Analysis of chloride plume migration based on aquifer characteristics, water well pumping rate, soil, and stratigraphic properties.

Abstract

A study was performed to detemine the fate and transport of chlorides in groundwater on and around a Central Kansas refinery. This was done by sampling monitoring wells over a four year period for chlorides and measuring static water level in these wells to determine groundwater flow. Groundwater modeling was also performed to determine the probable source location(s) for the chlorides noted in groundwater and project changes in chloride concentration and movement expected to occur in the future. In addition, chloride source identification was perfom1ed after the method of Whittemore to correlate and compare the chemical nature of groundwater taken from the monitoring wells with the postulated source regions. Source identification was also used to verify the results found as part of the groundwater modeling. Process water for the refinery is taken from four pumping wells located in the northeast part of the refinery facility and operated continuously. These wells pump between 1325 and 3220 liters/min (350 and 850 gpm) and have been pumping long enough that drawdown from these wells has stabilized. The radius on influence for the four wells, taken as a single system, is estimated at 1.3 km (0.8 mile) on the west side and up to 3.4 km (2 miles) on the east side of the facility. Chlorides located within the radius of influence of these are preferentially drawn to the wells over time. Based on sampling data and a previous study by Whittemore (1997) two primary chloride source regions were identified. One was a shallow source located on the refinery property, consisting of an unlined impoundment that had contained refinery process water until closure and removal in 1994. The other primary source was the Johnson Oil Field, located approximately 3.4 km (2 miles) east of the refinery. The field contains chloride injection wells and formerly had unlined impoundments holding salt water that had been separated from the crude oil. Groundwater modeling for the refinery and surrounding property confirmed these source regions and determined that the chloride plume was moving toward the refinery at the rate of approximately 67 meters (205 feet) per year. Source identification of groundwater taken from selected wells also verified these source regions and confirmed the modeling results. Based on this rate of movement, historic chloride data from wells on and around the facility, and the distance from the source to the pumping wells, it is estimated that the chloride plume from the Johnson Well Field reached the onsite pumping wells approximately ten years ago. Transport time from the shallow onsite source and the pumping wells was estimated at slightly over Se\'e11 years. The modeling results aIso indicate that the groundwater quality from the pumping wells will continue to decline over time without remedial action.