Tree species diversity in temperate and tropical forest gaps: The role of lottery recruitment

How canopy gaps promote tree species coexistence in temperate and tropical forests is reviewed. Given that evidence for traditional resource-based niche partitioning of canopy gaps is weak, lottery recruitment, whereby colonization of vacant living space is random with respect to species identity, may be key to maintaining diversity. Gap formation events are not highly predictable and species’ propagules tend to be patchily distributed. For these reasons the predictability of gap-phase recruitment is low and lottery principles apply. Recruitment limitation from discontinuities in species’ propagule supplies in space and time may permit lottery recruitment in temperate and tropical forests. However, the relative importance of recruitment limitation in species-rich versus species-poor communities is unclear. Although lottery models with purely random recruitment in vacant sites can be applied, relatively complex models appear to better capture the essential features of forest community dynamics. For example, models with recruitment probabilities weighted by species abundance can produce the non-random trajectories often observed in communities. Models with local competitive displacement by individuals of dominant species can also produce non-random trajectories. Models with spatial structure, localized competition, and dispersal limitations may provide further insight into the effects of biotic interactions and recruitment limitation on forest dynamics.     

The competition-colonization trade-off is dead; long live the competition-colonization trade-off

When applied at the individual patch level, the classic competition-colonization models of species coexistence assume that propagules of superior competitors can displace adults of inferior competitors (displacement competition). But if adults are invulnerable to displacement by propagules (as trees are to seeds), and propagules compete to replace adults that die for reasons independent of the outcome of juvenile competition (a lottery system), a competition-colonization trade-off alone is not able to produce coexistence. However, we show that coexistence is possible if patch density varies spatially, such that it becomes a niche axis. We also show how a dispersal-fecundity trade-off can partition variation in patch density. We discuss the application of these models to empirical systems. An important implication of communities coexisting via variation in patch density is that the amount of habitat loss necessarily interacts with the pattern of loss in affecting extinctions, invasions, and coexistence, in contrast to displacement competition models, for which the spatial pattern of loss is not important or is less important. Finally, with respect to mechanisms promoting coexistence, we suggest that trade-offs between different stages of colonization could be far more common in nature than a trade-off between competitive ability and colonization ability.     

Spatial heterogeneity of mortality and temporal fluctuation in fertility promote coexistence but not vice versa: a random-community approach

Both spatial heterogeneity and temporal fluctuation of the environment are important mechanisms promoting species coexistence, but they work in different manners. We consider many pairs of species with randomly generated survivorship and fertility in the lottery model, and examine how the variability in demographic processes affects the outcome of competition. The results are: [1] Coexistence is easier if habitat difference in mortality is greater, or if year-to-year variation in reproductive rate is larger. But neither habitat-difference in fertility nor temporal variation in mortality promotes coexistence. [2] Mean fertility does not affect the outcome if CV remains constant. In contrast, enhanced mean mortality decreases the fraction of coexisting pairs if the environment fluctuates temporally. [3] We also investigate the effect of limited dispersal of propagules between habitats. Compared with the complete mixing case, the fraction of coexisting pairs is clearly enhanced if the spatial heterogeneity is the major source of environmental variation, but shows slight increase if the temporal fluctuation is dominant. We conclude that spatial heterogeneity is likely to work more effectively in promoting species coexistence than temporal fluctuation, especially when the species suffer relatively high mortality, and disperse their propagules in a limited spatial scale.     

Variation in the degree of specialization can maintain local diversity in model communities

We hypothesize that the continuum between generalist and specialist adaptations is an important general trade-off axis in the maintenance of local diversity, and we explore this idea with a simple model in which there are patch types to which species arrive as propagules and compete. Each patch type is defined by a competitive ranking of all species. A highly specialist species is the top competitor in one patch type but has a relatively low average ranking across different patch types, while a generalist species has a high average rank across patch types but is not the top competitor in any patch type. We use random dispersal and vary the fecundity of all species together to vary total propagule density and therefore recruitment limitation and density-dependent mortality. When fecundity is very high, each patch becomes occupied by its specialist species and generalists go extinct, so the number of species at equilibrium is equal to the number of patch types. If fecundity is very low, generalists dominate and specialists go extinct. There is a range of fecundity levels in which specialists, generalists, and intermediates coexist, and the number of species is substantially greater than the number of patch types. While coexistence of specialists and generalists has been considered a problem in evolutionary ecology, our results suggest to the contrary that this trade-off contributes to the maintenance of local diversity.     

Life-cycle switching and coexistence of species with no niche differentiation

The increasing evidence of coexistence of cryptic species with no recognized niche differentiation has called attention to mechanisms reducing competition that are not based on niche-differentiation. Only sex-based mechanisms have been shown to create the negative feedback needed for stable coexistence of competitors with completely overlapping niches. Here we show that density-dependent sexual and diapause investment can mediate coexistence of facultative sexual species having identical niches. We modelled the dynamics of two competing cyclical parthenogens with species-specific density-dependent sexual and diapause investment and either equal or different competitive abilities. We show that investment in sexual reproduction creates an opportunity for other species to invade and become established. This may happen even if the invading species is an inferior competitor. Our results suggests a previously unnoticed mechanism for species coexistence and can be extended to other facultative sexual species and species investing in diapause where similar density-dependent life-history switches could act to promote coexistence.     

The age-structured lottery model

The lottery model of competition between species in a variable environmental has been influential in understanding how coexistence may result from interactions between fluctuating environmental and competitive factors. Of most importance, it has led to the concept of the storage effect as a mechanism of species coexistence. Interactions between environment and competition in the lottery model stem from the life-history assumption that environmental variation and competition affect recruitment to the adult population, but not adult survival. The strong role of life-history attributes in this coexistence mechanism implies that its robustness should be checked for a variety of life-history scenarios. Here, age structure is added to the adult population, and the results are compared with the original lottery model. This investigation uses recently developed shape characteristics for mortality and fecundity schedules to quantify the effects of age structure on the long-term low-density growth rate of a species in competition with its competitor when applying the standard invasibility coexistence criterion. Coexistence conditions are found to be affected to a small degree by the presence of age structure in the adult population: Type III mortality broadens coexistence conditions, and type I mortality makes them narrower. The rates of recovery from low density for coexisting species, and the rates of competitive exclusion in other cases, are modified to a greater degree by age structure. The absolute rates of recovery or decline of a species from low density are increased by type I mortality or early peak reproduction, but reduced by type III mortality or late peak reproduction. Analytical approximations show how the most important effects can be considered as simple modifications of the long-term low-density growth rates for the original lottery model.     

Aggregation and the Coexistence of Competing Parasitoid Species

In nature, many insect species are attacked by more than one specialized species of parasitoid. We examine whether parasitoid aggregation among patches containing hosts can promote the coexistence of specialized parasitoids on the same host species. We construct models to analyze the effects of three types of parasitoid aggregation: direct density-dependent, inverse density-dependent, and density-independent aggregation. All three types of aggregation may facilitate coexistence, provided the parasitoid species show behavioral differences that produce different patterns of aggregation. By deriving general conditions of coexistence of parasitoids, we show that all three types of aggregation act to facilitate coexistence in the same way—by increasing the covariance between the distributions of susceptible hosts and the least common parasitoid. Although they act in the same way, in general the effect of density-independent aggregation in facilitating coexistence is greater than either direct or inverse density-dependent aggregation. This suggests that density-independent aggregation may have the greatest potential to facilitate the coexistence of specialize parasitoids using the same host.     

Biodiversity in South African fynbos and Mediterranean heathland

Abstract. The theory of convergence predicts that, given similar selective regimes, both present and past, unrelated ecological communities will show similar attributes. Mild Pleistocene climate, highly infertile soils, and similar fire regimes explain the remarkable convergence between mediterranean‐type vegetation from South Africa (fynbos) and Australia (kwongan). Heathlands in the Aljibe Mountains, at the western end of the Mediterranean basin, constitute a single vegetation type within the Mediterranean region. We studied the association between endemism and plant life form in a flora from environmentally similar areas of the South African Cape region (fynbos) and the Aljibe Mountains by contingency table analysis. We included two non‐acid, neighbouring areas to the latter region in the analysis as contrasts. We also compared the patterns of variation in three components of biodiversity (species richness, endemism level and taxonomic singularity) of fynbos and Aljibe heathland woody plant communities along similar soil fertility gradients by means of two‐way ANOVAs. At the regional (flora) level, our results show two common features in the biological aspects of endemism between the two regions: (1) edaphic endemism and (2) association of endemism with the shrub growth form. At the community level, we detected strong similarities in the patterns of variation of endemism and taxonomic singularity of woody communities from both regions along an ecological gradient related to soil fertility. We interpret these similarities, both at the regional and community levels, as suggestive of convergence between fynbos and Aljibe heathland.     

What predicts the richness of seeder and resprouter species in fire‐prone Cape fynbos: Rainfall reliability or vegetation density?

In ecosystems subject to regular canopy fires, woody species have evolved two general strategies of post‐fire regeneration. Seeder species are killed by fire and populations regenerate solely by post‐fire recruitment from a seed bank. Resprouter species survive fire and regenerate by vegetative regrowth from protected organs. Interestingly, the abundance of these strategies varies along environmental gradients and across regions. Two main hypotheses have been proposed to explain this spatial variation: the gap dependence and the environmental‐variability hypotheses. The gap‐dependence model predicts that seeders are favoured in sparse vegetation (vegetation gaps allowing effective post‐fire recruitment of seedlings), while resprouters are favoured in densely vegetated sites (seedlings being outcompeted by the rapid crown regrowth of resprouters). The environmental‐variability model predicts that seeders would prevail in reliable rainfall areas, whereas resprouters would be favoured in areas under highly variable rainfall that are prone to severe dry events (leading to high post‐fire seedling mortality). We tested these two models using distribution data, captured at the scale of quarter‐degree cells, for seeder and resprouter species of two speciose shrub genera (Aspalathus and Erica) common in fire‐prone fynbos ecosystems of the mediterranean‐climate part of the Cape Floristic Region. Contrary to the predictions of the gap‐dependence model, species number of both resprouters and seeders increased with values of the Normalized Difference Vegetation Index (a widely used surrogate for vegetation density), with a more marked increase for seeders. The predictions of the environmental‐variability hypothesis, by contrast, were not refuted by this study. Seeder and resprouter species of both genera showed highest richness in environments with high rainfall reliability. However, with decreasing reliability, seeder numbers dropped more quickly than those of resprouters. We conclude that the environmental‐variability model is better able to explain the abundance of woody seeder and resprouter species in Southern Hemisphere fire‐prone shrublands (fynbos and kwongan) than the gap‐dependence model.     

Evidence for lottery recruitment in Mediterranean old fields

Abstract. The hypothesis of lottery establishment ( Sale 1977 ) explains coexistence of species with similar niches through processes of stochastic recruitment. This initial idea forms the basis for a variety of mathematical models, but has not been tested empirically. This study is a field investigation of lottery establishment for plants with a seed bank, using Canonical Correspondence Analysis to compare the compositions of the vegetation and the seed bank according to different hypotheses on the mechanisms of establishment. This method was used for a data set from old fields from southern France. The weighted lottery (i.e. a random draw from the seed pool, weighted by the frequencies of each species) appeared as the best suited hypothesis to explain the high degree of similarity between the vegetation and the seed bank and the relative spatial distributions of the species. Several mechanisms are probably interacting, depending on the life histories of the species. Modelling and experimental approaches are needed to further test the hypothesis of lottery recruitment.     

Fire season effects on the recruitment of non‐sprouting serotinous Proteaceae in the eastern (bimodal rainfall) fynbos biome,South Africa

Abstract Research in Mediterranean‐climate shrublands in both South Africa and Australia shows that recruitment of proteoid shrubs (non‐sprouting, serotinous Proteaceae) is best after warm‐season (summer and autumn) fires and worst after cool‐season (winter and spring) ones. This pattern has been attributed to post‐dispersal seed attrition as well as size of pre‐dispersal seed reserves. Here we investigate patterns of post‐fire recruitment for four proteoid species in the eastern part of South Africa's fynbos biome, which has a bimodal (spring and autumn) rainfall regime. Despite the lack of significant differences in recruitment between cool‐ and warm‐season burns, we find some evidence for favourable recruitment periods following fires in spring and autumn, immediately before, and coinciding with, the bimodal rainfall peaks. This suggests that enhanced recruitment is associated with conditions of high soil moisture immediately after the fire, and that rapid germination may minimize post‐dispersal seed attrition. In two of the species, we also find a shift from peak flowering in winter and spring in the Mediterranean‐climate part of the fynbos biome, to summer and autumn flowering in the eastern part. Because these two species are only weakly serotinous, warm‐season flowering would result in maximal seed banks in spring, which could explain the spring recruitment peak, but not the autumn one. We conclude that eastern recruitment patterns differ significantly from those observed in the western and central parts of the biome, and that fire management protocols for the east, which are currently based on data and experience from the winter‐rainfall fynbos biome, need to be adjusted accordingly. Fire managers in the eastern fynbos biome should be less constrained by requirements to burn within a narrow seasonal range, and should therefore be in a better position to apply the required management burns.     

How flocculation can explain coexistence in the chemostat

We study a chemostat model in which two microbial species grow on a single resource. We show that species coexistence is possible when the species which would normally win the exclusive competition aggregates in flocs. Our mathematical analysis exploits the fact that flocculation is fast compared to biological growth, a common hypothesis in floc models. A numerical study shows the validity of this approach in a large parameter range. We indicate how our model yields a mechanistic justification for the so-called density-dependent growth.     

How flocculation can explain coexistence in the chemostat

We study a chemostat model in which two microbial species grow on a single resource. We show that species coexistence is possible when the species which would normally win the exclusive competition aggregates in flocs. Our mathematical analysis exploits the fact that flocculation is fast compared to biological growth, a common hypothesis in floc models. A numerical study shows the validity of this approach in a large parameter range. We indicate how our model yields a mechanistic justification for the so-called density-dependent growth.     

The stabilizing effect of a random environment

It is shown that the lottery competition model permits coexistence in a stochastic environment, but not in a constant environment. Conditions for coexistence and competitive exclusion are determined. Analysis of these conditions shows that the essential requirements for coexistence are overlapping generations and fluctuating birth rates which ensure that each species has periods when it is increasing. It is found that a species may persist provided only that it is favored sufficiently by the environment during favorable periods independently of the extent to which the other species is favored during its favorable periods.Coexistence is defined in terms of the stochastic boundedness criterion for species persistence. Using the lottery model as an example this criterion is justified and compared with other persistence criteria. Properties of the stationary distribution of population density are determined for an interesting limiting case of the lottery model and these are related to stochastic boundedness. An attempt is then made to relate stochastic boundedness for infinite population models to the behavior of finite population models.     

Synchronized reproduction promotes species coexistence through reproductive facilitation

Theories for species coexistence often emphasize niche differentiation and temporal segregation of recruitment to avoid competition. Recent work on mutualism suggested that plant species sharing pollinators provide mutual facilitation when exhibit synchronized reproduction. The facilitation on reproduction may enhance species persistence and coexistence. Theoretical ecologists paid little attention to such indirect mutualistic systems by far. We propose a new model for a two-species system using difference equations. The model focuses on adult plants and assumes no resource competition between these well-established individuals. Our formulas include demographic parameters, such as mortality and recruitment rates, and functions of reproductive facilitation. Both recruitment and facilitation effects reach saturation levels when flower production is at high levels. We conduct mathematical analyses to assess conditions of coexistence. We establish demographical conditions permitting species coexistence. Our analyses suggest a “rescue” effect from a “superior” species to a “weaker” species under strong recruitment enhancement effect when the later is not self-sustainable. The facilitation on rare species may help to overcome Allee effect.     

The trade-off between vegetative and generative reproduction among angiosperms influences regional hydrochorous propagule pressure

Aim Local communities are subject to spatiotemporal contingencies of landscape processes; community assembly is thus often considered to be unpredictable and idiosyncratic. However, evolved trade‐offs of species’ life histories may set distinct constraints on the assembly of species communities. In plants, the recruitment and invasion success of species into communities depend primarily on the number of propagules available and on their generative or vegetative character. Life‐history trade‐offs prevent individual plants from producing large numbers of both generative and vegetative propagules, but it is not clear whether this constrains their availability at the landscape scale. We thus tested whether: (1) the observed relationship between generative and vegetative propagules deviates from the null expectation stating that species contributing the bulk of generative propagules to the propagule rain should also contribute the bulk of vegetative propagules; and (2) whether vegetative and generative propagule pressures are negatively correlated once species abundance in the regional pool is accounted for. Location A large riparian landscape in the Netherlands. Methods Analyses were based on an extensive trapping of floating propagules (214,049 propagules of 47 species), and a rough proxy of species abundance across the entire pool. We used both species and phylogenetically independent contrasts as data points, and accounted for variation in size of generative propagules. Results Both hypotheses were confirmed. Numbers of generative and vegetative propagules trapped per species were significantly negatively correlated (r = ?0.33; t₄₅ = ?2.61, P = 0.006) and thus strongly deviated from the null expectation. This was confirmed by analyses accounting for variation in species abundance across the species pool, and in the size of generative propagules. Main conclusions The results indicate that plant recruitment and community assembly across streams may be influenced by the way individual plants allocate their resources between competing life‐history functions. Life‐history evolution across angiosperms might thus have constrained the present‐day assembly of local communities.     

Density-dependent cooperation as a mechanism for persistence and coexistence

To overcome stress, such as resource limitation, an organism often needs to successfully mediate competition with other members of its own species. This may favor the evolution of defective traits that are harmful to the species population as a whole, and that may lead to its dilution or even to its extinction (the tragedy of the commons). Here, we show that this phenomenon can be circumvented by cooperation plasticity, in which an individual decides, based on environmental conditions, whether to cooperate or to defect. Specifically, we analyze the evolution of density-dependent cooperation. In our model, the population is spatially subdivided, periodically remixed, and comprises several species. We find that evolution pushes individuals to be more cooperative when their own species is at lower densities, and we show that not only could this cooperation prevent the tragedy of the commons, but it could also facilitate coexistence between many species that compete for the same resource.     

Species coexistence by permanent spatial heterogeneity in a lottery model

We studied the effect of permanent spatial heterogeneity in promoting species coexistence in a lottery model. The system consisted of multiple habitats, each composed of a number of sites occupied by adults of two species. Larvae produced from different habitats were mixed in a common pool. When an adult died, the vacant site became occupied by an individual randomly chosen from the larval pool. If there were n habitats, there could be up to n-1 internal equilibria with both species in addition to two single-species equilibria. These equilibria and their local stability can be calculated from a single function, indicating the difference among species in their average lifetime reproductive success. Our main result is that between-habitat variation in the ratio of mortalities of two species promotes coexistence, while that of reproductive rates does not. This conclusion is the opposite of the role of temporal variation in the standard lottery model, in which between-year variation in the reproductive rate, but not that in the mortalities, promotes coexistence.     

Conspecific density dependence in seedlings varies with species shade tolerance in a wet tropical forest

Density-dependent seedling mortality could increase with a species relative abundance, thereby promoting species coexistence. Differences among species in light-dependent mortality also could enhance coexistence via resource partitioning. These compatible ideas rarely have been considered simultaneously. We developed models of mortality as functions of irradiance and local conspecific density (LCD) for seedlings of 53 tropical woody species. Species varied in mortality responses to these factors, but mortality consistently increased with shading and LCD. Across species, density-dependent mortality on a per-neighbour basis was inversely related to species community abundance, but higher LCD in more common species resulted in a weak relationship between species abundance and density-dependent mortality scaled to species maximum LCD. Species mortality responses to shading and maximum LCD were strongly and positively correlated. Our results suggest that species differences in density-dependent mortality are more strongly related to physiologically based life-history traits than biotic feedbacks related to community abundance.     

How might recruitment research on coral-reef fishes help manage tropical reef fisheries?

Abstract ‘Ecologic’ reef fishes (basic research subjects) and ‘Economic’ reef fishes (exploited by humans) share fundamental early life-history attributes of small, widely dispersed planktonic eggs, larvae, and (for some species in both groups) pelagic juveniles. These attributes predispose the open populations of species in both groups to limitation resulting from environmentally induced fluctuations in recruitment from planktonic/pelagic to benthic stages. Rates of movement within and among reefs, one of several postrecruitment processes likely to be subject to density-dependent regulation, may differ between Ecologies (mostly small-bodied) and Economics (generally larger-bodied). This is because of differences between species in the two groups in size-related differences in the home ranges of individuals. Existing data, however, neither support the notion that natural growth and mortality rates basically differ between the adults of Ecological and Economic species, nor that the generally larger home ranges of larger-bodied adult Economics are more subject to density-dependent control. Further, the small-bodied young-of-year juveniles of both groups on average probably have similar growth and mortality rates and small individual home ranges that are equivalently affected by density dependence. In conclusion I argue that, because of fundamental similarities in the sizes and durations of planktonic propagules and spawning periodicities, certain Ecologies and Economics may comprise a single recruitment guild. Coefficients of growth and mortality for postsettlement Ecologies also may resemble, and be applied as preliminary proxies for, analogous coefficients for Economic species. The efficacy of management strategies such as harvest refugia may differ for Ecological and Economic species, however, depending on whether the refugia are used to counter growth or recruitment overfishing.