Big data and data mining have provided several breakthroughs in fields such as health informatics, smart cities and marketing. The same techniques, however, have not delivered consistent key findings for climate change.
Big data and data mining have provided several breakthroughs in fields such as health informatics, smart cities and marketing. The same techniques, however, have not delivered consistent key findings for climate change.
There are a few reasons why. The main one is that previous data mining work in climate science, and in particular in the analysis of climate teleconnections, has relied on methods that offer rather simplistic “yes or no” answers.
“It’s not that simple in climate,” said Annalisa Bracco, a professor in Georgia Tech’s School of Earth and Atmospheric Sciences. “Even weak connections between very different regions on the globe may result from an underlying physical phenomenon. Imposing thresholds and throwing out weak connections would halt everything. Instead, a climate scientist’s expertise is the key step to finding commonalities across very different data sets or fields to explore how robust they are.”
And with millions of data points spread out around the globe, Bracco said current models rely too much on human expertise to make sense of the output. She and her colleagues wanted to develop a methodology that depends more on actual data rather than a researcher’s interpretation.
Read more at Georgia Institute of Technology
Photo credit: jodylehigh via Pixabay
