New patterns of the tree beta diversity and its determinants in the largest savanna and wetland biomes of South America

Clear and data-driven bioregionalizations can provide a framework to test hypotheses and base biodiversity conservation.Here we used occurrence and abundance data in combination with objective analytical methods to propose two bioregionalization schemes for tree species of the Cerrado and the Pantanal in South America.We also evaluated the contribution of three sets of determinants of the occurrence-and abundance-based subregions.We compiled data on tree species composition from 894 local assemblages based on species occurrences,and from 658 local assemblages based on species abundances.We used an unconstrained community-level modelling approach and clustering techniques to identify and map tree subregions for the occurrence and the abundance data sets,separately.Hierarchical clustering analyses were conducted to investigate floristic affinities between the subregions and to map broader floristic regions.We used multinomial logistic regression models,deviance partitioning,and rank-sum tests to assess the main subregion correlates.We identified 18 occurrence-and four abundance-based subregions in the Cerrado-Pantanal.The hierarchical classifications grouped the occurrence-based subregions into nine floristic zones and abundance-based subregions into two broad floristic zones.Variation in subregions were explained mainly by environmental factors and spatial structure in both occurrence and abundance data sets.The occurrence-and abundance-based subregions are complementary approaches to disentangle macroecological patterns and to plan conservation efforts in the Cerrado and the Pantanal.Our findings based on occurrence data revealed more complex and interdigitated boundaries between subregions of tree species than previously reported.The environment,historical stability,and human effects act in a synergetic way on the distribution of the subregions.Finally,the relevance of contemporary environmental factors to the subregion patterns we found alert us to the profound impact global warming may have on the spatial organization of the Cerrado-Pantanal tree flora.

Long-lasting effects of unplanned logging on the seed rain of mixed conifer-hardwood forests in southern South America

Understanding the effects of disturbance on seed rain is critical to predict changes in forest species composition and diversity.Logging effects on seed rain in a mixed conifer-hardwood forest complex in southern Brazil were evaluated.One year of seed rain data were collected from a large-scale observational experiment in logged and protected forests and quantity and average seed size weighted by species abundance(CWM)were compared between old-growth and logged stands 55 years after logging activities.Using these data,variations in frequency of functional groups of species in the seed rain were examined to see if they could be attributed to logging.Results show that the number of seeds per trap was highly right-skewed,ranging from 13 to 12,788 seeds per trap in one year.Seed rain was affected by logging history,with seed traps in old-growth plots receiving significantly less seeds than traps in logged plots.All species included mean seed size weighted by species abundance were significantly smaller in logged than in old-growth forests.This difference persisted after the exclusion of Arawcaria a large-seeded pioneer which was intensively logged,although the difference of seed size between the two forest classes was greatly reduced.Species abundance in the seed rain differed significantly from the established tree community,between logged and oldgrowth stands.The composition of the seed rain was much more variable than the composition of the established tree community and its points more scattered over the ordination space than the points corresponding to the protected forests.The number of collected seeds across different functional groups of species significantly differed between logged and old-growth plots.The seed rain of logged forests reflects their arrested succes sion as indicated by reduced abundance of functional groups such as pioneers,large seeded pioneers and A raucaria,as well as reduced functional diversity.S eed rain differences between logged and old-growth stands reflect the intensive logging of A.angustifolia.

In the lack of extreme pioneers:trait relationships and ecological strategies of 66 subtropical tree species

Aims Despite the growing interest in the topic of functional ecology,there are still forest regions that have not been examined,as most work has been done in the tropics.unresolved issues include the strength of a growth-mortality trade-off in trees(originally identified for seedlings)and the nature of the association between plant traits and vital rates,if any.our objectives were to examine whether(i)ecological strategies in south american mixed forests are organ-ized along the fast competitor×slow stress-tolerator and height gradients as the main strategy axes depicted in the overall trait and vital rate correlation structure,and(ii)a tentative path model we proposed can explain the patterns of covariation among traits and vital rates.Methods We studied a different habitat(subtropical mixed conifer-hardwood forests)and region(brazilian atlantic Forest)from the majority of related studies in forests,carried out in the Neotropical region.Data on total height,stem slenderness,crown depth,wood density,specific leaf area,leaf and seed length,seed dispersal mode,annual mortality,diameter relative growth rate and relative growth rate under favorable conditions were measured in southern brazil for 66 tree species.Data were subjected to principal components analysis and path analysis.restricted data on saplings and treelets were analyzed through correlation.Important Findingsstudied traits were reduced to four principal components.Principal components analysis produced axes that fit the resource acquisition versus resource-conservation and the height-mortality trade-offs,although the former was split into two distinct axes.seed size and seed dispersal mode appeared independently of these axes.a path model showed that leaf length and specific leaf area caused direct changes in trunk slenderness and,indirectly through growth,affected mortality.Expected trade-offs between growth and survivorship and between wood density and stem slenderness trade-offs were not found.this may result from the lack of extreme pioneers and over-representation of slow-growing hardwood species found in atlantic subtropical forests of south america.this suggests that the fastest growing species in the region do not grow so fast as to compromise wood density and survivorship,but grow fast enough to benefit from increased size.relationships between traits and vital rates seem to be mediated by the assembly process of regional floras,and the relative importance of traits like SLA and wood density may vary between floristic regions.