Authors: Daniel Cassiday*,
Topics: Geographic Information Science and Systems, Hazards and Vulnerability, Spatial Analysis & Modeling
Keywords: GIS, big data analysis, GeoAnalytics Server, Global Forest Watch, fires, VIIRS, MODIS
Session Type: Paper
Start / End Time: 9:55 AM / 11:35 AM
Room: Tyler, Marriott, Mezzanine Level
Presentation File: No File Uploaded
Global Forest Watch (GFW) is an online platform established by the World Resources Institute and partner organizations to track worldwide forest loss. An important tool in this platform is GFW Fires, an interactive web application that visualizes active fires around the world in near-real-time using satellite imagery from NASA’s Fire Information for Resource Management System (FIRMS) team. The application includes tools for generating customized in-depth reports on fire activity at global, national, and local levels. GFW Fires data sets are very large, often on the order of 1 million active fires in a given week, with historical data archived in to provide users with opportunities for longitudinal analysis. Challenges to the use of these big datasets can include: -Data management issues when ingesting and updating data -Sluggish load and redraw times -Inability for users to interpret densely packed clusters of points at small scales -Difficulty performing data analysis in real time To address these issues, the author utilized an ArcGIS GeoAnalytics Server deployment with a spatiotemporal big data store to support enhanced visualization and analysis of fire data from the Visible Infrared Imaging Radiometer Suite (VIIRS) and Moderate Resolution Imaging Spectroradiometer (MODIS) sensors. GeoAnalytics Server leverages the power of distributed computing to open up new possibilities for data visualization and analysis, such as on-the-fly hexagonal binning and emerging hotspot analysis. This paper will discuss some of the real-world challenges to implementing big data analysis on the web, with an overview of analytic tools made possible by distributed computing.