a b s t r a c tLandscape changes can lead to bee species loss, what impairs proper landscape level pollination processes, impacting both nature conservation and human welfare. Although landscape heterogeneity can rescue bee communities from collapsing, these insects seem sensitive to reduced functional connectivity, hindering pollen transfer among plants. Our objective was to verify which of these two factors, landscape heterogeneity or functional connectivity, can better explain variations of bee abundance and richness in a fragmented Atlantic Forest region. We sampled flower-visiting bees in 12 landscapes with varying heterogeneity and functional connectivity measured using a Functional Ecological Corridors framework. Both richness and abundance were affected by landscape factors, reaching its highest levels at intermediate levels of functional connectivity in highly heterogeneous landscapes, indicating the existence of strong regime shifts in the system. In low-forested landscapes, conservation actions for pollinating bees should focus on implementing diversified environments with high quality which are interspersed among each other and with native vegetation.

Natural regeneration offers a cheaper alternative to active reforestation and has the potential to become the predominant way of restoring degraded tropical landscapes at large-scale. We conducted a meta-analysis for tropical regions and quantified the relationships between both ecological and socioeconomic factors and biodiversity responses in naturally regenerating areas. Biogeographic realms, past disturbance, and the human development index (HDI) were used as explanatory variables for biodiversity responses. In addition, we present a case study of large-scale natural regeneration in the Brazilian Atlantic Forest and identify areas where different forms of restoration would be most suitable. Using our dataset for tropical regions, we showed that natural regeneration was predominantly reported within: the Neotropical realm; areas that were intensively disturbed; and countries with medium HDI. We also found that biodiversity in regenerating forests was more similar to the values found in old growth forests in: countries with either low, high, or very high HDI; less biodiverse realms; and areas of less intensive past disturbance. Our case study from Brazil showed that the level of forest gain resulting from environmental legislation, in particular the Brazilian Forest Code, has been reduced, but remains substantial. Complementary market incentives and financial mechanisms to promote large-scale natural regeneration in human-modified agricultural landscapes are also needed. Our analysis provides insights into the factors that promote or limit the recovery of biodiversity in naturally regenerating areas, and aids to identify areas with higher potential for natural regeneration.Abstract in Portuguese is available with online material.

No abstract

Consideration of soil quality indicators is fundamental for understanding and managing ecosystems. Despite the evidence regarding the importance of soil for provision of local and global ecosystem services, such as water regulation and carbon sequestration, soil remains an under-investigated and undermined aspect of the environment. Here we evaluate to what extent soil indicators are taken into account in restoration. We focused on the Brazilian Atlantic Forest, a highly fragmented biome and a global biodiversity hotspot. We conducted a systematic literature review and we showed that the majority (59%) of the studies on restoration did not consider any soil indicator. Studies that demonstrated the importance of soil indicators most commonly reported soil pH (71%, n = 44), followed by potassium content (66%, n = 41) and phosphorus (64.5%, n = 40), while the least reported indicator was water retention (6.5%, n = 4). Only 40% of the retrieved studies included information about reference sites or project baseline information. We complement our literature review with a case study on restoration in two areas of the Atlantic Forest. We found a relation between soil indicators such as soil organic matter, nitrogen, sodium and sand content, and aboveground indicators, confirming a necessity to include soil screening in restoration. Moreover, we found that prior to restoration none of these soil indicators were analyzed. This study highlights the gap that exists in soil data in restoration in studies on the Brazilian Atlantic Forest. We urge scientists and practitioners to include basic soil analysis to maximize the successful outcomes of restoration.

Natural regeneration provides multiple benefits to nature and human societies, and can play a major role in global and national restoration targets. However, these benefits are context specific and impacted by both biophysical and socioeconomic heterogeneity across landscapes. Here we investigate the benefits of natural regeneration for climate change mitigation, sediment retention and biodiversity conservation in a spatially explicit way at very high resolution for a region within the global biodiversity hotspot of the Atlantic Forest. We classified current land-use cover in the region and simulated a natural regeneration scenario in abandoned pasturelands, areas where potential conflicts with agricultural production would be minimized and where some early stage regeneration is already occurring. We then modelled changes in biophysical functions for climate change mitigation and sediment retention, and performed an economic valuation of both ecosystem services. We also modelled how land-use changes affect habitat availability for species. We found that natural regeneration can provide significant ecological and social benefits. Economic values of climate change mitigation and sediment retention alone could completely compensate for the opportunity costs of agricultural production over 20 years. Habitat availability is improved for three species with different dispersal abilities, although by different magnitudes. Improving the understanding of how costs and benefits of natural regeneration are distributed can be useful to design incentive structures that bring farmers' decision making more in line with societal benefits. This alignment is crucial for natural regeneration to fulfil its potential as a large-scale solution for pressing local and global environmental challenges.

Xenarthrans—anteaters, sloths, and armadillos—have essential functions for ecosystem maintenance, such as insect control and nutrient cycling, playing key roles as ecosystem engineers. Because of habitat loss and fragmentation, hunting pressure, and conflicts with domestic dogs, these species have been threatened locally, regionally, or even across their full distribution ranges. The Neotropics harbor 21 species of armadillos, 10 anteaters, and 6 sloths. Our data set includes the families Chlamyphoridae (13), Dasypodidae (7), Myrmecophagidae (3), Bradypodidae (4), and Megalonychidae (2). We have no occurrence data on Dasypus pilosus (Dasypodidae). Regarding Cyclopedidae, until recently, only one species was recognized, but new genetic studies have revealed that the group is represented by seven species. In this data paper, we compiled a total of 42,528 records of 31 species, represented by occurrence and quantitative data, totaling 24,847 unique georeferenced records. The geographic range is from the southern United States, Mexico, and Caribbean countries at the northern portion of the Neotropics, to the austral distribution in Argentina, Paraguay, Chile, and Uruguay. Regarding anteaters, Myrmecophaga tridactyla has the most records (n = 5,941), and Cyclopes sp. have the fewest (n = 240). The armadillo species with the most data is Dasypus novemcinctus (n = 11,588), and the fewest data are recorded for Calyptophractus retusus (n = 33). With regard to sloth species, Bradypus variegatus has the most records (n = 962), and Bradypus pygmaeus has the fewest (n = 12). Our main objective with Neotropical Xenarthrans is to make occurrence and quantitative data available to facilitate more ecological research, particularly if we integrate the xenarthran data with other data sets of Neotropical Series that will become available very soon (i.e., Neotropical Carnivores, Neotropical Invasive Mammals, and Neotropical Hunters and Dogs). Therefore, studies on trophic cascades, hunting pressure, habitat loss, fragmentation effects, species invasion, and climate change effects will be possible with the Neotropical Xenarthrans data set. Please cite this data paper when using its data in publications. We also request that researchers and teachers inform us of how they are using these data.

No abstract

Epiphytes are hyper‐diverse and one of the frequently undervalued life forms in plant surveys and biodiversity inventories. Epiphytes of the Atlantic Forest, one of the most endangered ecosystems in the world, have high endemism and radiated recently in the Pliocene. We aimed to (1) compile an extensive Atlantic Forest data set on vascular, non‐vascular plants (including hemiepiphytes), and lichen epiphyte species occurrence and abundance; (2) describe the epiphyte distribution in the Atlantic Forest, in order to indicate future sampling efforts. Our work presents the first epiphyte data set with information on abundance and occurrence of epiphyte phorophyte species. All data compiled here come from three main sources provided by the authors: published sources (comprising peer‐reviewed articles, books, and theses), unpublished data, and herbarium data. We compiled a data set composed of 2,095 species, from 89,270 holo/hemiepiphyte records, in the Atlantic Forest of Brazil, Argentina, Paraguay, and Uruguay, recorded from 1824 to early 2018. Most of the records were from qualitative data (occurrence only, 88%), well distributed throughout the Atlantic Forest. For quantitative records, the most common sampling method was individual trees (71%), followed by plot sampling (19%), and transect sampling (10%). Angiosperms (81%) were the most frequently registered group, and Bromeliaceae and Orchidaceae were the families with the greatest number of records (27,272 and 21,945, respectively). Ferns and Lycophytes presented fewer records than Angiosperms, and Polypodiaceae were the most recorded family, and more concentrated in the Southern and Southeastern regions. Data on non‐vascular plants and lichens were scarce, with a few disjunct records concentrated in the Northeastern region of the Atlantic Forest. For all non‐vascular plant records, Lejeuneaceae, a family of liverworts, was the most recorded family. We hope that our effort to organize scattered epiphyte data help advance the knowledge of epiphyte ecology, as well as our understanding of macroecological and biogeographical patterns in the Atlantic Forest. No copyright restrictions are associated with the data set. Please cite this Ecology Data Paper if the data are used in publication and teaching events.