Bikesharing schemes are transportation systems that not only provide an efficient mode of transportation in congested urban areas, but also improve last-mile connectivity with public transportation and local accessibility. Bikesharing schemes around the globe generate detailed trip data sets with spatial and temporal dimensions, which, with proper mining and analysis, reveal valuable information on urban mobility patterns. In this paper, we study the London bicycle sharing dataset to explore community structures. Using a novel clustering technique, we derive distinctive behavioural patterns and assess community interactions and spatio-temporal dynamics. The analyses reveal self-contained, interconnected and hybrid clusters that mimic London's physical structure. Exploring changes over time, we find geographically isolated and specialized communities to be relatively consistent, while the remaining system exhibits volatility, especially during and around peak commuting times. By increasing our understanding of the collective behaviour of the bikesharing users, this analysis supports policy appraisal, operational decision-making and motivates improvements in infrastructure design and management.