Fate and Transport of Chloride after Produced Water Releases to the Land Surface

HICKS, RANDALL, R.T. Hicks Consultants, Ltd., Albuquerque, New Mexico
JAN HENDRICKX, Department of Earth and Environmental Science, New Mexico Tech, Socorro, New Mexico
GRACIELA RODRIGUEZ, R.T. Hicks Consultants, Ltd., Albuquerque, New Mexico
JIRKA SIMUNEK, University of California, Riverside, California

A modeling and field study identified the conditions under which large releases of produced water (100-10,000 bbls) may cause a threat to ground water quality. The study employed the unsaturated zone model HYDRUS 1D to simulate the movement of chloride from the ground surface to a water table aquifer. We predicted the impact to ground water quality using a simple mixing model, which matched predictions from the robust model MODFLOW/MT3D. We then compared the simulations with the results of a field investigation program. The simulated chloride profile in the unsaturated zone matched sampling data from produced water release sites. Our ground water simulations, which assumed a well adjacent to the release site (no dispersion due to transport), also agreed with our field observations.

Simulation of more than 2000 different release scenarios identified three factors that are most important in determining if a produced water release could cause unacceptable impairment of ground water quality in an adjacent well. In order of decreasing importance, these factors are: the chloride load onto the ground surface, the thickness of the receiving aquifer, and the amount of clay in the unsaturated zone. We developed the term “chloride load” as a description of the release characteristics. The value of chloride load can be calculated if an operator knows the volume of the release, the chloride concentration of the release, and the surface geometry of the spill. One can also calculate the chloride load (grams/meter2) from subsurface sampling data. Borehole data from nearby water wells or site studies provide the values for the thickness of the underlying aquifer and the amount of clay in the unsaturated zone. Climate and the depth to ground water had little effect on predictions of chloride concentrations in ground water. However, climate, depth to ground water and the amount of clay in the unsaturated zone were most important in determining the time required for a surface release to reach ground water.

Simulation experiments of heterogeneous unsaturated zones, which are most common in nature, predicted chloride in ground water would not exceed the commonly-accepted standard of 250 ppm in an adjacent monitoring well in most cases (53%) and would exceed 1000 ppm in only 3% of the 384 large spill volume scenarios tested. We found that very large spills (10,000 barrels) that are distributed over many acres pose no greater threat to ground water quality than a 100-barrel spill captured within a berm because the chloride load is essentially the same for both spill geometries. Brine releases (100 barrels or greater) onto sandy unsaturated zones that overlie aquifers that are 3 meters thick or less will generally cause unacceptable impairment of ground water quality in wells adjacent to the release – especially if the depth to ground water is less than 30 meters.