Search

Global environmental changes are expected to alter the functional characteristics of understorey herb-layer communities, potentially affecting forest ecosystem functioning. However, little is known about what drives the variability of functional traits in forest understories. Here, we assessed the role of different environmental drivers in shaping the functional trait distribution of understorey herbs in fragmented forests across three spatial scales. We focused on 708 small, deciduous forest patches located in 16 agricultural landscape windows, spanning a 2500-km macroclimatic gradient across the temperate forest biome in Europe. We estimated the relative effect of patch-scale, landscape-scale and macroclimatic variables on the community mean and variation of plant height, specific leaf area and seed mass. Macroclimatic variables (monthly temperature and precipitation extremes) explained the largest proportion of variation in community trait means (on average 77% of the explained variation). In contrast, patch-scale factors dominated in explaining community trait variation (on average 68% of the explained variation). Notably, patch age, size and internal heterogeneity had a positive effect on the community-level variability. Landscape-scale variables explained only a minor part of the variation in both trait distribution properties. The variation explained by shared combinations of the variable groups was generally negligible. These findings highlight the importance of considering multiple spatial scales in predictions of environmental-change effects on the functionality of forest understories. We propose that forest management sustainability could benefit from conserving larger, historically continuous and internally heterogeneous forest patches to maximise ecosystem service diversity in rural landscapes. (C) 2018 Gesellschaft fur Okologie. Published by Elsevier GmbH. All rights reserved.

Tree-lines and alleys are expected to operate as migration enhancing corridors for habitat-demanding species, but their functionality is limited by the set of ecological filters. We use multiple plant traits related to dispersal and establishment to identify the limiting filters for forest plants in the rural landscape of Estonia. We develop a set of quantitative metrics to rank indicator traits according to their distributional changes along migration distance. We implement a trait comparison to the potential optimum level suggested by two ecological reference groups of species to interpret these responses as filter driven convergence or divergence. The suggested set of metrics provided a clear ranking of traits and showed interpretational limitations of widely used short list of traits (e.g. seed weight, plant height and SLA). Results also demonstrate that there is no consistency between indicator metrics based on the shift in trait mean and those based on the reduction of variability, instead they provide complementary information. Unexpectedly, many characteristic traits of forest-specialist plants do not exhibit the expected responses. The response signal of many dispersal traits is too ambiguous to interpret because either (1) they do not have one clear optimum level, or (2) they indicate an establishment/persistence limitation instead, such as seed weight and flowering duration. Establishment traits indicate filtering clearly by improved light conditions. The pattern of trait means demonstrates that the dispersal filtering incrementally intensifies with distance, while establishment filtering occurs sharply at the forest-corridor ecotone. Consolidated results underscore that the migration of forest-restricted plants into corridors is limited by the habitat quality for dispersal vectors (e.g., for myrmecochores and zoochores), the scarcity of suitable microsites for seedling establishment, and the competition for light. A single optimal structure of wooded corridors cannot be suggested as forest-dwelling species exhibit various adaptations. Forest-biodiversity-enhancing wooded corridors should incorporate a diversity of shade levels and have structures that facilitate visits of forest insects, birds and mammals. We show that biased conclusions about the functional efficiency of habitats and limiting ecological filters can be avoided when indicator analytics include (1) multiple response metrics and analytical methods, (2) multiple seemingly redundant traits, and (3) several reference groups for interpretation. The proposed analytical approach adjusts for (1) the indictor choice subjectivity, and (2) the scaling bias implemented in several ecological indicator trait systems, which over-emphasise qualitative preferences of species rather than reflect their real niches.

International audience; Global forest loss and fragmentation have strongly increased the frequency of forest patches smaller than a few hectares. Little is known about the biodiversity and ecosystem service supply potential of such small woodlands in comparison to larger forests. As it is widely recognized that high biodiversity levels increase ecosystem functionality and the delivery of multiple ecosystem services, small, isolated woodlands are expected to have a lower potential for ecosystem service delivery than large forests hosting more species. We collected data on the diversity of six taxonomic groups covering invertebrates, plants and fungi, and on the supply potential of five ecosystem services and one disservice within 224 woodlands distributed across temperate Europe. We related their ability to simultaneously provide multiple ecosystem services (multiservice delivery potential) at different performance levels to biodiversity of all studied taxonomic groups (multidiversity), forest patch size and age, as well as habitat availability and connectivity within the landscape, while accounting for macroclimate, soil properties and forest structure. Unexpectedly, despite their lower multidiversity, smaller woodlands had the potential to deliver multiple services at higher performance levels per area than larger woodlands of similar age, probably due to positive edge effects on the supply potential of several ecosystem services. Biodiversity only affected multiservice delivery potential at a low performance level as well as some individual ecosystem services. The importance of other drivers of ecosystem service supply potential by small woodlands in agricultural landscapes also depended on the level of performance and varied with the individual ecosystem service considered. Synthesis and applications. Large, ancient woodlands host high levels of biodiversity and can therefore deliver a number of ecosystem services. In contrast, smaller woodlands in agricultural landscapes, especially ancient woodlands, have a higher potential to deliver multiple ecosystem services on a per area basis. Despite their important contribution to agricultural landscape multifunctionality, small woodlands are not currently considered by public policies. There is thus an urgent need for targeted policy instruments to ensure their adequate management and future conservation in order to either achieve multiservice delivery at high levels or to maximize the delivery of specific ecosystem services.

Fragmentation strongly shapes the distribution of organisms within forest patches through contrasting environmental conditions between the edge and interior habitat. Edge-to-interior distribution patterns are, however, poorly studied for litter- and soil-dwelling fauna, such as terrestrial gastropods, despite their high densities and significant impact on ecosystem processes, as both herbivores and detritivores. Therefore, we investigated edge-to-interior abundance patterns of terrestrial gastropods in 224 fragmented forest patches across Western Europe. Catching over 15,000 gastropods, we found that slug abundance is reduced in forest edges, while snail abundance shows no response on the edge effect. We hypothesize that these patterns could be explained by higher drought tolerance of snails, since forest edges have reduced air and soil humidity and elevated temperatures compared to forest interiors. Reduced slug abundance in forest edges potentially has ecological consequences for herbivory in and outside forest patches and nutrient cycling.

Aim Macroclimate is a major determinant of large-scale diversity patterns. However, the influence of smaller-scale factors on local diversity across large spatial extents is not well documented. Here, we quantify the relative importance of local (patch-scale), landscape-scale and macroclimatic drivers of herbaceous species diversity in small forest patches in agricultural landscapes across Europe. Location Deciduous forest patches in eight regions along a macroclimatic gradient from southern France to central Sweden and Estonia. Methods The diversity of forest specialists and generalists at three levels (whole forest patch, sampling plots within patches and between scales) was related to patch-scale (forest area, age, abiotic and biotic heterogeneity), landscape-scale (amount of forest, grasslands and hedgerows around the patch, patch isolation) and macroclimatic variables (temperature and precipitation) using generalized linear mixed models and variation partitioning for each group of variables. Results The total amount of explained variation in diversity ranged from 8% for plot-scale diversity of generalists to 54% for patch-scale diversity of forest specialists. Patch-scale variables always explained more than 60% of the explained variation in diversity, mainly due to the positive effect of within-patch heterogeneity on patch-scale and between-scale diversities and to the positive effect of patch age on plot-scale diversity of forest specialists. Landscape-scale variables mainly contributed to the amount of explained variation in plot-scale diversity, being more important for forest specialists (21%) than for generalists (18%). Macroclimatic variables contributed a maximum of 11% to the plot-scale diversity of generalists. Main conclusions Macroclimate poorly predicts local diversity across Europe, and herbaceous diversity is mainly explained by habitat features, less so by landscape structure. We show the importance of conserving old forest patches as refugia for typical forest species, and of enhancing the landscape context around the patches by reducing the degree of disturbance caused by agriculture.