Abstract Detail

Macroevolution

Cortez, Maria Beatriz [1], Shimizu, Gustavo [2], Barve, Vijay [3], Gitzendanner, Matt [4], Guralnick, Robert [5], Soltis, Douglas [6], Soltis, Pamela [7].

Retrieving and managing large datasets from REFLORA and other databases to obtain information about taxa occurring in the campos rupestres.

The campos rupestres constitute a vegetation type predominantly occurring across three phytogeographic regions in Brazil: Cerrado, Atlantic Forest, and Caatinga. Harsh environmental conditions and land discontinuity foster the occurrence of many microhabitats, promoting not only an incredibly high level of plant diversity, but also a high level of species endemism. While the campos rupestres occupy less than 1% of the Brazilian territory, they harbor more than 5000 angiosperm species, representing over 10% of Brazil’s flora. Despite harboring an incredible biodiversity, most patterns and processes of diversification in the campos rupestres are not well understood, especially at a macro-spatial scale. Fortunately, the scientific community has access today to REFLORA, recognized worldwide as one of the most complete databases harboring information on Brazil’s flora. REFLORA is a decade-long effort financed by the Brazilian government that has its virtual infrastructure housed and maintained by Rio de Janeiro Botanical Garden and Alberto Luiz Coimbra Post-Graduation and Engineering Research Institute – UFRJ. This database is organized in two parts, the Virtual Herbarium and the Flora and Fungi of Brazil. To assemble the latter, nearly 1,000 taxonomists compiled different types of information, including taxonomic descriptions, geographic distribution, life forms, nomenclature status, and endemicity for nearly 50,000 species across land plants and fungi. The information contained in REFLORA, alongside occurrence data stored in other important databases (iDigBio, GBIF, and SpeciesLink), facilitates the process of investigating how the flora is phylogenetically distributed in space, considering parameters such as richness and diversity in the campos rupestres. To achieve this goal, we organized the initial part of our workflow into three phases. The first consists of retrieving a list of angiosperm taxa occurring in the campos rupestres using REFLORA’s resources, available for public download as a Darwin Core Archive. The second phase requires creating a set of scripts both in python and R to manage the downloaded data to recover information on number of species, nomenclature, distribution, endemicity, life form, and nativity. When our list was finalized, we moved on to the last phase of retrieving occurrence data to develop species distribution models for the nearly 5,500 angiosperm species occurring in the campos rupestres. Though this workflow may seem straightforward, we faced several challenges, particularly in phase two. These challenges were mostly related to ensuring proper interpretation of the downloaded data and to the labor-intensive portion of manually inspecting nomenclature. We hope that by compiling and managing the multiple sources of data for this project, we can produce a unique dataset that will be useful for other researchers interested in the angiosperm flora of the campos rupestres.

1 - University Of Florida, Florida Museum Of Natural History–Dickinson Hall, 1659 Museum Road, Room 357, Gainesville, FL, 32611, United States
2 - Rua Raphael Fernandes, 153, Mogi Das Cruzes, SP, 08717280, Brazil
3 - University of Florida, Florida Museum of Natural History, Dickinson Hall 1659 Museum Road, Gainesville, Florida, 32611-7800, United States
4 - University Of Florida, Florida Museum Of Natural History, Po Box 117800, Gainsville, FL, 32611, United States
5 - Florida Museum Of Natural History, 358 Dickinson Hall, University O, 358 Dickinson Hall, University Of Florida, Gainesville, FL, 32611, United States
6 - University of Florida, Florida Museum of Natural History,, 3215 Hull Road, P. O. Box 2710, Gainesville, FL, 32611, USA
7 - University Of Florida, Florida Museum Of Natural History, Gainesville, FL, 32611.0, United States

Keywords:
campos rupestres
REFLORA
evolutionary patterns
large datasets
R
Python
Checklist.

Presentation Type: Oral Paper
Number: MACRO II011
Abstract ID:911
Candidate for Awards:None