The goal of this project is to significantly advance the state of the art in computational methods for analyzing climate, ecological and other environmental datasets. The spatio-temporal nature of the data poses a number of unique challenges for traditional analysis techniques, which often work with independent and identically distributed (i.i.d.) record-based, time series, or spatial data but rarely data that exhibits all three of these characteristics. The underlying hypothesis is that there may be hidden nuggets of knowledge within eco-climatic datasets which may complement physics-based insights from first principles both in terms of descriptive analysis and predictive modeling. To this end, we are developing innovative computational methods and tools to enable the analysis of large, complex multivariate datasets. In particular, we are focusing our efforts in the following four areas:

The figure below shows the interrelationship among these four primary focus areas.

Interrelationship among the four primary research focus areas Complex Networks Relationship Mining High Performance Computing Predictive Modeling