“Science is not linear.”
This sentence was spoken by one of the most brilliant and humane scientists I have ever had the privilege of knowing: Paul Baker. And indeed, science is not linear, neither in the way it is produced nor in the trajectories of those who produce it. It is precisely from this idea that I begin to tell the story behind our paper.
Who gets to do science?
I am Allan Sandes, a researcher and university professor from Niterói, Rio de Janeiro, Brazil, and I grew up in a favela. My trajectory was transformed by access to public higher education and by public policies designed to expand opportunities for students from historically marginalized backgrounds.
After completing my undergraduate degree in Geography and a Master’s degree in Marine Geology, I began my PhD in Ocean and Earth Dynamics at Universidade Federal Fluminense, under Professor Cleverson Guizan. Cleverson, one of the most respected marine geologists in Brazil, invited me to join a research collaboration with Duke University, and through Brazil’s Ciência sem Fronteiras programme, I went to carry out part of my PhD at Duke’s Nicholas School of the Environment, under Paul Baker.
During one of our first conversations, Paul asked me where I lived in Rio de Janeiro. Still struggling with English, I answered: “I’m from Niterói. I live in Morro do MIC, a peripheral neighbourhood.” Paul looked surprised. “You’re from a favela?” I said yes. He looked directly into my eyes and said something I have never forgotten:
“Do you realize what is happening in your life? You, a young man from a favela in Rio de Janeiro, are at one of the best universities in the world. Do you realize how remarkable that is?”
At that moment, I began to understand that my presence at Duke was not simply a matter of luck or individual effort. It was also made possible by public policies that opened opportunities for someone from a historically excluded background. That experience made me return to a question that still follows me: who produces science, and who gets to produce it? How many scientific talents remain invisible simply because they were never given the opportunity to enter these spaces?
When the sediment core surprised us
In February 2010, an expedition aboard the R/V Knorr (cruise KNR197-4) recovered sediment core CDH-79 from the Brazilian equatorial continental margin, in the Pará-Maranhão Basin, south of the Amazon Fan. The 32.2-metre-long core came from a seamount at roughly 2,345 metres depth. The project, led by Paul Baker and Cléverson Guizan, aimed to understand the climatic and biogeographic evolution of tropical South America.
At first, there was nothing to suggest how important this core would become. The first surprise came during biostratigraphic analyses: the results suggested an age of approximately two million years, far older than the other cores recovered during the same expedition. I repeated the analyses. The result held up. As additional chronological and oxygen-isotope data became available, the evidence became compelling: we had recovered the longest sedimentary record ever obtained from that stretch of the Brazilian equatorial margin, spanning roughly 1.93 million years of tropical South American climate history.
A sediment core is, in many ways, a natural archive. Layer by layer, material from the continent and signals of environmental change accumulate on the seafloor. CDH-79 gave us access to a part of Amazonian climate history that had previously been largely inaccessible.
A different story about a changing Amazon
One long-standing question in tropical paleoclimatology is how Amazonian hydroclimate changed during glacial periods. A common interpretation has been that glacial conditions meant a drier, more fragmented rainforest. But CDH-79 told a more complicated story.
By combining sedimentary geochemical proxies with chronological data and comparing the record with other climate archives, we found evidence for a progressive, long-term increase in precipitation and runoff over western Amazonia and the tropical Andes during the Quaternary, particularly after approximately one million years ago. The record also revealed strong variability on orbital and millennial timescales: glacial periods were frequently associated with enhanced sediment discharge from the Amazon system and an intensification of extremely humid events linked to climate changes in the North Atlantic.
The Amazon hydroclimate, it turns out, cannot be reduced to a simple wet-versus-dry narrative. Its history is deeply connected to changes far beyond the tropics. Shifts in North Atlantic circulation, for instance, were capable of reorganizing tropical atmospheric systems and fundamentally altering the hydrological regimes of South America. The Amazon of the past was not a static system responding simply to glacial and interglacial cycles; it was part of a highly interconnected climate system whose behaviour changed through time.
Science takes time
The core was collected in 2010. The first hint of its extraordinary age came during my doctoral research. Interpreting the record took years of analyses, comparisons with other archives, and repeated questioning of our own assumptions. Sometimes the most important part of research isn’t finding an unexpected result. It’s deciding not to dismiss it simply because it doesn’t fit what you expected.
At many moments while working on this paper, I felt that it reflected exactly what Paul Baker had told me years before: science is not linear—not in its results, not in its interpretations, and not in the human trajectories that build it.
This study, now published in Communications Earth & Environment, offers a two-million-year perspective on precipitation and runoff changes over western Amazonia and the tropical Andes, contributing to our understanding of how tropical South American climate has responded to large-scale shifts in the Earth system.
For me, though, the story of CDH-79 is inseparable from the story of how I became a scientist.
Coming full circle
Years after that conversation with Paul at Duke, I returned to Brazil and became a professor at the same public university where I earned my undergraduate degree. Today, as a researcher at a Brazilian public university, I also coordinate science education projects in peripheral public schools. The place where I once arrived as a student from a favela became the place from which I now help open that same door for others.
This is why Paul’s question still matters: who gets to do science? Scientific knowledge does not emerge independently of the people and societies that produce it. When public policies expand access to universities and research opportunities, they change more than individual lives. They also change who gets to ask questions and what kinds of knowledge can be produced.
Because in the end, science is also about belonging. About showing that young people from the periphery, from favelas, need not appear only as subjects in social or environmental statistics. They can also produce cutting-edge knowledge, lead scientific expeditions, publish in high-impact international journals, and help rewrite what we know about the climatic history of our planet.
My trajectory and the CDH-79 record share something fundamental: both challenged expectations. My story challenged the idea that scientific careers belong only to those who have historically occupied scientific spaces. The core challenged simplified ideas about how the Amazon responded to past climate change. Both reminded me that what appears unlikely is not necessarily impossible; sometimes, it is simply something we have not yet had the opportunity to see.
Science, definitively, is not linear.