HR: 08:30h
AN: H11K-03    [Abstracts]
TI: Combining Wireless Sensor Networks and Groundwater Transport Models: Protocol and Model Development in a Simulative Environment
AU: * Barnhart, K
EM: kbarnhar@mines.edu
AF: Division of Environmental Science and Engineering, Colorado School of Mines, 1500 Illinois St, Golden, CO 80401, United States
AU: Urteaga, I
EM: urteaga@gmail.com
AF: Department of Mathematical and Computer Sciences, Colorado School of Mines, 1500 Illinois St, Golden, CO 80401, United States
AU: Han, Q
EM: qhan@mines.edu
AF: Department of Mathematical and Computer Sciences, Colorado School of Mines, 1500 Illinois St, Golden, CO 80401, United States
AU: Porta, L
EM: lporta@mines.edu
AF: Division of Environmental Science and Engineering, Colorado School of Mines, 1500 Illinois St, Golden, CO 80401, United States
AU: Jayasumana, A
EM: Anura.Jayasumana@colostate.edu
AF: Department of Electrical and Computer Engineering, Colorado State University, Engineering Room B104 1373 Campus Delivery, Fort Collins, CO 80523, United States
AU: Illangasekare, T
EM: tissa@mines.edu
AF: Division of Environmental Science and Engineering, Colorado School of Mines, 1500 Illinois St, Golden, CO 80401, United States
AB: Groundwater transport modeling is intended to aid in remediation processes by providing prediction of plume location and by helping to bridge data gaps in the typically undersampled subsurface environment. Increased availability of computer resources has made computer-based transport models almost ubiquitous in calculating health risks, determining cleanup strategies, guiding environmental regulatory policy, and in determining culpable parties in lawsuits. Despite their broad use, very few studies exist which verify model correctness or even usefulness, and those that have shown significant discrepancies between predicted and actual results. Better predictions can only be gained from additional and higher quality data, but this is an expensive proposition using current sampling techniques. A promising technology is the use of wireless sensor networks (WSNs) which are comprised of wireless nodes (motes) coupled to in-situ sensors that are capable of measuring hydrological parameters. As the motes are typically battery powered, power consumption is a major concern in routing algorithms. By supplying predictions about the direction and arrival time of the contaminant, the application-driven routing protocol would then become more efficient. A symbiotic relationship then exists between the WSN, which is supplying the data to calibrate the transport model, and the model, which may be supplying predictive information to the WSN for optimum monitoring performance. Many challenges exist before the above can be realized: WSN protocols must mature, as must sensor technology, and inverse models and tools must be developed for integration into the system. As current model calibration, even automatic calibration, still often requires manual tweaking of calibration parameters, implementing this in a real-time closed-loop process may require significant work. Based on insights from a previous proof-of-concept intermediate-scale tank experiment, we are developing the models, tools, and protocols necessary for a closed-loop simulation online, combining work across multiple disciplines. This simulation environment will expedite software development and a large-scale experimental aquifer will be used for further validation of the techniques. The results presented here address: setup of a WSN simulator which cooperates with transport models, development of fault detection techniques into the WSN routing protocol which are particular to this application, and planned steps in building a transport model capable of working in the WSN context.
DE: 1832 Groundwater transport
DE: 1846 Model calibration (3333)
DE: 1848 Monitoring networks
SC: Hydrology [H]
MN: 2007 Fall Meeting