Changes in tree and liana communities along a successional gradient in a tropical dry forest in south-eastern Brazil

• Published in: Plant ecology Vol. 201; no. 1; pp. 291 - 304
• Main Authors: Madeira, Bruno G, Espírito-Santo, Mário M, Neto, Santos D'Ângelo, Nunes, Yule R. F, Arturo Sánchez Azofeifa, G, Wilson Fernandes, G, Quesada, Mauricio
• Format: Journal Article
• Language: English
• Published: Dordrecht Dordrecht : Springer Netherlands 01.03.2009; Springer; Springer Netherlands; Springer Nature B.V
• Subjects: Abundance; Applied Ecology; Biodiversity; Biomedical and Life Sciences; Brazil; Climate; Community & Population Ecology; Deciduous forests; Dominance; Dry forests; Ecology; Ecosystems; Forest ecology; Forest ecosystems; forest growth; Forest regeneration; Forest succession; Forestry; Forests; Land use; Lianas; Life Sciences; Light effects; Pasture; Pasture management; Plant Ecology; Plant species; plowing; Rainforests; regeneration; Roots; Soil types; Species composition; species diversity; Succession; Terrestial Ecology; Trees; Tropical environments; Tropical forests; Tropical rain forests; Woody vines; Brazil; Liana–tree interactions; Succession; Floristic composition; Forest structure; Land use history
• ISSN: 1385-0237; 1573-5052
• DOI: 10.1007/s11258-009-9580-9

We investigated changes in species composition and structure of tree and liana communities along a successional gradient in a seasonally dry tropical forest. There was a progressive increase in tree richness and all tree structural traits from early to late stages, as well as marked changes in tree species composition and dominance. This pattern is probably related to pasture management practices such as ploughing, which remove tree roots and preclude regeneration by resprouting. On the other hand, liana density decreased from intermediate to late stages, showing a negative correlation with tree density. The higher liana abundance in intermediate stage is probably due to a balanced availability of support and light availability, since these variables may show opposite trends during forest growth. Predicted succession models may represent extremes in a continuum of possible successional pathways strongly influenced by land use history, climate, soil type, and by the outcomes of tree-liana interactions.