Missing uncertainties

Sometimes, uncertainties are not reported with the concentration measurements. The uncertainties then have to be estimated by some other means. Two approaches wre used for the AMAP data. In some cases, the uncertainties were specified by the data provider based on expert judgement. In others, the uncertainties were estimated using fixed and relative standard deviations derived from the uncertainties reported to the ICES database by a wide range of analytical laboratories for the monitoring period 1998 to 2017 (data extraction: 15 February 2019). Specifically, suppose \(c\) is a concentration measurement reported with missing uncertainty. Its uncertainty \(u\) was then estimated to be:

\[u^2 = s^2 + v^2 c^2\]

where:

• \(s\) is the fixed standard deviation, taken to be one third of the detection limit when reported and 0.7833 \(\mu \text{g kg}^{-1}\) otherwise
• \(v\) is the relative standard deviation, taken to be 0.0791 (unitless)