Farmers in Belize were almost certainly growing maize 6,000 years ago, millennia before Maya villages

People in southern Belize were farming what was almost certainly maize by 6,000 years ago along the edges of wetlands, according to a study led by Keith Prufer of the University of New Mexico and Megan Walsh of Central Washington University. That puts dedicated farmers in the landscape thousands of years before the first large villages emerged, and long before the Classic Maya cities that made the region famous.

The work comes from the Maya Golden Landscape, a region of 35 traditional Maya communities near the ancient city of Uxbenka, where the research behind the study began in the early 2000s and has continued for decades. Its central finding is a story of movement: farmers started in wet ground, left it when conditions changed, and settled uplands that people still plant today.

Carbon isotopes in tropical soil as a record of maize

Maize is a C4 plant, a type of plant whose carbon chemistry differs from the forest vegetation around it, and its signal persists in tropical soils for thousands of years. The team sampled wetland and upland soils and used laboratory chemical and microscopic analyses, including carbon isotopes, to identify and map changes in food production. The results appear in a paper in Communications Earth & Environment, published on Sept. 22, 2026, which the university distributed to journalists on Oct. 8.

Prufer, a professor of anthropology who directs the Human Ecology and Radiocarbon Lab in UNM’s Center for Stable Isotopes, put the central result plainly in the EurekAlert news release: “people were farming what was almost certainly maize by 6,000 years ago” along the edges of wetlands. The qualifier matters, because the evidence is chemical, not a pile of preserved cobs.

The authors also compared the soil record with early climate records to test whether past rainfall changes lay behind farmers’ decisions. Walsh, a professor in Central Washington University’s Department of Geography, co-led the work with Prufer, and nine other researchers and authors, many of them current or former UNM students, are credited with it. The approach draws on archaeology, isotope ecology and paleoecology at once, which is how a signal in dirt becomes a map of who grew what, and where.

From wetland edges to uplands after a dry spell

Farmers first worked the wetland margins, where soil moisture is high. After a dry interval ending about 4,000 years ago, they shifted C4 crops to upland areas away from the wetlands and began establishing settlements that grew into larger communities. Wetland farming appears to have been largely abandoned after local populations peaked more than 2,000 years ago, at which point the upland fields carried the load.

The sequence, then, runs in three steps: wetland-edge maize about 6,000 years ago, an upland move after a dry interval that ended about 4,000 years ago, and wetland abandonment more than 2,000 years ago.

Earlier work at the same ground supports the broad timeline. A 2014 study in Quaternary Research by Walsh, Prufer and colleagues, listed in Central Washington University’s repository, analyzed charcoal, pollen and sediment from a lagoon near Uxbenka and dated the start of farming-related forest clearance to around 2,600 calendar years before present, with fire activity falling as the polity weakened during droughts in the Late and Terminal Classic.

Fields that are still in use

The upland plots tell the longest story. They were farmed through the Maya Classic Period and are still farmed today without loss of productivity, a continuity that bears on modern agriculture as much as on archaeology. The study included Jose Mes, a Mopan Maya-speaking farmer from Santa Cruz village, as a contributor, and the team spoke with local farmers about traditional ecological knowledge while sampling the fields.

That lagoon record described human land alteration in association with swidden agriculture, the slash-and-burn rotation of plots, and it showed fire activity at its lowest in the Postclassic period, roughly 950 to 430 years ago, before rising again as people resettled the region. Read next to the new study, it frames the farming landscape as one with several phases of use, abandonment and return.

Uxbenka itself was a Maya polity dating from roughly 1,900 to 1,200 years ago, according to Prufer’s project page at the University of New Mexico, which also describes ethnographic work with the Mopan Maya community of Santa Cruz, whose communal lands include the site. The new result pushes farming in the district back several thousand years before that city rose.

Prufer also said in the release that the research began in the early 2000s, and that maize is so central to people’s lives today. The study’s reach is the point: farmers changed where they planted when the climate changed, and the places they chose stayed productive for millennia. Whether the earliest maize at wetland edges was a staple or a supplement is a question the isotopes cannot answer, and the reported findings do not claim to. Readers who follow field archaeology can find similar releases in EurekAlert’s archaeology collection.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview