Jennifer Ryan

Modeling of physical phenomena such as environmental pollutants, volcanic eruptions, and hurricanes relies on the ability to connect information from widely different spatial scales. Simulating the complex physics that arise in these applications requires considerable computing time and significant hardware resources. This necessitates reliance on techniques that can only represent part of the data. The choice in how we represent data leads to different patterns in the data. These patterns contain "hidden information" that can be used to construct more accurate representations of physical phenomena.

Jennifer Ryan's research develops mathematical methods and computational filters to enhance our ability to extract information from given data and to provide insight into any underlying hidden information. The properties of the filters she designs are physics-based and grounded in rigorous mathematical theory, making them useful in areas requiring large-scale computational models. By detecting the small-scale anomalies in the data, it is possible to construct bespoke filtering methods that reduce the noise in the data, allowing for the extraction of more accurate approximations to the physical phenomena being modeled.