Bacterial benthic community composition in the Baltic Sea in selected chemical and conventional weapons dump sites affected by munition corrosion

Cytogenetic damage in native Baltic Sea fish species: environmental risks associated with chemical munition dumping in the Gotland Basin of the Baltic Sea

This study represents the first attempt to assess genotoxicity and cytotoxicity effects in herring (Clupea harengus membras), flounder (Platichthys flesus), and cod (Gadus morhua callarias) caught at 47 study stations, located close to chemical munition dumpsites in the Gotland Basin, the Baltic Sea. Herring sampled from stations located in the center of chemical munition dumpsites exhibited the highest levels of micronuclei (MN) and total genotoxicity (ΣGentox), which is defined as the sum of frequencies of such nuclear abnormalities as micronuclei, nuclear buds, nuclear buds on the filament, and bi-nucleated erythrocytes with nucleoplasmic bridges. Exceptionally high and high ΣGentox risks were determined for flounder (89.47%), herring (79.31%), and cod (50%) caught at the stations located close to the chemical munition dumpsites.

Presence of mercury and methylmercury in Baltic Sea sediments, collected in ammunition dumpsites

Methylmercury (MeHg) is the most toxic and dangerous form of mercury occurring in the environment. MeHg is highly bioaccumulative in organisms and undergoes biomagnification via the food chain. In the Baltic Sea munition dumpsites, methylmercury can be formed from mercury fulminate contained in primary explosives, as environmental conditions there favour methylation. MeHg in analysed sediments ranged from 19 to 2362 pg g^-1d.w., the concentration of mercury (Hg_TOT) ranged from 4 to 294 ng g^-1 d.w., and the values of MeHg/Hg ratio ranged from 0.1 to 2.0%. The obtained results confirmed that munition dumpsites are a source of mercury. The concentration of MeHg is elevated in a wider area than immediately next to dumped munitions. Presented results suggest that physical processes responsible for sediment and near-bottom water movement are diffusing MeHg signal, making munition dumpsites rather a diffuse source of MeHg than a number of point sources associated with particular munitions.

The Effectiveness of Using a Pretrained Deep Learning Neural Networks for Object Classification in Underwater Video

Video image processing and object classification using a Deep Learning Neural Network (DLNN) can significantly increase the autonomy of underwater vehicles. This paper describes the results of a project focused on using DLNN for Object Classification in Underwater Video (OCUV) implemented in a Biomimetic Underwater Vehicle (BUV). The BUV is intended to be used to detect underwater mines, explore shipwrecks or observe the process of corrosion of munitions abandoned on the seabed after World War II. Here, the pretrained DLNNs were used for classification of the following type of objects: fishes, underwater vehicles, divers and obstacles. The results of our research enabled us to estimate the effectiveness of using pretrained DLNNs for classification of different objects under the complex Baltic Sea environment. The Genetic Algorithm (GA) was used to establish tuning parameters of the DLNNs. Three different training methods were compared for AlexNet, then one training method was chosen for fifteen networks and the tests were provided with the description of the final results. The DLNNs were trained on servers with six medium class Graphics Processing Units (GPUs). Finally, the trained DLNN was implemented in the Nvidia JetsonTX2 platform installed on board of the BUV, and one of the network was verified in a real environment.