NSF Data Infrastructure Building Blocks: Scalable Capabilities for Spatial Data Synthesis

Massive spatial data collected from numerous sources are increasingly used to instrument our natural, human and social systems at unprecedented scales while providing us with tremendous opportunities to gain dynamic insight into complex phenomena. Though such big data streams play crucial roles in many scientific domains and promise to enable a wide range of decision-making practices with significant societal impacts, exploiting them successfully poses significant challenges. On one hand, spatial and location attributes serve as a common key to many types of data such as census and population, land use and cover, floodplain, and vegetation distribution. Oftentimes perceived as significant benefits, spatial data synthesis can be used to link disparate pieces of data that pertain to common spatial references and units. On the other hand, however, there are diverse spatial references and units for data collection and management and they are based on different representation models and assumptions.

To break through these challenges, this project aims to establish a suite of scalable capabilities for spatial data synthesis enabled by innovative cloud computing and cyberGIS and driven by multiple scientific communities. Such capabilities will also be designed to support integration with cyberGIS analytics and workflow for solving scientific problems. The project establishes core capabilities through a spiral approach by initially developing the capabilities for solving specific scientific problems and later moving on to engage broader communities for validating and improving the core capabilities. The scientific problems will revolve around two interrelated themes: 1) measuring urban sustainability based on a number of social, environmental, and physical factors and processes; and 2) examining population dynamics by synthesizing multiple states of the art population data sources with social media data.

NSF Grant Numbers: 1443080

People: Kiumars Soltani (soltani2@illinois.edu), Pierre Riteau (priteau@uchicago.edu), Hao Hu (haohu@illinois.edu) , Anand Padmanabhan (apadmana@illinois.edu), Kate Keahey (keahey@anl.gov), Shaowen Wang (shaowen@illinois.edu)