Total residual oxidants (TRO) in treated ballast water can produce various disinfection by-products (DBPs) depending on local conditions, such as salinity and organic matter content in water. Because TRO and DBPs are known to be harmful to aquatic organisms and humans, ecotoxicity tests have been proposed for screening the residual toxicity before discharging treated ballast water. In the present study, we aimed to address the decay rates and toxicity changes of TRO under various conditions in salinity, initial TRO concentrations, and residence time of TRO. In addition, the toxicological sensitivities of bioluminescent bacteria Vibrio fischeri and a commonly-used microalgae Skeletonema costatum relative to the residual toxicity of TRO and six selected DBPs were determined. Decay rate of TRO concentration increased as a function of salinity and was affected by the initial concentrations of TRO. Unexpectedly, significant bioluminescence inhibition was observed for hypochlorite-treated water at < 0.1 mg L-1 TRO (expressed as Cl2), which is a lower concentration than the maximum allowable discharge concentration (MADC) to marine waters established by the International Maritime Organization (IMO). The ecotoxicological thresholds of no observed effective concentration and median effect concentration for all tested DBPs were about 3-10 times lower for V. fischeri than for S. costatum. The results indicate that bioluminescent microbes possess an ecologically-relevant sensitivity to both TRO and DBPs in ballast water. In general, bioassay using V. fischeri was potentially more effective than microalgae for screening the total toxicity of TRO and DBPs in treated ballast water, especially given that ballast water usually contains a highly variable and complex mixture of toxicants.