Analytical uncertainties

The hydrochemical data generated from the site investigation programme in Forsmark constitute the following four types: 1. Data from chemical analyses. 2. Data from long term on-line measurements in flow through cells (Chemmac measurements). 3. Data from field measurements in surface waters and near surface groundwaters. 4. Special studies or experiments This quality evaluation assesses the chemical analyses performed by SKB laboratories (Aespoe chemical laboratory and the mobile field laboratory at the Forsmark site) and by contracted external laboratories. Analytical methods and reported measurement uncertainties are presented in Chapter 2. The discussions in this document focus on some selected constituents of special importance for hydrochemical interpretations and modelling as well as on methods/constituents that have been questioned or are known, from experience, to be somewhat unreliable. Of first concern, when interpreting the hydrogeochemical data, is contamination by drilling water in the groundwater samples. Not only the reliability of the uranine analyses (the dye used to trace the drilling water) and the size of the analytical errors, but also other conditions affect the confidence in the drilling water budget calculations as well as in drilling water contents in water samples and indirectly the judgement of sample quality/representativity. Second, there is a need to establish a consistent set of major constituents (mainly Na, Ca, Cl, SO4 and possibly Mg and HCO3) for each sample. Of fundamental importance are also the reliability of data for constituents and parameters that are indicators of groundwater origin (especially magnesium, bromide and oxygen-18) or constituents and parameters that are of importance in order to understand redox conditions (such as iron and sulphide). The sulphide concentration is also a critical parameter for safety assessment evaluations