Chicken farmers on the Delaware–Maryland–Virginia peninsula along the eastern shore of the United States are introducing between 20 and 50 metric tons of arsenic to the environment annually, and researchers aren’t sure where it’s ending up. At some point, however, this arsenic could become mobilized and contaminate surface and groundwater.

The poultry industry in the region—second in size only to the state of Arkansas—raises some 600 million chickens annually, according to the Department of Agriculture. In the process, these chickens are fed organic arsenic compounds, like roxarsone, to control infections and increase weight gain. Keeping within U.S. Food and Drug Administration limits, little of the roxarsone is retained in the meat. Most of it ends up unchanged in the roughly 1.5 million metric tons of manure excreted annually by the chickens.

Transporting this litter for disposal elsewhere costs money, so farmers tend to stockpile the litter on long rows on fields called windrows, or apply it directly to corn and soybean fields as fertilizer. Of course, the arsenic contained in the manure gets applied too and could be contaminating surface and groundwater, but whether it does and where the arsenic actually ends up is a mystery.

To determine the distribution, occurrence, transport, and fate of arsenic from the poultry feeding operations in this region, U.S. Geological Survey (USGS) researchers have been analyzing arsenic concentrations in soils from agricultural fields where poultry waste has been applied, soils from more pristine environments such as nearby forests, bed sediments from the Pocomoke River, water from the river itself, surface water from ditches throughout the river’s basin, and groundwater throughout the basin.

Dissolved arsenic concentrations in the Pocomoke were not much higher than 1 µg/L during base flow, but during storm events, levels increased to around 4 µg/L, says Tracy Connell Hancock, a USGS hydrologist. Arsenic concentrations in ditch water samples also rose during storm events from a base flow level of 0.5 µg/L to as high as 10 µg/L, indicating a surface source such as manure particles mobilized by storm runoff.

Hancock and colleagues also looked at levels of arsenic in pore water from cored sediments and shallow groundwater from piezometers (wells) collected beside an agricultural field. Here, arsenic concentrations were highest at the surface, fluctuating with depth, also indicating a surface source and possibly some near-surface concentration by evapotranspiration, Hancock says.

Although the arsenic levels observed at each of the monitoring points were higher than background, "the levels we’re seeing aren’t incredibly high," Hancock says. Also, the USGS investigators found mostly inorganic arsenic forms—As(V) and As(III)—rather than roxarsone.

In fact, the only place they’ve found roxarsone is directly in the poultry litter. Fresh litter samples contained between 30 and 50 mg/kg of arsenic, says John Garbarino, a USGS research chemist. Of that, at least 70% is easily mobilized with water, suggesting that roxarsone can readily be mobilized to the environment by either agricultural field irrigation or rainfall on uncovered windrows.

In degradation experiments, the USGS scientists leach the litter with water at room temperature—steadily monitoring the amount of roxarsone collected in the leachate. "After 24 to 48 hours, all of the roxarsone is gone, and it seems to change into a compound we haven’t identified yet," Garbarino says. The rate of transformation seems to be bacteria-related because when the litter extract was sterilized, the roxarsone remained stable for at least 10 days.

Indirect evidence indicates that roxarsone undergoes more extensive degradation after it is introduced into the soil. Arsenic speciation of soil extracts show that amended soils have 5–10 times more water-extractable arsenic than the controls, and that As(V) is the predominant species. In contrast, As(III) was the predominant species in the extract of bed sediments from a ditch adjacent to the agricultural field, making it likely that in anoxic environments, bacteria are promoting the reduction of arsenate to arsenite.

However, a one-time water extraction of the amended soils only mobilized about 20% of the total arsenic applied to the field, according to Garbarino, meaning that the arsenic isn’t leaching from the soils as quickly in one storm event.

But so far, "even though a lot of arsenic is being added to the environment, we can’t account for it all," Garbarino says. "Because we can leach roxarsone from the litter so easily, we know it’s getting mobilized on fields." But how does it get from being roxarsone to As(V) in the soils and As(III) in the sediments, he asks. "Those are all the steps we’re missing.”

“We know that the water-extractable arsenic correlates directly with the amount of litter applied to a field and that the litter is the source of most of the water-extractable arsenic," Garbarino says. "The good news is that so far, it seems that even though we’re applying lots of arsenic, it seems to be leaching from the soils at a gradual rate and not being released in a big slug that would cause big problems.”

But the arsenic is still being stored in the soils, associated with either organic matter or metal oxides or hydroxides. Arsenic extracted with water from poultry litter-amended soil was found to be associated with the organic matter and was primarily in the As(V) form. The greatest percentage of the total arsenic in this soil was sorbed to the metal oxides and hydroxides and could not be mobilized by water alone. Soil in corn or soybean fields is generally dry and oxic, thus promoting an environment where As(V) is favored. However, during unusually wet periods, soil conditions could become anoxic to promote microbial reduction of As(V) to As(III), a much more mobile form of arsenic. Consequently, at some point arsenic, in one form or another, might be mobilized to contaminate surface or ground water, Garbarino says.

Plus, the litter stored in uncovered windrows could be a point source. “Conceivably, if there’s a rain event onto a pile of this, it could mobilize a lot of arsenic quickly,” Garbarino says. “If we get 70% in the lab, you’d think it would be similar in the field.”

Hancock and her team will be monitoring storm events in ditches, river tributaries, and the Pocomoke River itself, looking at the level of both particle-bound and dissolved arsenic. "Arsenic often behaves like phosphorus—it’s sticky and adheres to particles, so it’s possible that it could be getting transported downstream during storm events while stuck on manure particles or sediment particles," she says. —Kris Christen