Published online 18 May 2009 | Nature | doi:10.1038/news.2009.488


Mercury traces expose Inca mining boom

Peruvian lake sediments reveal extensive activity of past civilizations at 'mine of death'.

Andes mountainsLake pollution holds clues to the long history of mercury mining in the Andes.C. Cooke

Evidence of massive mining of mercury ores in South America by the Spanish, Incas and others before them has been extracted from the bottom of Peruvian lakes.

Mines in the Huancavelica region of present-day Peru have long yielded cinnabar ore, or mercury sulphide. The region's Santa Barbara mine was known as the mina de la muerte (mine of death) by the Spanish because of the number of people who died working in it, many of them from mercury poisoning. Their massive mining of cinnabar ore for its mercury, which was used in silver production, also destroyed any traces of mining by previous civilizations that controlled the region.

However, from its use to make bright red body paint, archaeologists know that Andean societies from the Chavín through to the Incas mined cinnabar. Now they have some idea of just how much.

Colin Cooke of the University of Alberta in Canada and his colleagues extracted sediment cores from three lakes around the Santa Barbara mine. By analysing the history of mercury pollution in these lakes, which are different distances from the mine, they were able to estimate the levels of mercury pollution released by mining and how far this taint spread.

“The Incas are the big show, the Spanish are second to them but not by much.”

Colin Cooke
University of Alberta

At the two lakes closest to the Santa Barbara mine, "dramatic increases" in mercury are seen dating back to about 1400 BC. "The onset of cinnabar mining at Huancavelica ca. 1400 BC places our lake-sediment records among the earliest evidence for mining and metallurgy in the Andes," the researchers write in Proceedings of the National Academy of Sciences1.

By the time the Chavín are approaching the height of their empire in 600 BC, levels are ten times higher than background. After this, mercury levels decrease until around AD 1200 in one lake and AD 1450 in the other, then begin to increase again as the Incas take over, reaching more than 50 times background by AD 1550.

"Initially when mercury mining starts in 1400 BC it's restricted to the area around the mercury mine," says Cooke. "It's certainly clear once the Incas move into the region it goes up almost tenfold. The Incas are the big show, the Spanish are second to them but not by much."

A tale of three lakes

At the third lake, which is significantly further from the mine than the other two, mercury levels only start to rise around AD 1400. Mercury pollution caused by the Chavín seems to be limited to cinnabar dust, but to reach the 225 kilometres to the third lake at least some of the pollution generated later on must have been gaseous mercury, Cooke and his co-workers suggest.

The huge increase in mercury levels, and the fact that it has spread further, leads Cooke and his co-workers to suggest that the Incas must have been not just grinding up the cinnabar to make dye but also smelting it to extract the mercury.

This conclusion is likely to be controversial, as the smelting of ore to produce mercury for silver processing is currently associated with the colonial rule of the Spanish.


The study also shows the value of using lake cores as windows onto past mining activities, says Richard Burger, an anthropologist at Yale University.

"These mines were so heavily used during colonial times that any evidence of the early mining technology had been eliminated," he says. "When I read [the paper] I thought 'this is so elegant'. People have used lake cores widely in order to look at vegetation change or as a proxy for climate change. This is the first time I've seen people use it to look at mining."

Cooke agrees that sediment cores could have a wider role in archaeology than has so far been realized. In some regions, he says, "it may be the only archive that's left". 

  • References

    1. Cooke, C., Balcom, P. H., Biester, H. & Wolfe, A. P. Proc. Natl Acad. Sci. USA doi:10.1073/pnas.0900517106 (2009).
Commenting is now closed.