IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, TVCG-2008-07-0094.R1 Visualization of Simulated Urban Spaces: Inferring Parameterized Generation of Streets, Parcels, and Aerial Imagery Carlos A. Vanegas, Daniel G. Aliaga, Bedřich Beneš, Paul Waddell Abstract—Urban simulation models and their visualization are used to help regional planning agencies evaluate alternative transportation investments, land use regulations, and environmental protection policies. Typical urban simulations provide spatially distributed data about number of inhabitants, land prices, traffic, and other variables. In this article, we build on a synergy of urban simulation, urban visualization, and computer graphics to automatically infer an urban layout for any time step of the simulation sequence. In addition to standard visualization tools, our method gathers data of the original street network, parcels, and aerial imagery and uses the available simulation results to infer changes to the original urban layout. Our method produces a new and plausible layout for the simulation results. In contrast with previous work, our approach automatically updates the layout based on changes in the simulation data and thus can scale to a large simulation over many years. The method in this article offers a substantial step forward in building integrated visualization and behavioral simulation systems for use in community visioning, planning, and policy analysis. We demonstrate our method on several real cases using a 200 GB database for a 16,300 km 2 area surrounding Seattle, Washington. Index Terms— Computer Graphics, Information Visualization, Picture/Image Generation, Simulation Output Analysis, ——————————  —————————— 1 INTRODUCTION Urban simulation models and the visualization of their results are increasingly being used in city, county, and regional planning agencies to assess alternative transportation investments, land use regulations, and environmental protection policies. The amount of data generated by such a simulation model over a long forecasting horizon and over a large scale (e.g., 10 to 30 years for a city of several million people) is overwhelming and therefore difficult to easily interpret for planners, policy‐makers, and the public and even for the modelers running the simulation. Visualization techniques are essential to extract useful information from the large mass of data generated by such simulations. However, to date such simulation systems have been limited in their scope of visualization, in spite of providing very sophisticated economic and behavioral simulation engines. In this article and to the best of our knowledge, we propose a first method using the input data to an urban simulation, the output data of the simulation, and computer graphics techniques to automatically and interactively infer urban layouts (Figure 1). A city can be represented by its urban layout which we define to be the intricate collection of its man‐made structures arranged into parcels, blocks, streets, and neighborhoods (e.g., aerial images together with Geographical Information Systems (GIS) vector data such as that provided by Google Maps, MapQuest, etc.). Since urban layouts are difficult to model by hand because they are very complex, large, and widespread, we use a simulation and an automatic inference approach to create a layout of an existing or of a future urban space, therefore enabling multiple forms of visualizations for the aforementioned applications. Our work builds on a synergy of efforts in urban simulation, urban visualization, and computer graphics (Figure 2a). An urban simulation attempts to model and predict the complex socio‐economic interactions that govern the growth and development of an urban area. A typical output of such simulation is predictions of real estate development, prices, and location choices of households and firms at fine‐grained levels of geography such as grid cells or parcels, over entire metropolitan areas, and over planning horizons of several decades. Due to the magnitude and granularity of these simulations, it is difficult to adequately estimate all parameters, to automatically determine an exact city configuration at all stages of the simulation, and to intuitively visualize the significance of computed results. Urban visualization systems have been used to assist such simulations but a typical scenario is that a manual post‐ processing of simulation results needs to be done by a technical user. For instance, the user extracts summary indicators from the results, exports them from the ———————————————— • Carlos A. Vanegas and Daniel G. Aliaga are with the Department of Computer Science at Purdue University, West Lafayette, IN 47907. Email: cvanegas@cs.purdue.edu , aliaga@cs.purdue.edu . • Bedřich Beneš is with the Department of Computer Graphics Technology at Purdue University, West Lafayette, IN 47907. E-mail: bbenes@purdue.edu • Paul Waddell is with the Evans School of Public Affairs at the University of Washington, Seattle, WA 98195. E-mail: pwaddell@u.washington.edu Manuscript received