[View PDF]

Nagiza Samatova NCSU/ORNL
Fred Semazzi NCSU

Objectives

• Develop predictive forecasting methodology for climate extremes (e.g., hurricanes, droughts, rainfalls)
• Devise scalable algorithms for predictive mining of large-scale climate complex networks
• Provide mechanistic insights about the key factors contributing to extreme events variability
• Demonstrate high predictive skill for North Atlantic seasonal hurricane activity

Impact

• Provide policy makers more reliable information on seasonal climate extremes
• Scalable large-scale graph mining algorithms of broader applicability (e.g., bioenergy)
• Advance our understanding of the mechanisms that influence hurricane variability and behavior
• International impact managing meningitis epidemic outbreaks driven by climate extremes

2012 Accomplishments 15 percent more accurate forecast of seasonal hurricane activity Comparative climate networks analytics & machine learning methods "Novel data-driven methods promise to excel beyond the traditional methods in climate prediction tools" (Fred Semazzi, Nobel Prize co-winner, climate scientist) Z. Chen, W. Hendrix, H. Guan, I. Tetteh, A. Choudhary, F. Semazzi, N. Samatova, "Discovery of extreme events-related communities in contrasting groups of physical system networks," Data Mining and Knowledge Discovery, 27(2), p. 225-258, 2012.