The gears of the forest

The loss of more than 60% of native vegetation within a region leads to profound changes in how it operates

Deforestation creates border zones where effects such as heat and low moisture penetrate

Jos Barlow / University Of Lancaster

When a forest gives way to an open area or human uses, much more is lost than just the deforested area. So-called edge effects, which are largely determined by changes in sunlight and moisture, penetrate between 200 and 400 meters into the forest, altering the composition of animal and plant species as well as ecological function. A study published in November in the journal Nature by the ecologists Marion Pfeifer of the University of Newcastle and Robert Ewers of Imperial College London, both in the United Kingdom, indicates that 85% of species are affected by the edge effect. “The forest transforms,” emphasizes Brazilian biologist Cristina Banks-Leite, professor at Imperial College London and a coauthor of the study. “We still cannot say the extent to which the community can adapt to the new way the forest functions.” To understand this changing dynamic, researchers scrutinize what affects the ecological functioning of forests and what are known as ecosystem services, namely how the forests contribute to human activities (for example, by providing pollinators for crops).

One difficulty in measuring edge effects is that deforestation often does not follow regular lines, forming geometric islands of forest. “The Nature study is an important step in developing a technique that can measure the effect in irregular spots,” says Banks-Leite. The differential comes from study author and French mathematician Véronique Lefebvre of Imperial College, who developed the variables I (edge influence) and S (edge sensitivity) to quantitatively assess the configuration of the landscape (such as the degree of coverage offered by the treetops) and how animal species respond and move through the environment. “The method incorporates two aspects which were previously not accounted for in quantitative analyses: the contrast between habitats, and multiple edges—not just the nearest edge,” says Pfeifer.

The analysis of 1,673 vertebrates in 22 forests in the Americas, Africa, Australia, and Asia indicated that half of the species suffer from deforestation (and risk extinction) while the other half expand—invading plants and animals that do well in altered environments. At first glance, it might seem all the same, but the problem is that generalist species do not necessarily contribute to forest operations.

José Carlos Morante-Filho / UESC The scaled antbird (Drymophyla squammata) specializes in capturing insects in the understoryJosé Carlos Morante-Filho / UESC

Not all animals have the same sensitivity to edge effects. Amphibians, for example, risk drying out if they stray from the deep forest, especially if they are small. But snakes and lizards with their long bodies can bake in the sun if they are large. Among the mammals, bats seem to pass over less favorable areas and move on to others, while large land species are seen to be limited to the space required to obtain resources.

This limitation transcends the very survival of the animals, suggests an article coordinated by researchers from the Senckenberg Center for Research in Biodiversity and Climate in Germany which was published in January in Science. With contributions from collaborators in various countries including Brazil, the study evaluated 57 mammal species and found that in the areas most affected by human activities, animals move at half to one-third of their usual rate, mainly due to changes in individual behavior. Not only does this change hamper the animals’ ability to acquire food and other resources, it also affects the nutrient cycle in the ecosystem and seed dispersal, among other effects.

In Brazil, the study included several species monitored in the Pantanal wetlands, one of the analyzed regions with the lowest rate of manmade changes, explains biologist Nina Attias, who contributed with data obtained during her doctorate at the Federal University of Mato Grosso do Sul which involved the Brazilian Agricultural Research Corporation (EMBRAPA) Pantanal, on the movements of armadillos. “Armadillos and other mammals monitored in the Pantanal serve as a comparison to evaluate how animals move in areas with varied levels of human alteration.” On a more local scale in the Pantanal, areas with different levels of preservation change how animals use the space, says Attias, who today is a researcher at the Institute for Wildlife Conservation in Campo Grande, Mato Grosso do Sul.

Léo Ramos Chaves The forest provides a home for native bees, which also pollinate nearby plantationsLéo Ramos Chaves

According to Cristina Banks-Leite, the problem is that it is impossible to predict the sensitivity of a species based on its characteristics. Her doctoral student Jack Hatfield examined data from two studies conducted in the Atlantic Forest in São Paulo. In one, Banks-Leite used nets to capture birds in 65 study areas between 2001 and 2007. In the other, biologist Pedro Develey, who currently serves as executive director of the non-governmental organization BirdLife/SAVE Brasil, observed and documented birds in 32 locations between 2000 and 2003. Together, they included 196 species of birds, which Hatfield classified according to 25 morphological and behavioral characteristics. He next created a ranking of sensitivity to deforestation for each of the studies, and investigated which variables had a greater influence, as he described in an article published in January in the journal Ecological Applications. “The ranking is the same for both sets of data, indicating that the data collection mode has no effect,” says Banks-Leite.

But when statistically analyzing which variables affect this ranking, the results are almost random. The best indicators were the number of habitats frequented by the species, its ability to use open environments, and the habit of following streams of army ants on the move and eating the insects displaced by the large swarm. But none of these characteristics allow a coherent and consistent estimate of how birds behave when forests change. “Within one same species there are individuals that are more or less sensitive,” notes Banks-Leite. She believes that the plasticity of each species in adapting to environments, which allows the animals to act differently in different situations, may be more predictive. “The same function may be performed by very different organisms: a bird disappears and is replaced by a spider, for example.” But to understand these processes, knowledge about different organisms must be brought together, a rare activity.

This is exactly what is being done by a group led by biologist Deborah Faria, also a coauthor of the Nature study and coordinator of the Laboratory for Ecology Applied in Conservation at the State University of Santa Cruz (UESC) in Ilheus, Bahia. Through landscape-scale analyses of areas of remaining forest on farms over the five years of the project, she and her colleagues have characterized how deforestation produces changes in the Atlantic Forest in this region of southern Bahia. “We show that deforestation leads to degradation of the physical structure, and changes in the patterns of composition, abundance of species, and ecological processes in the remaining forest.”

Fragments found in landscapes with fewer forests, for example, produce less fruit compared to more forested areas, according to a 2017 article published in the journal Biotropica. This project was part of the doctorate work by biologist Michaele Pessoa, who under the guidance of biologist Eliana Pasqual collected and measured the fruit produced in 100 100-square-meter plots over one year, studying a total of 1 hectare. The loss of forest affected the variety and abundance of fleshy fruit, mostly because shade trees which are specialized to live deep within forests were seen to be sensitive to changes in the environment. In areas where less than 30% of the original vegetation remained, almost 60% of the trees were typical of sunny zones, more than twice the rate which is typical in the forest.

In this context, the deforested areas also have fewer birds that feed on fruits and insects, as shown in the work of the biologist José Carlos Morante-Filho, currently a professor at UESC. This study was published in 2015 in PLOS ONE, as part of his doctoral work guided by Faria. “The quantity and diversity of birds do not decrease, but there are different species,” says Faria. The forest birds abruptly become rarer and less diverse from a functional point of view when less than half of the original vegetation in a given area remains (which was measured within a 2-kilometer radius from the center of each of the sampling areas). Changes in the composition of bird fauna, which play a prominent role in transporting seeds, are part of a transformation in how these forests function.

Trees in thinned forests also become abruptly impoverished, as shown in an article by biologist Maíra Benchimol, a professor at UESC, published in 2017 in the journal Biological Conservation. Her group examined communities of mature and young trees and showed at least 35% of the forest is necessary to ensure diversity in the young trees. “When we look at a mature tree, we see the past: 90 years ago, perhaps, when it was established,” says Faria. She explains that the adult trees are not being restored in degraded areas, producing a forest composed of trees with less dense wood, which stores less carbon. In her assessment, some species such as the jequitibá-da-bahia (Cariana spp.) do not have sufficient individuals to promote recovery. “Only adults remain, they are almost living fossils.”

José Carlos Morante-Filho / UESC Pollinator: the hummingbird Thalurania glaucopis only lives in forestsJosé Carlos Morante-Filho / UESC

The threshold from which studies indicate that the composition of birds and trees changes significantly draws special attention, because the area of preservation established by law for private land by the new Forest Code (which was revised in 2012) is only 20%. “Deforestation exceeding 60% leads to a change in the regime of forests, which are now degraded and secondarized, causing losses or reduction in the ability of these forests to provide ecosystem services,” warns Faria. She is working on a summary of the project containing the conclusions in terms of ecological effects of deforestation. “One day we will have a set of long-term data for the Atlantic Forest, like the PDBFF for Amazonia,” she adds, referring to the Biological Dynamics of Forest Fragments Project, an experiment which has been running for 38 years in central Amazonia, 80 kilometers north of Manaus.

Today it is clear that the ecological changes observed in fragmented forests through the Amazonian experiment are the result of interactions between local effects (such as deforestation and hunting) and changes on other scales—even on the global scale, according to a commentary published this month in the journal Biological Reviews. It is the case of an acceleration in plant productivity as a response to increased carbon dioxide (CO2) in the atmosphere. “CO2 can fertilize the plants by increasing their growth, but it gives the dominant trees and vines an opportunity to suffocate some of the unique plant species in the Amazon,” explains biologist William Laurance of James Cook University in Australia.

He says that in the 1980s, it was surprising to see how quickly the border effects had an impact. There were not only losses: there are winners and losers. “The fragments are hyperdynamic, and many processes, like population changes in species and nutrient cycles, are accelerated.” In accordance with ecological function, the experiment indicates that the dispersion of larger seeds is reduced in the fragments by the disappearance of large animals. Added to ever more frequent intentional fires in the region and pressure from hunting in the Amazon region, the loss of forests has devastating effects.

Larissa Rocha-Santos / UESC Andira anthelmia also survives in pasturesLarissa Rocha-Santos / UESC

Where there are human interests involved, the ecological processes become ecosystem services. “I understand these ecosystem services as an inverted edge effect, in other words, the effect of the forest on areas for human use,” explains ecologist Jean Paul Metzger of the Institute of Biosciences at the University of São Paulo (IB-USP). His group investigated the benefits forest offer to coffee plantations in the border region between São Paulo and Minas Gerais (known as Mogiana and Sul de Minas), which produces a quarter of the country’s coffee. They found positive effects from this relationship, such as those promoted by birds and bats. By consuming pests, these animals reduce the loss of leaves in the coffee plants and increase the quantity of fruit, according to an article published in 2017 in the journal Landscape Ecology. The presence of bees, essential pollinators, is also affected by the structure of the landscape. A 2016 study published in the journal Agriculture, Ecosystems and Environment detected a 28% improvement in the productivity of coffee plants when 22 bee species were present, both native and Africanized bees. The data are sufficient to recommend that plantations be located no farther than 300 meters from some forest, the distance these insects can fly. Metzger warns that the inverted edge effect is not always good for humans, and in some cases can spread diseases such as dengue and yellow fever.

The original inhabitants of forests are the most interested in maintaining its function, but they are far from the only ones. Understanding these processes requires huge sets of data gathered and analyzed by teams in multiple specialties, an objective which is still under construction.

The Interface Project: relations between landscape structure, ecological processes, biodiversity, and ecosystem services (no. 13/23457-6); Grant Mechanism Thematic Project; Biota Program; Principal Investigator Jean Paul Walter Metzger (USP); Investment R$1,957,416.76.

Scientific articles
PFEIFER, M. et al. Creation of forest edges has a global impact on forest vertebrates. Nature. v. 551, p. 187-91. 9 nov. 2017.
HATFIELD, J. et al. Trait-based indicators of bird species sensitivity to habitat loss are effective within but not across data sets. Ecological Applications. v. 28, n. 1, p. 28-34. jan. 2018.
TUCKER, M. A. et al. Moving in the Anthropocene: Global reductions in terrestrial mammalian movements. Science. v. 359, n. 6374, p. 466-69. 26 jan. 2018.
PESSOA, M. S. et al. Fruit biomass availability along a forest cover gradient. Biotropica. v. 49, n. 1, p. 45-55. jan. 2017.
MORANTE-FILHO, J. C. et al. Birds in anthropogenic landscapes: The responses of ecological groups to forest loss in the Brazilian Atlantic Forest. PLOS ONE. v. 10, n. 6, e0128923. 17 jun. 2015.
BENCHIMOL, M. et al. Translating plant community responses to habitat loss into conservation practices: Forest cover matters. Biological conservation. v. 209, p. 499-507. mai. 2017.
LAURANCE, W. F. et al. An Amazonian rainforest and its fragments as a laboratory of global change. Biological Reviews. v. 93, n. 1, p. 223-47. fev. 2018.
LIBRÁN-EMBID, F. et al. Effects of bird and bat exclusion on coffee pest control at multiple spatial scales. Landscape Ecology. v. 32, n. 9, p. 1907-20. jul. 2017.
SATURNI, F. T. et al. Landscape structure influences bee community and coffee pollination at different spatial scales. Agriculture, Ecosystems and Environment. v. 235, p. 1-12. nov. 2016.