Asmund Gert, Vorkamp Katrin, Backus Sean, Comba Michael
Department of Arctic Environment, National Environmental Research Institute, Frederiksborgvej 399 DK4000 Roskilde, Denmark.
Sci Total Environ. 2004 Sep 20;331(1-3):233-45. doi: 10.1016/j.scitotenv.2004.03.031.
The majority of analytical results in the Greenland AMAP (Arctic Monitoring and Assessment Programme) are produced by laboratories that participate regularly in performance studies. This makes it possible to judge the quality of the results based on the objective measurements made by independent assessors. AMAP laboratories participated, while analysing the AMAP samples, in the QUASIMEME laboratory performance study programme, and in two other smaller laboratory performance study programmes. Here is a system presented where the laboratory performance studies are used to calculate a constant error and a proportional error for each analyte. The calculations are performed by the trial and error method that sets the proportional and the constant error so that the 95%, the 68.2%, and the 50% confidence limits fit as well as possible with the data. The constant error is roughly the same as the detection limit defined as the highest concentration where zero is inside the 95% confidence interval. The relative errors of the trace analyses, i.e. the relative deviation of the result obtained by the AMAP laboratory from the assigned value, are in most of the cases less than 25% which is regarded as acceptable by the QUASIMEME. Usually the errors, especially for trace elements, are less than 12.5%, while errors for trace organics below 1 microg kg(-1) may rise to 50% or more.
格陵兰AMAP（北极监测与评估计划）的大部分分析结果由定期参与性能研究的实验室得出。这使得能够根据独立评估者的客观测量来判断结果的质量。AMAP实验室在分析AMAP样本时，参与了QUASIMEME实验室性能研究计划以及其他两个较小的实验室性能研究计划。这里介绍了一种系统，其中实验室性能研究用于计算每种分析物的恒定误差和比例误差。计算通过试错法进行，该方法设定比例误差和恒定误差，以使95%、68.2%和50%的置信限尽可能与数据拟合。恒定误差大致与检测限相同，检测限定义为零在95%置信区间内的最高浓度。痕量分析的相对误差，即AMAP实验室获得的结果与指定值的相对偏差，在大多数情况下小于25%，QUASIMEME认为这是可接受的。通常误差，尤其是对于痕量元素，小于12.5%，而低于1微克/千克的痕量有机物的误差可能会升至50%或更高。
