Year 2017
Joonas Kahiluoto

Automating continuous turbidity field measurements with continuous measurement uncertainty estimation

The need for environmental monitoring and producing reliable information is more and more valuable in the everchanging world. Due to technological advancements, new technologies with better spatial and temporal resolutions are emerging, but insufficient knowledge about the reliability of these technologies hinders their wider adaptation.

In this thesis, a concept for reliable continuous field measurements was developed. The concept was then tested by designing and building an automated water turbidity measurement station and applying Nordtest TR537 for measurement uncertainty calculations from previous measurement results. The developed concept can be used for automated quality control in other continuous measurements as well in the boundaries set by available measurement devices and standard solutions.

Validation and utilization experiments were conducted in which the measurement station performed reliably. From the experimental results, it was evident that there are problems in the mixing and stability of the formazin standard solutions. The stability of the standard solutions was investigated in a separate experiment and the stability appeared to be satisfactory in laboratory conditions. Because of the problems in the standard solutions the bias and measurement uncertainty of the measurement devices was not accurately determined and the estimates presented are higher than they should be. The estimated measurement uncertainties are higher than what is recommended for the information in environmental monitoring databases, but addressing the known issues would probably lower the uncertainties to meet the criteria.

Results from the measurement devices were compared with laboratory results from samples taken from the same place at the same time and a strong liner correlation was found between the results. The results from Optosevens turbidimeter did not differ statistically significantly from the laboratory results. The difference between results from ABBs turbidimeter and laboratory results was statistically significant.

The work will continue with combining the control system (SYKE Cloud) and measurement uncertainty calculation software (AutoMUkit) and mixing studies.

Thesis electronical version can be downloaded from here

This info last modified 27 Apr 2024 by Jukka Kortela