Search

The process of community assembly is important to explain species coexistence and the maintenance of species diversity. Succession provides a good opportunity to study the role of the deterministic and stochastic processes during the community species composition. We investigated species composition and environmental factors and calculated the change in species diversity, phylogenetic diversity, net relatedness index, and nearest taxon index along a successional chronosequence, with a series of 0.5 × 0.5 m2 plots (n = 20) from five successional subalpine meadow plant communities (ages 1, 3, 5, 15, and 30) in the Qinghai-Tibetan Plateau. In the plots, we found a significant change in species composition and abiotic environmental factors. Our redundancy analysis showed that soil nitrate nitrogen and soil organic carbon were the key environmental factors that affect the community composition in succession gradient. The community showed an overdispersion phylogenetic structure along with the early and medium succession. The ends of the succession stage showed a random or convergent phylogenetic structure. The results showed that the construction process of plant communities in the early and middle stages of succession was mainly affected by habitat filtering. In the late stage of succession, the process of community construction is more complex, which could be affected by the limiting similarity and neutral driving process. Our findings are important to be used in managing ecological restoration and protection.

Describing the rules of community assembly is a central topic of ecology. Studying successional processes through a trait-based null model approach can help to better understand the rules of community assembly. According to theoretical considerations, at the beginning of succession - after getting over the dispersal limitation stage - community composition is primarily shaped by environmental filters (generating functional convergence), while in later stages limiting similarity (generating functional divergence) will be dominant. However, empirical evidence does not clearly support theoretical expectations. Our aim was to detect the presence and changes of trait-based assembly processes during old-field succession based on twelve traits. Changes in vegetation composition were evaluated by a combination of time series and space-for-time substitution: conducting three resurveys of permanent plots on four old-field age-groups. The individual dispersion of traits was transformed into effect size (i.e. departure from null model expectation). The impact of time since abandonment on effect sizes was tested by generalized additive mixed effect models. We detected a non-random pattern for each trait in at least some part of the succession. Departure from randomness did not change significantly over time for six traits: seed mass, lateral spread and pollination type were divergent, while leaf size, generative height and length of flowering were convergent. Six traits had changing patterns along the succession. Four of them showed increasing divergence (e.g. dispersal type, LDMC), which supports our hypothesis. While two (SLA, life form) displayed increasing convergence, contrary to expectations. We confirmed the general hypothesis that convergence is predominant initially and that divergence can be detected later in succession for four traits. However, the large variation found in trait dispersion indicates that complex processes operate during succession.

Quantifying niche space occupation based on assessments of multiple resource utilization traits can improve an understanding of plant species coexistence within communities of varying resource availability. We used an ‘occupied functional niche space (OFS)’ method based on multiple leaf traits to examine how plant communities accommodate changes in water availability within a desert-steppe ecotone spanning a limited geographic region. Our study area (western Inner Mongolia) spanned a range of conditions from highly water-limited (desert, D) to less water-limited (desert steppe, S), with two areas capturing the ecotonal boundary between these two vegetation types – partial desert (pD) and partial desert steppe (pS). We tested whether environmental filtering, heterogeneity, or limiting similarity drove trait convergence versus divergence of OFS across a desert-steppe ecotone. We quantified changes in plant community weighted means of both functional (specific leaf area and leaf dry matter content) and stoichiometric (leaf nitrogen content and leaf carbon concentration) leaf traits. OFS in the ecotonal areas (pS and pD) had larger volumes than the most mesic (S) subregion. However, divergence in the centroid of OFS was apparent for the driest (D) subregion, relative to the other three sub-regions. Underlying changes to OFS were changes in plant functional types, with more functional groups found in D and the fewest in S. For community-weighted leaf traits, SLA was highest in the ecotone area (pD and pS), while lowest in D. Further, community weighted means based on leaf dry matter content in D was the highest of all sub-regions, while both community weighted means based on leaf nitrogen content and community weighted means based on leaf dry matter content decreased as precipitation decreased, consistent with expectations. Our results suggest that communities alter both the volume and the centroid of functional trait space to adjust to decreasing water resource availability. We posit that limiting similarity (i.e. competitive exclusion) in water-resource use traits among species, supported by changes in life histories, led to OFS divergence in the ecotonal areas (pS and pD) and D, while environmental filtering may have led to convergence of OFS within the most mesic region (S) as water-stress lessons. In total, our results contribute knowledge as to how plant communities alter trait space in resource acquisition across climatic gradients, and underscores the utility of multi-trait approaches in understanding patterns of species coexistence in response to climatic changes.

Soil communities are impacted by increasingly frequent and intense climate extremes. To accurately predict the effects of these events on soil communities and effectively design recovery and management programs for soil biodiversity, it is necessary to understand the spatiotemporal patterns and mechanisms underlying the soil community responses. Here, we investigated a mangrove soil macrofaunal (mollusk) community before and after an extreme cold event (ECE) to determine the effects of the ECE on the spatial pattern, temporal succession, and deterministic and stochastic processes affecting the soil metacommunity. We used distance-decay and time-decay analyses to characterize the spatial pattern and temporal succession of soil mollusk community dissimilarity (e.g., species identity, abundance, and biomass) across a reginal scale (~350 km scale). We used null models to compare the relative importance of deterministic and stochastic processes in shaping the regional soil mollusk community assembly. The ECE resulted in a reorganization of the soil mollusk metacommunity reflected by the mortality and extinction of several original species, colonization by new opportunistic species, and a significant reduction in mollusk abundance and biomass. The ECE increased divergent temporal succession but decreased spatial segregation in the soil mollusks, consistent with changes in the community dissimilarity of species abundance, richness, and biomass. The ECE increased the numbers of non-random segregated and random structures of soil mollusks while decreased the number of non-random aggregated communities. Notably, non-random segregated communities were the most strongly affected. Thus, the response of the soil mollusk metacommunity to the ECE was mediated by an increase in limiting similarity processes and stochastic processes and a decrease in limiting dissimilarity processes. The dominant limiting similarity processes involved species competition and/or dispersal limitation. More broadly, our study indicated that alterations in the spatiotemporal dynamics of soil fauna in response to climate change should be studied at the metacommunity scale. These spatiotemporal dynamics should be further tested in an ecological processes framework to reliably evaluate soil community responses to climate change and to devise management strategies for global soils.

Despite the urgent demand for sustainable mining, the revegetation and rehabilitation of areas degraded by mining activities remain challenging. Uncertainties about species selection, the management and control of alien invasive species and monitoring procedures require ongoing multidisciplinary research. Modern functional and phylogenetic approaches in community ecology represent powerful tools to enhance the entire rehabilitation process and to overcome these challenges, but methodologies that incorporate up-to-date phylogenetic information or knowledge about intraspecific variation along rehabilitation trajectories must be developed. Functional and phylogenetic criteria may play an especially important role in guiding the definition of faster-growing, more resistant species mixtures that can be applied in mine land revegetation and rehabilitation. Limiting similarity approaches may help to reduce biological invasions, enabling native vegetation to outcompete alien invasive species and aiding in the development of enduring mechanisms for controlling such species in rehabilitated mine land. Finally, incorporating phylogenetic and functional aspects into monitoring of the success of mine land rehabilitation with additional tools, such as remote sensing or metabarcoding, can enrich scientific knowledge about ecosystem functions such as nutrient cycling, community assembly and resource availability while providing sound information about the success of revegetation activities. In the middle to long term, research activities on these topics can provide valuable technical recommendations for all aspects of the management of rehabilitated ecosystems, thus contributing to more sustainable mining.

Two major theories on community assembly (habitat filtering and limiting similarity) predict contrasting patterns in the spatial distribution of plant species. The respective distribution of species functional traits has received much less attention but is essential to understand facilitative and competitive species interactions. Here we apply a recently developed technique for phylogenetic analysis to study the spatial distribution of major functional traits during seven years of primary plant succession at a re-established catchment in a German lignite mine. Habitat filtering induced a significant temporal trend towards underdispersed functional trait diversity leading to patches of functionally similar vegetation. This functional convergence was not mirrored by a respective trend towards species aggregation. Differences in soil pH and nitrogen content were positively correlated with respective differences in trait expression. Separate analyses of trait distribution for aggregated and segregated species revealed trait convergence in co-occurring and random patterns in spatially segregated species pairs except for leaf size showing a strong temporal tendency for trait convergence in spatially segregated species. Contrasting patterns of species co-occurrence did not translate into respective contrasts in trait distribution. We interpret the prevalence of random trait associations in spatially segregated species as an indication that species occurrence patterns alone are weak predictors of assembly mechanisms. Mechanistic inference of species co-occurrence needs to be accompanied by a detailed assessment of the patterns of functional trait assembly.

Revegetation of roadsides is an opportunity for grassland restoration, yet these habitats are prone to be colonised by invasive alien plant species (IAS). Therefore, the selection of seed mixtures for revegetation should consider potential competition with IAS present in the soil seed bank or arriving by traffic-related seed rain. We investigated whether the limiting similarity hypothesis, in terms of plant seed-size-output strategy, could be used to design native grassland communities resistant to IAS. In a greenhouse experiment, a small- or a large-seeded IAS was sown into factorial combinations of two native communities with small or large seed-size-output strategies at two sowing densities. Height and aboveground biomass of the IAS were measured after four and eight weeks, respectively. Small-seeded native communities at high density were highly effective in suppressing the small- and large-seeded IAS, mostly controlled by a density effect. Thus, limiting similarity in seed-size-output strategy only partly explained resistance to IAS, while density-driven suppression was more effective.

In spite of recent advances, the relative roles of deterministic and stochastic processes in community assembly remains disputed. We tested community assembly hypotheses in a disturbance gradient in a logged mixed conifer-hardwood forest complex in southern Brazil. If abiotic heterogeneity was strong enough (Selection hypothesis), we expected niche clustering and reduced niche overlap between species and between ecological groups, phylogenetic clustering, significant β-diversity departure from a random expectation after controlling for α-diversity variation, and non-random trait distribution due to abiotic filtering and limiting similarity. Null expectations were predicted under the Neutral hypothesis. Under the Quasi-Neutral hypothesis, we expected niche clustering and reduced niche overlap as in the Selection hypothesis, but random phylogenetic distribution, random β-diversity variation, and random trait distribution due to niche-based specialization but weak habitat matching resulting from dispersal limitation and drift. We used a series of null models to provide an integrative evaluation of community assembly including species abundances, environmental preferences, phylogenetic relatedness, and trait distribution. Results supported the Quasi-Neutral hypothesis due to a combination of reduced niche overlap, niche clustering along soil and disturbance gradients, evidence for abiotic filtering of functional traits but absent phylogenetic structure and random β-diversity variation. Support to the Quasi-Neutral hypothesis provides evidence for niche-based species selection along environmental gradients, coupled with poor matches between species distributions and environmental factors due to stochastic processes. It also highlights the importance of an integrative approach to the evaluation of community patterns indicative of assembly processes. Had only a subset of the approaches used been employed and different conclusions would have been reached: niche overlap and trait distribution analyses would have indicated a stronger role for selection, while phylogenetic and abundance analyses would have indicated stochastic community assembly. Future research should include different disturbance regimes along with environmental variation in order to assess the interplay of the different drivers that shape the community assembly process, as well as the potential of the Quasi-Neutral hypothesis as an explanation for the assembly process of complex and species-rich subtropical forests.

Potentially invasive species can be identified before they start to spread by comparing their traits with those of successful invaders. A powerful tool for delimiting the traits associated with invasiveness are analyses of a number of species of the same genus, where it is more likely to detect relevant differences because of elimination of biases that constrain the search for such traits in whole floras. Since the influence of traits on invasion success may differ with respect to the stage of the plant's life cycle, comparative studies should address the whole life cycle, including early stages. Here we studied which biological traits are associated with the ability to naturalize within the genus Impatiens, how frequency of planting affects naturalization success, and whether naturalized species with biological traits similar to the native representative of this genus are more successful. The genus Impatiens includes a number of cultivated species popular in horticulture, among them several widespread invaders. We used one native and 10 alien annual taxa. This data set involved all commonly cultivated species, and representatives of different invasion status in Europe. In garden experiments and climatic chambers we measured seed mass, time to germination, percentage of seeds germinated, seedling growth rate, total seedling biomass, seedling root/total biomass ratio, adult biomass and fecundity. These traits and planting frequency were used to explain the invasion success of the species, expressed as (i) invasion status in Europe and (ii) the number of global temperate regions in which the species has been reported as naturalized. The frequency of planting was used as a proxy of propagule pressure to separate this potentially biasing factor known to affect plant invasiveness from the effect of plant traits. We found that both species traits and frequency of planting were correlated with naturalization. Species naturalized in many temperate regions of the world had heavier seeds, high seedling growth rate and allocated low proportion of seedling biomass to roots. Importantly, common planting was more strongly correlated with naturalization success than with biological traits. Impatiens species naturalized in Europe exhibited better seed germination in the common garden, and it took a longer time for the seeds to germinate. Species escaped from cultivation but occurring only as casuals in Europe had heavy seeds and invested more resources into shoots than roots, whereas species not escaping from cultivation were characterized by fast seed germination and light seed. In general, traits linked to early stages of the life cycle were more strongly associated with invasion success than those of the adults. Frequently planted species tend to naturalize more easily than those planted scarcely. The successful invaders share traits similar to the one native Impatiens species in Europe and those with traits distinct from it do not invade. Our results indicate that many Impatiens species represent potential invaders should their planting become more widespread; this prediction is supported by the fact that Impatiens species included in the experiment completed their life cycles in an experimental garden in central Europe.

Processes controlling the assembly of microorganism communities are poorly understood. In taxonomic groups where many organisms are only described from environmental DNA, phylogenetic relationships may provide a proxy for functional attributes. We used a null model approach to compare the phylogenetic diversity of arbuscular mycorrhizal (AM) fungal communities coexisting within the roots of plant individuals sampled worldwide. Parallel null models were constructed using taxon pools that were filtered according to some or all of spatial scale, ecosystem type and host plant identity. This allowed us to assess the importance of individual ecological gradients in shaping phylogenetic diversity and to identify potential mechanisms. In general, coexisting fungi were more phylogenetically clustered than the random communities defined by null models. However, clustering increased when null models incorporated large spatial scales, indicating that communities are assembled according to the dispersal ability of fungal taxa. Clustering also increased when null models incorporated a variety of ecosystems or host plant species, indicating effects of habitat filtering and mutualist partner selection at wide and local scales, respectively. We did not record phylogenetic overdispersion among coexisting fungi, providing no support for a role of limiting similarity in structuring AM fungal communities. However, the existence of significant clustering even in comparison with the finest-scale null model indicates that, in addition to larger-scale deterministic and neutral processes, the symbiosis may be further influenced by genotypic selectivity or stochastic processes at very small scales.

International audience; Climatic extreme events such as droughts (unpredictable), dry periods (predictable) or even flush floods, threaten freshwater ecosystems worldwide. The filtering mechanisms of these events and their strength on communities, however, can be different among regions. While time-for-adaptation theory defines whether or not water scarcity can be considered as disturbance, the stress-dominance theory predicts an increase in importance of environmental filtering and a decrease in the role of biotic interactions in communities with increasing environmental stress. Here, we tested whether environmental filtering (leading to trait convergence) or limiting similarity (leading to trait divergence) is the main assembly rule shaping the structure and trait composition of benthic diatom assemblages in Mediterranean (Portuguese) and continental (Hungarian) temporary and perennial streams. We assumed that the trait composition of diatom assemblages in the two stream types would be less different in the Mediterranean than in the continental region (addressed to time-for-adaptation theory). We also hypothesized that trait composition would be shaped by environmental filtering in the Hungarian streams while by biotic interactions in Portuguese streams (addressed to stress-dominance theory). Our results supported our first hypothesis since traits, which associated primarily to temporary streams were found only in the continental region. Our findings, however, only partially proved the stress-dominance hypothesis. In the continental region, where drying up of streams were induced by unpredictable droughts, biotic interactions were the main assembly rules shaping community structure. In contrast, environmental filtering was nearly as important as limiting similarity in structuring trait composition in the Mediterranean region during the predictable dry phase with no superficial flow. These analyses also highlighted that drought events (both predictable and unpredictable ones) have a complex and strong influence on benthic diatom assemblages resulting even in irreversible changes in trait composition and thereby in ecosystem functioning.

G. E. Hutchinson raised the paradox of how a number of phytoplankton species competing for the same resources are able to coexist in a relatively isotropic environment. As a key for solving the paradox, we focused on the limiting similarity which prohibits the coexistence of similar species. We expected that the limiting similarity will be mitigated by some factors which are not represented in traditional theoretical studies but can be represented if we use a three-dimensional (3D) model. The use of a 3D model enables us to explore the limiting similarity without using the controversial assumption connecting niche overlap with the competitive exclusion in previous theoretical studies. Furthermore, while constant or no dispersion is given in theoretical studies, it is explicitly represented in a 3D model. The coexistence of similar species in 3D environments was explored by dividing a target persistent species in a quasi-equilibrium community into 80 subspecies, which were slightly different from one another at the optimum temperature for population growth. In the experiments, we found cases in which several dozens of species having nearly-overlapping niches coexist stably at a point for over 80 years. The comparison of experiments with and without dispersion by oceanic currents revealed that dispersion negates the progress of competitive exclusion or significantly delays the exclusion. The result that species having nearly-overlapping niches are able to coexist also highlights the crucial role of the specific rate in determining the number of coexisting species. If the number of coexisting species is not determined by the limiting similarity, it will be determined by the frequency of speciation events, as in the neutral theory.

Phylogenetic analysis of plant communities is useful for inferring ecological mechanisms driving forest succession. However, such analysis has scarcely been undertaken in tropical dry environments, especially for the dynamics of demographic components (i.e., recruited, surviving and dead plants) affecting the successional process. Here, we combine chronosequence and dynamic data to study the role of habitat filtering and limiting similarity in the old-field succession of a seasonal tropical dry forest subjected to strong climatic events (e.g., the El Nino Southern Oscillation (ENSO) and heavy rains). We documented successional changes in the phylogenetic structure and phylodiversity of regenerative communities (shrubs and trees ≤ 1 m in height), assessed phylogenetic changes in demographic components during succession, and explored the effects of interannual rainfall variation on such changes. Over five years, we monitored the dynamics of regenerative plant communities in a chronosequence of nine abandoned pastures (fallow age of 0–12 years) and three old-growth sites. For each year and site, we quantified changes in the standardized mean phylogenetic distance (MPD.OBS.Z) and phylodiversity (PD) of entire regenerative communities and recruited, surviving and dead plants. For each phylogenetic metric and demographic component, we assessed the effect of successional age, time (years), and interannual rainfall variation. Large variation in MPD.OBS.Z and PD was observed among sites and years, especially in the recently abandoned pastures. Overall, the phylogenetic community structure was clustered early in succession and became more random as succession advanced. Species from the order Fabales made especially strong contributions to the phylogenetic clustering, especially during the first years of succession. In dead plants, the MPD.OBS.Z and PD increased in the drier year due to the high mortality of plants from several clades (orders). The high prevalence of clustered and random phylogenetic structure suggests that abiotic filtering and stochastic processes are major structuring mechanisms in this successional system. Due the strong contribution of Fabales in the phylogenetic structure of regenerative communities of tropical dry forest, the effects of strong climatic events (e.g., ENSO) on Fabales could influence the old-field succession of tropical dry forests.

Restoring invasion-resistant plant communities is critical for the successful control of invasive plant species. It is based on ecological principles, such as limiting similarity, and the diversity–invasibility hypothesis, which can be used to select optimal combinations of species and determine appropriate plant density for the effective suppression of invasion due to propagule pressure. However, no attempt has been made to combine these factors in a single research framework. Here, we show for the first time the relative importance of all significant factors, including seed density, limiting similarity, diversity effect, and propagule pressure, in the invasion mechanism of Sicyos angulatus, an invasive plant species. Our results suggest that seed density, rarely explored in previous studies, is as important a determinant of invasion success as limiting similarity, diversity effect, and propagule pressure. Thus, the density-mediated mechanism must be given careful consideration for the restoration of strong invasion-resistant native plant communities.

Several studies have demonstrated that topography, as a proxy for resource availability, plays an important role in shaping tree species distribution patterns in tropical forests. Here we take a functional approach by assessing changes in plant functional traits along topographical gradients in a 20-ha forest dynamics plot in Xishuangbanna, southern Yunnan, China. We hypothesized that observed changes in trait composition and diversity along the topographic gradient should reflect tree life-history strategies linked to resource (water and nutrients) availability. We measured and compiled data of the maximum height, leaf area, seed mass and wood density of 334 tree species in 500 subplots (20 � 20 m). We found that species composition and diversity changed significantly along the topographical gradient. This change was accompanied by increasing community average seed mass and wood density, and decreasing leaf area and tree maximum height with higher elevations and steeper slopes. Higher slopes and ridges were being dominated by tree communities that, possibly due to environmental filtering, converged on a combination of traits characteristic of slow growth and low turn-over. Lower slopes and valleys, on the other hand, were dominated by tree communities that showed high trait evenness and divergence, and tree species characterized by traits indicative of fast growth and turn-over, i.e. competition for resources may be responsible for complementary trait combinations and resource use in this habitat type. Our study shows that functional traits can provide a mechanistic understanding of the processes shaping tropical tree community assembly along topographical gradients and supplements information obtained using species composition and diversity data. Some practical outcomes of our study include: (1) plant species compositional trends in tropical forests can be predicted with GIS technology, focusing on topographical gradients; (2) forest communities on ridges may become more dominant and expand in the future due to increasing drought severity; (3) a considerable number of tropical tree species may currently be under threat of local extinction due to topographically biased deforestation trends and (4) trait based survey approaches improve our understanding of the mechanisms behind species compositional trends.

Ecological theory predicts that species coexisting locally may either be functionally more dissimilar than expected by chance due to niche differentiation ('trait divergence'), or rather be functionally similar given that species bearing traits associated with low competitive ability are likely to be excluded ('trait convergence'). How the importance of trait divergence and convergence varies with productivity and depending on different plant traits still remains controversial. However assessing the effect of site productivity on meadow species assembly can help defining proper management practice for these ecosystems. We sought evidences for trait convergence and divergence across a productivity gradient over 21 wet meadows in Železné hory Mts. (Czech Republic), considering 13 plant traits. In each meadow, we recorded species composition in four 1m2 plots divided into 100 subplots, and different environmental parameters and site characteristics linked to productivity (standing biomass, soil water table and soil nutrients content). Species spatial aggregation within plots (calculated by the V-score metric on the 100 subplots per plot averaged across all plots in each meadow) was correlated with trait dissimilarity between species in a rather limited number of cases. Significant divergence was observed when traits were considered in combination or when using phylogeny as a proxy of functional similarity. Moreover, plot level patterns of trait divergence and convergence depended on local environmental characteristics. Specifically, for several of the traits studied (e.g. lifespan, combined traits) there was evidence that functional divergence of species decreased with increasing productivity. Overall from these results at fine scale we cannot rule out neutrality, but, where significant, patterns suggest stronger effects of niche differentiation between coexisting species (trait divergence) rather than convergence of functionally similar species. Nevertheless our results also support the view that increased productivity, coupled with intense competition for resources (mainly for light), can lead to the coexistence of more functionally similar species. This is relevant because biodiversity enhancement is generally achieved by maintaining low productivity in these meadows. In general our study gives insights on how species coexistence is related to life history traits and environmental variation, which is also relevant for the conservation and management of species-rich meadows. © 2013 Elsevier B.V.

A traditional question in community ecology is whether species' traits are distributed as more-or-less regularly spaced clusters. Interspecific competition has been suggested to play a role in such structuring of communities. The seminal theoretical work on limiting similarity and species packing, presented four decades ago by Robert MacArthur, Richard Levins and Robert May, has recently been extended. There is now a deeper understanding of how competitive interactions influence community structure, for instance, how the shape of competition kernels can determine the clustering of species' traits. Competition is typically weaker for greater phenotypic difference, and the shape of the dependence defines a competition kernel. The clustering tendencies of kernels interact with other effects, such as variation in resource availability along a niche axis, but the kernel shape can have a decisive influence on community structure. Here we review and further extend the recent developments and evaluate their importance.

In this review, we examine two new trait-based models of community assembly that predict the relative abundance of species from a regional species pool. The models use fundamentally different mathematical approaches and the predictions can differ considerably. Maxent obtains the most even probability distribution subject to community-weighted mean trait constraints. Traitspace predicts low probabilities for any species whose trait distribution does not pass through the environmental filter. Neither model maximizes functional diversity because of the emphasis on environmental filtering over limiting similarity. Traitspace can test for the effects of limiting similarity by explicitly incorporating intraspecific trait variation. The range of solutions in both models could be used to define the range of natural variability of community composition in restoration projects.

The influence of seasonal environmental variation on species coexistence is an ecologically important factor. Its two aspects are how seasonal variation contributes to coexistence mechanisms, and, given a seasonally varying coexistence pattern, how sensitive that coexistence is to nonstationary external influences (such as climate change). Here we develop a formula for calculating the robustness of discrete-time periodic dynamics. Robustness is defined as the sensitivity of the position of the cycle in phase space to varying model parameters. Though the results are different, the main biological conclusions are in line with those from a similar study concerning continuous-time cycles ( Barabas et al., 2012a ): species segregation in the timing of resource use or predator avoidance increases community robustness in a way that is analogous to the effects of resource partitioning. We also connect this formalism with the widely used and successful framework of Chesson (1994) , demonstrating that the merging of these two perspectives yields simplified expressions for robustness more amenable to analytical treatment. As an example, we apply our results to a two-cycle in a model of two competing annual plants with seedbanks, using our formulas to calculate the range of parameters that allow for the coexistence of the competitors. This helps us understand which components of the environmental variation the coexistence is sensitive to; in our case, the model is fairly robust against changing seed survival, moderately so against changing the variance in seed germination, and quite sensitive to changing the mean seed germination rates.

a b s t r a c t Several multi-year biodiversity experiments have shown positive species richness-productivity rela- tionships which strengthen over time, but the mechanisms which control productivity are not well understood. We used experimental grasslands (Jena Experiment) with mixtures containing different num- bers of species (4, 8, 16 and 60) and plant functional groups (1-4; grasses, legumes, small herbs, tall herbs) to explore patterns of variation in functional trait composition as well as climatic variables as predictors for community biomass production across several years (from 2003 to 2009). Over this time span, high community mean trait values shifted from the dominance of trait values associated with fast growth to trait values suggesting a conservation of growth-related resources and successful reproduction. Increasing between-community convergence in means of several productivity-related traits indicated that environ- mental filtering and exclusion of competitively weaker species played a role during community assembly. A general trend for increasing functional trait diversity within and convergence among communities sug- gested niche differentiation through limiting similarity in the longer term and that similar mechanisms operated in communities sown with different diversity. Community biomass production was primarily explained by a few key mean traits (tall growth, large seed mass and leaf nitrogen concentration) and to a smaller extent by functional diversity in nitrogen acquisition strategies, functional richness in mul- tiple traits and functional evenness in light-acquisition traits. Increasing species richness, presence of an exceptionally productive legume species (Onobrychis viciifolia) and climatic variables explained an additional proportion of variation in community biomass. In general, community biomass production decreased through time, but communities with higher functional richness in multiple traits had high pro- ductivities over several years. Our results suggest that assembly processes within communities with an artificially maintained species composition maximize functional diversity through niche differentiation and exclusion of weaker competitors, thereby maintaining their potential for high productivity.

Plant community assembly from a regional pool is largely driven by two mechanisms: environmental filtering and niche partitioning, which result in trait convergence or divergence, respectively. Although empirical evidence for both assembly mechanisms exists, the environmental conditions and traits where each of the two assembly patterns is prevalent remain unclear. We studied community assembly mechanisms in herb layer communities of temperate forest patches in NW Germany, looking at distributions of competitive and reproductive traits along gradients of soil fertility and disturbance. We also examined how community assembly patterns changed over a time span of two decades. Canopy height converged toward taller species with increasing soil fertility and increasing light availability. Most reproductive traits diverged with an increasing degree of disturbance and with increasing fertility. Comparisons over time indicated that disturbance events induced the coexistence of species with different reproductive strategies and also selected for tall species as a result of enhanced competitive pressure. Our study demonstrates that in accordance with existing hypotheses, competitive traits (e.g., canopy height) can be convergent in favorable environments. However, this convergence is associated with a divergence of traits related to other challenges (e.g., reproduction), indicating that true functional redundancy within communities does not exist. Moreover, our study shows that the expected divergence of reproductive traits at disturbed sites can be accompanied by a convergence of other traits (e.g., canopy height), indicating that several assembly mechanisms can operate simultaneously.

We develop the theory of limiting similarity and niche for structured populations with finite number of individual states (i-state). In line with a previously published theory for unstructured populations, the niche of a species is specified by the impact and sensitivity niche vectors. They describe the population's impact on and sensitivity towards the variables involved in the population regulation. Robust coexistence requires sufficient segregation of the impact, as well as of the sensitivity niche vectors. Connection between the population-level impact and sensitivity and the impact/sensitivity of the specific i-states is developed. Each i-state contributes to the impact of the population proportional to its frequency in the population. Sensitivity of the population is composed of the sensitivity of the rates of demographic transitions, weighted by the frequency and by the reproductive value of the initial and final i-states of the transition, respectively. Coexistence in a multi-patch environment is studied. This analysis is interpreted as spatial niche segregation.

This paper addresses the factors that constrain the assembly of earthworm communities in temperate land use systems. An occurrence matrix containing 44 localities and 20 species was build from data sets of different studies carried out in Haute Normandie (France). The whole data set included species lists from the three main landscape units and from a range of land use types from unmanaged forests and grasslands to intensive pastures and annual crops. We used species pool, multivariate (correspondence analysis) and null model analyse to explore how species occurring in a given community are sorted from the regional species pool. Special attention was paid to identifying non-randomness in co-occurrence and morpho-ecological patterns and to describing community patterns in response to habitat constraints. Twenty species of Lumbricidae were identified in the total species pool. We found no significant effect of landscape units on species richness and density, while land use type had a significant impact on both variables, with the lowest values in crops and forests, and the highest values in grasslands. Correspondence analysis highlighted how broad land use types, soil gradients and land use intensity influence patterns of community composition and ecological coherence. Non-random patterns in guild proportion were found in a few situations, with differences between grasslands and forests. This shows that environmental filters select species at specific scales according to particular ecological traits, resulting in a convergence of these traits between species of a given community. Local species richness never exceeded nine species and species co-occurrence was significantly lower than expected by chance and much lower than described in the literature for other taxa. Although no significant patterns of limiting similarity were observed at the scale of our study, we conclude that earthworm communities are also highly structured by competition. Our study thus agrees with the idea that both habitat and competitive constraints operate simultaneously to determine how community assembly takes place.

Riverine freshwater mussel species can be found in highly diverse communities where many similar species coexist. Mussel species potentially compete for food and space as adults, and for fish host resources during the larval (glochidial) stage. Resource partitioning at the larval stage may promote coexistence. A model of resource utilization was developed for two mussel species and analyzed to determine conditions for coexistence. Mussel species were predicted to coexist when they differed in terms of their success in contacting different fish host species; very similar strategies offered limited possibilities for coexistence. Differences in the mussel species’ maximum infestation loads on the fish hosts that coincided with differences in their fish host contact success promoted coexistence. Mussel species with a given set of trade-offs in fish host use were predicted to coexist only for a subset of relative fish host abundances, so a shift in relative fish host abundances could result in the loss of a mussel species. An understanding of the conditions for freshwater mussel species coexistence can help explain high mussel diversity in rivers and guide ongoing conservation activities.

Simple patch-occupancy models of competitive metacommunities have shown that coexistence is possible as long as there is a competition-colonization tradeoff such as that of superior competitors and dispersers. In this paper, we present a model of competition between three species in a dynamic landscape, where patches are being created and destroyed at a different rate. In our model, species interact according to a linear non-transitive hierarchy, such that species Y(3) outcompetes and can invade patches occupied by species Y(2) and this species in turn can outcompete and invade patches occupied by the inferior competitor Y(1). In this hierarchy, inferior competitors cannot invade patches of species with higher competitive ability. Analytical results show that there are regions in the parameter space where coexistence can occur, as well as regions where each of the species exists in isolation depending on species' life-history traits associated with their colonization abilities and extinction proneness as well as with the dynamics of habitat patches. In our model, the condition for coexistence depends explicitly on patch dynamics, which in turn modulate the limiting similarity for species coexistence. Coexistence in metacommunities inhabiting dynamic landscapes although possible is harder to attain than in static ones.

The focus of this paper is a general relationship proposed by May ( Amer. Natur. 107 (1973)) between the stability properties of stochastic models incorporating environmental variation and the stability properties of the deterministic models from which they are derived. The concepts of stochastic stability underlying this conjectured relationship are discussed and compared to the standard definitions of deterministic stability as well as alternative criteria for stability in stochastic models. It is shown by example that May's qualitative stability criterion does not ensure stability in any sense unless restrictive conditions on the form of the model are satisfied. Even under these conditions, the criterion, which is based on linearization, generally provides information only about the local dynamics of multispecies models. The applicability of such information to stochastic limiting similarity theory is discussed and alternative methods of analysis are proposed.

Competition in a coarse-grained heterogeneous environment is considered. It is assumed that each individual adjust has position within the macrohabitat so as to try to maximize a utility whose value is a function of position within the habitat and of the population density at that position. Robust qualitative conclusions concerning the resulting population behavior relate niche breadth, niche shift, and resource partitioning to changes in population numbers. In particular populations avoid competition by reducing coarse-grained niche overlap so that overlap is a measure not of the degree of competition but of its absence. Specific models of exploitation or interference competition give specific quantitative results which are summarized. Competition equations are derived from underlying models which imply Lotka-Volterra equations in a fine-grained environment (MacArthur, R., and Levins, R. 1967. Amer. Natur. 101 , 377–385) but which are nonlinear in per capita growth in a coarse-grained one. The findings are relevant to theoretical studies of competition, limiting similarity, species packing, and the evolution of niche position as well as to practical problems of data analysis and resource management. The extension to predator-prey interactions is outlined.

A coevolutionary model of species packing is developed which allows evolutionary adjustment in both niche position and within-phenotype niche width of from one of three competing species. The environment is specified as a single resource dimension x and availability of resources along x is given by a Gaussian curve with parameters ( x R and σ R ). The model predicts that for S species the ratio of optimal niche width (ŵ) to σR is roughly independent of σR and can be approximated by 1 S when the competitors are completely resource limited. Niche separation ( d w ) increases only moderately with increases in resource diversity σR and is greater for two than for three competing species. To the extent that the competitors do not utilize all resources, both optimal niche separation and niche width decrease. Many of the general trends in niche width and niche separation predicted by this coevolutionary model parallel those from optimal foraging theory and limiting similarity models of community structure. The coevolutionary model stands out, however, in the singularly high values predicted for niche separation, making coevolved communities highly invadeable. Hence, the theory suggests, as some empirical evidence indicates, that coevolved competition communities can only exist as such on remote islands or in other habitats which might be free from invasion by outside species. For such communities, resource diversity (σR) should have much less effect on species packing than either species number itself or the extent that these competitors are resource limited.