J Plant Ecol ›› 2018, Vol. 11 ›› Issue (5): 740-750.DOI: 10.1093/jpe/rtx044

• Research Articles • Previous Articles     Next Articles

Are the cerrado and the seasonal forest woody floras assembled by different processes despite their spatial proximity?

Raquel Carolina Miatto1,2,* and Marco Antonio Batalha1   

  1. 1 Department of Botany, Federal University of S?o Carlos, PO Box 676, 13565–905, S?o Carlos, Brazil
    2 Department of Biology, University of S?o Paulo, 14040–901, Ribeir?o Preto, Brazil
  • Received:2017-05-17 Revised:2017-06-26 Accepted:2017-08-01 Published:2018-09-27



The Brazilian cerrado occupies land that could be occupied by seasonal forest, given current climatic conditions and their spatial proximity. Soil has been identified as one of the main determinants of cerrado and forest prevalence. We tested whether cerrado and seasonal forest woody floras were assembled by different processes. We postulated that soil nutrient availability would account for differences in the functional and phylogenetic structure of the cerrado and the seasonal forest.


In 100 5 × 5 m plots distributed along cerrado and seasonal forest patches in south-eastern Brazil, we sampled five individuals with a basal diameter ≥3 cm from 127 species and measured seven of their functional traits (plant height, basal area, tortuosity, leaf size, specific leaf area, leaf dry matter content, and leaf toughness). We constructed a phylogenetic tree and calculated the pairwise mean functional-phylogenetic distances (MFPDs), an approach that accounts for functional and phylogenetic information both separately and combined. We also sampled soils to a depth of 5 cm in each plot and had their chemical and physical properties determined. We related the MFPD to soil properties and compared MFPD between cerrado and forest species.

Important Findings

Phylogenetic distances were higher than functional distances in both cerrado and seasonal forest communities, suggesting trait convergence in both environments. Irrespective of the importance given to functional or phylogenetic information, most of the communities in the cerrado and in the seasonal forest fell within the null expectation, implying either that multiple assembly processes can occur simultaneously along the gradient of soil fertility, or that not all important traits were included. MFPD was related to soil nutrient status when only functional distances were considered. In this case, MFPD was lower in the cerrado than in the forest, indicating that soil nutrient availability influenced plant traits, with the low-nutrient soils in the cerrado constraining the range of variation in these traits. We found largely similar sorting mechanisms occurring in the cerrado and in the seasonal forest when we accounted for several important traits simultaneously along the phylogenetic distances. Nevertheless, we also found strong evidence for abiotic environmental filtering in the cerrado and for biotic filtering in the seasonal forest when relating MFPD to soil variables. Despite not being the only ecological force structuring these communities, soil nutrient status plays an important role in maintaining the functional distinctiveness of the two vegetation types.

Key words: competition, functional traits, environmental filter, nutrient limitation, phylogeny, savanna