濒危物种
生态学
灌木
系统发育多样性
栖息地
生物群落
生物
植被(病理学)
栖息地破坏
地理
系统发育树
生态系统
医学
生物化学
病理
基因
作者
Pablo Viany Prieto,Guilherme Dubal dos Santos Seger,Felipe S. M. Barros,Ary Teixeira de Oliveira‐Filho,Marinez Ferreira de Siqueira
标识
DOI:10.1016/j.foreco.2023.121352
摘要
Conservation assessments that aim to represent the pool of lineages that exist within a region are challenging due to incomplete knowledge of the spatial distribution and environmental preferences of distinct lineages. In this context, community-level models represent a powerful tool that can be used to predict spatial variation in phylogenetic composition. If lineages composition and habitat loss present a non-random spatial pattern within a region, we can expect that some lineages are more threatened by habitat loss than others. To test this hypothesis we combined multivariate analysis and a modelling approach to predict the phylogenetic composition of angiosperm tree/shrub communities in Rio de Janeiro State, southeastern Brazil, which is regarded as a plant endemism center within the Brazilian Atlantic forest biome. Then we correlated the resulting models with maps of vegetation cover and protected areas, to identify lineages subjected to higher habitat loss and clades that receive less protection. We found that angiosperm tree/shrub lineages are highly structured across distinct vegetation types and that a gradient of deforestation and land protection parallels the phylogenetic gradient formed by the distinct vegetation types. Consequently, some vegetation types and its associated angiosperm lineages are much more threatened by habitat loss and are less protected than others. Specifically, Ericales (asterids), Caryophyllales and Santalales are the most threatened clades in this region, along with their preferred habitats, seasonal forests and coastal white-sand forests (restingas). Protecting and restoring these dry habitats is critical for the conservation of tree and shrub phylogenetic diversity in the Brazilian Atlantic forest.
科研通智能强力驱动
Strongly Powered by AbleSci AI