A network of interconnected urban centers, complete with earthen platforms, plazas, and straight roads, has been detected beneath dense vegetation in Ecuador’s Upano Valley using airborne laser scanning. The settlements, according to findings published in Science, date back roughly 2,500 years, making them among the oldest known examples of organized urbanism in the Amazon basin. The discovery forces a rethinking of who lived in the pre-Hispanic Amazon, how they organized their societies, and what the forest itself looked like before European contact.
How lidar exposed a pre-Hispanic road grid in Ecuador’s Upano Valley
Lidar, short for light detection and ranging, works by firing millions of laser pulses from aircraft toward the ground. The pulses pass through gaps in the tree canopy and bounce back, producing detailed three-dimensional maps of the terrain below. In the Upano Valley, this technology revealed what decades of ground-level archaeology could not: a dense network of pre-Hispanic urban centers connected by large straight roads, surrounded by agrarian infrastructure that sustained the population for centuries.
The road system is not random. The straight alignments linking settlement clusters suggest deliberate planning on a regional scale. Earthen platforms and plazas appear at regular intervals, indicating shared architectural conventions across the valley. The agrarian fields flanking these corridors point to managed food production that went well beyond small-scale slash-and-burn farming. One open question is whether the orientation of these roads and fields tracked seasonal flooding patterns. If the road grid doubled as a hydraulic framework, aligning planting cycles with predictable water levels, then correlating lidar-derived field orientations with paleoclimate data from lake-sediment cores in the same valley could test that hypothesis directly. No published study has yet attempted that correlation, but the raw topographic data needed for it already exists in a publicly archived dataset.
Competing timelines and what the primary data actually shows
The peer-reviewed study in Science places the beginning of the Upano Valley urban network at roughly 2,500 years ago, or around 500 BCE. A separate account framed the same sites as having flourished 2,000 years ago, according to The Guardian. The difference is not necessarily a contradiction. The earlier date refers to when the settlements first appeared, while the later figure likely describes a peak period of occupation and expansion. But the gap matters for anyone trying to place these communities alongside contemporaneous civilizations in Mesoamerica or the Andes, because a 500-year difference shifts which cultures were developing in parallel.
The primary Science paper describes what the researchers call “garden urbanism,” a pattern in which dense residential clusters sit within intensively managed agricultural zones rather than behind defensive walls or on hilltops. This form of settlement challenges the long-standing assumption that the Amazon could not support large, stable populations before European arrival. The Upano Valley evidence shows that it did, and that the people living there engineered their environment to do so.
An institutional press release from CNRS, distributed through EurekAlert, summarized the findings as providing “evidence for early urbanism in the Upper Amazon.” That phrasing is careful. The researchers stopped short of calling these sites cities in the modern sense, instead describing a distributed urban pattern where no single center dominated the valley. The distinction is important because it suggests a different model of political organization than the centralized city-states found in parts of Mexico or Peru during the same era.
Gaps in the record and what researchers still cannot answer
Several questions remain open. The lidar survey mapped surface topography with precision, but it cannot tell researchers who these people were, what language they spoke, or how their society was governed. Ground-level excavation in the Upano Valley has been limited, and the dense vegetation that hid the sites for centuries also makes fieldwork slow and expensive. No named author quotes from the research team have been made publicly available beyond the institutional press release, which limits the ability to assess how the scientists themselves interpret the more speculative implications of their data.
Population estimates for the Upano Valley settlements have appeared in media coverage but are absent from the Science abstracts and the Zenodo dataset metadata. Comparisons to contemporaneous Mesoamerican centers, while suggestive, rely on secondary framing rather than primary archaeological evidence. Until excavation produces household-level data, including food remains, tools, and burial patterns, the actual number of people who lived in these settlements will remain an estimate at best.
The practical stakes extend beyond archaeology. The same lidar methods that revealed these sites also expose how modern deforestation and road construction are erasing the physical traces of pre-Hispanic settlement across the Amazon. Every hectare of forest cleared without prior survey risks destroying evidence that has survived for millennia. For researchers, the next step is straightforward: expand lidar coverage to adjacent valleys and begin systematic ground-truthing of the features already mapped. For policymakers in Ecuador and neighboring countries, the findings add a new dimension to debates over land use, because the forest is not just an ecological resource but also an archaeological archive that, once lost, cannot be recovered.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
