Reorganization of interaction networks modulates the persistence of species in late successional stages

On the limits to invasion prediction using coexistence outcomes

The dynamics of ecological communities in nature are typically characterized by probabilistic processes involving invasion dynamics. Because of technical challenges, however, the majority of theoretical and experimental studies have focused on coexistence dynamics. Therefore, it has become central to understand the extent to which coexistence outcomes can be used to predict analogous invasion outcomes relevant to systems in nature. Here, we study the limits to this predictability under a geometric and probabilistic Lotka-Volterra framework. We show that while individual survival probability in coexistence dynamics can be fairly closely translated into invader colonization probability in invasion dynamics, the translation is less precise between community persistence and community augmentation, and worse between exclusion probability and replacement probability. These results provide a guiding and testable theoretical framework regarding the translatability of outcomes between coexistence and invasion outcomes when communities are represented by Lotka-Volterra dynamics under environmental uncertainty.

Interaction Networks Help to Infer the Vulnerability of the Saproxylic Beetle Communities That Inhabit Tree Hollows in Mediterranean Forests

Insect communities are facing contrasting responses due to global change. However, knowledge on impacts of communities' reorganizations is scarce. Network approaches could help to envision community changes in different environmental scenarios. Saproxylic beetles were selected to examine long-term variations in insect interaction/diversity patterns and their vulnerability to global change. We evaluated interannual differences in network patterns in the tree hollow-saproxylic beetle interaction using absolute samplings over an 11-year interval in three Mediterranean woodland types. We explored saproxylic communities' vulnerability to microhabitat loss via simulated extinctions and by recreating threat scenarios based on decreasing microhabitat suitability. Although temporal diversity patterns varied between woodland types, network descriptors showed an interaction decline. The temporal beta-diversity of interactions depended more on interaction than on species turnover. Interaction and diversity temporal shifts promoted less specialized and more vulnerable networks, which is particularly worrisome in the riparian woodland. Network procedures evidenced that saproxylic communities are more vulnerable today than 11 years ago irrespective of whether species richness increased or decreased, and the situation could worsen in the future depending on tree hollow suitability. Network approaches were useful for predicting saproxylic communities' vulnerability across temporal scenarios and, thus, for providing valuable information for management and conservation programs.

Short-Term Temporal Patterns in Herbivore Beetle Assemblages in Polyculture Neotropical Forest Plantations

Although insect herbivorous communities in tropical forests are known to exhibit strong seasonality, few studies have systematically assessed temporal patterns of variation in community structure and plant-herbivore interactions in early successional arboreal communities. We assessed seasonal and interannual variation of the diversity and composition of herbivorous beetles and the tree-herbivore network in a recently established polyculture forest plantation, during the dry and the rainy seasons of 2012 and of 2013. Species richness was similar between years, while the ecological diversity was higher in 2012. Comparing seasons, no differences were found in 2012, whereas in 2013, the species richness and ecological diversity were higher during the dry season. The species composition differed radically across years and seasons. Moreover, a quantitative nested pattern was consistently found across both temporal scales, more influenced by species densities. We found temporal changes in the species strength, whereas connectance and interaction evenness remained stable. Rapid temporal changes in the structural complexity of recently established polyculture plantations and the availability and quality of the trophic resources they offer may act as drivers of beetle diversity patterns, promoting rapid variation in herbivore composition and some interacting attributes. Nonetheless, network structure, connectance, and interaction evenness remained similar, suggesting that reorganizations in the distribution of species may determine the maintenance of the patterns of interaction. Further work assessing long-term temporal dynamics of herbivore beetle assemblages are needed to more robustly relate diversity and interaction patterns to biotic and abiotic factors and their implications in management programs.

Generalizing game-changing species across microbial communities

Microbes form multispecies communities that play essential roles in our environment and health. Not surprisingly, there is an increasing need for understanding if certain invader species will modify a given microbial community, producing either a desired or undesired change in the observed collection of resident species. However, the complex interactions that species can establish between each other and the diverse external factors underlying their dynamics have made constructing such understanding context-specific. Here we integrate tractable theoretical systems with tractable experimental systems to find general conditions under which non-resident species can change the collection of resident communities-game-changing species. We show that non-resident colonizers are more likely to be game-changers than transients, whereas game-changers are more likely to suppress than to promote resident species. Importantly, we find general heuristic rules for game-changers under controlled environments by integrating mutual invasibility theory with in vitro experimental systems, and general heuristic rules under changing environments by integrating structuralist theory with in vivo experimental systems. Despite the strong context-dependency of microbial communities, our work shows that under an appropriate integration of tractable theoretical and experimental systems, it is possible to unveil regularities that can then be potentially extended to understand the behavior of complex natural communities.

Observed Ecological Communities Are Formed by Species Combinations That Are among the Most Likely to Persist under Changing Environments

AbstractDespite the rich biodiversity found in nature, it is unclear to what extent some combinations of interacting species, while conceivable in a given place and time, may never be realized. Yet solving this problem is important for understanding the role of randomness and predictability in the assembly of ecological communities. Here we show that the specific combinations of interacting species that emerge from the ecological dynamics within regional species pools are not all equally likely to be seen; rather, they are among the most likely to persist under changing environments. First, we use niche-based competition matrices and Lotka-Volterra models to demonstrate that realized combinations of interacting species are more likely to persist under random parameter perturbations than the majority of potential combinations with the same number of species that could have been formed from the regional pool. We then corroborate our theoretical results using a 10-year observational study, recording 88 plant-herbivore communities across three different forest successional stages. By inferring and validating plant-mediated communities of competing herbivore species, we find that observed combinations of herbivores have an expected probability of species persistence higher than half of all potential combinations. Our findings open up the opportunity to establish a formal probabilistic and predictive understanding of the composition of ecological communities.

Seeing through the static: the temporal dimension of plant-animal mutualistic interactions

Most studies of plant-animal mutualistic networks have come from a temporally static perspective. This approach has revealed general patterns in network structure, but limits our ability to understand the ecological and evolutionary processes that shape these networks and to predict the consequences of natural and human-driven disturbance on species interactions. We review the growing literature on temporal dynamics of plant-animal mutualistic networks including pollination, seed dispersal and ant defence mutualisms. We then discuss potential mechanisms underlying such variation in interactions, ranging from behavioural and physiological processes at the finest temporal scales to ecological and evolutionary processes at the broadest. We find that at the finest temporal scales (days, weeks, months) mutualistic interactions are highly dynamic, with considerable variation in network structure. At intermediate scales (years, decades), networks still exhibit high levels of temporal variation, but such variation appears to influence network properties only weakly. At the broadest temporal scales (many decades, centuries and beyond), continued shifts in interactions appear to reshape network structure, leading to dramatic community changes, including loss of species and function. Our review highlights the importance of considering the temporal dimension for understanding the ecology and evolution of complex webs of mutualistic interactions.

Structural forecasting of species persistence under changing environments

The persistence of a species in a given place not only depends on its intrinsic capacity to consume and transform resources into offspring, but also on how changing environmental conditions affect its growth rate. However, the complexity of factors has typically taken us to choose between understanding and predicting the persistence of species. To tackle this limitation, we propose a probabilistic approach rooted on the statistical concepts of ensemble theory applied to statistical mechanics and on the mathematical concepts of structural stability applied to population dynamics models - what we call structural forecasting. We show how this new approach allows us to estimate a probability of persistence for single species in local communities; to understand and interpret this probability conditional on the information we have concerning a system; and to provide out-of-sample predictions of species persistence as good as the best experimental approaches without the need of extensive amounts of data.

Towards a Probabilistic Understanding About the Context-Dependency of Species Interactions

Observational and experimental studies have shown that an interaction class between two species (be it mutualistic, competitive, antagonistic, or neutral) may switch to a different class, depending on the biotic and abiotic factors within which species are observed. This complexity arising from the evidence of context-dependencies has underscored a difficulty in establishing a systematic analysis about the extent to which species interactions are expected to switch in nature and experiments. Here, we propose an overarching theoretical framework, by integrating probabilistic and structural approaches, to establish null expectations about switches of interaction classes across environmental contexts. This integration provides a systematic platform upon which it is possible to establish new hypotheses, clear predictions, and quantifiable expectations about the context-dependency of species interactions.

Plant speciation in the age of climate change

BACKGROUND

Species diversity is likely to undergo a sharp decline in the next century. Perhaps as many as 33 % of all plant species may expire as a result of climate change. All parts of the globe will be impacted, and all groups of organisms will be affected. Hundreds of species throughout the world have already experienced local extinction.

PERSPECTIVES

While thousands of species may become extinct in the next century and beyond, species formation will still occur. I consider which modes of plant species formation are likely to prevail in the next 500 years. I argue that speciation primarily will involve mechanisms that produce reproductively isolated lineages within less (often much less) than 100 generations. I will not especially consider the human element in promoting species formation, because it will continue and because the conclusions presented here are unaffected by it. The impact of climate change may be much more severe and widespread.

CONCLUSIONS

The most common modes of speciation likely to be operative in the next 500 years ostensibly will be auto- and allopolyploidy. Polyploid species or the antecedents thereof can arise within two generations. Moreover, polyploids often have broader ecological tolerances, and are likely to be more invasive than are their diploid relatives. Polyploid species may themselves spawn additional higher level polyploids either through crosses with diploid species or between pre-existing polyploids. The percentage of polyploid species is likely to exceed 50 % within the next 500 years vs. 35 % today. The stabilized hybrid derivatives (homoploid hybrid speciation) could emerge within a hundred generations after species contact, as could speciation involving chromosomal rearrangements (and perhaps number), but the number of such events is likely to be low. Speciation involving lineage splitting will be infrequent because the formation of substantive pre- and post-zygotic barriers typically takes many thousands of years.

Structural stability as a consistent predictor of phenological events

The timing of the first and last seasonal appearance of a species in a community typically follows a pattern that is governed by temporal factors. While it has been shown that changes in the environment are linked to phenological changes, the direction of this link appears elusive and context-dependent. Thus, finding consistent predictors of phenological events is of central importance for a better assessment of expected changes in the temporal dynamics of ecological communities. Here we introduce a measure of structural stability derived from species interaction networks as an estimator of the expected range of environmental conditions compatible with the existence of a community. We test this measure as a predictor of changes in species richness recorded on a daily basis in a high-arctic plant-pollinator community during two spring seasons. We find that our measure of structural stability is the only consistent predictor of changes in species richness among different ecological and environmental variables. Our findings suggest that measures based on the notion of structural stability can synthesize the expected variation of environmental conditions tolerated by a community, and explain more consistently the phenological changes observed in ecological communities.

Will a small randomly assembled community be feasible and stable?

How likely is it that few species can randomly assemble into a feasible and stable community? Some studies have answered that as long as the community is feasible, it will nearly always be stable. In contrast, other studies have answered that the likelihood is almost null. Here, we show that the origin of this debate has been the underestimation of the association of the parameter space of intrinsic growth rates with the feasibility and stability properties of small randomly-assembled communities. In particular, we demonstrate that not all parameterizations and sampling distributions of intrinsic growth rates lead to the same probabilities of stability and feasibility, which could mistakenly lead to under- or overestimate the stability properties of feasible communities. Additionally, we find that stability imposes a filtering of species abundances "towards" more even distributions in small feasible randomly-assembled communities. This indicates that the stability of feasible communities is inherently linked to the starting distribution of species abundances, a characteristic that many times has been ignored, but should be incorporated in manageable lab and field experiments. Overall, the return to this debate is a central reminder that a more systematic exploration of the feasible parameter space is necessary to derive general conclusions about the stability properties of ecological communities.

Structural stability of nonlinear population dynamics

In population dynamics, the concept of structural stability has been used to quantify the tolerance of a system to environmental perturbations. Yet, measuring the structural stability of nonlinear dynamical systems remains a challenging task. Focusing on the classic Lotka-Volterra dynamics, because of the linearity of the functional response, it has been possible to measure the conditions compatible with a structurally stable system. However, the functional response of biological communities is not always well approximated by deterministic linear functions. Thus, it is unclear the extent to which this linear approach can be generalized to other population dynamics models. Here, we show that the same approach used to investigate the classic Lotka-Volterra dynamics, which is called the structural approach, can be applied to a much larger class of nonlinear models. This class covers a large number of nonlinear functional responses that have been intensively investigated both theoretically and experimentally. We also investigate the applicability of the structural approach to stochastic dynamical systems and we provide a measure of structural stability for finite populations. Overall, we show that the structural approach can provide reliable and tractable information about the qualitative behavior of many nonlinear dynamical systems.