Monday, June 11

MS7
Optimization and Subsurface Flow and Transport

3:00 PM - 5:00 PM
Room: Trailridge

This minisymposium will include presentations on a variety of optimization problems in the geosciences and algorithms that are particularly useful in those applications. The optimization problems are difficult because of the high computation cost of the underlying simulation, the presence of multiple local minima and nonsmoothness in the objective function, and the continual change in the models of the underlying physics. The speakers will discuss applications such as optimal design of remediation strategies for groundwater and parameter estimation. The methods range from gradient-based variations of Newton's method to several types of direct search and sampling methods. The minisymposium will be of interest to specialists in remediation and recovery, parameter identification, and optimization.

Organizer: Tim Kelley
North Carolina State University
Cass T. Miller
University of North Carolina at Chapel Hill

3:00-3:25 A Multiobjective Approach to Cost Effective Long-Term Groundwater Monitoring Design using an Elitist Nondominated Sorted Genetic Algorithm

David E. Goldberg, Patrick Reed, and Barbara S. Minsker, University of Illinois at Urbana-Champaign

3:30-3:55 Control of subsurface temperature with implicit filtering

Alton Patrick, Tim Kelley, and Kathleen R. Kavanagh, North Carolina State University; Cass T. Miller, University of North Carolina at Chapel Hill; Astrid Battermann, University of Trier, Germany; Joerg M. Gablonsky, The Boeing Company

4:00-4:25 Using Multiobjective Optimization to Construct Tradeoff Curves for Subsurface Remediation

Mark Erickson and Alex S. Mayer, Michigan Technological University; Jeffrey Horn, Northern Michigan University

4:30-4:55 Optimal Groundwater Remediation Design with Variable Design Periods

George Karatzas, Technical University of Crete, Greece; Maria Papadopoulou and George Pinder, University of Vermont