Identifying necessary steps to promote agreed upon procedures of quantifying uncertainty for in-situ ocean measurements
by Dr. Christoph Waldmann

Abstract

A number of approaches exist that help evaluate uncertainties for environmental measurements. WMO has established a dedicated expert team led by Andrea Merlone of INRiM to address this task. In ocean sciences, there obviously is a barrier that has to be overcome which is related to the fact that for each measurement variable and also used instrument an overall uncertainty budget has to be established. The sheer number of ocean variables makes this a huge task. However, as soon as an agreed upon evaluation scheme is available the calculation procedure can be implemented in MATLAB or Python using a defined algorithm which will alleviate the introduction. Examples for remote sensing data and for current measurements exists and can be used as templates. For each variable considered the measurand, in this case in- situ measurements of ocean parameters, has to be defined based on the existing process knowledge. This then forms the base to separate random fluctuations from the actual observed variation of interest. In this presentation the different considerations that shall lead to an agreed upon uncertainty evaluation will be presented and possible outstanding issues identified. Based on an experiment carried out at a small island in the German Bight in 2020 the evaluation of uncertainty through the example of temperature measurements will be discussed. Due to the different environmental conditions over the time interval of observation specific issues can be shown, for instance the occurrence of 1/f noise that makes the calculation of measurement means a challenge. Finally, a strategy on how to introduce uncertainty quantification procedures will be discussed that also addresses the role of WMO and IOC/GOOS in this framework.