EESA will host a DSSS lecture on October 11 in Building 66 Auditorium. Albert J. Valocchi focuses on modeling of pollutant fate and transport in porous media, with applications to groundwater resource sustainability, groundwater contamination, geological sequestration of carbon dioxide, and impacts of model uncertainty on groundwater resource management.
Albert J. Valocchi is the Abel Bliss Professor in the Department of Civil and Environmental Engineering at the University of Illinois at Urbana-Champaign. He received his B.S. in Environmental Systems Engineering from Cornell University in 1975, and his Ph.D. in Civil Engineering from Stanford University in 1981. He has been on the faculty at Illinois since 1981. From 2004 to 2012 he was Associate Head and Director of Graduate Studies.
Dr. Valocchi teaches undergraduate and graduate courses in water resources engineering, groundwater hydrology and contaminant transport, groundwater modeling, and computational methods. His research focuses on modeling of pollutant fate and transport in porous media, with applications to groundwater resources sustainability, groundwater contamination, geological sequestration of carbon dioxide, and impacts of model uncertainty on groundwater resources management.
Dr. Valocchi has received several awards in recognition of his research and teaching accomplishments. In 2009 he was recognized as Fellow of the American Geophysical Union. In 2011, he was named an Abel Bliss Professor in the College of Engineering at the University of Illinois. In 2013 he was received the Stanley H. Pierce Award from the College of Engineering in recognition of his work toward enhancing the graduate student experience.
We all aim to conduct research that will have major scientific impacts, but occasionally achieve more incremental progress. Sometimes we find that conventional approaches may be good enough in practice, or that hyped new techniques are sometimes less than what we hoped for.
1. Mixing controlled biogeochemical reactions in porous media—do we really need pore-scale experiments and modeling?
Over roughly the past decade we have used micro-fluidics experiments and pore-scale models to study mixing-controlled biogeochemical reactions in porous media. We have worked on the practically important case of transverse mixing, relevant for natural or engineered in-situ biodegradation and other scenarios. Unlike the longitudinal mixing scenario, we found that the classical theory of hydrodynamic dispersion and mixing is adequate in many cases. We also studied biofilm reactions that affect the pore geometry, and thereby cause complex feedbacks with flow and mixing. We have successfully used pore-scale flow and reactive transport models to simulate the microfluidics experiments. But in some cases we find that conventional continuum-scale models are ‘good enough.’
2. Combining physically-based and data-driven models to improve predictions of regional groundwater flow—what can we learn from machine-learning?
We have developed a framework that uses machine-learning/data-driven models to correct for errors when calibrated physically-based groundwater flow models are applied for forecasting. We have demonstrated that our approach improves accuracy of predictions of pumping impacts in real world case studies. Although the data-driven approach yields operational benefits, it has so far not provided new insights into the groundwater system or indicated ways to improve the physically-based model.