The role of lakes in the global methane budget seems to be higher than previously thought. Numerous pockmarks have been described for the marine environment, but at freshwater fluid seeps the geological, chemical and biological processes operating are largely unknown. Based on different observations of intense gas flow through the water column at Lake Constance two joint research cruises were conducted in winter 2005/06 to systematically search and study these structures. Several large fields of pockmark like structures were found in the eastern part of the lake (side scan sonar, echo sounder). Water and gas was selectively sampled and based on these preliminary data a joint DFG-funded research project was arranged. The objectives are to (1) locate, describe and map pockmark areas at Lake Constance, (2) identify the pockmark formation mechanism, (3) quantify the methane flux and the temporal variability and (4) identify the source of methane.So far more than 450 pockmarks larger > 2 m have been identified by side-scan sonar. The pockmarks vary largely in size (dm-range to maximum diameters of 15m) and spatial distribution. At some places, they are irregular spaced, but now and then smaller decimetre sized pockmarks are evenly spaced along small lineaments. Often, large pockmarks are located at morphological highs, such as channel rims or little hills at the lake floor. There is no morphological evidence for a catastrophic gas release (solid discharge at the rim, or irregular sediment patterns near the pockmarks). The observed structures point to a constant gas release from a deeper reservoir.Two representative locations (water depth 12m and 85m) close to the former Rhine estuary have been selected for further geochemical examinations. After deploying a digital horizontal sonar system at the lake floor to allow for exact positioning, water and sediment samples were taken within the pockmark with niskin bottles from a rosette and multicorer. In addition, sediment samples and gas samples were taken by divers across the shallow pockmark. Gas concentration and isotope ratios of dissolved gas in water were measured using standard GC-FID and GC-irMS techniques. Methane concentration in sediment samples was measured directly after coring by means of by a diffusion-based methane sensor.Preliminary results of the isotope analysis of methane (δ13C, δD) and CO2 (δ13C) as well as the absence of higher molecular weight alkanes strongly indicate a bacterial formation of the gases rather than a thermogenic origin of the methane. The results of the analysis of free gas of the sediments indicate the methyl formation as dominant pathway. However, up to now available data suggest a certain difference in the gases located at the deep pockmark, the corresponding reference site and gas sampled at the shallow pockmark. In the sediments of the deep pockmark high methane concentrations were recorded, however with no significant differences between the pockmark and control cores. At the shallow pockmark sediment cores could be positioned more precisely. Here methane concentrations vary within meters in order of magnitude. Several autonomous devices for gas flow measurements using different approaches were deployed in March 2008 at selected gas emanating pockmarks. Data are not yet available but will hopefully presented