- A 1991 hypothesis suggested that deforestation at Centinela in western Ecuador caused the immediate extinction of dozens of plant species believed to exist nowhere else.
- A 2024 reassessment finds that nearly all of these species occur beyond Centinela, indicating that earlier conclusions were shaped by limited sampling rather than true global extinction.
- The case highlights a broader issue in tropical ecology: species may appear rare or endemic simply because they have not yet been widely documented.
- While forest loss remains severe and risks persist, the evidence suggests biodiversity decline often unfolds more gradually, underscoring the need for stronger data to guide conservation decisions.
In 1991, botanists Calaway Dodson and Alwyn Gentry advanced a striking proposition. Surveying a rapidly deforested ridge in western Ecuador, they suggested that dozens of plant species known only from that site—Centinela—had likely vanished with the forest. The idea was later distilled into the “Centinelan extinction hypothesis”: that habitat clearing can trigger the immediate, global extinction of narrowly distributed species.
It was a powerful claim. It gave a concrete example of how biodiversity loss might unfold in tropical forests, where many species appear rare, localized, and poorly documented. It also rested on a deeper uncertainty. In such systems, what has not been recorded is often treated as if it does not exist.
A 2024 reassessment, published in Nature Plants, returns to Centinela using decades of additional collections and records. Drawing on herbarium records, literature, expert input, and targeted field surveys, the authors reconstruct what is known about the site’s flora. Their conclusion is straightforward. Nearly all of the species once thought endemic to Centinela have been found elsewhere. Of 98 putative microendemics, 99% are now known from other locations.
The original concern about forest loss remains valid. The forests of western Ecuador have been heavily cleared, and many species remain under threat. But the mechanism proposed—that clearing a single site could instantaneously eliminate dozens of globally unique plants—does not hold in this case. What looked like extinction was, in large part, a function of incomplete sampling.
This pattern is familiar in tropical botany, where baseline data are sparse. Even today, plant distributions are reconstructed from patchy records, unevenly collected across geography and time. At Centinela, the apparent concentration of endemics reflected where botanists had looked, not necessarily where species occurred. As collections expanded across northwestern South America, the share of species known only from Centinela declined steadily, approaching zero even before the original paper was published.
Other lines of evidence point in the same direction. Species once thought restricted to a narrow ridge are now documented across wide environmental gradients, from lowland forests to Andean slopes, spanning large ranges in temperature, precipitation, and elevation. There is little to suggest restricted dispersal: many species are carried by birds.
Fieldwork reinforces the pattern. Surveys in nearby forests have repeatedly located populations of species previously considered endemic. In some cases, they were found within hours of targeted searching. The difficulty was not just rarity. It was also that few people had looked in the right places.
The study also complicates another assumption embedded in the Centinelan narrative: that the site itself was entirely destroyed. While much of the forest has been converted to pasture and plantations, fragments remain. Small patches, scattered across the landscape, still harbor elements of the original flora. Their presence does not negate the scale of loss, but it weakens the premise of total disappearance.
The evidence changes how Centinela is understood. It is no longer a clear example of instantaneous extinction driven by hyper-local endemism. It is instead a case study in how limited data can produce strong conclusions, and how those conclusions can persist even as underlying knowledge improves.
Similar issues arise elsewhere in tropical ecology. Conservation science has long wrestled with the problem of inference under uncertainty. In biodiversity-rich but poorly surveyed regions, species appear rare because they are rarely observed. Models built on such data can overstate how restricted species really are.
The Centinela reassessment points to a more cautious interpretation. It does not argue that extinctions are overstated in general. Habitat destruction remains a dominant driver of biodiversity decline, particularly in tropical forests. But it does show that claims about immediate, site-specific extinction require stronger empirical grounding than has sometimes been available.
In this respect, the findings align with a broader shift in how biodiversity change is understood. Large-scale syntheses increasingly distinguish between different dimensions of loss. Species richness may decline, but community composition can shift in more complex ways. Local extinctions may be delayed, contingent on population size, dispersal, and landscape structure. The trajectory is rarely as abrupt as early formulations suggested.
This does not reduce the urgency of conservation work. If anything, it sharpens it. The study points to an “extinction debt” in fragmented landscapes: species that persist in small, isolated populations but are likely to disappear without intervention. The risk is not that extinctions have already occurred unnoticed, but that they are unfolding slowly, and may still be prevented.
It also highlights the continued discovery of new species in the region. Recent fieldwork has identified additional taxa, some already described, others awaiting formal classification. This is not evidence of stability. It shows how much remains unknown.
These gaps in knowledge have practical consequences. Conservation decisions depend on knowing what exists, where it occurs, and how it is changing. Without sustained investment in field surveys, herbarium collections, and taxonomic expertise, those questions remain only partially answered. The risk, as the authors note, is that real extinctions proceed undetected—not because they are invisible in principle, but because they fall outside the scope of observation.
Centinela does not overturn the reality of biodiversity loss. It clarifies how it should be understood. The original hypothesis captured a genuine concern about tropical deforestation, but it overstated the immediacy and scale of extinction at a single site. The revised picture is more measured. Species ranges are often larger than first assumed. Losses accumulate over time. Detection lags behind reality.
The distinction matters. Conservation depends not only on urgency, but on accuracy. Exaggerated claims can distort priorities, just as understated ones can delay action. The task is to navigate between them, grounding decisions in evidence that is both rigorous and incomplete.
At Centinela, that evidence now points to persistence rather than disappearance. The forest is diminished, fragmented, and under pressure. But its flora is not gone. What remains is a warning—and a chance to act while some of this system is still intact.
Citations:
- White, D.M., Pitman, N.C.A., Feeley, K.J. et al. (2024) Refuting the hypothesis of Centinelan extinction at its place of origin. Nat. Plants 10, 1627–1634. https://doi.org/10.1038/s41477-024-01832-7
- Keck, F., Peller, T., Alther, R. et al. (2025) The global human impact on biodiversity. Nature 641, 395–400. https://doi.org/10.1038/s41586-025-08752-2
Credits
