We investigated gas bubble emissions at the Don-Kuban paleo-fan in the northeastern Black Sea regarding their geological setting, quantities as well as spatial and temporal variabilities during three ship expeditions between 2007 and 2011. About 600 bubble-induced hydroacoustic anomalies in the water column (flares) originating from the seafloor above the gas hydrate stability zone (GHSZ) at ~700 m water depth were found. At about 890 m water depth a hydrocarbon seep area named "Kerch seep area" was newly discovered within the GHSZ. We propose locally domed sediments ('mounds') discovered during ultra-high resolution bathymetric mapping with an autonomous underwater vehicle (AUV) to result from gas hydrate accumulation at shallow depths. In situ measurements indicated spatially limited temperature elevations in the shallow sediment likely induced by upward fluid flow which may confine the local GHSZ to a few meters below the seafloor. As a result, gas bubbles are suspected to migrate into near-surface sediments and to escape the seafloor through small-scale faults. Hydroacoustic surveys revealed that several flares originated from a seafloor area of about 1 km2 in size. The highest flare disappeared in about 350 m water depth, suggesting that the released methane remains in the water column. A methane flux estimate, combining data from visual quantifications during dives with a remotely operated vehicle (ROV) with results from ship-based hydroacoustic surveys and gas analysis revealed that between 2 and 87 x 106 mol CH4 yr-1 escaped into the water column above the Kerch seep area. Our results show that the finding of the Kerch seep area represents a so far underestimated type of hydrocarbon seep, which has to be considered in methane budget calculations.

Detailed description:During the M84-2 cruise the Kongsberg EM122 multi-beam echo-sounder was used. The EM122 uses a nominal sounding frequency of 12 kHz. 288 beams or 400 soundings with a 1°/2° footprint are formed for each ping while the seafloor is detected using amplitude and phase information for each beam sounding. For further information consult https://www.km.kongsberg.com/. The EM122 was recording constantly within the permitted areas, meaning either designated bathymetry surveys or flare imaging-surveys. Problems of proper bottom detection occurred during Leg 2a on EM122, where the inner approx. 60 beams were forming a tube form profile. Therefore the surveys during Leg 2a were done mainly with <120% overlap and later during port time at Trabzon the PU-software was reinstalled to eliminate this problem. Responsible person during this cruise / PI: Paul Wintersteller (pwintersteller@marum.de), Christian Ferreira (cferreira@marum.de). Description of data processing:Postprocessing and products were conducted by the Seafloor-Imaging group of MARUM, responsible person: Paul Wintersteller (seafloor-imaging@marum.de). The open source software MB-system (Caress, D.W., and D.N. Chayes, MB-System Version 5, Open source software distributed from the MBARI and L-DEO web sites, 2000-2012.) was utilized for this purpose. A tide correction was applied, based on the Oregon State University (OSU) tidal prediction software (OTPS) that is retrievable through MB-System. In general, there is a tide differences of less than 20 cm in the Black Sea. Though CTD measurements were taken during the M84-2 cruise, these were not sufficient to represent the changes in the sound velocity throughout the transits and partial study areas, therefore a SVP correction had to be applied. The corrections of sound velocity is based on various SVP/CTD measurements throughout different cruises in the Black Sea within the last 15-20 years. In addition, the NATO Black Sea data base (http://sfp1.ims.metu.edu.tr/ODBMSDB/ODBMSDB.dll/querydb) was consulted for further CTD profiles in the given season/period of the year and in close position to the data set. The CTD data has then been used to calculate SVP by the usage of the UNESCO formula (G.S.K. Wong and S Zhu, Speed of sound in seawater as a function of salinity, temperature and pressure (1995) J. Acoust. Soc. Am. 97(3) pp 1732-1736). Changes in SVP were applied with the mbset function within mbsystem. Further roll, pitch and heave corrections were not applied for the M84-2 data, but a draft offset correction of -5 m was applied with mbset. Bathymetric data has been manually cleaned for existing artefacts with mbeditviz. NetCDF (GMT) grids of the product and the statistics were created using mbgrid. Quality assurance was furthermore proven by visual and statistical comparison of crossing survey-lines from different cruises throughout the Black Sea. No total propagated uncertainty (TPU) has been calculated to gather vertical or horizontal accuracy. The currently published bathymetric grid of the cruise has a resolution of 125 m. A higher resolution is, at least partly, achievable. The grid extended with _num represents a raster dataset with the statistical number of beams/depths taken into account to create the depth of the cell. The extended "_sd"-grid contains the standard deviation for each cell.