In compliance with European Union present legal state (Directive 2000, 2006, 2009) surface and ground water quality monitoring should also include identification of each source of uncertainty associated with analytical procedure. Moreover, directives (Directive 2009, RMŚ 2011) indicate necessity of implementation the quality assurance and quality control program (QA/QC) of research in water monitoring. They give the measurement uncertainty threshold value as well, which is assessed basing on adequate norms of environment quality and it's equal to 50% or less (k — 2). In this paper the empirical approach for total uncertainty and its components (geochemical, sampling and analytical) assessment was used (Witczak et al. 2006, Nordtest 2007, Kmiecik 2011, Drzymała 2012). In such approach, according to Ramsey et al. (1992), the percentage contribution of measurement variance in total variance cannot exceed 20%. Otherwise, the interpretation of hydrogeochemical data cannot be performed correctly. Moreover, the influence of sampling method on the measurement uncertainty was examined. For uncertainty estimation there were used results from analyses of normal and control (duplicate) samples, which were collected within monitoring of Upper Odra River Basin surface water and coalmine water inducted to it. The assessment of total uncertainty and its components was shown on the example of Br results delivered from normal and duplicate samples analyses. The influence of container type on measurement uncertainty was determined. For this aim two procedures of samples collecting were used, concerning two types of samples containers: polyethylene containers and dark glass containers. The bromide analyses were performed using ICP-MS method in accredited Hydrogeochemical Laboratory. Each sample was analysed twice. For total uncertainty and its components assessment the ROBAN program with rANOVA technique was applied. Basing on delivered results it was stated, that in the two groups of bromide results in samples collected neither into polyethylene containers nor to dark glass containers, the measurement variance doesn't exceed the threshold value of 20% of total variance and is equal to 0.08% and 0.02% respectively. In both cases, for two types of containers, also relative uncertainty of measurement doesn't exceed the threshold value of 50% and is equal to 5.39% and 3.05% respectively. However, the difference between values of relative uncertainty of measurement is significant and indicates the necessity of collecting samples for bromide determination into dark glass containers if possible.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.