Assessing the performance of four indicative analysis devices for ballast water compliance monitoring, considering organisms in the size range ≥10 to <50 μm.

To minimize the global transfer of harmful aquatic organisms and pathogens, the International Maritime Organization (IMO) has introduced the standard in Regulation D-2 to limit the number of viable organisms in ballast water discharged by ships. To meet the standard, many ships are installing ballast water management systems. Concurrently, regulators are looking for indicative analysis devices able to assess compliance with Regulation D-2, producing rapid, accurate and reliable results while being easy to operate. The purpose of this research is to compare four indicative analysis devices against detailed microscopy for measuring the size class of organisms ≥10 to <50 μm in minimum dimension (e.g., phytoplankton, including autotrophs, heterotrophs or mixotrophs), using field and laboratory tests. Comparisons were conducted on (treated) ballast water discharge samples collected across Canada during three consecutive years (2017–2019). During seven tests in 2019, paired ballast water uptake samples were also obtained, facilitating measurements before and after treatment was applied. Indicative analysis devices also were challenged with natural environmental samples containing different organism abundance levels, ranging from low (nominally <10 cells mL−1) to high (nominally >150 cells mL−1) during laboratory tests. While the indicative analysis devices examined during this research produced numeric estimates having weak correlations with the standard reference method, categorical outcomes (above/below the D-2 standard) had high agreement (89% or better) when assessing ballast water samples, but lower agreement (67% or poorer) during laboratory tests. There was a relatively high rate of false negative results measured by all devices during laboratory tests. Results provided by indicative analysis devices had higher uncertainty when organism abundances in ballast samples are below and close to the D-2 standards. • Indicative devices had weak correlation with microscopy based on numeric estimates. • Categorical outcomes had high agreement for ballast water samples. • Uncertainty was higher for abundances below and close to the D-2 standard. [ABSTRACT FROM AUTHOR]