Aqueous Pollutant Capture with Enhanced Filtration & On Estimation of Hydrometeorological Signals with Sparse Prior

Program
3:00 p.m. Refreshments
3:30 Welcome and Awards Presentation – Miki Hondzo, Professor of Civil Engineering and Associate Director of Research and Development, St. Anthony Falls Laboratory
3:45 Award Remarks and Seminar Presentations – Mohammad Ebtehaj and Andrew Erickson, Ph.D. candidates in Civil Engineering, St. Anthony Falls Laboratory
4:30 Question and Answer

Aqueous Pollutant Capture with Enhanced Filtration

Andrew Erickson,
Ph.D. candidate in Civil Engineering, St. Anthony Falls Laboratory,
College of Science and Engineering, University of Minnesota

A recent nationwide study reports that many stormwater pollutants such as phosphorus, cadmium, copper, zinc, and nitrogen are approximately 45% dissolved. Very few stormwater treatment practices can consistently capture dissolved pollutants over the life-cycle of a treatment practice, and therefore a large portion of the pollutant load is entering our impaired water bodies untreated. This presentation will discuss proven techniques for capturing dissolved pollutants and examine field applications of these techniques.

On Estimation of Hydrometeorological Signals with Sparse Prior

Mohammad Ebtehaj,
Ph.D. candidate in Civil Engineering, St. Anthony Falls Laboratory,
College of Science and Engineering, University of Minnesota

The past decades have witnessed a remarkable emergence of new sources of multi-scale multi-sensor geophysical data such as precipitation, soil moisture, cloud cover, and vegetation. For precipitation, these data include global spaceborne active and passive sensors, regional ground-based weather surveillance radars, and local rain gauges. Optimal integration and resolution enhancement of these multi-sensor data promise a posteriori estimates of precipitation fluxes with increased accuracy to be used for more accurate prediction of hydro-geomorphological events, such as floods and landslides. To this end, new methodologies are presented that capitalize on an important but overlooked property of precipitation, namely that of “sparsity”. “Sparsity” refers to the fact that when a signal is projected onto an appropriate domain (say, the gradient space) only a few of the projection coefficients are non-zero. Given the ubiquity of sparsity in many hydro-climatic processes, I intend to advance and redefine the conventional data fusion and assimilation techniques in hydrometeorological applications.