Overview

The LCP Chemicals Superfund is an area encompassing a variety of past industrial chemical sites active for most of this century. Manufacturing activity included bleach production, paint production, and oil refining. These practices left the ground and marsh soils heavily polluted with complex mixtures of toxic chemicals. Cleanup has been underway for several years and the contaminated upland (dry soil) regions are removed or stabilized. Marsh soils are the subject of clean up efforts.

Extensive groundwater plume analysis was conducted at the LCP Chemicals site in Glynn County, Georgia. These studies show several small plumes and pools begin beneath areas of past industrial activity and then flow toward the Turtle River. With the excavation, treatment, and redisposal of soil in the upland areas there are now no point sources of contamination feeding these plumes. Over time, the pollution should run down into the areas under the marsh. The driving forces moving the pollution are groundwater draining from rain onshore pushing the pollution to the southwest, and tide and river water pushing the pollution to the northeast. The pollution movement will eventually reach an "equilibrium" where no further movement can occur.

Some chemicals will elute from the ground and disperse in the tidal basin. Others will be trapped underground. Some chemicals, notably the brine solutions, may be left in the ground for thousands of years.

It is not yet clear if any attempt will be made to affect an underground cleanup. Since the pollution is trapped underground in an area where neither marsh biota nor humans can be exposed, it may be perfectly safe to leave the material in place.

The Glynn Environmental Coalition (GEC) received for review the Groundwater RI Study. This comprehensive document was examined for the GEC by Denahan and Associates. The report is excerpted here.

Three aquifer systems have been described at the site: the surficial, the shallow rock and the Floridan. A significant effort has been made to characterize the site geologically, hydrogeologically, and chemically. Most of the wells sample the upper, middle and lower portions of the surficial aquifer. This aquifer is reportedly not used for drinking. "Potable water wells tested by the USEPA showed no indication of contamination." Six deeper wells were installed to sample the shallow rock aquifer, which is reportedly used locally for domestic and light industrial water supplies. Also water samples were obtained from the Floridan Aquifer and analyzed for site contaminants

The surficial aquifer is determined to have two to three flow zones with low permeability materials sometimes present. One site contaminant, caustic brine, has a high density and a tendency to move as a unit. Therefore, the bottom flow zones in the surficial aquifer were mapped to determine the subsurface structural control of the caustic brine movement.

Ground water in the upper portion of the surficial aquifer flows from east to west across the site toward the marsh. The discharge feature for the surficial aquifer is Purvis Creek with Turtle River providing a regional ground water discharge boundary.

Ground water in the surficial aquifer at the site contains fresh and brackish water. Fresh water flows across the site from east to west and brackish water enters the aquifer from Purvis creek and the Turtle River.

Monitor wells at the site detect a tidal response. Tidal influences were considered in assessing water tables. The consultant used mathematical averaging on the tidal influenced data to assess long term trends. This in effect muted the short-term tidal fluctuations leaving other events in the record such as precipitation events, and "stages of the monthly tide cycle".

Reportedly, comparison of rainfall occurrence and water level fluctuations in the surficial aquifer show that the water levels in the surficial aquifer respond to rainfall events almost immediately.

Organic compounds, petroleum and solvents, are distributed widely over the site. Of the Volatile Organic compounds only the Benzene and Ethyl Benzene plumes are not contained within the monitor well array.

Naphthalene is found downgradient of the former naphtha plant; 1,4 Dichlorobenzene is found downgradient of the former distillation area; Pentachlorophenol is in the central portion of the site.

Natural attenuation studies were performed for the ground water. The studies indicate that natural attenuation is occurring in the subsurface with a reported 98% reduction in concentration of all petroleum organics along the examined transect.

Polychlorinated Biphenyls (PCBs) and pesticides are defined to within the site limits. None of the deeper wells were reported to contain PCB’s.

Nine pesticides were reported at the site. The RI states "there is no systematic distribution of pesticides and they are not of concern regarding the ground water due to extremely low concentrations of detected compounds, and low frequency of detection." One deeper well was reported to contain a pesticide, beta-BHC.

Inorganic analytes or metals are distributed widely over the site in the ground water. Inorganic compounds discussed in the RI are: antimony, arsenic, barium, beryllium, chromium, lead, mercury, nickel, selenium, and thallium. The industrial operations are directly related to the sources of mercury, barium and lead. The other metals-- antimony, arsenic, beryllium, chromium, nickel, selenium, and thallium, reportedly may have other sources or may not be a result of past activities. Reportedly, most of the other metals are naturally occurring. Many of the other metals are also found in paint and as impurities in unrefined petroleum and lead.

Most of the metals found in the groundwater at the site can be attributed to leaching of naturally occurring minerals by alkaline or acidic releases. Three areas at the site are defined in the RI as areas of "highly alkaline releases"; the cell building and caustic storage area; the brine mud impoundments; and bleach muds overlying acidic petroleum disposal sludge. Of the metals found in the ground water at the site, the RI reports barium mercury and nickel in the deeper wells. Detection of the metals is correlated with elevated pH in ground water.

Mercury distribution in groundwater is reported to match that of arsenic and chromium. "Elemental liquid mercury was observed in site soils within the surficial aquifer below an area of the former cell building". Elemental mercury is confined to the shallow soils due to its difficulty migrating through the small pore spaces in the subsurface. In ground water mercury exhibits sharp concentration gradients. The RI reports, "the extent of mercury in ground water is well bounded within the property by trace levels or non-detects". The RI states, "the trace levels of mercury are most likely attributable to contamination of the samples by airborne mercury".

The Caustic Brine Pool (CBP)

The Caustic Brine Pool (CBP) is beneath the former cell building. The CBP has distinct boundaries and does not form a plume. The RI reports the CBP pool has two driving forces-- the hydraulic gradient and gravity. Chemical reactions reportedly occur along the western boundary of the brine pool as the materials within the CBP encounter the groundwater, dropping the pH, causing precipitation and adsorption, resulting in a reduction of porosity in the aquifer materials.

The reduction of porosity in the aquifer materials reportedly should "form a trapping mechanism that inhibits the downgradient movement of the CBP and forms the sharp western edge". The CBP has a "high density causing it to sink (gravity) until it becomes trapped against a confining unit". The brine will move down slope on the confining unit pooling in low areas. Where confining materials are absent the CBP will "spill" over the edge and sink to the next lower confining unit. The upgradient "hydrodynamic driving force counteracts gravity spreading of the CBP". Due to the density of the CBP, and the reduced porosity around the margins, ground water tends to flow around the CBP.

The CBP was formed when purified brine and caustic was released into the environment. Brine from onsite releases does not contain potassium. The natural salt and brackish waters of the site can be differentiated from the released brine by the presence of potassium

The RI report describes the CBP as defined in its lateral and vertical extent. Reportedly the CBP has been present for at least 20 years and has reached a state of static equilibrium. Based on physical properties of the CBP it is said to be physically stable and the most significant driving force, caustic brine addition, has ceased. The RI states " the CBP will tend toward neutral background conditions with time, as neutral pH waters from recharge and upgradient sources mix".

Reportedly, the elevated pH in the CBP is the most significant factor in mobilizing the metals. The only mechanism for dissolved contaminants to exit the CBP is diffusion at the boundary, a very slow process. When the mobilized metals exit the CBP the pH drops and the metals become locked within a silica gel or get precipitated as insoluble solid or liquid phases.

The study was generally thorough. It is apparent that a lot of thought was given to the activities planned and the data required to determine source removal and future potential remediation planning.

The presentation of water quality data from various sampling episodes, and/or selected water quality results, due to "data quality objectives" for a particular data set presents only one map with contaminant concentrations unrelated to each other in time. This approach is inherently problematic in that any site treated in such fashion will have only one map and the contaminant concentrations will look static rather than dynamic as occurs in natural systems. While reviewers and responsible parties want to make decisions based on the best available data, the commonly practiced approach of developing ground water and contaminant concentration maps based on time equivalent data provides several views of contaminants through time, showing the dynamics of plume movement. This practice also provides real data sets to calibrate transport models and better determine the need for remedial measures. Separate concentration maps for each sampling event should also have been provided in the RI.

The ground water model was only calibrated to the steady state equilibrium scenario. In ground water modeling this is the most easily achieved calibration. Once a ground water model results in a good fit to the observed site water levels and gradients the model should them be further calibrated to known transient conditions to determine if the model will duplicate observed transient responses. If a ground water model is to be used to evaluate pump and treat remedial options it should be calibrated to accurately depict aquifer response to pumping.

The hydraulic conductivity values used in the model were predominately determined based on lithology and visual examination. D&A thinks that more aquifer testing should have been done to obtain the data required by the model or actual measured data should by been used and or cited in the RI. Also, even though the ground water model was a 3-dimensional model with 10 layers there was no discussion of vertical permeabilities versus horizontal permeabilities and whether or how vertical permeabilities were considered in the model. The model should have been calibrated to aquifer response to transient stresses, such as on-site pumping tests.

The presentation of vertical gradient data should have been more complete.

In D&A’s opinion the fate and transport study was very thorough in its assessment of the mobility of the contaminants and the conclusion that natural attenuation of the contaminants will prevent further significant migration is well supported.

Written by R. Kevin Pegg, Ph.D.; edited by Dr. Mary S. Saunders. Copies of the newsletter are available from the GEC, at the Glynn County library, or at www.NucleicAssays.com/tags on the Internet.