Abstract  City-wide urban infrastructures are increasingly  reliant on network technology to improve and expand their services. As a side effect of this digitalization, large amounts of data can be sensed and  analyzed to uncover patterns of human behavior. In  this paper, we focus on the digital footprints from  one type of emerging urban infrastructure: shared  bicycling systems. We provide a spatiotemporal  analysis of 13 weeks of bicycle station usage from  Barcelona's shared bicycling system, called Bicing.  We apply clustering techniques to identify shared  behaviors across stations and show how these  behaviors relate to location, neighborhood, and  time of day. We then compare experimental results  from four predictive models of near-term station  usage. Finally, we analyze the impact of factors  such as time of day and station activity in the  prediction capabilities of the algorithms