Predictive modelling of metals in Swedish reference lakes

Abstract: Metals are a natural part in our environment. Due to increased use of metals in our society the biogeochemical cycles of metals are being altered on local, regional and global scales. Metals can be emitted through direct use of some elements (Fe, Zn and Cu) or indirectly as a result of processes (e.g., burning of fossil fuels re-sulting in emission of V and Hg). Both the chemical and physical properties of the element affect the amount of metals distributed in the environment, but the main factor is the usefulness of an element for humans. (Mason, 2013) To be able to evaluate an anthropogenic impact on a lake of pollutants like metals, reference conditions are needed. Background concentration is one type of refer-ence condition, which can be separated into regional and a natural background concentration. E.g. in Sweden, 1860 is used as a reference year for acidification because the effect of acid deposition should have been at a minimum during this time period (SEPA, 2010). The aim of this study was to create a model that could predict the natural back-ground concentration of a metal in lakes based on easily measured variables such as TOC and pH. This was possible for Pb, Zn and Ni (if including sedimentation rate).The predicted background concentration in this study were higher for all three metals than the one currently used in the Swedish Environmental Protection Agency’s standards. However, beyond this it was also possible to model the cur-rent concentration of Al, Cu, Cr, Co, Fe, Mn and V in the studied lakes. Due to the fact that all lakes are to some extent affect by humans (through atmos-pheric deposition), using modelling approach can provide a better way to estimate the background concentrations in lakes and through that also a better estimation of the human impact.