Influence of QA/QC procedures on non-sampling error in deposition monitoring in forests

A Working Ring Test (WRT) was organised in the framework of the EU Regulation (EC) No 2152/2003 ("Forest Focus") and of the UN/ECE Program "ICP Forests" to evaluate the overall performance of the laboratories monitoring atmospheric deposition and soil solution in European Forests. Seven natural samples of atmospheric deposition and soil solutions and 5 synthetic solutions were distributed to 52 laboratories, which analysed them using their routine methods. Thirteen variables are considered in this paper: pH, conductivity, calcium, magnesium, sodium, potassium, ammonium, sulfate, nitrate, chloride, total alkalinity, total dissolved nitrogen and dissolved organic carbon. For each variable, the relative standard deviation of the results was evaluated, after outlier rejection, to estimate the analytical error of the measurements. The results are evaluated considering the Quality Assurance/Quality Control (QA/QC) procedure included in the ICP Forests monitoring manual: consistency check of the data and use of control charts and internal standards. A Data Quality Objective (DQO) is defined for each of the variables and the number of data meeting the DQOs are discussed in relation to the QA/QC procedures adopted. Although 38% of the results did not meet the DQO, the laboratories adopting QA/QC procedures produced a larger proportion of results meeting the objective and a consistent part of the outliers could be detected a posteriori checking analyses consistence.