Iron islands in the Amazon: investigating plant beta diversity of canga outcrops.
Ontology highlight
ABSTRACT: The world's largest mineral iron province, Serra dos Carajás, is home to an open vegetation known as canga, found on top of isolated outcrops rising out of the Amazon rainforest. Over one thousand vascular plants species have been recorded in these canga sites, including 38 edaphic endemics. A new survey adds to our investigation of biogeographic relationships between sixteen canga outcrops and the effect of the distance between site pairs on the number of shared species, regional species turnover and species distribution patterns. Plant collecting expeditions to the westernmost site, the Serra de Campos of São Félix do Xingu (SFX), were carried out followed by the identification of all collected specimens and the creation of a species database, built to perform biogeographical analyses. Floristic relationships among the sites were investigated regarding their similarity, using multivariate analyses. The correlation between canga areas and species richness was tested, as well as the geographical distance between pairs of outcrops and their shared species. Vascular plants at SFX total 254 species including 17 edaphic endemics. All canga sites are grouped with 25% of minimum similarity, and the SFX falls within a large subgroup of outcrops. The total species number shared between site pairs does not change significantly with geographical distance but is positively correlated with the area of each outcrop. Meanwhile, shared endemic species numbers between site pairs decline when geographical distance increases, possibly imposed by the barrier of the rainforest. Our data suggest higher shared similarity between the largest and species-richest sites as opposed to geographically nearby sites, and provide useful insight for drafting conservation and compensation measures for canga locations. The size of the canga outcrops is associated to higher floristic diversity but connectivity among islands also plays a role in their similarity.
SUBMITTER: Andrino CO
PROVIDER: S-EPMC7642173 | biostudies-literature | 2020
REPOSITORIES: biostudies-literature
ACCESS DATA