Last fall, the DigitalGlobe Foundation launched a pilot program that expanded its mission to foster the next generation of geospatial leaders. The pilot program offered access to DigitalGlobe’s Global Basemap (GBM) to two innovative universities: UC San Diego and the University of Minnesota. In this two-part blog series, we look at what GBM has enabled these universities to achieve and where they plan to go from here.
Global Basemap and UC San Diego’s Big Pixel Initiative
“It’s pretty incredible to realize the power of remote imaging in our modern age, where a top view perspective can capture and quantify with high fidelity the dynamics of our very existence,” said UC San Diego Research Scientist Dr. Albert Lin. “Then, when you get a sense of the depth of the archive, it becomes clear that we are on the brink of something big—a moment when vast scales of data combined with powerful computation will change the nature of the questions we can even ask.”
This is the basic principle of the Big Pixel Initiative (BPI) started at UC San Diego to provide researchers with unique access to DigitalGlobe’s Global Basemap, made possible by the partnership between the university and DigitalGlobe Foundation.
“Having access to DigitalGlobe Basemap will allow researchers to ask questions and derive answers at a scale that is truly global,” said Gordon Hanson, economics professor, director of the school’s Center on Global Transformation, and co-director the Big Pixel Initiative with Dr. Lin. “By combining geospatial data with state-of-the-art research in fields such as public health, economics and ecology, we should be able to glean insights that were not previously possible.”
DigitalGlobe Basemap in Action
To put this into action, the BPI team conducted a series of campus-wide activities with DigitalGlobe’s Global Basemap. Initially the team launched a cross-departmental faculty workshop to familiarize leaders from computer science, visual arts, political science and the Scripps Institution of Oceanography with satellite imagery. Soon after, the campus hosted a day-long GBM hackathon with PhDs and Masters students from both computer science and the social sciences. The theme: “What Will You Discover?”
Students competed to develop the most compelling research questions that could be explored through imagery at scale. Winning ideas included an algorithmic approach to global quantification of informal settlements, and a meshing of imagery and global ship tracking data to separate permitted from unpermitted boats to identify illegal smuggling routes.
Following these activities, eight mini-grants ($15,000) were provided to a broad cluster of researchers. The only requirement for the cross-disciplinary efforts was that they used satellite imagery to ask big questions. For example, one team of researchers from the math department developed an algorithm to identify refugee camps through high-resolution data—an effort to quantify the movements of a highly dynamic and hard to measure populations. Another project used imagery to quantify the expansion of urbanization across India over the last decade and its relationship to social economic dynamics.
“I am excited,” said Dr. Ran Goldblatt, a postdoctoral researcher at the Big Pixel Initiative. “Utilizing big data to understand our world, how it changes and the relationships that drive it is what I want my research carrier to focus on.”
Some early results from his work with Hanson using satellite imagery to survey 30,000 mines across the world has shown that global economic cycles and booms in mineral commodity pricing has a significant impact on forest cover loss, at scale.
“Throughout the year, one of the most exciting aspects of the initiative has been expanding the bridges between applied research and the leading tools of industry in GIS,” said Jessica Block, the lead GIS coordinator of BPI. In June, Block and others from UCSD spent several days at DigitalGlobe headquarters to explore new collaborative research opportunities between UCSD and DigitalGlobe within the pilot program.
From these discussions, they envisioned a path forward where the university and the remote sensing industry are partnered closely in the experimental application and development of tools that drive insight. Tools such as DigitalGlobe’s Geospatial Big Data platform or Tomnod crowdsourcing could help overcome the computational hurdles this type of thinking must address.
“It seems inevitable that the combination of big data and big analytics will produce insights that may shift how we see the world,” explained Lin. “But what excites me most is that through this partnership we launch collectively into the unknown. We enter into a new age of discovery.”
Our second post in this series will explore how the University of Minnesota has leveraged Global Basemap to ask big questions.