The dynamics of tropical forests are a focus of research for College of the Atlantic plant sciences professor <a href="/live/profiles/3291-susan-g-letcher">Dr. Susan Letcher</a>.The dynamics of tropical forests are a focus of research for College of the Atlantic plant sciences professor Dr. Susan Letcher.

As part of a group dedicated to understanding the way that Costa Rican forests were able to recreate themselves after major upheavals like clear cutting or forest fires, Letcher helped determine a new way of viewing ecosystem development. The knowledge could prove helpful anywhere that people seek to restore balance to downtrodden lands.

Letcher is a co-author on “Opposing mechanisms affect taxonomic convergence between tree assemblages during tropical forest succession,” published in the scientific journal Ecology Letters. While the paper pinpoints the plant species necessary for ecosystem restoration in a specific forest, the statistical tool that Letcher and her team created is potentially transferable to natural locations across the globe, she said.

“Scientists, policy experts, and conservationists could use this tool in other ecosystems such as sand dunes, northeastern forests, or coral reefs around the world to track which species groups are essential for returning disturbed or exploited ecosystem to their most healthy state,” Letcher said. “Upon having this information researchers can make informed decisions about how to reintroduce and preserve key species and successfully restore once devastated ecosystems.”

The context

The roots of the paper go back to conversations among the research group that Letcher worked with at the University of Connecticut, headed by Dr. Robin Chazdon. The lead author of the study, Dr. Natalia Norden, was conducting postdoctoral research there in 2008 when Letcher finished her doctorate. One major focus of the research group’s discussions was how ecologists might gain a clearer picture of how forests assemble.

Many College of the Atlantic students interested in plant sciences work in the <a href="/greenhouse/" target="_blank" rel="noopener noreferrer">Amos Eno Greenhouse</a>, or simply choose to spend time with the plants the greenhouse hosts.Many College of the Atlantic students interested in plant sciences work in the Amos Eno Greenhouse, or simply choose to spend time with the plants the greenhouse hosts.In the next few years, the research group and a broader group of colleagues made enormous advances in investigating this topic. In 2009, Letcher, Norden, and other forest ecologists and statisticians met at Tirimbina Biological Station in Northeastern Costa Rica and planned a series of research avenues that would take advantage of their decades of data on forest assembly. They drew up outlines for a series of papers, but some of the proposed analyses depended on tools that were not yet available. The statisticians at the meeting, along with colleagues in Taiwan, were then tasked to create a tool to simplify the analysis of the hundreds of species that influence forest assembly, which they did within two years.

The challenge

“The trouble with comparing biodiversity is always that you have different amounts of information about different sites,” Letcher said. “If you have 200 in one and 600 in another you know more about the composition of the larger site, and if these sites are really diverse and have different assemblages of species it’s kind of like comparing apples to oranges.”

“In this particular study, we approached that question by examining which groups of species contributed to convergence or divergence between secondary forests and old growth forests– that is, which species groups make recovering forests more similar to the forests that were there before the disturbance, and which species groups make the new forests less similar,” she said.

This was accomplished by classifying species into four different categories: old growth specialists, second growth specialists (found in young forests), generalists, and species too rare to classify with any confidence. The classification used the novel techniques published in 2011.

<a href="/live/profiles/3291-susan-g-letcher" target="_blank" rel="noopener noreferrer">Dr. Susan Letcher</a> was hired in 2017 as a professor of plant sciences at College of the Atlantic.Dr. Susan Letcher was hired in 2017 as a professor of plant sciences at College of the Atlantic.

Generalists are species that can survive in a variety of environmental conditions, while specialists are those species that live in very specific environmental conditions depending on their tolerance to shade, water, predators, and minerals. Previously, ecologists made the distinction between these groups in an ad hoc manner, based on observation of species in the field. The new mathematical tool used in this paper allows researchers to make a statistical determination, given abundance data from two habitats, of whether a species is a specialist in one habitat or the other, or a generalist. In this paper, the two habitats used were secondary forests — forests growing back after a major disturbance, in this case clear-cutting for cattle pasture — and old-growth forests, the remaining patches of forest that have not been subject to major disturbances for at least 300 years.

“Based on how they specialize in different successional stages we came up with a conceptual explanation for how succession unfolds in these landscapes — how forests reassemble after <a href="https://www.coa.edu/about/bar-harbor-maine/acadia-national-park/">Acadia National Park</a>, which neighbors COA, provides students interested in plant biology and ecology a prime experiential classroom.Acadia National Park, which neighbors COA, provides students interested in plant biology and ecology a prime experiential classroom. Credit: Ella Samuel ’16major disturbances based on these four big groups of species rather than the individual species out the roughly 400 species in our data set,” Letcher said.

Often in ecology assumptions are made that forest dynamics and growth are predictable, but when sites are actually tracked over time contradicting results sometimes arise, Letcher said. With their new statistical tool and decades of monitoring data, Letcher and her colleagues were able to uncover deep patterns in the changes in species composition that unfolded in this specific tropical forest in Costa Rica.

“To me it is so exciting that we can take thousands of trees and decades of research and use these statistical tools to compress that information down into something that is so elegant that makes so much sense,” Letcher said.

The verdict

Letcher and her colleagues ultimately found that generalist species are responsible for helping forests reassemble to their original historical compositions.

Letcher praised her peers that she worked on the study with. Chazdon, a professor of ecology and evolutionary biology at UConn, played an instrumental role, she said, providing over 20 years of data from secondary growth and old growth forests in Costa Rica. Chazon was Letcher’s graduate advisor and professor at UConn.

“She is just a phenomenal person for bringing people and ideas together,” said Letcher.

Chazdon’s field assistants on the project included Jeanette Paniagua, Bernal Paniagua, Marcos Molina, Juan Romero, and Enrique Salicetti.

“Those are the people who did the actual work of going out into the forests everyday, measuring these trees, and tracking them over time. It was a huge collaborative project,” Letcher said.

The study was funded by the Andrew Mellon Foundation, National Science Foundation, University of Connecticut, NASA, and the Blue Moon Foundation.

Opposing mechanisms affect taxonomic convergence between tree assemblages during tropical forest succession can be found here.