BIG Geospatial Data
WHAT IS SPATIAL BIG DATA? Defined in part by the context, use-case Data too big, complex for traditional desktop GIS Often seen as relating to user experience Three data attributes of unsatisfactory user experience
VOLUME Massive Globally distributed Unacceptable response time Example: Kriging crowd-sourced temperature data
VELOCITY Frequent data Real time Example: monitoring of smart phones, tweets Data loss System failure
VARIETY Multi-dimensional Large human effort to accomplish task Fusion of multiple data sources Example: mapping post-disaster situation on the ground
EXAMPLES OF BIG SPATIAL DATA Raster Global Climate Models Unmanned aerial vehicle data (drones) LiDAR Vector Volunteered Geographic information (OpenStreetMap) GPS Trace Data (tied to eco-routing)
Graph Spatial-Temporal Engine Measurement Data (vehicle sensors sensing elevation) Historical Speed Profiles (dynamic road routing)
Traditional Spatial DataBig Spatial Data Simple Use Cases Map of 2012 election voter preferences Real time maps of tweets, traffic ExamplesPoint, line, raster graph dataCheck-ins, drone videos, GPS tracks in phones Volume10 6 crime reports/year, gigabytes of roadmaps GPS traces VarietyRaster, vector, graphMoving objects, time-series VelocityLimited velocity (waiting for next Census) High velocity (real-time map of tweets)
SOURCES Directed surveillance Automated inherent Volunteered gifted
DATA PROCESSING A need to utilize data Integration Open data analytics
APPLICATIONS OF BIG SPATIAL DATA Eco-Routing UPS routes avoid left-turns to limit idling, save fuel Eco-routing could be extended across industries, help save fuel Climate Change models With more years of historical models, long-range climate models will be more robust CartoDB earth observation Disaster response Red Cross detected tornado in Texas by following tweets, seeing hotspot
IT CHALLENGES Data Intensity Lots of data Coming in fast! Formatting, structure, organization Computing Intensity Earth phenomena is complex Complex algorithms and models needed Often beyond standard computing capacity
Concurrent Intensity Allow use to millions of people at the same time, Emergency response capabilities Spatiotemporal Intensity Data must be intense across space and time Geographic, atmospheric, oceanic
OTHER ISSUES Trustworthy Privacy Ethical Technocracy Corporatization and technology lock-in
REFERENCES Evans, M. R., Oliver, D., Yang, K., & Shekhar, S. (2013). Enabling Spatial Big Data via CyberGIS: Challenges and Opportunities. CyberGIS: Fostering a New Wave of Geospatial Innovation and Discovery. Springer Book. Yang, C., Goodchild, M., Huang, Q., Nebert, D., Raskin, R., Xu, Y., & Fay, D. (2011). Spatial cloud computing: how can the geospatial sciences use and help shape cloud computing?. International Journal of Digital Earth, 4(4),