received in revised form 22 September 2007; accepted 2 October 2007 Available online 26 November 2007 Abstract Diffusive gradients in thin films (DGT) is a dynamic, in situ measuring technique that can be used to supply diverse information on concentrations and behaviour of solutes. When deployed in soils and sediments, quantitative interpretation of DGT measurements requires the use of a numerical model. An improved version of the DGT induced fluxes in soils and sediments model (DIFS), working in two dimensions (2D DIFS), was used to investigate the accuracy with which DGT measurements can be used to estimate the distri- bution coefficient for labile metal (K D ) and the response time of the soil to depletion (T C ). The 2D DIFS model was used to obtain values of K D and T C for Cd, Zn and Ni in three different soils, which were compared to values determined previously using 1D DIFS for these cases. While the 1D model was shown to provide reasonable estimates of K D , the 2D model refined the estimates of the kinetic param- eters. Desorption rate constants were shown to be similar for all three metals and lower than previously thought. Calculation of an error function as K D and T C are systematically varied showed the spread of K D and T C values that fit the experimental data equally well. These automatically generated error maps reflected the quality of the data and provided an appraisal of the accuracy of parameter estimation. They showed that in some cases parameter accuracy could be improved by fitting the model to a sub-set of data. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: DGT; DIFS; Distribution coefficient; Kinetics; Soil; Metals 1. Introduction Understanding the exchange of trace metals between the solid and the solution phase is critical when trying to esti- mate their bioavailability and mobility in soil (McLaughlin et al., 1998; Harper et al., 2000). Until recently most studies have focused on using either a linear distribution coefficient (K D ), or a non-linear adsorption isotherm (Freundlich or Langmuir) to describe the distribution of the metal between the two pools. The importance of the kinetics that control these equilibria has also been recognized (Ernstber- ger et al., 2005), as they can control metal supply to biota (Pinheiro et al., 2004; DeGryse et al., 2006; Lehto et al., 2006a,b). However, measurement of the sizes of the solu- tion and solid phase pools and the kinetics that control the rate of equilibration between them has proved prob- lematic. There is demand for a reliable method that can be used with minimal disturbance to the soil. Measurements using the technique of diffusive gradients in thin-films (DGT) have been used with DIFS (DGT Induced Fluxes in Soils), a dynamic numerical model describing the interaction of DGT with soil, to estimate the proportion of trace metal in the solid phase (K D ) and the response times with which the equilibration between the solid and solution phase concentrations takes place (T C )(Ernstberger et al., 2002, 2005). So far the inter- pretation of the DGT measurements in terms of K D and T C has relied on a version of DIFS that assumes that DGT devices can be represented by an infinite planar surface (a one dimension model). The recent release of an improved, freely available DIFS model, working in two dimensions (2D DIFS) (Sochaczewski et al., 2007), allows a more accurate determination of K D and T C based on DGT 0045-6535/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2007.10.003 * Corresponding author. Tel.: +44 1524 593899; fax: +44 1524 593985. E-mail address: h.zhang@lancaster.ac.uk (H. Zhang). www.elsevier.com/locate/chemosphere Available online at www.sciencedirect.com Chemosphere 71 (2008) 795–801