Research - Geochemical Modeling & Natural Attenuation

Background

The Ashumet Valley (AV) Plume is a large region of contaminated groundwater that extends more than 3 miles downstream of an effluent discharge from a wastewater treatment facility.  This site is located adjacent to the Massachusetts Military Reservation, as shown in Figure 1.  The wastewater disposal facility at this site maintained a discharge into the subsurface via effluent disposal beds from 1936 until 1995, at which time the discharge was relocated to another source.  Even though the source was discontinued, the highly contaminated region has remained downstream of this facility and the loads of phosphorous to the effected pond have continued.  Large portions of this contaminated region are even anoxic (i.e. with no oxygen present), which is important since oxygen is a critical water quality parameter and has an important effect on aquifer restoration and chemical conditions in the environment. Understanding these processes are critical for quantifying rates of restoration of contaminated groundwater, which is critical to our ability to protect our water supplies. 

Goals

The goal of this research is to understand the biogeochemical processes that are constraining the restoration. As part of this effort, Professor Mathisen has been collaborating with an interdisciplinary team of hydrologists, geochemists and microbiologists from the United States Geological Survey (USGS) to investigate the processes governing the restoration of sewage-contaminated groundwater after the removal of the contamination source.  This work has included field experiments and geochemical modeling. 

Field Experiment

In fall 2001, an experiment was completed at the Cape Cod field sitee. A steady injection of uncontaminated groundwater (with oxygen) was discharged into a contaminated zone of anoxic groundwater (with no dissolved oxygen) directly below the former sewage-effluent disposal bed.  By sampling from an array of 12 multi-level sampler wells, the dissolved oxygen and variety of other water quality parameters were monitored in the groundwater downstream of the injection.  Figure 3 includes a picture illustrating some of the researchers collecting groundwater samples for laboratory analysis.  The injection equipment is housed in the brown shed and the sampling array is shown in the foreground (the researchers are collecting samples from the first well downstream of the injection.)

 

Text Box: Figure 1 – Ashumet Valley Plume (from USGS) Text Box: Figure 2 – Aerial View of disposal beds and Ashumet Pond (From USGS) Text Box: Figure 3 – Sampling at the field site Text Box: Figure 4 – Observed and simulated dissolved oxygen concentrations

Geochemical Modeling

Using the laboratory and field results, a reactive transport model was developed that could characterize the flow, transport, and chemical interactions between the constituents in the water and on the soil surfaces.  This model was developing using the PHREEQC modeling package (Parkhurst and Apelo, 1999).  The simulations were used to show that the restoration of oxygen in this anoxic region is governed by biodegradation processes involving oxidation reactions. In these oxidation processes, bacteria consume oxygen and other constituents released from the sediments (e.g. organic carbon and ammonium).  Figure 4 shows dissolved oxygen (DO) concentrations in milligrams per liter (mg/l) measured 5 meters from the injection, along with the times affected by the startup and shutdown.  The injection DO (which ranges from 8 to 9 mg/l) exceeds the measured DO in Figure 4 by more than 1.3 mg/l throughout most of the test.  Therefore, the results show that the DO is being consumed. The solid green line in this figure represents the simulated DO obtained when a model is used to represent these oxidation processes.  The version of the model used for this simulation made use of simple “first-order” model formulations to represent some of the key processes.  However, results from more advanced simulations are being finalized and will be presented in a journal publication.  For the purposes of this summary, it is evident that the simulated DO closely matches the measured DO at this location, demonstrating the success of this type of model in characterizing these processes.   By accounting for these processes and properly representing them using the model, we are able to accurately simulate the concentration of oxygen observed in this test.

 

Other sites of interest:

Cape Cod Toxics Research Site