Chemical warfare agents and their risk assessment in Daphnia magna and fish in the Baltic Sea - 15 years of measurements

The Baltic Sea is one of the world's largest brackish waters, it is drained through the Danish Straits into the Kattegat, and almost enclosed by nine countries. The Baltic Sea is one of the world's most polluted water bodies thus facing a wide range of environmental threats towards its water resources, such as fish stocks, and coastal environments and economics. Chemical warfare agents (CWAs) that were dumped following the Second World War are known to occur in intact or degraded states in sediments and are documented to affect benthic fauna and fish as well as having injured fishermen having accidentally caught lumps of CWAs in their nets. However, a thorough mapping of remaining CWAs and degradation products and an understanding of the environmental risks does not exist although more than 75 years have passed. This study compiles and analyzes published/peer-reviewed data, generated since 2005 in five selected comprehensive studies, on sediment measurements of known dumped CWAs and degradation products in the Baltic Sea. As a worst-case approach, sediment con-centrations are transformed to concentrations in near-bottom water, which represents Predicted Environmental Concentrations (PECs) to marine biota. To investigate the accuracy and representativeness of toxicological values, which are fundamental in risk assessments, two cases are considered: Case 1 (specificity) uses toxico-logical data (EC50 or NOEC) for Daphnia magna and fish with applied assessment factors (AF) to derive Predicted No-Effect Concentrations (PNECs) for organoarsenical and non-arsenical CWAs; Case 2 (robustness) uses partly Danish Environmental Quality Standards (DK EQS) for arsenicals, and partly a Water Quality Criterion (WQC) for arsenicals, representing the marine environment. From 872 data points risk quotients (RQs=PEC/PNEC) are calculated. In Case 1 exceedances of risk for the sum of chemicals (sumRQ>1) occur 24 and 1 times for Daphnia magna and fish, respectively, without applying AFs. 263 and 120 exceedances are found for Daphnia magna and fish, respectively, when applying AFs. Case 2 shows 0 (WQC) and 25 (DK EQS) exceedances for arsenicals when using more robust toxicological values, however, at the expense of specificity of chemicals and target species. The results underline the importance of obtaining more representative and accurate toxicological data (lowering AFs) in order to increase the accuracy of the risk estimates. This quantitative state of risk towards representative marine species indicates that there are indeed potential risks, and it qualifies the understanding and debate on the challenges and future actions regarding dumped chemical munitions in the Baltic Sea.