A computational database will be investigated for cyber-enabled discovery of spatio-temporal transport and mixing dynamics in geophysical turbulent flows modeled through simulation. It will be possible to identify vortical flow structures and follow them in time to compute conditional flow quantities, thereby, quantifying the dynamical role of these structures. The database provides an efficient extensible framework for formulating and validating hypotheses. It combines the machinery of an imperative model of spatial data retrieval with powerful computational facilities tailored to dynamic representations of irregular time-dependent structures, overcoming limitations of relational databases in handling such data.
Broader impacts of this research are in its contributions to the national infrastructure for research and education in Computer Science, Geosciences and Mechanical and Aerospace Engineering; it will lead to a computational methodology applicable to any field that involves time-dependent, multi-scale physical phenomena and admits mathematical rules to characterize features that are posited to be dynamically important.
Two Ph.D. students will be trained and will work together with collaborators outside their own fields, enabling cross-disciplinary teaching and training. Research results will be integrated into high-performance computing courses and the database will be used as a tool in graduate courses in turbulence and computational fluid dynamics. The software produced by the investigation will be disseminated publicly. The investigators will actively recruit from under-represented groups.
