The proliferation of sensor devices in the emerging landscape of ‘smart cities’ provides new mechanisms with which to measure the built and natural environment. City governments increasingly rely on sensor data to monitor infrastructure, mobility patterns, environmental hazards, disasters, and more. At the same time, citizens have increasing access to tools with which to examine urban concerns outside of institutional means. By looking at the use of one specific category of sensor data, air quality, this thesis provides a critical analysis of the plurality of ways in which urban sensing data is generated and represented. Specifically, the thesis examines representations of air quality data intended for governmental to grassroots audiences, and how these representations may prove to be problematic in attempts to reconcile their myriad forms and meanings across contexts and constituencies. Urban planning and design, disciplines that rely on the interpretation of environmental data in order to propose strategies for shaping the built environment, serve as a unique point of convergence of the key tensions that persist in the use of sensor data in cities. Case studies of various urban sensing initiatives in the U.S. and abroad illustrate disjunctions between different modes of sensor data collection and the way that data is communicated, affecting the way that governments negotiate with citizen stakeholders and vice versa. The core research questions this thesis examines are twofold: (1) What are the ways in which air quality sensor data is represented and given meaning in city dashboards, data portals, and other graphic user interfaces for different audiences, and (2) How might sensor data be used in the context of urban planning and design to reveal new frameworks for environmental data collection and representation that promote collaboration between government and citizen stakeholders?