Mediterranean agro-forestry systems are undergoing rapid change due to abandonment. This turns formerly cultivated or grazed oak-tree parklands (i.e., savanna-type formations called “dehesas”) into flammable formations of scattered trees within a matrix of shrubs with open spaces. Wildfires can now occur, threatening the persistence of these formations. Fire-prone shrublands commonly regenerate after fire from seeds stored in the soil. Understanding the relationships between standing vegetation and the soil seed bank across microhabitats can help predict the response of the system in case of fire. Here we investigated these relationships in an abandoned Quercus suber (cork oak) dehesa in central Spain. Vegetation and soil were sampled and assigned to different microhabitats: under the trees (TRC), in dense shrub cover (DSC), and in low shrub cover (LSC). A heat shock was applied to half of each sample to simulate fire; the other half served as control. Both sets of samples were then germinated in a greenhouse. Almost 90% of the species were herbs, while the rest were woody shrubs. The number of species recorded in TRC and DSC was lower than in LSC, in which species richness, particularly herbs, was maximal. Heating increased the total number of species that germinated, but mean species richness per sample was not altered. Heating markedly increased the number of germinations in all microhabitats, particularly those of woody species. Furthermore, the germination of shrubby species increased in the 3 microhabitats, notably in TRC. While the standing plant community was well differentiated among microhabitats, this was not the case for the soil seed bank, which was homogeneous across microhabitats, with or without heating. We conclude that the high density of shrubby seeds found in TRC or in the other microhabitats presents substantial threats to the persistence of Q. suber parklands in case of fire.

In natural ecosystems, species assemblages of isolated ecological communities frequently exhibit a nested pattern. The rapid urbanization that has occurred in Tokyo, central Japan, has resulted in the formation of extensive isolated forest remnants. We examined how geographic factors and the life-history traits of butterflies affected the occurrence of nested distribution patterns in butterflies from 20 forest remnants in the city. The species inhabiting each remnant were surveyed using transect counts, and the geographic attributes of the forest remnants, such as remnant shape, isolation, and distance to a region of contiguous forest were characterized.The species life-history traits that were considered included host plant type, host plant range, voltinism, and adaptability of the butterflies to the matrix (i.e., areas outside forest remnants). Butterfly species with host plants that were cultivated within the matrix were defined as highly adaptable species. The results showed that the butterfly assemblages in the surveyed area were significantly nested. In addition, the nested rankings (NR) of remnants, which are used as indicators of extinction vulnerability, were correlated with remnant area but not with remnant shape, isolation, or distance to the continuous forest. The best model based on AIC_c revealed that species with short flying periods and a narrow host plant range consisting of woody plants that were not cultivated in the matrix had low associated NR values. Our findings showed that selective local extinction may contribute to the nestedness of butterflies in forest remnants, and that host plant type, host plant range, voltinism, and adaptability to the matrix appear to affect butterfly vulnerability to habitat fragmentation. From a conservation perspective, understanding the factors that influence extinction vulnerability has important implications, because it allows us to predict why some butterfly species are more susceptible to extinction than others.

Habitat restoration comprises the re-creation of suitable environmental conditions with the intention of recolonization by certain target species. In previously abandoned calcareous grasslands, however, many characteristic plant species have been reported missing even decades after the reinstatement of traditional mowing or grazing management. Such grasslands are said to exhibit a colonization credit. This may be particularly true for orchid species, which often rely on highly specialized pollination strategies and mycorrhizal associations for completion of their life-cycle. In this study, we investigated whether restored calcareous grasslands exhibited an orchid colonization credit, whether this credit was associated with the degree of grassland fragmentation, and with particular species' life history traits. Applying the Beals index as a quantitative method to identify suitable habitats, several orchid species were indeed found missing from grasslands deemed suitable. There was no relation, however, between the extent of the colonization credit and the spatial isolation or size of the grasslands. Of all life history traits examined, only a high degree of pollinator specialization could be related to delayed colonization. This may suggest that restoring the pollinator community is an important bottleneck in calcareous grassland restoration.

Exotic earthworms are entering previously uninhabited soils of boreal forests, their invasion largely facilitated through human activities. As ecosystem engineers, earthworms are capable of causing dramatic changes in above- and belowground forest composition, but whether they have the same effects in all forests remains unclear. Forest compositional changes caused by earthworms may be mediated by interactions between earthworms and mycorrhizal fungi. Specifically, tree seedling growth may be altered by the presence of exotic earthworms and their subsequent impact on mycorrhizal fungi. In this study, we investigate the effects of exotic earthworms on ectomycorrhizal colonization and seedling growth of the conifer Picea glauca (white spruce) in gray luvisolic soils from the Boreal Plains. Anecic Lumbricus terrestris and epigeic Dendrobaena octaedra earthworms were added to mesocosms each containing a white spruce seedling in a greenhouse experiment. Impacts on the composition of ectomycorrhizal fungi in the mesocosms were determined using a combination of morphological and molecular techniques, and effects on seedling growth were assessed through above- and belowground measurements. The proportion of ectomycorrhizal root tips and ectomycorrhizal community composition did not vary as a function of earthworm species or density. Similarly, exotic earthworms had no significant effect on spruce seedling growth or survival.

Over recent decades, deer populations have reached densities exceeding their natural range of variability in several regions worldwide. Abundant deer populations can change the composition and structure of understory plant communities in forests and trigger trophic cascades affecting animal communities. To investigate such indirect effects on boreal forest songbirds we compared the understory vegetation and songbird communities of balsam fir stands on Anticosti Island (Quebec, Canada), where white-tailed deer (Odocoileus virginianus) were introduced, to those of the neighbouring Mingan Archipelago, where deer are absent. Mean cover of trees at 0–2 m height was on average sevenfold higher on Mingan than on Anticosti, and songbirds dependent on the understory were more common on Mingan than on Anticosti. Songbird community composition was more variable on Mingan, suggesting that the opening of the forest understory on Anticosti through intensive browsing homogenized the songbird community by reducing the occurrence of understory-dependent species. These indirect effects on songbirds were similar, although less dramatic, than those described in temperate forests. We interpret this as a consequence of lower productivity and complexity in understory vegetation of boreal forests. Our results emphasize the negative effects of abundant deer populations on animal diversity and underline the role played in the dynamics of communities by top down factors that regulate herbivore populations.

Understanding the factors that affect community composition is essential for community ecology. The genetic similarity rule (GSR) identifies 3 variables (host genetic composition, phytochemistry, and the environment) that could affect community composition. Few studies have determined the relative influence of these variables on community composition. Using path analysis, we found that arthropod community similarity was better explained by geographic (56%) and environmental (32%) distance than genetic distance in clonal aspen (Populus tremuloides). Comparing our data with data from similar studies of poplars (P. fremontii and P. fremontii × P. angustifolia hybrids), we found that hybrid poplar stands had higher levels of genetic and arthropod diversity than did clonal aspen stands. We found a significant relationship between genetic and arthropod diversity only in hybrid stands and across Populus systems. Our findings agree with the GSR expectations that the importance of the genetic composition of the host in structuring communities depends in part on the amount of genetic variation present in the study system. This is relevant for management and restoration strategies of geographically restricted species and of disjunct populations of otherwise widespread species, as these tend to have lower effective population sizes and reduced levels of genetic diversity.

Elevation gradients provide ideal scenarios to study plant responses to environmental factors and to global warming. Physiological and morphological traits, growth, and reproduction in bilberry were investigated at 6 elevations along an elevation gradient from 350 to 2000 m asl. Chlorophyll content and growth increased with elevation, reaching a maximum at 950 m, and then decreased, with both variables being negatively influenced by high soil pH. By contrast, after removing the positive effect of tree canopy cover, the efficiency of photosynthesis did not show differences between elevations. The number of stomata per area increased with elevation, while leaf area reached maximum values at 950 m. Regarding reproductive traits, densities of flowers and fruits were not affected by elevation, and fruit set, seed set, and seed viability only varied between localities within elevation. Moreover, flower production was negatively correlated with soil pH. Reproductive success was not limited by pollen quantity at any of the elevations. However, elevation affected number of ovules, number of mature seeds per fruit, and fruit dry weight; these variables reached their highest values at around 1700 m. These results show that, while bilberry exhibited the most favourable vegetative performance at mid elevations, maximum reproductive output was observed at higher elevations.

We assessed habitat selection of moose in a study area containing 2 highways and a network of forest roads in Quebec, Canada. We tested the hypothesis that roadways would affect moose habitat selection and that moose behaviour would vary with time, proximity to roads, type of roadway, and environmental characteristics (i.e., habitat and topography). We equipped 47 moose with GPS telemetry collars and assessed habitat selection using resource selection functions. Moose searched primarily for areas with high forage availability, but they also avoided highways and forest roads, although avoidance usually disappeared beyond 100–250 m. Avoidance was not directly proportional to noise disturbance; moose systematically avoided the first 100 m adjacent to forest roads, while habitats adjacent to highway sides were sometimes used in proportion to their availability. The benefits of using habitats adjacent to highways may be greater than the costs to moose, which was not always the case for forest roads. The road-avoidance zone varied seasonally but was generally wider for males than females, suggesting that males were more sensitive to road disturbance. We believe that moose frequent highways and associated roadsides to find food and mineral salts, and possibly to reduce predation risk for females. Topography was also an important correlate of habitat selection by moose, especially for females, which had significant selection coefficients for altitude and slope in almost every season. Proper assessment of human impacts on ungulates in natural environments requires consideration of unpaved forest roads. In a study area with a small human footprint and low road density (0.16 km·km^-2), moose selection patterns suggested they minimized potential risks associated with highways at the coarse scale while seeking short-term benefits of highway roadsides, i.e., sodium in vegetation and pools, at the finer spatial scale.

Even though there is no reason to believe in a unique mechanism that would explain rarity, recognizing common patterns might aid in identifying effective conservation strategies. This study therefore approached the problem of rarity in spring habitats both at the level of multiple taxonomic groups (molluscs, oligochaeta, water mites, copepods, ostracods, chironomids, stoneflies, caddisflies, diatoms, and vascular plants) and, in more detail, at the level of a single group (bryophytes). The aim was to evaluate whether the proportion of rare species was associated with uncommon environmental conditions, while for the bryophytes an additional aim was to test whether common and rare species differed concerning their niche parameters (niche breadth and niche position), their biological traits (involved in dispersal processes), and their habitat preferences. Overall, 4 major results may be highlighted. 1) The significant concordance between the rarity of virtually all the taxonomic groups inhabiting spring habitats and their environmental conditions suggested that most of the rare species at community level might be explained by their uncommon resource requirements. 2) Rare bryophyte species not only had high niche positions, as did all the taxonomic groups, but also had narrow niche breadths. This result suggests an interesting distinction between resource use and resource availability. 3) Because the species traits linked to dispersal ability did not differ between common and rare species, the hypothesis of an important role being played by these traits in determining species distribution was not supported. 4) Among the rare bryophyte species, an undefined number were casual. Because they were less hygrophilous and less influenced by the environment, this result suggests that they might recruit essentially by chance from the surrounding habitats.

We examined the effect of long-term nitrogen (plus phosphorus and potassium) fertilization, resulting in increased shrub cover, on seasonal changes in understory light, soil temperature, and soil moisture in an ombrotrophic bog. An increase in leaf area index (LAI) was negatively correlated with light transmission through the canopy, decreasing photosynthetically active radiation (PAR) reaching the peat surface by up to an average of 77% compared to unfertilized plots. Owing to the denser shrub canopy, near surface soil temperature was cooler in summer and less spatially variable within the fertilized plots. A laboratory study of the environmental controls on Sphagnum capillifolium carbon dioxide (CO₂) exchange showed that there were significant interactions between moisture and temperature, but changes in CO₂ exchange in response to temperature or moisture were small compared to the influence of light. These results suggest that the absence of moss in the fertilized plots may be, in part, the result of decreased light availability. Alterations to the competitive balance between the shrub and moss layer could lead to changes in C storage in these ecosystems.