期刊
ENVIRONMENTAL MANAGEMENT
卷 59, 期 3, 页码 410-418出版社
SPRINGER
DOI: 10.1007/s00267-016-0800-5
关键词
Diversity recovery; Forest regeneration; Landuse effects; Human-modified ecosystems
Cocoa agroforests like the cabrucas of Brazil's Atlantic forest are among the agro-ecosystems with greatest potential for biodiversity conservation. Despite a global trend for their intensification, cocoa agroforests are also being abandoned for socioeconomic reasons especially on marginal sites, because they are incorporated in public or private protected areas, or are part of mandatory set-asides under Brazilian environmental legislation. However, little is known about phylogenetic structure, the processes of forest regeneration after abandonment and the conservation value of former cabruca sites. Here we compare the vegetation structure and composition of a former cabruca 30-40 years after abandonment with a managed cabruca and mature forest in the Atlantic forest region of Espirito Santo, Brazil. The forest in the abandoned cabruca had recovered a substantial part of its original structure. Abandoned cabruca have a higher density (mean +/- CI95 %: 525.0 +/- 40.3 stems per ha), basal area (34.0 +/- 6.5 m(2) per ha) and species richness (148 +/- 11.5 species) than managed cabruca (96.0 +/- 17.7; 24.15 +/- 3.9 and 114.5 +/- 16.0, respectively) but no significant differences to mature forest in density (581.0 +/- 42.2), basal area (29.9.0 +/- 3.3) and species richness (162.6 +/- 15.5 species). Thinning (understory removal) changes phylogenetic structure from evenness in mature forest to clustering in managed cabruca, but after 30-40 years abandoned cabruca had a random phylogenetic structure, probably due to a balance between biotic and abiotic filters at this age. We conclude that abandoned cocoa agroforests present highly favorable conditions for the regeneration of Atlantic forest and could contribute to the formation of an interconnected network of forest habitat in this biodiversity hotspot.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据