A wireless sensor system for validation of real-time automatic calibration of groundwater transport models

In this paper, we present the use of a wireless sensor network in a lab for subsurface contaminant plume monitoring with the objective of automatic calibration of groundwater transport models. A tank configured to simulate an aquifer was used as a testbed, and a 2D model was created based on the setup. To simulate a contaminant plume, an ion tracer was injected into the tank. Sensor probes capable of detecting the plume were buried inside the tank, and wireless motes used to take readings from the sensors and relay data to a base station. More importantly, a run-time fault detection and diagnosis for abnormal sensor readings is designed and integrated into the data acquisition system. Further, an adaptive data collection technique is integrated that is able to provide evidence about the effectiveness of the groundwater transport model in use. Results from the tracer tests are presented, as well as lessons gained.