Lovejoy Conservation Biology Research Fellows

The Lovejoy Conservation Biology Research Fellowship program provides financial support to graduate students to conduct fieldwork on their MSc or PhD theses at the Amazon Biodiversity Center’s research center (also known as the Biological Dynamics of Forest Fragmentation Project) in Manaus, Amazonas, Brazil.  Fellows conduct conservation research on the impacts of forest fragmentation and/or forest regeneration on plants, animals, ecological processes, and carbon sequestration.  Fellowships are made possible with the support of the Sacharuna Foundation.


2015 Fellows

Francisca Helena Aguiar da Silva
National Institute of Amazonian Research (INPA) — Ecology Program — Ph.D. candidate

Home range and resources for harpy eagle (Harpia harpyja) in the Brazilian Amazon

The present study intends to describe the ecological aspects of the Harpy Eagle (Harpia harpyja), the most powerful predatory bird in the Americas. In some regions of its distribution range they are critically endangered or probably extinct. In Brazil, Harpy Eagle nesting territories are in different kinds of anthropogenic land-uses scenarios, which gives us an opportunity to study aspects of landscape that influence ecological processes. This study aims to describe the food habits and reproductive behavior of the Harpy Eagle to contribute on its effective conservation. During breeding season, we used camera trapping as a non-intrusive method to describe behavior and prey remains in the nest to know about resource selection.

Harpy Eagle (Harpia harpyja), a large predatory bird of the Americas, interacts with several species of mammals, birds and reptiles. Understanding how changes in the landscape surrounding the nests interfere with its habits and behavior is the main objective of this study. Between March 2015 and March 2016 prey species were identified from traces (bones, fur, feathers and claws) collected in the nest and on the ground below, and also from camera trap photos taken of two nests located at the BDFFP reserves (Km41 and Porto Alegre). Thirteen species of mammals and birds had been consumed by this large eagle: Bradypus tridactylus (pale-throated  three-toed sloth), Choloepus didactylus (two-toed sloth), Alouatta macconnelli (Guyanan red howler), Sapajus apella (Guianan Brown Capuchin), Pithecia pithecia (White-faced sakis), Chiropotes sagulatus (Northern bearded sakis), Didelphis marsupialis (common opossum), Coendou melanurus (Black-tailed hairy dwarf porcupine), Nasua nasua (South American Coati), Potos flavus (kinkajou), Eira barbara (tayra), Ramphastos sp. and Galiformes. In addition to the direct interactions with their prey, the Harpy Eagle indirectly interacted with S. apella (a prey) and the birds Cathartes melambrotus, Ibycter americanus, Penelope marail which occasionally use the eagle nests, searching and eating prey remains. Inside a radius of 2.5 km surrounding both nests there are only forests, none of the anthropogenic land-uses as were measured in other nests along the Brazilian Amazonia. All these results will be analyzed in conjunction with data from other Harpy Eagle nests located in different landscapes in the Amazonia, Cerrado and Atlantic forests. Determining Harpy Eagle resilience based on historical and recent disturbances by deforestation in the vicinity of their nests can contribute to Harpy Eagle conservation actions and to (protecting?) the forests surrounding its nesting trees.


Raul Vasconcelos Rodrigues
National Institute of Amazonian Research (INPA) — Ecology — Masters degree candidate

The effect of forest fragmentation on the structure of the lizard community in Central Amazon

Forest fragmentation causes a loss of forested area and it is mainly caused by deforestation for different land uses such as road building or land for grazing and agriculture. This forest fragmentation process changes the climate of the area, since the decreasing of vegetation cover increases solar incidence in the region and, for example, increases environmental temperature. Consequently, forest fragmentation resulting from climate change and anthroprogenic land use could increasingly harm lizards.

Forest fragmentation is the result of deforestation and may result in the loss of species. This may occur due to change in the environment caused by alteration of the microclimate. In the forest fragments, the temperature tends to be higher due to a higher sunlight incidence. Many studies have shown that the structure of communities of different groups of animals and plants is negatively influenced by habitat fragmentation. Being ectothermic, abrupt changes in temperature can dramatically affect the behavior of lizards. Changes in the landscape caused by forest fragmentation can modify the availability of microhabitats for lizards that may cause changes in community structure. This study aims to evaluate how forest fragmentation has affected the structure of the lizard community in the Central Amazon. My plan is to assess whether there are changes in the composition, abundance and number of species between areas of forest fragments with different sizes, continuous forests and secondary forests. I expect that the forest fragmentation process should negatively affect the structure of the lizard community. I forsee that forest fragmentation processes will decrease the abundance and the number of lizard species in forest fragments. The study will be conducted on three farms in the study areas of the Biological Dynamics of Forest Fragments Project. On each farm I will survey lizards in three tracks of 100 m in a fragment of 1 ha, in a fragment of 10 ha, in a secondary growth forest adjacent to the fragments and in an area of continuous forest. I will use the visual search method limited by time and active search. The collection period was from November to December 2015, and I will collect again from August to November 2016. For statistical analysis, I will use NMDS to compare the four environments, as well as sorting and analysis of similarity.


Márlon Breno Costa Santos da Graça
Institution: National Institute of Amazonian Research (INPA) — Entomology — Ph.D. candidate

Taxonomic composition, functional and phylogenetic of frugivorous butterflies in three reserves of the BDFFP

Butterflies are charismatic insects known for their capacity to indicate anthropogenic environmental disturbances. Forest destruction due to phenomena such as habitat fragmentation results in severe damages in the populations of these insects. Many species are locally extinct because they could not adapt to the new environmental conditions. For example, one butterfly species will lose its larval food resource if the specific host plant is cut off from the forest. Hence, the individuals of this species would have to move to another area in order to find a suitable host for its larvae to avoid perishing. Notwithstanding, forest fragmentation destroys the natural patches throughout the native landscape, cornering the most sensitive species. For that reason, it is important that surveys conducted in such areas take into consideration every tool available to understand how butterfly communities are responding to habitat loss. Our aim is to evaluate these ecological responses, to make the case for more rigorous environmental legislation in Brazil.

The challenge of community ecology studies nowadays is to integrate ecological approaches that deal with distinct dimensions of biodiversity. With this project, we seek tot gather taxonomic, functional and phylogenetic diversities of a group of butterflies (fruit-feeding guild) in order to comprehend their spatial distribution within the BDFFP’s Amazonian reserves. Butterflies were sampled with bait traps at eight plots on four reserves: Dimona (10 ha), Porto Alegre (100 ha e 10 ha), Colosso (10 ha e 1 ha) and KM 41 (Trails A, J and U). The distance between sampling points was always greater than 1 km to keep samples independent from one another. Functional traits to be measured are body size (wings and thorax) and larval diet breadth. Phylogenetic diversity will be estimated using the Phylogenetic Species Variability Index (PSV). The area and fragmentation of the reserves, as well as the diffuse solar light gradient will be used to predict the turnover in taxonomic, functional and phylogenetic composition of fruit-feeding butterflies. The results of this study will be crucial for illustrating the importance of the BDFFP reserves and similar reserves for the community of invertebrates, making the case for the establishment of public policies that protect and conserve the Amazon rainforest.  Also, it will illustrate the necessity of investigating different dimensions of biodiversity, as a way to expand avenues of ecological research and their role for environmental conservation.


Ricardo Luís Spaniol
Federal University of Rio Grande do Sul (UFRGS) — Ecology — Ph.D. candidate

Functional organization of frugivorous butterflies guilds in a forest disturbance gradient in Central Amazon

Understanding the effects of forest fragmentation on biodiversity can go beyond a simple inventory of species. In our study we propose using taxonomic, functional and phylogenetic diversity aspects of frugivorous lepidopterans (butterflies) to follow the responses of this group of organisms to a forest disturbance gradient in Central Amazonia.

Understanding forest fragmentation effects on biodiversity can only be achieved by identifying biological community structure in these environments, evaluating multiple dimensions of diversity. Practices that generate this knowledge should go beyond static interpretations of species composition, abundance and/or richness values of a community, but also informing how organisms are related to ecosystem functioning. Thus, we propose a taxonomic, functional and phylogenetic diversity study of fruit-feeding Lepidoptera, following their responses to a forest disturbance gradient. We seek to answer the following questions: 1) Do fruit-feeding butterflies and moths assemblages differ in their taxonomic, functional and phylogenetic diversity between secondary forest, forest fragments and continuous forest? 2) Do morphological and behavioral attributes in fruit-feeding Lepidoptera vary along a disturbance gradient, following it? 3) Do patterns of attribute divergence and/or convergence operate in the structuring of fruit-feeding lepidopterans assemblages in fragmented landscapes in the Amazon Forest? 4) Do butterflies and moths respond similarly to this environmental gradient? Between August and September 2015, we installed traps containing bananas bait fermented in sugar cane juice in secondary forest, forest fragments and continuous forest in BDFFP study areas. Morphological and behavioral attributes of captured individuals have been compiled, and analyzed with environmental and landscape variables that may be associated with the assembly of these assemblages in the study sites. To date, a total of 404 frugivorous lepidopterans were recorded in the sampled BDFFP areas, distributed among 59 species butterflies and moths. This number may increase as data and materials are further processed. All lepidopterans sampled in this first year of work are in the final stages of mounting, identification and data organization. As these stages are concluded, individual lepidopterans have their functional attributes measured. Once all data on identification and attributes is obtained, we will be able to step into the next stage of quantitative analyses, so as to answer the proposed questions.


Danilo Roberti Alves de Almeida
University of São Paulo – College of Agriculture (ESALQ – USP) — Forest Sciences — Ph.D. candidate

Is it possible to predict diversity considering forest structure? Mapping structural differences and their relationship with the floristic diversity from LiDAR data

The work is about the discrimination of forest types and estimates of floristic diversity (alpha and beta) from the vegetation structure and its spatial heterogeneity, measured with LiDAR remote sensing.

The Amazon forest shows great complexity and differentiation in the structural attributes of vegetation and floristic diversity due to its wide spatial distribution and variation in environmental conditions. The structural dynamics and diversity are two of the primary principals of forest ecology for making important decisions in conservation, preservation and forest management, in the face of great pressure on natural forests and biodiversity. However, carry estimates in the field of structural variables and identification of species throughout the Amazon are not viable operationally and economically. Therefore, it is of fundamental importance the improvement and creation of remote sensing methods, working in a large scale to estimate the floristic diversity and to classify forest types. The purpose of this study is to explore the potential of transported air LiDAR (ALS) and portable terrestrial (LTP) to access and monitor the floristic diversity in forests of central Amazonia (BDFFP) from structural variables estimated by the LiDAR data. We will create a predictive model in order to predict floristic diversity both in continuous forests and in fragmented forests. In addition to a predictive model of floristic diversity from the forest structure I also intend to generate an automatic classification methodology of the different forest types from structural variables.

Marielle Natasha Smith
The University of Arizona — Ecology & Evolutionary Biology — Ph.D. candidate

Quantifying changes in three-dimensional forest canopy structure and light environments during tropical forest regeneration using LIDAR

Researchers at the BDFFP have found that two very different regeneration pathways exist in the secondary forests of the BDFFP reserves. Areas that have been deforested with no further land use are dominated by the tree genus Cecropia during regeneration. These areas go through a normal successionary sequence in which the numbers of pioneer, light-tolerant trees like Cecropia diminish over time, being replaced by more shade-tolerant species as the forest canopy develops. However, areas which have been used as pasture and burned become dominated by the tree genus Vismia. Unlike the Cecropia dominated forests, Vismia continues to proliferate in these areas for many years following abandonment. These contrasting successionary pathways likely give rise to very different forest structures – namely, the location and quantity of leaves. For example, leaf area may be higher in Vismia-dominated forests because young Vismia trees can continue to recruit underneath a Vismia canopy. I made measurements of forest structure using ground-based LiDAR in Vismia and Cecropia dominated secondary forests in order to understand whether these two different secondary forest types have distinctively different structures. I will pair the LiDAR measurements with existing biomass measurements of the secondary forest plots in order to quantify how biomass is related to metrics of canopy structure. If Vismia and Cecropia dominated forests do have contrasting structures, we might be able to identify secondary forest areas of different land use histories at large scales using airborne LiDAR, and ultimately, improve predictions of above ground biomass in these areas and how long secondary forest areas require to return to a state that resembles primary forest.

Amazon forests play a particularly important role in global atmospheric circulation, such that changes to these forests could alter cycles of water and carbon, and energy balance. Land-use change is a major driver of canopy structure and function change in Amazonian forests. The increasing prevalence of secondary forests in the Amazon and other tropical forests makes it critical for us to develop a detailed understanding of the three-dimensional changes that occur in canopy structure and canopy environments during regeneration. Such insights will increase our ability to predict the trajectory of regeneration pathways and to characterize the potentially important role that secondary forests play in biosphere-atmosphere energy balance. This proposal seeks to understand how three-dimensional forest canopy structure and function (canopy light environments) change in tropical forests regenerating from different land use histories. LiDAR (light detection and ranging) provides unprecedented three-dimensional information on canopy structure and can also be used to estimate light transmission. I will measure forest canopy structure using a ground-based profiling LiDAR instrument along a chronosequence of successional stages and across the major land use histories found at BDFFP. To characterize light environments, I will take hemispherical photos and make PAR (photosynthetically active radiation) measurements along the transects measured with LiDAR. I will determine how canopy structure is related to forest aboveground biomass over succession and land use history by comparing plot level aboveground biomass values against metrics of forest canopy structure. Together these measurements will reveal how forest biometric statistics (aboveground biomass, tree size) are related to quantitative differences in canopy structure and light environments through regeneration and contrasting land use histories. This project has important implications for improving our ability to predict the structural trajectory of forest successional change for areas with different previous disturbances. The findings could be scaled up to make regional scale predictions using survey data from airborne LiDAR. Ultimately, this work will help to inform our understanding of how alterations in forest structure due to land use changes might impact regional scale dynamics.

Patrícia de Oliveira Santos
Institution: State University of Londrina (UEL) — Biological Science — Ph.D. candidate

Phylogenetic analysis of Burseraceae and Cecidomyiidae: evidence of coevolution?

José Marcelo Pelloso Molina
State University of Londrina (UEL) — Biological Science — Ph.D. candidate

Pollination of Castilleae (Moraceae) for insects of the order Thysanoptera: a possible cospeciation process?


Caroline da Cruz Vasconcelos
National Institute of Amazonian Research (INPA) — Botany — Master degree candidate

Propagules and seedlings of Sapotaceae-Chrysophylloideae in the Amazon: a phylogenetic approach

Sapotaceae is a very important plant family, in an ecological as well as economic view. The trees are valuable components in tropical forests and their products like wood, latex and fruits are widely used by humans. The taxonomy of the family is not straightforward and it is often difficult to interpret. Over the years, different classifications have been proposed, based principally on floral characteristics. However, the generic delimitation is still fragile and new approaches are required for a more solid delimitation of the genera, as for example, the analysis of molecular characters. In this new proposal, the information on morphological characters is almost absent. In this context, this project suggests to broaden the morphological knowledge on fruits, seeds and seedlings of the neotropical genera of Chrysophylloideae subfamily and add, if necessary, new characters to help in the generic delimitation of Sapotaceae. The project has an evolutionary approach to understand the phylogenetic and morphological relations among the members of the study group. A biproduct of this project will be an illustrated guide to help in the identification of the trees, fruits and seedlings, and an interactive identification key with multiple access.

Sapotaceae Juss. comprises about 1,250 species of 58 genera. This family is widely distributed in tropical and subtropical regions of the whole world, and occurs especially in moist and low-land forests. In the Neotropics Sapotaceae has a great diversity and abundance, with 450 tree species of 11 genera. Nine of these genera belong to the subfamily Chrysophylloideae and are widely distributed in the Amazon. The objective of this work is to widen the knowledge on fruit, seed and seedling morphology of 30 species, belonging to the nine neotropical genera of Chrysophylloideae. Botanical samples (branches with ripe fruits) will be collected in two sites of Amazonas state: at BDFFP Reserve (Km37) and Adolpho Ducke Forest Reserve. In the laboratory of Seeds of the National Institute for Amazonian Research-INPA, all samples will be processed as follows: biometric data and morphological description of fruits and seeds; seed germination tests and finally morphological description of the seedlings and juveniles. During all stages of the work, photographic records will illustrate the main features of the species. All data will be analyzed in an evolutionary phylogenetic approach, based on the proposal of Faria et al. (pers. comm.) for the Chrysophylloideae. This work will result in an improved phylogenetic tree with informative morphological characters that may give a better understanding of the evolution of fruit, seed and seedlings characters among the members of this family - an unpublished work in the context of Sapotaceae to South America. Hopefully we will conclude an interactive identification key with multiple access to the 30 species, and an illustrated identification guide to support the recognition of the species in the field.

Paulo Ricardo Rodrigues Piovesan
National Institute of Amazonian Research (INPA) — Botany — Master degree candidate

Vegetative propagation and its influence on the relative abundance of lianas in terra firme forest of Central Amazon
Lianas are important components of tropical forests, but are rarely considered in forest inventories. It is a highly diverse group and composes a good part of the biomass of a forest. Over the last years there is evidence that the population of lianas is increasing both in forest fragments as in continuous forests. The population increase may be due to increase of atmospheric CO2 prompted by global climate change. In my study, I want to see if the propagative capability of lianas can give an indication of increase in abundance and hence population growth. Do all lianas have this feature?

The aim of the project is to evaluate whether the relative abundance of different species of lianas is associated with vegetative propagation capacity of the species. We also investigate the possibility that vegetative propagation capacity is independent of phylogeny among the angiosperm and gymnosperm lianas species. Liana species were collected in the Km 37 Reserve of the Biological Dynamics of Forest Fragments Project. Data on relative abundance were obtained from a census in the area realized by Dr. Robyn Burnham, and a total of 10 species among those of highest relative abundance was selected in the census compared with related and unrelated species of lower relative abundance. For each species evaluated, five individuals separated by at least 50 meters were tested. For each individual, a total of seven cuttings with four leaf nodes each were prepared. Cuttings will be planted horizontally in plastic boxes containing a layer of sand covered by coconut fiber and maintained for five months in the greenhouse of the National Institute of Amazonian Research (INPA - Campus V8). We will evaluate the presence or absence of rooting, number of roots, size of largest roots, mortality rate of cuttings, dry biomass of roots and leaves, and average rooting index.