RESEARCH

The geography of matching traits in the assembly of ecological networks – Ecological networks, where species are vertices and species interactions are edges, capture well the complexity and structure of multitrophic communities. A single network can represent the whole set of coexisting species assemblages and the interspecific interactions patterns. Species traits, that is the morphological, physiological, or behavioral characteristics of species can influence the structure of networks by determining the assembly of interspecific interactions. The matching of species traits between interacting species across trophic levels may promote or limit the occurrence and frequency of species interactions in the network. The co-evolution of matching traits between taxa can be positively selected because it provides an increase in interaction efficiency and a reduced competition for interaction partners among species. As such, many ecological networks show identifiable signals of trait matching in their interaction structure. However, biotic interaction filters are not the only forces driving dissimilarities in the structure of interactions across ecological networks. Historical contingencies such as extinction events differing across taxa or regions can influence the trait diversity of trophic assemblages and trait-matching patterns in ecological networks. Moreover, the type and strength of environmental filters shaping trait diversity of species assemblages in an ecological network can differ between trophic levels. However, the effects of historical contingencies or distinct environmental selection forces acting within-trophic levels shaping trait-matching patterns in networks across regions are rarely examined.

Macroecology of Palm-Frugivore seed dispersal interaction networks

Seed dispersal is a key ecosystem process. Animal mediated seed dispersal, is specially important for tropical forests where up to 90% of plant species rely on animals to disperse its seeds. As such, frugivory has evolved into a mutualism where the animal benefits from consuming the water and nutrient rich endocarp. I return, the plant gets to disperse its seed over large distances inside the animal gut. Palms are ubiquitous in tropical forests, and provide fruit resources for a great variety of animals. Palms have also experienced differential extinction patterns in different biogeographical realms and exhibit a wide array of fruit sizes, colors and shapes.

In this project, we tested wether large scale biogeographic differences in the structure of palm-frugivore seed dispersal interactions and functional trait-matching patterns was present between the Neotropics and Afrotropics. In addition, we quantified the geographical coverage, and knowledge gaps of interaction information on seed dispersal of palms at both regions. Moreover, we look to increase the digital mobility of biotic interactions data by making our dataset(s) available in global standardized interaction databases (e.g. GloBI).

In parallel, I also studied the seed dispersal interaction between the Yolillo palm (Raphia taedigera) and the Baird's tapir (Tapirus bairdii). This interaction is a remnant of the mostly-extinct megafaunal interactions in the Neotropics. However, the persistence of this seed dispersal interaction in Costa Rica and Nicaragua is threatened by defaunation and deforestation. Because the tapir is the only animal capable of dispersing Raphia seeds over large distances, the extinction of this interaction can lead to a reduce in the genetic exchange among Raphia taedigera populations. We identified patches of palms with higher priorities for conservation integrating literature-based natural history observations on tapir dispersal behavior with spatially explicit models of connectivity.

These projects, and BOM were developed during my Master period with the Biogeography and Macroecology (BIOMAC) research group in the Institute for Biodiversity and Ecosystem Dynamics (IBED) of the University of Amsterdam (UvA).

Discoverability of primary biodiversity data from literature

The answer of large scale questions in Ecology requires biodiversity data to be accesible and standardized. Most of this data is still locked as single biodiversity observations reported in literature. Literature on biological sciences, mostly found as PDFs in the internet hold a great amount of primary biodiversity information on species-species interactions or species functional traits. In this project we developed Biodiversity Observations Miner (BOM), a software to discover biodiversity observations from multiple PDF articles at a time. Thus, accelerating the process of biodiversity data discovery from textual observations in literature.