Pedro V. Eisenlohr

Identity and relationships of the Arboreal Caatinga among other floristic units of seasonally dry tropical forests (SDTFs) of north-eastern and Central Brazil

The tree species composition of seasonally dry tropical forests (SDTF) in north-eastern and central Brazil is analyzed to address the following hypotheses: (1) variations in species composition are related to both environment (climate and substrate) and spatial proximity; (2) SDTF floristic units may be recognized based on peculiar composition and environment; and (3) the Arboreal Caatinga, a deciduous forest occurring along the hinterland borders of the Caatinga Domain, is one of these units and its flora is more strongly related to the caatinga vegetation than to outlying forests. The study region is framed by the Brazilian coastline, 50th meridian west and 21st parallel south, including the Caatinga Domain and extensions into the Atlantic Forest and Cerrado Domains. Multivariate and geostatistic analyses were performed on a database containing 16,226 occurrence records of 1332 tree species in 187 georeferenced SDTF areas and respective environmental variables. Tree species composition varied significantly with both environmental variables and spatial proximity. Eight SDTF floristic units were recognized in the region, including the Arboreal Caatinga. In terms of species composition, its tree flora showed a stronger link with that of the Cerrado Dry Forest Enclaves. On the other hand, in terms of species frequency across sample areas, the links were stronger with two other units: Rock Outcrops Caatinga and Agreste and Brejo Dry Forests. There is a role for niche-based control of tree species composition across the SDTFs of the region determined primarily by the availability of ground water across time and secondarily by the amount of soil mineral nutrients. Spatial proximity also contributes significantly to the floristic cohesion of SDTF units suggesting a highly dispersal-limited tree flora. These units should be given the status of eco-regions to help driving the conservation policy regarding the protection of their biodiversity.

Florestas de restinga e de terras baixas na planície costeira do sudeste do Brasil: vegetação e heterogeneidade ambiental

It was evaluated the floristic similarity between two Atlantic Rainforest physiognomies in Brazilian coast area, herein called Restinga and Lowland forests. The hypothesis was that, due the differences in geomorphologic processes, these forests would differ on soil physical and chemical properties, floristic composition, live above-ground biomass and litterfall production. It was sampled 1 ha (100 × 100 m) for each site located in Ubatuba, Sao Paulo state, SE Brazil. Within each hectare it was recorded trees with DBH > 4.8 cm in all 10 × 10 m contiguous plots, and collected soil and litterfall samples. The cluster and ordination analyses indicated the two communities as distinct groups considering soil and floristic composition, agreeing with the initial hypothesis. Species diversity was higher (p 0.05) between the two forests. This apparent paradox could be explained assuming that, since different species establish themselves in the Restinga or Lowland forests and find a favorable spectrum of conditions and resources, they would tend to persist and to develop in that place; even so the edaphic conditions differ between the Restinga and Lowland forests, each species could respond in a particular way to these variations, and then both forests could reach similar values of biomass and litterfall production. It is probable that the environmental filter conditioned by soils has being important for the strong floristic segregation between these two forests.

Persisting challenges in multiple models: a note on commonly unnoticed issues regarding collinearity and spatial structure of ecological data

There is a growing need to heed some caveats in numerical data analysis. In 2013, I set out some issues regarding multiple regression frameworks. Here, I used both hypothetical and real data sets collected in Brazil to discuss the implications of, and provide suggestions for, some statistical issues regarding to collinearity and spatial structure of ecological data. For example, a weak treatment of collinearities might lead to discarding important variables for the model, and this can be avoided by a correct approach to collinearities before the model selection. Moreover, studies have demonstrated that the spatial structure in both predictor and response variables is an important point to be addressed, rather than the presence of this structure only in the residuals. Aiming to facilitate the controlling of such bias, I provide two fully explained scripts for R language. Considering the seriousness of spatial structure, my opinion is that no article that presents confirmatory analysis should be considered for publication if their authors do not heed that caveat; facing this issue, I strongly suggest that one performs a variance partitioning scheme.

Does size matter? Conservation implications of differing woody population sizes with equivalent occurrence and diversity of species for threatened savanna habitats

The Cerrado stands out from other savannas of the world for hosting the highest biodiversity and for its ongoing massive destruction. Savannas on plain relief and with deep soils (DS) in Central Brazil have been widely converted into agropastoral ecosystems. In contrast, savannas on steep relief with shallow and rocky soils (RS) have been considered as future biodiversity refuges in Central Brazil. In this study, we show that adjacent DS and RS savanna sites (each with ten 1-ha plots) differ based on their woody species population sizes, but not based on species occurrence or diversity. In addition, we discuss the implications of these results for species conservation. Our research indicates that the DS and RS savannas are complementary, but not equivalent, regarding their representation of savanna vegetation in Central Brazil. We hypothesize that if RS savannas become the only refuges for savannic vegetation, widespread biodiversity losses will occur in the short term (through loss of exclusive and habitat-specialist species) and long term (due to fragmentation, reductions in population size and loss of genetic variability). Thus, we suggest that the consideration of savannas on distinct substrates should be considered for improving conservation decision-making and initiatives and can be beneficial when expanding or creating new conservation units.

Climate-related variables and geographic distance affect fern species composition across a vegetation gradient in a shrinking hotspot

Background: The Atlantic Forest biome is a top hotspot for conservation priorities, but the variations in fern species composition and the factors driving these are still poorly known. The vegetation gradient found in the state of Santa Catarina, southern Brazil, constitutes an interesting environmental model for examining such patterns. Aims: We analysed the variation in fern species composition across a vegetation gradient in the southern Atlantic Forest and the variables that might influence it. Methods: We applied cluster analysis (TWINSPAN and WPGMA), ordination (DCA), regression (OLS and GWR) and variance partitioning to species occurrence, environmental and spatial data in 40 sample units (50 × 50 km2), using metadata. Results: Bioclimatic variables, mainly those related to humidity (water being an essential medium for fern reproduction), and geographic distance were related to variation in fern species composition. Dispersal constraints, probably related to neutral processes, could be evoked to explain the significance of the geographic distance. Overall, a great deal of uncertainty remained with regard to the determinants of floristic composition. Conclusions: Our study emphasises, for the first time in the Atlantic Forest, a significant role of environmental determinism and dispersal constraints on the variation of fern species composition.

Challenges in data analysis: pitfalls and suggestions for a statistical routine in Vegetation Ecology

Abstract The step of data analysis in a scientific work is not always a friendly universe. Here, I provide suggestions and warn of five pitfalls in a proposal of statistical routine focused on selection of predictor variables for multiple regression—a simple model used to answer questions commonly raised in Vegetation Ecology—and verification of assumptions of this method. I believe that this manuscript will clarify important points in the data analysis process and, therefore, contribute to make studies in Vegetation Ecology more competitive in the international scientific scenario.

Humidity, low temperature extremes, and space influence floristic variation across an insightful gradient in the Subtropical Atlantic Forest

The use of well-sampled regions and insightful environmental gradients can provide important theoretical knowledge for understanding the drivers of tropical forest patterns and processes, which are necessary for biological conservation. We investigated the variation in tree species composition across a vegetation gradient in the Subtropical Atlantic Forest, as well as various predictors that could impact such patterns. The exploratory and confirmatory analyses included 178 sampling units distributed among 13 different forest types containing binary occurrence records for 935 species from 25,546 trees. Confirmatory analyses were conducted considering spatial autocorrelation, a well-known factor responsible for type I error inflations. The floristic heterogeneity of the study area was revealed by clear patterns of floristic differentiation between vegetation types, as well as by the significant improvements in local models (GWR—geographically weighted regressions) over global models (OLS—ordinary least squares). We found that the observed floristic variation has most likely been driven by humidity from the Atlantic Ocean, low temperature extremes and geographic distance. The latter was revealed by the high contribution of spatial components to the partial models. These results from a shrinking biological hotspot, particularly those concerning the importance of temperature extremes, represent crucial data for identifying conservation priorities, especially considering the seriousness of the temperature shifts predicted by global climate change scenarios.

The Brazilian Atlantic Forest: new findings, challenges and prospects in a shrinking hotspot

The Brazilian Atlantic Forest has been explored and degraded for more than 500 years (Morellato and Haddad 2000; Ribeiro et al. 2011). The forest fragments that still exist, except in protected areas such as National Reserves and Biological Reserves, are concentrated on the tops of mountains and/or steeper slopes, where agricultural activity is difficult or unfeasible, either for access or due to the generally low soil fertility (Moreno et al. 2003). This biome is, however, recognized as one of the 35 world hotspots for conservation priorities (Myers et al. 2000; Zachos and Habel 2011), and has even been referred as a ‘‘hottest hotspot’’ (Laurance 2009), ‘‘shrinking hotspot’’ (Ribeiro et al. 2011), or ‘‘top hotspot’’ (Eisenlohr et al. 2013). It is for this reason that a special issue is devoted to these forests here. Researchers have responded to this scenario, addressing important floristic and phytogeographic features of the Atlantic Forest vegetation, and producing works revealing major implications for biodiversity conservation (e.g. Oliveira-Filho and Fontes 2000; Marques et al. 2011; Werneck et al. 2011). These studies focused on tree species, the component on which there is far more available data than on most other plant types. However, in recent years, the number of surveys dealing with other growth forms has increased, along with the inclusion of life-forms other than trees in sampling protocols and floristic surveys. Including non-tree life-forms in investigations contributes to our knowledge of the geographic limits and the historical processes acting in the Atlantic Forest, in addition to providing a sounder basis for the development and technical planning of conservation policies.

Tree species composition and richness along altitudinal gradients as a tool for conservation decisions: the case of Atlantic semideciduous forest

Altitude is one of the major environmental variables influencing the distribution of tree taxa around the world, and can be a useful parameter for the development of conservation strategies. Our objectives were to obtain an overview of the conservation status of taxa from the Atlantic semideciduous seasonal forests of southeastern Brazil and check, among the floristically consistent altitudinal zones, which had the largest number of endemic, threatened and indicator species. To accomplish this, we used species occurrence and geo-climatic data from the ‘NeoTropTree’ database. First, we checked which zones were floristically consistent using an ordination method (non-metric multidimensional scaling, NMS) followed by an analysis of variance. Then, we conducted an indicator species analysis, estimated species richness, calculated the taxonomic distinctness index for all zones and evaluated important aspects for conservation, such as endemism and threat of extinction. Taxa richness, and number of endemic and threatened species decreased with increasing altitude, while the amount of indicator species for each zone increased with altitudinal elevation. These variations related to the altitudinal gradient should be considered when developing conservation strategies. We also should prioritise the conservation of the entire gradient and, whenever possible, include endemic and endangered species in forest recovery programs.

Human Impacts Affect Tree Community Features of 20 Forest Fragments of a Vanishing Neotropical Hotspot

The loss in forest area due to human occupancy is not the only threat to the remaining biodiversity: forest fragments are susceptible to additional human impact. Our aim was to investigate the effect of human impact on tree community features (species composition and abundance, and structural descriptors) and check if there was a decrease in the number of slender trees, an increase in the amount of large trees, and also a reduction in the number of tree species that occur in 20 fragments of Atlantic montane semideciduous forest in southeastern Brazil. We produced digital maps of each forest fragment using Landsat 7 satellite images and processed the maps to obtain morphometric variables. We used investigative questionnaires and field observations to survey the history of human impact. We then converted the information into scores given to the extent, severity, and duration of each impact, including proportional border area, fire, trails, coppicing, logging, and cattle, and converted these scores into categorical levels. We used linear models to assess the effect of impacts on tree species abundance distribution and stand structural descriptors. Part of the variation in floristic patterns was significantly correlated to the impacts of fire, logging, and proportional border area. Structural descriptors were influenced by cattle and outer roads. Our results provided, for the first time, strong evidence that tree species occurrence and abundance, and forest structure of Atlantic seasonal forest fragments respond differently to various modes of disturbance by humans.