The harbour porpoise (Phocoena phocoena) is an important top predator and as such an indicator species for its environment. Before the beginning of the 21st century, little data existed on distribution and abundance of harbour porpoises in German waters of North and Baltic Sea. As a consequence, virtually nothing was known about important habitats, seasonal differences in distribution and environmental variables determining its distribution. Studies of habitat characteristics of harbour porpoises have been driven both by the need to support conservation and management actions and the increasing availability of suitable tools (e.g. GIS, remote sensing, regression modelling techniques). This thesis aimed to characterise the habitat of harbour porpoises in German waters by estimating abundance, examining spatial and seasonal patterns in distribution, deriving habitat prediction models and investigating the feeding ecology of porpoises in order to infer on predator-prey dynamics. A comprehensive database was set up based on results of aerial surveys conducted year-round in the course of five consecutive years (2002-2006), following standard line transect methodology. Robust abundance estimates for different surveys, that accounted for animals missed on the transect lines, could be derived for the Baltic and the North Sea. In the Baltic Sea, these estimates are especially important to evaluate the effect of bycatch that was found to be a major threat to porpoises throughout the western Baltic Sea. In the North Sea, presented abundance estimates will serve as a baseline for management decisions with respect to the projected construction of large offshore wind farms and its possible impacts on porpoises. Important habitats were detected in offshore waters of the German North Sea: in spring, the two hot spots ‘Borkum Reef Ground’ and ‘Sylt Outer Reef’ (SOR) were identified as key foraging areas for harbour porpoises. In summer, the large hot spot SOR persisted, causing a strong north-south density gradient. In autumn, porpoises were more evenly distributed and density was lower than during spring and summer. Differences in these seasonal hot spots were investigated in relation to several static (e.g. depth, slope) and dynamic (e.g. sea surface temperature, chlorophyll) predictors by applying generalised additive models (GAM) and mixed models (GAMM). The key habitat descriptors as selected by the models varied between seasons. Predictors explaining most of the variance were the hydrographical parameter ‘residual currents’ and proxies for primary production and fronts (chlorophyll and nutrients) as well as the interaction ‘distance to coast/water depth’. In order to address possible seasonal dietary shifts, two methods, the traditional method of stomach content analysis and the new method of quantitative fatty acid signature analysis (QFASA) were applied to elucidate the feeding ecology of harbour porpoises in German waters. A multivariate analysis revealed significant seasonal and between-year fluctuations in the relative importance of prey species: in spring, mainly sandeel, goby and herring contributed with high masses whereas cod was most important in summer. Cod seemed to be an important prey item throughout the study period. Goby and flatfish species appeared to be more important before 2001, whereas in the period 2002-2006 sandeel, herring and sprat contributed more to the diet. In conclusion, this thesis discovered important temporal and spatial patterns of habitat use and elucidated its underlying causes as harbour porpoise density were set in relation to the German Bight environment. This thesis could fill addressed gaps in knowledge and improved our understanding of harbour porpoise ecology in German waters and possibly beyond.