Carl Renshaw, Associate Professor of Earth Sciences at Dartmouth College, is collaborating with Joel Blum, formerly in the Department of Earth Sciences at Dartmouth and now Professor and Chair of the Department of Earth Sciences at the University of Michigan, on studies that examine how arsenic moves through the environment and into drinking water. The work is an extension of an earlier Dartmouth study demonstrating that arsenic in bedrock is the primary source of well-water arsenic contamination in New Hampshire. Renshaw, Blum and their colleagues are building complex geological models that can be used to predict where drinking-water arsenic is likely to be elevated using geochemical information collected in New Hampshire, computer models representing the movement of fluids through rock and theoretical models that describe how arsenic is transported through the environment.

Recent work is focusing on a nearby Superfund field site which is being used to develop and test their model: the former Coakley Landfill in southern New Hampshire that is known to be contaminated by arsenic as well as other organic and inorganic contaminants. Interestingly, this site also sits on top of a geological structure that naturally contains arsenic at high levels. What is not yet clear is whether the arsenic found in groundwater wells near the site has its origins in the landfill, the bedrock below the landfill, or in chemical processes involving a interaction of both sources.

Ironically, some of the same chemical conditions normally considered favorable for maintaining modern landfills may also be those that enhance mobilization of arsenic from bedrock into groundwater. One of the primary goals of the project is to develop a method for distinguishing between arsenic contamination from human activity and arsenic contamination from natural geological deposits. Because New Hampshire contains both natural arsenic deposits and anthropogenic or human-caused sources of arsenic, as well as many landfills situated over potential bedrock sources of arsenic, determining how these sources interact will be important in evaluating arsenic groundwater contamination and in assessing potential remediation strategies.

Buried in bedrock?

Early in the 1990s scientists began investigating clusters of elevated arsenic concentrations in bedrock wells in certain regions of New Hampshire. Former orchards where arsenic-based pesticides had been heavily used in the 1920's through 1940's were suspected sources of contamination; so were landfill sites where chemical wastes had been dumped years earlier. But systematic studies failed to find a link between human activities at these sites and the arsenic in well water.
In 1999, a study by Dartmouth geochemists Joel Blum and Steven Peters (both now at the University of Michigan) and Dartmouth epidemiologist Margaret Karagas found strong evidence that nature is the major contributor of the arsenic found in New Hampshire's bedrock wells. Using samples from 992 New Hampshire households, investigators identified a cluster of wells with unusually high arsenic concentrations in the region around the town of Bow. Using geological data, the researchers found that these clusters corresponded with the presence of a particular type of arsenic-rich rock typically found along the edges of granite formations. This led them to suggest that these formations, called pegmatites, might be the source of high arsenic concentrations in drilled wells in the Bow region and in other areas of the state.



A geological legacy

New Hampshire’s bedrock arsenic is a geological legacy that goes back some 450 million years, when two great continental plates collided along the eastern edge of what is now North America. At that time, the region we now know as New Hampshire was covered by an ancient ocean much like the modern Black Sea. Deep, oxygen-poor, and sulfur-rich, it was the ideal environment for creating arsenic-laced rock. Arsenic, which is naturally present in seawater, is strongly attracted to sulfur. Over time, the seawater arsenic became bound to sulfur in mineral complexes that sank to the sea floor.

As the collision progressed one plate buckled and lifted, forming New England's Appalachian mountains, and the other plunged the sea-bed mineral deposits deep into the earth, where they were transformed under high temperature and pressures into the abundant granites that underlie New Hampshire. The high temperatures and pressures also forced boiling-hot water into rock fractures, forming narrow mineral-rich streaks and veins of rock. Some veins contain a pure form of arsenic that is rarely found in nature. It was these veins, exposed at the Earth’s surface, that led early miners in the state to deposits. Renshaw and Blum believe that arsenic intruded into the region's bedrock in this way, forming scattered deposits throughout New England.

Arsenic and apples

Though natural arsenic deposits are relatively abundant in New England and particularly New Hampshire, there are also several human-caused sources of arsenic in the region. Arsenic was used extensively in apple orchards as a pesticide and also was employed as an embalming agent. It can therefore be found in higher concentrations in the soil of old orchards and graveyards.

A recent study by Renshaw and Dartmouth investigators Christine Wong, Xiahong Feng and Stefan Sturup suggests it stays there, in the topmost layer of soil, for as much as a century. The group compared undisturbed orchard sites in New Hampshire where lead arsenate pesticide was known to have been used to similar sites where lead arsenate had never been used. They found that at contaminated sites, almost all of the arsenic (and almost all the lead as well) remains in the top 20 centimeters of soil in the areas where it was originally applied. This was found to be true even in sites that had been tilled and replanted. The group is now investigating whether more extensive disturbances, such bulldozing old orchard sites for redevelopment, increase weathering and surface runoff and potentially mobilize the arsenic from old pesticides.

More lessons from Bow

Soon after identifying clusters of arsenic-contaminated wells in Bow, New Hampshire, Blum and colleagues began focusing on soils and rock in the region. One mineral found in high amounts at Bow, New Hampshire was arsenopyrite, a mineral composed of both arsenic and iron. Scientists know the proportions of iron and arsenic in this mineral, and the investigators expected to find a similar proportion of these dissolved minerals in well water in the region, reflecting the dissolution of the arsenopyrite into groundwater. But the investigators were surprised to find a very different pattern: as arsenic levels in the groundwater increased, iron levels decreased.

The researchers proposed a chemical model to describe these observations. They knew that when arsenopyrite dissolves, arsenic - or more precisely the arsenic ion (a 'cation' or positively charged particle) - combines with hydrogen and oxygen to create a negatively charged 'anion.' The iron, in turn, combines with oxygen and hydrogen to create a solid known as an iron oxy-hydroxide – a substance more commonly known as rust. The researchers noted that the groundwater samples at the Bow site with higher arsenic concentrations also had a higher pH. The team hypothesized that the pH of the water could affect this reaction by changing the surface charge of the iron oxy-hydroxide particle.

When pH is low, iron oxy-hydroxide particles are positively charged and are therefore attracted to the negatively charged arsenic anion. So the arsenic remains bound to the iron oxy-hydroxide rather than dissolving into the groundwater. However, when pH levels become high, the charge of the iron oxy-hydroxide particle becomes negative. Since two negative charges naturally repel each other the arsenic is repelled from the iron oxy-hydroxide, leaving it free to dissolve into groundwater.

The Bow study provided evidence that high pH could mobilize arsenic by forcing it to separate from its partner iron oxy-hydroxide particle.

The Coakley Landfill

In 2000 Carl Renshaw and graduate student Jamie deLemos began to examine another site in which the source of groundwater arsenic was open to question: the Coakley Landfill. Located in southern New Hampshire on the outskirts of Pease International Air Force Base, Coakley had been found to be a source of volatile organic contaminants (VOCs) that had seeped out of the old landfill. For this reason the United States Environmental Protection Agency placed Coakley on the National Priorities List in 1983, making it an official Superfund site. The site also exhibited high levels of arsenic contamination. In 1998 the landfill was capped, reducing seepage. Natural processes largely reduced the organic contaminants to levels considered safe.

However, groundwater from wells adjacent to the landfill continue to have high levels of arsenic, which was assumed to be originating from waste deposited at the old landfill. In fact, since the landfill was capped, the arsenic levels in some wells have increased tenfold - many times the level considered safe for drinking by the US Environmental Protection Agency.

Superfund legislation requires polluters to pay for cleanup of contaminated sites, and if the arsenic in local wells could be attributed to dumping at the landfill, the polluter could be required to pay for remediation. But what if the landfill was not the primary source? What if the arsenic originated in New Hampshire’s unique bedrock?

Capping and landfill chemistry

Renshaw and deLemos are trying to answer these questions and finding that, much like the geochemical model at the Bow site, the answer is not straightforward. The presence of pegmatite veins in the bedrock and the lack of any known arsenic dumping into the landfill suggested that bedrock could be the source of well-water arsenic in the region. If the arsenic was being mobilized from bedrock by the mechanism observed in Bow, which the investigators originally suspected, then the same inverse relationship between iron and arsenic should be present at the Coakley site. But no such inverse relationship has been found, and the mechanism of mobilization appears to be much more complicated.

The researchers knew that before the landfill was capped in 1998, substantially more water was able to flow through the landfill, essentially flushing the site of some of its arsenic and other metal contamination. When Renshaw and deLemos started studying the site in 2000, long after the pile was capped, they immediately noticed that arsenic levels in wells near the site were on the rise. The researchers suspected that arsenic already in the ground was being mobilized by some undetermined chemical reaction.

Renshaw believes the primary source of arsenic found in well water in the region may be the bedrock or overlying sediments; however, refuse dumped into the landfill may be playing a supporting role in releasing the bedrock arsenic into groundwater. There is evidence from studies in Bangladesh that, in addition to iron, the presence of organic material can greatly influence the mobilization of arsenic from geological sources. Ironically, capping the landfill - a common practice designed to reduce the spread of landfill contaminants to outlying areas - may have exacerbated the arsenic problem by decreasing both the amount and velocity of the water flowing through the landfill as well as the pH and the amount of oxygen in the groundwater. The slower water velocity allows chemical processes more time to strip the water of oxygen, resulting in highly "reduced" rather than oxygenated groundwater. The more reduced groundwater, in turn, frees more arsenic from the bedrock, enabling it to dissolve into groundwater.

Building a model

Renshaw and colleagues are currently building a groundwater flow and transport model for the Coakley site that will describe the rate at which water and arsenic flow through the site. By analyzing water, rock, and sediment samples collected at the site, and comparing these analyses to the model predictions, the team hopes to determine why arsenic levels are rising. Although it may be impossible to determine the exact source of all the arsenic, the study will explain the conditions favoring its release at the Coakley site and the possible impact of the capping on the arsenic contamination.

The model may prove valuable to environmental engineers looking to remediate other waste sites where capping is being considered as part of the remediation plan, stressing the types of conditions that could lead to arsenic release and accumulation.