For a safe and secure dive with ROV QUEST we require stable weather conditions with wave heights below 3 m for several hours. We are looking for these conditions for 2-3 days ahead throughout our entire working area on the South Sandwich microplate. The working area extends over 5 degrees of latitude and 8 degrees of longitude and our meteorologist is busy finding us the forecast for the best diving conditions in all of its regions. Because of the small-scale highs and lows and their movement, this is not always easy. On Monday morning, it became evident that on Tuesday and Wednesday diving conditions were only possible in the southernmost part, in the area of the Kemp Caldera. As the multiday forecast did not permit ROV dives in the North, RV Polarstern left at ones to steam southwards bound to the Kemp Caldera. This enabled us to add a wide-scale line of a multi-beam survey to the already existing survey of the central spreading ridge. In the evening of the 21st of May the next ROV dive was meant to commence.

However, we had to realise that because of unforeseeable technical problems with the ROV no further dives during this expedition are possible, immediately leaving towards Saunders Island. There our team of volcanologists was able to proceed with their photogrammetry survey on Wednesday, 22nd of May. While during the previous survey time, the north-eastern side of Cordelia Bay and part of the south-eastern flank of Ashen Hills were surveyed, this time drone flights to the western and northern areas of the island were achieved. Important data, which enable the making of a topographic map of down to 10 cm resolution, were collected by optical and infra-red cameras during the drone surveys. After sunset, RV Polarstern steamed to the northwest to the spreading ridge. On Thursday, 23rd of Mai, several sediment and water column sampling events happened, which showed plume-like structures in the water column. Already in the night to Friday, the next storm arrived, which we could not avoid because of time constrains. As the meteorologist office on board was able to predict time and composition of the storm, we were able to plan gear deployments in line with the meteorological forecast. Although the weather model of the German Weather Service (DWD) can only use a few terrestrial base stations as well as vessel and satellite data, it is extremely correct.