Research
We seek to understand how plant populations respond to chronic human disturbance and to changing biotic and abiotic environmental conditions. Human societies depend on Earth’s ecosystems for food, medicine, non-timber resources, livelihoods, and overall well-being. Studying these human–environmental relationships is therefore essential for projecting the effects of future social-ecological interactions and understanding their biological consequences, and it is a central goal of contemporary ecology and conservation biology.
Our work addresses key conservation challenges in tropical ecosystems, including deforestation, habitat fragmentation, frequent harvest, fire, and flooding, at study sites in the Amazon basin. We focus on the ecological consequences of wild plant harvest, and in particular on how unpredictability and variation in both environmental conditions and harvesting activities shape plant population responses over the short- and long-term. Beyond the Amazon, we are leading a global synthesis of how non-timber forest products recover from chronic harvest across life forms, species, and ecoregions.
In the Ucayali River region of the Peruvian Amazon, we are investigating how ayahuasca (Banisteriopsis caapi) responds to frequent harvest and to varying environmental conditions, including periodic drought and flooding. Our long-term demographic research has shown that heavily harvested populations shift toward clonal growth, but that this compensation may not be enough to prevent population decline. Building on these findings, we are now investigating sustainable harvest limits for ayahuasca populations in restored ecosystems, to help inform novel strategies for producing locally grown, sustainably sourced plant medicine.
Does harvest cause a decline in genetic diversity of ayahuasca lianas by increasing clonal reproduction in response to short- and long-term disturbance?
Ayahuasca is a tropical liana that can reproduce both sexually, through flowers and winged fruits (samaras), and asexually, through clonal growth. Sexual reproduction generally carries greater energetic costs, and disturbance often stimulates clonal growth. Lianas subjected to regular harvest for medicinal use are therefore expected to shift their reproductive investment toward clonality. Our demographic research has confirmed this shift, and has shown that it is not enough to prevent population decline. What remains unknown are its genetic consequences. If harvested populations rely increasingly on clonal reproduction, their genetic diversity may decline over time, potentially reducing population fitness and stability through processes such as inbreeding depression. Using population genomics, we are investigating how short- and long-term harvest, and the reproductive trade-offs it drives, affect the genetic diversity of ayahuasca populations.
What are the short- and long-term effects of harvest on ayahuasca variety populations in restored habitats?
Throughout the Amazon basin, botanists recognize a single species of liana, Banisteriopsis caapi, as the primary source of ayahuasca. Yet Indigenous and local communities distinguish several varieties of the vine, each with its own name in local vernacular taxonomy and with reported differences in morphology and anatomy. Whether these varieties represent forms of a single species, distinct genetic lineages, or separate species altogether is a question taxonomists and ethnobiologists have asked for a century. These varieties are harvested for their bark and combined with other plants to prepare ayahuasca, yet little is known about their population dynamics or how they respond to harvest over the short- and long-term. Without this knowledge, sustainable harvest limits cannot be established. At the Pacha Nishi Project in the Ucayali region of the Peruvian Amazon, we are investigating the systematics and population dynamics of four locally recognized ayahuasca varieties, Cielo, Dulce, Negra, and Sacha, planted in restored ecosystems along drought and flooding gradients.
What are the global medicine security implications of medicinal plant use?
Of the roughly 350,000 plant species classified by botanists, approximately 30,000 are used medicinally, and over 80% of the world’s population relies on ethnomedicine and medicinal plants as a primary source of healthcare. The use and international trade of medicinal plants also generate approximately $130 billion in revenue in global markets. Understanding how and why certain species are selected as medicines, and how they can be used sustainably over the long term, is therefore essential to human societies around the world.
A common misconception is that people historically selected plants and plant parts for medicine at random, developing remedies through trial and error. Theories in ethnobiology suggest otherwise: local people draw on extensive ecological knowledge to identify useful medicinal plants non-randomly. Closely related plant species are more likely to share medicinal uses than distantly related ones, because they often inherit similar chemical defenses from common ancestors, and those defenses shape their effectiveness as medicines. As a result, some plant families are expected to contain a greater proportion of medicinal species than others. Our own research has shown that phylogeny, rather than the availability of therapeutic substitutes, predicts which medicinal species face the greatest use pressure. People are also expected to favor plant parts rich in chemical compounds. Under optimal defense theory, young leaves, which are often rich in chemical defenses such as alkaloids, should be selected more often than older leaves, which tend to rely more on physical defenses.
Drawing on specimens at the Harvard University Herbaria, we are building a verified global database of medicinal plant use and a species-level phylogeny of medicinal plants to test these theories at a global scale with phylogenetic control for the first time. In doing so, we combine tools from ethnobiology, ecology, phylogenetic comparative methods, and mathematical modeling to understand the factors that shape medicinal plant selection, the use of specific plant organs, and the dynamics of local ecological knowledge. The first product of this work, a phylogenetic test of the doctrine of signatures, was published in 2026 with an undergraduate student as first author.
“
We must not let this precious opportunity to learn how to respectfully manage ayahuasca production slip through our fingertips for there is much at stake beyond the revenue streams of commercialization, subsistence, and local livelihoods. We must honor the wisdom of the ages and learn from the symbiotic relationships of the natural world to ensure that the diversity of ways of being and ways of knowing inspired by the interconnection between the human experience and ayahuasca teacher plants remain and continue their dialog in reverence. Therefore, it is critical that we develop a deeper understanding of the ecology and demography of these plants and how they respond to increasing harvest pressures and environmental variability over time.
- Michael Coe