EVALUATING POTENTIAL LOSS OF EXISTING MONITORING WELLS AT HANFORD USING A FIRST ORDER ANALYSIS OF UNCERTAINTY

During defense material production activities at the Hanford site in Southeastern Washington State, large volumes of wastewater were discharged to waste facilities from 1943 – 1987. These subsurface discharges resulted in large water table changes and created significant groundwater mounds (in excess of 20 m) under waste management facilities in the central part of the site. Since the mission at the Hanford Site changed from weapons production to environmental restoration in 1988, the decline in wastewater discharges has caused the groundwater mounds to dissipate. Monitoring the approximately 700 wells at the Hanford Site has become increasingly difficult in recent years due to a falling water table that is causing wells to go dry. Using the first order analysis in UCODE and ~76,000 historical observations of head, an initial implementation of the Hanford Site uncertainty methodology investigates which of the 700 monitoring wells are likely to go dry between now and 2050, and compares the success of model predictions with monitoring well data through the year 2003. Model predictions are compared to ~12,700 observations of head from 1997 – 2003. This analysis included a comparison of simulated head, as well as the confidence and prediction intervals bracketing the central value. Results demonstrated that confidence and prediction intervals are critical to the successful prediction of a dry well status, and that the uncertainty methodology can be used to assist in the development of a long-term strategy for their replacement.

Pacific Northwest National Laboratory is operated for the U. S. Department of Energy under Contract DE-AC06-76RL01830.