High and dry. Researchers erect a tower in Bangladesh to monitor water chemistry.

Charles Harvey

A Silent Killer in Bangladesh Wells

Every day, millions of people in Bangladesh drink poisoned water. Wells all over the country tap into shallow aquifers with high concentrations of arsenic. Now researchers report that they've figured out the cause of this contamination.

About 30 years ago, international aid agencies and the government of Bangladesh started installing wells throughout the country. For good reason: The ponds and rivers where people used to get their water also contained sewage--and deadly pathogens. But in the mid-1990s, other health problems started appearing. Those who drink the well water year after year develop lumps on their hands and feet and greatly increase their risk of cancer, especially lung cancer; epidemiologists say that drinking arsenic-contaminated well water is as bad as smoking.

Not all water in Bangladesh is equally contaminated, so scientists have been working on figuring out why some wells contain more arsenic than others. Hydrogeologist Charles Harvey of the Massachusetts Institute of Technology in Cambridge and his colleagues focused their investigation near one village outside of Dhaka, the country's capital. They drilled wells to measure the depth of the ground water and built a computer model of how water moves underground. Organic carbon in the water is also crucial, because microbes that feed on it release arsenic from the sediment into the water. So the researchers measured concentrations of organic carbon and many other dissolved materials in the aquifer and in rice fields and ponds.

Based on these data, the organic carbon in the aquifer most likely comes from humanmade ponds, the team reports online this week in Nature Geoscience. Villagers in Bangladesh dig ponds when they build a house; they pile up the clay to elevate the house above floodwaters, and the pond is used for growing fish and bathing. Rice paddies, on the other hand, don't seem to leak as much water straight down, Harvey says.

The new model probably won't apply to all of Southeast Asia, says hydrogeochemist Shawn Benner of Boise State University in Idaho; it's an intensive look at one site. But he says the study takes a big step forward in linking groundwater flow to the biogeochemistry that leads to the release of arsenic. Even if the study can only show where it would be safe to sink a well in that one village, that's a big advance, he says.

Still, biogeochemist Andy Meharg of the University of Aberdeen in the United Kingdom is skeptical that surface water like that from the ponds is a major source of carbon. Other studies have found that arsenic is often buried in the sediment along with ancient organic carbon, and Meharg points out that arsenic levels can be high in places with little or no human activity, such as mangroves.