Effects of Random and Systematic Errors on Freundlich Parameters for Pesticide Sorption

Abstract

Assessment of the Freundlich sorption parameters of pesticides is an important element of their environmental risk assessment (especially leaching to groundwater), and therefore these parameters must be accurately assessed. In this work, the effect of random and systematic errors on the accuracy of the Freundlich sorption parameters was assessed for pesticide–soil sorption studies using Monte Carlo simulations and analytical approximations. This assessment was performed for the so-called indirect method, which assumes that the mass in the system at the end of the study is equal to the added mass. The errors in these parameters appeared to depend strongly on the percentage decrease in the concentration in the liquid phase. The effects of random errors on these parameters was shown to be limited if the recommendations of the current Organisation for Economic Cooperation and Development (OECD) guideline were followed. However, the effect of the systematic error resulting from the loss of substance during the study might lead to observed Freundlich sorption coefficients that are approximately two times too high and Freundlich exponents that are 10% too high. Therefore, it is recommended that the possible effect of loss on the Freundlich sorption parameters should be assessed. This effect can be relatively simply calculated by combining the observed percentage decrease in the liquid phase with an upper limit of the percentage loss during the study. It is further recommended that the highest possible solid/liquid ratio should be used in pesticide–soil sorption studies. These recommendations are especially relevant for pesticides with low sorption for which the leaching assessment is most critical.